(navigation image)
Home American Libraries | Canadian Libraries | Universal Library | Community Texts | Project Gutenberg | Children's Library | Biodiversity Heritage Library | Additional Collections
Search: Advanced Search
Anonymous User (login or join us)
Upload
See other formats

Full text of "Malacologia"

HARVARD UNIVERSITY 




LIBRARY 

OF THE 

DEPARTMENT OF MOLLUSKS 

IN THE 

Museum of Comparative Zoology 



Gift of: 



VOL 38, NO. 1-2 1996 



MALACOLOGIA 



International Journal of Malacology 
Revista Internacional de Malacologia 
Journal International de Malacologie 
Международный Журнал Малакологии 
Internationale Malakologische Zeitschrift 



Publication dates 

Vol. 28, No. 1-2 19 Jan. 1988 

Vol. 29, No. 1 28 Jun. 1988 

Vol. 29, No. 2 16 Dec. 1988 

Vol. 30, No. 1-2 1 Aug. 1989 

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 MALACOLOGIA 1996 

CONTENTS 

FADWA A. ATTIGA & HAMEED A. AL-HAJJ 

Ultrastructural Study of Euspermiogenesis in Clypeomorus Bifasciata and Clypeo- 
morus Tuberculatus (Prosobranchia: Cerithiidae) With Emphasis on Acrosome 
Formation 47 

RÜDIGER BIELER & RICHARD E. PETIT 

Additional Notes on Nomina First Introduced by Tetsuaki Kira in "Coloured Illustra- 
tions of the Shells of Japan" 33 

M. E. CHASE & R. С BAILEY 

Recruitment of Dreissena Polymorpha: Does the Presence and Density of Conspe- 

cifics Determine the Recruitment Density and Pattern in a Population? 19 

KENNETH С EMBERTON 

Microsculptures of Convergent and Divergent Polygyrid Land-Snail Shells 67 

KENNETH С EMBERTON, TIMOTHY A. PEARCE & ROGER RANDALANA 

Quantitatively Sampling Land-Snail Species Richness in Madagascan Rainforests . 203 

MARÍA FERNANDA LÓPEZ ARMENGOL 

Taxonomic Revision of Potamolithus Agapetus Pilsbry, 1911, and Potamolithus 
Buschii (Frauenfeld, 1865) (Gastropoda: Hydrobiidae) 1 

MARTIN HAASE & ERHARD WAWRA 

The Genital System of Acochlidium fijiense (Opisthobranchia: Acochlidioidea) and its 
Inferred Function 1 43 

WALTER R. HOEH & MARK E. GORDON 

Criteria for the Determination of Taxonomic Boundaries in Freshwater Unionoids 
(Bivalvia: Unionoida): Comments on Stiven and Alderman (1992) 223 

G. M. KUCHENMEISTER, D. J. PRIOR & I. G. WELSFORD 

Ouantification of the Development of the Cephalic Sac and Podocyst in the Terres- 
trial Gastropod Umax Maximus L 1 53 

RICHARD E. PETIT & RÜDIGER BIELER 

On The New Names Introduced in the Various Printings of "Shells of the World in 
Colour" [Vol. I by Tadashige Habe and Kiyoshi Ito; Vol. II by Tadashige Habe and 
Sadao Kosuge] 35 

DR. F. D. POR & DR. R. M. POLYMENI 

A Call for a New International Congress of Zoology 229 

PETER D, ROOPNARINE 

Systematics, Biogeography and Extinction of Chionine Bivalves (Bivalvia: Veneridae) 

in Tropical America: Early Oligocene-Recent 1 03 

LUIZ RICARDO L. SIMONE 

Anatomy and Systematics of Buccinanops Gradatus (Deshayes, 1844) and Bucci- 
nanops Moniliferus (Kiener, 1834) (Neogastropoda, Muricoidea) From the Southeast- 
ern Coast of Brazil 87 

CHRISTINA M. SPOLSKY, GEORGE M. DAVIS & ZHANG Yl 

Sequencing Methodology and Phylogenetic Analysis: Cytochrome b Gene Sequence 
Reveals Significant Diversity in Chinese Populations of Oncomelania (Gastropoda: 
Pomatiopsidae) 213 

P. TATTERSFIELD 

Local Patterns of Land Snail Diversity in a Kenyan Rain Forest 161 

LAURA R. WHITE, BRUCE A. McPHERON, & JAY R. STAUFFER, JR. 

Molecular Genetic Identification Tools for the Unionids of French Creek, Pennsylva- 
nia 181 

DAZHONG XU & MICHELE G. WHEATLY 

CA Regulation in the Freshwater Bivalve Anodonta ImbecHis: I. Effect of Environmen- 
tal CA Concentration and Body Mass on Unidirectional and Net CA Fluxes 59 



AWARDS FOR STUDY AT 
The Academy of Natural Sciences of Philadelphia 

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 pri- 
marily intended to assist predoctoral and immediate postdoctoral students. 
Awards usually 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. 



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. 



MALACOLOGIA, 1996, 38, NO. 1-2 



38 NO. 1-2 



MALACOLOGIA 



1996 



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 MAU\COLOGIA. 

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 may be in English, 
French, German or Spanish. Papers in lan- 
guages other than English must include a 
translation of the Abstract in English. Authors 
desiring to have their abstracts published in 
other languages must provide the transla- 
tions (complete with main titles). Include all 
foreign accents. Both American and British 
spellings are allowed. 

4. Unless indicated otherwise below, 
contributors should follow the recommenda- 
tions in the Council of Biology Editors (CBE) 
Style Manual (ed. 5, 1983) available for U.S. 
$24.00 from CBE, 9650 Rockville Pike, 
Bethesda, MD 20814, 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 manuscript. Make the hierarchy of 
headings within the text simple and consis- 
tent. 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 
letters 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. 

1 0. Use the metric system throughout. Mi- 
cron should be abbreviated |.im. 

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 13.5 x 20.0 cm 
(preferably not as tall as this so that the cap- 
tion 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 Уз. Letters and numbers should be at 
least 3 mm high after reduction. Several 
drawings or photographs may be grouped 
together to fit a page. Photographs are to be 
high contrast. High contrast is especially im- 
portant for histological photographs. 

13. All illustrations are to be numbered 
sequentially as figures (not grouped as plates 
or as lettered subseries), and are to be ar- 
ranged as closely as possible to the order in 
which they are first cited in the text. Each 
figure 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 |.im," not "163 i-im"). 
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 summarize what is 
shown in an illustration, and should not du- 
plicate 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. 



VOL38, NO. 1-2 1996 



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 Editors: 



CAROL JONES 
Denver, CO 



JOHN B. BURCH 
University of Michigan 
Ann Arbor 



ANNE GISMANN 

Maadi 

Egypt 



MALACOLOGIA is published by the INSTITUTE OF MALACOLOGY, the Sponsor Members of 
which (also serving as editors) are: 



RÜDIGER BIELER 
Field Museum, Chicago 

JOHN BURCH 

MELBOURNE R. CARRIKER, 

President 

University of Delaware, Lewes 

GEORGE M. DAVIS 
Secretary and Treasurer 

CAROLE S. HICKMAN 
University of California, Berkeley 



ALAN KOHN 

University of Washington, Seattle 

JAMES NYBAKKEN 

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 JACKIE L VAN GOETHEM 

Secretary, UNITAS MALACOLOGICA Treasurer, UNITAS MALACOLOGICA 

Rijksmuseum van Natuurlijke Koninklijk Belgisch Instituut 

Historie voor Natuurwetenschappen 

Leiden, Netherlands Brüssel, Belgium 



J. FRANCIS ALLEN, Emérita 
Environmental Protection Agency 
Washington, D.C. 

KENNETH J. BOSS 

Museum of Comparative Zoology 

Cambridge, Massachusetts 



Emeritus Members 

ROBERT ROBERTSON 

The Academy of Natural Sciences 

Philadelphia, Pennsylvania 

W. D. RUSSELL-HUNTER 
Easton, Maryland 



Copyright © 1996 by the Institute of Malacology 



1996 
EDITORIAL BOARD 



J. A. ALLEN 

Marine Biological Station 

Millport, United Kingdom 

E. E. BINDER 

Muséum d'Histoire Naturelle 
Genève, Switzerland 

A. J. CAIN 

University of Liverpool 
United Kingdom 

P. С ALOW 

University of Shieffield 
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 CU\RKE 

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 

Université 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 

Mahne 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 

Association of Systematics Collections 

Washington, DC, U.S.A. 

B. HUBENDICK 

Naturhistohska 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 

Faculté des Sciences 

Brest, France 

C. MEIER-BROOK 
Tropenmedizinisches Institut 
Tübingen, Germany 



H. к. MIENIS 

Hebrew University of Jerusalem 

Israel 

J. E. MORTON 
The University 
Auckland, New Zealand 



A. STANCZYKOWSKA 
Siedlce, Poland 



F. STARMUHLNER 

Zoologisches Institut der Universität 
Wien, Austria 



J. J. MURRAY, Jr. 
University of Virginia 
Charlottesville, U.S.A. 



Y. I. STAROBOGATOV 

Zoological Institute 
St. Petersburg, Russia 



R. NATARAJAN 

Marine Biological Station 

Porto Novo, India 

J. 0KU\ND 
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 

Oingdao, People's Republic of China 



W. STREIFF 
Université de Caen 
France 

J. STUARDO 

Universidad de Chile 
Valparaiso 

S. TILLER 

Muséum National d'Histoire Naturelle 

Paris, France 

R. D. TURNER 

Harvard University 
Cambridge, Mass., U.S.A. 

J.A.M. VAN DEN BIGGEU\AR 
University of Utrecht 
The Netherlands 

J. A. VAN EEDEN 
Potchefstroom University 
South Africa 

N. H. VERDONK 
Rijksuniversiteit 
Utrecht, Netheriands 



D. G. REID 

The Natural History Museum 

London, United Kingdom 



B. R. WILSON 

Dept. Conservation and Land Management 

Kallaroo, Western Australia 



N. W. RUNHAM 

University College of North Wales 

Bangor, United Kingdom 

S. G. SEGERSTRLE 
Institute of Marine Research 
Helsinki, Finland 



H. ZEISSLER 
Leipzig, Germany 

A. ZILCH 

Forschungsinstitut Senckenberg 

Frankfurt am Main, Germany 



MALACOLOGIA, 1996, 38(1-2): 1-17 

TAXONOMIC REVISION OF POTAMOLITHUS AGAPETUS PILSBRY, 1911, AND 
POTAMOLITHUS BUSCHII (FRAUENFELD, 1865) (GASTROPODA: HYDROBIIDAE) 

Maria Fernanda López Arnnengol 

Instituto de Embriología, Biología e Histología, Facultad de Ciencias Médicas-CONICET, 
Universidad Nacional de La Plata, Calle 60 y 120 (1900), La Plata, Argentina 

ABSTRACT 

Potamolithus agapetus Pilsbry, 1911, and P. buschii (Frauenfeld, 1865) are related species 
that live sympatrically in Rio de la Plata. 

Studies carried out on populations of both species from Rio de la Plata show that P. agapetus 
presents a marked secondary sexual dimorphism on shell shape and size. 

The female shell is bigger than male shell, and its body whorl shape is subglobose, with a 
rounded angle at the basal periphery and another angle a short distance below the suture. The 
male body whorl shape is usually rounded without keels and seldom with a round angle at the 
basal periphery. 

Females of P. agapetus are very similar to the shell of P. buschii, which lacks secondary 
sexual dimorphism. For that reason, P. agapetus females were excluded from the original 
description by Pilsbry (1911), and seemingly included in subsequent enlarged descriptions of 
P. buschii. 

Both species share the same body whorl shape, but both present different degrees between 
angulose to globular shapes. They can be distinguished by shell color pattern, columella width, 
body whorl sculpture, head pigment pattern, eyebrow position, nuchal node size in females, gill 
filament number and range, and in the shape and number of cusps on the central and lateral 
radular teeth. 

Key words: Potamolittius, Hydrobiidae, taxonomy, sexual dimorphism. 



INTRODUCTION 

The genus Potamolithus comprises small 
(up to 7 mm long), thick-shelled gastropods 
that inhabit rivers and streams (Pilsbry, 1 91 1 ; 
López Armengol, 1985). 

This genus, exclusively South American 
and endemic in Ribeira, Itajaí-acú and Jacuhy 
rivers in southern Brazil and Uruguay River, 
part of the Paraná and Rio de la Plata drainage 
systems (López Armengol, 1985). 

Controversial aspects of authorship and 
type species (ICZN Case 2801 ; López Armen- 
gol & Manceñido, 1992; Kabat, 1993; Kabat & 
Hershler, 1993; Manceñido & López Armen- 
gol, 1993) have been cleared up by ICZN ac- 
tion (ICZN Opinion 1779, 1994) fixing Pilsbry 
& Rush as the authors of this genus and Pot- 
amolithus lapidum (d'Orbigny, 1835) as its 
type species. 

In 1911, Pilsbry presented a key to species 
and subspecies and a description of the 
known species which were arranged in four 
groups. Parodiz (1965) gave a description of 
Potamolithus species and added new char- 
acters and geographical data. Davis & Pons 
da Silva (1984) described the anatomy of P. 



ribeirensis from Feitoria River, Brazil, and dis- 
cussed phylogenetic relationships and con- 
vergence with other hydrobiid and pomatiop- 
sid genera. 

The descriptions of Potamolithus species 
were based on shell features, and only a few 
specimens were studied in some cases. 

Potamolithus agapetus Pilsbry, 1911, and 
Potamolithus buschii (Frauenfeld, 1865) are 
sympatric in Rio de la Plata (Pilsbry, 1911; 
López Armengol, 1985). According to Pilsbry 
(1911) both species belong to the "group of P. 
buschii" because they share the same gen- 
eral shell shape: both equally wide and high 
with a normal length spire, a simple lip, and a 
flattened columella. Juveniles of P. buschii are 
not always readily distinguishable from imma- 
ture P. agapetus. Potamolithus agapetus was 
originally described as the smallest Potamo- 
lithus known and has a globular-conic shell. 
On the other hand, P. buschii was originally 
described as having a wide and carínate shell. 

Studies carried out on populations of both 
species from Rio de la Plata show that spec- 
imens with a shell morphology agreeing with 
the original description of P. agapetus are all 
males. On the other hand, a great variability 



LÓPEZ ARMENGOL 




FIG 1 Scanning electron micrographs of the shell of P. agapetus. A, B, nnales; C, D, females. Body whorl 
penphery: A, globose, and C, angular. F, lateral view: note the concave body whorl base. G, enlargement 
of the shell showing the surface faintly marked with growth-lines and some pits. Scale bar A-E - 1 mm; F 
= 50 |am. 



TWO RELATED SPECIES OF POTAMOLITHUS 




FIG. 2. Variation in the shape of the shell in P. agapetus. A, females. B, males. Scale bar = 1 mm. 



TABLE 1 . Whorl number of P. agapetus. Frequency of males and females at each whorl stage present 
at the localities studied. % = percentage of population. 







Anchorena beach 




Rio San Ji 


jan 




Isla San Gabriel's 


Whorls 


ó 


9 


% 


â 


9 


% 


â 


9 


% 


eroded 


7 


15 


3.72 


2 


12 


18.92 


2 


2 


4.12 


2.50 


— 


— 


— 


— 


— 


— 


4 


1 


5.15 


3.00 


12 


1 


2.19 


5 


— 


6.76 


18 


7 


25.77 


3.25 


104 


17 


20.44 


7 


— 


9.46 


19 


1 


20.62 


3.50 


42 


68 


18.58 


1 


1 


2.70 


7 


6 


13.40 


3.75 


7 


221 


38.51 


2 


11 


17.57 


2 


20 


22.68 


4.00 


— 


86 


14.53 


2 


29 


41.89 


— 


8 


8.25 


4.25 


— 


12 


2.03 


— 


2 


2.70 


— 


— 


— 


N = 


172 


420 




19 


55 




52 


45 





was observed in P. buschii in such charac- 
ters as radula, head pigmentation, number of 
gill filaments, and the size of the female 
nuchal node. 

The aim of this work is to redeschbe both 
P. agapetus and P. buschii. 



MATERIALS AND METHODS 

Localities studied were: Rio de la Plata, 
Anchorena beach, Argentina (34"29'S, 
58°28'W), col.: López Armengol, 30-IV-1984, 
Colección Malacológica del Museo de La 
Plata, MLP 4652; mouth of Río San Juan 
where it empties into Río de la Plata, Uruguay 
(33^"17'S, 57 58'W), col.: Pérez Duhalde, 15- 
VII-1989, MLP 4986; Río de la Plata, Isla San 
Gabriel, Uruguay (34 29'S, 57 52'W),col.: Ló- 



pez Armengol-Casciotta, 17-111-1985, MLP 
4655. 

The samples were taken randomly and in- 
clude all individuals of all size classes at a 
single site in the river. The sample for the 
number of individuals and their sex for each 
whorl number consisted of 592 specimens of 
P. agapetus and 289 specimens of P. buschii. 
These samples were drawn from an initial 
population of 3,404 individuals (MLP 4652). 

Specimens were measured by ocular mi- 
crometer in a Wild M-5 stereoscopic micro- 
scope. All specimens studied were unpara- 
sitized. Measurements are those of Hershler 
& Landye (1988). The following ratios were 
formed using some of this data: shell length/ 
body whorl length; body whorl length/shell 
width; shell length/shell width and aperture 
length/shell length. 



LOPEZ ARMENGOL 

TABLE 2. Shell measurements (mm) and ratios for 29 males and 44 females of 3.50 
to 3.75 whorls of P. agapetus (MLP 4652). X ± standard deviation (range). * = 
significant difference between sexes, P< .001. 



Characters 




males 


females 


SD P < .001 


Shell length 




2.32 ±0.25 
(1.95 - 3.09) 


2.88 ±0.26 
(2.27 - 3.24) 


* 


Body whorl length 




2.08 ±0.24 
(1.76 - 2.77) 


2.56 ± 0.24 
(1.95 - 2.96) 


*. 


Spire length 




0.25 ±0.04 
(0.15 - 0.32) 


0.32 ± 0.05 
(0.19 - 0.44) 




Shell width 




2.28 ±0.26 
(1.95 - 3.09) 


2.81 ±0.27 
(2.20 - 3.21) 




Aperture length 




1.74 ±0.22 
(1.45 - 2.39) 


2.17 ±0.20 
(1.76 - 2.52) 




Aperture width 




1.22 ±0.16 
(1.01 - 1.76) 


1.51 ±0.15 
(1 .20 - 1 .83) 




Columella width 




0.24 ±0.04 
(0.16- 0.32) 


0.30 ±0.06 
(0.19 - 0.44) 




Umbilical area width 




0.18 ±0.07 
(0.06 - 0.38) 


0.22 ±0.08 
(0.06 - 0.38) 




Shell length/body whorl 


length 


1.12±0.02 
(1.08 - 1.18) 


1.13 ±0.02 
(1.08 - 1.17) 




Body whorl length/shell 


width 


0.91 ±0.03 
(0.84 - 0.99) 


0.91 ± 0.04 
(0.83 - 0.98) 




Shell length/shell width 




1.02 ±0.02 
(0.92 - 1.09) 


1.02 ±0.04 
(0.96- 1.07) 




Aperture length/shell length 


0.73 ±0.04 


0.75 ± 0.04 








(0.63 - 0.80) 


(0.63 - 0.83) 





Number of whorls was counted according 
to Emberton (1985), but 0.25, 0.5 and 0.75 
were the fractions considered. Body whorl 
and penultimate whorl convexity were calcu- 
lated following Hershler & Landye (1988). 
Whorl convexity value is directly proportional 
to whorl convexity. 

Shells and radulae were studied and pho- 
tographed using scanning electron micro- 
scope (Jeol JSM-T 100). Heads were dried 
by the critical point method. 

The position and distance between the 
base of penis or nuchal node with respect to 
the lobes of the eyes and the angle of the 
base of penis or nuchal node with respect to 
the mid-line of the neck were calculated on 
fixed material, following Davis et al. (1986). 

Statistical analyses were limited to calcu- 
lating the means, standard deviations, and 
standard 't' test for sexual dimorphism in 
shell measurements and ratios and gill fila- 
ment number. The significance level ac- 
cepted was P < .001. Xi^ was performed to 
evaluate sex ratio =1:1. 



RESULTS 
Potamolithus agapetus Pilsbry 1911 

Potamolithus agapetus Pilsbry 1911: 578, pi. 

40, fig. 10, 10a. 
Potamolithus agapetus Parodiz 1965: 9 

Type material: Academy of Natural Sciences 
of Philadelphia 69,683. 

Type locality: Rio de la Plata, at Isla San 
Gabriel, near Colonia, Colonia Department, 
Uruguay. 

Description 

The shell is globose-conic to subglobose 
(Figs. 1, 2) and solid but not thick. The color 
is uniform light brown. The surface is rather 
smooth, faintly marked with growth lines (Fig. 
1 F). The spire is 1 1 % of the shell length. The 
number of whorls is most frequently between 
3.00 and 4.00 (Table 1), slightly convex 
(penultimate whorl convexity = 0.20 and 
body whorl convexity = 0.14) in outline. The 



TWO REU\TED SPECIES OF POTAMOLITHUS 



4.75 

4.50 

4.75 

€ 4.00 
о 

i 

° 3.75 

0) 

E 

Z 3.50 

3.25 
3.00 



▲ AAA 



A AAAAAA 



2.0 



В 



4.25 



i2 400 



3,75 



350 



325 



20 



• • • «AAAAAA^AA 



i«*iii -AA A 



• •••••••^^•* A 



3.0 4.0 

Shell length (mm) 



5.0 



A • 



^ ^AAéééiéè^éAè 



A* ^Aikkkk.kkkkkkk A* 



A A 



• A 



3.0 4.0 

Shell length (mm) 



5.0 



FIG. 3. Scatter-diagram for the number of whorls and shell length. A, P. agapetus (82 males and 61 
females). B, P. buschii (128 males and 108 females). Note the sexual dimorphism in shell size in P. 
agapetus. Males (black circles), females (black triangles). One symbol may represents more than one 
specimen. 



LOPEZ ARMENGOL 




В 




FIG. 4. Pigment patterns of P. agapetus. A, head-neck, dorsal view. B, penis, right side. Scale bar = 1 mm. 



body whorl base is concave in dorsal view 
(Fig. IE). The aperture is oblique, inclined 
about 35 ' to 42 ■ (X = 39") towards the coiling 
axis, rounded-ovate, and angular at the top. 
The columella is wide and flattened (Fig. 1B, 
D). 

Shells with discontinous peristome have a 
thin outer lip, and the umbilical area can be 
present or absent. When it present, is narrow 
and bounded by an angle. In shells with con- 
tinous peristome (Fig. IB, D), the inner lip is 
heavily calloused and the outer lip is simple 
and thin. There is a rather conspicuous um- 
bilical area bounded by an angle or an acute 
ridge. Some specimens have an umbilical 
opening. 

There is sexual dimorphism in shape and 
shell size. The shape of the body whorl in 
males is usually globose (Figs. 1A, 28). How- 
ever, some males have a shell with a rounded 
angle below the suture or with two angles, 
one below the suture and the other at the 
basal periphery (Fig. 2B). Males have a 
rounded outer lip (Fig. IB). The female body 
whorl shape is usually subglobose, with two 
rounded angles, one below the suture and 
the other at the basal periphery (Figs. 1С, 
2A); the outer lip is sharp (Fig. ID). 

The females are larger than males with the 
same number of whorls (Fig. ЗА). No sexual 
dimorphism in umbilical area width and cal- 
culated ratios were observed. Statistics on 
shell dimensions for males and females of 



3.50 to 3.75 number of whorls are given in 
Table 2. 

There was no significant difference in num- 
ber of males and females at Anchorena 
Beach (0.76:1). 

The head can be unpigmented or with a 
band of melanin in the snout, or with two 
V-shaped bands orientated with the vertex 
pointing from the snout to the neck. Another 
band runs dorsally in the middle of each ten- 
tacle. There is a concentration of white 
spheric granules above and around the eyes 
("eyebrows") (Fig. 4A), and eye lobes are 
slight swellings at the base of each tentacle. 

The neck of females bears a protuberance 
called nuchal node (Davis & Pons da Silva, 
1984). The position of the nuchal node base 
is mainly to the right of the mid-line of the 
head. The nuchal node is X = 0.25 mm ± 0.02 
(0.24-0.30) high (Fig. 5A, B). The distance be- 
tween the base of the nuchal node and the 
eyes is X = 0.31 mm ± 0.09 (0.15-0.4). The 
angle of the base of the nuchal node (with 
respect to the mid-line of the neck) is X = 52° 
(34'-72''). 

The penis is simple, without appendages; 
with a black spot at the distal end (Figs. 4B, 
5C). The distance between the base of the 
penis and the lobes of the eyes is X = 0.14 
mm ±0.01 (0.12-0.15). The angle of the base 
of the penis (with respect to the mid-line of 
the neck) is X = 23° (14''-30 ). 

There are 19 to 28 gill filaments (Fig. 6), 



TWO REU\TED SPECIES OF POTAMOLITHUS 




FIG. 5. Scanning electron micrographs of the 
head-neck of P. agapetus. A, dorsal view of female 
head-neck, showing the nuchal node. B, right side 
of the head-neck of a female. C, left side of the 
head-neck and fully erect penis of a male. Scale 
bar = 200 цт. 



with no indication of sexual dimorphism in- 
their number (X = 23.00 ± 2.30 and 23.87 ± 
2.64 for males and females respectively). 

Radula typically taenioglossate (Fig. 7), the 
statistics and cusp formulae given in Table 3. 
Distinctive features are: the concave hollow 
in the middle of the anterior cusp of the cen- 
tral teeth (Fig. 70); the external edge of lateral 
angle of the central teeth is sometimes 
curved; the innermost pair of basal cusps 



arise from the face of the tooth; there is a 
concave hollow between the basal cusps and 
basal process, and the basal process is 
prominent. 

There were no differences among the 
blades of the lateral tooth (Fig. 7A, B, D, E). 
There is a pronounced posterior projection 
on the face of the lateral tooth (Fig. 7D). 

Potamolithus buschii (Frauenfeld, 1865) 

Lithoglyphus Buschii Frauenfeld, 1865, ex 

Dunker, in litt: 530, pi. 11 
Potamolitlius busctiil, Pilsbry & Rush 1896: 

80 
Potamolithus buschii, Pilsbry, 1896: 88 
Potamolithus buschii, Pilsbry, 1911: 580, pi. 

40, figs. 11-14, pi. 41b, fig. 2 
Potamolithus buschii, Parodiz, 1965: 28, figs. 

63-72 

Type matehal: Naturhistorisches Museum, 
Vienna, Austria. 

Type locality: "Erst kürzlich von Buenos-Ay- 
res [sic. Colonia Department, Uruguay] er- 
halten. Wird gefunden an der Mündung des 
St. Juan in den La Plata." 

Description 

The shell is imperforate, solid, subglobose 
to globose in shape (Figs. 8-10). The shell is 
green, with irregular buff zigzag streaks (Fig. 
11); some specimens (27% at Anchorena 
beach) have a dusky-brown band located su- 
tural and peripheral on the body whorl (Fig. 
11). The surface is smooth, although marked 
with growth-lines (Fig. 8F). The spire length is 
variable, between 9.60% and 15% of the 
shell length. The number of whorls is most 
frequently between 3.75 and 4.00 (Table 4), 
convex (penultimate whorl convexity = 0.17 
and body whorl convexity = 0.21) inoutline. 
The body whorl can be carínate, strongly an- 
gular, with a rounded angle, or globose at the 
basal periphery (Figs. 8A-D, 10). The body 
whorl is convex above the basal periphery, 
usually having a low keel or rounded angle at 
the back and a short distance below the su- 
ture (Fig. 9A-C). There is also, sometimes, a 
second spiral ridge below the upper one and 
a concavity between both called sulcus (Fig. 
9D). The base is flattened in dorsal view (Fig. 
9A-0). The aperture is oblique, inclined about 



LOPEZ ARMENGOL 



5 г 



А - 



3 - 



fc2 
E 



Г h 




*-- 


-f 






Pagapetus /\ 


P buschii 


/ 
/ 


\ 
\ 

\ 






N=20 M 


N = 18 

; 
1 


1 
1 
1 

1 
1 


i 
1 
i 

\ 
\ 
i 






Л r 


\ Ä 1 
\ /\ ' 

\ / \ ' 




1 


1 
\ 

i 

\ 

\ n 

\ 1 \ 
\ 1 \ 

\ ! \ 




\ J 


Vv 

— 1 1 )l 1 к L_ 


— 1 




1 1 


-Л — i 


1 1 1 1 1 1 



19 20 21 



22 23 TU 25 26 27 28 29 30 31 32 33 34 35 36 
Number of gill filaments 



FIG. 6. Gill filament number in P. agapetus and P. buschii. Scatter-diagram between number of gill filaments 
and number of individuals. 



40-^' to 54° (X = 46°53') towards the axis of 
coiling; basally rounded and angular at the 
top. Columella narrow and flattened or con- 
vex (Fig. 8E). 

In shells with a discontinous peristome, the 
outer lip is thin and may or may not have an 
umbilical area. When the umbilical area is 
present, it is narrow and bounded by an an- 
gle. In shells with a continous peristome, the 
inner lip is heavily calloused and the outer lip 
is thick (Fig. 8E). Sometimes the peristome is 
edged with a black line. There is a well-de- 
veloped concave umbilical area bounded by 
an angle or an acute ridge. Some specimens 
have an umbilical opening. 

Statistics on shell dimensions for males 
and females of 3.50 to 3.75 whorls are given 
in Table 5. No sexual dimorphism in shell size 
was evident; females and males at the same 
number of whorls have the same size (Fig. 
3B). 

There was no significant difference in the 
number of males and females at Anchorena 
Beach (0.90:1). 

The entire head is black (melanin), and 
there is a black band in the middle of each 
tentacle. Next to the eyes there is a hyaline 



band with white spheric granules on it (Fig. 
12 A). 

The nuchal node is located to the right of 
the mid-line and is 0.06 mm high (Fig. 13A, 
B). The distance between the base of nuchal 
node and the lobes of the eyes is X = 1.01 
mm ± 0.24 (0.63-1 .26). The angle of the base 
of nuchal node (with respect to the mid-line 
of the neck) is x = 48° (27°-69''). 

The penis is simple, without appendages 
(Fig. 13C). The penis bears two parallel 
bands of melanin running on both sides, one 
dorsal along the distal part and the other ven- 
tral (Fig. 12B). The distance between the 
base of penis and the lobes of the eyes is X = 
0.59 mm ±0.08 (0.45-0.75). The angle of the 
base of the penis (with respect to the mid-line 
of the neck) is X = 32 (20' -45 ). 

There are 28 to 36 gill filaments (Fig. 6), 
with no indication of sexual dimorphism in 
their number (X = 30.14 ± 1.57 and 32.45 ± 
1 .69 for males and females respectively). 

Radula tipically taenioglossate (Fig. 14). 
The statistics and cusps formulae given in 
Table 3. Distinctive features are: that the mid- 
dle of the anterior cusps of the central tooth 
is flat; the external edge of lateral angle is 



TWO REU\TED SPECIES OF POTAMOLITHUS 




FIG. 7. Radula of P. agapetus. A, Section of radular ribbon excluding left outer marginals. B, enlargement 
of central and right lateral teeth. C, central tooth. D, central and left lateral teeth. E, left lateral and marginal 
teeth. F, right inner and outer marginal teeth. Scale bar A = 50 цт; B-F = 10 цт. 



TABLE 3. Formulae for the most common cusps arrangements for the four radular teeth 
of P. agapetus and P. buschii. 



Tooth 



Formula (%) 



P. agapetus (4 radulae) 

Central 29 



Lateral 38 

Inner marginal 35 

Outer marginal 21 
P. buschii (2 radulae) 

Central 51 

Lateral 46 

Inner marginal 31 

Outer marginal 24 



6-1-6 6-1-5 5-1-6 

(79.30), (1 7.24), (3.45) 

3-3 3-3 3-3 

4-1-5 (44.74); 5-1-4 (31.58); 5-1-5 (23.68) 

15-22 

17-23 

4-1-4 4-1-4 4-1-5 

(33.30); (33.30); (33.30) 

2-2 2-3 2-2 

3-1-3 (80.43); 4-1-3 (10.87); 2-1-2 (8.70) 

9-11 

12-15 



10 



LÓPEZ ARMENGOL 




FIG. 8. Scanning electron micrographs of the shell of P. buschii. A-D, frontal view. Body whorl: A, carinated; 
B, strongly angular; C, rounded angle; D, globose. E, unnbilical view. F, enlargement of the shell showing 
the surface marked with growth-lines and some pits. Scale bar A-E = 1 mm; F = 50 um. 



TWO RELATED SPECIES OF POTAMOLITHUS 



11 




FIG. 9. Scanning electron micrographs of the shell of P. buschii. A-C, dorsal view. Note the flat body whorl 
base and the different degrees of subsutural carination: A, carinated; B, angular; C, globose. D, lateral view: 
note the sulcus between two ridges. Scale bar = 1 mm. 



straight, and the innermost pair of basal 
cusps arise from the face of the tooth (Fig. 
140). The ventral part of basal cusps is a little 
concave and the basal process is not prom- 
inent. The central blade of lateral teeth v\/id- 
ened with respect to the other cusps (Fig. 
14A, B, D). 

DISCUSSION AND CONCLUSIONS 

Potamolithus agapetus shows marked 
secondary sexual dimorphism in shell shape 



and size. The shape of the body whorl in 
males is usually rounded, whereas the female 
is subglobose, with a rounded angle at short 
distance below the suture and another angle 
at the basal periphery. Like other gastropods 
showing sexual dimorphism, the female shell 
is larger than the male shell. 

Potamolithus agapetus was described by 
Pilsbry (1911) as the smallest Potamolithus 
known, with body whorl evenly rounded, 
without keels or angles but his description 
did not include subglobose shells. Two 



12 



LOPEZ ARMENGOL 




FIG. 10. Variation in the shape of the shell in P. buschii. Scale bar = 1 mm. 



rounded angles are usually present in fe- 
males. 

Potamolithus buschii was described by 
Frauenfeld (1865) as having a wide, carinate 
shell, but in subsequent descriptions by Pil- 




bry (1911) and Parodiz (1 965) the concept of 
this species changed. Pilsbry (191 1) included 
the least angular forms of P. buschii from Isla 
San Gabriel (type locality of P. agapetus) and 
Parodiz (1965) stated that carinated shells 
were not the common form of the species. 

Because P. agapetus and P. buschii are 
related species and sympatric in Rio de la 
Plata, it is probable that P. agapetus females 
have been included in the descriptions of P. 
buschii by Pilsbry (1911) and Parodiz (1965). 
For example, Pilsbry (1911) showed in his 
Plate 40, fig. 14, the least angular form of P. 
buschii, which is very similar to the female 
form of P. agapetus. This is became both 
species share the body whorl shape ranging 
from angulose to globular, and broad um- 
bilical area circled by an angular or acute 
ridge. The features that reliably to distinguish 
both species, as redefined herein are listed In 
Table 6. 



Imtn 

FIG. 1 1 . Shell of P. buschii, showing the peripheral 
band and irregular buff zigzag streaks. 



ACKNOWLEDGEMENTS 

I wish to express my gratitude to Analia 
Amor, Instituto de Embriología, Biología e 



TWO HELMED SPECIES OF POTAMOLITHUS 



13 



TABLE 4. Whorl number of P. buschii. Frequency of males and females at each whorl stage present at 
the localities studied. % = percentage of population. 





Anchorena Beach 




Río San Ji 


uan 




Isla San Gabriel 


Whorls 


â 9 


% 


â 


9 


% 


â 


9 


% 


eroded 


25 19 


15.23 


— 


2 


4.55 


20 


19 


14.45 


2.25 


— — 


— 


— 


— 


— 


— 


3 


1.11 


3.00 


— — 


— 


— 


1 


2.27 


3 


8 


4.07 


3.25 


1 3 


1.38 


1 


1 


4.55 


15 


24 


14.45 


3.50 


14 4 


6.23 


— 


3 


6.82 


11 


19 


11.11 


3.75 


64 56 


41.52 


9 


5 


31.82 


32 


53 


31.48 


4.00 


29 68 


33.56 


1 


7 


18.18 


21 


37 


21.48 


4.25 


2 4 


2.08 


5 


9 


31.81 


1 


4 


1.85 


N = 


135 154 




16 


28 




103 


167 





TABLE 5. Shell measurements (mm) and ratios for 78 maies and 53 
females of 3.50 to 3.75 whorls of P. buschii. X ± standard deviation 
(range). There is no significant difference between sexes, P < .001. 



Characters 




males 


females 


Shell length 




3.90 ± 0.39 


3.77 ± 0.45 






(2.84 - 4.5) 


(2.34 - 4.68) 


Body whorl length 




3.49 ± 0.35 


3.39 ±0.41 






(2.52 - 4.14) 


(2.16 - 4.32) 


Spire length 




0.41 +0.09 


0.39 ± 0.08 






(0.18 - 0.63) 


(0.18 - 0.63) 


Shell width 




3.92 ± 0.39 


3.87 ±0.48 






(2.52 - 4.68) 


(2.61 - 4.86) 


Aperture length 




3.03 ± 0.29 


2.99 ±0.31 






(2.08 - 3.51) 


(2.16 - 3.69) 


Aperture width 




2.22 ± 0.23 


2.17 ±0.26 






(1.39 - 2.61) 


(1.35 - 2.70) 


Columella width 




0.32 ± 0.08 


0.34 ± 0.08 






(0.09 - 0.45) 


(0.18 - 0.54) 


Umbilical area width 




0.24 ±0.12 


0.25 ±0.13 






(0.04 - 0.54) 


(0.09 - 0.54) 


Shell length/body whorl 


length 


1.11 ±0.02 


1.11 ±0.02 






(1.06 - 1.20) 


(1.06 - 1.16) 


Body whorl length/shell 


width 


0.89 ± 0.04 


0.88 ± 0.03 






(0.79 - 1.03) 


(0.82 - 0.97) 


Shell length/shell width 




1.01 ±0.05 


1.02 ±0.04 






(0.89 - 1.13) 


(0.94 - 1.11) 


Aperture length/shell length 


0.78 ±0.05 


0.79 ±0.04 






(0.67 - 0.92) 


(0.69 ~ 0.92) 



Histología, and Miguel O. Manceñido, Facul- 
tad de Ciencias Naturales y Museo, for their 
valuable help and criticism of the manuscript. 
I am indebted to G. M. Davis and an anony- 



mous referee for critically reading the manu- 
script. I also want to thank Maria I. Braca- 
mente (CONICET) for the preparation of rad- 
ular material. 



14 



LOPEZ ARMENGOL 




В 




FIG. 12. Pigment patterns of P. buschii. A. head-neck, dorsal view. B, penis, right side. Scale bar = 1 nnm. 



TABLE 6. Characters distinguishing P. agapetus and P. buschii. 



Characters 



P. agapetus 



P. buschii 



Shell 

Irregular buff zigzag streaks 

Growth-lines 

Body whorl sculpture 



Sulcus in dorsal view 

Body whorl base in dorsal view 

Relationship between shell length 

and shell width 
Aperture inclination 
Columella 
Peristome 

External Features 

Head pigment pattern 
Eyebrows position 
Nuchal node size 
Penis pigment pattern 



no 
faintly marked 
rounded basal angle 
in females 

no 

concave 

longer than wider 

35 to 42" 

wide 

simple and thin 

unpigmented 

above and around the eyes 

0.25 mm 

black spot in distal end 



Gill Filaments 




Gill filaments number range 


19-28 


Radula 




Central teeth 




Middle of the anterior cusps 


concave hollow 


External edge of lateral angle 


sometimes curved 


Ventral part of basal cusps 


concave 


Basal process prominent 


yes 


Lateral teeth 




Central blade more developed 


no 


Sexual Dimorphism in Shell 


yes 



yes 

marked 

rounded angle, strongly angular, 

or carena subsutural and basal 

in both sexes 

no/yes 

flat 

wider than longer 

40 to 54^ 

narrow 

thicker, sometimes dark-edged 

entirely black 

in hyaline bands 

0.06 mm 

two parallel bands 

28-36 



flat 

straight 

less concave 

no 

yes 
no 



TWO RELATED SPECIES OF POTAMOLITHUS 



15 




FIG. 13. Scanning electron micrographs of the 
head-neck of P. buschii. A, dorsal view of female 
head-neck, showing the nuchal node (arrow). B, 
nght side of the female head-neck. C, left side of 
the head-neck and fully erect penis of a male. 
Scale bar A, В = 200 мт; С = 500 |.im. 



16 



LOPEZ ARMENGOL 




FIG. 14. Radula of P. buschii. A, section of radular ribbon excluding left outer marginals. B, enlargement of 
central and lateral teeth. C, central teeth. D, lateral teeth. E, left inner and outer marginal teeth. F, right inner 
and outer marginal teeth. Scale bar A = 50 цт; B-F = 10 цт. 



LITERATURE CITED 

DAVIS, G. M. & M. C. PONS DA SILVA, 1984, Pot- 
amolithus: morphology, convergence, and rela- 
tionships among hydrobioid snails. Malacologia, 
25: 73-108. 

DAVIS, G. M., N. V. SUBBA RAO & K. E. HOAG- 
LAND, 1986, In search of Tricula (Gastropoda: 
Prosobranchia): Tricula defined, and a new ge- 
nus described. Proceedings of the Academy 
of Natural Sciences of Philadelphia, 1 38: 426- 
442. 



EMBERTON, K. C, 1985, Seasonal changes in the 
reproductive gross anatomy of the land snail Trl- 
odopsis tridentata tndentata (Pulmonata: Po- 
lygyridae). Malacologia, 26: 225-239. 

FRAUENFELD, G. R. VON, 1865, Zoologische Mis- 
cellen. V. Verhandlungen der К. К. Zoologisch- 
Botanischen Gesellschaft in Wien, 15: 525-536. 

HERSHLER, R. & J. J. UXNDYE, 1988, Anzona Hy- 
drobiidae (Prosobranchia: Rissoacea). Snnithso- 
nian Contributions to Zoology, 459: 63 pp. 

ICZN, 1994, Opinion 1779. Potamolithus Pilsbry 
and Rush, 1896 (Mollusca, Gastropoda): placed 



TWO RELATED SPECIES OF POTAMOLITHUS 



17 



on the Official List with Paludina lapidum d'Or- 
bigny, 1835 as the type species. Bulletin of Zoo- 
logical Nomenclature, 51: 271-272. 

KABAT, A. R., 1993, Comments on the proposed 
designation of Potamolithus rushii Pilsbry, 1896 
as the type species of Potamolithus Pilsbry, 
1896 (Mollusca, Gastropoda) (1). Bulletin of Zoo- 
logical Nomenclature, 50: 52. 

KABAT, A. R. & R. HERSHLER, 1993, The proso- 
branch snail family Hydrobiidae (Gastropoda: 
Rissooidea): review of classification and su- 
praspecific taxa. Smithsonian Contributions to 
Zoology, 547: 94 pp. 

LÓPEZ ARMENGOL, M. F., 1985, Estudio 
sistemático y bioecológico del género Potamo- 
lithus (Hydrobiidae) utilizando técnicas numéri- 
cas. Facultad de Ciencias Naturales y Museo, 
UNLP, Tesis No. 455: 281 pp. unpublished. 

LÓPEZ ARMENGOL, M. F. & M. O. MANCEÑIDO, 
1992, Potamolithus Pilsbry, 1896 (Mollusca, 
Gastropoda): proposed confirmation of P. rushii 
Pilsbry, 1986 as the type species. Bulletin of 
Zoological Nomenclature, Case 2801, 49: 109- 
111. 



MANCEÑDO, M. O. & M. F. LÓPEZ ARMENGOL, 
1993, Comments on the proposed designation 
of Potamolithus ruhii Pilsbry, 1896 as the type 
species of Potamolithus Pilsbry, 1896 (Mollusca, 
Gastropoda) (2). Bulletin of Zoological Nomen- 
clature, 50: 53. 

PARODIZ, J. J., 1965, The hydrobid snails of the 
genus Potamolithus (Mesogastropoda-Rissoa- 
cea). Sterkiana, 20: 1-38. 

PILSBRY, H. A., 1896, Notes on new species of 
Ammicolidae collected by Dr. Rush in Uruguay. 
The Nautilus, 10: 86-89. 

PILSBRY, H. A., 1 91 1 , Non-marine Mollusca of Pa- 
tagonia. In: w.B. SCOTT, ed.. Reports of the 
Princeton University Expeditions to Patagonia, 
1896-1899, 3(2)(5): 566-602. 

PILSBRY, H. A. & W. H. RUSH, 1896, List, with 
notes, of land and fresh water shells collected by 
Dr. Wm. H. Rush in Uruguay and Argentina. The 
Nautilus, 10: 76-81. 



Revised Ms. accepted 1 January 1995 



MAU\COLOGIA, 1996, 38(1-2): 19-31 

RECRUITMENT OF DREISSENA POLYMORPHA: DOES THE PRESENCE 

AND DENSITY OF CONSPECIFICS DETERMINE THE RECRUITMENT DENSITY 

AND PATTERN IN A POPULATION? 

M. E. Chase & R. С Bailey 

Ecology and Evolution Group, Department of Zoology, University of Western Ontario, 
London, Ontario, Canada N6A 5B7 



ABSTRACT 

Results of a field experiment conducted to examine the density and spatial pattern of re- 
cruitment in a population of Dreissena polymorpha in Lake St. Clair were consistent with the 
hypothesis that recruitment is in response to a chemical cue released by conspecific adults. 
The number of recruits were significantly higher in treatments in which conspecific adults were 
present. Analysis of the distribution of adults and recruits in low and high density treatments 
showed a strong spatial correlation between adults and recruits. However, the distribution of 
recruits in the low density treatment was more aggregated in comparison to the high density 
treatment. Comparison of density and size distribution of recruits between low and high density 
treatments and the adjacent natural population revealed recruits in the natural population were 
smaller and less dense than recruits in the experimental treatments. This results suggests that 
although recruitment is in response to conspecific adults, recruitment into a population with 
lower adult densities, as represented by the experimental treatments, may result in enhanced 
growth of the new recruits. 

Key words: Dreissena polymorpiia, recruitment, spatial, density, conspecific. 



INTRODUCTION 

Larval settlement and juvenile recruitment 
are the initial processes determining the 
structure of populations of many sessile, 
aquatic species (Rodriguez et al., 1993). Suc- 
cessful recruitment of a sessile organism de- 
pends on the behavioral adaptation of early 
life stages to meet or avoid biological and 
physical hazards (Schubart et al., 1995). The 
location of settlement and potential recruit- 
ment can affect the performance and ultimate 
survival of a sessile organism. Stimuli neces- 
sary for settlement involve a combination of 
factors, including speed of fluids and con- 
tours of the substratum (e.g., Sebens, 1983; 
Wethey, 1986; Butman, 1989; Pawlik & Had- 
field, 1990; Pawlik et al., 1991; Johnson, 
1994), luminosity (e.g.. Crisp & Ritz, 1973; 
Young & Chia, 1982) and chemical cues (e.g., 
Morse & Morse, 1984, Pawlik, 1986; Rai- 
mondi, 1988). Perhaps the most widely ex- 
amined of all settlement stimuli are the exis- 
tence of chemical inducers associated with 
conspecific adults. Such cues are of great 
ecological importance, because the induction 
of settlement by conspecifics can account for 
the aggregated distribution of many benthic 
marine invertebrates (Rodriguez et al., 1993). 



Aggregated distributions may increase the 
probability of fertilization in individuals that 
either release their gametes into the water 
column (e.g., Pearse & Arch, 1969; Russo, 
1979; Pawlik, 1986) or have internal fertiliza- 
tion (e.g., Raimondi, 1991). Aggregation also 
acts as an effective defense mechanism (e.g., 
Garnick, 1978; Bernstein et al., 1981; Pawlik, 
1986; Hoffman, 1989), increases filter-feeding 
efficiency (e.g., Barnes & Powell, 1950) and 
results in decreased juvenile mortality (e.g., 
Highsmith, 1 982). Settlement induced by con- 
specific adults has been described in several 
benthic invertebrates, including polychaetes 
(Jensen & Morse, 1984; Pawlik, 1986), bar- 
nacles (Knight-Jones, 1953; Rittschof et al., 
1984; Raimondi, 1988; Johnson & Strath- 
mann, 1989; Crisp, 1990; Raimondi, 1991), 
echinoids (Highsmith, 1982; Burke, 1984) and 
molluscs (Seki & Kan-no, 1981). 

The majority of studies of settlement and/ 
or recruitment, however, have been confined 
to marine invertebrates. This is a reflection of 
the common planktonic larval stage charac- 
teristic of many benthic marine invertebrates. 
The recent invader Dreissena polymorpha 
(Pallas), the zebra mussel, is one of the few 
North American freshwater benthic inverte- 
brates with a planktonic larval stage. Other 



19 



20 



CHASE & BAILEY 



freshwater bivalves that possess a plank- 
tonic larval or juvenile stage include such ex- 
otic species as the quagga nnussel, Dreis- 
sena bugensis, and the asian clam Corbicula 
fluminea. Most species of North American 
freshwater bivalves reproduce either via a 
specialized parasitic larval stage called 
glochidia (e.g., Unionidae) or through incuba- 
tion of a small number of embryos that simply 
crawl away once ready for juvenile existence 
(e.g., Sphaeriidae) (Mackie, 1991). Dreissena 
polymorpha larvae may remain in the water 
column for 5 days to 5 weeks (Sprung, 1993) 
before settling onto hard substrata, undergo- 
ing metamorphosis and becoming juveniles. 
The incorporation of a new cohort or age 
class into the population is the stage of the 
larval life cycle referred to as recruitment 
(Connell, 1985). 

Since its introduction into North America, 
researchers have devoted considerable en- 
ergy to studying the ecology and the control 
of D. polymorpha. Extensive research has 
been conducted to determine the distribution 
(e.g., Hebert et al., 1991; Schaner et al., 
1991; Dermott& Munawar, 1993), predict the 
spread (e.g., Strayer, 1991; Neary & Leach, 
1992: Ramcharan et a!., 1992) and ultimate 
impact (e.g., Maclsaac et al., 1992; Bunt et 
al., 1993) of D. polymorpha on lake ecosys- 
tems. However, in order to predict the spread 
or impact of D. polymorpha, we must first 
understand what factors regulate popula- 
tions. For a sessile organism, population dis- 
tribution is determined by dispersal ability 
and the extent of passive transport at a large 
spatial scale and suitable settlement sites at 
a small spatial scale (Minchinton & Scheib- 
ing, 1991). As a result, a more appropriate 
predictor of population structure and com- 
munity patterns may be sought through the 
study of recruitment. 

The results of a 1 989 survey of D. polymor- 
pha in Lake St. Clair (Hebert et al., 1991) re- 
vealed a marked heterogeneity in size and 
cohort structure among sites, depending on 
the density of D. polymorpha. Hebert et al. 
(1991) suggested that veliger settlement may 
be cued by a chemical released by conspe- 
cific individuals that is an attractant at low 
concentrations and a repellent at high con- 
centrations. Wainman et al. (1995) also sug- 
gested that shell induced factor was the ex- 
planation for the difference in recruitment 
between experimental substrata with and 
without mussels present. Their experimental 
design, however, consisted of racks sus- 



pended below the water surface in the metal 
iforbay of a thermal generating station, and 
therefore their results may not be represen- 
tative of recruitment in the natural population. 
The objectives of this study were twofold. 
Firstly, we determined whether or not the 
presence of conspecifics may be a cue to 
induce recruitment into a population of D. 
polymorpha at Lake St. Clair. Secondly, we 
determined whether or not the spatial ar- 
rangement of recruits was influenced by the 
presence and density of conspecifics. Re- 
sults of recruitment in the experimental study 
were Chen compared to recruitment in the 
natural population in Lake St. Clair. 



METHODS 



Study Site 



The study site was approximately 1 km 
from shore at 42 19'57.0"N, 82-33'19.5"W at 
2.5 m depth near Stoney Point, Ontario, on 
the southeastern shore of Lake St. Clair 
(Fig. 1). Lake St. Clair is the smallest of the 
Great Lakes, with a total area of 1,114 km^ 
(Bolsenga & Herdendorf, 1993). The mean 
depth is only 3 m, with maximum natural 
depth of 6.4 m and maximum depth along a 
dredged shipping channel of 8 m (Bolsenga & 
Herdendorf, 1993). Average annual tempera- 
ture is 1 1 .9 0. Temperatures range from near 
freezing for most of the winter to their sum- 
mer average peak of 24" С in July and Au- 
gust. Substratum at the site was predomi- 
nately silt and clay, with some fine sand and 
approximately 40% hard substrata, consist- 
ing of mainly rocks and Unionidae shells. 
Dreissena polymorpha were found on most 
submerged hard substrata. Densities of D. 
polymorpha at this site (± SE) were 15,735 (± 
31 6)/m2 (1 992), 1 5,545 (± 31 ОУт" (1 993) and 
10,264 (± 109)/m^ (1994) (Chase, unpub- 
lished data). Recruitment occurred in August 
and October in 1992, August in 1993, and in 
October in 1994. Population data from previ- 
ous years showed good recruitment at this 
site (1992: 5,469 individuals/m^; 1993: 
10,393 individuals/m^ (Chase, unpublished 
data). 

Experimental Design 

The experimental substrata consisted of 
plexiglass plates (18 x 18 cm). Each plate 
was attached to a cement block with a stain- 



RECRUITMENT OF DREISSENA POLYMORPHA 



21 



^° .„ 20 30 \(sT. CLAIR 

)) RIVER 


MICHIGAN r-^^^ i In 


1 "^5 ^ 


/ ^^^^^a 


) LAKE ^^ 


1 ST. CLAIR S 


DETROIj:^/ 1 


/f^^^^^^^^ "^^^^-^ 


y/ WINDSOR ^STONEY POINT 


y(y ONTARIO 


Г ^ Ч. л ONTARIO 


/(/ ^ 


JHUR04/ *^0 Л ^^ 


) L^^ / 


^ 1 /lakTontariq) 
MICHIGAN J CÎ^p -^^^ч, / 




^ry^ ^ y/ NEW YORK 




___^N^3Î^- 


LAKE ERIE \l 


OHIO ¡ 



FIG. 1. Location of study site at Stoney Point, Ontario, Canada. 



less steel bolt (Fig. 2b). The plates were im- 
mersed in lake water 2-3 days before use to 
remove any manufacturing chemicals that 
might have been present and which could 
prevent initial attachment by adult mussels. 
On May 15, 1994, adult mussels (6-21 mm 
in length) were collected from Lake St. Glair 
using SCUBA, and subsequently returned to 
the laboratory. Mussels were randomly cho- 
sen and placed on the plexiglass plates in an 
aquaria to allow byssal thread attachment. 
Three experimental treatments were estab- 
lished: 

(1) No adult mussels, 

(2) Low adult density (average 167 ± 30 in- 
dividuals/m^), and 



(3) High adult density (average 4583 ±419 
individuals/m^). 

Mussels on the plates remained in the lab- 
oratory for 48 hours, during which time they 
were fed dried Chlorella sp. (Beta Green, Na- 
trol) ad lib. It was found that moderate tem- 
perature (approximately 15''G) and food ad- 
dition enhanced byssal thread attachment of 
the adults (M. Chase, personal observation). 

Data Collection 

On May 17, 1994, the plates were ran- 
domly arranged in Lake St. Clair using 
SCUBA in a 6 X 3 configuration, with a 1-m 
perpendicular distance between blocks (Fig. 



22 



CHASE & BAILEY 



ш. 



v~\ 



1 m 



il 



■ 



Сш) ■ ■ И ^ ■ 

no adult mussels □ high density treatment ■ low density treatment 




plexiglass plate 
stainless steel bolt 

adult mussel 
cement block 



i6X 6 cm 



I i2X2cm 

r-i 1 X 1 cm 



: ni 


.. s^ 


7i ^ 


^i Я^ 


^íS 


...... ^^ 




d»- -йк 


ж ^ 


T 


CS -r»- 




_¿J 




-Г 4- 


v> г 


II 


1 1 



6 cm 

FIG. 2a. Schematic diagram of experimental design, which consisted of three treatments of adult mussel 
density — no adults, low density and high density — in Lake St. Clair. FIG. 2b. Diagram of one cement block 
as placed in the field with a plexiglass plate (18 ■ 18 cm) attached with a stainless steel bolt. FIG. 2c. 
Diagram of a plexiglass plate from the low density treatment showing division into 324 (1 x 1 cm) squares 
and subsequent analysis at three spatial scales; 1 x 1 cm, 2 x 2 cm and 6x6 cm. Open symbols represent 
adult mussels, closed symbols represent recruits. 



RECRUITMENT OF DREISSENA POLYMORPHA 



23 



2a). The plates were monitored visually twice 
monthly for recruitment. 

The plates were removed on November 1 6, 
1994, 182 days after deployment, and re- 
turned to the laboratory for examination. Of 
the 18 plates deployed, 16 were recovered, 
including 6 plates from the high density treat- 
ment, 5 plates of the low density treatment, 
and 5 plates of the no mussel treatment. In 
addition to retrieving the plates, 10 rocks 
were randomly collected from the surround- 
ing area so that the density and size distribu- 
tion of recruits from the experimental treat- 
ments could be compared to the natural 
population. 

In the laboratory, each plate was divided 
into 324 (1 X 1 cm) squares (Fig. 2c). The 
number of adults and recruits in each square 
was recorded under lOx magnification using 
a Wild-Heerbrug microscope. Recruitment 
was defined as individuals between 0.8 and 4 
mm in shell length. Adults and recruits were 
then removed from the plates. 

Mussels from the natural population were 
removed from each of the rocks collected and 
preserved in ethanol. Densities of adults and 
recruits in the natural population were deter- 
mined using the method of Bailey et al. (1 995). 
The shell length of the recruits from both the 
plates and the natural population were mea- 
sured at 6.4x magnification using a digitizing 
tablet interfaced with an IBM personal com- 
puter (Roff & Hopcroft, 1986). Length was 
measured as the longest distance between 
the umbo and the ventral margin. 

Data Analysis 

Counts of the number of recruits per plate 
in each treatment were Log-io (x + 1) trans- 
formed, and a one-way ANOVA was per- 
formed followed by a Tukey-Kramer test to 
make a posteriori comparisons of means. 

Lengths of recruits on the plates were 
Log 10 transformed and then compared within 
and among treatments by use of one-way 
ANOVA. Lengths of recruits in each treat- 
ment were then compared to Log^o trans- 
formed lengths of recruits in the natural pop- 
ulation by one-way ANOVA. 

To determine the effect of adult density on 
the spatial arrangement of recruits, the dis- 
tribution of adults and recruits was examined 
at three spatial scales; 1 x 1 cm, 2x2 cm and 
6x6 cm (Fig. 2c). A nested analysis of co- 
variance was applied to determine how 
adults and recruits covaried at these scales. 



(Л 


2.50 -1 




1- 






Z3 






Œ. 
О 


2.00 


1 


LU 






CE 

Li. 


1.50 


1 


О 






>— V 






■^ 


1.00 




+ 






X 












о 


0.50 - 




о" 
о 

-J 


0.00 - 


i 






HIGH DENSITY LOW DENSITY NO MUSSELS 






TREATMENT 



FIG. 3. Number of recruits (Log^o (X + 1) trans- 
formed) (± SE) in each of the experimental treat- 
ments; no adult mussels (n = 5 plates), low density 
(n = 5 plates) and high density (n = 6 plates). 



This analysis quantified the strength and na- 
ture of the covariation of adults and recruits 
on the plates in the low and high density 
treatments. 



RESULTS 



Recruit Density 



One-way ANOVA on log^ (x + 1) trans- 
formed number of recruits showed significant 
(F = 83.82, DF = 2,13, p < 0.001) differences 
among the treatments (Fig. 3). Pairwise com- 
parisons of mean number of recruits in each 
treatment using the Tukey-Kramer test 
showed that the number of recruits in the no 
adult mussel treatment was significantly 
lower than the low density treatment (p < 
0.001) and the number of recruits in the low 
density treatment was significantly lower 
than the high density treatment (p = 0.036) 
(Fig. 3). Examination of the relationship be- 
tween recruit density and adult density within 
treatments revealed a positive linear relation- 
ship within the low density treatment (Fig. 4a) 
but a negative linear relationship within the 
high density treatment. Neither regression 
was significant (Low density: DF = 1,3, F = 
2.84, r^ = 0.49; high density: DF = 1,4, F = 
7.314, r^ = 0.65). 

Size Distribution 

One-way ANOVA of length of recruits 
between the five plates in the low density 
treatment was not significant (F = 1.26, 
DF = 4,203, p = 0.286); therefore, lengths 
of recruits in the low density treatment were 
pooled. Lengths of recruits from the six 



24 



CHASE & BAILEY 



E 

(Л 

—I 
< 
z> 
о 

> 

a 

z 



1400 - 


A 


• 


^^ • 


1200 - 




•^^ 




1000 - 


// 


^ • 




800 - 


• 






600 - 






1 



200 250 



^^ 














z 


5000 - 


В 


• 








ш 




^\^ 










1- 


4000 - 


• 


\» 








-) 








\ч,^ 






Π


3000 - 






\\ 






Ü 








• 




• 


Ш 


2000 - 










ce 


1000 - 






1 


• 


1 



3000 3500 4000 4500 5000 5500 6000 6500 

ADULT DENSITY ( INDIVIDUALS / m^) 

FIG. 4. Regression of recruit density versus adult 
density for the low density (A) and high density (B) 
treatments. 



plates in the high density treatments were 
also pooled, because the one-way ANOVA of 
length of recruits between the high density 
treatments was not significant (F = 2.12, DP = 
5,622, p = 0.062). 

One-way ANOVA of length of recruits in the 
low and high density treatments was signifi- 
cant (F = 4.89, DF = 1,834, p = 0.027). The 
mean length of recruits in the low density 
treatment (1.89 ± 0.03 mm) was significantly 
larger than the mean length of recruits in the 
high density treatment (1.81 ± 0.02 mm), de- 
spite the small difference in means. 

Because the lengths of recruits in the high 
and low density treatments differed, separate 
ANOVA's were performed to compare them 
to the natural population. One-way ANOVA 
revealed the length of the recruits in both the 
low (F = 175.04, DF = 1,353, p < 0.001) and 
the high (F = 197.81, DF = 1,773, p < 0.001) 
density treatments (Fig. 5) were significantly 
larger than the mean lengths in the natural 
population. The mean length of recruits in the 
natural population was 1.33 ± 0.03 mm (Fig. 
5), about 500 цт less than in both experi- 
mental treatments. 



Spatial Arrangement 

Across both low and high adult density 
treatments, and all plates (Fig. 6), correlation 
between adults and recruits was high (r = 
0.90 ± 0.07). There was no significant differ- 
ence between the correlation values at either 
the 1 x1, 2x2 or 6x6 cm spatial scales in 
the low adult density treatment, although the 
mean correlation at the 2x2 cm spatial scale 
was always the highest (2 x 2: r = 1 .1 4 ± 0.08; 
6 X 6: r = 0.87 ± 0.26). Although r > 1 is 
mathematically impossible in simple correla- 
tion analysis, such estimates are possible in 
nested covariance analysis. They should just 
be interpreted as high correlations at this 
scale. One-way ANOVA of correlation at the 
different spatial scales in the high adult den- 
sity treatment revealed no significant differ- 
ence between the 2 x 2 and the 6 x 6 cm 
spatial scales (F = 2.04, DF = 1 , 1 0, p = 0.1 83) 
although the mean correlation at the 6 x 6 cm 
scale (r = 0.88 ± 0.07) was always higher than 
the mean correlation in the 2 x 2 cm scale 
(r = 0.74 ± 0.08). Correlation at the 1 x 1 cm 
scale was significantly lower than either the 2 
X 2 or the 6 X 6 cm scales in the high density 
treatment (F = 17.01, DF = 2, 15, p < 0.001). 
Correlation at the 1 x 1 cm scale was also low 
in the low density treatments (mean r = 0.64 
± 0.5). Low correlation between adults and 
recruits at the 1 x 1 cm scale reflects the 
average length of adult mussels on the 
plates, which was 1.03 cm. 



DISCUSSION 

Recruitment Density 

Results of the experimental study showed 
that the density of recruitment increased with 
the density of adults. Little recruitment was 
observed on plates with no adult mussels 
present. Several explanations may account 
ifor the pattern of recruitment observed in this 
study, including differential deposition and 
attachment, post depositional movement, 
and differential mortality after settlement 
(Johnson, 1994). Because this study exam- 
ined only recruitment, it is difficult to deter- 
mine which of the possible mechanisms may 
be underlying the settlement of D. polymor- 
pha at Lake St. Clair. Recruitment is defined 
as the arrival of the first cohort or age class 
into the population (Connell, 1985), so it in- 
cludes any post-settlement movement or 



I 



RECRUITMENT OF DREISSENA POLYMORPHA 



25 





1- 


H 




z 


z 




Ш 


III 


^ 


s 


2 


о 


1- 


1- 


F- 


< 


< 


3 


Ш 

ce 


ш 

ОС 


D 


1- 


1- 


Û. 


> 


>. 


Ü 


к- 


H 


Q. 


(Л 


<ñ 


-J 


z 


7 


< 


ш 


III 


CC 


û 


Q 


< 


X 


о 


z 


I 


-1 



L 




О 




с 


■ 




a> 


^ 




I 
Ф 

■a 


Ю 




с 
га 


■ 




от 


со 




с 

0) 

E 
'S 


о 






со 




i 




^■^ 


О) 

о. 




1= 


X 

ш 


in 


E 


(Л 

с 
ф 




^~^ 


СП 
О 
СМ 

il 


о 


X 


5 


CN¡ 


H 


о 




o 


ГО 




z 


to 

II 


■ 


Ш 


m 



ю 

■ 

о 



о 

■ 

о 



0) ^ 



о о о о о о о 
г^ <û ю ^ со CsJ 1- 



INnOO 



Ф о 



d §- 



26 



CHASE & BAILEY 




0.00 



6X6 2X2 1X1 6X6 2X2 1X1 



HIGH 



LOW 



FIG. 6. Correlation (± SE) between adults and recruits in the high and low density treatments at each of the 
three spatial scales; 1 - 1 cm, 2 x 2 cm and 6x6 cm. 



mortality that may have occurred. It is possi- 
ble that settlement did occur on the clean 
plates but the mussels did not survive to the 
time of census. Alternatively, another factor 
may have been acting as a deterrent to pre- 
vent settlement to the clean plates. Even on 
plates with adults, recruitment occurred only 
on or very close to the adults. It was ob- 
served on several occasions that herbivorous 
snails were present on the clean plates. While 
the mussels are at no risk of prédation from 
the snails, their presence may still act as a 
deterrent to recruitment. Barnacle recruit- 
ment can be reduced in the presence of lim- 
pets (Denley & Underwood, 1979; Miller, 
1986) because of the biological disturbance 
(i.e., bulldozing) by the limpets. Johnson & 
Strathmann (1989) demonstrated reduced 
settlement of barnacle larvae as a result of 
prior occupation of the substratum. Their re- 
sults indicated that mucus secretions may 
have been responsible for the reduction in 
settlement, because they may have affected 
the adhesion of the larvae or caused an al- 
teration in the existing cues present on the 
substratum (Johnson & Strathmann, 1989). 
The most likely explanation of the recruitment 
pattern observed in this study, however, is 
that of differential deposition and attach- 



ment, that is, recruitment is in response to a 
cue released by conspecific adults. Wainman 
et al. (1995) also observed no settlement on 
treatments without mussels. In addition to 
treatments with and without adult mussels, 
Wainman et al. (1995) included treatments 
with mussel-sized stones. These treatments 
served as a control to ascertain whether re- 
cruitment was in response to a chemical cue 
released by the adults or simply in response 
to the heterogeneity of the substrata. Wain- 
man et al. (1995) found settlement and re- 
cruitment was significantly lower on mussel- 
sized stones than on live mussels. This 
pattern was maintained even after 10-12 
days despite reduction in the numbers of re- 
cruits. This pattern suggests that although 
there is post settlement mortality, the pattern 
of settlement with conspecifics was main- 
tained. A laboratory study on settlement and 
metamorphosis of larval zebra and quagga 
mussels (Baldwin, 1995) provides further ev- 
idence for the presence of a chemical cue 
associated with conspecific adults. Baldwin 
(1995) found that in the laboratory D. poly- 
morpha settled and metamorphosed more 
readily on natural substrata (adult shells) and 
in water from adult rearing tanks as opposed 
to water not exposed to adults. On the basis 



RECRUITMENT OF DREISSENA POLYMORPHA 



27 



of our study and the research described 
above, it appears that D. polymorpha re- 
sponds to some chemical cue released by 
adult conspecifics that enhances settlement 
and recruitment. The exact nature of the cue, 
however, was beyond the scope of this 
study. 

Our study also examined the effect of in- 
creased density of adult mussels on the den- 
sity and spatial pattern of recruitment. Prox- 
imity to adults allows for synchronization of 
spawning and increased fertilization of 
spawned gametes as well as local introduc- 
tion of food as a result of adult filtering activ- 
ity. Large aggregations have a better chance 
of surviving physical disturbance and conse- 
quently gain a longer adult life span and over- 
all increased fecundity (Pawlik, 1986). How- 
ever, at high population densities, organisms 
may experience intense intraspecific compe- 
tition for space (Wu, 1980; Hui & Moyse, 

1987) and resources (Russo, 1979) and in- 
creased rates of both prédation (Fairweather, 

1988) and parasitization (Blower & Roughgar- 
den, 1989). Therefore, at high population 
densities there may be selective pressure for 
individuals to avoid conspecifics at settle- 
ment (Satchell & Farrell, 1993). 

Within this study, recruitment density was 
highest in the high density treatment. How- 
ever, variation in the adult densities within 
each of the high and low density treatments 
enabled the examination of the relationship 
between recruit density and adult density 
within each treatment. This variation is the 
result of differential attachment of adult mus- 
sels within the laboratory and subsequent 
loss of adults during transportation and 
placement in the field site. Although the data 
are limited, it was observed that within the 
high density treatment (n = 6 plates) there is 
a negative relationship between adult mussel 
density and the density of recruits, whereas 
in the low density treatment (n = 5 plates), 
there is a positive relationship between adult 
mussel density and recruit density. It is pos- 
sible that there is reduced recruitment with 
higher adult densities, but the adult densities 
employed in the high density treatments 
were not large enough to elicit such a re- 
sponse. When comparison was made be- 
tween the recruitment densities from the ex- 
perimental treatments and the natural 
population, it was observed that the recruit- 
ment density in the natural population was 
1 ,257 ± 1 78 individuals per m^ [which is com- 
parable to the recruitment density in the low 



density treatment (1 ,1 98 ± 1 41 individuals per 
m^)] but much lower than the recruitment 
density in the high density treatment (3,189 ± 
637 individuals/m^). Adult densities in the 
natural population were 8,029 ± 506 individ- 
uals per m^ versus only 4,583 ±419 individ- 
uals per m^ for the high density treatments. 
This suggests that there may be some avoid- 
ance of the high adult density in the natural 
population. 

Spatial Arrangement 

Examination of the spatial arrangement by 
nested analysis of covariance revealed a 
strong correlation between adults and re- 
cruits, confirming the observation that at all 
scales we tended to find recruits when adults 
were present. When the spatial arrangement 
was examined on three spatial scales the low 
density treatment had the highest correlation 
at the 2 X 2 cm scale, whereas the high den- 
sity treatment the highest correlation was at 
the 6 X 6 cm scale. However, correlation in 
the 2x2 and the 6 x 6 cm scales were not 
significantly different within treatments. This 
pattern suggests that while in the low density 
treatment the recruits are found closer to the 
adults than in the high density treatment, 
both treatments show the same conclusion 
that the recruitment occurs in response to the 
presence of adult conspecifics. 

In a patchy environment (represented by 
the low density treatment), the recruits must 
be close to the adults to obtain whatever 
benefit — protection, enhanced feeding — 
that such an association would elicit. This is 
indicative of the higher correlation at the 2 x 
2 cm scale. Hoffman (1989) suggested that 
gregarious settlement reduces stress on the 
vulnerable meta individual. Clumps of barna- 
cles may also influence water flow in a way 
that enhances feeding (Barnes & Powell, 
1950). However, in a more homogenous en- 
vironment (represented by the high density 
treatment) such a close association may be 
detrimental because of competition for space 
and resources and the increased risks of pré- 
dation of parasitism. In the high density treat- 
ment, the adults and recruits covaried on a 
larger scale (6x6 cm) than in the low density 
treatment, indicating a more uniform distribu- 
tion. In addition, the ratio of recruits to adults 
was much lower in the high density treatment 
(0.8 ± 0.2) than the low density treatment (7.5 
± 0.8). Hebert et al. (1 991 ) observed a marked 
heterogeneity in size and cohort structure in 



28 



CHASE & BAILEY 



D. polymorpha at different sites in Lake St. 
Clair in 1989. Members of the 1988 cohort 
had the smallest shell sizes at sites with the 
highest density, suggesting that their growth 
rates were slowed by intraspecific competi- 
tion. 

Natural Population 

Comparison of the mean length of recruits 
revealed that recruits in the natural popula- 
tion were significantly smaller than recruits in 
either the low or the high density treatments. 
The largest mean length of recruits was in the 
low density treatment (1.89 mm), which may 
suggest that increased competition for food 
in the high density environment resulted in 
reduced growth of recruits. Such a scenario 
will confer an advantage of recruiting into a 
low density habitat with either more space to 
grow or decreased competition for food with 
larger mussels. This observation may also 
explain the reduction in growth of mussels at 
Lake St. Clair since their introduction. Popu- 
lation densities near Stoney Point were only 
0.5 and 4,500 individuals per m^ in 1988 and 
1989 respectively (Hebert et al., 1991). At that 
time, Mackie (1 991 ) reported that an overwin- 
tering young adult between 1-4 mm in shell 
length will attain a shell length of 1 5 to 20 mm 
by the end of the year. Our data have shown 
that overwintering young adults of similar 
size (1-4 mm) had shell lengths of only 9 mm 
(Chase, unpublished data) by the end of the 
next year. Population densities at Stoney 
Point now exceed 10,000 individuals per m^ 
(Chase, unpublished data). Similar restriction 
in growth rates and survival of recent recruits 
of the barnacle Pollicipes polymerus were 
determined to be the result of competition 
between the established adults and the re- 
cruits for food resources (Page, 1986). When 
large adults were expehmentally removed 
from an aggregate, the smaller barnacles 
were able to increase rapidly in size (Page, 
1986). Larger barnacles may also have inter- 
fered with the water flow that brings food to 
the smaller barnacles (Page, 1986). In the 
mussel Mytilus edulis, Kautsky (1982) also re- 
ported that growth was suppressed in small 
mussels by increased density of large mus- 
sels. However, differences in size distribution 
and abundance may also be the result of dif- 
ferential recruitment between the expehmen- 
tal and natural population. This observation 
may also be the result of the raised level of 
the bricks in the water column, which may 



enhance growth of recruits. Pontius & Culver 
(1 995) found that D. polymorpha higher in the 
water column had larger biomass, which may 
indicate they were better able to obtain food. 
However, the significant difference between 
the low and high density treatments suggests 
an explanation other than height in the water 
column. 



Conclusion 

It appears that for D. polymorpha at Lake 
St. Clair, the presence and density of con- 
specifics are important determinants of the 
recruitment density and pattern in the popu- 
lation. The presence of adult conspecifics 
may offer some chemical cue that induces 
recruitment into the population. However, re- 
cruitment into a low density habitat may be 
advantageous because it may enhance the 
growth of young recruits. Therefore, in D. 
polymorpha there appears to be a tradeoff 
between adult densities that are high enough 
to provide an attachment site and protection 
but low enough to enhance growth and pos- 
sibly survival. Larger mussels may have a 
better chance at surviving the winter, and be- 
cause fecundity is related to size in most 
benthic invertebrates (Hughes, 1971; Spight 
& Emien, 1976; Brousseau, 1978; Sprung, 
1987; Chase & Thomas, 1995), recruitment 
into a low density habitat may also enhance 
reproductive output, assuming this size dif- 
ferential is maintained. 

The objective of this study was to examine 
recruitment. As such, the extent of post set- 
tlement mortality is unknown. It is possible 
the post settlement mortality was higher in 
the low density treatments than in either the 
high density treatment or the natural popula- 
tion. Thus, although our study suggests that 
recruitment into low density habitats may be 
advantageous because of enhanced growth 
and survival, it may have also suffered from a 
greater initial mortality. However, in terms of 
the ultimate survival and population structure 
of D. polymorpha at Lake St. Clair our results 
are valid. 



ACKNOWLEDGEMENTS 

We are grateful to R. Coulas, S. MacPher- 
son, J. Mitchell and S. Wolfenden for their 
assistance in the field. Special thanks to M. 
Topping for reading drafts of this paper and 



RECRUITMENT OF DREISSENA POLYMORPHA 



29 



offering constructive criticism. This research 
was funded through Natural Science and En- 
gineering Research Council of Canada, On- 
tario Ministry of Natural Resources, and the 
Great Lakes University Research Fund (Lake 
Erie Trophic Transfer) grants to R. C. Bailey. 



LITERATURE CITED 

BAILEY, R. C, M. E. CHASE & J. P. BECHTOLD, 
1995, An improved technique for estimating the 
density of benthic macroinvertebrates on cob- 
ble. Journal of Freshwater Ecology, 10(2): 189- 
192. 

BALDWIN, B. S., 1995, Settlement and metamor- 
phosis of larval zebra and quagga mussels: im- 
plications for their colonization and spread. Pa- 
per presented at the 5th International Zebra 
Mussel and other Aquatic Nuisance Organisms 
Conference, Toronto, Ontario, 22 February 1995. 

BARNES, H. & H. T. POWELL, 1950, The develop- 
ment, general morphology and subsequent elim- 
ination of barnacle populations {Balanus crena- 
tus and ß. balanoides) after heavy settlement. 
Journal of Animal Ecology, 19: 175-179. 

BERNSTEIN, B. В., В. E. WILLIAMS & К. H. MANN, 
1981, The role of behavioral responses to pred- 
ators in modifying urchins' {Strongylocentrotus 
droebachiiensis) destructive grazing and sea- 
sonal foraging patterns. Marine Biology, 63: 39- 
49. 

BLOWER, S. M. & J. ROUGHGARDEN, 1989, Par- 
asites detect host spatial pattern and density — a 
field experimental analysis. Oecologia, 78: 138- 
141. 

BOLSENGA, S. J. & С E. HERDENDORF, 1993, 
Lake Erie and Lake St. Clair Handbook, Wayne 
State University Press, Detroit. 

BROUSSEAU, D. J., 1978, Spawning cycle, fecun- 
dity and recruitment in a population of a soft 
shelled clam, Mya arenaria, from Cape Ann, 
Massachusetts. Fishieries Bulletin United States 
Fist! and Wildlife Sen/ices, 76: 155-166. 

BUNT, С M., H. J. MACLSAAC & W. G. SPRULES, 
1993, Pumping rates and projected filtering im- 
pacts of juvenile zebra mussels {Dreissena poly- 
morpfia) in western Lake Erie. Canadian Journal 
of Fishieries and Aquatic Science, 50: 1017- 
1022. 

BURKE, R. D., 1984, Pheromonal control of meta- 
morphosis in the Pacific sand dollar, Dendraster 
excenthcus. Science, 225: 442-443. 

BUTMAN, С A., 1989, Sediment-trap experiments 
on the importance of hydrodynamical processes 
in disturbing settling invertet^rate larvae in near- 
bottom waters. Journal of Experimental l\/larine 
Biology and Ecology, 134: 37-88. 

CHASE, M. E. & M. L. H. THOMAS, 1995, The ef- 
fect of the rate and onset of temperature in- 
crease on spawning of the periwinkle, Littorina 



littorea (L.). Journal of Experimental Marine Biol- 
ogy and Ecology, 186: 277-287. 

CONNELL, J. H., 1985, The consequences of vari- 
ation in initial settlement vs. post-settlement 
mortality in rocky intertidal communities. Journal 
of Experimental Marine Biology and Ecology, 93: 
11-45. 

CRISP, D. J., 1990, Gregariousness and system- 
atic affinity in some North Carolinian barnacles. 
Bulletin of Marine Science, 47: 516-525. 

CRISP, D. J. & D. A. RITZ, 1973, Responses of 
cirripede larvae to light. I. Experiments with 
white light. Marine Biology, 23: 327-335. 

DENLEY, E. J. & A. J. UNDERWOOD, 1 979, Exper- 
iments on factors influencing settlement, survival 
and growth of two species of barnacles in New 
South Wales. Journal of Experimental Marine Bi- 
ology and Ecology, 36: 269-293. 

DERMOTT, R. & M. MUNAWAR, 1993, Invasion of 
Lake Erie offshore sediments by Dreissena, and 
its ecological implications. Canadian Journal of 
Fishieries and Aquatic Science, 50: 2298-2304. 

FAIRWEATHER, P. G., 1988, Consequences of 
supply-side ecology: manipulating the recruit- 
ment of intertidal barnacles affects of intensity of 
prédation upon them. Biological Bulletin, 175: 
349-354. 

GARNICK, E., 1978, Behavioral ecology of 
Strongylocentrotus droebactiiensis (Müller) 
(Echinodermata: Echinoidea). Aggregating be- 
haviour and Chemotaxis. Oecologia, 37: 77-84. 

HEBERT, P., С WILSON, M. MURDOCH & R. 
LAZAR, 1991, Demography and ecological im- 
pacts of the invading mollusc Dreissena poly- 
morpha. Canadian Journal of Zoology, 69: 405- 
409. 

HIGHSMITH, R. C, 1982, Induced settlement and 
metamorphosis of sand dollar {Dendraster ex- 
centhcus) larvae in predator-free sites: adult 
sand dollar beds. Ecology, 63: 329-337. 

HOFFMAN, D. L, 1989, Settlement and recruit- 
ment patterns of a pedunculate barnacle, Pollic- 
ipes polymerus Sowerby, off La Jolla, California. 
Journal of Experimental Marine Biology and 
Ecology, 125: 83-98. 

HUGHES, R. N., 1 971 , Ecological energetics of the 
key hole limpet, Fissurella barbadensis, Gmelin. 
Journal of Experimental Marine Biology and 
Ecology, 6: 167-178. 

HUl, E. & J. MOYSE, 1987, Settlement patterns 
and competition for space. Pp. 363-376, in: A. j. 
SOUTHWOOD, ed.. Barnacle Biology. fK. A. 
Balkema, Rotterdam. 

JENSEN, R. A. & D. E. MORSE, 1984, Intraspecific 
facilitation of larval recruitment: gregarious set- 
tlement of the polychaete Phragmatopoma cali- 
fornica (Fewkes). Journal of Experimental Marine 
Biology and Ecology, 83: 107-126. 

JOHNSON, L. E., 1994, Enhanced settlement on 
microtopographical high points by the intertidal 
red alga Halosaccion glandiforme. Limnology 
and Oceanography, 39(8): 1893-1902. 

JOHNSON, L. E. & R. R. STRATHMANN, 1989, 



30 



CHASE & BAILEY 



Settling barnacle larvae avoid substrata previ- 
ously occupied by a mobile predator. Journal of 
Experimental Marine Biology and Ecology, 1 28: 
87-103. 

KAUTSKY, N., 1 982, Quantitative studies on gonad 
cycle, fecundity, reproductive output and re- 
cruitment in a Baltic Mytilus edulis population. 
Marine Biology, 68: 143-160. 

KNIGHT-JONES, E. W., 1953, Laboratory experi- 
ments on gregariousness during settling in Bal- 
anus balanoides and other barnacles. Journal of 
Experimental Marine Biology and Ecology, 30: 
584-599. 

MACISAAC, H. J., W. G. SPRULES, O. E. JO- 
HANNSON & J. H. LEACH, 1992, Filtenng im- 
pacts of larval and sessile zebra mussel {Dreis- 
sena polymorpha) in western Lake Erie. 
Oecologia, 92: 30-39. 

MACKIE, G. L., 1 991 , Biology of exotic zebra mus- 
sel, Dreissena polymorpha, in relation to native 
bivalves and its potential impact in Lake St. Clair. 
Hydrobiologia, 219: 251-268. 

MILLER, K. M., 1986, The role of spatial and size 
refuges in the interaction between juvenile bar- 
nacles and grazing limpets. MSc. Thesis, Univer- 
sity of British Columbia. 

MINCHINTON, T. E. & R. E. SCHEIBLING, 1991, 
The influence of larval supply and settlement on 
the population structure of barnacles. Ecology, 
72: 1867-1879. 

MORSE, A. N. С & D. E. MORSE, 1984, Recruit- 
ment and metamorphosis of Haliotis larvae in- 
duced by molecules uniquely available at the 
surface of crustose red algae. Journal of Exper- 
imental Marine Biology and Ecology, 75: 191- 
215. 

NEARY, B. P. & J. H. LEACH, 1992, Mapping the 
potential spread of the zebra mussel {Dreissena 
polymorpha) in Ontario. Canadian Journal of 
Fisheries and Aquatic Sciences. 49: 406-415. 

PAGE, H. M., 1986, Differences in population 
structure and growth rate of the stalked barnacle 
Pollicipes polymerus between a rocky headland 
and an offshore oil platform. Mahne Ecology 
Progressive Series, 29: 157-164. 

PAWLIK, J. R., 1986, Chemical induction of larval 
settlement and metamorphosis in the reef-build- 
ing tube worm Phragmatopoma californica (Sa- 
bellariidae: Polychaeta). Manne Biology, 91: 59- 
68. 

PAWLIK, J. J. & M. G. HADFIELD, 1990, A sympo- 
sium on chemical factors that influence the set- 
tlement and metamorphosis of marine inverte- 
brate larvae: introduction and perspective. 
Bulletin of Mahne Science, 46: 450-454. 

PAWLIK, J. R., С A. BUTMAN & V. R. STARCZAK, 
1991, Hydrodynamic facilitation of gregarious 
settlement of a reef-building tube worm. Sci- 
ence, 251: 421-424. 
PEARSE, J. S. & S. W. ARCH, 1969, The aggrega- 
tion behaviour of Diadema (Echinodermata: 
Echinoidea). Micronesica, 51: 166-171. 
PONTIUS, R. A. & D. A. CULVER, 1995, Estimating 



Dreissena impact on pelagic food webs based 
on size-specific grazing rates and population de- 
mographics. Paper presented at the 5th Interna- 
tional Zebra Mussel and other Aquatic Nuisance 
Organisms Conference, Toronto, Ontario, 22 
February 1995. 

RAMCHARAN, С W., D. K. PADILLA & S. I. DOD- 
SON, 1992, A multivariate model for predicting 
population fluctuations of Dreissena polymorpha 
in North American Lakes. Canadian Journal of 
Fishehes and Aquatic Sciences, 49: 150-158. 

RAIMONDI, P. T., 1988, Settlement cues and the 
determination of the vertical limit of an intertidal 
barnacle. Ecology, 69: 400-407. 

RAIMONDI, P. T., 1991, Settlement behaviour of 
Chthamalus anispoma larvae largely determines 
the adult distribution. Oecologia, 85: 349-360. 

RITTSCHOF, D., E. S. BRANSCOMB & J. D. 
COSTLOW, 1984, Settlement and behaviour in 
relation to flow and surface in larval barnacles, 
Balanus amphitrite Darwin. Journal of Experi- 
mental Mahne Biology and Ecology, 82: 
131-146. 

RODRIGUEZ, S. R., F. P. OJEDA & С INESTROSA, 
1993, Settlement of benthic marine inverte- 
brates. Manne Ecology Progressive Series, 97: 
193-207. 

ROFF, J. С & R. R. HOPCROFT, 1986, High pre- 
cision microcomputer based measuhng system 
for ecological research, Canadian Journal of 
Fisheries and Aquatic Sciences, 43: 2044-2048. 

RUSSO, A. R., 1979, Dispersion and food differ- 
ences between two populations of the sea ur- 
chin Strongylocentrotus franciscanus. Journal of 
Biogeography, 6: 407-414. 

SATCHELL, E. R. & T. M. FARRELL, 1993, Effects 
of settlement density on spatial arrangement in 
four intertidal barnacles. Manne Biology, 116: 
241-245. 

SCHANER, T., R. С BAILEY, N. A. RUKAVINA & P. 
J. SAVIOE, 1991, Diving survey to determine 
densities of zebra mussel in Western Lake On- 
tario in 1991 . Lake Ontario Fisheries Unit Annual 
Report, Section 7. 

SCHUBART, С. D., V. В. LAWRENCE & G. MIYA- 
SATO, 1995, Recruitment of Balanus glándula 
(Crustacea: Cirripedia) into empty barnacle tests 
and its ecological consequences. Journal of Ex- 
perimental Marine Biology and Ecology, 186: 
143-181. 

SEBENS, K. P., 1983, The larval and juvenile ecol- 
ogy of the temperate octocoral Alcyonium side- 
hum Vernl. I. substratum selection by benthic 
larvae. Journal of Expehmental Mahne Biology 
and Ecology, 71: 73-89. 

SEKI, T. & H. KAN-NO, 1981, Induced settlement 
of the Japanese abalone, Haliotis discus hannai, 
veliger by the mucus of the juvenile and adult 
abalones. Bulletin of Tohoku Regulatory Fisher- 
ies Research Laboratory, 43: 29-36. 

SPIGHT, T. M. & J. EMLEN, 1976, Clutch size of 
two marine snails with a changing food supply. 
Ecology, 57: 1162-1178. 



RECRUITMENT OF DREISSENA POLYMORPHA 



31 



SPRUNG, M., 1987, Ecological requirements of 
developing Dreissena polymorpha eggs. Archiv 
für Hydrobiologie/Supplement, 79: 69-86. 

SPRUNG, M., 1993, The other life: an account of 
present knowledge of the larval phase of Dreis- 
sena polymorpha. Pp. 39-53, in: т. F. nalepa & D. 
w. SCHLOSSER, eds., Zebra mussels: biology, im- 
pacts and control. Lewis Publishers, Chelsea, 
Ml. 

STRAYER, D. L., 1991, The projected distribution 
of the zebra mussel, Dreissena polymorpha, in 
North America. Canadian Journal of Fisheries 
and Aquatic Sciences, 48: 1389-1395. 

WAINMAN, B. C, S. S. HINCKS, N. K. KAUSHINK 
& G. L. MACKIE, 1995, Biofilm and substrate 



preference in the dreissenid larvae of Lake Erie. 
Canadian Journal of Fisheries and Aquatic Sci- 
ences, In press. 

WETHEY, D. S., 1986, Ranking os settlement cues 
by barnacle larvae: influence of surface contour. 
Bulletin of Marine Science, 39: 393-400. 

WU, R. S. S., 1980, Effects of crowding on the 
energetics of the barnacle Balanus glándula Dar- 
win. Canadian Journal of Zoology, 58: 559-566. 

YONGE, C. M. & F. S. CHIA, 1982, Factors con- 
trolling spatial distribution of the sea cucumber 
Psolus chitonoides: settling and post-settling 
behaviour. Marine Biology, 69: 195-205. 

Revised Ms. accepted 28 November 1995 



MALACOLOGIA, 1996, 38(1-2): 33-34 

ADDITIONAL NOTES ON NOMINA FIRST INTRODUCED BY TETSUAKI KIRA IN 
"COLOURED ILLUSTRATIONS OF THE SHELLS OF JAPAN" 

Rüdiger Bieler^ & Richard E. Petit^ 



The taxa, both available and unavailable, 
first proposed by Tetsuaki Kira in the numer- 
ous printings of his "Coloured Illustrations of 
the Shells of Japan" and the English edition, 
"Shells of the Western Pacific in Color, Vol. I" 
were recently listed by us (Bieler & Petit, 
1990). At the time, we discussed 54 species- 
group and one genus-group name. Forty 
names were found to be available from this 
work, although only five of these had been 
formally designated as new taxa. The diffi- 
culty in recognizing some of these unan- 
nounced introductions is demonstrated by 
our having to add two new taxa that we over- 
looked despite extensive searching and 
comparing the various printings and editions 
in which some nude names have been intro- 
duced. There probably remain still others that 
have eluded us. Also, we add additional data 
on the previously listed genus-group name. 

Laevistrombus Abbott, 1960 

Laevistrombus Kira, 1955: 31 {nomen nu- 
dum). 

Laevistrombus Kira. Abbott, 1960: 47-48 
(type species designated: Strombus ca- 
narium Linné, 1758). 

This name first appeared in the 3rd printing 
of the 1st edition of "Coloured Illustrations of 
the Shells of Japan" as a subgenus for two 
nominal species of Strombus: S. (L.) canar- 
ium Linné, 1758, and S. (L.) Isabella Lamarck, 
1822. No description or statement of differ- 
entiation was given, as required by ICZN 
Code Article 13a, nor was a type species 
designated. Subsequent printings remained 
unchanged at least through the 6th printing 
of the 2nd edition (1 963). In the 9th printing of 
the 2nd edition (1964), Laevistrombus is ele- 
vated to genus-level and L. Isabella emended 
to L. canarium "forma" Isabella. The two in- 
termediate printings have not been seen, but 
have no effect on this discussion. 

When Abbott (1 960: 47-48) treated Laevis- 



trombus as a subgenus in his monograph 
of Strombus, he gave a brief description of 
Laevistrombus and designated S. canarium 
Linné as its type species. Although Abbott 
cited Kira as the author of Laevistrombus, the 
name had not previously been available and 
must take date and authorship from Abbott, 
1960 (ICZN Code Article 50a). 



Simplicifusus Kuroda & Habe, 1971 

Simplicifusus Kira, 1962: 85 (nomen nudum). 

Simplicifusus Kira, 1964: 77 (nomen nudum). 

Simplicifusus Kira. Kuroda & Habe, 1971: 
282, 184 (type species designated: Fusi- 
nus simplex Smith [sic; = Fusus simplex 
E. A. Smith, 1879]). 

Simplicifusus first appeared in Kira's 
"Shells of the Western Pacific in Color" 
(1962: 85) as a subgenus of Fusinus for two 
species: F. (S.) hyphalus M. Smith and F. (S.) 
simplex (Smith) [= Fusinus hyptialus Maxwell 
Smith, 1940, and Fusus simplex E. A. Smith, 
1879]. We cannot determine exactly when 
this name first appeared in the Japanese ver- 
sion of this work, "Coloured Illustrations of 
the Shells of Japan, Vol. I." It was not in the 
6th printing (1963) but was in the 9th printing 
(1964). We have not seen the two intermedi- 
ate printings. However, Kira (1962, 1964) 
gave no description or statement of differen- 
tiating characters as required by ICZN Code 
Article 13a. 

Kuroda & Habe (1971) cited Simplicifus 
Kira as a genus (Japanese text, p. 282) and 
as a subgenus of Fusinus (English text, p. 
1 84). A description of the genus is given (Jap- 
anese text, p. 282), and Fusinus simplex 
(Smith) is designated as type species (Japa- 
nese p. 282; English p. 184). Because Sim- 
plicifusus was not previously an available 
name, it must take date and authorship from 
Kuroda & Habe, 1971 (ICZN Code Article 
50a). 



■'Field Museum of Natural History, Roosevelt Road at Lake Shore Drive, Chicago, Illinois 60605, U.S.A. 
2p. O. Box 30, North Myrtle Beach, South Carolina 29582, U.S.A. 



33 



34 



BIELER & PETIT 



Pictodentalium Habe, 1963 

Pictodentalium Kira, 1959: 105 (and subse- 
quent years; nomen nudum). 

Pictodentalium Habe, 1963: 255 (genus de- 
scribed; type species designated: Den- 
talium (Pictodentalium) formosum hirasei 
Kira, 1959). 

In our previous paper (1990: 141), we 
showed that this genus-group name was a 
nomen nudum in all editions of Kira's works. 
It was treated in a systematic manner by 
Habe In 1963 (p. 255), who gave a descrip- 
tion of it as a subgenus. He attributed the 
name to Kira and gave the type-species as 
Dentalium {Pictodentalium) formosum hirasei 
Kira, stating that designation was by mono- 
typy. He then placed D. (P.) formosum hirasei 
Kira in the synonymy of D. (P.) formosum (A. 
Adams & Reeve, 1850). Because Pictoden- 
talium had not previously been made avail- 
able, it must take date and authorship from 
Habe, 1963 (ICZN Code Article 50a). 

We thank Dr. Alan R. Kabat for bringing to 
our attention the omission of Simplicifusus in 
our earlier paper, and an anonymous re- 
viewer for additional data on the availability 
of Pictodentalium. 



editions of Tetsuaki Kira's "Coloured Illustra- 
tions of the Shells of Japan" and "Shells of the 
Western Pacific in Color, Vol. I," with an anno- 
tated list of new names introduced. Malacologia, 
32: 131-145. 

HABE, T., 1963, A classification of the scaphopod 
mollusks found in Japan and its adjacent area. 
Bulletin of the National Science Museum (Tokyo), 
6: 252-281, pis. 37-38. 

[ICZN] International Commission on Zoological No- 
menclature, 1985, International Code of Zoolog- 
ical Nomenclature, 3rd ed. London, Berkeley, 
and Los Angeles, xx + 338 pp. 

KIRA, T., 1955, Coloured illustrations of the shells 
of Japan, [viii] + 204 pp., 67 pis.; Hoikusha, Os- 
aka [for additional printings, see Bieler & Petit, 
1990]. 

KIRA, T., 1959, Coloured illustrations of the shells 
of Japan. Enlarged & Revised Edition. [1st print- 
ing of Revised Edition: March 10, 1959]. [6] + vii 
+ [1] + 239 pp., [1] + 71 pis.; Hoikusha, Osaka. 
[6th printing: February 5, 1963; 9th printing: No- 
vember 1, 1964; for additional printings, see 
Bieler & Petit, 1990]. 

KIRA, T., 1962, Shells of the western Pacific in 
color, [vii] + 224 pp., 72 pis. Hoikusha, Osaka [for 
additional printings, see Bieler & Petit, 1990]. 

KURODA, J. & T. HABE, 1971, [Descnptions of 
genera and species] in Kuroda, Habe & Oyama, 
7779 sea shells of Sagami Bay. Maruzen, Tokyo, 
xix + 741 pp. [in Japanese], pis. 1-121, 489 pp. 
[in English], 51 pp. index, map. 



LITERATURE CITED 



ABBOTT, R. T, 1960, The genus Strombus in the 

Indo-Pacific. Indo-Pacific Mollusca, 1: 33-146. 
BIELER, R. & R. E. PETIT 1990, On the various 



Revised Ms. accepted 28 November 1995 



MAIJ\COLOGIA, 1996, 38(1-2): 35-46 

ON THE NEW NAMES INTRODUCED IN THE VARIOUS PRINTINGS OF "SHELLS 

OF THE WORLD IN COLOUR" [VOL. I BY TADASHIGE HABE AND KIYOSHI ITO; 

VOL. II BY TADASHIGE HABE AND SADAO KOSUGE] 

Richard E. Petit^ & Rüdiger Bieler^ 



ABSTRACT 

The two volumes of "Shells of the World in Colour" (Vol. I, "The Northern Pacific" by Habe 
& Ito; Vol. II, "The Tropical Pacific" by Habe & Kosuge) contain many gastropod and bivalve 
names denoted as new therein. Some of these are nomina nuda made available only in later 
publications. However, the volumes also contain new taxa that are made available but not 
indicated as such. The problem is compounded by the existence of multiple printings of both 
volumes in which unexplained nomenclatural changes have been made. Forty-four species- 
group names and two genus-group names date from these works. Twelve genus-group names 
indicated as new were not made available until later. All pertinent treatments of these taxa are 
listed. 



INTRODUCTION 

In our continuing efforts to determine the 
status and correct dates of publication of 
various taxa proposed by Japanese authors, 
this paper discusses names introduced in the 
tv^o volumes of "Shells of the World in Co- 
lour" (Vol. I, "The Northern Pacific" by Habe 
& Ito; Vol. II, "The Tropical Pacific" by Habe 
& Kosuge). Both of these volumes went 
through numerous printings, with changes 
being made that are not indicated as such. 

Neither book is easy to locate, and few 
workers have access to more than one print- 
ing (we have failed to locate any copies of 
some printings). This paper lists the changes 
between printings that affect zoological no- 
menclature. At least 14 genus-group and 44 
species-group names are involved, spanning 
many marine gastropod and bivalve families. 

Of particular importance is the determina- 
tion of when a particular taxon was made 
available for taxonomic purposes. The de- 
scriptions of the species and subspecies in 
the two works under consideration are in 
Japanese and usually very brief. These spe- 
cies-group taxa are, however, accepted as 
being validly proposed. The genus-group 
taxa present more serious problems because 
12 of the new names were introduced with- 
out fulfilling ICZN Code Article 13 require- 
ments of providing a fixation of type species, 
and a differentiating description or indication 



to such. They are here regarded as nomina 
nuda and became available only in later 
works. Two names, Harpofusus and Mega- 
crenella, appear to fulfill the minimal require- 
ments set by the Code and are here ac- 
cepted as dating from their first appearance. 

It is hoped that the following notes will be 
of value to systematists who must refer to 
these taxa. We have maintained original or- 
thography when possible, and have not indi- 
cated some typographical errors and incor- 
rect usages in order to avoid using "[sic]" as 
much as possible. Readers should be aware 
that in addition to these "new" names there 
are numerous changes between the editions 
involving generic or (for subspecies) specific 
allocations, re-identifications, and adjust- 
ments in spelling and latinization. The works 
apparently were newly typeset, at least in 
part, between printings, sometimes resulting 
in a compounding of problems. An example 
of the combination of intended and acciden- 
tal changes is Habe & Ito's reference to a 
species of Neptúnea (p. 66, pi. 33, fig. 8); this 
was initially identified as Neptúnea minor and 
later (1977) corrected to ^'Neptúnea Ruro- 
sio," a lapsus for N. kuroshio Oyama, 1958. 

An example of the taxonomic confusion in 
these works is the nominal subspecies stii- 
rogai, first introduced as "Coííiseíla pelta stii- 
rogai Habe et Ito (nov.)" in 1965. The 1977 
printing of the work, referring to the same 
illustration, not only still indicated it as being 



^P. O. Box 30, North Myrtle Beach, South Carolina 29582, U.S.A. 

^Field Museum of Natural History, Roosevelt Road at Lake Shore Drive, Chicago, Illinois 60605, U.S.A. 



35 



36 



PETIT & BELER 



new, but changed the name of the species: 
"Collisella cassis shirogai Habe et Ito (nov.)." 
An additional layer of difficulty was intro- 
duced by printer's errors. For instance, 
"SL/cc/num chishimananux Habe et Ito 
(nov.)" of 1965a was meant to introduce a 
new subspecies, nux, for the species chishi- 
manum. 

Another example that has perplexed au- 
thors is Harpa kawamurai Habe, first intro- 
duced in the 3rd printing of Habe & Kosuge 
(1972) with no indication that it was new. 
Harpa kajiyamai Habe, which appeared at the 
same time, has never before been correctly 
cited in the literature. 

Systematists are urged to cite these works 
by printing. The date of a particular printing 
can be easily determined from the colophon 
(inscription at end of each copy). For details 
on date determination, see Bieler & Petit 
(1990: 132). 



LISTING OF NEW NAMES 

(A) Habe & Ito, 1965 (in sequence of occur- 
rence in volume; the work in which each 
taxon is considered to have been made 
available is shown by the usage of 
1965a, 1965b, or a later date) 



Collisella pelta shirogai, 1 965a 

Omphalomargarites, 1 965b 

Cirsotrema kagayai, 1 965a 

Bulbus flavus elongatus, 1 965a 

Trophonopsis scitula empliaticus, 1 965a 

Boreotrophon paucicostatus, 1 965a 

Nodulotrophon, 1965b 

Molinia multicostata, 1 965a 

Ancistrolepis troclioidea ovoidea, 1 965a 

Fusipagoda, 1965b 

Buccinum cliishimanum nux, 1 965a 

Buccinum hosoyai, 1 965a 

Buccinum opisthoplectum microcon- 

clia, 1965a 

Buccinum felis shikamai, 1 965a 

Buccinum kawamurai, 1965a 

Clinopegma buccinoides, 1 965a 

Neoberingus, 1965b 

Beringion, 1965b 

Harpofusus, 1965a 

Volutopsion, 1965b 

Buccinum subreticulatum, 1965a 

Buccinum ferrugineum, 1965a 

Buccinum kinukatsugi Habe & Ito, 1 968 

Buccinum midori, 1 965a 

Boreomelon stearnsii ryosukei, 1 965a 



(26) Fulgoraria (Musashia) kanaka hayashii, 
1965a 

(27) Decollidrillia, 1965b 

(28) Decollidrillia nigra, 1965a 

(29) Megacrenella, 1965a 

(30) Adula californiensis chosenica, 1 965a 

(31) Megacardita ferruginosa koreana, 1965a 

(B) Habe & Kosuge, 1966 (in sequence of 
occurrence in volume; the work in which 
each taxon is considered to have been 
made available is shown by the usage of 
1966a, 1966b, or a later date) 

(32) Patelloida (Col I i sel Una) saccfiarinoides, 
1966a 

(33) Astralium yamamurae, 1 966a 

(34) Granulittorina, 1966b 

(35) Granulittorina philippiana, 1 966a 

(36) Clypeomorus batillariaeformis, 1 966a 

(37) Ficadusta, 1966b 

(38) Reticutriton, 1966b 

(39) Spinidrupa, 1966b 

(40) Pyrene testudinaria nigropardalis, 
1966a 

(41) Pyrene lacteoides, 1966a 

(42) Plicarcularia gibbosuloidea, 1 966a 

(43) Hemifusus carinifer, 1 966a 

(44) Latirus stenomphalus, 1 966a 

(45) Vexillum rubrocostatum, 1 966a 

(46) Nebularia yaekoae, 1 966a 

(47) Harpa kawamurai Habe, in Habe & 
Kosuge, 1972 

(48) Harpa kajiyamai Habe, in Habe & 
Kosuge, 1972 

(49) Volutoconus grossi mcmichiaeli, 1 966a 

(50) Brachytoma kurodai, 1 966a 

(51) Bractiytoma kawamurai, 1966a 

(52) Brachytoma vexillium, 1966a 

(53) Eglisia brunnea, 1 966a 

(54) Mantellum perfragile, 1 966a 

(55) Anomiostrea, 1966b 

(56) Laevicardium rubropictum, 1 966a 

(57) Vasticardium nigropunctatum, 1 966a 

(58) Macrotoma yamamurae, 1 966a 



DISCUSSION BY VOLUME 

"Shells of the World in Colour, Vol. I. The 
Northern Pacific." Tadashige Habe and 
Kiyoshi Ito 

The first printing of "Shells of the World in 
Colour, Vol. I" is dated June 1 , 1 965 (Habe & 
Ito, 1965a). A paper by Habe & Ito published 
in Venus (The Japanese Journal of Malacol- 
ogy) on July 31, 1965 (1965b) also contains 



SHELLS OF THE WORLD IN COLOR 



37 



descriptions of taxa, indicated as new 
therein, which had been shown as new in the 
book. In the next few years there were sev- 
eral printings of the book; Dr. Kosuge (per- 
sonal comm., March 15, 1995) advises that 
the 11th printing appeared in March 1991. 
Printings that we have seen: 



Printing 1 June 1, 1965 (1965a) 

2 September 1, 1970 

4 August 1, 1972 

5 January 20, 1974 
8 October 1, 1977 

The following new species appear (using 
the original arrangement of families). Impor- 
tant changes between printings and refer- 
ences from other sources are also listed. 



and with type species stated to be by 
original designation) 
Omphalomargarites Habe et Ito. Habe, 1977: 
90 (cited as of 1 965a, with type as Mar- 
gantes vorticifera (Dall, 1873) by mono- 
typy; 1965b mentioned only as a "cf." 
reference). 

Epitoniidae 

(3) Cirsotrema kagayai Habe & Ito, 1965a 

Cirsotrema kagayai Habe et Ito (nov.). Habe & 

Ito, 1965a: 29, pi. 7, fig. 25; 1970, 1972, 

1974, 1977: ibid. 
Cirsotrema kagayai sp. nov. Habe & Ito, 

1965b: 17, 30, pi. 2, fig. 9. 
Cirsotrema kagayai Habe et Ito. Habe, 1 977: 

56 (cited as of 1965a). 



GASTROPODA 



Acmaeidae 



(1) Collisella cassis shirogai Habe & Ito, 1965a 

Collisella pelta shirogai Habe et Ito (nov.). 

Habe & Ito, 1965a: 11, pi. 4, fig, 18; 

1970, 1972, 1974: ibid. 
Collisella pelta shirogai subsp. nov. Habe & 

Ito, 1965b: 16, 29, pi. 4, fig. 8. 
Collisella pelta shirogai Habe et Ito. Habe, 

1977: 111 (cited as of 1965a). 
Collisella cassis shirogai Habe et Ito (nov.). 

Habe& Ito, 1977: 11, pl. 4, fig. 18. 

Trochidae 

(2) Omphalomargarites Habe & Ito, 1 965b 

Omphalomargarites (nov) vorticifera (Dall, 
1873). Habe & Ito, 1965a: 17, pl. 6, figs. 
6, 7; 1970, 1972, 1974, 1977: ibid, (ge- 
nus-group name = nomen nudum). 

Omphalomargarites subgen. nov. Habe & Ito, 
1 965b: 1 7 (type species Margantes vor- 
ticifera (Dall, 1873), with no indication of 
genus in which it was to be placed). 

Omphalomargarites gen. nov. Habe & Ito, 
1965b: 30 (type species. Margantes vor- 
ticifera (Dall, 1873). 

Omphalomargahtes (gen. nov.) vorticifera 
(Dall). Habe & Ito, 1965b: 45 (plate cap- 
tion for pl. 2). 

Omphalomargarites Habe & Ito. Kuroda & 
Habe, 1971: 31(21) (with Habe & Ito, 
1965b, given precedence over 1965a, 



Naticidae 

(4) Bulbus flavus elongatus Habe & Ito, 1965a 

Bulbus flavus elongatus Habe et Ito (nov). 

Habe & Ito, 1 965a: 31 , pl. 8, fig. 8; 1 970, 

1972, 1974, 1977: ibid. 
Bulbus flavus elongatus subsp. nov. Habe & 

Ito, 1965b: 17, 31, pl. 3, fig. 2. 
Bulbus flavus elongatus Habe et Ito. Habe, 

1977: 38 (cited as of 1965a). 

Muricidae 

(5) Trophonopsis scitula emphaticus Habe & 

Ito, 1965a 

Trophonopsis scitula emphaticus Habe et Ito 

(nov.). Habe & Ito, 1965a: 36, pl. 10, fig. 

10; 1970, 1972, 1974, 1977: ibid. 
Trophonopsis scitula emphaticus subsp. nov. 

Habe & Ito, 1965b: 18, 31, pl. 2, fig. 1. 
Trophonopsis scitulus emphatica Habe et Ito. 

Habe, 1977: 38 (cited as of 1965a). 

(6) Boreotrophon paucicostatus Habe & Ito, 

1965a 

Boreotrophon paucicostatus Habe et Ito 
(nov: [sic]). Habe & Ito, 1965a: 37, pl. 10, 
fig. 13; 1970, 1972, 1974, 1977: ibid. 

Boreotrophon paucicostatus sp. nov. Habe & 
Ito, 1965b: 18, 32, pl. 2, fig. 10. 

Boreotrophon paucicostatus Habe et Ito. 
Habe, 1977: 95 (cited as of 1965a). 

(7) Nodulotrophon Habe & Ito, 1965b 

Nodulotrophon (nov.) dalli (Kobelt, 1878). 
Habe & Ito, 1965a: 37, pl. 10, fig. 14; 



38 



PETIT & BIELER 



1970, 1972, 1974, 1977: ibid. (genus- 

group name = nomen nudum). 
Nodulotrophon gen. nov. Habe & Ito, 1965b: 

19, 32 (with type species as Trophon 

da/// Kobelt, 1878). 
Nodulotrophon Habe et Ito. Habe, 1977: 87 

(cited as of 1965a, with type, by mono- 

typy, Trophon dalli Kobelt, 1878; 1965b 

not mentioned). 
Taxonomic note: This genus-group name 

must date from 1965b because there 

was no description in 1965a. 

Buccinidae 

(8) Mohnia multicostata Habe & Ito, 1 965a 

Mohnia multicostata Habe et Ito (nov.). Habe 

& Ito, 1965a: 45, pi. 13, fig. 12; 1970, 

1972, 1974, 1977: ibid. 
Mohnia multicostata sp. nov. Habe & Ito, 

1965b: 19, 33, pi. 2, fig. 2. 
Mohnia multicostata Habe et Ito. Habe, 1977: 

80 (cited as of 1965a). 

(9) Ancistrolepis trochoidea ovoidea Habe & 

Ito, 1965a 

Ancistrolepis trochoideus ovoideus Habe et 
Ito (nov.). Habe & Ito, 1965a: 46, pi. 13, 
fig. 18; 1970, 1972, 1974, 1977: ibid. 

Ancistrolepis trochoideus ovoideus subsp. 
nov. Habe & Ito, 1965b: 20, 33, pi. 2, fig. 
13. 

Ancistrolepis trochoidea [Bathyancistrolepis] 
ovoidea Habe et Ito. Habe, 1977: 92 
(cited as of 1965a). 

(1 0) Fusipagoda Habe & Ito, 1 965b 

Fusipagoda (nov.) exquisita Dali, 1913. Habe 
& Ito, 1 965a: 48; Habe & Ito, 1 970, 1 972, 
1974, 1977: ibid, (genus-group name = 
nomen nudum). 

Fusipagoda gen. nov.. Habe & Ito, 1965b: 21 
(with type species as Mohnia exquisita 
Dali). 

Fusipagoda Habe et Ito. Habe, 1977: 43 
(cited as of 1965b with type species as 
cited, by original designation; 1965a 
cited as "name only"). 

(11) Buccinum chishimanum nux Habe & Ito, 
1965a 

Buccinum chishimananux [sic] Habe et Ito 
(nov.). Habe & Ito, 1965a: 49, pi. 14, fig. 
2. 

Buccinum chishimanum nux subsp. nov. 
Habe & Ito, 1965b: 22, 36, pi. 2, fig. 7. 



Buccimum [sic] chishimana nux Habe et Ito 
(nov.). Habe & Ito, 1970:49, pi. 14, fig. 1; 
1972, 1974, 1977: ibid. 

Buccinum chishimanum nux Habe et Ito. 
Habe, 1977: 88 (cited as of 1965a). 

(12) Buccinum hosoyai Habe & Ito, 1965a 

Buccinum hosoyai Habe et Ito (nov.). Habe & 
Ito, 1965a: 49, pi. 14, fig. 2; Habe & Ito, 
1970, 1972, 1974, 1977: ibid. 

Buccinum hosoyai sp. nov. Habe & Ito, 
1965b: 23, 36, pi. 2, fig. 8. 

Buccinum hosoyai Habe et Ito. Habe, 1977: 
49 (cited as of 1965a). 

(13) Buccinum opisthoplectum microconcha 
Habe & Ito, 1965a 

Buccinum opisthoplectum microconcha 
Habe et Ito (nov.). Habe & Ito, 1965a: 50, 
pi. 14, fig. 7; 1970, 1972, 1974, 1977: 
ibid. 

Buccinum opisthoplectum microconcha 
subsp. nov. Habe & Ito, 1965b: 23, 37, 
pi. 2, fig. 6. 

Buccinum opisthoplectum microconcha 
Habe et Ito. Habe, 1977: 75 (cited as of 
1965a; stated to be a synonym of Buc- 
cinum japonicum A. Adams, 1861). 

(14) Buccinum fells shikamai Habe & Ito, 
1965a 

Buccinum fells shikamai Habe et Ito (nov.). 

Habe & Ito, 1 965a: 50; 1 970, 1 972, 1 974, 

1977: ibid. 
Buccinum fells shikamai subsp. nov. Habe & 

Ito, 1965b: 23, 37, pi. 2, fig. 5. 
Buccinum fells shikamai Habe et Ito. Habe, 

1977: 110 (cited as of 1965b). 

(15) Buccinum kawamurai Habe & Ito, 1965a 

Buccinum kawamurai Habe et Ito (nov.). 

Habe & Ito, 1965a: 52, pi. 15, fig. 1; 

1970, 1972, 1974, 1977: ibid. 
Buccinum kawamurai sp. nov. Habe & Ito, 

1965b: 26, 38, pi. 2, fig. 11. 
Buccinum kawamurai Habe et Ito. Habe, 

1977: 58 (cited as of 1965a). 

(16) Clinopegma buccinoides Habe & Ito, 
1965a 

Clinopegma buccinoides Habe et Ito (nov.). 

Habe & Ito, 1965a: 55, pi. 16, fig. 1; 

1970, 1972, 1974, 1977: ibid. 
Clinopegma buccinoides sp. nov.. Habe & 

Ito, 1965b: 27, 41, pi. 2, fig. 12. 



SHELLS OF THE WORLD IN COLOR 



39 



Clinopegma buccinoides Habe et Ito. Habe, 
1977: 28 (cited as of 1965a). 

(1 7) Neoberingius Habe & Ito, 1 965b 

Neobehngius (nov.) frielei Dall, (1895) [sic]. 
Habe & Ito, 1965a: 57, pi. 17, fig. 3. (ge- 
nus-group name = nomen nudum). 

Neoberingius gen. nov. Habe & Ito, 1965b: 
21, 35, pi. 3, fig. 7. (type species, Ber- 
ingius frielei Dall, 1894 [sic]). 

Neoberingius (nov.) frielei (Dall, 1895). Habe 
& Ito, 1970: 57, pi. 17, fig. 3; 1972, 1974, 
1977: ibid. 

Neoberingius Habe et Ito. Habe, 1977: 83 
(cited as of 1965b, with type species as 
cited, by original designation; 1965a 
cited as "name only"). 

(18) Beringion Habe & Ito, 1965b 

Beringion (nov.) marsfialli (Dall, 1919). Habe & 
Ito, 1965a: 58, pi. 17, fig. 4; 1970, 1972, 
1974, 1977: ibid, (genus-group name = 
nomen nudum). 

Beringion gen. nov. Habe & Ito, 1965b: 21, 
35, pi. 3, fig. 6 (with type species as Ber- 
ingius marshal li Dall, 1919). 

Beringion Habe et Ito. Habe, 1977: 27 (cited 
as of 1965b, with type species as cited, 
by original designation; 1965a referred 
to with comments: "f. 4 as Beringion 
(nov.) marshalli; f. 5, B. beringii, with a 
notice of 'the type-species of Beringion\ 
name only"). 

Taxonomic note: Habe's statement (1977: 
83) is ambiguous as the mention of 
"type species" in the Japanese text is in 
the context of ''Beringion type species is 
figured," which appears in discussion of 
B. beringii (Middendorff). He cited the 
new genus as of 1965b, which we con- 
sider to be correct. 

(19) Harpofusus Habe & Ito, 1965a 

Harpofusus (nov.) melonis (Dall, 1891). Habe 
& Ito, 1965a: 59, pi. 18, fig. 1; 1970, 
1972, 1974: ibid, (with type species, by 
monotypy, Harpofusus melonis (Dall, 
1891)). 

Harpofusus gen. nov. Habe & Ito, 1965b: 20, 
34 (with type species as Pyrulofusus 
melonis Dall, 1891 [as Pyrofusus on p. 
20]). 

Pyrulofusus {Harpofusus) melonis (Dall, 
1891). Habe & Ito, 1977:59, pi. 18, fig. 1. 

Harpofusus Habe et Ito. Habe, 1977: 46. 
(listed as a genus of Buccinidae, cited as 



of 1965a, with type species, by mono- 
typy, Pyrulofusus melonis (Dall, 1891) [= 
Strombella melonis Dall, 1891]). 
Taxonomic note: Habe & Ito (1965a: 59) 
move this species from its previous 
placement (Volutopsis, name given in 
Japanese only) into a new genus, based 
on the yellowish-orange aperture color- 
ation and the vertical shell folds and spi- 
ral ribs. Similarity to Pyrulofusus is also 
mentioned. This fulfills the ICZN Code 
requirements, and we date this taxon as 
of 1965a. 

(20) Volutopsion Habe & Ito, 1965b 

Volutopsion (nov.) castaneus (Mörch, 1858). 
Habe & Ito, 1965a: 62, pi. 20, fig. 6; 
1970, 1972, 1974, 1977: ibid, (genus- 
group name = nomen nudum). 

Volutopsion gen. nov. Habe & Ito, 1965b: 21, 
35, pi. 2, fig. 15 (with type species as 
Volutopsius castaneus Dall [sic]). 

Volutopsion Habe et Ito. Habe, 1977: 131 
(cited as of 1965a with type species, by 
monotypy, Volutopsion castaneus [-um] 
(Mörch, 1858); 1965b is mentioned only 
as a "cf." reference, with same type 
species indicated, but by original desig- 
nation. We consider this genus name to 
be available from 1965b, with type spe- 
cies, by original designation, Volutop- 
sion castaneum (Mörch, 1858) [= Neptú- 
nea castanea Mörch, 1858]. 

(21) Buccinum subreticulatum Habe & Ito, 
1965a 

Buccinum Subreticulatum [sic] Habe et Ito 

(nov.). Habe & Ito, 1965a: 73, pi. 27, fig. 

4. 
Buccinum subreticulatum sp. nov. Habe & 

Ito, 1965b: 24, 39, pi. 2, fig. 14. 
Buccinum subreticulatum Habe et Ito (nov.). 

Habe & Ito, 1970: 73, pi. 27, fig. 4; 1972, 

1974, 1977: ibid. 
Buccinum subreticulatum Habe et Ito. Habe, 

1977: 118 (cited as of 1965a). 

(22) Buccinum ferrugineum Habe & Ito, 
1965a 

(23) Buccinum kinukatsugi Habe & Ito, 1 968 

Buccinum ferrugineum Habe et Ito (nov.). 

Habe & Ito, 1965a: 76, pi. 28, fig. 8. 
Buccinum ferrugineum sp. nov. Habe & Ito, 

1965b: 25, 40, pi. 3, fig. 3. 
Buccinum kinukatsugi nom. nov. Habe & Ito, 



40 



PETIT & BIELER 



1968: 2, 5, pl. 1, fig. 4 (new name for 
Buccinum ferrugineum Habe & Ito, 1965, 
non Born, 1780 [sic; = 1778]). 

Buccinum kinukatsugi Habe et Ito (nov.). 
Habe & Ito, 1970: 76, pl. 28, fig. 8; 1972, 
1974, 1977: ibid. 

Buccinum l<inukatsugi Habe et Ito. Habe, 
1977: 63 (cited as of 1968). 

(24) Buccinum midori Habe & Ito, 1 965a 

Buccinum midori Habe et Ito (nov.). Habe & 

Ito, 1965a: 76, pl. 28, fig. 9; 1970, 1972, 

1974, 1977: ibid. 
Buccinum midori sp. nov. Habe & Ito, 1965b: 

25, 40, pl. 2, fig. 16. 
Buccinum midori Habe et Ito. Habe, 1977: 75 

(cited as of 1965a). 

Volutidae 

(25) Boreomelon stearnsii ryosukei Habe & 
Ito, 1965a 

Boreomelon stearnsii ryosukei Habe et Ito 

(nov.). Habe & Ito, 1965a: 77, pl. 29, fig. 

2; 1970, 1972, 1974: ibid. 
Boreomelon stearnsii ryosukei subsp. nov. 

Habe & Ito, 1965b: 26, 42, pl. 2, fig. 17. 
Boromelon [sic] stearnsii ryosukei Habe et Ito 

(nov.). Habe & Ito, 1977: 77, pl. 29, fig. 2. 
Boreomelon stearnsii ryosukei Habe et Ito. 

Habe, 1977: 103 (cited as of 1965a). 

(26) Fulgoraria (Musashia) kaneko hayashii 
Habe& Ito, 1965a 

Fulgoraria (Musashia) kaneko hayashii Habe 

et Ito (nov.). Habe & Ito, 1965a: 77, pl. 

29, fig. 4; 1970, 1972, 1974, 1977: ibid. 
Fulgoraria (Musashia) kaneko hayashii subsp. 

nov. Habe & Ito, 1965b: 26, 42, pl. 3, fig. 

5. 
Fulgoraria (Musashia) kaneko hayashii Habe 

et Ito. Habe, 1 977: 47 (cited as of 1 965a). 

Turridae 

(27) Decollidrillia Habe & Ito, 1 965b 

(28) Decollidhllia nigra Habe & Ito, 1 965a 

Decollidrillia nigra Hade [sic] et Ito (nov.). 
Habe & Ito, 1965a: 80, pl. 30, fig. 6. (ge- 
nus-group name = nomen nudum). 

Decollidhllia nigra gen. et sp. nov. Habe & Ito, 
1965b: 27, 43, pl. 4, fig. 6. 

Decollidrillia nigra Habe et Ito (nov.). Habe & 
Ito, 1970: 80, pl. 30, fig. 6; 1972, 1974, 
1977: ibid. 



Decollidrillia Habe et Ito. Habe, 1977: 35 
(cited as of 1965b, with type, by original 
designation, D. nigra; 1965a cited as 
"name only"). 

Decollidhllia nigra Hade [-be] et Ito. Habe, 
1977: 83 (species name cited as of 
1965a). 

Taxonomic note: We agree that this new ge- 
nus dates from 1965b, but type desig- 
nation is by monotypy (Articles 13c, 
68d). 



Bivalvia 



Mytilidae 



(29) Megacrenella Habe & Ito, 1965a 

Crenella (Megacrenella nov.) columbiana 
Dall, 1897. Habe & Ito, 1965a: 109, pl. 
35, fig. 11; 1970, 1972, 1974, 1977: ibid, 
(with type species, by monotypy, 
Crenella (Megacrenella) columbiana 
(Dall, 1897)). 

Megacrenella gen. nov. Habe & Ito, 1965b: 
28, 44, pl. 3, fig. 4 (with type species as 
Crenella columbiana Dall, 1897; 1965a 
listed as a "cf." reference). 

Megacrenella Habe et Ito. Habe, 1977: 74 
(cited as of 1965a, with type species, by 
original designation, Crenella columbi- 
ana Dall, 1897) 

Taxonomic note: We consider the type indi- 
cation as by monotypy. The two other 
nominal taxa mentioned in the Japanese 
text are clearly stated to be synonyms of 
Crenella columbiana. Habe & Ito (1965a: 
1 00, in Japanese) refer to something that 
translates to "type species group," 
which we cannot accept as original des- 
ignation. The authors discuss the posi- 
tion of the group, based on morpholog- 
ical characters, as standing between 
Solamen and Crenella (the latter name 
mentioned only in Japanese characters) 
and also indicate its relationship to 
Arvella. This appears to fulfill the ICZN 
Code requirements, and we date this 
taxon as of 1965a. 

(30) Adula californiensis chosenica Habe & 
Ito, 1965a 

Adula californiensis chosenica (Kuroda MS.) 
Habe et Ito (nov.). Habe & Ito, 1965a: 
114, pl. 37, fig. 4; 1970, 1972, 1974, 
1977: ibid. 



SHELLS OF THE WORLD IN COLOR 



41 



Adula californiensis chosenica subsp. nov. 

Habe & Ito, 1965b: 28, 43, pi. 3, fig. 1. 
Adula californiensis chosenica Habe et Ito. 

Habe, 1977: 31 (cited as of 1965a and 

stated to be a synonym of A. schmidti 

(Schrenck, 1867)). 

Carditidae 

(31) Megacardita ferruginosa koreana Habe & 
Ito, 1965a 

Megacardita ferruginosa koreana Habe et Ito 

(nov.). Habe & Ito, 1965a: 128, pi. 43, fig. 

8; 1970, 1972, 1974, 1977: ibid. 
Megacardita ferruginea [sic] koreana subsp. 

nov. Habe & Ito, 1965b: 28, 45 (plate 

caption), pi. 3, fig. 8. 
Megacardita ferruginea [sic] koreanica [sic] 

subsp. nov. Habe & Ito, 1965b: 44. 
Megacardita ferruginosa koreana Habe et Ito. 

Habe, 1977: 65 (cited as of 1965a). 



"Shells of the World in Colour, Vol. II. 

The Tropical Pacific." Tadashige Habe and 

Sadao Kosuge 

First published January 15, 1966 (1966a), 
this work preceded an article in Venus by the 
same authors (1966b) in which new taxa, first 
appearing in Volume II, are proposed. There 
is no indication in Volume II that these taxa 
are newly introduced therein. The authors 
stated (1966b) that these "genera and spe- 
cies were figured and briefly described" in 
1966a and that "they are redeschbed in de- 
tail herewith in the nomenclatural value." Dr. 
Kosuge (personal comm., March 15, 1995) 
has confirmed that the genera all must date 
from the Venus article. 

Dr. Kosuge also advises that there are at 
least ten printings of this work, the 10th ap- 
pearing in March, 1991. 

Printings that we have seen: 



Gastropoda 



Acmaeidae 



Printing 1 
2 
3 
5 
6 



January 15, 1966 (1966a) 
November 1, 1966 (1966c) 
February 1, 1972 
November 11, 1974 
September 1, 1976 



The following new species appear (using 
the original arrangement of families). Impor- 
tant changes between printings and refer- 
ences from other sources are also listed. 



(32) Patelloida (Collisellina) saccharinoides 
Habe & Kosuge, 1966a 

Patelloida (Collisellina) saccharinoides Habe 
et Kosuge. Habe & Kosuge, 1966a: 6, pi. 
2, fig. 10; 1966c, 1972, 1974, 1976: ibid. 

Patelloida (Collisellina) saccharinoides Habe 
et Kosuge (sp. nov.). Habe & Kosuge, 
1966b: 312. 

Patelloida (Collisellina) saccharioides [sic] 
Habe et Kosuge (sp. nov.). Habe & Ko- 
suge, 1966b: 326, pi. 29, fig. 6 (this spell- 
ing also on plate caption on same page). 

Patelloida (Collisellina) saccharinoides Habe 
et Kosuge. Habe, 1977: 103 (cited as of 
1966a). 

Turbinidae 

(33) Astralium yamamurae Habe & Kosuge, 
1966a 

Astralium yamamurai [sic] Habe et Kosuge. 
Habe & Kosuge, 1966a: 11; 1966c: ibid, 
(error in spelling corrected on page 121 
and in all later usages) 

Astralium (Distellifer [sic]) yamamurae Habe 
et Kosuge. Habe & Kosuge, 1966a: 121, 
pi. 45, fig. 11; 1966c, 1972, 1974, 1976: 
ibid. 

Astralium yamamurae Habe et Kosuge. Habe 
& Kosuge, 1972: 11; 1974, 1976: ibid. 

Astralium (Destellifer) yamamurae Habe et 
Kosuge (sp. nov.). Habe & Kosuge, 
1966b: 313, 327 (with reference to 
1966a, pi. 45, fig. 4 [s/c; error for fig. 11]). 

Astralium (Destellifer) yamamurae Habe et 
Kosuge. Habe, 1977: 133 (cited as of 
1966a; 1966b cited as "name only"). 

Littorinidae 

(34) Granulilittorina Habe & Kosuge, 1 966b 

(35) Granulilittohna philippiana Habe & Ko- 
suge, 1966a 

Granulilittorina philippiana Habe et Kosuge. 

Habe & Kosuge, 1966a: 20, pi. 6, fig. 13; 

1966c; ibid, (genus-group name = 

nomen nudum). 
Granulilittorina philippiana Habe et Kosuge 

(gen. et sp. nov.). Habe & Kosuge, 

1966b: 313, 328 (with reference to 

1966a, pi. 6, fig. 13). 



42 



PETIT & BIELER 



Granulilittorina millegrana (Philippi) Habe et 
Kosuge. Habe & Kosuge, 1972, 1974, 
1976: 20, pl. 6, fig. 13. 

Granulilittorina Habe et Kosuge. Habe, 1977: 
45. (cited as of 1966b, with type, by 
monotypy, G. pliilippiana Habe & Ko- 
suge; 1966a not mentioned). 

Granulilittorina philippiana Habe et Kosuge. 
Habe, 1977: 96 (cited as of 1966a; 
stated to be a synonym of G. millegrana 
(Philippi, 1848)). 

Taxonomic note: In 1966a no indication was 
given that this was a newly introduced 
genus-group name. Rosewater (1970: 
491-493) used Granulilittorina as a valid 
subgenus of Nodilittohna. In treating the 
genus-group name he cited both 1966a 
and 1966b. However, under the species 
name (in the synonymy of N. (G.) mille- 
grana) he listed as of 1966b with "[fig- 
ured in] Habe and Kosuge" 1966a 
(square brackets in quote are of Rose- 
water). 

Cerithiidae 

(36) Clypeomorus batillariaeformis Habe & 
Kosuge, 1966a 

Clypeomrus [sic] batillariaeformis Habe et 
Kosuge. Habe & Kosuge, 1966a: 23, pl. 
7, fig. 14; 1966c: ibid. 

Clypeomorus batillariaeformis Habe et Ko- 
suge (sp. nov.) Habe & Kosuge, 1966b: 
314, 328, pl. 29, fig. 1 3 (with reference to 
1966a, pl. 7, fig. 14). 

Clypeomorus batillariaeformis Habe et Ko- 
suge. Habe & Kosuge, 1 972, 1 974, 1 976: 
23, pl. 7, fig. 14. 

Clypeomorus batillariaeformis Habe et Ko- 
suge. Habe, 1977: 26 (cited as of 1966a; 
original misspelling of genus shown and 
corrected). 

Taxonomic note: Houbrick (1985: 51) treated 
this species in detail and attributed it to 
Habe & Kosuge, 1966b, without any 
mention of 1966a. 

Cypraeidae 

(37) Ficadusta Habe & Kosuge, 1966b 

Ficadusta pulchella (Swainson, 1823). Habe 

& Kosuge, 1966a: 40, pl. 14, figs. 15, 16; 

1966c: ibid, (genus-group name = 

nomen nudum). 
Ficadusta Habe et Kosuge (gen. nov.). Habe 

& Kosuge, 1966b: 314, 329 (with refer- 



ence to 1966a; type species: Cypraea 
pulchella Swainson). 

Ficadusta pulchella (Swainson). Habe & Ko- 
suge, 1966b: 326 (plate expl.), pl. 29, 
figs. 11, 12. 

Ficadusta pulchella (Swainson, 1923 [sic]). 
Habe & Kosuge, 1972: 40, pl. 14, figs. 
15, 16; 1974, 1976: ibid, (type reset in 
1972 to correct English common name). 

Ficadusta Habe et Kosuge, 1966. Habe, 
1977: 40 (cited as of 1966b, but with 
type by "monotypy," whereas in 1966b 
it was designated; 1966a listed as 
"name only"). 

Cymatiidae 

(38) Reticutriton Habe & Kosuge, 1 966b 

Reticutriton pfeiffehanum (Reeve, 1844). 

Habe & Kosuge, 1966a: 43, pl. 15, fig. 

14; 1966c: ibid, (genus-group name = 

nomen nudum). 
Reticutriton Habe et Kosuge (gen. nov.). 

Habe & Kosuge, 1966b: 315, 330 (with 

reference to 1 966a; type species: Triton 

pfeifferianus Reeve). 
Reticutriton pfeifferianus (Reeve, 1844). Habe 

& Kosuge, 1972: 43, pl. 15, fig. 14; 1974, 

1976: ibid. 
Reticutriton Habe et Kosuge. Habe, 1977: 

102 (cited as of 1966b; 1966a not men- 
tioned). 

Muricidae 

(39) Spinidrupa Habe & Kosuge, 1966b 

Spinidrupa eurantha [sic] (A. Adams). Habe & 
Kosuge, 1966a: 54, pl. 20, fig. 4; 1966c, 
1972, 1974, 1976: ibid, (genus-group 
name = nomen nudum). 

Spinidrupa Habe et Kosuge (gen. nov.). Habe 
& Kosuge, 1966b: 315, 330 (with refer- 
ence to 1966a; type species: Murex 
eurantha [sic] A. Adams [p. 315; as eu- 
racantha on p. 330; = Murex euracanthus 
A. Adams, 1851]. 

Spinidrupa Habe et Kosuge. Habe, 1977: 1 15 
(cited as of 1966b; 1966a listed as 
"name only"). 

Pyrenidae (Columbellidae) 

(40) Pyrene testudinaria nigropardalis Habe & 
Kosuge, 1966a 

Pyrene testudinalia [sic] nigropardalis Habe et 



SHELLS OF THE WORLD IN COLOR 



43 



Kosuge. Habe & Kosuge, 1966a: 57, pi. 
21, fig. 3; 1966c: ibid. 

Pyrene testudinaria nigropardalis Habe et Ko- 
suge (sp. nov.). Habe & Kosuge, 1966b: 
316, 331, pi. 29, fig. 7 (with reference to 
1966a). 

Pyrene testudinaria nigropardalis Habe et Ko- 
suge. Habe & Kosuge, 1 972, 1 974, 1 976: 
57, pi. 21, fig. 3. 

Pyrene testudinalia [sic] nigropardalis Habe et 
Kosuge. Habe, 1977: 83 (cited as of 
1966a). 

(41) Pyrene lacteoides Habe & Kosuge, 
1966a 

Pyrene lacteoides Habe et Kosuge. Habe & 
Kosuge, 1966a: 57, pi. 21, fig. 8; 1966c, 
1972, 1974, 1976: ibid. 

Pyrene lacteoides Habe et Kosuge (sp. nov.). 
Habe & Kosuge, 1966b: 316, 330, pi. 29, 
fig. 8 (with reference to 1966a). 

Pyrene lacteoides Habe et Kosuge. Habe, 
1977: 68 (cited as of 1966a). 

Nassahidae 

(42) Plicarcularia gibbosuloidea Habe & Ko- 
suge, 1966a 

Pliarcularia [sic] gibbosuloidea Habe et Ko- 
suge. Habe & Kosuge, 1966a: 60, pi. 22, 
figs. 5, 6; 1966c, 1972, 1974, 1976: ibid. 

Pliarcularia [sic] gibbosuloidea Habe et Ko- 
suge (sp. nov.). Habe & Kosuge, 1966b: 
317. 

Plicarcularia gibbosuloidea Habe et Kosuge 
(sp. nov.). Habe & Kosuge, 1966b: 326 
[pi. explanation], 331, pi. 29, figs. 2, 3 
(with reference to 1966a). 

Plicarcularia gibbosuloidea Habe & Kosuge. 
Habe, 1977: 44 (cited as of 1966a; orig- 
inal misspelling of genus shown and cor- 
rected). 

Galeoidae (Galeolidae in 1966a: 64 and 
1966c: 64; correct on p. 65 and in later 
printings) 

(43) Hemifusus carinifer Habe & Kosuge, 
1966a 

Hemifusus carinifera [sic] Habe et Kosuge. 

Habe & Kosuge, 1966a: 64, pi. 23, fig. 2; 

1966c, 1972, 1974, 1976: ibid. 
Hemifusus cariniferus Habe et Kosuge (sp. 

nov.). Habe & Kosuge, 1966b: 317, 332, 

pi. 29, fig. 17 (with reference to 1966a). 
Hemifusus cariniferus Habe et Kosuge. Habe, 

1977: 29 (cited as of 1966a). 



Note: Originally introduced as an adjective in 
the female form, the ending has to be 
adjusted to the masculine -fer {-fer, -fera, 
-ferum, meaning "bearing"; as opposed 
to -ferus-a-um, meaning "wild"). 

Fasciolariidae 

(44) Latirus stenomphalus Habe & Kosuge, 
1966a 

Latirus stenomphalus Habe et Kosuge. Habe 
& Kosuge, 1966a: 68, 122, pi. 45, fig. 16 
(with reference to Kira, [1954]: pi. 30, fig. 
16, which is the species Kira figured as 
Latirus recurvirostrum Schubert & Wag- 
ner); 1966a, 1972, 1974, 1976: ibid. 

Latirus stenomphalus Habe et Kosuge (sp. 
nov.). Habe & Kosuge, 1966b: 318, 334, 
(with reference to Latirus recurvirostrum 
Kira, 1954: pi. 30, fig. 16 [on p. 318] and 
to 1966a [p. 334]; misspelled sttnom- 
phalus on p. 318). This reference to Kira 
is to the species he figured as Latirus 
recurvirostrum Schubert & Wagner. 

Latirus stenomphalus Habe et Kosuge. Habe, 
1977: 116 (cited as of 1966a). 

Mitridae 

(45) Vexillum rubrocostatum Habe & Kosuge, 
1966a 

Vexillum rubrocostatum Habe et Kosuge. 

Habe & Kosuge, 1966a: 73, pi. 28, fig. 9; 

1966c, 1972, 1974, 1976: ibid. 
Vexillum rubrocostatum Habe et Kosuge (sp. 

nov.). Habe & Kosuge, 1966b: 319, 333, 

pi. 29, fig. 4 (with reference to 1966a). 
Vexillum rubrocostatum Habe et Kosuge. 

Habe, 1977: 102 (cited as of 1966a). 

(46) Nebularia yaekoae Habe & Kosuge, 
1966a 

Nebularia yaekoae Habe et Kosuge. Habe & 
Kosuge, 1966a: 76, pi. 28, fig. 34; 1966c, 
1972, 1974, 1976: ibid. 

Nebularia yaekoae Habe et Kosuge (sp. nov.). 
Habe & Kosuge. 1966b: 319, 333, pi. 29, 
fig. 10 (with reference to 1966a). 

Nebularia yaekoae Habe et Kosuge. Habe, 
1977: 131 (cited as of 1966a). 

Harpidae 

(47) Harps kawamurai Habe, in Habe & Ko- 
suge, 1972 

Harpa striata (Lamarck, 1816). Habe & Ko- 



44 



PETIT & BIELER 



suge, 1966a: 79, pl. 30, fig. 2; supple- 
mental pi. 1, fig. 2; 1966c: ibid. 

Harpa kawamurai Habe. Habe, in Habe & Ко- 
suge, 1972: 79, pl. 30, fig. 2; supplemen- 
tal pl. 1, fig. 2; 1974, 1976 (no indication 
that name is new). 

Harpa kawamurai Habe & Kosuge, 1973 [sic]. 
Habe, 1975b: 10 (listed as "invalid" and 
as "= Harpa major Röding, 1798"). 

Harpa kawamurai Habe & Kosuge, 1973 [sic]. 
Matsukuma & Okutani, 1986: 6. 

Taxonomic note: The 3rd printing of Habe & 
Kosuge, where this species first ap- 
pears, is rare, and we have located only 
one copy. Not listed by Habe (1977). The 
Japanese text of Habe (1975b: 10) 
states that, according to personal com- 
munication with Dr. Rehder, this nominal 
species is a form of Harpa major Röding, 
1798. 

(48) Harpa kajiyamai Habe, in Habe & Ko- 
suge, 1972 

Harpa cancellata (Röding, 1798). Habe & 
Kuroda, 1966a: 79, pl. 30, fig. 3; supple- 
mental pl. 1, fig. 3; 1966c: ibid. 

Harpa kajiyamai Habe. Habe & Kosuge, 1972: 
79, pl. 30, fig. 3; supplemental pl. 1, fig. 
3. 

Harpa kajiyamai Rehder, 1973: 244, pl. 188, 
figs. 3, 4 (described from specimens re- 
ceived from Habe, who was stated to 
have recognized the species as new and 
given it a provisional name, and re- 
quested that it be named for the collec- 
tor). 

Harpa kajiyamai Rehder. Habe & Kosuge, 
1974: 79, pl. 30, fig. 3; supplemental pl. 
1, fig. 3; 1976: ibid. 

Taxonomic note: Walls (1980: 191) in his list 
of Harpa species includes both H. kajiy- 
amai Habe, 1970 [sic], and /-/. kajiyamai 
Rehder, 1973, indicating that both are "= 
[Harpa] liarpa," a synonymy we do not 
endorse. This species name must be at- 
tributed to Habe (1972). Not listed by 
Habe (1977). 

Volutidae 

(49) Volutoconus grossi mcmichaeli Habe & 
Kosuge, 1966a 

Volutoconus grossi mcmichaeli Habe & Ko- 
suge. Habe & Kosuge, 1966a: 86, pl. 33, 
fig. 1; 1966c, 1972, 1974, 1976: ibid. 

Volutoconus grossi mcmichaeli Habe et Ko- 



suge (sp. nov.). Habe & Kosuge, 1966b: 
320, 335, pl. 29, fig. 1 9 (with reference to 
1966a). 
Volutoconus grossi mcmichaeli Habe et Ko- 
suge. Habe, 1 977: 74 (cited as of 1 966a). 

Turhdae 

(50) Brachytoma kurodai Habe & Kosuge, 
1966a 

Brachytoma kurodai Habe et Kosuge. Habe & 
Kosuge, 1966a: 96, pl. 38, fig. 13; 1966c, 
1972, 1974, 1976: ibid. 

Brachytoma kurodai Habe et Kosuge (sp. 
nov.). Habe & Kosuge, 1966b: 320, 335, 
pl. 29, fig. 14 (with reference to 1966a) 

Brachytoma kurodai Habe et Kosuge. Habe, 
1977: 66 (cited as of 1966a). 

(51) Brachytoma kawamurai Habe & Kosuge, 
1966a 

Brachytoma kawamurai Habe et Kosuge. 

Habe & Kosuge, 1966a: 96, pl. 38, fig. 

14; 1966c, 1972, 1974, 1976: ibid. 
Brachytoma kawamurai Habe et Kosuge (sp. 

nov.). Habe & Kosuge, 1966b: 321, 336, 

pl. 29, fig. 9 (with reference to 1966a) 
Brachytoma kawamurai Habe et Kosuge. 

Habe, 1977: 58 (cited as of 1966a). 

(52) Brachytoma vexillium Habe & Kosuge, 
1966a 

Brachytoma vexillium Habe et Kosuge. Habe 
& Kosuge, 1966a: 96, pl. 38, fig. 15; 
1966c, 1972, 1974, 1976: ibid. 

Brachytoma vexillum Habe et Kosuge (sp. 
nov.). Habe & Kosuge, 1966b: 321, 336, 
pl. 29, fig. 5 (with reference to 1966a). 

Brachytoma vexillum Habe et Kosuge. Habe, 
1977: 130 (cited as of 1966a; original 
spelling not mentioned). 

Taxonomic note: The original spelling of the 
specific name, although obviously a mis- 
spelling or typographical error, must be 
retained in accordance with ICZN Code 
Article 32. 

Epitoniidae 

(53) Eglisia brunnea Habe & Kosuge, 1 966a 

Eglisia brunnea Habe et Kosuge. Habe & Ko- 
suge, 1966a: 103, pl. 40, fig. 16; 1966c: 
ibid. 

Eglisia brunnea Habe et Kosuge (sp. nov.). 
Habe & Kosuge, 1966b: 322, 337, pl. 29, 
fig. 18 (with reference to 1966a). 



SHELLS OF THE WORLD IN COLOR 



45 



Eglisia lanceolata brunnea Habe et Kosuge. 

Habe & Kosuge, 1972: 103, pi. 40, fig. 

16; 1974, 1976: ibid. 
Eglisia brunnea Habe et Kosuge. Habe, 1977: 

28 (cited as of 1966a). 



Bivalvia 



Limidae 



(54) Mantellum perfragile Habe & Kosuge, 
1966a 

Mantellum perfragile Habe et Kosuge. Habe 

& Kosuge, 1966a: 144, 177, pi. 68, fig. 6; 

1966c, 1972, 1974, 1976: 144. 
Mantellunn perfragile Habe et Kosuge (sp. 

nov.). Habe & Kosuge, 1966b: 323, 338. 

(not figured; 1966a not referred to) 
Limaria perfragile Habe et Kosuge. Habe & 

Kosuge, 1972, 1974, 1976: 177, pl. 68, 

fig. 6 (as Mantellum on p. 144). 
Mantellum perfragile Habe et Kosuge. Habe, 

1 977: 95 (cited as of 1 966a and placed in 

Limaria {Platilimaria)). 

Ostreidae 

(55) Anomiostrea Habe & Kosuge, 1966b 

Anomiostrea pyxidata (Adams et Reeve, 
1850). Habe & Kosuge, 1966a: 144, pl. 
55, fig. 9; 1966c, 1972, 1974, 1976: ibid, 
(genus-group name = nomen nudum). 

Anomiostrea Habe et Kosuge (gen. nov.). 
1966b: 323, 338, with type designated 
as Ostrea pyxidata Adams et Reeve (with 
reference to 1966a). 

Anomiostrea Habe et Kosuge. Habe, 1977: 
23 (cited as of 1966b; 1966a not men- 
tioned). 

Taxonomic note: Listed under '^nomina du- 
bia" by Stenzel (1971: N1167, figs. 
J140a-c), who also showed the name of 
the type species to be preoccupied. 
Type species renamed Anomiostrea cor- 
alliophila Habe, 1975a (new name for O. 
pyxidata Adams & Reeve, 1848 [sic; = 
1850] non Born, 1780 [sic; = 1778]. 

Cardiidae 

(56) Laevicardium rubropictum Habe & Ko- 
suge, 1966a 

Laevicardium rubropictum Habe et Kosuge. 
1966a: 153, pl. 59, fig. 2; 1966c, 1972, 
1974, 1976: ibid. 



Laevicardium rubropictum Habe et Kosuge 
(sp. nov.). 1966b: 324, 339, pl. 29, fig. 20 
(with reference to 1 966a) 

Laevicardium rubropictum Habe et Kosuge. 
Habe, 1977: 102 (cited as of 1966a). 

(57) Vasticardium nigropunctatum Habe & 
Kosuge, 1966a 

Vasticardium nigropunctatum Habe et Ko- 
suge. Habe & Kosuge, 1966a: 154, pl. 
59, fig. 9; 1966c, 1972, 1974, 1976: ibid. 

Vasticardium nigropunctatum Habe et Ko- 
suge (sp. nov.). Habe & Kosuge, 1966b: 
324, 340, pl. 29, fig. 16 (with reference to 
1966a). 

Vasticardium nigropunctatum Habe et Ko- 
suge. Habe, 1 977: 84 (cited as of 1 966a). 

Mactridae 

(58) Macrotoma yamamurae Habe & Kosuge, 
1966a 

Mictrotoma [sic] yamamurae Habe et Ko- 
suge. Habe & Kosuge, 1966a: 166, pl. 
65, fig. 8; 1966c: ibid. 

Mactrotoma yamamurae Habe et Kosuge (sp. 
nov.). Habe & Kosuge, 1966b: 325, 340, 
pl. 29, fig. 15 (with reference to 1966a; 
original misspelling of genus noted). 

Mactrotoma yamamurae Habe et Kosuge. 
Habe & Kosuge, 1972: 166, pl. 65, fig. 8. 

Heterocardia gibbosula Philippi [sic; = De- 
shayes]. Habe & Kosuge, 1974: 166, pl. 
65, fig. 8; 1976: ibid. 

Mactrotoma yamamurae Habe et Kosuge. 
Habe, 1977: 133 (cited as of 1966a; 
stated to be a synonym of Heterocardia 
gibbosula Deshayes, 1855). 



ACKNOWLEDGEMENTS 

The following made copies of publications 
available or otherwise responded to our re- 
quests for data: Dr. E. V. Coan, Dr. R. N. 
Kilburn, Dr. H. G. Lee, Dr. J. H. McLean, Mr. 
Thomas С Rice, Dr. Gary Rosenberg, Mr. 
Walter Sage, and Dr. Emily H. Vokes. Dr. 
Sadao Kosuge corresponded with us con- 
cerning later printings of both volumes and 
the availability of the taxa. Dr. H. D. Cameron, 
University of Michigan, provided etymologi- 
cal advice. We are especially indebted to Dr. 
Takahiro Asami, Tachikawa College of To- 
kyo, whose translations from the Japanese 
helped us in deciding on the validity of taxon 



46 



PETIT & BELER 



descriptions, and to Dr. M. G. Harasewycli 
who, while in Japan, searched for and ob- 
tained for us a copy of the elusive 3rd printing 
of Habe & Kosuge. We also wish to thank two 
anonymous reviewers for their comments. 



LITERATURE CITED 

BIELER, R. & R. E. PETIT, 1990, On the various 
editions of Tetsuaki Kira's "Coloured illustra- 
tions of the shells of Japan" and "Shells of the 
western Pacific in color Vol. I," with an anno- 
tated list of new names introduced. Malacologia 
32: 131-145. 

HABE, T., 1975a, New name for Anomiostrea pyx- 
idata (Adams & Reeve) (Ostreidae). Venus 33: 
184 (April). 

HABE, T., ed., 1975b, Publication for commemo- 
rate 77th anniversary of the birth of l\/lr. Ryosuke 
Kawamura. Illustration of shells described by and 
dedicated to Mr. R. Kawamura. 20 pp., incl. 5 
pis. Tokyo (December). 

HABE, T., 1977, Catalogue of molluscan taxa de- 
scribed by Tadashige Habe during 1939-1975, 
with illustrations of hitherto unfigured species 
{for commemoration of his sixtieth birthday). 185 
pp., incl. 7 pis.; Tokyo, [compiled by T. Inaba and 
K. Oyama, but authorship credited to Habe on 
page 2] 

HABE, T. & К. ITO, 1965a, Shells of the world in 
colour. Vol. I. The northern Pacific, viii, [2 pp. 
map], 176 pp., 56 pis.; Hoikusha, Osaka [addi- 
tional printings listed in this paper]. 

HABE, T. & К. ITO, 1965b, New genera and spe- 
cies of shells chiefly collected from the North 
Pacific. Venus 24: 16-45, pis. 2-4 (July 31). 

HABE, T. & К. ITO, 1968, Buccinid species from 



Rausu, Hokkaido. Venus 27: 1-8, pl. 1 (August 
31). 

HABE, T. & S. KOSUGE, 1966a, Shells of the world 
in colour. Vol. I. The tropical Pacific, vii, [2 pp. 
map], 193 pp., pis. 1-68, supplemental pis. 1-2; 
Hoikusha, Osaka (January 15; additional print- 
ings listed in this paper). 

HABE, T. & S. KOSUGE, 1966b, New genera and 
species of the tropical and subtropical Pacific 
molluscs. Venus 24: 312-341, pl. 29 (May 17). 

HOUBRICK, R.S., 1985. Genus Clypeomorus 
Jousseaume (Cerithiidae: Prosobranchia). 
Smithsonian Contributions to Zoology 403: 
1-131. 

KIRA, T., 1954. [Coloured illustrations of the shells 
of Japan], [viii] + 172 + 24 pp., 67 pis; Hoikusha, 
Osaka (additional printings listed in Bieler & 
Petit, 1990). 

KURODA, T., T. HABE & К. OYAMA, 1 971 . The sea 
shells of Sagami Bay. Maruzen, Tokyo, xix + 741 
pp. [in Japanese], pis. 1-121, 489 pp. [in En- 
glish], 51 pp. index, map. 

MATSUKUMA, A. & T. OKUTANi, 1986. Studies on 
the Kawamura collection (Mollusca) in the Na- 
tional Science Museum, Tokyo-ll. Catalogue of 
type specimens, with description of Pinna cello- 
phana n. sp. (Bivalvia). Venus 45: 1-10 

REHDER, H. A., 1973, The family Harpidae of the 
world. Indo-Pacific Mollusca 3: 207-274. 

ROSEWATER, J., 1970, The family Littorinidae in 
the Indo-Pacific. Part I. The subfamily Littohni- 
nae. Indo-Pacific Mollusca 2: 417-528. 

STENZEL, H. В., 1971, Oysters. Treatise on Inver- 
tebrate Paleontology, Part N, Volume 3, Mol- 
lusca 6, Bivalvia. Pp. N953-N1224. 

WALLS, J. G., 1980, Conchs, tibias and harps. 
T.H.F. Publications Inc. Ltd., Neptune, New Jer- 
sey. 191 pp. 

Revised Ms. accepted 28 November 1995 



MAU\COLOGIA, 1996, 38(1-2): 47-58 

ULTRASTRUCTURAL STUDY OF EUSPERMIOGENESIS IN CLYPEOMORUS 

BIFASCIATA AND CLYPEOMORUS TUBERCULATUS (PROSOBRANCHIA: 

CERITHIIDAE) WITH EMPHASIS ON ACROSOME FORMATION 

Fadwa A. Attiga^ & Hameed A. Al-Hajj 
Department of Biological Sciences, University of Jordan, Amman, Jordan 

ABSTRACT 

The ultrastructure of euspermiogenesis and euspermatozoa of Clypeomorus bifasciata and 
C. tuberculatus are almost identical. Early spermatids have oval to spherical nuclei, sparse 
endoplasmic reticulum, few mitochondria, and a well-developed Golgi complex with many 
vesicles in its vicinity. Acrosome differentiation occurs anywhere within the cytoplasm, and 
begins with a proacrosomal vesicle, which becomes cup-shaped and plugged at its edges with 
a dense interstitial granule. Microtubules are embedded in the matrix between the outer and 
inner acrosomal membranes. The acrosomal vesicle becomes aligned parallel to the antero- 
posterior nuclear axis, and changes into an inverted flask shape, with two external supporting 
structures at its basal margins. The interstitial granule becomes hat-shaped, separating the 
acrosome from the nucleus. The mature acrosome consists of a flat cone with microtubules in 
its core, an acrosomal rod-like material, and a basal plate. Nuclear shape changes from spher- 
ical to hammer-head to club-shape, with a posterior invagination enclosing the initial axonemal 
portion. The fine chromatin material of early spermatids changes to fibrillar, lamellar, and finally 
very compact material. The euspermatozoan midpiece originates from fusion of spermatid 
mitochondria into four large spheres, which are later organized into four non-helical mitochon- 
drial elements, two of which are large and the other two are extremely small. A dense ring 
structure marks the junction between the midpiece and the glycogen piece. The latter consists 
of nine tracts of glycogen granules surrounding nine axonemal doublets. The results of this 
study suggest that acrosomal ultrastructure could be used to establish phylogenetic relation- 
ships in Cerithiacea at the generic level. 



INTRODUCTION 

Morphological diversity of spermatozoa in 
prosobranchs, as in other animal groups, has 
been considered as a tool that can be used to 
ascertain evolutionary paths, through building 
up phylogenetic and taxonomic affinities 
among species (Franzen, 1955, 1956, 1970; 
Nishiwaki, 1964; Healy, 1983a, 1988a; Koike, 
1985). Based on ultrastructural studies of 
spermiogenesis and sperm morphology, me- 
sogastropods as a part of caenogastropods 
(mesogastropods and neogastropods) are 
classified into two groups. Members of the 
first group have short nuclei with shallow 
basal invaginations, associated with conical 
or flattened acrosomes. The midpiece may 
show modification of cristae into parallel cri- 
stal plates, and it is separated from the gly- 
cogen piece by a dense ring structure. The 
glycogen piece consists of axonemal micro- 



tubules and nine tracts of glycogen granules, 
whereas the short end piece is composed of 
an axoneme surrounded only by a plasma 
membrane. This group of caenogastropods 
includes superfamilies Cerithiacea (Healy 
1982a, b, 1983a; Afzelius & Dallai, 1983 
Koike, 1985), Viviparacea (Griffond, 1980 
Koike, 1985), and Cyclophoracea (Selmi & 
Giusti, 1980; Healy, 1984; Kohnert & Storch, 
1 984a, b, Koike, 1 985). All other superfamilies 
in Caenogastropoda are classified into the 
second group, which shares with the first 
group similar glycogen pieces, dense ring 
structures and end pieces. On the other hand, 
members of this group have apical acrosomal 
vesicles and accessory acrosomal mem- 
branes, whereas their short or long tubular 
nuclei may be completely invaginated by the 
axoneme (Healy, 1988a). The midpiece ele- 
ments are helically coiled, with usually un- 
modified cristae (Healy, 1983a, 1986b; Max- 



^This work was conducted as part of Fadwa Attiga's Master thesis. The George Washington University, Columbian College 
and Graduate School of Arts and Sciences, Department of Biological Sciences, Ph.D. Program. Author to whom all 
correspondence should be mailed. Address: 2301 E St. NW, Apt # A406, Washington, DC 20037, U.S.A. 



47 



48 



ATTIGA & AL-HAJJ 



well, 1983; Kohnert & Storch, 1984a; Koike, 
1985; Jaramlllo et al., 1986). Furthermore, 
comparative sperm ultrastructure has been 
useful in establishing the affinities of many 
cerithiacean super-families of the Caenogas- 
tropoda (Healy, 1982a, b, 1983a, 1986a, b, 
1988a, b, 1990a, b, 1993; Houbrick, 1988). 

The present work deals with the ultrastruc- 
ture of euspermiogenesis and mature eu- 
sperm (typical sperm) in two species of the 
superfamily Cerithiacea (family Cerithiidae) 
that inhabit the rocky shore of the Gulf of 
Aqaba (Houbrick, 1985; Hulings, 1986). 
These are: Clypeomorus bifasciata (Sowerby, 
1855) [= С moniliferum (Kiener, 1841), 
auett.], and С tuberculatus (Linnaeus, 1758) 
[= С petrosa gennesi (Fisher & Vignal, 1901)]. 
Comparative study of spermiogenesis and 
sperm morphology of the two cerithiid spe- 
cies as well as other reported cerithiids aims 
to emphasize species-specific characters 
between cerithiids from different geographi- 
cal regions, and to establish the phylogenetic 
status of cerithiaceans among prosobranchs. 



MATERIALS AND METHODS 

Specimens were collected monthly for a 
year in the intertidal zone opposite to the Ma- 
rine Science Station of the Gulf of Aqaba. The 
shell was gently broken, and the testis, re- 
moved by dissection, was immediately im- 
mersed in 2.5% glutaraldehyde in filtered sea 
water for 2 hours at room temperature. The 
tissue was rinsed thoroughly in filtered sea 
water, post fixed in 1 % OSO4 solution in fil- 
tered sea water, dehydrated in acetone and 
embedded in Spurr's (1969) medium. Blocks 
were cut with Sorval MT 2B ultramicrotome 
using glass knives, and ultrathin sections 
(50-60 nm) were stained with uranyl acetate 
and lead citrate. Electron microscopic exam- 
inations were done with a Zeiss EM 10B 
transmission electron microscope operated 
at 60 KV. 



RESULTS 

The various stages of euspermiogenesis in 
Clypeomorus bifasciata and С tuberculatus 
are almost identical. Therefore, the following 
description applies for both species unless 
otherwise mentioned. 

Early spermatids are spherical to ovoidal. 



with eccentric nuclei. The chromatin material 
is granular, with some local aggregations of 
no specific pattern. The granular cytoplasm 
contains few cisternae of endoplasmic retic- 
ulum, few mitochondria, and a well-devel- 
oped Golgi complex with many vesicles at 
the extremities of its cisternae, indicating ac- 
tivity (Fig. 1). Nutritive cells can be seen in the 
intercellular space with many elongated 
pseudopodia (Fig. 1). 

Acrosome development can be divided 
into two major phases; the pre-attachment 
acrosome and the post-attachment one, in 
reference to its attachment to the nucleus. 
Acrosomal genesis during the first phase be- 
gins with a single proacrosomal vesicle as- 
sociated with Golgi complex, in addition to 
many nearby dense vesicles that are likely to 
be utilized in the production of the acrosomal 
elements (Fig. 2). Later, this vesicle attains an 
inverted U-shape due to posterior indenta- 
tion, and a dense interstitial granule plugs the 
prospective subacrosomal space (Figs. 3, 4). 
The dispersion of dense material from this 
granule and its deposition on the inner and 
outer acrosomal membranes assist in the ac- 
centuation of the acrosome (Figs. 4, 9). Two 
dense internal supporting structures appear 
within the acrosomal body, and microtubules 
constitute the skeleton of the acrosomal 
cone between the inner and outer acrosomal 
membranes (Figs. 4, 8, 9). 

The second phase of acrosomal develop- 
ment is demarcated by the attachment of the 
basal interstitial granule to a depression on 
the anterior pole of the fibrous nucleus, op- 
posite to the site of axoneme development 
(Fig. 10). The acrosome rotates 90' to be- 
come aligned parallel to the antero-posterior 
axis of the developing spermatid (Figs. 10, 
11). Following its attachment, the acrosome 
looks like an inverted flask due to a constric- 
tion at its posterior part (Figs. 11, 12). Two 
crescent-shaped external supporting struc- 
tures can be seen at the basal margins of the 
acrosome near its attachment point to the 
nucleus. The post-attachment acrosome is 
further elongated, while the dense interstitial 
granule gives rise to a basal plate between 
the acrosome and the nucleus (Figs. 11, 12). 

The acrosome of the mature euspermato- 
zoon in С bifasciata and С tuberculatus con- 
sists of three structures; acrosomal cone, ac- 
rosomal rod-like material, and basal plate 
(Figs. 16 inset, 17). The tapering cone may 
occasionally show parallel plate-like sub- 
structures, and it is characterized by basal 



ULTRASTRUCTURAL STUDY OF EUSPERMIOGENESIS 



49 






О 



. m . *- 



;г* ' 



йГ «1 



GC 



M 



f 



"■■•'fi 



v'%. 



А% 



-|Га 



u /o^ . 



^ 




^^' 



.5i- 



à. 



^Щ<:' 



©-: 



•V-Ííaa^^^^í. 



FIG. 1. Clypeomorus bifasciata. Early spermatid showing nucleus (N), mitochondnon (M), Golgi complex 
(GC) and associated vesicles (V). Notice pseudopodia (PR) of the nutritive cell (NC). -11,500 

FIG. 2. С bifasciata. Early spermatid showing peripheral chromatin lining the nucleus (N), Golgi complex 
(GC), and interstitial granule (IG). >31,250 



50 



ATTIGA & AL-HAJJ 




© 





--'9^lîl- 



хэ" I 



■ *.*r 


N 


i^ 


•t/ 




' — .il >: 



^"^^^ "0» 






FIG. 3. C. bifasciata. Early spermatid showing nucleus (N), two nnitochondrial (M) spheres at nuclear base, 
Golgi connplex (GC), vesicles (V), and differentiating proacrosomal vesicle (PAV). Notice cytoplasmic bridge 
(asterisk). ^ 16,000 

FIG. 4. С bifasciata early spermatid. Section showing a differentiating acrosome with microtubules (MT) in 
its cone, internal supporting structures (arrows) subacrosomal space (SAS), and interstitial granule (IG). 
x36,000 

FIG. 5. C. tuberculatus early spermatid. Section showing nucleus (N), with sites of the attachment of 
mitochondrial (M) spheres (arrows). x1 5,200 

FIG. 6. C. tuberculatus early spermatid. Section showing nuclear (N) base with mitochondrial (M) spheres, 
implantation fossa (IF), centriolar derivative (CD), and axoneme (AX). x56,700 



FIG. 7. C. bifasciata early spermatid. Section showing four mitochondrial (M) spheres surrounding the 
axoneme (AX) as the first stage of midpiece development. a1 5,000 



ULTRASTRUCTURAL STUDY OF EUSPERMIOGENESIS 



51 



bulges in the cone wall that cause a constric- 
tion in the subacrosomal space. The latter, 
which extends the whole length of the ac- 
rosomal cone, contains an acrosonnal rod- 
like nnaterial (Figs. 16 inset, 17). A dense 
basal plate linking the acrosome and the nu- 
cleus can be seen as a straight dense layer 
between the two structures. Cross sections 
in the mature acrosome indicate its flatness, 
and microtubules assume a zipper-like struc- 
ture in the matrix between the inner and outer 
acrosomal membranes (Fig. 18). 

Chromatin condensation starts with the 
formation of a uniformly thick layer at the pe- 
riphery of the nucleus (Figs. 2, 3). As spermi- 
ogenesis proceeds, the granular chromatin 
accumulates at the posterior nuclear pole, 
and the nucleus undergoes antero-posterior 
compression, leading to gradual increase in 
the nuclear width at the expense of its length 
(Fig. 8). In addition, the axoneme extends 
backwards from a centriolar derivative in the 
implantation fossa, so that vertical sections 
through a developing euspermatozoon at this 
stage present hammer-head and handle con- 
figurations (Fig. 10). 

A second phase of chromatin condensa- 
tion is evident in middle spermatids as the 
mid-anterior portion of the nucleus, opposite 
to the axoneme, begins a forward movement. 
During this phase, fibrils are arranged longi- 
tudinally parallel to the nuclear antero-poste- 
rior axis (Fig. 11). As fibrils increase in thick- 
ness, they start to fuse into fibers and 
subsequently into lamellae representing 
thereby the lamellar phase of chromatin con- 
densation (Figs. 13-15). Chromatin conden- 
sation culminates in a homogenous, com- 
pact club-like nucleus with no distinct 
ultrastructure (Fig. 19). 

Concomitant with nuclear condensation, a 
growing axoneme pushes the nucleus for- 
ward to increase its length, while the nuclear 
width is reduced under a force of lateral com- 
pression. This leads to progressive lengthen- 
ing of the antero-posterior nuclear axis, thus 
reversing its trend in the previous stages 
(Figs. 10, 11). The mature nucleus in С bifas- 
ciata and C. tuberculatus has a short poste- 
rior invagination accommodating the proxi- 
mal portion of the axoneme (Figs. 16, 19). 

The euspermatozoan tail in С bifasclata 
and С tuberculatus is composed of a middle 
piece, a glycogen piece and an end piece. 
The posterior nuclear envelope becomes in- 
dented at its center defining thereby the im- 
plantation fossa, which represents the point 



of axoneme development (Fig. 6). The gene- 
sis of the axoneme appears to be associated 
with a single dense structure (centriolar de- 
rivative), which does not seem to possess the 
common pattern of centriolar arrangement 
(Fig. 6). As nuclear condensation com- 
mences, mitochondrial fusion gives rise to 
four large spherical mitochondria at the pos- 
terior nuclear pole (Fig. 7). The association of 
these spheres with the nucleus is achieved 
by their attachment to four posterior nuclear 
depressions, and it represents the first step 
in midpiece development, which is concom- 
itant with the granular phase of nuclear con- 
densation. The mitochondrial cristae are 
modified into parallel cristal plates that have 
undergone considerable reorganization as 
mitochondria form a sheath around the typi- 
cal 9 + 2 axoneme (Figs. 10, 11). Transverse 
sections through the midpiece reveal four 
non-helically arranged mitochondrial ele- 
ments, two of which are semicircular large 
elements that are arranged perpendicular to 
the central pair of axonemal microtubules, 
and each reveals multiple cristal plates. The 
other two mitochondrial elements are ex- 
tremely small and are aligned with this central 
pair, showing at most one cristal plate (Fig. 
21). In addition, a ring of microtubules is ob- 
served surrounding the midpiece at late 
stages of its development (Fig. 21). Glycogen 
granules in the glycogen piece are organized 
in nine tracts; one per microtubular doublet 
(Fig. 24), and the transition zone between the 
midpiece and the glycogen piece is marked 
by a dense ring structure that is attached to 
the euspermatozoan plasma membrane (Fig. 
22). The latter continues to encircle the ax- 
onemal microtubules, forming the end piece 
of the tail (Fig. 25), without a distinct transi- 
tion structure between the glycogen piece 
and the end piece (Fig. 23). Cytoplasmic 
bridges connect adjacent developing sper- 
matids throughout various stages of eusper- 
miogenesis (Figs. 3, 11). 



DISCUSSION 

Euspermiogenesis as seen in Clypeomorus 
bifasciata and С tuberculatus includes many 
common features that were reported in all 
other cerithiaceans (Giusti, 1971; Healy, 
1982a, 1984; Koike, 1985; Afzelius et al., 
1989, Hodgson & Heller, 1990; Minniti 1993) 
as well as other mesogastropods and neo- 
gastropods (Giusti, 1969; Buckland-Nicks & 



52 



ATTIGA & AL-HAJJ 




ULTRASTRUCTURAL STUDY OF EUSPERMIOGENESIS 



53 



Chia, 1976; West, 1978; Griffond, 1980; 
Kohnert, 1980; Healy 1983b; Koike 1985). In 
general, acrosome formation is associated 
with Golgi complex and involves the produc- 
tion of a proacrosomal vesicle, which occurs 
anywhere in the cytoplasm, because there is 
no definite route of acrosome migration from 
the posterior to the anterior pole of the de- 
veloping spermatid. This situation was re- 
ported in other cerithiaceans (Healy, 1982a, 
1986a; Minniti, 1993), in contrast to many 
other mesogastropods and neogastropods, 
in which such a route is marked and linked to 
various stages of nuclear shaping (Buckland- 
Nicks & Chia, 1976; West, 1978; Jong-Brink 
et al., 1977; Buckland-Nicks et al., 1983; 
Jaramillo et al., 1986, Gallardo & Garrido, 
1989). Cerithiids, including those used in this 
study, are characterized by a high degree of 
development of the pre-attachment ac- 
rosome. Prior to its attachment to the nu- 
cleus, the acrosome bears an acrosomal 
cone, an acrosomal rod-like material and an 
interstitial granule, which gives rise to the 
basal plate; these constitute the elements of 
the mature acrosome. 

Because spermiogenesis involves several 
complex processes of cellular shaping and 
remodeling, microtubules and microfilaments 
are expected to play a crucial rule in these 
processes. In contrast to other reported cer- 
ithiids, microtubules were seen within the de- 
veloping acrosomal cones of Clypeomorus 
bifasciata and С tuberculatus. This was 



strongly suggested by the longitudinal sec- 
tions cutting through the developing ac- 
rosomes (Figs. 4, 9) as well as the hollow 
round structures seen in transverse sections 
(Fig. 18). Such an arrangement of microtu- 
bules within the cone is thought to provide it 
with a degree of rigidity, and aid in its elon- 
gation after it attaches to the nucleus, as was 
suggested in some non-cerithiacean meso- 
gastropods and neogastropods (Walker & 
MacGregor, 1968; Buckland-Nicks & Chia, 
1 976; Giusti & Mazzini, 1 973). Other cerithiids 
as well as other mesogastropods and neo- 
gastropods have a ring of microtubules 
surrounding developing acrosomes (Buck- 
land-Nicks & Chia, 1976; Huaquin & Bustos- 
Obergon, 1981; Buckland-Nicks et al., 1983; 
Healy, 1983b). In addition, microtubules were 
seen around midpieces of Clypeomorus bi- 
fasciata and С tuberculatus at late stages of 
development. Their appearance at such late 
stages in these two cerithiids as well as other 
mesogastropods (Jong-Brink et al., 1977; 
Kohnert, 1980; Griffond, 1980; Healy, 1982a, 
1983a, b, 1988b; Buckland-Nicks et al., 
1983; Afzelius et al., 1989; Al-Hajj & Attiga, 
1995) strengthens the idea that they are im- 
portant in sloughing the excess cytoplasm 
around midpieces as well as other parts of 
the euspermatozoon (Fawcett et al., 1971). 

Furthermore, the ornamentation of the ma- 
ture acrosomal cone with parallel plate-like 
substructures seen in Clypeomorus bifasci- 
ata and С tuberculatus was also reported in 



FIG. 8. C. tuberculatus. Early spermatid showing basal chromatin accumulation in nucleus (N), sites of 
mitochondrial (M) association with the nucleus (arrow heads), acrosomal cone (AC), interstitial granule (IG), 
and internal supporting structures (arrow) in acrosomal cone. x27,200 

FIG. 9. C. tuberculatus early spermatid. Section showing acrosomal cone with microtubules (arrow) and 
internal supporting structures (arrows), subacrosomal space (SAS), and interstitial granule (IG). x50,000 

FIG. 10. С bifasciata. Middle spermatid showing interstitial granule (IG), hammer-like nucleus (N), and 
midpiece (MP). x22,400 

FIG. 11. С tuberculatus. Middle spermatid with a cytoplasmic bridge (astehsk) showing acrosomal cone 
(AC), external supporting structure (ES), basal plate (BP), nucleus (N), and midpiece (MP). > 15,000 

FIG. 12. C. tuberculatus middle spermatid. Section showing acrosomal cone (AC), internal supporting 
structure (arrow heads), external supporting structure (ES), basal plate (BP), and fibrillar nucleus (N). 
V44.000 

FIG. 13. C. bifasciata middle-late spermatid. Cross section in fibrillar nucleus. --23,750 

FIG. 14. С bifasciata late spermatid. Cross section in nucleus showing islands of lamellae. -^35,000 

FIG. 15. С tuberculatus late spermatid. Cross section in nucleus with semi-fully condensed chromatin. 
■31,500 



54 







DRS 



• >' A «fît / >-^Учр Ä * -^ <_ 

v^v- -*''^- i®":; Í!l«r '• Î* 



' GP* * . 



/ 




^'^^ 






^ 



25)' 



,ïé#a«*!*»-. 



.^' 



FIGS. 16-25. 



ULTRASTRUCTURAL STUDY OF EUSPERMIOGENESIS 



55 



other mesogastropods and neogastropods 
(Giusti & Mazzini, 1973; Healy, 1983a, 1986b; 
Jaramillo et al., 1986; Afzelius et al., 1989; 
Al-Hajj & Attiga, 1995). Acrosomal mem- 
branes in Chorus giganteas (Jaramillo et al., 
1986), Truncatella subcylindrica (Giusti & 
Mazzini, 1973), and Melanopsis (Afzelius et 
al., 1989) were reported to have a scalloped 
appearance with regular periodicity. This or- 
namentation of the growing acrosome that is 
later hidden by electron-dense material 
seems to be of scarce occurrence. Giusti & 
Mazzini (1973) interpreted this periodicity as 
microtubular palisade, whereas Jaramillo et 
al. (1986) thought that it is due to the pres- 
ence of actin crests, which may play a role in 
acrosome reaction and egg penetration. 
However, the lack of knowledge about ac- 
rosome reaction and fertilization in gastro- 
pods makes it difficult to conclude the nature 
or function of these structures, pending fur- 
ther investigation. 

Chromatin condensation and nuclear shap- 
ing are two highly linked processes in sper- 



miogenesis of mesogastropods and neo- 
gastropods. Chromatin condensation passes 
through granular, fibrillar and lamellar phases, 
culminating in a homogeneous compact nu- 
cleus with no ultrastructure (Walker & Mac- 
Gregor, 1968; Buckland-Nicks & Chia, 1976; 
Feral, 1977; West, 1978; Huaquin & Bustos- 
Obergon, 1981; Healy, 1982a, b, 1983b, 
1988b; Buckland-Nicks et al., 1983; Jaramillo 
et al., 1986; Gallardo & Garhdo, 1989). The 
mature nucleus in the two cerithiids investi- 
gated in this study has a short basal invagi- 
nation accommodating the proximal portion 
of the axoneme, which is a common nuclear 
shape seen in many other mesogastropods 
and neogastropods (Giusti, 1 969, 1 971 ; Giusti 
& Mazzini, 1973; Reader, 1973, Griffond, 
1 980; Kohnert, 1 980; Koike & Nishiwaki, 1 980; 
Healy, 1982a, b, 1983a, 1986b). On the other 
hand, some mesogastropods and many neo- 
gastropods have intranuclear canals that in- 
vaginate the nucleus completely up to its apex 
(Walker &MacGregor, 1968; Buckland-Nicks, 
1973; Buckland-Nicks & Chia, 1976; West, 



FIG. 16. C. tuberculatus mature euspermatozoon, with acrosome (A) and nucleus (N). -48,000 Inset: 
Section showing acrosomal cone (AC) with plate-like substructure (arrow head), acrosomal rod-like material 
(AR), and basal plate (BP). -57,500 

FIG. 17. С bifasciata semi-mature acrosome. Section showing acrosomal cone (AC) with plate-like sub- 
structure (arrow head), acrosomal rod-like material (AR), and basal plate (BP). у63,000 

FIG. 18. С bifasciata semi-mature euspermatozoon. Section in acrosome showing acrosomal rod-like 
material (AR), microtubules (MT) in acrosomal cone, and surrounding pseudopodium (PP) of nutritive cell. 
x40,000 

FIG. 19. С tuberculatus semi-mature euspermatozoa showing nucleus (N), centriolar derivative (CD) in 
nuclear basal invagination and midpiece with mitochondrial sheath (M) surrounding the axoneme (AX). 
x60,000 

FIG. 20. C. tuberculatus semi-mature euspermatozoon. Cross section in nucleus (N) showing the proximal 
portion of the axoneme (arrows) within nuclear basal invagination. Notice pseudopodia (PP) of nutritive 
cells. -40,000 

FIG. 21 . С bifasciata late euspermatozoon. Cross section in midpiece showing two large and two extremely 
small mitochondrial (M) elements surrounded by microtubules (arrows). -48,000 

FIG. 22. C. tuberculatus mature euspermatozoon. Section showing dense ring structure (DRS) at the 
junction between midpiece with mitochondrial sheath (M) around axoneme (AX), and glycogen piece (GP). 
x38,000 

FIG. 23. C. bifasciata mature euspermatozoa. Sections showing the junction between glycogen piece (GP) 
and end piece (EP). -52,000 

FIG. 24. С bifasciata mature euspermatozoa. Cross section in glycogen piece showing nine tracts of 
glycogen granules associated with axonemal microtubular doublets. -37,500 

FIG. 25. C. bifasciata mature euspermatozoa. Cross section in end piece showing 9 + 2 pattern of micro- 
tubular arrangement surrounded by plasma membrane. -128,000 



56 



ATTIGA & AL-HAJJ 



1978; Huaquin & Bustos-Obergon, 1981; 
Buckland-Nicks et al, 1983; Healy, 1984; 
Jaramillo et al., 1986; Hodgson, 1993). Healy 
(1983a) interpreted this extreme structural di- 
versity in the extent of nuclear invagination to 
factors in the reproductive environment, es- 
pecially because some superfamilies include 
both types of nuclei among their species 
(Buckland-Nicks, 1973; Healy, 1988a). 

Centrioles were not seen in developing eu- 
spermatids in С bifasciata and C. tubercula- 
tus, and the proximal portion of the axoneme 
was attached to the posterior nuclear invagi- 
nation through a single dense structure that 
has no apparent microtubules. This centriolar 
derivative was reported in other caeno- 
gastropods (Buckland-Nicks, 1973; Healy, 
1982a, 1983a, 1986a, b, 1988b). 

The midpiece in euspermatozoa of С bi- 
fasciata and C. tuberculatus is consistent with 
those of other cerithiids and members of sub- 
group 1(i) of Healy's (1983a) classification of 
mesogastropods as a group of caenogastro- 
pods. Midpieces in these animals are char- 
acterized by four non-helically arranged ele- 
ments, two of which are extremely small, 
whereas the other two are large showing mul- 
tiple cristal plates. The non-helical arrange- 
ment of mitochondria around the axoneme in 
Cerithiacea is considered a primitive charac- 
ter compared to the helical mitochondrial 
sheath seen in other mesogastropods and 
neogastropods (Walker & MacGregor, 1968; 
Giusti, 1969, 1971; Anderson & Personne, 
1970; Buckland-Nicks, 1973; Giusti & Maz- 
zini, 1973; West, 1978; Koike & Nishiwaki, 
1980; Kohnert, 1980; Griffond, 1980). This 
feature supports Healy's (1982a) proposi- 
tion that cerithiaceans represent ancestral 
mesogastropods, acting as a linkage group 
between primitive spermatozoa of Archaeo- 
gastropoda on one hand and modified sper- 
matozoa of higher mesogastropods and neo- 
gastropods on the other. Such a position 
makes it an interesting group to study sper- 
matozoan evolution. 

In cerithiids, including those investigated in 
this study, the glycogen piece consists of 
nine tracts of glycogen granules associated 
with the nine doublets of the axonemal mi- 
crotubules. This arrangement, which is seen 
also in other mesogastropods and neogas- 
tropods (Giusti, 1969, 1971; Buckland-Nicks, 
1973; Reader, 1973; West, 1978; Koike & 
Nishiwaki, 1980; Kohnert, 1980; Huaquin & 
Bustos-Obergon, 1981; Healy, 1982a, 1986a, 
b, 1988a, c; Jaramillo et al., 1986; Gallardo & 



Garrido, 1989; Al-Hajj & Attiga, 1995), is 
thought to be linked to the euspermatozoan 
motility, because the nine glycogen tracts in 
the mature euspermatozoa obscures nine ra- 
dial links between the axonemal doublets 
and the plasma membrane in the immature 
spermatid (Healy, 1983a). 

Many investigators have suggested that 
acrosomal ultrastructure of gastropods could 
provide useful taxonomic data by being spe- 
cies-specific. In Cerithiacea, such a species- 
specific acrosome ultrastructure is not well 
presented. Species-specific features were 
seen when comparing acrosomes of Cerith- 
ium vulgatum (Giusti, 1971), С nodulosum 
(Koike, 1985), С rupestre (Minniti, 1993), С. 
caeruleum (Al-Hajj & Attiga, 1995), whereas 
Clypeomorus bifasciata, C. tuberculatus (this 
study), С moniliferus (bifasciata), and C. 
breviculus (Healy, 1 983a) have almost identi- 
cal acrosomes. In addition, differences in the 
ultrastructure of the acrosome were estab- 
lished at the generic level between Cerittiium, 
Rliinoclavis, Australaba (Healy, 1983a), and 
Clypeomorus (Healy, 1983a; this study), but 
were not seen between euspermatozoa of 
Conomurex and Lambis, which possess very 
similar acrosomes (Koike & Nishiwaki, 1980). 

In conclusion, comparative studies of 
sperm ultrastructure have proved to act as an 
acceptable guide for determining the affini- 
ties between major groups in gastropods as 
well as in many other animal groups. 



ACKNOWLEDGMENTS 

The authors are thankful to Dr. Saleem Al- 
Mograby from the Marine Science Station of 
the Gulf of Aqaba for help in providing access 
to collect the specimens from the station's 
rocky beach. 



LITERATURE CITED 

AFZELIUS, B. & R. DALLAI, 1983, The paired sper- 
matozoa of the marine snail Turritella communis 
Lamarck (Mollusca, Mesogastropoda). Journal 
of Ultrastructure Researcti, 85: 31 1-319. 

AFZELIUS, В., R. DALLAI & G. GALLAINI, 1989, 
Spermiogenesis and spermatozoa in Melanopsis 
(Mesogastropoda, Mollusca). Journal of Submi- 
croscopic Cytology and Pathology, 21: 187-200. 

AI-HAJJ, H. & F. ATTIGA, 1995, Ultrastructural ob- 
servations of euspermiogenesis and mature 
euspermatozoa in Cerithium caeruleum (Gas- 



ULTRASTRUCTURAL STUDY OF EUSPERMIOGENESIS 



57 



tropoda: Cerithiacea) from the Jordan Gulf of 
Aqaba. Invertebrate Reproduction and Develop- 
ment. 28: 113-124. 

ANDERSON, W. & P. PERSONNE, 1970, The lo- 
calization of glycogen in the spermatozoa of var- 
ious invertebrate and vertebrate species. The 
Journal of Cell Biology, 44: 29-51 . 

BUCKLAND-NICKS, J., 1973, The fine structure of 
the spermatozoon of Littohna (Gastropoda: 
Prosobranchia) with special reference to sperm 
motility. Zeitschrift für Zellforschung Mikrosko- 
pische Anatomie, 144: 11-29. 

BUCKLAND-NICKS, J., & F.-S. CHIA, 1976, Sper- 
matogenesis of a marine snail, Littohna sitkana. 
Cell and Tissue Research, 170: 455-475. 

BUCKLAND-NICKS, J., D. WILLIAMS, F.-S. CHIA, 
& A. FONTAINE 1983, Studies on the polymor- 
phic spermatozoa of a marine snail, 2. Genesis of 
the eupyrene sperm. Gamefe Research, 7: 19-37. 

FAWCETT, D., A. WINSTON, & D. PHILLIPS, 1 971 , 
Morphogenetic factors influencing the shape of 
the sperm head. Development Biology, 26: 220- 
251. 

FERAL, C, 1 977, Étude de la Spermatogenese typ- 
ique chez Ocenebra ehnacea. Mollusque, Gas- 
tèropode, Prosobranche. Bulletin de la Société 
Zoologique de France, 102: 25-30. 

FRANZEN, A., 1955, Comparative morphological 
investigation into the spermiogenesis among 
Mollusca. Zoologische Bijdragen Uppsala, 30: 
399-456. 

FRANZEN, A., 1956, On spermiogenesis, morphol- 
ogy of the spermatozoon and the biology of fer- 
tilization among invertebrates. Zoologische Bij- 
dragen Uppsala, 31: 356-482. 

FRANZEN, A., 1970, Phylogenetic aspects of the 
morphology of spermatozoa and spermiogene- 
sis. Pp. 3-46, in: B. Baccetti, editor. Compara- 
tive spermatology; proceedings of the interna- 
tional symposium held in Rome and Siena, 1-5 
July 1969, Academic Press, New York. 

GALLARDO, С. & О. GARRIDO, 1989, Spermio- 
genesis and sperm morphology in the marine 
gastropod Nucella crassilabrum with an account 
of morphometnc patterns of spermatozoan vari- 
ation in the family Muricidae. Invertebrate Repro- 
duction and Development, 15: 163-170. 

GIUSTI, F., 1969, The spermatozoon of a fresh- 
water prosobranch mollusc. Journal of Submi- 
croscopic Cytology and Pathology, 1: 263-273. 

GIUSTI, F., 1971, L'ultrastructura dello spermato- 
zoo nella filogenesi e nella sistemática dei mol- 
luschi gasteropodi. Atti della Società Italiana di 
Scienze Naturali e Museo Cívico di Stona Natu- 
rale Milano, 112: 381-402. 

GIUSTI, F. & M. MAZZINI, 1973, The spermato- 
zoon of Truncatella (s. str.) subcylindnca (L.) 
(Gastropoda, Prosobranchia). Monitore Zoo- 
lógico Italiano, N. S., 7: 181-201, 

GRIFFOND, В., 1980, Etude ultrastructurale de la 
Spermatogenese typique de Viviparus viviparus 
(L.), Mollusque, Gastropode. Archivum Biologie 
(Bruxelles), 91: 445-462. 



HEALY, J., 1982a, An ultrastructural examination 
of developing and mature euspermatozoa in 
Pyrazus ebeninus (Mollusca, Gastropoda, Pota- 
mididae). Zoomorp/io/ogy, 100: 157-175. 

HEALY, J., 1982b, Ultrastructure of parasper- 
matozoa, euspermatozoa and eusperm-like 
spermatozoa of Obtortio sp. (Prosobranchia: 
Cerithiacea). Helgolander Wissenschafthche 
Meeresuntersuchungen, 35: 489-500. 

HEALY, J., 1983a, Ultrastructure of euspermato- 
zoa of cerithiacean gastropods (Prosobranchia: 
Mesogastropoda). Journal of Morphology, 168: 
57-75. 

HEALY, J., 1983b, Ultrastructure of euspermiogen- 
esis in the mesogastropod Stenothyra sp. 
(Prosobranchia, Rissoacea, Stenothyridae). 
Zoología Scripta, 12: 203-214. 

HEALY, J., 1984, The ultrastructure of gastropod 
spermatozoa and spermiogenesis. Ph.D. Thesis, 
University of Queensland (Bhsbane). 

HEALY, J., 1986a, An ultrastructural study of eus- 
permatozoa, paraspermatozoa, and nurse cells 
of the cowrie Cypraea erronés (Gastropoda, 
Prosobranchia, Cypraedae). Journal of Mollus- 
can Studies, 52: 125-137. 

HEALY, J., 1986b, Euspermatozoa and parasper- 
matozoa of the relict cerithiacean gastropod 
Campanile symbolicum (Prosobranchia, Meso- 
gastropoda). Helgolander Wissenschafthche 
Meeresuntersuchungen, 40: 201-218. 

HEALY, J., 1988a, Sperm morphology and its sys- 
tematic importance in the Gastropoda. In; W. F. 
Ponder, ed, Prosobranch phylogeny. Proceed- 
ings of a symposium held at the 9th International 
Malacological Congress. Edinburgh, 1986. Mal- 
acological Review, 4, Supplement: 251-272. 

HEALY, J., 1988b, The ultrastructure of spermato- 
zoa and spermiogenesis in pyramidellid gastro- 
pods, and Its systematic importance. Helgolan- 
der Wissenschafthche Meeresuntersuchungen, 
42: 303-318. 

HEALY, J., 1988c, Sperm morphology in Serpulor- 
bis and Dendropoma and its relevance to the 
systematic position of the Vermetidae (Gas- 
tropoda). Journal of Molluscan Studies, 54: 295- 
308. 

HEALY, J., 1990a, Systematic importance of sper- 
matozeugmata in triphorid and cerithiopsid gas- 
tropods (Caenogastropoda: Triphoroidea. Jour- 
nal of Molluscan Studies, 56: 1 1 5-1 1 8. 

HEALY, J., 1 990b, Taxonomic affinities of the deep 
sea genus Provanna (Caenogastropoda): new 
evidence from sperm ultrastructure. Journal of 
Molluscan Studies, 56: 119-122. 

HEALY, J., 1993, Transfer of the gastropod family 
Piesiotrochidae to the Campaniloidea based on 
sperm ultrastructural evidence. Journal of Mol- 
luscan Studies, 59: 135-146. 

HODGSON, A., 1993, Spermatozoon structure and 
spermiogenesis in Nassahus kraussianus (Gas- 
tropoda, Prosobranchia, Nassariinae). Inverte- 
brate Reproduction and Development, 23: 1 1 5- 
121. 



58 



ATTIGA & AL-HAJJ 



HODGSON, A. & J. HELLER, 1990, Spermatogen- 
esis and sperm structure of the normally parthe- 
nogenetic snail Melanoides tuberculata. Israel 
Journal of Zoology, 37: 31-50. 

HOUBRICK, R., 1985, Genus Clypeomorus Jous- 
seaume (Cehtlniidae; Prosobranchia), Smithso- 
nian Contributions to Zoology, Number 403: 131 
pp. 

HOUBRICK, R., 1988, Cerithioidean phylogeny. 
Malacologlcal Review, 4, Supplement, 88-128. 

HUAQUIN, L. & E., BUSTOS-OBREGON, 1981 , Ul- 
trastructural analysis of spermatid differentiation 
in Concholepas concholepas. Archivum Biologie 
(Bruxelles), 92: 259-274. 

HULINGS, N., 1 986, Aspects of the reproduction of 
rocky intertidal molluscs from the Jordan Gulf of 
Aqaba (Red Sea). The Vellger, 28(3): 318-327. 

JARAMILLO, R., O. GARRIDO & В. JORQUERA, 
1986, Ultrastructural analysis of spermiogenesis 
and sperm morphology in Chorus giganteus 
(Lesson, 1929) (Prosobranchia: Muhcidae). The 
Vellger, 29: 217-225. 

JONG-BRINK, M., H. BOER, T. HOMME & A. 
KODDE, 1977, Spermatogenesis and the role 
of Sertoli cells in the freshwater snail Bl- 
omphalarla glabrata. Cell and Tissue Research, 
181: 37-58. 

KOHNERT, R., 1980, Zum Spermiendimorphismus 
der Prosobranchier: Spermiogenese und Ultra- 
struktureller Aufbau der Spermien von Blthynia 
tentaculata (L.). Zoologischer Anzeiger, 205: 
145-161. 

KOHNERT, R. & V. STORCH, 1 984a, Vergleichend- 
ultrasrtukturelle Untersuchungen zur morpholo- 
gie eupyrener Spermien der Monotocardia 
(Prosobranchia). Zoologische Jahrbucher, 11: 
51-93. 

KOHNERT, R. & V. STORCH, 1984b, Elektronen- 
mikroskopische Untersuchungen zur Spermio- 
genese der eupyrenen Spermien der Monotocar- 
dia (Prosobranchia). Zoologischer Jahrbucher, 
112: 1-32. 

KOIKE, К., 1985, Comparative ultrastructural stud- 
ies on the spermatozoa of the Prosobranchia 
(Mollusca: Gastropoda). The Science Report of 



Faculty Education, Gumma University, 34: 33- 
153. 

KOIKE, K. & S. NISHIWAKI, 1980, The ultrastruc- 
ture of dimorphic spermatozoa in two species of 
the Strombidae (Gastropoda: Prosobranchia). 
Venus, Japanese Journal of Malacology, 38: 
259-274. 

MAXWELL, W., 1983, Mollusca Pp. 1-312, in: К 
Adiyodi R Adiyodi, Reproductive biology of In- 
vertebrates, Vol. II: Spermiogenesis and sperm 
function, John Wiley and Sons, New York. 

MINNITI, F., 1993, A morphological and ultrastruc- 
tural study of euspermatozoa and parasperma- 
tozoa in Cerithium rupestre Risso (Caenogas- 
tropoda) and its phylogenetic significance. 
European Archive of Biology, 104: 7-19. 

NISHIWAKI, S., 1964, Phylogenetical study on the 
type of the dimorphic spermatozoa in Proso- 
branchia. The Science Report of the Tokyo Ky- 
olku Dalgaku, Sec. B, 2: 237-275. 

READER, T., 1973, Histological and ultrastructural 
studies on the testis of Bithynia tentaculata (Mol- 
lusca: Gastropoda), and on the effects of Cer- 
carla helvetica XII (Trematoda: Digenea) on this 
host organ. Journal of Zoology (London), 171: 
541-561. 

SELMI, M. & GIUSTI, F., 1980, Structure and func- 
tion in typical and atypical spermatozoa of 
Prosobranchia (Mollusca). I. Cochlostoma mon- 
tanum (Issel) Mesogastropoda (1). Atti della Ac- 
cademia del Fisiocritici Siena, IV Congresso (So- 
cietà IVIalacologica Italiana, Siena, 1978): 115- 
147. 

SPURR, A., 1969, A low-viscosity epoxy resin em- 
bedding medium for electron microscopy. Jour- 
nal of Ultrastructure Research, 26: 31-43. 

WALKER, M. & MACGREGOR, H., 1968, Sper- 
matogenesis and the structure of the mature 
sperm in Nucella lapillus (L.). Journal of Cell Sci- 
ence, 3: 95-104. 

WEST, D., 1978, Reproductive biology of Colus 
stimpsoni (Prosobranchia, Buccinidae) II. Sper- 
miogenesis. The Veliger, 21: 1-10. 

Revised Ms. accepted 28 November 1995 



MALACOLOGIA, 1996, 38(1-2): 59-65 

CA REGULATION IN THE FRESHWATER BIVALVE ANODONTA IMBECILIS: 

I. EFFECT OF ENVIRONMENTAL CA CONCENTRATION AND BODY MASS ON 

UNIDIRECTIONAL AND NET CA FLUXES 

Dazhong Xu^ & Michèle G. Wheatly^ 
Department of Zoology, University of Florida, Gainesville, Florida 32611, USA 

ABSTRACT 

The present paper reports unidirectional and net Ca fluxes of a freshwater bivalve, Anodonta 
imbecilis, as a function of external Ca concentration and body mass. Larger animals were better 
able to maintain Ca balance than smaller animals, which experienced net loss of Ca. External 
Ca concentration had no significant effect on net Ca flux. Unidirectional Ca influx decreased 
with body mass and increased with external Ca concentration. The relationship between ex- 
ternal Ca concentration and unidirectional Ca influx follows the Michaelis-Menten equation. The 
estimated half saturation Ca concentration for unidirectional Ca influx and the maximum uni- 
directional influx were 0.213 mM and 4.329 |imol g dry mass ^h \ respectively. External Ca 
concentration did not affect unidirectional Ca efflux of the animals. Unidirectional Ca efflux 
decreased with body mass. 

Key words: Calcium flux, calcium concentration, body mass, bivalve. 



INTRODUCTION 

While calcification has been relatively well 
studied in molluscs (reviev^/ed by Watabe, 
1983), the contribution nnade by whole body 
unidirectional Ca flux to Ca regulation in bi- 
valves has not been well established. 

In molluscs, various epithelia can take up 
external Ca (Van der Borght, 1963; Van der 
Borght & Van Puymbroeck, 1964; Green- 
away, 1971; Thomas et al., 1974), especially 
the mantle surface facing the mantle cavity 
(Jodrey, 1953; Horiguchi, 1958) and the gill 
(Horiguchi, 1958). The active transport of Ca 
was demonstrated in the freshwater gastro- 
pods Lymnaea stagnalis (Van der Borght & 
Van Puymbroeck, 1964; Greenaway, 1971) 
and Biomphalaria glabrata (Thomas et al., 
1974). Based on models for active Ca uptake 
into other freshwater species (for example te- 
leost fish, Flik et al., 1985), the epithelial up- 
take occurs in two stages: diffusion down an 
electrochemical gradient across the apical 
membrane into the cytosol, and active trans- 
port across the basolateral membrane into 
the haemolymph. If, as in other freshwater 
species, the Ca pump in molluscan epithelia 
is Ca-activated ATPase (Watabe, 1983), the 
rate of Ca uptake should be correlated with 
the Ca concentration in the ambient water. 



Previous studies showed that this is true for 
some molluscs (Greenaway, 1971; Thomas 
et al., 1974; Russell-Hunter, 1978; Pynnonen, 
1991) but not others (Hunter, 1975; Russell- 
Hunter et al., 1981). 

Allometry refers to the scaling of physio- 
logical function/morphological parameters to 
body mass. Interspecific allometry of captive 
aquatic molluscs described scaling of water 
flux to body mass (Nagy & Peterson, 1988). 
Significant relationships between dry body 
mass and shell length, shell height, gut-pas- 
sage time, gut content or metabolic fecal loss 
have also been reported (Hawkins et al., 
1990). The relationship between dry mass 
and oxygen consumption had a negative cor- 
relation (Dietz, 1974). 

The present study uses radiotracer tech- 
niques to determine the unidirectional Ca 
fluxes in a freshwater bivalve, Anodonta im- 
becilis, as a function of external Ca concen- 
tration and body mass. 

MATERIALS AND METHODS 

Experimental Animals and General 
Holding Conditions 

The freshwater bivalves Anodonta imbeci- 
lis (6-58 g) were collected 60 km from 
Gainesville, Florida, from a canal along the 



'Present Address; Department of Biological Sciences, Wright State University, Dayton, OH 45435, USA. 



59 



60 



XU & WHEATLY 



TABLE 1. Wet and dry mass of animals used to determine the effect of body mass on unidirectional 
and net Ca flux. 



group 1 



group 2 



group 3 



group 4 



wet mass (g) 
dry mass (g) 



11.15±0.89 
0.362 ± 0.041 



15.8810.56 
0.511 ±0.064 



25.98 ± 1.40 
1.061 ±0.076 



44.02 ± 2.03 
1.929 ±0.171 



Mean ± SEM. N (groups 2, 3, 4) = 10. N (group 1) = 9. 



Suwannee River at Fanning Springs. The an- 
imals collected were kept in aquaria with 
aged and well-aerated 21 'C Gainesville tap 
water with the following cationic composition 
(in mM): Na\ 0.55; K^ 0.04; Ca^", 0.60; 
Mg^", 0.42; and CI , 0.73. The pH was 7.7. 
Food was withheld for the holding period (up 
to two months), and animals were used 
within two months of collection. Animals 
were acclimated in aquaria for at least 10 
days before measurements were made. Only 
healthy animals (indicated by relatively heavy 
weight, active ventilation and powerful water 
ejection upon disturbance) were used in the 
experiments. All experiments were con- 
ducted at 21 C. 

Unidirectional and Net Ca Fluxes — Effect of 
External Ca Concentration 

Four groups of animals were used in the 
experiment. The Ca concentrations of the ex- 
perimental media were 0.27, 0.60, 1.00 and 
2.00 mM, respectively. The outer surface of 
the shell of the experimental animals was 
covered with wax to prevent direct Ca loss 
from the shell/water interface. Animals were 
acclimated in the experimental water for 3 
days before conducting the experiment. Me- 
dia with Ca concentration of 1.00 and 2.00 
mM were made by adding CaCl2 to Gaines- 
ville tap water (0.6 mM Ca). The medium with 
Ca concentration of 0.27 mM was made us- 
ing the following recipe (in mM): NaCI, 0.4; 
CaCl2, 0.27; NaHCOg, 0.2; and KCl, 0.04. An- 
imals were placed individually in experimen- 
tal flux chambers containing 300 ml medium 
and acclimated for more than 12 hours. At 
the beginning of a flux measurement, the wa- 
ter was drained from the chamber and 200 ml 
fresh medium were added. An initial water 
sample was taken from each chamber and 
then 1 |.iCi of ^^Ca (CaCU in water, 10 mCi 
ml \ Du Pont) was added to each chamber. 
Water samples were taken from each cham- 
ber at t = h and t = 6 h for determination of 
radioactivity and Ca concentration. These 
samples were used to estimate net and uni- 



directional Ca fluxes. At the end of the exper- 
iment, animals were sacrificed by cutting the 
adductor muscles using a dissecting knife. 
Soft tissues of each animal were then dis- 
sected out and dried to constant weight to 
determine dry mass. In a parallel experiment, 
empty shells (the valves sealed together and 
covered with wax on the outer surface) were 
bathed in an identical chamber to estimate 
the possible accumulation of ''^Ca by the 
shell surface. Throughout the paper, dry 
mass means the dried mass of the soft tis- 
sues (excluding shells), wet mass refers to 
the whole wet mass of the animals (including 
shells). 

Unidirectional and Net Ca Flux — Effect of 
Body Mass 

Bivalves with wet mass of 6-58 g (N = 39) 
and dry mass of 0.2-2.7 g were used in the 
experiment. Animals were divided into four 
groups according to wet mass (Table 1). 
Groups were numbered 1 to 4 (small to 
large). For each group, the flux volume and 
isotope addition were as follows: group 1, 
100 ml and 0.5 цС1 "*^Ca; group 2 and 3, 150 
ml and 0.8 |.iCi ''^Ca; group 4, 200 ml and 1 
|jCi "'^Ca. The experimental method was the 
same as described above. The Ca concen- 
tration of the media was 1 mM. 

Analytical Methods 

Water samples (3 ml) were mixed with 
ScintiVerse fluor (3 ml) and then radioactivity 
was measured using a liquid scintillation 
counter (LSC, Beckman LS5801). The Ca 
concentration of experimental water or extra- 
pallial fluid (EPF) was measured after appro- 
priate dilution (0.2 ml sample + 2 ml 2% LaClg 
+ 1.8 ml distilled water) using an atomic ab- 
sorption spectrophotometer (Perkin Elmer 
2100). 

Calculation 

The flux equation described by Wheatly 
(1989) was used to calculate unidirectional 
Ca influx: 



Ca REGULATION IN THE FRESHWATER BIVALVE ANODONTA IMBECILIS 



61 



JIn 



(Ri - Rf)V 
SAtW 



(1) 



Influx 



where Jin is unidirectional Ca influx (|лто1 g 
dry mass ^h ^), Ri and Rf are the initial and 
final radioactivity (cpm ml "') of respective 
water samples, V is the flux volume (ml), SA is 
the medium mean specific radioactivity (cpm 
)imol ^) calculated as the mean radioactivity 
divided by the mean Ca concentration, t is 
the elapsed time (h), and W is the dry mass of 
the animal (g). 
Net flux was calculated as: 



Jnet 



(Ci - Cf)V 

tw 



(2) 



where Jnet is Ca net flux (цто1 g dry mass ^ 
h ''), Ci and Cf are the initial and final me- 
dium Ca concentrations (mM), V is the flux 
volume (ml), t is the elapsed time (h), and W is 
the dry mass of the animal (g). A positive 
value for net flux indicates net Ca influx while 
a negative value indicates net Ca efflux. 

Unidirectional efflux was calculated using 
the conservation equation: 



Jout = Jin - Jnet 



(3) 



Statistical Analysis 



Data were expressed as mean and stan- 
dard error (± SEM). The statistical analysis 
was performed using StatView 4.01 and Su- 
per ANOVA computer packages. Correlation 
analysis was performed by calculating corre- 
lation coefficients (r values) and using Fish- 
er's r to z method to test the significance of 
correlation. One-factor ANOVA was used to 
analyze differences between groups and 
then Fisher's PLSD test was used when nec- 
essary to compare the means. ANCOVA was 
used to compare the slopes and intercepts of 
different linear relationships. The significance 
level for all statistical analyses was set at 
0.05. 



RESULTS 



W) 



о 




0.5 1 1.5 2 

Ca concentration in the medium (mmol Г ) 

FIG. 1. Unidirectional and net fluxes of Anodonta 
imbecilis in media of different Ca concentrations. 
Points represent mean and standard errors. The 
wet mass of animals used were as follows: group 1 
(Ca = 0.27 mM), 41 .59 ± 2.30 g (N = 9); group 2 (Ca 
= 0.6 mM), 40.14 ± 2.41 g (N = 10); group 3 (Ca = 
1 .0 mM), 44.02 ± 2.03 g (N = 1 0); group 4 (Ca = 2.0 
mM), 43.59 ± 2.07 g (N = 8). The equation of the 
curve fitting the influx data is: influx 4.329C/(0.213 
+ C), where С is the Ca concentration of the me- 
dium; r = 0.996. 



of the animals in media containing 0.60 mM 
(Fisher's PLSD, p = 0.0036), 1.00 mM (Fish- 
er's PLSD, p = 0.0007) and 2.00 mM Ca 
(Fisher's PLSD, p = 0.0001). There were no 
significant differences in Ca net flux in Ca 
concentrations of 0.60, 1.00, 2.00 mM (Fish- 
er's PLSD, p = 0.1500 for the largest differ- 
ence). 

There was a nonlinear relationship be- 
tween the unidirectional Ca influx and the ex- 
ternal Ca concentration (Fig. 1). The unidirec- 
tional influxes for animals in media containing 
0.27, 0.60, 1.00 and 2.00 mM Ca were sig- 
nificantly different (one-factor ANOVA, p = 
0.0132). The mean unidirectional influx in- 
creased as external Ca concentration in- 
creased, was partially saturable and could be 
approximately described by the Michaelis- 
Menten equation: 



Empty shells showed no significant accu- 
mulation of '^^Ca on the waxed outer shell 
surface. Ca net flux was affected by the Ca 
concentration in the medium (Fig. 1). Animals 
in medium containing 0.27 mM Ca showed a 
significant net Ca efflux of - 1 .63 ± 0.24 цто1 
g dry mass ^h ^ (N = 9) compared to those 



Influx = К 



Km + C 



(4) 



where К is the maximum rate of unidirec- 
tional Ca influx. Km is the Ca concentration in 
the medium at which half saturation is at- 
tained, and С is the Ca concentration in the 



62 



XU & WHEATLY 



CO 

CO 



Q 



3n 



1- 



dry mass = 0.047(wet mass) -0.184 



10 



20 



30 

Wet mass (g) 



40 



50 



60 



FIG. 2. Relationship between dry mass and wet mass of Anodonta imbecilis. Points represent the mass of 
individual animals; r = 0.909 for regression. Fisher's r to z, p < 0.0001, N = 39. 



medium. The calculated half saturation Ca 
concentration for unidirectional Ca influx 
(Km) was 0.213 mM. The maximum unidirec- 
tional influx (K) was 4.329 |jmol g dry mass 
^h \ Thus, the following equation can be 
used to describe the relationship between 
unidirectional influx and external Ca concen- 
tration: 



Influx = 4.329 



0.213 + С 



(5) 



Unidirectional Ca effluxes showed no sig- 
nificant difference among animals in media 
with different Ca concentration (one factor 
ANOVA, p = 0.7807; Fig. 1). 

The dry mass of bivalves (excluding shell) 
was positively correlated with wet mass (r = 
0.909, Fisher's r to z, p < 0.0001, Fig. 2). 
Larger animals generally maintained Ca bal- 
ance as indicated by the negligible net flux 
(Fig. 3). However, animals smaller than 0.5 g 
tended to exhibit a negative Ca balance as 
indicated by a net efflux (Fig. 3). Both the 
unidirectional Ca influx and efflux decreased 
with increase in dry body mass. Negative lin- 
ear relationships were derived between log 
unidirectional Ca influx and log dry mass (r = 

0.800, Fisher's r to z, p < 0.0001), and log 
unidirectional efflux and log dry mass (r = 
-0.862, Fisher's r to z, p < 0.0001; Fig. 4). 
The slopes and intercepts respectively were 
as follows: -4.03, 6.52 (unidirectional influx) 
and -8.34, 7.54 (unidirectional efflux). The 



slope and intercept for unidirectional efflux 
were both significantly higher than those for 
unidirectional influx (ANCOVA, p = 0.0001). 



DISCUSSION 

Active unidirectional Ca influx that follows 
enzyme saturation kinetics has been previ- 
ously demonstrated in freshwater snails. 
Greenaway (1 971 ) found in the snail Lymnaea 
stagnalis that active uptake of Ca was nec- 
essary below external levels of 0.5 mM. The 
uptake mechanism was half-saturated and 
near-saturated in external media containing 
0.3 and 1.0-1.5 mM Ca, respectively, and 
snails showed a positive Ca balance in media 
containing more than 0.062 mM Ca. For the 
snail Biomphalaria glabrata, the half and near 
saturated Ca concentration for Ca uptake 
were 0.267 and 1.0-2.0 mM, respectively, 
and the minimum equilibrium concentrations 
were 0.012-0.025 mM for a closed system 
and 0.25 mM for an open system (Thomas et 
al., 1974). Both animals exhibited a high af- 
finity Ca uptake mechanism. In the present 
study, the unidirectional Ca influx of Л. imbe- 
cilis seems to display the same kinetics as a 
function of external Ca concentration. The 
half saturation Ca concentration in the me- 
dium was 0.213 mM, lower than the value 
estimated for the freshwater snail L. stagnalis 
(0.3 mM; Greenaway, 1971) and the value es- 



Ca REGULATION IN THE FRESHWATER BIVALVE ANODONTA IMBECILIS 



63 





CO 




ГЛ 


-X 


cd 


HJ 

^ 


s 


■+^ 


h 


0) 

tí 




cd 


bJD 


О 


о 



-5^ 


• 
• • 


• 


• 


10^ 
15- 


• • 

• 

• 

• 






20^ 


• • 






25- 


• 






30-^ 




— 1 г — ^ ' — 


— ' ' 1 



О 



1 2 

Dry mass (g) 



FIG. 3. Relationship between Ca net flux and body dry mass of /4noc/onfa imbecilis. Points represent Ca net 
flux of individual animals. The Ca concentration of the medium is 1 mM. N = 36. 



100 



В 

ce 
О 



10 



log (influx) = -0.40281og (dry mass) + 0.6520 
log (efflux) = -0.83421og (dry mass) + 0.7548 





0.1 



10 



Dry mass (g) 



FIG. 4. Relationship between log unidirectional Ca influx and log dry mass (r = 0.800, Fisher's r to z, p < 
0.0001), and between log unidirectional Ca efflux and log dry mass (r - 0.862, Fisher's r to z, p < 0.0001) 
of Anodonta imbecilis. Points represent the unidirectional influx or efflux of individual animals. The slopes 
for log unidirectional influx and log unidirectional efflux are significantly different (ANCOVA, p = 0.0001). N 
= 36. 



timated for the freshwater snail Blomphalaria 
glabrata (0.267 mM; Thomas et al., 1974). 

If the unidirectional Ca efflux is purely by 
passive diffusion, one would expect unidirec- 
tional efflux to decrease as the external con- 
centration is raised. Because unidirectional 
Ca efflux was unaffected by change in exter- 



nal Ca concentration, other mechanisms may 
be involved. Greenaway (1971) suggested 
that part of the unidirectional Ca influx in L. 
stagnalis is due to exchange diffusion and 
that this component increases when external 
Ca concentration increases following en- 
zyme-saturation kinetics. If the unidirectional 



64 



XU & WHEATLY 



Ca efflux in A. imbecilis is attributable to ex- 
change diffusion, then unidirectional efflux 
would increase as external Ca rises. Any por- 
tion of unidirectional Ca efflux not attributed 
to exchange diffusion ("routine loss") would 
decrease with the increase in external Ca 
concentration because of the reduction in 
concentration gradient. The combined effect 
might be that unidirectional Ca efflux is un- 
changed. The net Ca flux of animals in media 
of different Ca concentration mirrored the 
change in unidirectional Ca influx because 
unidirectional Ca efflux remained constant. 
This pattern of Ca net flux is similar to the Ca 
net uptake pattern of L. stagnalis in media of 
different Ca concentration (Greenaway, 
1971). The difference between these two an- 
imals is that in A. imbecilis Ca net flux was 
negative as opposed to the net uptake exhib- 
ited by L. stagnalis. 

The relationship between wet and dry 
mass of A. imbecilis is linear, indicating the 
proportional increase of soft body tissue with 
shell and water content. A similar relationship 
was found between shell weight and fresh 
tissue weight in the freshwater snail L. stag- 
nalis (Greenaway, 1971). 

Few animals in this study exhibited a sig- 
nificant net uptake of Ca from the medium, 
suggesting that considerable Ca is obtained 
from dietary sources. The freshwater snail 
Lymnaea stagnalis was found to obtain 20% 
of its calcium from food (Van der Borght & 
Van Puymbroeck, 1966). This is also consis- 
tent with previous work, which demonstrated 
that freshwater bivalves obtain part of their 
Ca from food (Pynnonen, 1991). 

Larger animals are better able to maintain 
their Ca balance with the environment than 
smaller animals, which tend to lose Ca to the 
medium. This was due to the fact that unidi- 
rectional Ca efflux is larger than Ca influx in 
small animals and decreases at a greater rate 
with increase in body mass. This implies that 
smaller animals depend more on dietary Ca 
than larger animals, possibly since they cal- 
cify their shell more rapidly. 

Allometric regression showed that unidi- 
rectional Ca fluxes decreased with dry body 
mass of A. imbecilis. Thus, smaller animals 
exchange Ca with their environment more 
rapidly commensurate with an increased sur- 
face area to volume ratio. This result is similar 
to a recent study on crayfish, which revealed 
that the diffusional and active ion flux rates 
are both greater in small crayfish (Wheatly et 
al., 1991). An allometric study of Na fluxes in 



amphibia (Pruett et al., 1991) showed that re- 
gression lines for unidirectional Na influx and 
efflux had the same slope and intercept con- 
firming Na balance in animals of all size. In 
the present study, efflux decreased more 
than influx with increase of body dry mass 
resulting in significantly greater net efflux of 
Ca in smaller animals. 



ACKNOWLEDGMENTS 

This research was supported by NSF grant 
DCB 89 1 641 2 to MGW. We thank Drs. Karen 
Bjorndal, David Evans and Frank Nordlie for 
their suggestions on the research and for use 
of experimental equipment. We thank Dr. Jim 
Williams, Ms. Jane Brimbox and Mr. Ricardo 
Lattimore in the U.S. Fish and Wildlife service 
for their help in collecting experimental ani- 
mals. 



LITERATURE CITED 

DIETZ, T. H., 1974, Body fluid composition and 
aerial oxygen consumption in the freshwater 
mussel, Ligumia subrostrata (Say): effects of de- 
hydration and anoxic stress. Biological Bulletin, 
147: 560-572. 

FLIK, G. R., J. H, VANRIJS & S. E. WENDELAAR 
BONGA, 1985, Evidence for high-affinity Ca^*- 
ATPase activity and ATP-driven Ca^*-transport 
in membrane preparations of the gill epithelium 
of the cichlid fish Oreochromis mossambicus. 
Journal of Experimental Biology, 1 19: 335-347. 

GREENAWAY, P., 1971, Calcium regulation in 
freshwater mollusc Limnaea stagnalis (L.) (Gas- 
tropoda: Pulmonata). I. The effect of internal and 
external calcium concentration. Journal of Ex- 
perimental Biology, 54: 199-214. 

HAWKINS, A. J. S., E. NAVARROL & J. I. P. IGLE- 
SIAS, 1990, Comparative allometries of gut-pas- 
sage time, gut content and metabolic faecal loss 
in Mytilus edulis and Cerastoderma edule. Ma- 
rine Biology, 105: 197-204. 

HORIGUCHI, Y., 1958, Biochemical studies on 
Pteria (Pinctada) martensii (Dunker) and Hyriop- 
sis schlegeli (Martens) IV. Absorption and trans- 
ference of '''Ca in Hyriopsis schlegeii (Marterns). 
Bulletin, Japanese Society of Scientific Fishieries, 
23: 710-715. 

HUNTER, R. D., 1975, Variation in population of 
Lymnaea palustris in upstate New York. Ameri- 
can [[Midland Naturalist, 94: 401-420. 

JODREY, L. H., 1953, Studies on shell formation. 
III. Measurement of calcium deposition in shell 
and calcium turnover in mantle tissue using the 
mantle-shell preparation and ''^Ca. Biological 
Bulletin, 104: 398-407. 



Ca REGULATION IN THE FRESHWATER BIVALVE ANODONTA IMBECILIS 



65 



NAGY, K. A. & С С. PETERSON, 1988, Scaling of 
water flux rate in animals. University of California 
Press, Berkeley. 

PRUETT, S. J., D. F. HOYT & D. F. STIFFLER, 
1991, The allometry of osmotic and ionic regu- 
lation in amphibia with emphasis on intraspecific 
scaling in larval Ambystoma tigrinum. Physiolog- 
ical Zoology, 64: 1 1 73-1 1 99. 

PYNNONEN, K., 1 991 , Accumulation of ''^Ca in the 
freshwater Unionids Anodonta anatina and Unio 
tumidus, as influenced by water hardness, pro- 
tons, and aluminum. Journal of Experimental Zo- 
ology, 260: 18-27. 

RUSSELL-HUNTER W. D., 1978, Ecology of fresh- 
water pulmonates. Pp. 335-383, in: V. Fretter & 
J. Peake, eds., Pulmonates 2A, Academic Press, 
London. 

RUSSELL-HUNTER, W. D., A. J. BURKY & R. D. 
HUNTER, 1981, Interpopulation variation in cal- 
careous and proteinaceous shell components in 
the stream limpet, Ferrissia rivularis. Malacolo- 
gia, 20: 255-266. 

THOMAS, J. D., M. BENJAMIN, A. LOUGH & R. H. 
ARAM, 1974, The effects of calcium in the ex- 
ternal environment on the growth and natality 
rates of Biomphalaria glabrata. Journal of Animal 
Ecology, 43: 839-860. 

VAN DER BORGHT, O., 1963, In- and outflux of 



Ca-ion in freshwater gastropods. Archives Inter- 
nationales de Physiologie et de Biochimie, 71: 
46-50. 

VAN DER BORGHT, О. & S. VAN PUYMBROECK, 
1964, Active transport of alkaline earth ions as 
physiological base of the accumulation of some 
radio-nuclides in freshwater molluscs. Nature, 
204: 533-535. 

VAN DER BORGHT, O. & S. VAN PUYMBROECK, 
1966, Calcium metabolism in a freshwater mol- 
lusc: quantitative importance of water and food 
as supply for calcium during growth. Nature, 
210: 791-793. 

WATABE, N., 1983, Shell formation. Pp. 236-287, 
in: K. M. Wilbur & A. S. M. Saleuddin eds.. The 
Mollusca Vol. 4. Physiology, Academic Press, 
New York. 

WHEATLY, M. G., 1989, Physiological responses 
of the crayfish Pacifastacus leniusculus to envi- 
ronmental hyperoxia. Journal of Experimental Bi- 
ology, 143: 33-51. 

WHEATLY, M. G., F. P. ZANOTTO & A. T GAN- 
NON, 1991, Allometry of postmolt calcification 
and associated ion fluxes in crayfish. American 
Zoologist, 31: 119A. 



Revised Ms. accepted 1 January 1996 



MAUVCOLOGIA, 1996, 38(1-2): 67-85 

MICROSCULPTURES OF CONVERGENT AND DIVERGENT POLYGYRID 

LAND-SNAIL SHELLS 

Kenneth С Emberton 

Department of Malacology, Academy of Natural Sciences of Pfiiladelphia, 1900 Benjamin 
Franklin Parkway, Philadelptiia, Pennsylvania 19103-1195, U.S.A. 

ABSTRACT 

Polygyrid evolution has produced five pairs of closely convergent shell forms, four of which 
occur in sympatry. Scanning electron microscopy of the apertura! parietal and basal denticles 
(or regions) (at about 500 « ) in those ten species, and of the body whorl (at about 1 0Ov) in those 
and eleven more polygyrid species, reveals possible new microsculptural characters, homol- 
ogies, and radiations. Twelve informative new character states are tentatively proposed, of 
which half support, without homoplasy, previous shell-free phylogenetic hypotheses based on 
anatomy and allozymes. Two of the homoplastic characters actually enhance shell-form con- 
vergences, which are nonetheless distinguishable using other microsculptural features. Further 
SEM studies are warranted to test these proposed characters, to add others, and to test the 
hypothesis that shell micromorphology is much more informative than shell macromorphology 
for land-snail phylogenetics. 

Key words: Gastropoda: Stylommatophora, morphology, systematics, phylogenetics, cla- 
distics, SEM. 



INTRODUCTION 

Polygyrid shell-form evolution is unique for 
its multiple close convergences in sympatry 
and is also noteworthy for its sudden diver- 
gences; shell-sculpture evolution in polygy- 
rids is also of great interest for its repeated 
convergences on periostracal hairs and its 
divergences among sister taxa (Pilsbry, 1 940; 
Solem, 1976; Emberton, 1988, 1991a, 1994a, 
1995a, 1995b; Asami. 1988, 1993). These 
combinations of close convergences and 
rapid divergences make it virtually impossi- 
ble to reconstruct polygyrid phylogeny from 
gross shell morphology, even when develop- 
mental characters are viewed by x-ray (Em- 
berton, 1995b). Polygyrid shell convergences 
and divergences, however, have so far been 
compared only macroscopically, at magnifi- 
cations no greater than 50x. 

The purpose of this paper is a preliminary 
assessment of the the microsculptures of se- 
lected polygyrid shell convergences and di- 
vergences, using scanning electron micros- 
copy (SEM). 

Four species of polygyrids have previously 
been examined under SEM for microsculp- 
tural features of the apertural lip: Stenotrema 
barbatum (Clapp) (Solem, 1972: figs. 23, 24 
= Solem, 1974: fig. 5), as well as Daedalochila 



auriformis (Bland), Millerelix mooreana (Bin- 
ney), and M. doerfeuilliana sampsoni (Weth- 
erby) (Solem & Lebryck, 1976: figs. 33-46). 
All four species had fields of hexagonal to 
rounded crystalline plates, uplifted on one 
side, and those plates varied in size and dis- 
tributional pattern among species. Intraspe- 
cific variation was studied in D. auriformis, 
with the important discoveries that the pari- 
etal and the palatal apertural denticles dif- 
fered in microsculpture, and that a gerontic 
shell had more strongly developed micro- 
sculpture than a younger adult shell (Solem & 
Lebryck, 1976). 

Only a single polygyrid specimen has pre- 
viously been examined under SEM for shell 
periostracal microsculpture. Stenotrema bar- 
batum exhibited at 195x and 360x a regular 
array of gradually tapered, sharp-pointed 
hairs, seemingly round in cross-section and 
projecting perpendicularly from a surface 
field of subparallel, slightly anastomosing 
ridges (Solem, 1974: fig. 6). 

The present study is a preliminary survey, 
based on only one shell per species (al- 
though including several pairs of sister spe- 
cies), so the microsculptural characters dis- 
covered herein must be considered tentative. 
In order to minimize the known sources of 
intraspecific variation (Solem & Lebryck, 



67 



68 



EMBERTON 



1976), only gerontic shells were used and 
both parietal and palatal apertural denticles 
(or regions) were examined. 



MATERIALS AND METHODS 

Twenty-one polygyrid species were cho- 
sen for examination; Figure 1 presents their 
phylogenetic relationships as hypothesized 
from anatomical and biochemical data (Em- 
berton, 1988, 1991a, 1994a, 1995b). The 
species include North America's four most 
extreme cases of shell-form convergence in 
sympatry (Emberton, 1995b: fig. 1): globose 
Neohelix major and Mesodon normal is, um- 
bilicate Allogona profunda and Appalachina 
sayana, flat Xolotrema fosteri and Patera lae- 
vior, and tridentate Triodopsis fallax and /n- 
flectarius inflectus. A fifth shell-form conver- 
gence (Emberton, 1991b) was also included: 
"lipped" Neohelix dentifera and Inflectarius 
ferrissi. 

Additional polygyrid species were included 
for their periostracal-microsculpture diver- 
gences and convergences. Xolotrema deno- 
tata and X. obstricta are sister species (Em- 
berton, 1988) that can hybridize in the field 
(Vagvolgyi, 1968) and in the laboratory 
(Webb, 1980), but their differences in shell- 
whorl shape and sculpture are extreme. Xo- 
lotrema obstricta has a strongly keeled pe- 
riphery and is sculpted with large, strongly 
raised ribs, whereas X. denotata has a 
rounded periphery and is sculpted with hair- 
like processes (Pilsbry, 1940; Emberton, 
1988). The keeled, ribbed shell of X. obstricta 
is closely paralleled by that of Patera sargen- 
tiana (Pilsbry, 1940; Emberton, 1991a), with 
which it is sympatric in northern Alabama. 
Species of the Patera radiation (Emberton, 
1991a) have diverged primarily in their shell 
surface sculpture: P. laevior is smooth, P. 
sargentiana is ribbed, P. perigrapta bears in- 
cised spiral grooves, and P. appressa sculp- 
tior is pustulose (Pilsbry, 1940). 

Hair-like periostracal processes on the 
shell surface have arisen independently and 
convergently (Emberton, 1995b) in Xolotrema 
denotata, in some Vespericola such as V. Co- 
lumbiana pilosa, in the Stenotrema clade, 
and in the Inflectarius clade (Pilsbry, 1940). 
Stenotrema's radiation is marked by extreme 
divergence in shell hairs, the variation of 
which includes short and dense (e.g. S. max- 
illatum), and long and sparsely distributed 
(e.g. S. barbigerum) (Pilsbry, 1940). To a 



much smaller degree, the general shapes of 
shell hairs also seem (at 50x) to differ among 
species of Inflectarius: broad-based and 
sharp-tipped in /. inflectarius and /. magazin- 
ensis, acutely triangular in /. smithi, obtusely 
triangular in /. subpalliatus, and lost in /. fer- 
rissi, the sister species of /. subpalliatus (Em- 
berton, 1991a). 

All studied shells are in the collection of the 
Academy of Natural Sciences of Philadelphia 
(ANSP). Species authors and ANSP catalog 
numbers of vouchers are given in the figure 
captions. 

Shells were prepared for SEM using meth- 
ods modified slightly from Solem (1970): 
soaking overnight in a weak solution of de- 
tergent (Alconox), immersing for five to 20 
seconds in an ultrasonic cleaner, rinsing in 
distilled water, air-drying, and mounting — in 
standard position — on stubs using some 
combination (depending on the size of the 
shell) of double-sided conductive tape, car- 
bon paint, carbon cement, carbon paste, and 
custom-bent paper clips. Mounted shells 
were gold-coated and photographed with a 
Cambridge Stereoscan 200 SEM in one or 
more of the following views: (a) whole shell 
(or as much as possible, including the entire 
aperture) in apertural view; (b) edge of pari- 
etal denticle (or callus) at about 1 ,000x; (c) 
edge of basal denticle (or lamellum or lip) at 
about 1,000x; and (d) body-whorl sculpture 
at about 200x and at <100x if necessary. 

The resulting photographs were descrip- 
tively compared, then subjected to a stan- 
dard phylogenetic character analysis (Wiley, 
1 981 ; Brooks & McLennan, 1 991 : chapter 2). 
The states of each character were parsimo- 
niously mapped by hand on the polygyrid 
phylogenetic hypothesis (Emberton, 1988a, 
1991a, 1994a, 1995b). 

RESULTS 

Four Convergences in Sympatry 

Figures 2-5 show SEM photographs of the 
four pairs of convergent species that occur in 
sympatry, with triodopsins and Allogona on 
the left, and mesodontins on the right: a, b, 
Neohelix major and Mesodon normalis; c, d, 
Xolotrema fosteh and Patera laevior; e, f, Tri- 
odopsis fallax and Inflectarius inflectus; and 
g, h, Allogona profunda and Appalachina say- 
ana. Apertural views at lowest possible mag- 
nifications (Fig. 2) indicate extremely close 
convergences in apertural dentition between 



POLYGYRID SHELL MICROSCULPTURE 

I l Neohelix major 2a- 5a 

II G "-^ — N. dentifera 6a,c,e,g 

Ir=HI L 

■■ Xolotrema foster/ 2c-5c 

I l X. denotata 7a,c,e 

i | 
— il 

I ' X. obstricta 7b,d,f 

: Tri od op sis fall ax 2e-5e 
^ Vespericola columbiana 9a 



69 



T 



^= Allogona profunda 2g-5g 

= Stenotrema maxillatum 9c 

= Stenotrema ЬагЫдегитЭе.ЛОб 

= Patera perig rapta 8c 

= P. laevior 2d-5d 

= P. ap pressa 8e 

= P. sargentiana 8a,b,d 

= Inflectarius subpalliatus 9f 

f= I. ^emss/ 6b, d,f,h, 10a, с 

== /. magazinensis 9b 

= I. smithi 9d 

= I. inflectus 2f-5f,10b 



Ij-^ — Appalachina say an a 2h-5h 
I ' II и 

lí^B^ Mesodon normal/s 2b-5b 
G 

FIG. 1. Phylogenetic relationships of the 21 species examined for this study, as hypothesized from alloz- 
ymes and reproductive morphology and behavior (Emberton, 1988, 1991a, 1994a). Principal convergences 
are designated by five abbreviations for shell shape/size: G, globose; F, flat; T, tridentate; U, umbilicate; L, 
lipped (Emberton, 1991b, 1995b). Figure numbers of SEM photos are given for each species name; 
"2a-5a" = 2a, 3a, 4a, 5a. 



N. major and M. normalis (without dentition) 
and between X. fosteh and P. laevior (blade- 
like parietal denticles and basal lamellae); 
moderately close convergence between T. 



fallax and /. inflectus (parietal denticles similar 
but more curved in T. fallax, palatal denticles 
nearly identical, basal denticles similar but 
with a columellar buttress in T. fallax); and 



70 



EMBERTON 




FIG. 2. Apertural features of North America's four most extreme cases of polygyrid shell-form convergence 
in sympatry (Emberton, 1995b: fig. 1). A, Neohelix major (Binney), ANSP uncataloged, 5.65- . B, Mesodon 
normalis (Pilsbry), ANSP uncataloged, 5.50x. C, Xolotrema fosteri (F. С Baker), ANSP 117483, 4.43x. D, 
Patera /aewor (Pilsbry), ANSP 186465, 3.98-. E, Triodopsis fallax (Say), ANSP 192768, 4.95>. F, Inflectarius 
inflectus (Say), ANSP 91616, 6.30>. G, Mogona profunda (Say), ANSP 77867, 4.90 ^. H, Appalachina 
sayana (Pilsbry), ANSP 264654, 5.35x. 



POLYGYRID SHELL MICROSCULPTURE 



71 






У 

г 








' 


^/*' 









Ь-Ш' Ш- 


G 



FIG. 3. The parietal denticles or parietal regions of the same specimens as Fig. 2, at about 500- magni- 
fication. A, Neohelix major (Binney), ANSP uncataloged, 505 -. B, Mesodon normalis (Pilsbry), ANSP un- 
cataloged, 500 ■ . C, Xolotrema fosteri (F. С Baker), ANSP 1 1 7483, 51 ■ . D, Patera laevior (Pilsbry), ANSP 
186465, 500 ■. E, Triodopsis fallax (Say), ANSP 192768, 550/. F, Inflectarius inflectus (Say), ANSP 91616, 
565x. G, Mogona profunda (Say), ANSP 77867, 520 ■. H, Appalachina sayana (Pilsbry), ANSP 264654, 
51 5x. 



72 



EMBERTON 




FIG 4 The basal denticles or basal regions of the same specimens as Fig. 2, at about 500. magnification. 
A, Neohelix major (Binney), ANSP uncataloged, 515x. B, Mesodon normalis (Pilsbry). ANSP uncataloged, 
500x. C, Xolotrema fosteri (F. С Baker), ANSP 117483, 520.. D, Patera laevior (^''^Ьд), ANSP 186465, 
51 Ox E Triodopsis fallax (Say), ANSP 192768, 540x. F, Inflectarius inflectus (Say), ANSP 91616 540x. G, 
Allogona profunda (Say), ANSP 77867, 51 Ox. H, Appalachina sayana (Pilsbry), ANSP 264654, 520x. 



POLYGYRID SHELL MICROSCULPTURE 



73 




FIG. 5. The body-whorl sculptures of the same specimens as Fig. 2, at about 100- magnification. A, 
Neohelix major (Binney), ANSP uncataloged, 102.0-. B, Mesodon normalis (Pilsbry), ANSP uncataloged, 
101.5x. C, Xolotrema fosteri (F. С Baker), ANSP 117483, 104.54ts. D, Patera laevior (Pilsbry), ANSP 
186465, 102.5« . E, Triodopsis fallax (Say), ANSP 192768, 99.5-. F, Inflectarius inflectus (Say), ANSP 91616, 
102.5X. G, Mogona profunda (Say), ANSP 77867, 104.0-. H, Appalachina sayana (Pilsbry), ANSP 264654, 
102.5X. 



74 



EMBERTON 



only slight convergence between A\. pro- 
funda and Ap. sayana (basal node much 
broader in AI. profunda, parietal denticle 
lacking in AI. profunda. Convergences be- 
tween iterated pairs are consistently close in 
overall shell size and shape, and apertural 
size and shape (Fig. 2). 

Flat Shell Forms: On the parietal denticle or 
parietal region, convergences disintegrate at 
high magnification (Fig. 3), with the notable 
exception of X. fosteri and P. laevior, both of 
which have a smooth surface dotted with low 
mounds bearing wide, shallow, rough-bot- 
tomed craters; the mounds are virtually iden- 
tical between species in their sizes, densities, 
apparently random distributions, and struc- 
tural details (Fig. 3c, d). This microstructural 
convergence is all the more remarkable given 
the great variation in this region among the 
other six species. Background surfaces 
range from smooth (Fig. 3b, e, g, h) to floc- 
culent (Fig. 3a) to randomly pitted (Fig. 3f). 
Secondary structures range from large, 
straight escarpments both sparse (Fig. 3a) 
and dense (Fig. 3g), to small dense escarp- 
ments both straight (Fig. 3b) and polygonal 
(Fig. 3e), to rounded pustules (Fig. 3f), to 
shallow canals that are randomly sized and 
directed (Fig. 3h). 

On the basal denticle or basal region, mi- 
crosculpture in each sympatric-convergent 
species (Fig. 4) is similar to that on its parietal 
denticle or region, with the exceptions of X. 
fosteri and P. laevior. These two species 
share, on a smooth background, a pattern of 
polygonal escarpments (Fig. 4c, d) that are 
similar in density and dispersion to, but very 
different in structure from, the cratered 
mounds they share on the parietal denticles 
(Fig. 3c, d). These polygonal escarpments are 
similar in size and shape, but are slightly 
more uptilted in P. laevior than in X. fosteri. 
The other six species have virtually identical 
backgrounds on their basal and parietal den- 
ticles or regions, except that the flocculence 
in Figure 4a is less pronounced than in Figure 
3a. In secondary structures. Figures 4f and 3f 
are virtually identical, as are Figures 4e and 
3e, and Figures 4b and 3b. Figure 4a is sim- 
ilar to Figure 3a, but with the significant ad- 
dition of a new structure: a smooth-surfaced 
puddle overlying a low, bumpy knob. Figure 
4h has the canals of Figure 3h, but they are 
much smaller and sparser, and there is the 
addition in Figure 4h of low mounds. Figure 
4g differs from Figure 3g only in that its es- 



carpments are generally slightly shorter and 
less straight. 

Globose Shell Forms: The shell body-whorl 
surface at high magnification (Fig. 5) shows 
additional features previously undetected. 
The microsculptural convergence between 
Neohelix major and Mesodon normalis, indis- 
tinguishable under the dissecting micro- 
scope at 50x (Emberton, 1995a), is readily 
detectable under SEM. Although both have a 
matte-like surface produced by a pattern of 
large transverse ridges crossed by smaller 
spiral cords, the cords in N. major (Fig. 5a) 
are even in width, whereas the cords in M. 
normalis are variable in width and generally 
much wider than in N. major. Both have un- 
even, transverse micro-wrinkles, which are 
more pronounced in N. major, however. Fur- 
thermore, N. major has an even pattern of 
parallel, spiral microstriae that are totally 
lacking in M. normalis. 

These spiral microstriae appear in all three 
members of the Triodopsinae (Fig. 5a, c, e), 
as well as in Mogona (Fig. 5g), but in none of 
the four members of the Mesodontini (Fig. 
5b, d, f, h). The microstriae are periostracal 
structures only, as they do not appear in the 
underlying calcium carbonate layer in those 
patches where the periostracum has flaked 
off (Fig. 5a, c, g). 

Tridentate Shell Forms: Periostracum also 
seems to be the sole source of the unique 
pattern of jumbled wrinkles or folds that tend 
toward a spiral direction in Inflectarius inflec- 
tus (Fig. 5f ), the hairs of which (broken in this 
specimen) are simple outward extensions of 
arcuate periostracal folds, as well as of the 
unique network of transverse-trending wrin- 
kles in Xolotrema fosteri (Fig. 5c). The spirally 
oriented pustules of Patera laevior, on the 
other hand, are also part of the calcium car- 
bonate shell matrix, as evidenced by the pus- 
tules over which the periostracum had flaked 
off (Fig. 5d). The same seems to be true of 
the unusual nodulose spiral cords of Mog- 
ona profunda (Fig. 5g). 

The remaining outstanding features of 
comparative shell body-whorl microsculpture 
in Figure 5 are the transverse ridges, which 
vary in size, shape, density, and angle. In the 
convergent pair N. major and M. normalis 
(Fig. 5a, b), they are equally large (only one 
complete ridge appears in each photograph), 
low, and rounded in profile, but in N. major 
they are slightly more angled from the vertical 



POLYGYRID SHELL MICROSCULPTURE 



75 



than in M. normalis. In the convergent pairX. 
fosteri and P. laevior (Fig. 5c, d), the trans- 
verse ridges are about equal in angle, size, 
and density (three ridges in each photo- 
graph), but they are extremely weak in P. lae- 
vior. The shell convergence between T. fallax 
and /. inflectus breaks down entirely at nni- 
crosculptural level (Fig. 5e, f): T. fallax has 
strong, dense, well-separated, sharply an- 
gled transverse ridges, whereas /. inflectus 
nearly lacks them entirely. The transverse 
ridges of AI. profunda (Fig. 5g), although 
equal in density and angle to those of N. ma- 
jor (Fig. 5a), are about twice as broad, so 
broad in fact that they are adjacent, sepa- 
rated by only narrow gutters. Thus, they are 
quite different from the convergent Ap. say- 
ana, whose pronounced, narrow, well sepa- 
rated, mildly angled ridges are more like a 
stronger version of P. laevior (Fig. 5d). 

The N. dentifera — /. ferhssi Convergence 

The extreme shell-shape convergence be- 
tween Neohelix dentifera and Inflectarius fer- 
hssi (Fig. 6a, b; Emberton, 1991b) carries 
strong clues of its origins in its apertural and 
body-whorl microsculptures (Fig. 6c-h). In 
the case of N. dentifera, these clues conflict 
slightly with anatomical and allozymic evi- 
dence (Emberton, 1988, 1991b). Thus, al- 
though N. dentifera's parietal denticle (Fig. 
6c) shares a flocculent-textured background 
surface with its congener N. major (Fig. 3a), 
the dominant sculpture of short, polygonal 
escarpments is much closer to that of the 
related Triodopsis fallax (Fig. 3e), from which 
it differs only in the much steeper angles of 
its escarpments. The discrepancy is similar 
for N. dentifera's basal apertural region (Fig. 
6e), which has a flocculent ground similar to 
that of N. major (Fig. 4a), but has a sculpture 
of steep, polygonal escarpments which are 
similar in density and distribution to those of 
Xolotrema fosteri (Fig. 4c) and which are sim- 
ilar in their small size and steep polygons to 
those of Triodopsis fallax (Fig. 4e). In the dis- 
tinctly exaggerated degree of their steep- 
ness, however, the basal-lip polygons of 
Neohielix dentifera are convergently more 
similar to Patera laevior (Fig. 4d). Despite 
these discrepancies, apertural microsculp- 
ture clearly agrees with nonconchological 
data (Emberton, 1988) in placing N. dentifera 
in the Triodopsini. Body-whorl microsculp- 
ture, on the other hand, is entirely concor- 
dant: N. dentifera (Fig. 6g) is nearly identical 



to its congener N. major (Fig. 5a), from which 
it differs only in its broader spiral cords and 
weaker transverse wrinkles. 

The phylogenetic affinities of /. ferhssi (Fig. 
6b) based on shell microsculpture are clear 
and concordant with anatomical and allozy- 
mic data (Emberton, 1991a, b). Its parietal 
denticle has the same randomly pitted back- 
ground surface and the same rounded pus- 
tules as /. inflectus (Fig. 3f), except that the 
pustules are arrayed in unevenly parallel rows 
instead of distributed randomly as in /. inflec- 
tus. Inflectarius ferhssi's basal apertural re- 
gion (Fig. 6f) also seems homologous with 
the basal denticle of /. inflectus (Fig. 4f ), with 
a similar randomly pitted background surface 
(less evident in Fig. 6f due to low contrast), 
and with similar rounded pustules, which are 
more sparsely disthbuted in /. ferhssi. The 
unique body-whorl microsculpture of /. fer- 
hssi (Fig. 6h) consists of the same jumbled 
pattern of spiral-trending wrinkles as in /. in- 
flectus (Fig. 5f), but on a finer and lower size 
scale. 

Thus, shell body-whorl microsculpture 
and, to a lesser extent, shell apertural micro- 
sculpture tend to bear out generic relation- 
ships in Neohelix and Inflectarius, despite ex- 
treme divergences in shell size and shape. 

Divergences Within the Flat-Shell Clades 

Intrageneric divergence in body-whorl mi- 
crosculpture is extreme in Xolotrema (Figs. 
5c, 7) and in Patera (Figs. 5d, 8), with struc- 
tural homologies difficult if not impossible to 
decipher. Nevertheless, these two genera, 
which closely converge on each other in shell 
size and shape (Figs. 2c and 2d; Figs. 7a and 
8a) and in apertural microsculpture (Figs. 3c 
and 3d; Figs. 4c and 4d), are always clearly 
distinguishable in body-whorl microsculp- 
ture, as discussed below. 

The Xolotrema Clade: Shell divergence be- 
tween the presumably hybridizing sister spe- 
cies Xolotrema obsthcta and X. denotata is 
extreme, not only on a gross scale (Fig. 7a, 
b), but also at high magnification (Fig. 7c, d) 
and at very high magnification (Fig. 7e, f), at 
which they also can be seen to diverge from 
their congener X. fosteh (Fig. 5c). Xolotrema 
obsthcta's peripheral keel and transverse rib- 
bing are lacking in X. denotata, which bears 
periostracal hairs (all broken in this speci- 
men) lacking in X. obsthcta. Xolotrema fosteh 
lacks all these features, except transverse 



76 



EMBERTON 




FIG. 6. Two conchologically convergent species that are also are ecologically parallel and convergent 
(Emberton, 1991b). A, C, E, G, Neohelix dentifera (Binney), ANSP 90119: A, aperture, 4.56«; C, parietal 
denticle, 540- ; E, basal apertural lip, 505x; G, body-whorl sculpture, 104.0'. B, D, F, H, Inflectarius ferrissi 
(Pilsbry), ANSP 98085; B, aperture, 4.56x; D, parietal denticle, 505x; F, basal apertural lip, 51 Ox; H, 
body-whorl sculpture, 102.5x. 



POLYGYRID SHELL MICROSCULPTURE 



77 




FIG. 7. Two conchologically divergent but hybridizing sister species. A, C, E, Xolotrema obstncta (Say), 
ANSP 68553: A, body whorl to the left of the aperture, 6.25 « ; C, body-whorl sculpture, 39.0 ■ ; E, body-whorl 
sculpture, 101.0'. B, D, F, Xolotrema denotata (Férussac), ANSP 172721: B, body whorl to the left of the 
aperture, 6.3x; D, body-whorl sculpture, 38.7^; F, body-whorl sculpture, 104.0^. 



ribs, which are nonetheless lower, denser, 
and nnore angled than those of X. obstncta. 
Xolotrema fosteri's spiral microstriae are en- 
tirely lacking in both X. obstncta and X. de- 
notata. The strongest candidate for homol- 
ogy among the three species is in the 
periostral wrinkles, but these vary enor- 
mously. In X. fosteri, the wrinkles are small 
and appear only sporadically in dense retic- 
ulate patterns oriented transversely in the 
gullies between transverse ribs. In X. ob- 
stncta, the wrinkles are very large and appear 



universally in a dense, somewhat reticulate 
pattern oriented at an angle between trans- 
verse and spiral. The whnkles in X denotata 
are medium to large and universally distrib- 
uted in a sparse, somewhat reticulate pattern 
generally oriented transversely, and regularly 
punctuated at right angles by short, thick 
hairs that form cross patterns on locally 
thickened transverse wrinkles. Thus, these 
three types of wrinkles share a generally re- 
ticulate pattern, but even in this factor, their 
differences are so great (Figs. 5c, 7e, 7f ) as to 



78 



EMBERTON 




FIG. 8. A species convergent on and sympatric with Xolotrema obstricta (Fig. 7a, c, e), and two of its 
congeners. A, B, D, Patera sargentiana (Johnson & Pilsbry), ANSP 150249: A, body whorl to the left of the 
aperture, 6.30- ; B, body-whorl sculpture, 39.0/; D, body-whorl sculpture, 101. Ox. C, Patera perigrapta 
(Say), ANSP 160543, body-whorl sculpture, 104.5x. E, Patera appressa sculptior (Chadwick), ANSP 
128954, body-whorl sculpture, 102.5x. 



defy detection of homology. In fact, the peri- 
ostracal wrinkles of X. denotata (Fig. 7f) are 
much more similar to those of Inflectahus in- 
flectus (Fig. 5f ), a close convergence that dif- 
fers slightly in the general orientation of the 
wrinkles (spiral in /. inflectus and transverse in 
X. denotata) and that differs greatly in the 
form of periostracal hairs (broad upward ex- 
tensions of arcuate folds in /. inflectus; nar- 
row, buttressed blades at right angles to 
straight folds in X. denotata). The perio- 



stracum is quite thick in X. denotata, as evi- 
denced by the depth of the cracks appearing 
in Fig. 7f). 

The Patera Clade 

The intersubfamilial convergence in sym- 
patry between Patera sargentiana (Fig. 8a, b, 
d) and Xolotrema obstricta (Fig. 7a, c, e) is 
rather close. Both have large, flat, heavily 
ribbed shells with peripheral angulations 



POLYGYRID SHELL MICROSCULPTURE 



79 



(Figs. 7a, 8a), but in X. obstricta the angula- 
tion is a pronounced keel. The transverse ribs 
are equal in density and form (Figs. 7c, 8b), 
but are more raised in P. sargentiana. Both 
have spirally oriented ridges (Figs. 7e, 8d), 
but in X. obstricta these are periostracal wrin- 
kles or folds in a vaguely reticulate pattern, 
whereas in P. sargentiana they are shell-ma- 
trix pustulose cords in regularly parallel pat- 
tern, with an additional underlying substruc- 
ture of weak transverse cords. 

Patera sargentiana' s parallel spiral rows of 
micro-pustules (Fig. 8d) are clearly homolo- 
gous with those of P. laevior (Fig. 5d) de- 
scribed above. Another congener, P. peri- 
grapta (Fig. 8c), lacks pustules entirely, but 
bears parallel spiral grooves equal in density 
to the pustular rows of P. laevior. A fourth 
congener, P. appressa sculptior (Fig. 8e), 
combines the grooves of P. perigrapta (but 
weaker than in that species) with the pustules 
of P. laevior (but stronger than in that spe- 
cies): its microsculpture is one of pustules 
equally spaced within shallow, parallel, spiral 
grooves. Thus, it appears that spiral grooves, 
spiral rows of pustules, and spiral cords of 
pustules are all homologous in Patera, and 
concomitantly that the spiral gullies in P. sar- 
gentiana (Fig. 8d) are not homologous with 
the spiral grooves of P. perigrapta (Fig. 8c). 
Transverse ribbing in P. laevior, P. perigrapta, 
and P. appressa sculptior is similarly weak 
and variable, entirely unlike the strong ribs of 
P. sargentiana, but possibly homologous 
with the secondary, weak transverse ribs of 
that species. 

Shell Hairs: Convergence and Divergences 

Convergent periostracal hairs were de- 
scribed and compared above for Inflectarius 
inflectus (Fig. 5f) and Xolotrema denotata 
(Fig. 7f), although incompletely because the 
hair tips were broken off in both specimens. 
The periostracal hairs of Vespericola Colum- 
biana pilosa, which are shown unbroken and 
at the same high magnification in Figure 9a, 
are entirely different in structure. These hairs 
are thick, rigidly curved, and columnar, aris- 
ing from shallow, socket-like depressions, 
and are relatively unbuttressed. They arise 
from a unique background surface sculpture 
of both spirally and transversely oriented pat- 
terns of periostracal wrinkles, punctuated by 
large but weak transverse ribs and by small 
shelf-like protrusions (Fig. 9a). 

In Stenotrema (Figs. 9c, e; Fig. 10d), the 



periostracal hairs and surface sculptures of 
the two examined species are so different 
that they seem entirely non-homologous. 
Stenotrema maxillatum (Fig. 9c) has relatively 
small, dense, regularly arranged, backward- 
directed, elongate-conic hairs that are 
slightly buttressed transversely and that are 
marked by short, forward, spiral wrinkles in 
an otherwise transversely wrinkled, smooth 
background surface. Stenotrema barbigerum 
(Fig. 9e), on the other hand, bears relatively 
large, moderately dense, regularly arranged, 
transverse folds that arc with the concave 
side forward, that have tiny, low spines on 
their forward surfaces, and that lie in a back- 
ground surface of minute, densely packed, 
parallel, tranverse ridges overlying an uneven 
system of shallow, spiral grooves. The con- 
spicuous, micro-spinose, transverse folds of 
S. barbigerum vary in length and shape, de- 
pending on position (Fig. lOd): on the upper 
shell whorls they are low in profile, on the 
lower shell whorls their central regions are 
drawn outward and backward into thorn-like 
hairs, and on the shell's keeled periphery 
they extend outward into long, unevenly 
blade-like hairs (Fig. lOd; Fig. 9e: upper 
right). 

Unlike in Stenotrema, homologies among 
periostracal hairs of the genus Inflectarius 
(Figs. 5f; 6h; 9b, d, f) are much more evident, 
despite extensive morphological radiation. 
The hairs of /. subpalliatus (Fig. 9f) seem to 
be enlarged versions of the hairs described 
above of /. inflectus (Fig. 5f): both are long, 
arcuate, high-standing folds rising from a 
background of smaller, variously oriented 
wrinkles. Inflectarius subpalliatus's sister 
species, /. ferrissi (Fig. 6h), lacks hairs en- 
tirely, and carries only vestigial traces of folds 
and wrinkles in its relatively featureless body- 
whorl microsculpture. In stark contrast to this 
effacement, the hairs of /. magazinensis (Fig. 
9b) show increased complexity: the bottom 
of the arcuate fold is abruptly curled forward 
in a scoop-like fashion, the central high-point 
of the fold is thickened and curled over like a 
cresting wave, and extending from this crest 
is a downward arching secondary fold that 
continues onto the background surface as a 
rear-support buttress. These same modifica- 
tions are developed even further, to a re- 
markable degree, in /. smitfii (Fig. 9d). In this 
species, the entire arcuate fold is relatively 
deeply arched and scoop-like; the central 
high-point extends forward as a long, blunt, 
club-like structure with a surface sculpture of 



80 



EMBERTON 




FIG. 9. More periostracal hairs (body-whorl microsculpture). A, Vespericola columbiana pilosa (Henderson), 
ANSP 158355, 100.0-. C, Stenotrema maxillatum (Say), ANSP 170141, 99.0 -. E, Stenotrema barbigerum 
(Redfield), ANSP 1701 10, 102. Ox. B, Inflectarius magazinensis (Pilsbry & Ferries), ANSP 395865, 102.5x. D, 
Inflectarius smithi (Clapp), ANSP 160055, 100.5x. F, Inflectarius subpalllatus (Pilsbry), ANSP 171134, 
ЮЗ.Ох. 



minute, regular, adjacent pits; and the sec- 
ondary, downward-arching, rear-buttress 
fold is high-standing and strongly developed. 



Character Analysis 

All of the ten species that were examined 
for apertural microsculpture have patterns of 
escarpments, nodules, or mounds (their ab- 
sence in Fig. 3h is considered an artifact of 
the sparse distributions of mounds in that 
species: Fig. 4h). These microprojections 



seem to be homologous, with a basic mor- 
phology of inclined, crystalline platelets (Figs. 
3e, 4c-e, 6c, e) that is modified by various 
coating surfaces. A good example is the 
highly modified surface of the parietal denti- 
cle of Inflectarius ferrissi (Fig. 6d), which 
shows little evidence of crystalline platelets. 
At lower magnification, however, this surface 
can be seen to coat only the leading edge of 
the parietal denticle (Fig. 10a), the uncoated 
interior of which has a standard pattern of 
microplatelets (Fig. 10c). Similarly, in the 
closely related /. inflectus, the rounded nod- 



POLYGYRID SHELL MICROSCULPTURE 



81 




FIG. 10. A, С, apertural parietal denticle of Inflectahus ferrissi (Pilsbry), ANSP 98085: A, tip region, 43.05x; 
C, central region, 515- . B, edge of apertural parietal denticle of Inflectahus inflectus (Say), ANSP 91616, 
127.5X. D, mid region of the shell of Stenotrema barbigerum (Redfield), ANSP 170110, 12.35x. 



ules (Fig. 3f) are seen at a lower magnifica- 
tion to be crystalline microplatelets as they 
lose some of their coating away from the 
edge of the parietal denticle (Fig. 10b). 

Similar crystalline microprojections have 
been found in a wide variety of pulmonates, 
and in other gastropod groups as well 
(Solem, 1970, 1972, 1973; Solem & Lebryk, 
1976). They seem therefore to be construc- 
tional aspects of apertural deposition (Wilbur 
& Saleuddin, 1983; Watabe, 1988), although 
it has been argued that in some cases their 
shape is modified by natural selection for de- 
fense (Solem, 1972). 

The relative sizes and distributions of mi- 
croprojections seen in this study do not seem 
to be reliable systematic characters, because 
they correlate with shell size. Thus, three of 
the four large shells (Fig. 2a, g, h) have the 
largest and sparsest microprojections (Figs. 
3a, g, h; 4a, g, h); the two small shells (Fig. 2e, 
f) have the smallest and densest micropro- 
jections (Figs. 3e, f; 4e, f); and the four inter- 
mediate-sized shells (Figs. 2c, d; 6a, b) and 
one of the large shells (Fig. 2b) have micro- 
projections that are intermediate in size and 
density (Figs. 3b-d; 4b-d; 6c-f). 



Based on this analysis, a single, multi-state 
character can be proposed: 

Character 1. Apertural coating. 

State a. No coating on parietal or basal 
denticle/region; surface very smooth and 
featureless; microprojections crystalline and 
clean. Triodopsis fallax. 

State b. Thin, flocculent coating on both 
parietal and basal denticles/regions; surface 
flocculent; microprojections clean to partially 
coated. Neohelix major and N. dentifera. 

State b'. Same as state b, but on the basal 
region, the microprojections and their sur- 
roundings have an additional coating of thin, 
smooth material. N. major. 

State с Smooth-surfaced coating on the 
parietal denticle/region only; flanks but not 
tips of parietal microprojections coated; no 
coating on basal denticle/region. Xolotrema 
fosteri and Patera laevior. 

State d. Medium-thick, minutely pitted 
coating on both parietal and basal denticles/ 
regions; microprojections entirely covered on 
denticle edges, but with tips exposed away 
from denticle edges. Inflectahus inflectus and 
/. ferhssi. 



82 



EMBERTON 



State d'. Same as state d, but the coating 
covers only the edge of the parietal denticle, 
where it is thick and forms parallel rows. /. 
ferrissi. 

State e. Thick coating scored with shallow 
canals of random size and orientation, on 
both parietal and basal denticles/regions; mi- 
croprojections thickly and entirely covered. 
Appalachina sayana. 

State f. Thin, smooth coating on both pa- 
rietal and basal denticles/regions; micropro- 
jections completely to partially covered. Me- 
sodon normalis and Allogona profunda. 

Analysis of body-whorl microsculpture is 
based on the 21 species studied. Intrage- 
neric variation is so great in transverse ribs 
that they seem unreliable as systematic char- 
acters. The same must be said for the pat- 
terns of (but not necessarily for the presence 
of) periostracal wrinkles. The periostracum 
forms as a flexible, curtain-like sheath that 
later is sclerotized by quinone tanning 
(Saleuddin & Petit, 1983; Waite, 1983), so the 
pattern of wrinkles may be influenced by a 
number of environmental and constructional 
factors other than phylogenetic constraints. 
Periostracal hairs are always associated (in 
this sample) with periostracal wrinkles, how- 
ever. 

The following characters seem reliable. 

Character 2. Spiral microstriae. 

State a. Present. Neohelix major, N. dentif- 
era, Xolotrema fosteri, Triodopsis tridentata, 
Allogona profunda. 

State b. Absent. All other examined spe- 
cies. 

Character 3. Spiral cords, smooth unless tra- 
versed by spiral microstriae. 

State a. Present. Neohelix major, N. dentif- 
era, Mesodon normalis. 

State b. Absent. All other examined spe- 
cies. 

Character 4. Spiral cords, nodulose. 

State a. Present. Allogona profunda. 

State b. Absent. All other examined spe- 
cies. 

Character 5. Spiral grooves/pustules/pustu- 
late cords. 

State a. Present. Patera laevior, P. sargen- 
tiana, P. pehgrapta, P. appressa. 

State b. Absent. All other examined spe- 
cies. 



Character 6. Periostracal wrinkles. 

State a. Present. Xolotrema fosteri, X. ob- 
stricta, X. denotata, Vespericola columbiana, 
Stenotrema maxillatum, S. barbigerum, In- 
flectahus inflectus, I. ferrissi, I. smithi, I. mag- 
azinensis, I. subpalliatus. 

State b. Absent. All other examined spe- 
cies. 

Character 7. Periostracal hairs. 

State a. Thin, straight, cruciform base. Xo- 
lotrema denotata. 

State b. Thick and round, recurved, 
socket-like base. Vespericola columbiana. 

State с Thick, straight, simple base. 
Stenotrema maxillatum. 

State d. Thick to thin, straight to curved, 
long arcuate sculpted base. Stenotrema bar- 
bigerum. 

State e. Thin, curved, long arcuate smooth 
base. Inflectarius inflectus, I. subpalliatus, I. 
magazinensis, I. smithi. 

State e'. Same as state e, but with arched 
medial buttress and thickened central exten- 
sion. Inflectarius magazinensis, I. smithi. 

State e". Same as state e', but with medial 
buttress very large and central extension very 
long, clubbed, and sculpted. /. smithi. 

State f. Absent. All other examined spe- 
cies. 

Figure 1 1 maps the informative character 
states onto the phylogenetic hypothesis pre- 
viously shown in Figure 1 . There are 12 infor- 
mative character states, of which seven ap- 
pear homoplastic in Figure 11. One of these 
homoplasies (state 7e') is spurious, because 
it actually resolves a trichotomy by providing 
a new synapomorphy uniting Inflectarius 
magazinensis with /. smithi. Two of the ho- 
moplasies microsculpturally enhance the 
general shell convergences between X. fos- 
teri and P. laevior (state 1 c) and between N. 
major and M. normalis (state 3a). The ho- 
moplasies in state 6a involve multiple origins 
of wrinkles in the periostracum, yet this state 
is still informative in uniting the three species 
of Xolotrema, for example. The loss of peri- 
ostracal hairs in /. ferrissi (state 7f) accompa- 
nied its great evolutionary shifts in shell size 
and in ecology (Emberton, 1991b). The re- 
maining homoplasies involve the apertural 
coatings of AI. profunda and M. normalis 
(state If) and the spiral microstriae of most 
triodopsins and AI. profunda (state 2a). Thus, 
six of the twelve informative microstructural 
character states (lb. Id, 2b, 5a, 7e, 7e') sup- 



POLYGYRID SHELL MICROSCULPTURE 



83 



port the phylogenetic hypothesis without ho- 
moplasy. 



DISCUSSION 

Although this study is preliminary, it offers 
hope that the shells of polygyrids — and by 
inference the shells of other land-snail fami- 
lies — are not entirely useless for hypothesiz- 
ing phylogeny. Thus, although polygyrid 
gross shell morphology and ontogeny 
yielded virtually no phylogenetic resolution 
among subgenera (Emberton, 1995b: fig. 16), 
polygyrid microsculptural shell morphology 
has so far yielded potential new informative 
characters with a 50% (6 of 12) "success 
rate" in resolving a previously, robustly hy- 
pothesized phylogeny (this paper: Fig. 11). 
Verifying these characters will require much 
more work, which will also undoubtedly dis- 
close many new microsculptural characters. 

Of the new characters tentatively pro- 
posed, some are particularly intriguing. Spiral 
microstriae, a possible new synapomorphy 
for the tribe Triodopsini, may finally provide a 
means of distinguishing fossils of this tribe 
from those of the iteratively convergent thbe 
Mesodontini (Emberton, 1994a, 1995b). The 
homology among adult-shell spiral pustules, 
spiral pustular ridges, and spiral grooves pro- 
posed here for Patera is extreme, but is in line 
with Pilsbry's (1940: 576) remark that, in the 
embryonic sculpture of many polygyrid spe- 
cies, "many stages in the transition from 
striae to granules are found." 

The remarkable radiation of periostracal 
hairs in Inflectarius was unsuspected and 
raises questions concerning the function of 
such complex hairs as in /. smithi. Likewise, 
the great discrepancy in hair microstructure 
between Stenotrema maxillatum and S. bar- 
bigerum raises many questions regarding the 
origin(s), radiation(s), functions, and phyloge- 
netic-information content of shell hairs in this 
large genus, almost all species of which have 
hairs. 

Periostracal hair-like or scale-like pro- 
cesses on the shell (Kaiser, 1966; see 
Saleuddin & Petit, 1983, on the periostracum) 
have evolved numerous times within the Po- 
lygyridae. They evolved at least three times 
within the Mesodontini alone, for example 
(Emberton, 1991a). Polygyrid shell hairs date 
back to at least the Miocene (Roth & Ember- 
ton, 1994) and display a wonderful variation 
in size, disposition, microsculpture, and fra- 



gility (Pilsbry, 1940; Solem, 1974: fig. 6; Em- 
berton, 1995b: fig. 4; this paper). 

Thus, polygyrids provide an excellent sys- 
tem for testing functional hypotheses regard- 
ing shell hairs. These hypotheses, none of 
which have been tested, include the func- 
tions (a) "to repel moist particles" (Solem, 
1974) and prevent wet leaves from adhering 
to the shell; (b) to defend against predators 
(Webb, 1950); and (c) to camouflage the shell 
by trapping soil and debris (Pilsbry, 1940: p. 
761). Apparently, different functions are 
served by different shaped hairs (Fig. 9), but 
this remains to be investigated. 

The different types of coatings on the ap- 
erture found here were a marked addition to 
previous SEM discoveries (Solem, 1970, 
1 972, 1 973; Solem & Lebryk, 1 976); the com- 
positions and functions of these coatings are 
unknown. There is evidence, however, that 
coatings may change with shell age. Thus, in 
Daedalochila auriformis, the parietal denticle 
of a non-gerontic adult had a coating identi- 
cal to that of Inflectarius inflectus (Fig. 3F), 
but the parietal dentical of a gerontic D. au- 
riformis was smooth and coating-free (Solem 
& Lebryk, 1976: figs. 33, 37). Although such 
coatings could possibly be preservational ar- 
tifacts in the form of dried mucous films, they 
do not in any way resemble the mucous films 
illustrated in Solem (1970: figs. 12-15). 
Clearly, future studies should fully assess in- 
traspecific variation in aperture microsculp- 
ture if such characters are to have any value 
for phylogenetics. 

Two proposed microsculptural conver- 
gences are remarkable for actually enhanc- 
ing gross shell-form convergences in sympa- 
try. Thus, Neohelix major and Mesodon 
normalis (Emberton, 1994b, 1995a, 1995b: 
fig. 1) also converge in their body-whorl spiral 
cords, and Xo lot re ma foster! and Patera lae- 
wor (Emberton, 1995b: figs. 1, 17) also con- 
verge in their uncoated apertures, revealing 
nearly identical microsculptures of crystalline 
projections from a smooth surface. Both 
these pairs of species are separable, how- 
ever, by other shell microsculptural features 
(Fig. 11). 

For future work, particularly on periostracal 
hairs, it can be recommended to use very 
fresh, live-collected material, preferably alco- 
hol preserved. Dried periostracum can be 
quite brittle, and thus no unbroken hairs 
could be found on the shells of /. inflectarius 
or X. denotata selected for this study. On the 
positive side, however, most microstructures 



84 



EMBERTON 



2a 



lb 
3a 



1c 



6a I 



2bl 



6a 



If 2a 



6a 



Neohelix major 

N. den titer a 

Xo lot rema foster/ 

X. denotata 

X. obstricta 

Trio dop sis fa IIa X 

Vespericola columbiana 

Allogona profunda 

Stenotrema max ilia turn 

Stenotrema barbigerum 

Patera perigrapta 



5a 



1c 



■= P. laevior 
P. appressa 
P. sargentiana 
Inflectarius subpalliatus 



Id 



6a I 
7e| 



=^111= /. ferrissi 
7f 

--^^ I. magazinensis 
le' 
J"— — r= /• smithi 



II 7e' 

I' /. in flee tu s 



Appalachina say ana 



"— ^- Mesodon normalis 
If 
3a 

FIG. 1 1 . Map of tentatively proposed microsculptural character states onto the cladogram of Fig. 1 . See text 
for definitions. 



POLYGYRID SHELL MICROSCULPTURE 



85 



remained intact despite the relatively great 
age of many of the specimens used in this 
study. Regarding apertural microsculpture, 
Solem & Lebryk (1976) found clear, uneroded 
details in subfossil pupillid shells. In prepar- 
ing shells for SEM, Solem (1970) cautioned 
against the difficulty or impossibility of re- 
moving dried mucous films, which are not a 
problem in alcohol-preserved material. No 
such films were noticed in this study. 

Polygyrid shell-form evolution may be 
unique for the sympatry of its convergences 
(Emberton, 1995a, 1995b), but certainly not 
for the convergences themselves, which par- 
allel shell-form evolution in other stylommato- 
phoran groups, such as the Helicidae sensu 
lato, Bradybaenidae, and Camaenidae (Zilch, 
1959-1960). For phylogenetics of these and 
other land-snail groups, it can be hypothe- 
sized (from this study and Emberton, 1995b) 
that shell micromorphology is much more in- 
formative than shell macromorphology. 



ACKNOWLEDGEMENTS 

Supported by Academy of Natural Sci- 
ences discretionary funds and National Sci- 
ence Foundation grant DEB-9201060. Caryl 
Hesterman assisted on the SEM and printed 
and mounted the SEM photographs. 



LITERATURE CITED 

ASAMI, T., 1988, Temporal segregation of two 
sympatric species of land snails. Venus, 47: 
278-297. 

ASAMI, T., 1993, Divergence of activity pattern in 
coexisting species of land snails. Malacologia, 
35: 399-406. 

BROOKS, D. R. & D. A. MCLENNAN, 1991, Phy- 
togeny, ecology and behavior: a research pro- 
gram in comparative biology. University of Chi- 
cago Press, Chicago, 434 pp. 

EMBERTON, K. C, 1988, The genitalic, allozymic, 
and conchological evolution of the eastern North 
American Thodopsinae (Gastropoda: Pulmo- 
nata: Polygyhdae). Malacologia. 28: 159-273. 

EMBERTON, K. C, 1991a, The genitalic, allozymic, 
and conchological evolution of the Mesodontini, 
trib. nov. (Gastropoda: Pulmonata: Polygyridae). 
Malacologia. 33: 71-178. 

EMBERTON, K. C, 1991b, Ecology of a shell con- 
vergence between subfamilies of polygyrid land 
snails. Biological Journal of the Linnean Society, 
44: 105-120. 

EMBERTON, K. C, 1994a, Polygyrid land-snail 



phylogeny: external sperm exchange, early 
North American biogeography, iterative shell ev- 
olution. Biological Journal of the Linnean Soci- 
ety, 52: 1-31. 

EMBERTON, K. C, 1994b, Partitioning a morphol- 
ogy among its controlling factors. Biological 
Journal of the Linnean Society, 53: 353-369. 

EMBERTON, K. C, 1995a, Sympatric convergence 
and environmental correlation between two 
land-snail species. Evolution. 49: 469-475. 

EMBERTON, K. C, 1995b, When shells do not tell: 
145 million years of evolution in North America's 
polygyrid land snails, with a revision and conser- 
vation priorities. Malacologia. 37: 69-110. 

KAISER, P., 1966, Bau, Entwicklung und Regener- 
ation des Haarkleides von Trichia hispida (Lin- 
naeus) zugleich ein Beispiel für einfache Muster- 
bildung im Tierreich. Archiv für Molluskenkunde, 
95: 111-122. 

PILSBRY, H. A., 1940, Land Mollusca of North 
America (North of Mexico), Volume 1, Part 2. 
Monographs of the Academy of Natural Sci- 
ences, Philadelphia, 3: 575-994. 

ROTH, B. & K. С EMBERTON, 1994, "Extralimital" 
land mollusks (Gastropoda) from the Deep River 
Formation: evidence for mesic medial Tertiary 
climate. Proceedings of the Academy of Natural 
Sciences of Philadelphia, 145: 93-106. 

SALEUDDIN, A. S. M. & H. P. PETIT, 1 983, Mode of 
formation and structure of the periostracum. Pp. 

199-234 in: A. SALEUDDIN S. M. & K. M. WILBUR, 

eds.. The Mollusca, volume 4. Physiology, part 1. 
Academic Press, New York. 

SOLEM, A., 1970, Malacological applications of 
scanning electron microscopy. 1. Introduction 
and shell surface features. The Veliger, 12: 394- 
400. 

SOLEM, A., 1972, Microarmature and barriers in 
the apertures of land snails. The Veliger, 15: 81- 
87. 

SOLEM, A., 1973, Apertural barriers in Pacific Is- 
land land snails of the families Endodontidae and 
Charopidae. The Veliger. 15: 300-306. 

SOLEM, A., 1974, The shell makers: introducing 
mollusks. New York: Wiley-lnterscience, 289 pp. 

SOLEM, A., 1976, Comments on eastern North 
American Polygyridae. Nautilus, 90: 25-36. 

SOLEM, A. & S. LEBRYK, 1976, Apertural micro- 
projection size correlations in pupillid and po- 
lygyrid land snails. The Veliger, 19: 115-120. 

VAGVOLGYI, J., 1968, Systematics and evolution 
of the genus Tnodopsis (Mollusca: Pulmonata: 
Polygyridae). Bulletin of the Museum of Compar- 
ative Zoology, 136: 145-254. 

WAITE, J. H., 1983, Ouinone-tanned scleropro- 
teins. Pp. 467-504 in: P. W. Hochachka, ed.. 
The Mollusca, volume 1. Metabolic biochemistry 
and molecular biomechanics. Academic Press, 
New York. 

WATABE, N., 1988, Shell structure. Pp. 69-104 in: 
E. R. Trueman & M. R. Clarke, eds., The Mol- 
lusca, volume 11. Form and function. Academic 
Press, New York. 



86 



EMBERTON 



WEBB, G. R., 1950, Shell-spinules as defense 
structures. Nautilus, 63: 107. 

WEBB, G. R., 1980, The hybridization of the snails 
Xolotrema denotata and X. carolinensis in the 
laboratory (Pulmonta, Polygyridae, Triodopsi- 
nae). Gastropodia, 2: 1-2. 

WILBUR, K. M. & A. S. M. SALEUDDIN, 1983, Shell 
formation. Pp. 235-287 in: A. S. M. Saleuddin & 
K. M. Wilbur, eds., The Mollusca, volume 4, 
Physiology, part 1. Acadennic Press, New York. 



WILEY, E. O., 1981, Phylogenetics: the theory and 
practice of phylogenetic systematics. John Wiley 
and Sons, New York, 439 pp. 

ZILCH, A., 1959-1960, Gastropoda. Teil 2. Euthy- 
neura. Band 6. Pp. 1-834 in: O. H. Schindewolf, 
ed., Handbuch der Palaozoologie. Gebrüder 
Bornträger, Berlin. 



Revised Ms. accepted 10 January 1996 



MALACOLOGIA, 1996, 38(1-2): 87-102 

ANATOMY AND SYSTEMATICS OF BUCCINANOPS GRADATUS (DESMAYES, 

1844) AND BUCCINANOPS MONILIFERUS (KIENER, 1834) (NEOGASTROPODA, 

MURICOIDEA) FROM THE SOUTHEASTERN COAST OF BRAZIL 

Luiz Ricardo L. Simone 

Museu de Zoología da Universidade de Sao Paulo Caixa Postal 7172-01064-970 

Sao Paulo, Brazil 

ABSTRACT 

A morphological revision of species of the genus Buccinanops, endemic to South America, 
begins with the description of B. gradatus and 6. moniliferus. In an attempt to obtain data to 
resolve systematic problems from the family to the specific level in this group, a detailed 
anatomical description of the head-foot, pallia! organs, digestive system, including odontopho- 
ral muscles, and genital system are given. These animals are blind, have a vestigial valve of 
Leiblein and, in the case of S. moniliferus, there is sexual dimorphism, males being about half 
of the size of females. 



INTRODUCTION 

The systematic concepts on the South 
American neogastropod species Buccinan- 
ops gradatus (Deshayes, 1 844) and S. monil- 
iferus (Kiener, 1834) are confused at almost 
every level. 

There a controversy about their at the fam- 
ily-level placement; some authors (e.g., Ab- 
bott & Dance, 1983; Rios, 1994) have consid- 
ered these species to be Nassahidae, 
whereas others (e.g., Rios, 1985) have in- 
cluded the genus in the Buccinidae. Mean- 
while, Ponder (1973: 325) noted that anatom- 
ical characters for the separation of these 
two families have not been established. 

At the generic level, B. moniliferus was 
considered to belong to Dorsanum Gray, 
1847, by several authors (e.g., Carcelles & 
Parodiz, 1939; Rios, 1994) and Buccinanops 
Orbigny, 1841, by Calvo (1987) and Rios 
(1985), based on radular characters, and by 
Pastorino (1993) because of differences from 
the type species of the genus Dorsanum, D. 
miran (Bruguiére). Both species — B. monil- 
iferus and B. gradatus — were included in the 
South African genus Bullia Gray, 1834, in 
early literature (e.g.. Reeve, 1846) and by Ab- 
bott & Dance (1983) and Alimón (1990). 

At the species level, B. moniliferus in con- 
trast, is well established, due its distinctive 
conchological characters. Buccinanops gra- 
datus, on the other hand, is a variable spe- 
cies with several synonyms according to 
some authors (e.g., Rios, 1 975), whereas oth- 



ers consider these synonyms to be valid spe- 
cies. No convincing arguments have been 
given to support either position. The available 
species-group names are: B. lamarckii 
(Kiener, 1834), B. cochlidius (Dillwyn, 1817), 
B. uruguayensis (Pilsbry, 1897), and ß. de- 
formis (King & Broderip, 1832). Aggravating 
these problems is the fact that neither B. gra- 
datus or B. moniliferus were described with a 
specific type locality. 

A step in solving these systematic prob- 
lems may be an anatomical analysis of well 
localized and identified specimens. This pa- 
per includes anatomical descriptions of Buc- 
cinanops moniliferus and ß. gradatus, which 
will serve as the basis for future compari- 
sons. 

The specific names are changed to mas- 
culine gender herein, following Art. 30(a)ii of 
the ICZN Code for generic names ending in- 
ops. 



MATERIAL AND METHODS 

Part of the studied material belonged to 
Museu de Zoología da Universidade de Sao 
Paulo (MZUSP) and part was collected by ot- 
ter trawl by fishermen in Praia Grande, Sao 
Paulo, Brazil, and has been deposited in 
MZUSP, fixed in 70% ethanol. 

The anatomical dissections were made us- 
ing standard techniques. Some anatomical 
parts, such as the genital organs and anterior 
region of the digestive system, were dehy- 



87 



88 



SIMONE 



drated in ethanol series, stained in carmine, 
cleared and fixed in creosote. Radulae and 
protoconch were also examined using SEM 
in the Laboratorio de Microscopía Eletrônica 
do Instituto de Biociências da USP. All draw- 
ings were made with the aid of a camera lu- 
cida. 

The musculature of the odontophore was 
studied by means of dissection of three 
specimens of each species preserved with 
an extended proboscis. The jugal muscles 
and peroral muscles are not described in de- 
tail. For the most part, the muscles are 
named according to the terminology of Wils- 
mann (1942). 

The synonymic list of B. gradatus is not 
given here, because studies on possible syn- 
onymy are continuing. 



Abbreviations 



aa 


anterior aorta 


af 


anterior furrow of the foot 


ag 


albumen gland 


an 


siphoned anus 


ao 


anterior oesophagus 


au 


auricle 


bm 


mantle border 


eg 


capsule gland 


cm 


columellar muscle 


cv 


ctenidial vein 


da 


duct to anterior digestive gland 


dl 


duct of the gland of Leiblein 


dp 


duct to posterior digestive gland 


ft 


foot 


ga 


inner gland near anus 


gd 


gonopericardial duct 


gi 


gill 


gk 


glandular part of the kidney 


gl 


gland of Leiblein 


go 


gonad 


gp 


female genital pore 


in 


intestine 


Ic 


left cartilage 


m1 to 




m14 


odontophoral muscles 


me 


mid oesophagus 


mf 


muscular fibers 


mo 


mouth 


ne 


nephrostome 


ng 


nephridial gland 


nn 


nuchal node 


nr 


nerve ring 


nv 


nephridial vessel 


nw 


nephridial wall 


od 


odontophore 



oe posterior oesophagus 

OS osphradium 

pa posterior aorta 

pc pericardio walls 

pe penis 

pn proboscis nerve 

pp penial papilla 

ps penial sinuses 

pv proximal vertex of the cartilages 

pw proboscis wall 

ra radula 

re right cartilage 

rm radular membrane 

rn radular nucleus 

rt rectum 

sd salivary gland duct 

sg salivary gland 

si siphon 

St stomach 

sv seminal vesicle 

te tentacles 

ty gastric typhlosoles 

uc union between both cartilages 

va vas deferens aperture to palliai cavity 

vd vas deferens 

ve ventricle 

vl valve of Leiblein 

vm visceral mass 

vp villous part of the kidney 

Buccinanops gradatus (Deshayes, 1 844) 
(Figs. 1-3, 6, 9, 12-30) 

Diagnosis 

Shell generally homogeneous beige in 
color; subsutural carina generally present, 
without spines. Osphradium about 2/3 of gill 
length. Radular rachidian teeth with eight 
well-spaced cusps that are heterogeneous in 
size; two well-developed median cusps on 
lateral teeth. Odontophore with only one pair 
of "m9" muscles; and with double radular 
protractor muscle (ml 4). Both stomach 
typhlosoles longitudinal. Penis long, with a 
well-developed papilla. Female genital pore 
papillate, surrounded by two folds. 

Description 

Shell: Up to 60 mm in length, homogeneous 
beige, with up to 8 convex whorls (Figs. 1-3). 
Protoconch of about 2.5 whorls; first whorl 
glassy-smooth, semi-spherical; others with 
strong axial ridges. Limit between proto- 
conch and teleoconch not conspicuous. Te- 
leoconch to 5 whorls; two first whorls with 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONILIFERUS 



89 




FIGS. 1-11. Shells and radulae: 1, 2, dorsal and frontal view of female of Buccinanops gradatus (MZUSP 
28079), scale = 10 mm; 3, frontal view of a male of B. gradatus (MZUSP 28078), scale = 10 mm; 4, 5, dorsal 
and frontal view of two specimens of S. moniliferus (MZUSP 28191), scale = 10 mm; 6, radula of B. 
gradatus, SEM, scale = 0.2 mm; 7, profile of the protoconch and first teleoconch whorl of B. moniliferus, 
SEM, scale = 1 mm; 8, the same in apical view, scale = 0.5 mm; 9, detail of Fig. 6, scale = 0.1 mm; 1 0, radula 
of B. moniliferus, SEM, scale = 0.2 mm; 1 1 , dorsal view of a specimen of B. moniliferus without developed 
spines on the subsutural carina (MZUSP 28181), scale = 10 mm. 



axial ridges, similar to those of protoconch, 
gradually disappearing on subsequent 
whorls. Subsutural carina generally present, 
low, rounded (Fig. 3). Periostracum very thin, 
dark-brown, lost on body whorl. Aperture el- 
liptic; outer lip arched, sometimes notched 
by carina; inner lip concave, covered by thin 
callus. Canal short, broad, bordered exter- 
nally by well-developed carina. 



There is considerable shell variation; the 
most common form is shown in Figures 1-3, 
but specimens with a shorter or taller spire 
are common. The subsutural carina Is lacking 
in some specimens, resembling B. cochlidius 
and B. uruguayensis, whereas others have a 
well-developed carina and resemble B. de- 
formis. The lot MZUSP 28080 has specimens 
showing both conditions. Several specimens 



90 



SIMONE 




FIGS. 12-14. Buccinanops gradatus anatomy: 12, visceral mass and palliai cavity organs of a female, scale 
= 5 mm; 13, frontal view of a male head-foot, mantle removed, scale = 5 mm; 14, transversal section of the 
mid region of the anterior oesophagus, scale = 1 mm. 



have the spire without a carina and a well- 
developed carina on the last whorl. No nota- 
ble shell differences between males and fe- 
males were found. 

Operculum: Corneous, ovate-unguiculate, 
with terminal nucleus, partially sealing shell 
aperture. Muscle scar elliptic, near inner bor- 
der. Operculum deformation very common. 

Head-Foot: Homogeneous pale-beige in 
color. Head somewhat projecting. Tentacles 
long, lateral, without eyes (Figs. 13, 15). Foot 



large, with furrow along anterior edge for an- 
terior pedal glands (Fig. 13: af). Males with 
large penis, behind right tentacle (Fig. 13). 
Small posterior metapodial tentacle present. 

Mantle Border: Simple, slightly thick. Siphon 
developed, with smooth borders (Figs. 12, 
13). Without pigment or with scanty dark 
spots. 

Mantle Cavity: About one whorl In length 
(Fig. 12). Osphradium bipectinate, narrow, 
long (about 2/3 of the total gill length), with 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONIUFERUS 



91 



several leaflets on both sides. Gill monopec- 
tinate, somewhat elliptic, with numerous tri- 
angular, low leaflets. Hypobranchial gland a 
thin glandular mass covering mantle between 
gill and rectum. 

Circulatory and Excretory Systerns: Heart at 
posterior-right side of palliai cavity (Fig. 12); 
auricle fusiform; ventricle spherical, very- 
large. Anterior and posterior aorta as normal 
for caenogastropods (Fig. 12). Kidney large, 
behind posterior-left side of palliai cavity 
(Fig. 12). Internally, kidney with villous and 
glandular parts (Fig. 28); nephridial gland 
covering pericardial wall of kidney lumen (Fig. 
28: ng). Nephrostome a slit surrounded by 
muscle fibers, in mid region of kidney wall at 
posterior end of palliai cavity (Figs. 12, 28: 
ne). 

Digestive System: Proboscis pleurembolic, 
thick-muscular (Fig. 15), very-long (about 
same length as shell when extended). Buccal 
mass about half length of proboscis. Probos- 
cis opening surrounded by thick muscular 
sphincter. Mouth a vertical slit at distal end 
of proboscis. Proboscis structure (Fig. 15): 
odontophore in anterior half attached to inner 
ventral wall; muscles at posterior odonto- 
phore edge running posteriorly and attaching 
to ventral half of inner proboscis surface up 
to ventral face of rhynchodeal cavity (Fig. 15: 
mf). Aorta, paralleled in both sides by a pair 
of nerves, runs in mid line of ventral surface 
covering these muscles; oesophagus lies 
above all these structures, connected to pro- 
boscis by tridimentional net of thin muscle 
fibers. 

Odontophore muscles (Figs. 20-26): (ml) 
dorsal jugal muscles — origin: outer-proximal 
dorsal wall of odontophore; insertion: inner- 
dorsal peribuccal wall; (m2) transversal mus- 
cle — uniting dorsally outer edge of both car- 
tilages, involving dorsally other muscles of 
odontophore; (m3) pair of lateral retractor 
muscles of radula (retractor of pharynx) — or- 
igin: in dorsal region of foot, running attached 
to inner-ventral wall of proboscis; insertion: 
proximal vertex of each cartilage (pv); (m4) 
medial retractor muscle of radula — origin: 
partly in dorsal region of foot, between m3 
muscles, running attached to inner-ventral 
wall of proboscis also between the m3, and 
partly in ventral face of proximal vertex of 
each cartilage (mid tensor); insertion: mainly 
on ventral edge of radula; (m5) dorsal pro- 
tractor muscle of radula — origin: joined with 



medial retractor muscle (m4), bifurcating in 
mid region of odontophore; insertion: dor- 
sally on both sides of peroral wall; (m6) pair of 
tensor lateral muscles — lying on anterior half 
of the outer edge of both cartilages; (m7) pair 
of small muscles — origin: on outer edge of 
cartilages just proximal to m6 origin; inser- 
tion: on peribuccal wall just proximal to m5 
insertion; (m8) small muscle — origin: on outer 
edges of both cartilages just proximal to m7 
origin; near mid region of muscle both 
branches unite for a short distance and after 
they separate inserting on ventral region of 
peribuccal wall near mid line; (m9) pair of 
small muscles — origin: outer edge of carti- 
lages just proximal to the m8 origin; insertion: 
dorsal edge of radula; (mIO) pair of large lat- 
eral tensor muscles of radula — origin: dorsal 
face of proximal vertex of cartilages; inser- 
tion: mainly lateral-dorsal margin of radula, 
uniting with medial retractor muscle (m4) for 
about 2/3 of their length (Fig. 22); (mil) 
horizontal muscle — uniting ventrally inner 
edge of both cartilages; (ml 2) ventral jugal 
muscles — origin: outer-proximal-dorsal wall 
of odontophore; insertion: inner-ventral peri- 
buccal wall, some muscular fibers more de- 
veloped (Fig. 24); (m13) pair of large tensor 
ventral muscles — origin: ventral face of pos- 
terior vertex of each cartilage just at medial 
retractor muscle (m4) origin; insertion: ventral 
edge of radula; (ml 4) pair of small protractor 
muscles of radula — origin: mixed with medial 
retractor muscle (m4), distinguishable only 
near horizontal muscle (mil); insertion: ven- 
tral edge of radula between tensor ventral 
muscle (m13) insertion. 

Radula (Figs. 6, 9)— Rachidian flattened, 
arched, with eight well-spaced cusps that are 
smaller towards outer edges; lateral teeth ob- 
lique, each with four cusps, marginal cusp 
largest, middle two cusps smallest. 

Anterior oesophagus lumen "X" in section 
(Fig. 14), covered by net of radial and oblique 
muscles uniting oesophagus with inner sur- 
face of proboscis wall; salivary gland ducts 
running on either side of oesophagus (Fig. 
14: sd) and discharging into peroral chamber. 

Valve of Leiblein vestigial, anterior to nerve 
ring, poorly visible on outer surface of oe- 
sophagus (Fig. 15), marked internally by sud- 
denly change of inner longitudinal folds, 
forming a low valve (Fig. 16: vl). 

Two salivary glands clustered around 
nerve ring (Fig. 15: sg), their ducts on outer 
side of nerve ring, running to posterior half of 
anterior oesophagus and within muscular net 



92 



SIMONE 




FIGS. 15-19. Buccinanops gradatus anatomy: 15, anterior region of the digestive system and proboscis 
opened longitudinally along dorsal mid line and head mid line, scale = 1 mm; 1 6, detail of the region of mid 
oesophagus opened longitudinally, scale = 2 mm; 1 7, stomach in ventral view, scale = 2 mm; 1 8, the same 
opened longitudinally, scale = 2 mm; 19, detail of the anal region, terminal region of the rectum partially 
opened longitudinally to expose an inner gland, scale = 1 mm. 



of anterior half of anterior oesophagus (Fig. 
14). No accessory salivary glands present. 

Entire oesophagus a long, somewhat uni- 
fornn, thick muscular walled tube without 
crop (Figs. 1 5-1 7); internally with several lon- 
gitudinal folds (Fig. 16). Mid-oesophagus 
very short (Fig. 16: me). Gland of Leiblein 



long, thin, with short duct, running posteriorly 
close to posterior oesophagus (Figs. 15, 16), 
yellowish-brown in color. 

Stomach well developed; walls somewhat 
thick; two ducts to digestive glands, one dor- 
sal near insertion of oesophagus, the other 
ventral near opening to intestine (Fig. 17). In- 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONILIFERUS 



93 




FIGS. 20-23. Odontophore of Buccinanops gradatus: 20-23, successive dissection in dorsal view. 20, only 
proboscis wall opened and oesophagus removed. 21 , the outer layer of muscles removed. 22, second layer 
of muscles removed exposing the inner muscles. 23, most muscles removed to show the cartilages. 



94 



SIMONE 




FIGS. 24-26. Successive dissection in ventral view, proboscis entirely removed. 24, outer view of the 
odontophore; 25, same with first layer of muscles removed. 26, second layer of muscles removed, hori- 
zontal muscle (m1 1) opened longitudinally exposing a part of the dorsal muscles. Scales = 2 mm. 



ner stomach surface rich in folds; opposite to 
digestive gland ducts these folds converging; 
two ventral typhlosoles present between 
these ducts (Fig. 18). 

Intestine thin-walled, lying anteriorly to kid- 
ney (Fig. 28), in right side of pallia! cavity in 
males or close left side of pallia! oviduct in 
females (Fig. 12). Anus siphoned, slight back 
of mantle border (Figs. 12, 19: an). Internally, 
a sub-terminal glandular mass present (Fig. 
19). 



Genital System: Male. Testis in visceral mass 
near columella; vas deferens initially a narrow 
duct. Seminal vesicles greatly convoluted 
just postehor to palliai cavity (Fig. 29). Gono- 
perlcaldial duct present, small (Fig. 29: gd). In 



floor of palliai cavity, vas deferens a closed 
duct thickened by prostate gland, except in 
its posterior extremity, where there is a small 
aperture (Fig. 29: va). Penis narrow, long, in- 
ternally with a convoluted vas deferens and 
well-developed sinuses on both sides (Fig. 
29); tip rounded, with a small pointed papilla 
on right side in which the vas deferens opens 
(Figs. 13, 29). 

Female. Ovary in visceral mass mixed with 
digestive gland, mainly concentrated near 
columella. Oviduct extremely narrow, on right 
side of palliai cavity, with well-developed al- 
bumen-capsule glands (Fig. 27), both difficult 
to differentiate from one another, occupying 
about half of palliai cavity length (Fig. 27). 
Vestibule thin-walled, somewhat long. Fe- 
male genital aperture papillated, surrounded 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONILIFERUS 



95 




FIGS. 27-29. Buccinanops gradatus anatomy; 27, detail of the right side of the palliai cavity in inner view 
to show the palliai oviduct; 28, kidney chamber opened ventrally; 29, detail of a cleared penis and right side 
of the palliai cavity floor showing the mid and anterior regions of the male reproductive system. Scales = 
2 mm. 



by tvuo folds, right fold thin, left fold larger 
broad (Fig. 30), sited in the posterior-right 
side of anus (Figs. 12, 27, 30). 

Habitat. Sandy-nnud bottoms, from 5 to 25 
m depth. For data on posture and capsules, 
see Penchaszadeh (1973). 

Range. With certainty from Rio de Janeiro 
to Sao Paulo coast; specimens from other 
regions still under study. 



Examined specimens. BRAZIL, otter trawl. 
Rio de Janeiro: MZUSP 28184, 1 specimen, 
Cabo de Sao Thome (1 1/ii/1969); MZUSP 
15295, 1 specimen, Atafona, Sao Joâo da 
Barra. Sao Paulo: Ubatuba: MZUSP 28080, 8 
specimens, Itaquá Beach (i/1971, Mon- 
touchet col.); MZUSP 28185, 2 specimens, 
Cabras Is., Anchieta Is. (28/vi/1978); MZUSP 
28186, 2 specimens, Anchieta Is. (4/viii/1960, 



96 



SIMONE 



an 




FIG. 30. Buccinanops gradatus anatomy: detail of 
Fig. 27. showing the female genital pore, scale = 1 
mm. 



Clarimundo col.); MZUSP 28081. 1 speci- 
men, same (21/v/1979). Baixada Santista: 
MZUSP 28183, 5 specimens, from Barra de 
Santos to Guarujá (vii/1969. Instituto de 
Pesca col.); MZUSP 28192, 20 specimens. 
Pereque Beach, Guaruja (6/vi/1985); MZUSP 
28193, 8 specimens, Santos Bay (2/ix/1970); 
MZUSP 28187, 2 specimens, from Moela Is. 
to Ponta Pereque (17/v/1962, Clarimundo 
col.); MZUSP 28188, 5 specimens. Goes 
Beach, Guarujá (17/viii/1970, Colella col.); 
MZUSP 28189, 2 specimens, from Barra de 
Santos to Farol da Moela (vii/1969. Instituto 
de Pesca col.); MZUSP 28078, 28079, 
28082, 25 specimens, Barra de Santos (21/ 
ix/1970. Colella col.); MZUSP 28183. 14 
specimens. Moela Is., 15 m deep (17/v/1962, 
Clarimundo col.). Praia Grande, off Bo- 
queirâo Beach: MZUSP 28190. 12 speci- 
mens (i/1994, Simone col.); MZUSP 27319, 
12 specimens (10/1/1990, Simone col.). Total: 
1 17 specimens. 

Buccinanops monlliferus (Kiener, 1 834) 

(Figs. 4, 5, 7, 8, 10, 11. 31-42) 

Buccinum moniliferum Kiener, 1834: 2, pl. 3, 
fig. 8; Reeve, 1842: 234, pl. 268, fig. 4; 
Deshayes, in Lamarck, 1844: 191. 

Bullía armata Gray, 1854: 26; Reeve, 1846: pl. 
1 . fig. 2 [Hab. ?]; Adams & Adams, 1 858: 
113; Kobelt, 1877: 290; Tryon, 1882: 14, 



pl. 6, figs. 82, 83; Paetel. 1888: 116; 
Morretes, 1949: 98. 

Buccinanops moniliferum: Orbigny, 1845: 
199; Rios, 1985: 103, pl. 35, fig. 456; 
Calvo, 1987: 143, fig. 122 (radula); Pas- 
tonno, 1993: 160-165, figs. 1-3 (radula). 

Buccinum (Buccinanops) manilifcrum: (err.) 
Orbigny, 1846: 434. 

Buccinanops cochlidium: Gray, 1854: 40 
[non Dillwyn, 1817). 

Buccinum armatum: Küster. 1858: 90, pl. 15, 
fig. 20. 

Bullía (Buccinanops) moniliferum: Chenu, 
1859; 160, fig. 750; Abbott & Dance, 
1983; 117 (fig.). 

Dorsanum armatum: Cossmann, 1901; 218. 

Dorsanum moniliferum: Carcelles & Parodiz 
1939: 747, figs. 1, 2; Carcelles, 1944 
249; Rios, 1970: 92, pl. 28; Rios, 1975 
95, pl. 27, fig. 398; Penchaszadeh, 
1971a (posture and capsules); Pen- 
chaszadeh, 1971b: 480; Figueiras & Si- 
cardi, 1972: 179, pl. 13, fig. 176; Mar- 
torelli, 1991 (parasite); Castellanos, 
1994: 89, 96, fig. 31-4 (capsule); Rios, 
1994: 130, pl. 41, fig. 557. 

Diagnosis 

Shell generally with two spiral purple 
bands on each whorl; subsutural carina with 
regular-spaced spines. Osphradium about 
half of gill length. Radular rachidian teeth with 
nine cusps of homogeneous size; generally 
only one mid cusp on lateral teeth. Odonto- 
phore with two or three pairs of m9 muscles; 
and with single radular protractor muscle 
(ml 4). Typhlosoles of stomach perpendicular 
one another. Male about half of female size. 
Penis somewhat long, with a small node in 
tip. Female genital pore single, bordered by 
bulged thick muscular walls. 

Description 

Shell: Up to 50 mm in length, with up to 
seven convex whorls, generally pale-cream, 
with two broad spiral bands brown-purple on 
each whorl (Figs 4, 5). Protoconch of about 
2.5 whorls; first whorl smooth, others with 
strong axial ridges and subsutural furrow 
(Figs. 7, 8). Limit between protoconch and 
teleoconch not conspicuous. First two whorls 
of teleoconch with axial ridges, similar to 
those of protoconch, disappearing on subse- 
quent whorls. Subsutural carina present, with 
short, uniform, somewhat spaced, triangular 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONILIFERUS 



97 



spines turned distally and dorsally (Figs. 4, 5). 
Periostracum very thin, black, lost on body 
whorl. Aperture elliptic; outer lip arched, 
notched by carina, with a low anal sinus; in- 
ner lip concave, covered by a thin white cal- 
lus. Canal short, broad, bordered externally 
by well-developed carina. 

Shell variation is low compared with the 
preceding species, as shown in Figure 42. In 
rare specimens, absence of spines in subsu- 
tural carina were observed (e.g., MZUSP 
281 81 ; Fig. 1 1). In other specimens, there is a 
homogeneous purple color, in contrast to the 
common two spiral bands per whorl. Albino 
and sinistral specimens are also known. 

Operculum: Gorneus, ovate- unguiculate, 
with terminal nucleus, partially sealing shell 
aperture; muscle scar elliptic near inner bor- 
der. Operculum deformation very common, 
rarely lost. One female (MZUSP 28151) has 
two well-developed opercula side by side, in 
the normal position. 

Head-Foot: Homogeneous pale-beige in 
color. Head somewhat projecting; tentacles 
long, lateral, without eyes. Foot large, with 
furrow along anterior edge for anterior pedal 
glands (Figs. 31, 32: af). Small posterior 
metapodial tentacle present. 

Mantle Border: Simple, slightly thick (Fig. 
33). Siphon developed, with smooth borders 
(Figs. 31, 33), pigmented by dark-brown ir- 
regular spots. Siphon with well-developed 
muscular root. 

Mantle Cavity: About 1.5 whorls in length 
(Fig. 33). Osphradium bipectinate, narrow, 
long, with several short leaflets in both sides, 
lying along about half of gill length. Gill 
monopectinate, elliptic, long, with numerous 
triangular, low leaflets. Hypobranchial gland 
thin, poorly developed, near and anterior to 
anal region. 

Circulatory and Excretory Systems: As de- 
scribed for preceding species (Fig. 33). 

Digestive System: Radular rachidian teeth 
with nine cusps that are somewhat uniform in 
size and close one-another (Fig. 1 0); marginal 
teeth with only one mid cusp (Fig. 10) or 
rarely with two smaller cusps, the inner cusp 
longer. 

In odontophore, most part of muscles and 
other structures very similar to that of B. gra- 
datus, except that in B. moniliferus the small 



muscles originating on the outer edge of car- 
tilages and inserting on the dorsal edge of 
radula (called "m9" in preceding species) are 
multiple and vary from 2 to 3 successive sim- 
ilar-sized pairs. The small muscle that origi- 
nates with medial retractor muscle of radula 
(ml 4) and inserts on ventral edge of radula 
near mid line is single (Fig. 34: ml 4a) and has 
a part of its fibers inserting ventrally in 
beribuccal wall also near mid line (Fig. 34: 
m14b). 

Stomach (Fig. 37) similar to that of preced- 
ing species, except one typhlosoles is longi- 
tudinal, from the oesophagus to the intestine 
(fig. 38: tyl), whereas the other is transversal, 
lying duct to posterior digestive gland (Fig. 
38: ty2). 

All other studied characters of the diges- 
tive system of B. moniliferus are closely sim- 
ilar to preceding species (Figs. 35, 36), in- 
cluding characters of valve and gland of 
Leiblein (Fig. 35) and anus (Fig. 33). 

Genital System: Male. Testis in visceral mass 
near columella. Seminal vesicles greatly con- 
voluted (Fig. 39) just posterior to palliai cav- 
ity. A small aperture where vas deferens en- 
ters floor of palliai cavity (Fig. 39: va); 
remainder closed, thickened by prostate 
gland (Fig. 40). Penis narrow, long (Fig. 32), 
internally a convoluted vas deferens and two 
well-developed sinuses in both sides (Fig. 40: 
ps); rounded tip with a very small vesicle on 
right side in which vas deferens opens (Fig. 
40). 

Female. Ovary in columellar side of visceral 
mass not mixed with digestive gland (Fig. 33). 
Oviduct very narrow. Albumen and capsule 
glands well developed, difficult to distinguish 
from one another, occupying about half of 
palliai cavity length (Fig. 41); vestibule thin- 
walled, very short. Female genital aperture 
small, bordered by bulged thick muscular 
walls (Fig. 41). Two specimens (39.8 mm and 
33.0 mm length, MZUSP 28176) have a small 
node where penis occurs in males (Fig. 31). 

Sexual dimorphism. Mature males notably 
smaller than mature females. Mature male 
length: 20.3-27.8-36.8 mm. Mature female 
length: 31.0-43.5-49.5. 

Habitat. Sandy-mud bottoms, from 5 to 25 
m depth. 

Range. From Rio de Janeiro, Brazil, to San 
Matías Gulf, Argentina. 

Examined specimens. BRAZIL, otter trawl. 
Rio de Janeiro: MZUSP 19591, 1 specimen, 
sta. IV, 22''06'S, 41 "04'W, off Cabo de Sao 



98 



SIMONE 




FIGS. 31-33. Buccinanops moniliferus anatomy: 31, frontal view of a female (MZUSP 28176) with nuchal 
node, mantle partially opened, scale = 2 mm; 32, frontal view of the head-foot of a male, mantle and siphon 
removed, scale = 2 mm; 33, visceral mass and palliai cavity organs of a female in inner view, scale = 1 mm. 



Thome, 16nn(11/ii/1969, "W. Besnard" col.). 
Sao Paulo: off Ubatuba: MZUSP 28124, 17 
specimens, 22°05'50"S, 41 04'12"W, 10 m 
(vii/1991); MZUSP 28125, 19 specimens, 
22'^06'07"S, 41 '04'08"W, 13 m (3/1992). 
Baixada Santista: MZUSP 28179, 1 speci- 
men, from Barra de Santos to Guarujá (vii/ 
1969, Instituto de Pesca col.); MZUSP 
28181, 2 specimens. Goes Beach, Guarujá 
(17/viii/1970, Golella col.); MZUSP 28084, 10 
specimens, Pereque Beach, Guarujá (6/vi/ 



1985). Praia Grande, off Boqueiráo Beach: 
MZUSP 28191, 11 specimens (i/1994, Si- 
mone col.); MZUSP 26865, 2 specimens (10/ 
xi/1 970, Ribas col.); MZUSP 281 75, 20 spec- 
imens (i/1994, Simone col.); MZUSP 28176, 
46 specimens (i/1990, Simone col); MZUSP 
27320, 2 specimens (IO/i/1990, Simone col.); 
MZUSP 28177, 5 specimens (summer, 1994, 
Simone col); MZUSP 28151, 86 specimens 
(xii, 1991, Simone col); MZUSP 28152, 17 
specimens (summer, 1987, Simone col); 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONILIFERUS 

34 m3 



99 




FIGS. 34-38. Buccinanops moniliferus anatomy: 34, odontophoral muscles exposed by dissection (com- 
pare with the fig. 25), scale = 2 mm; 35, left view of the anterior region of the digestive system, scale = 2 
mm; 36, region of the proboscis opened longitudinally in dorsal mid line, scale = 8 mm; 37, stomach in 
ventral view, scale = 2 mm; 38, the same opened longitudinally, scale = 2 mm. 



MZUSP 28153, 18 зреаглепз (i/1987, Si- 
mone col). Itanhaénn: MZUSP 281 80, 1 spec- 
imen, Prainha Beach (18/11/1970, Vaz col.); 



MZUSP 28178, 13 specimens, Prainha 
Beach (18/ii/1970, Vaz col.). Total: 271 spec- 
imens. 



100 



SIMONE 




FIGS. 39-41 . Buccinanops moniliferus anatomy: 39, mid region of the male genital duct In ventral view; 40, 
dorsal view of a cleared penis and right side of the palliai floor with a detail of a section in mid region of the 
penis; 41, detail of the right side of the palliai cavity showing the palliai oviduct, scales: 1 mm. 



DISCUSSION 

Buccinanops gradatus differs anatomically 
from B. moniliferus in having: (1) the osphra- 
dium proportionally longer; (2) the rachidian 
teeth of the radula with fewer, more widely 
spaced cusp that are less uniform in size; (3) 
the lateral teeth with two well-developed in- 
termediate cusps (S. moniliferus generally 
has only one or two smaller cusps, see fig. 1 
of Pastorino, 1993); (4) only one pair of 
odontophoral "m9" muscles; (5) double 
"ml 4" muscle; (6) stomach with the typhlo- 
soles parallel one another; (7) absence of sex- 
ual dimorphism — in B. moniliferus, the mature 
male is smaller than the mature female; (8) 



penis proportionally longer, and with the pa- 
pilla more developed; (9) the female genital 
pore in form of a small papilla surrounded by 
two folds, whereas in B. moniliferus, it 
bulges, has thick walls, and is without papilla. 
Analysis of the anatomical characters of 
other species of Buccinanops is necessary 
for any systematic interpretation of the 
above-cited differences. Probably, based on 
number and degree of differences, both 
studied species may belong to close, but dif- 
ferent genera. Buccinanops moniliferus is 
maintained in the genus Buccinanops, but 
the generic attribution may change in future. 
Pastorino (1993) gave a strong argument in 
favor to the separation of this species from 



ANATOMY OF BUCCINANOPS GRADATUS AND B. MONILIFERUS 



101 



••9 



• • • *• 






• • • • 



, D ••• 

•• •• ••• 

Ч-. • 



•ODD a 



10 20 30 40 50 

FIG. 42. Graph length ■ width based on 203 specimens of Buccinanops moniliferus. 156 females (dark 
circles) and 48 males (squares). 



the genus Dorsanum, based on differences 
from its type species, D. miran from Africa 
(Alimón, 1990). 

The radula of B. moniliferus is similar to 
that of B. cochlidlus (see Pastorino, 1993: 
162, figs. 4-6), but differs in having more 
cusps on the rachidian, and its largest cusp 
on the right, not the left side. 

Both studied species have some morpho- 
logical similarity to the European Bucclnum 
undatum (Buccinidae) and Nassarius reticu- 
lata (Nassanidae) (Fretter & Graham, 1962: 
214-5, figs. 1 1 5-1 1 6), differing mainly in hav- 



ing tentacles without eyes and by reduction of 
the valve of Leiblein. The odontophoral mus- 
cles of both studied species are similar to 
those of Bucclnum undatum (see Wilsmann, 
1942), differring mainly in having: (1) the hor- 
izontal muscle (mil) shorter, (2) the dorsal 
protractor of the radula (m5) thinner, (3) the 
lateral tensor muscle (mIO) stronger, and (4) 
the minor dorsal muscles (m7, 8 and 9) dif- 
ferently arranged. No studies with this level of 
detail of the Nassarius odontophore exists. 
The ongoing comparative study on the ar- 
rangement of the odontophoral muscles of 



102 



SIMONE 



other Buccinanops and Nassahus species 
may add data useful in family-level distinc- 
tions. 



ACKNOWLEDGMENTS 

I am very grateful by Dr. Guido Pastorino, 
Faculdad de Ciencias Naturales y Museo, 
Universidad Nacional de La Plata, Argentina, 
for detailed reading. My special thank also for 
the anonymous referees for detailed revision 
and criticisms. 



LITERATURE CITED 

ABBOTT, R. T. & S. P. DANCE, 1983, Compendium 
of seashells. E. Dutton. New York. 411 pp. 

ADAMS, H. & A. ADAMS, 1858, The genera of Re- 
cent Molluscs; arranged according to their orga- 
nization, vol I. London. 

ALLMON, W. D., 1990, Review of the Bullía group 
(Gastropoda: Nassariidae), with comments on its 
evolution, biogeography, and phylogeny. Bulle- 
tins of American Paleontology, 99(335): 1-179. 

CALVO, I. S., 1 987, Radulas de gastropodes marin- 
hos brasilelros. Editora da Fundaçao Univer- 
sidade do Rio Grande. Rio Grande. 201 pp. 

CARCELLES, A., 1944, Catálogo de los moluscos 
marinos de Puerto Quequén (República Argen- 
tina). Revista del Museo de la Plata, (Nueva Se- 
rie), Zoología 3(23): 233-309. 

CARCELLES, A. & J. J., PARODIZ, 1939, Dorsan- 
inae argentinas y uruguayas. Physis, 17: 745- 
769. 

CASTELLANOS, Z. J. A., 1994, Moluscos, in Los 
invertebrados. Ediciones Estudio Sigma S.R.L. 
Buenos Aires 3(1): 1-206. 

CHENU, J. C, 1859, Manuel de conchyliologie et 
de paléontologie conchyliologique, vol. 1. Paris. 
508 pp. 

COSSMANN, A. E. M., 1901, Essais de paléocon- 
chologie comparée, Vol. 4, 293 Paris, pp., 10 pis. 

FIGUEIRAS, A. & O. E. SICARDI, 1972, Catálogo 
de los moluscos marinos del Uruguay. Comuni- 
caciones de la Sociedad Malacológlca del Uru- 
guay, 3(22): 169-188. 

FRETTER, V. & A. GRAHAM, 1962, British proso- 
branch molluscs. Ray Society. Lond., 755 pp. 

GRAY, J. е., 1854, List of shells of South America 
in the collection in the British Museum. Collected 
and described by M. A. d'Orbigny, in the "Voy- 
age dans l'Amérique Méridionale." London, i-iii 
+ 89 pp. 

KIENER, L. C, 1834, Species général et iconogra- 
phie des coquilles vivantes, famille des purpu- 
riféres. Vol. 2, 417 pp. Paris. 

KOBELT, W., 1877, Catalog der Gattung Bullía 
Gray, 1877. Jahrbücher der Deutschen Malako- 
zoologischen Gesellschaft, 4(4): 289-294. 

KÜSTER, H. C, 1858, Die Gattungen Buccinum, 



Purpura, Concholepas und Monoceros. In: Mar- 
tini und Chemnitz. Systematiches Conchylien- 
Cabinet. Nürnberg 3(1 A): 1-229 pp., 35 pis. 

LÛ.MARCK, J. P. В. M., 1844, Histoire naturelle des 
animaux sans vertèbres, 2nd ed.. Vol. 10, 639 
pp. Paris. 

MARTORELLI, S. R., 1991, Primera cita de una 
cercarla tricocerca parasita de Dorsanum monil- 
iferum (Mollusca; Buccinidae) para el Atlántico 
sudoccidental. Apontes al conocimiento de su 
ciclo de vida. Neotropica, 37(97): 57-65. 

MORRETES, F. L., 1949, Ensaio de catálogo dos 
moluscos do Brasil. Arquivos do Museu Para- 
naense, 7: 5-216. 

ORBIGNY, A., 1835-1846, Voyage dans l'Amé- 
rique Méridionale, Tome Cinquième, 3 partie: 
Mollusques. Paris, i-xliii + 758 pp., 85 pis. 

PAETEL, F., 1888, Catalog der Conchylien-Samm- 
lung von F. Paetel. Berlin. 639 pp. 

PASTORINO, G., 1993, The taxonomic status of 
Buccinanops d'Orbigny, 1841 (Gastropoda: 
Nassariidae). Veliger, 36: 160-165. 

PENCHASZADEH, P. E., 1971a. Observaciones 
sobre la reproducción y ecología de Dorsanum 
moniliferum (Valenciennes, 1834) (Gastropoda, 
Buccinidae) en la region de Mar del Plata. Neo- 
tropica. 17(53): 49-54. 

PENCHASZADEH, P. E., 1971b. Aspectos de la 
embriogenesis de algunos gasterópodos del 
genero Buccinanops d'Orbigny, 1841 (Gas- 
tropoda, Prosobranchiata, Buccinidae). Physis, 
30(81): 475-482. 

PENCHASZADEH, P. E., 1973, Nuevas observa- 
ciones sobre la reproducción de Buccinanops 
gradatum (Deshayes, 1844) (Gastropoda, Proso- 
branchiata, Dorsaninae). Physis, 32(84): 15-18. 

PONDER, W. F., 1973, The ongin and evolution of 
the Neogastropoda. Malacologia, 12: 295-338. 

REEVE, L. A., 1841-1842, Conchologia systemat- 
ica, or complete system of conchology: in which 
the lepades and conchiferous Mollusca are de- 
scribed, 2 vols. London. 

REEVE, L. A., 1845-1846, Monograph of the genus 
Bullia. Conchologia Iconica, Vol. 3: 4 pis. Lon- 
don. 

RIOS, E. C, 1970, Coastal Brazilian seashells. Fun- 
daçao Cidade do Rio Grande, MORG, Rio 
Grande, 255 pp., 60 pis. 

RIOS, E. C, 1975, Brazilian manne molusks ico- 
nography. Fundaçao Universidade do Rio 
Grande, Rio Grande, 331 pp., 91 pis. 

RIOS E. C, 1985, Seashells of Brazil. Fundaçao 
Universidade do Rio Grande, Rio Grande, 328 
pp., 102 pis. 

RIOS, E. C, 1994, Seashells of Brazil, second edi- 
tion. Fundaçao Universidade do Rio Grande, 368 
pp., 113 pis. 

TRYON, G. W., 1882, Manual of conchology, Vol. 
4. Philadelphia. 

WILSMANN, T., 1942, Der Pharynx von Buccinum 
undatum. Zoologische Jahrbücher, Abteilung für 
Anatomie und Ontogenie der Tiere, 61(1): 1-48. 

Revised Ms. accepted 16 January 1996 



MALACOLOGIA, 1996, 38(1-2): 103-142 

SYSTEMATICS, BIOGEOGRAPHY AND EXTINCTION OF CHIONINE BIVALVES 
(BIVALVIA: VENERIDAE) IN TROPICAL AMERICA: EARLY OLIGOCENE-RECENT 

Peter D. Roopnarine 

Department of Biology, Southeast Missouri State University, Cape Girardeau, 

Missouri 63701 U.S.A. 

ABSTRACT 

The genus Chione ranges from the Early Oligocène of the tropical western Atlantic to the 
Recent of the tropical western Atlantic and eastern Pacific. The genus is generally considered 
to comprise the subgenera Chione s.S., Chionista, Chiionopsis, lliochione, Lirophora, Pan- 
chione, and Puberella. A phylogenetic analysis of the subgenera suggests that the genus is 
paraphyletic, because its current definition excludes the related genera Anomalocardia, Pro- 
tothaca and Timoclea. This problem is resolved by converting the taxonomic classification to a 
phylogenetic one, constituting Puberella, Chionista, Chionopsis, lliochione and Lirophora, 
newly elevated to full generic status. 

The first genera to appear in the fossil record are Lirophora and Puberella in the Early 
Oligocène of the western Atlantic. The genus Chionopsis appears next, in the Late Oligocène 
of the western Atlantic. Chione and Panchione first occur in the Early Miocene, also of the 
western Atlantic. By the Early Pliocene, Chione, Chionopsis, Lirophora and Panchione were 
present in the eastern Pacific. During the Pliocene, diversities and distributions of these genera 
changed dramatically. In the Early-middle Pliocene (5.2-2.5 Ma), both Lirophora and Panchione 
suffered severe extinction in the western Atlantic. All supraspecific taxa present in the western 
Atlantic suffered elevated extinction during the Late Pliocene. Conversely, diversities increased 
in the eastern Pacific during the Pliocene, added to by the evolution of Chionista, and lliochione. 
The net result is much higher Recent diversity in the eastern Pacific compared to the western 
Atlantic. 

Key words: Chione, extinction, Neogene, systematics. 



INTRODUCTION 

The reasons for diversification and extinc- 
tion of a lineage are the results of complex 
interactions between phenotype and envi- 
ronment. While phylogenetic and morpholog- 
ical analyses can lead to an understanding of 
patterns of phenotypic change, the impetus 
for such change must come from the organ- 
ism's biotic and abiotic environments. The 
potential therefore exists for understanding 
the evolution of a clade, given knowledge of 
the relationships among its inclusive taxa, 
and an understanding of the ecological con- 
ditions under which the clade evolved. 

This approach is utilized in a phylogenetic 
and paleoecological study of the Cenozoic 
tropical American marine bivalve genus 
Chione (von Mühlfeld, 1811). Species as- 
signed to this genus are among the most 
abundant members of many shallow soft- 
bottom communities throughout Atlantic and 
Pacific tropical America (North, Central and 
South), and have been so since at least the 
Early Miocene. The genus has remained re- 



stricted to tropical America since its first ap- 
pearance in the Early Oligocène, and occurs 
today in the Atlantic from South Carolina to 
Brazil, and in the Pacific from southern Cali- 
fornia to northern Peru. The purpose of this 
study is to present an analysis of phylogeny, 
and a taxonomic revision, of Chione subgen- 
era using shell morphological characters. A 
biogeographic history, focusing on the late 
Neogene, is also presented using fossil oc- 
currences. The geographic range of the ge- 
nus is within a region that was subject to sev- 
eral significant geological and océanographie 
changes during the late Neogene, notably 
the uplift of the Isthmus of Panama, and the 
initiation of Northern Hemisphere cooling. 
These events are likely to have affected the 
diversities and distributions of Chione taxa. 
Many other molluscan taxa were affected 
adversely (Stanley, 1986; Vermeij & Petuch, 
1986; Vermeij, 1993), namely by extinction in 
the Western Atlantic and subsequent restric- 
tion to the tropical Eastern Pacific. 

The phylogenetic analysis was performed 
at the subgeneric level, because while mono- 



103 



104 



ROOPNARINE 



phyly of Chione is questionable (as dis- 
cussed below), species within each Chione 
subgenus share well-defined, discrete char- 
acter states with little interspecific variation 
(see, for example, Roopnahne, 1995). The 
ease with which species can be assigned to 
subgenera using the combination of the 
states present within them argues strongly 
for the monophyly of the currently defined 
subgenera. The phylogenetic analysis facili- 
tated a taxonomic revision of Chione, be- 
cause the current taxonomy is not consistent 
with the analysis. The new taxonomy will be 
more accurate phylogenetically, and there- 
fore offer a more accurate depiction of rela- 
tionships among the subgenera. 

Analyzing the distribution and diversity of 
Chione subgenera revealed patterns of 
change in geographic ranges. Placing the 
geological and geographic ranges of the sub- 
genera in a phylogenetic framework allowed 
changes in biogeography and diversity to be 
applied to understanding patterns of diversi- 
fication and extinction. Diversity and extinc- 
tion of Chione taxa were documented by ex- 
amining the available records of all Chione 
species of the late Neogene in both the West- 
ern Atlantic and the Eastern Pacific. These 
data will be used to test various hypotheses 
constructed to explain the late Neogene mol- 
luscan extinctions in the tropical Western At- 
lantic. Prior to hypothesis testing, though, it 
will be essential to present some information 
explaining the situation of tropical America 
during the Neogene. 

Late Neogene Extinctions in 
Tropical America 

Tropical America during the late Neogene 
was the site of dramatic océanographie 
changes, coupled with changes in faunal 
composition and diversity. The two most 
commonly cited causes of diversity changes 
are decreasing temperature (Stanley, 1986) 
and decline or disruption of planktonic pro- 
ductivity levels (Vermeij, 1978, 1987; Alimón 
et al., 1993; see also Jones & Hasson, 1985). 
Both mechanisms, however, are plausibly 
linked to two geological events, the uplift of 
the Isthmus of Panama, and the initiation of 
intense Northern Hemisphere cooling. 

Uplift of the Isthmus of Panama may have 
begun as early as the early Middle Miocene 
(>15.2 mya) (Duque-Caro, 1990). Termination 
of surficial circulation and final closure prob- 
ably did not occur, however, until the Early 



Pliocene (approximately 3.5 mya) (Coates et 
al., 1992). Separation of the oceans has long 
been associated with Plio-Pleistocene 
changes in faunal composition in both the 
tropical Western Atlantic and Eastern Pacific 
(Vermeij & Petuch, 1986; Jackson et al., 
1993). 

Severe late Neogene Northern Hemisphere 
cooling and glaciation, documented in the 
deep-sea stable isotopic record (Shackleton 
& Hall, 1985; Krantz, 1991), the stratigraphie 
record of North Atlantic coastal deposits 
(Krantz, 1991; Cronin, 1993) and microfaunas 
(Cronin, 1991), began about 2.5-2.4 Ma 
(Shackleton & Hall, 1985; Stanley, 1986; Cro- 
nin, 1991, 1993; Alimón et al., 1993). This 
event, along with later Late Pliocene-Early 
Pleistocene cooling events, has been linked 
to molluscan extinctions in the southeastern 
United States and the Caribbean region 
(Stanley, 1986). Weyl (1968) hypothesized 
that closure of the Panama seaway would 
have intensified the northward flowing Gulf 
Stream current. The resultant change in the 
heat distribution and precipitation in the 
Northern Hemisphere may have initiated 
the buildup of Arctic ice and the subsequent 
glaciations. Therefore, the two events may be 
linked. 

Initial observations of the changes in faunal 
composition were interpreted as a large- 
scale decline in the diversity of the Western 
Atlantic molluscan fauna, compared to the 
Eastern Pacific (Woodring, 1966; Vermeij, 
1978; Vermeij & Petuch, 1986). Recent com- 
pilations of Pliocene to Recent molluscan 
faunas in the Caribbean and the southeast 
United States have modified this interpreta- 
tion by noting that the Recent molluscan fau- 
nas of the tropical Western Atlantic are actu- 
ally as diverse, or more than their Pliocene 
counterparts (Alimón et al., 1993; Jackson, 
1993). Therefore, extinction must have been 
matched by speciation and/or invasions (Ver- 
meij, 1993; Vermeij & Rosenberg, 1994). 
Such observations do not, however, negate 
the fact that numerous molluscan taxa that 
were once widespread in the tropical West- 
ern Atlantic are now restricted to the Eastern 
Pacific (Vermeij & Petuch, 1986), or to a few 
"refuges" in the Caribbean Sea (Petuch, 
1982). 

A necessary step in describing the late 
Neogene biological history of this region is to 
explore plausible reasons why some mollus- 
can taxa suffered declines in diversity in the 
Western Atlantic during the Late Pliocene, 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



105 



while others were either unaffected or in- 
creased in diversity in the Western Atlantic or 
Eastern Pacific. One possible approach to 
this problem is to document the changing di- 
versities within a phylogenetic context or 
framework, and observe if the nature of di- 
versity change is uniform throughout the 
clade. Analysis of geographic and character 
distributions with respect to diversity change 
may subsequently yield some clues to 
whether extinction was phylogenetically ran- 
dom, or if it followed a character-based pat- 
tern. 

A basis for selective extinction is not nec- 
essarily recognizable in the results of a phy- 
logenetic analysis if the pattern of extinction 
is related to a non-phylogenetic property 
(strictly speaking, for example, geographic 
distribution). This reason deserves attention 
because it implies that the basis for extinc- 
tion could be an environmental perturbation 
with effects that transcend patterns of evo- 
lutionary relationship. The two causes of ex- 
tinction suggested above, cooling and de- 
cline of planktonic productivity, could belong 
potentially to this class of environmental per- 
turbation. To demonstrate the action of this 
type of variable, one could document rele- 
vant physical (geological) evidence, or more 
important, predict and test the effect(s) of 
these agents on the organisms under study. 
Tests of cooling and declining planktonic 
productivity as agents of selection are ex- 
plored more fully in the following sections. 

Cooling and Extinction 

Stanley (1984) argues that the decimation 
of tropical bivalves relative to cooler-water 
species would be an effect of a cooling- 
based extinction mechanism. This hypothe- 
sis requires that an established latitudinal 
temperature gradient exists at the time of 
cooling. The gradient would, prior to cooling, 
have determined the development of recog- 
nizable temperate and tropical faunal prov- 
inces. Temperate biotas can migrate equa- 
torward during times of global cooling, but 
tropical biotas have no refuges. Moreover, 
the provinces would presumably be compos- 
ites of stenothermal and eurythermal taxa, 
the differential survivals of which indicate the 
occurrence of a cooling-related extinction. 
For example, Stanley (1984, 1986) suggests 
that the relatively higher survival of euryther- 
mal bivalves from the Upper Pliocene Pine- 
crest (Upper) Beds of Florida is the result 




FIG. 1 . Geographic ranges of Recent species of 
Chione. 



of a middle-Late Pliocene cooling event. The 
recognition of the provinces is crucial to the 
hypothesis, for only then can comparisons 
be made between provinces located in dif- 
ferent thermal regimes. Such a comparison 
could possibly be made in the late Neogene 
between the tropical Pacific and Atlantic Ga- 
tunian province, and the more northerly, sub- 
tropical Caloosahatchian province (Petuch, 
1982; Jones & Hasson, 1985) (Fig. 1). Rec- 
ognizing the provinces has depended tradi- 
tionally on the identification of resident and 
endemic species and their ranges (Woodring, 
1966; Petuch, 1982), and inferences about 
the thermal tolerances of these species 
(Stanley, 1986). 

The middle Pliocene (-4.0-2.5 Ma), 
though not a formal stratigraphie subdivision, 
is recognizable by indications of a period of 
global warming following Late Miocene and 
Early Pliocene cooling events (summarized 
in Cronin, 1991, and Krantz, 1991). The end 
of the interval may be marked by a major 
regression associated with Northern Hemi- 
sphere glaciation, and quite evident in 
coastal deposits of the North Atlantic, for ex- 
ample in the southeast United States (Krantz, 
1 991 ). Despite data indicating that the middle 



106 



ROOPNARINE 



Pliocene was a time of relative global 
warmth, the Gatunian and Caloosahatchian 
provinces were unexpectedly cooler than 
other contemporaneous regions, and were 
only as warm or slightly cooler than they are 
today (Jones & Hasson, 1985; Cronin, 1991). 
Gatunian waters were, however, absolutely 
warmer than Caloosahatchian during this 
time, by as much as 10°C in the winter, and 
6°C in the summer (Cronin, 1991). 

Cronin's (1991) observations do not sup- 
port Stanley's (1986) cooling hypothesis, be- 
cause they imply a reduced latitudinal ther- 
mal gradient in the Western Atlantic (Cronin, 
1991) at a time when the hypothesis would 
require a highly developed gradient. For ex- 
ample, during the middle Pliocene, summer 
temperatures south of Cape Hatteras, North 
Carolina (Fig. 1) were approximately 2.6°C 
warmer than temperatures north of the cape. 
The temperature difference today is 8.6°C 
(Cronin, 1991). Similarly, summer tempera- 
tures differed between the southern Carib- 
bean and Florida during the middle Pliocene 
by approximately 3.8-C, but as much as 
4.3°C today (Cronin, 1991). 

The cooler temperatures in the Caloosa- 
hatchian and Atlantic Gatunian regions dur- 
ing a time of global warmth seem anomalous, 
but it has been suggested that they could be 
explained by the existence of upwelling 
zones (Vermeij, 1978; Jones & Hasson, 1985; 
Vermeij & Petuch, 1986; Cronin, 1991). Ac- 
cording to Cronin's (1991) data, winter tem- 
peratures in the Caribbean were an average 
1.1 °C cooler than today, and in the Caloosa- 
hatchian an average of 1.2°C cooler. The 
cooling hypothesis, however, does not re- 
quire that middle Pliocene temperatures be 
absolutely warmer than Pleistocene and Re- 
cent temperatures, simply that temperatures 
declined during the time of extinction. 

A simple, faunally-based test of the ex- 
planatory power of the cooling hypothesis 
would be to document the geographic distri- 
butions of Chione subclades in the Gatunian 
and Caloosahatchian provinces during the 
Pliocene and post-Pliocene. The test de- 
pends on the ability to distinguish between 
the provinces on the basis of faunal compo- 
sition (in this case excluding Chione taxa). 
Petuch (1982) demonstrated that the two 
provinces maintained high levels of ende- 
mism, with respect to gastropods, through- 
out the Early Miocene to Pleistocene. Cata- 
strophic cooling would, in accordance with 
Stanley's (1984) argument, affect taxa that 



were restricted to the more southerly Gatu- 
nian province more severely than taxa resi- 
dent in the Caloosahatchian province. In ad- 
dition, one should expect to observe the 
migration of Caloosahatchian taxa to Gatu- 
nian waters as temperatures in the Gatunian 
approached pre-cooling temperatures of the 
Caloosahatchian. 

Modern species of Chione are at least par- 
tially resthcted in their latitudinal distributions 
by temperature. No species range beyond 
tropical and sub-tropical regions (Fig. 2), and 
it can be demonstrated that species' ranges 
have changed in response to changing ther- 
mal regimes. For example, the species 
Chione undatella (Sowerby, 1835), is abun- 
dant in the Upper Pleistocene, interglacial 
Millerton Formation of northern California. 
The northernmost extent of the species today 
is southern California, approximately 640 km 
to the south. The modern boundaries of mol- 
luscan faunal provinces are frequently asso- 
ciated with temperature gradients (Vermeij, 
1978), but also with barriers to circulation, for 
example Cape Hatteras. 

Productivity and Extinction 

Unlike the cooling hypothesis, a hypothe- 
sis of extinction resulting from declining 
planktonic productivity is not based on any 
well-documented geological event. Evidence 
for higher levels of planktonic productivity in 
the tropical Western Atlantic during the 
Pliocene is mostly indirect. Petuch (1982) 
identified Recent "primary" relict molluscan 
faunas of Mio-Pliocene systematic affinities, 
off the coast of Venezuela, in cool, upwelling 
areas of high planktonic productivity. He in- 
ferred that the survival of these communities 
in upwelling areas was indicative of the wide- 
spread occurrence of these areas in the Mi- 
ocene, when the Caribbean was dominated 
by such communities. The presence of sys- 
tematically related "secondary" relict com- 
munities off the Yucatan Peninsula and 
Roatan also support this contention (Petuch, 
1982). Stanley (1986) hypothesized that a 
zone of strong upwelling off the southern ex- 
tent of the Florida peninsula during the Early 
Pliocene could explain faunal differences be- 
tween the Caloosahatchian province and the 
nearby Bahamanian fauna, an idea sup- 
ported by Cronin's (1991) ostracod paleo- 
temperature data. More recently, Jones & 
Alimón (1995) have suggested the occur- 
rence of extensive upwelling off the west 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



107 



.0/-0.4 

YTTX Caloosahatchian 
*^ш Province 



Gatunian 

Province 




FIG. 2. Petuch's (1982) provincial configuration of tropical America during the middle Pliocene. Arrows 
indicate that the provinces extend further in those directions. Numbers indicate winter and summer pa- 
leotemperature estimates (Cronin, 1991). Map adapted from Jones & Hasson (1985). 



coast of Florida during the Early-middle 
Pliocene, as evidenced by the ontogenetic 
stable isotopic records of various molluscan 
taxa. Cronin's (1991) data suggest strongly 
the existence of extensive upwelling systems 
off the southeast United States and Central 
America, during the Pliocene. Cronin (1991) 
(see also Raymo et al., 1990) speculated that 
during the middle Pliocene, overall global 
warmth and reduced amounts of sea ice 
resulted in a relatively stronger Gulf Stream 
gyre, which in turn caused upwelling along 
much of the east coast of North America. 
Moreover, given the westward direction of 
surficial flow from the Atlantic to the Pacif- 
ic, during the Early and middle Pliocene, 
through what is now Panama, it is conceiv- 
able that extensive upwelling existed in sur- 
rounding shelf areas (Stanley, 1986). Alimón 
et al. (1993) proposed that declines in pro- 
ductivity were responsible for the decimation 
of the rich Pliocene molluscan fauna of 
southern Florida. 



Declining levels of planktonic productivity 
would result in different levels of extinction 
among different biogeographic regions if the 
ecological crisis was more severe in one re- 
gion than in another, or altogether absent in 
one or more regions. This, for example, would 
be the case if the Caribbean in the Early 
Pliocene was a region with extensive up- 
welling, and has suffered a subsequent de- 
cline in the number of upwelling areas and 
therefore planktonic productivity (Stanley, 
1986; Cronin, 1991). Distinguishing the action 
of this agent of extinction from others is dif- 
ficult, because there is no reliable, direct 
method for assessing levels of productivity in 
fossil communities. Changes in planktonic 
productivity could, however, affect commu- 
nity composition. For example, more produc- 
tive habitats may comprise more species than 
less productive ones (Brown, 1975; Pianka, 
1975), perhaps because higher production 
provides more resources for successful utili- 
zation by more species (MacArthur, 1965). 



108 



ROOPNARINE 



This type of observation or speculation how- 
ever, cannot generally be measured in fossil 
communities, because many organisms and 
materials (for example, organic detritus) that 
form the resource bases of ecological com- 
munities have low fossilization potentials 
(Dodd & Stanton, 1990). 

One possibly relevant parameter that can 
be measured easily in the fossil record is 
the distribution of body sizes within a com- 
munity. Under conditions of declining sus- 
pended food supply, a reasonable prediction 
is that all suspension-feeding bivalves would 
be affected. A further prediction is the dis- 
proportionately higher extinction of large 
bodied bivalves. In support of this prediction 
are observations that: (1) animal body-size 
scales negatively with population size (Pe- 
ters, 1983), making larger animals more sus- 
ceptible to extinction (Vermeij, 1987) under 
conditions of ecological duress, (2) the rate of 
nutrient intake scales positively with body 
size, and (3) conditions of limited resource 
supply may lead to communities dominated 
by small bodied taxa (Thiel, 1975; Peters, 
1983). Vermeij (1978) also points out that 
molluscs inhabiting cool, upwelling waters off 
Venezuela are significantly larger than con- 
specifics elsewhere in the Caribbean. He at- 
tributes the differences to higher juvenile 
growth rates afforded molluscs in more pro- 
ductive waters, but evidence in support of 
this remains circumstantial. 

This paper presents analyses that demon- 
strate that the Chione subgenera collectively 
reveal a pattern of greater extinction in the 
late Neogene of the Western Atlantic relative 
to the Eastern Pacific, coupled with higher 
rates of origination in the Eastern Pacific. 
Phylogenetic revision of the complex, how- 
ever, further demonstrates that the extinc- 
tions were not distributed equally among the 
subgenera, but seemingly at random. Several 
closely related subgenera were affected 
more severely than others, while other sub- 
genera had higher rates of origination in the 
Eastern Pacific. 

A summary of the current taxonomic infor- 
mation of Chione is presented in the follow- 
ing section. A compilation of the geological 
histories of the subgenera is also presented. 
Hypotheses of extinction caused by cooling 
or declining planktonic productivity were 
tested by first examining paleobiogeographic 
distribution patterns, and then evaluating 
body size distributions of late Neogene spe- 
cies. 



Taxonomic Status of Chione 

There are approximately 67 described ex- 
tinct and extant species of Chione (although 
some of these are undoubtedly synonyms) 
(Table 1). The species have been assigned 
traditionally, correctly or not. to one of seven 
subgenera composing the genus-C/i/one s.S., 
Chionopsis Olsson, Puberella Fischer-Piette, 
Lirophora Conrad, Panchione Olsson, llio- 
chione Olsson, Timoclea Brown, and Chion- 
ista Keen. While ambiguity of taxonomic ranks 
confuses the relationships among subgenera, 
this paper will demonstrate that the genus is 
paraphyletic. The paraphyletic nature of the 
genus can be resolved by revising Chione, 
and changing the definition and composition 
of the genus. 

The above subgenera are all recognized as 
members of the subfamily Chioninae Frizzell, 
1936, but various authors have argued that at 
least some subgenera are different enough 
from the definition of Chione (see Palmer, 
1927; Olsson, 1961; Keen, 1971, for defini- 
tions of Chione) to deserve status as separate 
genera. Therefore, Olsson (1961) regarded 
Chionopsis as a distinct genus, and both 
Woodring (1982) and Ward (1992) treat Liro- 
phora similarly. The latter authors also con- 
sider Panchione a subgenus of Lirophora, 
contrary to Keen (1969). In addition, it is 
unclear from a morphological perspective 
whether some taxa, such as Chionista, are 
members of the Chione clade (as currently 
defined), or are nested within other taxa, such 
as the related genus Protothaca Dall (Keen, 
1971). From a taxonomic viewpoint, the 
boundaries between Chione and Protothaca 
become obscured when such subgenera as 
Chionista and Leul<oma Römer are consid- 
ered. These problems underscore the value of 
a phylogenetic approach, in which unique 
characteristics of taxa are emphasized less, 
and relationship among them stressed in- 
stead. Recently, Harte (1992a, b) has empha- 
sized, using morphological and immunologi- 
cal distance data, the close relationship of 
Chione to the other chionine genera Merce- 
naria Schumacher, and Anomalocardia Schu- 
macher. Because of these complications, the 
Chione subgenera will be referred to through- 
out the rest of the paper collectively as the 
Chione complex. 

Other Chione subgenera are also problem- 
atic from both taxonomic and biogeographic 
perspectives. For example, Timoclea, if it 
were indeed a subgenus of Chione (see for- 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



109 



TABLE 1 . List of documented and described species assigned originally to Chione subgenera. Abbreviations 
for Localities/Range refer to geological formation, and are explained in Table 3. Asterisked species were 
personally examined by the author. Species that have been reassigned to new higher taxa are noted as such 
under Comments. Sources consulted frequently were Gardner (1926), Palmer (1927), Grant & Gale (1931), 
Parker (1 949), Hertlein & Strong (1 948), Olsson & Harbison (1 953), Olsson (1 961 , 1 964), Perhllat (1 963), Jung 
(1969), Keen (1971), Abbott (1974), Woodring (1982) and Ward (1992). 



Species 


Age 


Localities/Range 


Author(s) 


Comments 


Chione 










araneosa' 


E.— L. 

Pliocene 


EM, BU 


Olsson, 1942 


Described from the Burica 
Fm. (U. Pliocene) (Olsson, 
1942), but also present in 
Esmeraldas Fm. (L, 
Pliocene) 


californiensis' 


L. Pliocene — 


Baja Sur, 


Broderip, 






Recent 


Mexico, L. Pleistocene; 
Pt. Mugu, 
Calif.— Panama, Recent 


1835 




cancellata' 


E. Pliocene — 


Most shallow 


Linnaeus, 






Recent 


water deposits in tropical 
western Atlantic from m. 
Pliocene on. 


1767 




chipolana* 


E. Miocene 


CH 


Dall, 1903 




compta' 


Recent 


G. of Calif. — Bayovar, 
Peru 


Brodenp, 1835 




erosa' 


m.— L. 
Pliocene 


CA, JB, PB 


Dall, 1903 




guatulcoensis' 


Recent 


Pt. Guatuico, 
Mexico — Panama Bay 


Hertlein & Strong, 
1948 




mazyckii' 


Recent 


N. Carolina — Cape San 
Roque, Brazil 


Dall, 1902 




pailasana 


E.— m. 
Pliocene 


Venezuela 


Weisbord, 1964 




primigenia 


E. Pliocene 


Dominican Rep. 


Pilsbry & Johnson, 
1917 


= cancellata? 


quebradillensis 


E. (?) 
Pliocene 


Puerto Rico 


Maury, 1920 


= cancellata? 


santodomingensis 


E. Pliocene 


Dominican Rep. 


Pilsbry & Johnson, 
1917 


= cancellata? 


undatella' 


L. Pliocene — 


SD; s. Calif.— 


Sowerby, 


allisoni (Hertlein & 




Recent 


Paita, Peru, Recent 


1835 


Grant, 1972) = undatella 


Chione 










subimbricata' 


Recent 


G. of Calif.— Paita, Peru 


Sowerby, 1835 




tu mens' 


L. Pliocene — 
Recent 


Baja Calif., Plio.; 

MZ, Pleist.; Baja Calif., 

Pacific coast and G. of 

California 


Vernll, 1870 




vaca' 


E.— m. 
Pliocene 


EM, BU 


Olsson, 1942 




Chionista 










cortezi' 


Recent 


Pacific coast 
Baja Calif., and Gulf of 
Calif, 


Carpenter, 
1864 




fluctifraga' 


L. Pliocene — 


Upper Pliocene, 


Sowerby, 






Recent 


Baja Calif.; s. Calif.— Gulf 
of Calif. 


1853 




Chionopsis 










amathusia' 


Recent 


G. of Calif.— 
Mancora, Peru 


Philippi, 1844 




eurylopas 


M. Miocene 


BO 


Woodring, 1982 




gnidia' 


Recent 


G. of Calif.— Peru 


Broderip & Sowerby, 
1829 




jamaniana 


Pliocene — 


Pliocene, 


Pilsbry & 






Recent 


Ecuador; Punta Pasado, 
Ecuador, Recent 


Olsson, 1941 




omatissima 


Pliocene — 


Pliocene, 


Broderip, 






Recent 


Ecuador; 
Panama — Ecuador, 
Recent 


1835 




posorjensis 


L. Oligocène 


UB 


Olsson, 1931 




procancellata' 


т.— L. 
Pliocene 


PB, JB, CA 


Mansfield, 
1932 


formerly Chione 


propinqua 


M. Miocene 
— E. Pliocene 


BO, ZO, DA 


Speiker, 1922 


(continued) 



110 



ROOPNARINE 



TABLE 1 . {Continued) 



Species 


Age 


Localities/Range 


Author(s) 


Comments 


rowleei 


L. Miocene 


GT 


Olsson, 1922 




tegulum' 


L. Miocene 


GT 


Brown & Pilsbry, 
1911 




Chionopsis 










walli 


Miocene(?) 


Manzanilla, 
Trinidad 


Guppy, 1866 


formerly Ctiione 


woodwardr 


E.— L. 
Pliocene 


BW, GB, AX; 
Cumana, Venezuela 


Guppy, 1866 


formerly Ciiione 


lliochione 


L. Pliocene — 


Baja Sur, 


Wood, 1828 




subrugosa' 


Recent 


Mexico, L. Pliocene: G. of 
Calif. — Peru, Recent 






Lirophora 










athleta- 


E.— L. Pliocene 


PB, CA 


Conrad, 1862 




alveata' 


L. Miocene 


SM 


Conrad, 1831 




ballista 


E. Oligocène 


SI 


Dall, 1903 




carlottae 


E. Pliocene 


CE, GB 


Palmer, 1927 




caroniana 


L.(?) 
Miocene 


Springvale, 
Trinidad, W.l. 


Maury, 1925 




chiriquiensis 


E. Pliocene 


LI 


Olsson, 1922 




clenchr 


L. Pleistocene — 
Recent 


L. Pleist., Louisiana; 

Texas 
— G. de Campeche, 

Mexico, Recent 


Pulley, 1952 




dalli 


L. Miocene 


ES 


Olsson, 1914 




discrepans 


Recent 


Nayarit, Mexico — Islay, 
Peru 


Sowerby, 1835 




ebergenyi 


E. Pliocene 


PN 


Bose, 1906 




falconensis 


L. Miocene — 
E. Pliocene 


AL, AN, GT, UR 


Hodson, 1927 




hendersoni' 


L. Pliocene 


BW 


Dall, 1903 




lat ill rata' 


E. Miocene — 


BE. CA, CV, JB, 


Conrad, 1841 


numerous 




Recent 


PB, WA, YT, ZO; N. 
Carolina — Brazil, Recent 




morphs probably representing 
different species 


mariae' 


Recent 


G. of Calif. — Guayaquil, 
Ecuador 


d'Orbigny, 1846 




obliterata 


Recent 


Pacific 
Mexico — Рапагла(?) 


Dall, 1902 




paphia' 


Recent 


West Indies— Brazil 


Linnaeus, 1767 




quirosensis 


M. Miocene 


CU, BO 


Hodson, 1927 




riomaturensis 


E. Pliocene 


MA 


Maury, 1925 




sellardsi' 


E. Miocene 


CH 


Gardner, 1926 




tembia 


L. Miocene 


AN 


Olsson, 1964 




victoria 


E. Oligocène 


Vicksburg Group, RB, MS, 
Mississippi 


Dall, 1903 




vrendenburgi 


L. Miocene 


ES 


Ward, 1992 




Panciiione 










burnsii' 


E. Miocene 


CH 


Dall, 1900 


formerly Lirophora 


fur)iakensis' 


E. Miocene 


CH 


Gardner, 1926 


formerly Liroptiora 


liolocyma' 


E. Miocene — 


CH, GT 


Brown & 






L. Miocene 


Chesapeake region 


Pilsbry, 1911 


formerly Lirophora 


hotelensis 


L. Miocene 


GT 


Olsson, 1922 


formerly Lirophora 


keiiettii' 


E. Pliocene- 


PN, G. of Calif.— 


Hinds, 1845 


formerly 




Recent 


Peru 




Lirophora: not Mercenaria 
(as in Harte, 1992a) 


mactropsis' 


L. Miocene 


AN, GT 


Conrad, 1855 


formerly Lirophora 


parkeria 


E— M. 
Miocene 


CV, PM 


Glenn, 1904 


formerly Lirophora 


Pancliione 










trimeris' 


E. Miocene 


CH 


Gardner, 1926 


formerly Lirophora 


ulocyma' 


m. Pliocene 


JB, PB 


Dall, 1895 


formerly Lirophora 


Puberella 










bainbridgensis' 


E. Oligocène 


E. Oligocène of 
Mississippi, Alabama, 
Georgia and Florida 


Dall, 1916 


= spenceri, Cooke, 1919; 
formerly Chione 


cortinaria' 


E. Miocene 


CH; Murfreesboro Stg.. 
Virginia 


Rogers, 1835 


formerly Chione 


cribraria' 


m.— L. 
Pliocene 


PB, CA; Duplin 
Stg., N. Carolina 


Conrad, 1843 


formerly Chione 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



111 



TABLE 1 . (Continued) 



Species 


Age 


Localities/Range 


Author(s) 


Comments 


intapurpurea' 


Recent 


N. Carolina — Brazil 


Conrad, 1843 


formerly Cliione 


montezuma' 


Recent 


Costa Rica — Panama 


Pilsbry & Lowe, 1932 


formerly Ciiionopsis 


morsitans' 


L. Pliocene 


CA 


Olsson & Harbison, 1953 


formerly Ctilone 


olssoni 


Recent 


Ecuador 


Fischer-Piette, 1969 




púbera 


L. Miocene 
(?)— Recent 


L. Miocene, 
Trinidad(?); West 
Indies — Brazil, Recent 


Valenciennes, 1827 




pulicaria' 


Recent 


G. of Calif. — Tumaco, 
Colombia 


Brodenp, 1835 




purpurissata 


Recent 


G. of Calif. — Ecuador 


Dall, 1902 




sawkinsi' 


E.— L 

Pliocene 


CE, BW 


Woodnng, 1925 


formerly Ciiione 



example, Keen, 1 971 ), would be the only sub- 
genus to range beyond the Americas, being 
found also in the Indo-Pacific and eastern At- 
lantic. However, there are several fundamen- 
tal morphological differences between Timo- 
clea and other members of the Chione 
complex; the sculpture is almost entirely ra- 
dial, except for irregularly raised points on the 
radial lines, which together form an apparent 
cancellate pattern. Also, the palliai sinus 
tends to be deeper than the normal condition 
for Chione, and the hinge plate is not bowed 
ventrally, an uncommon condition in Chione. 

Similarly, the species C. (Chione) subimbri- 
cata (Sowerby) and C. (Chione) tumens (Ver- 
rill), were originally classified in the Western 
Atlantic and Indo-Pacific genus Anomalocar- 
dia (Hertlein & Strong, 1 948), but were reclas- 
sified in Chione s. s. (Olsson, 1961). Olsson 
(1961) argued correctly that С subimbricata 
possesses the hinge characteristics and 
some of the sculptural characters of Chione. 
Chione tumens is essentially a larger version 
of С subimbricata, and has been considered 
by some to be a subspecies of С subimbri- 
cata (for example Keen, 1971). Olsson (1942) 
also described a closely related species from 
the Lower Pliocene Burica Formation of Pan- 
ama, Chione vaca Olsson. Phylogenetic anal- 
yses reported later in this paper support 
Olsson's (1961) assignment of these taxa to 
Chione s. s., despite the obvious and unique 
nature of their sculpture. 

Given the obvious taxonomic confusion of 
Chione and related genera, it is doubtful that 
the examination of diversity patterns and ad- 
aptation within the current genus could yield 
evolutionarily meaningful results. Such an ex- 
amination presumes that the genus is mono- 
phyletic, but this presumption cannot be le- 
gitimized until a phylogenetic analysis of the 
taxa (subgenera) within the genus is under- 
taken, and the taxonomic relationships and 



character transformations within the clade 
established. 

Geological History 

The Chione complex ranges geologically 
from the Early Oligocène (Rupelian Stage) 
(Dockery, 1982) to the Recent, and appears 
to be restricted to tropical, primarily shallow. 
New World waters (Olsson, 1961). The earli- 
est occurrences are of Puberella and Liro- 
phora in the Lower Oligocène Byram Forma- 
tion of Mississippi (Dockery, 1982), and 
Puberella in Lower Oligocène strata of An- 
tigua (Dockery, 1982). Dall described one of 
the Byram species as Chione (Chione) bain- 
bridgensis (= spenceh Cooke). The type of 
surface sculpture and the depth of the palliai 
sinus, however, suggest that this species is 
more closely related to Puberella púbera 
(Saint-Vincent). Lirophora had extended its 
range to the Eastern Pacific by at least the 
Late Miocene, occurring in the Zorritos For- 
mation of Peru (Woodring, 1982; see Duque- 
Caro, 1990, for age of formation). Puberella, 
on the other hand, while diverse today in the 
Eastern Pacific, has no documented fossil 
record in that region. 

Chionopsis first occurs in the Late Oli- 
gocène (Chattian Stage?) Upper Bohio For- 
mation of Panama (Caribbean side) (Wood- 
ring, 1982). It was widespread in both the 
Eastern Pacific and Western Atlantic during 
the Miocene, but is today a paciphilic taxon, 
having become extinct in the Western Atlan- 
tic by the end of the Pliocene. The earliest 
documented occurrence of Chione s.s. is in 
the species-rich Lower Miocene (Burdigalian 
Stage) Chipóla Formation of northwestern 
Florida [Chione chipolana Dall, 1903). The 
taxon reached the Eastern Pacific by at least 
the Early Pliocene (Roopnarine, unpub- 



112 



ROOPNARINE 



TABLE 2. Species used in the phylogenetic analysis. Chione subgenera are listed first. Specimens 
belonging to all species, with the exception of Lirophora victoria, were exannined by the author. 



Genus/subgenus 



Species 



Locality/Formation 





chipolana 


Ciiionisata 


fluctifraga 


Chionopsis 


amathusia 


lliochione 


subrugosa 


Liropliora 


victoria 




athleta 


Panchione 


mactropsis 




ulocyma 


Puberella 


cribraha 


Chione 


tumens 


Anomalocardia 


auberiana 




flexuosa 


Mercenaria 


mercenaria 


Protothaca 


asperrima 


Timoclea 


marica 



Jamaica (Recent) 

Chipóla Fm., Florida (Lower Miocene) 

Gulf of California (Recent) 

Pacific Panama (Recent) 

Pacific Panama (Recent) 

Lower Oligocène (Dockery, 1982) 

Caloosahatchee Fm., Florida (Upper Pliocene) 

Gatun Fm., Panama (Upper Miocene — Lower Pliocene) 

Lower Pinecrest Beds, Florida (middle Pliocene) 

Waccamaw Fm., South Carolina (Upper Pliocene) 

Gulf of California (Recent) 

Florida, Recent 

Brazil (Recent) 

South Carolina (Recent) 

Pacific Panama (Recent) 

Guam (Recent) 



lished), and is today widespread in both tlie 
tropical Western Atlantic and Eastern Pacific. 

The Chipóla Formation is also the first oc- 
currence of Panchione, even though Pan- 
chione species have at times been classified 
as Lirophora species (Gardner, 1926). Wood- 
ring (1982) however, likens the type of 
Panchione, P. mactropsis (Conrad) (Late 
Miocene, Gatun Formation, Panama), to an- 
other species, P. ulocyma (Dall) from the Chi- 
póla Formation. It is clear from the descrip- 
tion of Panchione (Olsson, 1964) that the 
earliest species occur in the Chipóla Forma- 
tion. Woodhng (1982) incorrectly states that 
Panchione persists in the Western Atlantic 
only until the Late Miocene, for species sim- 
ilar (if not identical) to P. ulocyma occur in the 
Upper Pliocene Caloosahatchee Formation 
of Florida. Panchione is today paciphilic, be- 
ing represented by a single Eastern Pacific 
species P. kelletii (Hinds) (Woodring, 1982). 

The two remaining subgenera, Chionista 
and lliochione have brief fossil records. Both 
have their earliest documented occurrences 
in Upper Pliocene deposits of Baja California 
(Durham, 1950). Also, the earliest docu- 
mented occurrence of the Chione vaca-sub- 
imbricata-tumens species trio is in the Lower 
Pliocene Esmeraldas Formation of Ecuador 
(C. vaca, personal observation). All three taxa 
are today restricted to the Eastern Pacific. 

MATERIALS AND METHODS 
Phylogenetic Analyses 

Twenty-seven morphological characters 
were described for species from all the sub- 



genera discussed above. Type species were 
used whenever specimens were available. 
Type species examined include: Chionopsis 
amathusia (Philippi), Chione cancellata (Lin- 
naeus), Panchione mactropsis (Conrad), llio- 
chione subrugosa (Wood), Chionista flucti- 
fraga (Sowerby), Mercenaria mercenaria 
(Linnaeus), Anomalocardia flexuosa (Lin- 
naeus), and Protothaca (Leukoma) asperrima 
(Sowerby). The type species of Timoclea, T. 
ovata (Pennant), was not available, so the 
Indo-Pacific species T. (Glycydonta) manca 
(Linnaeus) was used instead. A complete list 
of the species used in the analysis is given in 
Table 2. Characters were obtained from both 
left and right valves, utilizing the asymmetry 
typical of chionine valves, and are discussed 
in more detail in Appendix I. The numerically 
coded data set is presented in Appendix II. 
Specimens from the following collections 
were examined: California Academy of Sci- 
ences, Field Museum of Natural History, Flor- 
ida Museum of Natural History, Tulane Uni- 
versity Geological Collections (the collections 
of Drs. Emily and Harold Vokes), University of 
California Berkeley (Museum of Paleontol- 
ogy), the private collection of Dr. Geerat Ver- 
meij, and the author's own collection. 

The data were analyzed, and phylogenetic 
hypotheses constructed, using PAUP 3.11 
(upgrade of PAUP 3.0, Swofford, 1991). The 
branch-and-bound algorithm was used to 
provide an exact solution to the search for a 
most parsimonious cladogram. All equally 
most parsimonious solutions were retained. 
Non-binary characters were unordered and 
scaled to equal weight based on the number 
of states per character (PAUP option 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



113 



WEIGHT SCALE; Swofford, 1991). This 
method of weighting ensures that characters 
with three or more states do not dominate the 
resulting trees (Swofford, 1985). 

The ingroup comprises the subgenera 
Chione, Chionopsis, Puberella, Lirophora, 
Panchione, lliochione, Chionista and the 
Chione tumens group. The genera Merce- 
naria, Anomalocardia, Protothaca and Timo- 
clea were treated as outgroup taxa. The out- 
group taxa were not constrained to be 
outgroups with respect to the ingroup (i.e., 
the cladograms were not rooted with the out- 
groups), nor was an a priori hypothesis of 
relations among the outgroup taxa included 
in the analysis. In fact, preliminary analyses 
(Roopnahne, unpublished) constraining 
these genera to be outgroups indicated that 
the ingroup cannot be monophyletic with re- 
spect to the outgroup taxa. The cladograms 
of the present analysis are unrooted. 

Optimal character state trees were recon- 
structed after analysis according to the 
ACCTRAN and DELTRAN criteria. ACCTRAN 
maximizes character reversals, and mini- 
mizes convergences (Swofford, 1991). This 
criterion is therefore a conservative test of 
parallel and convergent evolution, which is 
suspected for many chionine character 
states. DELTRAN forces character transfor- 
mations in the opposite direction, favoring 
convergence over reversals. A comparison of 
the transformations formulated by each algo- 
rithm allows a comparison of alternative ev- 
olutionary pathways of chionine characters. 
Character transformations that are sup- 
ported by both algorithms could be consid- 
ered particularly robust. Finally, the results of 
the analysis were used as a basis to revise 
the current taxonomic arrangement of 
Ctiione taxa. 

Biogeographic Analysis 

The purpose of this analysis was to exam- 
ine the geological and geographic distribu- 
tions of species within each of the Chione 
subgenera, and thereby arrive at conclusions 
about subgeneric survival and restriction dur- 
ing the late Neogene. The primary focus was 
a consideration of differences between the 
Eastern Pacific and the Western Atlantic, and 
between the Gatunian and Caloosahatchian 
provinces. Based on previous work (Wood- 
ring, 1966; Vermeij & Petuch, 1986; Stanley, 
1986), higher levels of extinction are pre- 
dicted to occur in the Western Atlantic com- 
pared to the Eastern Pacific. More recent 
work (Alimón et al., 1993; Jackson et al., 



1993) suggests that the extinctions in the 
Western Atlantic might be matched by spe- 
ciation, whereas the Eastern Pacific should 
have a higher overall rate of origination. 

A list (Table 1) was compiled comprising 
all species assigned to the following taxa: 
Chione, Chionopsis, Puberella, Lirophora, 
Panchione, lliochione, and Chionista. The ge- 
nus /Anoma/ocard/a was also included, based 
on the results of the phylogenetic analyses. 
Several species were re-assigned to new 
subgenera based on a reconsideration of 
subgeneric definitions (Table 1), and some 
species names were synonomized on the ba- 
sis of the information gathered by specimen 
examination and literature descriptions. All 
these changes are noted on the species list 
for each subgenus (Table 1). Of the 90 spe- 
cies listed in Table 1 , specimens belonging to 
47 of them were examined by this author. 
Thirty-nine of the 47 species examined orig- 
inated after the Miocene. Literature sources 
listed in Table 1 were used to obtain informa- 
tion pertaining to the time of first appearance, 
and geological and geographical ranges of 
each species. A complete listing of geologi- 
cal formations considered is given in Table 3, 
along with ages, and source of age informa- 
tion. 

Many Neogene deposits in tropical and 
sub-tropical America remain poorly dated, 
due to a combination of poor stratigraphie 
resolution, the lack of continuous sequences 
with index fossils, and discontinuous geolog- 
ical study. Several recent advances have 
started to resolve the situation, but on prima- 
rily regional scales (Duque-Caro, 1990; Jones 
et al., 1 991 ; Krantz, 1 991 ; Coates et al., 1 992; 
Jones, 1995). Therefore, "consensus" ages 
were assigned to many of the formations 
listed, based on faunal characteristics and 
the most recent age estimates available. 

The paleobiogeographic data cover the 
Early Oligocène to the Recent. In order to 
summarize the general biogeographic history 
of the subgenera, each chronological epoch 
was divided into sub-epochs, according to 
Harland et al. (1990). Chronological stages 
were not used because that resolution is sim- 
ply not yet available for many formations. In 
each sub-epoch interval, the number of spe- 
cies in each subgenus was listed, along with 
the geographic locations or range of the spe- 
cies (Table 1). Each location or range was 
assigned to one or both of two general re- 
gions: the tropical Western Atlantic or East- 
ern Pacific. Central American deposits and 
the species therein which pre-date Isthmian 
uplift (middle Pliocene) were considered to 



114 



ROOPNARINE 



TABLE 3. List of all geological deposits and formations considered in this study. Abbreviations are used 
throughout the text. References were used as sources of most current age assignments or 
réévaluations. 



Formation 


Abbreviation 


Age 


Location 


Reference 


Aguequexite 


AX 


middle Pliocene 


Atlantic Mexico 


Jackson et al., 1993 


Alhajuehla 


AL 


Late Miocene 


Pacific Panama 


Woodring, 1982 


Angostura 


AN 


Late Miocene 


Ecuador 


Duque-Caro, 1990 


Bermont 


BM 


Early Pleistocene 


Florida 


Lyons, 1991 


Bowden 


BW 


Late Pliocene 


Jamaica 


Stanley, 1986 


Burica 


BU 


middle Pliocene 


Pacific Panama 


Coates et al-, 1992 


Byram 


BY 


Early Oligocène 


Mississippi 


Dockery, 1982 


Caloosahatchee 


CA 


Late Pliocene 


Florida 


Lyons, 1991 


Calvert 


CV 


E. — Middle Miocene 


Maryland 


Ward, 1992 


Cercado 


CE 


Early — middle Pliocene 


Dominican Republic 


Saunders et al., 1986 


Chipóla 


CH 


Early Miocene 


Florida 


Bryant et al., 1992 


Culebra 


CU 


Middle Miocene 


Atlantic Panama 


Duque-Caro, 1990 


Daule 


DA 


L. Miocene — E. Pliocene 


Ecuador 


Duque-Caro, 1990 


Duplin 


DU 


Early — middle Pliocene 


South Carolina 


Krantz, 1991 


Eastover 


ES 


Late Miocene 


Virginia 


Ward, 1992 


Esmeraldas 


EM 


Early Pliocene 


Ecuador 


Duque-Caro, 1990 


Gatun 


GT 


Late Miocene 


Atlantic Panama 


Coates et al., 1992 


Gurabo 


GB 


Early — middle Pliocene 


Dominican Republic 


Saunders et al., 1986 


Jackson Bluff 


JB 


middle Pliocene 


Florida 


Lyons, 1991 


La Boca 


BO 


Middle Miocene 


Pacific Panama 


Woodring, 1982 


Limon Group 


LI 


middle — Late Pliocene 


Atlantic Costa Rica 


Coates et al-, 1992 


low/er Pinecrest 


PB 


middle Pliocene 


Florida 


Jones et al., 1991 


Beds 










Matura 


MA 


Early Pliocene 


Trinidad, W. 1. 


Jung, 1969 


Mint Spring 


MS 


Early Oligocène 


Mississippi 


Dockery, 1982 


Montezuma 


MZ 


Early Pleistocene 


Pacific Costa Rica 


Coates et al., 1992 


Murfreesboro 




Early Miocene 


Maryland 


Ward, 1992 


Stage 










Penita 


PN 


Early — middle Pliocene 


Pacific Panama 


Coates et al., 1992 


Red Bluff 


RB 


Early Oligocène 


Mississippi 


Dockery, 1982 


Rio Banano 


RN 


Early — middle Pliocene 


Costa Rica 


Coates et al., 1992 


San Diego 


SD 


Late Pliocene 


California 


Hertlein & Grant, 1972 


Silex Beds 


SI 


Early Oligocène 


Florida 




St. Mary's 


SM 


Late Miocene 


Virginia 


Ward, 1992 


Upper Bohio 


ив 


Late Oligocène 


Pacific Panama 


Woodring, 1982 


Upper Pinecrest 


PB 


Late Pliocene 


Florida 


Jones et al., 1991 


Beds 










Urumaco 


UR 


M(?)— L(?) Miocene 


Venezuela 




Waccamaw 


WA 


Late Pliocene 


South Carolina 


Krantz, 1991 


Yorktow/n 


YT 


Early — middle Pliocene 


Virginia 


Krantz, 1991 


Zorritos 


ZG 


Late Miocene 


Peru 


Duque-Caro, 1990 



belong to a region based on characteristics 
of faunal composition and whether they are 
located on the Pacific or Atlantic sides of the 
isthmus. Contemporaneous Neogene depos- 
its on opposite sides of the isthmus are often 
very different faunistically (Duque-Caro, 
1990; Coates et al., 1992), possibly reflecting 
the action of océanographie barriers, the age 
of initial isthmus uplift, and differences in tec- 
tonic and sedimentary histories. The diversity 
within each region was then documented by 
summing the number of species in each of 



the two regions for successive sub-epoch in- 
tervals. Levels of speciation and extinction 
were assessed as the numbers of first and 
last occurrences per region per sub-epoch. 
Note that by adopting this approach, extinc- 
tions can only be constrained to the sub-ep- 
och following a last occurrence. Biostrati- 
graphic discontinuity for a subgenus does 
not necessarily represent true extinction, but 
is also dependent on stratigraphie continuity, 
which as already noted is problematic for 
much of Neogene tropical America. 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



115 



Testing the hypothesis that cooling caused 
the late Neogene extinctions requires a fine- 
scale latitudinal resolution of biogeographic 
distributions. Due both to the absence of 
widespread, stratigraphically continuous late 
Neogene sections in tropical America, and 
the coupled uncertainties in paleobiogeo- 
graphic boundaries, latitudinal resolution is 
linnited. Using Petuch's (1982) scheme, I have 
divided the entire region into three areas (Fig. 
2): the tropical Pacific Gatunian Province, 
which was roughly equivalent to the Recent 
Panamic Pacific region; the Atlantic Gatunian 
Province, which encompassed the modern 
day Caribbean Sea; and the Caloosahatchee 
Province, which extended from the Florida 
peninsula to South Carolina. The Florida pen- 
insula during the Pliocene may have repre- 
sented a zone of transition between the trop- 
ics and sub-tropics (Stanley, 1986). Species 
from Florida were assigned, at least initially, 
to the Caloosahatchian category, a decision 
supported in Roopnarine (1995). 

The last occurrences of species within 
each subgenus were placed within this 
framework for the Early-middle (5.2-2.5 mya) 
and latest Pliocene (2.4-1.6 mya). A general 
analysis of extinction levels was performed 
by summing the number of species within 
each province during each sub-epoch. The 
analysis was also performed using time inter- 
vals based on the latest age estimates of rel- 
evant deposits. A hypothesis of cooling, as 
formulated by Stanley (1984), cannot be re- 
jected if a higher level of extinction is noted in 
the tropical Atlantic Gatunian Province com- 
pared to the subtropical Caloosahatchian 
Province, after 2.4 Ma. 

Body Size 

As noted earlier, declining levels of plank- 
tonic productivity in the late Neogene tropical 
Western Atlantic has been cited as a proxi- 
mal cause of the extinctions. Changing the 
nutrient supply to a community should alter 
the trophic composition of the community. 
Hypothetically, larger-bodied species are af- 
fected more adversely than smaller, trophi- 
cally equivalent (all suspension-feeders) spe- 
cies. This statement is based on the following 
observations: large-bodied species, while 
numerically inferior with respect to smaller 
species, nevertheless account for compara- 
ble quantities of biomass (Stanton & Nelson, 
1 980; Stanton et al., 1 981 ; Staff et al., 1 985); 
and while smaller poikilotherms ingest rela- 



tively larger quantities of food than do larger 
ones (the amount of food ingested by poiki- 
lotherms is roughly four times their metabolic 
rates), the rate of nutrient intake scales pos- 
itively with body size (Peters, 1983). 

I = 0.78W° ^2 

where I = ingestion rate and W = body weight 
(Peters, 1983). 

In order to test a hypothesis of declining 
levels of planktonic primary production, max- 
imum body size was documented for as 
many Atlantic Pliocene, and Atlantic and Pa- 
cific post-Pliocene chionine species as were 
available. Body size was defined simply as 
maximum valve height, and was measured 
with digital calipers to the nearest 0.01 mm. 
The number of specimens examined for each 
species was noted (see Results, Table 4), to 
caution that this type of analysis is prone to 
sample-size bias. Articulated valves were 
counted as single individuals. Species col- 
lections with many specimens may have in- 
creased chances of containing very large 
specimens (due purely to sampling bias), 
analogous to the rarefaction relationship be- 
tween species richness and sample size ob- 
served in ecological studies (Sanders, 1968). 
This potential source of bias is compensated 
for, however, by the observation that small- 
bodied species are more abundant in depos- 
its than larger species, are therefore generally 
represented by larger samples, and hence 
have an increased probability of containing 
specimens near the maximum size of the 
population from which the sample was de- 
rived. Empirical rarefaction curves could not 
be used in this instance to verify the obser- 
vation quantitatively, because the derived 
curves would not be independent of time 
(Raup, 1975). 

The maximum body size of a species was 
defined as the body size of the largest spec- 
imen measured. In cases when a species 
was not represented adequately in one of the 
museum or author's collection, maximum 
size was taken from a literature description of 
the species, primarily from Abbott (1974), 
Keen (1971) and Palmer (1927). 

Many fossil species are known from single 
localities or very restricted geographic areas. 
A review of Tatsle 1 , however, will show that 
many extant species have very large geo- 
graphic ranges, and therefore undoubtedly 
exhibit ecologically based variation. There- 
fore, when documenting body size for Recent 



116 



ROOPNARINE 






•Mercenaria mercenaria 

'Pubereüa cribraria 

• Chionopsis amatkusia 

. Timoclea marica 
Protoütaca aspenima 
Ckionistafluctifraga 
Chionetumens 
Chione coßiceüata 
Chione chipolana 
lUochione subrugosa 
Anomalocardia fUxuosa 
AnomaJocartäa auberiana 
Lirophora athleta 
Ijrophora victoria 
Panchione mactropsis 
Panckione ulocyma 




Ч 



Mercenaria mercenaria 
Pubereüa cribraria 
Chionopsis amatkusia 
Umodea marica 
Protothaca asperrima 
Chionistaßuctifraga 
Chionetumens 
Chione cancellata 
Chione chipoiana 
ItíochJone subrugosa 
Anomatocardia flexuosa 
Anomalocardia auberiana 
lÀrophara aüdeta 
Lirophora victoria 
Panckione mactropsis 
Panchione ulocyma 




-Mercenaria mercenaria 
-Pubereüa cribraria 
^Chionopsis amathusia 
-Timoclea marica 
-Prtrtothaca aspenima 
Chionistaflucüfraga 
netumens 
necanceüata 
'hione chipolana 
nomaiocardia auberiana 
•Anomalocardia flexMosa 
'ШосЫопе subrugosa 
•Lirophora athleta 
ihora victoria 
Panchione mactropsis 
■Panchione ulocyma 



FIG. 3. Cladograms resulting from analysis of Chione shell characters. Incongruency among the cla- 
dograms is entirely dependent upon the placement of lliochione. 



species, geographic range and location were 
considered. 



RESULTS 
Phylogenetic Analyses 

Analysis of the character data by PAUP re- 
sulted in three equally parsimonious cla- 
dograms (Fig. 3), each 106 steps in length, 
and with consistency indices of 0.552, ho- 
moplasy indices of 0.448 and retention indi- 
ces of 0.602. The cladograms are similar in 
topology, differing only in the placement of 
the species lliochione subrugosa. The spe- 
cies lliochione subrugosa and Lirophora ath- 
leta (Conrad) have zero branch lengths. They 
could thus be identified as potential ances- 
tors, or as possessing the same combination 
of character states as an ancestor. The zero 
branch length for lliochione subrugosa re- 
sults, however, from the exclusion of charac- 
ter #26 (radial indentation of posterior valve 
surface), which is autapomorphic in /. subru- 
gosa, from the analysis. The character was 
included in the data matrix because it may 



become informative if more characters or 
taxa are added at a later time, or if the data 
set is later analyzed at a higher level of uni- 
versality (Wiley, 1981; S. Carlson, personal 
communication). It also indicates that /. sub- 
rugosa possesses ancestral states for many 
characters, but is most likely not an ancestral 
taxon. Lirophora athleta has a zero branch 
length, possibly because its sister species L. 
victoria (Dall) has been coded with several 
missing character states. Lirophora victoria is 
the earliest documented species of Lirophora 
(Early Oligocène), and not all character states 
could be coded with confidence. The results 
of the analysis could therefore be interpreted 
to indicate L. athleta as a potential ancestor. 
No autapomorphies could be identified to 
distinguish the two species, reflecting a com- 
mon problem when dealing with morpholog- 
ical "species" of Lirophora (see, for example. 
Ward, 1992). All three cladograms identify 
species in the same subgenus as sister spe- 
cies: Chione cancellata and С chipolana; 
Lirophora athleta and L. victoria; Panchione 
ulocyma and P. mactropsis; and Anomalo- 
cardia auberiana and A. flexuosa. 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 1 1 7 

-^— -^^^— — — ^^^^^^— ^^^^^^^— Mercenaria mercenaria 



С 




PubereUa cribraria 
Chionopsis anmthusia 
Timoclea marica 
Prototkaca aspirrima 
Chiomstafluctifiraga 
Chione tumens 
Cfüone canceUata 
Chione chipolana 
ШосЫопе subrugosa 
Anomalocardiaßexuosa 
Anomalocardia auberiana 
Lirophora athleta 
Lirophora victoria 
Punchione mactropsis 

Panchione ulocyma 

FIG. 4. Strict consensus tree of cladograms illustrated in Fig. 3. 



С 




Strict and 50% majority rule consensus 
trees were used to sumnnarize the informa- 
tion of the three cladograms (Fig. 4, 5). Both 
trees are well resolved and identical in topol- 
ogy because of the high congruency among 
the source cladograms. The incongruencies 
among the cladograms result in a polytomy 
from which branch Anomalocardia, llio- 
chione, and the sister taxa Lirophora and 
Panchione. The ingroup is divided into two 
sister clades containing the following spe- 
cies: first, Chionista fluctifraga, Chione tu- 
mens, C. cancellata and С chipolana; and 
secondly, Anomalocardia auberiana, A. flex- 
uosa, lliochione subrugosa, Lirophora 
atheleta, L. victoria, Panchione mactropsis. 



and P. ulocyma. The inclusion of Anomalo- 
cardia in the large Chione clade, and the sur- 
prising exclusion of Chionopsis amathusia 
and Puberella cribraria, indicate that Chione, 
as currently defined, is a paraphyletic genus. 
The branching position of Timoclea confirms 
its status outside the main clade of Chione 
subgenera. The structure of the cladograms 
and the consensus trees suggest strongly the 
need for a thorough taxonomic revision of 
Chione. Revising the taxonomy of Chione on 
the basis of a phylogenetic analysis pre- 
sumes that the phylogenetic hypotheses are 
more informative about the interrelationships 
of the Chione subgenera than is the tradi- 
tional taxonomy. 



118 



ROOPNARINE 



100 



100 



100 



с 



100 




100 



100 



с 



100 



100 



с 



100 



с 



' Mercenaria mercenaria 
' PubereUa cribraria 
' Chionopsis amaihusia 
Timoclea marica 
Protothaca aspirrima 
Chionistafluctífraga 
Chione tumens 
Chione cancellata 
Chione chipolana 
lUoctüone subrugosa 
Anomalocardiaßexuosa 
Anomalocardia auberiana 
Liropkora atMeta 
Lirophora victoria 
Panchione mactropsis 
Panchione ulocyma 



FIG. 5. 50% majority rule consensus tree of cladograms illustrated in Figure 3. This tree is identical in 
topology to the strict consensus tree (Fig. 4). 



Character Optimization and Evolution 

The three cladograms are very similar in 
topology, but the ACCTRAN and DELTRAN 
reconstructions differ for several significant 
characters. Perhaps a useful manner in which 
to view the data will be to examine the sup- 
port for the three major clades that are a con- 
sistent feature of all the cladograms; the 
overall Chione clade (including Anomalocar- 
dia, but excluding Cliionopsis and Puberella), 
the Chione-Chionista clade (hereafter Chione 
subclade) and the Lirophora-Panchione- 
Anomalocardia-lliochione clade (hereafter Li- 
rophora subclade). Character transforma- 
tions supported by several cladograms can 



be regarded as robust, but unresolved differ- 
ences have to await the addition of more 
character information to the analysis. The fol- 
lowing character transformations are all illus- 
trated on cladogram 1 . Interior nodes are la- 
belled on Figure 6, and the synapomorphies 
uniting taxa are listed in Appendix III. 

Chione Clade 

Two characters are apomorphic at the an- 
cestral node of the Chione clade on all cla- 
dograms, and support the monophyly of the 
clade. Two additional characters are apo- 
morphic on the basis of ACCTRAN and DEL- 
TRAN reconstructions respectively. 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 1 1 9 

— ^^-^— ^— ^— — — — ^■^^^— — — ^— Mercenaria mercenaria 



30 



29 



27 



28 



19 



18 



26 



PubereUa cribraria 
Ckionopsis amatkusia 
Timoclea marica 
Protothaca asperrima 
Chionistafluctífiraga 
> Chione tumens 
Chione canceUaia 
Chione chipolana 
IKockione subrugosa 
Anomalocartüaßexuosa 
Anomalocanüa auberiana 
Lirophora athleta 
Lirophora victoria 
Panchione mactropsis 

Panchione ulocyma 

FIG. 6. Character state tree of Character #1 (depth of palliai sinus). Key to character states: (0) solid 
line— reduced; (1) double line — short; (2) double line, upper (left) thick— deep; (3) double line, lower (right) 
thick — very deep. Transformations at labelled interior nodes are listed in Appendix III. 



24 



E 



25 



С 



Ö2 



21 



Character #20 (condition of right middle 
cardinal tooth) (CI = 0.667) is apomorphic at 
the ancestral node on all three cladograms 
(node 26) and is reconstructed identically by 
the ACCTRAN and DELTRAN algorithms. 
The character is mapped onto cladogram 1 
to illustrate the changes (Fig. 6). The plesio- 
morphic state is a tooth with a shallow, surf- 
icial groove. At node 26, the tooth becomes 
smooth. The only exception is Chionista fluc- 
tifraga, which evolves a bifid tooth. This con- 
dition is apparently convergent with the bifid 
tooth of Mercenaria. 

Character #7 (distal edge of concentric 
sculptural lamellae) (CI = 0.500) is apomor- 



phic at node 26 on all three cladograms, but 
is reconstructed differently by ACCTRAN and 
DELTRAN. Based on the ACCTRAN recon- 
struction (Fig. 7), the plesiomorphic state is a 
concentric sculptural element with a sharp 
distal edge. At the Chione ancestral node, the 
sculptural edges become smoother, but re- 
vert (converge in the DELTRAN reconstruc- 
tion) to a sharp morphology at the ancestral 
node of the Clilone subclade. Both Chione 
chipolana and Puberella cribraria have sculp- 
tural elements with reinforcing ridges on the 
distal edges. This similarity is indicated to be 
a homoplasy. 
Character #8 (orientation of concentric 



120 



ROOPNARINE 



Mercenaria mercenaria 
Puberella cribraria 
Chionopsis amaíhusia 
Timoclea marica 
Prototkaca asperrima 
Oúonistafluctifraga 
Chione tumens 
Chione canceUata 
Chione chipolana 
Riockione subrugosa 
Anomalocanuaßexuosa 
Anomalocardia auberiana 
Lirophora atkleta 
Lirophora victoria 
Panchione mactropás 

Panchione ulocyma 

FIG. 7. Character #7 (summit of concentric sculpture). Key (lines same as for Fig. 6): (0)— sharp; (1)— 
smooth; (2) — summit reinforced. 




sculptural elenaents) (CI = 0.667) is apomor- 
phic at the Chione clade's ancestral node 
(Fig. 8), but is reconstructed differently by the 
ACCTRAN and DELIRAN algorithms. The 
plesiomorphic state is vertical sculpture, 
which reappears in the Chione (Chione) sub- 
clade. The apomorphic state is folded sculp- 
ture which is flattened in the dorsal direction. 
This type of concentric sculpture is typical of 
the Lirophora subclade, and is useful in de- 
fining the group. Chione chipolana, however, 
also has sculpture that appears to be partially 
folded. The sculpture is vertical at its base, 
but is foliaceous and becomes folded toward 
the summit. This variation may be synapo- 
morphic with the ancestral sculpture of the 



Lirophora subclade, but such a hypothesis 
cannot be demonstrated with the material 
currently available. The rounded nature of the 
sculpture in Chione tumens is shared only by 
С vaca and С. subimbhcata (not included in 
this study). 

Lirophora Subclade 

There are four characters that strongly 
support the Lirophora subclade; #1 , depth of 
palliai sinus, CI = 0.600; #2, type(s) of sculp- 
ture on valve surface, CI = 0.800; #8, orien- 
tation of concenthc sculptural lamellae, CI = 
0.667 (discussed above); #1 1 , morphology of 
nymph, CI = 0.500. All these were recon- 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 121 

— ^— — ^— — ^^— — ^^^— — — — ^^ Mercenaria mercenaria 



С 




PubereUa cribraria 
Chionopsis amaihHsia 
Timoclea marica 
Protothaca aspirrima 
Chionistafluctifraga 
Cßüone tumens 
Chione canceUata 
Chione chipolana 
lUockione subrugosa 
Anomalocartüaflexuosa 
Anomalocardia auberiana 
Lirophora athleta 
: Lirophora victoria 
: Panchione mactropsis 

Panchione ulocyma 

FIG. 8. Character #8 (orientation of concentric sculpture). Key (as in previous figures): (0) — vertical; (1) — 
folded dorso-ventrally; (2) — rounded. 



structed identically by ACCTRAN and DEL- 
TRAN (Figs. 9, 10), with the exception of 
character #8. Character #8 was already dis- 
cussed as being apomorphic at the ancestral 
node of the overall Chione clade, but can be 
used to distinguish between the Lirophora 
subclade and the Chione subclade. 

The palliai sinus is generally reduced in all 
chionine taxa, compared to other venerid 
subfamilies. It is further reduced in the Liro- 
phora subclade, being present but extremely 
small. This is probably a reflection of the 
shallow burial of these clams during life. The 
palliai sinus is completely absent in Anoma- 
locardia flexuosa, a species used in this anal- 
ysis, but is present in other species assigned 



to Anomalocardia, for example A. auberiana. 
The state in these other species is identical to 
the state in the other taxa of the Lirophora 
subclade. The presence of much reduced or 
absent palliai sinuses in the Chione subclade 
is convergent with the states in the Lirophora 
subclade. 

Perhaps the character most diagnostic of 
the Lirophora subclade is the morphology of 
the nymph. This is a binary character, the 
nymph being either smooth, or roughened, a 
condition described as "rugose." All taxa 
nested within the Lirophora subclade have 
rugose nymphs. The only other taxon with a 
rugose nymph is Mercenaha, but this is ap- 
parently a convergent condition. 



122 



ROOPNARINE 



Mercenaria mercenaria 
Puberdla cribraria 
Chionopsis omaíhHsia 
Tlmoclea marica 
Protothaca asperrima 
Chionistafluctífraga 
Chione tumens 
Chione canceUata 
Chione chipolana 
ШосЫопе subrugosa 
AnomaUfcanüaßexuosa 
Anomaiocanüa auberiana 
Lirophora athieta 
Liropkora victoria 
Panckione mactropsis 

Panckione ulocyma 

FIG. 9. Character #2 (type of sculpture present). Key (as in previous figures): (0) — concentric only; (1 ) — radial 
and concentric, concentric dominant; (2) both, radial sub-obsolete; (3) triple line — both sub-obsolete; (4) 
double line, upper thick — both, radial donninant. 




Like nymph morphology, the morphology 
of the sculptural elements is diagnostic of the 
Lirophora subclade. Taxa comprising the Li- 
rophora subclade have both radial and con- 
centric sculptural elements, but the radial el- 
ements tend to be faint to obsolete. 
Lirophora itself has no radial elements 
present, but this can be viewed as a com- 
plete loss of radial sculpture, the concentric 
sculpture being synapomorphic with the rest 
of the subclade. The absence of radial sculp- 
ture in both Lirophora and Mercenana should 
therefore be recognized as homoplastic, 
both states being the result of the loss of a 
character. 



Chione Subclade 

The Chione subclade is supported strongly 
by at least three characters, #1, #8 and #10 
(anterior cardinal tooth of left valve, CI = 
0.500) (Fig. 11), two of which (1 and 10) are 
reconstructed identically by ACCTRAN and 
DELTRAN. The palliai sinus is altogether ab- 
sent in all species of Chione s.S., and Chion- 
ista, with the exception of Chione chipolana. 
Chione chipolana's palliai sinus appears to 
be convergent with the state in the Lirophora 
subclade. The plesiomorphic condition of 
character #10 is a relatively wide tooth, but 
the tooth is noticeably narrow in both the 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 123 

— ^^— — ^— — ^— ^— — ^^— — ^^ Mercenaria mercenaria 

z=^=^=^==:==^^=z=^^ РиЬегеПа cribraria 

— Chionopsis anuiihusia 

^= limoclea marica 

=1 Protothaca asperrima 



Chionistaßuctifraga 
Chione tumens 
Chione cancellata 
Chione chipolana 
niochione subrugosa 
Anomalocardiaflexuosa 
Anomalocanüa auberiana 
Lirophora athleta 
Liropkora victoria 
Panchione mactropsis 

Panchione ulocyma 

FIG. 10. Character #11 (condition of nymph). Key as in previous figures. (0) — rugose; (1) — smooth. 




С 



Chione subclade and Mercenaria (the result 
of convergence). 

The orientation of the concentric lamellae 
is vertical in Chione s.S., and this is probably 
the retention of a plesiomorphic character 
state (contrary to the ACCTRAN reconstruc- 
tion). Chione tumens has a truly unique type 
of sculpture, and it is unclear how it is derived 
from any of the character states in the Chion- 
inae. 

Biogeographic Analyses 

All the Chione subgenera (with the excep- 
tion of Chionista) first appear in the Western 
Atlantic. The number of species described 



from the Early Oligocène through the Middle 
Miocene (35.4-16.3 Ma) is low (Fig. 12). The 
Early Miocene is an exception, but most of 
the species recorded here (85.7%) are from 
the very hch Chipóla Formation of northwest 
Florida. Woodring (1982) suggested that 
some Panchione species documented there 
may be synonyms and require revision. Three 
of the four Chipóla Panchione species how- 
ever were examined by this author, and all 
are consistently recognizable. It should also 
be noted that the number of well-described 
Oligocène and Lower Miocene soft-sediment 
deposits in tropical America is relatively 
small, and accurate chronological and bio- 
stratigraphic dating is problematic. The num- 



124 



ROOPNARINE 



<^ Mercenaria mercenaria 
PubereUa cribraria 
Chionopsis amaihusia 
Timoclea marica 
Protothaca asperrima 
Chionistaßuctifraga 
Chione tumens 
Chione cancellata 
Chione chipolana 
Iliochione subrugosa 
Anomalocartüaflexuosa 
Anomtdocardia auberiana 
I Lirophora atkleta 

Lirophora victoria 
Panchione mactropsis 

Panchione ulocyma 

FIG. 1 1 . Character #10 (condition of left valve anterior cardinal tooth). Key as in previous figures. (0)— tooth 
wide; (1) — tooth narrow. 




ber of species reported here for this period is 
therefore probably not a dependable reflec- 
tion of actual diversity. 

There is a steady increase in diversity from 
the Middle Miocene to the Early Pliocene 
(16.3-5.2 Ma) in both the Western Atlantic 
and Eastern Pacific (Fig. 12), corresponding 
in part to an increase in the number of de- 
posits available for sampling. During this pe- 
hod, increasing diversity in the Western At- 
lantic is accounted for primarily by the 
appearance of new species, which is accom- 
panied by an increasing level of last appear- 
ances (Fig. 12). In the Eastern Pacific on the 
other hand, increasing diversity is the result 
of first appearances, coupled with a low rate 



of extinction, and therefore higher species 
longevities. 

The level of extinction in the Western At- 
lantic continues to increase into the Late 
Pliocene, and the proporlion of species that 
last appear in the Late Pliocene greatly ex- 
ceeds the number of first appearances (Fig. 
13). The result is a dramatically lower diver- 
sity in the Recent. In contrast, the rate of ex- 
tinction in the Eastern Pacific decreases dur- 
ing the late Neogene (Fig. 13). The proportion 
of new species, though also decreasing, is 
never exceeded by the proportion of species 
going extinct. There are many species (17) 
which first appear in the Eastern Pacific dur- 
ing the Pleistocene and Recent; this results in 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



125 



(/) 

Ф 

О 
Ш 

a 

*^- 
o 

6 




Chionopsis 



О О о О 





Sub-epoch divisions 



FIG. 12. Total species diversity of Chione subgenera since the Early Oligocène. Chronological (horizontal 
axis) intervals are geological sub-epochs. Graphs on upper left represents sum diversity of traditional 
subgenera. Upper right — Chione clade (as defined in this paper); lower left — Chionopsis; lower right — 
Puberella. Circles — total number of species present; squares — number of first appearances; triangles — 
number of last appearances. 



a higher diversity than in the Western Atlan- 
tic. The Pliocene is also the time of first ap- 
pearance of the strictly Eastern Pacific sub- 
genera, Chionista, lliochione, and the Chione 
vaca-subimbricata-tumens trio. 

In summary, these data agree well with ob- 
servations that the Western Atlantic suffers 
heavier extinction during the Pliocene (82.6%) 
than does the Eastern Pacific (38.5%) (Wood- 
ring, 1966; Stanley, 1986; Vermeij & Petuch, 
1986) (Fig. 13). The Eastern Pacific exhibits a 
higher rate of origination during the post- 
Pliocene compared to the Western Atlantic, 
resulting in a higher diversity in the Eastern 
Pacific (17 vs. 7 new species). The overall 
pattern is reflected by the individual subgen- 
era. All exhibit higher levels of extinction in the 
Western Atlantic relative to the Eastern Pa- 



cific, and higher levels of origination in the 
Eastern Pacific during the post-Pliocene. Ex- 
tinction within these Chione subgenera in the 
Western Atlantic was not matched by speci- 
ation. 

These results cover the entire Pliocene 
though, obscuring the relative timing of the 
disappearances, and hence the action of an 
extinction agent such as cooling. In order to 
focus on the time of the extinctions, it be- 
comes necessary to assign estimated ages 
to sampled deposits. Given the contentious 
nature of aging Neogene tropical American 
deposits, the most current age estimates 
available were relied upon, coming primahly 
from the following references; Coates et al. 
(1992), Duque-Caro (1990), Jones et al., 
(1991) and Krantz (1991). These works either 



126 



ROOPNARINE 



82.6% Pliocene 




25 



post-Pliocene 

66.7% 




FIG. 13. Late Neogene changes in diversity. Upper 
graph illustrates levels of extinction during the 
Pliocene in the western Atlantic and the eastern 
Pacific. Shaded areas represent number of extinct 
species, unshaded represent survivors. Lower 
graph illustrates levels of origination in the Pleis- 
tocene and Recent. Shaded areas represent new 
species. 



re-evaluate and assign ages, or sunnmarize 
current estinnates. The Early Pliocene (5.2- 
3.4 Ma) of the Atlantic Gatunian is covered 
in this set of data by species fronn the 
Agueguexquite Fornnation of Mexico (Perril- 
lat, 1963), the Cercado and Gurabo deposits 
of the Dominican Republic (Saunders et al.. 



1986), the Rio Banano Formation of the 
Limon Group in Costa Rica (Coates et al., 
1992) and the Tubara Formation of Colombia 
(Duque-Caro, 1990). All these formations 
probably pre-date the initiation of Northern 
Hemisphere cooling (~2.4 Ma), or end shortly 
thereafter. They could therefore be listed as 
Early-"middle" Pliocene. The Lower Pliocene 
deposits of the Caloosahatchian Province, 
namely the Jackson Bluff Formation, the 
lower Pinecrest Beds, the Duplin Formation, 
the Raysor Formation and the Yorktown For- 
mation, all range past the official stage 
boundary of the Late Pliocene (Jones et 
al., 1991; Krantz, 1991), 3.4 Ma, but termi- 
nate about 2.5 Ma. Therefore they too could 
be considered technically as Early-middle 
Pliocene. The Atlantic Gatunian Bowden For- 
mation is Late Pliocene in age (Stanley, 
1986), as are the Caloosahatchian Caloosa- 
hatchee Formation, upper Pinecrest Beds, 
the Murfreesboro Stage (Ward, 1992) and the 
Waccamaw Formation (Lyons, 1991; Jones 
et al., 1991). These formations are all < 2.5 
million years old. Early-middle Pliocene East- 
ern Pacific deposits comprise formations 
such as the Burica, Esmeraldas and Pro- 
greso formations, while the Late Pliocene is 
represented phmarily by deposits from Baja 
and southern California. 

Placed in a temporal framework of Early- 
middle and Late (latest) Pliocene categories, 
the pattern of extinction is sinking. Only 50% 
of all species in the Early-middle Pliocene of 
the Western Atlantic were extinct by the Late 
Pliocene, but 73.3% of Late Pliocene species 
are absent from the Pleistocene (Fig. 14). On 
the other hand, extinction levels in the East- 
ern Pacific are 57.1% and 11.1% respec- 
tively for the Early-middle and Late Pliocene. 
The extinctions would therefore seem to be 
concentrated in the Late Pliocene of the 
Western Atlantic. This result is consistent 
with the compilations of Alimón et al. (1993) 
and Jackson et al. (1993). At the subgeneric 
level however, a different pattern emerges 
(Fig. 15). In the Western Atlantic, the subgen- 
era Llrophora and Panchione experience 
much higher levels of extinction (60% and 
100%) respectively) during the Early-middle 
Pliocene, than in the Late Pliocene. The sub- 
genera Chione, Chionopsis and Puberella do 
not exhibit heightened levels of extinction un- 
til the Late Pliocene. 

The difference of timing among the sub- 
genera is not explained easily. The data 
could be biased biogeographically if the sub- 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



127 



73.3% 




Atlantic 



Pacific 



FIG. 14. Timing of Pliocene extinctions in Western 
Atlantic and Eastern Pacific. In each category, bar 
on the left summarizes data for the Early-middle 
Pliocene, bar on right the Late Pliocene. See text 
for explanation of chronological intervals. Percent- 
ages refer to percent of diversity extinct in next 
chronological interval. Shading as in Fig. 13. 



genera were not distributed evenly between 
the Atlantic Gatunian region and the 
Caloosahatchian Province, and if one of the 
two regions experienced more severe extinc- 
tion. The extinctions are probably not related 
to phylogenetic history. It is interesting to 
note that Lirophora and Panchlone occupy a 
separate subclade that could have been dec- 
imated by a selective extinction agent (Fig. 4). 
If the extinction was biased against Lirophora 
and Panchione however, then Anomalocardia 
should exhibit the same pattern. It does not, 
and it seems therefore that any hypothesis of 
phylogenetic selectivity of the extinction can 
be rejected. 

In order to examine the role of biogeo- 
graphic distribution in the extinction, it is nec- 
essary to assign species to one or both of 
two biogeographic regions, the Atlantic Ga- 
tunian region and the Caloosahatchian Prov- 
ince. Interestingly, all Pliocene species of 
Chione were endemic to one of the two re- 
gions, with the possible exception of Chione 
cancellata. Even this species, however, may 
in fact be two separate taxa (Roopnarine, 
1995). Species can therefore be placed easily 
into one or two of the following categories: 
Early-middle Pliocene Atlantic Gatunian; 
Early-middle Pliocene Caloosahatchian; Late 
Pliocene Atlantic Gatunian; Late Pliocene 
Caloosahatchian. The results indicate that 
extinction of Early-middle Pliocene species in 



the Atlantic Gatunian region was 53.8%, 
while in the Caloosahatchian Province it was 
only 18.2% (Fig. 16). By the latest Pliocene, 
however, extinction declined slightly in the 
Atlantic Gatunian to 50%, but climbed to 
58.3% in the Caloosahatchian. 

The high level of extinction in the Atlantic 
Gatunian during the Early-middle Pliocene is 
due almost entirely to the extinction of spe- 
cies assigned to Lirophora and Panchione 
(Fig. 17). Both subgenera also experience 
higher levels of extinction at this time, in the 
Caloosahatchian Province, relative to other 
subgenera. Furthermore, it is apparent that 
Lirophora, with its numerical dominance of 
the species diversity of both the Atlantic Ga- 
tunian and Caloosahatchian, contributes the 
most to the extinction. During the latest 
Pliocene all surviving subgenera experience 
high levels of extinction in both geographic 
regions, with the exception of Anomalocar- 
dia. 

Body Size 

Early-middle Pliocene species of the 
Caloosahatchian Province all exceed 35 mm 
in height (Table 4). The largest species is 
Chionopsis procancellata (Mansfield) from the 
lower Pinecrest Beds, the largest specimen of 
which measured 57.75 mm. Lirophora athleta 
from the lower Pinecrest Beds reached a 
height of at least 35.5 mm. Species from the 
same deposit include Chione erosa, which 
reached a height of 47.35 mm, and Panchione 
ulocynna, which reached a height of 43.30 mm. 
Puberella cortinaria (Rogers) from the Jack- 
son Bluff Formation was measured at 37.03 
mm. Species from the northern regions of the 
Atlantic Gatunian Province tend to be much 
smaller. There are no records of Early-middle 
Pliocene specimens of Chionopsis wood- 
wardi (Guppy) or Puberella sawkinsi (Wood- 
ring) from the Cercado or Gurabo Forma- 
tions exceeding 30 mm in height, nor of 
specimens of Chione primigenia Pilsbry & 
Johnson or C. santodomingensis Pilsbry & 
Johnson (Palmer, 1927). Interestingly, there is 
some evidence that species from the south- 
ern portion of the province, for example the 
Rio Banano Formation of Costa Rica, were 
larger. Woodring (1982) recorded a specimen 
of Panchione mactropsis of 51.5 mm height 
from the Rio Banano. Chionopsis tegulum 
(Brown & Pilsbry), while not measured for this 
study, is a large species, attaining heights in 



128 



ROOPNARINE 



6 

5 

Vi 

.2 4 
о 

(О о 



• 2 -^ 

"1 

О 






i и 



;^ 



I 

Si 

1 






к?^ v^^ 



0) 

■^4 _1 

ф 
(О 3 - 



" 1 

О 



Е.-т. Pliocene 



у у 
2_ 2 



J 

V / 

•у / 

Í 



i 






Ví^^^^*^^ 



FIG. 15. Subgeneric breakdown of data presented 
in Fig. 14. 



14 

12 -I 



фЮ 
и 
g. 8 

о 4 



53.8% 





А^* 



tí* 



Х?^ 






FIG. 16. Temporal and biogeographic categoriza- 
tion of Pliocene extinctions in the western Atlantic. 
Shading represents extinct species. Abbreviations: 
E.-m. — Early and middle Pliocene; L. — Late 
Pliocene; Gatunian — Atlantic Gatunian region; 
Caloosa. — Caloosahatchian Province. 



excess of 45 mm in the Upper Miocene Ga- 
tun Formation. It also persists into the Lower- 
middle Pliocene Rio Banano Formation. The 



5 - 



0) 

Ф 
(0 3 



Late Pliocene 



Ш 



VM 



и^ 



1. 



Ш 



^ 






FIG. 17. Subgeneric breakdown of data presented 
in Fig. 16. Per subgenus, bars on right represent 
the Caloosahatchian Province, and on left the At- 
lantic Gatunian region. Key to bars as in previous 
figures. 



only Early Pliocene sizes available from the 
Eastern Pacific are of the species Chionopsis 
amathusia and Chione vaca (Olsson) from the 
Esmeraldas Formation of Ecuador. Both ex- 
ceeded 30 mm in valve height. 
The situation alters significantly in the lat- 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



129 



TABLE 4. Valve heights measured for samples of chionine bivalves. Maximum valve height recorded for 
each sample. 



Species 



Age 



Locality/Formation 



Chione erosa 


E.— middle 




Pliocene 




E. — middle 




Pliocene 




E. — middle 




Pliocene 




E. Pliocene 




L Pliocene 




L. Pliocene 


Chione cancellata 


L. Pliocene 


(United States) 






E. Pleistocene 




M. Pleistocene 




L. Pleistocene 




Recent 




Recent 




Recent 


Chione cancellata 


Recent 


(West Indies) 






Recent 




Recent 




Recent 


Chione mazyckii 


Recent 


Chione undatella 


Recent 




Recent 




Recent 


Chionopsis 


E. Pliocene 


procancellata 






L. Pliocene 




L. Pliocene 




L. Pliocene 


Chionopsis woodwardi 


L. Pliocene 


Chionopsis amathusia 


E. Pliocene 




Recent 


Chionopsis gnidia 


Recent 


Lirophora athleta 


E. Pliocene 




E. Pliocene 




L. Pliocene 




L. Pliocene 




L. Pliocene 


Lirophora latilirata 


L. Pliocene 




E. Pleistocene 




Holocene 


Lirophora hendersoni 


L. Pliocene 


Lirophora paphia 


Recent 


Panchione ulocyma 


E. — m. Pliocene 


Puberella cribraria 


L. Pliocene 




L. Pliocene 


Puberella intapurpurea 


Recent 


Puberella morsitans 


L. Pliocene 


Puberella sawkinsi 


L. Pliocene 


Puberella pulicaria 


Recent 


Ch ion i St a cortezi 


Recent 


Chionista flucti fraga 


Recent 


lliochione subrugosa 


Recent 


Chione tumens 


Recent 


Chione raca 


E. Pliocene 



Raysor Fm., South Carolina 

Pinecrest Beds, Bird Rd., Florida 

Pinecrest Beds, Collier Co., Florida 

Pinecrest Beds, Sarasota, Florida 
Caloosahatchee Fm., Palm, Beach Co., 

Florida 
Caloosahatchee Fm., Palm Beach Co., 

Florida 
Waccamaw Fm., South Carolina 

Bermont Fm., Highlands Co., Florida 
Ft. Thompson Fm., Collier Co., Florida 
Anastasia Fm., Palm Beach Co., Florida 
Sanibel Is., Florida 
Palm Beach Co., Florida 
South Carolina 
Bahamas Islands 

Jamaica 

St. Thomas, U.S. Virgin Is. 

Venezuela 

South Carolina 

Laguna San Ignacio, Baja California, 

Mexico 
Bahia San Luis, Baja California, Mexico 
San Carlos, Baja California, Mexico 
Lower Pinecrest Beds, Sarasota Co., 

Florida 
Highlands Co., Florida 
Palm Beach Co., Florida 
Okeechobee Co., Florida 
Bowden Fm., Jamaica 
Esmeraldas Fm., Ecuador 
Baja California, Mexico 
Guaymas, Mexico 
Sarasota Co., Florida 
Collier Co., Florida 
Collier Co., Florida 
Caloosahatchee Fm., Florida 
Caloosahatchee Fm., Florida 
Waccamaw Fm., North Carolina 
Bermont Fm., Palm Beach Co., Flohda 
Mississippi delta, Louisiana 
Bowden Fm., Jamaica 
Jamaica 

Pinecrest Beds, Sarasota Co., Florida 
Waccamaw Fm., South Carolina 
Waccamaw Fm., North Carolina 
Bahamas 

(Olsson and Harbison, 1953) 
Bowden Fm., Jamaica 
San Felipe, Baja California, Mexico 
San Felipe, Baja California, Mexico 
Gulf of California, Baja California, Mexico 
Panama 

Gulf of California, Baja California, Mexico 
Esmeraldas Fm., Ecuador 





Max. 


No. of 


height 


specimens 


(mm) 


7 


39.83 


68 


41.46 


27 


38.07 


11 


47.34 


100 


32.41 


34 


36.86 


15 


34.48 


39 


38.37 


100 


26.91 


38 


31.58 


50 


26.87 


100 


36.24 


9 


38.76 


30 


22.07 


58 


27.49 


34 


26.03 


40 


35.35 


3 


12.97 


5 


37.43 


27 


47.99 


16 


52.17 


5 


57.75 


13 


52.31 


4 


55.84 


11 


52.14 


9 


21.33 


1 


34.53 


6 


41.61 


11 


75.15 


5 


35.5 


18 


27.72 


32 


27.86 


1 


30.83 


10 


24.05 


13 


28.99 


35 


32.65 


42 


31.68 


42 


24.46 


19 


22.47 


10 


43.30 


40 


40.56 


8 


38.91 


3 


30.09 


1 


41.50 


14 


22.37 


1 


35.50 


12 


64.70 


8 


49.04 


29 


34.26 


35 


39.01 


1 


33.94 



130 



ROOPNARINE 



est Pliocene. Caloosahatchian descendants 
of Early Pliocene conspecifics remain quite 
large, all exceeding 30 mm in height. New 
species, for example Puberella morsltans 
(Olsson & Harbison), are also large. In the 
Atlantic Gatunian, however, species from the 
Upper Pliocene Bowden Formation do not 
exceed 25 mm in height. The species in this 
formation, Chionopsis woodwardi, Puberella 
sawkinsi and Lirophora hendersoni (Dalí), 
range in size from 20-25 mm. 

Species in the Caribbean today remain rel- 
atively small. The two most common species, 
Chione cancel lata and Lirophora pap h la (Lin- 
naeus) rarely exceed 30 mm in height. Nota- 
ble exceptions occur in upwelling areas (Ver- 
mel], 1978), for example off the coast of 
northern Venezuela. This region yields large 
specimens of C. cancellata. The largest one 
measured in this study was 35.35 mm in 
height. Species from the coastal waters of 
the United States, however, are comparable 
in size to Early and Late Pliocene Caloosa- 
hatchian species. All species examined had 
specimens over 30 mm. Palmer (1927) de- 
scribes a specimen of the Caribbean species 
Puberella púbera with a valve height of 51 
mm, but does not give detailed locality infor- 
mation. 

It is noteworthy that no Recent species in 
neither the Atlantic Gatunian nor the 
Caloosahatchian Provinces exceeds 40 mm 
in height. Table 4 indicates that there were at 
least four Early Pliocene species {Chione 
erosa Dal I, Chionopsis tegulum, Chionopsis 
procancellata, and Panchione ulocyma), and 
three Late Pliocene Caloosahatchian species 
{Chionopsis procancellata, Puberella cribraria 
(Conrad), and P. morsltans) which did exceed 
40 mm in height. All these species are ex- 
tinct, and the subgenera Chionopsis and 
Panchione are today paciphilic. In contrast. 
Recent Pacific chionine species measured 
for this study are very large. Chione californ- 
iensis Broderip and C. undatella both attain 
heights in excess of 50 mm, while the spe- 
cies Chionopsis amathusia and C. gnidia 
(Broderip & Sowerby) have maximum sizes of 
41.61 mm and 75.15 mm respectively. One 
specimen of Puberella pulicaha (Broderip) 
from the Gulf of California measured 35.50 
mm. Moreover, the subgenera that have 
evolved in, and are restricted to the Eastern 
Pacific, Chionista, lliochione and Chione tu- 
mens are also quite large; for example, 
Chionista cortezi (Carpenter), 64.70 mm; 
Chionista flucti fraga, 49.04 mm; lliochione 



E80 

*-70 
D) 

"S 60 - 



0) 

> 

CO 

>40 

3 

E 
"x 
<5 20 



50 - 



30 




/^^^ 



О 



О 



.o*^ 






,o® 



áí^ 



<>• 



iS> 



V 



ri.^ 



<b' 



FIG. 18. Maximum valve heights of chionine spe- 
cies during the late Neogene. Open circles — Atlan- 
tic Gatunian region; filled circles — Caloosahatch- 
ian Province; open triangles — Pacific Gatunian 
region. 



subrugosa, 34.26 mm; Chione tumens, 39.01 
mm. In general. Pacific species are larger 
than their Atlantic congeners. Figure 1 8 sum- 
marizes these data, and Figure 19 presents 
the data for individual subgenera. 



DISCUSSION 
Evolution of Chione 

Assuming that the paleontological record 
of chionine species is reasonably well docu- 
mented, the common ancestor of the Chione 
subgenera, as well as Anomalocardia, had 
evolved by the Late Eocene, possibly earlier. 
Some subgenera bear morphological resem- 
blances to others, for example Lirophora and 
Panchione, but all appear in the fossil record 
essentially fully developed, with taxon-defin- 
ing synapomorphies present. 

Within the entire clade (comprising Anom- 
alocardia, Chione, Chionista, lliochione, Liro- 
phora, and Panchione) there are examples of 
convergent and parallel evolution exhibited 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



131 



55 -, 

50 

45 

40 -J 

35 

30 

25 



Chione m 



36 

32 A 
28 
24 
20 



Lirophora 



44 
40 
36 
32 - 
28 - 
24 



Panchione 



80 
70 
60 
50 
40 
30 
20 
10 



Chionopsis 




.if^ 



.ié> 



Ь 



^ 



V 



,<í> 



V 



.\^ 



^ 



<^" 



42 
40 
38 - 
36 - 
34 
32 -\ 
30 




Puberella 



FIG. 19. Subgeneric breakdown of data presented in Fig. 18. Circles— Pacific Gatunian region; triangles- 
Atlantic Gatunian region; squares— Caloosahatchian Province. Note extinction of Chionopsis and Pan- 
chione in Atlantic regions. 



132 



ROOPNARINE 



by related taxa inhabiting similar habitats. 
Two characters, however, define the clade 
strongly and consistently. The smooth con- 
dition of the right middle cardinal tooth dis- 
tinguishes the clade from the outgroups Mer- 
cenaria, Protothaca and Timoclea, as well as 
Chionopsis and Puberella. These outgroup 
taxa have teeth that are grooved, with the 
exception of Mercenaria, which has a bifid 
tooth. That condition is convergent with the 
bifid tooth of Chionista. The concentric 
sculpture is also diagnostic of the clade. The 
distal edge of the concentric lamellae tend 
to be smooth in all members of the clade, 
except Chione chipolana, which has a rein- 
forcing ridge along the lamellar edge, and C. 
cancellata, which has a sharp edge (the com- 
mon condition in Cliione s.S.). The common 
outgroup condition (plesiomorphic) is sharp 
edged lamellae. Puberella however, also has 
distally reinforced lamellae. 

The two subclades are likewise well de- 
fined. The Chione subclade is defined prima- 
rily by three characters; an extremely shallow 
palliai sinus, vertical orientation of the con- 
centric lamellae, and a narrow left anterior 
cardinal tooth. Taxa of the Lirophora sub- 
clade on the other hand, have four strong, 
diagnostic synapomorphies; a short palliai si- 
nus, the dominance of concentnc sculpture 
relative to radial sculpture, folding of the con- 
centric sculpture, and the possession of a 
rugose nymph. The characters that define the 
two subclades contrast strongly between 
them, but it is interesting to note that only two 
characters are the common strong apomor- 
phic definitions of these two clades. Consid- 
ering the levels of homoplasy in characters 
that distinguish the subclades, it would be 
imprudent to make inferences of relationship 
and clade membership based on analysis of 
a single or few characters. 

A note of caution must be expressed at this 
point against the use of seemingly consis- 
tent, functionally significant single characters 
as descriptors of phylogeny or taxonomic 
classification. The possession of a function- 
ally significant character state by several taxa 
could be indicative of common ancestry, but 
might also signify the homoplastic nature of a 
trait that is in "high demand." Fisher (1981, 
1985) argued this with respect to the multi- 
ple, independent evolution of functionally 
significant traits. An example relevant to this 
study is the condition of the nymph. The 
nymph as described above is a platform on 
the valve margin, just posterior to the poste- 



rior cardinal tooth. It is an attachment site for 
the elastic ligament between both valves, 
and is typically smooth. In several chionine 
taxa {Mercenaria and the entire Lirophora 
subclade), however, the nymph is noticeable 
rugose. The distinct difference between the 
two states of this character has made it ven/ 
useful in defining species and subgeneric 
boundaries. Moreover, its role as an attach- 
ment site for the ligament has lent it some 
notion of functional significance, though how 
smoothness and rugosity may affect liga- 
ment biomechanics is unknown. Harte 
(1992a) used the roughened nymph of Liro- 
phora kellettii (revised in this study to Pan- 
chione kellettii) as the basis for assigning the 
species to Mercenaria. Implicit in her study 
was the untested assumption that the tradi- 
tional taxonomic genus Chione is a mono- 
phyletic clade, provided that the genus Mer- 
cenaria is nested within it. The phylogenetic 
analysis presented here does not support the 
nesting of Mercenaria with Chione subgen- 
era, and has demonstrated therefore that 
Chione is not a monophyletic genus, and that 
the roughened nymphs of Lirophora and 
Mercenaria are most likely homoplastic. 

Overall, the entire Chione clade has under- 
gone tremendous diversification since the 
Early Oligocène. There are 90 species re- 
corded in this paper. Perhaps even more 
striking is the number of distinct morpholo- 
gies, or "bauplans" (as defined by Hall, 
1992), that are nested within the clade. 
Though the entire clade is supported by sev- 
eral characters, for example prominent con- 
centric sculpture, the subgenera as men- 
tioned earlier can all be distinguished easily 
from each other. These terminal taxa are not 
necessarily distinguished by autapomorphic 
characters, but more often by possessing 
unique combinations of synapomorphic 
character states. Moreover, the taxa, or the 
synapomorphic combinations that they rep- 
resent, appear in the fossil record abruptly 
and fully defined. In general, the original sy- 
napomorphic combinations still define the 
subgenera in the Recent, and allow the easy 
assignment of species to proper subgeneric 
clades. It would seem therefore that above 
the species level the rate of character evolu- 
tion in the Chione clade was initially very 
high, probably during the Late Eocene, but 
character innovation has since stabilized or 
fallen to near zero. To-date there is very little 
fossil evidence to shed light on the origins of 
individual subgenera. 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



133 



Taxonomic Revision 

Revision of a taxonomic classification on 
the basis of phylogenetic analysis should 
meet at least two important criteria. First, the 
new classification must reflect the sister- 
group relationships implied by the analysis 
(deOueiroz & Gauthier, 1992). Secondly, re- 
visions of the existing classification should 
be minimized, with the only alterations of 
rank and taxon membership being those ne- 
cessitated by the phylogenetic analysis 
(Wiley et al., 1 991 ). This critehon ensures that 
rank-based studies are affected as little as 
possible, but are consistent with the phylo- 
genetic analysis. Rank taxa would therefore 
represent monophyletic clades, and as such 
would remain convenient, phylogenetically 
meaningful units for use in studies of diversi- 
fication and extinction. The simplest method 
for achieving these goals is to convert the 
taxonomic classification to the tree that it 
supports, and then evaluate the logical con- 
sistency of the taxonomic tree with the phy- 
logenetic tree. Changes made to the taxo- 
nomic tree to make it logically consistent with 
the phylogenetic tree are subsequently trans- 
lated to modifications of the rank taxonomic 
classification (Wiley et al., 1991). 

The three cladograms and the resulting 
consensus trees all show that the genus 
Chione is paraphyletic because of the recog- 
nition of Anomalocardia as a separate genus. 
If the prevailing classification is converted to 
a phylogenetic tree (Fig. 20), it is immediately 
obvious that it is logically inconsistent with 
the consensus tree. Not only does the taxo- 
nomic classification not imply relationships 
among the Chione subgenera, but there are 
no hypotheses concerning the relationships 
among the chionine genera. The simplest 
resolution of the problem is the revision of 
Chione. Harte (1 992a) suggested that Merce- 
naria be subsumed under Chione as a new 
subgenus. That solution, however, would 
disturb the long-standing understanding of 
Mercenaria as a genus. For reasons of con- 
sistency, it would also necessitate the inclu- 
sion of Anomalocardia as a Chione subge- 
nus. Such alterations would violate the 
criterion of minimal changes outlined above. 

Alternatively, several Chione subgenera 
could be elevated to genus rank. Based on 
the treatment of several of these subgenera 
as genera by previous authors, the elevations 
would minimize the changes necessary to 
convert the taxonomic classification to a phy- 





Anomalocardia 

Mercenaria 

Protothaca 

Timoclea 

Chione 

Chionista 

Chionopsis 

Iliochione 

Lirophora 

Panchione 

Ruber ella 



Mercenaria 

Puherella 

Chionopsis 

Protothaca 

Timoclea 

Chionista 

Chione 

A nomalocardia 

Iliochione 

Lirophora 

Panchione 



FIG. 20. Upper figure illustrates the maximum 
amount of phylogenetic information that can be 
derived from the traditional rank classification. 
Lower figure is the consensus tree of this study, 
showing genera only. 

logenetic one. As discussed earlier. Wood- 
ring (1982) and Ward (1992) both consid- 
ered Lirophora to be of genus rank, and 
Olsson (1964) treated Chionopsis similarly. 
Keen (1969) also elevated Panchione to ge- 
nus rank. Elevations of these subgenera to 
genus rank would therefore not be uncon- 
ventional. 

The changes required to convert the taxo- 
nomic classification to a phylogenetic classi- 
fication begin with the exclusion of Timoclea 
from consideration as a subgenus of Chione 
and recognition of it as a distinct genus. 
Next, the subgenera Chione, Lirophora, Ilio- 
chione, and Panchione are elevated to genus 
rank. Puberella, originally designated a sub- 
genus of Chionopsis (Fischer-Piette & Vuka- 
dinovic, 1977), is elevated to generic rank. In 
the monophyletic clade comprising Liro- 
phora, Anomalocardia, Iliochione and Pan- 
chione, both Iliochione and Panchione must 
be considered separate genera, due primarily 



134 



ROOPNARINE 



TABLE 5. Comparison of traditional classification (left columns) to new phylogenetically 
based classification. Subgenera listed on right are genera sensu stricto. 



OLD CLASSIFICATION 


NEW CLASSIFICATION 


GENUS 


SUBGENUS 


GENUS 


SUBGENUS 


Chione 


Chione 


Chione 


Chione 




Chionista 




Chionista 




Chionopsis 


lliochione 






lliochione 


Lirophora 






Lirophora 


Panchione 






Panchione 


Puberella 






Puberella 


Chionopsis 




Anomalocardia 




Anomalocardia 




Mercenaria 




Mercenaria 




Protothaca 




Protothaca 




Timoclea 




Timoclea 





to the uncertain position of lliochione. The 
alternative would require the subsuming of 
both genera, plus either Lirophora or Anom- 
alocardia, into a large genus classified as ei- 
ther Lirophora or Anomalocardia, on the ba- 
sis of historical precedence. The subgeneric 
rank of Chionista is supported by the reclas- 
sification. Table 5 compares the old taxo- 
nomic classification with the new phyloge- 
netic one. It is important to note here the 
number of taxa (genera) which were formerly 
subsumed in the paraphyletic Chione, and 
the exclusion of other genera on the basis of 
autapomorphies, not patterns of relationship. 

Pliocene Extinctions 

Since Woodring's (1966) initial claim, the 
general picture of late Neogene molluscan di- 
versity in tropical America has been one of 
high Pliocene extinction in the tropical West- 
ern Atlantic compared with the Eastern Pa- 
cific (Vermeij, 1978; Vermeij & Petuch, 1986). 
Coupled with this is the documentation of ex- 
tensive contemporaneous extinctions in the 
subtropical waters off the southeast United 
States (Stanley, 1986). Several mechanisms 
have been suggested to explain the extinc- 
tions, the most commonly cited ones being 
disruption of planktonic productivity levels 
(Vermeij, 1978; Vermeij & Petuch, 1986; Ali- 
món et al., 1993) and Northern Hemisphere 
cooling (Stanley, 1986). It has also been ar- 
gued that the extinctions were in fact faunal 
turnovers, and that losses of diversity have 
been overcompensated for by new origina- 
tions and invasion (Alimón et al., 1993; Jack- 
son et al., 1993; see also Vermeij & Rosen- 



berg, 1993). Jackson et al. 's (1993) data 
(comprising subgeneric rank taxa) also sup- 
port the initiation of the extinctions at 2.4 Ma, 
well into the Late Pliocene. 

The history of the chionine taxa considered 
in this paper agree in general with the sce- 
nario constructed by Jackson et al. (1993), 
with several exceptions. During the entire 
Pliocene, extinction of Atlantic chionine spe- 
cies (restricted from here on to species be- 
longing only to the former Chione subgenera 
and Anomalocardia) exceeded origination 
(Fig. 12). Origination decreases in the Pacific 
during the Pliocene, but is never outweighed 
by extinction. This observation, and a high 
origination rate in the Pacific during the Pleis- 
tocene and Holocene, result in a higher total 
diversity of chionine species in the Eastern 
Pacific today compared to the Western At- 
lantic. Two new supraspecific taxa, Chionista 
and lliochione, also first appear in the Eastern 
Pacific during the Pliocene. Overall, the ex- 
tinction level of chionine species during the 
entire Pliocene was 82.6% in the tropical 
Western Atlantic, but only 38.5% in the Pa- 
cific Gatunian (Fig. 13). Today there are 21 
chionine species in the Eastern Pacific, ver- 
sus 13 in the Western Atlantic {Chione can- 
cellata is considered to comprise two distinct 
"morphological" species, based on Roop- 
narine, 1995). Of the five subgenera present 
in both regions during the Pliocene, all sur- 
vive in the Eastern Pacific today, but only 
three in the Western Atlantic. The conclu- 
sions to be drawn from this study therefore, 
are that extinction of chionine species during 
the Pliocene was significantly higher in the 
Western Atlantic than in the Eastern Pacific, 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



135 



and the loss of diversity was not compen- 
sated for by new originations. 

The timing of a large portion of the extinc- 
tion remains ambiguous. Of Early-middle 
Pliocene Chione species in the Western At- 
lantic 45% were absent from the Late 
Pliocene (Fig. 15). Extinction in the Eastern 
Pacific at that time was a slightly higher 57% 
(though species diversity was lower than in 
the Western Atlantic). During the Late 
Pliocene, however, extinction in the Western 
Atlantic increased to 66.7%, but fell in the 
Eastern Pacific to 11.1%. Dating the extinc- 
tion to the Late Pliocene on the basis of these 
summary calculations would be incorrect, 
though, because the pattern is not reflected 
by all the individual subgenera. Lirophora and 
Panchione exhibit much higher extinction in 
the Early-middle Pliocene (60% and 100% 
respectively) than do the other subgenera 
(Fig. 1 6). It therefore seems that there were at 
least two episodes of extinction. 

The first extinction episode, in the Early- 
middle Pliocene, was associated primarily 
with the subgenera Lirophora and Pan- 
chione. The extinctions were also concen- 
trated in the Atlantic Gatunian region. The 
reason(s) for the differential extinction among 
the subgenera is not obvious. Lirophora and 
Panchione are nested within a subclade sep- 
arate from the subclade to which Chione be- 
longs, but the characters distinguishing the 
two subclades currently have limited func- 
tional interpretations. There are furthermore 
no obvious ecological differences among the 
subgenera. All inhabit fairly shallow, coarse 
to medium-grained sediments, and are sym- 
patric today. Lirophora and Panchione prob- 
ably dwell deeper in the sediment than does 
Chione, as evidenced by their deeper palliai 
sinuses. Anomalocardia, however, disrupts 
any phylogenetic pattern because it experi- 
ences minimal levels of extinction. The 
greater proportion o^ Anomalocardia diversity 
resided in the Caloosahatchian Province, not 
the Atlantic Gatunian region, unlike Lirophora 
and Panchione, and is suggestive of and 
consistent with a geographic extinction pat- 
tern. This possibility is explored more fully 
below. 

Mechanisms of Extinction 

Two expectations of extinction caused 
by cooling would be more severe extinction 
of Atlantic Gatunian species relative to 
Caloosahatchian species, and the higher sur- 



vival of eurythermal (biprovincial) species 
(Stanley, 1984, 1986). According to Cronin's 
(1991, 1993) data, the Northern Hemisphere 
cooling event reported by Stanley (1986) ac- 
tually began 2.6-2.4 Ma. This event coin- 
cided with, and was probably causative of, a 
major regression recorded on the Atlantic 
Coastal Plain, and allows the categorization 
of the Pliocene deposits into Early-middle 
and Late Pliocene groups. Of the species in 
the Atlantic Gatunian during the Early-middle 
Pliocene 40% did not survive into the Late 
Pliocene (Fig. 17). The extinction level in the 
Caloosahatchian Province at that time was 
only 18.75%. Stanley's prediction would 
therefore seem to be supported. Of the Late 
Pliocene species in the Atlantic Gatunian and 
Caloosahatchian 36.4% and 50% respec- 
tively were extinct by the Early Pleistocene. 
At this level of the analysis, the data cannot 
reject Stanley's cooling hypothesis. How- 
ever, the possibility that the surviving species 
were eurythermal cannot be entertained, be- 
cause there were no biprovincial chionine 
species in the Late Pliocene. A possible ex- 
ception would be Chione cancellata, except 
that, as noted above, the Caloosahatchian 
and Caribbean forms may be separate spe- 
cies. The Caloosahatchian form first appears 
in the Upper Pliocene/Pleistocene Wacca- 
maw Formation of South Carolina, and has 
not been documented further south than the 
Florida peninsula. Therefore, no evidence ex- 
ists to support the higher survival of euryther- 
mal species, nor the survival of Caloosahat- 
chian species by southward migration. 

The complete pattern of Pliocene extinc- 
tion of chionine species does not support a 
hypothesis of cooling as a mechanism. Ex- 
tinction is higher in more tropical areas dur- 
ing or soon after the initiation of cooling, but 
the higher level is due to the non-random ex- 
tinction of Lirophora and Panchione species. 
By the Late Pliocene, extinction was distrib- 
uted fairly evenly among all the subgenera. It 
is interesting to note, however, that the two 
Recent paciphilic chionine genera, Chlonop- 
sis s.S. and Panchione, were distributed 
throughout the Atlantic Gatunian, but only as 
far north as Florida during the Pliocene. 
These are the only two taxa to suffer com- 
plete extinction in the Atlantic. Their extinc- 
tions do not appear to support a hypothesis 
of cooling though, because these taxa failed 
to find refuge by migrating equatorward. 

The indirect test of declining planktonic 
productivity as a mechanism of extinction 



136 



ROOPNARINE 



yields more intriguing results. All Early-mid- 
dle Pliocene Atlantic Caloosahatchian spe- 
cies exceeded 35 mm in maximum valve 
height. Only two of eight contemporary At- 
lantic Gatunian species, Chionopsis tegulum 
and Panchione mactropsis, exceeded 35 
mm, and these in fact exceeded 40 mm. Only 
one of the Caloosahatchian species did not 
survive beyond the middle Pliocene. Three of 
the Atlantic Gatunian species survived into 
the Late Pliocene, but they were all small, 
being less than 25 mm in height. Caloosahat- 
chian species, survivors and new Late 
Pliocene species on the other hand remained 
relatively large, all exceeding 30 mm in max- 
imum valve height, and three of six exceed- 
ing 40 mm. Recent Caloosahatchian species 
regularly exceed 30 mm in valve height, but 
none are known to attain heights of 40 mm or 
more. Caribbean species have remained 
small, generally not exceeding 25 mm. Nota- 
ble exceptions occur in areas of upwelling, 
and hence relatively high planktonic produc- 
tivity. Recent Pacific species commonly at- 
tain heights in excess of 50 mm (Fig. 19); 
Chionopsis gnidia is the largest described 
chionine species dealt with in this paper. In 
summary, size distributions have not 
changed very much in the Caloosahatchian 
Province since the Early Pliocene, despite 
changes in species composition. In the Ga- 
tunian Province though, the change has been 
more dramatic. Large species in the Atlantic 
Gatunian region did not survive into the Late 
Pliocene. Species size distribution there re- 
mains small because of the differential ex- 
tinction of the larger Early-middle Pliocene 
species, and apparently because no large 
(> 35 mm) species have evolved in that re- 
gion since. The opposite is true of the Pacific 
Gatunian, or Panamic Province. Post Early- 
middle Pliocene species there tend to be very 
large, perhaps a consequence of the wide- 
spread coastal upwelling in that region. 

Given the equivocal nature of the results of 
the above tests, any hypothesized cause(s) 
of chionine extinction during the Pliocene in 
the Western Atlantic is speculative. More- 
over, the data are limited to only those gen- 
era examined, and the availability of material 
documented in the field, museum collec- 
tions, and the literature. Regardless, the pat- 
tern of extinction can be explained tentatively 
by current geological data which appear to 
support a decline in productivity, at least in 
the Early-middle Pliocene. 

The loss of Early-middle Pliocene large- 



and small-bodied species from the Atlantic 
Gatunian, representing 40% of the species, 
contrasts strongly with the contemporary 
Caloosahatchian. Only 18.75% of Early-mid- 
dle Pliocene Caloosahatchian species did 
not survive into the Late Pliocene, and there 
is no reduction in overall body size. There is 
therefore an indication of an Early-middle 
Pliocene episode of extinction in the Atlantic 
Gatunian that did not have a great impact on 
the Caloosahatchian Province. The Late 
Pliocene extinctions in the Atlantic Gatunian 
are matched in severity, however, by the Late 
Pliocene Caloosahatchian extinctions, but 
there is no change in overall body size in ei- 
ther province. It is therefore possible that 
there were two episodes of extinction in the 
tropical Western Atlantic during the Pliocene 
(Petuch, 1995). 

The earlier extinction in the Atlantic Gatu- 
nian probably followed the final closure of the 
Panama seaway (-3.5 Ma). Final closure oc- 
curred during the earliest Pliocene when the 
shallowing of the seaway was already dra- 
matic (< 100 m depth, Duque-Caro, 1990; 
see also Coates et al., 1992). The shallowing 
undoubtedly resulted in the fragmentation of 
once contiguous and widespread popula- 
tions, but perhaps more importantly, it also 
changed the océanographie configuration of 
the Caribbean region. It has not yet been de- 
termined how changes in circulation and the 
decline and eventual termination of flow from 
the Atlantic to the Pacific may have affected 
local diversity. The presence of seasonal cool 
water in the Caribbean during the middle 
Pliocene (Cronin, 1991) unlike today, the ex- 
istence of relict communities in the Carib- 
bean in areas of upwelling today (Petuch, 
1982), and the loss of large bodied species 
from the Atlantic Gatunian (this paper) how- 
ever, all suggest a decline in planktonic pro- 
ductivity. Interestingly, though. Lower-middle 
Pliocene deposits in the Caloosahatchian 
Province show no signs of significant extinc- 
tion or loss of large bodied species. This 
would suggest that the extinctions did not 
extend to, or did not affect the Caloosahat- 
chian Province significantly. 

The Late Pliocene extinctions of the Atlan- 
tic Gatunian and Caloosahatchian provinces 
coincide with the initiation of Northern Hemi- 
sphere cooling (2.5-2.4 Ma) (Stanley, 1986; 
Cronin, 1991, 1993). Neither extinction, how- 
ever, exhibits pattern expected of cooling 
scenarios. The only noticeable difference be- 
tween Late Pliocene and Recent faunas, be- 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



137 



sides the lower diversity of modern faunas, is 
the absence of very large-bodied species 
(> 40 mm) in the Recent. Large species, or 
large specimens (30-35 mm), survive in the 
Caloosahatchian region, and in areas of up- 
welling in the Caribbean. This observation 
suggests that declining planktonic productiv- 
ity may again have been a factor, but perhaps 
not the only, or primary one. 



SUMMARY 

(1) Phylogenetic analysis shows that the 
genus Chione as currently defined is para- 
phyletic. A monophyletic clade comprises 
the genera Chione (excluding the subgenera 
Chionopsis and Puberella), and Anomalocar- 
dia. Chione has been revised in order to con- 
struct a taxonomic classification that is con- 
sistent with the results of the phylogenetic 
analysis. The genus was revised and re- 
placed by the genera (formerly subgenera) 
Chione (s.S.), Chionopsis, Lirophora, II io- 
chione, Panchione and Chionista. The treat- 
ments of Chionopsis as a genus (Olsson, 
1964), similarly Lirophora (Woodring, 1982; 
Ward, 1992), and the consideration of Pan- 
chione as a subgenus separate from Liro- 
phora (Keen, 1969), support the revision. 

(2) Sculptural characters, primarily the re- 
tention of radial sculpture, and the marginal 
elaborations of concentric sculpture, are di- 
agnostic of the "Chione subclade" (comphs- 
ing the genera Chione and Chionista). The 
"Lirophora subclade" is defined more 
strongly by a combination of characters de- 
scribing internal as well as external valve 
morphology, most notably aspects of con- 
centric sculpture and nymph rugosity. 

(3) During the late Neogene the origination 
of new chionine species rose steadily in the 
Eastern Pacific. The number of last appear- 
ances increases briefly during the Pliocene, 
but never exceeds the number of first ap- 
pearances. Of Early-middle Pliocene species 
in the Atlantic Gatunian Province 40% do not 
survive into the Late Pliocene, compared with 
18.75% of Early-middle Pliocene Caloosa- 
hatchian species. The higher extinction in the 
Atlantic Gatunian region is accounted for pri- 
marily by the seemingly non-random extinc- 
tion of species assigned to the genera Liro- 
phora and Panchione. During the Late 
Pliocene however, 36.4% and 50% of spe- 
cies in both the Atlantic Gatunian and 
Caloosahatchian Provinces respectively are 



last appearances, and these extinctions 
seem to be distributed randomly among the 
surviving genera. This wave of extinction be- 
gan at least 2.5 million years ago. 

(5) A hypothesis of cooling as a mechanism 
of extinction is not supported by these data. 
Extinctions caused by cooling would have 
been restricted to the Late Pliocene because 
of the late date of the onset of Northern 
Hemisphere refrigeration (Cronin, 1991, 
1993). Extinction is not more severe in the 
fully tropical Atlantic Gatunian Province than 
the sub-tropical Caloosahatchian. Moreover, 
there is no evidence that Caloosahatchian 
species survived the crisis by southward mi- 
gration. Some equivocal support might be 
provided by the observation that neither of 
the two paciphilic genera examined, Chion- 
opsis (s.S.) and Panchione, extended further 
north than Florida during the Pliocene. 

(6) The loss of all large bodied (> 35 mm) 
species from the Early-middle Pliocene At- 
lantic Gatunian Province is consistent with a 
hypothesis of declining planktonic productiv- 
ity. During the Late Pliocene, however, body 
size did not seem to be a very discriminating 
factor, since small bodied Atlantic Gatunian 
species also suffered extinction. Modern 
species have maximum valve heights, in the 
Caloosahatchian and Caribbean, ranging 
from 25 to 35 mm. On the other hand, Late 
Pliocene Caloosahatchian species exceed- 
ing 40 mm in maximum valve height suffered 
100% extinction. Therefore, it seems that 
large bodied species did suffer differentially 
higher levels of extinction during the 
Pliocene. The mechanism responsible is hy- 
pothesized to be a disruption of planktonic 
productivity patterns and levels. Additional, 
indirect evidence is provided by the Late and 
post-Pliocene evolution of very large chion- 
ine species in the upwelling-hch Panamic 
Province. 



ACKNOWLEDGMENTS 

I wish to thank the following individuals for 
guidance throughout the completion of this 
project: Geerat Vermel], Sandra Carlson and 
Mark Patterson. The following individuals 
were also helpful in providing comments and 
assistance: Jay Schneider, Kim Driver, Scott 
Gardner, Stanley Bürsten, William Coles, 
Eleanor Villar and Richard Cowen. I thank R. 
Bieler (Field Museum of Natural History), H. 
Lescinsky (University of California, Davis), R. 



138 



ROOPNARINE 



Porteil (Florida Museum of Natural History), 
G. Vermeij (University of California, Davis) 
and E. Vokes (Tulane University) for the kind 
loan and contribution of specimens. This 
work was supported by grants from the De- 
partment of Geology, University of California 
Davis, The Center for Population Biology, 
University of California Davis, the Lerner- 
Gray Fund for Research, Sigma Xi and the 
Paleontological Society. 



LITERATURE CITED 

ABBOTT, R. T., 1974, American seashells. Von 
Nostrand Reinhold. New York. 

ALLMON, W. D., G. ROSENBERG, R. W. PORTELL 
& K. S. SCHINDLER, 1993, Diversity of Atlantic 
Coastal Plain mollusks since the Pliocene. Sci- 
ence, 260: 1626-1629. 

BROWN, J. H., 1975, Geographical Ecology of 
Desert Rodents. Pp. 31 5-341 , in: M. L. Cody & J. 
M. Diamond, eds., Ecology and evolution of 
connmunities. The Belknap Press of Harvard Uni- 
versity. 

BRYANT, J. D., B. J. MACFADDEN & P. A. MUEL- 
LER, 1992, Innproved chronologic resolution of 
the Hawthorn and the Alum Bluff groups in 
northern Florida: implications for Miocene chro- 
nostratigraphy. Geological Society of America 
Bulletin, 104: 208-218. 

COATES, A. G., J. B. С JACKSON, L. S. COLLINS, 
T M. CRONIN, H. J. DOWSETT, L. M. BYBELL, 
P. JUNG, J. A. OBANDO, 1992, Closure of the 
Isthmus of Panama: the near-shore marine 
record of Costa Rica and western Panama. Geo- 
logical Society of America Bulletin, 104: 
814-828. 

CRONIN, T M., 1991, Pliocene shallow water pa- 
leoceanography of the North Atlantic Ocean 
based on marine ostracodes. Quartenary Sci- 
ence Reviews, 10: 175-188. 

CRONIN, T. M., 1993, The Yorktown Formation of 
Virginia: implications for Late Pliocene climate 
and sea level history. Geological Society of 
America, Abstracts w/ith Programs, 25(6): A-334. 

DEQUEIROZ, K. & J. GAUTHIER, 1992, Phyloge- 
netic taxonomy. Annual Review/ of Ecology and 
Systematics, 23: 449-480. 

DOCKERY, D. T. 1982, Lower Oligocène Bivalvia 
of the Vicksburg Group in Mississippi. Missis- 
sippi Department of Natural Resources. Bureau 
of Geology, Bulletin, 123: 261 pp., 77 pis. 

DODD, J. R. & R. J. STANTON, Jr., 1990, Paleo- 
ecology. Concepts and applications. John Wiley 
& Sons. 

DUQUE-CARO, H., 1990, Neogene stratigraphy, 
paleoceanography and paleobiogeography in 
northwest South America and the evolution of 
the Panama Seaway. Palaeogeography, Palaeo- 
climatology, Palaeoecology, 77: 203-234. 



DURHAM, J. W., 1950 Megascopic paleontology 
and marine stratigraphy. Thie Geological Society 
of America, Memoir, 43: 216pp. 

FISCHER-PIETTE, E., & A. D. VUKADINOVIC, 
1977, Suite des revisions des Veneridae (mol- 
lusques lamellibranches) Chioninae, Samarangi- 
nae et complement aux Venus. /Wemo/res du 
l\/luseum National d'Histoire Naturelle (A-Zoolo- 
gie), (п. s.) 106: 186 pp., 22 pis. 

FISHER, D. С, 1981 , The role of functional analysis 
in phylogenetic inference: examples from the 
history of the Xiphosura. American Zoologist, 21 : 
47-62. 

FISHER, D. C, 1985, Evolutionary morphology: be- 
yond the analogous, the anecdotal, and the ad 
hoc. Paleobiology, 11: 120-138. 

GARDNER, J., 1926, The molluscan fauna of the 
Alum Bluff Group of Florida. Pt. IV. Veneracea. 
United States Geological Survey, Professional 
Paper, 142D: 151-185, pis. 24-28. 

GRANT III, U. S. & H. R. GALE, 1 931 , Pliocene and 
Pleistocene mollusca of California and adjacent 
regions. /Wemo/rs of the San Diego Society of 
Natural History, 1 . 959Pp. 

HALL, B. K., 1992, Evolutionary developmental bi- 
ology. Chapman and Hall. 

HARLAND, W. В., R. L. ARMSTRONG, A. V. COX, 
L. E. CRAIG, A. G. SMITH, & D. G. SMITH, 1990, 
A geological time scale 1989. Cambridge Univer- 
sity Press, Cambridge. 

HARTE, M. е., 1992a, An eastern Pacific [Cerce- 
narla and notes on other chionine genera (Bi- 
valvia: Veneridae). The Veliger, 35: 137-140. 

HARTE, M. е., 1 992b, A new approach to the study 
of bivalve evolution. American Malacological 
Bulletin. 9: 199-206. 

HERTLEIN, L. G. & A. G. STRONG, 1948, Mollusks 
from the west coast of Mexico and Central 
America. Zoológica, 33 (pt. 4: 13). 

HERTLEIN, L. G. & U. S. GRANT III, 1972, The 
geology and paleontology of the marine Pliocene 
of San Diego, California; Paleontology, Pelecy- 
poda. San Diego Society of Natural History, 
Memoir, 2B: 144-409. 

JACKSON, J. B. C, P. JUNG, A. G. COATES & 
L. S. COLLINS, 1993, Diversity extinction of 
tropical American mollusks and emergence of 
the Isthmus of Panama. Science, 260: 1624- 
1626. 

JONES, D. S. & W. D. ALLMON, 1995, Records of 
upwelling, seasonality and growth in stable-iso- 
tope profiles of Pliocene mollusk shells from 
Flonda. Lethaia, 28(1): 61-74. 

JONES, D. S. & P. F. HASSON, 1985, History 
and development of the marine invertebrate fau- 
nas separated by the Central American isthmus. 
In: F. G STEHLi & s. D^ WEBB, eds., The great Amer- 
ican biotic interchange. Plenum Press. New 
York. 

JONES, D. S., B. J. MACFADDEN, S. D. WEBB, P. 
A. MUELLER, D. A. HODELL, & T M. CRONIN, 
1991, Integrated geochronology of a classic 
Pliocene fossil site in Florida: linking marine and 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



139 



terrestrial biochronologies. The Journal of Geol- 
ogy, 99(5): 637-648. 

JUNG, P., 1969, Miocene and Pliocene mollusks 
from Trinidad. Bulletins of American Paleontol- 
ogy, 55: 293-657. 

KEEN, A. M., 1969, Treatise on Invertebrate Pale- 
ontology. Part N. Mollusca 6. Bivalvia, Vol 2. R. 
С Moore, ed. The Geological Society of America 
and The University of Kansas. 

KEEN, A. M., 1971, Sea shells of tropical west 
Annerica. Stanford University Press, California. 
1064 pp. 

KRANTZ, D. E., 1991, A chronology of Pliocene 
sea-level fluctuations, U. S. Atlantic Coastal 
Plain. Quaternary Science Reviews, 1 0: 1 63-1 74. 

LYONS, W. G., 1991, Post-Miocene species of 
Latlrus Montfort, 1810 (Mollusca: Fasciolariidae) 
of southern Florida, with a review of regional ma- 
rine biostratigraphy. Bulletin of Florida Museum 
of Natural History, 35: 131-208. 

MACARTHUR, R. H., 1965, Patterns of species di- 
versity. Biological Review, 40: 510-533. 

OLSSON, A. A., 1942, Tertiary and Quaternary fos- 
sils from the Burica Peninsula of Costa Rica and 
Panama. Bulletin of American Paleontology, 
26(106): 106 pp. 

OLSSON, A. A., 1961, Mollusks of the tropical 
Eastern Pacific. Panamic-Pacific Pelecypoda. 
Paleontological Research Institution. Ithaca, 
New York. 574 pp. 

OLSSON, A. A-, 1964, Neogene mollusks from 
northwestern Ecuador. Paleontological Re- 
search Institution. Ithaca, New York. 256 pp. 

OLSSON, A. A. & A. HARBISON, 1953, Pliocene 
Mollusca of southern Florida. The Academy of 
Natural Sciences of Philadelphia Monographs, 8: 
457 pp. 

PALMER, K., 1927, The Veneridae of Eastern 
America, Cenozoic and Recent. Palaeonto- 
graphica Americana, 1: 209-522. 

PARKER, P., 1 949, Fossil and recent species of the 
pelecypod genera Chione and Securella from the 
Pacific Coast. Journal of Paleontology, 23: 577- 
593. 

PERRILLAT, M. d. C, 1963, Moluscos de la For- 
macion Aquequexquite (Miocene medio) del 
Istmo de Tehuantepec, Mexico. Paleontología 
Mexicana, 14: 45 pp., 6 pis. 

PETERS, R. H., 1983, The ecological implications 
of body size. Cambridge Studies in Ecology. 
Cambridge University Press. Cambridge. 

PETUCH, E. J., 1982, Geographical heterochrony: 
Contemporaneous coexistence of Neogene and 
Recent molluscan faunas in the Americas. Palae- 
ogeography, Palaeoclimatology, Palaeoecology, 
37: 277-312. 

PIANKA, E. R., 1975, Niche Relations of desert liz- 
ards. Pp. 292-314, in: M. L. CODY & J. M. DIA- 
MOND, eds., Ecology and evolution of communi- 
ties. The Belknap Press of Harvard University 
Press. 

RAUP, D. M. 1975, Taxonomic diversity estimation 
using rarefaction. Paleobiology, 1: 333-342. 



RAYMO, M. E., D. RIND & W. F. RUDDIMAN, 1990, 
Climatic effects of reduced Arctic sea ice limits 
in the GISS II General Circulation Model. Paleo- 
ceanography, 5: 367-382. 

ROOPNARINE, P. D., 1995, A re-evaluation of sta- 
sis between the bivalve species Chione 
erosa — C. cancellata (Bivalvia: Veneridae). Jour- 
nal of Paleontology, 69: 280-287. 

SANDERS, H. L. 1968, Marine benthic diversity: a 
comparative study. American Naturalist, 102: 
243-282. 

SAUNDERS, J. В., P. JUNG & B. BIJU-DUVAL, 
1986, Neogene paleontology in the northern Do- 
minican Republic, I. Field surveys, lithology, en- 
vironment, and age. Bulletins of American Pale- 
ontology, 89(323): 1-79. 

SHACKLETON, N. J. & M. A. HALL, 1985, Oxygen 
and carbon isotope stratigraphy of Deep Sea 
Drilling Project Hole 552A: Plio-Pleistocene gla- 
cial history. In: D. G. ROBERTS & D. Schnitker, 
eds., Initial Reports of the Deep Sea Drilling 
Project, 81 : 599-609. 

STAFF, G. M., E. N. POWELL, R. J. STANTON JR. 
& H. CUMMINS 1985, Biomass: Is it a useful tool 
in paleocommunity reconstruction? Lethaia, 18: 
209-232. 

STANLEY, S. M., 1984, Marine mass extinctions: a 
dominant role for temperatures. Pp. 69-117 in: 
M. H. NiTECKi, ed.. Extinctions. The University of 
Chicago Press, Chicago. 

STANLEY, S. M., 1986, Anatomy of a regional 
mass extinction: Plio-Pleistocene decimation of 
the Western Atlantic bivalve fauna. Palaios, 1 : 
1 7-36. 

STANTON, R. J., JR. & P. NELSON, 1980, Recon- 
struction of the trophic web in paleontology: 
community structure in the Stone City Formation 
(Middle Eocene, Texas). Journal of Paleontology, 
54: 118-135. 

STANTON, R. J., JR., E. N. POWELL & P. NELSON 
1981, The role of carnivorous gastropods in the 
trophic analysis of a fossil community. Malaco- 
logia, 20: 451-469. 

SWOFFORD, D. L., 1985, PAUP Version 2.4.1. 

SWOFFORD, D. L., 1 991 , Phylogenetic analysis us- 
ing parsimony (PAUP), version 3.0s. Illinois Nat- 
ural History Survey. Champaign. 

THIEL, H., 1 975, The size structure of the deep-sea 
benthos. International Revue der Gesamten Hy- 
drobiologie, 60: 575-606. 

VERMEIJ, G. J., 1978, Biogegraphy and adapta- 
tion: patterns of marine life. Harvard University 
Press, Cambridge. 

VERMEIJ, G. J., 1987, Evolution and escalation, an 
ecological history of life. Princeton University 
Press. Princeton, New Jersey, xiii + 332 pp. 

VERMEIJ, G. J., 1993, The biological history of a 
seaway. Science, 260: 1603-1604. 

VERMEIJ, G. J. & E. J. PETUCH, 1986, Differential 
extinction in tropical American molluscs: ende- 
mism, architecture, and the Panama land bridge. 
Malacologia, 17: 29-41. 

VERMEIJ, G. J. & G. ROSENBERG 1993, Giving 



140 



ROOPNARINE 



and receiving — The tropical Atlantic as donor 
and recipient region for invading species. Amer- 
ican Malacological Bulletin, 10: 181-194. 

WARD, L. W., 1992, Molluscan Biostratigraphy of 
the Miocene, Middle Atlantic Coastal Plain of 
Nortfi America. Virginia Museum of Natural His- 
tory, Memoir, 2. 159 pp. 

WEYL, P. K., 1968, The role of the oceans in cli- 
matic change: a theory of the ice ages. Meteo- 
rological Monographs, 8(30): 37-62. 

WILEY, E. O., 1981, P^y/ogfenef/cs. The theory and 
practice of phylogenetic systematics. Wiley-Liss. 
John Wiley & Sons, Inc. 

WILEY, E. O., D. SIEGEL-CAUSEY, D. R. BROOKS 
& V. A. FUNK, 1991, The compleat cladist: a 
primer of phylogenetic procedures. The Univer- 
sity of Kansas Museum of Natural History, Spe- 
cial Publication 19. 

WOODRING, W. P., 1966, The Panama land bridge 
as a sea barrier. American Philosophical Society 
Transactions, 110: 425-433. 

WOODRING, W. P., 1982, Geology and Paleontol- 
ogy of Canal Zone and adjoining parts of Pan- 
ama. United States Geological Survey Profes- 
sional Paper, 306-F: 757 pp. 

Revised Ms. accepted 6 February 1996 



APPENDIX I 



Left Valve 



1 . Depth of pallia! sinus: = greatly reduced 
to absent; 1 = present but short; 2 = palliai 
sinus deep, anterior end to beneath posterior 
cardinal tooth; 3 = very deep, past beneath 
posterior cardinal. 

The most common condition of the palliai 
sinus in chionine bivalves is moderately to 
very deep. Examples can be found in the 
genera Mercenaria and Timoclea. Many other 
taxa, for example Lirophora, have very short 
sinuses. An intermediate condition can be 
found in Puberella and Chionopsis. Neither 
Chione nor Petenopsis have a recognizable 
palliai sinus. 

2. Types of sculpture present: = concentric 
elements only; 1 = radial and concentric 
elements, co-dominant; 2 = radiais and con- 
centrics, concentrics dominant, radiais sub- 
obsolete; 3 = radiais and concentrics, both 
sub-obsolete; 4 = radiais and concentrics, 
radiais dominant. 

There are two types of sculptural elements 
in the Chioninae, radial and concentric. Both 
forms are present and co-dominant in Chione 
and Chionopsis (including Puberella). The 



concentric elements in these genera are thin, 
raised lamellae, whereas the radiais are 
raised and cordlike. The radiais are absent in 
Mercenaria, but this appears to be a derived 
autapomorphic condition, not homologous 
with the "concentric-dominant" sculpture of 
Lirophora, Anomalocardia, lliochlone and 
Panchione; Mercenaria's concentrics are of 
the Chione and Chionopsis type. Lirophora, 
Anomalocardia, lliochione and Panchione all 
have greatly reduced or absent radial ele- 
ments, and the development of thick, folded 
concentric lamellae. LeuKoma and Timoclea, 
on the other hand, lack concentric elements 
almost entirely, but have radial sculpture very 
similar to the Chione and Chionopsis. The 
sculptures of Chionista and Petenopsis seem 
to be autapomorphic. 

3. Spacing of concentric elements: = close 
(narrow spacing); 1 = widely spaced; 2 = wide 
spacing on older, juvenile shell, but tending 
to become more narrowly spaced with in- 
creasing age. 

4. Posterior development of concentric ele- 
ments: = flared in ventral direction; 1 = 
weakly flared, but flattened; 2 = no flare. 

The posterior end of concentric elements is 
highly developed in many chionine taxa, pos- 
sibly for functional reasons associated with 
shallow burrowing. 

5. Anterior development of concentric ele- 
ments: = no anterior projection; 1 = devel- 
oped anteriorly; 2 = weak anterior develop- 
ment. 

The concentric elements are developed 
anteriorly in many chionine taxa, and these 
may aid during burrowing, probably acting as 
ratchets, but this has not been rigorously 
tested. 

6. Structure of concentric elements: = con- 
centric elements smooth; 1 = ventral surface 
of concentric elements bearing closely 
packed vertical ribs; 2 = ventral surface of 
concentric elements bearing widely spaced 
vertical ribs; 3 = concentric elements present 
only as raised structures on radial elements. 

7. Distal edge of concentric elements: = 
sharp; 1 = smooth; 2 = smooth and rein- 
forced. 

The distal edges of the concentric ele- 
ments are either thin and sharp, or thickened 



NEOGENE EXTINCTION OF TROPICAL AMERICAN BIVALVES 



141 



and smooth. Some, as in Puberella, are thin, 
but have a thickened, reinforced summit. 

8. Orientation of concentric elements: = 
vertical; 1 - folded; 2 = step-like and 
rounded. 

This character describes the orientation of 
concentric elements relative to the surface of 
the valve. 

9. Definition of escutcheon: = escutcheon 
developed and set off from rest of valve by 
obvious ridge ("keel"); 1 = escutcheon de- 
veloped but weakly keeled; 2 = escutcheon 
not developed. 

10. Width of anterior cardinal tooth: = wide; 
1 = narrow. 

The anterior cardinal is either well devel- 
oped and wide, as in Chione, or thin and 
blade-like, as in Chionopsls. 

11. Nymph: = rugose; 1 = smooth. 

The nymph, a platform posterior to the 
posterior cardinal tooth, houses the ligament 
in venerid species. The nymph is generally 
smooth, but can have a roughened or "rug- 
ose" surface. The state of this character is 
sometimes given predominant weight in de- 
termining chionine relationships (for exam- 
ple, Harte 1992a), but nymph rugosity, of 
variable morphology, is present in non-chio- 
nine taxa, for example Pitar (Lamelliconcha) 
Dall, 1902. Nymph rugosity is therefore ho- 
moplastic at some levels. 

12. Ventral margin crenulation: = large; 1 = 
fine; 2 = intermediate between and 1 , and 
regular. 

13. Lunule sculpture: = numerous concen- 
tric elements; 1 = few concentric elements; 2 
= smooth; 3 = concentric elements with sub- 
dominant radial ribs; 4 = sub-obsolete con- 
centric and radial elements; 5 = numerous 
radial ribs; 6 = lunule not developed. 

14. Middle cardinal tooth morphology: = 
bifid; 1 = smooth; 2 = smooth, except for 
dorso-ventral groove. 

15. Posterior cardinal tooth shape: = 
straight; 1 = curved; 2 = weakly curved. 

16. Shape of hinge plate margin: = plate 
very bowed beneath anterior cardinal tooth; 1 
= plate weakly bowed, not obvious; 2 = plate 



bowed, not exaggerated as in 0; 3 = plate 
straight. 

Many chionine species have hinge plates 
that are noticeably bowed, or curved, be- 
neath the anterior cardinal tooth. An extreme 
example of this can be found in Mercenaria. 
Other taxa have straight margins, and there is 
a continuum between the two states. 

Right Valve 

17. Definition of escutcheon: = escutcheon 
developed, but not demarcated noticeably 
from rest of valve surface; 1 = escutcheon 
developed and demarcated from rest of valve 
by keel; 2 = escutcheon not developed. 

18. Sculpture of escutcheon: = concentri- 
cally sculptured; 1 = escutcheon smooth; 2 = 
sub-obsolete concentric sculpture; 3 = es- 
cutcheon absent; 

19. Width of middle tooth: = tooth narrow; 
1 = tooth wide. 

20. Condition of middle cardinal tooth: = 
tooth bifid; 1 = tooth grooved on dorsoventral 
axis; 2 = tooth smooth. 

21. Condition of posterior cardinal tooth: = 
tooth bifid; 1 = tooth grooved on dorsoventral 
axis; 2 = tooth smooth. 

22. Orientation of groove on posterior margin: 
= groove just overlaps ventral tip of posterior 
cardinal tooth; 1 = groove distal to posterior 
cardinal tooth; 2 = groove abuts posterior car- 
dinal tooth; 3 = groove overlaps posterior car- 
dinal tooth significantly; 4 = groove absent. 

This groove is present in many chionine 
taxa on the right valve, and houses the pos- 
terior margin of the left valve when the shell is 
closed. 

23. Condition of radial ribs between concen- 
tric sculpture: = no radial ribs present; 1 = 
radial ribs prominent; 2 = radial ribs present 
but fine, tending to become obsolete. 

General Shell Morphology 

24. Escutcheon symmetry: = escutcheon 
symmetric between valves; 1 = escutcheon 
asymmetric between valves; 2 = escutcheon 
absent. 

The escutcheon is generally of different 



142 



ROOPNARINE 



morphology between valves in chionine taxa, 
but is sometimes identical. 

25. Development of valve surface near pos- 
terior margin: = identical to rest of valve 
surface; 1 = sculpture and valve surface dif- 
ferentially developed as posterior margin ap- 
proached. 



Some taxa have significant changes of 
sculpture towards the posterior margin. 
Much of the change is in concentric sculp- 
tural morphology, as in Panchione mactrop- 
sis, where the folded concentric sculpture 
becomes lamellar. Iliochione has an indenta- 
tion of the valve. 



APPENDIX II 
(see APPENDIX I for explanation of numerical codes) 



"9"= missing data 
Chione cancellata 
Chione chipolana 
Chionopsis amathusia 
Puberella cribraria 
Lirophora athleta 
Lirophora victoria 
Panchione mactropsis 
Panchione ulocyma 
Iliochione subrugosa 
Chionista fluctifraga 
Petenopsis tumens 
Mercenaria mercenaria 
Anomalocardia flexuosa 
Protothaca asperrima 
Timoclea marica 
Anomalocardia auberiana 



0120210000103221001212110 
1100212000106121001212110 
2110110011101011110101101 
21 001 1 2001 1 000200001 1 021 
1 002001 101012121 01 021 3001 
1000001101912121019919001 
1201211101012221011210201 
1 20221 1 1 01 01 222301 1 21 0201 
1202011101012101011223201 
2302021 1 201 25002231 0041 21 
0112011200112121011222100 
200000000000000000000001 
0202001 1 1 1 01 21 03020221 001 
2402030001 1 1 4201 011112111 
3401 030001 1 041 031 21 1 241 00 
1202021111011101021221201 



APPENDIX III 

Synapomorphies for interior nodes, cladogram #1. Results of both ACCTRAN and DELIRAN 
routines are listed. Interpret listings as (x:y) = (character:state). ACCTRAN 











Synapomorph 


es 


Node 


ACCTRAN 








DELTRAN 


17 


4:0, 5:2, 7:0, 13:3, 18:0, 


24:1 




4:0, 5:2, 18:0, 24:1 


18 


1:0, 8:0, 15:2, 25:0 








1:0, 15:2, 25:0 


19 


10:0 








10:0 


20 


2:0, 6:0, 19:0, 23:0 








2:0, 6:0, 23:0 


21 


5:2, 14:2, 22:0 








5:2, 14:2, 22:0 


22 


15:2 








15:2 


23 


6:0, 9:1, 18:2, 22:1 








9:1, 18:2, 22:1 


24 


21:2 








21:2 


25 


1:1, 2:2, 11:0, 12:1, 


22:3 


, 23:2 




1:1, 2:2, 8:1, 11:0, 12:1, 22:3, 23:2 


26 


7:1, 8:1, 20:2 








7:1, 13:2, 20:2 


27 


2:4, 6:3, 13:4 








2:4, 6:3, 13:4 


28 


4:2, 13:2, 14:1, 19:1 


, 21 


1, 22:2 




4:2, 14:1, 19:1, 21:1, 22:2 


29 


13:1, 16:1, 18:1, 22:1, 23:1, 24:0, 


25:1 


16:1, 18:1, 23:1, 24:0, 25:1 



MAUXCOLOGIA, 1996, 38(1-2): 143-151 

THE GENITAL SYSTEM OF ACOCHLIDIUM FÍJENSE (OPISTHOBRANCHIA: 
ACOCHLIDIOIDEA) AND ITS INFERRED FUNCTION 

Martin Haase^ & Erhard Wawra^ 

ABSTRACT 

The genital system of the adolescent-gonochoric freshwater opisthobranch Acochlidium 
fijiense is described from histological serial sections of five individuals and dissection of a sixth 
animal in full detail. The penis has a characteristic armature consisting of an ascending spiral 
of chitinous spines on the edge of the glans. The basal finger in association with the parapros- 
tate probably functions as stimulatory organ analogous to the gypsobelum of pulmonate gas- 
tropods. The presence of sperm in the haemocoel and the kidney of one specimen and the 
penial armature suggest that /4. fijiense transfers sperm through hypodermic impregnation. The 
most peculiar feature is the connection of the genital system with the digestive system. A duct 
with unknown function connects the digestive gland with the distal gonoduct. In addition, in one 
individual the ampulla which stores autosperm had an opening into the digestive gland. This 
opening is interpreted as a temporary structure established only when required in order to 
digest excess autosperm, thus compensating the lack of a gametolytic gland. However, it 
cannot be ruled out that this connection seen in a single individual was an abnormality. 

Key words: Acochlidium, genital system, hypodermic impregnation, Opisthobranchia, sperm 
transfer, stimulatory organs 



INTRODUCTION 



MATERIALS AND METHODS 



The vast majority of opisthobranch gastro- 
pods are marine. Up to now only seven spe- 
cies are known from freshwater habitats. 
These seven species all belong to the order 
Acochlidioidea. All marine acochlidioidean 
species are smaller than 5 mm. All but one 
species of the freshwater forms, on the other 
hand, exceed 15 mm. The exception is the 
Caribbean Tantulum elegans Rankin, 1979, 
which lives interstitially (Rankin, 1979). The 
large species — Strubellia paradoxa (Strubell, 
1892) and five species of the genus Acoch- 
lidium Strubell, 1 892 [Following Wawra (1 989), 
we use a conservative classification and reject 
Rankin's (1979) taxonomic splitting.] — occur 
on islands in the Pacific region (Haynes & 
Kenchington, 1991). Despite the size of the 
Pacific freshwater species, which would make 
anatomical investigation and maintenance 
and observation in aquaria rather easy com- 
pared to small, interstitial snails, relatively little 
is known on both their anatomy and biology. 
The present study gives a detailed anatomical 
description of the genital system of /A. fijiense 
Haynes & Kenchington, 1991, and allows in- 
ferences on its function and the reproductive 
biology of this species. 



Five individuals from the series of para- 
types of A. fijiense from Vanua Levu, Fiji, de- 
posited in the mollusc collection of the Mu- 
seum of Natural History in Vienna by Haynes 
& Kenchington (1991) under the inventory 
number 84901 were embedded in Paraplast 
and serially sectioned, one specimen at 7 цт 
and the remaining four at 10 цт. The series 
were stained with Heidenhain's Azan. The 
fixed (Bouin) snails measured 6.1 mm, 7.4 
mm, 7.9 mm, and 8.03 mm respectively. One 
specimen could not be measured because it 
had the visceral hump turned down. These 
individuals belonged to the largest among 
the series of paratypes. In the following, they 
will be referred to as snails number 1 to 5 
beginning with the smallest animal. We do 
not proceed on the assumption that these 
snails represent a developmental sequence, 
because they might have had contracted to 
differing degrees at fixation. The genital sys- 
tem was reconstructed using the computer 
program PC3D of Jandel Scientific. The pe- 
nis of a sixth paratype from the same lot was 
dissected, critical point dried and investi- 
gated by scanning electron microscopy 
(SEM). 



'Institut für Zoologie der Universität Wien, Althanstrasse 14, A-1090 Wien, Austna. 

^Erhard Wawra, 3. Zoologische Abteilung, Naturhistorisches Museum Wien, Burgring 7, A-1014 Wien, Austria. 



143 



144 



HAASE & WAWRA 




FIG. 1. Reconstruction of the genital system ex- 
cept gonad from dorsal, am = ampulla; de = ductus 
ejaculatorius; dgd = distal gonoduct; fgm = female 
gland mass; pd = paraprostatic duct; pg = prae- 
ampullary gonoducts; ppr = paraprostate; pr = 
prostate; ps = penial sheath; to = [presumptive (see 
Discussion)] temporary opening of the ampulla into 
the digestive gland; vd = vas deferens. Scale bar = 
500 |im. 




FIG. 3. Schematic representation of the genital 
system and its connection with the digestive gland. 
Accessory organs not drawn, am = ampulla; dd = 
duct connecting digestive gland and distal gono- 
duct; de = ductus ejaculatorius; dg = digestive 
gland; dgd = distal gonoduct; fgm = female gland 
mass; go = gonad; in = intestine; oe = oesophagus; 
pg = praeampullary gonoduct; po = postampullary 
gonoduct; pr = prostate; st = stylet; to = [presump- 
tive (see Discussion)] temporary opening of the 
ampulla into the digestive gland; vd = vas deferens. 



dg 



Г, ,;k ■■«««, .' «Si 



d*"^ 




FIG. 2. Distal genital system from latero-dorsal. dd 
= duct connecting digestive gland and distal gono- 
duct; de = ductus ejaculatorius; dgd = distal gono- 
duct; fgm = female gland mass; pr = prostate; ps = 
penial sheath; vd = vas deferens. Scale bar = 500 
цт. 



RESULTS 

The description of the genital systenn fol- 
lows the route the gametes take from their 
place of origin in the gonad to the genital 
openings, that is from posterior to the ante- 
rior end of the snail. The gonad is covered by 
the lobes of the digestive gland and consists 
of a large number of acini (Fig. 16). Oocytes 



4 in b .,^\.r---iW~W!^X-\iSi 

FIGS. 4, 5. Origin of duct connecting digestive 
gland and distal gonoduct. Increment between 
Figs. 4 and 5 = 30 цт. ao = aorta; dd = duct con- 
necting digestive gland and distal gonoduct; dg = 
digestive gland; dgd = distal gonoduct; in - intes- 
tine; V = ventricle. Scale bars = 100 цт. 



and spermatocytes mature in the same aci- 
nus. The gametes produced in these acini are 
collected through a branching net of ciliated 
preampullary gonoducts, which open into the 
ampulla (Figs. 1, 16, 17). This ampulla con- 
sists of a number of communicating cham- 
bers. Its epithelium lacks cilia. The glandular 
postampullary gonoduct connects the am- 
pulla with the large female gland mass. The 
duct enters the gland mass ventrally on the 
left side. This gland mass has two histologi- 
cally distinct portions, which probably func- 
tion as albumen and mucous gland, respec- 
tively. At the right side, the ciliated distal 
gonoduct leaves the gland mass and 
traverses the body wall to the anterior end, 
where it opens close to the mouth (Figs. 1 , 2). 
There are two ducts branching off the dis- 
tal gonoduct. Proximally, a shorl duct con- 
nects the gonoduct with the digestive gland. 



ACOCHLIDIUM fíjense 



145 




FIGS. 6-11. Penis. 6. Apical view. Scale bar = 100 цш; 7, 8. Lateral views. Scale bars = 100 |am; 9. Large 
spine in epidermal sheath. Arrow indicates distal end of the sheath. Scale bar = 10 \im; 10. Bulge of the 
edge of the penial glans with small spines. Arrow indicates distal end of an epidermal sheath. Scale bar = 
20 lam; 1 1 . Cross-section through the tip of the ejaculatory duct. Scale bar = 50 цт. bf = basal finger; de 
= ductus ejaculatorius; ep = epithelium; ps = penial sheath; st = stylet; th = thorn. 



The junction of this short duct with the diges- 
tive gland is close to the origin of the intestine 
(Figs. 2-5). Distally, the vas deferens, ven- 
trally attached to the penial sheath, leads 
backwards to the prostate (Figs. 1-3, 14). 

The muscular ductus ejaculatorius origi- 
nates in the middle of the ventral side of the 
prostate (Fig. 2). It enters the muscular penis 



at its base after several coils between the 
lobes of the prostate and the paraprostate 
(see below) and around the penis (Figs. 1, 8, 
14, 15). Distally, the ductus ejaculatorius 
leaves the penis at the left side and rests on 
its external wall (Figs. 6, 8, 12, 13) (In snail no. 
3 the ductus ejaculatorius was completely re- 
tracted into the penis.). The opening through 



146 



HAASE & WAWRA 



de 





FIG. 12. Schematic representation of the penis. 
Chitinous elements are solid blacl<. bf = basal fin- 
ger; de = ductus ejaculatorius; ip = intrapenial 
gland; pd = paraprostatic duct; ppr = paraprostate; 
pr = prostate; th - thorn. 



which the ductus ejaculatorius exits the penis 
is guarded by a chitinous thorn (Figs. 7, 8, 12, 
13). This thorn is associated with a gland ly- 
ing in the penis, which is hereafter referred to 
as intrapenial gland (Fig. 12). At its tip, the 
ductus ejaculatorius bears a chitinous stylet 
which is in fact a groove closed by epithelium 
(Fig. 11). Basally, at the dorsal side, the penis 
bears a finger, the basal finger, armed with a 
corneous stylet (Figs. 6, 7, 12). This hollow 
stylet is connected with a gland, which we 
call paraprostate because of its position ven- 
tral of the actual prostate, by the parapros- 
tatic duct (Figs. 1, 14, 15). The glans penis is 
armed with chitinous spines, too (Figs. 6-10, 
12). Two types of spines in an ascending spi- 
ral on the edge of the glans can be distin- 
guished. This spiral comprises almost an en- 
tire whorl. In the specimen dissected, 12 
large spines formed the lower semi-circle 
and 24 finer spines completed the spiral. The 
spines stand in a single row except in the 



distal-most part where the edge of the penial 
glans is broadened to a bulge (Figs. 6, 10, 
12). All spines, the stylets of the ductus ejac- 
ulatorius and the basal finger, and the thorn 
at the opening through which the ductus 
ejaculatorius leaves the penis are partly cov- 
ered by an epidermal sheath (Figs. 7, 9, 10, 
13), which, in the case of the spines on the 
penial glans, bears bundles of presumably 
sensory cilia (Fig. 9). The penis can be pro- 
truded through a sheath (Figs. 1, 2, 13, 14), 
which opens behind the right rhinophore. 

The gonad of snail no. 1 contained only 
spermatogonia and spermatozoa. Snails nos. 
4 and 5 had in addition yolk material, but no 
oocytes, whereas spermatognia, spermato- 
zoa, yolk material and oocytes were found in 
snails nos. 2 and 3. Specimen no. 3 had sig- 
nificantly more oocytes than no. 2. In these 
latter two individuals, the digestive gland 
contained spermatozoa (Fig. 17). The am- 
pulla of snail no. 2 had a distinct opening into 
the digestive gland (Fig. 17). No such open- 
ing was found in the remaining specimens. 
This opening in snail no. 2 is no preparatorial 
artefact as indicated by the extension of the 
mass of sperm in the digestive gland far in 
front of and behind the opening. A bundle of 
spermatozoa lay in a fold of the foot close to 
the anus of animal no. 3 (Fig. 1 8). These sper- 
matozoa were obviously expelled through the 
intestine at fixation. Spermatozoa were also 
found in the posterior third of the visceral 
hump in the kidney and in the haemocoel of 
snail no. 2 (Fig. 19). 



DISCUSSION 

Acochlidium fijiense was described as her- 
maphroditic (Haynes & Kenchington, 1991). 
However, from the different states of gonadal 
maturation we conclude that adolescent 
gonochorism, that is beginning as a male (or 
female, which does not apply in this case) 
and then becoming a simultaneous her- 
maphrodite (Ghiselin, 1987), may be a more 
precise characterization of the reproductive 
strategy of A. fijiense. 

Our findings of the penial morphology dif- 
fer in several aspects from the description of 
Haynes & Kenchington (1991). These authors 
observed neither the basal finger nor the true 
course of the ductus ejaculatorius. Their 
statement on number and position of the 
spines on the edge of the penial glans varies 
as well. These discrepancies are probably 



ACOCHLIDIUM FIJIENSE 



147 




;A;^2r?Nâ^^2^. 




FIGS. 13-15. Male genital organs. 13. Ductus ejaculatorius leaving penis; 14. Distal region of prostate and 
paraprostate, penis and sperm conducting ducts. Arrows indicate spines, arrow heads the penial sheath; 
15. Proximal region of prostate and paraprostate with ejaculatory duct, de = ductus ejaculatorius; dg = 
digestive gland; oe = oesophagus; p = penis; ppr = paraprostate; pr = prostate; ps = penial sheath; sg = 
salivary glands; th = thorn; vd = vas deferens. Scale bars = 100 |.im. 



due to different states of contraction of the 
penis after fixation. Besides, Haynes & Ken- 
chington (1991) neither had the opportunity 
of SEM investigations nor did they section 
the penis. 

For similar reasons, comparison with other 
species of the genus Acochlidium are difficult. 
What Bucking (1933) described as an oviduct 
in A. ambolnense Strubell, 1 892, is clearly the 
paraprostatic duct with the basal finger. And 
in A. sutteri Wawra, 1979 (Wawra, 1979), the 
smaller thorn appears to represent the arma- 
ture of the basal finger or its homologue. The 
descriptions of A. bayerfehlmanni Wawra, 
1 980 (Bayer & Fehlmann, 1 960; Wawra, 1 980) 
and A. weberi (Bergh, 1 896) (Bergh, 1 896) are 
too superficial to allow a detailed comparison. 

Organs similar in structure and position to 
the paraprostate and the basal finger of A. 
fijiense have been described in another aco- 
chlidioidean species, the interstitial Pseu- 
dunela cornuta (Challis, 1970) (Challis, 1970: 



37). The penial gland and the "complex mus- 
cular organ equipped with a single, hollow, 
curved spine" of P. cornuta are probably re- 
spective homologues. 

The basal finger probably functions as a 
stimulatory organ analogous to the dart (gyp- 
sobelum) of some stylommatophoran land 
snails or other organs in some other pulmo- 
nates (Tompa, 1984). Adamo & Chase (1988, 
1 990) found that in Helix aspersa O. F. Müller, 
1774, dart shooting decreased courtship du- 
ration. In that species, the active substance is 
secreted by the digitiform glands which pro- 
duce a mucus that coats the dart. This mu- 
cus is only effective if it is injected into the 
body cavity (Adamo & Chase, 1990; Chung, 
1 986). We assume that in A. fijiense the basal 
finger in association with the paraprostate 
has a similar function during copulation. 

The three strategies of sperm transfer oc- 
curring in the order Acochlidioidea, namely 
copulation, injection, and transfer by sper- 



148 



HAASE & WAWRA 



•^fjip"^"" 



V"' / 






go 



dg 



wi^ 






to 



am 



16 




^"X'M^i^^^ '^Vf' 



'<Ф^^>'' '^'' 








in 



/i?'*^. 





к 


'¡iß. 


i 

* 


,'■ 


dg 


.*' с ч 


д 


•i; go;: 

ш ч 


*■ ' 




'Ж^,^ s. 19 



FIGS. 16-19. Spermatozoa. 16. Gonad; 17. Ampulla opening into digestive gland; both organs are filled 
with sperm; 18. Sperm in a fold of the foot. Arrow indicates the anus; 19. Sperm in haemocoel and kidney. 
Arrows indicate blood cells; Figs. 16, 17 and 19 are interference contrast photographs, am = ampulla; с = 
haemocoel; dg = digestive gland; dgd = distal gonoduct; go = gonad; in = intestine; к = kidney; pg = 
praeampullary gonoducts; sp = spermatozoa; to = [presumptive (see Discussion)] temporary opening of the 
ampulla into the digestive gland. Scale bars = 100 (jm. 



matophores, are briefly discussed by Wawra 
(1992). In A. fijiense, the penial armature, 
the stylet-bearing ductus ejaculatohus and 
the fact that we found spermatozoa in the 
haemocoel and in the kidney of snail no. 2 
indicate that sperm are injected into the 
haemocoel at copulation. The injection of 
sperm into the kidney was probably an acci- 
dent. 

Hypodermic impregnation is the presump- 
tive mode of sperm transfer in another aco- 



chlidioidean species, the interstitial Hedylop- 
sis spiculifera (Kowalewsky, 1901) (Wawra, 
1989). Hypodermic injection of sperm is fur- 
ther known in some Sacoglossa (Baba & Ha- 
matani, 1970; Gascoigne, 1956, 1975, 1976, 
1978, 1993; Hand & Steinberg, 1955; Jen- 
sen, 1986; Marcus, 1973; Reid, 1964; Trow- 
bridge, 1995) and two nudibranch species 
(Rivest, 1984). Gascoigne (1993) distin- 
guished between precise and imprecise hy- 
podermic injection. In the first, more common 



ACOCHLIDIUM fíjense 



149 



mode, sperm are injected through the body 
wall directly into parts of the genital system, 
while in the imprecise mode, spermatozoa are 
released into the haemocoel. Imprecise hy- 
podermic impregnation is only reported for 
the sacoglossans Elysia maoria (Powell, 1 937) 
(Reid, 1964), E. subornata Verrill, 1901 
(Jensen, 1986), Bosellia cohnneae Marcus, 
1973 (Marcus, 1973), and Alderja modesta 
(Lovén, 1844) (Hand & Steinberg, 1955). 
Hedylopsis spiculifera and our study organ- 
ism, Acochlidium fijiense, probably practice 
the imprecise mode, too. In none of the im- 
precisely injecting species is the fate of the 
transferred sperm known. That holds also for 
those acochlidioideans in which sperm trans- 
ferred in spermatophores attached to the 
body wall enter the haemocoel through lysis 
of the recipient's epidermis (Doe, 1 974; Had- 
field & Switzer-Dunlap, 1984; Morse, 1976; 
Swedmark, 1968a, b). 

In the nudibranchs Palio zosterae (O'Dono- 
ghue, 1924) and P. dubia (Sars, 1824), sperm 
must be injected into the gonadal acini. 
Sperm that are released into the haemocoel 
are phagocytosed by blood cells (Rivest, 
1984). In the specimen of Л. fijiense impreg- 
nated with sperm, we found accumulations 
of blood cells, too. But because these sper- 
matozoa are presumably intended to fertilize 
eggs, the blood cells have probably a differ- 
ent function such as nourishment or guid- 
ance to the fertilization site. 

Like most acochlidioideans, A. fijiense 
lacks both a seminal receptacle for storage 
of allosperm and a gametolytic gland (bursa 
copulathx) to digest excess alio- and auto- 
sperm and other surplus substances and 
products of the genital system (Hadfield & 
Switzer-Dunlap, 1984). Because these sperm- 
receiving organs are typical of the Bauplan of 
genital systems of opisthobranchs (e.g., Sal- 
vini-Plawen, 1991), we consider their loss to 
be secondary (see below). The loss of the 
gametolytic gland appears to be compen- 
sated by the digestive gland. In the five spec- 
imens that we sectioned, only the ampulla of 
snail no. 2 had an opening into the digestive 
gland. Both this snail and individual no. 3 had 
sperm in the digestive gland. Based on the 
fact that ampulla and digestive gland were 
connected in only one individual but two 
snails had sperm in the digestive gland, and 
supported by the consideration that both di- 
gestion of food and release of gametes would 
be hampered by a permanent opening of the 
ampulla into the digestive gland, we conclude 



that this connection is transient, established 
only when required. Whether the digestive 
gland also digests allosperm cannot be told. 
This might be the case if sperm were acci- 
dently injected into the digestive gland at cop- 
ulation. If A. fijiense copulated through the 
genital pore, sperm could reach the digestive 
gland through the duct connecting distal 
gonoduct and digestive gland. However, the 
distal gonoduct has no vaginal characteristics 
which would indicate reception of a copula- 
tory organ and sperm. Circumstantial evi- 
dence (penial armature, sperm in haemocoel 
and kidney of snail no. 2) suggests that the 
mode of sperm transfer is hypodermic injec- 
tion. 

Because the connection of ampulla and di- 
gestive gland was seen in only a single indi- 
vidual, one might argue that this opening was 
an abnormality, and consequently, the sperm 
in the digestive gland of individual no. 3 
would be allosperm (see above). But the fact 
that this opening was seen in only a single 
snail is not a strong argument against the 
presumed regularity of the temporary con- 
nection of digestive gland and ampulla, sim- 
ply because of the improbability to detect a 
transient structure. Until further evidence we 
intuitively prefer the first interpretation of the 
findings in snail no. 2. 

The loss of seminal receptacle and game- 
tolytic gland in the genital system of acoch- 
lidioidean species appears to be correlated 
with the mode of sperm transfer. Strubellia 
paradoxa is the only acochlidioidean species 
possessing both organs (Wawra, 1988). Only 
the gametolytic gland is present in Pseu- 
dunela cornuta (Challis, 1970). In both spe- 
cies, allosperm must enter the genital system 
through the genital opening in order to get to 
the receptacle or to the gametolytic gland. All 
other species for which the genital anatomy 
is described, including those of the genus 
Hedylopsis Thiele, 1931 (contra Odhner, 
1937, and Rankin, 1979, see Wawra, 1989), 
have lost both the receptacle and game- 
tolytic gland. In all these species, sperm are 
or are presumed to be transferred either by 
hypodermic impregnation or through sper- 
matophores attached to the body wall of the 
mating partner. All these species are either 
small, interstitial forms or relatively large 
freshwater snails of the genus Acochlidium 
(not sensu Rankin, 1979; the species are 
listed in Haynes & Kenchington, 1991). Be- 
cause of the poor state of knowledge on the 
majority of the species of the order Acochlid- 



150 



HAASE & WAWRA 



ioidea, it is too early to speculate how often 
these modes of sperm transfer accompanied 
by the loss of the receptacle and the game- 
tolytic gland evolved in this group. This holds 
also for the question whether form and func- 
tion of the genital system of Acochlidium are 
autapomorphies of the genus or inherited 
from a marine, possibly even interstitial, an- 
cestor. 

The bundle of sperm observed in a fold of 
the foot near the anus of snail no. 3 was ob- 
viously expelled from the intestine at fixation. 
In this way, the animal decreased its volume 
and thus could contract more efficiently. The 
decrease of volume of the digestive gland 
could in addition be achieved through the by- 
pass to the distal gonoduct. However, the 
true function of the duct connecting the di- 
gestive gland and the distal gonoduct re- 
mains a matter of speculation. A noteworthy 
analogy exists in some turbellahans where 
the ductus genitointestinalis connects the 
genital system with the digestive system 
(e.g., Reisinger, 1968). The true function of 
this ductus genitointestinalis is also unclear. 

The peculiar course of the vas deferens 
was described earlier for A. sutteri (Wawra, 
1979) and A. bayeiiehlmanni (Wawra, 1980), 
but it remained unclear through which duct 
eggs are laid. In A. fijiense — and we assume 
that the same holds for the above mentioned 
species — the distal gonoduct continues be- 
yond the branch to the vas deferens to open 
anteriorly. It thus provides the passage for 
the eggs to the exterior. Egg masses and lar- 
vae of A. fijiense were described by Haynes & 
Kenchington (1991). 

Some of our conclusions as to the function 
of the genital system of Л. fijiense are neces- 
sarily speculative. Observations of and ex- 
perimentation with living animals have to 
complement our findings and will confirm or 
falsify our hypotheses. 

ACKNOWLEDGEMENTS 

We are greatful to Dr. H. Hilgers for sec- 
tioning one specimen. Dr. H. Kothbauer, Dr. 
B. Ruthensteiner, Dr. L. Salvini-Plawen and 
two anonymous referees made helpful com- 
ments on the manuscript. 



LITERATURE CITED 

ADAMO, S. A. & R. CHASE, 1988, Courtship and 
copulation in the terrestrial snail ¡Helix aspersa. 
Canadian Journal of Zoology, 66: 1446-1453. 



ADAMO, S. A. & R. CHASE, 1990, The "love dart" 
of the snail Helix aspersa injects a pheromone 
that decreases courtship duration. Journal of Ex- 
perimental Zoology, 255: 80-87. 

BABA, К. & I. HAMATANI, 1970. The anatomy of 
Ercolania boodleae (Baba, 1938) from Seto, Kii, 
Middle Japan (Opisthobranchia: Sacoglossa). 
Publications of the Seto Marine Biological Lab- 
oratory, 18: 215-222. 

BAYER, F. M. & H. A. FEHLMANN, 1960, The dis- 
covery of a freshwater opisthobranch mollusk, 
Acochlidium amboinense Strubell, in the Palau 
Islands. Proceedings of the Biological Society of 
Washington, 73: 183-194. 

BERGH, L. S. R., 1896, Die Hedyliden, eine Familie 
der kladohepatischen Nudibranchien. Verhand- 
lungen der Zoologisch-Botanischen Gesellschaft 
Wien, 45: 1-12, 2 pis. 

BUCKING, G., 1933, Hedyle amboinensis (Stru- 
bell). Zoologisches Jahrbuch, Abteilung System- 
atik, 64: 549-582, 1 pl. 

CHALLIS, D. A., 1970, Hedylopsis cornuta and /W/- 
crohedyle verrucosa, two new Acochlidiacea 
(Mollusca: Opisthobranchia) from the Solomon 
Islands Protectorate. Transactions of the Royal 
Society of New Zealand, Biological Sciences, 1 2: 
29-40. 

CHUNG, D. J. D., 1 986, Stimulation of genital ever- 
sión in the land snail Helix aspersa by extracts of 
the glands of the dart apparatus. Journal of Ex- 
perimei Zoology 238: 129-139. 

DOE, D. A., 1974, A new species of the order Aco- 
chlidiacea (Opisthobranchia: Microhedylidae) 
from New England. Transactions of the American 
Microscopical Society, 93: 241-247. 

GASCOIGNE, T., 1956, Feeding and reproduction 
in the Limapontiidae. Transactions of the Royal 
Society of Edinburgh, 63: 1 29-1 51 . 

GASCOIGNE, T., 1975. The radula and reproduc- 
tive system of Olea hansineensis Agersborg, 
1923 (Gastropoda: Opisthobranchia: Sacoglos- 
sa). The Veliger, 17: 313-317. 

GASCOIGNE, T., 1976, The reproductive systems 
and classification of the Stiligeridae (Opistho- 
branchia: Sacoglossa). Journal of the Malaco- 
logical Society of Australia, 3: 157-172. 

GASCOIGNE, T., 1978, The internal anatomy of 
Stiliger fuscatus (A. Gould, 1870) (Opisthobran- 
chia: Sacoglossa). Zoological Journal of the Lin- 
nean Society 63: 265-274. 

GASCOIGNE, T., 1993, Stylets, styles, and other 
cuticular extensions of the male duct in the or- 
ders Sacoglossa (Gastropoda). Bollettino Mala- 
cologico, 29: 139-146. 

GHISELIN, M. T., 1987, Evolutionary aspects of 
marine invertebrate reproduction. Pp. 609-666, 
in: A. С GlESE, J. S. Pearse & V. B. Pearse, 
Reproduction of marine invertebrates, 9, Palo 
Alto. 

HADFIELD, M. J. & M. SWITZER-DUNLÛ.P, 1984, 
Opisthobranchs. Pp. 209-350, in: A. Tompa, N. 

H. VERDONK & J. A. M. VAN DEN BiGGELAAR, The 

Mollusca, 7. Reproduction, Orlando. 



ACOCHLIDIUM fíjense 



151 



HAND, С. & J. STEINBERG, 1955, On the occur- 
rence of the nudibranch Alderia modesta (Lovén, 
1844) on the central California coast. The Nau- 
tilus, 69; 23-29. 

HAYNES, A. & W. KENCHINGTON, 1991, Acoch- 
lidium fijiensis sp. nov. (Gastropoda: Opistho- 
branchia: Acochlidiacea) from Fiji. The Veliger, 
34: 166-171. 

JENSEN, K. R., 1986. Observations on copulation 
in two species of Elysia from Florida (Opistho- 
branchia: Ascoglossa). Ophelia, 25: 25-32. 

MARCUS, E. d. B.-R., 1973, On the genus Bosellia 
(Mollusca: Gastropoda: Ascoglossa). Bulletin of 
Marine Science, 23: 811-823. 

MORSE, M. P., 1976, Hedylopsis risen sp. п., a 
new interstitial mollusc from the New England 
coast (Opisthobranchia, Acochlidiacea). Zoológ- 
ica Scnpta, 5: 221-229. 

ODHNER, N. H., 1937, Hedylopsis suecica n. 
sp. und die Nacktschneckengruppe Acochlidia- 
cea (Hedylacea). Zoologischer Anzeiger, 120: 
51-64. 

RANKIN, J. J., 1979, A freshwater shell-less mol- 
lusc from the Caribbean: structure, biotics, and 
contribution to a new understanding of the Aco- 
chlidioidea. Royal Ontario Museum Life Sciences 
Contributions, 116: 1-123. 

REID, J. D., 1964, The reproduction of the saco- 
glossan opisthobranch Elysia maoria. Proceed- 
ings of the Zoological Society of London, 143: 
365-393. 

REISINGER, E., 1968, Xenoprorhynchus ein Mo- 
dellfall für progressiven Funktionswechsel. 
Zeitschrift für Zoologische Systematik und Evo- 
lutionsforschung, 6: 1-55. 

RIVEST, B. R., 1984, Copulation by hypodermic 
injection in the nudibranchs Palio zosterae and 
P. dubia (Gastropoda, Opisthobranchia). Biolog- 
ical Bulletin, 167: 543-554. 

SALVINI-PUWVEN, L., 1 991 , Ongin, phylogeny and 



classification of the phylum Mollusca. Iberus, 9: 
1-33. 

SWEDMARK, В., 1968a, Deux espèces nouvelles 
d'Acochlidiacées (Mollusques Opisthobranches) 
de la faune interstitielle marine. Cahiers de Biol- 
ogie Marine, 9: 175-186. 

SWEDMARK, В., 1968b, The biology of interstitial 
Mollusca. Symposium of the Zoological Society 
of London, 22: 135-149. 

TOMPA, A., 1984, Land snails (Stylommatophora). 
Pp. 48-141 , in: A. Tompa, N. H. Verdonk & J. A. 
M. VAN DEN BiGGELAAR, The Mollusca, 7. Repro- 
duction, Orlando. 

TROWBRIDGE, С D., 1995. Hypodermic insemi- 
nation, oviposition, and embryonic development 
of a pool-dwelling ascoglossan (= sacoglossan) 
opisthobranch: Ercolania felina (Hutton, 1882) on 
New Zealand shores. The Veliger, 38: 203-211. 

WAWRA, E., 1979, Acochlidium sutteri nov. spec. 
(Gastropoda, Opisthobranchia, Acochlidiacea) 
von Sumba, Indonesien. Annalen des Naturhis- 
torischen Museums Wien, 82: 595-604, 4 pis. 

WAWRA, е., 1980, Acocholidium bayerfehlmanni 
spec. nov. (Gastropoda: Opisthobranchia: Aco- 
chlidiacea) from Palau Island. The Veliger, 22: 
215-218, 2 pis. 

WAWRA, E., 1988, Strubellia paradoxa (Strubell 
1892) (Gastropoda: Opisthobranchia) von den 
Salomon-lnseln. Zoologischer Anzeiger, 220: 
163-172. 

WAWRA, е., 1989, Zur Kenntnis der interstitiellen 
Opisthobranchierart Hedylopsis spiculifera 
(Kowalewsky) (Mollusca, Gastropoda). Zoológ- 
ica Scripta, 18: 397-403. 

WAWRA, E., 1992, Sperm transfer in Acochlidia- 
cea. P. 103, in: F. GiuSTi & G. MANGANELLi, Ab- 
stracts of the Eleventh International Malacologi- 
cal Congress, Siena. 

Revised Ms. accepted 20 February 1996 



MAU\COLOGIA, 1996, 38(1-2): 153-160 

QUANTIFICATION OF THE DEVELOPMENT OF THE CEPHALIC SAC AND 
PODOCYST IN THE TERRESTRIAL GASTROPOD UMAX MAXIMUS L. 

G. M. Küchenmeister', D. J. Prior^ & I. G. Welsford^ 

ABSTRACT 

During embryonic development in the terrestrial gastropod Umax maximus L., an anterior 
cephalic sac and a posterior podocyst are elaborated. At 4 ± 1.2 d (X ± SD; n = 25), rhythmic 
contractions began in the podocyst. By 6.3 ± 0.8 d of development, the cephalic sac began 
rhythmic contractions in antiphase to those of the podocyst; a behavior we have termed 
cephalopedal pumping. Pumping preceded regular heart activity, which began at 8 ± 1.8 d of 
development, and progressively decreased in frequency throughout embryonic development, 
ceasing by hatching. Cephalopedal pumping was capable of redistribution of dye throughout 
the embryo. The two structures were capable of sustained, independent, rhythmic contrac- 
tions. Because pumping precedes regular heart activity and can, presumably, redistribute 
hemolymph, cephalopedal pumping may serve as a primordial circulatory system, separate 
from the developing cardiovascular system. 

Key words: circulatory system, development, Limax maximus, slug, gastropod, heart 



INTRODUCTION 

Terrestrial gastropods such as slugs un- 
dergo direct development and are easily 
raised in the laboratory. Due to the combina- 
tion of these characteristics, slugs drew early 
attention as model developmental systems 
(e.g., Laurent, 1837; Jourdain, 1884; Hench- 
man, 1890; Cuenot, 1892; Kofoid, 1895; Car- 
dot, 1924). It has been known for over 100 
years that slug embryos elaborate an anterior 
cephalic sac (= anterior vesicle, Laurent, 
1837; Jourdain, 1884; = vesicle, Kofoid, 
1898; = ectodermal sac, Simpson, 1901) and 
a posterior podocyst during development. It 
has also been widely reported that, once 
formed, these structures undergo rhythmic 
contractions. It has been proposed that the 
rhythmic contractions serve a circulatory 
function in embryos (e.g., Jourdain, 1884). Al- 
ternatively, the contractions of the cephalic 
sac and podocyst have been suggested to 
control movement of the embryo within the 
egg (Laurent, 1837; Jourdain, 1884), a circu- 
latory function (e.g., Cuenot, 1892; Cardot, 
1924) and/or serve a respiratory or osmoreg- 
ulaton/ function (Laurent, 1837; Jourdain, 
1884; Kofoid, 1895). 

Although the development of the podocyst 
and cephalic sac has been used as a quali- 



tative component of a staging scheme for 
some species of slugs (Carrick, 1938), the 
development of these structures has yet to 
be quantified. Furthermore, there exists very 
little data on the potential physiological sig- 
nificance of these structures and/or their 
contractile behavior. We have initiated inves- 
tigations into the development of the cepha- 
lic sac and podocyst of the terrestrial slug 
Limax maximus L. and compared this with 
the development of heart in an attempt to 
quantify both the morphology and contractile 
behavior of these structures during develop- 
ment. A preliminary report of these data has 
appeared in abstract form (Welsford & Prior, 
1988). 



MATERIALS AND METHODS 

Animals 

Sexually mature (according to the criteria 
of Sokolove & McCrone, 1978) L. maximus, 
which had either been collected from the field 
or raised from eggs in a laboratory culture, 
were kept in vented containers lined with wa- 
ter-saturated paper towels under a regulated 
light cycle (L:D 14:10 or 13:11), a regulated 
temperature cycle (18 С during lights on and 



'Department of Biology, Bradley University, Peoria, Illinois 61625, U.S.A. 

^Northern Michigan University, Marquette, Michigan 49855, U.S.A. 

^Department of Biology, Keene State College, Keene, New Hampshire 03435-2001, U.S.A., to whom correspondence 

should be addressed. 



153 



154 



KUCHENMEISTER, PRIOR & WELSFORD 



12°C during lights off) and approximately 
100% RH. Slugs were fed ad libitum on lab- 
oratory food pellets (Purina Rat Chow). Umax 
maximus lays its eggs in discrete clumps or 
masses of 20-250 eggs, thus facilitating 
identification of eggs laid by an individual an- 
imal (Prior, 1 983). Egg masses were collected 
daily from containers and placed on water- 
saturated filter paper-lined petri dishes at a 
constant temperature (either 15°C or 20°C) 
under a light cycle (L:D 14:10). 

Morphological Measurements 

Every 2 d, embryos were decapsulated 
(i.e., removed from the eggs) in a dish of ster- 
ile slug saline (55.6 mM Na^ 4.2 mM K^ 7 
mM Ca"", 4.6 mM Mg"", 80.3 mM CI , 0.2 
mM H2PO4 , 5.0 mM HCO3, 5.0 mM Dex- 
trose, pH 7.3-7.41, 139-145 mOsm/kg H2O; 
Prior & Grega, 1982). Decapsulated embryos 
were transferred by pipette to a depression 
slide and viewed under a compound micro- 
scope fitted with an ocular micrometer (Nikon 
Optiphot II or Olympus Bmax). The following 
measurements were taken (in цт); (1) maxi- 
mal width of the cephalic sac; (2) maximal 
length of the cephalic sac; (3) maximal length 
of the foot; (4) maximal length of the 
podocyst; (5) maximal width of the podocyst; 
(6) maximal width of the pericardial chamber 
(Fig. 1A,B). To allow for comparisons of rel- 
ative structural dimensions at varying embry- 
onic stages, all measurements were normal- 
ized to the length of the foot because this 
structure is readily identifiable at all develop- 
mental stages. 

To determine whether cephalopedal con- 
tractions could distribute dye throughout the 
embryo, a microcapillary tube was filled with 
2% Blue Dextran (MW approximately 2X10*^; 
Sigma Chemical Co., Inc.) and inserted into 
either the cephalic sac or podocyst with the 
aid of a micromanipulator (Leitz or WPI inc.). 
Dye was injected using low pressure (not 
greater than 0.5 atm). To control for potential 
damage during injection, only data from em- 
bryos which exhibited normal contraction ac- 
tivity for at least 20 min after injection were 
reported. To more clearly delineate the re- 
gions throughout which contractions were 
distributing dye, a saturated carmine solution 
was injected into either the cephalic sac or 
podocyst. Following injection with carmine, 
embryos were placed in a saturated solution 
of chlorobutanol for 30 min to ensure mus- 
cular relaxation (Kempf, personal communi- 



cation), fixed in paraformaldehyde, dehy- 
drated, embedded in paraffin, sectioned at 
10 \xm thickness and mounted onto slides. 
Sections were stained with Harhs' Hematox- 
ylin and counter stained with Eosin Y follow- 
ing standard protocols (Schleicher, 1953). 
Stained sections were observed under light 
microscopy and the distribution of carmine 
particles was determined. Sections were 
drawn with the aid of a drawing tube attach- 
ment to a Nikon Optiphot II. 

Physiological Measurements 

To ensure that decapsulation had no effect 
on the contractions of the cephalic sac, 
podocyst or (in later stages) heart, the rate of 
heart and/or cephalopedal contractions was 
determined prior to and following decapsula- 
tion. Contractile activity was measured by 
placing the embryo on a modified depression 
slide which allowed the embryo to be contin- 
uously superfused with saline. Embryo be- 
havior was recorded on VHS videotape using 
a videomicroscope (Zeiss or Olympus). 

Statistical Analysis 

Comparisons between developmental 
times at varying temperatures and between 
heart rate and cephalopedal contraction rate 
were performed using a T-test for indepen- 
dent samples. Trends for heart and cephalo- 
pedal contraction during development were 
determined by calculating linear regressions. 
Statistical analyses were performed using 
either PsiPlot (PsiPlot Inc.) or DataDesk (Data- 
Desk Inc.), and graphs were constructed 
using either PsiPlot or Kaleidograph (Kalei- 
dograph Inc.). In all comparisons, a probabil- 
ity values of less than 0.05 was considered 
significant. 



RESULTS 

Development of the Cephalic Sac, 
Podocyst and Heart 

Although there was considerable variation 
in the developmental time of different animals 
within a clutch and between clutches, embry- 
onic development in L maximus averaged 
26.6 ± 13.6 (X ± SD; n = 15 clutches) d at 
20' C. Development was temperature-depen- 
dent, taking significantly longer at 15''C (35.6 
± 19.2 d; t = 4.97, p = 0.00011). The cephalic 



CEPHALOPEDAL PUMPING IN L. MAXIMUS 



155 



A 



Podocyst 




— Cephalic Sac 



Tentacles 



Mouth 



100 |im 



В 




1 = Maximum Width of Cephalic Sac 
2= Maximum Length of CephaHc Sac 
3 = Maximum Length of Foot 
4= Maximum Length of Podocyst 
5= Maximum Width of Podocyst 
6= Maximum Width of Pericardium 

FIG. 1 . A: Drawing of a slug embryo at 1 5 d of development indicating the relative positions of the podocyst, 
foot, cephalic sac and pericardial area. B: Indicates the morphological measurements taken at each 
embryonic stage. Scale bar is 100 цт. 



156 



KUCHENMEISTER, PRIOR & WELSFORD 



sac developed within 4 d of egg laying and 
decreased progressively in size relative to the 
foot throughout embryonic development (Fig. 
2A). The cephalic sac invaginated within 
22-26 d of development and, consistent with 
the observations of Carrick (1938) for Agrioli- 
max agrestis, was observed to form compo- 
nents of the internal organs including the al- 
bumin glands and reproductive structures. 
The sac was absent by hatching in all animals 
observed (n = 350). 

The podocyst also developed within the 
first 4 d, and increased in size relative to the 
foot throughout the first 20-22 d of develop- 
ment. The podocyst decreased in size after 
this time, usually disappearing prior to hatch- 
ing (Fig. 2B). Occasionally a hatchling slug 
retained a remnant of the podocyst after 
hatching, but this was lost in all observed 
cases within 3 d post-hatch (n = 350). 

The heart first appeared between 10-12 d 
of development from a dorsal evagination of 
the embryo just caudal to the cephalic sac. 
The heart increased in size relative to the foot 
until approximately 18 d of development then 
decreased in relative size thereafter (Fig. 20). 

Cephalopedal Contractions 

By 4 ± 1.2 (X ± SD; n = 25) d of develop- 
ment, the podocyst initiated irregular, low fre- 
quency contractions (16.2 ± 5.4 beats per 
minute [BPM]; X ± SEM; n = 45; Fig. ЗА). By 
day 5 (56.4 ± 1.8 d), the cephalic sac began 
contractions in antiphase to the podocyst. 
Cephalopedal contractions peaked in fre- 
quency at approximately 4 d of development 
and decreased in frequency throughout de- 
velopment (r^ = 0.932; p < 0.0001; Fig. ЗА). 
After 18 d of development, there was a 
marked increase in the variance of the 
cephalopedal contractions (Fig. ЗА.). 

The rate of cephalopedal contractions was 
unaffected by decapsulation (Table 1). The 
antiphasic nature of the cephalopedal con- 
tractions led us to investigate the nature of the 
oscillatory control of contractions. The con- 
tractions of both the podocyst and cephalic 
sac were unaffected by rupture or removal of 
the other structure (Table 1) or, indeed, by 
complete transection of the embryo, with the 
exception that the structures ceased to con- 
tract in antiphase to one another (Table 1 ). The 
independent contractions in each structure in 
completely transected embryos lasted for an 
average of 2 h (2 ± 8.7 h; n = 15) and in one 
instance, continued for 20 h after transection. 



Cephalic Sac Development 




^ 1.5 



E 0.5 



2 4 6 8 101214 1618202224262Í 
Days of Development 



Podocyst Development 







2 4 6 8 101214 1618202224262Í 

Days of Development 



Heart Development 




2 4 



6 8 10121416182022242628 
Days of Development 



FIG. 2. Means of normalized measures of cephalic 
sac width (A), podocyst width (B) and pericardial 
area width (C) during embryonic development. 
Each bar represents the mean value of 55 animals. 



CEPHALOPEDAL PUMPING IN L. MAXIMUS 



157 



A. 



25 




20 


г 




15 
10 
5 



- 


1 , , , , 1 


, , , 1 , , , , 1 , 


% 



20 



25 



30 



В. 



150 



100 



50 



\,J\] 




20 



Days of Development 



25 



30 



FIG. 3. A: Frequency (in BPM) of cephalopedal 
pumping in embryos of varying ages (measured at 
20-C). Each point represents the mean (± SD) re- 
sponse of 45 animals. B: Frequency (in BPM) of 
heart activity in embryos of varying ages. Each 
point represents the mean (± SD) response of 45 
animals. 



Heart Contractions 

In contrast to the cephalopedal activity, 
heart activity did not begin until 9.2 ± 1 .4 d of 
development (Fig. 3B). Heart rate was signif- 
icantly greater than cephalopedal contrac- 
tions at all stages (mean across all stages for 
hearl was 57.4 ± 156.7 BPM vs. 7.1 ± 25.7 
BPM; t = 6.545; p < 0.0001). Heart rate 
peaked at 9 d of development and decreased 
progressively throughout development (r^ = 
0.7; p < 0.0001), but with a significantly dif- 
ferent slope than that of cephalopedal con- 
tractions (-2.0 BPM/d for heart vs. -0.64 
BPM/d for cephalopedal contractions; Fig. 
3A,B). Initially, heart rate was strongly cou- 
pled to pumping activity, cyclically increas- 
ing and decreasing with each cephalopedal 
contraction (Fig. 5A). After 14 d of develop- 
ment, heart activity became independent of 
cephalopedal activity (Fig. 5B). Heart activity 



was highly variable throughout the first 18 d 
of development, but the variance of heart ac- 
tivity decreased from 18-20 d of develop- 
ment to hatching (Fig. 3B). 

Staining of Intra-Embryonic Regions 

Cephalopedal contractions were capable 
of redistributing dye throughout the embryo 
(Fig. 6). However, the portion of the embryo 
that was stained by this procedure depended 
upon the age of the embryo. Injection into the 
podocyst in embryos younger than 10 d, re- 
sulted in staining of primarily the entire pos- 
terior region of the embryo, including a region 
surrounding the CNS and heart (summarized 
in Fig. 6). The cephalic sac was stained by 
podocyst injections only lightly and only after 
prolonged injections (>5 min). Injections into 
the cephalic sac in embryos younger than 1 
d resulted primarily in staining a restricted 
internal space distinct from that stained with 
the podocyst injection and continuous with 
the stomach, reproductive organs and he- 
patopancrease (Fig. 6). Injection into the 
cephalic sac only lightly stained the podocyst 
and heart and only after prolonged injections. 
In embryos older than 12 d, injection into the 
podocyst or cephalic sac did not stain the 
heart. In addition, in embryos older than 1 2 d, 
injections into the podocyst in excess of 10 
min failed to stain the regions of the embryo 
continuous with the cephalic sac (Fig. 6). At 
all stages of embryonic development tested, 
histological sections confirmed the dye sep- 
aration of the internal regions continuous 
with either sac or podocyst in embryos older 
than 12 d (Fig. 6). 



DISCUSSION 

The observations in the present study are 
consistent with qualitative observations on 
the development of Agriolimax agrestis (by 
Carrick, 1938) and Umax maximus (by Simp- 
son, 1901) and thus support the utility of the 
use of Carrick's (1938) scheme (developed 
for use with embryos of A. agrestis) for the 
staging of embryos of Umax maximus. Be- 
cause preliminary observations suggest that 
this scheme also holds for the slugs Lelima- 
nia valentiana and Agriolimax (=Deroceros) 
reticulatus, this scheme may be generalizable 
to all slug species. The availability of a stag- 
ing scheme for slugs may aide in the approx- 
imate aging of embryos for which the date of 



158 



KUCHENMEISTER, PRIOR & WELSFORD 



TABLE 1. The effect of rupture or removal of either the cephalic sac or podocyst on the rate of cephalic 
sac, podocyst and heart contractions is shown, In addition, the effect of complete embryo transection 
on cephalic sac and podocyst contractions is shown. Data are mean (± SD) response of 15 animals. 



Structure 



Rate in Egg Case 
(BPM) 



Rate in Saline 
(BPM) 



Rate 60 min Post- 
Rupture of Opposing 
Structure (BPM) 



Rate 60 min Post- 
Transection (BPM) 



Podocyst 
Cephalic Sac 
Heart 



25 + 12.6 
23 ±14.3 
134 ±23 



23 ±13 (91%) 

22 ±18 (98%) 

122 ±28 (91%) 



22 ±14 (96%) 

19 ±17 (86%) 

119 ±47 (98%) 



20 ±16 (97%) 

21 ±19(110%) 
121 ±59(101%) 



A. Heart Activity at 10d of Development 




Relaxed Contracted 

Podocyst Contraction Cycle 



B. Heart Activity at 20d of Development 



80 



Ф 60 




'1 



ЛМ 



Relaxed Contracted 

Podocyst Contraction Cycle 

FIG. 4. Differences in sensitivity of heart activity to 
cephalopedal activity at 10 d of development (A) 
and 22 d of development (B). In each figure, the 
mean (± SD) instantaneous heart rate is shown (ex- 
pressed in BPM) during podocyst contraction and 
relaxation. Each bar represents the mean (± SD) 
response of 45 animals. 

laying of a clutch is unknown and/or which 
have developed under variable thermal con- 
ditions, because the development is mark- 
edly temperature sensitive. 

The cephalic sac and podocyst developed 
earlier than the heart and began rhythmic, 



antiphasic contractions prior to the onset of 
heart activity. The frequency of cephalopedal 
contractions was significantly lower than that 
of heart and the rate of decrement in activity 
throughout development between heart and 
cephalopedal contractions differed signifi- 
cantly. In addition, heart and cephalopedal 
contractions demonstrated differing patterns 
of variance, with heart becoming more regu- 
lar as cephalopedal contractions became 
less regular. Because podocyst and cephalic 
sac contractions were capable of redistribut- 
ing dye throughout the embryo, contractions 
of these structures could accomplish the cir- 
culation of hemolymph throughout various 
regions of the embryo prior to maturation of 
the cardiovascular system. We have thus 
termed the rhythmic antiphasic contractions 
of the cephalic sac and podocyst, cephalo- 
pedal pumping. 

The fact that heart activity was significantly 
affected by cephalopedal pumping at early 
stages suggests that, early in development, it 
may be continuous with the cephalic sac and 
podocyst, but that this connection is lost as 
the embryo matures. This hypothesis was 
supported by dye injections into the cephalic 
sac and podocyst, which stained the lumen 
of the heart in early stages, but not in later 
stages, of L. maximus embryos. 

Both the cephalic sac and the podocyst 
can contract independently of one another. 
Thus, each structure may be driven be sep- 
arate oscillators that are coupled to one an- 
other. Because the CNS reportedly develops 
within approximately 9 d in embryos of L. 
maximus (Henchman, 1890), the nature of 
such putative oscillators remains uncertain 
as does their fate in hatchling slugs. Work on 
these oscillators is ongoing. 

Gastropod mollusks have served as useful 
model systems for the study of the develop- 
ment of the central nervous system and be- 
havior due to the relatively small number of 
neurons present within the CNS and the fact 



CEPHALOPEDAL PUMPING IN L. MAXIMUS 



159 



Cephalic sac 



A 



Mouth 




— Podocyst 



CNS Foot 



100 ^im 



Cephalic sac 



Mouth 




Podocyst 



Foot 



100 цт 



С 




Podocyst 



TENT 



Mouth BM 



100 \im 



FIG. 5. Drawings of histological sections of embryos at 5 d of development (A), 1 5 d of development (B) and 
25 d of development (C). In each figure, the stippled areas denote the regions stained by injection of 
carmine solution into the podocyst (A and B) or cephalic sac (C). Scale bar is equal to 100 |дт. SIM = 
stomach, ALB = albumin gland, BM = buccal mass, CNS = central nervous system, TENT = tentacles, HRT 
= heart. 



160 



KUCHENMEISTER, PRIOR & WELSFORD 



that neurons are frequently large and easily 
identifiable from individual to individual. 
However, considerably less information is 
available on the development of peripheral 
structures in these organisms. Investigations 
of the development of the peripheral struc- 
tures are hampered in many mollusks by a 
metamorphosis that includes a veliger stage 
and can entail dramatic alterations in central 
and peripheral morphology and physiology. 
Furthermore, the housing and maintenance 
of these organisms frequently requires mari- 
culture facilities. Terrestrial gastropods are 
easily raised the laboratory and thus may 
serve as useful model systems for certain de- 
velopmental investigations. 



ACKNOWLEDGMENTS 

This work was supported in part by the 
OTEFD at Bradley University, a Whitehall 
Foundation Grant to IGW an MBRS grant to 
DJP and NSF DUE— 935273 (IGW). The au- 
thors wish to thank RR Stephens for com- 
ments on the manuscript and Drs. G. E. Gos- 
low and D. Blinn for the kind use of facilities 
during the initial stages of these investiga- 
tions. 



LITERATURE CITED 

CARRICK, R., 1938, The life history and develop- 
ment o\ Agriolimax agrestis L. the gray field slug. 
Transactions of the Royal Society of Edinburgh, 
59 (Part 3)(21):563-597. 

CARDOT, H., 1924, Observations physiologiques 
sur les embryons des gastéropodes pulmones. 
Journal de Physiologie et de Pathologie Gen- 
erale, 22:575-586. 

CUENOT, L., 1892, Etudes physiologiques sur les 



gastéropodes pulmones. Archives de Biologie, 
12:683-740. 

HENCHMAN, A, P., 1890, The origin and develop- 
ment of the central nervous system in Umax 
maximus. Bulletin of the Museum of Compara- 
tive Zoology (Han/ard University), 20:169-209. 

JOURDAIN, M. S., 1884, Sur les organes segmen- 
tares et le podocyste des embryones de Lima- 
ciens. Comptes Rendus Hebdomadaires des Sé- 
ances de l'Académie des Sciences, 97:308-310. 

KOFOID, G. A., 1895, On the early development of 
Umax. Bulletin of the IVIuseum of Comparative 
Zoology (Han/ard University), 27:35-135. 

LAURENT, M. 1837, Observations sur le developp- 
ment des oeufs de la limace grise et de la hmace 
rouge. Comptes Rendus Hebdomadaires des 
Séances de l'Académie des Sciences, 4:295- 
297. 

PRIOR, D. J. 1983, The relationship between age 
and body size of individuals in isolated clutches 
of the terrestrial slug, Umax maximus (Linnaeus, 
1758). Journal of Experimental Zoology, 225: 
321-324. 

PRIOR, D. J. & D. S. GREGA, 1 982, Effects of tem- 
perature on the endogenous activity and synap- 
tic interactions of the salivary burster neurones 
in the terrestrial slug, Umax maximus. Journal of 
Experimental Biology, 98:415-428. 

SCHLEICHER, E. M., 1953, An improved hematox- 
ylin-eosin method for sections of bone marrow. 
Stain Technology 28■.^^9-^23. 

SIMPSON, G. В., 1901 , The anatomy and physiol- 
ogy of Polygyra albolabris and Umax maximus, 
and the embryology of Umax maximus. Bulletin 
of the American Museum of Natural History, 
8:237-311. 

SOKOLOVE, P. G. & E. J. MCCRONE, 1978, Re- 
productive maturation of the slug, Umax maxi- 
mus, and the effects of artificial photoperiod. 
Journal of Comparative Physiology, 125:317- 
325. 

WELSFORD, I. G. & D. J. PRIOR, 1988, Embryonic 
cephalo-pedal pumping in the terrestrial slug, 
Umax maximus. American Zoologist, 28:25A. 

Revised Ms. accepted 8 April 1996 



MALACOLOGIA, 1996, 38(1-2): 161-180 



LOCAL PATTERNS OF IJ\ND SNAIL DIVERSITY IN A KENYAN RAIN FOREST 



P. Tattersfield 

Bettfield dough Cottage, Castleton Road, Chapel-en-le-Frith Stockport SK12 6PE, 

United Kingdom 



ABSTRACT 

Terrestrial molluscs were sampled in indigenous forest and plantation plots in Kakamega 
Forest, western Kenya, which is the eastern-most patch of Guineo-Congolian rain forest in 
Africa. Fifty species (one slug and 49 snails) were recorded from 27 indigenous forest plots, and 
the mean species per plot was 23.4. The majority of the species present in the fauna were small, 
litter dwellers, with 52% having a major shell dimension of less than 5 mm. Overall, species 
richness and faunal composition were relatively uniform throughout the forest system. How- 
ever, forest edge plots, including plots located along large rivers and in smaller blocks of forest, 
had a deficiency of some minute, litter-dwelling species but supported a higher frequency of 
some large-shelled taxa. The four plantations sampled supported fewer species per plot (15.25 
species/plot) and also lacked several of the small, litter-dwelling species found in the indige- 
nous forest. 

Many other species of mollusc have been previously reported from Kakamega Forest. The 
reported mollusc fauna of Kakamega Forest represents about 5.8-9.5% of the total known East 
African forest mollusc fauna, thus suggesting that there must be considerable taxonomic re- 
placement of species throughout the region. The recorded molluscan diversity in Kakamega 
Forest is high in a worldwide context. Kakamega Forest is not old in geological terms, the Lake 
Victoria basin having received a much more arid climate during periods of extended glaciation 
at higher latitudes. Its forest fauna must have colonised since the last glacial maximum in Africa, 
approximately 14000 years BP; the composition of the recorded fauna supports the view that 
recolonisation was mainly from forest refugia in central Africa. The conservation implications of 
the findings are discussed. 

Key words: land snails. Gastropoda, biodiversity, rain forest, Africa, Kenya, Kakamega 
Forest. 



INTRODUCTION 

Comparisons of land snail diversity at a re- 
gional scale indicate that there are sometimes 
large differences between tropical and tem- 
perate zones (Cameron, 1995). At the local 
scale, faunas from temperate sites are often 
rich in species but quite uniform over large 
geographical areas. However, there is little 
comparable information about local diversity 
patterns in tropical areas. Solem (1984) pro- 
posed a model to account for world-wide land 
snail diversity levels, but noted that much fur- 
ther information is required about levels of 
sympatric diversity in many parts of the world, 
especially in the tropics. This lack of informa- 
tion clearly has implications when trying to 
assess the impact of habitat loss on mollus- 
can biodiversity, and consequently for con- 
servation planning (Cameron, 1995). 

Much has been written about the East Af- 
rica terrestrial mollusc fauna, but there have 
been few investigations on molluscan as- 



semblages in different habitat types. This 
study investigates the patterns of land snail 
diversity in a relatively restricted area of rain 
forest habitat in western Kenya. It examines 
areas of forest that have been subjected to 
varying levels of human disturbance and ex- 
ploitation and also surveys the mollusc fauna 
in plantations of both exotic and indigenous 
tree species. 

THE SITE 

The Kakamega Forest complex (about 
0°15'N, 34^54'E) (Fig. 1) lies mostly to the 
west of Kakamega town in west Kenya, 
about 40 km north of Lake Victoria. It com- 
prises several separate blocks of forest (Mu- 
riuki & Tsingalia, 1990), of which Kakamega 
Forest itself is by far the largest; the smaller, 
isolated areas of Kisere and Miaba forests lie 
to the north of the main forest block. Bunyala, 
Maragoli and Kaimosi forests are situated to 
the northwest, south and southwest of Kaka- 



161 



162 



TATTERSFIELD 



SUDAN v. 




' /\ 




Л ETHIOPIA 




"s^ /"îo» Vmlaba 

\ r* ,' ^ ''OREST 




Л "^^^^ 


/ 




UGANDA \ 




\ 




MlElgonA^ »CUVA 
/ gKital* KENYA 


SOMALIA 


\ 
\ 




/ i,-^AKAMEGA FOREST 








LikTK ,-•„.»„„„ AMIKeny« 


Ч / 


i, KISERE 
^i />>v FOREST 




/ \. ■Nairobi 


У 


' J¿>^^ V^ 







TAN 
N 
100 2 


KiHmanjaroS^^ 1 
ZANIA 1 
OOkm /^ 


/ 


¡ ByunguV ^-— '^ ■^^ 

'' ^/'^^•Ч5 12 / Л '\ 










/ 


4^ 


\ ^y^^"^ 1 KAKAMEGA ^^ ', 
^^^--^-^ 1 FOREST ^^4 / 


Vihiga 




KEY 


ForesI boundary 


1 \ 13, 14 ^-''' 

Ч X lsecheno->r ^^15 ^^ 
S..a,y.-' <^f9-^^lV^7l*-. - 

f 25^.o26 

31° 2;rx^ 

У „ ^л_Р^2. \ 


7 

\ 

1 









• 


Indigenous forest plot 










О 


Plantation plot 










• ' 


Road or track 










■ 


Village 








9 


1 2 3 4кгл ,' NJ,'-' 










' ' ' ' ^-4 У 



FIG. 1. The location of Kakamega Forest and the sampling plots. 



mega Forest, respectively. The area has a 
relatively flat or gently undulating topogra- 
phy at an altitude of approximately 1500- 
1680 m. Two major rivers, the Yala and Isi- 



ukhu, as well as many smaller watercourses, 
run through the forest. 

The forest complex covers an area of ap- 
proximately 265 km^. About 45 km^ is pro- 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



163 



tected for wildlife as national reserve and na- 
ture reserve, and some of the area is currently 
proposed as a national park. The remainder is 
gazetted as forest reserve and is managed 
by the Forestry Department; current policy is 
to encourage indigenous forest rather than 
plantation. According to Muriuki & Tsingalia 
(1 990), about 48% of the forest complex sup- 
ports indigenous forest stands, the remainder 
containing plantations and grassland clear- 
ings of both natural and anthropogenic origin. 

Parts of the forest were selectively logged 
in the 1930s, 1940s, late 1970s and early 
1980s (Tsingalia, 1990). The region contains 
some of the densest population in Kenya, and 
until a presidential decree in 1 986, many parts 
of the forest were subject to shifting cultiva- 
tion by local people. Kokwaro (1984) noted a 
rapid decline in the extent of indigenous for- 
est. Such activities have now been stopped, 
but other forms of illegal exploitation still take 
place, especially the removal of plant prod- 
ucts for medicinal use and firewood, stock 
grazing and poaching for small game. If 
caught, offenders face a stiff punishment, 
with fines for a first offence equivalent to one 
month's income or three months in prison. 

The climate can be described as relatively 
hot and humid, although rainfall is seasonal, 
with most falling between March and July 
and again in October and November. Muriuki 
& Tsingalia (1990) report mean annual rainfall 
as 221 6 mm, although Zimmerman (1 972) re- 
corded over 3500 mm during 1963, indicat- 
ing substantial fluctuation from year to year; 
mean monthly maximum temperature ranges 
from 18-29 'C (Muriuki & Tsingalia, 1990). 

The site is biogeographically important, 
being situated at the edge of several regional 
vegetation zones. White (1983) classifies it as 
transitional rain forest, noting that it supports 
several Guineo-Congolian plant species at 
their easternmost African limit. The fauna 
also has strong central and west African af- 
finities (Faden, 1970; Zimmerman, 1972). Lu- 
cas (1968) describes Kakamega Forest as 
"the most easterly point of the West African- 
Congo type forest." It supports many plant 
and animal species which are not found else- 
where in Kenya (Faden, 1970). 

METHODS 

Plot Selection and Description 

Thirty one plots (1-31), each approximately 
40 m X 40 m, were sampled in Kakamega, 
Kisere and Miaba forests; 27 of the plots 



were in mixed, indigenous forest, the other 
four being in plantations (Fig. 1). Maximum 
plot separation was approximately 30 km. 
Forest edges were avoided during selection 
although three plots were situated at the 
edge of the Yala and Isiukhu rivers (Plots 6, 7, 
22) and contained marginal, riverine forest 
vegetation locally dominated by light de- 
manding tall herbaceous species; plot 23 lay 
within about 50 m of the Yala River but did 
not contain forest edge. 

Twenty three of the indigenous forest plots 
contained mature, mixed stands (canopy 
height generally about 20-30 m). All had been 
exploited for wood, timber or other resources 
to some degree, although observations on 
forest structure and other signs, such as the 
presence of saw pits, freshly cut wood and 
evidence of cattle grazing indicated that five 
of these plots (Plots 10, 11, 14, 16 and 29) 
were substantially more disturbed than the 
others. The other four indigenous forest plots 
(5, 9, 12, 30) contained relatively young, 
mixed, stands. These plots were all charac- 
terised by a relatively low canopy (about 
10-20 m), high light penetration at ground 
level, and a mixed and generally well-devel- 
oped grassy herbaceous field layer. Some of 
them contained Acanthus arbórea Forsskal, a 
characteristic plant species of disturbed sites 
requiring high illumination. Details of former 
land use are unknown, but it is possible that 
these areas had been cut and subsequently 
regrown or been recolonised by indigenous 
forest. 

All four plantations (Plots 20, 25, 26 and 
31) were probably about 20-40 years old 
and all had a well-developed canopy. The 
trees were well spaced and had presumably 
been thinned. Two contained monoculture 
stands of the non-indigenous Bischofia jav- 
anica Blume, one had been planted with 
mixed, non-indigenous conifers {Pinus patula 
Schlecht & Chamisso and Cupressus lusitan- 
ica Mill.) and the other contained a monocul- 
ture of the indigenous, Guineo-Congolian 
lowland rain forest tree species Maesopsis 
eminii Engl. The detailed history of these ar- 
eas is not known, although it is probable that 
they were formerly covered in indigenous for- 
est, possibly with a period of cultivation prior 
to conversion to plantation. 

Physical plot characteristics were re- 
corded including topography, inclination, as- 
pect, and presence and relative abundance 
of such potential molluscan microhabitats as 
dead wood, fallen trees, and rocks. Forest 



164 



TATTERSFIELD 



structure was described by estirnating can- 
opy height and the percentage cover of dif- 
ferent vegetation strata (tall, medium and low 
tree components and scrub, herb and liane 
categories) for each site. Many of the indig- 
enous plots had a dense understorey shrub 
layer of Dracaena afromontana Mildbr. Plant 
species were not generally recorded, al- 
though notes were made on common domi- 
nant species where they could be identified. 
Mutangah et al. (1992) provide further infor- 
mation about the vegetation of Kakamega 
Forest. 

Mollusc Sampling 

Sampling for molluscs was undertaken by 
a combination of direct search and litter siev- 
ing methods. Each plot was searched for at 
least 30 minutes, ensuring that all potential 
microhabitats, such as dead wood, rocks, 
tree trunks, leaf litter and living vegetation, 
were examined. It was not possible to survey 
the forest canopy directly, but fallen trees 
and canopy branches supporting epiphytic 
orchids, mosses and lichens were examined 
when available. All molluscs found were col- 
lected. Up to three local guides assisted with 
the direct searching and therefore sampling 
effort varied amongst the plots. However, the 
mean number of species per plot does not 
differ significantly between the sites sampled 
by one or three people (F, 23 df = 3.70, P > 
0.05), and thus this vahation in sampling ef- 
fort does not appear to affect the assessment 
of diversity levels. About 4 litres of surface 
leaf litter and soil were taken from each plot 
and passed through a coarse sieve (4 mm 
mesh size). Large species retained in the 
sieve were removed. The fine fraction was 
then dried and passed through two further 
sieves (mesh sizes 2 mm and 0.5 mm). These 
sieve fractions were searched separately un- 
der good illumination until no further mol- 
luscs could be found (generally about 30-45 
minutes). Material passing through the 0.5 
mm mesh for the first few sites was searched 
for snails, but because none were found, this 
fraction was discarded for subsequent sam- 
ples. Some specimens were preserved in 
70% ethanol, the others were stored dry. 
These sampling methods are similar to those 
used in other studies (e.g.. De Winter, 1995). 

Identification and Analysis 

Most molluscs have been identified to spe- 
cies level and nomenclature is provided in 
Appendix I. A reference collection has been 



sent to the National Museums of Kenya, 
Nairobi, and the remainder of the material will 
be deposited in the National Museum of 
Wales, Cardiff (NMW Z 1993.062). Several 
species in the urocyclid genus Thapsia are 
present in the samples. Two of these {Thap- 
sia microleuca Verdcourt and T. eucosmia 
Pilsbry) are distinctive, but the other species, 
of which there are at least two, are difficult to 
separate and have been aggregated for the 
purposes of analysis. Further notes on iden- 
tification are provided in Appendix I. 

The number of individuals (separated into 
living and dead specimens based on the 
presence of body tissues and shell condition) 
has been recorded for each of the samples. 
However, since sampling effort varied 
amongst the plots, it does not provide a mea- 
sure of absolute species abundance or allow 
direct comparisons between the plots. Most 
of the analyses have therefore been based on 
presence and absence data. Nevertheless, 
the number of individuals does represent a 
measure of the relative abundance of species 
in the fauna and is therefore of some interest. 
The analysis of diversity patterns has fol- 
lowed Cameron's (1992) methods. Two mea- 
sures have been adopted, Whittaker's (1975) 
Index (/), the ratio of overall species number 
(S) to the mean number of species per plot 
{^), provides a measure of between plot 
differences. An index of 1 reveals identical 
faunas, whereas higher values demonstrate 
increasing differentiation. High values of / can 
either result from the geographical replace- 
ment of taxa within the same habitat or along 
habitat gradients (Cody, 1986). Following 
Cameron's (1992) methods, these effects 
have been examined by calculating the ratio 
of the variance of the number of sites per 
species to the maximum variance possible 
for the same values of S and =<. Where re- 
placement effects are important, as opposed 
to random effects due to sampling error, the 
achieved variance is low compared with the 
maximum possible. 



RESULTS 

Table 1 lists the 53 species recorded from 
the 31 plots during the survey. These consist 
of one slug and 52 snail species. Seven spe- 
cies were recorded only as dead shells and 
one of these, Cecilioides species, has been 
excluded from the analyses because it was 
only found on Plot 7, which lies adjacent to 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



165 



TABLE 1. Species of molluscs recorded in the 31 sampling plots in Kakamega Forest. 



Species 



12 3 4 5 6 7 



9 10 11 12 13 14 15 16 17 1! 



Elgonocyclus koptaweliensis 

Maizania elatior 

Succinea sp. 

Truncatellina ninagongonis 

Nesopupa bisulcata 

Pupisoma harpula 

Pupisoma orcula 

Pupisoma sp. A 

Pupisoma sp. В 

Acanthinula sp. 

Rhactiidina ciiiradzuluensis var. virginea 

Conulinus rutshuruensis major 

Cerastua trapezoidea lagariensis 

Micractaeon koptawelilensis 

Nothapalus sp. 

Subulona Clara 

Oreohomorus iredalei 

Pseudoglessula elegans 

Pseudopeas cf. yalaensis 

Curvella sp. A 

Curvella cf. babauiti 

Achatina stuhlmanni 

Limicolaria cf. saturata 

Punctum ugandanum 

Punctum sp. A 

Punctum sp. В 

Trachycystis iredalei 

Trachycystis ariel 

Prositala butumbiana 

Kaliella barrakporensis 

Kaliella iredalei 

carínate species 

Guppya quadnsculpta 

Afroconulus iredalei 

Trochozonites cf. medjensis 

Thapsia eucosmla 

Thapsia microleuca 

Thapsia spp. 

Gymnarion aloysiisabaudiae 

Chlamydarion oscitans 

urocyclid slug 

Halolimnohelix percivali 

Halolimnohelix plana 

Gonaxis elgonensis 

Gulella woodhousei 

Gulella osborni 

Gulella impedita 

Gulella ugandensis 

Gulella lessensis 

Gulella handeiensis 

Gulella disseminata 

Streptostele bacillum 

Cecilioides sp. 



+ + + + 



+ 


+ 


+ 




+ 


+ 


+ 


+ 










+ 


+ 


+ 
















+ 


+ 






+ 


+ 


+ 




+ 






+ 


+ 


+ 








+ 


+ 


+ 
+ 




+ 


+ 
+ 


+ 
+ 




+ 


+ 










+ 






+ 


+ 







+ + + + + 



+ + + + 
+ + + + 



+ + + + 



+ + + + + + 



+ + + + 



+ + 
+ + 
+ + + + + + + + + 



+ + + + 



+ + 

+ 
+ 
+ 
+ + + + 



+ + + + 

+ + + + + + 



+ + + + + + 



+ + + + 



+ + + + 



Site totals 



30 22 24 20 19 33 27 27 22 24 30 18 19 19 23 20 23 24 



{continued) 



166 



TATTERSFIELD 



TABLE 1. Species of molluscs recorded in the 31 sampling plots in Kakamega Forest, (continued) 



Species 



19 20 21 22 23 24 25 26 27 



Species 
28 29 30 31 totals 



Elgonocyclus koptaweliensis 

Malzania elatior 

Succinea sp. 

Truncatellina ninagongonis 

Nesopupa bisulcata 

Pupisoma harpula 

Pupisoma orcula 

Pupisoma s p. A 

Pupisoma sp. В 

Acanthinuia sp. 

Rhachidina chiradzuluensis var. virginea 

Conulinus rutshuruensis major 

Cerastua trapezoidea lagariensis 

Micrataeon koptawelilensis 

Nothapalus sp. 

Subulona Clara 

Oreohomorus iredalei 

Pseudoglessula elegans 

Pseudopeas cf. yalaensis 

Curvella sp. A 

Curvella cf. babauiti 

Achatina stuhlmanni 

Limicolaria cf. saturata 

Punctum ugandanum 

Punctum sp. A 

Punctum sp. В 

Trachycystis iredalei 

Trachycystis ariel 

Prositala butumbiana 

Kaliella barrakporensis 

Kaliella Iredalei 

carinate species 

Guppya quadrisculpta 

Afroconulus iredalei 

Trochozonltes cf. medjensis 

Thapsia eucosmia 

Thapsia microleuca 

Thapsia spp. 

Gymnarion aloysilsabaudiae 

Chlamydarion oscitans 

urocyclid slug 

Halolimnohelix percivali 

Halolimnohelix plana 

Gonaxis elgonensis 

Gulella woodhousei 

Gulella osborni 

Gulella impedlta 

Gulella ugandensis 

Gulella lessensis 

Gulella handeiensis 

Gulella disseminata 

Streptostele bacillum 

Cecilioides sp. 

Site totals 



+ + + 



+ + 
+ + 













2 


+ 










12 
6 


+ 










13 


+ 


+ 


+ 


+ 




13 




+ 






+ 


12 






+ 




+ 


9 


+ 










9 


+ 




+ 






7 
2 
2 

7 
7 


+ 


+ 




+ 




24 
1 


+ 


+ 


+ 


+ 




14 


+ 










5 




+ 


+ 


+ 


+ 


29 






+ 




+ 


24 


+ 




+ 


+ 




20 


+ 


+ 


+ 


+ 


+ 


24 


+ 








+ 


7 
3 


+ 


+ 


+ 


+ 




16 


+ 


+ 






+ 


19 


+ 


+ 








8 






+ 


+ 

+ 


+ 


16 

4 






+ 


+ 




14 


+ 


+ 


+ 


+ 


+ 


24 


+ 


+ 


+ 


+ 




27 


+ 


+ 


+ 


+ 




26 


+ 


+ 


+ 


+ 


+ 


27 








+ 


+ 


12 
4 






+ 


+ 




8 




+ 








12 


+ 
+ 


+ 


+ 
+ 


+ 


+ 


31 
22 


+ 
+ 




+ 




+ 


11 
1 
1 
8 


+ 






+ 


+ 


16 




+ 


+ 


+ 


+ 


20 


+ 


+ 


+ 


+ 


+ 


26 


+ 


+ 


+ 


+ 




19 


+ 






+ 


+ 


11 




+ 








11 
1 


+ 


+ 


+ 


+ 




26 




+ 


+ 


+ 


+ 


20 

1 



29 5 26 18 13 26 16 22 27 21 24 24 If 



the Islukhu River. Cecilioides is an open- 
country genus, and the shells were probably 
deposited onto the plots by the river. The 



other species only recorded as dead shells 
were Cerastua trapezoidea lagariensis, Gule- 
lla handeiensis, Nothapalus sp., Trachycystis 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



167 



Mean 

cumulative 

percentage 



Ol 

total 

species 

in 

the 

fauna 



50 



5 10 

Number 



15 

Ol Plots 



FIG. 2. Proportion of total indigenous forest fauna 
(50 species) as a function of the number of plots 
sanripled. 



ahel, Pupisoma sp. A and Rhachidina chirad- 
zuluensis var. virgínea; these have been in- 
cluded in the analyses because they are 
forest-dwelling species. Two species are 
confined to the plantation plots and thus, 
overall species number (S) in the indigenous 
forest is 50. A plot of cumulative species 
number against plot number (three random- 
ized plot orderings) reaches an asymptote af- 
ter 15 sites (Fig. 2). Because 27 indigenous 
forest plots were sampled, it is clear that the 
sampling detected the great majority of the 
species present in the plots, unless some im- 
portant micro-habitats containing specialist 
species were overlooked. 

The Fauna in Indigenous Forest 



faunas of most of the indigenous forest plots 
are closely similar, except for a group of 
seven plots which includes the Yala and Isi- 
ukhu river plots (Plots 6, 7, 22, 23 and 24) and 
Plots 8 antd 10, which lie in Kisere and Miaba 
forests respectively. This smaller group 
therefore appears to represent a forest-edge 
fauna, which is either found in riverine forest 
or in the smaller forest blocks; it is referred to 
as the "riverine forest" group. Six species 
are more frequent in the riverine forest group, 
whereas four are more frequent in the larger 
subset of indigenous forest plots (Table 2). 
Plot 22 is particularly isolated on the ordina- 
tion; it contains the only record for Notha- 
palus sp. and is the only indigenous forest 
plot to contain Limicolaria cf. satúrala. The 
species occurring in excess in the riverine 
forest group are relatively large species, 
whereas the four found in excess in the main 
forest group are small litter-dwellers. The 
plots containing disturbed or young indige- 
nous forest fall within the main cluster of in- 
digenous forest plots on the ordination, sug- 
gesting that their fauna does not differ 
substantially from that of the more mature 
and less disturbed stands. However, one 
species, Gulella ¡mpedita, is significantly less 
frequent in the young forest than in the ma- 
ture, undisturbed category (Fisher's Exact 
Test, P < 0.05). 

Species distribution throughout the study 
area has also been investigated by plotting 
site occupancy for each species on maps of 
the forest, and by examining species fre- 
quency in seven geographical groups of plots 
(Table 3). These analyses essentially confirm 
the findings of the ordination and show that 
most species are widespread throughout the 
forest. However, one species, Gulella ugan- 
densis, is widespread in the survey area, oc- 
curring from Miaba to the Yala River, but is 
apparently absent from the eight plots near 
Isecheno; this pattern does not obviously re- 
late to any of the environmental factors re- 
corded and the reason for it is not known. 



Possible groupings of plots have been in- 
vestigated using Reciprocal Averaging Ordi- 
nation (RAO) based on the presence and 
absence data from all 31 plots. Species dis- 
tribution and diversity levels have also been 
examined in geographically restricted groups 
of plots. RAO (Hill, 1973) is an ordination 
method which arranges the plots along arti- 
ficial axes according to their species comple- 
ments. The ordination shows (Fig. 3) that the 



Patterns of Diversity in the 
Indigenous Forest 

Total species number varies amongst the 
indigenous forest plots, but does not relate 
obviously to geographical position or any of 
the habitat or other environmental factors re- 
corded. Mean species per plot (a diversity) 
does not differ significantly amongst the 
seven geographical groups (Table 3; Fq^q ^, 



168 
ЮОг 

90 

80 

70 h 



TATTERSFIELD 



60 

Axis 2 

50 
40 
30 
20 
101- 






Riverine Forest Group 



f 



о • 

PinusCupressus 



Bischofia 



KEY 



Indigenous forest plots 
Plantation plots 

I Bischo 



10 



^ 



20 30 40 50 60 70 80 90 100 

Axis 1 

FIG. 3. Reciprocal averaging ordination of the plots. 



= 1.22, P = 0.34), between the riverine forest 
group and the main indigenous forest groups 
identified by the ordination (Table 3; F, 25 df 
= 0.16 P = 0.69) or between the undisturbed 
mature indigenous forest, the young forest 
and the highly disturbed indigenous stands 
(Table 4; F2 24 df. = 0-85, P = 0.44). There is 
thus no evidence of significant variation in 
alpha diversity levels throughout the survey 
area or between the different age and distur- 
bance categories of indigenous forest. Table 
4 also gives values for Whittaker's Index, /, 
and the proportion of maximum variance 
achieved for the sites per species statistic. 
Neither of these measures suggests that 



there are large differences between the plots 
or that there is significant geographical re- 
placement of taxa across the forest system. 

Other Characteristics of the Fauna 

The majority of the species recorded are 
small litter-dwellers and were retrieved by the 
sieving. Few species appeared to be re- 
stricted to specific microhabitats. However, 
Thapsia eucosmia was found almost exclu- 
sively on living vegetation and on tree trunks, 
Nesopupa bisulcata was most frequently re- 
corded on the underside of dry, large, fallen 
leaves on the forest floor, and Maizania elatior 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



169 



TABLE 2. Species recorded in significant excess in the riverine forest and otiner indigenous forest 
groups identified by the ordination 







Nu 


mber of Plots 










Riverine ■ 


forest 


Other 


indigenous 




Species 


(Max = 


-7) 


(M; 


3X = 


20) 


Probability* 


More frequent in riverine forest group 














Maizania elatior 


7 






4 




0.000 


Cerastua trapezoidea lagahensis 


5 






2 




0.005 


Chlamydarion oscitans 


5 






4 




0.023 


Thapsia eucosmia 


5 






3 




0.011 


Conulinus rutshuruensis major 


5 






1 




0.006 


Gulella ugandensis 


5 






4 




0.023 


More frequent in other indigenous forest 


group 












Nesopupa bisulcata 









13 




0.006 


Curvella cf. babauiti 


2 






19 




0.001 


Curvella s p. A 


2 






16 




0.023 


Punctum ugandanum 


1 






15 




0.009 



'Probability based on Fisher's Exact Test. Null hypothesis that frequencies are equal in both groups. 



TABLE 3. Mean number of species in geographical groups of plots and in {\no groups identified by 
the ordination 



Group 



Plots 



N Mean S.E. 



Geographical Groups 
Miaba 
Kisere 
Byungu'' 
Isiukhu 
Isecheno 
Yala 
Vihiga 

Ordination Groups 
Rivenne Forest 
Other indigenous forest 



10, 11 

8, 28 

1, 2, 3, 4, 9 

6, 7 

13, 14, 15, 16, 17, 18, 19, 21 

22, 23, 24, 27 

29, 30 



6, 7, 8, 10, 22, 23, 24 

1, 2, 3, 4, 5, 9, 11, 12, 13, 14, 15, 16, 17, 18, 19, 21, 

27, 28, 29, 30 



2 


27.00 


3.00 


2 


24.00 


3.00 


5 


23.60 


1.72 


2 


30.00 


3.00 


8 


22.88 


1.25 


4 


21.00 


3.34 


2 


24.00 


0.00 


7 


24.00 


2.49 


20 


23.20 


0.83 



^Excludes the young forest plots 5 and 12 



was found on the surface of the forest floor 
and appeared to have a highly aggregated 
distribution. No snails were found amongst 
epiphytes or on fallen tree branches, and no 
evidence was detected that molluscs live in 
the forest canopy. 

About 73% of the species in the indigenous 
fauna have shell sizes (maximum dimension) 
of 1 mm or less and 52% are less than 5 mm. 
About 1 7% have shell size exceeding 1 5 mm; 
Achatina stuhlmanni is the largest species in 
the fauna with a shell length in excess of 100 
mm. In total, 3,723 specimens were collected 
from the 27 indigenous plots, of which 1,504 
(approximately 40%) were classified as living. 
Guppya quadhsculpta constituted ^0% and 



1 1 .25%) of the living and overall totals respec- 
tively and was the commonest species. Only 
six species (12%o of the total recorded fauna) 
exceeded 5%o of all the specimens (i.e., living 
plus dead) recorded (Table 6) and these col- 
lectively represented 46.23%o of the overall 
total. Twenty five species (50%o of the fauna) 
each contributed less than 1 % to the total 
number of specimens. The numbers of dead 
and living specimens were very roughly equiv- 
alent for most species, except for Subulona 
clara and the aggregated Thapsia species, 
of which many more dead shells than liv- 
ing ones were found. During the fieldwork, it 
was noted that substantial accumulations 
of mostly dead Subulona clara shells were 



170 TATTERSFIELD 

TABLE 4. Estimates of mollusc diversity in indigenous forest and plantation 







Indigenous Forest 








Undisturbed, 
mature 


Disturbed, 
mature 


Young, recent 
colonisation 


All indigenous 


Plantation 


No. Plots 
Total species, S 


18 


5 


4 


27 
50 


4 
33 


Species per site 
Mean ± SE, a 
Range 

Whittaker's Index, / 


24.0 ±1.1 
13-33 


23.4 ±1.9 
19-30 


20.75 ±1.4 
18-24 


23.4 ±0.9 

13-33 

2.1 


15.25 13.6 

5-22 

2.2 


Sites per species 
Variance 
Maximum var. 
% achieved 


— 


— 


— 


60.64 

167.84 

36.1% 


— 



TABLE 5. Species recorded in significant excess in the plantation and indigenous 
forest categories 







N 


umt 


Der of Plots 












Indigenous 






Plantation 


Forest 




Species 


(Mi 


ax = 


4) 


(Max = 27) 


Probability* 


More frequent in plantation 












Limicolaha cf. saturata 




2 




1 


0.037 


More frequent in indigenous forest 












Micractaeon koptawelilensis 




Ü 




24 


0.001 


Punctum ugandanum 




Ü 




16 


0.043 


Gulella osborni 




1 




25 


0.008 


Gulella disseminata 




1 




25 


0.008 


Gulella impedita 









19 


0.016 


carínate species 









26 


0.000 



•Probability based on Fisher's Exact Test. Null hypothesis that frequencies are equal in both plan- 
tation and indigenous forest groups. 



sometimes found In the leaf litter beneath rot- 
ting and fallen logs. 

Plantation Faunas 

The small number of plantations sampled 
inevitably means that conclusions are tenta- 
tive. However, mean species per plot (Table 
4) is significantly lower in the plantations than 
in the indigenous forest (all 27 plots com- 
bined) (Fi 29 df = 1005, P = 0.004). Total 
species number is exceptionally low in the 
Pinus-Cupressus plantation. However, nei- 
ther the Pinus-Cupressus nor the Maesopsis 
plantations ordinate separately from the main 



cluster of forest plots, although the two 
Bischofia javanica plantations do (Fig. 3). Ta- 
ble 4 gives Whittaker's Index (I) for the 
plantation plots; it does not differ substan- 
tially from the value for the indigenous for- 
est. 

Six species are significantly more frequent 
in the indigenous forest than the plantations 
(Table 5), and all of these are small, litter- 
dwellers; they are all also relatively abundant 
in the indigenous forest (all being repre- 
sented by at least 60 shells, Table 6). One 
large species, Limicolaha cf. saturata, is sig- 
nificantly more frequent in the plantations 
than indigenous forest, and two others, the 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



171 



TABLE 6. Number of specimens (living plus dead), proportion of total specimens collected and rank 
order of each species in indigenous forest and plantation habitats 





Species 


Indigenous 


forest 


Plantation 




Rank 


Number 


% 


Number 


% 


Rank 


1 


Guppya quadrisculpta 


419 


11.25 


15 


4.5 


6 


2 


Thapsia spp. 


393 


10.56 


36 


10.81 


4 


3 


Gonaxis elgonensis 


248 


6.66 


47 


14.41 


2 


4 


Pseudoglessula elegans 


238 


6.39 


7 


2.1 


14 


5 


Subulona Clara 


215 


5.78 


3 


0.9 


22 


6 


Pseudopeas cf. yalaensis 


208 


5.59 


10 


3 


7 


7 


Kaliella barrakporensis 


174 


4.67 


21 


6.31 


5 


8 


Kaliella iredalei 


159 


4.27 


4 


1.2 


20 


9 


carinate species 


153 


4.11 





— 


— 


10 


Maizania elatior 


140 


3.76 


1 


0.3 


29 


11 


Curvella cf. babauiti 


129 


3.47 


8 


2.4 


10 


12 


Gulella disseminata 


100 


2.69 


5 


1.5 


18 


13 


Thapsia microleuca 


96 


2.58 


3 


0.9 


22 


14 


Punctum ugandanum 


76 


2.04 





— 


— 


15 


Gulella impedita 


75 


2.02 





— 


— 


16 


Micractaeon koptawelilensis 


73 


1.96 





— 


— 


17 


Gulella ugandensis 


71 


1.91 


49 


16.72 


1 


18 


Punctum sp. A 


63 


1.69 


9 


2.7 


8 


19 


Gulella osbomi 


62 


1.67 


1 


0.3 


29 


20 


Gymnarion aloysiisabaudiae 


59 


1.59 


7 


2.1 


14 


21 


Gulella woodhousei 


58 


1.56 


8 


2.7 


8 


22 


Curvella sp. A 


57 


1.53 


8 


2.4 


10 


23 


Nesopupa bisulcata 


52 


1.4 





— 


— 


24 


Thapsia eucosmia 


50 


1.35 





— 


— 


25 


Streptostele bacillum 


44 


1.18 


3 


0.9 


22 


26 


Trachycystis iredalei 


33 


0.89 


1 


0.3 


29 


27 


Prositala butumbiana 


22 


0.59 


2 


0.6 


26 


28 


Conulinus rutshuruensis major 


21 


0.56 


4 


1.2 


20 


29 


Chlamydahon oscitans 


20 


0.54 


8 


2.4 


10 


29 


Pupisoma harpula 


20 


0.54 


2 


0.6 


26 


31 


Afroconulus iredalei 


19 


0.51 


6 


1.8 


16 


32 


Punctum sp. В 


17 


0.46 





— 


— 


32 


Gulella lessensis 


17 


0.46 





— 


— 


32 


Truncatellina ninagongonis 


17 


0.46 





— 


— 


35 


Pupisoma sp. A 


15 


0.43 





— 


— 


36 


Gerast и a trapezoidea lagariensis 


14 


0.38 





— 


— 


36 


Halolimnohelix plana 


14 


0.38 


6 


1.8 


16 


36 


Pupisoma orcula 


14 


0.38 


5 


1.5 


18 


39 


Elgonocyclus koptaweliensis 


10 


0.27 





— 


— 


39 


Achatina stuhlmanni 


10 


0.27 


3 


0.9 


22 


39 


Pupisoma sp. В 


10 


0.27 





— 


— 


42 


Succinea sp. 


9 


0.24 





— 


— 


43 


Oreohomorus iredalei 


7 


0.19 


1 


0.3 


29 


44 


Limicolaria cf. saturata 


6 


0.16 


39 


11.71 


3 


45 


Trachycystis ariel 


4 


0.11 





— 


— 


45 


Trochozonites cf. medjensis 


4 


0.11 


2 


0.6 


26 


47 


Rhachidina chiradzuluensis 


3 


0.08 





— 


— 


48 


Acanthinula sp. 


2 


0.05 





— 


— 


49 


Gulella handeiensis 


1 


0.03 





— 


— 


49 


Nothapalus sp. 


1 


0.03 





— 


— 



urocyclld slug and Halolirvnohelix percivali, 
are confined to the Bischofia plantation (al- 
though in one plot only). The abundance of 
several species also differs substantially be- 



tween the plantations and indigenous forest 
(Table 6). In particular, Gulella ugandensis is 
relatively much more abundant in the planta- 
tions, whereas Guppya quadrisculpta and 



172 



TATTERSFIELD 



Pseudoglessula elegans rank higher in the in- 
digenous plots than in the plantations. 



DISCUSSION 

The Fauna of Kakamega Forest 

The analysis of cumulative species number 
suggests that the sampling has provided an 
accurate picture of indigenous forest diver- 
sity levels. However, many additional species 
of terrestrial molluscs have been reported 
previously from the Kakamega Forest area 
(Germain, 1923; Pickford, 1995; Pain, 1957; 
Verdcourt, 1 983, 1 988). A comprehensive ap- 
praisal of these species would require critical 
examination of all the available original ma- 
terial and is beyond the scope of this study. 
However, some species can be eliminated 
from the list for the indigenous forest on the 
grounds of conspecificity (e.g., see Pseudo- 
peas yalaensis in Appendix I), and others 
have almost certainly been misidentified. For 
example, Germain (1923) lists Trachycystis 
planulata Preston from "les bords de la 
rivière Yala," but because this species was 
deschbed from 9,000-10,000' on Mount 
Kenya, it seems highly improbable that it oc- 
curs at Kakamega. Verdcourt (1962) dis- 
cusses another error made by Germain. 
Some of these species may be associated 
with non-forest habitats or ecotones and are 
therefore not part of the indigenous forest 
fauna. Furthermore, the need for major revi- 
sions of some taxa, such as the genus Thap- 
sia, means that it is not possible to make pre- 
cise estimates of S for the forest. However, 
by taking these factors into account as far as 
possible, the total list for the forest might be 
estimated at roughly 70-80 species. This 
substantially exceeds the total species num- 
ber (S) of 50 found in the indigenous forest 
plots. 

Local Patterns and Levels of Diversity 

Solem (1984) has reviewed the evidence 
on worldwide land snail diversity and has 
shown that sympatric levels in most parts of 
the world are low, with most sites typically 
supporting less than ten species. However, a 
few places are known to support much richer 
faunas. The highest site diversity reported to 
date is on the Manukau Peninsula, North Is- 
land, New Zealand, where more than 60 spe- 
cies have been found to exist microsympat- 



rically in lowland patches of relict forest 
(Solem et al., 1981). Solem (1984) also cites 
the work of Fred Thompson and John Stani- 
sic, who have recorded 25-30 species and 
40 species from sites in the Greater Antilles 
(Hispaniola and Jamaica) and from rain forest 
in Queensland, New South Wales, respec- 
tively. At least 50 species are present in some 
Tanzanian coastal and upland forests (K. Em- 
berton, pers. comm., and Tattersfield, un- 
published). Alpha diversity can also be high in 
rich temperate sites. Cameron (1986) reports 
a median species number of 15 in coniferous 
forest with mull humus soils in British Colum- 
bia; Wäreborn (1969) assessed there to be a 
mean snail species number of 16.58 per plot 
in his richest Swedish woodlands; and 
Waiden (1981) reported mean snail species 
as 25.2 in five broad-leaved woodland site on 
calcareous moraine in Sweden. Old wood- 
lands in the Pennines (Cameron, 1978a) sup- 
port a mean snail number of 28.25 species. 
Values of S can also be high in optimum tem- 
perate forests. Coney et al. (1982) reported 
57 species from 37 forest sites in Tennessee; 
Tattersfield (1990) found 31 species in En- 
glish woodland sites on both limestone and 
acidic geologies; and Branson & Batch (1970) 
recorded a total of 45 species in Kentucky. 
Based on these studies, it is therefore appar- 
ent that the mean plot diversity of 24 species/ 
site and the overall total of 50 species from 
the indigenous Kakamega stands are rela- 
tively high in a worldwide context. Microhab- 
itat specializations can help account for high 
snail diversity in some faunas (Cameron, 
1978b), but further work would be needed to 
assess whether this is important in the Kaka- 
mega fauna. However, the current study re- 
vealed very few examples of possible micro- 
habitat specializations. 

Calculation of the Shannon-Weaver diver- 
sity index (H) (Poole, 1974) and index of 
evenness (J = H/Hmax), which take into 
account the number of specimens contrib- 
uted by each species as well as S, also indi- 
cates that the Kakamega fauna is more di- 
verse than almost all other woodland/forest 
faunas that have been studied (Table 7); in- 
deed, these indices show that it is on a par 
with the richest known fauna described from 
New Zealand (Solem et a!., 1981). 

Verdcourt, cited in Solem (1984), consid- 
ered that "the wet forests of East Africa may 
yield up to 20-25 species from a small area 
although more frequently such a collection 
yields only 10-15." The only systematically 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



173 



TABLE 7. Overall species number (S) and Shannon-Weaver diversity (H) and evenness 
(J) indices for molluscan faunas from Kakamega Forest and other forest systems 



Area/Forest type 


S 


H 


J 


Source 


Kakamega Indigenous Forest 










Live snails 


43 


3.23 


0.86 


This study 


All snails 


50 


3.27 


0.84 




New Zealand 










Live snails 


45 


3.22 


0.85 


Solem et al. (1981) 


All snails 


56 


3.26 


0.81 




Tennessee, U.S.A. 


57 


2.93 


0.73 


Coney et al. (1982) 


British Columbia* 










Mull litter sites 


26 


2.62 


0.80 


Cameron (1986) 


Intermediate litter sites 


16 


2.17 


0.78 




Mor litter sites 


9 


1.72 


0.78 




Finland (islands) 










Deciduous woodland Site 9 


21 


2.60 


0.85 


Valovirta(1984) 


Sites 


21 


2.35 


0.77 




Site 6 


19 


2.26 


0.77 




Sites 


20 


2.25 


0.75 




Sweden 










Moist meadow woods 


31 


2.31 


0.67 


Wäreborn (1969) 


Drier mixed woods 


9 


1.41 


0.64 




Moist mixed woods 


9 


1.23 


0.56 




Shropshire, United Kingdom 


12 


1.75 


0.64 


Cameron (1982) 



'excludes slugs 



collected information about mollusc diversity 
patterns in African forests is from De Winter's 
(1995) series of 20 lowland rain forest plots 
(from a total area of approx. 48 km^) in the 
Ogoou, Maritime region of Gabon, West Af- 
rica. Using similar methods to those adopted 
here (snails extracted from approx. 4 I of lit- 
ter), he reported mean species number per 
litter sample as 3.4 (range 0-7, median 3), 
with the number of specimens ranging from 
0-20 (mean 7.9). It was estimated that the 
litter zone supported 28 (± 2.2) species. A 
similar number of species (about 25) in the 
Kakamega fauna can be classified as small 
(shell size < 5 mm) litter dwellers. Large spe- 
cies were not sampled in plots so it is not 
possible to compare directly alpha diversity 
levels, but it is evident that many more spe- 
cies and individuals were found in the Kaka- 
mega litter samples than in Gabon; the aver- 
age number of shells recovered from the 
Kakamega indigenous forest plots was 137, 
most of which were small litter-dwellers. 

De Winter (1995) recorded a total of 32 
species (including four freshwater species) 
and estimated that the whole forest might 
support 39 terrestrial species by taking into 
account previous records; these values of S 
are also lower than at Kakamega. Taking into 
account the litter dwellers only, which form 
82% and 51 % of the recorded fauna in Kaka- 



mega Forest and Gabon respectively, the ra- 
tio S/c< is substantially greater in Gabon than 
at Kakamega. This indicates that there is 
substantially more variation in the fauna 
amongst the Gabonese plots than in Kaka- 
mega Forest, but it is not possible to estab- 
lish whether this is related to habitat variation 
or to the geographical replacement of sister 
species. Such a conclusion was also tenta- 
tively made by De Winter (1995), who noted 
that there were few species in common from 
similar forest types over a distance of less 
than 50 km, but that a fair number of addi- 
tional species were found. Solem (1984) pre- 
dicted that the median total linear range of all 
land snails worldwide would be found to be 
less than 100 km and probably less than 50 
km, and Cameron (1992) has reported taxo- 
nomic replacement effects over short dis- 
tances in semi-arid habitats in Western Aus- 
tralia. The commonest of De Winter's (1995) 
species was found in only half of the litter 
samples, whereas 1 8 (36.7%) of the 49 indig- 
enous forest species in Kakamega Forest 
were found in more than 50% of the plots; 
this also suggests that the fauna of the Ga- 
bonese Forest is very much more heteroge- 
neous than at Kakamega. In Kakamega 
Forest, both / and the proportion of the max- 
imum variance of the sites per species sta- 
tistic achieved are broadly similar to the val- 



174 



TATTERSFIELD 



ues reported by Cameron (1992) for faunas 
from woodland in the English South Downs 
(maximum separation 75 km) and from rock 
habitats in the Pennines (separation 30 km). 
They differ from British Columbian coastal 
forests (maximum site separation of 300 km), 
which have a more homogeneous fauna, and 
from faunas from the Oscar and Napier 
ranges of Western Australia (about 160 km 
maximum separation), where both / (9.82) 
and the proportion of variance achieved (8%) 
revealed strong replacement effects in large 
camaenid taxa (Cameron, 1992). 

Origins of the Kakamega Forest Fauna 

The African climate has been unstable dur- 
ing the Pleistocene, and this had a strong 
influence on forest cover. Many parts of equa- 
torial Africa were dry and cool during the last 
ice age, and Lake Victoria was almost non- 
existent at around 14000 BP, during the Last 
Glacial Maximum in Afhca (Kendall, 1 969; Liv- 
ingstone, 1980). With the absence of mois- 
ture-laden convection currents from the Lake 
Victoria waterbody, the Kakamega area 
would not have supported forest cover. There 
is evidence (reviewed in Hamilton, 1982) that 
forest in equatorial Africa became confined to 
a relatively small number of discrete areas 
during times of extended glaciation and that 
the current distribution patterns of many 
groups of forest species can be accounted for 
by subsequent expansion from these refugia 
(Hamilton, 1982; Kingdon, 1990). Former for- 
est refugia are now often rich in endemic spe- 
cies. The Gabonese forest studied by De 
Winter (1995) falls within or close to such a 
refugium at the Gabon-Cameroon border 
(Kingdon, 1990), whereas Kakamega Forest 
does not. The mollusc fauna of Kakamega 
Forest must have recolonised after the cli- 
matic amelioration and redevelopment of for- 
est. Palynological evidence (Hamilton, 1972; 
Kendall, 1968; Livingstone, 1967) shows that 
at 12000-10000 years BP, forest spread from 
a refugium in eastern Zaïre (possibly extend- 
ing into west Uganda) across what is now 
Uganda and the Kakamega area of west 
Kenya. Furthermore, studies on the avifauna 
(Zimmerman, 1972) and tree flora (Hamilton, 
1982) of Kakamega Forest have shown that 
they are impoverished versions of the Central 
African biotas, which also strongly suggests 
that recolonisation was from west or central 
Africa. The mollusc fauna of Kakamega Forest 
would suggest a similar route of recolonisa- 



tion because Pain (1957) has shown that 
Achatina stuhlmanni is commonest in Zaïre 
west of the Upper Ituri River and the genera 
Prositala, Pseudoglessula (Ischnoglessula), 
Oreohomorus, Nothapalus, Conulinus, Tro- 
chozonites {Zonitotrochus), and Gulella (sect. 
Silvigulella) all have west and central African 
affinities (Verdcourt, 1972). 

Regional Patterns and Levels of Diversity 

The mollusc fauna of East Africa (Kenya, 
Uganda and Tanzania) contains about 1,015 
terrestrial species of which about 844 (83%) 
are forest dwellers (Verdcourt, 1972). Based 
on the current survey results plus other 
records it can therefore be assessed that the 
fauna of Kakamega Forest supports about 
5.8-9.5% of the potential forest fauna of the 
region. This low proportion indicates that 
there must be substantial geographical re- 
placement of taxa throughout the region. As 
discussed above, this situation is probably 
not unusual worldwide; Solem (1984) sum- 
marised the available evidence which sug- 
gests that allopatric diversity is exceptionally 
high amongst land snails. The regional levels 
and patterns of terrestrial snail diversity are 
very different in temperate northwest Europe, 
where the fauna is relatively homogeneous 
over large geographical areas (Cameron, 
1 995). Kerney & Cameron (1 979) cover a land 
area about 38% larger than East Africa, but it 
supports only 279 terrestrial molluscs (Ker- 
ney & Cameron, 1 979), of which about 1 52 or 
54.5% are forest or woodland species. Rich 
sites in Britain may support a large propor- 
tion of the national gastropod fauna; for ex- 
ample, Whitcombe Wood on Jurassic lime- 
stone in Gloucestershire (Boycott, 1934) 
supports 37 gastropods (28 snails and nine 
slugs), which represent approximately 33% 
of the total British land gastropod fauna (or 
32% of the snail fauna). 

Bernard Verdcourt (in Rogers & Home- 
wood, 1982) lists 111 terrestrial gastropod 
species and subspecies known from the 
Usambara Mountains in northern Tanzania. 
These mountains support lowland and inter- 
mediate rain forest communities and have 
rainfall (1919 mm/year at Amani (91 1 m asi)) 
and temperature (21.7-28.3''C mean maxi- 
mum at Amani) regimes broadly rather similar 
to Kakamega Forest. However, the list in 
Rogers & Homewood (1982) has only seven 
species in common with the total reported 
fauna (i.e., all records) from Kakamega For- 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



175 



est. Unlike Kakamega Forest, the Usambaras 
are thought to have have supported forest 
cover for millions of years, probably since be- 
fore the Miocene, and contain very high num- 
bers of endemic species (Rogers & Home- 
wood, 1982). The richer overall fauna in the 
Usambaras may be related to this long period 
of forest stability, plus its greater extent and 
more diverse physical geography and habi- 
tats. However, at individual sites in the East 
Usambara, values of both S and ^ are 
broadly similar to those found in Kakamega 
Forest (Tattersfield, unpublished). 

Plantation Faunas 

Impoverished snail faunas and low mollus- 
can diversity levels have been reported from 
plantations elsewhere (Cameron, 1978a). The 
very small number of plantations examined 
restricts firm conclusions, but there is evi- 
dence that both the Maesopsis and Bischofia 
plantations also have impoverished faunas, 
and that the latter also appears to be com- 
positionally different from the indigenous for- 
est. Differences in the diversity and species 
composition in several groups of soil arthro- 
pod have been reported between primary 
forest and Maesopsis eminil plantation in the 
East Usambaras in Tanzania (Mahunka, 
1989). Maesopsis eminii, which is not an in- 
digenous species in northeast Tanzania and 
where it is considered to be an ecological 
weed, has also been shown to alter radically 
the characteristics of the litter and topsoil 
(Hamilton, 1989) and to be associated with a 
loss of the organic soil horizons (Macfadyen, 
1989; Binggeli & Hamilton, 1993). Whether 
Maesopsis itself or other factors, such as 
canopy loss, drainage or disturbance, is re- 
sponsible for organic horizon loss is not clear 
(Macfadyen, 1989), although such changes 
might be expected to have a large effect on 
the small, litter-dwelling snails, in line with 
that reported here. The six species (Table 5) 
that are less frequent in the plantations may 
thus, tentatively, be regarded as indicators of 
indigenous forest, in the same way that some 
species can be used to assist in the differen- 
tiation of ancient and secondary woodland in 
Britain (Kerney & Stubbs, 1980). Lovejoy et 
al. (1986) have demonstrated that substantial 
changes occur in microclimate at the edges 
of recently cleared rainforest. It is interesting 
in this context that the species that are sig- 
nificantly less frequent in both the riverine 
(forest edge) indigenous plots and in the 



plantations are all small, litter dwellers that 
might be expected to be more suceptible to 
such changes. 

Conservation Implications 

The mollusc fauna of Kakamega Forest 
does not contain the high numbers of en- 
demic species found in some other forest 
systems in East Africa (for example the Tan- 
zanian Usambara ranges (Rogers & Home- 
wood, 1982)). However, in common with the 
bird and butterfly faunas and the flora, its ma- 
lacofauna does support central and west Af- 
rican elements, which are scarce or absent 
from most of Kenya and Tanzania. The mol- 
luscs of Kakamega Forest are therefore of 
some biogeographical interest, and they sup- 
plement this previously acknowledged con- 
servation importance of the Kakamega For- 
est system. 

Further information is required about local 
diversity patterns in other African forests, but 
if faunas from other forest systems are rela- 
tively uniform like in Kakamega Forest, then 
this, and the high level of allopatric diversity 
throughout East African forests, have several 
potential conservation implications. Notwith- 
standing the probable importance of edge ef- 
fects in forest fragments, the majority of mol- 
lusc species found during the study could 
probably be conserved in a relatively small 
area of forest. Of course, there are many im- 
portant reasons why the size of protected ar- 
eas should be maximised, but based on 
these conclusions and from a solely mollus- 
can perspective, it is apparent that the pro- 
tection of a large number of widely distrib- 
uted, small forest blocks might be more 
effective at conserving regional molluscan 
biodiversity than would the retention of a 
smaller number of large forest areas of equal 
extent. The degree to which regional mollusc 
biodiversity could be maintained by conserv- 
ing the endemic-rich forest systems needs 
further survey and analysis. 



ACKNOWLEDGEMENTS 

This project could not have been under- 
taken without the help given by many people, 
all of whom I would like to thank. In particular, 
I would like to thank Dr. Bernard Verdcourt 
(Kew), who helped with the identifications 
and shared his immense knowledge about 
East African molluscs. Drs. Mary Seddon and 



176 



TATTERSFIELD 



Graham Oliver of the National Museum of 
Wales, and Dr. Peter Mordan, Fred Naggs 
and Kathie Way of The Natural History Mu- 
seum (Malacology Section) kindly made their 
extensive collections available and also of- 
fered advice and assistance. Professor Rob- 
ert Cameron (Sheffield) and Dr. Laurence 
Cook (Manchester) provided advice and sup- 
port, and Dr. A. С van Bruggen (Leiden) as- 
sisted with identification. Computer and 
other facilities were made available by Penny 
Anderson Associates. Gene Hammond 
helped with the drawings. Mollusc speci- 
mens were kindly loaned by V. Heros (MNHN, 
Paris), W. K. Emerson (AMNH, New York), 
Tanya Kausch (MCZH, Harvard) and Dr. R. 
Kilias (Museum für Naturkunde der Humboldt 
Universität, Berlin). 

I would like to thank the Office of the Pres- 
ident of Kenya for permission to undertaken 
the research. Dr. Richard Bagine and Mu- 
sombi Kibberenge (National Museums of 
Kenya) kindly provided advice and help when 
organising the fieldwork. The Kenya Wildlife 
Service gave support in the field. Dr. Glyn 
Davies and other members of the KIFCON 
team provided information about Kakamega 
and other Kenyan forests. I am also very 
grateful to my local assistants, Nixon Sagita, 
Solomon Astweje and Caleb, who both 
helped with sampling and were invaluable 
guides in the forest. 

The work was kindly supported by grants 
from the British Ecological Society/Coal- 
bourn Trust, the People's Trust for Endan- 
gered Species and the Percy Sladen Fund of 
the Linnean Society. Ross Lab of Maccles- 
field and J. G. Stanniar & Co., Manchester 
generously supplied equipment. 



LITERATURE CITED 

ADAM, W., 1954, Etudes sur les mollusques de 
l'Afrique centrale et des régions voisines. I. Ver- 
tiginidae et Valloniidae. Volume Jubilaire Victor 
Van Straelen, Bruxelles, 2: 725-817. 

ADAM, W., 1957, Etudes sur les mollusques de 
l'Afrique centrale et des régions voisines. II. Ver- 
tiginidae et Valloniidae (Addendum). Institut 
Royal des Sciences Naturelles de Belgique, Bul- 
letin, 33: 1-26. 

BINGGELI, P. & A. С HAMILTON, 1993, Biological 
invasion by Maesopsis enninii in the East Usam- 
bara forests, Tanzania. Opera Botánica, 121: 
229-235. 

BOYCOTT, A. E., 1934, The habitats of land Mol- 
lusca in Britain. Journal of Ecology, 22: 1-38. 



BRANSON, B. A. & D. L. BATCH, 1970, An eco- 
logical study of valley forest gastropods in a 
mixed mesophytic situation in northern Ken- 
tucky. The Veliger, 12: 333-350. 

BRUGGEN, A. С VAN, 1969, Studies on the land 
molluscs of Zululand with notes on the distribu- 
tion of Southern Africa. Zoologische Verhan- 
delingen (Leiden), 103: 1-116. 

BRUGGEN, A. С VAN & B. VERDCOURT, 1993, 
Notes on a sinistral specimen of Nesopupa bisul- 
cata (Jickeli) (Gastropoda Pulmonata: Vertig- 
inidae) from Kenya. Basteria, 57: 103-105. 

CAMERON, R. A. D., 1978a, Terrestrial snail faunas 
of the Malham area. Field Studies, 4: 715-728. 

CAMERON, R. A. D., 1978b, Differences in the 
sites of activity of co-existing 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(2): 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: 271-285. 

CAMERON, R. A. D., 1995, Patterns of diversity in 
land snails. The effects of environmental history. 

In: A. C- VAN BRUGGEN, S. M. WELLS, & TH. С M. 

KEMPERMAN, eds.. Biodiversity and conservation 
of the Mollusca. Backhuys, Oegstgeest-Leiden, 
Netherlands. 

CODY, M., 1986, Diversity, rarity and conservation 
in Mediterranean-climate regions. In м. E. soule, 
ed.. Conservation Biology: 123-152. Sunder- 
land, Mass: Sinauer. 

CONEY, С С, W. A. TARPLEY, J. С WARDEN & 
J. W. NAGEL, 1982, Ecological studies of land 
snails in the Hiwassee River Basin of Tennessee, 
USA. rvialacological Review, 15: 69-106. 

DE WINTER, T., 1995, Gastropod diversity in a rain 
forest in Gabon, West Africa. In: A. с van 

BRUGGEN, S. M. WELLS & TH. С M. KEMPERMAN, 

eds.. Biodiversity and conservation of the Mol- 
lusca. Backhuys, Oegstgeest-Leiden, Nether- 
lands. 

FADEN, R. В., 1970, A preliminary report on the 
Kakamega Forest in Kenya. Unpublished report 
from the East African Herbarium, Nairobi, Kenya. 

GERMAIN, L., 1923, Voyage de M. Guy Babault 
dans L 'Afrique Orientale Anglaise. Resultats sci- 
entifique. Mollusques terrestres et fluviátiles, 
Paris. 

HAMILTON, A. C, 1972, The interpretation of pol- 
len diagrams from highland Uganda. Palaeo- 
ecology of Africa, 7: 45-149. 

HAMILTON, A. C, 1982, Environmental history of 
East Africa. A study of the Quaternary. Academic 
Press, London. 

HAMILTON, A. C, 1989, Some effects of Maesop- 
sis on litter and soil on the East Usambaras. In: a. 
с HAMILTON & R. BENSTED-SMiTH, eds., Forest 
conservation in the East Usambara Mountains, 
Tanzania. lUCN, Switzerland and Cambridge, 
United Kingdom. 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



177 



HILL, M. O., 1973, Reciprocal averaging ordina- 
tion: an eigenvector method of ordination. Jour- 
nal of Ecology, 61 : 237-249. 

KENDALL, R. L., 1969, An ecological history of the 
Lake Victoria Basin. Ecological Monographs, 39: 
121-176. 

KERNEY, M. P. & R. A. D. CAMERON, 1979, A fte/d 
guide to the land snails of Britain and North-west 
Europe. Collins, London. 

KERNEY, M. P. & A. E. STUBBS, 1980, The con- 
servation of snails, slugs and freshwater mussels. 
Nature Conservancy Council, Peterborough. 

KINGDON, J., 1990, Island Africa. The evolution of 
Africa's rare animals and plants. Collins, London. 

KOKWARO, J. O., 1984, The Kakamega Forest of 
Kenya: the easternmost relic of the equatorial 
rain forests of Africa. In a report on a workshop 
on the plant communities of Kenya. EMTA, 
Nairobi, Kenya. 

LIVINGSTONE, D. A., 1967, Postglacial vegetation 
of the Ruwenzori Mountains in Equatorial Africa. 
Ecological Monographs, 37: 25-52. 

LIVINGSTONE, D. A., 1980, Environmental 
changes in the Nile headw/aters. In: м. a. j. 
WILLIAMS & H. FAURE, eds.. The Sahara and the 
Nile. Balkema, Rotterdam. 

LOVEJOY, T. E., R. O. BIERREGAARD, A. B. RY- 
LANDS, J. R. MALCOLM, С E. OUINTELA, L. H. 
HARPER, K. S. BROWN, A. H. POWELL, G. V. N. 
POWELL, H. O. R. SCHUBART, & M. В. HAYS, 
1986, Edge and other effects of isolation on Am- 
azon forest fragments. In: м. E. soule, ed., Сол- 
servation biology, 257-285. Sinauer, Sunder- 
land, Mass. 

LUCAS, G. LI., 1968, Kenya. In: i. hedberg & О 
HEDBERG, eds.. Conservation of vegetation in Af- 
rica south of the Sahara. Acfa Phytogeographica 
Suecica, 54. 

MACFADYEN, A., 1989, A bnef study of the rela- 
tionships between Maesopsis and some soil 
properties in the East Usambaras. In: a. c. Hamil- 
ton & R. BENSTED-SMiTH, eds.. Forest conserva- 
tion in the East Usambara Mountains Tanzania. 
lUCN, Switzerland and Cambridge, United King- 
dom. 

MAHUNKA, S., 1 989, Preliminary studies of the soil 
fauna of primary and secondary submontane 
rain forests on the East Usambaras. In: a. c. 
HAMILTON & R, BENSTED-SMITH, Forest Conserva- 
tion in the East Usambara Mountains Tanzania. 
lUCN, Switzerland and Cambndge, UK. 

MURIUKI, J. W. & M. H. TSINGALIA, 1990, A new 
population of de Brazza's monkey in Kenya. 
On/x, 24: 157-162. 

MUTANGAH, J. G., O. MWANGANI & P. K. 
MWAURA, 1992, Kakamega Forest: a vegetation 
survey report. Unpublished report by: Kenya In- 
digenous Forest Conservation Project, National 
Museums of Kenya, Nairobi. 

PAIN,T., 1957, On the occurrence of Achaf/nasfuh- 
Imanni V. Martens in Kenya. Revue de Zoologie 
et Botanique Africaine, 56: 345-348. 

PICKFORD, M., 1995, Fossil landsnails of East Af- 



rica and their palaeoecological significance. 
Journal of African Earth Sciences, 20: 167- 
226. 

PILSBRY, H. A., 1919, A review of the land mol- 
lusks of the Belgian Congo chiefly based on the 
collections of the American Museum Congo Ex- 
pedition, 1909-1915. Bulletin of the Amehcan 
Museum of Natural History, 40: 1-370. 

POOLE, R. W., 1974, An introduction to quantita- 
tive ecology. McGraw-Hill, Tokyo. 

ROGERS, W. A. & K. M. HOMEWOOD, 1982, Spe- 
cies richness and endemism in the Usambara 
mountain forests, Tanzania. Biological Journal of 
the Linnean Society, 18: 197-242. 

SOLEM, A., 1984, A world model of land snail di- 
versity and abundance. In: World-wide snails. 
Biogeographical studies on non-marine Mol- 
lusca. E. J. Brill, Leiden. 

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. 

TATTERSFIELD, P., 1990, Terrestnal mollusc fau- 
nas from some south Pennine woodlands. Jour- 
nal of Conchology, 33: 355-374. 

TATTERSFIELD, P., 1995, A new species of Trun- 
catellina (Gastropoda: Vertiginidae) from Mount 
Kenya and new distributional records of Trunca- 
tellina ninagongonis (Pilsbry) from East Africa. 
Journal of Conchology, 35: 241-245. 

TSINGALIA, M. H., 1990, Habitat disturbance, se- 
verity and patterns of abundance in Kakamega 
Forest, Western Kenya. African Journal of Ecol- 
ogy, 28: 213-226. 

VERDCOURT, В., 1962, A note on M. Guy Ba- 
bault's specimens of Achatina stuhlmanni von 
Martens. Journal of Conchology, 25: 86. 

VERDCOURT В., 1964, The genus Maizania Bgt. 
(Gastropoda, Maizaniidae) in eastern Africa. 
Journal of the East Africa Natural History Society, 
24(5): 1-22. 

VERDCOURT, В., 1970, Notes on Kenyan land and 
freshwater snails. 10. Variation in the Gulella 
ugandensis complex (Streptaxidae). Basteria, 
34: 45-54. 

VERDCOURT, В., 1972, The zoogeography of the 
non-marine Mollusca of East Africa. Journal of 
Conchology, 27: 291-348. 

VERDCOURT, В., 1981, Contributions to the 
knowledge of the Helicidae-Bradybaeninae of 
Zaïre. Revue de Zoologie Africaine, 95: 525-559. 

VERDCOURT, В., 1982a, A new operculate genus 
from East Africa — Elgonocyclus (Mollusca, 
Cyclophoridae), a preliminary diagnosis. Baste- 
ria, 46: 77-78. 

VERDCOURT В., 1982b, Notes on East Afncan 
land and freshwater snails, 12-15. Zoologische 
Mededelingen, 56: 217-236. 

VERDCOURT, В., 1983, A list of the non-manne 
Mollusca of East Africa (Kenya, Uganda, Tanza- 
nia, excluding Lake Malawi). Achatina, 11: 200- 
239. 

VERDCOURT, В., 1988, The rediscovery of Pune- 



178 



TATTERSFIELD 



tum ugandanum (E. A. Smith) (Pulmonata: Punc- 
tidae). Archiv für Molluskenkunde, 118:1 49-1 51 . 

VERDCOURT, В., 1990, A new genus and species 
of land snail, Micractaeon kakamegaensis, from 
Kakamega Forest, Kenya (Pulmonata: ?Ferrus- 
saciidae). Archiv für Molluskenkunde, 121: Si- 
se. 

VERDCOURT, В., 1991a, Illustrations of E/gonocy- 
clus koptaweliensis (Germain). Conchologists' 
Newsletter, 118:409. 

VERDCOURT, В., 1991b, Some notes on East Af- 
rican Endodontoid snails — Part 1 . Concholo- 
gists' Newsletter, 116: 352-359. 

VERDCOURT, В., 1991c, Some notes on East Af- 
rican Endodontoid snails — Part 2. Concholo- 
gists' Newsletter, 117: 371-378. 

WALDÉN, H., 1981, Communities and diversity of 
land molluscs in Scandinavian woodlands. 1. 
High diversity communities in taluses and boul- 
der slopes in SW Sweden. Journal of Conchol- 
ogy, 30: 351-372. 

WAREBORN, I., 1969, Land molluscs and their en- 
vironments in an oligotrophic area in southern 
Sweden. Oikos, 20: 461-479. 

WHITE, F., 1983, The vegetation of Africa: a de- 
scriptive memoir to accompany the Unesco/ 
AETFAT/UNSO vegetation map of Africa. 
Unesco, Paris. 

WHITTAKER, R. H., 1975, Evolution of species di- 
versities in land communities. Evolutionary Biol- 
ogy, 10: 1-68. 

ZIMMERMAN, D. A., 1972, The avifauna of the 
Kakamega Forest, Western Kenya, including a 
bird population study. Bulletin of the American 
Museum of Natural History, 149: 255-340. 

Revised Ms. accepted 1 March 1996 



APPENDIX I. SPECIES NOMENCU\TURE 
AND NOTES ON IDENTIFICATION 

Museums are referred to in this appendix as 
follows: 

BMNH — Natural History Museum, London, 
United Kingdom 

NMW — National Museum of Wales, Cardiff, 
United Kingdom 

AMNH — American Museum of Natural His- 
tory, New York, U.S.A. 

MCZH — Museum of Comparative Zoology, 
Harvard, Cambridge, Mass., U.S.A. 

MNHN — Muséum Nationales d'Histoire Na- 
turelles, Paris, France 

Systematic List 

Elgonocyclus koptaweliensis (Germain) — 

Verdcourt (1982a, 1991a). 
Maizania elatior (von Martens) — Verdcourt 

(1964). 



^^Succinea'' sp. — A revision of African Suc- 
cineidae is required before the genera 
present can be elucidated or the species 
named consistently. The genus Suc- 
cinea Drap, probably does not occur in 
Africa (Verdcoun, 1972). 

Truncatellina ninagongonis (Pilsbry) — The 
Kakamega material matches the holo- 
type (MCZH 77268) collected from Mt. 
Ninagongo, 9000 ft. (north of Lake Kivu), 
Zaire (Tattersfield, 1995). 

Nesopupa bisulcata (Jickeli) — Adam (1954) 
and Bruggen & Verdcourt (1993). 

Pupisoma (Salpingoma) harpula (Reinhardt) 
(= Pupisoma japonicum Pilsbry) — Iden- 
tification by Dr. A. C. van Bruggen (Lei- 
den). Also Adam (1957). This widespread 
species is not listed in Verdcourt (1983). 

Pupisoma orcula (Benson) — Adam (1 957) 
and material in NMW from Bekar, India. 

Pupisoma sp. A — Shell almost globular (c. 
2.2 X 2 mm), brownish olive-green, with 
fine radial sculpture. 

Pupisoma sp. В — Probably a Pupisoma but 
matches none of the species in Adam 
(1954, 1957). However, the species (ap- 
prox. 2 X 1 .4 mm) has a moderately large 
umbilicus and traces of lamellae on the 
periphery, which make it rather endo- 
dontoid in appearance. 

Acanthinula sp. — The Kakamega material 
has only a few small spines. It lacks the 
spiral striation on the first whorl of both 
Preston's expathata Preston (holotype, 
BMNH 1937.12.30.2085) and Pllsbry's 
azorica; it may be underscribed. 

Rhiachidina chiradzuluensis var. virgínea 
(Preston) — Matches Preston's cotypes 
(BMNH) from Mount Kenya. 

Conulinus rutshuruensis major Verdcourt — 
Matches paratype (BMNH 196731, 
Nandi Forest, Kenya). 

Cerastua trapezoidea lagahensis (E. A. Smith) 

Micractaeon koptawelilensis (Germain) (= 
kakamegaensis Verdcourt) — Verdcourt 
(1990). 

Nothapalus sp. — Shell yellow, 19.5 x 6.4 
mm. It seems unwise to name this spe- 
cies on the basis of the single specimen. 
The shell of Preston's iredalei is narrower 
and his suturalis is larger. However, it is 
not disimilar in shape to either of these 
species. It has not been possible to 
compare the shell with material of N. ba- 
baulti (Germain) or N. paucispirus (von 
Martens). 

Subulona Clara Pilsbry — The Kakamega 



MOLLUSCAN DIVERSITY IN A KENYAN RAIN FOREST 



179 



species matches material of clara in 
BMNH. 

Oreohomorus ¡redalei Preston — The size 
and shape of the Kakamega specimens 
match material from the Belgian Congo 
and Mount Elgon in NMW, although 
shells of the former have more white co- 
louration. This species is conspecific 
with O. nitidus (von Martens) (Verdcourt, 
1983). Examination of a syntype of O. 
alblnl Germain (MNHN), which was de- 
scribed from Kakamega Forest, indi- 
cates that it is close or identical to the 
material collected in 1993, although it 
has lost its periostracum. The strong 
crenulations and spiral sculpture on the 
early whorls described by Germain 
(1923) and shown on the illustration are 
not visible even though the syntype ex- 
amined is clearly the illustrated shell. 

Pseudoglessula (Ischnoglessula) elegans 
(von Martens) — Kakamega material 
matches syntypes of elegans (BMNH), 
although all the types are in very poor 
condition with bleached or missing peri- 
ostracum. The material from Kakamega 
also appears identical to the illustration 
of P. subfuscidula Pilsbry, 1919, which is 
probably conspecific (Verdcourt, 1983). 
Verdcourt (1983) lists P. mutandana 
Connolly from Kakamega Forest but this 
species (syntype in BMNH) is larger and 
has much finer ribbing on the first four 
whorls than elegans; it was not found 
during the survey. 

Pseudopeas cf. yalaensls Germain — The 
Kakamega specimens match syntype 
material o^ yalaensls (MNHN); the types 
have very faint and barely perceptible 
spiral sculpture on the first whorl. Opeas 
euschemon Connolly may be conspe- 
cific (Verdcourt, 1983); however, the six 
shells of this species (NMW, Melvill and 
Tomlin coll., Mt. Mikeno) are larger than 
the types of yalaensls, and there is no 
trace of spiral sculpture. Their general 
shape is however similar and further ma- 
terial would be needed to confirm 
whether they are the same species. 

Cun/ella sp. A — shell broadly conic, thin, 
transluscent white, approx. 7 x 3.5 mm. 
Apical whorls smooth, remainder with ir- 
regular, arcuate growth lines. Outer lip 
curved in profile, arching forward in the 
centre. Columella curving smoothly into 
the basal margin of the shell mouth, not 
truncate. 



Cun/ella cf. babaultl Germain — The Kaka- 
mega specimens appear identical to the 
syntype (MNHN). The Kakamega mate- 
rial also matches Pseudopeas ke- 
kumeganum Connolly (syntype in 
BMNH), which Verdcourt (1983) sug- 
gests may be conspecific. The Kaka- 
mega shells have faint spiral micro- 
sculpture indicating that the species 
belongs in Pseudopeas. 

Achatlna stuhlmanni von Martens — Pain 
(1957). 

Limlcolaha cf. saturate Smith — The Kaka- 
mega material has a similar shape to the 
saturate holotype although the shells are 
smaller. Material oí saturate in NMW also 
generally has larger shells than the 
Kakamega specimens and further inves- 
tigations are desirable to confirm identi- 
fication. 

Punctum ugandanum (E. A. Smith) — Verd- 
court (1988). 

Punctum sp. A — The shells (approx. 1 .2 x 
0.8 mm) have rather regular ribbing and 
a characteristic spiral micro-sculpture 
suggesting that the species is in Punc- 
tum. Possibly close to hottentotum 
(Melvill & Ponsonby) but spire more ele- 
vated. 

Punctum sp. В — Shell (c. 1 .6 x 0.9 mm) with 
lamellae and possibly in Trachycystls. 
Smaller, less elevated spire and without 
the very broad lamellae of E. A. Smith's 
lamelllfera. 

Trachycystls Iredalel Preston — Verdcourt 
(1991b, c). 

Trachycystls arlel (Preston) — Agrees with 
paratype in BMNH. Also illustration in 
Bruggen (1969). 

Prosltala butumblana (von Martens) — Verd- 
court (1991b, c). 

Kallella barrakporensis (Pfeiffer) 

Kallella Iredalel Preston 

Carínate species (undescribed) — This dis- 
tinctive but undescribed, minute species 
is distributed widely across Africa 
(Malawi, Zaïre, Angola, Ghana and vari- 
ous other west African countries) (pers. 
comm., A. С van Bruggen). The shell is 
discoid and has six, spiral lamellae. The 
Kakamega material appears to be poly- 
morphic for shell colour with both white/ 
transparent and red-brown shells. 

Guppya quadrlsculpta (Connolly) 

Afroconulus Iredalel (Preston) — Micro- 
scopic and larger-scale shell sculpture 
and shape agree very well with cotype of 



180 



TATTERSFIELD 



iredalei (BMNH, Mt. Kenangop, Aber- 
dares, Kenya). However, the Kakamega 
material also does not differ significantly 
from urguessensis (Connolly), which may 
therefore be conspecific. The type of A. 
diaphanus (Connolly) could not be found 
in BMNH. 

Trochozonites (Zonitotrochus) cf. medjensis 
Pilsbry — The angle of the shell apex is 
smaller and the ribbing stronger in the 
Kakamega material than in the holotype 
(AMNH, Medje, Belgian Congo). How- 
ever, overall shape and size agree well. 
The holotype shell of expathata Preston 
(BMNH, Mt. Mikeno, Belgian Congo) has 
a much flatter base than both medjensis, 
and the Kakamega species and is clearly 
different. 

Thapsia eucosmia Pilsbry — Agrees well with 
the holotype (AMNH, Medge, Belgian 
Congo). This Thapsia has a "nipple-like" 
apex to its shell and shouldered whorls, 
which appear effectively to separate it 
from the other species collected during 
the study. 

Thapsia microleuca Verdcourt — Verdcourt 
(1982b). 

Thapsia spp. — Preston figured many spe- 
cies in this genus, which appear to be 
barely separable even with type matehal 
side by side for comparison. In the ab- 
sence of a full revision of the genus, it 
seems unwise to assign names to the 
Kakamega material. In addition to eu- 
cosmia and microleuca, the following 
species have been recorded from Kaka- 
mega forest previously: 

Thapsia elgonensis (Preston) 
Thapsia cinnamomeozonata Pilsbry 
Thapsia densesculpta (Preston) 
Thapsia karamwegasensis Germain 
Thapsia yalaensis Germain 
Thapsia gerstenbrandti (Preston) ?= elgonen- 
sis (Prest.) 
Thapsia mime (Preston) 



There appears to be at least two unidentified 
species in the Kakamega material. One 
has strong spiral striae and tight whorls 
without any evidence of a shell band. 
The other has faint spiral microsculpture 
and a brown shell with a faint band; this 
species is close to mime (Preston). 

Gymnarion aloysiisabaudiae (Pollonera) 

Chlamydahon oscitans (Preston) 

Urocyclid slug 

Halolimnohelix percivali (Preston) — Con- 
firmed by dissection by B. Verdcourt. 
Both holotype and paratypes (BMNH) 
are juvenile and lack a reflected peris- 
tome, but the shells otherwise agree with 
the Kakamega material. 

Halolimnohelix plana Connolly — Agrees 
with holotype in BMNH; also Verdcourt 
(1981). 

Gonaxis elgonensis (Preston) — material 
agrees well with paratypes in BMNH. 

Gulella woodhousei (Preston), = babauiti 
Germain, ?= perturbata Preston. 

Gulella osbomi Pilsbry — Illustrated by Pils- 
bry (1919). 

Gulella impedita Connolly — agrees perfectly 
with holotype in BMNH. 

Gulella lessensis Pilsbry — Illustrated by Pils- 
bry (1919). 

Gulella handeiensis Verdcourt 

Gulella disseminata (Preston) — The Kaka- 
mega material agrees well with the holo- 
type of var. kekumegaensis Connolly 
(BMNH). 

Gulella ugandensis (E. A. Smith) — See Verd- 
court (1970). 

Streptostele bacillum Pilsbry — The Kaka- 
mega matehal has spiral microsculture 
on the apical whorls. It matches perfectly 
with the holotype (AMNH, Bequaert 
Coll.) collected from Ituh Forest, Penge, 
Zaïre (Pilsbry, 1919). 



MALACOLOGIA, 1996, 38(1-2): 181-202 

MOLECULAR GENETIC IDENTIFICATION TOOLS FOR THE UNIONIDS OF 
FRENCH CREEK, PENNSYLVANIA 

Laura R. White,^ Bruce A. McPheron,^ & Jay R. Stauffer, Jr.^ 

ABSTRACT 

A molecular genetic key to the unionids of French Creek, Pennsylvania, an Allegheny River 
tributary, is presented here. The key is an integral part of a new approach to identifying unionid 
glochidia larvae attached to host fishes in the drainage. Working with tissue from adult union- 
ids, we used the polymerase chain reaction (PCR) followed by restriction enzyme digests to find 
species-specific genetic "fingerprints" for the 25 species in the drainage. We have demon- 
strated the utility of the key by using it to identify 70 glochidia attached to fishes collected in the 
French Creek drainage. 

Key words: Unionoidea, glochidial identification, PCR, RFLP analysis, ITS regions. 



INTRODUCTION 

North America's freshwater mussels (Bi- 
valvia: Unionoidea) are declining precipi- 
tously in richness and abundance (e.g., Den- 
nis, 1987; Anderson et al., 1991; Nalepa et 
al., 1 991 ; Williams et al., 1 992, 1 993). Sizable 
gaps in knowledge of unionid reproductive 
requirements hamper current preservation 
efforts. Information on the identities of the 
host fishes upon which unionid glochidia lar- 
vae are obligate parasites is especially in- 
adequate. Traditional methods of gathering 
such data have a variety of drawbacks. 

To date, lists of unionid host fishes have 
been derived primarily in two ways. The first, 
which has its roots in artificial propagation 
efforts (e.g., Lefevre & Curtis, 1910, 1912; 
Coker et al., 1921), involves inoculating pu- 
tative hosts with glochidia taken from gravid 
females of the unionid species of interest. 
Fishes in aquaria that ultimately contain meta- 
morphosed juveniles are considered suitable 
hosts (e.g., Zaie & Neves, 1 982; Waller & Hol- 
land-Bartels, 1988). Unsuitable hosts launch 
immune responses that thwart glochidial en- 
cystment, preventing further development 
and causing glochidia to be shed (Arey, 
1923a, 1932). 

As the completion of metamorphosis re- 
quires a week to several months of attach- 
ment (ZaIe & Neves, 1982), this approach is 
often time-consuming. It is also ill-suited to 
systems with large numbers of potential host 



fishes. Moreover, drawing inferences from in- 
oculation studies can be complicated by the 
fact that "suitable" host fishes can appar- 
ently acquire immunity to glochidia with re- 
peated exposure, the duration and species 
specificity of which are poorly established 
(Reuling, 1919; Arey, 1923b; Fuller, 1974). To 
obtain unambiguous results, it is often nec- 
essary to collect putative hosts from unionid- 
free streams or to inoculate naïve fishes bred 
and raised in the laboratory. Finally, while 
artificial inoculation methods are appropriate 
if laboratory propagation of unionids is the 
only goal, the results of such studies might 
be inapplicable to organisms in their natural 
environments. Such studies disregard micro- 
habitat preferences and specialized mor- 
phologies and behaviors (e.g., the waving of 
fish-like mantle flaps by gravid female Lamp- 
silis species; Ortmann, 1911; Kraemer, 1970) 
that might modulate unionid-fish interactions 
in situ. 

To circumvent these problems, several in- 
vestigators (e.g.. Wiles, 1975; Stern & Felder, 
1978) have attempted morphology-based 
identification of glochidia attached to fishes. 
Such determinations have thus far entailed 
identifying the glochidia using dissecting mi- 
croscopes or compound light microscopes. 

There are drawbacks to this approach as 
well. Glochidia are less than 1 mm in diameter. 
Encystment makes them difficult to observe 
and might influence their shapes in unpredict- 
able ways (Wiles, 1975). Closely related spe- 



School of Forest Resources and Intercollege Graduate Degree Program in Ecology. 2C Ferguson BIdg., The Pennsylvania 
State University, University Park, Pennsylvania 16802, U.S.A. 

^Department of Entomology and Institute of Molecular Evolutionary Genetics. 535 ASI BIdg., The Pennsylvania State 
University, University Park, Pennsylvania 16802, U.S.A. 



181 



182 



WHITE, McPHERON & STAUFFER 



cies, such as Villosa nebulosa (Conrad) and 
Villosa vanuxemensis vanuxemensis [vanux- 
emi] (I. Lea) (Zaie & Neves, 1982), are difficult 
to distinguish from each other and are easily 
misidentified. Hoggarth (1992) reported that 
glochidia photographed by Wiles (1975) and 
identified by the author as Pyganodon [An- 
odonta] Cataracta (Say) were actually Alasmi- 
donta undulata (Say). Clarke (1981, 1985), 
Rand & Wiles (1982), and Hoggarth (1988) 
dennonstrated that scanning electron micros- 
copy can be used to distinguish among 
glochidia taken from gravid females. Whether 
their techniques can be adapted for species- 
level identification of glochidia from host 
fishes remains to be investigated, however. 
The objective of the research described 
herein was to develop a new method for 
identifying glochidia attached to fishes, a 
method that exploits genetic differences 
among unionid species. The method utilizes 
restriction fragment length polymorphism 
(RFLP) analysis of polymerase chain reaction 
(PCR) products. In combination, PCR and 
RFLP analysis are useful for performing sen- 
sitive analyses of minute quantities of DNA 
(e.g., Whitmore et al., 1992; Simon et al., 
1993), such as those present in single 
glochidia. In short, a diagnostic suite of re- 
striction sites (or "genetic fingerprint") is 
sought for each unionid species in the drain- 
age of interest. Glochidia on host fishes are 
then identified on the basis of the "finger- 
pnnts" they possess. 



MATERIALS AND METHODS 

Study Site 

The aquatic system for which the glochidial 
identification method was developed is the 
French Creek drainage, in southwestern New 
York and northwestern Pennsylvania (Fig. 1). 
French Creek is a fourth-order tributary to the 
upper Allegheny River. It drains approxi- 
mately 3,000 km^. Twenty-five unionid spe- 
cies (C. Bier, pers. comm.) and 53 fish spe- 
cies (J. Stauffer, unpubl. data) have been 
collected from the French Creek drainage re- 
cently, making its fish and molluscan faunas 
the richest in Pennsylvania. Two of the drain- 
age's unionid species, Epioblasma torulosa 
rangiana (I. Lea) and Pleurobema clava (La- 
marck), are federally endangered and have 
no known hosts. Two additional species are 
considered globally threatened and seven 



are of special concern (Williams et al., 1993); 
of these nine, five have no known hosts. 

LeBoeuf Creek is thought to harbor higher 
densities of P. clava than any other part of the 
drainage (A. Bogan, pers. comm.). To assess 
the utility of the identification technique, 
fishes were collected from LeBoeuf Creek at 
Moore Road bridge, just east of Route 19, 3 
km south of LeBoeuf Gardens, Pennsylvania 
(Fig. 1). Full descriptions of the site and col- 
lection procedures are given by White (1994). 



Specimen Collection and Preservation 

Adult unionids. Adult unionids were col- 
lected throughout the French Creek drainage 
(Fig. 1, Table 1) in 1991, 1992, and 1993. 
Numbers of unionids collected ranged from 
one to 23 per species, with a median of six. 
Adult Lasmigona costata (Rafinesque), Am- 
blema plicata (Say), and Lampsilis siliquoidea 
(Barnes) specimens were also collected from 
West Virginia (Dunkard Creek) and Ohio 
(lower Muskingum River, Little Muskingum 
River, and Big Darby Creek), so that their ge- 
netic "fingerprints" could be compared with 
those of French Creek specimens to evaluate 
the key's applicability to other drainages. 

Adult unionids were collected using masks 
and snorkels or Plexiglas-bottomed buck- 
ets. Nonendangered species were trans- 
ported to the laboratory either alive (wrapped 
in cheesecloth in chlorine-free ice water) or 
frozen on dry ice. In the laboratory, live union- 
ids were either killed and frozen at 80''C, or 
maintained in aquaria in which currents were 
established. Two small (5- to 100-mg) pieces 
of foot tissue were excised from each indi- 
vidual in the laboratory using a sterile scalpel 
blade or scissors. Both samples were frozen 
at 80 C, one for nucleic acid extraction and 
the other for voucher material. The remaining 
tissue was preserved with the valves in 70% 
ethanol, also as voucher material. To facilitate 
future molecular genetic examination, the lat- 
ter tissue was not fixed in formalin. All voucher 
material was deposited into the mollusc col- 
lection of the Academy of Natural Sciences in 
Philadelphia upon completion of the research 
(Dry Catalog # 398499-398500; Alcohol Cat- 
alog # A18354-A18438; Frozen Catalog # 
F100-F118). 

For endangered unionids, a single tissue 
sample was obtained from each specimen at 
streamside by relaxing its adductor muscles 
in soda water and clipping off a 5- to 50-mg 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



183 




Conneaut 
Lake 



10 km 



FIG. 1. Location of French Creek drainage collection sites. (For site descriptions, see Table 1 and White 
1994). 



184 



WHITE, McPHERON & STAUFFER 



о 



LT) CM 



о 



О 



ü 



CNJ 1- ■* •^ 



CO 1- M- 



1- CO CO CO 



T- -.-■'- CO 

+ + 



CD СЧ О -^ 



та о 



о 

i5 о 



cu "О 



0-2 Э о 



ш 



с ;2 с 

о р Щ v¿ 

о -2 Е с 

'í 'í «=1: "=!: 



i^ та та й 

TD 01 со та 

â -Q -Q Я 
.О- 9 9 ^ 
5 Q. О. 3 
ÜJ Uj Uj U. 



У та 5 

со (О .Í2 .<2 

¡À V) <Л <У) 

О. Q. Q. Q. 

£ Е Е Е 

та та та та 

-J -J -J -J 



S та 

та Í0 

с со 

-Э Ï 2 
Q. О. та 
Е Е ^ та 
о о о t; 
о о и g 
та та та та 
с с: с с 
о о о m 
D5 CD 05■~ 
È È È i 

со со с/) fi 

та та та У 

-J ~j — J -J 



о 
.2 ^ 



§ 2 

та с 
и й 



>< со 



00 

та 5 
о та 

"О "~ 



та о '^ Í0 5 
п; "О -S 5 та 

о о ?з Р 



о та та 
Ï Е Е 

(И Ф 

та хз о 

È S 2 
3 3 3 



í^ с ^ -5 та та 
о g, та р о о 



II 3 „ - 
О 2 5> « 



(О ¿ 



Œ. 



га W g 

F ^ о 
ГО О -о 

Е 



> с 

ir >- 5 

га ¡2 о 
О 



СП ■ 
О .!5 



О. ф О 



Ç II 



со 



3 ^ ^^э 

яге II .- 

с с О о 
^ ^^ '¿. 
О ÍJ га ^ 



и "9 ^S 



О с>^ 



^ ^ (Л И 

У ■ - ч^ ^ 



*- > Щ ш 

га "О о. у 

• га ^ *; 

х: :^ с о 

с ° - ¿ 

О) ^ Ü tn ^ 

li го -а §-'^ 

• - О ° ^ -'^ 

^ ^ :z о о 

го о о. _;- "5; 
О ^ •■сой) 
^ CJ ¿^ с 3 

О Ï °" 



ф 



O^J-Ö 



г= ^^ о 



СО 0) 

'=> Е 



о ОС 



< О 

О - 



ш о 



го ОС CJ) -fc ^ 

р 5- О) со га 

.£ t = о. СП 
О) 3 < => 

> с II t ■ - 

^ :2 о ^ Л 

§ ^ ш ^ о 
о 

— с i: 

— О) CD 



-I > 



m 



о 



-' го II «¡ 
L: ^ с\1 s 
О О =) Il 
rö CD 

-^ Il ^^ с 

Il >¿ 2 

Ш 2 Й Ч) 

-!.§•> 

l: о' -s; 

ш £ Е ™ 

Е О ^ - 

о . ей О 

О ос с^ х: 

>- ^ ^ " 
О) Е Л с 

■^ = и m 



О tÓ 3 

^ о О)^ U- 

¿pLL С О 
^ С II 



го M $ 



га - - 

Q ^ 

cnCJ ^ II - 

rt^ »- га 

CD ^ О) ц о 

II ^ ^^Ц 

О Ï ^ J =' 

CD и. II О О 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



185 



piece of foot using a sterile scalpel blade or 
scissors (Pennsylvania Fish and Boat Com- 
nnission permit number 142 (Type I); proce- 
dure reviewed prior to permitting by the 
United States Fish and Wildlife Service). Tis- 
sue samples were frozen immediately on dry 
Ice for transportation to the laboratory, where 
they were kept at -80" С pending nucleic 
acid extraction. After a 10- to 15-min recov- 
ery period in a bucket of streamwater, the 
specimens were photographed and returned 
to natural positions in the substrate as close 
to their original locations as possible. 

Fishes. Fishes were collected throughout 
French Creek by kick-seining and were trans- 
ported to the laboratory on dry ice. In the 
laboratory, a 5- to 100-mg piece of muscle 
was excised from the body wall of each and 
was frozen at -80 С prior to nucleic acid 
extraction. The remainder of each specimen 
was also frozen at -80" С as voucher mate- 
rial. 

Glochidia. Glochidia of known identity were 
obtained from marsupia of gravid nonendan- 
gered female unionids collected and frozen 
as described above. Glochidia of unknown 
identity were obtained from fishes collected 
throughout French Creek by kick-seining. 
The fishes were transported to the laboratory 
alive, maintained in an aquarium for one 
week, then killed and frozen at -80'C; un- 
encysted glochidia were presumed to have 
been shed during the holding period. En- 
cysted glochidia were removed as described 
below. 

Laboratory Techniques 

Nucleic acid extraction. For adult unionids, 
unattached glochidia, and fishes, a standard 
phenol-chloroform extraction protocol (after 
Kocher et al., 1989) was used to isolate total 
nucleic acids. Each tissue sample was 
minced over ice using a sterile scalpel blade, 
then transferred to a 1.5-ml microfuge tube 
and homogenized in 500-800 ц! of extraction 
buffer (100 mM Ths-HCI, pH 8.0: 10 mM 
EDTA; 125 mM NaCI; 0.1% SDS; 50 mM 
DTT; 5 ng/ial proteinase K) using a flame- 
sealed 1000-|.il pipette tip; different scalpel 
blades and pipette tips were used for each 
sample, to prevent cross-contamination. Ho- 
mogenized samples were incubated 2-24 hrs 
at 37-C, then extracted sequentially with 
equal volumes of Tris-buffered phenol, 50% 



phenol-50% chloroform, and chloroform (= 
24 chloroform: 1 isoamyl alcohol, v:v; Sam- 
brook et al., 1989). Samples were centrifuged 
4-5 min at 16,000 x g during each extraction 
to separate the phases. After the final extrac- 
tion, 0.05 volume of 5 M ammonium acetate 
and two volumes of cold absolute ethanol 
were added to each sample. Samples were 
placed at -80 С for 15-30 min, then spun 
1 5-45 min at 1 6,000 x g at 4' C. Supernatants 
were decanted and pellets were dried in a 
Savant SpeedVac Concentrator. Pellets were 
resuspended in 10-25 |il of sterile distilled 
water, depending on their size, and stored at 

20°C. Even when no pellet was visible in a 
tube, 10 |il of sterile distilled water was 
added and the sample was stored at -20°C. 
Extractions were assayed on 0.8%-agarose 
minigels stained with ethidium bromide and 
were diluted 0-1000x depending upon esti- 
mated DNA concentration. 

For glochidia attached to fishes, an extrac- 
tion protocol similar to that deschbed by 
Martin et al. (1992) for fish oocytes was used. 
Each glochidium was removed from its host 
over ice using sterile forceps and a dissecting 
light microscope, then transferred with a 
200-|jl pipette tip to a 1 .5-ml microfuge tube 
containing 30 |jl of buffer (50 mM KCl: 10 mM 
Tris-HCI, pH 8.3; 1 |ig/ml proteinase K; 1 
pg/ml bovine serum albumin). Nonidet P-40 
was added to a final concentration of 1%. 
Solutions were heated to 95 С for 5 min in a 
thermal cycler, diluted to a final volume of 50 
|il with sterile distilled water, and stored at 4° 
or -20°C. Extractions were not assayed 
prior to amplification, as they contained too 
little DNA to be visualized with ethidium-bro- 
mide staining (data not shown). 

Amplification. Reaction volumes of 50 or 1 00 
|.il were used. Reaction mixtures consisted of 
0.5-2.0 |il of diluted template DNA; 1 [.iM of 
each primer (0.2 (.iM of each RAPD primer); 
0.1 mM each of dATP, dCTP, dGTP, and 
dTTP; 2.0-2.5 units of Perkin-Elmer Cetus 
Taq polymerase: and manufacturer-supplied 
buffer at 1x final concentration (10 mM Tris- 
HCI, pH 8.3; 50 mM KCl; 15 mM MgCIsi 
0.01 % (w:v) gelatin). For glochidia from host 
fishes, 1-10 |.il of undiluted template was 
used. 

Primer sequences were as follows. ITS-1 
of nuclear rDNA: 5'-TAACAAGGTTTCCG- 
TAGGTG-3' (18S region) and 5'-AGCTRGC- 
TGCGTTCTTCATCGA-3' (5.8S region); ITS-1 
through ITS-2: 5'-TCCGTAGGTGAACCTGC- 



186 



WHITE, McPHERON & STAUFFER 



GG-3' (ITS1 of Lee & Taylor, 1 992; 1 8S region) 
and 5'-TCCTCCGCTTATTGATATGC-3' (ITS4 
of Lee & Taylor, 1992; 28S region); 12S mi- 
tochondrial rDNA: 5'-TAATAATAAGAGGGA- 
CGGGCGATGTGT-3' (adapted from HI 478 
of Kocher et al., 1 989 using sequence data for 
Drosophila yakuba Burla (Clary & Wolsten- 
holme, 1985)) and 5'-TAATAAAAAACTAGG- 
ATTAGATACCCTATTAT-3' (adapted from 
LI 091 of Kocher et al., 1989); RAPD primer 
A-02: (5'-TGCCGAGCTG-3'; Operon Tech- 
nologies, Inc., Alameda, CA). Rationales for 
primer choices are discussed in White (1994) 
and White et al. (1994). 

Thirty-four amplification cycles were per- 
formed (1 min at 93°C, 1 min at 50''C, and 2 
min at 72'-C) followed by one cycle with in- 
creased extension time (9 min). For RAPD 
PCR, 45 amplification cycles of 1-min dena- 
turation at 94' C, 1-min reannealing at 36' C, 
and 2-min extension at 72"C were per- 
formed. Reaction products were assayed 
on 0.8-2.0% agarose minigels stained with 
ethidium bromide. 

Restriction Enzyme Digestion. Restriction en- 
zyme digests were performed in 10- to 20-|.il 
reaction volumes consisting of 8-1 2 |.il of PCR 
product, 5-1 5 units of restriction enzyme, and 
the manufacturer-supplied buffer at a final 
concentration of 1x. Digests were conducted 
at the manufacturer-recommended tempera- 
ture (usually 37°C) for 4-48 hrs. Restriction 
fragments and uncut PCR products were as- 
sayed on 2.0%-agarose gels stained with 
ethidium bromide. Efforts to separate poorly- 
resolved fragments with 6-10% Polyacryl- 
amide or 2-4% MetaPhor high-resolution 
agarose met with limited success and were 
ultimately abandoned. 



RESULTS 

Key to the Unionids of French Creek 

The following key was developed for iden- 
tification of French Creek unionid glochidia. 
One proceeds through the key by amplifying 
the genomic region indicated in bold text, 
digesting the PCR product with the restriction 
enzyme listed after the x, and assigning a let- 
ter to the resulting restriction fragment pattern 
(by referring to the accompanying figure 
and/or to the fragment size data in Appendix 
1 ). Assaying undigested PCR products along- 



side digested products facilitates pattern in- 
terpretation and is highly recommended. Su- 
perscripts refer to notes that follow the key. 
While the key likely reflects phylogeny to 
some extent, the data from which it was con- 
structed are insufficient for testing specific 
hypotheses about relationships; thus, the key 
should be considered artificial. 

1. ITS-1 X Mspl (Fig. 2) 

A 2^ 

В Ligumia nasuta (Say) 

С 9'" 

D Amblema plicata (Say)'^ 

E Quadrula cylindhca (Say) 

F 12 

G Strophitus undulatus (Say)*^ 

H Alasmidonta marginata Say 

2. ITS-1 X Sau96l (Fig. 3) 

A 3 

В 8 

3. 12S X Haelll (Fig. 4) 

A 4 

В . . . . Actinonaias ligamentina (Lamarck) 
С Lampsilis siliquoidea (Barnes)^ 

4. ITS 1-2 X Mspl' (Fig. 5) 

A 5 

A' 7 

В Lampsilis fasciola Rafinesque 

5. 12S X Rsal (Fig. 6) 

A Villosa iris (I. Lea) 

Epioblasma spp 6^ 

6. ITS 1-2 X Mboll (Fig. 7) 

A Epioblasma torulosa rangiana 

(I. Lea)^ 
A' . . . . Epioblasma triquetra (Rafinesque) 

7. ITS-1 X Aval (Fig. 8) 

A Lampsilis cardium Rafinesque, 

Lampsilis ovata (Say)' 
Ligumia recta (Lamarck) 

8. ITS-1 X AccI (Fig. 9) 

A Ptychobranchus fasciolahs 

(Rafinesque) 
Villosa fabalis (I. Lea) 

9. ITS-1 X BstEII (Fig. 10) 

A Elliptio dilatata (Rafinesque) 

10 

10. ITS 1-2 X Mspl* (Fig. 11) 

A 11 

A' Fusconaia subrotunda (I. Lea) 

11. RAPD A-02 (Fig. 12) 

A Pleurobema clava (Lamarck)'" 

В . . . Pleurobema sintoxia [= coccineum] 
(Rafinesque) 

12. ITS-1 X BamHI (Fig. 13) 

A Lasmigona costata (Rafinesque) 

13 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 187 

13. ITS-1 X Hinfl (Fig. 14) O 15 

A . . . Anodontoides ferussacianus (I. Lea) 15. US 1-2 x Mspl* (Fig. 16) 

В 14 А Pyganodon [= Anodonta] grandis 

14. ITS-1 X Mboll (Fig. 15) (Say) 

A Lasmigona compressa (I. Lea) A' Lasmigona complanata (Barnes) 




a. uncut "^ 



10 



15 



20m m25 



b. Mspl-cui 




-1500 



600 



-100 



DE 
Anf 



G H Pattern 
Tribe 



FIG. 2. ITS-1 PCR products from 25 French Creek unionid species digested with Mspl. Restriction fragment 
patterns (A-H) separate species into their respective tribes. Tribe Am = Amblemini, An = (subfamily) An- 
odontinae, L = Lampsilini, P = Pleurobemini. Tribe Lampsilini (patterns A, B): 1 = Epioblasma torulosa 
rangiana, 2 = Epioblasma thquetra, 3 = Lampsilis cardium, 4 = Lampsilis fasciola, 5 = Lampsilis ovata, 6 = 
Lampsilis siliquoidea, 7 = Villosa fabalis, 8 = Villosa iris, 9 = Actinonaias ligamentina, 1 = Ptycliobranchus 
fasciolaris, 1 1 = Ligumia recta, 12 = Ligumia nasuta: tribe Pleurobemini (pattern C): 13 = Elliptio dilatata, 14 
= Pleurobema clava, 15 = Pleurobema sintoxia, 16 = Fusconaia subrotunda; tribe Amblemini (patterns D, E): 
1 7 = Amblema plicata, 1 8 = Quadrula cylindrica; subfamily Anodontinae (patterns F, G, H): 1 9 = Anodontoides 
ferussacianus, 20 = Pyganodon grandis, 21 = Lasmigona compressa, 22 = Lasmigona costata, 23 = Las- 
migona complanata, 24 = Strophitus undulatus, 25 = Alasmidonta marginata. Tribe designations follow 
Vaught (1989). Gels shown throughout key are 2.0% agarose. Size marker used throughout key is 100-bp 
ladder. 



WHITE, McPHERON & STAUFFER 




-1500 



-600 



100 



a. uncut 



12 3456789 10 11 




1500 



600 



100 



b. Sau96l-cut 



В 



FIG. 3. ITS-1 PCR products from "1-A" species digested with Sau96l. 1 = Epioblasma torulosa rangiana, 
2 = Epioblasma triquetra, 3 = Actlnonaias ligamentina, 4 = Lampsilis cardium, 5 = Lampsilis fasciola, 6 = 
Lampsilis ovata, 7 - Lampsilis siliquoidea, 8 = Ligumia recta, 9 = Villosa iris, 10 = Villosa fabalis, 11 = 
Ptychobranchus fasciolaris. 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



189 




a. uncut 



1234567 89 




-1500 



-600 



-100 



b. НаеШ -cut 



В С 



FIG. 4. 12S PCR products from "2-A" species digested with Haelll. 1 = Lampsilis cardium, 2 = Lampsilis 
fasciola, 3 = Lampsilis ovata, 4 = Ligumia recta, 5 = Epioblasma torulosa rangiana, 6 = Epioblasma triquetra, 
7 = Villosa iris, 8 = Actinonaias ligamentina, 9 = Lampsilis siliquoidea. 



Notes to Accompany the Key 

^Includes Ptychobranchus fasciolaris, in 
contradiction to White et al., 1994; the spec- 
imen identified in White et al. (1994) as P. 
fasciolaris is almost certainly EIHptio dilatata. 

'^The Pleurobema sintoxia specimen from 
Foster Corner exhibited a unique pattern (Fig. 
17). 

""One of the 1 8 Amblema plicata specimens 



from the lower Muskingum River, Ohio, ex- 
hibited a unique pattern quite similar to that 
of Ligumia nasuta (Fig. 18). 

^In contradiction to White et al., 1994; the 
specimen identified by White et al. (1994) as 
Strophitus undulatus was subsequently re- 
identified as Pyganodon grandis by A. E. 
Bogan. 

''Two Lampsilis siliquoidea specimens from 
the French Creek drainage (one of the three 



190 



WHITE, McPHERON S STAUFFER 



uncut 



Msp i cut 



1234567 1234567 




1500 



600 



-100 



La J L 



В 



FIG. 5. ITS 1-2 PCR products from "3-A" species digested with Mspl. 1 = Epioblasma torulosa rangiana, 
2 = Epioblasma triquetra, 3 = Villosa iris, 4 = Lampsilis cardium, 5 - Lampsilis ovata, 6 = Ligumia recta, 7 
= Lampsilis fasciola. 



from Venango and the one from Conneaut 
Outlet) exhibited patterns with three bands 
instead of two (Fig. 19). 

*A and A' are most reliably distinguished by 
digesting samples of known DNA and assay- 
ing them in lanes adjacent to the unknown 
DNA. Digesting several samples of each 
known and unknown DNA is recommended, 
as it allows one to intersperse samples of 
each type on a single gel for easier detection 
of subtle length differences. Assays should 
be run on at least a 2%-agarose gel, for as 
long as possible, to achieve maximal sepa- 
ration. 

^Couplet 6 reliably separates two of the 
four Epioblasma torulosa rangiana speci- 
mens examined (one of the two from 
Venango and the one from Utica) from the 
three Epioblasma triquetra specimens exam- 



ined. The broader utility of this couplet is un- 
certain; it should be used with caution. Also 
see note f. 

*^federally endangered species 

'Lampsilis cardium and Lampsilis ovata 
specimens could not be distinguished from 
each other using any of the primers and re- 
striction enzymes tried (White, 1994: appen- 
dix B2). It is conceivable that these species 
hybridize in French Creek; some specimens 
exhibited intermediate shell morphologies 
and could not be identified to species with 
certainty on the basis of external characters 
(A. E. Bogan, pers. comm.). 

Reliability of the Key 

The key was tested extensively using adult 
unionids identified morphologically. In its an- 
notated form, it proved valid for all French 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



191 



— UnCUtn Г- 

12 3 1 



Rsal 
cut- 

2 : 




1500 



600 



100 



FIG. 6. 12S PCR products from "4-A" species di- 
gested with Rsal. 1 = Epioblasma torulosa rangi- 
ana, 2 = Epioblasma triquetra, 3 = Villosa iris. 



Creek specimens examined. It was also valid 
for all Ohio and West Virginia A. plicata (Fig. 
20), L. siliquoidea, and L. costata specimens 
examined. Moreover, glochidia obtained 
from a gravid French Creek female L. costata 
followed the key, exhibiting restriction frag- 
ment patterns identical to those of adult L. 
costata specimens, as expected (data not 
shown). 

Identification of Unknown Glochidia with 
the Key 

Four unknown glochidia from the gills of a 
tippecanoe darter (Etheostoma tippecanoe 
Jordan & Evermann) collected 20 July 1993 in 
French Creek downstream of Utica, Pennsyl- 
vania, exhibited restriction fragment patterns 
identical to those of adult V. fabalis specimens 
(unpubl. data). In a larger-scale test of the 
technique's utility, all glochidia found on 
fishes collected 6 June 1994 at the LeBoeuf 
Creek site were analyzed. Of the 115 glochidia 



Mboll 
r-uncirt-i|-cirt-i 

1 2 3 12 3 




-1500 



600 



-100 



FIG. 7. ITS 1-2 PCR products from "5-0" species 
digested with Mboll. 1 = Epioblasma torulosa ran- 
giana, 2 = Epioblasma triquetra, 3 = E. triquetra. 




-1500 



-600 



-100 





FIG. 8. ITS-1 PCR products from "4-A"' species 
digested with Aval. 1 = Lampsilis cardium, 2 = 
Lampsilis ovata. 3 = Ligumia recta. 



192 



WHITE, McPHERON & STAUFFER 




-1500 



600 



-100 



A 



FIG. 9. ITS-1 PCR products from "2-B" species 
digested with Accl. 1 = Ptychobranchus fasciolahs, 
2 = Villosa fabalis. 




-1500 



600 



-100 



FIG. 10. ITS-1 PCR products from "1-C" species 
digested with BstEII. 1 = Elliptio dilatata, 2 = 
Fusconaia subrotunda, 3 = Pleurobema clava, 4 = 
Pleurobema sintoxia. 



DISCUSSION 



processed, 72 (63%) were amplified success- 
fully (i.e., their ITS-1 PCR products were vis- 
ible on an agarose gel stained with ethidium 
bromide). Of these, 66 (92%) were identifi- 
able; the other six yielded resthction frag- 
ments too faint to be seen. Fourteen of the 43 
glochidia not amplified successfully were in 
the first set of samples, extracted using a pro- 
tocol that differed slightly from that ultimately 
adopted. Disregarding this flawed first at- 
tempt, the amplification success rate was 72 
out of 102 (71%). 

All 66 glochidia identified exhibited the 
restriction fragment patterns characteristic 
of Ptychobranchus fasciolahs (Fig. 21), a 
species for which no hosts are currently 
known (Hoggarth, 1992). Four Etheostoma 
blennioides Rafinesque, three Etheostoma 
flabellare Rafinesque, five Etheostoma ni- 
grum Rafinesque, and one Etheostoma zon- 
ale (Cope) harbored the glochidia. These four 
darter species are therefore suggested ten- 
tatively to be P. fasciolahs hosts, pending 
verification through laboratory inoculation 
studies. 



Distinguishing Unionid DNA from Host Fish 
DNA 

Without exception, the ITS-1 regions of the 
fishes examined are markedly different in 
length from those of the unionids. For single 
individuals of five of the six darter species 
examined (E blennioides, E. flabellare, E. tip- 
pecanoe, Etheostoma vahatum Kirtland, and 
E. zonale), the product is approximately 690- 
710 bp; for the sixth darter, Etheostoma 
maculatum Kirtland, the product is approxi- 
mately 410 bp long (White, 1994: Fig. 2.3). 
Among most of the unionids, the ITS-1 prod- 
uct ranges from approximately 580 to 625 
bp; for Alasmidonta marginata Say and Stro- 
phitus undulatus (Say), it is approximately 
950-1,050 bp long (see uncut products in 
Fig. 2). Because the length ranges for fishes 
and unionids are non-overlapping, any host- 
fish DNA contaminating glochidial DNA is 
easily recognized as such. Furthermore, 
when ITS-1 PCR products of the six darter 
specimens are digested with Mspl, they yield 
restriction fragment patterns different from all 
unionid patterns. Hence, even if the glochid- 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



193 



Mspl 
cut 



uncut — I 1— 

1 2 3 4 12 3 4 




-1500 



600 



-100 



FIG. 11. ITS 1-2 PCR products from "9-0" species digested with Mspl. 
Pleurobema sintoxia, 3 = Fusconaia subrotunda, 4 = F. subrotunda. 



1 = Pleurobema clava, 2 = 



iai identification nnethod described herein 
were applied to fishes (e.g., salmonids) 
whose ITS-1 regions are close to the union- 
ids' in length (Pleyte et al., 1992), contannina- 
tion could be detected reliably by digesting 
the host fish's ITS-1 product and assaying it 
alongside the digested products of the 
glochidia it harbored. The contaminating 
DNA could be factored out of the RFLP anal- 
yses by disregarding restriction fragments 
present in both gel lanes. 

Current Limitations of the Technique 

Identifying glochidia on naturally infected 
fishes is a hit-or-miss approach to discover- 
ing hosts of a particular unionid species of 
interest. To maximize the chances of suc- 
ceeding, it is important to collect fishes from 
sites where the unionid species of interest is 
abundant relative to other species (or at least 
where it is abundant relative to other sites). 
As the preliminary LeBoeuf Creek study 



12 3 4 5 6 




-1500 



-600 



-100 



FIG. 12. RAPD A-02 PCR products from "10-A" 
species. 1-3 - Pleurobema clava, 4-6 = Pleu- 
robema sintoxia. 



194 



WHITE, McPHERON & STAUFFER 



- uncut 1 (— 

2 3 4 5 1 



BamHI cut — 
2 3 4 5 




-1500 



-600 



-100 



FIG. 13. ITS-1 PCR products from "1-F" species digested with BamHI. 1 = Anodontoides ferussacianus, 
2 - Pyganodon grandis, 3 = Lasmigona complanata, 4 = Lasmigona compressa, 5 = Lasmigona costata. 




— uncut — 
1 2 3 



Mboll cut-, 
1 2 3 



-1500 



600 



-100 




FIG. 14. ITS-1 PCR products from "12-0" species 
digested with Hinfl. 1 = Anodontoides ferussa- 
cianus, 2 = Pyganodon grandis, 3 = Lasmigona 
complanata, 4 = Lasmigona compressa. 



FIG. 15. ITS-1 PCR products from "13-B" species 
digested with Mboll. 1 = Pyganodon grandis, 2 - 
Lasmigona complanata, 3 - Lasmigona com- 
pressa. 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



195 



— uncut 

12 3 4 5 6 



1 г 



1 



Mspl cut 1 

2 3 4 5 6 




-1500 



-600 



-100 



^A— ' A^A 



' 1_Л_1 



FIG. 16. ITS 1-2 PCR products from "14-0" species digested with Mspl. 1-3, 5, 6 = Pyganodon grandis; 
4 = Lasmigona complanata. 

Mspl 
uncut n ^cut 



-,f-oui-^ 




-1500 



-600 



Mspl 
uncut— I r— cut 



-100 




-1500 



600 



12 12 



FIG. 17. ITS-1 PCR products from anomalous and 
standard Pleurobema sintoxia specimens digested 
with Mspl. 1 = anomalous pattern, 2 = standard 
pattern. 



-100 



12 3 12 3 

FIG. 18. ITS-1 PCR products from anomalous and 
standard Amblema plicata specimens digested 
with Mspl. 1 = Ligumia nasuta, 2 = anomalous /Am- 
blema plicata, 3 = standard A. plicata. 



196 



WHITE, McPHERON & STAUFFER 



Haelll 
j-uncut-j J— cut— I 




1 



FIG. 19. 1 2S PCR products from anomalous Lamp- 
silis siliquoidea specimens digested with Haelll. 
Specimens exhibit 3-banded pattern instead of 
standard 2-banded pattern. 



demonstrated, this will not guarantee suc- 
cess, however. Additionally, fishes should be 
collected repeatedly throughout the full du- 
ration of the unionld's breeding period. 

The glochidial amplification procedure cur- 
rently has a success rate below 100%. Most 
unsuccessful amplification attempts were 
likely the result of glochidia being lost during 
transfer from host to extraction buffer; once 
excised from the host, glochidia are ex- 
tremely difficult to see. Improvements in the 
transfer technique could increase the ampli- 
fication success rate dramatically. The iden- 
tification success rate, already quite high, 
could probably be increased by gel-purifying 
and reamplifying very faint PCR products 
prior to restriction enzyme digestion. 

Extending the Key Beyond French Creek 

To apply the method to an aquatic system 
other than the French Creek drainage, some 
preliminary work is required. First, tissue 
samples must be obtained from several indi- 
viduals of each unionid species found in the 
study system. Ideally, each species should 



be represented by specimens collected at a 
variety of sites. 

Next, the reliability of the key, for the 
study-system species included in it must be 
assessed. All specimens of each such spe- 
cies should be analyzed using the key, to see 
whether they yield the expected restriction 
fragment patterns for each enzyme (as did 
the West Virginia and Ohio specimens we ex- 
amined). If they do not, the key will have to be 
modified accordingly. 

The key will also have to be extended to 
include any study-system species not found 
in French Creek. This is most easily accom- 
plished as follows: first, analyze a single 
specimen of each new species, using the 
French Creek key. If a specimen yields a 
novel restriction fragment pattern for a cer- 
tain couplet, test all individuals of the species 
to see if they share the pattern; if they do, 
modify the key accordingly. If a specimen 
yields no novel patterns, proving indistin- 
guishable from a species already included in 
the key (or from another new species), screen 
single individuals of the indistinguishable 
species pair (or group) with a variety of prim- 
ers and restriction enzymes until a diagnostic 
difference is found. (Consulting Appendix В 
of White, 1994, might prove useful in this re- 
gard.) Alternately, sequence a moderately 
variable region of the genome of each spe- 
cies and scan the sequence data for restric- 
tion site differences. Finally, verify that the 
differences found apply to all individuals of 
the species, then modify the key accordingly. 
Publish modified versions of the key 
promptly to save other investigators precious 
time and resources. 

Overall Assessment of the Technique 

Using a molecular genetic key to identify 
glochidia attached to fishes has distinct ad- 
vantages over traditional means of identifying 
putative unionid hosts (White et al., 1994). 
The laboratory procedures are relatively fast 
and easy to perform. Once a key has been 
developed, glochidia can be identified in one 
or two days; the techniques involved can be 
learned (if not mastered) in a week. The 
method is also relatively inexpensive, partic- 
ularly if one has access to a laboratory al- 
ready equipped for molecular genetic re- 
search (see White, 1 994: appendix C, for cost 
analysis). 

The data generated to develop keys are 
potentially valuable to unionid systematists. 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



197 




a. uncut 



1 ' ' I 5 



I I I 



10 ' ' • • 15 




b. Aispl-cut 



FIG. 20. ITS-1 PCR products from 15 Amblema plicata specimens from three drainages, digested with 
Mspl. 1-5 = French Creek specimens, 6-10 = Dunkard Creek specimens, 11-15 = Muskingum River 
specimens. 



198 



WHITE, McPHERON & STAUFFER 



Mspl Sau96l Ace I 
p uncut ^(— cut— 1 (-cut— II— cut 




-1500 



-600 



-100 



1 2 12 12 12 



FIG. 21. ITS-1 PCR products from LeBoeuf Creek glochidium and adult Ptychobranchus fasciolaris, di- 
gested with Mspl, Sau96l, and Accl. The glochidium exhibits restriction fragment patterns identical to those 
of the adult P. fasciolaris. The glochidium was removed from the gills of an Etheostoma flabeilare specimen. 
1 = glochidium, 2 = adult P. fasciolaris. 



as well. For exannple, the RFLP analysis of 
ITS-1 shown in Figure 2 suggests that pat- 
terns of site gain and loss could demarcate 
tribal boundaries. In many organisms, this 
sort of information has been used to recon- 
struct phylogenetic relationships (reviewed in 
Avise, 1994). Our study was not designed to 
provide the complete matrix necessary to an- 
alyze this question, but our data (summarized 
in Appendix 2) do provide a starting point for 
systematists wishing to pursue the issue of 
higher relationships. (Note that many of the 
results presented in Appendix 2 are unrepli- 
cated and/or based on small numbers of 
specimens.) 

The method is well suited to conservation 
work. It does not entail killing adult unionids 
and hence can be used with endangered 



species. It yields results that are relevant to 
natural communities. It can even furnish in- 
sights into subtle ecological matters, such as 
patterns of host-fish partitioning among 
unionids. Finally, it can be applied to diverse 
systems with large numbers of fish and 
unionid species. 



ACKNOWLEDGMENTS 

This material is based upon work sup- 
ported under a National Science Foundation 
Graduate Fellowship. Any opinions, findings, 
conclusions, or recommendations expressed 
in this publication are those of the authors 
and do not necessarily reflect the views of 
the National Science Foundation. This re- 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 



199 



search was also supported by a grant to Jay 
R. Stauffer, Jr., from the Pennsylvania Wild 
Resources Conservation Fund. G. M. Davis 
and A. E. Bogan encouraged us to pursue the 
research and provided valuable feedback 
throughout the project. M. E. Gordon, M. С 
Hove, R. J. Neves, C. Saylor, W. С Starnes, 
and J. D. Williams contributed important in- 
sights. С Bier, A. E. Bogan, and G. T. Wat- 
ters assisted with specimen collection and 
identification. Field assistance was also pro- 
vided by N. J. Bowers, K. L. Bryan, С 
Gatenby, M. J. Gutowski, E. A. Hale, K. A. 
Kellogg, J. A. Lee, T. Proch, R. Shema, G. A. 
Smith, T. D. Stecko, and E. S. vanSnik. J. 
Clayton provided access to West Virginia col- 
lection sites; С Copeyon assisted in obtain- 
ing endangered-species permits. N. J. Bow- 
ers and W. S. Sheppard provided primer and 
restriction enzyme samples. Laboratory as- 
sistance and advice were furnished by N. J. 
Bowers, E. Carlini, С. L. Crego, D. Cox-Fos- 
ter, and H. -Y. Han. 



LITERATURE CITED 

ANDERSON, R. M., J. B. LAYZER & M. E. GOR- 
DON, 1 991 , Recent catastrophic decline of mus- 
sels (Bivalvia: Unionidae) in the Little South Fork 
Cumberland River, Kentucky. Brimleyana, 17: 
1-8. 

AREY, L. В., 1923a, A microscopical study of 
glochidial immunity. The Anatomical Record, 26: 
350. 

AREY, L. В., 1923b, Observations on an acquired 
immunity to a metazoan parasite. Journal of Ex- 
perimental Zoology, 38: 377-381 . 

AREY, L. В., 1932, A microscopical study of 
glochidial immunity. Journal of Morptiology, 53: 
367-379. 

AVISE, J. C, 1994, Molecular markers, natural his- 
tory and evolution. Chapman and Hall, New 
York, 511 pp. 

CLûiRKE, A. H., 1981, The freshwater molluscs of 
Canada. National Museum of Natural Sciences, 
National Museums of Canada, Ottawa, Canada, 
446 pp. 

CLARKE, A. H., 1985, The tribe Alasmidontini 
(Unionidae: Anodontinae), Part II: Lasmigona 
and Simpsonaias. Smithsonian Contributions to 
Zoology, 399: 1-75. 

CLARY, D. O. & D. R. WOLSTENHOLME, 1985, 
The mitochondrial DNA molecule of Drosophila 
yakuba: nucleotide sequence, gene organiza- 
tion, and genetic code. Journal of Molecular Ev- 
olution, 22: 252-271. 

COKER, R. E., A. F. SHIRA, H. W. CLARK & A. D. 
HOWARD, 1921, Natural history and propaga- 



tion of fresh-water mussels. Bulletin of the 
United States Bureau of Fisheries, 37: 77-181. 

DENNIS, S. D., 1987, An unexpected decline in 
populations of the freshwater mussel, Dysnomia 
(= Epioblasma) capsaeformis, in the Clinch River 
of Virginia and Tennessee. Virginia Journal of 
Science, 38: 281-288. 

FULLER, S. L. H., 1974, Clams and mussels (Mol- 
lusca: Bivalvia). Chapter 8, pp. 215-273, in: с w. 
HART, JR. & 8. L. H. FULLER, eds.. Pollution ecology 
of freshwater invertebrates. Academic Press, 
New York, 389 pp. 

HOGGARTH, M. A., 1988, The use of glochidia in 
the systematics of the Unionidae (Mollusca: Bi- 
valvia). Doctoral Dissertation, The Ohio State 
University, 340 pp. 

HOGGARTH, M. A., 1992, An examination of the 
glochidia-host relationships reported in the lit- 
erature for North American species of Unionacea 
(Mollusca: Bivalvia). Malacology Data Net, 3(1- 
4): 1-30. 

KOCHER, T. D., W. К. THOMAS, A. MEYER, S. V. 
EDWARDS, S. PAABO, F. X. VILLABLANCA & 
A. C. WILSON, 1989, Dynamics of mitochondnal 
evolution in animais: amplification and sequenc- 
ing with conserved primers. Proceedings of the 
Academy of Natural Sciences of the United 
States of America, 86: 6196-6200. 

KRAEMER, L. R., 1970, The mantle flap in three 
species of Lampsilis (Pelecypoda: Unionidae). 
Malacologia, 10: 225-282. 

LEE, S. B. & J. W. TAYLOR, 1992, Phylogeny of 
five fungus-like protoctistan Phytophthora spe- 
cies, inferred from the internal transcribed spac- 
ers of ribosomal DNA. Molecular Biology and Ev- 
olution, 9: 636-653. 

LEFEVRE, G. & W. С CURTIS, 1910, Experiments 
in the artificial propagation of freshwater mus- 
sels. Bulletin of the United States Bureau of Fish- 
eries, 28: 615-626. 

LEFEVRE, G. & W. С CURTIS, 1912, Studies on 
the reproduction and artificial propagation of 
freshwater mussels. Bulletin of the United States 
Bureau of Fisheries, 30: 105-201. 

MARTIN, A. P., R. HUMPHREYS & S, R. PALUMBI, 
1992, Population genetic structure of the armor- 
head, Pseudopentaceros wheeleri, in the North 
Pacific Ocean: application of the polymerase 
chain reaction to fisheries problems. Canadian 
Journal of Fisheries and Aquatic Sciences, 49: 
2386-2391. 

NALEPA, T. F., В. A. MANNY, J. С ROTH, S. С 
MOZLEY & D. W. SCHLOESSER, 1991, Long- 
term decline in freshwater mussels (Bivalvia: 
Unionidae) of the western basin of Lake Erie. 
Journal of Great Lakes Research, 17: 214-219. 

ORTMANN, A, E., 1911, A monograph of the na- 
jades of Pennsylvania, pp. 279-347, In: w. J. Hol- 
land, ed., Memoirs of the Carnegie Museum, 
4(6). Carnegie Institute, Pittsburgh, Pennsylva- 
nia. 

PLEYTE, K. A., S. D. DUNCAN & R. B. PHILLIPS, 
1992, Evolutionary relationships of the salmonid 



200 



WHITE, McPHERON & STAUFFER 



fish genus Salvelinus inferred from DNA se- 
quences of the first internal transcribed spacer 
(ITS 1) of ribosomal DNA. Molecular Phylogenet- 
ics and Evolution, 1 : 223-230. 

RAND, T G. & M. WILES, 1982, Species differen- 
tiation of the glochidia of Anodonta Cataracta 
Say, 1817 and Anodonta implicata Say, 1829 
(Mollusca: Unionidae) by scanning electron mi- 
croscopy. Canadian Journal of Zoology, 60: 
1722-1727. 

REULING, F. H., 1919, Acquired immunity to an 
animal parasite. Journal of Infectious Diseases, 
24: 337-346. 

SAMBROOK, J., E. F. FRITSCH & T MANIATIS, 
1989, Molecular cloning: a laboratory manual, 
2nd ed. Cold Spring Harbor Laboratory Press, 
Cold Spring Harbor, New York. 3 vol. 

SIMON, C, С MC INTOSH & J. DENIEGA, 1993, 
Standard restriction fragment length analysis of 
the mitochondrial genome is not sensitive 
enough for phylogenetic analysis or identifica- 
tion of 17-year periodical cicada broods 
(Hemiptera: Cicadidae): the potential for a new 
technique. Annals of the Entomological Society 
of America, 86: 228-238. 

STERN, E. M. & D. L. FELDER, 1978, Identification 
of host fishes for four species of freshwater mus- 
sels (Bivalvia: Unionidae). American Midland 
Naturalist, 100: 233-236. 

VAUGHT, K. C, 1989, A classification of tfie living 
Mollusca. American Malacologists, Inc., Mel- 
bourne, Florida, 195 pp. 

WALLER, D. L. & L. E. HOLLAND-BARTELS, 1988, 
Fish hosts for glochidia of the endangered fresh- 
water mussel Lampsilis higginsi Lea (Bivalvia: 
Unionidae). Malacological Review, 21: 119-122. 



WHITE, L. R., 1994, Identification of fresfiwater 
mussel glochidia on host fishes from French 
Creek, Pennsylvania, using molecular tech- 
niques. Doctoral dissertation, The Pennsylvania 
State University, 75 pp. 

WHITE, L. R., B. A. MC PHERON & J. R. 
STAUFFER, JR., 1994, Identification of freshwa- 
ter mussel glochidia on host fishes using restric- 
tion fragment length polymorphisms. Molecular 
Ecology, 3: 183-185. 

WHITMORE, D. H., T. H. THAI & С M. CRAFT, 
1992, Gene amplification permits minimally inva- 
sive analysis of fish mitochondhal DNA. Trans- 
actions of the American Fisheries Society, 121: 
170-177. 

WILES, M., 1975, The glochidia of certain Union- 
idae (Mollusca) in Nova Scotia and their fish 
hosts. Canadian Journal of Zoology, 53: 33-41. 

WILLIAMS, J. D., S. L. H. FULLER & R. GRACE, 
1992, Effects of impoundment on freshwater 
mussels (Mollusca: Bivalvia: Unionidae) in the 
main channel of the Black Warrior and Tombig- 
bee rivers in western Alabama. Bulletin of the 
Alabama Museum of Natural History, 13: 1-10. 

WILLIAMS, J. D., M. L. WARREN, JR., K. S. CUM- 
MINGS, J. L. HARRIS & R. J. NEVES, 1993, Con- 
servation status of freshwater mussels of the 
United States and Canada. Fisheries, 18(9): 
6-22. 

ZALE, A. V. & R. J. NEVES, 1 982, Fish hosts of four 
species of lampsiline mussels (Mollusca: Union- 
idae) in Big Moccasin Creek, Virginia. Canadian 
Journal of Zoology, 60: 2535-2542. 



Revised Ms. accepted 20 May 1996 



MOLECULAR GENETIC IDENTIFICATION OF UNIONIDS 201 

APPENDIX 1. Estimated sizes of restriction fragments used in the key (excluding fragments shorter than 
100 bp). 

Couplet Pattern Fragment Size (bp) Couplet Pattern Fragment Size (bp) 

1 A= (275 - 285) + 185 5 
B= 305 + 185 
C= 275 + 140* 6 
D= 305 + 140" 
E= 465 + 140 7 
F= (495 - 505) + 140 
G = 960 8 
H= 895 
'anomalous P. sintoxia = 9 

205 + 140 

''anomalous A. plicata = 10 

305 + 170 

2 A= 350 + 230 11 
B= 240 + 225 + 185 

3 A= 240 + 165 12 
B= 200 + 165 
C= 235 + 195' 13 
'anomalous L. siliquoidea = 

1 = 245 + 205 + 170 14 

2 = 275 + 245 + 200 

4 A= 390 + 270 + 170 15 
A'= 375 + 265 + 170 
B= 310 + 265 + 170 



= 


415 


A = 


250+ 180 


A = 


575 + 280 + 205 


A' = 


575 + 290 + 195 


= 


615 


A = 


510 


= 


575 


A = 


370 + 225 


= 


565 


A = 


340 + 225 


A = 


385 + 270+ 195 


A' = 


400 + 275+ 195 


A = 


585 


B = 


555 


= 


(590 - 620) 


A = 


430 + 190 


A = 


455 


B = 


520 


= 


(600 - 610) 


A = 


370 + 240 


A = 


500 + (445 - 461 




195 


A' = 


515 + 460 + 195 



202 



WHITE, McPHERON & STAUFFER 



gl 

0) 
GC 



< < 



Ci^ ij!. Ю !£¿ CO 

OO 0< О 



t;^ ^ го 00 i£2. í::^ OJ 

m < < < < < ü 



- о ü CD CD I 



CD 



£i £2- 1:^ 2^ t- £i 
Q ü О О О ü 



2- со ^ 2;, !£5. in £i 

X LJ_ U. LL LL LL CD 



о о < CÛ о CÛ 



о < cû m cû m Q 



û LL 



ШШШШ <<<<cû<<Q<o<< 



<<<<<<< 



< m 



Qû 



'3^ 

со Cû Cû Qû 



<<<<<<<<<CÛCÛ< 



Й 52. ::^$£.S.^ z:^ 3- S2. 52- iíl î^ £2- n^ 3^ i^ :;^ 2- 
Ош OOÜO <<<<<<< Cû <<< < 



О OOOO OO OOOO OO О 



О О 



OOOOOOOOO 



■,- CD COLO 

<ooo 



z:^ О £i £i n^ C- £i 
OOO OO OO 



OOO < О О 



OO 



о о 



íü. S2. £2- Ci SL Ei £i H^ zz, 
OOOOOOO О О 



< < 



т-^ c\j^ сч, со, ГЧ ¡^ ГЧ, 2^ 

<<<<<<< о 



< < 



< < < < 



< < < < < < 



со 1^ ,^ 
< О < 



та 
с 

■^ та со 



та о та 
' та m 



с ел 



о gi2 
^ та^ 
с Е со 

E "О С 

•û й о 
3 -Э с 
(Л ч -=]; 



та 


? 


Q. CL 


ь 


Ь 


о 


сч 


о 


о 


та 


та 


I.- 


г 


о 


о 


О) О) 


F F 


со 


0, 


та 


tn 


-J 


-J 



CTJ;g 5 

m ^ -Э 
-2 С ^ 

8 о^§ 

о ^ -2 

Е !§■ 

со О) Р 
та ^ ?- 
-J CL со 



ш 

та 

с 

1._ 

Е 0) 

га ^ 
1^ 



та F 
о с 

IÎ2 
о. о^^ 
^ та Û. 

■^ та lb 

'ч: о 



5 о 



Я -Q ой 



_та 



со та та 

та S Е 

та 3Í 3í 

с -Q -Q 

о g g 

ùj u. s; 5; 



с 

._ QJ 

11 

E со 

tu та 

-> та 

<^ S 



та P 
о Ф 

^ 2. 



~ ™. та 

та та о 

E e J^ 

со со :~ 

-Э -2 55 

-Q -Q Q. 

О О g 

АД та 

Uj Uj -J 



О 

G 

со 

чЭ 
со 


та 

S 

о 

.со 


о 
.о- 

СО 

со 


а nasuta 
а recta 
branchus 
fabilis 
iris 


О 


со 

Q 


О 


P S í¿ та та 

t с -С со со 


ь 

та 

—1 


та 

—1 


ь 

та 


О) О) ^ .= ^5 

:а ^j 0- i> i> 



MAUXCOLOGIA, 1996, 38(1-2): 203-212 

QUANTITATIVELY SAMPLING LAND-SNAIL SPECIES RICHNESS IN 
MADAGASCAN RAINFORESTS 

Kenneth С Emberton, Timothy A. Pearce & Roger Randalana 

Molluscan Biodiversity Institute, 21 6- A Haddon Hills, Haddon field, New Jersey 08033, U.S.A. 

and Institute for the Conservation of Tropical Environments, B.P. 3715, Tsimbazaza, 

Antananarivo 101, Madagascar 

ABSTRACT 

Land-snail species richness in tropical rainforests tends to be high but difficult to assess 
because of low densities and often small shell sizes. We tested three quantitative sampling 
methods in primary rainforests of southeastern Madagascar. Timed searching yielded seven 
times as many micro-snail species (species that during at least part of their life have shells < 5 
mm maximum dimension) per person-hour as either litter sampling or soil-plus-litter sampling. 
The number of species found in 20 m ■ 20 m during three person-hours of searching, however, 
was boosted a maximum of 38% by one eight-liter sample each of litter and soil-plus-litter. 
Litter sampling and timed searching both yielded more than 1.5 times the proportion of live- 
collected species as soil-plus-litter sampling. Sampling method was unbiased toward 1 2 of the 
20 commonest species, but three large, presumed arboreal species were favored by timed 
searches; two minute, presumed burrowers by soil-plus-litter sampling; and three minute, 
cryptically colored species by both litter and soil-plus-litter sampling. A 1.2-mm sieve caught 
at least 78% of the total specimens and passed adults of 7% of species, of which the smallest 
adult dimension was 1.0 mm. These results suggest that the best sampling strategy is timed 
searching for micro-snails, while incidentally collecting macro-snails and litter-plus-soil for later 
picking of the 5.5-1.2 mm and the 1.2-0.85 mm, dry-sieved fractions. This strategy should be 
transferable to other tropical-rainforest land-snail faunas. 

Key words: Gastropoda, tropical biodiversity, leaf-litter biota, soil biota. 



INTRODUCTION 

Land snail faunas of tropical rainforests 
tend to be quite diverse (maximum reported: 
52 species per 4 ha) despite often low den- 
sities (Emberton, 1995a [and citations there- 
in]; Tattersfield, 1994, in prep.; F. Thompson, 
pers. commun.; despite Solem's [1984] un- 
documented statement to the contrary). Much 
of this diversity consists of micro-gastropods 
(< 5 mm greatest dimension), the collection 
of which can be extremely labor-intensive 
(Emberton, 1994, 1995a, 1996; DeWinter, 
1995; F. Thompson, pers. commun.; P. Tat- 
tersfield, pers. commun.; R. Ramirez, pers. 
commun.). Because most tropical rainforests 
are vastly undercollected for micro-gastro- 
pods and are undergoing irreversible defor- 
estation, great urgency attaches to collecting 
these mostly undiscovered, undescribed mol- 
luscs as efficiently and thoroughly as pos- 
sible. Because of the prime importance in 
land-snail systematics of preserving anato- 
mies and DNA in ethanol, sampling methods 
should maximize live collections. Because 
land snails are generally so patchily distrib- 



uted, even within seemingly uniform forest, 
sampling should probably avoid random- 
quadrat methods (Emberton, 1995a). 

Timed searches by experienced collectors 
are a well-proven method of quantitatively 
sampling patchily distributed organisms (Cod- 
dington et al., 1991). One of us has recently 
advocated timed searches as the most effi- 
cient collecting method for tropical rainforest 
micro-snails (Emberton, 1995a), and has ap- 
plied such data toward assessing conserva- 
tion priorities (Emberton, 1996). The efficacy 
of timed searches for collecting all or a sub- 
stantial portion of the micro-gastropod fauna, 
however, has never been tested, to our knowl- 
edge. 

Collection of measured quantities of se- 
lected leaf litter is another quantitative sam- 
pling method that has proven effective for 
tropical-rainforest land-snail communities 
(Tattersfield, 1994). Soil-plus-litter samples 
also often yield species that are collected in 
no other way (F. Thompson, 1995). Some 
species may be soil specialists, other species 
may take refuge in soil from drying litter, and 
soil can accumulate dead shells of litter spe- 



203 



204 



EMBERTON, PEARCE & RANDALANA 



cialists (pers. observ.; Burch & Pearce, 1990). 
Processing of soil-plus-litter samples, how- 
ever, is more labor-intensive than processing 
of litter samples. 

The purpose of this paper is to compare 
the performances of (a) timed searching, (b) 
litter sampling, and (c) soil-plus-litter sam- 
pling for determining the species richness of 
and obtaining live material of the micro-land- 
snail fauna of Madagascan rainforests, and 
to arrive thereby at the most efficient overall 
sampling strategy. 



METHODS AND MATERIALS 

We sampled 48 plots, each 20 m x 20 m, at 
16 stations on three widely separated moun- 
tains in southeastern Madagascar (Fig. 1 , Ta- 
ble 1). Localities and stations were chosen to 
serve both for this study and for testing di- 
versity patterns between the Vohimena and 
Anosy mountain chains (Emberton, 1996, 
Emberton et al., in review). Stations were at 
100 m elevation intervals from 100 m to 500 
m and at 200 m elevation intervals above 500 
m, with a station at the highest or a local 
summit. 

Stations were restricted to primary forest 
that had no more than limited selective cut- 
ting. For each station, we recorded the ele- 
vation (average of two Thommen Altitrek al- 
timeters, calibrated from topographic maps), 
latitude and longitude (from topographic 
maps), and the topography (summit, ridge, 
slope, or valley). For more extensive data on 
these stations, see Emberton (in review). 

At each station, we sampled three adjacent 
20 m X 20 m replicate plots, each marked off 
with flagging tape. We sampled 25 January 
to 7 February 1995, during the rainy season, 
within one week of heavy rains, when snails 
and slugs seemed likely to be most active 
and therefore perhaps easier to find. We in- 
cluded only micro-snails, which for the pur- 
poses of this study we defined as those spe- 
cies that during at least part of their life have 
shells that are smaller than 5 mm maximum 
dimension (the vast majority remain below 
this size as adults). 

Timed searching was for three person- 
hours per plot: one-half hour by six collec- 
tors. Three of these collectors (RR and two 
assistants who had been trained by all three 
authors) were constant over all stations and 
plots, and the other three were hired locally 
and trained by RR. As incentives, small cash 



prizes were offered for the most snails and 
the smallest snail collected in each plot. Mi- 
cro-molluscs were hand-collected into 
30-ml, snap-cap vials, drowned overnight, 
then fixed and preserved in 70-90% ethanol. 

Litter samples and soil-plus-litter samples 
were each eight I in volume per plot, col- 
lected over a 30-minute period by KCE and 
TAP, respectively. Both types of sampling 
were from moist, sheltered microhabitats 
such as beside logs, between buttress roots 
of trees, within Asplenium and Pandanus ro- 
settes, under and near piles of Ravenala and 
palm fronds, and in moist depressions (Em- 
berton & Arijaona, in press: fig. 2). Litter and 
litter-plus-soil samples were collected into 
four-mill plastic bags and kept as cool as 
possible until processing, a maximum of 
three days later, with daily opening of each 
bag for aeration. 

All litter and soil-plus-litter samples were 
wet-sieved through three mesh sizes: 11.5 
mm, 5.5 mm, and 1 .2 mm. We used wet siev- 
ing (i.e. washing the samples with water) in 
order to process quickly samples wet from 
recent or current rains, and to assure live re- 
covery of slugs, semislugs, and thin-shelled 
species. Sieve boxes for the first three size 
fractions consisted of large plastic storage 
boxes (55 X 48 >■ 35 cm) from which the bot- 
toms had been cut (leaving a 3.8-cm margin), 
then covered with hardware cloth (1 1 .5 mm), 
hardware mesh (5.5 mm), or hardware screen 
(1.2 mm) (the latter two supported by hard- 
ware cloth) held in place with duct tape. The 
three sieve boxes were nested over an intact 
box to catch effluent during washing of a lit- 
ter sample and were transferred to a second 
box if the first filled. Whenever the litter or 
soil-plus-litter samples were not too wet, as 
much dry-sieving as possible was performed 
prior to wet-sieving. The first two sieve frac- 
tions were picked immediately for all inverte- 
brates by the authors, aided by teams of local 
workers, each of whom was carefully trained 
and monitored by at least one of the authors. 
The third fractions (retained by the 1.2-mm 
sieve) were fixed and stored for no longer 
than three weeks in an equal or greater vol- 
ume of 90% ethanol (the resulting ethanol 
concentration averaged about 60%). The ef- 
fluent was caught by pouring all sieved wash 
water from the bottom box or boxes through 
two nested nylon stockings, from which ex- 
cess water was squeezed gently, then which 
were fixed and stored in an equal or greater 
volume of 90% ethanol. 



RAINFOREST IJ\ND-SNAIL SAMPLING 



205 



Z4 



^C.'30' 



47*30* 



25°- 





♦^MAHERMANA 



LAPIRY 



TO LAG NJ ARO 

VASIHA 

о 



30 KM 




FIG. 1. The three mountains sampled in the Anosy and Vohimena chains, southeastem-most Madagascar 
(see inset). Contours are shown at 500 m and 1 ,000 m. The dashed line indicates Andohahela Reserve. The 
dot indicates the city of Fort Dauphin (= Tolagnaro). 



All > 1.2-mm sieve fractions were picked 
for all inverlebrates by RR and six assistants, 
each of whom was trained by all three au- 
thors and monitored by RR. Non-molluscan 
invertebrates are being distributed among in- 
terested specialists. Only molluscs are anal- 
ysed in this paper. 

To test the efficiency of the 1 .2-mm sieve 
at catching snails, the sieving effluent (i.e., all 
that passed through the 1.2-mm sieve) from 
one plot per station (the plot whose upper 



sieve fractions yielded the greatest number 
of species) was further sieved through U.S.A. 
Standard Testing Sieves Nos. 20 and 30 
(0.85 mm and 0.60 mm). Both these fine frac- 
tions were picked for snails and shells by RR 
and four trained, monitored assistants, wear- 
ing Optivisor magnifying lenses of 2x magni- 
fication. Picking of all sieve fractions was 
performed on a white or light-gray, hard sur- 
face. Those snails from the < 1 .2 mm fraction 
were used only for testing the sieve effi- 



206 



EMBERTON, PEARCE & RANDALANA 



TABLE 1. Stations sampled for land snails in southeastern Madagascar. Elv 
meters, r/s/v = ridge, slope, and valley. 



elevation in 



# 


Mountain 


Elv 


Latit. S 


Long. E 


Topogr 


1 


Mahermano 


340 


24.26.12 


47.13.13 


summit 


2 


Mahermano 


300 


24.26.17 


47.13.10 


slope 


3 


Mahermano 


200 


24.26.15 


47.13.04 


slope 


4 


Mahermano 


100 


24.26.22 


47.12.41 


valley 


5 


llapiry 


540 


24.51.40 


47.00.20 


summit 


6 


llapiry 


500 


24.51.33 


47.00.27 


ridge 


7 


llapiry 


400 


24.51.27 


47.00.38 


r/s/v 


8 


llapiry 


300 


24.51.36 


47.00.40 


slope 


9 


llapiry 


200 


24.51.39 


47.00.46 


slope 


10 


Vasiha 


860 


24.55.18 


46.44.19 


summit 


11 


Vasiha 


700 


24.55.23 


46.44.27 


slope 


12 


Vasiha 


500 


24.55.19 


46.44.45 


slope 


13 


Vasiha 


400 


24.55.25 


46.44.45 


valley 


14 


Vasiha 


300 


24.55.37 


46.44.49 


slope 


15 


Vasiha 


200 


24.56.13 


46.45.13 


slope 


16 


Vasiha 


100 


24.56.20 


46.46.07 


slope 



ciency, and were not included in the main 
data matrix or data analysis. 

All snails and shells were sorted and iden- 
tified to morphospecies by KCE. For each 
morphospecies, a relatively intact adult rep- 
resentative was chosen and was photo- 
graphed in two to five diagnostic views at 
standard magnifications, using a Polaroid 
camera mounted on a Wild dissecting micro- 
scope. The resulting reference collection and 
file of photographs were used to identify all 
specimens, both adults and juveniles, except 
for some juveniles of the most minute sieve 
fractions, which were identified only to genus 
or family. Systematic treatments of the mor- 
phospecies, 85% of which are new, are in 
progress; vouchers are in the collection of 
the Molluscan Biodiversity Institute, with 
types and references to be placed in the 
Madagascar national museum (Pare Bota- 
nique et Zoologique de Tsimbazaza, An- 
tananarivo) and in the Academy of Natural 
Sciences of Philadelphia. (Patterns of diver- 
sity, distribution, and abundance of the mor- 
phospecies are treated in a separate paper 
[Emberton, in review].) 

To compare efficiencies of the three meth- 
ods, we calculated the number of person- 
hours required to collect and — in the case of 
litter and litter-plus-soil — to wet-sieve and to 
pick an average sample (for all invertebrates). 
We then computed the mean numbers of 
molluscan specimens and of species ob- 
tained per person hour by each method. We 
were not able to calculate the percent of 
picking time devoted to molluscs alone, so 



our person-hour calculations were overesti- 
mates. 

We used analysis of variance (ANOVA) by 
least-squares estimation (Wilkinson, 1990) to 
evaluate differences among the three sam- 
pling methods in (a) number of species col- 
lected per plot, (b) percent of the total spe- 
cies that were found in each plot, and (c) 
percent of species collected live. For the per- 
cent of the total species collected within 
each plot, we used the entire data set in a 
one-way ANOVA. For species number and 
percent live, however, we factored out the 
effects of locality (mountain) and elevation by 
including them in a three-way ANOVA on the 
largest possible subset of the data including 
all three mountains (see Emberton et al., in 
review: fig. 2), which had to be limited to 200 
m and 300 m elevations (Table 1). 

For each species representing at least one 
percent of the total specimens, we used chi- 
square analysis to test among the three sam- 
pling methods for equal numbers of speci- 
mens. Predicted frequencies were based on 
the total number of specimens resulting from 
each method. Probability estimates were 
Bonferroni-adjusted to allow for multiple 
tests. 



RESULTS 

Including the macro-snail species that 
showed up in the upper sieve fractions, we 
collected a total of 87 species (also see be- 
low). Taxonomically, these species were dis- 



RAINFOREST LAND-SNAIL SAMPLING 



207 



TABLE 2. Average time investments and productivities of three sampling methods. Collect = collecting 
within a 20 m ' m plot, Sieve = wet sieving of an eight-liter sample from a 20 m >; 20 m plot, Pick = 
picking all invertebrates (not just gastropods) from the > 1.2-mm sieved sample. Total hours = total 
person-hours per plot sample, Specm. /p-hr = mean number of specimens obtained per person hour, 
Spp./p-hour = mean number of species obtained per person hour, Spp. /specm. = proportion of mean 
species to mean specimens. 





Person-H 


ours per 


Task 


Total 
hours 


Specm. 
p-hr 


Spp./ 
p-hr 


Spp./ 
specm. 


Method 


Collect 


Sieve 


Pick 


Timed search 
Litter sample 
Soil-plus-litter 


3.0 
0.5 
0.5 


0.0 
4.8 
4.8 


0.0 
4.7 
9.8 


3.0 
10.0 

15.1 


9.36 
0.88 
0.91 


3.03 
0.46 
0.41 


0.32 
0.52 
0.45 



tributed as follows, with higher classification 
following Abbott & Boss (1989) for "Proso- 
branchia" and Gymnomorpha and Nordsleck 
(1986) for Pulmonata: 

"Subclass PROSOBRANCHIA" 
Order MESOGASTROPODA 
Superfamily CYCLOPHOROIDEA 
Cyclophoridae 

Boucardicus 17 

Cyathopoma 1 

Hainesia 1 

Diplomrnatinidae 

Malahnia 1 

Superfamily LITTORINOIDEA 
Pomatiasidae 

Tropidophora 3 

Superfamily RISSOOIDEA 
Assimineidae 

Omphalotropis 2 

Subclass GYMNOMORPHA 
Order SOLEOLIFERA 

Veronicellidae 1 

Subclass PULMONATA: Order 
STYLOMMATOPHORA 
Suborder ORTHURETHRA 

Superfamily CHONDRINOIDEA 
Orculidae 

Fauxulus 2 

Suborder SIGMURETHRA 
Infraorder ACHATINIDA 
Superfamily ACHATINOIDEA 

Subulinidae 3 

Superfamily STREPTAXOIDEA 

Streptaxidae 14 

Superfamily ACAVOIDEA 
Acavidae 

Ampelita 1 

Clavator 1 

Helicophanta 1 

Superfamily PUNCTOIDEA 

Charopidae 9 



Infraorder HELICIDA 
Superfamily HELICARIONOIDEA 
Helicarionidae: Sesahnae 

Kaliella 1 

Helicarionidae: Microcystinae 

Microcystis 10 

Helicarionidae: Arlophantinae 

Kalidos 7 

Malagahon 1 

Helicarionidae: Macrochlamydinae 

Sitala 9 

We excluded from analysis all specimens of 
the one slug species (Veronicellidae) and of 
the six snail species that were considered al- 
ways too large, even as juveniles, to qualify 
as micro-molluscs (< 5 mm): the one Haine- 
sia, two of the three Tropidophora, and all 
three acavids. 

Distributions of the 80 analyzed species 
among samples, totalling 2,430 specimens, 
are archived at the Molluscan Biodiversity In- 
stitute (MBI) and the Academy of Natural Sci- 
ences of Philadelphia (ANSP). 

The three sampling methods required 
drastically different investments of time to 
acquire gastropods (Table 2). Timed search 
was by far the most efficient, yielding about 
ten times the number of specimens and 
seven times the number of species per per- 
son-hour as either litter sampling or soil-plus- 
litter sampling. These advantages are inflated 
somewhat, however, because we took time 
to pick all invertebrates. 

The litter and soil-plus-litter methods were 
more diverse than timed search, yielding 
about half again as many species per speci- 
men (also see below). 

Table 3 gives ANOVA results for number of 
species collected per 20 m x 20 m plot. Sam- 
pling method had a highly significant effect 
when the less significant effect of elevation 



208 



EMBERTON, PEARCE & RANDALANA 



TABLE 3. Analysis of variance in the number of species collected per 20 m ^ 20 nn plot, with 
least-squares estimates of means. Independent variables are sampling method (timed search vs. litter 
sample vs. soil-plus-litter sample), elevation (200 m vs. 300 m), and location (one of three mountains). 





Sum of 


Degrees of 


Mean 




Probability 


Source 


Squares 


Freedom 


Square 


F-Ratio 


of Equality 


Sampling 


172.0 


2 


86.0 


16.83 


0.000*** 


Elevation 


29.6 


1 


29.6 


5.80 


0.021* 


Locality 


28.3 


2 


10.2 


1.99 


0.152 


Sam X Elv 


16.1 


2 


8.1 


1.58 


0.220 


Sam X Loc 


38.3 


4 


9.6 


1.88 


0.136 


Elv X Loc 


0.9 


2 


0.5 


0.09 


0.914 


Sx ExL 


10.6 


4 


2.7 


0.52 


0.722 


Error 


184.0 


36 

Number of Species 


5.1 








Mean 


Std. Error 


N 




Sampling: 










Timed 


9.0 


0.5 


18 






Litter 


4.7 


0.5 


18 






Soil-Lit 


6.3 


0.5 


18 






Elevation: 












200m 


7.4 


0.4 


27 






300m 


5.9 


0.4 


27 






Locality: 












Mahermano 


5.9 


0.5 


18 






llapiry 


7.4 


0.5 


18 






Vasiha 


6.6 


0.5 


18 







'p < 0.05. 



0.001. 



was partitioned out (see Emberton et a!., in 
review, concerning elevational variation). 
Timed searching within 20 m x 20 m for three 
person-hours averaged 9.0 species. This was 
about twice as many species as occurred in 
an eight-liter sample of litter selected from 
the same area (4.7 species), and was about 
half again as many species as occurred in an 
equivalent soil-ptus-litter sample (6.3 spe- 
cies). Thus, this timed searching method pro- 
duced more species than the other two sam- 
pling methods. When considered in the 
context of time invested, the productivity of 
timed searching by this method was even 
more pronounced (see above). 

Timed searching alone, however, fell far 
short of assessing total number of species 
collected. ANOVA results in Table 4 indicate 
that timed searching produced on average 
only 72% of the species sampled within a 20 
m X 20 m plot. Thus, the number of species 
found in a plot during three person-hours of 
searching was boosted 39% (28%/72%) by 
one eight-l sample each of litter and soil- 
plus-litter. Most of these additional species 
occurred in soil-plus-litter samples, which 
yielded half of the total, as opposed to the 



litter samples, which yielded only somewhat 
over a third of the total sampled species. 

On the other hand, Table 5 shows that for 
sampling live-collected individuals, litter 
sampling was equivalent to timed searching 
(51 .6 ±5.6 = 46.4 ±5.2) and significantly more 
efficient than soil-plus-litter sampling. Thus, 
nearly half of the litter-sample and timed- 
search species were represented by at least 
one live-collected individual, whereas only 
somewhat over a fourth of the soil-plus-litter- 
sample species were. This result was not sur- 
prising because soil can accumulate dead 
shells of snails living in litter or trees (pers. 
observ.; Burch & Pearce, 1990). In other 
words, litter sampling and timed searching 
both yielded more than 1 .5 times the propor- 
tion of live-collected species as soil-plus-lit- 
ter sampling. 

Table 6 shows the total live-plus-dead 
number of each species collected by each of 
the three sampling methods. Twenty species 
(25%) were represented by at least 1 % (> 24) 
of the total specimens. Chi-square tests on 
these species indicated that 12 (60%) of 
them had equal (not significantly different) 
representation among sampling methods. Of 



RAINFOREST U\ND-SNAIL SAMPLING 



209 



TABLE 4. Analysis of variance in the percent of species that were collected within each replicate plot, 
with least-squares estimates of means, the independent variable is sampling method (timed search vs. 
litter sample vs. soil-plus-litter sample). 



Source 



Sum of 
Squares 



Degrees of 
Freedom 



Mean 
Square 



F-Ratio 



Probability 
of Equality 



Sampling 
Error 


30,530.9 
43,868.3 


2 

141 

Percent of Species 


15,265.4 
311.1 




Mean 


Std. Error 


N 


Sampling: 
Timed 
Litter 
Soil-Lit 


72.4% 
37.1% 
50.1% 


2.5% 
2.5% 
2.5% 


48 
48 
48 


*•* p « 0.001. 









49.1 



0.000"* 



TABLE 5. Analysis of variance in the percent of species represented by at least one live-collected 
individual, with least-squares estimates of means. Independent variables are sampling method (timed 
search vs. litter sample vs. soil-plus-litter sample), elevation (200 m vs. 300 m), and location (one of 
three mountains). 





Sum of 


Degrees of 


Mean 




Probability 


Source 


Squares 


Freedom 


Square 


F-Ratio 


of Equality 


Sampling 


5057.0 


2 


2528.5 


5.24 


0.010" 


Elevation 


165.1 


1 


165.1 


0.34 


0.562 


Locality 


922.9 


2 


461.5 


0.96 


0.394 


Sam ■ Elv 


381.0 


2 


190.5 


0.40 


0.677 


Sam ' Loc 


1446.5 


4 


361.6 


0.75 


0.565 


Elv ■ Loc 


712.9 


2 


356.4 


0.74 


0.485 


S - E . L 


2627.2 


4 


656.8 


1.36 


0.268 


Error 


16405.0 


34 

Percent Live Species 


482.5 








Mean 


Std. Error 


N 




Sampling: 










Timed 


46.4% 


5.2% 


18 






Litter 


51 .6% 


5.6% 


16 






Soil-Lit 


28.4% 


5.2% 


18 






Elevation: 












200m 


43.9% 


4.5% 


25 






300m 


40.3% 


4.2% 


27 






Locality: 












Mahermano 


47.8% 


5.4% 


17 






llapiry 


41.3% 


5.2% 


17 






Vasiha 


37.3% 


5.4% 


17 







0.01. 



the remaining eight species, Boucardicus sp. 
9 and Microcystis sp. 4 were significantly 
more prevalent in both litter and soil-plus- 
litter samples than in timed-search samples, 
and Sitala sp. 7 was present in the litter sam- 
ples in greater proportions than expected in 
the chi-square test. All three of these are both 
dark brown in color (matching the color of 



litter and soil) and minute in size (adult great- 
est dimensions 2.2 mm, 2.2 mm, and 1.8 
mm, respectively). 

Two species — Streptaxidae spp. 9 and 
1 3 — were predominant in soil-plus-litter sam- 
ples and notably scarce in litter only samples. 
Both these species are high-spired (height/ 
diameters 2.5 and 2.2), very small (adult 



210 



EMBERTON, PEARCE & RANDALANA 



TABLE 6. Numbers of snails of each of 80 species collected using three different sampling methods: t = 
timed search, I = litter sample, s = soil-plus-litter sample. Chi-square tests for equal frequencies among 
sampling methods were calculated for each species with > 24 specimens: * p < 0.05, Bonferroni 
adjusted. GnSp = genus or family and numbered morphospecies. Genera and families in taxonomic 
order are: BO, Boucardicus; CY, Cyathopoma; MN, Malarinia; TR, Tropidophora; OM, Omphalotropis; 
FA, Fauxulus; SU, Subulinidae; ST, Streptaxidae; CH, Charopidae; KL, Kaliella; Ml, Microcystis; KD, 
Kalidos; MG, Maiagarion; and SI, Sitala. 







Nurr 


iber 




Chi-Sq 


GnSp 




Nurr 


iber 






GnSp 


t 


1 


s 


Total 


t 


1 


s 


Total 


Chi-Sq 


BO01 


131 


35 


43 


209 


5.4 


CHOI 


28 


4 


3 


35 


9.0 


BO02 


47 


11 


15 


73 


2.5 


CH02 


75 


34 


59 


168 


8.2 


BO03 


3 





2 


5 


— 


CH03 


6 





3 


9 


— 


BO04 


11 


5 


8 


24 


— 


CH04 


14 


1 


12 


27 


5.9 


BO05 


2 





1 


3 


— 


CH05 


14 


2 


5 


21 


— 


BO06 


6 


3 





9 


— 


CH06 


18 


2 


8 


28 


2.1 


BO07 


32 


10 


9 


51 


2.4 


CH07 


3 








3 


— 


BO08 


2 





3 


5 


— 


CH08 


1 








1 


— 


BO09 


1 


39 


33 


73 


102.3* 


CH09 


4 








4 


— 


ВОЮ 


1 


1 


1 


3 


— 


KL01 


8 


6 


1 


15 


— 


B011 


7 





5 


12 


— 


MI01 


15 


3 


4 


22 


— 


B012 


1 





1 


2 


— 


MI02 


7 





1 


8 


— 


B013 


5 


5 


6 


16 


— 


MI03 


31 


9 


14 


54 


0.1 


B014 


2 


2 


2 


6 


— 


MI04 


1 


9 


20 


30 


34.6* 


B015 





1 





1 


— 


MI05 


1 





2 


3 


— 


B016 


1 


1 





2 


— 


M 106 


2 


1 





3 


— 


B017 





1 





1 


— 


MI07 





1 





1 


— 


CY01 


13 


9 


12 


34 


4.3 


MI08 


2 


4 


2 


8 


— 


MN01 








1 


1 


— 


MI09 


3 


2 





5 


— 


TR01 


195 


20 


56 


271 


33.0* 


MHO 


2 





1 


3 


— 


OM01 





1 


2 


3 


— 


Mill 


1 








1 


— 


OM02 


1 


7 


1 


9 


— 


MI12 


1 








1 


— 


FA01 


1 


2 


1 


4 


— 


KD01 


705 


9 


21 


135 


27.7* 


FA02 








1 


1 


— 


KD02 


22 


2 





24 


— 


SU01 


104 


18 


54 


176 


6.3 


KD03 


7 


1 





8 


— 


SU02 


3 


2 


2 


7 


— 


KD04 


6 


4 





10 


— 


SU03 


2 


5 


6 


13 


— 


KD05 


1 








1 


— 


ST01 


9 





1 


10 


— 


KD06 


7 


1 


1 


9 


— 


ST02 


6 


1 


3 


10 


— 


KD07 


9 


11 


1 


21 


— 


ST03 


11 


2 





13 


— 


MG01 


9 


3 


1 


13 


— 


ST04 


9 


3 


8 


20 


— 


SI01 


11 


5 


3 


19 


— 


ST05 


2 








2 


— 


SI02 


7 





3 


10 


— 


ST06 


88 


25 


53 


166 


2.0 


SI03 


1 


1 


4 


6 


— 


ST07 


20 


2 


7 


29 


2.9 


SI04 


5 


1 


3 


9 


— 


ST08 


13 


1 


4 


18 


— 


SI05 


49 





2 


51 


34.2* 


ST09 


13 


1 


25 


39 


27.8* 


SI06 


4 








4 


— 


ST10 


2 


11 


7 


20 


— 


SI07 


102 


70 


78 


250 


27.8* 


ST11 


2 


2 


5 


9 


— 


SI08 


1 








1 


— 


ST12 


6 


1 


3 


10 


— 


SI09 








1 


1 


— 


ST13 


10 


8 


27 


45 


26.7* 














ST14 


3 





1 


4 


— 


Tot 


1348 


420 


662 


2430 





heights 3.9 mm and 3.6 mm), and with 
glossy, fusiform, small-apertured shells sug- 
gestive of a soil-burrowing niche. In contrast, 
Tropidophora sp. 1 , Kalidos sp. 1 , and Sitala 
sp. 5 all occurred predominantly in timed 
searches and were significantly under-repre- 
sented in litter and soil-plus-litter samples. All 



three of these species are relatively large 
(adult greatest dimensions 13.1 mm, 33.5 
mm, and 7.3 mm). Tropidophora sp. 1 is of- 
ten if not exclusively arboreal, and K. sp. 1 
juveniles are at least partially arboreal, as 
they frequently show up in vegetation-beat- 
ing samples (Emberton, unpublished); S. sp. 



RAINFOREST LAND-SNAIL SAMPLING 



211 



5 has a fragile, light-colored shell that is high- 
spired for the genus (height/diameter 1 .0), all 
suggestive of arboreality. 

A total of 101 specirnens passed through 
the 1.2-mm sieve. (Distributions of these 
specimens among species and samples are 
archived at MBI and ANSP.) Thus, the 
1.2-mm sieve caught a minimum of 78% of 
the specimens in the litter and litter-plus-soil 
samples of each plot. 

The 1 .2-mm sieve caught representatives 
of all species in the samples, however, ex- 
cept for one: Streptaxidae sp. 15. This is a 
minute, high-spired species (adult height 2.4 
mm, diameter 1.0 mm), of which only two 
specimens were obtained. In addition, the 
sieve passed at least one adult of the five 
smallest species of Boucardicus, some in 
substantial numbers. Thus, adults of six spe- 
cies (8% of total) passed through the sieve at 
least in part. The smallest adult dimension of 
any of these species was 1 .0 mm. 



DISCUSSION 

Sieving of litter and litter-plus-soil may at 
first seem superior to timed searches for 
sampling land-snail diversities because it 
yields higher ratios of species to individuals. 
Practically, however, timed searches are the 
most expedient by far, yielding species at 6.6 
times the rate per person hour of either 
sieved sampling method. The degree of this 
advantage is surely an overestimate, be- 
cause of our labor-intensive method of wet- 
sieving then picking for all invertebrates; nev- 
ertheless, even if we could halve or quarter 
our litter-processing time, time searching 
would be 3.3 or 1.7 times as efficient. Timed 
searching also requires minimal equipment 
and minimal weight and volume of samples 
to transport (critical factors in expeditions 
that require extensive backpacking). 

Nevertheless, our method of timed search- 
ing yielded fewer than three-fourths of the 
total species collected per plot. A more thor- 
ough sampling strategy must, therefore, in- 
clude some litter or soil-plus-litter sampling. 
Both these methods were roughly equivalent 
in their species richness and number of spec- 
imens per person-hour of effort. There were 
different advantages to each. Litter samples 
were 50% faster to process and yielded more 
live-represented species, whereas soil-plus- 
litter samples collected burrowing species 



that were otherwise missed. Thus a sample 
of litter-plus-soil seems preferable. 

Thus, for greatest efficiency in assessing 
species number and obtaining live speci- 
mens, a good strategy seems to be collect- 
ing litter-plus-soil samples during timed 
searches, taking them from places that are 
yielding good numbers of micro-molluscs. To 
be quantifiable, samples should be taken to a 
constant or measurable volume. 

Because only minute, cryptic or burrowing 
species were missed by timed searching, be- 
cause the 1.2-mm sieve passed both adults 
and identifiable juveniles, and because 1.0 
was the smallest adult dimension we en- 
countered, we recommend in processing the 
supplemental litter-plus-soil samples that the 
5.5-mm sieve fractions be discarded, and 
that both the > 1.2-mm and the > 0.85-mm 
fractions be retained and picked for micro- 
molluscs. Only a few of the minutest juveniles 
will be missed, at least for these Madagascar 
samples. Because wet-sieving is very labor- 
intensive (Table 2) and logistically difficult, we 
recommend dry-sieving, either on-site when 
litter and soil are dry enough, or later when 
the samples have been stored in, for exam- 
ple, muslin bags long enough to dry suffi- 
ciently without dehydrating slugs and semi- 
slugs. 

Macro-molluscs (young juveniles > 5 mm) 
tend to comprise only a small part of the 
Madagascan rain-forest land-snail fauna, in 
this case 7% (6/88) of the species. Also, 
macro-snails have been the most extensively 
collected in the past (Emberton, 1995b), so 
are least likely to yield new biogeographic or 
systematic information. Therefore, for great- 
est efficiency in sampling total species rich- 
ness, we recommend emphasizing the col- 
lection of micro-snails, and collecting macro- 
snails only as they are encountered during 
micro-snail searches. 

Thus, in sum, timed searches for micro- 
snails, incidentally collecting macro-snails 
and litter-plus-soil for dry-sieving and picking 
the > 1 .2-mm and > 0.85-mm fractions, seem 
best for quantitatively sampling Madagascan 
rainforest land-snails. This strategy should 
be transferable, with local modifications, to 
other tropical rainforests. 

ACKNOWLEDGMENTS 

We are grateful to the U.S. National Sci- 
ence Foundation and USAID for funding 
(grant DEB-9201060 to KCE); to the Mada- 



212 



EMBERTON, PEARCE & RANDALANA 



gasear Département des Eaux et Forets 
(DEF) for permission to collect and export 
specimens; to the staff of Ranomafana Na- 
tional Park Project (RNPP) (especially P. 
Wright, B. Andriamahaja, L. Robinson, and 
Madame Aimée) for logistic support; to Rich- 
ard laly, Edmond of Ambatolahy, and our 
other collectors and guides from the villages 
of Esetra, Mahialambo, and Malio for their 
superb work and their stamina under trying 
conditions; to Gervais of RNPP for his amaz- 
ing feats of driving; to the officials and other 
residents of Esetra, Mahialambo, and Malio 
for their very generous logistic support; to our 
sorting assistants in Fort Dauphin, Ranoma- 
fana National Park, and Antananarivo for their 
careful labor; to M. Fenn (World Wide Fund 
for Nature, Fort Dauphin) for recommending 
Malio and its forests; to the DEF chef. Fort 
Dauphin, for recommending Esetra Forest; 
and to С Hesterman (MB!) for help with data 
analysis. 



LITERATURE CITED 

ABBOTT, R. T. & K. J. BOSS, eds., 1989, A clas- 
sification of thie living Mollusca. American Mala- 
cologists Inc., Melbourne, Florida. 

BURCH, J. B. & T. A. PEARCE, 1990, Gastropoda. 
Pp. 201-309, in: D. R. DINDALL, ed., So/7 biology. 
New York: Wiley Interscience. 

CODDINGTON, J. A., С E. GRISWOLD, D. S. DA- 
VILA, E. PEÑARANDA & S. F. LARCHER, 1991, 
Designing and testing sampling protocols to es- 
timate biodiversity in tropical ecosystems. Pp. 
44-60, in: E. С DUDLEY, ed., The unity of evolu- 
tionary biology: Proceedings of the Fourth Inter- 
national Congress of Systennatic and Evolution- 
ary Biology. Portland, Oregon, Dioscorides. 

DEWINTER, A. J., 1995, Gastropod diversity in a 
rain forest in Gabon, West Africa. Pp. 223-228, 

in: A. с VAN BRUGGEN, 8. M. WELLS & TH. С M. KEM- 

PERMAN, eds.. Biodiversity and conservation of 

the l\Aollusca. Oegstgeest-Leiden, the Nettier- 

lands, Backtiuys Publishiers. 
EMBERTON, K. C, 1994, Thiirty new species of 

Madagascan land snails. Proceedings of the 

Academy of Natural Sciences of Philadelphia, 

145: 147-189. 
EMBERTON, K. C, 1995a, Land-snail community 

morphiologies of the highest-diversity sites of 



Madagascar, North America, and New Zealand, 
with recommended alternatives to height-diam- 
eter plots. Malacologia, 36: 43-66. 
EMBERTON, K. C, 1995b, On the endangered bio- 
diversity of Madagascan land snails. Pp. 69-89, 

in: A. с VAN BRUGGEN, 8^ M. WELLS & TH. С M. KEM- 

PERMAN, eds., Biodiversity and conservation of 
the Mollusca. Oegstgeest-Leiden, the Nether- 
lands, Backhuys Publishers. 

EMBERTON, K. C, 1996, Conservation phorities 
for forest-floor invertebrates of the southeastern 
half of Madagascar: evidence from two land- 
snail clades. Biodiversity and Conservation, 5: 
729-741. 

EMBERTON, K. C, in review, Diversities, distribu- 
tions, and abundances of 80 species of minute- 
sized land snails in southeastern-most Mada- 
gascan rainforests, with a report that lowlands 
are richer than highlands in endemic and rare 
species. Submitted to Biodiversity and Conser- 
vation. 

EMBERTON, K. C, & R. ARIJAONA. in press, Land 
snails. In: p. wright, ed., Ranomafana National 
Park project: a case study of conservation in 
Madagascar Chicago, University of Chicago 
Press. 

EMBERTON, K. C, T A. PEARCE & R. RAN- 
DALANA, in review, Richer molluscan diversity in 
the unprotected Vohimena than in the protected 
Anosy Mountain Chain, southeast Madagascar. 
Submitted to Biological Conservation. 

NORDSIECK, H., 1986, The system of the Stylom- 
matophora (Gastropoda), with special regard to 
the systematic position of the Clausiliidae, II. Im- 
portance of the shell and distribution. Archiv für 
Molluskenkunde, 1 1 7: 93-1 1 6. 

SOLEM, A., 1984, A world model of land snail di- 
versity and abundance. Pp. 6-22, in: A. solem a 
& A. с VAN BRUGGEN, eds., World-wide snails: 
Biogeographical studies on non-marine Mol- 
lusca. Leiden, E.J. Brill/Dr. W. Backhuys. 

TATTERSFIELD, P., 1994, Patterns of mollusc di- 
versity in Kakamega Forest, Kenya. Bulletin of 
the British Ecological Society, July, 1990: 172- 
176. 

THOMPSON, F. G., 1995, New and little known 
land snails of the family Spiraxidae from Central 
America and Mexico (Gastropoda, Pulmonata). 
Bulletin of the Florida Museum of Natural History, 
Biological Sciences, 39(2): 45-85. 

WILKINSON, L, 1990, SYSTAT: The System for 
Statistics. Evanston, Illinois, SYSTAT, Inc. 



Revised Ms. accepted 28 January 1996 



MAlJ\COLOGIA, 1996, 38(1-2): 213-221 

SEQUENCING METHODOLOGY AND PHYLOGENETIC ANALYSIS: CYTOCHROME 

b GENE SEQUENCE REVEALS SIGNIFICANT DIVERSITY IN CHINESE 

POPULATIONS OF ONCOMELANIA (GASTROPODA: POMATIOPSIDAE) 

Christina M. Spolsky,^ George M. Davis^ & Zhang Yi^ 

ABSTRACT 

The evolution of the snail species Oncomelania hupensis. a vector for transmission of Schis- 
tosoma japonicum in Asia, is tightly linked to the evolution of its parasite. We report here on 
studies of the evolution of O. hupensis on the mainland of China, using sequence divergence 
of the mitochondrial cytochrome b gene. The cytochrome b gene was amplified by PCR, cloned 
in pBluescript, and sequenced; these methods are described in detail. The sequences for three 
populations of two subspecies of this prosobranch gastropod were aligned and used in re- 
constructing phylogenetic trees. The phylogenetic analyses confirm the divergence of Chinese 
Oncomelania into subspecies and provide a finer tool for further genetic discrimination. Com- 
parison of Oncomelania cytochrome b sequences to published sequences for two pulmonale 
gastropods shows a greatly increased divergence rate in the pulmonates relative to that of 
Oncomelania and of other metazoan groups. 

Key words: Oncomelania. sequencing methodology, phylogenetics, PCR, cloning, cy- 
tochrome b, infraspecific diversity 



INTRODUCTION 

The rissoacean genus Oncomelania is of 
particular importance to the field of tropical 
medicine because one of its two species, the 
polytypic Oncomelania hupensis, is involved 
in the transmission of Schistosoma japoni- 
cum in Asia. 

Oncomelania minima is restricted to north- 
western Honshu, Japan. Polytypic O. hupen- 
sis, in cont'-ast, is distributed from northern 
Burma (fossil) throughout southern China, 
Japan, the Philippines and Sulawesi. The 
polytypic status of Oncomelania has been 
reviewed (Davis, 1994) with the following 
subspecies recognized: O. h. hupensis (China 
mainland): O. h. formosana and O. h. chlul 
(Taiwan): O. h. nosophora (Japan): O. h. qua- 
ciras/ (Philippines): O. h. Ilndoensis (Sulawesi). 
More recently, Davis et al. (1995) recognized 
three subspecies on the mainland of China on 
the basis of allozyme molecular genetics, 
shell morphology and biogeography: O. h. hu- 
pensis; O. h. robertsoni and O. h. tangl. 

A coevolved relationship between snail lin- 
eages and the genus Schistosoma extends 
back to the Gondwanaland origin of these 
taxa. Davis (1992), in reviewing the patterns 



and processes of this coevolution, made the 
point that transmission of the parasite now is 
population-specific in many instances. The 
hypothesis is that as populations of Oncomel- 
ania have dispersed and diversified in the di- 
rection from Burma-Yunnan, China, through- 
out China to Japan and the Philippines (Davis, 
1979), the parasite has had to modify genet- 
ically with the genetically changing snail pop- 
ulations or become regionally extinct. This hy- 
pothesis predicts that genetic distances 
among parasite populations parallel genetic 
distances among snail populations. 

Allozyme electrophoretic data demonstrate 
strong population divergence among popu- 
lations of Oncomelania throughout China 
(Davis et al., 1 995). The problem with the elec- 
trophoretic approach, however, is that as one 
increases the number of populations com- 
pared, errors in assigning the homology of 
alleles increase. Because one must always 
run a control population as a standard for 
determining the identity of alleles at each of 30 
or more loci, one needs exponentially increas- 
ing numbers of cross-comparisons among 
populations; the experimental labor and need 
to keep many fresh frozen populations be- 
comes prohibitive. 



^The Academy of Natural Sciences of Philadelphia, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania, USA 
19103 

^Chinese National Center of Systematic Medical Malacology, Chinese Academy of Preventive Medicine, Shanghai, Peo- 
ple's Republic of China 



213 



214 



SPOLSKY, DAVIS & Yl 



TABLE 1. Localities and collecting information for three populations of Oncomelania hupensis in China. 
Latitudes and longitudes are given. Catalog numbers given are for the Chinese Institute of Parasitic 
Diseases (CIPD) and the Academy of Natural Sciences of Philadelphia (ANSP). 

1. Sichuan (SC): Sichuan Province; TianQian County; Xing Hua District; Xia Len Village 2nd group. 

102 46.0'E; 30 5.02'N; CIPD 0338 

2. Yunnan (DA): Yunnan Province; Dali City; Da Jin Ping Zi Ran Village. Ditch. 

100 12.4'E; 25 27.6'N; CIPD 0349 

3. JiangXi (JX): JiangXi Province, Pengze County, WangLing District, JingWang Village. 

116 30.0'E; 29 55.0'N; ANSP 399275 

collected 7 Dec. 1993 by Zhang Jian Guo and Quo Gang Qiang 



Gene sequencing provides an alternative 
method for reconstructing and evaluating 
phylogenetic relationships among a group of 
organisms. Once a gene is sequenced, that 
sequence is permanently available for com- 
parisons, and does not need to be repeated 
as taxa are added to the analysis. We have 
looked for a gene that evolves rapidly enough 
to distinguish populations of Oncomelania 
to the same as or a greater degree than allo- 
zymes do. We have not used RFLD (restric- 
tion fragment length differences) methods 
because they are too crude and because 
they also rely on cross-comparisons of elec- 
trophoretic mobility. We have not used mi- 
crosatellites because they give too fine a res- 
olution: they provide excellent discrimination 
at the population level and below, but are too 
sensitive a tool for inter-population and inter- 
specific comparisons. Sequences from the 
appropriate gene potentially provide the 
most powerful set of discrete character data 
for phylogenetic analysis. 

This paper presents detailed methods and 
preliminary results of amplifying and se- 
quencing the mitochondrial cytochrome b 
gene from Oncomelania hupensis, and com- 
paring the results of a three-taxon analysis 
with results from the allozyme study by Davis 
et al. (1995). Problems encountered in PCR 
amplification of cytochrome b from On- 
comelania are stressed. Numerous system- 
atic studies of cytochrome b gene sequence 
document that the rate of evolution of this 
gene is appropriate to demonstrate differ- 
ences among species and infraspecific taxa. 
Sufficient phylogenetically informative char- 
acters are present in such comparisons so 
that phylogenetic analyses are very robust. In 
contrast to hbosomal RNA genes, there are 
few or no insertions and deletions among cy- 
tochrome b genes, and thus no problems 
with alignment even among taxa as divergent 
as molluscs, insects, and mammals. Be- 
cause correct alignment is essential to ob- 



taining the correct phylogeny (Thorne & Kish- 
ino, 1992), comparison of genes coding for 
proteins allows one to reconstruct phyloge- 
nies that are more likely to reflect true evolu- 
tionary relationships for those genes. 



MATERIALS AND METHODS 

Specimens Studied 

Snails were collected from three localities 
in China: in Sichuan and Yunnan Provinces in 
northwestern China, and in JiangXi Province 
in the east (Table 1). Snails were brought to 
the United States alive in an estivating state. 
Once in the laboratory, snails were activated 
by placing them on moist filter paper in Petri 
dishes and kept at 4-C. Immediately prior to 
isolation of DNA, the snails were quick-frozen 
at -80 С by placing them individually in the 
wells of a ceramic depression plate previ- 
ously chilled to -80C. 

DNA Preparation 

The methods used for preparing DNA from 
individual snails were modified from those of 
Spolsky & Uzzell (1984, 1986) and of Doyle & 
Doyle (1987). Briefly, a frozen snail (4-9 mm 
shell length) was crushed, the whole individ- 
ual immediately dropped into 600 ul of lysis 
buffer (0.02 M Tris, 0.1 M EDTA, 0.5% Sar- 
kosyl) containing 200 ug/ml proteinase K, 
and incubated at 55 С overnight. One hun- 
dred ul of each of 5 M NaCI and СТАВ ex- 
traction solution (5% СТАВ, 0.5 M NaCI) 
were added, and the resulting solution ex- 
tracted with an equal volume of chloroform. 
800 ul of СТАВ precipitation buffer (1% 
СТАВ, 50 mM Tris pH 8.0, 10 тМ EDTA) 
were added to the aqueous phase, mixed, 
and placed at room temperature for 30 min. 
The CTAB-DNA precipitate was pelleted (15 
min at 14K rpm), redissolved in NTE (1.2M 



CYTOCHROME b DIVERSITY IN CHINESE ONCOMELANIA 



215 



NaCI, 10 mM Tris, 1 mM EDTA) containing 
100 ug/ml RNase, and again precipitated by 
addition of two volumes of ethanol. The DNA 
pellet was washed with 70% ethanol in TE, 
then redissolved in 100-200 ul of water or 
0.1 X TE. Aliquots (3 ul) of each DNA were 
subjected to electrophoresis through a 0.8% 
agarose gel in TBE and stained with ethidium 
bromide to obtain a rough estimate of the 
DNA concentration and quality. The amount 
of DNA was more precisely quantified using a 
Hoefer TKO100 fluorometer. Concentration 
of each DNA preparation was adjusted to 50- 
100 ng/ul. Using this protocol, we obtained 
between 5 and 65 ug of high molecular 
weight DNA per individual snail. 

DNA Amplification 

PCR was used to amplify the mitochondrial 
cytochrome b gene using the primer pair 
On5L (forward: 5'-CATTTAGGTCTGCGGTC- 
CAC) and ОпбН (reverse: 5'-GGCGTAAC- 
TAGTGGGTTAGCTGG). These Oncomela- 
n/a-specific primers define a fragment 61 bp 
in length. Preliminary sequence for O. hupen- 
sis from Sichuan was obtained using mollus- 
can phmers SUP1 and SUP2 (a gift from T. 
Collins). These primers, although not optimal, 
provided sufficient sequence to enable de- 
sign of On5L and ОпбН. Optimal sequence 
for the latter primers was determined using 
PRIMER version 0.5 (Lincoln et al., 1991) and 
was based on the preliminary Oncomelania 
sequence in combination with comparisons 
of conserved cytochrome b regions for a 
number of molluscan, echinoderm, and ver- 
tebrate taxa. Each PCR reaction contained 
approximately 50-100 ng of template DNA, 
200 цМ of each dNTP, 30 pmole of each 
primer, and two units of Taq polymerase 
(Promega), in 50 ul of supplier-provided 
buffer at a magnesium concentration of 2.5 
mM. The PCR conditions consisted of 40 cy- 
cles of denaturation at 94-C for 45 sec, an- 
nealing at 43'"C for 1 min, and extension at 
72 С for 1 min 20 sec on an M-J Research 
model PTC-100 thermal controller. 



was used for ligation into the polycloning re- 
gion of the Plasmid vector pBluescript SK 
(Stratagene), previously prepared for ligation 
by linearizing with EcoRV and ddT-tailing us- 
ing a modification of Holten & Graham's 
(1 991 ) method. For the latter protocol, the lin- 
earized vector was incubated with ddTTP 
and terminal transferase at 37'C for one 
hour; 20 ng of this ddT-tailed vector was 
used per ligation. Each ligation reaction con- 
tained, in addition to PCR product and pre- 
pared vector, 4% polyethylene glycol 8000 
and 0.2-0.5 unit of T4 DNA ligase (Promega) 
in the appropriate buffer. Ligations were al- 
lowed to proceed overnight at 15"C, then 
drop-dialyzed by placing each reaction on a 
Millipore type VS25 membrane floating in a 
Petri dish on 0.1 x ТЕ. One third of a ligation 
was used for transformation via electropora- 
tion (BioRad puiser) of the E. coli host cell line 
XL1 Blue. Bacterial colonies with recombinant 
Plasmids were identified by plating on selec- 
tive Luria agar plates containing 100 mg/ml 
ampicillin, 40 ug/ml X-gal, and 40 ug/ml 
IPTG. Putative positive colonies were grown 
overnight at 37^0 in 2 ml of LB + ampicillin. 
Minipreps (Sambrook et al., 1989) of these 
growths were screened for the presence of 
inserts of the correct size by cutting the insert 
out of the recombinant plasmid with Hindlll 
and PstI, followed by electrophoretic analy- 
sis. Confirmed positive colonies were grown 
in larger scale liquid cultures (15 ml), and 
plasmid isolated from them on Qiagen tip- 
100 columns following the manufacturer's 
protocol. Sequences of the cloned fragments 
were determined by automated cycle se- 
quencing on an ABI 373A sequencer with 
Stretch upgrade, using commercially avail- 
able vector primers T3 and T7. Using the au- 
tomated sequencer, these two primers pro- 
vide completely overlapping sequence for 
each strand of the 61 bp cytochrome b frag- 
ment. To prevent incorrect nucleotide calls 
caused by occasional random misincorpora- 
tion of nucleotides during amplification, at 
least three clones of each ligation were se- 
quenced. 



Cloning, Screening and Sequencing 



Data Analyses 



Amplified DNA products were separated 
on a 1% agarose gel. Bands corresponding 
to fragments of the correct size were cut out, 
purified using Geneclean (Bio 101) glass 
beads, and quantified by fluorometry. A 
50-ng aliquot of the purified PCR product 



Sequences for each individual were as- 
sembled by visual inspection using the se- 
quence editor ESEE version 1.09e (Cabot & 
Beckenbach, 1989). ESEE was also used to 
align Oncomelania sequences with each 
other and with cytochrome b sequences 



216 



SPOLSKY, DAVIS & Yl 



available from Genbank. Aligned sequences 
were formatted appropriately for phyloge- 
netic analyses using EAT (Cabot, 1993). Pair- 
wise maximum-likelihood distances were 
calculated using program DNADIST of the 
phylogenetic analysis package PHYLIP ver- 
sion 3.57 (Felsenstein, 1989, 1993); these 
distances were estimated under the Felsen- 
stein maximum-likelihood model, which 
takes into consideration unequal frequencies 
of nucleotides, unequal rates of transitions 
and transversions, and multiple substitutions 
at individual sites. Distance, parsimony, and 
maximum-likelihood trees were calculated 
using programs FITCH, DNAPENNY, and 
DNAML of PHYLIP, programs that do not as- 
sume equal rates of change along the 
branches of a tree. Optimal FITCH and 
DNAML trees were found by running 20 rep- 
etitions of each program with randomized in- 
put order and optimization by global branch 
rearrangement. Bootstrap and delete-half- 
jackknife estimates (1,000 replicates) of con- 
fidence intervals for the maximum-likelihood 
analyses were made using program SEQ- 
BOOT, in conjunction with DNAML and CON- 
SENSE. 



RESULTS 

Relationships Among the Populations of 
Oncomelania hupensis 

Figure 1 presents aligned cytochrome b 
sequences for individual O. hupensis from 
three populations in China (Sichuan, Yunnan, 
and JiangXi); each sequence represents the 
consensus from at least three clones for that 
individual. Sequences were obtained for two 
individuals from the locality in Sichuan; diver- 
gence between the two specimens was less 
than 0.4% (two site differences). This low in- 
trapopulation divergence is consistent with 
low intrapopulationat variability in morphol- 
ogy as well as in other genetic measures. 
Given the low intrapopulation relative to in- 
terpopulation variability, we concentrated in 
this preliminary study on obtaining measures 
of cytochrome b divergence among, rather 
than within, populations. Oncomelania se- 
quences were also aligned with published 
sequences for the gastropods Albinaha co- 
erulea and Cepaea nemoralis. Nucleotides 
corresponding to the primer regions have 
been trimmed from the sequences, resulting 
in alignment of 572 nucleotides. In this region 



of cytochrome b, Cepaea and Albinaria share 
an extra nucleotide triplet at positions 109- 
111; in addition, Cepaea alone has an inser- 
tion of two nucleotides at positions 521-522. 
For the phylogenetic analyses, all nucleotide 
positions were included. The three O. hupen- 
sis sequences have relatively few changes 
among them; a total of 70 variable sites were 
detected in this region of the cytochrome b. 
In contrast, Albinaria and Cepaea each differs 
from Oncomelania at numerous nucleotide 
positions. Sequence divergence estimates 
are given in Table 2. Divergences between 
the closest pair, Sichuan and Yunnan, are 
3.8%, whereas between either of these and 
JiangXi, distances are 10.2 and 11.9% re- 
spectively. Distances between Oncomelania 
and Albinaria average 53%, between On- 
comelania and Cepaea, 58.8%. 

Phylogenetic Analyses 

For the phylogenetic analyses, the three 
Oncomelania hupensis populations were 
compared to Albinaria caerulea, Cepaea ne- 
moralis, the sea urchin Strongylocentrotus 
purpuratus, and the mammal Homo sapiens. 
The stylommatophoran pulmonate gastro- 
pods Albinaria and Cepaea are the closest 
relatives to Oncomelania for which published 
cytochrome b sequence is available. Strongy- 
locentrotus and Homo were included in the 
analyses to provide rooting for the gastropod 
clade. The optimal transition/transversion ra- 
tio, that is, the ratio which minimizes the like- 
lihood measure, was determined empirically 
using DNAML (P. Beerii, pers. comm.); for the 
set of taxa used, the optimal ratio was 1.1. For 
comparison, phylogenies were also recon- 
structed using distances and parsimony. The 
topology of the phylogenetic tree obtained by 
each of the three methods was the same (Fig. 
2). The validity of each tree was tested by both 
bootstrap and jackknife resampling of the 
data. For both resampling methods with all 
tree building strategies, each node of the tree 
was strongly supported by high bootstrap val- 
ues (minimum of 95% confidence level). 

DISCUSSION 

Although cytochrome b sequencing has 
been used in numerous systematic studies in 
organisms ranging from vertebrates to ar- 
thropods to echinoderms, it was not a simple 
matter to apply those techniques to mol- 
luscs. Because of the very ancient branching 



CYTOCHROME b DIVERSITY IN CHINESE ONCOMELANIA 



217 



ATCTCTCGTG ATGTAAACTA 

T. . 

. .T T. . 

. . TAT CCTGG 

. . TATG . . С . . С. . . CCAGG 



TGGTTGACTT TTACGGGCAC 

A. . . . 

С A. . . . 

A . . A . . . T . . С . T . . TTT . . 
C..G..G... G.T..A...T- 



TGGGGCCAGC 

A. .T 

T. . . 

TCT 

. . . с . . ATCG 



TATTGGCGTA ATTCTTTTAT TTATGACTAT GGGGACAGCT 



AA. . .ATA. . 
AA.A.ATA. . 

TTTCT.AGGGT 
. .C A. 



ATATACCACC 

T. 

ТА . . . 

. .ce. T. CA. 
T . GC . G . CA . 

ACGTTTTACC 
... .С 



СТТ CT 

ATG 

ACACATGAAA 



GTGT T 

GG CT 

GTGAGGTCAG 



GG. А 
CG. А 



. CAA . . . . T . 
. .ACGA . . . T . 



. . . .A А С 

. GG . T . GC . . . . CT . . G . . . 
. .AC . A . G ce . . T . . A 



. С . . GC . CA 



TTGAATGGGT TTGAGGTGGT TTTGCTGTTG ATAATGCAAC 
A 



.A 

.A. .CG. G. . 
GT . . . G . . . . 

ATTAACTCGA 



. T . . AC . T . . 
.T. .C 

GTATGTGGGA 

T 

T. . . 

с . .T.T. .c 
A. .CT. A. .T 

TTTTTTACTC 

R.C. 

. .C 

CT. A. 

... .A. T. .T 



A A 

A. .G A 

T A. .A 

AAAATATTAG 
С . . 



CCC.G.A.G. 
G . T . GGC . С . 

TTCATTTTGT 



.C. 

.A. 



.T. 

.T. 



Sichuan ATCTCTCGTG ATGTAAACTA TGGTTGACTT TTACGGGCAC TTCATGCAAA TGGGGCCAGC TGATTTTTTA 

Yunnan 
JiangXi 
ALBINARIA 

CEPAEA . . TATG . . С . . С . . . CCAGG С . . G . . G . . . G . T . . A . . . T- CC 

Sichuan TCTGCATTTA TTTTCATATT GGGCGGGGGA TGTATTAT-- -GGGTCATTT 

Yunnan T -- -..A 

JiangXi с ..A.. A A -- 

ALBINARIA . G . TT . . G . . . GCC С . . A . . T . . AC . A . . С . . . CA 

CEPAEA . GCTT . . G . . . GCC T . . T . . AG . A . . С . . . CA 

Sichuan 
Yunnan 
JiangXi 
ALBINARIA 

CEPAEA CG . A . . G . GT . .ACGA . . . T . . .AC . A . G CC . . T . . A T . . С 

Sichuan ATATCTTTCT ggggtgcc.ac tgtaattact aatttattat cggctattcc 

Yunnan 

JiangXi T. .A G. 

ALBINARIA A. .T. .A T. 

CEPAEA AT. .A. .C. .T. 

Sichuan 
Yunnan 

JiangXi G . . A 

ALBINARIA A G..A..G ..CT GGC . . . . T . . CC.G.A. 

CEPAEA .GAC...A.. G..G..CT T....AA ..CA.. С A. 

Sichuan ACTACCTTTC ggcattgcag cattgactgt attacatttg ctattcctac acgagtccgg ctcaaataac 

Yunnan A a .... 

JiangXi GT A CA T. . . 

ALBINARIA . T . . . . с . T CTT . . . AGG . GG . . AG . AC. . С T . . . A . T A . T . 

CEPAEA G . . с . с . . . ТА . . . TT . . T . . . TGTGT . .G.C.C T. 

Sichuan 
Yunnan 

JiangXi с. 

ALBINARIA 
CEPAEA 

Sichuan 

Yunnan - - 

JiangXi G..C..T. --....G... T 

ALBINARIA T.A.. AG..T..GG. G.T..G.T.A --..A. CAC . T . T . . CAGC . 

CEPAEA TG. . .CAT. GCCTG.... T ATG TT . TG . T . . G . . . . A . CAC . 

Sichuan GAAAATTTCA TT 

Yunnan 

JiangXi T. 

ALBINARIA A.. .. 

CEPAEA С .TT A. 

FIG. 1. Alignment of sequences of a 572 nucleotide fragment of cytochrome b from three populations of 
Oncomelania hupensis in China and from two pulmonate gastropods. The top line contains sequence for 
cytochrome b of Oncomelania hupensis from Sichuan Province, China. For the remaining sequences (O. 
hupensis from Yunnan and JiangXi Provinces, Albinana coerulea. Cepaea nemoralls), nucleotides were 

given only for sites that differed from the Sichuan sequence. A dot (.) indicates the presence of the same 
nucleotide as in the top sequence; a dash (-) indicates the absence of a nucleotide at that position. 
Sequences have been deposited with Genbank. 



. T . . . . с . T CTT . . . AGG . GG . . AG . AC. . С T . . . A . T 

G . . С . . С . . . ТА . . . TT . . T . . . TGTGT . .G.C.C... 

CCATTAGGAT TAAATAGAGA CGGAGAAAAG GTACCATTTC 

. . .e.G. .G. . . .T G 



ACGAGTCCGG 
.T 

.T. .A. .T. . 
. T . . TAAG . . 
.T. .CAAA. . 



A 

G . . . TC 
T . . GTC . 



.чТТеТТ.АТТ.А CAGCTTTAAA GATTTAGTTG 



. TA ATTTATTTC . 
, .A AT.TGTCCC 



T A 

. TT . AGG . . A AA . . . . 
T . TGTC ... A . . CAG . 



. .G. .G 

GATTTATTAT TTTACTTTTT ATAC ГТТСАТ 



.ce. 
.ce. 



. . T G G. 

. T . . C.AA AGG . . . . 

. T . . CA . GG . . G AT . . . GTGG . 



-TACTAGATA CTATTTGCAC CCCAA-^iTACT AAC.AGACCCA 



.T G. . 

. TTCTT. . . . 
.TA. T. . .T. 



G. . . . 

.CT. . 
T. .G. 



.G 
.C 
.C 



of molluscs from the basal phylogeriy, their 
cytochrome b sequences are divergent 
enough so that the so-called "universal prim- 
ers" for cytochrome b (Kocher et al., 1989) 
do not amplify this gene from molluscs. In 
fact, there is enough divergence within the 
molluscs to make designing a universal 
primer for Mollusca difficult. The PCR prim- 
ers we designed, OnSL and ОпбН, work for 



most, but not all, populations of Chinese On- 
comelania, and variably well for the related 
Tricula. Because of the variable yield and pu- 
rity of the product from different taxa, we 
cloned all amplified products prior to se- 
quencing. Although more labor-intensive, 
this procedure provided such excellent qual- 
ity sequence that in the long run time was 
saved by not having to do multiple repetitions 



218 




SPOLSKY, DAVIS & Yl 

Strong ylocen tro tus 
/ Yunnan 

97 

100 ^ Sichuan 
JiangXi 

100: 



Cepaea 
A I bin aria 



Homo 



FIG. 2. A maximum-likelihood tree of relationships among three populations of the mesogastropod proso- 
branch Oncomelania hupensis and the stylommatophoran pulmonates Alblnaria caerulea and Cepaea 
nemoralis. Both Homo sapiens and Strongylocentrotus purpuratus were used as outgroups for the analysis 
in order to provide rooting for the gastropod node and allow for calculation of a bootstrapping value for the 
node. Bootstrap values are listed to the left of each node; these indicate the number of times that node 
occurred among 1,000 bootstrap replicates. 



and by the clarity and thus accuracy of the 
sequence obtained. 

The three populations of Oncomelania hu- 
pensis form, as expected from divergence 
levels, a very closely related monophyletic 
group. Previous conchological, biogeo- 
graphic, and electrophoretic analyses (Davis 
et al., 1995) supporl division of O. hupensis 
into three subspecies: the Sichuan and Yun- 
nan populations belong to the subspecies O. 
h. robertsoni, whereas the JiangXi population 
belongs to the subspecies O. h. hupensis. 
The phylogenetic analyses of nucleotide se- 
quence of cytochrome b presented here are 
consistent with this subspecies concept: the 
Sichuan and Yunnan populations are the 
most closely related (Fig. 2, Table 2). These 
populations are geographically close, and 
have smooth shells with no varix. Distances 
of either of these populations to the JiangXi 
population, ribbed and with a strong varix, 
are almost threefold more. 

The two pulmonate gastropods, Albinaha 
and Cepaea, form a separate group that di- 
verged over 300 million years ago from the 
prosobranch Oncomelania. Both gastropod 
groups, Oncomelania and pulmonates, share 
a common node relative to non-molluscan 
taxa. One striking feature of the phylogenetic 
tree is the very long branch lengths for the 
pulmonates, particularly for Cepaea (Fig. 2), 
suggesting many more changes along the 
branches leading to Albinaha and Cepaea 



than along any other branches. This is also 
evident in pairwise distance comparisons 
(Table 2). In comparisons using either Ce- 
paea or Albinaha, results are what one would 
expect: distances to Oncomelania are less 
than to Homo or Strongylocentrotus; on the 
other hand, using Oncomelania, pairwise dis- 
tances to Homo and to Strongylocentrotus 
are less than to the more closely related gas- 
tropod taxon Cepaea. This is not a phenom- 
enon restricted to the cytochrome b gene: 
phylogenetic analyses of cytochrome oxi- 
dase (Hoeh et al., 1996) have also demon- 
strated a longer branch length for Albinaha. 
In agreement with the large divergences we 
find for the pulmonates, the mitochondrial 
gene order for these two taxa also appears to 
have changed extensively, both from that of 
other molluscs as well as from other meta- 
zoan groups (Lecanidou et al., 1994; Hatzo- 
glou et al., 1995). An increased rate of evo- 
lution of the mitochondrial genome and of its 
gene order has also been observed in the 
bivalve mollusc Mytlius edulis (Hoffmann et 
al., 1992; Hoeh et al., 1996), but not in the 
polyplacophoran mollusc Kathahna tunicata 
(Boore & Brown, 1994). 

The increased cytochrome Ь divergence 
in pulmonates is puzzling. Some of the extra 
length of the pulmonate branches may pos- 
sibly be a result of frameshift-causing mis- 
readings of the nucleotide sequence. For 
example, if we translate the nucleotide se- 



CYTOCHROME b DIVERSITY IN CHINESE ONCOMELANIA 



219 



TABLE 2. Pairwise comparisons of sequence divergence over 572 nucleotide positions in the 
cytochrome b gene. Sequence divergences were estimated using the program DNADIST of PHYLIP 
version 3.57, under Felsenstein's (1989) maximum likelihood method. 



Y 



JX 



CEP 



ALB 



HOMO 



STR 



Sichuan 


— 










Yunnan 


0.0382 


— 








JiangXi 


0.1021 


0.1186 


— 






CEPAEA 


0.5969 


0.5853 


0.5899 


— 




ALBINARIA 


0.5364 


0.5427 


0.5256 


0.5148 


— 


HOMO 


0.5158 


0.5032 


0.5067 


0.7180 


0.6190 


STRONGYLO 


0.5409 


0.5304 


0.5324 


0.7225 


0.6177 



0.4663 



quence of Cepaea downstream from the 
point of the two-nucleotide insertion (Fig. 1), 
and compare it to the Sichuan Oncomelania 
sequence, we find virtually no homology 
among the 14 amino acids coded (Fig. 3); 
however, if we delete these two nucleotides 
from the Cepaea sequence, homology of the 
translated sequences becomes significantly 
higher. Possible sequence misreadings re- 
sulting in sequential insertions and deletions 
relative to other sequences may be difficult to 
detect if the sequence eventually goes back 
into alignment. If such shifts are not allowed 
for, however, this would result in significantly 
longer pairwise distances between taxa. Al- 
ternatively, if the apparent frameshift muta- 
tions reflect the real sequence rather than 
misreadings of the sequence, then the mito- 
chondrial cytochrome b sequences in these 
two cases appear to behave as if they were 
nuclear pseudogenes (cf. Collura & Stewart, 
1995). Because these sequences were ob- 
tained from cloned mitochondrial genomes, it 
is unlikely that these do represent nuclear 
pseudogenes. In this case, the apparent in- 
creased rate of evolution of the mitochondrial 
cytochrome b in the pulmonates suggests a 
decrease in functional constraints for this 
protein. 

Some workers have avoided the apparent 
high labor costs of gene sequencing by using 
RFLD^ analyses instead. This technique, 
however, has many of the same problems as 



^Although the abbreviation RFLP is often used to refer to 
restriction fragment analysis of sequence variation, we 
consider this inappropriate usage of the term polymor- 
phism as applied to genetic analysis. Genetic polymor- 
phism has a very specific meaning (cf. Ford, 1965) involv- 
ing discontinuous variation, that is, distinct alleles, of 
specific genes (genetic loci). In the restriction fragment 
analysis method commonly termed RFLP, fragment mo- 
bility differences cannot be assigned to specific loci. We 
therefore intentionally use the term RFLD (resiriction frag- 
ment length difference) for this method. 



does genetic analysis of allozyme mobility 
differences, and it often provides even less 
information than allozyme analysis does. Two 
additional serious weaknesses of RFLD anal- 
ysis, not present in allozyme studies, are: (1) 
the inability to identify genetic loci and there- 
fore to know what is allelic to what; because 
of this, one does not know which characters 
are independent, a requirement for phyloge- 
netic analysis. One therefore can at best get 
only a very crude measure of similarity; (2) the 
paucity of characters on which to base a phy- 
iogenetic analysis. A valid analysis requires a 
larger number of variable characters than the 
number of taxa; this is not the case for many 
RFLD studies. These weaknesses probably 
account for the discrepancy between the 
conclusions of Hope & McManus (1995) 
based on RFLD analyses and our conclu- 
sions based on phylogenetic analyses of 
cytochrome b gene sequence: among popu- 
lations from a number of Oncomelania sub- 
species from China, the Philippines, and Ja- 
pan, the largest difference Hope & McManus 
(1995) found is between the Sichuan and 
Yunnan populations of Chinese Oncomela- 
nia. Not only is this at variance with our elec- 
trophoretic and sequencing results, but it is 
also in conflict with conchological and bio- 
geographic data. Our conclusions, on the 
other hand, are strongly supported: there is 
strong correspondance between our se- 
quence data and biogeographic, concholog- 
ical, and extensive allozyme data. 

Ribosomal RNA sequencing has been 
used extensively in determining phylogenetic 
status among molluscs. In Oncomelania, se- 
quences from the D6 domain and the 5' ter- 
minus of 23S-like rRNA were useful in deter- 
mining the phylogenetic relationship of the 
genus to a number of molluscan taxa, but 
were not helpful in resolving finer differences 
at the species level (Emberton et al., 1990; 
Rosenberg et al., 1994). The present report 



220 



SPOLSKY, DAVIS & Yl 



Sichuan 
CEPAEA 
CEPAEA-2 



LVLFAPQMLTDPENF I 
LLCYITL . YLRTPKTF 
V . . YH . N . F Y 



FIG. 3. Alignment of translated sequences for a 
portion of the cytochrome b gene 3' from position 
523 of Fig. 1. The invertebrate genetic code was 
used for translation. Amino acids were given only 
for sites that differed from the Sichuan amino acid 
sequence; a dot indicates the presence of the 
same amino acid as in the Sichuan sequence. Ce- 
paea = amino acid sequence translated from nu- 
cleotide sequence stored in Genbank, accession 
U23045; Cepaea-2 = two "extra" nucleotides, at 
positions 521 and 522 of the amplified cytochrome 
b sequence, were removed prior to translation. 



confirms the utility of nucleotide sequencing 
of cytochrome b; this gene provides the res- 
olution necessary to determine intraspecific 
relationships among populations of On- 
comelania, and to cluster the populations ac- 
cording to subspecies status. Work is in 
progress, using cytochrome b gene se- 
quencing, on the degree and patterns of ge- 
netic divergence among Oncomelania hu- 
pensis, other species of Oncomelania, and 
Tricula. This work will serve to establish pat- 
terns of divergence, to substantiate the sub- 
species concepts for Oncomelania in China, 
to place the differentiation of Oncomelania in 
a phylogeographic context, and to corrobo- 
rate the existence of two diverging subfami- 
lies in the pomatiopsid assemblage, that is, 
the Pomatiopsinae {Oncomelania) and Tricu- 
linae {Tricula). 



ACKNOWLEDGEMENTS 

This work was supported in part by N.I.H. 
grant TMP AI 11 373 to Davis and, in part, by 
a TMPC grant 1 P50 AI39461-01. The sup- 
port of the Institute of Parasitic Diseases, 
Chinese Academy of Preventive Medicine, is 
gratefully acknowledged. We thank Thomas 
Uzzell and two anonymous reviewers for crit- 
ical comments on the manuscript. 



LITERATURE CITED 

BOORE, J. L. & W. M. BROWN, 1994, Complete 
DNA sequence of the mitochondrial genome of 
the black chiton, Katharina tunlcata. Genetics, 
138: 423-443. 



CABOT, E. L., 1993, EAT (Eyeball Analysis Tool- 
box). Distributed by the author, Simon Fraser 
University, Vancouver, Canada. 

CABOT, E. L & A. T. BECKENBACH, 1989. Com- 
puter Applications in the Biosciences. 5: 233- 
234. 

COLLURA, R. V. & С В. STEWART, 1995, Inser- 
tions and duplications of mtDNA in the nuclear 
genomes of Old World monkeys and hominoids. 
Nature, 378: 485-489. 

DAVIS, G. M., 1 979, The origin and evolution of the 
gastropod family Pomatiopsidae, with emphasis 
on the Mekong River Triculinae. Monograph of 
The Academy of Natural Sciences of Philadel- 
phia. 20: 1-120. 

DAVIS, G. M., 1992, Evolution of prosobranch 
snails transmitting Asian Schistosoma; coevolu- 
tion with Schistosoma: a review. Progress in 
Clinical Parasitology. 3: 145-204. 

DAVIS, G. M., 1994, Molecular genetics and taxo- 
nomic discrimination. The Nautilus, Supplement, 
2, 3-23. 

DAVIS, G. M., Z. Yl, G. Y. HUA & С. SPOLSKY, 
1 995, Population genetics and systematic status 
of Oncomelania hupensis (Gastropoda: Pomati- 
opsidae) throughout China. Malacologia. 37: 
133-156. 

DOYLE, J. J. & J. L. DOYLE, 1987, A rapid DNA 
isolation procedure for small amounts of fresh 
leaf tissue. Phytochemicai Bulletin, 19: 11-15. 

EMBERTON, K, C, G. S. KUNCIO, G. M. DAVIS, S. 
M. PHILLIPS, K. M. MONDEREWICZ & Y, H. 
GUO, 1990, Comparison of recent classifica- 
tions of stylommatophoran land-snail families, 
and evaluation of large-ribosomal RNA sequenc- 
ing for their phylogenetics. Malacologia. 31: 
327-352. 

FELSENSTEIN, J., 1989, PHYLIP— Phylogeny In- 
ference Package (Version 3.2). Cladistics. 5: 
164-166. 

FELSENSTEIN, J., 1993, PHYLIP (Phylogeny Infer- 
ence Package). Version 3.5c. Distributed by the 
author. Department of Genetics, University of 
Washington, Seattle, Washington, USA. 

FORD, E. В., 1965, Genetic polymorphism. 101 
pp., MIT Press. Cambridge, Massachusetts. 

HATZOGLOU, E., G. С RODAKIS & R. LECANI- 
DOU, 1995, Complete sequence and gene orga- 
nization of the mitochondrial genome of the land 
snail Albinaria caerulea. Genetics. 140: 1353- 
1366. 

HOEH, W. R., D. T. STEWART, B. W. SUTHER- 
LAND & E. ZOUROS, 1996, Cytochrome с oxi- 
dase sequence comparisons suggest an unusu- 
ally high rate of mitochondrial DNA evolution in 
Mytiius (Mollusca: Bivalvia). Molecular Biology & 
Evolution. 13: 418-421. 

HOFFMANN, R. J., J. L BOORE & W. M. BROWN, 
1992, A novel mitochondrial genome organiza- 
tion for the blue mussel, Mytilus edulis. Genetics, 
131: 397-412. 

HOLTON, T. A. & M. W. GRAHAM, 1 991 , A simple 
and efficient method for direct cloning of PCR 



CYTOCHROME Ь DIVERSITY IN CHINESE ONCOMELANIA 



221 



products using ddT-tailed vectors. Nucleic Acids 
Research. 19:11 56-1 1 58. 

HOPE, M. & D. P. McMANUS, 1994, Genetic vari- 
ation in geographically isolated populations and 
subspecies of Oncomelania liupensis deter- 
mined by a PCR-based RFLP method. Acta 
Tropica, 57: 75-82. 

KOCHER, T D., W. K. THOMAS, A. MEYER, S. V. 
EDWARDS, J. PAABO. F. X. VILIJ\BIJ\NCA & A. 
С WILSON, 1989, Dynamics of mitochondrial 
DNA evolution in animals: amplification and se- 
quencing with conserved primers. Proceedings 
of the National Academy of Sciences USA. 86: 
6196-6200. 

LECANIDOU, R., V. DOURIS & G. С RODAKIS, 
1994, Novel features of metazoan mtDNA re- 
vealed from sequence analysis of three mito- 
chondrial DNA segments of the land snail Albi- 
naria turrita (Gastropoda: Clausiliidae). Journal of 
Molecular Evolution, 38: 369-382. 

LINCOLN, S. E., M. J. DALY & E. S. LANDER, 
1991, PRIMER: a computer program for auto- 
matically selecting PCR primers. Distributed by 
the authors. Whitehead Institute for Biomedical 



Research, Massachusetts Institute of Technol- 
ogy, Cambridge, Massachusetts, USA. 

ROSENBERG, G., G. M. DAVIS, G. S. KUNCIO & 
M. G. HARASEWYCH, 1994, Preliminary riboso- 
mal RNA phylogeny of gastropod and unionoid- 
ean bivalve mollusks. The Nautilus. Supplement, 
2: 111-121. 

SAMBROOK, J., E. F. FRITSCH & T. MANIATIS, 
1 989, Molecular cloning, a lab manual, second 
edition. Cold Spring Harbor Laboraton/ Press. 

SPOLSKY, С & T. UZZELL, 1984, Natural inter- 
species transfer of mitochondrial DNA in am- 
phibians. Proceedings of the National Academy 
of Sciences USA, 81: 5802-5805. 

SPOLSKY, С &T. UZZELL, 1986, Evolutionary his- 
tory of the hybridogenetic hybrid frog Rana es- 
culenta as deduced from mtDNA analyses. Mo- 
lecular Biology & Evolution. 3: 44-56. 

THORNE, J. L. & H. KISHINO, 1992, Freeing phy- 
logenies from artifacts of alignment. Molecular 
Biology & Evolution, 9: 1148-1162. 



Revised Ms. accepted 1 July 1996 



MALACOLOGIA, 1996, 38(1-2): 223-227 



LETTERS TO THE EDITOR 



CRITERIA FOR THE DETERMINATION OF TAXONOMIC BOUNDARIES 

IN FRESHWATER UNIONOIDS (BIVALVIA: UNIONOIDA): 

COMMENTS ON STIVEN AND ALDERMAN (1992) 

Walter R. Hoeh' & Mark E. Gordon^ 



The southern Atlantic Slope region of the 
United States is an area characterized by high 
species richness and considerable local en- 
demism in the freshwater fauna (e.g., fresh- 
water mussels [Johnson, 1970: Burch, 1975; 
Davis et al., 1981; Kat, 1983: Hoeh, 1990], 
snails [Burch & Tottenham, 1980; Thompson, 
1968, 1984], crayfish [Hobbs, 1989], and fish 
[Lee et al., 1980]). Kat (1983) demonstrated 
species-level divergence within the regional 
Lampsilis radiata (Gmelin 1791) complex (i.e., 
L. radiata s.S., L. sp. [nowL. fu//er/iaf/ Johnson 
1984], and L. splendida [Lea, 1838]). Using 
allozymic and morphological comparisons, 
Stiven & Alderman (1 992; hereafter referred to 
as SA) examined nine populations of union- 
oids from North Carolina in order to, among 
other objectives, reassess and resolve the 
taxonomic status of three nominal taxa of 
Lampsilis: L. radiata radiata, L. radiata con- 
spicua (Lea 1872), and L. fullerkati (taxa as 
listed byTurgeon et al., 1988). SA concluded, 
based primarily on genetic distance criteria, 
that the former two taxa, and probably L. 
fullerkati as well, should "be considered sim- 
ply as allopatric 'populations' of Lampsilis ra- 
diata.'' 

A critical reading of SA reveals inconsis- 
tencies and errors in methodology and data 
interpretation. We believe that the following 
detailed discussion of SA is necessary for 
two reasons: (1) The taxonomic revisions 
suggested by SA, if implemented, would 
likely have a significant impact on the level of 
legislative protection afforded at least two of 
the taxa in question, and (2) the high visibility 
of this paper among malacologists and envi- 
ronmental resource professionals may lead 
to its use as an exemplar for studies of pop- 



ulation structure, taxonomic boundaries, and 
phylogenetic relationships in unionoid bi- 
valves. Thus, because of the potential impact 
of this paper, aspects of the research pre- 
sented therein should be carefully re-evalu- 
ated. 

Morphological Analyses 

The data presented in SA (table 2) suggest 
that there are statistically significant differ- 
ences in length, height, and width among 
Lampsilis r. conspicua, L. r. radiata, and L. 
fullerkati shells. However, similarity in slope 
and y-axis intercept for plots of length versus 
height for L. r. conspicua and L. r. radiata 
(specimens of L. fullerkati not plotted) was 
used as an indicator of conspecificity in SA 
(fig. 2). Furthermore, the statistically signifi- 
cant mensurable differences among the 
above three Lampsilis taxa were downplayed 
by reference to "site effects" in Lampsilis 
cariosa and Leptodea ochracea (SA: 366). 

Appeals to site effects (phenotypic plastic- 
ity) as explanations for the observed concho- 
logical differences among unionoid popula- 
tions, without the appropriate substantiating 
data, are simply hypotheses to be tested. 
Contrary to statements in the text, the Deep 
River Lampsilis cariosa population is not al- 
lozymically identical to the other two popula- 
tions of L. cariosa (SA: table 3). Therefore, the 
data presented in SA cannot discount ge- 
netic effects for the size differences among 
populations of L. cariosa. Furthermore, be- 
cause nine of the alleles detected in Lep- 
todea ochracea were found only in one of the 
two populations analyzed (again contrary to 
the text; see SA: table 3), genetic effects can- 



^ Department of Biology, Daltiousie University. Halifax, Nova Scotia B3H 4J1, Canada. Corresponding Authior & Current 
Address; Walter R. Hoeti, Department of Zoology, Miami University, Oxford, Otiio 45056, U.S.A. 
^Zoology Section, Campus Box 315. University of Colorado Museum, Boulder. Colorado 80309, U.S.A. 



223 



224 



HOEH & GORDON 



not be ruled out in this instance either. Even 
if two populations of a single species are al- 
lozymically identical, distinct genetic deter- 
minants for conchological morphology may 
be present. 

We wonder if the "site effect analyses" 
presented in SA controlled for age and/or 
gender (sexual dimorphism). These poten- 
tially confounding factors were not discussed 
in SA (no explanation of the methodology uti- 
lized in the conchological analyses was pre- 
sented in the "Methods" section). For exam- 
ple, assuming no genetic component for the 
observed size variation, were the Deep River 
individuals of LampslHs cariosa larger be- 
cause of site effects or because they were 
older? The Deep River ". . . population is 
comprised of only large old specimens and is 
thought to be declining" (SA: p. 356). We be- 
lieve this statement to be suggestive of an 
age effect that could have confounded the 
conchological analyses. In summary, the sig- 
nificant mensurable differences between L. r. 
conspicua, L. r. radiata, and L. fullerkati, com- 
bined with no substantiation for site effects, 
are consistent with the hypothesis that these 
three taxa are distinct evolutionary lineages. 
The similarity in slope and y-axis intercept for 
plots of length versus height for L. r. con- 
spicua and L. r. radiata is irrelevant given the 
significant morphological differences be- 
tween these taxa. 

Sampling Strategy for the Allozyme Study 

"We note also that when distance mea- 
sures are relatively large between pairs of 
species, and heterozygosity is low, the con- 
struction of phenograms can be carried out 
fairly reliably with only a few representative 
individuals for a species (Nei, 1978)" (SA: 
357-358). However, the particular sampling 
strategies used in SA for organisms and loci 
will often produce unreliable estimates of ge- 
netic distances. Three problems are outlined 
below. 

Although only single populations of Lamp- 
silis r. conspicua and L. r. radiata were ex- 
amined allozymical'y in SA, it is desirable to 
use multiple populations to represent each 
taxon in analyses of taxonomic boundaries 
and among-taxa relationships (e.g., Baver- 
stock & Moritz, 1990). This is especially true 
for groups, such as the Unionoida, that are 
known to contain cryptic species (e.g., Davis 



et al., 1981; Davis, 1983, 1984). Furthermore, 
representing species by single populations 
can produce phylogenetically misleading re- 
sults (e.g., Smouse et al., 1991). 

Species of Drosophila are typically consid- 
ered to have "high" heterozygosities (15- 
20%; e.g., Gorman & Renzi, 1979). Therefore, 
the heterozygosities reported in SA (table 3) 
for the nine unionoid populations are not gen- 
erally "low" (range: 5.1% to 31.8%, mean = 
18.3%). Thus, estimates of genetic distances 
reported in SA may be inaccurate due to the 
small number of individuals used in certain 
comparisons (e.g., L7\P was assayed for a 
single individual of Lampsilis fullerkati [SA: ta- 
ble 3]). 

Nei (1978) makes it clear that a relatively 
large number of assayed loci are required for 
accurate estimates of genetic distance. Nei 
(1978: 583) stated the following: "Nei and 
Roychoudhurry (1974) concluded that for es- 
timating average heterozygosity and genetic 
distance a large number of loci rather than a 
large number of individuals per locus should 
be used. . . ." What quantity does Nei imply 
with the phrase, "a large number of loci"? "In 
fact, less than 30 loci were studied in most 
recent protein surveys. This number is small; 
ideally, more than 50 loci should be used ..." 
(Nei, 1978: 587). The import of a relatively 
large number of loci for estimation of genetic 
distances has been empirically substantiated 
(e.g., Gorman & Renzi, 1979). Both Nei's and 
Roger's distances are subject to large stan- 
dard errors especially at relatively small dis- 
tances (as is potentially the case among 
Lampsilis fullerkati and the two populations 
representing L. r. conspicua and L. r. radiata 
analyzed in SA), and the major factor influ- 
encing the standard errors is the number of 
loci sampled (e.g., Nei, 1978, 1987; Nei & 
Chesser, 1983; Richardson et al., 1986; 
Chakraborty & Leimar, 1987; Baverstock & 
Moritz, 1990). The sampling of eleven loci, as 
was the case in SA, does not give rigorous 
estimates of genetic distance. Therefore, the 
estimates of absolute genetic divergence pre- 
sented in SA should be considered tentative 
as should any taxonomic revision based on 
those estimates. 

Lack of Reference to Types 

Neither Lampsilis r. radiata nor L. r. con- 
spicua from their respective type localities or 
type locality drainages (Potomac and Yadkin 



TAXONOMIC BOUNDARIES IN FRESHWATER UNIONOIDS 



225 



rivers, respectively) were allozymically ana- 
lyzed in SA. Not utilizing topotypic or near- 
topotypic material for molecular evaluation 
compounded with the lack of reference to 
type material of any sort should preclude the 
taxonomic revisions suggested in SA. Taxo- 
nomic revisions must be based on reference 
to "types." Molecular analyses do not obvi- 
ate this necessity. As it now stands, the 
specimens of L. r. radiata and L. r. conspicua 
utilized in SA cannot be confirmed as actually 
representing the taxa implied. 

Taxonomic Concepts 

Although often referring to conchological 
data, SA displays a strong reliance on levels 
of genetic distance for the determination of 
taxonomic boundaries. However, cogent ar- 
guments, based on theoretical and opera- 
tional criteria, have been made against the 
use of genetic distances for delimiting taxa 
(e.g.. Frost & Hillis, 1 990). A major operational 
problem discussed in Frost & Hillis (1990) is 
the arbitrary nature of genetic distance mea- 
sures. Because allozyme loci evolve at differ- 
ent rates (e.g., Sarich, 1977; Skibinski & 
Ward, 1982), genetic distance estimates are 
sensitive to the particular loci analyzed. Even 
if identical loci are scored, distance estimates 
may differ dramatically from one analysis to 
another due to the use of different electro- 
phoretic conditions (e.g., Singh et al., 1976). 
The use of non-identical suites of allozyme 
loci combined with different electrophoretic 
conditions for some of the loci in common 
between studies may partly explain the dis- 
crepancy in reported genetic distance esti- 
mates between LampsHis fullerkati and L. ra- 
diata radiata (SA: Nei's D = 0.049; Kat, 1983; 
Mean Nei's D = 0.129). Moreover, genetic 
distances are not appropriate measures of 
taxonomic status for recently diverged union- 
oid populations (e.g., Davis et al., 1981). "In 
no case is the species concept based on ge- 
netic distance alone" (Davis, 1983). A sim- 
plistic reliance on genetic distance-based 
taxonomic concepts should be abandoned. 

Regarding genetic identity levels, SA (p. 
366) states that ". . . Davis et al. (1981) ar- 
gued that values > 0.9 could be found among 
sympatric freshwater mussel species that re- 
cently underwent speciation. However, these 
two so-called subspecies of L. radiata are 
currently not sympatric. . . ." This statement. 



combined with SA's emphasis on genetic 
distance, implies that allopatric populations, 
in order to be recognized taxonomically, 
must be more divergent allozymically than 
sympatric populations. Can there not be dis- 
tinct allopatric species with absolute genetic 
identities greater than 0.9? Theoretically, any 
number of assayed allozyme loci could indi- 
cate genetic identity between two distinct 
species. This may be expected for relatively 
recently diverged taxa (e.g., Johnson et al., 
1977; Woodruff & Gould, 1980; Davis et al., 
1981; Carson, 1982; Kat, 1983). However, 
genetic differences could still exist at non- 
allozyme loci. How are these potential differ- 
ences evaluated? They are evaluated by ref- 
erence to morphological, ecological, and 
phenological data. "The following species 
concept is used here; a species of unionid is 
a single lineage comprised of one or more 
populations that diverge from other lineages. 
Divergence is shown by significant morpho- 
logical, cytological, reproductive biological 
and/or ecological differences. . . . The case 
for species status is strengthened if repro- 
ductive isolation is highly probable due to 
drainage system differences . . ." (Davis, 
1983). We believe that the lack of reference 
to genetic distance in this unionid species 
concept is appropriate. 

The concepts of (1) divergence (and, there- 
fore, diagnosability) and (2) a lineage of pop- 
ulations (both following Davis, 1983) should 
be incorporated in evaluations of unionoid 
taxonomic boundaries. The limited data 
available in SA (tables 2 & 3, range and hab- 
itat data provided in the text) suggest that 
both Lampsilis r. conspicua and L. fullerl<ati 
are diagnosable from L. r. radiata. Despite the 
small number of loci assayed and minimal 
genetic distances reported in SA, the allo- 
zyme data set does suggest that there are 
diagnostic alleles (i.e., those alleles found 
in only one of the three taxa of interest here; 
L. r. radiata, one diagnostic allele; L. r. con- 
spicua, eight diagnostic alleles; and L. fuller- 
kati, four diagnostic alleles). The presence of 
unique genetic elements in a particular pop- 
ulation is not consistent with a hypothesis of 
current genetic interchange with the other 
populations. Therefore, if these alleles remain 
diagnostic after adequate sampling of popu- 
lations of the three nominal taxa, this would 
be strong evidence consonant with the dis- 
tinctness of these populations. This sugges- 



226 



HOEH & GORDON 



tion of distinctness is further evinced by the 
morphological and ecological data presented 
in SA. 



CONCLUSIONS 

The significant conchological, habitat, and 
range differences reported in SA (e.g., both in 
the text and table 2), suggest that the popu- 
lations identified as L. fullerkati, L. r. radiata 
and L. r. conspicua represent distinct evolu- 
tionary lineages. Apparent non-identity of al- 
lozymic composition for these populations 
(SA: table 3) is consonant with this hypothe- 
sis. We believe that SA's "suggestion ... of 
regrouping the two and possibly three previ- 
ously recognized allopatric subspecies/spe- 
cies into one species complex, based upon 
very high levels of genetic identity as well as 
similar conchologies, is probably an uncom- 
mon event, yet appropriate until more dis- 
tinctive biological species properties become 
evident" (SA: 367) is not appropriate at this 
time. Given the above discussion and the 
rapid decline of North American unionoid 
populations (e.g.. Bogan, 1993), we believe 
that it is a scientifically justifiable and conser- 
vative course of action to continue to recog- 
nize L. r. conspicua {sensu SA) and L. fuller- 
kati as taxa distinct from L. r. radiata. Data 
indicating need for further revision within the 
L. radiata group may become available; how- 
ever, any revisions should necessarily be 
based on rigorous analyses of all the avail- 
able characteristics of individuals selected by 
appropriate sampling criteria. 

ACKNOWLEDGMENTS 

We thank A. E. Bogan, J. B. Burch, S. L 
Neale, E. E. Spamer, and D. T. Stewart for 
their criticisms of earlier versions of this com- 
mentary. We also appreciate the comments 
of an anonymous reviewer as well as those of 
the editor, George M. Davis. 



LITERATURE CITED 

BAVERSTOCK, P. R. & С MORITZ, 1990, Sam- 
pling design. Pp. 13-24, In D, м. hillis & с 
MORITZ, eds.. Molecular systematics. Sinauer As- 
sociates, Inc., Sunderland, Massachusetts, 
U.S.A. 

BOGAN, A. E., 1993, Freshwater bivalve extinc- 



tions (Mollusca: Unionoida): a search for causes. 
American Zoologist, 33: 599-609. 

BURCH, J. В., 1975, Freshwater unionacean clams 
(Mollusca: Pelecypoda) of North America. Mala- 
cological Publications, Hamburg, Michigan 
U.S.A. 204 pp. 

BURCH, J, B. & J. L. TOTTENHAM, 1980, North 
American freshwater snails: species list, ranges 
and illustrations. Waikerana. 1: 81-215. 

CARSON, H. L., 1982, Evolution of Drosophila on 
the newer Hawaiian Islands. Heredity, 48: 3-26. 

CHAKRABORTY, R. & O. LEIMAR, 1987, Genetic 
variation within a subdivided population, pp. 89- 
120. In N. RYMAN & F. UTTER, eds., Population ge- 
netics and fishery management. University of 
Washington Press, Seattle, WA, USA. 

DAVIS, G. M., 1983, Relative roles of molecular 
genetics, anatomy, morphometries and ecology 
in assessing relationships among North Ameri- 
can Unionidae (Bivalvia), pp. 193-222. In G. S. 
OXFORD & D. ROLLiNSON, eds., Protein polymor- 
phism: adaptive and taxonomic significance. Ac- 
ademic Press, London. 

DAVIS, G. M., 1984, Genetic relationships among 
some North American Unionidae (Bivalvia): sib- 
ling species, convergence, and cladistic relation- 
ships. Malacologia, 25: 629-648. 

DAVIS, G. M., W. H. HEARD, S. L. H. FULLER & С 
HESTERMAN, 1981, Molecular genetics and 
speciation in Elliptio and its relationships to other 
taxa of North American Unionidae (Bivalvia). Bi- 
ological Journal of the Linnean Society, 15: 131- 
150. 

FROST, D. R. & D. M. HILLIS, 1990, Species in 
concept and practice: herpetological applica- 
tions. Herpetologica, 46: 87-104. 

GORMAN, G. С & J. RENZI, 1979, Genetic dis- 
tance and heterozygosity estimates in electro- 
phoretic studies: effects of sample size. Copela, 
1979: 242-249. 

HOBBS, H. H., 1989, An illustrated checklist of the 
Amehcan crayfishes (Decapoda: Astacidae, 
Cambaridae, and Parastacidae). Smithsonian 
Contributions to Zoology, 480: 1-236. 

HOEH, W. R., 1990, Phylogenetic relationships 
among eastern North American Anodonta (Bi- 
valvia: Unionidae). Malacologicai Review, 23: 
63-82. 

JOHNSON, M. S., B. CLARKE & J. MURRAY, 
1977, Genetic variation and reproductive isola- 
tion in Partula. Evolution, 31: 116-126. 

JOHNSON, R. I., 1970, The systematics and zoo- 
geography of the Unionidae (Mollusca: Bivalvia) 
of the southern Atlantic Slope region. Bulletin of 
the Museum of Comparative Zoology, 140: 263- 
449. 

KAT, P. W., 1983, Morphologic divergence, genet- 
ics, and speciation among Lampsilis (Bivalvia: 
Unionidae). Journal of Molluscan Studies, 49: 
133-145. 

LEE, D. S., С R. GILBERT, С H. HOCUTT, R. E. 
JENKINS, D. E, MCALLISTER & J. R. STAUFFER, 
1 980, Atlas of North American freshwater fishes. 



TAXONOMIC BOUNDARIES IN FRESHWATER UNIONOIDS 



227 



North Carolina State Museum of Natural History, 
Raleighi, North Carolina. 854 pp. 

NEI, M., 1978, Estimation of average heterozygos- 
ity and genetic distance from a small number of 
individuals. Genetics, 89: 583-590. 

NEI, M., 1987, Molecular evolutionary genetics. 
Columbia University Press, New York. 512 pp. 

NEI, M. & R. K. CHESSER, 1983, Estimation of 
fixation indices and gene diversities. Annals of 
Human Genetics. 47: 253-259. 

RICHARDSON, B. J., P. R. BAVERSTOCK & M. 
ADAMS, 1986, Allozyme electrophoresis. Aca- 
demic Press, Sydney. 410 pp. 

SARICH, V. M., 1977, Electrophoresis in evolution- 
ary studies: rates, sample sizes and the neutral- 
ity hypothesis. Nature, 265: 24-28. 

SINGH, R. S., R. С LEWONTIN & A. A. FELTON, 
1976, Genie heterogeneity within electrophoretic 
"alleles" of xanthine dehydrogenase in Droso- 
phila pseudoobscura. Genetics, 84: 609-629. 

SKIBINSKI, D. O. F. & R. D. WARD, 1982, Corre- 
lations between heterozygosity and evolutionary 
rate of proteins. Nature, 298: 490-492. 

SMOUSE, P. E., T. E. DOWLING, J. A. TWOREK, 
W. R. HOEH & W. M. BROWN, 1991, Effects of 
intraspecific variation on phylogenetic inference: 
A likelihood analysis of mtDNA restriction site 



data in cyprinid fishes. Systematic Zoology. 40: 
393-409. 

STIVEN, A. E. & J. ALDERMAN, 1 992, Genetic sim- 
ilarities among certain freshwater mussel popu- 
lations of the Lampsilis genus in North Carolina. 
Malacologia. 34: 355-369. 

THOMPSON, F. G., 1968, The aquatic snails of the 
family Hydrobiidae of peninsular Florida. Univer- 
sity of Florida Press, Gainesville, Florida. 268 pp. 

THOMPSON, F. G., 1984, Freshwater snails of flor- 
ida, a manual for identification. University of Flor- 
ida Press, Gainesville, Florida. 94 pp. 

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 Fisher- 
ies Society. Special Publications, 16: 1-277. 

WOODRUFF, D. S. & S. J. GOULD, 1980, Geo- 
graphic differentiation and speciation in Ce- 
rion — A preliminary discussion of patterns and 
processes. Biological Journal of the Linnean So- 
ciety, 14: 389-416. 



Revised Ms. accepted 16 April 1996 



MALACOLOGIA, 1996, 38(1-2): 229-230 

A CALL FOR A NEW INTERNATIONAL CONGRESS OF ZOOLOGY 

Dr. F. D. Por^ & Dr. R. M. Polymeni^ 



We are looking for response concerning 
the feasibility of a New International Con- 
gress of Zoology, possibly to be convened in 
Athens, Greece, sometime during 1999 or 
2000. 

The First International Congress of Zool- 
ogy was held in Paris, in 1889. Seventy years 
later, the XVIth Congress in Washington rec- 
ommended the discontinuation of the con- 
gresses because of the feeling that zoology 
had split into too many unrelated, specialized 
fields. Nonetheless, a last XVIIth rump Con- 
gress was held in 1972 in Monte Carlo. The 
relatively few participants of this meeting 
unanimously, but in vain, asked for the con- 
tinuation of the congresses. The idea was ad- 
vanced that the International Conferences on 
Systematic and Evolutionary Biology would 
replace the defunct Zoological Congresses 
at a higher, integrative level. After several 
meetings of the ICSEB, it became evident 
that they did not live-up to this expectation. 
In contrast, the International Congresses of 
Botany have continued undisturbed and suc- 
cessfully. 

One of the unhappy consequences of the 
cessation has been the fact that the Interna- 
tional Commission of Zoological Nomencla- 
ture, once accountable to the congresses, 
came under the formal responsibility of the 
General Congresses of International Union of 
the Biological Sciences (lUBS). But the more 
painful and long-lasting consequence was a 
general depreciation of zoology in the aca- 
demic world as such, and the replacement of 
this discipline by a plethora of euphemistically 
more fashionable designations. 

However zoology at the end of this century 
is more alive than ever and rich in new ideas 
and achievements. A multispeciality ex- 
change of views is more necessary than ever 
before. Not unexpectedly, this is also the con- 
sequence of the extreme parochialism of the 
different splinter fields and the ignorance of 
general zoological issues which it generated. 
There is a long list of such issues that cut 



across the lines of all the zoological special- 
ities, some of them of important philosophical 
and practical significance. 

The widely circulated "Systematics Agenda 
2000" emphasizes our present incapacity to 
describe scientifically a zoological biodiver- 
sity that suddenly appears to be one order of 
magnitude largerthan envisaged in the 1 970s. 
This is not only a matter of quantities or of time 
needed, but a matter that calls for the restruc- 
ture of zoological research world over. A crit- 
ically depleted and weakened community of 
zoological systematics cannot live up to the 
task to investigate and possibly protect the 
heritage of the animal world. 

On the positive side, there have been many 
developments during the last three decades 
that need to be appreciated by an interna- 
tional forum of all the zoologists. Confined to 
the pages of specialized journals, these im- 
portant developments often do not reach the 
attention of peers in other zoological special- 
ties. In the field of more classical zoology, it 
would be useful to acquaint our colleagues 
with such topics as the recently discovered 
new animal phyla and classes, new concepts 
of vertebrate evolution, zoology of clonal 
animals, and present views on the Protozoa. 
A sample of subjects of wider implications 
are sociobiology, cladistics, molecular taxon- 
omy, modern embryology, the new vision on 
the Cambrian revolution, the neo-catastro- 
phism, vicariance zoogeography, in situ and 
ex situ conservation, cryopreservation, and 
cloning. This is a different zoology from that 
which ended with a whimper in Monte Carlo. 

We are ready to try to bring forward again 
the rich and unifying aspects of zoology and 
to reassert its general global, human and 
philosophical role. We are hoping for the ap- 
proval and support of the zoological dias- 
pora. The best encouragement will be to 
send us suggestions regarding the themes 
and the structure of the proposed New Inter- 
national Congress of Zoology. More impor- 
tantly, we need personal commitments to 



''Department of Evolution, Systematics and Ecology, Hebrew University of Jerusalem, РОВ 91904, Jerusalem, Israel 
^Department of Biology, Section of Zoology, University of Athens, РОВ 15784, Attiens, Greece. 
Please contact Dr. Rosa Polymeni; Tel. 30.1.7284364, Fax 30.1.7284604, email: rpolime@atlas.uoa.gr 



229 



230 



POR & POLYMENI 



help organizing symposia, workshops and 
hints of possible funding sources. We shall 
also need to establish an active and repre- 
sentative Action Committee. Understand- 
ably, we shall be able to appeal for funding 
only after having obtained convincing public 
support and after having a prestigious 
committee in place. 



The editor-in-chief of Malacologia welcomes let- 
ters that comment on vital issues of general im- 
portance to the field of Malacology, or that com- 
ment on the content of the journal. Publication is 
dependent on discretion, space available and, in 
some cases, review. Address letters to; Letter to 
the Editor, Malacologia, care of the Department 
of Malacology, Academy of Natural Sciences, 
19th and the Parkway. Philadelphia, PA 19103. 



MALACOLOGIA, 1996, 38(1-2): 231-239 

INDEX 



Page numbers in italics indicate 
illustrations of taxa. 

Acanthinula 165, 166, 171, 178 
Achatina stuhlmanni 165, 166, 169, 

171, 174, 179 
Acoctilidium amboinense 147 

bayerfetilmanni 147, 150 

fijiense 143-151; 144, 145, 147, 148 

sutteri 147, 1 50 

weberi 147 
Actinonaias ligamentina 184, 186-189, 

202 
Adula californiensis chosenica 36, 40, 41 

schmidti 4 1 
Afroconulus diaphanus 180 

iredalei 165, 166, 171, 179 
agapetus, Potamolithus 1-17; 2, 9 
ag Testis, Agrio Umax 156, 157 
Agriolimax agrestis 156, 157 

re tic и la tus 157 
Alasmidonta marginata 184, 186, 187, 

192, 202 

undulata 182 
A I bin a ha caerulea 216-219 
Alderja modesta 149 
Allogona profunda 67-86; 70-73 
aloysiisabaudiae, Gymnahon 165, 166, 

171, 180 
alveata, Li гор hora 1 1 О 
amathusia, Chionopsis 109, 112, 116- 

124, 128-130, 142 
Amblema plicata 182,184,186,187, 

189, 195, 197, 201, 202 
amboinense, Acochlidium 147 
Ampelita 207 
Ancistrolepis trochoidea ovoidea 36, 38 

trochoidea [Bathy ancistrolepis] ovoidea 
38 

trochoideus ovoideus 38 
Anodonta imbelicis 59-65 
Anodontoides ferussacianus 184, 187, 

194, 202 
Anomalocardia 1 08, 111, 113, 127, 1 33- 

135, 140 

auberiana 1 16-124, 142 

flexuosa 112, 1 16-124, 142 
Anomiostrea 36, 45 

coralliophila 45 

pyxidata 45 
Appalachina sayana 67-86; 70-73 
appressa. Patera 67-86 
araneosa, Chione 109 
ahel, Trachycystis 167, 179 
arm a ta. Bullía 96 
arma tum, Buccinum 96 
arm a tum. Dors a пит 96 



asperrima. Protot ha с а (Leucoma) 1 1 2 
asperhma, Protothaca 112, 116-1 24, 

142 
a s persa. Helix 1 47 
Astralium yamamurae 36, 41 

yamamurai 41 
Astralium (Destellifer) yamamurae 41 
Astralium (Distellifer) yamamurae 41 
athleta, Lirophora 110, 112, 116-124, 

127, 129, 142 
au ben ana, Anomalocardia 1 1 6-1 24, 1 42 
auhformis, Daedalochila 67, 83 
Australaba 56 

babauiti, Curvella 165, 166, 169, 171, 

179 
bacillum, Streptostele 165, 166, 171, 

180 
bainbridgensis, Chione (Chione) 1 1 1 
bainbhdgensis, Puberella 110 
bal lista, Lirophora 1 1 
barba tum, Steno trema 67 
barbigerum, Stenotrema 67-86; 80, 81 
barrakporensis, Kali ella 165, 166, 171, 

179 
batillariaeformis, Clypeomorus 36, 42 
batillariaeformis, Clypeomrus 42 
bayerfehlmanni, Acochlidium 147, 150 
beringii, Bering Ion 39 
Beringion 36, 39 

beringii 3 9 

mars ha Hi 39 
bifasciata, Clypeomorus 47-58; 49, 50, 

52, 54 
Biomphalaria glabra ta 59, 62, 63 
bisulcata, Nesopupa 165, 166, 168, 169, 

171, 178 
Boreomelon stearnsii ryosukei 36, 40 
Boreotrophon paucicostatus 36, 37 
Boromelon stearnsii ryosukei 40 
Bosellia corinneae 149 
Boucardicus 207, 209, 210 
Brachytoma kawamurai 36, 44 

kurodai 36, 44 

vex i Ilium 36, 44 
Bradybaenidae 85 
breviculus, Clypeomorus 56 
brunnea. Eg lisia 36, 44, 45 
brunnea, Egiisia lanceolata 45 
Buccimum chishimana nux 38 
Buccinanops cochlidium 96 

cochlidius 87 

deform is 87 

gradatus 87-]02; 89, 90, 92-96 

lamarckii 87 

moniliferus Ъ7-У 02; 89, 98-1 00 

monliferum 96 



231 



232 



INDEX 



uruguayensis 87 
buccinoides, Clinopegma 36, 38, 39 
Buccinum armatum 96 

chishimananux 35, 38 

chishimanum nux 36, 38 

felis shikamai 36, 38 

ferrugineum 36, 39, 40 

hosoyai 36, 38 

japonicum 38 

kawamurai 36, 38 

kinukatsugi 36, 39, 40 

mido ri 36, 40 

moniliferum 96 

opisthoplectum microconcha 36, 38 

subreticulatum 36, 39 

undatum 101 
Buccinum (Buccinanops) maniliferum 96 
Bulbus f lav us elongatus 36, 37 
Bu Ilia 87 

arm a ta 96 
Bu Ilia (Buccinanops) moniliferum 96 
burnsii, Panchione 1 1 
buschii, Lithoglyphus 7 
bushchii, Poîamolithus 1-17; 70-72, 74- 

16 
butumbiana, Prosita la 165, 166, 171, 

179 

caeruleum, Cerithium 56 

californiensis, Chione 109, 130 

Camaenidae 85 

canarium, Strom bus 3 

cancellata, Chione 1 09, 112, 11 6-1 24, 

127, 129, 130, 132, 135, 142 
cancellata, Harpa 44 
cardium, Lampsilis 184, 186-191, 202 
carinifer, Hemifusus 36, 43 
carinii era, Hemifusus 43 
cariosa, Lampsilis 223, 224 
carlottae, Lirophora 1 1 
caroniana, Lirophora 1 1 
castanea, Neptúnea 39 
castaneum, Volutopsion 39 
castaneus, Volutopsion 39 
castaneus, Volutopsius 39 
Cataracta, Pyganodon 182 
Cecilioides 164-166 
Ceras tua trapezoidea la да riens is 165, 

166, 169, 171, 178 
Cepaea nemoralis 216-219 
Cerithium 5 6 

caeruleum 56 

nodulosum 56 

rupestre 56 

vulgatum 56 
C/?/o/7e 103-142 

ara neos a 1 09 

californiensis 109, 130 

cancellata 109,112,116-124,127, 
129, 130, 132, 135, 142 

chipolana 109,111,112,116-124, 
132, 142 



compta 1 09 

erosa 109, 129, 130 

guatulcoensis 109 

mazyckii 109, 129 

pai lasaña 1 09 

primigenia 109, 127 

quebradillensis 1 09 

ласа 129 

santodomingensis 109, 127 

s и bim bric ata 1 09, 111, 125 

Шше/75 109,112,113,116-125,129, 
130 

undatella 106, 109, 129 

i /эса 125, 128 
Chione (Chione) bainbridgensis 1 1 1 

spenceri 1 1 1 

subimbricata 1 1 1 

tumens 1 1 1 
Chioninae 1 08 
Chionista 108, 113, 125, 133, 134, 137 

солГег/ 109, 129, 130 

eurylopas 109 

fluctifraga 109, 112, 116-124, 129, 
130, 142 

g nidi a 1 09 

jamaniana 1 09 

ornatissima 109 

posorjensis 1 09 

procancellata 1 09 

propincua 109 

rowleei 110 

tegulum 1 1 
Chionopsis 108, 111, 113, 126, 131, 

132-134, 137, 140 

amathusia 109,112,116-124,128- 
130, 142 

y/7/íy/a 129, 130, 136 

lliochione 1 1 

procancellata 127, 129, 130 

subrugosa 1 1 

tegulum 127, 130, 136 

iva/// 1 1 

woodwardi 110, 127, 129, 130 
chipolana, Chione 1 09, 111, 112, 116- 

124, 132, 142 
chiriquiensis, Lirophora 1 1 
chishimananum, Buccinum 35, 38 
c/7/ty/, Oncomelania hu pens i s 2 1 3 
Chlamydarion oscitans 165, 166, 169, 

171, 180 
Chorus giganteus 55 
chosenica, Adula californiensis 36, 40, 

41 
cinnamomeozonata , Thapsia 1 80 
Cirsotrema kagayai 36, 37 
с/ала, Subulona 165, 166, 169, 171, 178 
c/ava, Pleurobema 182, 184, 186, 187, 

192, 193, 202 
Clav ato r 207 
clenchi, Lirophora 1 1 
Clinopegma buccinoides 36, 38, 39 
Clypeomorus batillariaeformis 36, 42 



INDEX 



233 



bifasciata 47-58; 49, 50, 52, 54 

breviculus 56 

moniliferum 48 

moniliferus 5 6 

petrosa gennesi 48 

tuberculatus 47-58; 50, 52, 54 
Clypeomrus batillariaeformis 42 
coccineum, Pleurobema 186 
cochlidium, Buccinanops 96 
cochlidius, Buccinanops 87 
сое ru le a. Al bin a ri a 216-219 
Collisella cassis shirogai 35, 37 

pelta shirogai 35, 37 
colombiana, Crenel la 40 
colombiana, Megacrenella 40 
Columbiana, Vespericola 67-86 
complanata, Lasmigona 184, 187, 194, 

195, 202 
compressa, Lasmigona 184, 187, 194, 

202 
compta, Chione 1 09 
Conomurex 56 
Co nu lin us 1 74 

rutshuruensis major 165, 166, 169, 
171, 178 
conspicua, Lampsilis radia ta 223-226 
coralliophila, Anomiostrea 45 
Corbicula fluminea 20 
corinneae, Bosellia 149 
cornuta, Pseudunela 147, 149 
cortezi, Chionista 109, 129, 130 
cortina ri a, Puberella 110, 127 
costata, Lasmigona 182, 184, 186, 187, 

191, 194, 202 
Crenel la colombiana 40 

cribarla, Puberella 110, 112, 1 1 6-1 24, 

129, 130, 142 
Curvella 165, 166, 171, 179 

babaulti 165, 166, 169, 171, 179 
Cyathopoma 207, 210 
cylindrica, Quadrula 184, 186, 187, 202 
Cypraea pu le bel la 42 

Daedalochila auriformis 67, 83 
c/a///, Lirophora 1 1 
сУэ///, Nodulotrophon 37 
сУа///, Trop hon 38 
Decollidrillia 36, 40 

л/^глэ 36, 40 
deformis, Buccinanops 87 
denotata, Xolotrema 67-86; 77 
densesculpta, Thapsia 1 80 
Dentalium (Pictodentalium) formosum 34 

formes um hirasei 34 
den ti fera. Neo helix 67-86; 76 
Deroceros reticulatus 1 57 
diaphanus, Afroconulus 180 
dilatata. Elliptic 184, 186, 187, 189, 

192, 202 
discrepans, Lirophora 1 1 
disseminata, Gulella 165, 166, 170, 171, 

180 



doerfeuilliana, Millerelix 67 
Dorsanum 87 

arm a tum 96 

т/лал 87, 101 

moniliferum 96 
Dreissena polymorpha 1 9-31 
c/iyb/a, Pa//o 149 

ebergenyi, Lirophora 1 1 
есУа//5, My ti lus 28, 218 
Eg lisia brunnea 36, 44, 45 

lanceo lata brunnea 45 
е/аГ/ол, Maizania 165, 166, 168, 169, 

171, 178 
elegans, Pseudoglessula (Ischnoglessula) 

179 
elegans, Pseudoglessula 165, 166, 171, 

172 
elegans, Tantulum 143 
El до no с y с lus к op ta we liens is 165, 166, 

171, 178 
f///pr/o c^//arara 184, 186, 187, 189, 192, 

202 
elongatus, Bulbus fia vus 36, 37 
elongensis, Gonaxis 165, 166, 171, 180 
elongensis, Thapsia 1 80 
Elysia maoria 1 49 

suborna ta 1 49 
emphatica, Trophonopsis scitulus 37 
emphaticus, Trophonopsis scitula 36, 37 
emphaticus, Trophonopsis scitulus 37 
Epioblasma 1 8 6 

torulosa 184, 202 

torulosa rangiana 182, 187-191 

triquetra 184, 186-191, 202 
erasa, СЛ/оле 109, 129, 130 
eucosmia, Thapsia 164-166, 168, 169, 

180 
euracantha. Murex 42 
euracanthus. Murex 42 
eu rant ha. Murex 42 
eurantha, Spinidrupa 42 
eu ry lop as, Chionista 1 09 
exquisita, Fusipagoda 38 
exquisita, Mohnia 38 

^aba//s, W//osa 184, 186-188, 191, 192, 

202 
falconensis, Lirophora 1 1 
fallax, Triodopsis 67-86; 70-73 
fasciola, Lampsilis 184, 186-190, 202 
fasciolaris, Ptychobranchus 1 84, 186- 

189, 192, 198, 202 
Faux и lus 207, 210 
ferrissi, Inflectarius 67-86; 76, S 7 
ferrugineum, Buccinum 36, 39, 40 
ferussacianus, Anodontoides 184, 187, 

194, 202 
Fi с ad us ta 36, 42 
Ficadusta pule bel la 42 
fijiense, Acochlidium 143-151; 744, 745, 

747, 745 



234 



INDEX 



flexuosa, Anomalocardia 112, 116-1 24 

142 
flucti fraga, С hi on is ta 109, 112, 116-124 

129, 130, 142 
fluminea, Corbicula 20 
formosana, Oncomelania hupensis 2 1 3 
formosum, Dentalium (Pictodentaliumj 34 
fosteri, Xo lot re m a 67-86; 70-73 
frielei, Neoberingus 39 
Fu Ig orarla (Musashiaj kaneko hayashii 36, 

40 
fullerkati. Lamps I I is 223-226 
funiakensis, Panchione 1 1 
Fusconaia subrotunda 184, 186, 187 

192, 193, 202 
Fusinus 3 3 

hyphalus 33 
Fusinus (SimpHcifusus) hyphalus 33 
Fusipagoda 36, 38 

exquisita 38 
Fusus simplex 33 

gennesi, Clypeomorus petrosa 48 

gerstenbrandti, Thapsia 1 80 

gibbosula, Heterocardia 45 

gibbosuloidea, Plica rularía 36, 43 

giganteus, Chorus 55 

g la brat a, Biomphalaria 59, 62, 63 

g nidi a, Chionista 109 

gnidia, Chionopsis 129, 130, 136 

Gonaxis elongensis 165, 166, 171, 180 

gradatus, Buccinanops 87-102; 55, 50 
52-56 

grandis, Pyganodon 184, 187, 189, 194 
195, 202 

Granulittorina 36,41,42 
m i Heg rana 42 

Granulittorina philippiana 36, 41, 42 

guatulcoensis, Chione 109 

Gulella disseminata 165, 166, 170 171 
180 

disseminata var. kekumegaensis 1 80 
handeiensis 165, 166, 171, 180 
impedita 165-167, 170, 171, 180 
lessensis 165, 166, 171, 180 
osborni 165, 166, 170, 171, 180 
ugandensis 165-167, 169, 171, 180 
woodhousei 165, 166, 171, 180 

Gulella (sect. Silvigulella) 174 

Guppya quadrisculpta 165, 166, 169, 

171. 179 

Gymnarion aloysiisabaudiae 165, 166 

171. 180 

Haines i a 207 

Halolimnohelix percivali 165, 166, 171 

plana 165, 166, 171, 180 
handeiensis, Gulella 165, 166, 171, 180 
Harpa cancellata 44 

harpa 44 

kajiyamai 36, 44 

kawamurai 36, 43, 44 



major 44 

striata 43 
harpa, Harpa 44 
Harpo fusus 35, 36, 39 

meló ni s 39 
harpula, Pupisoma (Salpingoma) 1 78 
harpula, Pupisoma 165, 166, 171 
hayashii, Fu Igor aria (Mus as hi a) kaneko 

36, 40 
Hedylopsis 1 49 

s pi culi fera 148, 149 
Helicidae 85 
Helicophanta 207 
A/e//x a s pars a 1 47 
Hemifusus cari ni fer 36, 43 

cari ni fera 43 
hendersoni, Lirophora 110, 129, 130 
Heterocardia gibbosula 45 
hirasei, Dentalium (Pictodentaliumj 

formosum 34 
holocyma, Panchione 1 1 O 
hosoyai, Buccinum 36, 38 
hotelensis, Panchione 110 
hupensis, Oncomelania 213-218,220 
hupensis, Oncomelania hupensis 2 1 3 
hyphalus, Fusinus (SimpHcifusus) 33 
hyphalus, Fusinus 33 

Iliochione 108, 113, 125, 133, 134 137 
140 

subrugosa 112,116-124,129,130 
142 
imbelicis, Anodonta 59-65 
impedita, Gulella 165-167,170 171 

180 
Inflectarius ferrissi 67-86; 76, 5 7 

inflectarius 68 

inflectus 67-86; 70-75, 5 7 

magazinensis 67-86; 50 

5ш/Г/7/ 67-86; 50 

subpallia tus 67-86; 50 
inflectarius, Inflectarius 68 
inflectus, Inflectarius 67-86; 70-73, 81 
intapurpurea, Puberella 111, 129 
iredalei, Afroconulus 165, 166, 171, 179 
iredalei. Ka Helia 165, 166, 171, 179 
iredalei, Oreohomorus 165, 166, 171 

179 
iredalei, Trachycystis 165, 166, 171 
/>/s, l////osa 184, 186-191, 202 
Isabella, Laevistrombus 33 
Isabella, Laevistrombus canarium "forma" 



33 

ja maní ana, Chionista 109 
japonic um, Buccinum 38 
japonicum, Pupisoma 1 78 

kagayai, Cirsotrema 36, 37 
kajiyamai. Harpa 36, 44 
Kalidos 207, 210 
Kaliella 207, 210 



INDEX 



235 



barrakporensis 165, 166, 171, 179 
iredalei 165, 166, 171, 179 
karamwegasensis, Thapsia 180 
Katharina tu ni cata 218 
kawamurai, Brac/iytoma 36, 44 
kawamurai, Buccinum 36, 38 
kawamurai, ¡Harpa 36, 43, 44 
kekumegaensis, Gulella disseminata var. 

180 
kekumeganum, Pseudopeas 179 
kelletii, Panciiione 110, 112 
kinukatsugi, Buccinum 36, 39, 40 
koptaweililensis, IVlicractaeon 165, 166, 

170, 171, 178 

koptaweliensis, Elgonocyclus 165, 166, 

171, 178 

koreana, Megacardita ferruginea 41 
koreana, Megacardita furriginosa 36, 41 
koreanica, IVIegacardita ferruginea 41 
kurodai, Brachytoma 36, 44 
kurosfiio, Neptúnea 35 

/acteoides, Pyrene 36, 43 
Laevicardium rubropictum 36, 45 
laevior, Patera 67-86; 70-73 
Laevistrombus 33 

canarium "forma" isa bella 33 

Isabella 3 3 
lagariensis, Ceras tua trapezoidea 165, 

166, 169, 171, 178 
lamarckii, Buccinanops 87 
Lam bis 56 
Lampsilis cardium 184, 186-191, 202 

cariosa 223, 224 

fasciola 184, 186-190, 202 

fullerkati 223-226 

ovata 184, 186-191, 202 

radia ta 223 

radiât a conspicua 223-226 

radiât a radia ta 223-226 

siliquoidea 182, 184, 186-189, 191, 
196, 201, 202 

splendida 223 
lapidum, Potamolithus 1 
Lasmigona complanata 184, 187, 194, 

195, 202 

compressa 184, 187, 194, 202 

costata 182, 184, 186, 187, 191, 
194, 202 
la tili rata, Lirophora 110, 129 
Latirus recurvirostrum 43 

stenomphalus 36, 43 

sttnomphalus 43 
Lehm a nia valentiana 1 57 
Leptodea ochracea 223 
lessensis, Gulella 165, 166, 171, 180 
Leukoma 108, 140 
ligamentina, Actinonaias 184, 186-189, 

202 
Ligumia nasuta 184, 186, 187, 189, 

195, 202 



лесгэ 184, 186-191, 202 
Lim aria perfragile 45 
Lim aria (Platilimaria) perfragile 45 
Limax maximus 153-160; 755, 755 
Limicolaria saturata 165-167, 170, 171, 

179 
lindoensis, Oncomelania hupensis 2 1 3 
Lirophora 1 08, 111, 113, 126, 127, 1 30, 

131, 133-135, 137, 140 

alveata 1 1 O 

athleta 110, 112, 116-1 24, 1 27, 1 29, 
142 

bal I is ta 1 1 О 

carlottae 1 1 О 

caroniana 1 1 О 

chiriquiensis 1 1 О 

с lene hi 1 1 О 

dalli 1 1 О 

discrepans 1 1 О 

ebergenyi 1 1 О 

falconensis 1 1 О 

hendersoni 110, 129, 1 30 

la ti II га ta 110, 129 

mariae 1 1 О 

oblitera ta 1 1 О 

paphia 1 10, 129, 130 

quirosensis 1 1 О 

riomaturensis 1 1 О 

sellardsi 1 1 

tembla 1 1 О 

victoria 110, 112, 1 1 6-1 24, 1 42 

vrendenbergi 1 1 О 
Lithoglyphus buschii 7 
lliochione, Chionopsis 1 1 
Lymnaea stagna lis 59, 62-64 

Macrotoma yamamurae 36, 45 
mactropsis, Panchione 110, 112, 116- 

124, 127, 136, 142 
magazinensis, Inflectarius 67-86; 80 
Maizania elatior 165, 166, 168, 169, 

171, 178 
major. Сопи H nus rutshuruensis 165, 166, 

169, 171, 178 
major, Harpa 44 
major, Neohelix 67-86; 77-75 
Malaga ri on 207, 210 
M a la ri nia 207, 210 

maniliferum, Buccinum (Buccinanops) 96 
Mantellum perfragile 36, 45 
maori a, Elysia 149 
Margantes vorticifera 37 
margina ta, Alasmidonta 184, 186, 187, 

192, 202 
mariae, Lirophora 1 1 
marica. Timo с lea (Glycydonta) 1 1 2 
marica, Timoclea 112, 1 1 6-1 24, 1 42 
mars ha Hi, Beringion 39 
maxillatum, Stenotrema 67-86; 80 
maximus, Limax 153-160; 755, 755 
mazyckii, Chione 109, 129 



236 



INDEX 



mcmichaeli, Volutoconus grossi 36, 44 
medjensis, Trochozonites (Zonitotrochusj 

180 
medjensis, Trochozonites 165, 166, 171 
Megacardita ferruginea /<oreana 41 

ferruginea Icoreanica 41 

ferruginosa koreana 36, 41 
IVIegacrenella 35, 36, 40 
iVIegacrenella colombiana 40 
IVIelanopsis 5 5 
melonis, Harpofusus 39 
me/onis, Pyrulofusus (¡Harpofusus) 39 
melonis, Strom bell a 39 
Mercenaria 113, 132-1 34, 1 40 

mercenaria 112, 1 1 6-1 24, 1 42 
mercenaria, Mercenaria 112, 11 6-1 24, 

142 
Mesodon normalis 67-86; 70, 72, 73 
Mesodontini 83 
Micractaeon koptaweililensis 165, 166, 

170, 171, 178 
microconcha, Buccinum opisthoplectum 

36, 38 
Microcystis 207, 209, 210 
microleuca, Thapsia 164-166, 180 
Microtoma yamamurae 45 
midori, Buccinum 36, 40 
millegrana, Granulittorina 42 
millegrana, Nodilittorina (Granulittorina} 

42 
Millerelix doerfeuilliana 67 

doerfeuilliana sampsoni 67 

mooreana 67 
mime, Thapsia 1 80 
minima, Oncomelania 2 1 3 
miran, Dorsanum 87, 101 
modesta, Alderja 1 49 
Mohnia exquisita 38 

multicostata 36, 38 
moniliferum, Buccinum 96 
moniliferum. Bullía (Buccinanops) 96 
moniliferum, Clypeomorus 48 
moniliferum, Dorsanum 96 
moniliferus, Buccinanops 87-102; 89, 98- 

100 
moniliferus, Clypeomorus 56 
monliferum, Buccinanops 96 
montezuma, Puberella 1 1 1 
mooreana, Millerelix 67 
morsitans, Puberella 111, 129, 1 30 
multicostata, Mohnia 36, 38 
Murex euracantha 42 

euracanthus 42 

eu rant ha 42 
mutandana, Pseudoglessula 179 
My ti I US edulis 28, 218 

Nassarius reticulata 101 

nasuta, Ligumia 184, 186, 187, 189, 

195, 202 
Nebula ria yaekoae 36, 43 



nebulosa. Vi llosa 182 
nemoralis, Cepaea 2 1 6-2 1 9 
Neoberingus 36, 39 

frielei 3 9 
Neohelix dentifera 67-86; 76 

Алэуол 67-86; 7 7-75 
Neptúnea 35 

castanea 39 

kuroshio 35 

rurosio 3 5 
Nesopupa bisulcata 165, 166, 168, 169, 

171, 178 
nigra, Decollidrillia 36, 40 
nigropardalis, Pyrene testudinalia 42, 43 
nigropardalis, Pyrene testudinaria 36, 42, 

43 
nigropunctatum, Vasticardium 36, 45 
ninagongonis , Trunca tal lina 165, 166, 

171, 178 
nitidus, Oreohomorus 179 
Nodilittorina 42 

Nodilittorina (Granulittorina) millegrana 42 
nodulosum, Cerithium 56 
Nodulotrophon 36-38 

сУа/// 37 
normalis, Mesodon 67-86; 70, 72, 75 
nosophora, Oncomelania hupensis 213 
Nothapalus 165, 166, 171, 174, 178 
A76/X, Buccimum chishimana 38 
/7t/x, Buccinum chishimanum 36, 38 

oblitera ta, Li гор hora 1 1 О 
obstricta, Xolotrema 67-86; 77, 75 
ос bracea, Leptodea 223 
oissoni, Puberella 1 1 1 
Omphalomargarites 36, 37 

vorticifera 207, 210 
Omphalomargarites (Omphalomargarites) 

37 
Oncomelania 213-221 

hupensis 213-218, 220 

hupensis chiui 2 1 3 

hupensis formosana 213 

hupensis hupensis 2 1 3 

hupensis lindoensis 213 

hupensis nosophora 2 1 3 

hupensis quadrasi 2 1 3 

hupensis robertsoni 213, 218 

hupensis tangí 2 1 3 

minima 2 1 3 
orcula, Pupisoma 165, 166, 171, 178 
Oreohomorus 1 74 

/лесУэ/е/ 165, 166, 171, 179 

nitidus 1 7 9 
ornatissima, Chionista 109 
osborni, Gulella 165, 166, 170, 171, 180 
os ci tans, Chlamydarion 165, 166, 169, 

171, 180 
Ost rea pyx ida ta 45 
oi/aía, Lampsilis 184, 186-191, 202 
ovata, Timoclea 1 1 2 



INDEX 



237 



ovoidea, Ancistrolepis tro с ho idea 36, 38 
ovoidea, Ancistrolepis troclioidea 

[Bathyancistrolepis] 38 
ovoideus, Ancistrolepis trochoideus 38 

pai lasaña, Chione 1 09 
Palio dubia 1 49 

zosterae 149 
Panchione 108,113,126,127,131, 

133-135, 137, 140 

burnsii 1 1 

funiakensis 1 1 

holocyma 1 1 

hotelensis 1 1 

kelletii 110, 112 

mactropsis 110, 112, 116-1 24, 1 27, 
136, 142 

parkeria 1 1 

ulocyma 112,116-124,127,129, 
130, 142 
paphia, Li гор hora 110, 129, 130 
paradox a, St rubel lia 143, 149 
parkeria, Panchione 1 1 
Patelloida (Collisellina) saccharinoides 36, 

41 

saccharioides 4 1 
Patera appressa 67-86 

appressa sculptior 78, 79 

laevior 67-86; 70-73 

perigrapta 67-86; 78 

sargentiana 67-86; 78 
paucicostatus, Boreotrophon 36, 37 
percivali, Halolimnohelix 165, 166, 171 
perfragile, Limaria (Platilimaria) 45 
perfragile, Limaria 45 
perfragile, Mantellum 36, 45 
perigrapta. Patera 67-86; 78 
Petenopsis 1 40 
- tu mens 142 
pfeifferianus, Reticutriton 42 
philippiana, Granulittorina 36,41,42 
Pic to den ta Hum 3 4 

pilosa, Vespericola columbiana 67-86; Ô0 
р/э/7э, Halolimnohelix 165, 166, 171, 

180 
planulata, Trachycystis 172 
Pleurobema clava 182, 184, 186, 187, 

192, 193. 202 

coccineur.. 186 

sintoxia 184, 186, 187, 189, 192, 
193, 195, 201, 202 
Plica rula ha gibbosuloidea 36, 43 
plicata, Amblema 182, 184, 186, 187, 

189, 195, 197, 201, 202 
Polygyridae 83 
polymorpha, Dreissena 19-31 
Pomatiopsinae 220 
posorjensis, Chionista 109 
Potamolithus agapetus 1-17; 2, 9 

bushchii 1-17; 70-72, 14-16 

lapidum 1 
primigenia, Chione 109, 127 



procancellata, Chionista 109 
procancella ta , Chionopsis 127, 129, 130 
profunda, Allogona 67-86; 70-73 
propinqua, Chionista 1 09 
Pros Ítala 1 74 

butumbiana 165, 166, 171, 179 
Protothaca 108,113,132-134 

asperrima 112, 1 1 6-1 24, 1 42 
Protothaca (Leucoma) asperrima 1 1 2 
Pseudoglessula elegans 165, 166, 171, 

172 

m и tanda na 179 

subfuscidula 1 79 
Pseudoglessula (Ischnoglessulaj 1 74 

elegans 1 7 9 
Pseudopeas kekumeganum 179 

yaiaensis 165, 166, 171, 172, 179 
Pseudunela cornuta 147, 149 
Ptychobranchus fasciolaris 184, 186- 

189, 192, 198, 202 
púbera, Puberella 1 1 1 
Puberella 108, 111, 113, 126, 131, 132- 

134, 137, 140, 141 

bainbridgensis 1 1 O 

cortinaria 110, 127 

cribarla 110, 112, 1 1 6-1 24, 129, 1 30, 
142 

intapurpurea 111, 129 

montezuma 1 1 1 

morsitans 111, 129, 1 30 

oissoni 1 1 1 

púbera 1 1 1 

pulicaria 111, 129, 130 

purpurissata 1 1 1 

sawkinsi 111, 127, 129, 130 
pulchella, Cypraea 42 
pu le bella, Ficadusta 42 
pulicaria, Puberella 111, 129, 1 30 
Punctum ugandanum 165, 166, 169, 

170, 171, 179 
Pupisoma harpula 165, 166, 171 

japonicum 1 78 

OACív/a 165, 166, 171, 178 
Pupisoma (Salpingoma) harpula 1 78 
purpurissata, Puberella 1 1 1 
Pyganodon Cataracta 182 

grandis 184, 187, 189, 194, 195, 202 
Pyre ne lácteo i des 36, 43 

testudinalia nigropardalis 42, 43 

testudinaria nigropardalis 36, 42, 43 
Pyrulofusus 39 

Pyrulofusus (Harpofusus) melonis 39 
pyxidata, Anomiostrea 45 
pyxidata, Ost rea 45 

quadrasi, Oncomelania hupensis 2 1 3 
quadrisculpta, Guppya 165, 166, 169, 

171, 179 
Quadrula cylindrica 184, 186, 187, 202 
quebradillensis, Chione 1 09 
quirosensis, Li гор hora 1 1 O 



238 



INDEX 



raca, Chione 1 29 

г a di at a, La m psi lis 223 

radiata, Lampsilis radiata 223-226 

rangiana, Epioblasma to rulos a 182, 187- 

191 
recta, Ligumia 184, 186-191, 202 
recurvirostrum, La t i rus 43 
reticulata, Nassari us 101 
reticulatus, A g rio lim ax 1 57 
reticulatus, Deroceros 157 
fíe tic и triton 36, 42 

pfeifferianus 42 
fíhachidina chiradzuluensis var. virgínea 

165-167, 171, 178 
fíhinoclavis 56 
riomaturensis, Li гор hora 110 
robertsoni, Oncomelania hupensis 213, 

218 
row I eel, Chionista 110 
rubocostatum. Vex i Hum 36, 43 
rubropictum, Laevicardium 36, 45 
rupestre, Cerithium 56 
rurosio. Neptúnea 35 
ryosukei, Boreomelon stearnsii 36, 40 
ryosukei, Boromelon stearnsii 40 

saccharinoides, Patelloida (Collisellina) 

36, 41 
saccharioides, Patelloida (Collisellina) 41 
sampsoni, Millerelix doerfeuilliana Ql 
santodomingensis, Chione 109, 127 
sargentiana, Patera 67-86; ZS 
saturate, Limicolaria 165-167, 170, 171, 

179 
sawkinsi, Puberella 111, 127, 129, 130 
sayana, Appalachina 67-86; 70-73 
schmidti. Adula 41 
sculptior. Patera appressa 78, 79 
sellardsi, Lirophora 1 1 
shikamai, Buccinum fells 36, 38 
s hi год ai, Collisella cassis 35, 37 
s hi года i , Collisella pelta 35, 37 
siliquoidea, Lampsilis 182, 184, 186-189, 

191, 196, 201, 202 
simplex, Fusus 33 
Simplicifusus 33, 34 
sintoxia, Pleura bema 184, 186, 187, 

189, 192, 193, 195, 201, 202 
Sítala 207, 209, 210 
s m it hi, Inflectarius 67-86; 80 
spenceri, Chione (Chione) 1 1 1 
spiculifera, Hedylopsis 148, 149 
Spinidrupa 36, 42 

eu rant ha 42 
splendida, Lampsilis 223 
stagnalis, Lymnaea 59, 62-64 
stenomphalus, Latirus 36, 43 
Stenotrema barbatum 67 

barbigerum 67-86; 50, 5 7 

m ax i I latum 67-86; 50 
Streptostele bacillum 165, 166, 171, 180 
striata. Harpa 43 



St rom bella melonis 39 

Strombus canarium 3 

Strophitus undula tus 184, 186, 187, 

189, 192, 202 
Strubellia paradoxa 143,149 
sttnomphalus, Latirus 43 
stuhlmanni, Achatina 165, 166, 169, 

171, 174, 179 
subcylindrica, Truncatella 55 
subfuscidula, Pseudoglessula 179 
subimbricata, Chione (Chione) 1 1 1 
subimbricata, Chione 109, 111, 125 
suborna ta, Ely si a 149 
subpall'atus, Inflectarius 67-86; 50 
subreticulatum, Buccinum 36, 39 
subrotunda, Fusconaia 184, 186, 187, 

192, 193, 202 
subrugosa, Chionopsis 1 1 
subrugosa, lliochione 112, 116-1 24, 1 29, 

130, 142 
Subulona Clara 165, 166, 169, 171, 178 
Succinea 165, 166, 171, 178 
sutteri, Acochlidium 147, 150 

tangi, Oncomelania hupensis 2 1 3 

Tantulum elegans 143 

tegulum, Chionista 1 1 

tegulum, Chionopsis 127, 130, 136 

tembia, Lirophora 1 1 

Thapsia cinnamomeozonata 1 80 

densesculpta 1 80 

elongensls 1 80 

eucosmia 164-166, 168, 169, 180 

gerstenbrandti 1 80 

karamwegasensis 1 80 

microleuca 164-166, 180 

mime 1 80 

yalaensis 1 80 
Timoclea 108, 111, 113, 132-134, 140 

/77ЭЛ/СЭ 1 12, 1 16-124, 142 

ovata 1 1 2 
Timoclea (Glycydonta) marica 1 1 2 
torulosa, Epioblasma 184, 202 
Trachycystis ariel 167, 179 

/лесУа/е/ 165, 166, 171 

planulata 1 72 
/"r/cty/a 217, 220 
Triculinae 220 
Triodopsini 83 

Triodopsis fallax 67-86; 70-73 
triquetra, Epioblasma 184, 186-191, 202 
Trochozonites medjensis 165, 166, 171 
Trochozonites (Zonitotrochus) 1 74 

medjensis 1 80 
Trop hon dalli 38 
Trophonopsis scitula emphaticus 36, 37 

scitulus emphatica 37 

scitulus emphaticus 37 
Tropidophora 207,210 
Truncatella subcylindrica 55 
Truncatellina ninagongonis 165, 166, 

171, 178 



INDEX 



239 



tuberculatus, Clypeomorus 47-58; 50, 

52, 54 
tu mens, Chione (Chionej 1 1 1 
tumens, Chione 1 09, 112, 113, 116- 

125, 129, 130 
tumens, Pete no psi s 142 
tunica ta, Kattiarina 218 



165, 166, 169, 



165-167, 169, 171, 
1 12, 1 16-124, 127, 



ugandanum. Punctum 

170, 171, 179 
ugandensis, G и /ella 

180 
ulocyma, Panchione 

129, 130, 142 
undatella, Chione 106, 109, 129 
undatum, Buccinum 101 
undula ta, Alasmidonta 182 
undula tus, Strophitus 184, 186, 187, 

189, 192, 202 
Unionoida 224 
uruguayensis, Buccinanops 87 

vaca, Chione 125, 128 

valentiana, Lehm a nia 1 57 

vanuxemensis, Villosa vanuxemensis 1 82 

vanuxemi, Villosa 182 

Vasticardium nigropunctatum 36, 45 

Vespericola columbiana 67-86 

Columbiana pilosa 67-86; 80 
vexillium, Brachytoma 36, 44 
Vexillum rubocostatum 36, 43 
Victoria, Lirophora 110, 112, 1 1 6-1 24, 

142 
Villosa fabalis 184, 186-188, 191, 192, 

202 

iris 184, 186-191, 202 

nebulosa 1 82 

vanuxemensis vanuxemensis 1 82 

vanuxemi 182 
virgínea, Rhachidina chiradzuluensis var. 

165-167, 171, 178 
Volutoconus grossi mcmichaeli 36, 44 
Volutopsion 36, 39 

castaneum 39 

castaneus 39 
Volutopsius castaneus 39 
vorticifera, M arg a rites 37 
vorticifera, Omphalomargarites 207, 210 
vrendenbergi, Lirophora 1 1 
vu Ig a tu m, Cerithium 56 

walli, Chionopsis 1 1 
weberi, Acochlidium 147 
woodhousei, Gulella 165, 166, 171, 180 
woodwardi, Chionopsis 110, 127, 129, 
130 

Xo lot rem a denotata 67-86; 77 
fosteri 67-86; 70-73 
obstricta 67-86; 77, 78 



yalaensis, Pseudopeas 165, 166, 171 

172, 179 
yalaensis, T ha psi a 1 80 
yamamurae, Astralium (Destellifer) 41 
yamamurae, Astralium (Distellifer) 41 
yamamurae, Astralium 36, 41 
yamamurae, Macrotoma 36, 45 
yamamurae, Microtoma 45 
yamamurai, Astralium 41 

zosterae. Palio 149 



yaekoae, Nebularia 36, 43 



17. Tables are to be used sparingly, and 
vertical lines not at all. 

1 8. References cited in the text nnust be in 
the Literature Cited section and wee versa. 
Follow a recent issue of MALACOLOGIA for 
bibliographic style, noting especially that se- 
rials are cited unabbreviated. Supply pagina- 
tion for books. Supply infornnation on plates, 
etc., only if they are not included in the 
pagination. 

19. In systematic papers, synonymies 
should not give complete citations but 
should relate by author, date and page to the 
Literature 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 re- 
turned should request this at the time of or- 
dering rephnts. Otherwise, illustrations will be 
maintained for six months only after publica- 
tion. 



REPRINTS AND PAGE COSTS 

23. When 100 or more reprints are or- 
dered, an author receives 25 additional cop- 
ies free. Reprints must be ordered at the time 
proof is returned to the Editorial Office. Later 
orders cannot be considered. For each au- 
thors' change in page proof, the cost is U.S. 
$3.00 or more. 

24. When an article is 10 or more phnted 
pages long, MALACOLOGIA requests that an 
author pay part of the publication costs if 
grant or institutional support is available. 



SUBSCRIPTION PURCHASE 

25. Effective Jan. 1 995, 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 38, NO. 1-2 MALACOLOGIA 1996 

CONTENTS 

MARÍA FERNANDA LÓPEZ ARMENGOL 

Taxonomic Revision of Potamolithus Agapetus Pilsbry, 1911, and Potamolithus 
Buschii (Frauenfeld, 1865) (Gastropoda: Hydrobiidae) 1 

IVI. E. CHASE & R. С BAILEY 

Recruitment of Dreissena Polymorpha: Does the Presence and Density of Conspe- 

cifics Deterrлine the Recruitnnent Density and Pattern in a Population? 19 

RÜDIGER BIELER & RICHARD E. PETIT 

Additional Notes on Nomina First Introduced by Tetsuaki, Kira in "Coloured Illustra- 
tions of the Shells of Japan" 33 

RICHARD E. PETIT & RÜDIGER BIELER 

On The New Names Introduced in the Various Printings of "Shells of the World in 
Colour" [Vol. I by Tadashige Habe and Kiyoshi Ito; Vol. II by Tadashige Habe and 
Sadao Kosuge] 35 

FADWA A. ATTIGA & HAMEED A. AL-HAJJ 

Uitrastructural Study of Euspermiogenesis in Clypeomorus Bifasciata and Clypeo- 
morus Tuberculatus (Prosobranchia: Cerithiidae) With Emphasis on Acrosome For- 
mation 47 

DAZHONG XU & MICHELE G. WHEATLY 

CA Regulation in the Freshwater Bivalve Anodonfa Imbecilis: I. Effect of Environmen- 
tal CA Concentration and Body Mass on Unidirectional and Net CA Fluxes 59 

KENNETH С EMBERTON 

Microsculptures of Convergent and Divergent Polygyhd Land-Snail Shells 67 

LUIZ RICARDO L SIMONE 

Anatomy and Systematics of Buccinanops Gradatus (Deshayes, 1844) and Bucci- 
nanops Moniliferus (Kiener, 1834) (Neogastropoda, Muricoidea) From the Southeast- 
ern Coast of Brazil 87 

PETER D. ROOPNARINE 

Systematics, Biogeography and Extinction of Chionine Bivalves (Bivalvia: Veneridae) 

in Tropical America: Early Oligocene-Recent 103 

MARTIN HAASE & ERHARD WAWRA 

The Genital System o^ Acochlidium fijiense (Opisthobranchia: Acochlidioldea) and its 
Inferred Function 1 43 

G. M. KUCHENMEISTER, D. J. PRIOR & I. G. WELSFORD 

Quantification of the Development of the Cephalic Sac and Podocyst in the Terres- 
trial Gastropod Umax Maximus L 1 53 

P. TATTERSFIELD 

Local Patterns of Land Snail Diversity in a Kenyan Rain Forest 161 

LAURA R. WHITE, BRUCE A. McPHERON, & JAY R. STAUFFER, JR. 

Molecular Genetic Identification Tools for the Unionids of French Creek, Pennsylva- 
nia 181 

KENNETH С EMBERTON, TIMOTHY A. PEARCE & ROGER RANDALANA 

Quantitatively Sampling Land-Snail Species Richness in Madagascan Rainforests . 203 

CHRISTINA M. SPOLSKY, GEORGE M. DAVIS & ZHANG Yl 

Sequencing Methodology and Phylogenetic Analysis: Cytochrome b Gene Sequence 
Reveals Significant Diversity in Chinese Populations of Oncomelania (Gastropoda: 
Pomatiopsidae) 213 

LETTERS TO THE EDITOR 

WALTER R. HOEH & MARK E. GORDON 

Criteria for the Determination of Taxonomic Boundaries in Freshwater Unionoids 
(Bivalvia: Unionoida): Comments on Stiven and Alderman (1992) 223 

DR. F. D. POR & DR. R. M. POLYMENI 

A Call for a New International Congress of Zoology 229