(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 

Museum of 

Comparative Zoology 



VOL 32 1990,1991 



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 January IS 


Vol. 


29, 


No. 1 28 June 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 



MALACOLOGIA, VOL. 32 

CONTENTS 

RÜDIGER BIELER & RICHARD E. PETIT 

On the Various Editions of Tetsuaki Kira's "Coloured Illustrations of the 
Shells of Japan" and "Shells of the Western Pacific in Color Vol. I," With 
an Annotated List of New Names Introduced 131 

PHILIPPE BOUCHET 

Turrid Genera and Mode of Development: The Use and Abuse of Pro- 
toconch Morphology 69 

MALCOLM EDMUNDS 

Introduction 205 

MALCOLM EDMUNDS 

Does Warning Coloration Occur in Nudibranchs? 241 

JOSÉ С. GARCIA-GÓMEZ, ANTONIO MEDINA & RAFAEL COVEÑAS 

Study of the Anatomy and Histology of the Mantle Dermal Formations 
(MDFs) of Chromodoris and Hypselodoris (Opisthobranchia: 
Chromodorididae) 233 

TERRENCE M. GOSLINER 

Morphological Parallelism in Opisthobranch Gastropods 313 

GERHARD HASZPRUNAR 

Towards a Phylogenetic System of Gastropoda Part I: Traditional Meth- 
odology — A Reply 1 95 

KATHE R. JENSEN 

Comparison of Alimentary Systems in Shelled and Non-Shelled 

Sacoglossa ( + Ascoglossa) (Gastropoda: Opisthobranchia) 209 

ALAN R. KABAT 

Predatory Ecology of Naticid Gastropods with a Review of Shell Boring 
Prédation 1 55 

YURI I. KANTOR 

Anatomical Basis for the Origin and Evolution of the Toxoglossan Mode 

of Feeding 3 

E. ALISON KAY 

Turrid Faunas of Pacific Islands 79 

ALAN J. KOHN 

Tempo and Mode of Evolution in Conidae 55 

JAMES NYBAKKEN 

Ontogenetic Change in the Conus Radula, its Form, Distribution Among 

the Radula Types, and Significance in Systematics and Ecology 35 

JOSE ANGEL ALVAREZ PEREZ, MANUEL HAIMOVICI & 

JOÀO CARLOS BRAHM COUSIN 

Sperm Storage Mechanisms and Fertilization in Females of Two South 
American Eledonids (Cephalopoda: Octopoda) 1 47 



BERNARD E. PICTON 

Cumanotus beaumonti (Eliot, 1906), A Nudibranch Adapted for Life in a 
Shallow Sandy Habitat? 219 

MATHIEU POULICEK, MARIE-FRANÇOISE VOSS-FOUCART & 
CHARLES JEUNIAUX 

Regressive Shell Evolution Among Opisthobranch Gastropods 223 

L. VON SALVINI-PLAWEN 

The Status of the Rhodopidae (Gastropoda: Euthyneura) 301 

ROGER R. SEAPY 

The Pelagic Family Atlantidae (Gastropoda: Heteropoda) From Hawai- 
ian Waters: A Faunistic Survey 1 07 

GAMIL N. SOLIMAN 

A Comparative Review of the Spawning, Development and 

Metamorphosis of Prosobranch and Opisthobranch Gastropods with 

Special Reference to Those from the Northwestern Red Sea 257 

JOHN D. TAYLOR 

Introduction 1 

JOHN D. TAYLOR 

The Anatomy of the Foregut and Relationships in the Terebridae 19 

CHRISTOPOHER D. TODD 

Larval Strategies of Nudibranch Molluscs: Similar Means to the 

Same End? 273 

RITA TRIEBSKORN & С KÜNAST 

Ultrastructural Changes in the Digestive System of Deroceras Reticula- 
tum (Mollusca; Gastropoda) Induced by Lethal and Sublethal Concen- 
trations of the Carbamate Molluscicide Cloethocarb 89 

RICCARDO CATTANEO VIETTI & ANDREA BALDUZZI 

Relationship Between Radular Morphology and Food in the Doridina 
(Mollusca: Nudibranchia) 21 1 

RICCARDO CATTANEO VIETTI & RENATO CHEMELLO 

The Opisthobranch Fauna of a Mediterranean Lagoon (Stagnone di 
Marsala, Western Sicily) 291 



MCZ 
VOL 32, NO. 1 LIBRARY 1990 

DEC 1 2 1990 

HARVARD 
UNlVERSiTY 



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 

Nineteenth Street and the Parkway 

Philadelphia, Pennsylvania 19103, U.S.A. 



EUGENE COAN 

California Academy of Sciences 

San Francisco, CA 



Co-Editors: 



Assistant Managing Editor: 

CARYL HESTERMAN 

Associate Editors: 



CAROL JONES 
Pay son, AZ 



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: 



KENNETH J. BOSS, 

Museum of Comparative Zoology 

Cambridge, Massachusetts 

JOHN BURCH, President-Elect 

MELBOURNE R. CARRIKER 
University of Delaware, Lewes 

GEORGE M. DAVIS 
Secretary and Treasurer 

CAROLE S. HICKMAN, Vice-President 
University of California, Berkeley 



JAMES NYBAKKEN, President 
Moss Landing Marine Laboratory 
California 

CLYDE F. E. ROPER 
Smithsonian Institution 
Washington, D.C. 

W. D. RUSSELL-HUNTER 
Syracuse University, New York 

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. 

ELMER G. BERRY, 
Germantown, Maryland 



Emeritus Members 

ROBERT ROBERTSON 

The Academy of Natural Sciences 

Philadelphia, Pennsylvania 

NORMAN F. SOHL 
U.S. Geological Survey 
Reston, Virginia 



Copyright ^j 1 990 by the Institute of Malacology 



1990 
EDITORIAL BOARD 



J. A. ALLEN 

Marine Biological Station 

Millport. United Kingdom 

E. E. BINDER 

Museum d'l-iistoire Naturelle 

Genève, Switzerland 

A. J. CAIN 

University of Liverpool 
United Kingdom 

P. CALOW 

University of Sheffield 
United Kingdom 

A. H. CLARKE, Jr. 
Portland, Texas, U.S.A. 

B. С CLARKE 
University of Nottingham 
United Kingdom 

R. DILLON 

College of Charleston 

SC, U.S.A. 

С J. DUNCAN 
University of Liverpool 
United Kingdom 

E. FISCHER-PIETTE 

Muséum National d'Histoire Naturelle 

Paris, France 

V. FRETTER 
University of Reading 
United Kingdom 

E. GITTENBERGER 
Rijksmuseum van Natuurlijke Historie 
Leiden, Netherlands 

F. GIUSTI 

Université di Siena, Italy 

A. N. GOLIKOV 
Zoological Institute 
Leningrad, U.S.S.R. 

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



A. V. GROSSU 
Universitatea Bucuresti 
Romania 

T. HABE 
Tokai University 
Shimizu, Japan 

R. HANLON 

Marine Biomedical Institute 

Galveston, Texas, U.S.A. 

A. D. HARRISON 
University of Waterloo 
Ontario, Canada 

J. A. HENDRICKSON, Jr. 
Academy of Natural Sciences 
Philadelphia, PA, U.S.A. 

K. E. HOAGLAND 

Association of Systematics Collections 

Washington, DC, U.S.A. 

B. HUBENDICK 
Naturhistoriska Museet 
Göteborg. Sweden 

S. HUNT 

University of Lancaster 

United Kingdom 

R. JANSSEN 

Forschungsinstitut Senckenberg, 
Frankfurt am Main, Germany 
(Federal Republic) 

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. J. KOHN 

University of Washington 

Seattle, U.S.A. 



Y. KONDO 



A. LUCAS 

Faculté des Sciences 
Brest. France 

C. MEIER-BROOK 
Tropenmedizinisches Institut 
Tübingen. Germany (Federal Republic) 

H. K. MIENIS 

Hebrew University of Jerusalem 

Israel 

J. E. MORTON 
The University 
Auckland, New Zealand 

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

R. NATARAJAN 
Marine Biological Station 
Porto Novo. India 

J. OKLAND 
University of Oslo 
Norway 

T. OKUTANI 
University of Fisheries 
Tokyo. Japan 

W. L. PARAENSE 

Instituto Oswalde Cruz, Rio de Janeiro 

Brazil 

J. J. PARODIZ 
Carnegie Museum 
Pittsburgh, U.S.A. 

W. F. PONDER 
Australian Museum 
Sydney 

R. D. PURCHON 

Chelsea College of Science & Technology 

London, United Kingdom 

01 Z. Y. 

Academia Sínica 

Qingdao, People's Republic of China 

N. W. RUNHAM 

University College of North Wales 

Bangor, United Kingdom 



S. G. SEGERSTRÁLE 
Institute of Marine Research 
Helsinki. Finland 

F. STARMÜHLNER 

Zoologisches Institut der Universität 

Wien, Au st na 

Y. I. STAROBOGATOV 
Zoological Institute 
Leningrad, U.S.S.R. 

W. STREIFE 
Université de Caen 
France 

J. STUARDO 
Universidad de Chile 
Valparaiso 

8. TILLIER 

Muséum National d'Histoire Naturelle 

Paris, France 

R. D. TURNER 
Harvard University 
Cambridge, Mass., U.S.A. 

W. S. S. VAN BENTHEM JUTTING 
Domburg, Netherlands 

J. A. VAN EEDEN 
Potchefstroom University 
South Africa 

N. H. VERDONK 
Rijksuniversiteit 
Utrecht, Netherlands 

B. R. WILSON 

Dept. Conservation and Land Management 

Netherlands, Western Australia 

H. ZEISSLER 

Leipzig, Germany (Democratic Republic) 

A. ZILCH 

Forschungsinstitut Senckenberg 
Frankfurt am Main, Germany (Federal 
Republic) 



UNITAS MALACOLOGICA 
10th International Malacologicai Congress Syrnposium 



BIOLOGY AND EVOLUTION OF TOXOGLOSSAN GASTROPODS 



John D. Taylor 
Organizer 



29 August & 1 September 1989, 
Tubingen, Federal Republic of Germany 



Malacologia Guest Editor 
John D. Taylor 



CONTENTS 

J. D. Taylor. Introduction 

Yu. I. Kantor. Anatomical basis for the origin and evolution of the toxoglossan 

mode of feeding 3 

J. D. Taylor. The anatomy of the foreaut and relationshins with:: 

the Terebridae 19 

J. Nybakken. Ontogenetic change in the Conus radula, its iorm, distnbution 

amongst radula types, and Rlnnifiranrp in ^VRtpmRtinc; 

and ecology 

A. J. Kohn. Tempo and mode of evolution in Conidae. 

P. Bouchet. Turrid genera and mode of development: the use and abus 
protoconch morphology 

E. A. Kay. Turrid faunas of Pacific islands. 



MALACOLOGIA, 1990, 32(1): 1 



INTRODUCTION 



John D. Taylor 



Department of Zoology. The Natural History Museum, 
Cromwell Road, London SW7 5BD, U.K. 



The families of the Conoidea ( = Toxoglo- 
ssa), namely the Turhdae, Conidae, Tere- 
bridae and Pervicaciidae, are probably a 
monophyletic group, which share the autapo- 
morphy of possessing a venom gland and 
muscular bulb (presumed lost in some taxa). 
As is well known, many taxa also have highly 
modified, radular teeth which may be used 
singly at the proboscis tip for the hypodermic 
injection of venom. The relationships of the 
Conoidea to other prosobranch gastropods 
are uncertain, with some characters suggest- 
ing a relationship with the Neogastropoda, 
whilst others indicate a separate derivation 
from the mesogastropods. 

The four families differ greatly in the state of 
current knowledge. Much attention has been 
given to the species-level taxonomy of the 
Conidae, but the description of putative new 
species continues unabated. Far more is 
known about the biology and ecology of Co- 
nus than any other toxoglossan group, includ- 
ing details of their feeding, habitats and repro- 
duction. However, apart from the radula and 
the venom apparatus, there have been few 
anatomical studies, and there is no under- 
standing of relationships amongst the species 
groups or clades of Conus. 

For the Terebridae, there has been a recent 
taxonomic monograph, but little is known of 
the anatomy and biology of the family. Rela- 
tionships both within the family and with the 
other conoideans are uncertain. The anatom- 
ical data available suggest a lack of congru- 
ence between shell and anatomical charac- 
ters. Controversy surrounds the status of the 
Pervicaciidae, first proposed by Rudman for 
terebrid-like gastropods with solid radular 
teeth and no venom apparatus. 

The Turridae are an immensely diverse 
family with daunting taxonomic problems at 



all levels, with at least fifteen subfamilies in 
current use. The biology of only a few species 
is known in any detail, and the limited amount 
of anatomical work suggests an amazing di- 
versity of foregut structures within the family. 
These anatomical characters have yet to be 
incorporated into any phylogenetic analysis or 
classification. Further understanding of the 
origin and evolution of the toxoglossan feed- 
ing mechanism clearly depends upon further 
anatomical studies of more turrid species. 

It is clear from this brief summary that 
further progress in understanding the evolu- 
tion of the Conoidea depends upon a much 
more detailed knowledge of relationships, 
both within and between the conoidean fam- 
ilies. Shell characters have proved to be gen- 
erally unsatisfactory in determining relation- 
ships and much more attention needs to be 
given to the analysis of anatomical char- 
acters. In both the Turridae and Terebridae 
there are many examples of gastropods with 
similar shells having quite different internal 
anatomies. Concurrent studies on biology 
and feeding behaviour are essential to any 
understanding of the functional significance of 
both anatomical and shell characters. Addi- 
tionally, studies of larval development, partic- 
ularly in the Turridae and Conidae, are con- 
tributing data of both systematic and 
biogeographical importance. 

There has recently been an upsurge of in- 
terest in the systematics and evolution of tox- 
oglossan gastropods, and the objectives of the 
Tübingen Symposium were to bring together 
workers specializing in different groups and 
aspects of the Conoidea, to review present 
research, and to highlight areas of importance 
and interest for the future. Six of the nine pa- 
pers presented at the Symposium are pub- 
lished here. 



MALACOLOGIA, 1990, 32(1): 3-18 

ANATOMICAL BASIS FOR THE ORIGIN AND EVOLUTION OF THE 
TOXOGLOSSAN MODE OF FEEDING 

Yuri I. Kantor 

A.N.Severtzov Institute of Animal Evolutionary Morphology and Ecology. 
Academy of Sciences of the USSR, LenlnskI Prospekt 33, Moscow 117071, USSR 

ABSTRACT 

Five types of feeding mechanism can be recognized in the Toxoglossa. The mechanism by 
which separate marginal teeth are used at the proboscis tip for stabbing and poisoning the prey 
with secretions from the venom gland originated in "lower" turrids possessing a radular mem- 
brane, solid marginal teeth, a central tooth and sometimes lateral teeth. The morphological 
prerequisite of the appearance of toxoglossan mode of feeding was firstly the appearance of the 
venom gland, which initiated the formation of the specialized intraembolic type of proboscis with 
the buccal mass situated at its base. Hollow marginal teeth originated repeatedly and indepen- 
dently in different phylogenetic lineages of Toxoglossa. It is supposed that the ancestors of 
Toxoglossa were primitive mesogastropods with a short acrembolic proboscis and taenioglos- 
san radula. The separation of Toxoglossa from the Rachiglossa occurred at an early evolution- 
ary stage, when the common ancestor had seven radular teeth in each transverse row. 

Key words: Toxoglossa, evolution, feeding, radula. 



INTRODUCTION 

The order Toxoglossa is large, diverse, and 
well differentiated from the other prosobranch 
gastropods. It includes four Recent families: 
Turhdae, Conidae, Pervicaciidae, and Tere- 
bridae. One of the most outstanding and well- 
known features of the order is the specialized 
feeding mechanism of its higher representa- 
tives. That is the use of separate hollow mar- 
ginal teeth at the proboscis tip for stabbing 
and subsequent poisoning the prey, with the 
venom produced by a usually well-developed, 
tubular venom gland. Most representatives of 
the order (the "higher" Turridae, Conidae and 
part of the Terebridae) lack the radula mem- 
brane, and the radula itself consists of only 
hollow marginal teeth. The teeth being formed 
in the radula sheath are finally stored in the 
short arm of the radula sac, which is probably 
a homologue of the sublingual pouch. 

At the same time, many toxoglossans 
(mainly "lower" turrids) have a normally de- 
veloped radular membrane with two to five 
radular teeth per transverse row. Information 
on feeding mechanisms and the morphology 
of these "lower" toxoglossans is very limited, 
although the functional analysis of their diges- 
tive system and feeding mechanism may elu- 
cidate the pathways of ongin and evolution of 
"toxoglossan" mode of feeding. 

One of the most interesting problems is the 



origin of "toxoglossan" feeding mechanism. 
Does it have a single or repeated ongin in 
evolution, and what are the morphological 
prerequisites for its appearance? The main 
aim of this study was to clarify these prob- 
lems. 



MATERIALS AND METHODS 

Materials for the study were obtained 
mainly from the collections of the Zoological 
Museum of Moscow State University and In- 
stitute of Oceanology of the USSR Academy 
of Sciences (Moscow). Other material was 
kindly provided by Dr. James H. McLean (Los 
Angeles County Museum of Natural History, 
USA); the late Dr. Virginia O. Maes (Academy 
of Natural Sciences, Philadelphia, USA); Dr. 
Anders Waren (Naturhistohska Riksmuseet, 
Sweden); and Dr. R. N. Kilburn (Natal Mu- 
seum, South Africa). 

The morphology of the digestive tract was 
studied using sections 8-10 fxm thick, which 
were cut after dehydration and embedding in 
paraffin wax. The sections were usually 
stained with Massons triple stain. Its second 
solution, which contains orange-G and aniline 
blue, was used for staining the radula. Large 
specimens were also dissected under the ste- 
reomicroscope. In total, the morphology of 18 
species of Turridae belonging to six subfam- 
ilies was studied. 



KANTOR 



RESULTS AND DISCUSSION 

Within the Toxoglossa there is significant 
variability both in the morphology of the rad- 
ular teeth and their number in a transverse 
row (the radular formulae: 1-1-1-1-1, 1-0-1- 
0-1, 1-1-0-1-1, 1-0-0-0-1). The morphological 
changes in the radular apparatus and associ- 
ated structures of the anterior digestive sys- 
tem form the main evolutionary trends of the 
order. Several authors have tried to classify 
the radular types of Toxoglossa according to 
both the morphology and probable mecha- 
nism of function (Thiele, 1929: Powell. 1966: 
Morrison, 1966: McLean. 1971). The most 
complex classification was proposed by 
Shimek & Kohn (1981), who isolated six func- 
tional types of toxoglossan radula, four of 
which are found in lower "non-toxoglossate" 
turrids (those with solid marginal teeth). How- 
ever, one can say that only two general feed- 
ing mechanisms include all the isolated types: 
"toxoglossate" for those gastropods which 
have only hollow marginal teeth and lack a 
radular membrane, and "non-toxoglossate" 
for lower turrids. In the first feeding type, sep- 
arate, hollow marginal teeth are used at the 
proboscis tip for stabbing and poisoning the 
prey: in the second type, the radula is used as 
a whole organ only within the buccal cavity. In 
their analysis, Shimek & Kohn (1981) used 
mainly isolated radulae. without taking into 
account the morphology of the digestive tract, 
and this led to some misinterpretation (Sy- 
soev & Kantor, 1987). 

A functional morphological analysis of the 
digestive system of the species studied sug- 
gests that there are at least four different 
types of feeding mechanism for toxoglossans 
possessing a radula and one type for radula- 
less species. 

General Anatomy of Toxoglossa 

Before a more detailed analysis of the feed- 
ing mechanism, a brief description of the an- 
terior part of the digestive system of the Tox- 
oglossa is necessary. One of the outstanding 
features of the order is the specialized in- 
traembolic type of proboscis (Smith, 1967), 
which is characterized by the position of the 
buccal mass at the base of the proboscis or 
even behind it. This precludes the use of the 
radula as a whole organ for rasping and graz- 
ing, as in other gastropods. The second fea- 
ture is the presence of the well-developed tu- 
bular venom gland entering the anterior 



oesophagus behind the buccal cavity. It has 
been shown that the venom gland produces a 
venom that immobilizes or kills prey animals 
(Kline. 1956: Pearce, 1966: Miller, 1980: 
Shimek & Kohn, 1981: Kohn, 1956, 1959, 
1968, many others). The buccal tube leads 
from the buccal cavity to the mouth, which 
opens at the proboscis tip. The buccal tube 
has thick muscular walls in "lower" toxoglos- 
sans, but is thin-walled and practically lacking 
muscular fibres in higher representatives. 

It should be noted that the functional anal- 
ysis was carried out mainly using the anatom- 
ical evidence, because data on feeding be- 
haviour and diet are scarce and chiefly 
concern species of Conidae, Terebridae and 
some higher Turridae. As our knowledge of 
the morphology of turrids becomes more pre- 
cise, the proposed classification may change. 

Feeding Mechanism Type 1 

The first functional type of digestive system 
and feeding mechanism, that in which the 
radula is used as a whole organ only within 
the buccal cavity, was found among species 
of Pseudomelatominae (Turndae). This is an 
endemic subfamily from central west Amer- 
ica, which includes three genera and several 
species (McLean, in Keen, 1971). The anat- 
omy of two species — Pseudomelatoma peni- 
cillata (Carpenter, 1864) and Hormospira 
maculosa (Sowerby, 1834) — indicates the 
isolated position of the group among the Tox- 
oglossa (Kantor, 1988). This is obvious, in 
particular, from the presence of long curve of 
the antehor part of the digestive tract, a rarely 
found and undoubtedly secondary feature in 
turrids. The curve is formed either by elonga- 
tion of the part of the oesophagus between 
the nerve ring and the buccal mass (in 
Pseudomelatoma penicillata (Fig. 1 ), the buc- 
cal mass is situated at the proboscis base and 
far ahead of the nerve ring) or by the elonga- 
tion of the posterior part of the buccal tube (in 
Hormospira maculosa, the buccal mass is sit- 
uated in front of the nerve ring, distant from 
the proboscis base). 

Both species have a well-developed venom 
gland, longer in H. maculosa (its length com- 
prises 0.5 of the shell height). Although the 
diet of Pseudomelatominae is unknown, the 
presence of the large venom gland testifies to 
predatory mode of feeding. The gastropods 
have a muscular proboscis with a wide oral 
opening in the form of triangular or transverse 
slit and lack an oral sphincter. The radula of 



EVOLUTION OF TOXOGLOSSAN FEEDING 




FIG. 1. Anatomy of Pseudomelatoma penicillata (Carpenter). A — semidiagrammatic longitudinal section of 
the anterior part of the molluscan body. Salivary glands with the duct and convolutions of the venom gland 
together with the nerve ring are not shown. B, С — organs of the body haemocoel (B: from the left, C: from 
above). 



Pseudomelatominae consists of a large and 
well-developed central tooth, flanked by 
large, sharply pointed, scythe-like marginal 
teeth. Thus, although the morphology of the 
marginal teeth is primitive, the absence of 
lateral teeth indicates that the group has 
deviated greatly from the toxoglossan ances- 
tor. 

From the morphology of the digestive tract, 
one can suggest that prey capture probably 
occurs with the aid of the proboscis tip and is 
facilitated by the presence of a wide and 
highly extensible oral opening. The enveno- 
mation of the prey probably occurs in the an- 
terior part of the proboscis, and this facilitates 
the transportation of the prey through the 
buccal tube into the buccal cavity by the peri- 
staltic movements of well-developed circular 
muscles in walls of the buccal tube. The pres- 
ence of a very large odontophore (the largest 
of all the turrids studied) suggests that the 
radula tears the prey in the buccal cavity. 
Thus, the radula of Pseudomelatominae is of 
the slicing-rasping type as determined by 



Shimek & Kohn (1981). The large inner vol- 
ume of the buccal cavity and the curve of the 
anterior part of the digestive tract suggests 
that the prey is partially digested in the ante- 
rior part of the digestive tract. 

In summary, the main features of this feed- 
ing mechanism are; prey capture with the aid 
of the proboscis tip, without using marginal 
teeth (since the oral opening lacks a sphincter 
and the shape of the marginal teeth prevents 
their being held at the proboscis tip); use of 
the large and powerful radula for slicing and 
rasping the prey; and, what is probably a sec- 
ondary feature, at least partial digestion of the 
prey in the anterior part of the digestive tract. 
This feeding mechanism is the true "non-tox- 
oglossate" and was probably characteristic 
of ancestors of the Toxoglossa. In my opinion, 
it is widespread among turrids, and occurs 
probably in the Clavinae and other taxa lack- 
ing an oral sphincter (for example, Clavatula 
diadema), although digestion of the prey in 
the anterior part of the digestive system Is 
uncertain. 



KANTOR 



)Ь Ш'^^-'^'''''^^ 




FIG. 2. Morphology of the digestive system of Aiona spp. A — C: Aiorla abyssalis Sysoev et Kantor (A — 
semidiagrammatic section of the anterior part of the digestive system; В — magnified tip of the proboscis; 
С — radula): D — magnified tip of the proboscis of Aforia kupnyanovi Sysoev et Kantor. 



Feeding Mechanism Type 2 

The second functional type of digestive 
system IS found in some turrids with a well- 
developed radular membrane (subfamilies 
Turriculinae, Clavinae) (Sysoev & Kantor, 
1987, 1989). Its typical feature is the use of 
marginal teeth, which become detached from 
the radular membrane during its degeneration 
(in the sublingual pouch), at the proboscis tip 
for stabbing the prey. Meanwhile, the radula 
as a whole organ has a different function in 
the buccal cavity. This type of feeding mech- 
anism can be probably found amongst spe- 
cies of almost all subfamilies of Turndae, ex- 
cept the Pseudomelatominae, Zonulispirinae 
and probably the Clavatulinae. 

Since turrids belonging to this type have 
varied anatomies, it is difficult to distinguish 
morphological features common to all repre- 
sentatives of the group. For the species stud- 
ied {Aforia spp., Antiplanes spp., Splendnllia 



chathamensis Sysoev & Kantor, 1 989) the fol- 
lowing features can be noted: a large or me- 
dium-sized odontophore, with well-developed 
radular muscles; a sac-like enlargement of 
the anterior part of the buccal tube; and a 
well-developed oral sphincter. 

Individual solid marginal teeth were found 
at the proboscis tip, either held by the oral 
sphincter as in Afona (Fig. 2 B,D), or attached 
by their bases to the "mat" of epithelial cells in 
the enlargement of the buccal tube as in 
Splendnllia chathamensis (Fig. 3B). It should 
be noted that separate teeth were not found in 
the sublingual pouch. This seems to indicate 
that the marginal teeth are not used at the 
proboscis tip of Afona in every feeding act. On 
the contrary, the mechanism of tooth fixation 
in Splendnllia testifies to the long-term occur- 
rence of the tooth at the proboscis tip, i.e. the 
enlargement of the anterior part of the buccal 
tube may be considered as a functional ana- 
logue of the short arm of the radular sac. 



EVOLUTION OF TOXOGLOSSAN FEEDING 
rs sg Sd 




FIG. 3. Anatomy of Splendrillia chatamensis Sysoev & Kantor. A — semidiagrammatic longitudinal section of 
the anterior part of molluscan body; В — magnified tip of the proboscis. 



Transportation of teeth to the proboscis tip in 
Aforia may occur with the flow of venom dur- 
ing contraction of the muscular bulb or also by 
peristaltic movements of circular muscle fi- 
bres of the buccal tube. Splendrillia chatha- 
mensis has an additional, well-developed 
sphincter in the middle part of the buccal tube 
(Fig. ЗА, spt), which probably takes part in the 
transportation of the tooth. The marginal tooth 
is detached from the membrane and is 
pushed into the buccal cavity by the contract- 
ing walls of the buccal sac. The tooth length is 
about 1/3-1/4 of the contracted proboscis 
length. During the contraction of the proximal 
part of the proboscis, the tooth becomes held 
by the additional sphincter. When the distal 
part of the proboscis contracts, the tooth is 
passed into the oral sphincter. 

The function of the radula as a whole organ 
within the buccal cavity is most probably for 
the transport of food from the cavity to the 
oesophagus. This may be confirmed, in par- 
ticular, by the observations of Maes (1981), 
who noted the presence of intact sipunculans 
in the posterior part of the oesophagus of Dril- 
lia cydia (Bartsch, 1943) (Clavinae), although 
the large, pectinate lateral teeth might at first 



sight be thought to serve for tearing or rasping 
the prey. 

The use of marginal teeth at the proboscis 
tip in turrids with a well-developed radular 
membrane is probably a widespread phe- 
nomenon amongst the Turhdae. This may ex- 
plain the origin of hollow marginal teeth in 
different groups possessing the radular mem- 
brane and odontophore. For example, Ima- 
clava (Clavinae), most probably also uses the 
teeth at the proboscis tip for stabbing the prey 
in a way similar to higher toxoglossans. 

In summary, the main features of this feed- 
ing mechanism are: the detachment of mar- 
ginal teeth from the radular membrane during 
its degeneration; transportation of the teeth to 
the proboscis tip; and their use for damaging 
and poisoning the prey with the venom. A fea- 
ture of the proboscis is the sac-like enlarge- 
ment of the anterior part of the buccal tube, 
with the sphincter holding the base or the mid- 
dle part of the tooth. The function of the radula 
as a whole organ is mainly for the transport of 
the food from the buccal cavity to the oesoph- 
agus, although in some turrids it may be used 
also for tearing and rasping. This could be 
confirmed by the investigation of the prey ob- 



KANTOR 



Sd bm 




FIG. 4. Morphology of the digestive system of Toxiclionella túmida (Sowerby). A — semidiagrammatic lon- 
gitudinal section of the anterior part of the digestive system. The convolution of the venom gland and 
posterior part of the radular sac are not shown, the arrow indicates the entrance of the venom gland in the 
oesophagus: B— marginal tooth: C— the tip of the tooth, enlarged. 



tained from the buccal cavity and anterior oe- 
sophagus. Thus, the slicing, slicing-rasping, 
and slicing-stabbing types of radula described 
by Shimek & Kohn (1981) belong principally 
to the same functional type, which may be 
named ■stabbing-transporting" type. 

Feeding Mechanism Type 3 

The third feeding mechanism has been 
found so far only in a single species of Tur- 
ridae, Toxiclionella túmida (Sowerby, 1870) 
(Clavatulinae), although it probably exists in 
other species of this endemic south African 
genus. A feature of the morphology of the di- 
gestive system IS the position of the buccal 
mass, with the odontophore near the probos- 
cis tip (Fig. 4A). The oral sphincter is absent. 
The gastropod has a well-developed, long 
venom gland and an unpaired salivary gland 
with paired ducts, which is situated in the pos- 
terior part of the proboscis. The radula con- 
sists only of hollow, marginal teeth which are 
morphologically similar to the teeth of higher 
turrids (Fig. 4B,C); a radular membrane is 
present. The teeth are sufficiently long (the 



tooth, at the same scale, is figured above the 
proboscis on Fig. 4) that during protraction of 
the odontophore the tips would protrude 
through the oral opening. This leads to the 
conclusion that the mollusc uses the radula 
as a whole organ for stabbing the prey. 

The main difference of this mechanism 
from all others in which a marginal tooth is 
used for stabbing and poisoning the prey, is 
that the radula is used as a whole organ, not 
as separate teeth. It is possible that a similar 
mechanism occurs in Turricula nelliae spurius 
(Hedley, 1922) (Taylor, 1985), which has a 
similar proboscis morphology. 

Feeding Mechanism Type 4 

The fourth feeding mechanism was found 
in higher toxoglossans that lack a radular 
membrane, i.e. higher Turridae, Conidae and 
some Terebridae. The main feature of the 
mechanism is the use of individual, hollow 
marginal teeth at the proboscis tip for stab- 
bing the prey, and the completely reduced 
function of the radula as a whole organ within 
the buccal cavity. The feeding and the diet of 



EVOLUTION OF TOXOGLOSSAN FEEDING 




FIG. 5. Anatomy of Teretiopsis abyssalis Kantor & Sysoev. A — semldiagrammatic longitudinal section of thie 
anterior part of thie molluscan body; В — enlarged part of the section through the body wall and rhyn- 
chodaeum. 



species with this functional type is well 
known, and it is unnecessary to describe it in 
detail. Only the most important morphological 
features should be noted. These are the ves- 
tigial, or completely reduced, radular mem- 
brane; the absence of an odontophore; the 
presence of the short arm of the radular sac, 
where the fully formed marginal teeth are 
stored; and a well-developed, oral sphincter 
for tooth fixation. The radula is represented 
only by hollow marginal teeth, with the most 
specialized and complex morphology found 
within the prosobranchs. The tooth ligament 
(long flexible stalk attached to the tooth base) 
is probably the rudiment of the radular mem- 
brane. Amongst molluscs of this functional 
group, the enlarged rhynchostomal lips ap- 
peared. In some species, the lips are able to 
invert (i.e. to form a pseudoproboscis) and 
this is used in prey capture. It should be noted 
that in some representatives of the group — 
some vermivorous species of Conus (Marsh, 
1970) and С geographus L., 1758 (Johnson 
& Stablum, 1971) — stabbing is not a neces- 
sary part of each feeding act. 

Judging from the morphology of the diges- 
tive system, Zonulispirinae occupy an inter- 
mediate position between the gastropods of 
the second and the fourth functional groups. 
They have hollow marginal teeth, attached to 



a rather strong radular membrane. This may 
indicate that separate teeth are used at the 
proboscis tip. Moreover, the gastropods have 
very small odontophore (Maes, 1983); this in- 
dicates that the function of the radula as a 
whole organ within the buccal cavity is prob- 
ably rudimentary. 

Feeding Mechanism Type 5 

The fifth and last functional type is found 
among those Toxoglossa lacking a radula. 
Gastropods of this group belong to higher 
Turridae (according to the shell morphology) 
and some Terebridae. The most important 
features are: a reduced or completely absent 
proboscis; and absence of a radular sac, and 
venom and salivary glands. Most representa- 
tives of this group have either well-developed 
rhynchostomal lips or a large pseudoprobos- 
cis (Terebridae — Miller, 1975; Philbertia lin- 
earis (Montagu), Turridae — Sheridan et al., 
1973). Some turrids {Cenodagreutes spp. — 
Smith, 1967; Abyssobella atóxica Kantor & 
Sysoev — Kantor & Sysoev, 1986; Teretiopsis 
spp. — Kantor & Sysoev, 1989), lacking a 
pseudoproboscis, have a vast rhynchocoel 
and have developed a cavity between the 
rhynchodaeum and body walls, which are 
connected by numerous muscles in the cavity 



10 



KANTOR 



(Fig. 5). Species of the genus Taranis lack 
both a pseudoproboscis and a cavity. 

The feeding mechanism is known for tere- 
brids (Miller, 1970, 1975). Thus, species with 
a relatively short pseudoproboscis feed on 
the enteropneust Ptychodera flava, and spe- 
cies with a long pseudoproboscis feed on 
polychaetes. The capture and engulfment of 
the prey occurs with the aid of the pseudopro- 
boscis. Turrids lacking a pseudoproboscis, 
but with a cavity between the rhynchodaeum 
and the body walls, probably engulf the prey 
with the aid of negative pressure, which 
arises in the rhynchocoel during contraction 
of the radial muscle fibres (at that moment the 
inner volume of the rhynchocoel increases). It 
is difficult at present to say anything certain 
about the feeding mechanism of Taranis. 

The feeding of such aberrant groups as 
Strictispinnae (Turridae) is unclear. These 
gastropods lack a venom gland and have a 
very large odontophore. According to the fig- 
ure of Maes (1983), Stnctispira paxillus 
(Reeve, 1845) has a short buccal tube. Thus, 
there is a possibility that it can protrude the 
radula through the mouth opening and use it 
pincer-like, teanng off small pieces of food. 

Origin of the Toxoglossan Mode of Feeding 

In my opinion, the development of the 
unique 'toxoglossan" mode of feeding is con- 
nected with certain morphological prerequi- 
sites. These were the appearance of the 
venom gland and the intraembolic type of the 
proboscis. 

The mobile proboscis, which in the con- 
tracted state is situated in the special cavity of 
the body haemocoel, or proboscis-like struc- 
tures (for example, the extrovert formed by 
the walls of the buccal cavity in Janthinidae — 
Graham, 1965) appeared independently in 
different groups of marine predatory gastro- 
pods. The presence of the proboscis allows 
an increase in the mobility of the buccal mass, 
and this is achieved by its shift from the ven- 
tral side of the head (as in herbivorous gas- 
tropods) in the terminal (axial) position. This 
also allows ■distant" feeding, i.e. to feed on 
prey hidden in burrows, crevices, etc., and 
also on animals with external skeletons, for 
example on bivalves (inserting the proboscis 
between the open valves or through a drilled 
hole). 

Usually three types of proboscis are de- 
fined; acrembolic, pleurembolic, and intraem- 
bolic, these are differentiated by the position 



of the buccal mass and the mode of eversión. 
Only the latter two types are found among 
Neogastropoda. In gastropods with the pleu- 
rembolic proboscis, the buccal mass with rad- 
ular sac is situated near the proboscis tip, and 
proboscis eversión occurs with the aid of 
the posterior invaginable part of the rhyn- 
chodaeum (wall of the proboscis sheath or 
rhynchocoel). In many neogastropods with 
this proboscis type, the entire or nearly entire 
rhynchodaeum takes part in proboscis ever- 
sión. On the contrary, in gastropods with the 
intraembolic proboscis, the buccal mass is sit- 
uated at the proboscis base or even behind it 
{Pseudomelatoma penicillata, Turridae — Fig. 
1), the invaginable part of the rhynchodaeum 
is absent, and the proboscis eversión results 
only from its stretching. Recently, a proboscis 
somewhat intermediate between the typical 
pleurembolic and intraembolic types was de- 
scribed in Turricula nelliae spun us (Taylor, 
1985) and Toxiclionella túmida (herein). In 
these gastropods, the buccal mass is situated 
near the proboscis tip, and the rhynchodaeum 
is capable of partial eversión. 

Usually, the Neogastropoda are considered 
as a monophyletic group (Ponder, 1973; Tay- 
lor & Morris, 1988). On the other hand, doubts 
on the monophyletic origin of neogastropods 
were expressed by Golikov & Starobogatov 
(1975), with moreover the Toxoglossa (sensu 
Golikov & Starobogatov who included Mitroi- 
dea along with Conoidea and Terebroidea in 
the order) were separated from the rest. The 
problem of the ancestral group is also essen- 
tial to the argument. Ponder (1973) consid- 
ered that the Neogastropoda onginated from 
archaeogastropods or phmitive mesogastro- 
pods. Thus, the proboscis of neogastropods 
in general and of Toxoglossa in particular 
should be considered as de novo structure. 
Taylor & Morris (1988), on the contrary, sug- 
gested the possibility of the ongin of Neogas- 
tropoda from higher, advanced Mesogas- 
tropoda and their probóscides thus should be 
homologous with the pleurembolic proboscis 
of predatory Mesogastropoda. Finally, Sheri- 
dan et al. (1973) stated that the intraembolic 
type of the proboscis originated from the 
acrembolic type. 

For more careful consideration of the ques- 
tion some comments on the morphology of 
the buccal muscles are necessary. 

In archaeogastropods and primitive meso- 
gastropods lacking a proboscis, there are nu- 
merous buccal muscles that are connected to 
the columellar and pedal muscles. On the 



EVOLUTION OF TOXOGLOSSAN FEEDING 



11 



contrary, in Mesogastropoda and Neogas- 
tropoda with a developed pleurembolic pro- 
boscis, the buccal muscles have lost such a 
connection and are attached to the proboscis 
walls (Graham, 1973; herein). In a species of 
Clavinae, which are considered to be the 
least-derived Toxoglossans, there is such a 
connection of supramedian, radular tensor 
and columellar muscles (Fig. ЗА). In my opin- 
ion, this undoubtedly confirms the original 
basal position of the buccal mass in Clavinae. 
In the opposite case, the connection of the 
buccal and columellar muscles would be lost. 
Thus, one can state that the intraembolic pro- 
boscis has evolved independently from the 
pleurembolic type and not from the latter (by 
the shift of the buccal mass to the proboscis 
base) and that the ongin of Toxoglossa and all 
Neogastropoda in general (if they are consid- 
ered as a monophyletic group) from higher 
probosciferous mesogastropods is improba- 
ble. Probóscides of different groups of Neo- 
gastropoda probably appeared indepen- 
dently, and the detailed morphological studies 
of some poorly known groups would corrobo- 
rate this supposition. 

The appearance of the intraembolic pro- 
boscis in Toxoglossa may be connected with 
appearance and development of the venom 
gland. It is very likely that toxoglossan ances- 
tors were carnivorous gastropods with a short 
acrembolic proboscis. The acrembolic pro- 
boscis is found among various primitive gas- 
tropods (for example, Naticidae, Triphoridae, 
Cerithiopsidae) and principally may be con- 
sidered as an elongated buccal tube that has 
an ability to evert through the mouth opening 
as a glove finger. In the inverted position, the 
buccal mass is situated at the base of the 
proboscis, while in an everted position it is 
located at the proboscis tip (Fig. 6 A). During 
proboscis eversión the oesophagus is pulled 
through the nerve ring. 

The elongation of the acrembolic proboscis 
allows gastropods to feed on animals hidden 
in deep burrows, crevices or tubes, for exam- 
ple on polychaetes. At the same time, the 
elongation of the proboscis limits the size of 
the oesophageal glands, which have to be 
pulled through the nerve ring during eversión. 

It could be suggested that at early evolu- 
tionary stages, these gastropods started to 
use the secretion produced by the dorsal 
glandular folds of the oesophagus and 
squirted through the mouth for immobilization 
of the prey. This simplified the capture and 
swallowing of actively moving prey. After the 



appearance of such feeding mechanism, the 
proboscis may have elongated by the devel- 
opment of a tube in front of the mouth open- 
ing, which was situated in the sheath formed 
by the walls of introvert of the acrembolic pro- 
boscis (Fig. 7B). The main function of the pro- 
boscis was not to move the buccal mass for- 
ward, but to form the tube through which the 
venom reaches the prey. 

Such elongation of the proboscis appears 
closely related to the enlargement of the dor- 
sal oesophageal folds; as the inner volume of 
the proboscis grew, more venom was neces- 
sary to fill it. Gradually the glandular folds 
stripped off from the oesophagus and formed 
a tube, i.e. the venom gland. In the initial 
stages of the formation of the new proboscis 
type, the introvert was probably able to evert, 
but the enlarged size of the venom gland pre- 
vented its being pulled through the nerve ring. 
Finally, this caused fixation of the buccal 
mass in front of the nerve ring at the probos- 
cis base, and the introvert ceased to evert. At 
that moment, the newly formed proboscis 
possessed all features of the intraembolic 
type (Fig. 60). The functions of the radula 
were the same as in other gastropods (tearing 
and rasping the prey and its transportation to 
the oesophagus), but it acted only within the 
buccal cavity. 

If this proposed scheme of origin of the in- 
traembolic proboscis is accepted, then one 
can suppose that the rhynchodaeum is a ho- 
mologue of the introvert wall of the acrembolic 
proboscis and the proboscis itself is de novo 
structure that is not homologous with the 
pleurembolic proboscis of other neogastro- 
pods. 

The discovery of a mechanism by which 
individual solid marginal teeth are used at the 
proboscis tip in turrids with a well-developed 
radular membrane, allows us to reconstruct 
the development of the typical "toxoglossan" 
mode of feeding. In the process of radula 
growth, anterior (the oldest) rows of teeth are 
detached from the radular membrane, which 
in turn degenerates in the sublingual pouch. It 
is reasonable to suppose that some detached 
teeth are not removed through the digestive 
tract (as usually occurs in gastropods) but are 
somehow transported to the proboscis tip 
where they are used for damaging the prey 
integument. This intensifies the efficiency of 
venom action. Fixation of such a mechanism 
in evolution created the prerequisites and ne- 
cessity of the appearance of hollow marginal 
teeth. This was an important stage in toxo- 



12 



KANTOR 




FIG 6 A scheme for the origin and evolution of the proboscis of Toxoglossa. The dotted arrows indicate 
hypothetical connections. The hypothetical morphological stage is given on the dotted background. A— 
acrembolic proboscis of the ancestral group; B— intermediate morphological stage between the acrembolic 
and intraembolic proboscis types; C— the basal type of the intraembolic proboscis; D— origin of rhynchos- 
tomal lips; E— reduction of the proboscis, radula and venom and salivary glands; F— ongin of pseudopro- 
boscis- G— reduction of the proboscis, radula and venom and salivary glands; H— displacement of the 
buccal mass toward the proboscis tip and formation of the curve of the digestive tract; I— formation of the 
radial folds at the proboscis base. 



EVOLUTION OF TOXOGLOSSAN FEEDING 

rs 



13 




FIG. 7. The anatomy of Benthobia n.sp. A — semidiagrammatic longitudinal section of the anterior part of the 
molluscan body; В — radula. 



glossan evolution. As the mechanism of prey 
stabbing and poisoning by the teeth at the 
proboscis tip improved, the functions of the 
radula as a whole organ within the buccal 
cavity became less and less important. This 
finally led to reduction of the odontophore, 
central and lateral teeth, and as a final stage, 
the radular membrane. 



Up until now the intraembolic proboscis has 
been found only amongst Toxoglossa. How- 
ever, a similar proboscis type was found by 
the author in a species of the family Pseudo- 
lividae (Fig. 7), Benthobia n.sp. The buccal 
mass in this species is situated at the probos- 
cis base; moreover, there is a connection of 
the buccal muscles with the columellar mus- 



14 



KANTOR 



de (this confirms the primary position of the 
buccal mass, see above). This gastropod also 
has a very large gland of Leiblein. According 
to Ponder (1973), the venom gland of the 
Toxogiossa and the gland of Leiblein were 
formed independently, but in similar way. by 
the stripping off of the glandular folds from 
the oesophagus. Thus, one can state that 
Pseudolividae and Toxogiossa are not related 
groups, and the similar proboscis type ap- 
peared independently. The radular morphol- 
ogy of Benthobia (Fig. 7B) (very similar to 
Olividae), as well as details of the morphology 
of the anterior part of the digestive tract, indi- 
cates that the marginal teeth are not used at 
the proboscis tip by Benthobia. Thus, the de- 
velopment of the venom gland rather than the 
position of the buccal mass at the proboscis 
base, was the main factor conditioning the ap- 
pearance of "toxoglossan" mode of feeding in 
evolution. 

The origin of hollow marginal teeth took 
place repeatedly and independently in differ- 
ent phylogenetic lineages of Toxogiossa. Hol- 
low marginal teeth appeared at least twice 
among turrids, simultaneously with the reten- 
tion of the radular membrane and the central 
and sometimes lateral teeth. Radulae of this 
type are found among Imaclava unimaculata 
(Sowerby, 1843) (Clavinae) (Shimek & Kohn, 
1981) and Toxiclionella elstoni (Barnard, 
1962) (Clavatulinae) (Kilburn, 1985). 

The main trends of subsequent evolution of 
the Toxogiossa are variable and character- 
ized by morphological changes in the anterior 
part of the digestive system. Thus, three mam 
pathways of the morphological evolution of 
the proboscis may be defined. Some Toxo- 
giossa have circular folds formed by the pro- 
boscis in the contracted state (Fig. 61). This 
reduces the length of the contracted probos- 
cis, and probably simplifies the transportation 
of the individual marginal teeth from the rad- 
ular sac to the proboscis tip. 

The second lineage is connected with ori- 
gin and development of the mobile rhyncho- 
stomal lips, which take part in the prey cap- 
ture (Fig. 6D). The progressive development 
of lips into an introvert results in the 
pseudoproboscis (Fig. 6F) of some turrids 
and most Terebridae. The action of prey cap- 
ture gradually transferred from the proboscis 
to the rhynchostomal lips or pseudoprobos- 
cis, and this finally led to the complete reduc- 
tion of the true proboscis (Fig. 6E, G). The 
process is evolutionarily connected with the 
complete reduction of the radula, venom and 



salivary glands, a process that occurred inde- 
pendently in different phylogenetic lineages. 

Finally, the third, less studied trend is con- 
nected with the shift of the buccal mass to- 
ward the proboscis tip. Also the rhyn- 
chodaeum secondarily evolved the capability 
of partial eversión (this was made possible by 
the elongation of the oesophagus between 
the buccal mass and the nerve ring and for- 
mation of the curve of the oesophagus, as it 
takes place in Rachiglossa). The tendency is 
best seen in Tumcula nelliae spurius (Turri- 
culinae) and Toxiclionella túmida (Clavatuli- 
nae). An intermediate morphological stage is 
found in Clavatula diadema (Kiener, 1840) 
(Fig. 8), m which the buccal mass is situated 
inside the proboscis nearer to its base, 
and the rhynchodaeum is capable of partial 
eversión. The presence of two consecutive 
morphological stages in the same subfamily 
confirms the secondary character of this ev- 
olutionary lineage. 

In conclusion, one more question should be 
discussed, the resolution of which may shed 
light on the ancestral group of Neogastropoda 
and Toxogiossa, Usually, the radula of Clav- 
inae (Turndae), with a central tooth, flanked 
by pairs of lateral and marginal teeth, is con- 
sidered as a plesiomorphic condition in neo- 
gastropods (Taylor & Morris, 1 988). In pectini- 
branchiate gastropods, the radula is folded 
lengthways in the radular sac. The folds in the 
gastropods with differentiated groups of teeth 
are situated between the marginal and lateral 
teeth and between the lateral and central 
teeth. Thus, in gastropods with marginal and 
lateral teeth, there are two pairs of folds. A 
similar condition is observed in Olivella, ex- 
cept for the Clavinae, the only genus of neo- 
gastropods which has five teeth in a trans- 
verse row (Fig. 9A). Nevertheless, the clavine 
radula has only one pair of folds, which are 
situated between the central and lateral teeth 
(Fig. 9B). Species of the genus Antiplanes 
have the central formations which were con- 
sidered as a reduced central tooth. Investiga- 
tions of the radula indicate that it has only one 
fold (Fig. 9C). This may indicate that tradi- 
tional interpretation of the radular teeth of 
Clavinae is wrong and their radula is formed 
by central and two pairs of marginal teeth, 
which have become greatly differentiated in 
evolution. In Antiplanes, the central tooth is 
possibly completely reduced, and the central 
formations are the rudiments of the inner pair 
of the marginal teeth. 

Thus, one can suppose that the toxoglossan 



EVOLUTION OF TOXOGLOSSAN FEEDING 



15 



pg oe 




FIG. 8. The anatomy of Clavatula diadema (Kiener). A — semidiagrammatic longitudinal section of the an- 
terior part of the moiiuscan body; В — magnified base part of the proboscis. 



16 



KANTOR 





A 







FIG. 9. The radular folding m different Neogastropoda. At the left— the shape of radular teeth, at the 
right— diagrammatic transverse section of the radula sheath, A—Olivella; B^Splendrillia (Clavinae); C— 
Antiplanes (Turriculinae). 



radula originated from the taenioglossan (2- 
1-1-1-2) by the reduction of true lateral teeth 
and differentiation of the nnarginals. (The rad- 



ular fornnula of Clavinae should be 2-0-1-0-2, 
of Antiplanes. 2-0-0-0-2.) On the contrary, in 
Olivella the radula is formed by the rudiments 



EVOLUTION OF TOXOGLOSSAN FEEDING 



17 



of the marginal teeth, by a pair of laterals and 
a central, i.e. it originated from the taenio- 
glossan by a reduction of the pair of marginal 
teeth (formula: 1-1-1-1-1). The present hy- 
pothesis supposes that if the Neogastropoda 
is a monophyletic group, their ancestor had 
the taenioglossan radula, and the derivation 
of Rachiglossa and Toxoglossa occurred at 
an early stage, when the ancestor had seven 
teeth per transverse row. 



CONCLUSIONS 

(1) The evolution of Toxoglossa as a sep- 
arate taxon was connected with the origin and 
development of the venom gland. The devel- 
opment of the venom gland determined the 
appearance of the specialized intraembolic 
type of proboscis and the specific "toxoglos- 
san" mode of feeding. 

(2) The ancestors of Toxoglossa were 
probably lower mesogastropods with a short 
acrembolic proboscis and taenioglossan rad- 
ula. 

(3) In higher Toxoglossa, the specific "tox- 
oglossan" mode of feeding, using separate, 
hollow marginal teeth at the proboscis tip, has 
originated repeatedly and independently in 
the Turridae. A similar feeding mechanism 
with the use of solid marginal teeth at the pro- 
boscis tip in some lower turrids with a well- 
developed radular membrane and odonto- 
phore may be considered as the intermediate 
evolutionary stage. 

(4) In "higher" Toxoglossa with well-devel- 
oped rhynchostomal lips or with a pseudopro- 
boscis, a decrease of the proboscis size usu- 
ally occurs and this leads finally to the 
complete reduction of the radula, venom and 
salivary glands. 



ABBREVIATIONS 

asg — accessory salivary gland; be — buccal 
cavity; bm — buccal mass; bt — buccal tube; 
cm — columellar muscle; cu — cuticle; epm — 
"mat" of epithelial cells; gf — glandular folds of 
the oesophagus; m — buccal muscle, con- 
nected to the columellar muscle; mb — mus- 
cular bulb of the venom gland; nr — nerve ring; 
od — odontophore; oe — oesophagus; pg — 
venom gland; pr — proboscis; prr — proboscis 
retractor muscles; ps — proboscis sphincter; 
pw — proboscis wall; r — radula; rd — rhyn- 
chodaeum; rm — radial muscles, connecting 



the rhynchodaeum and the body wall; rs — 
radular sac; rt — marginal tooth, held at the 
proboscis tip; sd — salivary duct; sg — salivary 
gland; sip — sublingual pouch; spt — intermedi- 
ate sphincter of the buccal tube; sr — rhyncho- 
stomal sphincter. 



ACKNOWLEDGEMENTS 

The author is greatly indebted to Dmitri 
Ivanov, Zoological Museum of Moscow State 
University, James H. McLean (Los Angeles 
County Museum of Natural History), Anders 
Waren (Naturhistohska Riksmuseet), and R. 
N. Kilburn (Natal Museum) who kindly pro- 
vided materials for the study, and to Dr. Alex- 
ander Sysoev for his valuable comments on 
the manuscript. 



LITERATURE CITED 

GOLIKOV, A.N. & Y.I. STAROBOGATOV. 1975. 
Systematics of prosobranch gastropods. Malaco- 
logia. 15: 185-232. 

GRAHAM, A., 1965, The buccal mass of janthinid 
prosobranchs. Proceedings of the Malacological 
Society of London. 36: 323-338. 

GRAHAM, A., 1973, The anatomical basis of func- 
tion in the buccal mass of prosobranch and am- 
phineuran molluscs. Journal of Zoology. London. 
169: 317-348. 

JOHNSON, C. R. & W. STABLUM, 1 971 , Observa- 
tions on the feeding behaviour of Conus geogra- 
phius (Gastropoda: Toxoglossa). Pacific Science. 
25: 109-111. 

KANTOR, Yu. I., 1988, On the anatomy of 
Pseudomelatominae (Gastropoda, Toxoglossa, 
Turridae) with notes on functional morphology 
and phylogeny of the subfamily. Apex. 3: 1-19. 

KANTOR, Yu. I. & A. V. SYSOEV, 1986, A new 
genus and new species from the family Turridae 
(Gastropoda, Toxoglossa) in the northern part of 
the Pacific Ocean. Zoologichesklj Zhiurnal. 65: 
485-498. (In Russian). 

KANTOR, Yu. I. & A. V. SYSOEV, 1989, On the 
morphology of toxoglossan gastropods lacking a 
radula, with a description of new species and ge- 
nus of Turridae. Journal of t\/lolluscan Studies. 
55: 537-549. 

KILBURN, R. N.. 1985, Turridae (Mollusca: Gas- 
tropoda) of southern Africa and Mozambique. 
Part 2. Subfamily Clavatulinae. Annals of Natal 
Museum. 26: 417-470. 

KLINE, P., 1956, Notes on the stinging operation of 
Conus. Nautilus. 69: 76-78. 

KOHN, A. J., 1956, Piscivorous gastropods of the 
genus Conus. Proceedings of the National Acad- 
emy of Sciences. 42: 1 68-1 71 . 



Il 



KANTOR 



KOHN, A. J., 1959. The ecology of Conus in Ha- 
waii. Ecological Monographs. 29: 47-90. 

KOHN. A. J.. 1968. Microhabitats. abundance and 
food of Conus on atoll reefs in the Maldive and 
Chagos Islands. Ecology. 49; 1046-1061. 

MAES. V. O.. 1983. Observations on the systemat- 
ics and biology of a turrid assemblage in the Brit- 
ish Virgin Islands. Bulletin of Manne Sciences. 
33; 305-335. 

MARSH, H., 1970. Preliminary studies of the ven- 
oms of some vermivorous Conidae. Toxicon. 8: 
271-277. 

McLEAN, J. H-, 1 971 . A revised classification of the 
family Turridae. with the proposal of new subfam- 
ilies, genera, and subgenera from the eastern 
Pacific. Veliger. 14; 114-130. 

McLEAN. J. N., 1971, Family Turridae. In KEEN. A. 
M. Sea shells of tropical West America: marine 
molluscs from Baja California to Peru. Second 
ed. Stanford; Stanford University Press, pp. 686- 
766. 

MILLER. B. A.. 1970. Studies on the biology of 
some Indo-Pacific Terebndae. Ph. D. thesis. Uni- 
versity of New Hampshire. Dover. 213 pp. 

MILLER. B. A., 1975. The biology of Terebra gouldi 
Deshayes and a discussion of life history similar- 
ities among other terebrids of similar proboscis 
type. Pacific Science. 29; 227-241. 

MILLER. B. A.. 1980. The biology of Hastula incon- 
stans (Hinds. 1844) and a discussion of life his- 
tory similarities among other hastulas of similar 
proboscis type. Pacific Science. 33; 289-306. 

MORRISON. J. P. E., 1966, On the families of Tur- 
ridae. Annual Report of the American Malacolog- 
ical Union for 1965: 1-2. 

PEARCE. J. В.. 1966. On Lora treveliana (Turton) 
(Gastropoda; Turridae). Ophelia. 3; 81-91. 

PONDER. W. F.. 1973. Origin and evolution of the 
Neogastropoda. Malacologia. 12; 295-338. 

POWELL. A. W. В.. 1966. The molluscan families 



Speightiidae and Turridae. An evaluation of the 
valid taxa, both Recent and fossil, with lists of 
characteristic species. Bulletin of the Auckland 
Institute and Museum. 5; 184 pp. 

SHERIDAN. R.. J. -J. VAN MOL & J. BOUILLON, 
1973. Etude morphologique du tube digestif de 
quelques Turridae de la région de Roscoff. Cah- 
iers de Biologie Manne. 14; 159-188. 

SHIMEK. R. L. & A. J. KOHN, 1981, Functional 
morphology and evolution of the toxoglossan 
radula. Malacologia. 20; 423-438. 

SMITH. E. N.. 1967. The proboscis and oesopha- 
gus of some British turrids. Transactions of the 
Royal Society of Edinburgh. 67; 1-22. 

SYSOEV. A. V. & Yu. I. KANTOR, 1987. Deep-sea 
gastropods of the genus Afona (Turridae) of the 
Pacific; species composition, systematics, and 
functional morphology of the digestive system. 
Veliger. 30; 105-126. 

SYSOEV. A. V. & Yu. I. KANTOR. 1989. Anatomy 
of molluscs of genus Splendrillia (Gastropoda; 
Toxoglossa; Turridae) with description of two 
new bathyal species of the genus from New 
Zealand. New Zealand Journal of Zoology. 16; 
205-214. 

TAYLOR. J. D., 1985. The anterior alimentary sys- 
tem and diet of Turricula nelliae spunus (Gas- 
tropoda; Turndae). In; MORTON, B. & D. DUD- 
GEON, eds.. Proceedings of the Second 
International Workshop on the Malacofauna of 
Hong Kong and Southern China. Hong Kong 
1983. Hong Kong University Press, pp. 175-190. 

TAYLOR, J. D. & N. J. MORRIS, 1988, Relation- 
ships of neogastropods. Malacological Review, 
Supplement 4: 167-179. 

THIELE, J., 1929 [-1931]. Handbuch der system- 
atischen Weichtierkunde. Jena. Gustav Fischer, 
1929-1935; 1154 pp. 

Revised Ms. accepted 21 June 1990 



MALACOLOGIA, 1990, 32(1): 19-34 

THE ANATOMY OF THE FOREGUT AND RELATIONSHIPS IN THE TEREBRIDAE 

John D. Taylor 

Department of Zoology. The Natural History Museum, Cromwell Road, London 
SW7 5BD, United Kingdom 

ABSTRACT 

A study of foregut anatomy in the gastropod family Terebrldae shows that two major groups 
of species are represented. Members of one group have hypodermic radular teeth, venom 
apparatus and an extensible buccal tube. Terebrids of the second group have a very short 
buccal tube, a radula consisting of two rows of solid, curved teeth and no venom apparatus. 
Furthermore, there are many terebrid species lacking a radula, venom apparatus and buccal 
tube and these could be derived from either group. 

It is suggested that the two groups of Terebridae were derived independently from the Tur- 
ridae, and each group should be given family status. This study confirms Rudmans's (1969) 
proposal of the family Pervicaciidae for Terebra tnstis. and the family should now be extended 
to include perhaps all species of Duplicaría and a number of species currently referred to 
Terebra. 

Key words: Terebridae, anatomy, feeding, functional morphology, radula. 



INTRODUCTION 

The Terebridae is a family of conoidean 
gastropods characterised by high-spired, 
multiwhorled shells with relatively small aper- 
tures. There are about 300 living species 
which inhabit soft-substrate habitats at tropi- 
cal and sub-tropical latitudes (Bratcher & Cer- 
nohorsky, 1987). The family is particularly di- 
verse and abundant in shallow-water sandy 
habitats of Indo-Pacific coral reefs (Miller, 
1970; Taylor, 1986). The biology of Tere- 
bridae has been little studied and not only is 
little known about relationships within the 
family, but the relationship of terebrids with 
other conoideans is also obscure. 

Anatomical work on the Terebridae has 
been very limited. Details of individual spe- 
cies have been described by Risbec (1953), 
Marcus & Marcus (1960), Rudman (1969), 
Auffenberg & Lee (1988), and Taylor & Miller 
(1990). By far the most significant studies 
were by Miller (1 970, 1 971 , 1 975, 1 979). As a 
result of these studies. Miller (1970, 1971) 
proposed a classification of proboscis types 
within the family Terebridae which he thought 
represented natural groupings, although he 
made no attempt to examine relationships 
within the family. 

Miller (1970, 1971) recognised three main 
proboscis types amongst the species he stud- 
ied. They are briefly defined as follows: 



Type la species have a long, extensible la- 
bial tube or introvert, a short buccal tube, no 
radula, a pair of salivary glands, and no 
venom gland or muscular bulb. 

Type lb species are similar to type la in 
foregut anatomy, but have a very long labial 
tube, which is folded upon itself when re- 
tracted into the rhynchodeal cavity. 

Type IIa species have a medium length la- 
bial tube, a long proboscis and buccal tube, a 
pair of fused salivary glands, a radula sac and 
caecum, with hollow hypodermic radular teeth, 
a large venom gland and muscular bulb. 

Type lib species are similar, but the buccal 
tube is shorter and the rhynchodeum may be 
partitioned by a septum. 

Type III species have a labial tube of me- 
dium length, a very short or no buccal tube, 
salivary glands are vestigial or absent, and 
there is no radular or venom apparatus. A fea- 
ture of this group of species is the presence of 
a club-shaped accessory feeding organ at- 
tached to the left wall of the rhynchodeal cav- 
ity. 

Since Miller's scheme was published, the 
anatomy of a number of terebrid species has 
been described that do not fit into the classi- 
fication, and it is clear that some reappraisal 
is necessary. Furthermore, the recent mono- 
graph by Bratcher & Cernohorsky (1987) has 
highlighted the inconsistencies and problems 
in the classification of the family, with apparent 



19 



20 



TAYLOR 



incongruence between the anatomy and 
the shell characters used in the generic divi- 
sions. 

Apart from a phonetic study of Miocene 
species (Davoli. 1977). there have been no 
previous attempts to establish phylogenetic 
relationships within the Terebridae, and only 
very generalised comments on the relation- 
ship of the terebrids with other toxoglossans. 
Because a number of terebrid species pos- 
sess the toxoglossan feeding apparatus of 
hypodermic radular teeth, venom gland, and 
muscular bulb, the relationship with the fam- 
ilies Turridae and Conidae has long been es- 
tablished. However, many terebrid species 
lack a radular and venom apparatus, and the 
structures are presumed to have been evolu- 
tionanly lost. The Terebndae are usually as- 
sumed to have been derived from the Tur- 
ndae. although details of this relationship are 
obscure. Powell (1966. fig. 1) suggested, pre- 
sumably following Cossmann, 1896), a deri- 
vation from the turrid subfamily Clavatulinae. 
probably based upon the superficial similari- 
ties of the smooth, rather elongated shells of 
Pusionella with some terebrids. 

Because of the general uniformity of shell 
morphology, the Terebridae have been as- 
sumed to be monophyletic. The only serious 
dissenter from this view is Rudman (1969), 
who described the anatomy of the New 
Zealand and southern Australian species 
Pervicacia tristis (Deshayes, 1859) {Terebra 
tristis in Bratcher & Cernohorsky, 1987). This 
species has a radula consisting of two rows of 
short, solid and slightly curved marginal teeth 
and an odontophore, but lacks a venom ap- 
paratus. Rudman considered the species suf- 
ficiently distinct from other terebrids to justify 
the erection of a new family, the Pervicaci- 
idae. which he thought was derived from the 
Turridae independently from the rest of the 
terebrids. The latter together with the Conidae 
were derived from the Turridae after the evo- 
lution of the hypodermic type of radular teeth. 
Subsequently, other authors (Ponder, 1973: 
Bratcher & Cernohorsky, 1987) have consid- 
ered that the characters of T. tristis fall within 
the range of other Terebridae and that the 
separation as a separate family was not jus- 
tified. 

The objectives of this paper are to review 
what is known of foregut anatomy in the Tere- 
bridae, and to examine the use of these an- 
atomical characters in determining relation- 
ships both within the family and with other 
conoideans. 



MATERIALS AND METHODS 

The basic data for this study were obtained 
from dissections and serial sections of the 
foregut made from 18 species of Terebridae. 
These are listed below, nomenclature follow- 
ing Bratcher & Cernohorsky, 1987, which for 
convenience is also followed throughout the 
paper: Hastula aciculina (Lamarck, 1822); H. 
albula Menke, 1843; H. bacillus (Deshayes, 
1859): H. hectica (Linnaeus, 1758); H. sal- 
tearía (Deshayes, 1859); H. solida (De- 
shayes, 1857): Terebra affinis Gray, 1834; T 
babylonia Lamarck, 1822; T capensis Smith, 
1873; T. cerithina Lamarck, 1822; T. dimidiata 
(Linnaeus, 1758), T funiculata Hinds, 1844; 
T maculata Linnaeus, 1758; T. nassoides 
Hinds, 1844; T subulata (Linnaeus, 1758); T. 
tristis Deshayes, 1844; Duplicarla duplicata 
(Linnaeus, 1758); D. spectabilis (Hinds, 
1844). 

REVIEW OF FOREGUT ANATOMY 

This section consists of brief descriptions of 
the major features of the foregut anatomy of a 
number of terebrid species chosen to repre- 
sent the variety of form so far known in the 
family. Further details of some of the species 
can be found in the publications cited. 

Hastula cinerea (Born, 1778), H. galleana, 
and H. inconstans (Hinds, 1844) (see Marcus 
& Marcus, 1960; Miller, 1979; Taylor, unpub.) 

These three species have a similar anat- 
omy and are characteristic of Miller's type IIa 
foregut and of most other Hastula species. All 
three species have a short- to medium-length 
labial tube, a long buccal tube which can ex- 
tend outside the rhynchodeum, a radular sac 
and caecum with hollow, hypodermic radular 
teeth, a pair of salivary glands, and a well- 
developed venom gland and muscular bulb. 

Hastula tiectica has a basically similar 
anatomy, but the labial tube is longer. The 
most interesting feature of this species is in 
the structure of the radular teeth. The large 
barbed and hollow radular teeth have an ap- 
parently unique feature. The mid-section of 
the tooth is formed of a network structure 
rather like chicken-wire (Figs. 1c, 2). The 
tooth is hollow but has a well-developed ori- 
fice near the tip which in other toxoglossans is 
where the venom emerges on penetration of 
the prey. The function of the perforated mid- 
section of the tooth is not known but may al- 



ANATOMY OF TEREBRIDAE 



21 




FIG. 1. Single radular teeth of a. Hastula bacillus 
x95; b. Terebra babylonia x217; с Hastula hectica 
x718. 



low the delivery of venom along the full length 
of the tooth rather than just at the tip. 

Terebra /m/íaír/x Auffenberg & Lee, 1988 

This recently described species has a shell 
morphology very similar to species of the 
Hastula cinerea group but differs in anatomy. 
T. imitatrix has a large, spoon-shaped labial 
tube; a large, club-shaped accessory probos- 
cis structure with two rows of papillae on the 
ventral side; a very short buccal tube which 
probably cannot be extended out of the mouth 
of the rhynchodeum; a relatively short venom 
gland and a small muscular bulb. There is a 
radular sac and a caecum with rather elon- 
gate, slightly curved, barbless teeth. No sali- 
vary glands were located by the authors. 

Auffenberg & Lee (1988) were puzzled by 
the combination of characters in this species 
and for this reason hesitated to include it in 
the genus Hastula. The shell and much of the 
anatomy resembles that of other Hastula spe- 
cies, but the presence of the large accessory 
proboscis structure they thought to be a fea- 




FIG. 2. Detail of mid-section of the middle shaft of 
the radular tooth of Hastula hectica. Scale bar 10 
|j.m. 



ture usually restricted to terebrids with Miller's 
Type III proboscis. However, species in this 
latter group lack a buccal tube, radular appa- 
ratus, salivary glands, venom gland, and 
muscular bulb. 

Hastula bacillus (see Taylor & Miller, 1990) 

This small species was found abundantly 
on the surf beaches of the western side of 
Phuket Island, Thailand. There is a short, ex- 
tensible labial tube, a long extensible buccal 
tube, and a large, muscular, branched acces- 
sory proboscis structure (aps), which is an- 
chored to the left wall of the rhynchodeum 
(Fig. 3). The aps can be extended some dis- 
tance out of the rhynchodeum, but when re- 
tracted is bent into an "s" shape within the 
cavity. Entering the buccal cavity is a radular 
sac, without a caecum. A small odontophore 
is present, and the radula consists of teeth 
that are scroll-like at the base but taper into a 
pointed barb-less, knife-like, blade at the tip 
(Fig. la). There is a pair of salivary glands 
and a single (right) accessory salivary gland. 
The venom gland is large, with a well-devel- 
oped muscular bulb. 

S.E.M. studies of the accessory proboscis 
structure show that regularly spaced tufts of 
short, stiff cilia are distributed over the suface. 
Associated with each tuft are pairs or triplets 



22 



TAYLOR 



rw 



apm 




FIG. 3. Dissection of the head of Hastula bacillus showing the extended accessory proboscis structure 
and extended labial tube. The buccal tube is retracted into the rhynchodeal cavity but can be extended 
beyond the mouth of the cavity, apm, accessory proboscis retractor muscles; aps, accessory proboscis 
structure; asg, accessory salivary gland; be, buccal cavity; bt, buccal tube; m, mouth; mb, muscular 
bulb; oe, oesophagus; pm. proboscis retractor muscles; rs, radular sac; rw, rhynchodeal wall; vg, venom 
gland. 



ANATOMY OF TEREBRIDAE 



23 




FIG. 4. Branch of the accessory proboscis structure of Hastula bacillus showing the ciliary tufts with pairs and 
triplets of domes. SEM of critical point dried material. Scale bar 10 цт. 

FIG. 5. Detail of a ciliary tuft, showing short stiff cilia and the accompanying microvilli-covered domes. Scale 
bar 1 \i.m. 



of microvilli-covered domes (Figs. 4, 5). The 
ciliary tufts are similar to those seen on the 
palliai or cephalic tentacles of other molluscs 
and which are thought to be either chemo- or 
mechanosensory in function. Because Has- 
tula bacillus lives in a wave-disturbed habitat, 
Taylor & Miller (1990) suggested that the ac- 
cessory proboscis structure was more likely 
to be chemosensory in function and used in 
finding the preferred prey of Scolepis, a spio- 
nid polycheate. 

Although similar in basic anatomy to Has- 
tula cinerea, H. bacillus shows a number of 
important differences. The presence of the 
odontophore and accessory salivary glands 
are plesiomorphic characters. The radular 
teeth are much simpler than the hollow, 
barbed, hypodermic teeth of H. cinerea. Ad- 
ditionally, the branched accessory proboscis 
structure is a unique feature, but almost cer- 
tainly homologous with the club-shaped struc- 
ture in Terebra imitatrix and T. affinis. Prior to 
its discovery in H. bacillus and T. imitatrix, the 
accessory proboscis structure was consid- 
ered an advanced character, found only in 
species with Miller's type III proboscis. 

Terebra subulata 

Along with T babylonia and T guttata, this 
species has a long multiwhorled shell of more 



than 18 whorls. These species have an anat- 
omy typical of Miller's type lib. Terebra subu- 
lata has a labial tube introvert; a long buccal 
tube; a septum dividing the rhynchodeal cav- 
ity (Fig. 13); a short radular sac; a radular 
caecum; long, thin, hypodermic radular teeth, 
with small barbs and a constricted neck near 
the base of the tooth (Fig. 12f); a pair of sal- 
ivary glands; a pair of accessory salivary 
glands (Fig. 6); and a large venom gland and 
muscular bulb. 

Terebra maculata (see Miller, 1970) 

This species possesses Miller's type lb fo- 
regut. It has a very long labial tube introverf 
which is coiled within the rhynchodeal cavity 
(Fig. 13); a short buccal tube; a pair of sali- 
vary glands; no radula and no venom appa- 
ratus. 

Terebra gouldi Deshayes, 1857 (see Miller, 
1975) 

This species which represents Miller's type 
la foregut has a medium length labial tube 
introvert; a very short buccal tube; a pair of 
salivary glands, with no radula or venom ap- 
paratus. 



24 



TAYLOR 



asg 




2mm 

FIG. 6. Main organs of the buccal mass of Terebra subulata. asg, accessory salivary glands; с radular 
caecum; oe, oesophagus; rs, radular sac; sd, salivary duct; sg, salivary gland; vd, venom duct. 



Terebra affinis (see Miller, 1970) 

This species represents Miller's Type III 
foregut. It possesses a long labial tube intro- 
vert: no buccal tube; no salivary glands; no 
radula or venom apparatus. Arising from the 
left wall of the rhynchodeum is the accessory 
proboscis structure (Fig. 13), an extensible 
muscular stalk, with a mace-like, papillate 
head. 

Duplicaría spectabilis and D. duplicaría 

Both of these species lack eyes and ceph- 
alic tentacles. They possess a long labial tube 
introvert, but the buccal tube is extremely 
short (Figs. 7, 8). The buccal cavity is large, 
and opening into it is a small radular sac, with 
an odontophore with a radular ribbon consist- 
ing of two rows of solid, sickle-shaped radular 
teeth (Fig. 9). Salivary ducts from a pair of 
salivary glands open either side of where the 



radular sac joins the buccal cavity. There is 
no venom gland or muscular bulb. 

Terebra nassoides 

This is a small species collected from inter- 
tidal sand patches in Oman. The anatomy is 
basically similar to that of the above Dupli- 
caría species. It has no eyes or cephalic ten- 
tacles. The labial tube is hood-shaped, the 
dorsal part being much larger than the ven- 
tral. The buccal tube is very short, and there is 
a small odontophore (Fig. 1 1) and short rad- 
ular ribbon with two rows of curved, solid 
teeth (Fig. 10). There are no salivary glands 
and no venom apparatus. 

Terebra tristis 

The anatomy of this New Zealand-southern 
Australian species was described by Rudman 
(1 969) as Pervicacia tristis. There are no eyes 



ANATOMY OF TEREBRIDAE 



25 




FIG. 7, Longitudinal section through the extended labial tube and buccal mass of Duplicarla spectabills. m, 
mouth; od, odontophore: oe, oesophagus; re, rhynchodeal cavity; rd, radula; rm, mouth of rhynchodeum; rw, 
rhynchodeal wall; sm, sphincter muscle; sg, salivary gland. 



or tentacles. The dorsal part of the labial tube 
extends anteriorly over the ventral surface, 
forming a hood. Rudman stated that the labial 
tube appeared incapable of withdrawal, but 
sections of preserved specimens I have stud- 
ied show that both dorsal and ventral parts of 
the labial tube can be folded back into the 
rhynchodeum. The buccal tube is very short. 
There is a radula ribbon and odontophore 
with two rows of solid, slightly curved radular 
teeth. There is a pair of fused salivary glands, 
and two salivary ducts enter the buccal cavity. 
There is no venom apparatus. 



REVIEW OF FOREGUT CHARACTERS 

In this section I review the distribution and 
variation in the main organs of the foregut 
amongst the terebrid species for which the 
anatomy is known. 

Labial tube 

The possession of an extensible labial tube 
(introvert formed by the extension of the walls 
of the rhynchodeum) is perhaps characteristic 
of all species of the Terebridae. This charac- 
ter is not confined to the terebrids, but is 
found in some turrids (subfamily Daphnelli- 



nae) which have a polyembolic proboscis, 
such as Philbertia linearis (Sheridan et al., 
1973) and Cenodagruetes (Smith, 1967). 

There are some differences in the form of 
the labial tube which may be important. In Te- 
rebra tristis and T. nassoldes, the dorsal part 
of the tube is much larger than the ventral and 
when extended appears hood-like (Rudman 
1967, Taylor, unpub. observ.). The ventral 
part of the tube in T. nassoldes probably does 
not retract. 

Other variation mainly concerns the length 
of the tube. In those species having a long 
buccal tube and hypodermic radula, the labial 
tube is relatively short. However, in those 
forms with a short buccal tube and also lack- 
ing a radula, the labial tube is much longer, 
and in Terebra maculata and similar species 
the labial tube, when withdrawn, is folded on 
itself several times within the rhynchodeal 
cavity (Miller, 1970). 

In his account of feeding in Terebra gouldl, 
Miller (1975) has shown how this species, 
which lacks a radula and venom apparatus, 
and has only a very short buccal tube, uses 
the extensible labial tube to capture the en- 
teropneust Ptychodera and transfer the prey 
to the short buccal tube. Similarly, the long 
labial tube of T maculata probes in the sand 
for the capitellid polychaete Dasybranchus. 



26 



TAYLOR 



iwr 




brm 



FIG. 8. Section through the buccal mass of Duplicaría spectabilis showing the short, muscular, buccal tube, 
the odontophore and radular sac. brm, buccal retractor muscle, bt, buccal tube; iwr, inner wall of rhyncho- 
deum: m, mouth; od odontophore; ode odontophoral cartilage; ps, proboscis sheath; rdt, radular tooth; rs, 
radular sac; sd, salivary duct. 



Buccal Tube 

The buccal tube or true proboscis is long 
only in those species with hollow, hypodermic 



radula teeth. During the feeding process sin- 
gle teeth are transferred from the radular cae- 
cum to the tip of the buccal tube, where they 
are gripped by the sphincter muscle (e.g. Has- 



ANATOMY OF TEREBRIDAE 



27 




FIG. 9. Disaggregated radular teeth of Duplicaría spectabilis. Scale bar 20 ixm. 

FIG. 10. Side view of part of row of radular teeth of Terebra nassoides. Scale bar 20 \xvr\. 




FIG. 11. Section through the odontophore of Tere- 
bra nassoides showing the pair of odontophoral 
cartilages. SEf\/l of critical point dried material. 
Scale bar 30 |xm. 



tula inconstans: Miller, 1979). In these spe- 
cies, the proboscis wall is muscular and is 
capable of being extended well beyond the 
rhynchodeal cavity. Although they did not ob- 
serve any living animals, Auffenberg & Lee 



(1988) thought that the short, but muscular, 
buccal tube of Terebra imitatrix could not be 
extended out of the rhynchodeal cavity. 

Those terebrid species with solid radula 
teeth fixed on a ribbon have only a very short 
buccal tube (Fig. 13). This also applies to 
those species that have completely lost the 
radular apparatus. 

Accessory Proboscis Structure 

Miller (1970) described in Terebra affinis a 
retractile, club-shaped structure, which he 
called the accessory feeding organ. (Taylor & 
Miller [1990] prefer the functionally neutral 
term of accessory proboscis structure.) This 
consists of a muscular stalk with a distal papil- 
late head that is attached at the base to the 
posterior left side of the wall of the rhyn- 
chodeal cavity. This structure occurs in spe- 
cies otherwise lacking a buccal tube, radula, 
salivary glands and venom apparatus. Miller 
thought, but without direct observations, that 
the organ was involved in prey capture. 

The possession of the accessory proboscis 
structure is a major character defining Miller's 
Type III foregut. Miller (1970) mentioned, but 
did not illustrate, a number of other terebrids 
as having an aps. 

Recently, an accessory proboscis structure 
was described from two other terebrid spe- 
cies, Hastula bacillus (by Taylor & Miller, 



28 



TAYLOR 



1990) and Terebra imitatrix (by Auffenberg & 
Lee, 1 988). These species otherwise possess 
an extensible buccal tube, hypodermic radu- 
lar teeth, salivary glands and venom appara- 
tus. The anatomy is similar to other Hastula 
species and the Type IIa foregut of Miller 
(1970, 1971). Hastula bacillus has a long, 
branching aps, but in T. imitathx it is club 
shaped. Both muscular structures are at- 
tached to the left side of the rhynchodeal cav- 
ity and are probably homologous with the 
structure in Terebra affinis. 

As shown in Figures 4 and 5, the aps in 
Hastula bacillus possesses numerous ciliary 
tufts which T.E.M. analysis suggests are sen- 
sory, probably chemosensory, structures. Be- 
cause H. bacillus and T imitatrix both have a 
long functional buccal tube (true proboscis) it 
seems more likely that the aps is a sensory 
device, rather than part of the food-gathering 
apparatus. However, the fine structure of the 
aps in T. affinis and 7. imitatrix has not been 
investigated. Nevertheless, it is now clear that 
the possession of an aps is not an autapo- 
morphy of terebrids having the Type III 
foregut, but it can occur in terebrids, which 
compared with the outgroups in the Turrridae 
(Miller, 1989), are the least denved for the 
family. 

Radula 

There are basically two main types of rad- 
ula found in the Terebridae: (1) radulas of 
solid, sickle or dagger-shaped teeth; (2) rad- 
ulas of hollow, harpoon-like hypodermic teeth. 
Additionally, there are many terebrid species 
that have no radular apparatus at all. 

Relatively few radulae of the solid-toothed 
variety have been described. Figures 9 and 
10 illustrate the radula in two species, which 
are basically similar in morphology. The rad- 
ula consists of two rows of marginal teeth at- 
tached to the short radular ribbon. In all spe- 
cies there IS an odontophore with two 
odontophoral cartilages (Fig. 11). In Terebra 
nassoides and T. tristis the teeth are solid, 
broader at the base and curved: in Duplicarla 
spectabilis and D. duplicarla they are sickle 
shaped. Additionally, teeth like those in T tris- 
tis are found in Duplicarla kieneri and D. fictilis 
from South Australia (radula mounts in 
BM(NH)). Troschel (1866) illustrates an un- 
usual radula for Myurella lamarckii Kiener ( 
Duplicarla lamarckii. considered by Bratcher 
& Cernohorsky, 1987, as a form of D. dupli- 
cata). This has long sickle-shaped teeth as in 



D. duplicata, but with spur-like projections 
near the distal end of each tooth. I examined 
the radula of D. lamarcki from Kenya but 
found the teeth to be simple, with no sign of 
the spur-like cusps. 

The solid, sickle-shaped marginal teeth of 
these terebrids resemble those found in some 
turrids, particularly from the subfamily 
Pseudomelatominae (e.g. Tiariturris libya 
(Dall, 1919), Shimek & Kohn, 1981, fig. 2; 
Pseudomelatoma penicillata (Carpenter, 
1864), Kantor, 1988, fig. 1D-F). However, 
these turrids also have a large, unicuspid cen- 
tral tooth. The marginal teeth are erected into 
a basket structure as the radular passes over 
the bending plane. Shimek & Kohn (1981) 
thought that the central tooth functions as a 
slicing tooth and that the marginal teeth lac- 
erate the prey, tearing off fragments and con- 
veying them to the oesophagus. 

Amongst those species having the hollow 
hypodermic teeth, there is some variety of 
form (Figs. 1 , 12). Both the simplest and most 
complex forms are those of some Hastula 
species. In Hastula bacillus the teeth are 
rolled and hollow, but the anterior half con- 
sists of a knife-like blade. In Hastula cinerea, 
H. penicillata and H. salleana the teeth are 
robust with a harpoon-like, barbed tip, a large 
aperture near the tip and a broad base with a 
flared rim (Marcus & Marcus, 1960; Bändel, 
1984). An additional feature of the teeth in 
these three species is the presence of 
"screw-thread"-like flanges separating the 
rolls of the tooth (Fig. 1 2). Bändel (1 984) sug- 
gested that these flanges gave rigidity to the 
tooth by separating the rolls. Essentially sim- 
ilar teeth are seen in T. taurinus and T. pre- 
texta (Bändel, 1 984, figs. 31 3, 314). The teeth 
of Hastula fiectica with the perforated mid- 
section are apparently unique. In Terebra 
subulata, T. guttata, T. succinea, and T. anilis, 
the radula teeth are long and thin, with 
pointed tips and small barbs (Mills, 1977a; 
Bändel, 1984; personal observations). Those 
of Terebra babylonia are similar but lack the 
barbed tips (Fig. 1). Additionally, all these lat- 
ter species have a marked concavity near the 
base of the tooth produced by twisting of the 
tooth (Fig. lb). Of the species with the hypo- 
dermic teeth, Hastula bacillus and H. acicu- 
lina are the only species so far found with an 
albeit small odontophore and odontophoral 
cartilages. 

In Hastula bacillus and H. aciculina. the 
radular sac is relatively long and there is no 
radular caecum for the storage of detached 



ANATOMY OF TEREBRIDAE 



29 



¡H*' 




a 



9 



FIG. 12. Range of form found in the hypodermic-type radula teeth of Terebridae. Traced from original 
scanning electron micrographs or from references cited, a. Hastula bacillus. Thailand; b. H. cinerea Colom- 
bia (Bändel, 1984); с. H. hectica Kenya; d. Terebra pretexta Colombia (Bande!, 1984); е. T. babylonia Guam; 
f. T. subulata Maldives; g. T. guttata Queensland (Mills, 1977a); h. T. taunnus Colombia (Bande!, 1984); i. 
T. succinea Queensland (Mills, 1977a). Not to scale. 



radular teeth. In other species (e.g. Hastula 
cinerea and H. inconstans), the radular sac is 
shorter and there is a well-developed caecum. 
In Terebra subulata and T. babylonia the rad- 
ular sac is very short, with a large caecum. 

The hollow barbed radular teeth found in 
some terebrids are similar in form to those 
found in the Conidae and some turrids (sub- 
family Borsoninae) and which are considered 
to be the most derived type of radula to be 
found in the Conoidea. A great diversity of 
radular types is found in the Turridae, and 
Shimek & Kohn (1981) have developed an 
adaptive scenario to explain the evolution of 
the radula, from the primitive condition of five 
teeth in a row found in some Clavinae, to the 
advanced hypodermic marginal teeth of the 
Borsoninae. The main evolutionary elabora- 
tions concern the marginal teeth and include 
increasing size and complexity, with con- 
comitant loss of the central and lateral teeth. 

Although considerable variation is seen in 
the radula amongst the various subfamilies of 
Turhdae (Powell, 1966; -McLean, 1971; Shi- 
mek & Kohn, 1981), variation in both the 
Conidae and Terebridae is relatively small 
and concerns details of the form of the hypo- 
dermic teeth. Indeed, variation in the hypo- 
dermic teeth in the Terebridae is probably no 



greater than found in one subfamily of Tur- 
ridae, the Oenopotinae (Bogdanov, 1989). No 
convincing intermediate condition between 
the solid teeth and the hypodermic teeth of 
terebrids has been seen. The teeth in Hastula 
bacillus, with the hollow, rolled proximal end 
and the solid, blade-like distal end, might be 
an intermediate condition or a variant of the 
hypodermic tooth. 

Accessory salivary glands 

Accessory salivary glands are an apomor- 
phic character of the Neogastropoda and are 
known in some turrids and some Conus spe- 
cies (Ponder, 1973; Marsh, 1971; Schultz, 
1983). Recently, a single accessory gland 
was found in Hastula bacillus (Taylor & Miller, 
1990), but not in several other Hastula spe- 
cies examined. Furthermore, dissection and 
thin sections have also revealed a pair of ac- 
cessory glands in Terebra subulata (Fig. 6), 
but only single glands in Terebra babylonia 
and T. funiculata. 

Salivary glands 

A pair of salivary glands is present in most 
terebrids. The glands are usually partially 



30 



TAYLOR 



fused together and appear as two distinct 
lobes of one mass. Separate salivary ducts 
enter the buccal cavity at the base of the rad- 
ula sac. where this is present. Rudman (1969) 
described an apparently unusual feature of 
Pervicacia tristis where the salivary ducts 
fuse, entering the buccal cavity as a single 
duct. But my observations of serial sections of 
this species show two ducts entering the buc- 
cal cavity. 

Salivary glands and ducts are present in 
many terebrids where the radula. buccal tube 
and venom apparatus has been lost. They are 
however, absent in some of Miller's Type III 
species, such as Terebra affinis. 

Venom apparatus 

The venom apparatus of venom gland and 
muscular bulb is an autapomorphic character 
of the Conoidea (Taylor & Morns, 1988). In 
the Terebridae it is present only in those spe- 
cies with hollow, hypodermic radular teeth 
and a long buccal tube. There is some varia- 
tion in the length of the venom gland and the 
size of the muscular bulb. For instance, the 
venom apparatus is particularly large in Tere- 
bra subulata (Mills, 1977b). By contrast, 
Auffenberg & Lee (1988 p. 155) consider that 
the muscular bulb in Terebra imitatrix ". . . is 
weak, seemingly vestigial. . . . ". Mills (1977b) 
reported differences in the secretory epithelia 
of the venom gland between Terebra and Co- 
nus, but these need further investigation. 

Rhynchodeal septum 

Miller (1 970, 1 971 ) briefly mentioned a sep- 
tum across the rhynchodeal cavity in some 
terebrids with his Type lib proboscis. In Tere- 
bra subulata. this structure divides the rhyn- 
chodeal cavity into two compartments. It con- 
sists of an invagination of the inner wall of the 
rhynchodeum with a central aperture. When 
withdrawn, the buccal tube lies to the poste- 
rior and the labial tube to the anterior of the 
septum. When extended, the buccal tube 
passes through the central aperture of the 
septum. The function of the septum is un- 
known, but may be concerned with retaining 
prey that has been pulled into the rhyncho- 
deum by the buccal tube. 

GEOLOGICAL HISTORY 

No adequate analysis has been made of 
the geological history of the Terebridae. Var- 



ious authors (Cossmann, 1896, Wenz, 1938; 
Taylor et al., 1980) have considered, with 
varying degrees of confidence, the Creta- 
ceous (Santonian) species "Fusus" cingula- 
tus Sowerby, from Gosau, Austria, to be an 
early terebnd (as Strioterebrum). However, 
the species appears to have an elongate si- 
phonal canal and is more likely to be a mem- 
ber of the Turhdae. Otherwise, the earliest 
terebrids appear to be Hastula species from 
the Eocene of France and England (Coss- 
mann, 1896). The Terebridae diversified ex- 
tensively in the Miocene, with the appearance 
of many of the shell forms seen amongst Re- 
cent species (Davoli, 1977). 



DIET OF TEREBRIDAE 

Most available dietary information for the 
Terebridae is from the genus Hastula, and 
nearly all the species investigated seem to 
feed upon spionid polychaetes. Miller (1979) 
gives a detailed account of the Hawaiian spe- 
cies Hastula inconstans. which feeds exclu- 
sively upon Dispio magna. Also in Hawaii, H. 
hectica and H. strigillata feed upon Nerinides 
sp. and H. penicillata upon an unidentified 
spionid. Marcus & Marcus (1960) report H. 
cinerea as feeding upon Nerinides agilis, as 
does H. salleana (Stewart, in Miller, 1979) 

In Thailand, Hastula bacillus feeds upon 
Scolepis sp. (Taylor & Miller, 1990). The only 
exception to the spionid diet of Hastula is H. 
solida from Guam, which feeds upon a cirrat- 
ulid polychaete, probably Cirratula sp. (Tay- 
lor, unpub.). 

Species with Miller's foregut type lib also 
eat spionids. Miller (1970) reports Terebra 
textilis from Hawaii as eating Prionospio 
malmgreni. and from Guam Taylor (1987) re- 
ports T cingulifera and T subulata feeding 
upon Laonice cirrata. 

There is little dietary data available for tere- 
brids with solid radular teeth. Dissection of 
many Dupllcana spectabilis from Hong Kong 
revealed no recognisable food remains. The 
only information available is for Terebra nas- 
soides from Salalah, Oman, where three in- 
dividuals contained setae of a capitellid poly- 
chaete (Taylor, unpub.). 

Of the terebrids with no radula and venom 
apparatus, Miller (1975) has descnbed feed- 
ing in Terebra gouldi. which eats the enterop- 
neust Ptychodera flava. This diet is shared by 
Terebra dimidiata, T crenulata. and T. areo- 
lata. 



ANATOMY OF TEREBRIDAE 



31 



Amongst those species with a very long la- 
bial tube and no radula or venom apparatus, 
Terebra felina, T. macúlala and T. chlorata all 
eat the capitellid polychaete Dasybranchus 
caducus (Miller, 1970, 1975). 

Finally, the diet of the species with an ac- 
cessory proboscis structure but no radula and 
venom apparatus is unknown. The digestive 
tract of T. afflnis frequently contains amor- 
phous red-brown material, which Miller (1970) 
and Taylor (1986) thought might be the bran- 
chial tentacles of cirratulid polychaetes. 



DISCUSSION & CONCLUSIONS 

From the foregoing descriptions it is clear 
that a wide range of foregut anatomies are 
present in the Terebhdae (summarized in Fig. 
13). There is clearly more complexity to be 
accounted for than in Miller's (1970, 1971) 
classification. Furthermore, only a small pro- 
portion of the nearly 300 living species have 
been examined anatomically, and the discov- 
ery of further foregut types is to be expected. 

In considering the evolutionary relation- 
ships of the Terebridae, the first question to 
be asked is whether the family comprises a 
monophyletic group. Evidence from foregut 
anatomy suggests that there are two major 
divisions within the family. Firstly, there is the 
group of species with solid radular teeth, and 
a well-developed radular ribbon. These spe- 
cies have a short buccal tube, lack a venom 
apparatus and have no cephalic tentacles or 
eyes. Secondly, there is the group comprising 
the species possessing radular teeth of the 
hypodermic type. These species also pos- 
sess a venom apparatus and elongate buccal 
tube. Additionally, there are terebrids which 
lack a radula and venom apparatus and have 
a very short buccal tube. These species could 
be derived from one or other of the radulate 
groups. 

It is suggested that the two groups of tere- 
brids represent separate derivations from the 
Turridae. The group with solid teeth comprise 
some species classified in the genus Dupli- 
carla by Bratcher & Cernohorsky (1987), as 
well as Terebra nassoldes, T. capensis and T 
tristis, and probably many others. 

In their analysis of the toxoglossan radula, 
Shimek & Kohn (1981) considered that the 
most derived condition was the hypodermic 
type consisting of long, hollow, barbed mar- 
ginal teeth with only a vestigial radular mem- 
brane. The hypodermic radulae of such tere- 



brids as Hastula cinerea, H. salleana and 
Terebra subulata are similar to those found in 
the Conidae and Borsoniinae. However, Has- 
tula bacillus and H. aclcullna have a less de- 
rived condition, with simpler radular teeth with 
no barbs, a small odontophore and odonto- 
phoral cartilages and a more substantial rad- 
ular membrane. In H. bacillus there is no cae- 
cum to the radular sac in which teeth can be 
stored. However, thin sections showed a rad- 
ular tooth held at the proboscis tip. This is 
similar to the situation in some turrids, where 
detached, non-hypodermic, solid marginal 
teeth are held at the proboscis tip (Sysoev & 
Kantor, 1987). The knife-like distal portion of 
the tooth is more likely a stabbing structure, 
rather than a true hypodermic tooth. The 
presence of this less-derived radular appara- 
tus in H. bacillus suggests that the hypoder- 
mic radulae of the Terebridae and Conidae 
are parallel but independent developments. 

Compared to outgroups in the Turridae and 
Conidae (Miller, 1989), species of the genus 
Hastula are the least derived for the family. 
They all possess the basic intraembolic pro- 
boscis condition, with a long buccal tube, 
venom apparatus, with in many species a true 
hypodermic radula. A major variant is seen in 
Hastula bacillus, which possesses an elon- 
gate, branching accessory proboscis struc- 
ture. This may be homologous with the club- 
shaped structure seen in Terebra Imltatrix, 
which has a reduced buccal tube and a small 
venom apparatus. 

Terebra subulata and similar species (7. 
guttata, T babylonia, T funlculata) have long, 
thin radular teeth, and shells with many (18 + ) 
whorls. This group of species can be simply 
derived from the Hastula condition. 

Many terebrids, however, lack a radula, 
venom apparatus and buccal tube, and the 
foregut provides less useful evidence of rela- 
tionships. The trend in terebrids for the loss of 
most of the foregut structures except for the 
labial tube introvert, results in the condition 
known as the polyembolic proboscis (Smith, 
1967). The whole foregut is essentially sim- 
plified into a muscular tube that ingests 
prey. The extensible labial tube becomes the 
main organ of prey capture and ingestion, the 
true proboscis having disappeared. This trend 
is paralleled in the Turridae, where the poly- 
embolic proboscis occurs in some species of 
the subfamily Daphnellinae (Smith, 1967; 
Kantor & Sysoev, 1989). The simplified fore- 
gut of the terebrids could have been derived 
via a number of evolutionary routes. Corrob- 



32 



TAYLOR 



solid-toothed radula 



hypodermic radula 




FIG 13. Diagrammatic representation of the mam types of foregut found in the Terebndae, with raduiar 
teeth. Key to abbreviations: aps, accessory proboscis structure; asg, accessory salivary gland; rs, radu- 
iar sac- s rhynchodeal septum; sg, salivary glands; sm, sphincter muscle at mouth of labial tube; va, 
venom apparatus. Key to species: a. Duplicaría spectabilis, b. Duplicaría duplicata, с. Hastula cinerea, d. 
Hastula bacillus, e. Terebra subulata. f. Terebra gouldi and Terebra dimidiata. g. Terebra macúlala, h. 
Terebra affinis. 



ANATOMY OF TEREBRIDAE 



33 



orative evidence from other anatomical char- 
acters is needed to establish the relationships 
of these terebrids. For example, Terebra 
gouldi has the simplified foregut structure, but 
it has the shell characters of the genus Dupli- 
caría (placed there by Bratcher & Cerno- 
horsky, 1 988) and may conceivably have been 
derived from the solid-toothed group of tere- 
brids. By contrast, T. dimidiata and T. crenu- 
lata similarly have the simple foregut, but 
shell characters more like (perhaps superfi- 
cially) those in the Terebra subulata group of 
species. 

The mace-like accessory proboscis struc- 
ture described from Terebra affinis (Miller, 
1970, 1971) occurs in a species that other- 
wise lacks a buccal tube, radula, venom ap- 
paratus or salivary glands. It was thought to 
be an autapomorphic character of Miller's type 
III proboscis. However, the discovery of prob- 
ably homologous structures in the otherwise 
less-derived Hastula bacillus (Taylor & Miller, 
1990) and Terebra imitatrix (Auffenberg & 
Lee, 1988), suggests that this structure could 
be more widespread amongst the Terebridae. 

More species of Terebridae need to be ex- 
amined using more characters before an ad- 
equate phylogenetic analysis can be made. 
However, the main conclusion of this paper is 
that Rudman (1969) was essentially correct in 
separating the Pervicaciidae as a separate 
family. What is now clear is that the family 
should accommodate many more "terebrids," 
perhaps all of the Duplicaría species and 
probably many others, such as Terebra nas- 
soides, T. capensis, T. kieneri. and T fictilis. 
The family Terebridae should accomodate all 
the species with hypodermic radular teeth 
and venom apparatus and derivatives from 
these. The great range of morphology found 
in the terebrid foregut coupled with the appar- 
ent incongruence between shell characters 
and anatomy will make detailed classification 
of the families difficult. It is clear that shell 
characters are a poor guide to relationships in 
the Terebridae. 

A pressing problem concerns the anatomy, 
as yet unknown, of the south Australian spe- 
cies, T albida Gray, 1834. This is the type 
species of the genus Acus Gray, 1847, on 
which the family Acusidae Gray, 1853, is 
based. If this species turns out to have solid 
radular teeth and no venom apparatus, then 
the name Acusidae will have priority over Per- 
vicaciidae. Additionally, anatomical material 
is needed of Pervicacia ustulata, the type 
species of the genus Pervicacia. 



ACKNOWLEDGEMENTS 

I am very grateful for the help of David Coo- 
per who made the sections. Kurt Auffenberg, 
Dick Kilburn, Ian Loch, and Bruce Marshall 
generously donated specimens. Alison Kay, 
Bill Rudman and Yuri Kantor kindly read and 
commented on the mansucript. I thank John 
Miller, David Reid, Noel Morris, and Winston 
Ponder for useful discussion. 



LITERATURE CITED 

AUFFENBERG, K. & H. G. LEE, 1988. A new spe- 
cies of intertidal Terebra from Brazil. Nautilus. 
102: 154-158. 

BÄNDEL, К., 1984. The radulae of Caribbean and 
other Mesogastropoda and Neogastropoda. Zo- 
ologische Verhandelingen. 214: 1-188. 

BOGDANOV, J. P., 1989. Morphological transfor- 
mation in radula and protoconch of Oenopotinae 
Bogdanov, 1987. La Conchiglia. 21: 237-240. 

BRATCHER, T. & W. O. CERNOHORSKY, 1987. 
Living terebras of the world. A monograph of the 
Recent Terebridae of the world. Melbourne, Flor- 
ida: American Malacologists Inc. 

COSSMANN, M., 1896. Essais de Paléoconcholo- 
gie comparée. 2: 1-179. 

DAVOLI, F., 1977. Terebndae (Gastropoda). Parte 
1-1 Molluschi tortoniani di Montegibbio. Palaeon- 
tographica Itálica. 70: 135-169. 

KANTOR, Y. I., 1988. On the anatomy of the 
Pseudomelatominae (Gastropoda, Toxoglossa, 
Turridae) with notes on functional morphology 
and phylogeny of the family. Apex. 3: 1-19. 

KANTOR, Y. I. & A. V. SYSOEV, 1989. The mor- 
phology of toxoglossan gastropods lacking a rad- 
ula, with a description of a new species and ge- 
nus of Turridae. Journal of Molluscan Studies. 
55: 537-550. 

MCLEAN, J. H., A revised classification of the fam- 
ily Turridae, with the proposal of new subfamilies, 
genera, and subgenera from the eastern Pacific. 
1/е/;дег 14:1 14-130. 

MARCUS, E. & E. Marcus, 1960. On Hastula ci- 
nerea. Boletim da Faculdade de Filosofía. Cien- 
cias e Letras. Universidade de Sao Paulo (Zoo- 
logia). 23:25-66. 

MARSH, H., 1971. The foregut glands of vermivo- 
rous cone shells. Australian Journal of Zoology. 
19- 313-326. 

MILLER, B. A., 1970. Studies on the biology of 
Indo-Pacific Terebridae. Ph.D. Thesis. University 
of New Hampshire. Durham. 213pp. 

MILLER, B. A., 1971. Feeding mechanisms of the 
family Terebridae. Reports of the American Mal- 
acological Union AMU Pacific Division. 1970: 
72-74. 

MILLER, B. A., 1975. The biology of Terebra gouldi 
Deshayes, 1859, and a discussion of life history 



34 



TAYLOR 



similarities among other terebrids of similar pro- 
boscis type. Pacific Science. 29: 227-241. 

MILLER. B. A., 1979. The biology of Hastula incon- 
stans (Hinds, 1844) and a discussion of life his- 
tory similarities among other hastulas of similar 
proboscis type. Pacific Science. 33: 289-306. 

MILLER. J. A.. 1989. The toxoglossan proboscis: 
structure and function. Journal of Molluscan 
Studies. 55: 167-182. 

MILLS, P. M.. 1977a. Radular tooth structure in three 
species of Terebndae. Veliger. 19: 259-265. 

MILLS. P. M.. 1977b. On the venom gland of tere- 
brid molluscs. Proceedings of the Tfiird Interna- 
tional Coral Reef Symposium. Miami, Florida: 
631-637. 

PONDER, W. F., 1973. The ongin and evolution of 
the Neogastropoda. Malacologia. 12: 295-338. 

POWELL, A. W. В., 1966. The molluscan families 
Speightiidae and Turndae. An evaluation of the 
valid taxa, both Recent and fossil, with lists of 
characteristic species. Bulletin of the Auckland 
Institute and Museum. 5: 1-184. 

RISBEC, J.. 1953. Observations sur I'anatomie du 
Terebndae Néo-Calédonnais. Bulletin du t\/íu- 
seum National d'Histoire Naturelle. Pahs, series 
2, 25: 576-583. 

RUDMAN, W. В., 1969. Observations on Pervicacia 
tnstis (Deshayes, 1859) and a comparison with 
other toxoglossan gastropods. Veliger. 12: 53- 
64. 

SCHULTZ, M. С, 1983. A correlated light and elec- 
tron microscopic study of the structure and secre- 
tory activity of the accessory salivary glands of 
the marine gastropods. Conus flavidus and C. 
vexillum (Neogastropoda, Conacea). Journal of 
r^orphology. 176: 89-111. 



SHERIDAN, R., J. J. VAN MOL & J. BOUILLON, 
1973. Etude morphologique du tube digestif de 
quelques Turndae (Mollusca-Gastropoda-Proso- 
branchia-Toxoglossa) de la region de Roscoff, 
Cahiers de Biologie Manne. 14: 159-188. 

SHIMEK. R. L. & A. J. KOHN, 1981. Functional 
morphology and evolution of the toxoglossan 
radula. Malacologia. 20: 423-438. 

SMITH, E. H., 1967. The proboscis and oesopha- 
gus of some British turrids. Transactions of the 
Royal Society of Edinburgh. 67: 1-22. 

SYSOEV, A. V. & Y. I. KANTOR. 1987. Deep-sea 
gastropods of the genus Afona (Turndae) of the 
Pacific: species composition, systematics, and 
functional morphology of the digestive system. 
Veliger. 30: 105-126. 

TAYLOR. J. D.. 1986. Diets of sand-living predatory 
gastropods at Piti Bay, Guam, Asian Mahne Bi- 
ology. 3: 47-58. 

TAYLOR, J. D. & J. A. MILLER, 1990. A new type 
of gastropod proboscis; the fore-gut of Hastula 
bacillus (Deshayes) (Gastropoda: Terebndae). 
Journal of Zoology. 220: 603-617. 

TAYLOR. J. D. & N. J. MORRIS, 1988. Relation- 
ships of neogastropods. Malacological Review. 
Supplement 4: 167-179. 

TROSCHEL, F. H. 1866. Das Gebiss der 
Schnecken zur Begründung einer natuhichen 
Classification, vol. 2 Nicolai, Berlin. 

WENZ, W. 1938-1944. Gastropoda. Allegemeiner 
Teil und Prosobranchia. Handbuch der Paläozo- 
ologie. 6, part 1, 1639 pp. 



Revised Ms. accepted 21 June 1990 



MALACOLOGIA, 1990, 32(1): 35-54 

ONTOGENETIC CHANGE IN THE CONUS RADULA, ITS FORM, 

DISTRIBUTION AMONG THE RADULA TYPES, AND SIGNIFICANCE 

IN SYSTEMATICS AND ECOLOGY 

James Nybakken 

Moss Landing Marine Laboratories; Post Office Box 450; Moss Landing, 
California, U.S.A. 95039 

ABSTRACT 

The radula teeth of some species of the carnivorous genus Conus undergo a morphological 
change during their ontogeny. This change is documented for four species in two feeding 
categories. The form of change differs in the different feeding types, but the initial juvenile tooth 
appears similar in all. The change from juvenile to adult tooth appears to occur quickly and 
presumably results from the initiation of the activity of the superior epithelial tissue. Not all 
species of Conus show an ontogenetic change in the radula tooth. For those species showing 
a change, however, there is a correlation of tooth morphology and diet that suggests a close 
coupling of the two. It is suggested that ontogenetic change in morphology will occur in those 
species in which there is a marked change in the diet between juveniles and the adult, but not 
in those in which the prey does not change. 

Key words: Conus, radula, ontogeny, diet, evolution. 



INTRODUCTION 

The radula morphology of prosobranch 
gastropods is known to show a certain 
amount of morphological variability within a 
species (Bändel, 1 974; Borkowski, 1 975; Car- 
riker, 1943; Cernohorsky, 1970; Houbrick, 
1978; Howe, 1930; Merhman, 1967; Rosewa- 
ter, 1970). Such variability has been most of- 
ten reported for adult animals and has been 
occasionally correlated with sex (Arakawa, 
1958, 1959; Maes, 1966; Robertson, 1971). 
Reports of the occurrence of morphological 
change in gastropod radular teeth during on- 
togeny have been less frequent. Reported on- 
togenetic changes in prosobranch and 
opisthobranch gastropods have included in- 
creases in number of teeth per row (Bertsch, 
1976; Robertson, 1985) and alterations in the 
morphology of single teeth in a given row 
(Carhker, 1943; Fujioka, 1985; Hickman, 
1980; Hollister, 1954; Page & Willan, 1988; 
Thompson & Brown, 1984). Despite these 
studies, until recently there do not appear to 
have been any studies that follow ontogenetic 
changes within a species from post-metamor- 
phic juveniles to adults to ascertain when and 
how rapidly the changes occur, and to try to 
ascertain the reason for the changes. 

Ontogenetic change in the radula of Conus 



magus was reported by Nybakken & Perron 
(1988), and a second suspected case in Co- 
nus patricius was reported by Nybakken 
(1988). Rolan (1986) has also reported a dif- 
ference in radula tooth structure between ju- 
venile and adults of C. ermineus. 

Two of the three species for which ontoge- 
netic change has been demonstrated in Co- 
nus, С. magus, and С ermineus, are pisci- 
vores in which the juvenile is too small to 
consume fish. Hence it is probably not unex- 
pected that the radula tooth morphology 
should change with change in diet as Nybak- 
ken & Perron (1988) have demonstrated for 
С magus. However, the finding of a change 
in tooth morphology in a vermivore, С patri- 
cius, suggests that ontogenetic change might 
be more widespread within the genus. This, 
coupled with the knowledge that the different 
radula types in Conus can be associated with 
certain diets (Lim, 1969; Nybakken, 1970), 
led me to embark on a study of the radula 
tooth morphology within a broad size range of 
Conus species representing as many of the 
different tooth types and feeding types as 
were available. The objects of this study were 
to see if I could uncover further instances of 
ontogenetic change, if these changes were 
correlated with a particular tooth type and diet 
or whether such changes were universal 



35 



36 



NYBAKKEN 



throughout the genus, if the juvenile tooth was 
similar in all instances or different, and finally, 
to suggest or speculate as to the reasons for 
the observed changes. 

In order to accomplish the above tasks, it 
was necessary to establish a somewhat more 
elaborate scheme of classification of tooth 
types than that originally established by Lim 
(1969) in order to accommodate all the mor- 
phological types of teeth known to occur in 
Conus. 



METHODS AND MATERIALS 

The Conus specimens used in this study 
came from a number of sources; Conus ma- 
gus were furnished by Frank Perron and ei- 
ther collected by SCUBA in the field in Palau 
or raised in the laboratory from egg capsules; 
С pennaceus were collected in Hawaii by 
Frank Perron. С patricius specimens were 
obtained from the 1967 Pillsbury Expedition 
to the Gulf of Panama (Nybakken, 1 971 ), and 
Los Angeles County Museum, the Academy 
of Natural Sciences of Philadelphia, and Alex 
Kerstitch; Conus fergusoni were obtained 
from the Los Angeles County Museum and 
Alex Kerstitch. Juvenile specimens of C. 
ebraeus, С miliaris, and С. coronatus were 
furnished by Alan Kohn. Juvenile С pulcher 
were furnished by Constance Boone. All 
other adult and juvenile specimens from the 
eastern Pacific were from the Pillsbury Expe- 
dition to the Gulf of Panama (Nybakken, 
1 971 ), or from the author's collection from the 
Gulf of California (Nybakken, 1979) and the 
Galapagos (Nybakken, 1978). Indo-Pacific 
and West African specimens were either in 
the author's collection or from Alan Kohn. 

Each specimen was measured for total 
shell length with a vernier calipers. The shell 
was broken, the animal extracted, and the sex 
recorded. The radula sac was dissected out, 
transferred to a depression slide, and the rad- 
ula teeth freed by treatment with a solution of 
bleach. Freed teeth were washed in two 
rinses of water and mounted directly from wa- 
ter into a polyvinyl-alcohol lactophenol me- 
dium on standard glass slides. Radulae were 
examined under a compound microscope 
equipped either with a differential interference 
contrast system after Normarski or Hoffman 
Modulation Contrast optics. Drawings of indi- 
vidual teeth were made using a drawing tube. 

For specimens smaller than 2 mm in shell 
length, a different technique was employed. 



They were first measured using an eyepiece 
micrometer in a dissecting microscope. The 
shell was then broken away using a fine pair 
of forceps and the animal extracted entire. 
The whole animal was then transferred to the 
first depression on a spot plate and rinsed 
with water; then transferred to the next de- 
pression with acid fuchsin and left for 10-15 
minutes to stain the radula. Next the animal 
was moved through four successive rinses of 
water, which removed much of the stain ex- 
cept from the radula. The animal was then 
soaked in tissue solubilizer (Beckman BTS- 
450 0.5N Quaternary Ammonium Hydroxide 
in Toluene) and heated on a slide warming 
tray at 40-65 С for 2-4 hours. The animal 
was then transferred through Toluene into 
70% ethanol where the radula was usually 
visible as a series of red dots. The radula sac 
was excised and mounted in PVA-K on a slide 
for observation. 

Radula teeth used for scanning electron mi- 
croscopy were removed from water, air dried, 
and placed on double stick tape on standard 
stubs. They were coated with gold in a Po- 
laren sputter-coater unit and examined with 
an IS! SX30 SEM. The SEM mounts were 
used primarily to verify the three-dimensional 
structure of the Conus radula. 

A total of 89 species of Conus were exam- 
ined for the establishment of radula types in 
the genus. They represented species from all 
oceans. (A complete list of species and their 
radula type is found in Appendix 1.) 

In order to examine the ontogeny of the 
radula, I was more limited, both by time and 
by the availability of specimens. I used only 
specimens that I was able to identify or that 
were verified for me. Small specimens of any 
Conus species are not easy to come by; 
hence, this study is not as complete as I 
would have liked it to be. I was able to inves- 
tigate a complete size series of specimens 
from post-metamorphic juveniles to adult only 
for С magus. The only other species for 
which post-metamorphic juveniles and adults 
were available was С pennaceus, but here 
the series was not as complete. Other spe- 
cies for which size series were available 
were; С arcuatus (16.3-41.9 mm), С chal- 
deus (7.4-25.1 mm), C. coronatus (8.2-20.1 
mm), С ebraeus (7.5-32.5 mm), С fergusoni 
(26.1-51.9 mm), C. lucidus (14.4-38.5 mm), 
С patricius (27.1-83.5 mm), С pulcher 
(11.6-80.7 mm), C. tornatus (8.2-20.1 mm), 
and С virgatus (14.9-56.5 mm). Fortunately, 
these species encompass all of the common 



ONTOGENY OF CONUS RADULA 



37 



Barb 1 



Barb 2 
Serration 

Waist 



ANTERIOR 
V----TÍP 



Blade 



Ligament • 




POSTERIOR 



FIG. 1 : A diagrammatic Conus radula tooth illustrat- 
ing the various terms used in describing the mor- 
phology. 



radula types except Type 3, so it was possible 
to obtain a good overview of the potential 
changes in the different radula types. 



RESULTS 

Morphological Classification of the Conus 
Radula 

The individual Conus radula tooth is asym- 
metrical, three dimensional, and may be mor- 
phologically complex (Nybakken, 1970b). A 
system of terminology for the various parts of 
the tooth was provided by Nybakken (1970b) 
and Kohn et al. (1972) and is the one followed 
here. These terms are illustrated in Figure 1 . 

Lim (1967) first recognized that there were 
three morphologically different groups of rad- 
ulae in Conus. These three groups were cor- 
related with the three different feeding types 
within the genus, piscivores, molluscivores. 



and vermivores. Of these three, the vermi- 
vores are the most numerous and also the 
most diverse in terms of tooth morphology. 
Whereas the structure of the teeth of both the 
molluscivores and piscivores is unique and 
consistent, that of the vermivores is not. Ver- 
mivores include a number of different mor- 
phological types, only one of which, that pos- 
sessed by those species which prey on 
amphinomid worms, has been directly corre- 
lated with a specific diet (Nybakken, 1970a). 

In order to undertake this study, it was nec- 
essary to attempt to group the various differ- 
ent radula morphologies into a few more man- 
ageable groups. Because this had already 
been done for the molluscivores and pisci- 
vores by Lim (1969), that left only the vermi- 
vores. Personal observation of the radulas of 
89 species of Conus from all feeding types 
and oceans, coupled with the analysis of an- 
other 21 species that have been illustrated in 
the literature (Bergh, 1895; Piele, 1939; 
Warmke, 1960), plus an unpublished analysis 
of 179 species by Tucker (personal commu- 
nication), of which 113 were different from 
those I studied, suggested that the vermi- 
vores could be grouped into a relatively few 
morphological types, leaving only a few that 
did not fit and that I have chosen to call 
"unique" types. Those with which I am famil- 
iar are illustrated in Figure 2 and described 
here. 

Group 1 radulae are the most common 
among all Conus species and were found in 
34 of the 89 species examined (Fig. 2a). The 
individual tooth has the anterior and posterior 
parts (demarcation by the waist) approxi- 
mately equal. The anterior half is terminated 
by a single barb and has opposite the barb, a 
blade that extends posteriorly more than half- 
way to the waist. It may or may not be termi- 
nated by a barb. A serration is present that 
extends posteriorly to the level of the end of 
the blade, or to the level of the waist. The 
serration usually terminates in a prominent 
cusp. Scanning electron microscopy shows 
that the serration and terminal cusp are in fact 
internal (Fig. 3). The posterior half of the tooth 
is usually slightly greater in diameter than the 
anterior and has a slightly enlarged, usually 
rounded base bearing a prominent but small 
spur. For the 17 species for which the food is 
known, all are vermivores (Table 1). 

Two additional tooth types are similar to 
those of Group 1 . In fact. Tucker groups them 
together with Group 1. However, they are 
both morphologically distinct and readily dis- 



38 



NYBAKKEN 





0.5 mm 



d 



0.1 mm 




g- 



0.0625 mm 



FIG. 2: The major morphological groups of Conus radulae. a. Type 1. b. Type la. с Type 1b. d. Type 2. e. 
Type 3. f. Type 4. g. Radula of C. ebraeus. 



ONTOGENY OF CONUS RADULA 



39 




FIG. 3. a. Scanning electron micrograph of the tooth of С virgatus showing the internal position of the serration, b. 
Scanning electron micrograph of the waist of a tooth of С princeps showing the internal position of the cusp. 



40 



NYBAKKEN 



TABLE 1 . Conus Radula Types and Reported Food 



Conus Species 



Food 



Reference 



Radula Type 1 
С chaldeus 

С miles 



С abbreviatus 



С sponsalis 



С rattus 



С tiaratus 



С I i tie rat us 
С leopardus 
С taeniatus 



С arenatus 
С ceylanensis 



С scabriusculus 
С capitaneus 

С vexillum 
С balteatus 
С nniliaris 



С coron at us 



С nux 



Platy nereis dumereiii 
Palola siciliens i s 
Palola siciliensis 
Lys id i ce collaris 
Eunice antennata 
Perinereis helleri 
Platynereis dumereli 
Lysidice collaris 
Eunice antennata 
Eunice cariboea 
Eunice filamentosa 
Marphysa sanguínea 
Lumbnnereis sarsi 
Arabella iricolor 
Nereis jacksoni 
Perenereis helleri 
Platynereis dumereli 
Lysidice collaris 
Eunice cariboea 
Lumbrlnereis sarsi 
Perenereis helleri 
Eunice antennata 
Eunice afra 
Nereis jacksoni 
Neanthes spp. 
Eunice afra 
Eunice biannulata 
Lumbrlnereis sp. 
Dasybranchus caducus 
Ptychodera flava 
Ceratonereis 
Platynereis 
Eunicidae 
Perinereis 
Ceratonereis 
Palola 
Perenereis 
Nereis 
Ceratonereis 
Lysidice 

Palola siciliensis 
Nereis 
Eunice afra 
Eunice australis 
Eunice afra 
Onuphis sp. 
Perinereis spp. 
Lysidice collaris 
Eunice afra 
Eunice rubra 
Palola siciliensis 
Eunice cariboea 
Palola siciliensis 
Lysidice collaris 
Arabella iricolor 
Glycera tessellata 
Nereidae 
Eunicidae 
Syllidae 



Kohn, 1959 

Kohn, 1959; Kohn & Nybakken, 1975 

Kohn, 1959 



Kohn, 1959 

Kohn, 1959 
Nybakken, 1978 



Kohn, 1980 
Kohn, 1980 
Taylor & Reid, 1984 



Kohn & Nybakken, 1975 



Kohn & Nybakken, 1975 



Kohn & Nybakken, 1975 
Kohn & Nybakken, 1975 

Kohn & Nybakken, 1975 
Kohn & Nybakken, 1975 
Kohn, 1968a 



Marsh, 1971 



Kohn & Nybakken, 1975 
Nybakken, 1979 
Nybakken, 1971 



ONTOGENY OF CONUS RADULA 



41 



TABLE 1 . {Continued) 



Conus Species 


Food 


Reference 




Polynoidae 




С gladiator 


Eunice afra 
Eunice filamentosa 
PI aty nereis polyscalma 


Nybakken, 1979 


С regularis 


Nothria elegans 


Nybakken, 1979 


С pulicarius 


Dasybranchus caducus 


Kohn, 1959 




Nematonereis unicornis 


Taylor, 1986 


С vexillum 


Eunice antennata 
Marpliysa sanguínea 


Kohn, 1959 


Radula Type la 






C. princeps 


Eunice afra 
Eunice filamentosa 
Palola siciliensis 


Nybakken, 1979 


С fl avid и s 


Dasybranchus caducus 


Marsh, 1971 




Ampharetidae 


Kohn, 1959 




Capitellidae 






Maldanidae 






Terebellidae 






Eunicidae 




С. frigidus 


Dasybrancfius caducus 
Terebellidae 


Kohn, 1968b 


С. virgo 


Loimia medusa 


Kohn & Nybakken, 1975 


С. patricius 


Aphroditidae 
Spionidae 


Nybakken, 1988 


С. ventricosus 


Perinereis 

Palola 

Capitella 


Taylor, 1987 


Radula Type lb 






С. dalli 


gastropods 


Nybakken, 1968 


С. pennaceus 


Cypraea, Dolabrifera 


Kohn & Nybakken, 1975 


С. marmoreus 


molluscs 


Kohn, 1980 


С. textile 


Conus 


Kohn, 1968 


С. episcopus 


gastropods 


Kohn & Nybakken, 1975 


Radula Type 2 






С. catus 


fish 


Kohn& Nybakken, 1975 


С. striatus 


fish 


Kohn & Nybakken, 1975 


С. purpurascens 


fish 


Nybakken, 1967 


С. magus 


fish 


Nybakken & Perron, 1988 


Radula Type 3 






С brunneus 


amphinomid worms 


Nybakken, 1970; Nybakken, 1979 


C. zonal us 


Eurythoe 


Kohn & Nybakken, 1975 


C. imperialis 


Eurythoe 


Kohn, 1959 


Radula Type 4 






С lucidus 


Ampharete 

Lygadamis 

Sabellariidae 

Sabellinae 

Capitellidae 


Nybakken, 1978 




Nereidae 


Nybakken, 1979 


С arcuatus 


Onuphidae 




Radula Type — Unique 






С ebraeus 


Nereidae 


Kohn & Nybakken, 1975 


С lividus 


Terebellidae 


Kohn, 1959 




Nereidae 


Nybakken, 1979 




Cirratulidae 






Ptychodera 






Platynereis 






Phyllodocidae 






Maldanidae 





(continued) 



42 



NYBAKKEN 



TABLE 1. (Continued) 



Conus Species 



Food 



Reference 



С diadema 



С. californicus 



С. tor natu s 



Eunicidae 

Eurytiioe 

Terebellidae 

gastropods 

gastropods 

bivalves 

cephalopods 

polychaetes 

amphipods 

fish 

Nepfityidae 



Nybakken, 1979 



Kohn, 1966 



Nybakken, 1979 



cernable from Type 1 ; hence, I choose to give 
them separate status. The tooth type I have 
designated as Type la differs from Type 1 in 
that the anterior part of the tooth is elongated, 
the serration is proportionately longer, and 
the blade shorter than Type 1 (Fig. 2b). Usu- 
ally the cusp is more prominent also. As with 
Type 1 , the serration is internal for most of its 
length. For the five species for which food is 
known, all are vermivores, usually taking 
tube-dwelling polychaetes. I found this type in 
14 species investigated. 

The second related type, which I have des- 
ignated Type lb, is more different from Type 
1 than is Type 1 a (Fig. 2c). In this radula type, 
the anterior portion is extremely elongated, 
usually several times the length of the poste- 
rior region, and the serration runs the entire 
length of the anterior part. The serration may 
be external, as in С pennaceus (Fig. 4a), or 
completely internal, as in С dalli (Fig. 4b). 
There is no waist. In addition, the antenor tip 
has two unequal-sized barbs, the smaller of 
which is inflated laterally so as to appear as a 
spear blade (Fig. 4c). There is no blade and 
no spur. This tooth type is characteristic mol- 
luscivores (Table 1). I found this radula in 14 
species studied. 

The radula tooth type I have designated as 
Type 2 is unique to piscivorous Conus (Fig. 
2d, Table 1). Each tooth is very large, con- 
sisting of a very elongated shaft with no evi- 
dence of a waist or a serration. The anterior 
tip of the tooth has an armature of two oppos- 
ing barbs followed posteriorly by a very large 
third barb that protrudes outward. There is no 
spur. There appears to be a slight difference 
in the teeth of Indo-Pacific piscivores and 



those of the eastern Pacific and Atlantic. 
Those from the former have the tip of the 
large third barb recurved at the end, whereas 
those from the latter do not. I have found this 
tooth type in eight species. 

Another highly distinctive tooth type is that 
designated as Type 3 (Fig. 2e). Teeth of this 
morphological construction are characteristic 
of cones feeding on amphinomid worms (Ny- 
bakken, 1970a; Table 1). Diagnostic of these 
teeth is the presence of four barbs near the 
shortened anterior end. One of these barbs 
juts out from the tooth to form a prominent 
angle with the shaft. This barb also bears a 
short serration. The most prominent barb is 
the one with the greatest length and termi- 
nates a large blade. All barbs are pointed and 
none are recurved or hooked. Posterior to the 
barbs is a slight waist. Posterior to the waist 
the shaft expands to its maximum diameter 
and ends in a massive base that bears a large 
spur. There is no cusp. Tucker includes this 
tooth with Type 1 . I have found this tooth type 
in eight species. 

The tooth type designated as Type 4 is 
characterized by a shortened anterior section 
that bears two or three barbs but no serration 
and no blade (Fig. 2f). The barb nearest the 
antenor tip is always pointed, but the remain- 
ing one or two may be pointed or blunt. In 
addition, these teeth usually show evidence 
of a peculiar anterior fold. The waist is usually 
prominent, and the posterior part of the shaft 
is usually longer and broader than the anterior 
half. The base is very large and bears a very 
large spur. There is no cusp. Tucker has di- 
vided this group into two different groups 
based upon barb number. Conus species 



ONTOGENY OF CONUS RADULA 



43 




В 




i В0к X. 



15км газ 




FIG. 4. а. Scanning electron micrograph of the anterior tip of the tooth of a С pennaceus showing the 
external position of the serration, b. Scanning electron micrograph of the anterior tip of the tooth of a C. dalli 
showing the internal position of the serration, с Scanning electron micrograph of the anterior tip of the tooth 
of a С pennaceus. showing the inflated barb. 



44 



NYBAKKEN 



showing this radula type appear to be mainly 
deeper water dwellers on soft substrates. 
Food data for this tooth type are quite sparse, 
but they appear to feed on worms. I found this 
type in only five species, but this may be due 
to the lack of deeper water cones available for 
study. 

There are also some teeth that I would des- 
ignate as unique, apparently confined to one 
or two species. Those of which I am aware 
are shown in Figures 2g and 5. Of these, C. 
californicus has the most catholic diet, feed- 
ing upon molluscs, polychaetes, Crustacea 
and fish ; С ebraeus feeds on nereid polychae- 
tes: С tornatus on nephtyid polychaetes; and 
С lividus and С diadema both feed primarily 
on polychaetes (Table 1). 

It is by no means certain that the above 
designations represent all the tooth types 
present in the living Conidae. However, given 
the probable number of species (350-380) 
and the total number of species examined 
here (110), it does not seem likely that any 
large radula group remains unrepresented. 

Radula Classification and Correlation with 
Food in the Vermivorous Conus 

Whereas the molluscivores and piscivores 
seem to each possess a single characteristic 
radula morphology, the same does not ap- 
pear to be true for vermivores. Types 1 , 1a, 3, 
4, and at least two of the unique types (C 
lividus and С ebraeus) are all tooth types as- 
sociated with vermivory (Table 1). With the 
exception of radula Type 3, which seems to 
be specific for polychaetes of the family Am- 
phinomidae, the other vermivores cannot be 
correlated with a single worm family. The rad- 
ula tooth type represented by С tornatus may 
yet prove to be associated with a single poly- 
chaete family, because the only food remains 
found have been of the family Nephtyidae, a 
family otherwise not present in any other Co- 
nus for which food data are available (Nybak- 
ken, 1979). 

Analysis of Table 1 suggests that most of 
Conus species with Type 1 radulae feed on 
errant polychaetes primahly of the families 
Nereidae and Eunicidae. Those with Type la 
radulas feed more often on polychaetes of the 
families Terebellidae and Capitellidae. How- 
ever, because one species with this radula 
morphology, С princeps, feeds on eunicids 
and nereids, this evidence is hardly conclu- 
sive. 

Food data are very scarce for Conus spe- 



cies with radula Type 4. The data for С luci- 
dus indicated that cones with this tooth type 
are vermivores feeding on several families of 
sedentan/ polychaetes (Table 1). 

Conus lividus and С diadema share a rad- 
ula tooth morphology that so far has not been 
found in any other Conus species. It is most 
like Type lb but lacks a serration. For vermi- 
vores, these two species consume the widest 
vanety of food taxa, six families of polychae- 
tes, both errant and sedentary, as well as en- 
teropneusts and other gastropods. 

The most catholic diet of any Conus is that 
of С californicus, which is known to consume 
other gastropods, bivalves, fish, worms, and 
Crustacea. This species also has a unique 
radula tooth. 

Ontogenetic Change 

In order to attempt to document whether or 
not an ontogenetic change occurred within 
each of the major radula types, it was neces- 
sary to dissect as broad a size range of indi- 
viduals in each category as possible. Unfor- 
tunately, very small cones are not abundant in 
collections and immediate post-metamorphic 
specimens are even more rare. As a result of 
these difficulties, I was able to obtain post- 
metamorphic individuals of only radula Types 
1 b and 2. Good size ranges were available for 
some species of radula Types 1 and la, and 
that of С ebraeus, but there were no small 
specimens of Type 3. 

The most conclusive data documenting a 
profound ontogenetic change in the radula are 
found in those Conus species that consume 
fish (Type 2 tooth). This change was docu- 
mented by Nybakken & Peron (1988) for Co- 
nus magus and Rolan (1 986) for С ermineus. 
Both showed that the juvenile radula differed 
from that of the adult in lacking all three barbs, 
in size, and in presence of a spur. 

Since that time, a set of post-metamorphic 
specimens of С magus has become avail- 
able, and dissection of these animals, all be- 
low 2.0 mm in shell length, has revealed the 
presence of even more changes. The imme- 
diate post-metamorphic radula tooth is less 
than 0.08 mm in length (in an animal of 1.7 
mm shell length), has no barbs or blades, and 
is only slightly folded such that the central lu- 
men appears to be at least partially open for 
the entire length. The base is large but lacks 
the spur found in the later juvenile tooth (Fig. 
6). 

Analysis of a size senes of Conus penna- 



ONTOGENY OF CONUS RADULA 



45 




0.1 mm 



FIG. 5. Unique radula types, a. Radula of С lividus 
and C. diadema, b. Radula of С ximenes and С 
mahogani. c. Radula of С californicus. d. Radula of 
С tornatus. 



ceus, a molluscivore, from 9.6 mm to 33.4 
mm revealed no change in the radula mor- 
phology (Fig. 7a, b). However, a series of 
post-metamorphic Conus ornaría was avail- 
able from Perron. These animals had shell 
lengths of about 1.4 mm. Dissection of these 
animals revealed a radula tooth almost iden- 
tical to that found in the post-metamorphic С 
magus (Fig. 7c). This tooth was less than 0.06 
mm in length, had no barbs, blades, or serra- 
tions, and the central lumen was open 
throughout its length. 

Although radula Type 1 is by far the most 
common, very small specimens of species 
exhibiting this type were difficult to come by. I 




FIG. 6. Ontogeny of the radula tooth in С magus, a. 
Radula tooth of a post-metamorphic juvenile, b. 
Radula tooth of a juvenile, с Transitional radula 
tooth, d. Radula tooth of an adult. 



examined a series of teeth from С virgatus 
ranging in size from 56.5 mm to 14.9 mm but 
found no change. I examined a series of С 
chaldeus ranging in size from 25.1 mm down 
to 7.4 mm. In this senes, the smallest individ- 
ual had a tooth that differed from the adult in 
lacking a serration and a blade (Fig. 8b). The 
tooth was folded but did have a very large 
anterior lumen. 

A series of specimens of С ebraeus from 
33.2 mm down to 7.5 mm in shell length was 
dissected. In this case, the smallest individual 
possessed a radula clearly of the juvenile 
type without serration, barb, or blade (Fig. 8c, 
d) and resembling the juvenile tooth of Types 
1 and 1a. 

For radula tooth Type 1 a, the presence of a 
juvenile tooth differing from the adult has 
been described for С pathcius by Nybakken 
(1988) (Fig. 9a, b). In this study two additional 



46 



NYBAKKEN 



0.02 mm 



0.02 mm 



0.02 mm 



FIG. 7. Ontogeny of the radula tooth in C. penna- 
ceus. a. Radula tooth of an adult (shell length 33.4 
mm), b. Radula tooth of an animal of shell length 
9.6 mm. с Radula tooth of a post-metamorphic С 
ornarla of shell length 1 .4 mm. 




b. 




FIG. 8. a. Adult tooth of С chaldeus. b. Juvenile 
tooth of C. chaldeus. c. Adult tooth of С ebraeus. d. 
Juvenile tooth of С ebraeus. 




0.1 mm 



FIG. 9. a. Tooth of an adult С patrlclus. b. Tooth of 
a juvenile С patrlclus. 

No very small Conus bearing Type 4 teeth 
were available for dissection. A size series of 
C. arcuatus from 41.9 mm down to 21.0 mm 
failed to reveal any radula change. Similarly, 
a size series from C. lucidus of 38.5 mm down 
to 14.4 mm also failed to reveal any radula 
change. Since no animals of 10 mm or less in 
shell length were available, we cannot at 
present assess if there is an ontogenetic 
change in this radula type. 



DISCUSSION 



species, С pulcher and С fergusoni. have 
also been discovered to have different juve- 
nile teeth (Figs. 10 and 11). In C. fergusoni. 
the smallest individual dissected was only 
26.1 mm and the tooth was similar to tooth 
Type 1 (Fig. 10a). In С pulcher \he smallest 
specimen dissected was 1 1 .6 mm. Each tooth 
in this specimen was more similar to that of 
the juvenile С patrlclus and lacked a serra- 
tion, cusp, and blade and had a single barb 
(Fig. 11c). 



These studies have established that there 
is an ontogenetic change in the radula of Co- 
nus species of a number of different tooth 
types that include all three main feeding 
types. These changes are documented for 
the greatest range of shell size for Type lb 
(molluscivores) and Type 2 (piscivores) 
where immediate post-metamorphic juveniles 
were available for study. The striking similar- 
ity between the tooth types of the post-meta- 
morphic juveniles in these two otherwise very 



ONTOGENY OF CONUS RADULA 



47 



a. 



0.1 mm 



0.1 mm 



0.1 mm 




0.025 mm 



0.1 mm 



FIG. 10. a. Tooth of a juvenile С fergusoni. b. Tooth 
of an adult С fergusoni. 



different feeding groups, as opposed to the 
differences in the adult teeth, strongly sug- 
gests that perhaps all immediate post-meta- 
morphic juvenile Conus have similar teeth. It 
also suggests that the food may well be sim- 
ilar, because Nybakken & Perron (1988) have 
demonstrated that the food of juvenile C. ma- 
gus is worms, not fish. 

The post-metamorphic tooth does not ap- 
pear to be strongly chitinized, as it only 
weakly takes up dyes specific for chitin (acid 
fuchsin). Furthermore, it is only slightly rolled, 
such that the lumen is open for the entire 
length. Whether or not the tooth is functional 
is not known, but Shimek (personal commu- 
nication) has observed in certain turrids a 
similar type of tooth in which the tooth is rolled 
to form a tube by the proboscis before use. 

If the teeth of immediate post-metamorphic 



FIG. 11. a. Radula tooth of an adult С pulcher o^ 
shell length 79.2 mm. b. Radula tooth of an animal 
of shell length 27.8 mm. с Radula tooth of a juve- 
nile of shell length 11.6 mm. 



specimens eventually prove to have the same 
structure in all Conus species, then the differ- 
ence in morphology observed among adult 
teeth would seem to be initiated in the juvenile 
teeth as this study suggests. Although teeth 
different in morphology from the adult were 
found in juveniles in all three major feeding 
types, the morphology of tooth was not the 
same in each juvenile. The most prevalent 
juvenile tooth type, and very similar in adult 
Types 1 , 1 a, 2, 4, is a tooth with a single fold 
closing the lumen but leaving a very large an- 
terior opening; having no, or at most one, 
barb; no blade; no serration, and a large 
base. In C. magus, Nybakken & Perron 



48 



NYBAKKEN 




FIG. 12, Scanning electron micrograph of a juvenile С magus showing the development of the bulge that 
will become the large, recurved barb of the adult. 



(1988) have demonstrated that this tooth is 
correlated with feeding on syllid poiychaetes. 
In the moliuscivores (Type lb tooth), dissec- 
tion of a series of specimens of С pennaceus 
from 33.4 down to 9.6 mm failed to reveal any 
change from the adult radula. Yet, as noted 
above, the immediate post-metamorphic ani- 
mals of a closely related species, C. omaria. 
have a very similar tooth to post-metamorphic 
C. magus of the same shell length. Perron 
(personal communication) has noted that ju- 
veniles of С pennaceus feed on small gas- 
tropods, so the food type is similar to the 
adult. It is not in С magus. Therefore, it may 
be suggested that the reason a different juve- 
nile tooth is not seen in С pennaceus is that 
there is not significant change in diet such as 
seen in С magus. However, until we can ex- 
amine radulae from specimens of С penna- 
ceus between 9.0 mm and 2.0 mm, this must 
remain only a suggestion. 
In these juvenile teeth, although similar, 



there are indications of the morphological 
changes that will shape the adult teeth. Thus, 
in С magus there is at this stage the devel- 
opment of a prominent bulge on the tooth 
at the level where the adult wilt have the 
characteristic recurved barb (Fig. 12). Conus 
pathcius has an elongated anterior portion of 
the tooth corresponding to the elongated an- 
terior portion in the adult (Fig. 13). In С fer- 
gusoni, although we lack small specimens, 
those with shell lengths of 26 mm show an 
advanced juvenile tooth indistinguishable 
from a Type 1 tooth (Fig. 1 0). In this case, the 
adult tooth would seem to be derived from the 
juvenile by simple differential growth of the 
anterior region. 

A significant gap in the current analysis is 
the unavailability of very small specimens of 
cones with radula Types 3 and 4. It is espe- 
cially unfortunate for radula Type 3 because 
of our knowledge that the adults consume 
only one family of poiychaetes, Amphinomi- 



ONTOGENY OF CONUS RADULA 



49 




FIG. 13. Scanning electron micrograph of the juvenile tooth of С patricius showing the slender, elongated 
anterior end. 



dae. Although a fairly good size series of spe- 
cies with Type 4 radula was available in C. 
arcuatus and С lucidus. no change in tooth 
morphology was discernable; the need for 
srnaller sizes and post-metamorphic speci- 
mens is apparent. 

Observation of all of the above radula types 
and compahson of the adult and juvenile sug- 
gests that the location of the greatest change 
from juvenile to adult is in the distal half of the 
tooth. The basal portion seems to change lit- 
tle from juvenile to adult (Figs 8-1 1 ), whereas 
the distal half undergoes significant changes. 

It is also significant to note that it appears 
that the serration, where it is present, may be 
external or internal. If internal, it is of little or 
no use in any cutting or penetrating action. 
Where it is exposed, such as in С penna- 
ceus, it lies very close to the overlapping fold, 
suggesting that only a little change in growth 
or folding could make it internal. However, rel- 
atively few molluscivores were available for 
SEM work, so the extent of either internal or 



external serrations In this group is not known. 
It is also not known if this internal or external 
position of the serration correlates with any 
particular molluscan prey item. 

What initiates the radula change? This is 
unknown at present, but Marsh (1977) has 
shown that the Conus tooth is actually the 
product of two tissues, the odontoblasts, 
which make the initial tooth, and the superior 
epithelium which finishes the tooth. It is pos- 
sible that the form of the juvenile tooth is the 
product of the odontoblasts and that the adult 
tooth represents the finishing work of the su- 
perior epithelium. 

The establishment of the existence of a 
change in radula morphology within a single 
species with ontogeny and the finding that 
this change is widespread in the genus 
among all food types has implications for sys- 
tematics and ecology in Conus. 

In the first place, Nybakken & Perron 
(1987) and Nybakken (1988) have demon- 
strated that the change in radula morphology 



50 



NYBAKKEN 



can occur rapidly and within a narrow range of 
shell lengths. This means that specimens of 
different shell lengths that have different rad- 
ulas may not necessarily be different species. 
Now that we have established the likely mor- 
phology of the juvenile tooth in this study and 
that morphology seems to have certain rec- 
ognizable characteristics, it would seem that 
this should facilitate juvenile recognition and 
reduce errors with respect to use of the radula 
in taxonomy. It is also of importance, there- 
fore, to note the shell length of any specimens 
used in radula studies and when comparing 
specimens to compare only specimens of the 
same shell size. Given the fact that in C. pat- 
ricius the juvenile radula may persist into an- 
imals with shells and shell lengths of the adult 
aspect and morphology, this means there 
may be no a priori way to predict whether in 
some species a specimen of a given size will 
have the adult radula. However, this may be a 
feature restncted to only a few feeding types, 
because there is no evidence for it over a 
rather large size range in molluscivores (C. 
pennaceus) and some vermivores (C iuci- 
dus. С virgatus. С arcuatus). 

The fact that Nybakken & Perron (1988) 
conclusively demonstrated that the juvenile 
radula in piscivores is correlated with a differ- 
ent diet than the adults suggests that many of 
the other species in which the juvenile radula 
is different from the adult may also prey upon 
different food items when small. However, at 
this time no data exist to prove or disprove 
this contention. 

The results from Nybakken & Perron (1988) 
and Nybakken (1988) with С magus and С 
patricius. respectively, have also demon- 
strated that the observed radula change is not 
due to sexual dimorphism. Juvenile radulae 
were found in both sexes. 

Given the vanous radula types and their 
distribution among the species, which is likely 
the most primitive type and which are the de- 
rived? There are several ways of looking at 
this problem. The simplest is to employ the 
commonality principle, otherwise stated as 
"common equals primitive" (Wiley, 1981). 
This principle simply states that if a character 
is widely distributed within a taxon, then the 
character is likely primitive. Employing this 
principle and considering that Type 1 is by far 
the most common type suggests it is also the 
most primitive. 

Another criterion for determining primitive 
or derived is that of ontogenetic precedence 
(Hennig, 1966). This criterion assumes that 



the ontogenetic transformation toward a par- 
ticular character reflects the phylogenetic de- 
velopment of that ontogeny. Employing this 
criterion and observing all thie juveniles in this 
study suggests that the development of at 
least one anterior barb is a primitive feature 
and that it comes before the development of a 
serration. 

My tentative conclusion, therefore, is that a 
modified Type 1 tooth would seem to be the 
most primitive, perhaps without a serration, 
and that the others are derived. The likelihood 
that the primitive tooth is without a serration is 
also given support by the fact that the toxo- 
glossan teeth in the more primitive family Tur- 
ridae also lack a serration (Shimek & Kohn, 
1981). 



LITERATURE CITED 

ARAKAWA, K. Y., 1958, On the геглагкаЫе sexual 
dimorphism of the radula of Drupella. Venus. 19: 
206-214. 

ARAKAWA, K. v., 1969, On some problems of sex- 
ually dimorphic radulae. Venus. 28:125. 

BÄNDEL, К. 1974. Studies on the Littorinidae from 
the Atlantic. Veliger. 17:92-114. 

BERGH, R., 1 895, Beiträge zur Kenntniss der Coni- 
den. Nova Acta der Kaiserlich. Leopoldinisch- 
Carolinsch Deutsclien Akademie der Naturfor- 
sciier. 65(2):69-214: 13 pis. 

BERTSCH, H. 1976. Intraspecific and ontogenetic 
radula variation in opisthobranch systematics 
(Mollusca: Gastropoda). Systematic Zoology. 25: 
117-122. 

BORKOWSKI, T. V. 1975. Vanability among Canb- 
bean Littorinidae. Veliger. 17:369-377. 

CARRIKER, M. R., 1943, Vanability, developmental 
changes and denticle replacement in the radula 
of Lymnaea stagnalis appressa Say. Nautilus, 
57:52-59. 

CERNOHORSKY, W. O., 1970, Systematics of the 
families Mitridae and Volutomitridae (Mollusca: 
Gastropods). Bulletin of the Auckland Institute 
and Museum. 8:1-190. 

FUJIOKA, Y., 1985, Seasonal abberant radular for- 
mation in Thais bronni (Dunker) and T. clavigera 
(Kunster) (Gastropods: Muricidae). Journal of 
Experimental Marine Biology and Ecology, 90: 
43-54. 

HENNIG, W., 1966, Phylogenetic systematics. Uni- 
versity of Illinois Press, Urbana, 263 pp. 

HICKMAN, C. S., 1980, Gastropod radulae and the 
assessment of form in evolutionary paleontology. 
Paleobiology, 6:276-294. 

HOLLISTER, S. C., 1954, Some notes on the rad- 
ula. Nautilus. 68:44-46. 

HOUBRICK, R. S., 1978, The family Centhiidae in 
the Indo-Pacific, Part 1 . The genera Rhinoclavus, 



ONTOGENY OF CONUS RADULA 



51 



Pseudoverlagus and Clavocehthium. Mono- 
graphs of Marine Mollusca. 1 :1-130. 

HOWE, S. W,, 1 930, A study of the variations in the 
radula of a snail with particular reference to the 
size of the median teeth. Nautilus, 44:53-56. 

KOHN, A. J., 1959, The ecology of Conus in Ha- 
waii. Ecological Monographs, 29:47-90. 

KOHN. A. J., 1966, Food specialization in Conus in 
Hawaii and California. Ecology, 47:1041-1043. 

KOHN, A. J., 1968a, Ecological shift and release in 
an isolated population: Conus miliaris at Easter 
Island. Ecological Monographs, 48:323-336. 

KOHN, A. J., 1968b, Microhabitats, abundance and 
food of Conus on atoll reefs in the Maldive and 
Chagos islands. Ecology, 49:1046-1062. 

KOHN, A. J., 1980, Abundance, diversity, and re- 
source use in an assemblage of Conus species in 
Eniwetak lagoon. Pacific Science, 34:359-367. 

KOHN, A, J. & J. NYBAKKEN 1975, Ecology of 
Conus on eastern Indian Ocean fringing reefs: 
diversity of species and resource utilization. Ma- 
rine Biology, 29:21 1-234. 

KOHN, A. J., J. W. NYBAKKEN & С J. VAN MOL, 
1972, Radula tooth structure of the gastropod 
Conus imperialis elucidated by scanning electron 
microscopy. Science, 176:49-51. 

LIM, С. F., 1969, Identification of the feeding types 
in the genus Conus Linnaeus. Vellger, 12:160- 
164. 

MAES, V. O., 1966, Sexual dimorphism in the rad- 
ula of the genus Nassa. Nautilus, 79:73-80. 

MARSH, H., 1971, Observations on the food and 
feeding habits of some vermivorous Conus on 
the Great Barrier Reef. Vellger, 14:45-53. 

MARSH, H., 1977, The radula apparatus of Conus. 
Journal of Molluscan Studies, 43:1-1 1 . 

MERRIMAN, J. A. 1967. Systematic implications of 
the radular structures of West Coast species of 
Tegula. Vellger, 9:399-403. 

NYBAKKEN, J., 1967, Preliminary observations on 
the feeding behavior of Conus purpurascens Bro- 
derip, 1833. Vellger, 10:55-57. 

NYBAKKEN, J., 1968, Notes on the food of Conus 
da/// Stearns, 1873. Vellger. 11:50. 

NYBAKKEN, J., 1970a, Correlation of radula tooth 
structure and food habits of three vermivorous 
species of Conus. Vellger, 12:316-318. 

NYBAKKEN, J., 1970b, Radular anatomy and sys- 
tematics of the West American Conidae. Ameri- 
can Museum of Natural History Novitates, 2414: 
1-29. 

NYBAKKEN, J., 1 971 , The Conidae of the Pillsbury 
Expedition to the Gulf of Panama. Bulletin of Ma- 
rine Science, 21:93-110. 

NYBAKKEN, J., 1978, Population characteristics 
and food resource utilization of Conus in the 



Galapagos Islands. Pacific Science, 32:271- 
280. 

NYBAKKEN, J., 1979, Population characteristics 
and food resource utilization of Conus in the Sea 
of Cortez and West Mexico. Journal of Molluscan 
Studies, 45:82-97. 

NYBAKKEN, J., 1988, Possible ontogenetic 
change in the radula of Conus patriclus of the 
Eastern Pacific. Veliger, 31 :222-225. 

NYBAKKEN, J. & F. PERRON, 1988, Ontogenetic 
change in the radula of Conus magus (Gas- 
tropoda). Marine Biology, 98:239-242. 

PAGE, A. J. & R. С WILLAN, 1988, Ontogenetic 
change in the radula of the gastropod Epitonium 
bllleeana (Prosobranchia: Epitoniidae). Veliger, 
30:222-229. 

PIELE, A. J., 1939, Radula notes, VIII. Conus. Pro- 
ceedings of the Malacological Society of London, 
23:348-356. 

ROBERTSON, R., 1971, Sexually dimorphic ar- 
chaeogastropods and radulae. Annual Report of 
the American Malacological Union for 1970: 75- 
78. 

ROBERTSON, R., 1985, Archaeogastropod biol- 
ogy and the systematics of the genus Tricolia 
(Trochacea: Tricoliidae) in the Indo-West Pacific. 
Monographs of Mahne Mollusca, 3:1-103. 

ROLAN, е., 1986, Estudio de la radula de Conus 
ermineus Born, 1 778 desde el periodo juvenil al 
adulto. Publicacoes Ocasionáis da Sociedade 
Portuguesa de Malacologla, 6:23-28. 

ROSEWATER, J., 1970, The family Littorinidae in 
the Indo-Pacific. Part 1. The subfamily Littorini- 
nae. Indo-Pacific Mollusca, 2:417-479. 

SHIMEK, R. L. & A. J. KOHN, 1981, Functional 
morphology and evolution of the toxoglossan 
radula. Malacologla, 20:423-438. 

TAYLOR, J. D., 1 986, Diets of sand-living predatory 
gastropods at Piti Bay, Guam. Asian ti/lahne Bi- 
ology, 3:47-58. 

TAYLOR, J. D., 1987, Feeding ecology of some 
common intertidal neogastropods at Djerba, Tu- 
nisia. Vie et Milieu, 37:13-20. 

TAYLOR, J. D. & D. G. REÍD, 1984, The abun- 
dance and trophic classification of molluscs upon 
coral reefs in the Sudanese Red Sea. Journal of 
Natural History, 18:175-209. 

THOMPSON, T. E. & G. BROWN, 1984. Biology of 
opisthobranch molluscs, Vol. II. The Ray Society, 
London, 229 pp. 

WARMKE, G. L., 1960, Seven Puerto Rico cones: 
Notes and radulae. Nautilus, 73:119-124. 

WILEY, E. O. 1981. Phylogenetics, the theory and 
practice of phylogenetic systematics. John Wiley 
& Sons, New York, 439 pp. 

Revised Ms. accepted 21 June 1990 



52 



NYBAKKEN 



APPENDIX 1 
Alphabetical List of Conus Species Examined for Radula Tooth Structure 



C. abbreviatus 


Type 1 


С nigropunctatus 


Type 2 


С aemulus 


9 


С nux 


Type 1 


С amadis 


Type lb 


С omaria 


Type 1b 


C. ambiguus 


Type 1 


C. orion 


Type 1 


С araneosus 


Type 1b 


С paniculus 


Type 1b 


С archon 


Type 3 


С pat rictus 


Type la 


С arcuatus 


Type 4 


С pennaceus 


Type 1b 


C. arenatus 


Type 1 


С perplexus 


Type 4 


С aurora 


Type 1 


С piperatus 


Type 1 


С bal teat us 


Type 1 


С poormani 


Type 1 


С bartschi 


Type 3 


С princeps 


Type la 


С brunneus 


Type 3 


С. pulcher 


Type la 


С bulbus 


7 


С. purpurasceus 


Type 2 


С californicus 


Unique 


С. rattus 


Type la 


С capitaneus 


Type 1 


С. recurvus 


Type 4 


С cat us 


Type 2 


С. regularis 


Type 1 


С centurio 


Type 4 


С. scabnusculus 


Type 1 


С ceylanensis 


Type 1 


С. scalaris 


Type 1 


С chaldeus 


Type 1 


С. scitulus 


Type 1 


С CO ron at и s 


Type 1 


С. simplex 


Type 1 


С. dalli 


Type 1b 


С. stercomuscarum 


Type 2 


С. diadema 


Unique 


С. stnatellus 


Type 1 


С. dispar 


Type 1 


С. textile 


Type 1b 


С. distans 


Modified Type 3 


С. tiaratus 


Type 1 


С. dorriensis 


Type 1 


С. tornatus 


Unique 


С. ebraeus 


Type 5 


С. tulipa 


Type 1b 


С. elongatus 


Type 1 


С. varius 


Type 1 


С. emaciatus 


Type la 


С. ventricosus 


Type la 


С. episcopus 


Type lb 


С. venulatus 


Type 1 


С. ermineus 


Type 2 


С. vexillum 


Type 1 


С fergusoni 


Type la 


С. victor 


Type 1 


С. frigidus 


Type la 


С. victonae 


Type 1b 


С. furvus 


Type lb 


С. vidua 


Type la 


С. genuanus 


Type 3 


С. virgatus 


Type 1 


С. geographus 


Type lb 


С. VI tt at и s 


Type 1 


С. gladiator 


Type 1 


С. ximenes 


Unique 


С. gloriamaris 


Type lb 


С. zonatus 


Type 3 


С. gubernator 


Type 2 






С. imperialis 


Type 3 






С. litteratus 


Type 1 


Total Examined = 89 




С. lividus 


Unique 






С. loroisi 


Type 1 


By Type: 




С. lucidus 


Type 4 


Type 1 = 34 




С. magus 


Type 2 


Type la = 11 




С. mahogoni 


Unique 


Type 1b = 15 




С. marmoreus 


Type lb 


Type 2 = 7 




С. mercator 


9 


Type 3 = 7 




С. miles 


Type 1 


Type 4 = 5 




С. miliaris 


Type 1 


Type 5 = 1 




С. monile 


Type 1a 


? or unique = 9 




С. natalensis 


Type lb 






С. nicobaricus 


Type la 







ONTOGENY OF CONUS RADULA 



53 



APPENDIX 2 
Conus Radulae Studied from Drawings and Photos in the Literature 



С acuminatus 
С anemone 
С betulinus 
С daucus 
C. ermineus 
С generalis 
С. inschptus 
С. jaspideus 
С. juliae 
С. leopardus 
С. mercator 
С. mindanus 
С. monachus 
С. mucronatus 
С. nussatella 
С. pulicarius 
С. regius 
С. spuhus 
С. striatus 
С. taeniatus 
С. tessulatus 



Type 1b 
Type la 
Type 1 
Type 1 
Type 2 
Type 1a 
Type 4 
Type 4 
Type 1 
Type 1 
Type 1 
Type 4 
Type 2 
Type 2 
Type 1 
Type 1 
Type 3 
Type la 
Type 2 
Type 1 
Type 1 



(Piele, 1939) 

(Bergh, 1895, as С maculosus) 

(Piele, 1939) 

(Warmke, 1960) 

(Warmke, 1960, as С ranunculus) 

(Piele, 1939, as C. maldivus) 

(Piele, 1939) 

(Warmke, 1960) 

(Warmke, 1960) 

(Bergh, 1895, as С millipunctatus) 

(Bergh, 1895) 

(Piele, 1939, as С agassizif) 

(Piele, 1939) 

(Bergh, 1895) 

(Piele, 1939) 

(Bergh, 1895) 

(Warmke, 1960) 

(Warmke, 1960) 

(Bergh, 1895) 

(Bergh, 1895) 

(Piele, 1939) 



Total = 21 



By Type 
1 = 9 
1a = 3 
1b = 1 
2=4 
3 = 1 
4=3 
5=0 



Total Conus spp. examined = 110 



MALACOLOGIA, 1990,32(1): 55-67 

TEMPO AND MODE OF EVOLUTION IN CONIDAE 

Alan J. Kohn 
Department of Zoology, University of Washington, Seattle, Washington 98195, U.S.A. 

ABSTRACT 

I examined the paleontological literature of the Conidae, here limited to the genus Conus, in 
order to detect temporal patterns in its evolutionary history. All 20 Mesozoic species originally 
described as Conus are likely opisthobranchs, or if Conus they are from strata now known to be 
of Eocene age. The few reports of Paleocene species are probably also incorrect. The earliest 
bona fide fossils of Conus appear to be from the Lower Eocene (50-55 mybp) of England and 
France, where the contemporaneous land flora indicates tropical climatic conditions. Collation of 
the paleontological literature places the first radiation of Conus in the Middle and Upper Eocene. 
Diversity decreased in the Oligocène, one or more major radiations occurred in the Miocene, and 
diversity decreased again in the Pliocene, followed by very rapid increase to the present ap- 
proximately 500 species. The observed pattern is compared with four alternative models of 
taxonomic diversification: exponential, logistic, and exponential interrupted by periods of stasis 
or by periods of reduced diversity. The data fit the last model most closely, as do gastropods and 
other fossilizable marine invertebrates in general from the same era. Important evolutionary 
trends in Conus include (1) increasing shell size, thickness, and ratio of diameter to length, and 
decreasing spire height, and (2) inhabiting shallower, higher-energy marine environments. Shell 
form may have been the most important innovation leading to the major radiations. 

Key words: fossil history, evolutionary origins, adaptive radiation, Conidae. 



"One finds very few papers that give us an 
objective account of the evolution and adaptive 
radiation of any group of Mollusca." 

G. M. Davis (1981) 



INTRODUCTION 

During the past 35 years, comparative bio- 
logical studies of the Conidae have elucidated 
many aspects of the habitats and habits, 
feeding and reproductive biology, and diver- 
sity and community ecology of assemblages 
of co-occurring species (Table 1). This geo- 
logically youthful yet unusually species-rich 
family of gastropods is now probably as well 
known biologically as any tropical marine in- 
vertebrate taxon. Yet despite this wealth of 
neontological information, and the existence 
of a quite respectable fossil record, the origin 
and evolutionary history of the family, and the 
historical and ecological factors that have 
been important in its remarkable evolutionary 
radiation, remain virtually unknown. 

CLASSIFICATION OF THE 
FAMILY CONIDAE 

Conidae as a family-level taxon was first 
validly proposed by John Fleming (1 822), and 



has been generally accepted since about 
1850, but its scope has been variably per- 
ceived. Twentieth-century workers generally 
consider the neogastropod Superfamily Co- 
noidea (or Conacea or Toxoglossa) to com- 
prise two families, Conidae and Terebridae, 
(Thiele, 1 931 ; Seed, 1 983), or three, with sep- 
aration of Turridae from Conidae (Wenz, 
1942; Powell, 1966; Ponder & Waren, 1988). 
As this paper focuses on the evolution of 
Conidae in the narrower sense, I employ the 
latter classification here. 

Characterizing the distinction between Tur- 
ridae and Conidae is not straightforward. 
Several genera intermediate in shell form, 
most with extant representatives, appear to 
link the two families, and different authors 
have drawn different lines between them. 
Cossmann (1896) and Powell (1966) de- 
scribed each of these genera and noted their 
similarities and distinguishing features. Most 
20th century authors include in the Conidae 
only Conus, and Hemiconus if they consider 
this extinct genus or subgenus (Table 2). 



55 



56 KOHN 

TABLE 1. Summary of comparative biology of Indo-Pacific Conus 



Aspect 



Geographic distribution 
Species diversity 

Habitat occupation* 

Habits and activity 
Feeding biology* 

Reproductive biology* 



Patterns 



Species ranges: narrowly endemic to entire Indo-Pacific region. 

Range extent: correlated with dispersal ability of planktonic larvae. 

Lowest (1 species) in geographically peripheral regions. 

Low (5-9 species) in topographically simple, intertidal habitats. 
Highest (12-27 species) in complex subtidal coral reefs. 

Fine to coarse soft sediments; algal turfs; rubble to reef limestone. 
Some differential specialization by co-occurring species. 

Infaunal to epifaunal. Nocturnally active; diurnally sheltered. 

Predators on worms (mainly polychaetes), gastropods, or fishes. 
Markedly differential specialization by co-occurring species. 

Embryonic period 7-24 days. 

Precompetent planktonic veliger larval period 0-30 days. 
Strong gradients in developmental patterns related to egg size. 



'Species of Conus vary widely from specialists to generalists m these aspects. 



Cossmann's (1896) criteria that distinguish 
Coninae and Cryptoconinae are clear and 
applicable to both fossil and Recent forms. 
Partial resorption of inner walls, a hallmark of 
Conus (Kohn et al., 1979), also occurs in 
Conorbis and Hemiconus (Coninae) but not 
In Cryptoconus (Cryptoconinae), The spire 
and aperture in Cryptoconus each comprise 
about half the total shell length, whereas the 
spire of Conorbis is always shorter than the 
aperture length. In Coninae, shell form is gen- 
erally conic or biconic with the sides of the 
aperture parallel. Hemiconus and Conorbis 
thus share important shell features with Co- 
nus. Members of the Cryptoconinae do not 
resorb inner shell walls and have fusiform 
shells with ovate apertures. Here I follow 
Cossmann's (1896) distinction between Cryp- 
toconinae and Coninae, but in agreement 
with other 20th century workers I assign the 
Cryptoconinae to Turridae. I restrict the 
Conidae to the genera compnsing the sub- 
family Coninae. 

Conorbis (Eocene-Miocene and the Recent 
C. coromandelicus Smith) and Hemiconus 
(Middle Eocene-Lower Miocene) each in- 
clude about 20 species (Gilbert, 1 960; Powell, 
1966), Thus neither of these genera has un- 
dergone a striking radiation. Traditionally, 
they are considered the most closely related 
to Conus (e.g. Thiele, 1931; Powell, 1966), 
but we remain ignorant of the evolutionary 
history and phylogenetic relationships of 
these groups. I consider only Conus in the 
rest of this discussion. 



ORIGIN OF CONUS 

The oldest fossils described as Conus are 
from the Lias of Normandy near Caen; they 
are of Pliensbachian age (1 88-1 96 mybp; ab- 
solute dates in this paper are taken from Har- 
land et al., 1982, and Haq & Van Eysinga, 
1987). Charles Lyell read an informal account 
of the discovery of these specimens in a pub- 
lication of the Linnaean Society of Normandy 
([Eudes-Deslongchamps], 1837). He collect- 
ed at the site in 1 840, and later that year (Lyell, 
1840), in collaboration with G. B. Sowerby, he 
described Conus cadonensis and С conca- 
vus. Deshayes & Milne Edwards (1845:7) 
questioned the assignment of these species to 
Conus, and soon d'Orbigny (1850, 1852) con- 
firmed their suspicion by demonstrating in sec- 
tioned specimens that the last whorls are thin 
and that internal shell walls are not resorbed. 
In Conus, the last whorl is thick and inner walls 
are often reduced to 50 |xm (Kohn et al., 1979), 
in Eocene as well as modern species (Kohn, 
1982). Conus abbreviatus Eudes-Deslong- 
champs, 1849, and С caumontll Eudes-Des- 
longchamps, 1849, were described later from 
the same formation. Conus minimus D'Ar- 
chiac, 1843, was described from the Middle 
Jurassic (Bajocian, 172-177 mybp) of Aisne, 
but it was assigned questionably to Conus. It 
is clearly an opistobranch, on the same 
grounds. D'Orbigny (1850, 1852) assigned all 
of these taxa to the opisthobranch genus Ac- 
taeonlna, and they have generally been con- 
sidered members of the family Actaeonidae 



EVOLUTION OF CONIDAE 
TABLE 2. Boundaries between Turridae and Conidae of selected authors. 



57 



Genera: 



Genota Cryptoconus Conorbis Hemiconus Conus 



Geologic Eocene- Eocene- Eocene- Eocene- Eocene- 

Range; Recent Miocene Recent Pliocene Recent 



CONIDAE: CRYPTOCONINAE 



CONIDAE: CYTHARINAE 



J J CONIDAE: CONINAE j ^^^^^^^ ^^ дЭб) 
] [ CQNIDAE j Thiele (1931) 



CONINAE 



TURRIDAE: CRYPTOCONINAE 
TURRIDAE: CONORBIINAE 



][- 



CONIDAE 



Wenz (1942), 
Gilbert (1960) 

Powell (1966) 



TURRIDAE: CONORBIINAE 



] [ 



CONIDAE 



Ponder & Waren (1988) 



ever since (Meek, 1863; Cossmann, 1895; 
Zilch, 1959) (Table 3). 

The remaining Mesozoic fossils originally 
described as Conus are all reported as Cre- 
taceous. The next oldest is С verneullll Vil- 
anova, from the Neocomian of Spain (123- 
140 mybp). This also appears to be an 
opisthobranch (Tomlin, 1937). One species, 
С primitivus Collignon, was described from 
the Albian (98-109 mybp) of Madagascar. 
Definitely Albian (N. Sohl, in litt.), the single 
specimen is a partial internal mold lacking 
any shell material. It is too fragmentary to as- 
sign to any genus with confidence, but it is 
possibly an opisthobranch. The one Creta- 
ceous species described from Italy, С schi- 
osensis Böhm, 1895 (Cenomanian-Turonian; 
89-98 mybp), is also an opisthobranch (Sohl 
& Kollmann, 1985). 

Three species of Conus were described 
from the Cretaceous of France, one Turonian 
(88-91 mybp) and two Santonian (83-86 
mybp). Conus marticensis Matheron, 1843, 
was described from the Turonian at Mar- 
tigues. The original description and figures in 
Matheron (1843) do not permit its rejection 
from Conus, and its source formation, "Craie 
ligno-marneuse," is definitely Cretaceous. 
One of the Santonian species, С tubercula- 
tus Dujardin, 1837, from the Touraine, was 
the first Cretaceous Conus to be described. 
The great French protozoologist Dujardin de- 
scribed the species mainly from molds, but in 
the original figured specimen a partial cast 
replaced some of the original shell. This spec- 



imen is now assigned to the genus Gosavia, 
family Volutidae (Cossmann, 1896; Wenz, 
1943). The second Santonian species, С se- 
nesse/ Delpey, 1938, from Corbières, is prob- 
ably neither a conid nor a turrid. 

Three additional species described from 
the Upper Cretaceous, Conus cylindraceus 
Geinitz, 1850, from Silesia, С semicostatus 
Goldfuss, 1843, from Westphalia, and С la- 
tus Eichwald, 1869, from the Crimea, are also 
opisthobranchs. Geinitz (1850) stated that his 
generic assignment of his 4 mm-long fossil 
was doubtful. 

Of the six remaining species originally de- 
scribed as Cretaceous Conus, С. canalis 
Conrad, 1858, from Mississippi is now placed 
in the Volutidae, and the remaining five (three 
from California and two from Brazil) are now 
known to be from Eocene and Miocene strata 
(Table 3). 

Finally, one Cretaceous species was origi- 
nally assigned to Conorbis. Powell (1966) re- 
tained this species, С mcnairyensis Wade, 
1917, in Conorbis and thus extended the 
range of that genus from Cretaceous to Re- 
cent. However, the shell aperture of С mc- 
nairyensis is not straight and its sides are not 
parallel, and there is no evidence either of the 
exhalent sinus or the arcuate outer lip char- 
acteristic of the Turridae, including Conorbi- 
nae. Sohl (1964) retained С mcnairyensis in 
the Turridae, questionably assigning it to 
Cryptoconus (Table 3). 

Thus all 20 species of Conus and the one of 
Conorbis originally described from Mesozoic 



58 



KOHN 



1Л — ^ in — ^ LO 

CO m CO u^ CO 



С ^ CO ГГ 

>, >. г § 







Ol 


_ 






00 


С) 


E 






■^ 






¡Tt 


о 


I: 


с 










:^ 








in j- 


я 




ев 


?S ■=- 


fc 


с 






(Л ^^ ел 



(Л ^-, (Л 



л< го< га< та< ™<ф<а) 



< Ф 



с m с CD 
о о о О 



I Р X р I 



(^ и^ щ UJ 



G у. ч 



Ol со 
is о. 
■ О 



(1) (Л (1) ел о 

а Е а Е & 

"-> п '^ п ■-> 



О с 
Í 1 

(Л Ф 



S ^ о -О "С 

í^ I ^^ 5 § 
о -S о 2 - 



Е s 






фффффффф 
пзоязогоогоо 
0£0£0505 



го с -о ir Ä Ol 



о ^- о •.- 2 



об g об 5 (Л Ь 1Л 

= о = о Ф ? Ф 

ф (Л Ф (Л "5 о 'S 

_] _1 ш ш 



£ со (5_ _□. 



EVOLUTION OF CONIDAE 



59 



с 


О 


о 


03 X 


f^ 


с 1- 


5 


-Ç со 


о 


M 



О) ш 



^f5 






ел 



со О) 

§ со 



§ g If 



о 

о 


ф 

(Л 

ел 


со 


^ 

S 


(0 


™' 


с 
со 


>, 


ф 


ел 

CJ) 


о 


со 


го 


Е 


я 


с 


с 




> 


о. 


(1) 


Z 


(5 


Iñ 


со' 
■п 


_о 


m 


О 



LL о г ¿Í о 
^ CD О ^ ф 






CÛ я Ш 



СО 2 .!5 2 
ОТ о Ь_ со û_ со 



60 



KOHN 



strata are probably either not Mesozoic or not 
Conidae or both. I conclude that the family 
Conidae originated after the Cretaceous-Ter- 
tiary boundary. 

The few reports of Paleocene Conus are 
equally suspect. Only С rouauiti D'Archiac, 
1850, from the "Groupe Nummulitique" of 
southern France, occurs in beds that are 
mainly Eocene but may possibly include 
Thanetian material (Danizot, 1957). Conus 
rouauiti and the quite similar С concinnus J. 
deC. Sowerby, 1821, from the Lower Eocene 
(Ypresian) of France and England respec- 
tively, seem to represent the earliest bona 
fide records of Conus, s. s. In addition. Lower 
Eocene fossils of two other species have 
been reported from Sind, Pakistan (Coss- 
mann & Pissarro, 1909). 



SMALL BEGINNINGS? 

New groups of animals tend to originate 
from small ancestors, and a common trend in 
the evolutionary history of a supra-specific 
taxon is increase in body size. This is Cope's 
well known "law, " proposed for vertebrates 
but shown to hold quite generally as well for 
invertebrates with fossilizable hard structures 
(Newell, 1949). Does a pattern of increasing 
shell size characterize the evolutionary radia- 
tions within Conus? If so, does the overall dis- 
tribution of shell size in the genus shift up- 
ward, or, as Gould (1988) proposed as more 
likely, is apparent size increase due mainly to 
increasing variance in shell size? If С concin- 
nus and С rouauiti are accepted as early, if 
not the earliest, members of the genus, their 
size can be compared with later species, par- 
ticularly those at and just before the initiation 
of major radiations. Detailed analyses remain 
to be carried out, but preliminary data suggest 
trends. The type specimen of С rouauiti is 1 1 
mm, and that of С concinnus is 14 mm, in 
shell length. The largest specimen of the lat- 
ter species in The Natural History Museum, 
London, is 32 mm, and the mean length of the 
largest specimen in five Lower Eocene 
(London Clay) lots is 26 mm. 

Although many Conus species of about the 
same size occur in Middle Eocene deposits, 
much larger species are also prominent then, 
the time of the first known radiation of the 
genus. For example, a syntype and several 
other specimens of С edwardsi Cossvnann in 
The Natural History Museum, London, from 
the Middle Eocene Bracklesham Beds in 



southern England are about 70 mm long. This 
species also has a relatively shorter spire 
than C. concinnus, averaging about 16% of 
total shell length vs. 30°o in the latter species. 

While further analyses of the size fre- 
quency distributions of Paleogene and Neo- 
gene Conus remain to be carried out, the 
sizes of extant species, representing the most 
diverse radiation in the group's history, are 
reasonably well known. The size of modern 
Conus species varies markedly in different 
habitat types. In the Indo-Pacific region where 
most species occur, median shell lengths are 
23 mm on intertidal benches, 35 mm on sub- 
tidal coral reefs, and 80 mm in subtidal sandy 
bays (Kohn, 1980, 1981). Species similar in 
size to the hypothesized ancestral species 
thus persist commonly today. Coral reef plat- 
forms support the highest modern diversity; 
these species are of somewhat larger body 
size. Shallow reef-associated lagoons rank 
next in Conus diversity; they support assem- 
blages mainly of even larger species (Kohn, 
1981), including the largest extant species, 
with shell length of more than 200 mm. A few 
Indo-Pacific Conus species with maximum 
shell size less than the Lower Eocene species 
also occur today, and more exist in other 
parts of the geographic range of the genus. 

Thus size increase has characterized the 
evolution of Conus, in the sense of increasing 
upper size limits and increasing size variance. 
As in other groups of organisms to which 
Cope's law applies (Bonner, 1988), small 
species not only persist but may be quite di- 
verse. 



PATTERNS OF EVOLUTION IN 
TIME AND SPACE 

Alternative Hypotheses of Diversity Patterns 

As a simplified model of diversification 
rates, Stanley (1979) plotted the logarithm of 
the number of extant species in a taxon 
against the time since its origin. The slope of 
a line drawn between this point and the origin 
(assuming a single species initiated each 
taxon) can be interpreted as the exponential 
rate of species proliferation. This method un- 
derestimates, as Stanley (1979) noted, be- 
cause it omits extinct species. Moreover, di- 
versification rates are unlikely to remain 
constant over long periods, for reasons in- 
volving earth history, ecological factors, and 
the inherent attributes of evolutionary lin- 



EVOLUTION OF CONIDAE 



61 



a. 



О) 

E 




Time Since 
Origin of Тахоп 

FIG. 1. Alternative patterns of species proliferation 
of a diversifying taxon in time; semilogarithmic 
plots. A, Exponential. dDícñ = r^D^. Speciation rate 
per species (Г3) and species extinction rate (r^) are 
constant; r^ = Гд-г^. В, Logistic. dD eft = r^D 
(1-D/D). Го = initial diversification rate. D = equi- 
librium value. C, Periods of exponential diversifica- 
tion alternate with periods of stasis, r^ varies, as r^ 
> Ге. D, Periods of diversification (Гд > Ге) alternate 
withi periods of net extinction (r^ < r^). 

eages. Figure 1 shows several possible alter- 
native patterns. The simplest case is the ex- 
ponential or log-linear model (Stanley, 1975, 
1 979) (Fig. 1 A), for speciation by each daugh- 
ter species at a constant rate and a constant 
species survival rate. A constant exponential 
rate of increase damped by the imposition of 
a saturation value results in a logistic curve 
(Sepkoski, 1 978; Walker, 1 985) (Fig. 1 B). The 
remaining models in Figure 1 incorporate fac- 
tors that cause periods of diversification to al- 
ternate with periods of stasis or slow net 
change (Fig. 1С), or with periods of reduced 
diversity due to extinction rate exceeding spe- 
ciation rate (Fig. ID). 

Temporal Characteristics of the Conus 
Fossil Record 

In order to determine the history of taxo- 
nomic diversification in Conus, I developed a 
database of all records I could locate in the 
paleontological literature (2,500 from 1792 to 
the present) that indicated stratigraphie age 
and geographic location of fossil Conus spe- 
cies. For Middle Eocene-Pleistocene records, 
I generally accepted at face value the species 
identification and stratigraphy of original au- 
thors; no effort has yet been made toward 



critical evaluation of the data, and all of the 
biases that charactehze paleontological data 
in general apply (see e.g. Raup, 1976a). Fig- 
ure 2 shows the result in the form of a spindle 
diagram of the number of Conus species 
through the Cenozoic, including Lyellian per- 
centages for each epoch. At times when the 
number of species is increasing from stage to 
stage, turnover becomes an important aspect 
of diversification. Is increasing richness due 
to a modest number of originations combined 
with persistence of most species from the pre- 
vious stage, or to modest persistence and the 
origination of many new species? 

Figure 3 presents patterns of Conus spe- 
cies turnover during the Cenozoic; these data 
are also accepted uncritically from the pale- 
ontological literature. First and last appear- 
ances are expressed in absolute numbers 
(plotted on a logarithmic scale; Fig. ЗА), rela- 
tive to the total number of species present 
during the interval (Fig. 3B), and as apparent 
rates of speciation and extinction (Fig. 3C). 
Figure 3D shows turnover calculated as the 
number of originations plus extinctions per 
species present and the rate of diversifica- 
tion per species per million years, because 
the intervals used vary considerably in abso- 
lute time. The notations used (after Sepkoski, 
1978) are; 

S = number of first appearances in interval 
(apparent speciations); 

E = number of last appearances in interval 
(apparent extinctions); 

D = number of species present in interval 
(apparent diversity); 

At = duration of interval in millions of years; 

Turnover = (S + E)/D; 

Diversification rate = r^ = Гд - r^, 
where 

Rate of speciation Г3 = S/(DAt), and 

Rate of extinction r^ = E/(DAt). 

The main aspects of the patterns that 
emerge from analyses of the data as origi- 
nally reported are; 

(1) The genus Conus originated during 
Lower Eocene time. Mesozoic and Paleoceno 
records are rejected or dubious. 

(2) The first real radiation of the genus oc- 
curred in the Middle Eocene (Fig. 2). Many 
species persisted into the Upper Eocene, but 
75% of all Upper Eocene species are first re- 
ported then (Fig. 3B). In all, about 100 spe- 
cies are recorded from this epoch. Species 
turnover was maximal in Middle Eocene be- 
cause of the large numbers of both origina- 



62 



KOHN 



No. Of % 
Species Extant 



mybp 




FIG. 2. Kite diagram showing the number of species of Conus throughout Tertiary and Quaternary time, 
based on an uncritical analysis of the paleontological literature. Ages of epochs and subdivisions from 
Harland et al. (1 982) ; the two columns at left give the total number of species reported from each epoch and 
the fraction extant. Numbers of species are: Lower Eocene: 5: Middle Eocene: 42; Upper Eocene: 42: Lower 
Oligocène: 28: Upper Oligocène: 19; Lower Miocene: 1 27; Middle Miocene: 11 1 : Upper Miocene: 1 58; Lower 
Pliocene: 43; Upper Pliocene: 53; Pleistocene: 124. 



tions and disappearances (Fig. ЗА), but be- 
cause the interval was long, the rate of 
diversification was low (Fig. 3D). "Per spe- 
cies ' rates of origination and extinction were 
higher in the Upper Eocene (Fig. 3C). 

(3) Species diversity decreased in the Oli- 
gocène, a pattern common to the Gastropoda 
and marine invertebrates in general (Raup, 
1976b). Numbers and rates of originations 
and extinctions declined (Figs. 3A,C), as did 
species turnover (Fig. 3D): about 70% of spe- 
cies present originated in the Lower and Up- 
per Oligocène while extinction rates were 
52% and 32%, respectively (Fig. 3B). 

(4) One or more major radiations occurred 
in the Miocene. Nearly 300 species are re- 
corded from this epoch. Originations of new 
species increased to 82% of all species 



present in the Lower Miocene, when turnover 
was second only to the Middle Eocene (Figs. 
3B,D). As in that radiation, the Lower Miocene 
diversification rate was low because the inter- 
val was long (Figs. 3C,D). The rates were 
higher although absolute and relative num- 
bers of originations and extinctions declined 
in the shorter Middle Miocene (Figs. 3A,B,C). 
Originations then declined to about 50% of 
species present by Upper Miocene, extinc- 
tions increased, but rates of both declined 
(Figs. 3A,B,C). 

(5) Species diversity again declined in the 
Pliocene (152 species recorded). Generally 
reduced diversity of gastropods and other in- 
vertebrates is characteristic of this epoch 
(Raup, 1976b). Both the numbers (Fig. ЗА) 
and proportions (Fig. 3B) of originations and 



EVOLUTION OF CONIDAE 



63 




1000 



100- 



Eocene Oligocène Miocene A A 
Pliocene^ y 
Pleistocene -^ 

FIG. 3. A. The numbers of originations (•) and ex- 
tinctions { ) ot Conus species, determined from re- 
ports of first and last appearances in the fossil 
record, throughout the Cenozoic. B. Species turn- 
over, calculated as the sum of the numbers of orig- 
inations (S) and extinctions (E) (from Fig. ЗА) di- 
vided by the number of species reported during 
each interval (D) (see text). С The apparent rates 
of speciation (origination) (Гд = S/Dt; •) and extinc- 
tion (rg = E/Dt;D) during each interval. D. Species 
turnover ([S + E]/D; ) and species diversification 
rate (Гн = r,- r„;»). 



extinctions of Conus species declined, but 
their rates increased (Fig. 3C) in this tempo- 
rally brief epoch. 

(6) Very rapid speciation (Figs. 3A,B) and a 
large disparity between origination and ex- 




Eocene Oligocène 



Miocene AAA 
Pliocene'' / / 
Pleistocene ^/ 
Recent -^ 



FIG. 4. The numbers of species of Conus reported 
from Tertiary and Quaternary epochs and stages. 
Data from Fig. 2, plotted to conform with the models 
in Fig. 1. 

tinction rates (Fig. 3C) occurred during the 
even briefer Pleistocene, leading to modern 
high diversity. Nearly 65% of Pleistocene spe- 
cies are not known from earlier in the fossil 
record. 

(7) Only 11% and 33% respectively of Mi- 
ocene and Pliocene species survive, but 77% 
of Pleistocene species are extant. 

In his graph of nine extant radiating marine 
prosobranch clades according to model A 
(Fig. 1), Stanley (1979) gave a mean net ex- 
ponential rate r = 0.067 my \ At this rate, 
the number of species in a clade doubles in 
10.3 my. For Conidae, he indicated 500 ex- 
tant species and an age of 70 my, or r = 
0.103 my \ equivalent to a doubling time of 
6.7 my. Taking the age of Conidae as 55 my 
based on the evidence presented here in- 
creases r to 0.113 my ^ and decreases the 
doubling time to 6.1 my. As noted above, this 
model inevitably underestimates the rate of 
diversification. 

If the data in Figure 2 are replotted accord- 
ing to Figure 1, the resulting pattern ap- 
proaches model D most closely (Fig. 4); the 
fossil record of Conus indicates alternating 
periods of rapid radiation and of reduced di- 
versity. As this closely parallels the temporal 
diversity patterns of other marine invertebrate 
groups during Cenozoic time (Raup, 1976b), 
extrinsic environmental factors were likely im- 
portant causes. 

Geographic Features of the Conus 
Fossil Record 

Preliminary paleobiogeographic analysis of 
the Conus fossil record suggests that: 



64 



KOHN 



(1 ) The Lower Eocene origin of the genus is 
coastal European; the earliest verified 
records are from England and France. 

(2) The first real radiation, in the Middle 
Eocene, likely occurred in the same geo- 
graphic region: about 3 4 of Middle Eocene 
species are from Britain and Europe, but the 
genus also expanded its range broadly. Mid- 
dle Eocene species are also recorded from 
Egypt, Nigena, Pakistan, California, and the 
U.S. Gulf Coast. 

(3) After the Middle Eocene, extinctions of 
Conus species outpaced originations in the 
European seas, but modest, localized diver- 
sity increases occurred on the Indo-Australian 
plate (Upper Eocene), and in the Asian region 
of the Eurasian plate (Upper Oligocène). 
Whether or not different Conus species as- 
semblages occurred in each geographic re- 
gion remains to be addressed. Piccoli (1984) 
concluded that Paleogene molluscan assem- 
blages of the Mediterranean region were gen- 
erally Indo-Pacific in composition; Rosen 
(1988) emphasized regional differences in 
contemporaneous corals and echinoids. 

(4) Europe and the Indo-Australian regions 
were the sites of major Miocene radiations, 
the latter predominantly early in the epoch 
and the former continuing throughout Mi- 
ocene time. 

(5) Throughout its history, the genus Conus 
appears to have been confined to warm seas, 
with all of the major radiations occurring in 
tropical conditions. In Britain, "in Eocene 
times the climate, as reflected in the land 
flora, was that of tropical lowlands such as 
those of south-east Asia today" (Melville & 
Freshney, 1982). The same likely applies to 
the Miocene seas occupied by the genus, and 
its modern geographic distribution remains 
predominantly tropical. 

(6) The data fail to reveal the geographic 
sources of the group's most important radia- 
tion, resulting in several hundred extant spe- 
cies. Most species known as Pleistocene fos- 
sils are distributed similarly to their modern 
counterparts, with the majority in the Indo- 
Australian plate and western Pacific regions, 
and fewer but substantial numbers in the 
Americas. This remains a critical area for fu- 
ture investigation. 



to gather relevant data to test the hypotheses 
advanced. 

New Adaptive Zones? 

An adaptive zone, as the concept was in- 
troduced by Simpson (1953) and clarified by 
Van Valen (1971), is the "niche" of a taxon 
above the species level. Its two basic, more or 
less independent components comprise the 
resources used by the members of the focal 
taxon, and their resistance to prédation and 
parasitism. Did the evolutionary radiations of 
Conus depend on successful invasion of a 
different adaptive zone and different ways of 
acquiring resources and defending against 
enemies, from those of the ancestral and sis- 
ter taxa? 

The answer is by no means clear. During 
the early evolutionary history of Conus, the 
fossils are typically associated with fine sed- 
iments characteristic of continental shelf and 
greater depths, similar in general to the hab- 
itats of many species in the hypothetically an- 
cestral family Turridae. (Such habitats are 
also particularly favorable sites for fossiliza- 
tion.) Successful invasions of shallower bay 
and lagoon environments probably began in 
Middle Eocene time. The conical shape of the 
last whorl with the apex of the cone anterior 
would certainly facilitate locomotion by the 
gastropod through soft substrata. Evolution- 
ary trends toward (1) increased shell size, 
thickness and ratio of diameter to length, and 
decreased spire height, and (2) occupation of 
ever shallower and high energy marine envi- 
ronments probably occurred during all Middle 
Eocene and Miocene radiations of Conus but 
at present remain largely undocumented. 
These changes in shell form could well have 
expanded the taxonomic and size ranges of 
suitable prey organisms without sacrificing 
defensive shell strength. Simultaneously, the 
habitat shifts likely involved use of hard as 
well as soft but coarser substrata associated 
with coral reefs, a biogenic environment in- 
creasing in complexity and geographic extent 
contemporaneously with the radiations of Co- 
nus (Rosen, 1988). 

Key Innovations? 



NEW ADAPTIVE ZONES? 
KEY INNOVATIONS? 

Of necessity I address these topics with a 
high degree of speculation, and I urge others 



The rapid evolutionary radiation of a taxon 
is often assumed and sometimes docu- 
mented (e.g, Liem, 1973) to be associated 
with the origin of key evolutionary novelties, 
I.e. the development of new, usually morpho- 



EVOLUTION OF CONIDAE 



65 



logical, attributes that satisfy several criteria 
(as modified frorn Herrera, 1989): 

(1) The novel feature is significant to the 
taxon and absent from its sister or ancestral 
groups; 

(2) Taxa with the feature diversify early in 
their evolutionary history; 

(3) Taxa with the feature become structur- 
ally and taxonomically more diverse than sis- 
ter taxa lacking it. 

Coddington (1988) uses a cladistic frame- 
work to test whether innovations are adapta- 
tions in the strict sense (of Gould & Vrba, 
1982) of selection on a specific function pro- 
moting the origin, spread and maintenance of 
the innovative attribute, and dhving taxo- 
nomic diversification. Lauder & Liem (1989) 
provide additional criteria and an explicit pro- 
cedure for testing the key innovation hypoth- 
esis. It involves mapping the hypothesized 
key innovation onto a phylogeny of the taxon 
and comparing morphometric analyses of this 
taxon and of outgroups. 

What innovative features of Conus might 
qualify? Hallmarks of the genus include: 

(1) the detachable, hollow, barbed, har- 
poon-like radular tooth individually injected 
via an extensile intraembolic proboscis during 
prey capture (Kohn et al., 1972); 

(2) the peptide venoms injected through the 
tooth that rapidly immobilize the prey (Olivera 
et al., 1985); 

(3) the characteristically broadly conical or 
biconical shell with parallel-sided aperture, 
typically with the last whorl covering most of 
the prior whorl so that the spire is quite low; 
and 

(4) the thick, heavy and strong crossed- 
lamellar last whorl of the shell, with the pro- 
tected inner walls later mainly dissolved away 
during extensive interior renovation (Kohn et 
al., 1979). 

At present, data are lacking in Conus to test 
these features against even those predictions 
of the key innovation hypothesis that do not 
require phylogenetic evidence. Moreover, the 
present lack of detailed comparative anatom- 
ical information on Conus means that other 
important innovative characters may remain 
to be discovered. At best we can indicate the 
present status of knowledge: 

(1) The general features of the Conus rad- 
ular tooth mentioned above are shared by nu- 
merous turrids, especially the subfamily Bor- 
soniinae; whether this is a sister taxon of 
Conus, its ancestral taxon, or neither is un- 
known. The functional morphology of the pro- 



boscis of Conus and turrids is also very sim- 
ilar (Greene & Kohn, 1989; Kantor, 1990). 

(2) A venom apparatus morphologically 
similar to that of Conus also occurs in many 
taxa of Turridae, but nothing is known of the 
chemical nature of the venom in the latter 
family. 

(3) Shell size, shape and thickness may 
meet all of the criteria. 

(4) Internal wall resorption occurs, but to a 
lesser extent than in Conus, in some Olividae 
as well as in Hemiconus and Conorbis. 

Key innovations may be quite subtle. As 
Mayr (1 960) said, "Most evolutionary changes 
take place without the origin of new 
structures. . . . Most differences are merely 
shifts in proportions, fusions, losses, second- 
ary duplications, and similar changes," that 
nevertheless can lead to "evolutionary ava- 
lanches." In addition, the causal relationship of 
key innovations with subsequent taxonomic 
diversification may be quite indirect. Anatom- 
ical changes that provide selective advan- 
tages early in the evolution of a taxon may 
fortuitously support organisms of larger size at 
a later time in the evolution of a clade, as 
Bonner ( 1 988) points out, as well as promoting 
speciation and the clade's radiation in a new 
adaptive zone. 

Shell form is the most likely candidate for 
the critical key evolutionary innovation of Co- 
nus. The depressed spire and broadly conical 
form permits the aperture to expand, particu- 
larly anteriorly and posteriorly. This in turn 
may have permitted thickening of the last 
whorl without reducing aperture size, and in- 
ternal wall thinning to retain a large living 
space within the shell, thus accommodating 
large prey organisms. The latter is likely es- 
pecially important in a predator that must en- 
gulf and swallow whole prey. A thick, resistant 
shell is an important defense against both 
crushing predators and physical factors in 
shallow, high-energy Cenozoic marine envi- 
ronments. The development of a shell form 
with these features, involving no anatomically 
new structures but mainly changes in propor- 
tions, and in combination with prior posses- 
sion of a well-developed harpoon-like radular 
tooth and venom apparatus, may have been 
the key innovation leading to the major radi- 
ations of Conus. At present this hypothesis is 
speculative, and tests, such as those pro- 
posed by Coddington (1988) and Lauder & 
Liem (1989), must await improved knowledge 
in an area presently completely unknown but 
certainly amenable to study, the phylogeny of 



66 



KOHN 



the Conidae. However, available evidence 
from the Paleogene and especially Neogene 
radiations suggest that strongly but stylishly 
shelled. Conus is a young, upwardly mobile, 
progressional genus, albeit at a snail's pace. 



ACKNOWLEDGMENTS 

This research was supported by NSF Grant 
BSR 8700523. I thank J. D. Taylor. R. J. 
Cleevely, C. P. Nuttall and N. J. Morris for 
discussion and for providing research facili- 
ties at The Natural History Museum. London, 
and N. F. Sohl. G. J. Vermeij and R. M, Lin- 
sley for helpful comments. 



LITERATURE CITED 

BONNER. J. T.. 1988. The evolution of complexity. 
Princeton University Press. Princeton. 260 pp. 

CODDINGTON. J. A.. 1988. Cladistic tests of ad- 
aptational hypotheses. Cladistics. 4: 3-22. 

COSSMANN. M.. 1895-1896. Essais de paléoco- 
nchologie comparée. 1:58-69; 2:140-163. 
Cossmann. Pans. 

COSSMANN. M. & G. PISSARRO. 1909. The Mol- 
lusca of the Ranikot Series. Memoirs of the Geo- 
logical Survey of India. New Series. 3: 1-83. 

DANIZOT. G.. 1957. Nummulitique. Lexique Strati- 
graphique International, 1(4)a: 143-144. 

DAVIS. G. M.. 1981. Introduction to the second In- 
ternational Symposium on Evolution and Adap- 
tive Radiation of Mollusca, f^alacologia. 21: 1-4. 

DESMAYES, G. -P. & MILNE EDWARDS, H., 
1845, Histoire naturelle des animaux sans ver- 
tèbres. 1 1 . Paris. 665 pp. 

[EUDES-DESLONGCHAMPS. E.]. 1837. Séance 
publique de la Société Linnéenne de Normandie, 
tenue a Honf leur, le 28 juin 1 837. Mémoires de la 
Société Linneénne de Normandie. 6: 41-42. 

FISCHER. P.. 1887. Manuel de conchyliologie. F. 
Savy. Pans. 1369 pp. 

FLEMING. J., 1822, The philosophy of zoology, or 
a general view of the structure, functions, and 
classification of animals. Archibald Constable & 
Co.. Edinburgh. 565 pp. 

GEINITZ. H. В.. 1850. Charakteristik der Schichten 
und Petrefacten des sächsisch-böhmischen Krei- 
degebirges. Arnold. Leipzig. 116 pp. 

GLIBERT, M.. 1960. Les Conacea fossiles du cèn- 
ozoïque étranger des collections de llnstitut 
Royal des Sciences Naturelles de Belgique. In- 
stitut Royal des Sciences Naturelles de Belgique 
Mémoires. 2nd ser.. 64: 1-132. 

GOULD, S. J., 1988. Trends as changes in vari- 
ance: a new slant on progress and directionality 



in evolution. Journal of Paleontology. 62: 319- 
329. 

GOULD. S. J. & E. S. VRBA. 1982. Exaptation— a 
missing term in the science of form. Paleobiol- 
ogy. 8: 4-15. 

GREENE, J. L. & A. J. KOHN. 1989, Functional 
morphology of the Conus proboscis (Mollusca: 
Gastropoda). Journal of Zoology. 219: 487-493. 

HAQ. B. U. & W. B. VAN EYSINGA, 1987. Geolog- 
ical time table. 4th ed. Elsevier. Amsterdam. 

HARLAND. W. В.. A. V. COX, P G. LLEWELLYN, 
С A. G. PICKTON. A. G. SMITH & R. WAL- 
TERS. 1982, A geologic time scale. Cambridge 
University Press, Cambridge, 131 pp. 

HERRERA, С. M., 1989. Seed dispersal by ani- 
mals: a role in angiosperm diversification? Amer- 
ican Naturalist. 133: 309-322. 

KANTOR. Y.. 1990. Anatomical basis of ongin and 
evolution of toxoglossan mode of feeding. Mala- 
cologia. 32: 3-18. 

KEEN. A. M. & H. BENTSON. 1944. Check list of 
California Tertiary marine Mollusca. Geological 
Society of America Special Papers. 146: 1-280. 

KOHN, A. J.. 1980. Conus kahiko. a new Pleis- 
tocene gastropod from Oahu, Hawaii. Journal of 
Paleontology. 54: 534-541. 

KOHN. A. J.. 1981. Abundance, diversity, and re- 
source use in an assemblage of Conus species 
in Enewetak Lagoon. Pacific Science, 34: 359- 
369. 

KOHN, A. J.. 1982, Evolution of Indo-Pacific 
Conidae: A preliminary report. In KUSSAKIN, O. 
G. & A. I. KAFANOV. eds., Fauna and hydrogra- 
phy of the shelf zones of the Pacific Ocean. Pro- 
ceedings of the XIV Pacific Science Congress, 
Section Manne Biology.' Issue 4: 84-88. 

KOHN, A. J.. E. R. MYERS & V. R. MEENAKSHI, 
1979, Interior remodeling of the shell by a gas- 
tropod mollusc. Proceedings of the National 
Academy of Sciences of the U.S.A.. 76: 534- 
541. 

KOHN, A. J.. J. W. NYBAKKEN & J. J. VAN MOL, 
1972, Radula tooth structure of the gastropod 
Conus impenalis elucidated by scanning electron 
microscopy. Science. 176: 49-51. 

LAUDER. G. V. & K. F. LIEM. 1989. The role of 
historical factors in the evolution of complex or- 
ganismal functions. In WAKE. D. B. & G. ROTH, 
eds.. Complex organismal functions: integration 
and evolution in vertebrates, 63-78. Dahlem 
Konferenzen. Chichester: John Wiley & Sons. 
Ltd., in press. 

LIEM. K. F., 1973, Evolutionary strategies and mor- 
phological innovations: chichlid pharyngeal jaws. 
Systematic Zoology. 22: 425-441. 

LYELL. С. 1840. On the occurrence of two species 
of shells of the genus Conus in the Lias, or Infe- 
rior Oolite, near Caen in Normandy. The Annals 
and Magazine of Natural History. 6: 292-296. 

MATHERON. P. 1842-1843. Catalogue méth- 
odique et descriptif des corps organisés fossiles 
due Département des Bouches-du-Rhone. Mar- 
seille, Carnaud, 269 pp. 



EVOLUTION OF CONIDAE 



67 



MAYR, е., 1960, The emergence of evolutionary 
novelties. In TAX, S., ed. Evolution after Darwin, 
1 : 349-380. Chicago, University of Chicago 
Press. 

MEEK, F. В., 1863, Remarks on the family Actae- 
onidae, with descriptions of some new genera 
and sub-genera. American Journal of Science. 
2nd ser., 35: 84-94. 

MELVILLE, R. V. & E. С FRESHNEY, 1982, The 
Hampshiire basin and adjoining areas. Britisti Re- 
gional Geology. Institute of Geological Sciences. 
London, HMSO, 146 pp. 

NEWELL, N. D., 1949, Phyletic size increase, an 
important trend illustrated by fossil invertebrates. 
Evolution. 3: 103-124. 

olivera, в. m., w. r. gray, r. zeikus, j. m. 
Mcintosh, j. varga, j. rivier, v. doSan- 

TOS & L. J. CRUZ, 1985, Peptide neurotoxins 
from fish-hunting cone snails. Science, 230: 
1338-1343. 

DORBIGNY, A., 1850. Prodrome de paléontologie 
stratigrapfiique universelle des animaux mol- 
lusques & rayonnes, etc., V. I. V. Masson, Paris. 

D'ORBIGNY, A., 1852, Paléontologie française. Ire 
sér., 2:162-163. D'Orbigny, Pans. 

PICCOLI, G., 1984, Paleoecology and paleobio- 
geography: An example based on Paleogene 
shallow benthic mollusks from Venetian region 
(NE Italy). Geobios. Mémoire Spécial No. 8: 341- 
347. 

PONDER, W. F. & A. WAREN, 1988, Classification 
of the Caenogastropoda and Heterostropha — A 
list of the family-group names and higher taxa. 
Malacological Review. Supplement Л: 288-317. 

POWELL, A. W. В., 1966, The molluscan families 
Speightiidae and Turridae. Bulletin of ttie Auck- 
land Institute and Museum, 2, 1-188. 

RAUP, D. M., 1976a, Species diversity in the Phan- 
erozoic: an interpretation. Paleobiology. 2: 289- 
297. 

RAUP, D. M., 1976b, Species diversity in the Phan- 
erozoic: a tabulation. Paleobiology. 2: 279-288. 

ROSEN, B. R., 1988, Progress, problems and pat- 
terns in the biogeography of reef corals and other 
tropical marine organisms. Helgoländer Meere- 
suntersuchungen, 42: 269-301. 

SEED, R., 1983, Structural organization, adaptive 
radiation, and classification of molluscs. In 



HOCHACHKA, P. W., ed., The Mollusca. 1: 1- 
52. Academic Press, New York. 

SEPKOSKI, J. J. 1978. A kinetic model of Phaner- 
ozoic taxonomic diversity, I. Analysis of marine 
orders. Paleobiology. 4: 223-251. 

SIMPSON, G. G., 1953, The major features of ev- 
olution. New York. Columbia University Press, 
434 pp. 

SOHL, N. F., 1964, Neogastropoda, Opisthobran- 
chia and Basommatophora from the Ripley, Owl 
Creek, and Prairie Bluff Formations. U.S. Geo- 
logical Survey Professional Paper. 331-B: 152- 
344. 

SOHL, N. F. & H. A. KOLLMANN, 1985, Creta- 
ceous actaeonellid gastropods from the western 
hemisphere. U.S. Geological Survey Profes- 
sional Paper. 1304: 1-104. 

STANLEY, S. M. 1975. A theory of evolution above 
the species level. Proceedings of the National 
Academy of Sciences. U.S.A. 72: 646-650. 

STANLEY, S. M., 1979, Macroevolution. W. H. 
Freeman, San Francisco. 332 pp. 

STEWART, R. В., 1927, Gabbs California fossil 
type gastropods. Proceedings of the Academy of 
Natural Sciences of Philadelphia. 78: 287-505. 

THIELE, J.. 1931, Handbuch der systematischen 
Weichtierkunde. 1 . Gustav Fischer, Jena, 778 pp. 

TOMLIN, J. R., leB., 1937, Catalogue of recent and 
fossil cones. Proceedings of the Malacological 
Society of London. 22: 205-330. 

VAN VALEN, L., 1971, Adaptive zones and the or- 
ders of mammals. Evolution, 25: 420-428. 

WALKER, T D., 1985. Diversification functions and 
the rate of taxonomic evolution. In VALENTINE, 
J. W., ed., Phanerozoic diversity patterns, pp. 
311-334. Princeton University Press, Princeton. 

WARING, С A., 1917, Stratigraphie and faunal re- 
lations of the Martinez to the Chico and Tejon of 
Southern California. Proceedings of the Califor- 
nia Academy of Sciences. Series 4, 7: 41-124. 

WENZ, W., 1942, Gastropoda. Allgemeiner Teil 
und Prosobranchia. Handbuch der Paläozool- 
ogie, 6, part 1, 1639 pp. 

ZILCH, A.. 1959-1960, Gastropoda. Teil 2. Euthy- 
neura. Handbuch der Paläzoologie, 6, part 2, 832 
pp. 

Revised Ms. accepted 21 June 1990 



MALACOLOGIA, 1990,32(1): 69-77 

TURRID GENERA AND MODE OF DEVELOPMENT: 
THE USE AND ABUSE OF PROTOCONCH MORPHOLOGY 

Philippe Bouchet 
Museum National d'Histoire Naturelle, 55 Rue Button, 75005 Paris, France 

ABSTRACT 

Two contrasted protoconch morphologies (multispiral vs. paucispiral) are correlated in Tur- 
ridae, as in other Caenogastropoda, with modes of larval development (respectively plank- 
totrophic vs. non-planktotrophic). The multispiral vs. paucispiral dichotomy has been used ex- 
tensively to denote phylogeny and recognize genera, a practice unique to Turridae that resulted 
in Powell's "phenomenon " of "turnd pairs. " Because planktotrophy can be lost repeatedly and 
independently from a given ancestor, Powell's system fails to provide a phylogenetic classifi- 
cation: it leads to artificial polyphyletic genera, each characterized by a single protoconch type, 
but not necessarily deriving from a common ancestor. A group of turrid species possessing 
multispiral and paucispiral protoconchs should not be split into different (sub)genera when their 
teleoconch, radula and anatomical characters otherwise indicate that a single clade is involved. 
However, sculptural types among planktotrophic type protoconchs are considered to have tax- 
onomic utility at supraspecific levels. 

Key words: Turridae, protoconch, larval development, taxonomy, phylogeny. 



INTRODUCTION 

With 679 genus-group taxa and probably 
as many as 10,000 Recent and fossil nominal 
species, the family Turridae ranks as the most 
speciose of all marine gastropods. Tryon 
(1884) was the last author to try to mono- 
graph the Recent species of the family world- 
wide. This task has probably been considered 
unrealistic by later authors, who have mostly 
been working on a regional and/or strati- 
graphic basis. Species-level taxonomy in the 
Turridae offers no more problem than in other 
marine gastropod families, except perhaps 
that large series of specimens are only rarely 
available for an appraisal of variation. By con- 
trast, considerable difficulty in recognizing 
good classificatory characters at supraspe- 
cific levels has been frequently expressed in 
the literature, emphasis being variously laid 
by different authors on conchological, radu- 
lar or anatomical characters. 

Controversy as to the value of the proto- 
conch as a guide to phylogeny has continued 
unabated over the years (Kilburn, 1983). In 
the present paper, I review how the proto- 
conch has been used by paleontologists and 
zoologists in supraspecific turrid taxonomy. I 
demonstrate that mode of larval development 
alone cannot be used to recognize genera or 
subgenera. I conclude that the so-called 
"turrid pairs" of genera are most likely to be 
polyphyletic and should be abandoned in 



taxonomic practice in the family. Finally, I 
discuss the taxonomic utility of styles of orna- 
mentation and conclude that such morpholog- 
ical details are useful in planktotrophic proto- 
conchs. 



REVIEW 
The "Phenomenon" of ' Turhd-pairs" 

As other marine prosobranchs, Turridae 
exhibit two major types of protoconchs: (a) a 
multispiral — Powell (1942) also used the term 
"polygyrate" — protoconch, with a small proto- 
conch I, and a protoconch II consisting of 2-5 
whorls, often with an elaborate sculpture of 
ribs and cords; and (b) a paucispiral proto- 
conch, with no distinction between proto- 
conchs I and II, consisting of 1 -2 whorls with a 
large nucleus, and a simpler, stouter sculp- 
ture, or no sculpture at all. 

Although protoconchs had been previously 
used by malacologists before in gastropod 
taxonomy, it was certainly Powell who first 
formalized a system using protoconchs in 
routine supraspecific taxonomy in the Tur- 
ridae. Through his monumental work on New 
Zealand and worldwide turrids (Powell, 1942, 
1966), Powell has profoundly influenced 
modern taxonomic practice, and these two 
papers are cited in almost every paper on tur- 



69 



70 



BOUCHET 



rid supraspecific taxonomy published in the 
last few decades. 

Powell's (1942) opinion is worth citing in full 
here: "A certain number of genera appear to 
occur in parallel series, being alike in adult 
shell features and evidently of common origin, 
but by their respective protoconchs they are 
separable into polygyrate and paucispiral se- 
ries. In all these instances I have treated 
these parallel developments as distinct gen- 
era, for differences in the embryo are surely of 
basic biological importance." And further: 
"Much criticism has been levelled at the em- 
ployers of protoconch criteria in the family, but 
in all these objections the fault seems to lie in 
the failure of rigid application of these criteria. 
If we refuse to admit more than one style of 
protoconch in a genus these anomalies dis- 
appear." 

On Powell's authority, this opinion later be- 
came an established, almost unchallenged 
dogma entrenched in turrid taxonomic prac- 
tice, and genera have been and are being 
recognized based on this single character 
(e.g. Powell, 1942, 1964; van Aartsen & Fehr- 
de Wal, 1978: Gougerot & Le Renard, 1981; 
van Aartsen et al., 1984; Bernasconi & 
Robba. 1984; van Aartsen, 1988). Examples 
of such "turrid pairs" include: 

Mangelia Risso, 1826 (multispiral) ' Mangi- 

liella Bucquoy, Dautzenberg & Dollfus, 1883 

(paucispiral) (Fig. 1). 

Raphitoma Bellardi, 1847 (multispiral) 

Philbertia Monterosato, 1884 (paucispiral). 

Bela Gray, 1847 (multispiral) Fehria van 

Aartsen, 1988 (paucispiral). 

Lophiotoma Casey, 1904 (multispiral) Lo- 

phioturris Powell, 1964 (paucispiral). 

Parasyrinx Finlay, 1924 (multispiral) / Lira- 

syrinx Powell, 1942 (paucispiral). 

Protoconch Morphology and Turrid Larval 
Development 

A general correlation between protoconch 
morphology and mode of larval development 
has been demonstrated (Thorson, 1946; 
Shuto, 1974; Robertson, 1976; Jablonski & 
Lutz, 1980, 1983) and may be examined here 
with reference to what is known of turrid larval 
biology. 

Most described turrid egg-capsules are 
ovoid and lenticular, with a central dorsal 
plug, and attached by the ventral side to the 
substrate (Lebour, 1934; Thorson, 1946; 
Knudsen, 1950; Bändel, 1976; Bouchet & 



Waren, 1 980). A notable exception is the egg- 
capsules of the subfamily Clavatulinae, which 
are stalked and purse-shaped (Kilburn, 
1985). A capsule contains several dozens to 
a few hundred eggs. Nurse eggs have not 
been reported. 

As is frequent with species with plank- 
totrophic larval development, the complete 
development, from oviposition to metamor- 
phosis, has not been followed for any single 
specimen, but evidence can be derived from 
numerous scattered and fragmentary data. 
Turrid veligers, although never abundant, are 
frequently recorded in meroplankton samples 
(Franc, 1950; Richter & Thorson, 1975; Le- 
bour, 1934; Thinot-Quiévreux, 1969, 1972) 
and exhibit a broad range of morphological/ 
sculptural types (Kay, 1979, and personal 
observations). From published observations, 
the behaviour of such veligers is similar to 
that in other gastropod planktotrophic vel- 
igers, i.e. the larvae actively swim in the epi- 
pelagic layers of the water column while feed- 
ing on phytoplankton. These veligers have 
protoconchs of the multispiral type (Fig. la); 
the larvae of the many deep-sea turrids that 
undertake ontogenetic vertical migrations 
(Bouchet & Fontes, 1981; Killingley & Rex, 
1985) enter into this category. The planktonic 
phase is a period of active feeding and very 
active growth, and the whole protoconch II is 
secreted during this planktonic planktotrophic 
phase. The total length of the planktonic 
phase is not known with precision and cer- 
tainly varies between species, but by compar- 
ison with other prosobranch families a range 
from three to eight weeks is a likely estimate. 

The complete larval development of Oeno- 
pota levidensis (Carpenter, 1864), a species 
with paucispiral larval shell, has been de- 
scribed with considerable detail based on 
laboratory observations (Shimek, 1986). De- 
velopment to a veliger occurs within the egg- 
capsule in about 50 days. After hatching, the 
larvae swim actively for a period of a few 
days, and then live a demersal life on the bot- 
tom of the culture vessel. The larvae were 
experimentally fed with algal suspensions; 
shell and velar dimensions increase dunng 
the swimming phase, after which the veliger 
does not get appreciably larger, and by the 
15th posthatching day, the protoconchs are 
fully formed. Demersal development then 
continues without shell growth and the 
veligers metamorphose after 25 days. 

Based on these observations, Shimek 
(1986) concludes that paucispiral proto- 



TURRID PROTOCONCH 



71 




FIG. 1. The multispiral (la) and paucispiral protoconch (Id) define respectively the genera Mangelia and 
Mangiliella. There is no evidence that Mangelia striolata Risso, 1826, (lb) and M. vauquelini (Payraudeau, 
1826) (1c) are more closely related to each other than they are to Mangiliella multilineolata (Deshayes, 1833) 
(le) or M. taeniata (Deshayes, 1833) (If). Mangiliella should be synonymized with Mangelia. lb = 7.8 mm; 
1c = 9.0 mm; le = 5.7 mm; If = 5.5 mm. Scale lines 200 (xm. 
All specimens from Calvi, Corsica. 



conchs in turrids cannot be interpreted as ev- 
idence for lack of a planktonic stage, as 
claimed by Thorson (1935, 1946). A develop- 
ment with intracapsular metamorphosis had 
been inferred in Atlantic Arctic Oenopota by 
Thorson from the contents of egg capsules, 
because there is no appreciable size differ- 
ence between the shells of intracapsular em- 
bryos ready to hatch and the smallest benthic 
juveniles. Shimek's observations certainly 
demonstrate that a long demersal phase may 
occur in species with paucispiral larval shells. 
However they do not, in my opinion, weaken 



the distinction between planktotrophic and 
non-planktotrophic larval development. I fully 
admit that the duration of the post-hatching 
phase in O. levidensis may probably equal 
the duration of some of the shorter-lived 
planktonic planktotrophic veligers. The short 
initial free-swimming planktonic phase aside 
(though this is admittedly important in terms 
of dispersal), the biology of the veliger of O. 
levidensis is markedly different from that of 
planktotrophic veligers: it does not swim, ex- 
cept for brief moments following artifical stim- 
ulation, and it does not grow or secrete pro- 



72 



BOUCHET 



toconch shell material. By contrast with 
species with multispiral protoconchs, the iso- 
topic composition of the shells of deep-water 
turrids with paucispiral protoconchs indicates 
that a vertical ontogenetic migration does not 
occur (Killingley & Rex, 1985). 

Just as the duration of the planktonic phase 
varies among species with planktotrophic de- 
velopment, the duration of the swimming and 
demersal phases may be expected to vary 
considerably between species with non- 
planktotrophic development. Whether the 
case of Oenopota levidensis represents an 
average duration or an extreme is not known. 
Kllburn's (1985) observation that in the pau- 
cispiral protoconch of Clavatula tripartita 
(Weinkauff, 1876) "the defining varix pre- 
cedes the veliconch lip by about one-sixth of a 
whorl" may be an indication that a swimming 
phase is also present. Clearly, more data are 
needed on the the larval biology of additional 
species. In particular, it would be of great 
interest to know if the larval biology of warm- 
water turrids with paucispiral larval shells con- 
forms to the pattern described for the cold- 
water Oenopota levidensis. Knudsen (1950) 
described the egg-capsules and contained 
embryos of several West African continental 
shelf turrids, and inferred lecithotrophic or 
"direct" development, Non-planktotrophic lar- 
val development has been inferred from pro- 
toconch morphology in numerous temperate, 
tropical and deep-water turrids. 

Available evidence in the family Turridae 
therefore confirms the general correlation be- 
tween protoconch morphology and mode of 
development. This correlation, which had 
been assumed by Powell (1942 and later pa- 
pers) based on the data then available in 
other prosobranch families, is applicable to 
Recent as well as fossil turnds. 

Polanty of Changes in Protoconch 
Morphology 

Although the mode of development, and 
hence protoconch morphology, is a species- 
specific character throughout the range of a 
species (Hoagland & Robertson, 1988; 
Bouchet, 1989), it is known to change through 
time in monophyletic lineages (see refer- 
ences below). 

Powell (1942), certainly influenced by Fin- 
lay (1931), believed the paucispiral proto- 
conch / "sedentary larva" type to represent 
the ancestral condition, and the multispiral 
protoconch "free swimming larva" type to 



represent the derived condition. He moreover 
believed this change ot be irreversible: 
"When once the radical embryonic change 
from a sedentary to a free swimming larva 
takes place, both types appear to develop in- 
dependently, for there is no evidence sug- 
gesting indiscriminate change and rechange 
between these two types of embryos. It would 
seem rather that the Sinusigera' apex is an 
evolutionary culmination from the less effi- 
cient paucispiral type." This is, I think, a good 
example of circular reasoning. How can 
changes in protoconch morphology be recog- 
nized if, by definition, they are used to distin- 
guish genera? 

Contrary to Powell's assumption, the evi- 
dence throughout marine invertebrates is that 
planktotrophic larval development represents 
the ancestral (plesiomorphic) condition and 
the loss of planktotrophy is a derived (apo- 
morphic) condition (Strathmann, 1978, 1985). 
Strathmann (1978), however, argues that the 
loss of planktotrophy is in theory reversible as 
long as the larval ciliary opposed band sys- 
tem of feeding is not lost during intracapsular 
development. Such is the case in Oenopota 
levidensis (Shimek, 1986), so that reacquisi- 
tion of planktotrophy is in theory possible in a 
descendant of that species. 

The final answer to this question can only 
be found in the study of modes of develop- 
ment in fossil and Recent species, because 
biological time is too short for an approach 
other than theoretical. However, in the Tur- 
ridae, there are presently no available data 
that combine (a) description of a small lineage 
through time, and (b) description of proto- 
conch types without a preconceived idea on 
their significance in classification. 

Fragmentary data are scattered throughout 
the taxonomic literature on Caenogas- 
tropoda: in Neogene to Recent eastern Atlan- 
tic Terebridae (Bouchet, 1 981 ); in Paris Basin 
Eocene Tritons (Gougerot & Le Renard, 
1980); in Pliocene to Recent Mediterranean 
nassariids (Martinell & Cuadras, 1977) and 
Trophon (Bouchet & Waren, 1985); in Neo- 
gene to Recent eastern American Ficus 
(Smith, 1945); in North American Paleogene 
Volutidae (Hansen, 1983). All these papers 
point out many examples of changes from 
multispiral to paucispiral protoconchs (loss of 
planktotrophy), but not a single case of 
change from paucispiral to multispiral proto- 
conchs (reacquisition of planktotrophy) is re- 
corded. Therefore, although reacquisition of 
planktotrophy is theoretically embryologically 



TURRID PROTOCONCH 



73 



feasable, evidence is still wanting, and in this 
paper my working hypothesis is that the loss 
of planktotrophy in Turridae is, as a rule, not a 
reversible phenomenon. 

Refutation of Mode of Development as a 
Generic Character 

If Powell's system of genera was applicable 
to Turridae, it should equally be applicable, on 
the same basis, to other families of marine 
gastropods. As reviewed above, protoconch 
morphology correlates with mode of develop- 
ment. If the planktotrophic / non-plankto- 
trophic dichotomy is to be given generic value 
in the Turridae, then logically it should also be 
given the same value in other Caenogas- 
tropoda. However, there are many examples 
of genera (e.g. Nassahus, Chlcoreus, Alva- 
nia, Littorina) that are believed to be mono- 
phyletic and that contain both species with 
planktotrophic and non-planktotrophic larval 
development. As a matter of fact, there are 
sibling species in many genera that are dis- 
tinguished only on the basis of their mode of 
development (Hoagland & Robertson, 1988; 
Bouchet, 1989). Recognition of genera based 
only on the paucispiral vs. multispiral dichot- 
omy has been explicitly rejected by Robert- 
son (1976) in general and by Marshall (1978, 
1983) in the families Cerithiopsidae and 
Thphoridae. The practice is indeed quite re- 
stricted to Turridae (but has already been 
challenged by Kilburn, 1983), but it is not jus- 
tified by any larval biology feature that would 
be unique to turrids. 

Genera that are established on the pau- 
cispiral vs. multispiral dichotomy are prone to 
be artificial and polyphyletic. I have in Figures 
2 and 3 presented a hypothetical evolutionary 
tree starting from an ancestor with multispiral 
larval shell. Through speciation and anage- 
netic evolution, this ancestral species (P 1) 
gives rise to local shorter-lived species that 
have lost planktotrophy (L 1 to L 4), and 
longer-lived species that retain planktotrophy 
(P 2 to P 5). That species with non-plank- 
totrophic development are more local and 
outlived by species with planktotrophic devel- 
opment has been both predicted (Scheltema, 
1977), and noticed (Powell, 1942; Hansen, 
1978, 1980, 1983; Jablonski, 1982, 1986) in 
the fossil record. Loss of planktotrophy can 
occur independently several times through 
various evolutionary scenarios, such as insu- 
lar endemism or relict distribution resulting 
from changes in the climatic and/or geo- 





FIGS. 2-3. Hypothetical evolutionary tree of a turrid 
lineage. In Powell's system of genera (Fig. 2, 
above), the species with paucispiral protoconchs 
are grouped in one genus, although they do not 
share a common ancestor, and the species with 
multispiral protoconchs in another genus. Such 
genera are respectively polyphyletic and paraphy- 
letic. In the absence of detailed knowledge on the 
tree (and this is by far the most frequent situation in 
Turridae), the most parsimonious solution is a sin- 
gle genus comprising species both with multispiral 
and paucispiral protoconchs (Fig. 3, below). 



74 



BOUCHET 







FIG. 4. Loss of planktotrophy may permit an adaptive radiation. This character may then be expressed in 
classification. A genus defined by a paucispiral protoconch is monophyletic and perfectly acceptable. All 
species with paucispiral protoconch are derived from a common ancestor. 



graphic environments. According to Powell's 
system, all species with paucispiral larval 
shells (L 1 to L 4) are classified into one ge- 
nus, and species with multispiral larval shells 
(P 1 to P 5) in another. It is obvious that any 
one of species L 1 , L 2, or L 4 is more closely 
related to its immediate ancestor with multi- 
spiral larval shell than it is to other species 
with the same protoconch morphology. The 
genus L is not based on parental affinities and 
should therefore be rejected. I would there- 
fore regard as valid a single genus encom- 
passing all the species descended from P 1, 
whatever their mode of larval development. 

In other words, when teleoconch, radula 
and anatomical characters indicate that a sin- 
gle evolutionary clade is involved, protoconch 
morphology should not be used to split it into 
different genera based on the mode of devel- 
opment. Subgenera based on the same con- 
cept are no more acceptable, since their poly- 



phyletism would not be more justifiable. This 
view is not revolutionary in turrid taxonomy, 
and a number of authors have already al- 
lowed differing modes of development to co- 
exist within a single turrid genus (e.g. Bouchet 
& Waren, 1980; Kilburn, 1983; Maes, 1983). 

It may be worth noting, at this point, that 
even if the loss of planktotrophy was, contrary 
to my assumption, reversible, then the above 
conclusion remains valid. 

This does not of course necessarily mean 
that all turrid genera have or should have both 
species with paucispiral and species with 
multispiral protoconchs. Loss of planktotro- 
phy may represent the onset of an adaptive 
radiation, such as in an insular or polar envi- 
ronment (Fig. 4). All known Recent Oenopota 
have paucispiral protoconchs (Bogdanov, 
1989), and all Pliocene Oenopota apparently 
already had the same mode of development 
(Harmer, 1914-19; Beets, 1946). Although I 



TURRID PROTOCONCH 



75 



fully expect that somewhere in the ancestry of 
Oenopota there is a species with multispiral 
larval shell, it is obvious that the loss of plank- 
totrophy has permitted an adaptive radiation 
in the arctic/subarctic environment where 
planktotrophy is selected against, and a pau- 
cispiral protoconch is now certainly a hallmark 
of Oenopota. 

Admittedly, each of the successive specia- 
tion events by loss of planktotrophy in Figure 
2 represents a discrete radiation event, and 
one could imagine a nomenclature with L 1, 
L 2 + L 3, and L 4 each in a different genus. 
In my view, this should not be recommended 
in the present state of our knowlege: much 
of our genus-level and species-level taxon- 
omy is based on shell characters only, and 
this is all we will ever have in the many fossil 
taxa. The state of the art in turrid taxonomy is 
such that these discrete monophyletic genera 
cannot be easily recognized. Splitting and 
recognition of many monotypic genera with 
paucispiral larval shell will not help our under- 
standing of turrid evolution and, considering 
the sheer size of the family, is most likely to 
result in absolute chaos. 

What Future for the Turrid Protoconch as a 
Supraspecific Character? 

After all that has been said above, is there 
a future left for the use of protoconch in turrid 
taxonomy? That the paucispiral vs. multispiral 
dichotomy has no supraspecific taxonomical 
significance does not imply that just any kind 
of protoconch is to be expected in a turrid 
genus. 

The consequence of the genus concept ad- 
vocated here is that paucispiral protoconchs 
should only be compared with paucispiral pro- 
toconchs, and multispiral protoconchs with 
multispiral protoconchs. 

Accompanying the loss of planktotrophy, 
paucispiral protoconchs have few distinctive 
characters (see however Bodganov, 1989), 
and extensive parallelism and convergence 
between distant, unrelated taxa is the rule. It 
is very unlikely that these protoconchs have a 
promising future in turrid classification. By 
contrast, as has been noted above, the mor- 
phology and sculpture of planktotrophic turrid 
veligers is remarkably diverse and present a 
vast array of characters that have not yet 
been fully appreciated. I believe that different 
sculptural types within the multispiral proto- 
conch have a profound taxonomic meaning, 
just as teleoconch characters happen not to 



be random within a genus. Neopleuroto- 
moides was segregated from Pleurotomella 
despite extremely similar teleoconchs, be- 
cause the sculpture of their multispiral proto- 
conchs differ fundamentally (Shuto, 1 971 ): re- 
spectively two spiral keels with axial pillars, 
and diagonal cancellation extending over 
most of the whorl. (It is worth noting here that, 
at this stage of our knowledge, a species with 
paucispiral larval shell cannot be attributed to 
Pleurotomella or Neopleurotomoides, an ob- 
vious difficulty with the genus concept advo- 
cated in the present paper.) 

What remains for the future is to identify 
what are the basic sculptural types, evaluate 
the degree of convergence (for instance, has 
the diagonally cancellated sculpture ap- 
peared only once?), and understand how sec- 
ondary sculptural types can be derived from 
more fundamental ones. There is certainly a 
rich and rewarding future use of turrid proto- 
conchs in taxonomy. 



LITERATURE CITED 

AARTSEN, J. J. VAN, 1988, European Mollusca: 
Notes on less well-known species. XII. Bela men- 
khorsti nom. nov. = Pleurotoma nana Scacchi, 
1836 non Deshayes, 1835 and Fehna (nov. gen.) 
zenetouae nov. spec. La Conchiglia. 232 233: 
30-31. 

AARTSEN, J. J. VAN & M. С FEHR- DE WAL, 1978, 
The subfamily Mangeliinae Fischer, 1887 in the 
Mediterranean, Conchiglie. 14: 97-110. 

AARTSEN, J. J. VAN, H. P. MENKHORST & E. GIT- 
TENBERGER, 1984, The manne Mollusca of the 
Bay of Algeciras, Spain, with general notes on 
Mitrella. Marginellidae and Turridae. Bastena, 
supplement 2: 1-135. 

BÄNDEL, К., 1976, Spawning, development and 
ecology of some higher Neogastropoda from the 
Caribbean Sea of Colombia (South America). 
Veliger, 19: 176-193. 

BEETS, C, 1946, The Pliocene and lower Pleis- 
tocene gastropods in the collections of the Geo- 
logical Foundation in the Netherlands (with some 
remarks on other Dutch collections). Mededeelin- 
gen van de Geologische Stichting. (С, 4) 1 (6): 
1-166, 6 pis. 

BERNASCONI, M. P. & E. ROBBA, 1984, The 
Pliocene Turridae from Western Liguria. I. Cla- 
vinae, Turrinae, Turriculinae, Crassispirinae, Bor- 
soniinae, Clathurellinae. Bollettino del Museo 
Regionale di Scienze Naturali, Torino. 2: 257- 
358. 

BOGDANOV, J. P., 1989. Morphological transfor- 
mation in radula and protoconch of Oenopotinae 
Bogdanov, 1987. La Conchiglia. 233 236: 35- 
47. 



76 



BOUCHET 



BOUCHET, P., 1981, Evolution of larval develop- 
ment in eastern Atlantic Terebridae, Neogene to 
Recent. Malacologia. 21:363-369. 

BOUCHET, P., 1989. A review of poecilogony m 
gastropods. Journal of Molluscan Studies. 55: 
67-78. 

BOUCHET, P. & J. С FONTES, 1981, tVligrations 
verticales des larves de gastéropodes abyssaux: 
arguments nouveaux dus à l'analyse isotopique 
de la coquille larvaire et postlarvaire. Comptes 
Rendus des Seances de l'Académie des Sci- 
ences. Pans. (3)292: 1005-1008. 

BOUCHET, P & A. WAREN, 1980, Revision of thie 
Northi-East Atlantic bathyal and abyssal Turridae. 
Journal of Molluscan Studies, supplement 8: 1- 
120. 

BOUCHET. P. & A. WAREN, 1985, Revision of the 
Northeast Atlantic bathyal and abyssal Neogas- 
tropoda. excluding Turridae. Bollettino l\/lalaco- 
logico. supplement 1: 121-296. 

FINLAY, H. J.. 1931, On Austrosassia. Austro- 
tiarpa. and Austrolitties. new genera; with some 
remarks on the gastropod protoconch. Transac- 
tions of the New Zealand Institute. 62: 7-19. 

FRANC. A., 1950, Ponte et larves planctoniques de 
Ptiilbertia purpurea (Montagu). Bulletin du Labo- 
ratoire Maritime de Dinard. 33: 23-25. 

GOUGEROT. L. & J. LERENARD, 1980, Clefs de 
determination des petites espèces de gas- 
téropodes de l'Eocène du bassin parisien. XII. La 
famille des Triphondae. Cahiers des Naturalistes. 
35: 41-59. 

GOUGEROT, L. & J. LERENARD, 1981, Clefs de 
détermination des petites espèces de gas- 
téropodes de l'Eocène du bassin parisien. XIX. 
Le genre Pleurotomella. Cahiers des Natura- 
listes. 37: 81-92. 

HANSEN. T.. 1978. Larval dispersal and species 
longevity in lower Tertiary gastropods. Science. 
199: 885-887. 

HANSEN. T., 1980, Influence of larval dispersal 
and geographic distribution on species longevity 
in neogastropods. Paleobiology. 6: 193-207. 

HANSEN. T.. 1983. Modes of larval development 
and rates of speciation in early Tertiary neogas- 
tropods. Science. 204: 501-502. 

HARMER, F. W., 1914-1919, The Pliocene Mol- 
lusca of Great Britain, volume 1. Palaeonto- 
graphical Society, London. 

HOAGLAND, K. E. & R. ROBERTSON, 1988, An 
assessment of poecilogony in marine inverte- 
brates: phenomenon or fantasy? Biological Bul- 
letin. 174: 109-125. 

JABLONSKI, D., 1982. Evolutionary rates and 
modes in late Cretaceous Gastropods: role of lar- 
val ecology. Proceedings of the Third North Amer- 
ican Paleontological Convention. 1 : 257-262. 

JABLONSKI. D., 1986, Larval ecology and macro- 
evolution in marine invertebrates. Bulletin of Ma- 
rine Science. 39: 565-587. 

JABLONSKI, D. & R. LUTZ, 1980, Molluscan larval 
shell morphology. Ecological and paleontological 
applications. In RHOADS, D.C. & R. LUTZ, eds.. 



Skeletal Growth of Aquatic Organisms. 323-377. 
Plenum Press, New York, 

JABLONSKI. D. & R. LUTZ, 1 983, Larval ecology of 
marine benthic invertebrates: paleobiological im- 
plications. Biological Reviews. 58: 21-89. 

KAY. A.. 1979. Hawaiian manne shells. B. P. 
Bishop Museum Special Publication, 64: 1-653. 

KILBURN. R. N.. 1983. Turndae of southern Afnca 
and Mozambique. Part 1. Subfamily Turrinae. 
Annals of the Natal Museum. 25: 549-585. 

KILBURN. R. N.. 1985, Turridae of southern Africa 
and Mozambique. Part 2. Subfamily Clavatuli- 
nae. Annals of the Natal Museum. 26: 417-470. 

KILLINGLEY. J. & M. REX. 1985, Mode of larval 
development in some deep-sea gastropods indi- 
cated by oxygen-18 values of their carbonate 
shells. Deep Sea Research. 32:809-818. 

KNUDSEN, J., 1950, Egg capsules and develop- 
ment of some marine prosobranchs from tropical 
West Africa. Atlantide Report. 1 : 85-130. 

LEBOUR, M., 1934, The eggs and larvae of some 
British Turridae. Journal of the Manne Biological 
Association of the United Kingdom. 19: 541-554. 

MAES, V. O., 1983, Observations on the systemat- 
ics and biology of a turrid gastropod assemblage 
in the British Virgin Islands. Bulletin of Manne 
Science. 33: 305-335. 

MARSHALL, В., 1978, Cerithiopsidae of New 
Zealand, and a provisional classification of the 
family. New Zealand Journal of Zoology. 5: 47- 
120. 

MARSHALL, В., 1983, A revision of the Recent 
Triphondae of Southern Australia. Records of the 
Australian Museum, supplement 2: 1-119. 

MARTINELL. J. & С CUADRAS, 1977, Bioestadis- 
tica у analysis multivariable aplicados a la com- 
paración de una población actual y otra fósil, 
atribuidas a Sphaeronassa mutabilis (Linné): 
aportación a la sistemática del genero Sphaero- 
nassa Locard, 1886. Studia Geológica (Sala- 
manca), 13: 89-103. 

POWELL, A. W. В., 1942, The New Zealand Re- 
cent and fossil Mollusca of the family Turridae, 
Bulletin of the Auckland Institute and Museum. 2: 
1-188. 

POWELL, AW. В., 1964, The family Turndae in the 
Indo-Pacific. Part 1. The subfamily Turrinae. 
Indo-Pacific Mollusca. 1(5): 227-345. 

POWELL, A. W. В., 1966, The Molluscan families 
Speightiidae and Turridae. Bulletin of the Auck- 
land Institute and Museum, 5: 1-184, 23 pis. 

RICHTER, G. & G. THORSON, 1975, Pelagische 
Prosobranchier-Larven des Golfes von Neapel. 
Ophelia. 13: 109-185. 

ROBERTSON, R., 1976 (■1974"), Manne proso- 
branch gastropods: larval studies and systemat- 
ics. Thalassia Jugoslavica. 10: 213-238. 

SCHELTEMA, R., 1977, Dispersal of marine inver- 
tebrate organisms: paleobiogeographic and bio- 
stratigraphic implications. In KAUFFMANN, E. G. 
& J. E. HAZEL, eds., Concepts and Methods of 
Biostratigraphy. 78-122. Dowden, Hutchinson 
and Ross, Stroudsburg. 



TURRID PROTOCONCH 



77 



SHIMEK, R. L., 1986, The biology of the northwest- 
ern Pacific Turridae. V. Demersal development, 
synchronous settlement and other aspects of the 
larval biology of Oenopota levidensis. Interna- 
tional Journal of Invertebrate Reproduction and 
Development. 10: 313-333. 

SHUTO, T., 1971 , Taxonómica! notes on the turrids 
of the Siboga-collection originally deschbed by 
M. M. Schepman, 1931 (part III). Venus. 30(1): 
5-22. 

SHUTO, T., 1974, Larval ecology of prosobranch 
gastropods and its bearing on biogeography and 
paleontology. Lethaia. 7: 239-256. 

SMITH, В., 1945, Observations on gastropod pro- 
toconchs. Paleontographica Americana. 3: 1-48. 

STRATHMANN, R., 1978, The evolution and loss 
of feeding larval stages of marine invertebrates. 
Evolution. 32: 894-906. 

STRATHMANN, R., 1985, Feeding and nonfeeding 
larval development and life-history evolution in 
marine Invertebrates. Annual Review of Ecology 
and Systematics. 16: 339-361. 

THIRIOT-QUIEVREUX, C, 1969, Caractenstiques 



morphologiques des véligères planctoniques de 
gastéropodes de la région de Banyuls-sur-mer. 
Vie et Milieu, (8)20: 333-336. 

THIRIOT-QUIEVREUX, C, 1972, Microstructures 
de coquilles larvaires de prosobranches au mi- 
croscope électronique à balayage. Arcfiives de 
Zoologie Expérimentale et Générale. 113: 553- 
564. 

THORSON, G., 1935, Studies on the egg-capsules 
and development of arctic marine prosobranchs. 
Meddelelser от Grönland. 100: 1-71. 

THORSON, G., 1946, Reproduction and larval de- 
velopment of Danish marine bottom inverte- 
brates, with special reference to the planktonic 
larvae in the sound. Meddelelser fra Kommis- 
sionen for Danmarks Fiskeri- og Havunderso- 
gelser (Plankton). 4: 1-523. 

TRYON, G. W., 1884, Manual of Conchology. vol. 
VI. Conidae, Pleurotomidae. 413 pp., 34 pis. Phil- 
adelphia. 



Revised Ms. accepted 21 June 1990 



MALACOLOGIA, 1990, 32(1): 79-87 

TURRID FAUNAS OF PACIFIC ISLANDS 

E. Alison Kay 
Department of Zoology, University of Hawaii, Honolulu, Hawaii, U.S.A. 96822 

ABSTRACT 

Of the more than 300 species in 45 genera and seven subfamilies of Turridae described or 
recorded from the shallow waters (• 100 m) of the tropical Pacific, 160 species are recognized 
as occurring on the islands of the central Pacific. Compared with the turrids of continental 
shorelines (tropical west America and New Zealand), the Daphnellinae are better represented 
and the Clavinae less well represented on central Pacific islands. The turrids in general comprise 
a lesser proportion of the gastropod fauna, and shells in all turrid subfamilies are smaller on 
average. Species distribution is patchy and endemism is high in the Hawaiian Islands (47%) and 
at Easter Island (80%). Less than 30% of the central Pacific turrids are widespread in the 
Indo-Pacific, nearly 40% also occur in the western Pacific, and about 30% are known only from 
the Pacific Plate. Nearly 50% of the genera represented on the Pacific Plate apparently lack a 
fossil record. 

Key words: Turridae, biogeography, Indo-Pacific, distribution, island faunas. 



INTRODUCTION 

Members of the neogastropod family Tur- 
ridae are among the most numerous of ma- 
rine gastropods. The family may include as 
many as 2,000 Recent species (Kilburn, 
1983), 550 (Powell, 1966; McLean, 1971) to 
679 (Bouchet, 1990) generic and subgeneric 
units, and from 9 to 15 subfamilies (Powell, 
1966; McLean, 1971). Turrids are found 
throughout the world from the poles to the 
tropics, and from intertidal coral reefs to 
depths of more than 5,000 m in the abyssal 
regions of the sea. They can be both abun- 
dant and speciose. In the deepwaters of the 
Atlantic, they are the most abundant mollusc 
group in terms of both number of specimens 
and number of species (Bouchet & Waren, 
1980), and more than 30% of the described 
Alaskan, Arctic, and North Pacific boreal shal- 
low-water gastropods are turrids (Shimek, 
1986). Turrids of the tropical Pacific Ocean 
are not so easily quantifiable, despite the fact 
that more than 900 species names have been 
proposed or recorded for turrids in that area. 

This study represents an initial analysis of 
turrid species composition and distribution in 
three major island groups in the central Pa- 
cific, the Marshall Islands (Enewetak), the Ha- 
waiian Islands, Tahiti and the Tuamotus 
(French Polynesia). The turrids from these is- 
lands are also examined in terms of several of 
the generalizations now recognized as per- 
taining to the distribution of marine organisms 



in the Pacific (cf. Kay, 1980; Kay, 1984). Is 
there an attenuation of species and higher 
taxonomic groups from west to east across 
the Pacific? Are the relationships of Pacific 
island turrids to the west rather tfian the east? 
Is there a distinctive Pacific Plate element in 
the fauna? 



MATERIALS AND METHODS 

Collections 

A long-term study of the turrids of Pacific 
islands, particularly those of the Hawaiian Is- 
lands; Enewetak, Marshall Islands; Fanning 
Island, Line Islands; and Guam, Mariana Is- 
lands, has permitted collection in the field and 
comparison of field-collected material with 
type and reference collections in the Acad- 
emy of Natural Sciences Philadelphia (Pease 
and Garrett types); Australian Museum, Syd- 
ney (Hedley and Laseron collections); B. P. 
Bishop Museum, Honolulu, Hawaii (Thaanum 
collection of Okinawan turrids and material 
identified by Dall); The Natural History Mu- 
seum, London (Cuming collection, f^eeve 
types, Melvill and Standen material); the 
Musée d'Histoire Naturelle, Paris (Hervier, 
Crosse and Souverbie types); the Museum of 
Comparative Zoology, Harvard University 
(Pease collection); the National Museum of 
Natural History, Smithsonian Institution, 
Washington, D.C. (Dall, Pease and Garrett 



79 



80 



KAY 



TABLE 1. Numbers of turrid species described from the Pacific 1800-present. Western Pacific includes 
the Philippines, Queensland, Malayan archipelago. Lifu, Fiji. Fossil species included. 





WESTERN 


PRINCIPAL 




CENTRAL 


PRINCIPAL 


YEAR 


PACIFIC 


AUTHORS 




PACIFIC 


AUTHORS 


1800-1825 


3 


Lamarck 








1826-1850 


138 


Reeve, Hinds 




14 


Reeve 


1851-1875 


48 


Garrett, Gould, Souverbie 


61 


Garrett, Pease, Dunker 


1876-1900 


203 


Nervier, Melvill and Standen 


8 


Dall, Smith 


1901-1925 


241 


Hedley, Schepmann 




4 


Dall 


1926-Present 


74 


Ladd, McNeil, Shuto, 


Noda 


26 


Kay, Powell 



material: Ladd Enewetak and Bikini fossil 
types): and the National Museum of Wales, 
Cardiff (Tomlin collection). I rely also on the 
published work of Dautzenberg & Bouge 
(1933), Hedley (1922), Hervier (1896-1898), 
Melvill & Standen (1895-1897), and Richard 
(1982) to augment species lists compiled 
from museum and field collections. 

Turrid Taxonomy 

Turrid taxonomy remains a major problem 
in the understanding of turrid systematics and 
distribution. The importance of radular studies 
in the Turridae has been discussed by 
McLean (1971) and Kilburn (1983, 1985, 
1988). The nature of the material that serves 
as the bases for this study for the most part 
precluded study of radulae, and placement of 
turrids in subfamilies herein is largely based 
on conchological features recognized by 
Powell (1942, 1966), that is, position of the 
sutural gap and protoconch type. Although it 
is widely recognized that shell type can no 
longer be considered a valid generic or sub- 
generic criterion, I nevertheless find that pro- 
toconch form does seem to provide insight 
into subfamilial placement of many turrid spe- 
cies. 

Seven subfamilies are recognized here as 
occurring in the island Pacific, the Turrinae of 
Powell (1966, 1967, but excluding Turn- 
drupa); Crassipinnae (of Kilburn, 1983, in- 
cluding Turhdrupa); the Drilliinae sensu 
McLean (1971) and Kilburn (1988): the 
Mangeliinae and Daphnellinae sensu Powell 
(1966): the Borsoniinae, as recognized by 
Kilburn (1986): and the Cochlespirinae (of 
Powell, 1966). 

To arrive at an estimate of the numbers of 
turrid species associated with the tropical Pa- 
cific sensu lato and specifically with the is- 
lands of the central Pacific, the number of tur- 
rid species described from the western 



Pacific (Philippines, Queensland, Lifu, New 
Caledonia, Fiji) and the central Pacific (Ha- 
waii. French Polynesia, Samoa, etc.) was 
counted, and other species referred to the Pa- 
cific included in the list. More than 850 turrid 
species have been described from the Pacific 
(sensu lato) since 1800, of which less than 
15% were described from islands in the cen- 
tral Pacific (Table 1). An additional 75-100 
species have been recorded from the tropical 
Indo-Pacific, for example species described 
from Réunion by Deshayes and the northern 
Indian Ocean by Nevill. The total list was then 
culled for synonyms and other errors and is 
here reduced to about 160 species referable 
to the central Pacific islands. 

Pacific Island Biogeography 

Definitions utilized to delimit the islands of 
the central Pacific and their associated bio- 
geographic regions, the Indo-Pacific and the 
western Pacific (Fig. 1) are as follows. By 
Indo-Pacific is meant "the Indian Ocean in- 
cluding contiguous seas, and the Pacific 
Ocean as far east as Easter Island but ex- 
cluding the area occupied by the coast and 
offshore islands , . . of the Western Hemi- 
sphere" (Springer, 1982). The western Pa- 
cific is distinguished as the Pacific Ocean 
west of the western margin of the Pacific litho- 
spheric plate which includes such inland seas 
as the South China Sea, Arafura Sea, and 
Coral Sea (Springer, 1982), and the islands of 
Lifu (Loyalty Islands), Fiji, Okinawa (Ryukyu 
Islands) and Guam (Mariana Islands). The 
"Pacific islands" of this study (Enewetak, 
Marshall Islands: Hawaii: and Tahiti and Tu- 
amotus, French Polynesia) are primarily on 
non-marginal portions of the Pacific Plate, 

THE TURRIDS OF PACIFIC ISLANDS 

The distribution of the currently recognized 
160 species in six subfamilies is shown in Fig- 



TURRID FAUNAS OF PACIFIC ISLANDS 



81 




FIG. 1 . The Pacific Ocean showing the margins of the Pacific Plate and major island groups. 



ure 2. About 90% of the Pacific island turrlds 
represent four subfamilies: the Mangeliinae 
(43%), the Daphnellinae (29%), the Dhlliinae 
(9%), and Turrinae (7%). The remaining tur- 
rids include the Crassispirinae (7%), Borsoni- 
inae (4%), and Cochlespihnae (1%). Subfam- 
ily representation confirms Powell's (1966) 
summary of turrid distribution in the Pacific: 
"In the atolls, reefs and small isolated island 
groups of the Indo-Pacific, large turrids are 
either absent or poorly represented: Lo- 
phiotoma acuta is almost the only exception, 
the faunules being composed mainly of small 
mangelinids and daphnellids. The most char- 
acteristic mangelinid genera of the islands of 
the Indo-west Pacific are Eucithara, Lienar- 
dia, Etrema. Macteola, . . .". 

More than 80% of the turhd species asso- 
ciated with central Pacific islands are small 
(<5 mm in length), and the shells are sturdy 
and colorful. Except in the turrines and the 
deep-water daphnellines, shell form is 
marked by a short siphonal canal and usually 
reinforced labial extremity. Most of the shells 
also represent subfamilies that are character- 
ized by elaborate protoconchs (i.e. Daphnel- 



linae and Mangeliinae). Veligers of daphnel- 
lines and mangelines in Hawaii have been 
raised for periods of several weeks (J. B. Tay- 
lor, 1975). Iredalea exilis (Pease, 1868) (Dril- 
liinae) among the shallow water turrids is the 
only species with the protoconch of the type 
associated with direct or lecithotrophic devel- 
opment. 

Most of the central Pacific island turrids are 
known from depths of less than 30 m, and are 
commonly found at depths of 10-20 m on the 
fore-reef in rubble and sand. Only one spe- 
cies, Iredalea exilis, is frequently found in the 
intertidal on reef flats. Eleven species are re- 
corded from depths of more than 300 m, nine 
species dredged in the Hawaiian Islands and 
\\NO-Pleurotomella dubia Schepmann, 1913, 
and P. allisoni Rehder & Ladd, 1973-dredged 
from depths of more than 1 ,000 m on guyots 
in the mid-Pacific mountains (Rehder & Ladd, 
1973). 

The shallow water turrids of central Pacific 
islands are neither numerous nor abundant. 
Turrids feed mainly on polychaetes (J. D. Tay- 
lor, 1977; Maes, 1983), and they comprise 
about 15% of the predatory gastropod spe- 



82 



KAY 



E 

CO 

z¡ 
CO 



60 



50 



40 



30 



20 - 



10 



^^ш 



if 



Ш Turr 
й Crass 
S Drill 



Ш Mang 
BHlDaph 
Ш Borson 




QLD 



OKIN 



GUM 



ENE 



HAW 



FPOL 



FIG. 2. Subfamily composition m Pacific Turridae. QLD, Queensland: OKIN, Okinawa; GUM, Guam; ENE, 
Enewetak. HAW. Hawaii; FPOL, French Polynesia. Turr, Turrinae; Crass. Crassispirinae; Drill, Drilliinae; 
Mang. Mangeliinae; Daph, Daphnellinae; Borson, Borsoniinae. 



cies, in contrast to 27°o and 25% respectively 
in New South Wales and New Zealand (Ire- 
dale & McMichael, 1962; Powell, 1979); 31% 
on the tropical west American coast (Keen, 
1971 ); and an average of 7% on islands in the 
Indian Ocean (J. D. Taylor, 1977). Little is 
known of turrid biology in the Pacific, although 
J. D. Taylor (1984, 1986) notes the polycha- 
ete diet of four species on Guam, and refers 
(J. D. Taylor, 1984) three species to the poly- 
chaete feeding guild in a partial food web on a 
fringing reef on Guam. 

The distnbution of species among the is- 
lands is patchy (Table 2). Only 10 species 
(6%) are recorded from all three island 
groups. Enewetak and Hawaii share 19% of 
the 114 species occurring in both island 
groups, and French Polynesia and Hawaii 
share 15% of 123 species. 

From west to east across the Pacific, the 
numbers of turrid species decrease from 180 
species recorded along the Queensland 
coast of Australia (Hedley, 1922) to 76 spe- 
cies in French Polynesia, and from 100 spe- 
cies in the Philippines and Okinawa (Kuroda, 
1960) to about 70 species respectively at 
Enewetak (Kay & Johnson, 1987) and in the 
Hawaiian Islands (Kay, 1979) (Figure 3). The 
decrease in number of turrid species is not 
the same in all subfamilies. The Turrinae 
gradually increase in percentage composition 
across the Pacific from 5% on the Queens- 
land coast to 18% in Hawaii, but are appar- 
ently reduced in importance in French Poly- 



nesia, where only two species are recorded 
(Richard, 1982; Salvat & Rives, 1979). The 
Mangeliinae, representing 57% of the Queens- 
land turrid list decrease to 21% of the Hawai- 
ian list and increase to 39% of the list in 
French Polynesia (Fig. 2). 

Biogeographic Components 

Three biogeographical components are 
identified; 39% of the Pacific island turrids 
also occur in the western Pacific but are not 
reported further west in either the Indo-Ma- 
layan archipelago or the Indian Ocean; 31% 
are recorded only from the Pacific Plate; and 
29% are widespread within the Indo-Pacific, 
many of them found as far west as the coast 
of Natal. Two species, Microdaphne trichodes 
Dall, 1910 (McLean, 1971), and Kermia mac- 
ulosa (Pease, 1860) (Shasky, 1983), also oc- 
cur on the west coast of the Americas; both 
appear to be Indo-Pacific species that have 
crossed the East Pacific Barrier. 

The relative importance of the subfamilies 
differ among the three regional components 
of the fauna; in the Indo-Pacific faunal com- 
ponent, 47% of the turrids are daphnellines; 
and in the western Pacific and Pacific Plate 
components, 45% and 35% of the turrids re- 
spectively are mangelines (Fig. 4). There is a 
conspicuous component of Pacific Plate en- 
demism in all the subfamilies, but it is espe- 
cially noticeable in the Turrinae (56%), all of 
which are endemic to the Hawaiian Islands. 



TURRID FAUNAS OF PACIFIC ISLANDS 



83 



TABLE 2. Distribution records for 25 turrid species recorded from central Pacific Islands. Single island 
occurrences are not included. Xs represent occurrence. 



SPECIES 



ENEWETAK, 
MARSHALL IS. 



HAWAIIAN 
ISLANDS 



FRENCH 
POLYNESIA 



Mitromorpha metula 

(Hinds, 1843) 
Carinapex minutissima 

(Garrett, 1873) 
Iredalea exilis 

(Pease, 1868) 
Daphnella ornata 

(Hinds, 1844) 
Lienardia crassicostata 

(Pease, 1860) 
Daphnella flammea 

(Hinds, 1843) 
Kermia clandestina 

(Deshiayes, 1863) 
Psudodaphnella tincta 

(Reeve, 1846) 
С lav и s pica 

(Reeve, 1843) 
Xenuroturns cingulifera 

(Lamarck, 1822) 
Lophiotoma acuta 

(Perry, 1811) 
Xenuroturns kingae 

(Powell, 1964) 
Turridrupa albofasciata 

(Smith, 1877) 
Tritonoturns amabilis 

(Hinds, 1843) 
Daphnella olyra 

(Reeve, 1845) 
Mitromorpha alphonisana 

(Hervier, 1899) 
Kermia pumila 

(Mighels, 1845) 
Clav us exilis 

(Pease, 1868) 
Lienardia lutea 

(Pease, 1860) 
Lienardia m ig he I si 

(Iredale, 1917) 



xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 



xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 



xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 



xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 



xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 
xxxxxxxxx 



Guam, in the southern Marianas Islands, 
and Okinawa, both of which are on the Phil- 
ippine Plate, have turrid faunas that are sim- 
ilar in species composition and habit to those 
of the central Pacific islands (Fig. 2). As with 
the central Pacific island turnds, there is a 
strong Daphnellinae and Mangeliinae compo- 
nent in these faunas. 

Fossil Record 

There is no clear direction to the fossil 
record of Pacific island turrids, unless it is that 
about 50% of the genera represented in the 



islands of the central Pacific apparently lack a 
fossil record. There are Miocene and Plio- 
cene records for Mitromorpha (Borsoniinae) 
and Daphnella and Philbertia (Daphnellinae) 
in Okinawa and Europe respectively, for 
Inquisitor and Clavus (Drilliinae) in the Plio- 
cene of Java and Miocene of Borneo respec- 
tively, for Anacithara, Etrema, Eucithara, and 
Lienardia (Mangeliinae), and Miocene and 
Pliocene records for Gemmula, Lophiotoma, 
Turns (Turrinae) and Turridrupa (Claviinae) 
(Powell, 1966; Robba, 1987; Shuto, 1984). 
Among the genera and subgenera for which a 
fossil record is apparently lacking are: Eucy- 



84 



KAY 



180 



160 



(О 

ф 

о 140 

ф 

а 

со 

о 

^ 120 

0) 

п 
Е 

^ 100 



80 



60 



QLD 



LIFU 



О Above Equator 
# Below Equator 



OKIN 




FPOL 



ENE 




FIG. 3. Numbers of turrid species from west to east across the Pacific. Locality abbreviations as in Figure 
2. 





50 




45 




40 


^—^ 

# 


35 


(П 


30 


0) 




1 


25 


«n 




ь 


20 


D 




CO 


15 




10 




5 









Daph 
Ш Borson 




IP 



WP 



РАС 



FIG. 4. Subfamily composition of Indo-Pacific species. Western Pacific species, and Pacific Plate species. 
IP, Indo-Pacific, WP, Western Pacific, PAG, Pacific Plate. Subfamily abbreviations as in Figure 2. 



clotoma. Kermia. Microdaphne. Pseudodaph- 
nella. Tritonoturhs. Iredalea. Macteola. and 
Paramontana. 



The evolution of the Indo-Pacific nnolluscan 
fauna can be traced to the ancient Tethys 
seaway which, from Triassic to Miocene time, 



TURRID FAUNAS OF PACIFIC ISLANDS 



85 



connected the Mediterranean and Indo- 
Pacific across what is now the Middle East 
and Pakistan (Kay, in press). The seaway 
was closed in Early Miocene (Adams, 1981), 
and the faunas of the Mediterranean and the 
Indo-Pacific evolved separately. Among the 
earliest turrids of Pakistan-India are Drilliinae 
in the Palaeocene Cardita beaumonti beds of 
the Upper Ranikot of Pakistan. In Indonesia, 
the earliest recognizable molluscan faunas 
are middle Miocene (Shuto, 1976), and there 
are representatives of the Drilliinae, Turrinae, 
Borsoniinae in that area. In the Pacific, sev- 
eral Miocene turrids (Daphnella. Clavus, In- 
quisitor, Etrema. Anacithara. Eucithara, Lie- 
nardia, Gemmula, Lophiotoma, and Turris) 
are recorded from Fiji and Okinawa. Eu- 
clathurella, Gemmula. and Lophiotoma are all 
recorded from the Pliocene of Fiji. The fur- 
thest east that fossil turrids are known is in the 
Marshall Islands, where Ladd (1982) de- 
scribed Eucithara marshellensis from the Mi- 
ocene of Bikini and Enewetak. 



DISCUSSION 

Pacific island turrid faunas are character- 
ized by a suite of characters consonant with 
insular coral reef habitats separated one from 
another by great distances. The relatively 
high proportions of mangelines and daphnel- 
lines result in faunas with small, sturdy shells 
and with protoconchs indicative of long larval 
life. Indeed, Kilburn (1988) has suggested 
that the shell form of short, sturdy shells with 
reinforced labial lips may have evolved as an 
adaptation to a reef or under-rock existence, 
as opposed to shells with a produced sipho- 
nal canal and non-reinforced labial extremity 
(the 'turrid facies') of the predominantly sand- 
dwelling Turrinae and Cochlespihnae. The 
Pacific island turrid assemblages, with their 
prominent daphnelline and mangeline com- 
ponents, with colorful shells and long larval 
lives, and comprising less than 15% of the 
predatory gastropods, contrast with a Carib- 
bean island assemblage described by Maes 
(1983), which is rich in drillines with dark 
shells, most of which apparently have direct 
development. 

The pattern of turrid distribution among Pa- 
cific islands follows a generally recognized 
pattern of a marine fauna of the Indo-Pacific 
(Kay, 1 980): decreasing diversity west to east 
in the Pacific Ocean (Kay, 1980; Springer, 
1982); patchy distnbution (Kay, 1980; 1984); 



disproportionate representation in certain 
groups; and a recognizable component of 
species that are endemic to the Pacific Plate 
(Kay, 1980; Springer, 1982). On the Pacific 
Plate, the pattern of endemism follows that of 
other mollusks, with a group of species that is 
widespread, and with others of the turrids en- 
demic to two foci, the Hawaiian Islands where 
60% of the turrids are reported as endemic, 
and French Polynesia with 9%. No single is- 
land endemics as distinguished by Springer 
(1972) have been identified. 

Although in broad outline the Pacific island 
turrid faunas fit the pattern of history and dis- 
tribution of other marine mollusks, three as- 
pects of the current review of Pacific island 
turrid faunas are vulnerable to the criticism of 
insufficient and biased collection; (1) the ap- 
parent concentration of species in the west- 
ern Pacific; (2) the records of patchy distribu- 
tion; and (3) the apparent lack of fossil record. 

The prominent western Pacific element in 
the distributional pattern may be an artifact of 
collecting simply because of the enormous 
numbers of species described from the Phil- 
ippines, Loyalty Islands, New Caledonia, and 
Queensland by Reeve, Hervier, Melvill and 
Standen, Hedley, and others. As the taxon- 
omy of the several hundred species is worked 
out, many of the species may be shown to be 
invalid, and new collecting records will possi- 
bly extend the presently known ranges. The 
apparent patchy distribution of turrid species 
among the islands may also be misleading 
and with further collection such anomalies as 
the virtual absence of Turrinae in Tahiti may 
be rectified. The fossil history of Pacific island 
turrids is similarly subject to criticism in that so 
few fossil are actually known from Pacific is- 
lands. 



ACKNOWLEDGMENTS 

I am particularly grateful to the many cura- 
tors and curatorial assistants who have so 
graciously allowed me to work with the turrid 
collections under their care: P. Bouchet 
(Paris), G. Oliver (Cardiff), W.F. Ponder (Syd- 
ney), the late Joseph Rosewater (Washing- 
ton, D. C), G. Davis (Philadelphia), and K. 
Boss (Harvard). And again I must thank the 
staff in the Mollusca Section, BM(NH), for 
their patience and hospitality over the years 
as I have worked in the superb collections in 
that institution. 



86 



KAY 



LITERATURE CITED 

ADAMS, С. G.. 1981. Speciation. phylogenesis, 
tectonism. climate and eustacy: factors in the ev- 
olution of Cenozoic larger foraminiferal bioprov- 
inces. In L. R. M. COCKS, ed., The evolving 
Earth. British Museum (Natural History). London, 
and Cambridge University Press, pp. 255-289. 

BOUCHET, P. H.. 1990. Turrid genera and mode of 
development: the use and abuse of protoconch 
morphology. Malacologia. 32: 69-77. 

BOUCHET, P. H. & A. WAREN, 1980, Revision of 
the north-east Atlantic bathyal and abyssal Tur- 
ridae (Mollusca, Gastropoda). Journal of Mollus- 
cen Studies. Supplement 8:1-119. 

DAUTZENBERG, P. & J. L. BOUGE. 1933. Les 
mollusques testacés marins des établissements 
Français de l'Océanie. Journal de Conchyliolo- 
gie. 77:41-108: 145-326; 351-469. 

HEDLEY, C. 1922, A revision of the Australian Tur- 
ridae. Records of the Australian Museum. 13: 
213-259. 

HERVIER, R. P. J.. 1896, Descriptions d'espèces 
nouvelles de mollusques, provenant de I Archipel 
de la Nouvelle-Calédonie. Journal de Conchyli- 
ologie. 44:51-96. 

HERVIER, R. P. J.. 1897, Descriptions d'espèces 
nouvelles de mollusques, provenant de l'Archipel 
de la Nouvelle-Calédonie. Journal de Conchyli- 
ologie. 45:89-121; 165-191. 

HERVIER, R. P. J., 1898, Descnptions d'espèces 
nouvelles de mollusques, provenant de l'Archipel 
de la Nouvelle-Calédonie. Journal de Conchyli- 
ologie. 45:225-248. 

IREDALE, T & D. F. MCMICHAEL, 1962, A refer- 
ence list of the marine Mollusca of New South 
Wales. Memoirs of the Australian Museum. 1 1 : 
1-109. 

KAY, E. A., 1979. Hawaiian marine shells. B. P. 
Bishop Museum Special Publication. 64(4): 
1-653. 

KAY, E. A., 1980, Little worlds of the Pacific. An 
essay on Pacific Basin biogeography. Harold L. 
Lyon Arboretum Lecture No. 9. University of Ha- 
waii. 40 pp. 

KAY, E. A., 1984, Patterns of speciation in the Indo- 
west Pacific. In P. H. RAVEN, F. J. RADOVSKY 
& S. H. SOHMER, eds.. Biogeography of the 
tropical Pacific. Association of Systematics Col- 
lections and the Bernice P . Bishop Museum, pp. 
15-31. 

KAY, E. A., In Press. Biogeography and Cenozoic 
history of the Cypraeidae of the Pacific. Bulletin 
of Manne Science. 

KAY, E. A. & S. JOHNSON, 1987, Mollusca of 
Enewetak Atoll. In D. M. DEVANEY. E. S. 
REESE, B. L. BURCH, & P. HELFRICH, The Nat- 
ural History of Enewetak Atoll. Vol. 1 U. S. Dept. 
of Energy, pp. 105-148. 

KEEN, A. M., 1971, Sea shells of tropical west 
America. Stanford University Press. 1064 pp. 

KILBURN. R. N.. 1983, Turridae (Mollusca: Gas- 
tropoda) of southern Africa and Mozambique. 



Part 1 . Subfamily Turrinae. Annals of the Natal 
Museum. 25:549-585. 

KILBURN, R. N.. 1985, Turndae (Mollusca; Gas- 
tropoda) of southern Africa and Mozambique. 
Part 2. Subfamily Clavatulinae. Annals of the Na- 
tal Museum. 26:417-470. 

KILBURN. R. N.. 1986. Turridae (Mollusca: Gas- 
tropoda) of southern Africa and Mozambique. 
Part 3. Subfamily Borsoniinae. Annals of the Na- 
tal Museum. 27:633-720. 

KILBURN, R. N.. 1988, Turndae (Mollusca: Gas- 
tropoda) of southern Africa and Mozambique. 
Part 4. Subfamilies Dnlliinae, Crassispirinae and 
Strictispinnae. Annals of the Natal Museum. 29; 
167-320. 

KU RODA, T., 1960, A catalogue of molluscan 
fauna of the Okinawa islands. Privately printed, 
Japan. 

LADD, H. S., 1982, Cenozoic fossil mollusks from 
western Pacific islands; gastropods (Eulimidae 
and Volutidae through Terebridae). U. S. Geolog- 
ical Survey Professional Paper. 1 1 71 ;1 -1 00. 

MAES, V. O., 1983, Observations on the systemat- 
ics and biology of a turrid gastropod assemblage 
in the British Virgin Islands. Bulletin of Manne 
Science. 33:305-335. 

MCLEAN, J. H., 1971, A revised classification of 
the family Turridae. with the proposal of new sub- 
families, genera and subgenera from the eastern 
Pacific. Veliger 14:114-130. 

MELVILL, J. С & R. STANDEN, 1895-1897, Notes 
on a collection of shells from Lifu and Uvea, Loy- 
alty Islands, formed by Rev. James and Mrs. 
Hadfield, with list of species. Journal of Conchol- 
оду. 8:84-132; 273-315. 

POWELL, A. W. В.. 1942, The New Zealand Re- 
cent and fossil Mollusca of the family Turridae. 
Bulletin of the Auckland Institute and Museum 
2:1-192. 

POWELL, A. W. В.. 1966. The molluscan families 
Speightiidae and Turridae. Bulletin of the Auck- 
land Institute and Museum. 5:1-184. 

POWELL, A. W. В., 1964-1967, The family Tur- 
ndae in the Indo-Pacific. Part 1. The subfamily 
Turrinae. Indo-Pacific Mollusca, 1 ;227-345; 
409-443. 

POWELL, A. W. В., 1979. New Zealand Mollusca. 
Collins. 500 pp. 

REHDER, H. A. & H. S. LADD. 1973, Deep and 
shallow-water mollusks from the Central Pacific. 
Science Reports Tohoku University. 2nd Ser. 6; 
37-49. 

RICHARD, G.. 1982, Mollusques lagunaires et rec- 
ifaux de Polynésie française. Thesis for the Doc- 
teure of Science Naturelle, Université Pierre et 
Mane Curie, Pans. 

ROBBA, E. 1987. The final occlusion of Tethys: its 
bearing on Mediterranean benthic Mollusca. In 
MCKENZIE, KG., ed.. Proceedings of the Inter- 
national Symposium on Shallow Tethys. A.A. 
Balkema, Rotterdam & Boston, pp. 405-426. 

SALVAT, B. & С RIVES, 1975, Coquillages de 



TURRID FAUNAS OF PACIFIC ISLANDS 



87 



Polynésie. Editions du Pacifique Papeete; 391 
pp. 

SHASKY, D. R., 1983, New records of Indo-Pacific 
Mollusca from Cocos Island, Costa Rica. Nau- 
tilus. 97:144-145. 

SHIMEK, R. L., 1986, The biology of thie northieast- 
ern Pacific Turridae. V. Demersal development, 
synchironous settlement and other aspects of the 
larval biology of Oenopota levidensis. Interna- 
tional Journal of Invertebrate Reproduction and 
Development. 10:313-333. 

SHUTO, T., 1976, Correlation of Neogene forma- 
tions of Southeast and South Asia by means of 
molluscan faunas. Proceedings of tfie First Inter- 
national Congress on Pacific Neogene Stratigra- 
phy Tokyo. Science Council of Japan Geological 
Society of Japan, pp. 133-144. 

SHUTO, T., 1984, A revision of the Burmese Ter- 
tiary turrids. Memoirs of tfie Faculty of Sciences, 



Kyushiu University Series D (Geology). 25:1 15- 
157. 

SPRINGER, V. G., 1982, Pacific Plate biogeogra- 
phy with special reference to shorefishes. Smith- 
sonian Contributions to Zoology. 367:1-182. 

TAYLOR, J. В., 1975, Planktonic prosobranch 
veligers of Kaneohe Bay. Ph.D. Dissertation. Uni- 
versity of Hawaii, Honolulu. 

TAYLOR, J. D., 1977, Food and habitats of preda- 
tory gastropods on coral reefs. Reports Under- 
water Association 2:1 7-34. 

TAYLOR, J. D., 1984, A partial food web involving 
predatory gastropods on a Pacific fringing reef. 
Journal of Experimental (Marine Biology and 
Ecology 74:273-290. 

TAYLOR, J. D., 1 986, Diets of sand-living predatory 
gastropods at Piti Bay, Guam. Asian (Marine Bi- 
ology 3:47-58. 

Revised Ms. accepted 21 June 1990 



MALACOLOGIA, 1990, 32(1): 89-106 

ULTRASTRUCTURAL CHANGES IN THE DIGESTIVE SYSTEM OF 

DEROGERAS RETICULATUM (MOLLUSCA; GASTROPODA) 

INDUCED BY LETHAL AND SUBLETHAL CONCENTRATIONS OF 

THE CARBAMATE MOLLUSCICIDE CLOETHOCARB 

Rita Triebskorn' & С Künast^ 

ABSTRACT 

Specimens of the grey garden slug, Dereceras reticulatum. were fed lethal (2%, 1%, 0.5%, 
0.1%) or sublethal concentrations (0.01%, 0.001%) of the carbamate molluscicide Cloethocarb 
(BASF) as either pellets or wheat-germ agar. To investigate the influence of the chemical on the 
ultrastructure of the ceils in the digestive tract, samples of oesophagus, crop, stomach, intestine 
and hepatopancreas were taken at six time intervals. Reactions of lethal intoxication (e.g. 
elongation of cells, damage to nuclei and mitochondria, destruction of membranes) were dis- 
tinguished from those appearing most intensely after sublethal intoxication (e.g. reactions of the 
endoplasmic reticulum, Golgi apparatus, mucous cells) and finally from features that appear 
after both lethal and sublethal poisoning (e.g. reduction of storage products). 

It could be shown that higher concentrations of the pesticide do not necessarily produce 
stronger effects at the ultrastructural level. Because it is effective in elucidating cellular injury 
following lethal and sublethal intoxication, electron microscopy is a sensitive method for diag- 
nosing the animals' response to stress. 

Key words: molluscicide, carbamate, ultrastructure, digestive tract. Gastropoda, Dereceras 
reticulatum. 



INTRODUCTION 

The increasing importance of slugs as field 
and garden pests (Martin & Kelly, 1986), de- 
mands continued efforts to identify and pro- 
duce more effective and selective mollusci- 
cides. For commercial reasons, however, 
most molluscicides have been detected only 
incidental to screening programmes for the 
development of insecticides (Henderson & 
Parker, 1986). It is highly probable, therefore, 
that most potential molluscicides will also be 
insecticides having considerable side-effects 
on useful animals, as soil arthropods and an- 
nelids. 

Little is known about the mollusc-specific 
effects of even the most widely used commer- 
cial molluscicides, metaldehyde and methio- 
carb, owing to the fact that most pesticide re- 
search is restricted to LD^o tests, which 
provide information about lethal or non-lethal 
effects of the substances tested. Such acute 
toxicity tests have demonstrated the advan- 
tages of carbamate or metaldehyde applica- 
tion under specific conditions (Kemp & New- 
ell, 1985; Glen & Orsman, 1986; Prystupa et 
al., 1987), and have also revealed optimal 



molluscicida! concentrations for different ac- 
tive substances (Wright & Williams, 1980). 
Such tests, however, do not yield any further 
information about either the targets for the 
molluscicides or the mollusc-specific mecha- 
nisms induced in the slugs' bodies. Because 
this knowledge is essential for the develop- 
ment of new, more selective substances, 
much basic research is required. 

Basic biochemical and physiological stud- 
ies, for instance, have shown the inhibitory 
effect of methiocarb on cholinesterases 
(Pessah & Sokolove, 1983; Young & Wilkins, 
1989) and the influence of metaldehyde on 
feeding motoneurons in the buccal ganglia 
(Mills et al., 1989). 

In the present study, the influence of lethal 
(>0.1%) and sublethal (<0.01%) oral doses 
(OD) of the carbamate Cloethocarb on the 
cells of the digestive system of Deroceras re- 
ticulatum was investigated by electron mi- 
croscopy. Sublethal concentrations are de- 
fined as those not leading to any mortality 
during the test (up to 30 h). 

Diagnosis at the cellular level was chosen 
for its high sensitivity (Braunbeck, 1989), and 
for its providing information about reactions to 



'Zoologisches Institut I, Universität Heidelberg, Im Neuenheimer Feld 230, D-6900 Heidelberg, Germany 
^BASF Aktiengesellschaft, Landwirtschaftliche Versuchsstation, D-6703 Limburgerhof, Germany 



89 



90 



TRIEBSKORN & KÜNAST 



TABLE 1. Amount of ingested food containing various concentrations of Cloethocarb and calculated 
values for absolute quantity of active substance ingested g wet weight. 



active substance 



Food 

ingested 

(mg g wet weight) 



Active substance 

ingested 
(ixg g wet weight) 



2% Cloethocarb (pellet) 
2% Cloethocarb (agar) 
0.1% Cloethocarb 
0.01% Cloethocarb 
0.001% Cloethocarb 
Control 



70.0±50.5 

69.0±25.2 

83.9 144.0 

105.2147.5 

105.3129.9 

102.8138.3 



1400 

1380 
83.9 
10.5 
1.05 



both lethal and sublethal intoxication. Thus, it 
should be possible to distinguish primary car- 
bamate-specific cell reactions from symptoms 
resulting from cell death, and to differentiate 
irreversible damage from such cellular injury 
that might be compensated by the animal's 
detoxification mechanisms. 



MATERIALS AND METHODS 

The carbamate molluscicide Cloethocarb 
(BASF) was given to laboratory-reared Dero- 
ceras reticulatum by a single feeding of either 
pellets produced by BASF containing 2% of 
the active substance, phenol-2-(2-chloro-1- 
methoxyethoxy)-methylcarbamate, or wheat- 
germ agar containing 2%, 1%, 0.5%, 0.1%, 
0.01%, or 0.001% of this toxin. The amount of 
food ingested was determined by weighing the 
treated food, which was dried before and after 
feeding. Finally, the quantity of active sub- 
stance taken up with the food was calculated 
(Table 1). Doses in the 0.02% and 0.001% 
formulation proved to be sublethal. 

Behavioral and macroscopic changes of 
the animals were recorded during the first 
hour after the beginning of feeding. For deter- 
mination of cellular reactions, three animals 
feeding at each of the molluscicide concen- 
trations were dissected after 30 min, 1 h, 3 h, 
5 h, 24 h and 30 h. For primary fixation of 
excised tissue, a 2% glutaraldehyde solution 
in cacodylate buffer (0.01 M, pH 7.4) was in- 
jected into the body cavity. The oesophagus, 
crop, stomach, intestine and hepatopancreas 
were isolated under fixative and fixed for 2 h 
at 4 C. The tissues were then rinsed in ca- 
codylate buffer and postfixed in 1% osmium 
ferrocyanide (Karnovsky, 1971) for 2 h at 4 С. 
After rinsing in cacodylate and maléate buffer 
(0.05 M, pH 5.2), the specimens were stained 



en bloc overnight in 1% uranyl acetate dis- 
solved in maléate buffer (0.05 M, pH 5.2) at 
4 C. The samples were rinsed in maléate 
buffer, dehydrated and embedded in Spurr's 
medium (Spurr, 1969). Semithin and ultrathin 
sections were cut on a Reichert ultramicro- 
tome. Semithin sections were stained with 
methylene blue-azur (Richardson et al., 1960) 
and used for light microscope overviews. Ul- 
trathin sections were counterstained with lead 
citrate for 30 sec. The tissues were examined 
in a Zeiss EM 9. The following cell types of the 
digestive system were investigated; 

Oesophagus: storage cells, secretory cells 
of an eccnne type, secretory cells of a holo- 
crine type (mucous cells); 

Crop; storage cells; 

Stomach; storage cells, secretory cells of a 
holocnne type (mucous cells); 

Intestine; storage cells, secretory cells of 
an eccrine type, secretory cells of a holocrine 
type (mucous cells); 

Hepatopancreas; digestive cells, crypt cells, 
excretory cells. The muscle and nerve layers 
underlying the epithelia were also studied. 



RESULTS 
Macroscopic Observations 

Table 2 shows the mortality after Cloetho- 
carb application. More animals were killed af- 
ter the application of low lethal concentrations 
(0.1-1%) than after that of 2%. After ingestion 
of the 2% agar, however, animals died sooner 
than after that of all other concentrations. Fur- 
thermore, the mortality after 2% agar was 
higher than that after 2% pellets. 

The macroscopically visible reactions of 
animals to intoxication with Cloethocarb cor- 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



91 



TABLE 2. Mortality (absolute number of dead animals) after application of Cloethocarb. For each 
concentration, 20 animals were tested. 



Concentration 










Time 








Cloethocarb 


0.5 h 


1 h 


3h 


5h 


lOh 


16 h 


24 h 


30 h 


Control 


























0.001% 


























0.01% 


























0.1% 




















3 


10 


0.5% 

















2 


4 


14 


1% 




















3 


16 


2% agar 











1 


2 


2 


2 


8 


2% pellet 




















1 


4 



respond to typical symptoms of carbamate in- 
toxication (Godan, 1979): ten minutes after 
taking up pellets or wheat-germ agar contain- 
ing 2%, 1%, 0.5% or 0.1% of the poison, the 
animals show muscle convulsions that be- 
come more intense in the following 30 min- 
utes. During this period, they lose large 
amounts of a lucent mucus. Whereas during 
the first 30 minutes after intoxication, the an- 
imals still move actively, after 1 h, the animals 
are alternately active and immobile. The an- 
terior part of the body begins to swell while 
the posterior part flattens. After a period 
longer than 16 h, most of the animals lie mo- 
tionless on their sides and only occasionally 
move. Reactions to 0.1%, 0.5% and 1% OD 
are as intense as those to 2%, whereas be- 
havioral reactions to both sublethal concen- 
trations are absent. 

Electron Microscope Investigations 

Table 3 summarizes the most important re- 
actions of investigated components in cells of 
the digestive system of Deroceras reticulatum 
after intoxication with different concentrations 
of Cloethocarb. 

Thirty minutes after the onset of poisoning, 
reactions were confined to single cells and 
especially to the anterior part of the digestive 
tract (oesophagus and crop). During the fol- 
lowing hours, reactions spread over the epi- 
thelia, then appeared in the cells of the pos- 
terior part of the digestive system (stomach, 
intestine and hepatopancreas), with lapse of 
time corresponding to the rate of transport of 
toxic feedstuff by the alimentary canal or by 
the hemolymph (Triebskorn et al., 1990). 

In general, cells of the hepatopancreas are 
more strongly damaged than are those of the 
digestive tract. In the cells of the crop, the 
reactions are less severe than those in epi- 



thelia characterized by high percentages of 
mucous cells. 

In most cases, there are no differences in 
the cellular responses to 2%, 1%, 0.5% or 
0.1% OD. Such differences as exist appear 
less intense after 2% than after 0.1% (Table 
3: indented arrows). Cellular reactions of 
slugs exposed to pellets containing 2% of the 
carbamate substance are similar to those of 
animals fed the 2% treated agar. 

Cellular Outline 

In control animals, most of the epithelial 
cells of the digestive tract are columnar (Figs. 
1, 2). Their apical surfaces are characterized 
by microvilli (storage and secretory cells of 
oesophagus and crop, digestive cells of the 
midgut gland; Figs. 6, 8) or by cilia and mi- 
crovilli (storage and secretory cells of oesoph- 
agus, stomach and intestine; Fig. 9). Infold- 
ings of the basal surfaces of these cells are 
small or absent, and the basal membrane is 
very thin (Fig. 12). 

The mucus-producing cells of oesophagus, 
stomach and intestine are pyriform (Fig. 2). 
They bear small microvilli. Immature mucous 
cells do not reach the lumen. 

In the hepatopancreas, two other cell types 
can be distinguished in addition to the diges- 
tive cells; conical crypt cells (Fig. 3), with a 
microvillous border and a prominent basal 
labyrinth, and the bellied excretory cells, char- 
acterized by large excretory vacuoles and 
long microvilli. 

The molluscicide produces cytopathologi- 
cal changes in the general outline of cells and 
in their apical and basal surfaces. 

Most of the cells in the digestive system 
change their typical cellular outline after 5 h, 
and more intensely 24 h and 30 h after the 
application of food containing between 0.1% 
and 2% poison. The columnar cells of the di- 



92 



TRIEBSKORN & KÜNAST 



TABLE 3. The most striking reactions of cellular components after lethal and sublethal intoxication with 
Cloethocarb. 





CLOETHOCARB CONCENTRATIONS 




2% 


1% 


0.5% 


0.1% 


0.01% 


0.001% 
















CELL 
OUTLINE 


ÎI 


Stretching of the cells м 

Irregular cell shape м 


— 


— 




1 ! ' ; 1 1 


CELL APEX 


►- 


Irre 

Rec 

Sur 
Sur 


guiar shape of microvilli ^ 

Juction of microvilli ^ 

face blebs M 

face coat ^ 


— 










■ I ! 1 




CELL BASE 


►- 
►- 
►- 
►- 


Basal cell extensions ^ 

Dilation of basal labyrinth ^ 

Gaps 4 

Thickening of basal membrane ^ 


— 




1 1 1 






b. 




Crystalline inclusions 


— 






^ 


— 


NUCLEI 


►- 


Karyolysis Ц 




III 1 II 


IVIITO- 
CHONDRIA 


►- 
►- 

►- 


Dis 

Sw 

Rec 

Ruf 


placement ^ 


— — 


Juction of cristae ^ 

Dture of membranes ^ 




III, II 


ENDOPLASMIC 
RETICULUM 


►— 


^ Dec 

► Pro 

Me 
► Tub 


granulation, dilation of gER 

iteration, vesiculation of ER 

mbrane whorls 


< 

< 

< 












Destruction, rupture of membranes 






1 1 1 [ ! 1 




►— 

к 


Irregular arrangement of cister 








■-ti->rmQ ^ 


GOLGI 






^ 




APPARATUS 


Dilation of trans-face cisternae 
Destruction of membranes 






1 1 1 




VACUOLAR 


►- 
►- 


Increased fusion rate 

Increased membrane lability ^ 

^ Increased production of large mucous vacuok 


< 


— 






SYSTEM 


?s 


'4 




III 1 II 


STORAGE 
PRODUCTS 


►- 
►- 


Decrease of storage products - 

Increase of electron-dense ves 




^ 


ides 


< 




III 1 II 


MUSCLE 
TISSUE 


►- 


Muscle envelopes without filaments ^ 

Fragmentation M 

Irregular orientation of filaments < 


— 




1 1 1 1 




NERVE 
TISSUE 


►- 


Increased number of neurosecretory vesicles M 


— 




; ; 1 1 





EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



93 




•^Л^^ - 30 /xm 



30 ¿um 



FIG. 1 . Crop (control). Storage (sc) and mucous cells (muc) in epithelium of crop. Storage cells contain large 

amounts of lipid (li) and bear microvilli (mv). LM. 

FIG. 2. Stomach (control). Storage (sc) and mucous cells (muc) in epithelium of stomach, c: cilia. LM. 

FIG. 3. Hepatopancreas (control). Light-microscopial overview of crypt cell characterized by round nucleus 

(n) with prominent nucleolus (nu) and numerous vesicles (v). 

FIG. 4. Crop (2% Cloethocarb, 5h). Epithelial cells elongated (arrows), nuclei condensed, nt = nerve tissue. 

FIG. 5. Stomach (2% Cloethocarb, 5h). Gap (arrow) between muscle tissue (mt) and epithelial cells. 



gestive tract become greatly elongated, espe- 
cially after 0.1% OD, and often gaps open 
between the epithelial cells and the underly- 
ing muscle and nerve tissue (Figs. 4, 5). The 
cells of the hepatopancreas are already irreg- 
ularly shaped after 1 h (Fig. 7). Sublethal con- 
centrations (0.01%, 0.001%) do not affect cell 
shape. 

The most striking reactions of the cell api- 
ces are reduction of microvilli, formation of 
surface blebs and production of a hyaline sur- 
face coat. Thirty minutes after the application 
of a lethal concentration, reduced microvilli 
and apical cytoplasmic protrusions (blebs) al- 
ready can be observed in the columnar cells 



of the oesophagus and crop. Both reactions 
occur in all epithelia of the digestive system 
after 5, 24 and 30 h (Fig. 1 0), most severely in 
the cells of the intestine 24 h after ingestion of 
0.1% toxin and in the hepatopancreas as 
soon as 1 h after ingestion of any lethal con- 
centration. Food containing 0.01% poison 
leads to formation of small blebs and irregu- 
larly shaped microvilli in isolated cells. Inges- 
tion of food containing 0.001% poison pro- 
duced no evident reaction. 

A further phenomenon appearing after both 
sublethal and lethal intoxication is a surface 
coat consisting of a hyaline material overlying 
the microvilli or the cilia, or both (Fig. 1 1 ). This 



94 



^sJ^^V.-,, ©^>. 



TRIEBSKORN & KÜNAST 



i 



sb 



<í0 



mv 



*^i 
"У.Г. с - • . 4 -Г 



ЛИ 



•.-^'■0 



mv 



*■• 



со 



1 /xm 






mvr 






filote - 

bm*' 




1 /xm 0.5 дт 



.1 .'И 



' ■••(■ 



ñS^ 



'íí-^ 



^ 



i n 



1 /¿m 






/Ltm 




1 /xm V 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



95 



surface coat is less electron-dense in the di- 
gestive tract than in the hepatopancreas. It is 
already present in the anterior parts of the 
tract 1 h after administration of the poison and 
is very prominent in all regions of the diges- 
tive system after 5 h. 

Application of 0.001% of the molluscicidal 
agent does not produce a detectable surface 
coat. 

The most striking changes in the basal sur- 
face of the cell are development of basal celt 
extensions, thickening of the basement mem- 
brane, and development of gaps between the 
epithelium and the basement membrane. 

After 30 minutes, small basal infoldings al- 
ready have formed in the oesophagus, partic- 
ularly after 0.1% OD. After 5 and 24 h, how- 
ever, bases of the cells of the digestive tract, 
but not of the hepatopancreas, show consid- 
erable extensions (Figs. 13, 14). These ex- 
tensions are most prominent in the region of 
the stomach and intestine. 

In the hepatopancreas, the basal labyhnth 
of the crypt cells is extended. Thirty hours after 
the application of 0.01% OD the basal infold- 
ings are comparable to those in the oesoph- 
agus 30 minutes after lethal intoxication. 

From 24 h to 30 h after ingestion of lethal 
concentrations, gaps form between the basal 
parts of the epithelial cells and the underlying 
connective, muscle and nerve tissues (Fig. 
15). 

A further reaction detectable after the in- 
gestion of sublethal concentrations is the 



thickening and increased electron-density of 
the basement membrane (Figs. 14, 16), 
which appears as early as 1 h after the inges- 
tion of food containing >0.01% Cloethocarb. 

Nuclei 

In the storage and secretory cells of the 
digestive tract and in the digestive cells of the 
hepatopancreas, the nuclei are ovoid (Fig. 
17). In the crypt cells of the hepatopancreas 
and in the mucuous cells, however, they are 
roundish (Figs. 2, 3). The nuclei of the crypt 
cells are rich in heterochromatin and have a 
prominent, round nucleolus (Fig. 3). 

The most typical cytopathological changes 
in the nuclei are lightening of the karyoplasm, 
reduction of the heterochromatin, dilation of 
the nuclear envelope and formation of crys- 
talline inclusions. 

Thirty minutes after the onset of intoxica- 
tion, the nuclear envelope is locally dilated 
and the karyoplasm becomes more electron- 
lucent in individual cells. After 5 h these phe- 
nomena occur in many cells (Fig. 18). The 
nuclei show the effects of lytic processes. In 
the crypt cells the nucleoli assume irregular 
shapes. 

The most severe nuclear damage occurs in 
the posterior part of the digestive tract and in 
the cells of the hepatopancreas. Twenty-four 
hours after ingestion of food containing 0.1% 
or 0.01% of the molluscicide, large crystalline 
inclusions form in the nuclei, especially in the 
region of the stomach and the intestine (Fig. 



FIG. 6. Hepatopancreas (control). Digestive ceils, characterized by rnicrovillous border (mv) and system of 
digestive vacuoles (dv). 

FIG. 7. Hepatopancreas (0.1% Cioetiiocarb, 1ti). Irregularly shaped digestive cells. Digestive vacuoles (dv) 
contain electron-dense material, mv: microvilli. 

FIG. 8. Crop (control). Apex of a storage cell bearing microvilli (mv). Beneatti microvillous border mitochon- 
dria (mi) visible. 

FIG. 9. Stomach (control). Apices of two storage cells bearing microvilli (mv) and cilia (c). 
FIG. 10. Oesophagus (2% Cloethocarb, 5h), Apex of storage cell with reduced microvilli (mv) and surface 
blebs (sb). Mitochondria (mi) severely damaged. 

FIG. 11. Oesophagus (2% Cloethocarb, 1h). Storage cells with coat (со) overlying microvilli (mv). 
FIG. 12. Crop (control). Basal part of storage cell with thin basal membrane (bm). 
FIG. 13. Stomach (2% Cloethocarb, 5h). Basal parts of mucous (muc) and storage cells (sc) with basal cell 
extensions (arrows). 

FIG. 14. Crop (2% Cloethocarb, 24h). Storage cell with basal cell extensions (long arrows), electron-dense 
cytoplasm (cyt) and thickened basal membrane (bm, short arrows). 

FIG. 15. Crop (2% Cloethocarb, 30h). Gap between epithelium and basal membrane (bm, short arrows). 
FIG. 16. Intestine (0.1% Cloethocarb, 5h). Thickening of basal membrane (bm). 
FIG. 17. Oesophagus (control). Nucleus (n) of storage cell. 

FIG. 18. Stomach (2% Cloethocarb, 5h). Nuclei (n) with envelope dilated (arrows), karyoplasm lightened and 
heterochromatin reduced. 

FIG. 19. Stomach (0.1% Cloethocarb, 5h). Nucleus (n) with crystalline inclusion (ci). 
FIG. 20. Stomach (0.1% Cloethocarb, 5h). Crystalline inclusion (ci) filling greater part of karyoplasm. 



96 



TRIEBSKORN & KÜNAST 



19). In some cases, these crystals occupy an 
appreciable part of the nucleus (Fig. 20). 

A molluscicide concentration of 0.001% 
failed to induce any reaction in the nuclei. 

Mitochondria 

The storage cells have a layer of mitochon- 
dria beneath the microvillous border (Figs. 8, 
9), whereas in the other cells, these or- 
ganelles are irregularly dispersed throughout 
the cytoplasm. 

Reduction of cristae and swelling are the 
most important reactions of mitochondria to 
Cloethocarb intoxication. One hour after le- 
thal intoxication, the regular arrangement of 
mitochondria beneath the microvillous border 
begins to be disturbed. The organelles swell, 
the cristae become reduced (Figs. 21, 22), 
and after 5 h, the outer membranes rupture 
(Fig. 23). Damage to mitochondna is most se- 
vere in the stomach and intestine. 

After sublethal intoxication, changes in the 
mitochondria could not be observed. 

Endomembrane System 

Large amounts of granular endoplasmic re- 
ticulum (ER) occur in the secretory cells of the 
oesophagus and intestine, in the mucous 
cells of oesophagus, crop, stomach and in- 
testine (Fig. 24), and in the crypt cells of the 
hepatopancreas (Fig. 25). The cisternae are 
almost parallel, mostly within the basal or me- 
dio-basal parts of the cells. 

The granular endoplasmic reticulum of the 
mucous cell is of the wide-luminar type, the 
width of the cisternae ranging from 1 20 to 280 
nm. In the lumen of the cisternae, there are 
typical tubular structures with an average di- 
ameter of 30 nm (Fig. 24). In the other cell 
types, only small amounts of granular endo- 
plasmic reticulum are present. Additionally, 
there are some cisternae of smooth endo- 
plasmic reticulum in the storage cells and ex- 
cretory cells. 

The most prominent cytopathologlcal 
changes in the endoplasmic reticulum after 
Cloethocarb intoxication are dilations of the 
cisternae, degranulation of the granular ER, 
proliferation and vesiculation of ribosome-free 
ER, and formation of membrane whorls and 
tubular structures. 

Within 30 minutes after ingestion of lethal 
and sublethal concentrations, degranulation 
of the granular endoplasmic reticulum and di- 
lation of the cisternae occur (Fig. 26). After 1 
h, the cisternae of both granular and degran- 
ulated/smooth endoplasmic reticulum are 



greatly dilated (Fig. 27). Furthermore, the 
amounts of degranulated smooth endoplas- 
mic reticulum in the storage, secretory, crypt 
and excretory cells have increased. In the 
storage cells, the cisternae often touch (Fig. 
28) or surround lipid droplets (Figs. 29, 30). 
After 5 h, vesicles of endoplasmic reticulum 
occur throughout the cytoplasm of storage, 
secretory and crypt cells. This reaction is also 
evident after 0.01% OD (Fig. 31), but is most 
intense after intoxication with the sublethal 
concentration of 0.001 %. The proliferation be- 
comes stronger with time. In addition to the 
dilation of the cisternae, characteristic con- 
centric whorls of the endoplasmic reticulum 
and other membrane whorls form 5, 24 and 
30 h after all lethal and both sublethal con- 
centrations (Figs. 29, 30). After ingestion of 
lethal concentrations, the membranes of the 
endoplasmic reticulum often rupture (Fig. 32). 

Another phenomenon appearing 5 h after 
both lethal and sublethal intoxication is a sys- 
tem of tubules arising from and connected 
with the degranulated/smooth endoplasmic 
reticulum (Figs. 33, 34). It occurs especially in 
the storage and excretory cells. After 0.001% 
OD, there are fewer tubules than after 0.01 %, 
whereas their number after 0.01 % is similar to 
that after lethal intoxication. 

Large Golgi fields characterize the mucous 
and crypt cells (Figs. 35, 36). in both kinds of 
cells, small vesicles originating from the gran- 
ular endoplasmic reticulum fuse with the cis- 
face cisternae. Most trans-face cisternae also 
fuse with small vesicles of unknown origin, 
become spherical, and finally as large vacu- 
oles become free from the Golgi fields. 

In the other cell types, the Golgi complex is 
less prominent. 

Disorganisation of the cisternae, compres- 
sion of the cis-face and dilation of the trans- 
face cisternae and destruction of membranes 
are the most common cytopathologlcal re- 
sponses of the Golgi apparatus to Cloetho- 
carb. 

One hour after both lethal and sublethal 
concentrations of the toxin are ingested, the 
normally regular arrangement of the cisternae 
in large and small Golgi apparatus is dis- 
rupted. The cis-face cisternae become tightly 
stacked (Figs. 37, 40). In the mucous cells, 
very many mucous vacuoles originate from 
the trans-face cisternae, and vesicles arising 
from the endoplasmic reticulum become more 
numerous (Fig. 38). With sublethal intoxica- 
tion these reactions occur after 30 h. 

Five hours after lethal intoxication, how- 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



97 



ever, damage to large and small Golgi com- 
plexes becomes greater. The trans-face cis- 
ternae, especially those of the small Golgi 
apparatus in the storage and secretory cells, 
are grossly swollen (Fig. 39) and the mem- 
branes often rupture (Figs. 37, 41). 

Within the digestive cells of the hepatopan- 
creas are typical vacuolar and lysosomal sys- 
tems, the vacuoles of which fuse with each 
other and with lysosomes, vary in size, and 
are generally largest towards the basal re- 
gions of the cells (Fig. 42). The small endocy- 
totic vesicles, located in the most apical parts 
of the cells and the lysosomes, are more elec- 
tron-dense than the large vacuoles (Fig. 43). 

A second type of cell that is dominated by 
vacuoles is the mucous cell (Fig. 2). Its vac- 
uoles fuse on their way from the base to the 
apex and are thus largest towards the apical 
part of the cell. 

In immature mucous cells, only a few mu- 
cous vacuoles occur and seem not to fuse 
with one another. 

The most striking cytopathological reac- 
tions of the vacuolar system are the intensi- 
fied fusion of vacuoles and the increased la- 
bility of membranes. 

Within one hour after lethal intoxication, re- 
action of the digestive vacuoles and the ly- 
sosomes is already evident. Endocytotic ves- 
icles in the apical part of the cell are fewer, 
and the large vacuoles contain material of ap- 
preciable electron-density (Fig. 7). From 3 to 
5 h, wide cisternae appear as a result of the 
intensified fusion of small vesicles and of 
small and large vacuoles (Figs. 44, 45). 
Whereas the membranes of the resulting 
large cisternae often rupture, those of the re- 
maining small vesicles remain intact. After 16 
h, most of the vacuolar membranes are 
heavily damaged (Fig. 46) and the vacuolar 
system thus is disrupted. 

The molluscicide acts on the mucous cells 
to produce a greater number of cells entirely 
filled with mucous vacuoles. Furthermore, 
even in immature mucous cells, many large 
vacuoles fuse. This intensified production of 
mucous occurs after lethal oral dosage but is 
more intense after sublethal intoxication (Fig. 
47). 

Storage Products 

In control animals, there are large deposits 
of lipid and glycogen in storage and crypt cells 
(Fig. 48), but few storage products occur in 
the secretory, digestive and excretory cells. In 
the storage cells, most lipid droplets are 



slightly electron-dense and only a few lipid- 
containing vesicles are totally electron-dense. 

Cloethocarb intoxication results in a reduc- 
tion of storage products and a concomitant 
increase in electron-dense vesicles. As soon 
as 1 h after ingestion of all lethal concentra- 
tions and of 0.01% Cloethocarb (Fig. 49), the 
glycogen content is slightly reduced in the 
storage cells of the oesophagus and crop, 
and lipid droplets fuse and appear less elec- 
tron-dense. 

After 3 h, lipid droplets have become fewer 
while electron-dense vesicles have become 
more numerous (Fig. 50). Peroxisomes are 
frequently associated with lipid droplets and 
ER cisternae surround them (Figs. 28-30). 

After 5, 16 and 24 h, the amount of storage 
products is obviously diminished. There is still 
some lipid present, but very little glycogen re- 
mains in the storage cells. 

The reduction of storage products is less 
intense after application of sublethal concen- 
trations than after that of any of the lethal con- 
centrations. Both the decrease in lipid and 
glycogen content and the increase in the 
number of electron-dense vesicles can be re- 
lated to the concentration of molluscicide in 
the food ingested. 

Muscle and Nerve Tissue 

In control animals, a conspicuous layer of 
muscle tissue, the filaments of which are lon- 
gitudinal and transverse, underlie the epithe- 
lia of the oesophagus, stomach and intestine 
(Fig. 51). The filamentous portions are sur- 
rounded by a plasma membrane and an en- 
velope of connective tissue. Underlying the 
epithelia of the crop and hepatopancreas is a 
very thin layer of muscle. 

Nerves, characterized by various neuro- 
secretory vesicles, lie close to the muscle tis- 
sue. The neurosecretory vesicles vary from 
electron-dense to electron-lucent and are sur- 
rounded by an electron-lucent halo. 

Connective tissue occurs between nerves, 
muscles and the epithelia of the digestive sys- 
tem. 

The most prominent cytopathological re- 
sponses of muscles and nerve tissues to the 
molluscicide are disorientation of muscle fila- 
ments, appearance of muscle envelopes de- 
void of muscle filaments and augmentation of 
neurosecretory vesicles. 

At 1 , 3 and 5 h after lethal intoxication, the 
muscle filaments are irregularly oriented in 
the oesophagus, stomach and intestine (Fig. 
52). After 24 h, the muscle tissue is frag- 



98 

© 



TRIEBSKORN & KÜNAST 



¥W f^- 



\ 



ger 




1 ju.m 



0.2 /xm 0.5 /am- .J4;^ 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



99 



merited and plasma membranes surround cy- 
toplasm lacking muscle filaments (Fig. 53). 
The most severe damage observed is in the 
stomach and the intestine. 

Furthermore, neurosecretory vesicles be- 
come more numerous (Figs. 54-56), and 
dense connections form between muscle and 
nerve tissue (Fig. 56). 

Intoxication with 0.01% molluscicide also 
disarrays the muscle fibers. In some cases, 
muscle envelopes without muscle fibres un- 
derlie the epithelium of the intestine after 24 h 
and 30 h. The number of neurosecretory ves- 
icles also increase. After 0.001% OD, there is 
no apparent reaction of either muscle or 
nerve tissue. 



DISCUSSION 

The present paper was designed as a 
baseline study of cellular reactions in the 
slugs' bodies to carbamate intoxication. Be- 
cause the molluscicide was orally applied, the 
cells of the digestive system were investi- 
gated as targets for effects of poisoning. In a 
prior investigation, the passage of ^'Re- 
labeled Cloethocarb through the digestive 
system could be traced and labeled material 
could be shown to penetrate the cells (Trieb- 
skorn et al., 1990), thus indicating that all 
cells of the digestive tract are in direct contact 
with the poison. 

Whereas earlier workers report the influ- 
ence of carbamates on the nervous system 
alone (Pessah & Sokolove, 1983; Young & 



Wilkins, 1 989), we showed that carbamate in- 
toxication produces conspicuous cellular re- 
actions in the digestive system as well, even 
though sublethal poisoning did not induce 
overt reactions at the macroscopic, organis- 
mic, level. The results of the study support the 
statement of Armstrong & Millemann (1974) 
that "damage to the nervous system through 
Cholinesterase inhibition may not be the only 
or even the primary cause of death of ex- 
posed clams." 

In the present study, three types of cellular 
reaction in the digestive system of Deroceras 
reticulatum can be distinguished: those that 
are detectable only after lethal intoxication, 
those that occur with the same intensity after 
lethal and sublethal poisoning, and reactions 
that are most intense after the application of 
sublethal concentrations. 

Reactions typical of lethal intoxication are 
damage to nuclei and mitochondria; alter- 
ations to the general cell outline, to the basal 
cell surfaces and to muscle and nerve tissue; 
and formation of clefts between the epithelia 
and the underlying connective, muscle and 
nerve tissues. Most of these reactions are ab- 
sent after 0.001% OD, but they appear at 
0.1% OD in a few cells and with less intensity 
after 30 h. The fact that these reactions be- 
come visible only after lethal intoxication 
might indicate either that these cellular struc- 
tures have a lower sensitivity to the poison, or 
that those reactions are secondary responses 
of the cells. In the first case, only high con- 
centrations of the toxin would lead to unspe- 
cific stress reactions that finally induce cell 



FIG. 21. Oesophagus (control). Mitochondrion (mi) in storage cell (arrows: cristae). 
FIG. 22. Intestine (2% Cloethocarb, 1h). Mitochondria (mi) swollen, cristae (arrows) partly reduced, ger: 
granular endoplasmic reticulum. 

FIG. 23. Stomach (2% Cloethocarb, 5h). Mitochondria (mi) with ruptured membranes (arrows). 
FIG. 24. Intestine (control). Wide-luminar granular ER (ger) of mucous cell. 
FIG. 25. Hepatopancreas (control). Granular ER (ger) of crypt cell, n: nucleus. 

FIG. 26. Hepatopancreas (0.01 % Cloethocarb, 1 h). Degranulation of granular ER (ger) in crypt cell (arrows). 
FIG. 27. Oesophagus (2% Cloethocarb, 1h). Cisternae of granular ER (ger) in mucous cell grossly dilated. 
FIG. 28. Crop (2% Cloethocarb, 5h). Cisternae of ER (er) touching lipid droplet (li). 
FIG. 29. Hepatopancreas (2% Cloethocarb, 1h). Cisternae of ER (er) surrounding lipid droplet (li), mito- 
chondria (mi) and vesicles (v) in crypt cell. 

FIG. 30. Hepatopancreas (2% Cloethocarb, 16h). Membrane whorls of ER (er) surrounding lipid droplet (li) 
and vesicles (v). 

FIG. 31. Hepatopancreas (0.01% Cloethocarb, 5h). Vesicles of ER (arrows) in basal part of crypt cell. 
FIG. 32. Hepatopancreas (0.1% Cloethocarb, 30 min). Ruptured membranes (arrows) of ER. li: lipid; mi: 
mitochondrion. 

FIG. 33. Intestine (0.1% Cloethocarb, 5h). Transverse and longitudinal section of tubular system arising from 
ER. Arrows: lumen of cisternae. 

FIG. 34. Intestine (0.1% Cloethocarb, 5h). Transverse and longitudinal section of tubular system arising from 
ER. Tubules open into wide-luminar ER cisterna (er). 



100 



TRIEBSKORN & KÜNAST 






get 



V- 



cf 



\ 



\ 



0.5 fxm 



0.Э' 



vac 



muv 



cf 



4t f 



tf 



vac 




0.5 /xm 



1 /xm 



у 



X- ( 



/ 



% 




0.5 /am 
40^ 



/ 



0.5)Ltm 



er 



/ 



» •• 



; v.* 



/ \ 





ел 



0.5 /u. m 



Л 



0.5 дт 






env ее 




ß'i 






dv 



\ 



1 /am 



-*r 



ly''',^ 0.5 /am 

4?) (^ 



/ .•"7 4^"^ 



ly 



ly 



dv 



^^/ 









^¡тл : : > 



0.5 /am 0.5 /am 



'■ v^i?-' ^j«x -• '^^■~^' .:>..?■--; 5 um 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



101 



death. In the second case, interaction of the 
toxin with other targets in the slug's body 
might cause the reactions and these would 
follow other symptoms of cell death. 

Carbamates, as nerve toxins, induce un- 
controlled muscle contractions that do not ap- 
pear after sublethal concentrations. As a con- 
sequence of these muscle convulsions, the 
epithelial cells might be stretched, leading 
to basal cell extension, such as to induce 
detachment of the cells from the basal mem- 
brane. This phenomenon has also been de- 
scribed by Vogt (1986) for the hepatopan- 
creas of Penaeus monodon after exposure to 
dimethoate, which is also an inhibitor of cho- 
linesterases. It might be that the toxin distorts 
the cytoskeleton, thereby changing the shape 
of the cell and displacing the mitochondria. 

Besides damaged nuclei, fully intact nu- 
culei and others with conspicuous crystalline 
inclusions in the karyoplasm occur. This ob- 
servation accentuates the importance of the 
heterogeneity of the cellular reaction. 

The crystalline inclusions in the karyoplasm 
might result from either intensified productiv- 
ity or serious injury to metabolic or regulatory 
processes. 

The reactions of the mitochondria, swelling 
and reduction of the cristae, are often consid- 
ered unspecific stress symptoms (Rez, 1 986). 
We have demonstrated in earlier studies 
(Thebskorn, 1988; 1989a), however, that 
there are several other modes of mitochon- 



drial response to different molluscicides. 
Swelling of the organelles and reduction of 
cristae can ensue immediately upon intoxica- 
tion, but can also result from other reactions 
such as an increase in number or size of in- 
tramitochondrial granules or the appearance 
of glycogen-like particles in the matrix. Fur- 
thermore, swelling of the mitochondria could 
also be induced in other cellular systems, 
such as the fish liver, by poisoning or by cer- 
tain diets (Braunbeck et al., 1989; Segner et 
al., 1987). We assume, therefore, that swell- 
ing of the organelles and reduction of the cris- 
tae can be induced in various ways by exog- 
enous or endogenous stresses. Even if the 
symptoms are similar, the causes of the re- 
sponse might be totally different. One attempt 
to explain the reaction is that of Goyer & 
Rhyne (1975), who propose that the swelling 
of the organelles results from inhibition of ion 
transport and protein synthesis. It also seems 
possible that the toxin interacts with the mito- 
chondrial membrane so as to change its per- 
meability to ions. 

Reactions that are discernible after the 
sublethal molluscicide concentration of 0.01 % 
and are intense after lethal intoxication are; 
reduction of microvilli, often associated with 
formation of apical protrusions of the cyto- 
plasm (surface blebs); presence of a coating 
upon the apical surfaces of the cells; thicken- 
ing of the basal membrane; intensification of 
fusion between small vesicles and vacuoles 



FIG. 35. Stomach (control). Golgi apparatus of mucous cell. Small vesicles (v) arising from granular ER (ger) 
fuse with cis-face cisternae. On trans-face (tf), small vesicles (v) and mucous vacuole (muv) are visible. 
FIG. 36. Hepatopancreas (control). Golgi apparatus producing large vacuoles (vac), cf: cis-face; If: trans- 
face. 

FIG. 37. Crop (0.1% Cloethocarb, 5h). Golgi apparatus in mucous cell with cis-face cisternae (arrows) 
closely stacked, vac: vacuole. 

FIG. 38. Intestine (2% Cloethocarb, 5h). Increased number of small vesicles (arrows) surrounding Golgi 
apparatus in mucous cell. 

FIG. 39. Oesophagus (2% Cloethocarb, 5h). Storage cell. Trans-face cisternae of small Golgi apparatus 
greatly inflated (arrows), er: granular ER. 

FIG. 40. Oesophagus (2% Cloethocarb, 5h). Storage cell. Small Golgi apparatus with cis-face cisternae 
tightly stacked (arrows). 

FIG. 41. Oesophagus (2% Cloethocarb, 30h). Storage cell. Small Golgi apparatus with disorganized cister- 
nae; membranes irregularly arranged and sometimes ruptured (arrows). 

FIG. 42. Hepatopancreas (control). Apex of digestive cell with microvilli (mv), endocytotic channels (ec), 
endocytotic vesicles (env), lysosomes (ly) and digestive vacuoles (dv). 

FIG. 43. Hepatopancreas (control). Digestive vacuoles (dv) and lysosomes (ly) in digestive cell. 
FIG. 44. Hepatopancreas (0.1% Cloethocarb, 5h). Lysosomes (ly) fusing with vacuolar system (vac; arrows) 
in digestive cell. 

FIG. 45. Hepatopancreas (0.1% Cloethocarb, 5h). Fusion of lysosomes (ly) with vacuoles (arrows) in 
digestive cell. 

FIG. 46. Hepatopancreas (2% Cloethocarb, 16h). Autolytic digestive cell with severely damaged vacuolar 
system, (arrows: intact vacuoles; n: nucleus). 




TRIEBSKORN & KÜNAST 



j^*. 



\ *# 







mJmL 



i' У^ 









( 



i»' ^î^. 




\ 



à 



^4 






1 /Ltm 



mt ij^- ^->, 










1 /u.m 



^ 1 /um 1 ixxv\ 



,»* 



* 1 •• 






mt 




^/ . 



!' 



1 ^im 1 дт 



//■I 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



103 



in the digestive cells of the hepatopancreas; 
reduction of storage products; condensation 
of the cis-face stacks of the Golgi apparatus 
and formation of membrane whorls by ER cis- 
ternae. 

The first four reactions described are re- 
garded as effects induced by the direct con- 
tact of toxin and cell surfaces. The thicken- 
ing of the basement membrane and the 
formation of an apical coat protect the apical 
and basal surfaces of the cell, preventing fur- 
ther penetration of the toxin from either the 
lumen of the digestive tract or from the 
hemolymph space. The origin of the coat is 
not known. Perhaps the cells themselves 
produce it but maybe the mucous cells of the 
stomach and the intestine form it, inasmuch 
as exudation of mucus intensifies after intox- 
ication (Thebskorn & Ebert, 1989). 

The reduction of microvilli, the formation of 
blebs and the increased fusion of vesicles 
and vacuoles in the digestive cells of the 
hepatopancreas might result from the inter- 
action of the lipophilic molluscicide with 
membranes. This interaction might induce 
changes in composition, fluidity and finally 
stability of the membranes (Axiak et al., 1988; 
Moore, 1982, 1985; Moore et al., 1982). 

Although the reactions mentioned above 
have often been described as unspecific cell 
responses to any stress (Rez, 1986), e.g. 
starvation (Segner et al., 1987), we assume 
that the plasma and the lysosomal mem- 
branes are very unstable and sensitive sys- 
tems that react quickly to any alteration of cel- 
lular homeostasis. Lysosomal instability in 
mussels has been used as a measure of en- 
vironmental pollution (Moore, 1982, 1985; 
Moore et al., 1982; Lowe et al., 1981). 

In such studies, the reduction of membrane 



stability has been investigated with light mi- 
croscopy and enzyme-histochemistry but not 
with the electron microscope. In the present 
study, we could show that the membranes of 
the small vesicles in the digestive cells of the 
hepatopancreas were not affected. As a con- 
sequence of the intensified fusion between 
small and large vacuoles, the number of large 
vacuoles increases after intoxication. Moore 
et al. (1982) assumed that changes in mem- 
brane fluidity induce this altered rate of vesi- 
cle fusion. The membranes of the resulting 
large autolysosomes are less stable and often 
rupture. Bayne et al. (1985) distinguish be- 
tween these autolysosomes, a typical re- 
sponse to stress, and the heterophagosomes, 
large vacuoles involved in pinocytosis and in- 
tracellular digestion in untreated animals. 

The decrease in stability of the ER and 
Golgi membranes, especially after ingestion 
of lethal concentrations, might also result 
from the capacity of the carbamate to interact 
with membranes. The more important reac- 
tions of the ER, however, are those to suble- 
thal concentrations; these will be discussed 
later. 

An increased call upon energy resources to 
initiate protective or detoxification processes 
might lead to a quick reduction of lipid and 
glycogen in the storage cells. Recio et al. 
(1988) also describe a reduction of storage 
products in Arion ater induced by zinc. They 
characterize the reaction to intoxication as 
similar to the effects of starvation. Because in 
our recent study peroxisomes and ER cister- 
nae were often observed in close contact with 
lipid droplets, and because our histochemical 
enzyme tests revealed that catalase was in- 
duced by Cloethocarb (Triebskorn, 1989b), it 
seems possible that ß-oxidation, or peroxida- 



FIG. 47. Oesophagus (0.001% Cloethocarb, 30 min). Increased пиглЬег of mature mucous cells (muc). n: 
nucleus. 

FIG. 48. Crop (control). Lipid (li) and glycogen storage (gl) in storage cell. 
FIG. 49. Crop (2% Cloethocarb pellet, 1h). Reduction of glycogen, fusion of lipid droplets (li). 
FIG. 50. Oesophagus (2% Cloethocarb, 3h). Decrease of lipid storage (li) and increased numbers of elec- 
tron-dense vesicles (ev). 

FIG. 51. Oesophagus (control). Muscle (mt) and nerve tissues (nt) underlying epithelium. 
FIG. 52. Oesophagus (2% Cloethocarb, 5h). Irregularly oriented muscle filaments (arrows). 
FIG. 53. Crop (2% Cloethocarb, 24h). Fragmentation of muscle tissue (mt) and muscle envelope lacking 
muscle filaments (arrows). 

FIG. 54. Oesophagus (2% Cloethocarb, 30 min). Nerve with increased numbers of neurosecretory vesicles 
(nv). mt: muscle tissue. 

FIG, 55. Stomach (0.5% Cloethocarb, 5h). Nerve with increased number of neurosecretory vesicles (nv). 
FIG. 56. Oesophagus (2% Cloethocarb, 30 min). Dense connection between muscle (mt) and nerve tissues 
(nt; arrow). 



104 



TRIEBSKORN & KÜNAST 



tive processes, or both, are involved in the 
reduction of lipid stores. Furthermore, there is 
perhaps a relation between peroxidation and 
membrane destruction, as has often been de- 
scribed for vertebrates (Tappel, 1975; Reck- 
nagel. 1967). 

The intensity of the alterations and damage 
after lethal intoxication are shown to be more 
severe than after sublethal intoxication. Al- 
though, in this case, a dose-response rela- 
tionship is obvious, no positive correlation 
could be found between dose and effect, if the 
reactions after low and high lethal concentra- 
tions were observed. That 0.1% OD fre- 
quently causes more severe damage than 2% 
could be explained by the fact that high con- 
centrations induce protective mechanisms or 
potential defense reactions, such as exuda- 
tion of mucus, more quickly than low concen- 
trations. Given the capacity of the mucus to 
dilute the toxin with the passage of time, the 
relative amount of the chemical in the lumen 
of the digestive tract might therefore be lower 
after 2% than after 0.1%. even if a higher con- 
centration were ingested. Furthermore, Bo- 
wen & Jones (1985) assume that high con- 
centrations of molluscicides prevent the 
animal from taking up lethal doses of the pes- 
ticide owing to quickly induced paralysis of 
the crop. A higher concentration of the pesti- 
cide thus might not be related necessarily to a 
higher efficiency as postulated by Fries & 
Tnpp (1976). 

In the third category are responses to lethal 
oral doses that are more intense after suble- 
thal intoxication. Such is the case of degran- 
ulation and dilation of the granular endoplas- 
mic reticulum (ER), the proliferation and 
vesiculation of the ER, the formation of a tu- 
bular system and of membrane whorls by the 
ER and the production of large mucous vac- 
uoles. 

Reactions of the ER to intoxification similar 
to those described in this study have often 
been seen in both vertebrates (Sivarajah et 
al.. 1978; Klaunig et al.. 1979) and mussels 
(Nott & Moore. 1987). Because transitions be- 
tween smooth and granular ER were visible, 
especially in the crypt cells of the hepatopan- 
creas, and because Klaunig et al. (1979) de- 
scribe a continuity between two forms of ER, 
we hesitate to refer to degranulated ER as 
smooth ER. It is unclear, moreover, whether 
ribosome-free ER necessarily functions as 
smooth ER. Klaunig et al. (1979) interpret the 
circular arrays of ER as a response to sub- 
stances that induce enzymes of the mixed 



function oxygenases system. A similar con- 
clusion can be drawn from our enzyme-his- 
tochemical tests, which showed an increase 
of NADPH-neotetrazolium reductase in cell 
areas in which ER-proliferation and whorls of 
ER cisternae occurred (Triebskorn, 1989a, 
b). 

Reactions of the endoplasmic reticulum are 
stronger after sublethal intoxication than after 
a lethal dose, most probably because de- 
structive effects are less important than in- 
duced reactions of a potentially protective na- 
ture. Any new molluscicide developed should 
not induce such defence mechanisms. 

A second mechanism that intensifies after 
sublethal intoxication is the production of 
large mucous vacuoles owing to an increased 
activity of the secretory system (ER, Golgi ap- 
paratus). Such large mucous vacuoles oc- 
curred even in cells having the shape typical 
of immature mucous cells, which generally 
have prominent Golgi complexes, large 
amounts of granular ER and only few mucous 
vacuoles. Large amounts of mucus are ex- 
uded as an immediate response to the inges- 
tion of lethal doses of the molluscicide. As 
already mentioned above, the mucus might 
serve the animal to dilute the toxin. Moreover, 
as shown in an earlier study, the animals are 
capable both of increasing the quantity, and 
of varying the quality, i.e. the chemical com- 
position, of the mucus (Triebskorn & Ebert, 
1989). In the case of Cloethocarb, the exuda- 
tion of acidic mucus can be regarded as a 
kind of incidental detoxification, because the 
toxin is less stable under acidic conditions (Kü- 
nast, pers. comm.). Nevertheless, the reason 
for the alteration in the chemical composition 
of the mucus is not known. Because we could 
demonstrate activity of y-glutamyltransferase 
and an increase in amount of SH-groups in 
the mucous cells of the digestive system 
(Triebskorn, 1988), we assume that conjuga- 
tion processes (glutathione conjugation) 
might be related to secretion of mucus. An 
increase in the number of mucous cells, such 
as described by Neff et al. (1987) as a re- 
sponse of Arctic marine bivalves to experi- 
mentally spilled oil, could not be detected. 

Although it might serve slugs as a defense 
mechanism, intensified exudation of mucus 
can also kill them; specific molluscicides can 
not only enhance secretion of mucus but also 
damage the ultrastructure of cells, especially 
immature ones. That is the reason that induc- 
tion of mucus secretion would finally lead to a 
desiccation of the animal and loss of mucous 



EFFECT OF CARBAMATE TOXIN ON SNAIL GUT 



105 



cells would prevent production of the very mu- 
cus that protects the surface of the animal 
from desiccation and that is necessary for di- 
gestion in the intestinal lumen. We therefore 
agree with the conclusions of Airey et al. 
(1989), who regard the mucous cells as one 
of the targets for specific molluscicial interfer- 
ence. 

Bowen & Jones (1985) also advocate pur- 
suing baseline studies in the development of 
new substances and suggest that mollusci- 
cides should be "packaged so as to effect a 
slow release and combined with a phagostim- 
ulant and pinocytosis inducer." 

We think that such basic studies revealing 
specific sites with which molluscicides could 
interfere are a necessary adjunct to screening 
programs in industrial research, of which the 
objectives are the discovery of new, more 
specific and less hazardous molluscicides. 



ACKNOWLEDGMENTS 

This study was partly supported by the Ger- 
man Research Council (DFG Sto 75/9). Per- 
sonal thanks go to Thomas Braunbeck and to 
Günter Vogt for the revision of the paper and 
to Rainer Günzler for his helpful assistance in 
the arrangement of the tables. 



LITERATURE CITED 

AIREY, W. J., I. F. HENDERSON, J. A. PICKET, G. 
С SCOTT, J. W. STEPHENSON & С M. 
WOODCOCK, 1989, Novel chemical approaches 
to mollusc control. Proceedings of the British 
Crop Protection Council. 41 : 301-308. 

ARMSTRONG, D. A. & R. E. MILLEMANN, 1974, 
Pathology of acute poisoning with the insecticide 
Sevin in the bent-nosed clam Macoma nasuta. 
Journal of Invertebrate Pathology. 24: 201-212. 

AXIAK, v., J. J. GEORGE & M. N. MOORE, 1988, 
Petroleum hydrocarbons in the marine bivalve 
Venus verrucosa: Accumulation and cellular re- 
sponse. Marine Biology. 97; 225-230. 

BAYNE, B. L., D. A. BROWN, K. BURNS. D. R. 
DIXON, A. IVANOVICI, D. R. LIVINGSTONE, D. 
M. LOWE, M. N. MOORE, A. R. D. STEBBING & 
J. WIDDOWS, 1985, The effects of stress and 
pollution on marine animals. Praeger Scientific, 
New York. 

BOWEN, J. D. & G. W. JONES, 1985, Getting pes- 
ticides into cells. Industrial Biotechnology 
(Wales). 5: 29-32. 

BRAUNBECK, T, V. STORCH & R. NAGEL, 1989, 
Sex-specific reactions of liver ultrastructure in ze- 
bra fish (Brachydanio rerio) after prolonged sub- 



lethal exposure to 4-nitrophenol. Aquatic Toxicol- 
ogy. 14: 185-202, 

BRAUNBECK, T., 1989, Cytopathologische Ve- 
ränderungen in der Fischleber durch Umwelt- 
chemikalien. Beiträge zur Ökotoxikologie. Dis- 
sertation, Heidelberg. 

FRIES, С & M. R. TRIPP, 1976, Effects of phenol 
on clams. Marine Fishery Reviews.. 38: 10-11. 

GLEN, D. M. & J. A. ORSMAN, 1986, Comparison 
of molluscicides based on metaldehyde, methio- 
carb or aluminum sulphate. Crop Protection. 5: 
371-375. 

GODAN, D., 1979, Schadschnecken. Ulmer Ver- 
lag, Stuttgart. 

GOYER, R. A. & B. С RHYNE, 1975, Toxic 
changes in mitochondrial membranes and mito- 
chondnal function. Pp. 383-428 in: TRUMP, B. 
F. & A. U. ARSTILA, eds.. Pathology of all mem- 
branes. Vol. 1, Academic Press, New York. 

HENDERSON, I. F. & K. A. PARKER, 1986, Prob- 
lems in developing chemical control of slugs. As- 
pects of Applied Biology. 13: 341-347. 

KARNOVSKY, M. J., 1971, Use of ferrocyanide- 
reduced osmium tetroxide in electron micros- 
copy. Journal of Cell Biology. 51 : 284. 

KEMP, N. J. & P. F. NEWELL, 1985, Laboratory 
observations on the effectiveness of methiocarb 
and metaldehyde baits against the slug Dere- 
ceras reticulatum (Müller). Journal of Molluscan 
Studies. 51: 228-230. 

KLAUNIG, J. E., M. M. LIPSKY, B. F. TRUMP & D. 
E. HINTON, 1979, Biochemical and ultrastruc- 
tural changes in teleost liver following subacute 
exposure to PCB. Journal of Environmental Tox- 
icology. 2: 953-963. 

LOWE, D. M., M. N. MOORE & R. CLARKE, 1981, 
Effects of oil on digestive cells in mussels: quan- 
titative alterations in cellular and lysosomal struc- 
ture. Aquatic Toxicology. 1: 213-226. 

MARTIN, T. J., & J. R. KELLY, 1986, The effect of 
changing agriculture on slugs as pests of cereals. 
British Crop Protection Council of Pest Diseases. 
411-424. 

MILLS, J. D., E. R. BAILEY, M. A. WEDGWOOD & 
С R. McCROHAN, 1989, Effects of mollusci- 
cides on feeding behaviour and neuronal activity. 
Proceedings of the British Crop Protection Coun- 
cil. 41: 77-84. 

MOORE, M. N.. 1982, Lysosomes and environ- 
mental stress. Manne Pollution Bulletin. 13: 42- 
43. 

MOORE, M. N.. R. K. PIPE & S. V. FARRAR, 1982, 
Lysosomal and microsomal responses to envi- 
ronmental factors in Littonna littorea from Sullom 
Voe. Manne Pollution Bulletin. 13: 340-345. 

MOORE, M. N.. 1985, Cellular responses to pollut- 
ants. Marine Pollution Bulletin. 16(4): 134-139. 

NEFF, J. M., R. E. HILLMANN, R. S. CARR, R. L. 
BUHL & J. J. LATHEY, 1987, Histopathological 
and biochemical responses in arctic marine bi- 
valve molluscs exposed to experimentally spilled 
oil. Arctic. 40, Supp.(l): 220-229. 

NOTT, J. A., & M. N. MOORE, 1987, Effects of 



106 



TRIEBSKORN & KÜNAST 



polycyclic aromatic hydrocarbons on molluscan 
iysosomes and endoplasmic reticulum. His- 
tochemical Journal. 19: 357-368. 

PESSAH. J. N. & P. G. SOKOLOVE. 1983, The 
interaction of organophosphate and carbamate 
insecticides with cholinesterases in the terrestrial 
pulmonate Umax maximus. Comparative Bio- 
chemistry and Physiology. 74C(2): 291-297. 

PORT, С M. & G. R. PORT, 1986, The biology and 
behaviour of slugs in relation to crop damage and 
control. Agncultural Zoology Reviews. 1: 253- 
297. 

PRYSTUPA, B. D., N. J. HOLLIDAY & R. B. WEB- 
STER, 1987, Molluscicide efficacy against the 
marsh slug Deroceras laeve (Stylommatophora: 
Limacidae) on strawberries in Manitoba. Journal 
of Economic Entomology. 80(4): 935-943. 

RECIO, A., J. A. MARIGOfHEZ, E. ÁNGULO & J. 
Г\/10УА, 1988, Zinc treatment of the digestive 
gland of the slug Anon ater L. 2. Sublethal effects 
at the histological level. Bulletin of Environmental 
Contamination and Toxicology. 41: 865-871. 

RECKNAGEL, R. O., 1967, Carbon tetrachloride 
hepatotoxicity. Pharmacological Reviews. 19: 
145. 

REZ, G,, 1 986. Electron microscopic approaches to 
environmental toxicity. Acta Biológica Hungarica. 
37(1): 31-45. 

RICHARDSON, K. C, L. JARRET & E. H. RINKE, 
1960, Embedding m epoxy resins for ultrathin 
sectioning in electron microscopy. Stain Technol- 
ogy. 35: 313-325. 

SEGNER. H., P. BURKHARDT, E. M. AVILA, J. V. 
JUARIO & V. STORCH, 1987, Nutrition-related 
histopathology of the intestine of milkfish Chanos 
Chanos fry. Diseases of Aquatic Organisms. 2: 
99-107. 

SIVARAJAH, K., С S. FRANKLIN & W. P. 
WILLIAMS, 1978, Some histopathological effects 
of Arochlor 1254 on the liver and gonads of rain- 
bow trout Salmo gairdnen and carp, Cyprmus 
carpió. Journal of Fish Biology. 13: 41 1-414. 

SPURR, A. R., 1969, A low viscosity embedding 



medium for electron microscopy. Journal of Ul- 
trastructural Research. 26: 31-43. 

TAPPEL, A. L., 1975, Lipid peroxidation and fluo- 
rescent molecular damage to membranes. In: 
TRUMP, B. F. & A. ARSTILA, eds., Pathobiology 
of all membranes. Vol. I. Academic Press, New 
York, pp. 145-170. 

TRIEBSKORN, R., 1988, Molluskizid-induzierte 
Reaktionen im Verdauungstrakt von Deroceras 
reticulatum (MúWer). Verhandlungen der Deutsch- 
en Zoologischen Gesellschaft. 81 : 332. 

TRIEBSKORN, R., 1989a, Ultrastructural changes 
m the digestive tract of Deroceras reticulatum in- 
duced by a carbamate molluscicide and by met- 
aldehyde. Malacologia. 31(1): 141-156. 

TRIEBSKORN, R., 1989b, The influence of a mol- 
luscicide on two cell types in the digestive system 
of Deroceras reticulatum (Müller). 16th Interna- 
tional Malacological Congress. Tübingen, in 
prep. 

TRIEBSKORN, R. & D, EBERT, 1989, The impor- 
tance of mucus production in slugs' reaction to 
molluscicides and the impact of molluscicides on 
the mucus producing system. Proceedings of the 
British Crop Protection Council. 41 : 373-379. 

TRIEBSKORN, R., С KÜNAST, R. HUBER & G. 
BREM, 1990. The tracing of a '"C-labeled car- 
bamate molluscicide through the digestive sys- 
tem of Deroceras reticulatum (Müller). Pesticide 
Science. 28: 321-330. 

VOGT, G., 1986, Monitoring of environmental pol- 
lutants such as pesticides in prawn agriculture by 
histological diagnosis. Aquaculture. 67: 157-164, 

WRIGHT, A. A. & R. WILLIAMS, 1980, The effects 
of molluscicides on the consumption of bait by 
slugs. Journal of t\Aolluscan Studies. 46: 265- 
281. 

YOUNG, A. G. & R. M. WILKINS, 1989, The re- 
sponse of invertebrate acetylcholinesterase to 
molluscicides. Proceedings of the British Crop 
Protection Council. 41: 121-128. 

Revised Ms. accepted 13 February 1990 



MALACOLOGIA, 1990, 32(1): 107-130 

THE PELAGIC FAMILY ATLANTIDAE (GASTROPODA: HETEROPODA) 
FROM HAWAIIAN WATERS: A FAUNISTIC SURVEY 

Roger R. Seapy 

Department of Biological Science, California State University, Fullerton 
Fullerton, California 92634, U.S.A. 

ABSTRACT 

The atlantid heteropod fauna of Hawaiian waters is composed of 13 species, including Oxy- 
gyrus keraudreni, Protatlanta souleyeti and 1 1 species of Atlanta. Species characterizations are 
accompanied by scanning electron micrographs of the shells for all species, color photographs 
of six live species and a key to the Hawaiian atlantids. 

Keywords: Atlantidae, Heteropoda, Hawaii, taxonomy, shell morphology, operculum morphol- 
ogy, eye morphology. 



INTRODUCTION 

Among the several groups of pelagic gas- 
tropods, the family Atlantidae (Heteropoda: 
Mesogastropoda) is perhaps the most poorly 
understood taxonomically by most Zooplank- 
ton biologists. This is undoubtedly due to their 
highly similar shell morphologies and the fact 
that species identifications have been based 
almost entirely on shell structure. In addition, 
the atlantids are difficult to work with because 
of their microscopic size: shell diameters of 
most individuals from Hawaiian waters range 
from about 0.6-0.7 mm in recently metamor- 
phosed individuals to about 2 mm in adults, 
although two species are as much as 4 mm 
and one species is nearly 9 mm in diameter. 
Identifications are usually made under a dis- 
section microscope. The important taxonomic 
features of the shell spire (the number, shape 
and sculpture of the whorls) are often difficult 
to see, however, even at higher magnifica- 
tions. In this and other recent studies, dis- 
cussed below, the scanning electron micro- 
scope (SEM) has proven to be essential in 
resolving fine details of spire structure. 

The first comprehensive review of the taxon- 
omy of the atlantids was that of Tesch (1 949). 
He reduced the number of recognized species 
from about 30 to ten based on the voluminous 
collections made during the Dana Expedition 
to the Atlantic, Pacific and Indian Oceans. Two 
of the genera, Oxygyrus and Protatlanta, in the 
family Atlantidae were monotypic and re- 
mained unchanged, although Tesch reduced 
the number of species in the third genus, At- 
lanta, to eight. Since the monograph by Tesch, 



only one major faunistic study (Richter, 1974) 
and one taxonomic review (van der Spoel, 
1 976) have been completed. In addition to the 
eight species recognized by Tesch, Richter 
(1974) included six others. Of these six, A. 
oligogyra Tesch, 1906, A. gibbosa Souleyet, 
1852, and A. affinis Tesch, 1906, had been 
described prior to Tesch's revision. The re- 
maining three, A. echinogyra Richter, 1972, A. 
plana Richter, 1972, and A. meteori Richter, 
1972, were subsequently described. In his re- 
view of the atlantids in 1976, van der Spoel 
also added six species, A. pacifica Tokioka, 
1955, A. peresi Frontier, 1966, A. gibbosa, A. 
tokiokai van der Spoel & Troost, 1 972, A. echi- 
nogyra and A. plana, to the eight recognized 
by Tesch in 1949. Among these six, however, 
only three, A. gibbosa, A. echinogyra and A. 
plana, overlapped with those identified by 
Richter (1974) from the Indian Ocean. 

Identification of the various species of At- 
lanta has been based almost exclusively on 
shell morphology, although eye, opercular 
and radular morphology can be very impor- 
tant characteristics for the recognition of cer- 
tain species. Tokioka (1961) described the 
opercula of a number of atlantids. He found 
that the opercula of most species were similar 
and differed only in overall shape and location 
of the gyre (or spiral portion). In two species, 
however, the opercular gyres were uniquely 
ornamented. Tokioka characterized the oper- 
culum of A. Inf lata Souleyet, 1852, as having 
a spiral row of claw-like structures around the 
central portion of the gyre and that of A. tur- 
riculata d'Orbigny, 1836, as having two rows 
of short spines that spiral outward from the 



107 



108 



SEAPY 




С 




Type a 



Type b 



Type с 



FIG. 1 . The three morphological types of opercula found in the Atlantidae (after Richter, 1974). The opercula 
were drawn from specimens collected during this study. A. Type a (macro-oligogyre) operculum is from a 1 .6 
mm Protatlanta souleyeti. B. Type b (micro-oligogyre) operculum is from a 2.2 mm Atlanta meteori. С Type 
с (monogyre) operculum is from a 1.4 mm Atlanta helictnotdes. G: opercular gyre. Scale bar is 0.2 mm. 



center of the gyre. Richter (1972) subse- 
quently showed that the operculum of A. in- 
flata lacked spiral sculpture on the gyre, how- 
ever, and that Tokioka had actually figured 
the opercula of two undescribed species, 
which Richter had collected from the Indian 
Ocean and had named A. echinogyra and A. 
plana. Three basic types of opercula, termed 
macro-oligogyre, micro-oligogyre and mono- 
gyre, were recognized by Richter (1961). In 
his 1 974 paper Richter termed these Types a, 
b and с They differ in overall shape and in the 
position and number of turns of the gyre (Fig. 
1). Among the Indian Ocean species, seven 
had Type b opercula, while five had Type с 
and three had Type a opercula (Table 1). 

Eye morphology has been little used in dis- 
tinguishing the species of atlantids. This may 
be due to difficulties encountered in seeing 
details of eye structure through the shell of 
preserved animals, inasmuch as the shell can 
become very opaque following preservation. 
Van der Spoel (1972) described a procedure 
for clearing such specimens without destroy- 
ing the shell so that the soft parts can be seen. 
In this same paper van der Spoel illustrated the 
eyes of nine species of Atlanta, among which 
only one (A. helicinoides Souleyet, 1852) pos- 
sessed a distinctive morphology. This eye type 
was characterized by a very broad pigmented 



base into which the spherical lens was re- 
cessed. Richter (1974) concluded that three 
basic eye types (termed Types a, b and c) 
could be distinguished among the species of 
atlantids (Fig. 2). The broad-based eye of A. 
helicinoides (termed Type c) is markedly dif- 
ferent from the more cuboidal shape of the 
other two eye types. In all three eye types the 
lens rests in a cup of pigmented tissue. This 
pigmented tissue is continuous (Type c) or is 
interrupted by an approximately triangular, un- 
pigmented window (Types a and b). The latter 
two eye types are easily distinguished by the 
presence (Type b) or absence (Type a) of a 
narrow, transverse slit in the distal portion of 
the pigmented tissue (Fig. 2). Among those 
species identified by Richter (1974) from the 
Indian Ocean, Type с eyes occurred in only A. 
helicinoides and O. keraudreni, while Type a 
and b eyes were equally distributed among the 
remaining species (Table 1). 

Radular morphology has been largely dis- 
regarded as a taxonomic character in the At- 
lantidae (Tesch, 1949; van der Spoel, 1976), 
although Richter (1986, 1987, 1990) has used 
radular differences to separate species hav- 
ing very similar shell morphologies. Earlier, 
Richter (1961) characterized the radulae of 
nine species of Atlanta and concluded that 
two types (I and II) could be distinguished 



HAWAIIAN ATLANTIDAE 



109 



TABLE 1. Species in the family Atlantidae recognized by Richter (1974, 1986, 1987, 1990). Whorl 
number refers to shell whorl in which whorl width increases rapidly (see text). Eye types (a, b and c) and 
opercular types (a, b and c) are those characterized by Richter (1974). Radular types (I and II) are those 
described for nine species by Richter (1961). 



Species 



Whorl 
number 



Eye 
type 



Opercular 
type 



Radular 
type 



'Oxygyrus keraudreni (Lesueur, 1817) 
*Protatlanta souleyeti (Smith, 1888) 
'Atlanta lesueuri Souleyet, 1852 
'Atlanta oligogyra Tesch, 1906 
'Atlanta peroni Lesueur, 1817 

Atlanta gaudichaudi Souleyet, 1852 
'Atlanta plana Richter, 1972 
'Atlanta echlnogyra Richter, 1972 
'Atlanta fusca Souleyet, 1852 
'Atlanta turriculata d'Orbigny, 1836 
'Atlanta Inf lata Souleyet, 1852 
' Atlanta helicinoides S>ou\eye{, 1852 

Atlanta inclinata Souleyet, 1852 
'Atlanta tokiokai van der Spool & Troost, 1972 

Atlanta gibbosa Souleyet, 1852 
'Atlanta meteori Richter, 1972 



». 


с 


*** 


1 


3 


a 


a 


1 


3 


b 


b 


II 


3 


a 


b 




4 


b 


b 


II 


4 


b 


b 


II 


4 


a 


b 




4 


a 


с 




5 


a 


a 


1 


5 


a 


a 




5 


a 


с 


1 


5 


с 


с 


1 


5 


b 


с 




6 


b 


с 


II 


6 


b 


b 




6 


b 


b 





*Denotes species identified from Hawaiian waters 

"Whorl counts not made because this species has involute spire 

"'Operculum broadly triangular to trapezoidal; not comparable with the opercula of Protatlanta and Atlanta 





uw 



Type a 



Type b 



Type с 



FIG. 2. The three morphological types of eyes found in the Atlantidae (after Richter, 1974). Illustrations of 
Type a and b eyes modified from drawings in Richter (1 974, Fig. 3). Type с eye is from a specimen of Atlanta 
helicinoides (shell length = 1.6 mm) from Hawaiian waters. DP: distal portion of pigmented tissue; L: lens; 
PP; proximal portion of pigmented tissue; TS: transverse slit in distal pigment; UW: unpigmented window. 
Scale bar (0.2 mm) applies only to Type с eye. Sizes of Type a and Id eyes not given by Richter (1974). 



(Table 1). These two species groups were 
also distinguished on the basis of eye and 
opercular morphology (Richter, 1974); those 
species with Type I radulae had Type a or с 
eyes and Type a or с opercula, while species 
with Type II radulae had Type b eyes and 
Type b opercula, except for A. tokiokai, which 
had a Type с operculum. 



The present paper characterizes Oxygyrus 
keraudreni (Lesueur, 1817), Protatlanta soul- 
eyeti (Smith, 1888), and 1 1 species of Atlanta 
based on material from plankton net samples 
collected off the island of Oahu between 1 984 
and 1986. Species descriptions are accom- 
panied by scanning electron micrographs of 
all species and color photographs of live an- 



110 



SEAPY 



TABLE 2. Numbers of specimens per sample examined from plankton net tows taken during cruises off 
western coast of Oahu in April 1984. March 1986. August 1986, and November 1986, and off northern 
coast of island of Hawaii in August 1986. See text for types of nets used, depths and volumes of water 
filtered during tows. 





Apr 


Mar 


Aug 


Aug 


Nov 




Species 


1984 


1986 


1986* 


1986** 


1986 


Total 


Atlanta lesueuri 


1.060 


75 


202 


181 


513 


2.031 


Atlanta turnculata 


171 


115 


555 


58 


186 


1.085 


Atlanta plana 


299 


388 


144 


198 


30 


1.059 


Atlanta Inflata 


352 


274 


209 


104 


113 


1.052 


Atlanta peroni 


332 


437 


44 


43 


7 


863 


Protatlanta souleyeti 


306 


191 


9 


19 


64 


589 


Atlanta meteor! 


120 


75 


51 


90 


2 


338 


Atlanta oligogyra 


21 


9 


52 


18 


94 


194 


Atlanta helicinoides 


58 


35 


42 


8 


30 


173 


Atlanta fusca 


34 


18 


2 


3 





57 


Atlanta echlnogyra 








27 


1 


19 


47 


Atlanta tokiokai 


2 


15 


5 


3 





25 


Oxygyrus keraudreni 


11 


3 











14 



•Oahu 
''Hawaii 



imals for six species. A key to the Hawaiian 
atiantids is included at the end of the paper. 



MATERIALS AND METHODS 

A total of 7,527 specimens of atiantids were 
examined (Table 2). The animals were re- 
moved from plankton samples collected dur- 
ing cruises of research vessels from the Uni- 
versity of Hawaii in waters off the western 
coast of Oahu (21 15N. 158 20'W) and off 
the northwestern shore of Hawaii (19 43'N. 
156 06'W). During a 10-14 Apnl 1984 cruise 
off Oahu. 40 tows were taken with paired, 
opening-closing Bongo nets (70 cm mouth di- 
ameter), constructed of 0.5 mm mesh Nytex 
gauze. Oblique tows of 30 min duration were 
taken within 50 m depth intervals between the 
surface and 200 m and from 200 to 300 m. An 
average of 2,600 m^ of water was filtered dur- 
ing each tow. During a 22-29 March 1986 
cruise off Oahu, 15 oblique tows between the 
surface and about 300 m were taken with an 
open ring net (226 cm mouth diameter), con- 
structed of 0.5 mm mesh Nytex gauze. The 
tows averaged 35 min in duration, and an av- 
erage of 7.300 m^ of water was filtered during 
each tow. Oblique tows to 300 m were taken 
using the 226 cm ring net during a 6-9 Au- 
gust 1986 cruise off Oahu (three tows) and off 
Hawaii (three tows). Average tow duration 
was 35 min, and the average volume of water 
filtered was 7,600 m^. During daytime hours 



on 23 November 1986 off Oahu, oblique tows 
to a target depth of 50 m were taken with the 
226 cm ring net (three tows; average of 5.100 
m^ filtered) and open 70 cm Bongo nets 
(three tows: average of 1 .800 m^ filtered). Un- 
less used to obtain specimens for observation 
or photography, plankton samples were pre- 
served aboard ship in 4% formalin solution In 
buffered sea water immediately after collec- 
tion and were transferred to 40% isopropanol 
within 14 days. All shell measurements were 
made to the nearest 0.1 mm with an ocular 
micrometer in a Wild M5 dissection micro- 
scope. Because the keel of the shell was fre- 
quently damaged, all shell diameters were 
measured exclusive of the keel. 

During the March and August 1986 cruises, 
specimens were sorted from the fresh plank- 
ton samples for live photography using a 
Zeiss dissection microscope with Koda- 
chrome 64 color slide film and Kodak VRG 
100 color negative film. Specimens were 
placed in filtered sea water in clear glass petri 
dishes. Vivitar 285 strobes were positioned 
on either side of the microscope stage and 
were angled obliquely to produce a dark 
background. 

A minimum of four specimens of each spe- 
cies were examined under a JEOL JSM-35CF 
scanning electron microscope (SEM). The 
shells were mounted on aluminum stubs to 
which double-sided tape had been attached, 
and were then cold sputter-coated with gold- 
palladium (60:40). 21 nm thickness, in a 



HAWAIIAN ATLANTIDAE 



111 



Peleo Model 3 sputter coater. Photographs 
were taken on Kodak T-MAX 120 black-and- 
white negative film. During preparation of 
specimens of Atlanta, drying did not produce 
any changes in the shape of the calcareous 
shell (composed of aragonite; Batten & Du- 
mont, 1976). However, because the adult 
shell and keel of Oxygyrus keraudreni and the 
keel of Protatlanta souleyeti are made of 
conchiolin (Richter, 1974; Batten & Dumont, 
1976), drying resulted in shriveling and col- 
lapse of these organic shell components. To 
retain their original shape, a critical point dry- 
ing procedure was used prior to sputter- 
coating. Briefly, individual specimens were 
held between filter paper hats in specimen 
holders. They were transferred from ethanol 
solutions of 30% to 50% to 90% and to 100% 
(three times at each concentration), then 
placed in Freon 113 (transferred three times) 
and then critical point dried in carbon dioxide. 
Complete synonymies of the species char- 
acterized in this paper were given by van der 
Spoel (1976) and are not repeated here. 
Voucher specimens of each species were de- 
posited with the Bishop Museum, Honolulu, 
Hawaii, and the National Museum of Natural 
History, Smithsonian Institution, Washington, 
D.C. 



RESULTS AND DISCUSSION 

A total of 13 species of atlantids were re- 
corded from Hawaiian waters (Table 1). Two 
of the genera {Oxygyrus and Protatlanta) are 
monospecific, while the third (Atlanta) in- 
cludes the remaining eleven species. De- 
scriptions of these species are presented be- 
low. 

Because the shell morphologies of the lar- 
vae and adults of a species are quite different, 
and because the larvae and adults commonly 
occur together in plankton samples, shell dif- 
ferences are described here before proceed- 
ing to the species characterizations. The most 
conspicuous difference is that the keel of the 
adults is lacking in the larvae (Fig. 3A-D). In 
addition, the shell terminates in an apertural 
lip that is quite different in the adults and lar- 
vae. In adults the aperture is approximately 
triangular to oval in cross-sectional outline 
and is formed by the two halves of the outer- 
most (final) shell whorl and the base of the 
preceding shell whorl. In the larval shell the 
aperture is formed by two large lobes that are 
separated from the preceding shell whorl by 



broad lateral notches (Fig. 3A-D). The surface 
sculpture of the larval and adult shells of each 
species can be quite different. The larval 
shells of eight of the Hawaiian species pos- 
sess raised sculpture, which ranges in the ex- 
tent of development from simple (e.g. A. 
plana [Fig. ЗА], with a small number of 
weakly-elevated spiral ridges) to complex 
(e.g. A. echinogyra [Fig. 3C], with prominent 
spiral ridges, angled cross-ridges and punc- 
tae). The postlarval whorls of the adult shell 
generally lack elevated ridges, although 
punctae are present in some species. Thus, 
the transition from the larval to the adult shell 
is often very distinct (e.g. A. echinogyra; Fig. 
8E,F). 

Oxygyrus Benson, 1835 

Oxygyrus keraudreni (Lesueur, 1817) 
(Fig. 3E-H) 

Matehal: A total of 14 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than 1.1 mm) to a 3.4 mm adult. Four 
individuals, ranging from 1 .4 to 2.7 mm, were 
examined under the SEM. 

Species characterization: The adult shell 
and keel are of conchiolin, although the larval 
shell is calcareous and has prominent, zigzag 
spiral sculpture (Fig. 3E,G). The shell spire is 
involute (Fig. 3G). With age, the calcareous 
larval shell is overgrown by the conchiolin 
adult whorls. The conchiolin keel is tall and 
terminates abruptly at the shell aperture (Fig. 
3E). Also, the keel is truncate along its ante- 
rior margin. The color of the adult shell and 
keel is a translucent, light bluish-purple. The 
eyes are large and Type с (Fig. 2C). The 
operculum (Richter, 1961: Fig. 18; van der 
Spoel, 1976: Fig. 133C) is very different from 
those of other atlantids. It is broadly triangular 
(nearly trapezoidal) and lacks the spiral por- 
tion (or gyre). 

Discussion: This species is collected infre- 
quently and in low numbers in Hawaiian wa- 
ters (Table 2). A maximal shell diameter of 10 
mm was reported by Tesch (1949) and van 
der Spoel (1 976). Richter (1 982) recorded an- 
imals between 3 and 8 mm from the guts of 
immature dolphin fish. The largest specimen 
captured in the present study was only 3.4 
mm. 

The shell of O. keraudreni is unique among 
the atlantids because it has an involute spire, 
rather than the outwardly-produced spire on 
the right side of the shell of the other two gen- 



112 



SEAPY 




FIG. 3. Scanning electron micrographs of larval shells of Atlanta plana (A,B) and A. echinogyra (CD), and 
of adult shell of Oxygyrus keraudreni (E-H). All photographs are of right side of the shell taken either 
perpendicular to the shell plane or at a 60 tilt. Scale bars are 0.1 mm for larval shells; 0.5 mm for O. 
keraudreni a\ low magnification (E,F), and 0.1 mm for O. keraudreni at high magnification (G,H). 



HAWAIIAN ATLANTIDAE 



113 




FIG. 4. Scanning electron micrographs of Protatlanta souleyeti (A-D) and Atlanta peroni (E-H). For each 
species four views are included; low magnification of right side of shell (upper left) and at 60 tilt (lower left); 
high magnification of spire (upper right) and at 60" tilt (lower nght). Scale bars are 0.5 mm for low magni- 
fication, 0.1 mm for high magnification. 



114 



SEAPY 



era. Further, the adult shell is composed en- 
tirely of conchiolin (Richter, 1974; Batten & 
Dumont, 1976), the sclerotized protein that 
forms the outer periostracum layer of the gas- 
tropod shell (Hyman, 1967). In young individ- 
uals, such as the 1.9 mm shell illustrated by 
Tesch (1949: Fig. 1С) and in the 2.2 mm 
specimen shown here in Fig. 3E-H, the junc- 
tion between the larval and adult shell is 
clearly marked. During the critical point drying 
procedure used in this study, this junction was 
exaggerated by the partial elevation of the 
adult shell from the underlying larval shell. 

Protatlanta Tesch, 1908 

Protatlanta souleyeti (Smith, 1888) 
(Figs. 1A. 4A-D, 5A) 

Material: A total of 589 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than 0.7 mm) to a 1.9 mm adult. Eight 
individuals, ranging from 0.9 to 1.7 mm were 
examined under the SEM. 

Species characterization: The shell is cal- 
careous and the keel is of conchiolin. The 
shell spire is smooth, lacking sculpture, and is 
slightly elevated (Fig. 4B,D). The keel has a 
glass-like transparency (Fig. 5A) and a 
curved, rectangular shape (Fig. 4A), extend- 
ing from the shell aperture to about one-half 
the circumference of the shell. The anterior 
margin of the keel is sharply truncate. The 
digestive gland is contained within the shell 
spire and is usually a light brownish-orange to 
reddish-brown (Fig. 5A). The eyes are Type 
a. and the operculum is Type a. 

Discussion: The maximal size of P. soul- 
eyeti in this study was only 1 .9 mm, although 
this is greater than the adult size range of 1 .0 
to 1.5 mm cited by Tesch (1949) and van der 
Spool (1976). 

The transparency and shape of the keel im- 
mediately distinguishes this species from all 
other atlantids. Protatlanta souleyeti is most 
similar in appearance to Atlanta lesueuri Soul- 
eyet, 1852, and A. oligogyra, for all three spe- 
cies have a compact spire comprised of a low 
number of smooth whorls. When the keel has 
broken off, as sometimes happened in the 
Hawaiian material, P. souleyeti can be some- 
what difficult to separate from these two spe- 
cies of Atlanta. In such instances the color of 
the shell spire can be used to separate P. 
souleyeti (brownish-orange to reddish-brown) 
from A. lesueun (clear to light pink) and A. 
oligogyra (light violet). 



Atlanta Lesueur, 1817 

The genus Atlanta differs from Oxygyrus 
and Protatlanta in having a shell and keel that 
are both calcareous. Separation of the species 
of Atlanta has been based largely on shell 
characteristics, as discussed above. One of 
the features of the shell that is easy to deter- 
mine and has been used commonly in the past 
is the total number of whorls comprising the 
adult shell. This number is not constant, how- 
ever, but increases with shell growth. Alterna- 
tively, the number of whorls comprising the 
inner portion of the shell is a feature that is not 
affected by the size of the adult animal at the 
time of capture. In her review of the hetero- 
pods, Thiriot-Quiévreux (1973) referred to the 
number of whorls comprising the spire for each 
of the atlantid species. In the present paper, a 
similar approach is used. 

Under the dissection microscope, the whorl 
in which the shell morphology of atlantids 
changes from that of the larva to that of the 
adult is often clearly demarcated. Even if this 
point of change cannot readily be detected, 
the dramatic increase in overall whorl size, 
indicated by a rapid increase in whorl width, 
that begins in the last larval whorl and contin- 
ues following metamorphosis is very distinc- 
tive. The shell whorl in which this region of 
rapid increase in whorl width occurs is used 
here as a taxonomic character. To make 
whorl counts, the shell must be oriented in a 
consistent manner. The specimen must be ro- 
tated until the protoconch is directed away 
from the viewer (Fig. 6). In A. lesueuri (Fig. 
6A), for example, the protoconch comprises 
most of the first whorl and is followed by a 
narrow second whorl and a rapidly expanding 
third whorl. In A. peroni Lesueur, 1817 (Fig. 
6C), on the other hand, the second and third 
whorls are narrow and the fourth whorl ex- 
pands rapidly. Among the 1 1 Hawaiian spe- 
cies, the whorl that expands rapidly is the 
third shell whorl in A. lesueun (Fig. 6A) and A. 
oligogyra (Fig. 68); the fourth whorl in A. per- 
oni (Fig. 6C), A. plana (Fig. 6E) and A. echi- 
nogyra (Fig. 6F); the fifth whorl in A. fusca 
Souleyet, 1852 (Fig. 6G), A. turriculata (Fig. 
6H), A. inflata (Fig. 61) and A. helicinoides 
(Fig. 6J); and the sixth whorl in A. tokiokai 
(Fig. 6K) and A. meteon (Fig. 6L). For pur- 
poses of comparison with A. peroni and A. 
plana, a sketch of A. gaudichaudi Souleyet, 
1852 (Fig. 6D), from Australian waters is in- 
cluded, although this species was not col- 
lected from Hawaiian waters. 



HAWAIIAN ATLANTIDAE 



115 




FIG. 5. Laboratory photographs of live atlantids collected frorn southwest side of Oahu May 1987. A. 
Protatlanta souleyeti (0.8 mm). B. Atlanta lesueuri (1.2 mm). С A. turriculata (0.9 mm). D. A. tokiokai (1.5 
mm). E. A. echinogyra (1.0 mm). F. A. inflata (0.9 mm). 



116 



SEAPY 




FIG. 6. Sketches of atlantid shell spires viewed at right angles to axis of spire and onented with protoconch 
directed upwards. Dashed line in each sketch to aid in counting of shell whorls. A. A. lesueuri. B. A. 
oligogyra C. A. peroni. D. A. gaudichaudi. E. A. plana. F. A. echinogyra. G. A. fusca. H. A. tumculata. I. A. 
inflata. J. A. helicinoides. K. A. tokiokai. L. A. meteori. Scale bars are 0.5 mm. All sketches from specimens 
of atlantids collected off Hawaii, except for that of A. gaudichaudi, which was based on animals from 
Australian waters. 



HAWAIIAN ATLANTIDAE 



117 




FIG. 6G-L. 



118 



SEAPY 



Atlanta /esuetvn Souleyet, 1852 
(Figs. 5B, 6A, 7A-D) 

Material: A total of 2,031 specimens was 
examined (Table 2), which ranged from larvae 
(less than about 0.6 mm) to 1.9 mm adults. 
Seven specimens, ranging from 1,0 to 1.7 
mm, were examined under the SEM. 

Species characterization: The shell spire is 
compact and low (Fig. 7B). The spire whorls 
are smooth, lacking any sculpture (Fig. 7C). 
The sutures between the whorls are incised 
(Fig. 7D). Rapid increase in whorl width oc- 
curs in the third shell whorl (Fig. 6A). The keel 
IS high with a truncated anterior edge (Fig. 
7A). Because it is quite fragile, however, the 
keel IS commonly damaged and the truncated 
anterior margin might not be evident. The 
shell of the animal is clear (Fig. 5B), although 
the inner whorls can take on a light pink color 
(darkest in the sutures) in older specimens. 
The body is also clear (Fig. 5B), except for 
purple-red pigmentation at the end of the pro- 
boscis and at middle and distal locations on 
the opercular lobe. This pigmentation be- 
comes darker and more prominent in older 
individuals. The eyes are Type b, the opercu- 
lum is Type b. 

Discussion: Atlanta lesueuri is among the 
most abundant species of atlantids in Hawai- 
ian waters (Table 2). It resembles only one 
other species, A. oligogyra. Features that dis- 
tinguish the two species are discussed below 
under A. oligogyra. 

Atlanta oligogyra Tesch, 1906 
(Fig. 7E-H) 

Material: A total of 194 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than 0.6 mm) to a 1.9 mm adult. Four 
specimens, ranging from 0.9 to 1 .9 mm, were 
examined under the SEM. 

Species characterization: The shell spire is 
compact and low (Fig, 7F). The spire whorls 
are smooth, lacking any sculpture (Fig. 7G). 
The sutures between the first and second 
whorls are shallow, while those between the 
second and subsequent whorls are incised 
(Fig. 7G,H). The sutures have a very light vi- 
olet color. The keel is moderately tall and 
rounded in lateral profile (Fig. 7E). The eyes 
are Type a and the operculum is Type b. 

Discussion: The shell of A. oligogyra can be 
difficult to distinguish from that of the preced- 
ing species, A. lesueuri. Features used by 
Richter (1974) to characterize the shell of A. 
oligogyra and to separate it from that of A. 



lesueun included a lower keel, a brown keel 
base and light violet inner surface of the ap- 
erture (in adults). The eyes of the two spe- 
cies, however, are distinctly different: Type a 
and small in A. oligogyra, and Type b and 
large in A. lesueun. Despite these differ- 
ences, van der Spool (1976) treated A. oli- 
gogyra as a synonym of A. lesueuri. In de- 
fense of his separation of the two species in 
his 1974 paper, Richter (1986) expanded on 
the species characterizations and included 
differences in the radulae. 

In the Hawaiian material, A. oligogyra could 
be distinguished from A. lesueuri on the basis 
of the Type a eyes (Type b in A. lesueuri), the 
relatively lower and more rounded keel (tall 
and truncated at the anterior edge in A. 
lesueuri). and the light violet color of the spire 
sutures (clear in A. lesueuri). The keel base of 
Hawaiian A. oligogyra is clear, not brown as 
reported by Richter for Indian Ocean speci- 
mens. 



Atlanta peroni Lesueur, 1817 
(Figs. 4E-H, 6C) 

Material: A total of 863 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than about 0.7 mm) to an 8.4 mm adult. 
Six specimens, ranging from 1.6 to 3.8 mm, 
were examined under the SEM. 

Species characterization: The shell spire is 
low (Fig. 4H). The whorls comprising the spire 
are smooth, lacking any sculpture (Fig, 4G). 
The sutures between the first and second 
whorls are shallow, while those between sub- 
sequent whorls are deeply incised (Fig. 4H), 
Rapid increase in whorl width occurs in the 
fourth shell whorl (Fig, 6C), The moderately 
tall keel is rounded in profile (Fig. 4E). The 
keel base is clear in young individuals (i.e. 
less than about 2 mm), but changes from a 
light to a dark golden-brown in progressively 
older animals. The keel inserts between the 
fifth and sixth whorls in individuals greater 
than about 2 to 3 mm. The eyes are Type b, 
the operculum is Type b. 

Discussion: This species achieves the larg- 
est size (up to a diameter of 10 mm) in the 
genus Atlanta (Richter, 1974; van der Spool, 
1976). The largest specimen that I collected 
in Hawaiian waters was 8.4 mm, which ap- 
proaches a total diameter of 10 mm when the 
keel is included. 

Except for the golden-brown pigmentation 
that develops at the base of the keel with in- 



HAWAIIAN ATLANTIDAE 



119 




FIG. 7. Scanning electron micrographs of Atlanta lesueuri (A-D) and A. oligogyra (E-H). Views and scale 
bars as in Fig. 4. 



120 



SEAPY 



creased age. most of the shells examined 
were clear. In some instances, however, the 
shell in juveniles was found to be light pink, 
which presumably persists and would ac- 
count for the light pink color seen in some 
adults (also reported by Thiriot-Quiévreux, 
1973). 

The present characterization of the shell 
spire for Hawaiian A. peroni agrees with the 
descriptions of Tesch (1949) and van der 
Spoel (1976). In the A. peroni irom the Indian 
Ocean examined by Richter (1974: Fig. 7), 
however, the whorls comprising the shell 
spire had an elevated, thin spiral ridge along 
the outer margin of each whorl. This elevated 
ridge was also indicated in the drawing by 
Frontier (1966: Fig. 4) of A. peroni from the 
Indian ocean. Such a ridge would appear to 
be lacking in the A. peroni from the Pacific 
and Atlantic oceans. Van der Spool's de- 
scription of the species and illustrations of a 
specimen from the Atlantic Ocean (1976: Fig. 
135A,B) do not include this ndge, nor does 
the scanning electron micrograph of a speci- 
men in Thiriot-Quiévreux (1 973: Fig. 1 C). Pre- 
sumably, the specimen used by Thiriot- 
Quiévreux came from the North Atlantic or the 
Mediterranean Sea. Tesch's (1949) descrip- 
tion and drawing (1949: p. 16-17, Fig 9) oi A. 
peroni also did not indicate such a ridge. The 
presence of an elevated spiral ridge on the 
shells of specimens collected from the Indian 
Ocean is problematical. The species identi- 
fied by Frontier and Richter as A. peroni could 
represent a morphological variant of A. peroni 
that is unique to the Indian Ocean. However, 
Richter (pers. comm.) now thinks that it is ei- 
ther an undescribed species or is a species 
that was described previously and is not cur- 
rently recognized. 

The shell morphology of A. peroni is close 
to that of only one other species from Hawai- 
ian waters, A. plana. Differences between 
these two species and A. gaudichaudi will be 
discussed below under A. plana. 

Atlanta plana Richter, 1972 
(Figs. 3A-B, 6E, 8A-D, 9A-B) 

Material: A total of 1,059 specimens was 
examined (Table 2), which ranged from larvae 
(less than about 0.7 mm) to a 3.4 mm adult. 
Six specimens, ranging from 1.3 to 2.5 mm, 
were examined under the SEM. 

Species characterization: The shell spire 
forms a low cone (Fig. 8D). Under the dissec- 
tion microscope the spire can appear to lack 



spiral sculpture. Under the SEM, however, 
two weakly-developed spiral ridges are seen 
on the second and third whorls (Fig. 8C,D). In 
the last half of the third whorl, the spiral ridges 
break up and are replaced by spirally ar- 
ranged rows of small punctae (Fig. ВС). The 
sutures of the spire are violet. The shell whorl 
in which a rapid increase in width occurs is 
the fourth (Fig. 6E). The keel is rounded and 
somewhat low (Fig. 8A). The keel base is a 
copper-brown to golden-brown color. The 
eyes are Type a. The operculum is Type b 
(Fig. 9A) and possesses a low gyre with about 
20 flattened, outwardly directed spines (Fig. 
98). 

Discussion; Like Richter (1974), I consider 
A. plana to be most similar in appearance to 
A. gaudichaudi. In turn, these two species are 
perhaps most similar to A. peroni. In all three 
species the shell whorl that expands rapidly in 
width is the fourth (Figs. 6C,D,E). All three 
species also have Type b opercula (Table 1). 
The operculum of A. plana is unique, how- 
ever, in that it has a spinose gyre (Fig. 9B). 
Further, the Type a eyes distinguish A. plana 
from A. gaudichaudi and A. peroni (Type b 
eyes), and the whorl sculpture (two weakly 
developed spiral lines) in A. plana is lacking in 
the other two species. The violet suture pig- 
mentation of the spire in A. gaudichaudi dis- 
tinguishes this species from A. peroni, which 
has clear to light pink sutures. 

Richter reported both A. gaudichaudi and 
A. plana from the Indian Ocean. Newman 
(pers. comm.) has recorded both species in 
waters off Heron and Lizard islands, Austra- 
lia, and has indicated that A. gaudichaudi is 
the most common species of atlantid. It is 
therefore surprising that I have never identi- 
fied A. gaudichaudi irovn Hawaiian waters, al- 
though I have routinely checked the eye type 
and, periodically, the operculum of specimens 
identified as A. plana. 



Atlanta echinogyra Richter, 1972 
(Figs. 3C-D, 5E, 6F, 8E-H, 9C,D) 

Matenal: A total of 47 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than 0.7 mm) to a 1.7 mm adult. Four 
specimens, ranging from 1.1 to 1 .7 mm, were 
examined under the SEM. 

Species characterization: The spire of this 
small species forms a low cone that can be 
slightly tilted relative to the shell plane (Fig 
8F,H). The shell whorl that expands rapidly is 






HAWAIIAN ATLANTIDAE 

С / 



121 




FIG. 8. Scanning electron micrographs of Atlanta plana (A-D) and A. echinogyra (E-H). Views and scale bars 
as in Fig. 4. 



122 



SEAPY 




FIG. 9. Photographs of operculum and spiral portion (gyre) of operculum in Atlanta plana (A,B), A. echin- 
ogyra (CD), A. tuniculata (E,F). Scale bars are 0.1 mm. 



the fourth (Fig. 6F). The second, third and 
about half of the fourth whorl bear prominent 
spiral ridges and secondary sculpture (Fig. 
8G,H). The shell spire is a unifornn reddish- 
brown, whereas the remaining whorls of the 
shell and keel are clear (Fig. 5E). The eyes 
are Type a, and the operculum is Type с The 
opercular gyre is elevated and bears about 1 2 
broad-based spines (Fig. 9C,D). 

Discussion: This species is immediately 
recognized in fresh material by the moder- 
ately elevated, reddish-brown shell spire. The 
characteristic spire sculpture is also clear un- 
der the high magnification of a dissection mi- 



croscope. The elevated opercular gyre with 
the broad-based spines is also unique to this 
species. When the operculum is mounted on 
a slide (for viewing beneath a compound mi- 
croscope), the tips of the spines bend under 
the pressure of the cover slip, causing them to 
have the hook-like appearance seen in Figure 
9C,D. 

I have collected A. echinogyra infrequently 
and in comparatively low numbers from Ha- 
waiian waters (Table 2). Richter (1974), how- 
ever, reported this species to be abundant in 
the western Indian Ocean. Richter (1987) 
also recorded a larger maximum shell size 



HAWAIIAN ATLANTIDAE 



123 



(2.5 mm) for A. echinogyra than I obtained 
from Hawaiian waters (1.7 mm). 

Atlanta fusca Souleyet, 1852 
(Figs. 6G, 10A-D) 

Material: A total of 57 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than 0.6 mm) to a 1.9 mm adult. Four 
specimens, ranging from 1 .0 to 1 .6 mm, were 
examined under the SEM. 

Species characterization: The shell spire 
forms a strongly elevated cone (Fig. 10B,D) 
The spire whorls are sculptured by a complex 
pattern of ornamentation (Fig. 10C,D). A 
prominent spiral ridge is located along the 
outer margin of the spire whorls (Fig. 10C,D). 
This ridge is low on the second shell whorl 
and progressively increases in height to a 
maximum on the fourth and fifth whorls. The 
complex spire ornamentation ends on the lar- 
val shell and is replaced by rows of small 
punctae on the adult shell (Fig. 10A,C). The 
keel istall and rounded (Fig. 10A). In animals 
larger than about 1 .5 mm, such as the animal 
used in Fig. 10A-D, the keel inserts between 
the fifth and sixth shell whorls. The shell is 
yellowish-brown (amber) to brown. The eyes 
are Type a and the operculum is Type a. 

Discussion: The largest specimen of A. 
fusca collected in the present study ap- 
proaches the maximal size of 2 mm reported 
by Richter (1974) for this species from the 
Indian Ocean. Van der Spoeí (1 976) indicated 
an upper size limit of 4 mm, however. 

This species is distinguished by its conspic- 
uous brown to yellowish-brown color and by 
its tall, conical spire. The species that is most 
similar in appearance to A. fusca is A. turricu- 
lata (see below). 

Atlanta turhculata d'Orbigny, 1 836 
(Figs. 5C, 9E-F, 10E-H) 

Material: A total of 1,085 specimens was 
examined (Table 2), which ranged from larvae 
(less than 0.6 mm) to a 1.7 mm adult. Six 
specimens, ranging from 1 .0 to 1 .6 mm, were 
examined under the SEM. 

Species characterization: The shell spire 
protrudes laterally from the right side of the 
shell as an elongate turret' (Fig 10F), formed 
by the strongly elevated second and third 
whorls capped by the protoconch (Fig. 10H). 
When the shell is viewed at right angles to the 
shell plane (Fig. 10G), a prominent spiral 



ridge is evident along the periphery of the 
spire whorls. When oriented in the plane of 
the shell (Fig. 10H), however, this spiral hdge 
is seen to be situated in the middle of the 
second and third whorls. The spiral ridge also 
increases in height to a maximum on the 
fourth and fifth whorls (Fig. 10H). The light 
reddish-brown color of the shell spire grades 
into a clear outer shell whorl and keel (Fig. 
5C). The keel is well developed and rounded 
in lateral profile (Fig. 1 0E). The eyes are Type 
a, the operculum is Type a. The operculum is 
unique in having two parallel rows of numer- 
ous short spines that spiral outward from the 
gyre center (Fig. 9E,F). 

Discussion: The strongly turreted, light red- 
dish-brown shell spire and the spinose oper- 
culum immediately distinguish A. turhculata 
from the other species of atlantids. The spe- 
cies of atlantid that is most similar to A. tur- 
riculata in appearance is A. fusca. Both spe- 
cies are small (maximal size of 2.0 mm or less 
in Hawaiian waters), have pigmented spires, 
have an elevated ridge in the same position 
on the spire whorls and have Type a eyes and 
Type a opercula. 

Atlanta Inf lata Souleyet, 1852 
(Figs. 5F, 61, 11A-D) 

Material: A total of 1,052 specimens was 
examined (Table 2), which ranged from larvae 
(less than 0.6 mm) to 1 .5 mm adults. Twenty- 
eight specimens, ranging from 0.9 to 1 .4 mm, 
were examined under the SEM. 

Species characterization: The shell of this 
small species is laterally inflated (shell width 
is about 40% of shell diameter). The spire is 
relatively flat (Fig. IIB.D). The shell whorl 
that increases rapidly in width is the fifth (Fig. 
61). The spire whorls and sutures are weakly 
defined owing to the presence of thick, evenly 
spaced spiral ridges (Fig. 1 1 A,C). The keel is 
tall (Fig. IIA) and its anterior margin is trun- 
cate in undamaged specimens (not shown by 
the specimen used in Figure 1 1 A-D, but illus- 
trated clearly in Figure 4 of Richter, 1974). 
The digestive gland, contained within the 
shell spire, is mottled reddish-brown to yel- 
lowish-brown (Fig. 5F). The eyes are Type a, 
the operculum is Type с 

Discussion: A second color morph of A. ín- 
flala was common in the Hawaiian fauna. It 
was immediately distinguished in fresh and 
recently preserved specimens by a uniform 
violet to light purple color of the spire. This 
appears only to be a color variant, however, 



124 



SEAPY 




FIG. 10. Scanning electron micrographs of Atlanta fusca (A-D) and A. tumculata (E-H). Views and scale bars 
as in Fig. 4. 



HAWAIIAN ATLANTIDAE 



125 




FIG. 1 1 . Scanning electron micrographs of Atlanta inflata (A-D) and A. helicinoides (E-H). Views and scale 
bars as in Fig. 4. 



126 



SEAPY 



because eye and opercular morphologies 
were indistinguishable fronn those of typical A. 
Ínflala, as were the shells when viewed under 
the SEM. 

Like Richter (1987), I have observed con- 
siderable variability in the presence and de- 
gree of expression of the spiral ridges on the 
whorls of the shell spire. In the majority of the 
individuals that I examined under the SEM, 
however, the spiral ridges were well devel- 
oped. The shell spire of A. inflata is very sim- 
ilar in appearance to that of A. heliclnoides. 
and these two species can be easily confused 
unless other taxonomic characters are used 
(discussed below). 

Atlanta heliclnoides Souleyet, 1852 
(Figs. 1С. 2C. 6J, 11E-H) 

Matenal: A total of 173 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than about 0.7 mm) to a 2.0 mm adult. 
Twelve specimens, ranging from 1.1 to 1.9 
mm, were examined under the SEM. 

Species characterization: The shell of this 
small species is laterally inflated (shell width 
is about 40% of the shell diameter). The spire 
is slightly elevated (Fig. 1 1 H). The shell whorl 
that increases rapidly in width is the fifth (Fig. 
6J). The second through fourth whorls have 
evenly spaced, thin spiral ridges (Fig. 11G). 
Because the spiral ridges are relatively nar- 
row, the sutures can be clearly distinguished, 
particularly under a dissection microscope. 
On the fifth whorl the spiral ridges break down 
and are replaced by rows of low, small punc- 
tae (Fig. 11G). The keel is rounded and mod- 
erately tall in undamaged specimens. The 
eyes are Type с and the operculum is Type с 

Discussion: In the Hawaiian fauna two dis- 
tinct color morphs of A. heliclnoides were en- 
countered in approximately equal propor- 
tions: a light yellow-tan form and a light 
purple-pink form. I could not see any struc- 
tural differences in eye, opercular or shell 
morphologies that would justify their taxo- 
nomic separation, however. 

Referring to A. heliclnoides, Tesch (1949: 
19) stated, "This species is at first sight so 
extremely like the preceding one (A. inflata) 
that it requires considerable attention to dis- 
tinguish them." Richter (1987) also noted the 
strong similarities of the two species, particu- 
larly in the appearance of the shell spire. In 
SEM photographs (compare Figs. IIA, 11E), 
the gross morphologies of the shells can be 
seen to be nearly identical. The spires are 



relatively flat and about the same size, and 

the number and spacing of the spiral ridges 
on the spire whorls are the same (compare 
Figs, lie, 11G). Further, both species are 
small and the shells are laterally inflated. The 
only obvious differences between the shells 
of the two species are, first, the spiral ridges 
on the spire whorls are thinner and less prom- 
inent in A. heliclnoides than in A. inflata, with 
the result that the whorls comprising the spire 
are more clearly defined in A. heliclnoides, 
and, second, the keel of A. heliclnoides is 
somewhat low and rounded, whereas that of 
A. inflata is tall and truncate along the anterior 
edge. The two species can be immediately 
separated on the basis of their eyes, how- 
ever, which are Type с in A. helicinoides and 
Type b in /A. inflata. 

Atlanta tokiokai van der Spool & Troost, 
1972 (Figs. 5D, 6K, 12A-D) 

Material: A total of 25 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than about 0.8 mm) to a 2.6 mm adult. 
Four specimens, ranging from 1 .2 to 2.4 mm, 
were examined under the SEM. 

Species characterization: The shell spire is 
tilted (or inclined) relative to the shell plane 
(Fig. 12A). The spire is globose in side view 
(Fig. 12B,D), forming an apical angle of about 
80". Spirally arranged rows of small, low 
punctae are present on the spire (Fig. 12D). 
The punctae become more prominent on the 
last whorls of the spire. The shell whorl that 
increases rapidly in width is the sixth (Fig. 
6K). The shell is a light yellow-tan color (Fig. 
5D). The keel is tall and rounded (Fig. 12A). 
The eyes are Type b, and the operculum is 
Type с 

Discussion: Specimens of A. tokiokai col- 
lected from Hawaiian waters were small: the 
largest individual measured only 2.6 mm. This 
maximal size is close to that (2.8 mm) re- 
ported by Richter (1990). 

Pnor to Richters 1 990 revision of the group 
of Atlanta species having tilted spires, he 
(1974) and, subsequently, I (1990a, 1990b) 
had identified this species as A. inclinata Soul- 
eyet, 1852. Among the atlantids from Hawai- 
ian waters, this and the next species, A. me- 
teon, are the only two species belonging to 
the group of four species having tilted spires. 
Features that distinguish A. tokiokai from A. 
meteori are given below under the latter spe- 
cies. 

The species that is most similar to Atlanta 



HAWAIIAN ATLANTIDAE 



127 




FIG. 1 2. Scanning electron micrographs of Atlanta tokiokai (A-D) and A. meteori (E-H). Views and scale bars 
as in Fig. 4. 



128 



SEAPY 



tokiokai is A. inclinata (Richter, 1990). Both 
species have a globose spire, with an apical 
angle of about 80 . However, the spire of A. 
inclinata consists of four whorls, while that of 
A. tokiokai is comprised of five whorls. In A. 
inclinata the last shell whorl is colorless and 
the spire is a weak rose color, whereas the 
entire shell of A. tokiokai is a light yellow-tan. 
When viewed with the SEM, the prominent 
spiral rows of punctae on the spire whorls of 
A. tokiokai are greatly reduced or lacking on 
A. inclinata. with the result that the shell sur- 
face is essentially smooth. However, when 
viewed with the light microscope, the spire 
whorls of A. inclinata show fine radial mark- 
ings, which are due to the internal shell struc- 
ture. Lastly, A. inclinata attains a larger size 
(to 6 mm) than A. tokiokai (2.8 mm). 

Atlanta meteori Richter, 1 972 
(Figs. IB. 6L. 12E-H) 

Matenal; A total of 338 specimens was ex- 
amined (Table 2), which ranged from larvae 
(less than about 0.7 mm) to a 3.7 mm adult. 
Nine specimens, ranging from 1 .2 to 2.8 mm, 
were examined under the SEM. 

Species characterization: The shell spire is 
tall and strongly tilted (Fig. 12E). Viewed in 
the plane of the shell (Fig. 12F), the spire is 
conical and steep sided, with an apical angle 
of about 70 . The spire whorls are smooth and 
relatively flat, and are separated by very shal- 
low sutures that are difficult to resolve under 
the SEM (Fig. 12G,H). The shell whorl that 
increases rapidly in width is the sixth (Fig. 6L). 
The prominent keel is rounded in lateral pro- 
file, except at the anterior edge, where it is 
truncate (Fig. 12E). The eyes are Type b, the 
operculum is Type b. 

Discussion: In live material A. meteon is the 
clearest and most glass-like of the Hawaiian 
atlantids. Richter (1972, 1974) also com- 
mented upon this feature in A. meteori from 
the Indian Ocean. Atlanta meteon is most 
similar to A. gibbosa (Richter, 1990). Both 
species are clear and glass-like, but the spire 
of A. gibbosa forms a pointed cone (apical 
angle of about 85 ), and the spire whorls are 
rounded with incised and distinct sutures. The 
umbilicus is conspicuously wider in A. gib- 
bosa than in A. meteori. 

As indicated above, the only two species in 
the Hawaiian fauna with distinctly tilted spires 
are A. meteon and A. tokiokai. In the former 
species the shell is clear, whereas in the latter 
species it is light yellowish-tan. Further, the 



spire of the former species forms a tall cone 
(about a 70 apical angle) and has smooth 
whorls, whereas that of the latter species is 
lower, globose (about an 80 apical angle) 
and is ornamented by spiral rows of numer- 
ous, small punctae. The opercula are also dif- 
ferent: Type b in A. meteori and Type с in A. 
tokiokai. 



CONCLUSIONS 

The atlantid fauna of Hawaiian waters is 
highly diverse, and includes 1 3 of the 1 6 spe- 
cies reported by Richter (1974) from the ex- 
tensive plankton sampling program of the Me- 
teor Expedition to the Indian Ocean. The 
number of worldwide species in the genus At- 
lanta is at least twice as many as the eight 
species recognized by Tesch in 1949. The 
total of 16 species does not include two spe- 
cies described since 1949 (A. peresi and A. 
pacifica), which were characterized in the tax- 
onomic review of van der Spool (1976). Nei- 
ther species was reported from the Indian 
Ocean by Richter (1974) or from the central 
Pacific Ocean in the present study. In his 
study on atlantid opercula, Tokioka (1961) 
concluded that A. pacifica was not a valid 
species. The validity of these two species re- 
mains uncertain and confirming studies are 
needed. 



KEY TO HAWAIIAN ATLANTIDAE 

1 . a. Spire whorls involute, projecting 

spire lacking from right side of shell; 
adult shell and keel of conchiolin 
. . . Oxygyrus keraudreni (Fig. 3E-H) 
b. Spire projects laterally, to varying 
degrees, from the right side of the 
shell; adult shell calcareous; keel 
calcareous or of conchiolin 2 

2. a. Keel of conchiolin and transparent 

. . . Protatlanta souleyeti (Fig. 4A-D) 

b. Keel calcareous and translucent 

(genus Atlanta) 3 

3. a. Shell whorl that increases rapidly in 

width is the third 4 

b. Shell whorl that increases rapidly in 
width is the fourth, fifth or sixth . . 5 

4. a. Eyes Type b; shell and keel base 

unpigmented, although inner whorls 
can become light pink in older 
specimens: keel tall with anterior 
edge truncated . . . . A. lesueuri (Fig. 
8A-D) 



HAWAIIAN ATLANTIDAE 



129 



b. Eyes Type a; shell unpigmented 11. a. Eyes Type a; spire with prominent 

except spire whorls (faint violet) and spiral ridges on whorls; sutures 

sutures (light violet); keel moderately separating second from third whorls 

elevated and rounded . . A. oligogyra and third from fourth whorls difficult 

(Fig. 8E-H) to distinguish; keel tall, with anterior 

5. a. Shell whorl that rapidly increases in edge truncated A. inflata 

width is the fourth or fifth; axis of (Fig. 12A-D) 

spire not inclined relative to the shell b. Eyes Type c; spire with weakly 

plane 6 developed spiral ridges on whorls; 

b. Shell whorl that rapidly increases in sutures separating second from third 

width is the sixth; axis of spire whorls and third from fourth whorls 

inclined relative to the shell distinct; keel moderately low and 

plane 12 rounded in profile . . . A. helicinoides 

6. a. Shell whorl that rapidly increases in (Fig. 12E-H) 

width is the fourth 7 12. a. Shell light yellow-tan color; spire 

b. Shell whorl that rapidly increases in globular and moderately inclined 

width is the fifth 9 relative to the shell plane; surface of 

7. a. Spire whorls smooth, lacking spiral spire whorls with numerous, small 

sculpture; spire slightly rounded; and regularly spaced spiral rows of 

sutures of spire whorls unpigmented punctae that are most strongly 

or light pink; eyes Type b A. developed on the fourth and fifth 

peroni (Fig. 4E-H) whorls; operculum Type с A. 

b. Spire whorls with weak to tokiokai (Fig. 9A-D) 

well-developed spiral ridges; spire b. Shell clear and glass-like; spire tall, 

forms a low cone; spire whorls clear conical and steeply inclined relative 

with violet sutures or reddish-brown; to the shell plane; surface of whorls 

eyes Type a 8 smooth, lacking punctate sculpture; 

8. a. Spire weakly conical; second and operculum Type b A. meteori 

third spire whorls with two weakly (Fig. 9E-H) 

expressed spiral ridges and violet 

sutures; gyre of Type b operculum 

with about 20 narrow, projecting 

spineMF-g. 11B) . . . . Л. plana (Fig. ACKNOWLEDGMENTS 

b. Spire forms a low cone; spire red- 
dish-brown, with well-developed spiral The support provided by the officers, crew 
ridges and secondary sculpture on the and members of the scientific party during 
second through fourth whorls; gyre of cruises of the R/V KANA KEOKI and R/V 
Type с operculum elevated, with about KILA, University of Hawaii, are gratefully ac- 
12 broad-based, thick projecting knowledged. I am especially indebted to 
spines (Fig. 1 1 D) ... /A. echinogyra Richard Young of the University of Hawaii for 
(Fig. 10E-H) his collaboration and assistance in the collec- 

9. a. Spire projects conspicuously as a tion of the samples on which this study was 

high cone or a turret 10 based. For the loan of the Bongo nets used 

b. Spire flattened, not conical or during two of the cruises, I thank Jed Hirota. 

turreted 11 My deep gratitude is extended to Steven Karl 

10. a. Spire forms a high cone (spire angle for operation of the SEM, critical-point prepa- 

about 65-75'); shell distinctive yel- ration of specimens, photography and print- 

lowish-brown (or amber) color; gyre ing. Specimens of Atlanta gaudichaudi were 

of operculum lacking ornamenta- graciously provided by Leslie Newman. To 

tion A. fusca (Fig. 13A-D) Gotthard Richter I am most appreciative for 

b. Spire turreted and steep-sided (spire the verification of my species identifications, 
angle about 35-45°); shell (especially Critical reviews of the manuscript were pro- 
the spire) light reddish-brown; gyre vided by Richard Young, Leslie Newman and 
of operculum with double row of an anonymous reviewer. This study was sup- 
short, projecting spines (Fig. 11F) ported by National Science Foundation Grant 
A. turhculata (Fig. 13E-H) OCE-8500593. 



130 



SEAPY 



LITERATURE CITED 

BATTEN, R. L. & M. P. DUMONT. 1976, Shell ul- 
trastructure of the Atlantidae (Heteropoda, Meso- 
gastropoda) Oxygyrus and Protatlanta. with com- 
ments on Atlanta inclinata. Bulletin of the 
American Museum of Natural History. 157:263- 
310. 

FRONTIER, S., 1966. Notes morphologiques sur 
les Atlanta récoltées dans le plancton de Nosy 
Bé (Madagascar). Cafiiers Office de la Re- 
cfiercfie Scientifique et Tectonique Outre-t\/ler. 
Sene Oceanograptiie. 4:131-139. 

HYMAN, L. H., 1967, Tfoe invertebrates. Volume VI. 
Mollusca I. McGraw-Hill, New York. 792 p. 

RICHTER, G., 1961, Die Radula der Atlantiden 
(Heteropoda, Prosobranchia) und ihre Bedeu- 
tung für die Systematik und Evolution der Fami- 
lie. ZeitscfiriH für Morphiologie und Ökologie der 
Tiere. 50:163-238. 

RICHTER, G.. 1972, Zur Kenntnis der Gattung At- 
lanta (Heteropoda: Atlantidae). Archiiv für Mol- 
luskenkunde. 102:85-91. 

RICHTER, G., 1974, Die Heteropoden der "Me- 
teor'-Expedition in den Indischen Ozean, 1964 
65. "Meteor" Forschung-Ergebnisse. (D), 17:55- 
78. 

RICHTER, G., 1982, Mageninhaltsuntersuchungen 
an Oxygyrus keraudreni (Lesueur) (Atlantidae, 
Heteropoda). Beispiel einer Nahrungskette im 
tropischen Pelagial. Senckenbergiana Maritima. 
14:47-77. 

RICHTER, G., 1986, Zur Kenntnis der Gattung At- 
lanta (II). Atlanta lesueun Souleyet und Atlanta 
oligogyra Tesch. Archiv für Molluskenkunde. 
117:19-31. 



RICHTER, G., 1987, Zur Kenntnis der Gattung 
Atlanta (III). Atlanta inflata. A. helicinoides. A. 
echinogyra und A. plana (Prosobranchia: Het- 
eropoda). Archiv für Molluskenkunde. 117:177- 
201. 

RICHTER, G., 1990, Zur Kenntnis der Gattung At- 
lanta (IV). Die Atlanta inclinata-Gruppe (Proso- 
branchia: Heteropoda). Archiv für Mollusken- 
kunde. 119:259-275. 

SEAPY, R. R., 1990a, Patterns of vertical distribu- 
tion m epipelagic heteropod molluscs off Hawaii. 
Manne Ecology Progress Series. 60:235-246. 

SEAPY, R. R., 1990b, Sampling requirements for 
epipelagic heteropod molluscs. American Mala- 
cological Bulletin, 8:45-52. 

TESCH, J. J., 1949, Heteropoda. Dana-Report, 34: 
54 pp., 5 pi. 

THIRIOT-QUIÉVREUX, C, 1973, Heteropoda. 
Oceanography and Marine Biology. Annual Re- 
view. 11:237-261. 

TOKIOKA, T, 1 961 , The structure of the operculum 
of the species of Atlantidae (Gastropoda: Het- 
eropoda) as a taxonomic criterion, with records of 
some pelagic molluscs in the North Pacific. Pub- 
lications of Seto Manne Biological Laboratory, 9: 
267-332. 

VAN DER SPOEL, S., 1972, Notes on the iden- 
tification and speciation of Heteropoda (Gas- 
tropoda). Zoologische Mededelingen Rijksmu- 
seum van Natuurlijke Historie te Leiden, Al: 
545-560. 

VAN DER SPOEL, S., 1976, Pseudothecosomata, 
Gymnosomata and Heteropoda (Gastropoda). 
Bohn. Scheltema & Holkema. Utrecht, 484 p. 

Revised Ms. accepted 2 January 1990 



MALACOLOGIA, 1990, 32(1): 131-145 

ON THE VARIOUS EDITIONS OF TETSUAKI KIRA'S 

"COLOURED ILLUSTRATIONS OF THE SHELLS OF JAPAN" 

AND "SHELLS OF THE WESTERN PACIFIC IN COLOR VOL. I," 

WITH AN ANNOTATED LIST OF NEW NAMES INTRODUCED 

Rüdiger Bieler^ & Richard E. Petit^ 

ABSTRACT 

At least 37 printings of Tetsuaki Kira's "Coloured Illustrations of the Shells of Japan" (CISJ) 
and nine of its English-language version, "Shells of the Western Pacific in Color Vol. !" (SWPC), 
published between 1954 and 1989, are in existence. Some of the editions, often incorrectly 
dated on the title page, differ greatly in text and illustrations. They are of taxononnic importance 
as they contain descriptions of new taxa and so-called "manuscript names," spanning 30 mol- 
luscan families. Of the 54 species-group and one genus-group names in question, five had been 
introduced before Kira's work, two are not available for taxonomic purposes, and eight were 
subsequently introduced by other authors after Kira had listed them as nude names only. The 
remaining 40 species-group names, only five of which were designated as new taxa, were made 
available with this work. The majority, 34 names, date from Kira (1959, CISJ), two from Kira 
(September 1 954, CISJ), two from Kira (1 960, CISJ), one from Kira (1 962, SWPC), and one from 
Kira & Habe in Kira (1962, SWPC). New taxa and "manuscript names" are traced through the 
various editions, and annotated lists of the printings and of the nominal taxa are supplied. 



INTRODUCTION 

Tetsuaki Kira's "Coloured Illustrations of the 
Shells of Japan," first published in 1954, in 
Japanese, has become one of the standard 
reference works on Indo-Pacific mollusks. A 
revised edition in English, first published in 
1962, is known as "Shells of the Western Pa- 
cific in Color <Vol. I>." For convenience, they 
will hereafter be referred to as CISJ and 
SWPC, respectively. The work is of taxonomic 
interest as it contains the descriptions of at 
least 40 new species-group names. One new 
genus-group name appears, but is not validly 
introduced. However, the frequent use of 
manuscript names, the lack of dates given with 
the taxa, rampant misspellings, and the erratic 
use of parentheses around authors' names, 
make use of the book extremely difficult and 
have resulted in many erroneous citations by 
subsequent authors. While working on such 
problems, we discovered that there are at least 
37 Japanese and nine English-language print- 
ings of this work, published between 1 954 and 
1989. Of the Japanese version, two editions 
are generally recognized, the original and the 
"enlarged & revised" edition, which was first 
published in 1959. We found, however, many 
of the so-called new printings to be revised 
editions. They were often newly typeset, and 



contained not only technical corrections and 
additions (such as indexes) but had major tax- 
onomic changes in the figure captions, and 
figures were replaced or renumbered on some 
of the plates. Only five new species are clearly 
designated as such, but about 50 manuscript 
names, inconsistently assigned to various au- 
thors, appear in the different editions (there 
may be additional nude names or available 
descriptions of taxa wrongly assigned to other 
authors in this work that have escaped us). A 
further complicating fact is that the date on the 
English-language title page of the Japanese 
editions is often incorrect, as can be shown by 
the Japanese date (day, month and year of 
emperor Hirohito's reign) in the colophon (in- 
scription at the end) of each copy. Some En- 
glish-language title pages were apparently 
used for a number of subsequent printings 
("1 955," for instance, still appears on the 1 958 
printing). Issues of CISJ and SWPC of the 
same year do not necessarily agree in taxo- 
nomic treatment, as the two were apparently 
independently revised and edited (Kira men- 
tions in the SWPC preface that "Dr. Habe took 
the trouble of revising the Latin names into the 
latest and authorized way of use"). 

The taxonomic mess created by author and 
publisher of this work proves the inappropri- 
ateness (1) of introducing manuscript names 



^Delaware Museum of Natural History, Current address: 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. 



131 



132 



BIELER & PETIT 



into the literature, and (2) of describing new 
taxa in commercial book publications. 

LISTING OF NEW AND 
■MANUSCRIPT" NAMES 

'Cantharidus kanekonis Oyama" 
Gaza sericata Kira, 1959 
Pseudastralium henicus gloriosum Kira, 
1959 

Galeoastraea millegranosa Kuroda & 
Habe in Habe, 1958 
Papyriscala bifasciata Kira & Habe in 
Kira, 1962 

Cirsotrema (Elegantiscala) ku rodai Kira, 
1960 

Canarium microurceum Kira, 1959 
Neverita (Glossaulax) hosoyai Kira, 
1959 

Simnia xanthochila Kuroda, 1928 
Pustularia margarita tetsuakii Kira, 1959 
Erosana tomlini maturata Kira, 1959 
Semicassis persimilis Kira, 1959 
Bursa dunkeri Kira, 1959 
Eudolium inf latum Kuroda & Habe, 1 952 
Murex kiiensis Kira, 1959 
Ceratostoma (Pteropurpura) vespertilio 
Kira, 1959 

Coralliobia akibumii Kira, 1960 
Coralliophila pyriformis Kira, 1959 
Latiaxis kawamurai Kira, 1959 
Babylonia pallida Kira, 1959 
Neptúnea fukueae Kira, 1959 
Buccinum isaotakii Kira, 1959 
Nassanus (Zeuxis) kiiensis Kira, 1959 
Granulifusus kiranus Shuto, 1958 
Fusinus gemmuliferus Kira, 1959 
Fusinus crassiplicatus Kira, 1959 
Turrancilla apicalis Kira, 1959 
Baryspira urasima Kira, 1959 
Oliva hirasei Kuroda & Habe, 1952 
Mitropifex hirasei Kira, 1962 
Mitra (Cancilla) yagurai Kira, 1959 
Ancistrosynnx pulchernssima Kira, 1959 
Daphnella nobilis Kira, 1959 
Chelyconus kinoshitai Kuroda, 1956 
Asprella (Conasprella ?) ichinoseana 
Kuroda, 1956 

Euhadra roseoapicalis Kira, 1959 
Euhadra grata gratoides Kira, 1959 
Cadulus (Platyschides) novilunatus 
Kira, 1959 

Entalina majestica Kira, 1959 
Dentalium (Episiphon) candelatum Kira, 
1959 
(41) Dentalium (Pictodentalium) formosum 
hirasei Kira, 1959 



(42) Genus-group name Pictodentalium 

(43) Acila schencki Habe, ^958 

(44) Nuculana (Thestyleda) acinacea Habe, 
1958 

(45) Limopsis tajimae emphaticus K\ra, 1959 

(46) Fragum loochooanum K\ra, 1959 

(47) Clinocardium uchidai Habe, 1955 

(48) Vasticardium compunctum K\ra, 1959 

(49) "Vasticardium serricostatum" 

(50) Leukoma japónica K\ra, 1954 

(51 ) Irus ishibashianus Kuroda & Habe, 1 952 

(52) Solecurtus dunken K\ra, ^959 

(53) Nuttallia solida Kira, 1953 

(54) Heteromacoma oyamai Kna, 1954 

(55) Lanceolaria oxyrhyncha cuspidata Kira, 
1959 



LISTING OF PRINTINGS AND 
REVISED EDITIONS 

Table 1 lists the various printings and edi- 
tions encountered by us. We were unable to 
examine in detail all of the printings for which 
we established printing dates. To properly 
date a copy, locate the bottom line of the date 
information in the colophon (e.g. "..35. .6. .5.."), 
and add 1925 to the first number which is the 
Year of the Showa Era; the second number is 
the month, and the third number is the day. 
This example translates to June 5th, 1960, 
which is the second printing of the second 
edition (the number of printing appears in kanji 
to the right of the last Arabic numeral; begin- 
ning with the tenth printing, the number of 
printing is shown in Arabic numerals). 

Table 1 . List of printings and revised editions 
of Kira's "Coloured Illustrations of the Shells 
of Japan " (CISJ) and "Shells of the Western 
Pacific " (SWPC). Asterisks (*) mark those in 
which taxonomic names were first made avail- 
able. 

(a) Coloured Illustrations of the Shells of 
Japan 

All Japanese editions are cloth-bound oc- 
tavo volumes, with gold lettering in Japanese 
and a muricid-icon imprinted on the spine. 
The First Edition (before 1 959) has a greenish 
cloth binding with photographs of three ranel- 
lid specimens on the front cover; the Second 
Edition has a blue binding featuring a Latiaxis 
photograph on the front cover. The dust cover 
has a small buccinid (Pusiostoma) photo- 



TETSUAKI KIRA'S ILLUSTRATIONS OF SHELLS 



133 



graph on the spine and a Spondylus photo- 
graph in front; the cardboard slip case of the 
First Edition features photographs of Chlamys 



and Pleurotomaria shells, and in the Second 
Edition photographs of an Architectonica and 
two Strombus shells. 



Printing 



Publication date 



Notes 



First Edition 

1 

2 

3 

4 

5 

6 
Second Edition 

1 
* 2 

3 

4 

5 

6 

7-9 
10 
11 
12 
13 

14-15 
16 
17 
18 
19 
20 

21-24 
25 
26 
27 

28-30 
31 



1954a (September 5) 
1954b (November 1) 

1955 (August 15) 

1956 (June 1) 

1957 (June 15) 

1958 (May 1) 

1959 (March 10) 

1960 (June 5) 

9 

1961 (October 1) 
? 

1963 (February 5) 
? 

1965 (August 1) 

1966 (August 1) 
? 

1968 (May 1) 
7 

1971 (March 1) 
7 

1972 (October 1) 
7 

1974 (July 1) 
7 

1979 (July 1) 
7 

1981 (April 1) 
7 

1989 (February 1) 



[not seen] 
[not seen] 
[not seen] 

8 

[not seen] 
[not seen] 

9 

[not seen] 

[not seen] 
[no; seen] 
[not seen] 

[not seen] 

[not seen] 



Notes: 

'Contents {1954a): Japanese title page; [8 pp ] preface and introduction, dated August 1954; [1 p.] schematic drawings; pp. 
1-135 figure captions [p. 1 = series title page], including 67 pis.; [p. 136 blank]; pp. 137-172 discussion of plates in 
Japanese, witfi black-and-white drawings (incl. description of two new taxa); pp. 1-4 index of family names in Latin and 
Japanese; pp. 5-24 index of figured taxa in Japanese. Apparently due to misnumbering, neither this nor any other printing 
of the first edition contains pages 41, 42. or 105, 106. 

^1954b. Changes from preceding (1954a); Correction of many printer's errors, correction of gender in Latin species names, 
re-identifications, change of genus-group names as well as authorships (affecting most plates); renumbering of figures (pis. 
9, 29, 55, 65, 66); correction of transposed figure legends (pis. 27, 37); change of family names (pis. 23, 39); replacement 
of two figures by photographs of different specimens {Oliva emicator, Oliva erythrostoma: pi. 31 figs. 10, 14); change and 
introduction of manuscript names; slight changes in text and indexes reflecting changes in plates. 

■^Title page design changed and title given in English for the first time. The Japanese title did not change, but the translation 
of the title of the first two printings was sometimes rendered as "Illustrations of Japanese Shells in Natural Colour" when 
cited by Japanese authors. Other changes from preceding printings: consecutive page numbering (pp. [8], 1-204), new 
index of generic names in Latin (pp. 177-184); very few corrections/changes in figure captions, most differences due to 
printer's error. 

"Changes from 1955 (1956 not examined): minor correction in text (p. 36, Sulcerato): background color of plates changed 
from light-blue to gray or black (pi. 40). 

^Changes from preceding (1957): Minor technical adjustments (such as color of pi. 64). 

^JapaneseEnglish title page now stating "Enlarged & Revised Edition." Contents: [4 pp ] + [1 p. new foreword in Japa- 
nese); i-vii introduction in Japanese and index of family names in Japanese and Latin; [2 pp.] schematic drawings; pp. 
1-195 figure captions, descriptions in Japanese (including black-and-white photographs and line drawings) plus 1 unnum- 
bered (rare cowries) and 71 numbered pis.; pp. 196-210, introduction to shell collecting (including line drawings and 



134 



BIELER & PETIT 



black-and-white photographs): pp. 211-218 index of generic names m Latin; pp 219-239 index of figured taxa in Japanese 
- [241] colophon. 

Major changes from preceding (1958); Figure captions now containing descriptions; 5 additional color plates (1 unnum- 
bered, pis. 68-71); black-and-white photographs and line drawings in text; many re-identifications and new printer's errors; 
use of additional family names (e.g. Stomatiidae, p. 16, pi. 8); plates renumbered (pis. 16, 50); and many replacements of 
photographs in plates (pi. 4; Tugali gigas: pi. 7; Monodonta perplexa. M. nentoides: pi. 11; Clython sowerbyanus: pi. 12; 
Siliquaria cumingi. Serpulorbis imbncatus: pi. 14; Xenophora corrugata: pi 15; Tibia fusus: pi. 20; Errónea hirasei replaced 
by E. caurica; pi. 21 ; Cassis cornuta: pi. 30; Baryspira urasima. Oliva emicator. О erythrostoma; pi. 50; Sporidylus cruentus. 
addition of Spondylus sanguineus: pi. 55; Tridacna squamosa: pi. 57; Sunetta concinna). 

^Changes from preceding (1959): re-identifications, frequently on the generic level, and corrections of typographic errors for 
most plates. 

^Figures 5-16 on plate 8 have been renumbered to accommodate Stomatia rubra in the Stomatiidae. 

^This IS the first printing that we have seen which gives dates (years only) of previous printings on the verso of the title page. 
It is also the first pnnting seen which has "<<Vol. !>> " added to the title. 



(b) Shells of the Western Pacific in Color 

All English-language editions are larger 
(quarto) cloth-bound volumes, with gold let- 
tering in English and a Lai/ax/s-icon imprinted 
on the front cover. The dust cover features 
color photographs of various gastropod and 
scaphopod shells. An odd feature of ■ • Vol. 
!>>, in contrast to the later -Vol. Il>> by 



Habe not covered here, is that the odd-num- 
bered pages are on the left instead of the 
right. 

In contrast to the many different Japanese 
versions of this work, the English-language 
"editions" do not differ from each other, with 
the exception of minor technical details (such 
as the loss of figure-number 16 on plate 12 
after 1962, probably due to pnnter's error). 



Printing 



Publication date 



Notes 



First Edition 
' 1 
2 
Second (or "Revised") Edition 



Third (or ■'Second") Edition 



1962 (September) 
1965 (Novernber 1) 

1965 (October) 

1966 

1967 

1968 

1970 

1972 (May) 
1975 



[not seen]^ 
[not seen]^ 



Notes: 

"Inscribed '2nd printing, 1965 with copyright date of 1962. 

^The Revised Edition copyright date is October 1965. 

^From dates of printing listed in 1968 printing. 

"Stated to be "Third Edition January 1967," but it bears only the 1962 copynght date and not the copynght date of the 
Second Edition. As the copyright date for a Third Edition' is later given as May 1972,"' this is believed to be an error for 
"Third Printing," but we have no explanation for the 1962 copyright date 

^This is the first printing seen which has "<<Vol. I>> ' appended to the title. 

®The 1972 pnnting is referred to as the "Third Edition" on the verso of the title page, whereas the 1970 and earlier printings 
show Revised Edition." 

''The 1975 printing is again referred to as the "Second Edition" on the verso of the title page. 



Neither the new copyright date (1965), nor 
the term "Revised Edition" are understood by 
us, as the 1962 and 1968 printings we have 



examined differ only in the 1968 (and later) 
being styled "- ■ Vol. I > ■," but the 1967 
printing still omits this addition to the title. 



TETSUAKI KIRAS ILLUSTRATIONS OF SHELLS 



135 



ANNOTATED LISTING OF NEW TAXA 
AND "MANUSCRIPT NAMES" 

List of names that were recognized as ei- 
ther of Kira's or as "manuscript names" in 
Kira's work, following Kira's arrangement of 
molluscan families.^ 

Such names were traced through the print- 
ings and editions available to us. If published 
elsewhere, the proper citation of the original 
description is given in brackets. Printings in 
which the new names were first made avail- 
able are indicated by asterisks (*). It should 
be noted that the listing is not a synonymy but 
reflects a chronological order of figure cap- 
tions. In cases where the first two printings of 
CISJ differ, they are marked as "1954a" or 
"1954b." 



GASTROPODA 



Trochidae 



(1) 'Cantharidus kanekonis Oyama' 
nudum] 



\nomen 



1954-1958 (CISJ): Cantharidus kanekonis 

Oyama, MS.; p. 15, pi. 7, fig. 5 [nom. nud.]. 
1959 (CISJ): Canthandus yokohamensis (Bock); 

p. 13, pi. 7, fig. 5. 
1960-1989 (CISJ): Canthandus (Kanekotro- 

chus) infuscatus (Gould); p. 13, pi. 7, fig. 5. 
1962-1968 (SWPC): Canthandus (Kanekotro- 

chus) infuscatus (Gould); p. 10, pi. 8, fig. 5. 

Taxonomic note: Apparently the name was 
never made available. 

(2) Gaza sericata Kira, 1959 

1954-1958 (CISJ): Gaza sencata Kuroda, MS.; 
p. 16, pi. 8, fig. 13 [nom. nud.]. 

*1959 (CISJ): Gaza sericata Kuroda, MS.; p. 17, 
pi. 8, fig. 13 [with short description in Japa- 
nese]. 

1960-1963 (CISJ): Gaza sencata Kira; p. 17, pl. 
8, fig. 13. 

1962-1968 (SWPC): Gaza sencata Kira; p. 14, 
pl. 9, fig. 12. 

1965-1989 (CISJ): Gaza sencata Kira; p. 17, pl. 
8, fig. 12. 

Turbinidae 

(3) Pseudastralium henicus gloriosum Kira, 
1959 



1954a (CISJ): Pseudastralium henicus gloriosa 
Kuroda et Habe, MS.; p. 20, pl. 10, fig. 2 [nom. 
nud.]. 

1954tD-1958 (CISJ): Pseudastralium henicus 
gloriosum Kuroda et Habe, MS.; p. 20, pl. 10, 
fig. 2 [nom. nud.]. 

*1959 (CISJ): Pseudastralium henicus gloriosum 
Kuroda et Habe, MS.; p. 19, pl. 10, fig. 2 [with 
short description in Japanese]. 

1960-1961 (CISJ): Guildfor or dia [sic] (Pseu- 
dastralium) gloriosa Kira; p. 19, pl. 10, fig. 2. 

1963 (CISJ): Guildforordia [sic] henicus glorio- 
sum Kira; p. 19, pl. 10, fig. 2. 

1962-1968 (SWPC): Pseudastralium henicus 
gloriosum (Kira); p. 18, pl. 11, fig. 2. 

1965-1989 (CISJ): Pseudastralium henicus glo- 
riosum (Kira); p. 19, pl. 10, fig. 2. 

(4) Galeoastraea millegranosa Kuroda & 
Habe in Habe, 1 958 [1 958a, Venus, 20(1 ): 45, 
pl. 3, fig. 1] 

1954-1958 (CISJ): Bolma ? millegranosa 
Kuroda et Habe, MS.; p. 20, pl. 10, fig. 3 [nom. 
nud.]. 

1959 (CISJ): Bolma ? millegranosa Kuroda & 
Habe, MS. [sic]; p. 20, pl. 10, fig. 3. 

1960-1963 (CISJ): Galeoastraea (Harisazaea) 
millegranosa Habe [sic]: p. 20, pl. 10, fig. 3. 

1962-1968 (SWPC): Galeostraea [sic] millegra- 
nosa Habe [sic]; p. 18, pl. 1 1, fig. 3. 

1965-1989 (CISJ): Galeoastraea millegranosa 
Habe [sic]; p. 20, pl. 10, fig. 3. 

Epitoniidae 

(5) Papyriscala bifasciata Kira & Habe in Kira, 
1962 

1954a (CISJ): Epitonium (Papyriscala) hali- 

mense Makiyama; p. 27, pl, 13, fig. 16. 
1954b-1958 (CISJ): Epitonium (Papyriscala) sp.; 

p. 27, pl. 13, fig. 16. 
1959-1963 (CISJ): Epitonium (Papyriscala) sp.; 

p. 31, pl. 13, fig. 16. 
*1962 (SWPC): Papyriscala bifasciata Kira et 

Habe (п. sp.); p. 30, pl. 14, fig. 16. 
1965-1968 (SWPC): Papyriscala bifasciata Kira 

et Habe (п. sp.) [sic]; p. 30, pl. 14, fig. 16. 
1965-1989 (CISJ): Papyriscala bifasciata Kira et 

Habe; p. 31, pl. 13, fig. 16. 

Taxonomic note: Listed in synonymy of '^Pa- 
pyriscala yokoyamai (Suzuki & Ichikawa, 
1936)" by Kuroda et al. (1971: 257). Also 
listed this way by Inaba & Oyama (1977: 27), 
but authorship of P. yokoyamai is correctly 
given as "Suzuki & Ichimura, 1936." 



^Kira's collection Is now located in the Osaka City IVIuseum, Japan (Inaba & Оуагла, 1977: 36). 



136 



BIELER & PETIT 



(6) Cirsotrema (Elegantiscala) ku rodai Kira, 
1960 

1954-1958 (CISJ): Cirsotrema (Elegantiscala) 

sp.: p. 27, pi. 13, fig. 17. 
1959 (CISJ): Cirsotrema (Elegantiscala) sp.; p. 

32, pi. 13, fig. 17. 
*1960 (CISJ): Cirsotrema (Elegantiscala) kurodai 

Kira: p. 32, pi. 13, fig. 17. 
1961-1963 (CISJ): Cirsotrema (Elegantiscala) 

kurodai Kira: p. 32, pi. 13, fig. 17. 
1962-1968 (SWPC): Cirsotrema (Elegantiscala) 

vancosum Kuroda: p. 30, pi. 14, fig. 17. 
1965-1989 (CISJ): Cirsotrema (Elegantiscala) 

rugosum Kuroda et Ito: p. 32, pi. 13, fig. 17. 

Taxonomic note: In the original description of 
Cirsotrema (Elegantiscala) rugosum Kuroda & 
Ito, 1961 , the authors list Cirsotrema (Elegan- 
tiscala) sp. of Kira in the synonymy. Hanshin 
Shell Club (1986: 100-101) does not mention 
Kira as one of those "persons who dedicated 
taxonomic names to Dr. T. Kuroda." 

Strombidae 

(7) Cananum microurceum Kira, 1959 

1954-1958 (CISJ): Cananum microurceum 

Kuroda, fvlS.: p. 31, pi. 15, fig. 5 [nom. nud.]. 
'1959 (CISJ): Cananum microurceum Kuroda, 

MS.; p. 35, pi. 15, fig. 5 [with short description 

in Japanese]. 
1960-1989 (CISJ): Cananum microurceum Kira; 

p. 35, pi. 15, fig. 5. 
1962-1968 (SWPC): Cananum microurceum 

Kira; p. 34, pi. 16, fig. 5. 

Taxonomic note: Cited by Habe & Kosuge 
(1964a: 4) as of Kira (1958). 

Naticidae 

(8) Neverita (Glossaulax) hosoyai Kira, 1959 

*1959 (CISJ); Neverita (Glossaulax) hosoyai 
Kuroda & Kira, MS.; p. 42 [with short descrip- 
tion in Japanese]. 

1960-1989 (CISJ): Neverita (Glossaulax) ho- 
soyai Kira: p. 42. 

1962-1 968b (SWPC): Neverita (Glossaulax) ho- 
soyai Kira: p. 43. 

Taxonomic note: Listed as synonym of Glos- 
saulax didyma (Röding, 1798) by Majima 
(1987: 62). 

Amphiperatidae ' Ovulidae [depending on 
printing] 

(9) Simnia xanthochila Kuroda, 1928 [Venus, 
1(1): pi. 1. fig. 5; Venus, 1(3): 78 (1929)] 



1954a (CISJ): Pellasimnia xanthochila Kuroda, 
MS. [sic]: p. 36, pi. 18, fig. 13. 

1954b-1958 (CISJ): Pellasimnia hirasei xan- 
thochila (Kuroda); p. 36, pi, 18, fig. 13. 

1959-1989 (CISJ): Pellasimnia hirasei xan- 
thochila (Kuroda); p. 44, pi. 18, fig. 13. 

1962-1968 (SWPC): Pellasimnia hirasei xan- 
thochila (Kuroda); p. 45, pi. 19, fig. 13. 

Taxonomic note: Placed in genus Xandaro- 
vula Cate, 1973, by Cate (1973: 35) and 
Azuma (1976: 115). 

Cypraeidae 

(10) Pustularia margarita tetsuakii Kira. 1959 

1954-1958 (CISJ): Pustularia margarita tetuakii 
[sic] Kuroda, MS.; p. 36, pi. 18, fig. 17 [nom. 
nud]. 

'1959 (CISJ): Pustulana margarita tetsuakii 
Kuroda, MS.; p. 45, pi. 18, fig. 17 [with short 
description in Japanese]. 

1960-1989 (CISJ): Pustularia margarita tetsuakii 
Kuroda [sic]\ p. 45, pi. 18. fig. 17. 

1962-1968 (SWPC): Pustularia margarita tet- 
suakii Kuroda [sic]: p. 46, pi. 19, fig. 17. 

Taxonomic note: Considered "Japanese- 
Hawaiian race" of Pustularia cicércula (Linné, 
1758) by Cernohorsky (1967: 72). Listed as 
subspecies of Pustularia (Pustularia) globulus 
(Linné, 1758) by M. Schilder & F. A. Schilder 
(1971: 57). 

(11) Erosaria tomlini maturata Kira, 1959 

1954-1958 (CISJ): Erosana tomlini maturata 

Kuroda, MS.; p. 39, pi. 19, fig. 11 [nom. nud.]. 
'1959 (CISJ): Erosana tomlini maturata Kuroda, 

MS.; p. 47, pi. 19, fig. 1 1 [with short description 

in Japanese]. 
1960-1963 (CISJ): Erosana tomlini maturata 

Kira; p. 47, pi. 19, fig. 11. 
1962-1968 (SWPC): Erosaria tomlini ogasawa- 

rensis Schilder; p. 48, pi. 20, fig. 11. 
1965-1989 (CISJ): Erosana tomlini ogasawaren- 

sis Schilder; p. 47, pl. 19, fig. 11. 

Taxonomic note; Listed in synonymy of Ero- 
saria tomlini ogasawarensis Schilder, 1944, 
by Kuroda et a!. (1971: 105). 

Cassididae [Cassidae] 

(12) Semicassis persimilis Kira, 1959 

1954a (CISJ): Semicassis persimile [sic] Kuroda, 
MS.; p. 43, pl. 21, fig. 3 [nom. nud.]. 

1954b-1958 (CISJ): Semicassis persimilis 
Kuroda, MS.; p. 43, pl. 21, fig. 3 [nom. nud.]. 

'1959 (CISJ): Semicassis persimilis Kuroda, 



TETSUAKI KIRA'S ILLUSTRATIONS OF SHELLS 



137 



MS.; p. 52, pi. 21, fig. 3 [with short description 

in Japanese]. 
1960-1963 (CISJ); Semicassis persimilis Kira; p. 

52, pi. 21, fig. 3. 
1962-1968 (SWPC): Semicassis persimilis Ku- 

roda [sic]; p. 54, pi. 22, fig. 3. 
1965-1989 (CISJ): Semicassis persimilis Ku- 

roda [sic]\ p. 52, pi. 21, fig. 3. 

Taxonomic note: Abbott (1968: 129) lists this 
as synonym of Phalium bisulcatum (Schubert 
& Wagner, 1829) and states "Kira's type of 
persimilis may be lost." 

Bursidae 

(13) Bursa dunkeri Kira, 1959 

1954-1958 (CISJ): Bursa dunkeri Kuroda, MS.; 
p. 43, pi. 21, fig. 18 [nom. nud.]. 

*1959 (CISJ): Bursa dunkeri Kuroda, MS.; p. 54, 
pi. 21, fig. 18 [with short description in Japa- 
nese]. 

1960-1989 (CISJ): Bursa dunkeri Kira; p. 54, pi. 
21, fig. 18. 

1962-1968 (SWPC): Bursa dunken Kira; p. 57, 
pi. 22, fig. 18. 

Taxonomic note: Listed as of Kira (1962) by 
Kuroda et al. (1971: 133). 

Tonnidae 

(14) Eudolium inflatum Kuroda & Habe, 1952 
[Check List, p. 56] 

1954a (CISJ): Eudolium lineatum inflatum 

Kuroda et Habe; p. 44, pi. 22, fig. 4. 
1954b-1958 (CISJ): Eudolium lineatum inflatum 

Kuroda et Habe, MS. [sic]; p. 44, pi. 22, fig. 4. 
1959-1960 (CISJ): Eudolium lineatum inflatum 

Kuroda et Habe, MS. [sic]; p. 55, pi. 22, fig. 4. 
1963 (CISJ): Eudolium lineatum inflatum Kuroda 

et Habe; p. 55, pi. 22, fig. 4. 
1962-1968 (SWPC): Eudolium inflatum Kuroda 

et Habe; p. 59, pi. 23, fig. 4. 
1965-1989 (CISJ): Eudolium inflatum Kuroda et 

Habe; p. 55, pi. 22, fig. 4. 

Taxonomic note: Introduced by Kuroda & 
Habe (1 952: 56) as a new name for Eudolium 
lineatum Schepman as figured by Osima 
(1 943, Conch. Asiat. 1 , pi. 5, fig. 1 ). As Osima's 
figure is accompanied by a description, Ku- 
roda & Habe's (1952) name can be accepted. 

Muricidae 

(15) Murex kiiensis Kira, 1959 

1954-1958 (CISJ): Murex kiiensis Kuroda, MS.; 
p. 47, pi. 23, fig. 10 [nom. nud.]. 



*1959 (CISJ): Murex Kiiensis [sic] Kuroda, MS.; 

p. 58, pi. 23, fig. 10 [with short description in 

Japanese]. 
1960-1989 (CISJ): Murex kiiensis Kira; p. 58, pi. 

23, fig. 10. 
1962-1968 (SWPC): Murex kiiensis Kira; p. 63, 

pi. 24, fig. 10. 

Taxonomic note: Listed as of Kira (1962) by 
E. H. Vokes (1971: 62). 

(16) Ceratostoma (Pteropurpura) vespertilio 
Kira, 1959. 

1954a (CISJ): Ceratostoma (Pteropurpura) ves- 
pertilis [sic] Kuroda, MS.; p. 48, pi. 24, fig. 10 
[nom. nud]. 

1954b-1958 (CISJ): Ceratostoma (Pteropur- 
pura) vespertilio Kuroda, MS.; p. 48, pl. 24, fig. 
10 [nom. nud.]. 

*1959 (CISJ): Ceratostoma (Pteropurpura) ves- 
pertilio Kuroda, MS.; p. 61, pl. 24, fig. 10 [with 
short description in Japanese]. 

1960-1963 (CISJ): Ceratostoma (Pteropurpura) 
vespertilio Kira; p. 61, pl. 24, fig. 10. 

1962-1968 (SWPC): Pteropurpura vespertilio 
Kira; p. 66, pl. 25, fig. 10. 

1965-1989 (CISJ): Pteropurpura vespertilio 
(Kira); p. 61, pl. 24, fig. 10. 

Taxonomic note: Listed as of Kuroda in Kira 
(1955) by E. H. Vokes (1971: 115). 

Rapidae [Coralliophilidae] 

(17) Coralliobia akibumii Kira, 1960 

1954a (CISJ): Coralliobia inflata (Dunker); p. 51, 

pl. 25, fig. 3. 
1954b^1958 (CISJ): Coralliobia sp.; p. 51, pl. 25, 

fig. 3. 
1959 (CISJ): Coralliobia sp.; p. 63, pl. 25, fig. 3. 
*1960 (CISJ): Coralliobia akibumii Kira; p. 63, pl. 

25, fig. 3 [with short description in Japanese]. 
1961-1989 (CISJ): Coralliobia akibumii Kira; p. 

63, pl. 25, fig. 3. 
1 962-1 968 (SWPC): Coralliobia akibumii Kira (n. 

sp.) [sic]; p. 68, pl. 26, fig. 3. 

Taxonomic note: Listed as of Kira (1959) in 
synonymy of Coralliophila inflata (Dunker, 
1847) by Kosuge & Suzuki (1985: 34). 

(18) Coralliophila pyriformis Kira, 1959 

1954-1958 (CISJ): Coralliophila pyriformis 

Kuroda, MS.; p. 51, pl. 25, fig. 12 [nom. nud.]. 
*1959 (CISJ): Coralliophila pyriformis Kuroda, 

MS.; p. 64, pl. 25, fig. 12 [with short description 

in Japanese]. 
1960-1989 (CISJ): Coralliophila pyriformis Kira; 

p. 64, pl. 25, fig. 12. 
1965-1968 (SWPC): Coralliophila pyriformis 

Kira; p. 69, pl. 26, fig. 12. 



138 



BIELER & PETIT 



Taxonomic note: Listed in synonymy of Cor- 
alllophila radula (A. Adams, 1855) by Kosuge 
& Suzuki (1985: 39). 

(19) Latiaxis kawamurai Kira, 1959 

1954-1958 (CISJ): Latiaxis l<awamurai Kuroda, 

MS.: p. 51, pi. 25, fig. 20 [пот. nud.]. 
'1959 (CISJ): Latiaxis kawamurai Kuroda, MS.: 

p. 65, pi. 25, fig. 20 [with short description in 

Japanese]. 
1960 (CISJ): Latiaxis kawamurai Kira; p. 65, pi. 

25, fig. 20. 
1961-1989 (CISJ): Laticxis [sic] kawamurai Kira: 

p. 65. pi. 25, fig. 20. 
1962-1968 (SWPC): Latiaxis kawamurai Kira: p. 

70, pl. 26, fig. 20. 

Taxonomic note: Kuroda, in October 1958 
[Venus, 20(2)], published an illustration and 
figure caption of whiat he intended to describe 
as "Latiaxis kawamurai Kuroda, n.sp." The 
text describing the new species appeared 
only in November 1959 [Venus, 20(4)], eight 
months after Kira had given a description in 
CISJ. Kuroda (1959), recognizing Kira's prior- 
ity, correctly cited "Latiaxis kawamurai Kira, 
1 959" in the text (1 959: 31 9). Placed in genus 
Babelomurexby Kosuge & Suzuki (1985: 14). 

Buccinidae 

(20) Babylonia pallida Kira, 1959 [preoccu- 
pied, replaced by B. kirana Habe, 1965] 

1954-1958 (CISJ): Babylonia pallida Kuroda, 
MS.: p. 52, pl. 26, fig. 28 [nom. nud.]. 

*1959 (CISJ): Babylonia pallida Kuroda, MS.: p. 
69, pi. 26, fig. 28 [with short description in Jap- 
anese]. 

1960-1963 (CISJ): Babylonia pallida Kira: p. 69, 
pl. 26, fig. 28. 

1962-1968 (SWPC): Babylonia pallida Kira: p. 
75, pl. 27, fig. 28. 

1965-1989 (CISJ): Babylonia pallida Kuroda 
[sic]: p. 69, pl. 26, fig. 28. 

Taxonomic note: Placed in synonymy of 
Babylonia kirana n.sp. by Habe (1965: 119), 
who found Kira's name preoccupied by An- 
cilla pallida Perry, 1811 (secondary homon- 
ymy). Altena & Gittenberger (1981: 28-29) 
considered B. kirana as a nomen novum for 
B. pallida Kira, non Perry, and selected the 
shell figured by Kira (1959: pl. 26, fig. 28) as 
the lectotype of B. pallida Kira, and, conse- 
quently, of B. kirana Habe. However, ICZN 
Article 72(e) [a replacement name for a prior 
species-group name has the same name- 
bearing type] demands that "an author pro- 



poses a new species group-name expressly 
as a replacement name for a prior one. ' It can 
be argued that B. kirana Habe was not indi- 
cated "expressly as a replacement name," as 
it is only after Habe (1965: 119) has illustrated 
and descnbed the species and named a type 
specimen, that he mentions that the species 
had previously been named. Altena & Gitten- 
berger (1981: 29) also state that "ß. pallida 
Hirase, 1934, and ß. pallida Kira, 1959, have 
been introduced independently for the same 
species and, therefore, are primary hom- 
onyms and synonyms," and they select the 
specimen figured by Hirase (1934: pl 104, fig. 
9) as the lectotype of B. pallida Hirase. The 
listing and illustration of B. pallida Hirase, 
1934, however does not fulfill the require- 
ments of ICZN Article 13(a) [criteria of avail- 
ability to be satisfied by new names published 
after 1930] and has to be regarded as a 
nomen nudum. The nude name "Babylonia 
pallida Hirase" may have been what Kira 
meant when he first listed "Babylonia pallida 
Kuroda, MS." 



(21) Neptúnea fukueae Kira, 1959 

1954-1958 (CISJ): Neptúnea fukueae Kuroda, 
MS.: p. 55, pl. 27, fig. 4 [nom. nud.]. 

"1959 (CISJ): Neptúnea fukueae Kuroda, MS.; p. 
69, pl. 27, fig. 4 [with short description in Jap- 
anese]. 

1960-1963 (CISJ): Neptúnea fukueae Kira: p. 
69, pl. 27, fig. 4. 

1962-1968 (SWPC): Neptúnea fukueae Kira: p. 
76, pl. 28, fig. 4. 

1965-1989 (CISJ): Neptúnea fukueae Kuroda 
[sic]: p. 69, pl. 27, fig. 4. 

Taxonomic note: Listed as of Kira (without 
date) by Habe & Sato (1973: 2) when they 
made it type species of the new genus Goli- 
kovia. 



(22) Buccinum isaotakii Kira, 1959 

1954-1958 (CISJ): Buccinum leucostoma 

(Lischke): p. 55, pl. 27, fig. 8. 
'1959 (CISJ): Buccinum isao-takii Oyama, MS.; 

p. 70, pl. 27, fig. 8 [with short description in 

Japanese]. 
1960-1989 (CISJ): Buccinum isao-takii Kira; p. 

70, pl. 27, fig. 8. 
1962-1968 (SWPC): Buccinum isaotakii Kira; p. 

76, pl. 28, fig. 8. 



Nassariidae 



TETSUAKI KIRA'S ILLUSTRATIONS OF SHELLS 

Olividae 



139 



(23) Nassahus (Zeuxis) kiiensis Kira, 1959 

1954a (CISJ): Nassahus (Alectrion) kiiensis 

Kuroda, MS.; p. 56, pi. 28, fig. 21 [nom. nud.]. 
1954b-1958 (CISJ); Nassarius (Zeuxis) kiiensis 

Kuroda, MS.; p. 56, pi. 28, fig. 21 [nom. nud.]. 
*1959 (CISJ): Nassarius (Zeuxis) kiiensis 

Kuroda, MS.; p. 73, pi. 28, fig. 21 [with short 

description in Japanese]. 
1960-1963 (CISJ): Nassanus (Zeuxis) kiiensis 

Kira; p. 73, pi. 28, fig. 21. 
1962-1968 (SWPC): Zeuxis kiiensis Kira; p. 81, 

pi. 29, fig. 21. 
1965-1989 (CISJ); Zeuxis kiiensis (Kira); p. 73, 

pi, 28, fig. 21. 

Taxonomic note: Listed in synonymy of Nas- 
sahus castus (Gould, 1850) by Cernohorsky 
(1984: 131). 

Fasciolariidae 

(24) Granulifusus kiranus Shuto, 1958 [Trans. 
Proc. Paleont. Soc. Japan, 31: 258, pi. 38, 
fig. 1] 

1954a (CISJ); Fusinus kirana [sic] Kuroda, MS.; 

p. 60, pi. 30, fig. 3 [nom. nud.]. 
1954b-1958 (CISJ); Granulifusus kiranus 

Kuroda, MS.; p. 60, pi. 30, fig. 3 [nom. nud.]. 
1959 (CISJ); Granulifusus kiranus Kuroda, MS. 

[sic]; p. 77, pi. 30, fig. 3. 
1960-1989 (CISJ); Granulifusus kiranus Shuto; 

p. 77, pi. 30, fig. 3. 
1962-1968 (SWPC); Granulifusus kiranus 

Shuto; p. 85, pi. 31, fig. 3. 

(25) Fusinus gemmuliferus Kira, 1959 

1954-1958 (CISJ); Fusinus gemmuliferus 

Kuroda, MS.; p. 60, pi. 30, fig. 5 [nom. nud.]. 
*1959 (CISJ); Fusinus gemmuliferus Kuroda, 

MS.; p. 77, pi. 30, fig. 5 [with short description 

in Japanese]. 
1960-1989 (CISJ); Fusinus gemmuliferus Kira; 

p. 77, pi. 30, fig. 5. 
1962-1968 (SWPC): Fusinus gemmuliferus Kira; 

p. 85, pi. 31, fig. 5. 

(26) Fusinus crassiplicatus Kira, 1959 

1954-1958 (CISJ); Fusinus crassiplicatus 

Kuroda, MS.; p. 60, pi. 30, fig. 6 [nom. nud.]. 
*1959 (CISJ); Fusinus crassiplicatus Kuroda, 

MS.; p. 78, pi. 30, fig. 6 [with short description 

in Japanese]. 
1960-1989 (CISJ); Fusinus crassiplicatus Kira; 

p. 78, pi. 30, fig. 6. 
1962-1968 (SWPC): Fusinus crassiplicatus Kira; 

p. 85, pi. 31, fig. 6. 



(27) Turrancilla apicalis Kira, 1959 

1954-1958 (CISJ); Turrancilla apicalis Is. Taki, 

MS.; p. 63, pl. 31, fig. 2 [nom. nud.]. 
*1959 (CISJ); Turrancilla apicalis Is. Taki, MS.; p. 

79, pl. 31, fig. 2 [with short description in Jap- 
anese]. 

1960-1989 (CISJ); Turrancilla suavis (Yo- 

koyama); p. 79, pl. 31, fig. 2. 
1962-1968 (SWPC); Turrancilla suavis (Yo- 

koyama); p. 88, pl. 32, fig. 2. 

(28) Baryspira urasima Kira, 1959 

1954a (CISJ); Baryspira hinomotoensis 

(Yokoyama); p. 63, pl. 31, fig. 3. 
1954b-1958 (CISJ); Baryspira urasima Is. Taki, 

MS.; p. 63, pl. 31, fig. 3 [nom. nud.]. 
*1959 (CISJ): Baryspira urasima Is. Taki, MS.; p. 

80, pl. 31, fig. 3. [From here on different shell 
figured! With short description in Japanese]. 

1960-1989 (CISJ): Baryspira urasima Kira; p. 

80, pl. 31, fig. 3. 
1962-1968 (SWPC); Baryspira hinomotoensis 

(Yokoyama); p. 88, pl. 32, fig. 3. 

Taxonomic note: Listed as synonym of 
Baryspira hinomotoensis (Yokoyama, 1922) 
by Kuroda et al. (1971: 195). 

(29) Oliva hirasei Kuroda & Habe, 1952 
[Check List, p. 74] 

1954-1958 (CISJ); Oliva hirasei Kuroda, MS. 

[sic]: p. 63, pl. 31, fig. 8. 
1959 (CISJ): Oliva hirasei Kuroda, MS. [sic]; p. 

80, pl. 31, fig. 8. 
1960-1989 (CISJ); Oliva hirasei Kira [sic]; p. 80, 

pl. 31, fig. 8. 
1962-1968 (SWPC): Oliva hirasei Kira [sic]; p. 

89, pl. 32, fig. 8. 

Taxonomic note: Kuroda & Habe (1952: 74) 
give this name for a figure in Hirase (1909: pl. 
4, fig. 26). As Hirase's work (1909: 45, 46) 
contains a description of the cited figure, 
"Oliva ihsans Lam. Var. ?," the name can be 
accepted as of Kuroda & Habe (1952) [ICZN 
Art. 13(a)(ii)]. Petuch & Sargent (1986) list 
this name in their index as of Kuroda & Habe 
(1952), but in their text as of Kira (1959). 

Mitridae 

(30) Mitropifex hirasei Kira, 1962 

1954a (CISJ): Mitra (Scabricola) japónica A. Ad- 
ams; p. 68, pl. 34, fig. 2. 

1954b-1958 (CISJ); Vexillum (Uromitra) sp.; p. 
68, pl. 34, fig. 2. 



140 



BIELER & PETIT 



1959-1963 (CISJ): Vexillum (Uromitra) sp.: p. 

88. pi. 34. fig. 2. 
*1962 (SWPC): Mitropifex hirasei Kira (n.sp.); p. 

98, pi. 35. fig. 2. 
1965-1968 (SWPC): Mitropifex hirasei Kira 

(n.sp.) [sic]: p. 98, pi. 35. fig. 2. 
1965-1989 (CISJ): Mitropifex fiirasei Kira: p. 88. 

pi. 34, fig. 2. 

(31) Mitra (Cancilla) yagurai Kira, 1959 

1954a (CISJ): Mitra (Cancilla) yagurai Kuroda. 

MS.: p. 68, pi. 34, fig. 3 [nom. nud.]. 
19545-1958 (CISJ): Mitr [sic] (Cancilla) yagurai 

Kuroda, MS.; p. 68, pi. 34, fig. 3 [nom. nud.]. 
*1959 (CISJ): Mitra (Cancilla) yagurai Kuroda. 

MS.; p. 88, pi. 34, fig. 3 [with short description 

in Japanese]. 
1960-1963 (CISJ): Mitra (Tiara) yagurai Kira; p. 

88, pi. 34, fig. 3. 
1962-1968 (SWPC): Tiara yagurai (Kira); p. 98, 

pi. 35, fig. 3. 
1965-1989 (CISJ): Tiara yagurai (K\ra); p. 88, pi. 

34, fig. 3. 

Taxonomic note: Listed as synonym of Mitra 
interlirata Reeve, 1844, by Cernohorsky 
(1970: 46). 

Turridae 



1954a (CISJ): Ancistrosynnx pulcherissima [sic] 
Kuroda, MS.; p. 71, pi. 35, fig. 1 [nom. nud.]. 

1954t>-1958 (CISJ): Ancistrosynnx pulcherns- 
sima Kuroda, MS.; p. 71. pi. 35, fig. 1 [nom. 
nud.]. 

*1959 (CISJ): Ancistrosynnx pulchernssima 
Kuroda: p. 90, pi. 35, fig. 1 [with short descrip- 
tion in Japanese]. 

1960-1963 (CISJ): Ancistrosynnx pulcfierns- 
sima Kira: p. 90, pi. 35, fig. 1. 

1962-1968 (SWPC): Ancistrosynnx pulchernssi- 
mus Kira; p. 100, pi. 36, fig. 1. 

1965-1989 (CISJ): Ancistrosynnx pulchernssi- 
mus Kira: p. 90, pi. 35, fig. 1. 

Taxonomic note: Listed as of Kuroda, 1958, 
by Powell (1966: 42). 

(33) Daphnella nobilis Kira, 1959 

1954a (CISJ): Daphnella nofilis [sic] Kuroda, 
MS.; p. 71, pi. 35, fig. 4 [nom. nud.]. 

1954b-1958 (CISJ): Daphnella nobilis Kuroda, 
MS.; p. 71. pi. 35. fig. 4 [nom. nud]. 

*1959 (CISJ): Daphnella nobilis Kuroda, MS.; p. 
90, pi. 35. fig. 4 [with short description in Jap- 
anese]. 

1960-1989 (CISJ): Daphnella nobilis Kira; p. 90, 
pi. 35, fig. 4. 

1962-1968 (SWPC): Daphnella nobilis Kira; p. 
100, pi. 36, fig. 4. 



Conidae 

(34) Chelyconus kinoshitai Kuroda, 1956 [Ve- 
nus, 19(1): 6, text-fig. 7] 

1954-1955 (CISJ): Floraconus ? kinoshitai 

Kuroda, MS.; p. 75, pi. 37, fig. 18 [nom. nud.]. 

1957-1958 (CISJ): Floraconus ? kinoshitai 

Kuroda. MS. [sic]: p. 75, pi. 37, fig. 18. 

1959-1989 (CISJ): Chelyconus kinoshitai 

Kuroda; p. 97, pi. 37, fig. 18. 

1962-1968 (SWPC): Chelyconus kinoshitai 

(Kuroda); p. 108, pi. 38, fig. 18. 

(35) Asprella (Conasprella ?) ichinoseana 
Kuroda, 1956 [Venus, 19(1): 10, pi. 1, fig. 5] 

1954a (CISJ): Asprella (Conasprella) ichinose- 
ana Kuroda [sic]: p. 76, pi. 38, fig. 3 [nom. 
nud]. 

1954b-1955 (CISJ): Asprella ichinoseana 
Kuroda, MS.; p. 76, pi. 38, fig. 3 [nom. nud.]. 

1957-1958 (CISJ): Asprella ichinoseana Kuroda, 
MS. [sic]: p. 76, pi. 38, fig. 3. 

1959-1989 (CISJ): Asprella (Conasprella) ichi- 
noseana Kuroda; p. 98, pi. 38, fig. 3. 

1962-1968 (SWPC): Asprella (Conasprella) ichi- 
noseana Kuroda; p. 109, pi. 39, fig. 3. 



(32) Ancistrosynnx pulchernssima Kira, 1959 Bradybaenidae 



(36) Euhadra roseoapicalis Kira, 1959 

1954a (CISJ): Euhadra brandtii (Kobelt); p. 132, 
pi. 66, fig. 16. 

1954b^1958 (CISJ); Euhadra brandtii (KobeW): p. 
132, pi. 66, fig. 15 [renumbered]. 

*1959 (CISJ): Euhadra roseoapicalis Kuroda, 
MS.; p. 182, p. 66, fig. 15 [with short descrip- 
tion in Japanese]. 

1960-1963 (CISJ); Euhadra brandti [sic] (Ko- 
belt); p. 182, pi. 66, fig. 15. 

1962-1968 (SWPC): Euhadra brandti roseoapi- 
calis Kira; p. 197, pi. 67, fig. 15. 

1965-1989 (CISJ): Euhadra brandti roseoapica- 
lis Kira; p. 182, pi. 66, fig. 15. 

(37) Euhadra grata gratoides Kira, 1959 

1954a (CISJ): Euhadra grata gratoides Kira, MS.; 

p. 132, pi. 66, fig. 19 [пот. nud.]. 
1954b-1958 (CISJ): Euhadra grata gratoides 

Kira, MS.: p. 132, pi. 66, fig. 17 [renumbered] 

[nom. nud.]. 
*1959 (CISJ): Euhadra grata gratoides Kira; p. 

182, pi. 66, fig. 17 [with short description in 

Japanese]. 
1960-1989 (CISJ): Euhadra grata gratoides 

Kira; p. 182, pi. 66, fig. 17. 
1962-1968 (SWPC); Euhadra grata gratoides 

Kira; p. 197, pi, 67, fig. 17. 



TETSUAKI KIRA'S ILLUSTRATIONS OF SHELLS 



141 



SCAPHOPODA 
Siphonodentallldae 

(38) Cadulus (Platyschides) novilunatus Kira, 
1959 

1954a (CISJ): Gadila novilunata Kuroda, MS.; p. 

80, pi. 40, fig. 2 [nom. nud.]. 
1954b-1958 (CISJ): Cadulus (Platyschides) 

novilunatus Kuroda, MS.; p. 80, pi. 40, fig. 2 

[nom. nud.]. 
*1959 (CISJ): Cadulus (Platyschides) noviluna- 
tus Kuroda, MS.; p. 104, pl. 40, fig. 2 [withi shiort 

description in Japanese]. 
1960-1963 (CISJ): Cadulus (Platyschides) 

novilunatus Kira; p. 104, pl. 40, fig. 2. 
1962-1968 (SWPC): Gadila (Platyschides) 

novilunata (Kira); p. 116, pl. 41, fig. 2. 
1965-1989 (CISJ): Pulsellum virginalis (Boisse- 

vain); p. 104, pl. 40, fig. 2. 

Taxonomic note: Listed as "Gadila novilunata 
Kira, 1959," in synonymy of Platyschides vir- 
ginalis (Boissevain, 1906) by Habe (1964: 
49), and in synonymy of "Cadulus (Poly- 
schides) virginalis Biossevain [sic]" by Habe 
& Kosuge (1964b: 12); listed as "Gadila 
(Platyschides) noviluna [sic] Kira, 1959," un- 
der Polyschides (Platyschides) virginalis by 
Habe (1977: 342). 

(39) Entalina majestica Kira, 1959 

1954-1958 (CISJ): Entalina majestica Kuroda, 
MS.; p. 80, pl. 40, fig. 3 [nom. nud.]. 

*1959 (CISJ): Entalina majestica Kuroda, MS.; p. 
105, pl. 40, fig. 3 [withi shiort description in Jap- 
anese]. 

1960-1963 (CISJ): Entalina majestica Kira; p. 
105, pl. 40, fig. 3. 

1962-1968 (SWPC): Entalina majestica Kira; p. 
116, pl. 41, fig. 3. 

1965-1989 (CISJ): Entalina quadriangularis [sic] 
(Boissevain); p. 105, pl. 40, fig. 3. 

Taxonomic note: Listed as synonym of Enta- 
lina quadriangularis [error for quadrangularis] 
Boissevain, 1906, by Habe (1964; 39; 1977: 
339), and of E. quadrangularis by Habe & Ko- 
suge (1964b: 8). 

Dentaliidae 



Kuroda, MS.; p. 105, pl. 40, fig. 5 [withi short 

description in Japanese]. 
1960-1963 (CISJ): Dentalium (Episiphon) can- 

delatum Kira; p. 105, pl. 40, fig. 5. 
1962-1968 (SWPC): Episiphon candelatum 

(Kira); p. 116, pl. 41, fig. 5. 
1965-1989 (CISJ): Episiphon candelatum (Kira); 

p. 105, pl. 40, fig. 5. 

(41) Dentalium (Pictodentalium) formosum 
hirasei Kira, 1959 

1954-1958 (CISJ): Dentalium formosum Adams 
et Reeve; p. 80, pl. 40, fig. 11. 

*1 959 (CISJ): Dentalium (Pictodentalium) formo- 
sum hirasei Kuroda, MS.; p. 105, pl. 40, fig. 1 1 
[with short description in Japanese]. 

1960-1963 (CISJ): Dentalium (Pictodentalium) 
formosum hirasei Kira; p. 105, pl. 40, fig. 11. 

1962-1968 (SWPC): Pictodentalium formosum 
hirasei (Kira): p. 117, pl. 41, fig. 11. 

1965-1989 (CISJ): Dentalium (Pictodentalium) 
formosum hirasei (Kira); p. 105, pl. 40, fig. 1 1 . 

Taxonomic note: Placed in synonymy of Den- 
talium formosum Adams & Reeve, 1850, by 
Habe (1964: 15; 1977: 332) and Habe & Ko- 
suge (1964b: 4). 

(42) Genus-group name Pictodentalium 

Taxonomic note: With his introduction of Den- 
talium formosum hirasei, Kira (1959 and 
following years, see above) used the genus- 
group name "Pictodentalium," without de- 
scription or indication that it was new. Author- 
ship for this name was credited to Kira (1959) 
by Habe & Kosuge (1964b: 4) and Palmer 
(1974: 118); the latter lists D. formosum 
hirasei as type species. However, Kira's use 
of the name does not fulfill the ICZN require- 
ments [Articles 13(a),(e), criteria of availability 
to be satisfied by new names published after 
1930], and Pictodentalium Kira has to be con- 
sidered a nomen nudum. As ICZN Article 
13(a)(i) does not apply any degree to differ- 
entiation, the name might date from Palmer 
(1974: 118) who lists the name for a group 
comprising "the multicoloured dentaliids. ' 



BIVALVIA 



(40) Dentalium (Episiphon) candelatum Kira, 
1959 

i 1954-1958 (CISJ): Dentalium (Episiphon) can- 
delatum Kuroda, MS.; p. 80, pl. 40, fig. 5 [пот. 
nud.]. 
*1959 (CISJ); Dentalium (Episiphon) candelatum 



Nuculidae 

(43) Acila schencki Habe, 1958 [1958b, Pub!. 
Seto Mar. Biol. Lab., 6(3): 243] 

1954-1958 (CISJ): Acila schencki Kuroda, MS.; 
p. 83, pl. 41, fig. 6 [nom. nud.]. 



142 



1959 (CISJ): Acila schencki Kuroda, MS. [sic]: p. 

107, pi. 41, fig. 6. 
1960-1963 (CISJ): Acila schencki Kira [sic]: p. 

107, pi. 41, fig. 6. 
1962-1968 (SWPC): Acila schencki Kira [sic]: p. 

119, pi. 42, fig. 6. 
1965-1989 (CISJ): Acila schencki Kuroda [sic]: 

p. 107, pi. 41, fig. 6. 

Taxonomic note: Habe (1958b: 243) lists this 
as "Acila schencki Kuroda (MS)," and in syn- 
onymy shows that it is "Acila divaricata sub- 
mirabilis Makiyama" Schenck, 1936, non 
Makiyama, 1926 (p. 151, pi. 12, fig. 9). As 
there is a description of submirabilis in 
Schenck (1936: 88-90), the name A. 
schencki is here considered valid as of Habe 
(1958) [ICZN Art. 13(a)(ii)]. Habe (1977: 15) 
lists it as "Acila (Acila) divaricata schencki 
Kira, 1959.' 

Nuculanidae 

(44) Nuculana (Thestyleda) acinacea Habe, 
1958 [1958b, Pub!. Seto Mar. Biol, Lab., 6(3); 
247] 

1954-1958 (CISJ): Nuculana (Thestyleda?) aci- 
nacea Habe. MS.: p. 83. pi. 41, fig. 9 [nom. 
nud.]. 

1959 (CISJ): Nuculana (Thestyleda?) acinacea 
Habe. MS. [sic]: p. 107, pi. 41, fig. 9. 

1960-1963 (CISJ): Nuculana (Thestyleda?) aci- 
nacea Habe: p. 107, pi. 41, fig. 9. 

1965-1989 (CISJ): Nuculana (Thestyleda) aci- 
nacea Habe: p. 107, pi. 41, fig. 9. 

1962-1968 (SWPC): Nuculana (Thestyleda) 
acinacea Habe: p. 119. pi. 42, fig. 9. 

Limopsidae 

(45) Limopsis tajimae emphaticus Kira, 1959 

1954a (CISJ): Limopsis tajimae emphaticus 
Kuroda. MS.: p. 88, pi. 44, fig. 4 [nom. nud.]. 

1954b-1958 (CISJ): Limopsis tajimae emphati- 
cus Kira [sic], MS.: p. 88. pi. 44, fig. 4 [nom. 
nud.]. 

'1959 (CISJ): Limopsis tajimae emphaticus Kira; 
p. 112, pl. 44, fig. 4 [with short description in 
Japanese]. 

1960-1989 (CISJ): Limopsis tajimae emphaticus 
Kira: p. 112, pl. 44, fig. 4. 

1962-1968 (SWPC): Limopsis tajimae emphati- 
cus Kira: p. 125, pl. 45, fig. 4. 

Taxonomic note; Listed in synonymy of 
Limopsis tajimae Sowerby, 1914, by Kuroda 
et al. (1971; 338) and Habe (1977; 48). 



BIELER & PETIT 

Cardiidae 



(46) Fragum loochooanum Kira, 1959 

1954a (CISJ): Fragum loochooanus Kuroda, 
MS.: p. 108, pl. 54, fig. 13 [nom. nud.]. 

1954b-1958 (CISJ): Fragum loochooanum 
Kuroda, MS.: p. 108, pl. 54, fig. 13 [nom. nud.]. 

*1959 (CISJ): Fragum loochooanum Kuroda, 
MS.; p. 137, pl. 54, fig. 13 [with short descrip- 
tion in Japanese]. 

1960-1989 (CISJ): Fragun [sic] loochooanum 
Kira; p. 137. pl. 54. fig. 13. 

1962-1968 (SWPC): Fragum loochooanum Kira; 
p. 154, pl. 55, fig. 13. 

(47) Clinocardium uchidai Habe, 1955 [Publ. 
Akkeshi Mar. Biol. Stat,, 4:11, pl. 2, figs. 5, 6] 

1954-1955 (CISJ): Clinocardium uchidai Habe, 

MS.; p. Ill, pl. 55, fig. 1 [nom. nud.]. 
1957-1958 (CISJ): Clinocardium uchidai Habe, 

MS. [sic]: p. Ill, pl. 55, fig. 1. 
1959 (CISJ): Clinocardium uchidai Habe, MS. 

[sic]: p. 138, pl. 55, fig. 1. 
1960-1989 (CISJ): Clinocardium uchidai Habe; 

p. 138, pl. 55, fig. 1. 
1965-1 968b (SWPC): Clinocardium uchidai 

Habe; p. 156, pl. 56, fig. 1. 

Taxonomic note; Placed in synonymy of Cli- 
nocardium californiense (Deshayes, 1839) by 
Habe (1977; 172). 

(48) Vasticardium compunctum Kira, 1959 

1954-1958 (CISJ): Vasticardium compunctum 

Kuroda, MS.; p. 1 1 1 , pl. 55, fig. 9 [nom. nud.]. 
'1959 (CISJ): Vasticardium compunctum 

Kuroda, MS.; p. 139, pl. 50, fig. 9 [with short 

description in Japanese]. 
1960-1989 (CISJ): Vasticardium compunctum 

Kira: p. 139, pl. 55, fig. 9. 
1962-1968 (SWPC): Vasticardium compunctum 

Kira; p. 156, pl. 56, fig. 9. 

(49) "Vasticardium serricostatum" 

1954-1958 (CISJ); Vasticardium serricostatum 
Kuroda, MS.; p. 1 1 1 , pl. 55, fig. 1 1 [nom. nud.]. 

1959 (CISJ): Vasticardium serricostatum Kuroda, 
MS.; p. 139, pl. 55, fig. 11 [with short descrip- 
tion in Japanese]. 

1960-1963 (CISJ): Vasticardium serricostatum 
Melvill et Standen [sic], var.; p. 139, pl. 55, fig. 
11. 

1962-1968 (SWPC): Vasticardium okinawaense 
Kuroda; p. 157, pl. 56, fig. 11. 

1965-1989 (CISJ): Vasticardium okinawaense 
Kuroda: p. 139, pl. 55, fig. 11. 

Taxonomic note; The name "serricostatum 
Kuroda" is an error for what was meant to be 
described as a variety of Cardium (Trachycar- 



TETSUAKI KIRA'S ILLUSTRATIONS OF SHELLS 



143 



dium) serricostatum Melvill & Standen, 1899 
(p. 191, pi. 11, fig. 20). It was subsequently 
named by Kuroda (1960: 82) as "Trachy- 
cardium (Acrostehgma) {serricostatum Melvill 
& Standen var?} okinawaense Kuroda (nov.)," 
referring to Kira's illustration (1959: pi. 55, fig. 
11). Peculiarly, Fischer-Piette (1977) lists 
"Vasticardium serricostatum Kuroda" as il- 
lustrated by Kira (1955) in the synonymy of 
Laevicardium (Vasticardlum) flavum (Linné, 
1758), but has "Vasticardium okinawaense 
Kuroda" as illustrated by Kira (1 962), using the 
same figure, in synonymy with Cardium enode 
Sowerby, 1834. 

Veneridae 

(50) Leukoma japónica Kira, 1954 '" 

1954a (CISJ): Leucoma [sic] japónica Kira, MS. 

p. 115, pi. 57, fig. 17 [nom. nud.]. 
*1954a (CISJ): Leukoma japoniac [sic] sp. nov. 

p. 163. 
1954b-1958 (CISJ): Leukoma japónica Kira, MS 

[sic]: p. 115, pi. 57, fig. 17. 
1954b-1958 (CISJ): Leukoma japónica Kira sp 

nov. [sic]; p. 163. 
1959-1963 (CISJ): Leukoma japónica Kira: p 

147, pi. 57, fig. 17. 
1962-1968 (SWPC): Glycydonta marica japó- 
nica (Kira): p. 164, pi. 58, fig. 17. 
1965-1989 (CISJ): Glycydonta marica japónica 

(Kira); p. 147, pi. 57, fig. 17. 

Taxonomic note: As "Leuboma [sic] marica 
japónica Kira, 1954," placed in synonymy of 
Glycodonta marica (Linné, 1758) by Habe 
(1977: 250). 

(51) Iras ishibashianus Kuroda & Habe, 1952 
[Check List: 21] 

1954-1958 (CISJ): Irus ishibastiianus Kuroda et 

Habe; p. 115, pl. 57, fig. 25. 
1959 (CISJ): Irus ishibashianus Kuroda, MS. 

[sic]] p. 148, pl. 57, fig. 25. 
1960-1963 (CISJ): Irus ishibashianus Kuroda et 

Habe: p. 148, pl. 57, fig. 25. 
1962-1968 (SWPC): Notirus ishibashianus 

(Kuroda et Habe): p. 165, pl. 58, fig. 25. 
1965-1989 (CISJ): Notirus ishibashianus 

Kuroda et Habe; p. 148, pl. 57, fig. 25. 

Taxonomic note: Kuroda & Habe (1952: 21) 
introduced this as a new name for "Venerupis 
irus (Linné)," Yokoyama, 1924 (1924: 44, pl. 
2, fig. 23), non Donax irus Linné, 1758. 
Kuroda et al. (1971 : 427), Hanshin Shell Club 
(1986: 45), Inaba & Oyama (1977: 52), and 
Habe (1977: 268) list it as of Kuroda & Habe, 



1952. Habe (1981: 166) lists it as of Kira 
(1959). 

Asaphidae/Psammobiidae [depending on 
printing] 

(52) Solecurtus dunkerl Kira, 1959 

1954-1958 (CISJ): Solecurtus dunken Kuroda, 

MS.; p. 116, pl. 58, fig. 22 [nom. nud.]. 
*1959 (CISJ): Solecurtus dunken Kuroda, MS.; p. 

152, pl. 58, fig. 22 [with short description in 

Japanese]. 
1960-1963 (CISJ): Solecurtus dunkeri Kira; p. 

152, pl. 58, fig. 22. 
1962-1968 (SWPC): Solecurtus dunken Kira; p. 

169, pl. 59, fig. 22. 
1965-1989 (CISJ): Solecurtus dunken Kuroda, 

MS. [sic]; p. 152, pl. 58, fig. 22. 

Taxonomic note: Listed in synonymy of Sole- 
curtus divaricatus (Lischke, 1869) by Habe 
(1977: 224). 

(53) Nuttallia solida Kira, 1953 [Venus, 17(3): 
149, figs. 1c, Id, 2c] 

1954-1958 (CISJ): Nuttallia solida Kira; p. 119, 

pl. 59, fig. 10. 
1959-1989 (CISJ): Nuttallia solida Kira; p. 154, 

pl. 59, fig. 10. 
1 962-1 968 (SWPC): Nuttallia solida Kira; p. 1 71 , 

pl. 60, fig. 10. 

Taxonomic note: Listed in synonymy of Nut- 
tallia japónica (Reeve, 1857) by Habe (1977: 
224). 

Tellinidae 

(54) Heteromacoma oyamai Kira, 1954 

1954a (CISJ): Heteromacoma oyamai Kira, MS.; 

p. 119, pl. 59, fig. 21 [nom. nud.]. 
*1954a (CISJ): Heteromacoma oyamai sp. nov.; 

p. 164. 
1954b-1958 (CISJ): Heteromacoma oyamai 

Kira, MS. [sic]; p. 119, pl. 59, fig. 21. 
1954b-1958 (CISJ): Heteromacoma oyamai Kira 

sp. nov. [sic]; p. 164. 
1959 (CISJ): Heteromacoma oyamai Kira: p. 

155, pl. 59, fig. 21. 
1960-1963 (CISJ): Macoma oyamai (Kira); p. 

155, pl. 59, fig. 21. 
1962-1968 (SWPC): Heteromacoma oyamai 

Kira; p. 172, pl. 60, fig. 21. 
1965-1989 (CISJ): Heteromacoma oyamai 

(Kira); p. 155, pl. 59, fig. 21. 

Taxonomic note: Listed as Heteromacoma 
irus oyamai Kira, 1959, by Habe (1977: 210). 



144 



BIELER & PETIT 



Unionidae 

(55) Lanceolaria oxyrhyncha cuspidata Kira, 
1959 

1954-1958 (CISJ): Lanceolaria oxyrhyncha cus- 
pidata Kuroda, MS.: 127, pi. 63. fig. 18 [nom. 
nud.]. 

"1959 (CISJ): Lanceolana oxyrhyncha cuspidata 
Kuroda, MS.; p. 172, pi. 63, fig. 18 [with short 
description in Japanese]. 

1960-1989 (CISJ); Lanceolana oxyrhyncha cus- 
pidata Kira; p. 172, pi. 63, fig. 18. 

1962-1968 (SWPC); Lanceolana oxyrhyncha 
cuspidata Kira; p. 187, pi. 64, fig. 18. 

Taxonomic note: Listed in synonymy of Lan- 
ceolana grayana (Lea, 1834) by Habe (1977: 
115). 



ACKNOWLEDGEMENTS 

The following made copies of Kira publica- 
tions available or otherwise responded to re- 
quests for data: Dr. Arthur E. Bogan, Dr. Phi- 
lippe Bouchet, Mr. Richard B. Forrer, Dr. 
Michael G. Hadfield, Dr. Richard S. Houbrick, 
Dr. Ronald Janssen, Ms. Anne Joffe, Dr. E. 
Alison Kay, Mr. José H. Leal, Mrs. Jo Anne 
Little, Dr. James H. McLean, Ms. Paula M. 
Mikkelsen, Mr. Edward Nieburger, Mr. David 
M, Pugh, Dr. Robert Robertson, Mr. Walter E. 
Sage III, Mr. Paul H. Scott, Dr. Donald R. 
Shasky, and Dr. Emily H. Vokes. For provid- 
ing translations and other data, we are in- 
debted to Dr. Akihiko Matsukuma, National 
Science Museum, Tokyo, Japan. Also, we 
thank two anonymous reviewers for construc- 
tive criticism on the manuscript. The publish- 
ing company (Hoikusha, Osaka) did not re- 
spond to our inquiries. 



LITERATURE CITED 

ABBOTT, R. T., 1968, The helmet shells of the 
world (Cassidae). Part 1. Indo-Pacific Mollusca, 
2(9): 15-202. 

ALTENA, С. О. VAN REGTEREN & E. GITTEN- 
BERGER, 1981, The genus Babylonia (Proso- 
branchia, Buccinidae). Zoologische Verhandelin- 
gen uitgegeven door het Rijksmuseum van 
Natuurlijke Historie te Leiden, 188; 1-57, pis. 1- 
11. 

AZUMA, M., 1976. Systematic studies on the Re- 
cent Japanese family Ovulidae (Gastropoda) — 
IV. Venus. 35(3); 106-117, pi. 1. 

CATE, С. N., 1973, A systematic revision of the 
Recent cypraeid family Ovulidae (Mollusca; Gas- 



tropoda). The Vellger. 15, Supplement, iv + 116 
pp.. 51 pis. 

CERNOHORSKY, W. O., 1967, Marine shells of 
the Pacific. Pacific Publications Pty. Ltd., Syd- 
ney. 248 pp., illus. 

CERNOHORSKY, W. O., 1970, Systematics of the 
families Mitridae & Volutomitridae (Mollusca: 
Gastropoda). Bulletin of the Auckland Institute 
and Museum. 8; i-iv, 1-190. 

CERNOHORSKY. W. O., 1984, Systematics of the 
family Nassariidae (Mollusca; Gastropoda). Bul- 
letin of the Auckland Institute and Museum. 1 4; iv 
- 356 pp. (incl. 51 pis.). 

FISCHER-PIETTE, E., 1977, Révision des Cardi- 
idae (Mollusques Lamellibranches). Mémoires 
du Museum National d'Histoire Naturelle. Nou- 
velle Série, série A, Zoologie, 101; 1-212, pis. 
1-12. 

HABE, T., 1955, Fauna of Akkeshi Bay XXI. Pele- 
cypoda and Scaphopoda. Publications of the 
Akkeshi Manne Biological Station, 4; 1-31, pis. 
1-7. 

HABE, T., 1958a, On the radulae of Japanese ma- 
rine gastropods (4). Venus. 20(1); 43-60, pis. 
2-3. 

HABE, T., 1958b, Report on the Mollusca chiefly 
collected by the S.S. Sôyô-Maru of the Imperial 
Fisheries Experimental Station on the continental 
shelf bordering Japan during the years 1922- 
1930. Part 3. Lamellibranchia (1 ). Publications of 
the Seto Manne Biological Laboratory. 6(3); 241- 
280, pis. 11-13. 

HABE, T, 1964, Fauna Japónica. Scaphopoda 
(Mollusca). Biogeographical Society of Japan. 59 
pp., 5 pis. 

HABE, T., 1965, Notes on the ivory shell genus 
Babylonia Schlüter (Mollusca). Bulletin of the Na- 
tional Science Museum Tokyo, 8(2); 115-124, 
pi. 1. 

HABE, T., 1977, Systematics of Mollusca in Japan. 
Bivalvia and Scaphopoda. Japan, xiii + 372 pp. 

HABE, T., 1981, A catalogue of molluscs of 
Wakayama Prefecture, the Province of Kii. I. Bi- 
valvia, Scaphopoda and Cephalopoda: based on 
the Kuroda's manuscript and supervised by 
Tadashige Habe. Publications of the Seto Marine 
Biological Laboratory, Special Publications Se- 
ries, 7(1); i-xx, 1-301. 

HABE, T. & S. KOSUGE, 1964a, A list of the Indo- 
Pacific molluscs, concerning to the Japanese 
molluscan fauna (2). Class Gastropoda (Super- 
family Stromboidea). National Science Museum, 
Tokyo. 14 pp. 

HABE, T. & S. KOSUGE, 1964b, A list of the Indo- 
Pacific molluscs, concerning to the Japanese 
molluscan fauna (3). Class Scaphopoda. Na- 
tional Science Museum, Tokyo. 14 pp. 

HABE, T. & J. SATO, 1973, A classification of the 
family Buccinidae from the North Pacific. Pro- 
ceedings of the Japanese Society of Systematic 
Zoology, 8; 1-8, pis. 1-2 ("1972", published in 
1973 teste Inaba & Oyama, 1977). 

HANSHIN SHELL CLUB, 1986, Bibliography of Dr 



TETSUAKI KIRA'S ILLUSTRATIONS OF SHELLS 



145 



Tokubei Kuroda (for commemoration of his 99th 
birthday). Nishinomya, 103 pp., 33 pis. 

HIRASE, Y., 1909, On Japanese marine Mollusca 
(XXV). The Conchological ¡Magazine. 3(2); 41- 
46, pi. 4. 

HIRASE, S., 1934, [See HIRASE, S. 1936]. 

HIRASE, S., 1936, A collection of Japanese shells 
with illustrations in natural colors. Fifth edition. 
Tokyo, pp. 1-14, 1-2, pis. 1-129, pp. 1-217 [The 
1936 printing seen by us is titled "Fifth Edition" 
but contains a 1934 preface. As it is unlikely that 
this work went through five editions in this period 
of time, the "edition" probably means printing. 
We were unable to procure a copy of the 1934 
printing and assume, as did Altena & Gitten- 
berger (1 981 ), that the two printings do not differ 
substantially]. 

[ICZN] International Commission on Zoological No- 
menclature, 1985, International Code of Zoolog- 
ical Nomenclature. Third Edition. London, Berke- 
ley, and Los Angeles, xx + 338 pp. 

INABA, T & K. OYAMA, (compilers), 1977, Cata- 
logue of molluscan taxa described by Tadashige 
Habedunng 1939-1975, with illustrations of hith- 
erto unfigured species (for commemoration of his 
sixtieth birthday). 185 pp. (incl. 7 pis.); Tokyo. 

KIRA, T., 1953, On the Japanese species of the 
genus /Vt7fia///a(Pelecypocla). Venus, 17(3): 144- 
151. 

KIRA, T., 1954, [Coloured illustrations of the shells 
of Japan], [viii] + 172 + 24 pp., 67 pis.; 
Hoikusha, Osaka [title in Japanese; additional 
printings listed in this paper]. 

KIRA, T., 1959, Coloured illustrations of the shells 
of Japan. Enlarged and Revised Edition. [6] + vii 
-b [1] + 239 pp., [1] + 71 pis.; Hoikusha, Osaka 
[additional printings listed in this paper]. 

KIRA, T, 1962, Shells of the western Pacific in 
color, [vii] + 224 pp., 72 pis.; Hoikusha, Osaka 
[additional printings listed in this paper]. 

KOSUGE, S. & M. SUZUKI, 1985, Illustrated cata- 
logue of Latiaxis and its related groups. Family 
Coralliophilidae. Institute of (Malacology of Tokyo 
Special Publication, 1; 18 pp., 50 pis. 

KURODA, T, 1928, [plate only]. The Venus. 1(1); 
pi. 1 (November). 

KURODA, T, 1929, New Japanese shells. I. The 
Venus. 1(3): 77-82 (May). 

KURODA, T, 1956, New species of the Conidae 
(Gastropoda) from Japan. Venus. 19(1): 1-16, pi. 
1. 

KURODA, T, 1958, [plates only]. Venus. 20(2); pis. 
20-21 (October). 

KURODA, T, 1959, Descriptions of new species of 
marine shells from Japan. Venus. 20(4); 317- 
335 (November). 

KURODA, T, 1960, A catalogue of molluscan 
fauna of the Okinawa Islands (exclusive of Ceph- 
alopoda), iv + 104 pp., 3 pis. 



KURODA, T & T. HABE, 1952, Check list and bib- 
liography of the Recent mahne Mollusca of Ja- 
pan. L. W. Satch, Tokyo, 210 pp. 

KURODA, T, T HABE & К. OYAMA, 1971, The 
sea shells of Sagami Bay. Maruzen, Tokyo, xix 
+ 741 pp. [in Jap.], pis. 1-121, 489 pp. [in Engl.], 
51 pp. index, map. 

KURODA, T. & K. ITO, 1961 , Molluscan shells from 
southern Kii. Venus. 21(3); 243-267, pis. 16-18. 

MAKIYAMA, J., 1926, Tertiary fossils from North 
Kankyô-dô, Korea. t\Aemoirs of the College of 
Science. Kyoto Impenal University. (B)2(3); 143- 
160, pis. 12-13. 

MAJIMA, R., 1987, Taxonomic study of Japanese 
species of Glossaulax (Gastropoda; Naticidae). 
Venus. 46(2); 57-74. 

MELVILL, J. С & R. STANDEN, 1899, Report on 
the marine Mollusca obtained during the first ex- 
pedition of Prof. A. С Haddon to the Torres 
Straits, in 1888-89. Journal of the Linnean So- 
ciety of London. Zoology. 27: 150-206, pis. 10- 
11. 

OSIMA, K., 1943, Familia Tonnidae. Conchologia 
Asiática. 1(4); 111-136, pis. 1-5. 

PALMER, С P., 1974, A supraspecific classification 
of the scaphopod Mollusca. The Veliger. 17(2); 
115-123. 

PETUCH, E. J. & D. M. SARGENT, 1986, Atlas of 
the living olive shells of the world. The Coastal 
Education & Research Foundation, Charlottes- 
ville, Virginia, pp. xiii, 253. 

POWELL, A. W. В., 1966, The molluscan families 
Speightiidae and Turridae. An evaluation of the 
valid taxa, both Recent and fossil, with lists of 
characteristic species. Bulletin of the Auckland 
Institute and ¡Museum. 5: 1-184, pis. 1-23. 

SCHENCK, H. G., 1936, Nuculid bivalves of the 
genus /4c/7a. Geological Society of America. Spe- 
cial Papers. 4; i-xiv, 1-149, pis. 1-18. 

SCHILDER, M. & F. A. SCHILDER, 1971. A cata- 
logue of living and fossil cowries. Taxonomy and 
bibliography of Triviacea and Cypraeacea (Gas- 
tropoda Prosobranchia). Institut Royal des Sci- 
ences Naturelles de Belgique ÍMemoires. (2)85: 
1-246. 

SHUTO, T., 1958, Granulifusus from the Miyazaki 
group (Palaeontological study of the Miyazaki 
group — V). Transactions and Proceedings of the 
Palaeontological Society of Japan. N.S. 31 ; 253- 
264, pi. 38. 

VOKES, E. H., 1971, Catalogue of the genus /Wurex 
Linné (Mollusca; Gastropoda); Muricinae, 
Ocenebrinae. Bulletins of American Paleontol- 
ogy. 61(268): 1-141. 

YOKOYAMA, M., 1924, Mollusca from the coral- 
bed of Awa. Journal of the College of Science, 
Tokyo Impenal University, 45(1): 1-62, pis. 1-5. 

Revised Ms. accepted 19 March 1990 



MALACOLOGIA, 1990, 32(1): 147-154 

SPERM STORAGE MECHANISMS AND FERTILIZATION IN FEMALES OF TWO 
SOUTH AMERICAN ELEDONIDS (CEPHALOPODA:OCTOPODA). 

José Angel Alvarez Perez\ Manuel Haimovici^ & Joáo Carlos Brahm Cousin^ 

ABSTRACT 

Octopod species of the genus Eledone do not have spernnathecae in the oviducal glands. 
Sperm masses are found within the ovary, where fertilization takes place. In two South American 
species, Eledone massyae and Eledone gaucha, unusual filamentous structures were observed 
in the animal pole of the oocyte and were entangled greatly with the sperm masses. These 
structures are extensions of the surrounding layers of the oocytes. The sperm penetrate the 
filaments forming agglomerates surrounded by a layer of follicular cells. The filaments shorten 
as the oocyte grows, drawing the enclosed spermatozoa to the ooplasm, in which fertilization 
occurs. These filaments allow sperm to be stored for long periods and might be analogous to 
spermathecae in the oviducal gland of Octopodinae. 

Key words: Cephalopoda, Eledone. fertilization, sperm storage, Octopodidae, Brazil. 



INTRODUCTION 

Female incirrate octopods can store sperm 
for a long time after mating. In two of the four 
Octopodidae subfamilies, the Octopodinae 
and Bathypolypodinae, internal sperm stor- 
age mechanisms and fertilization have been 
described (Petersen, 1959; Froesch & Mar- 
thy, 1975; Wells & Wells, 1977, O'Dor & Ma- 
lacaster, 1983). Less is known about the Ele- 
doninae and Graneledoninae. In Eledone 
cirrhosa. a time lag between copulation and 
spawning has been established in aquaria ex- 
periments (Mangold et al., 1971) but mecha- 
nisms of sperm storage are uncertain 
(Boyle, 1983). 

Eledone massyae Voss, 1964, and Ele- 
done gaucha Haimovici,1988, were described 
recently and are little known. Haimovici & An- 
driguetto (1986) and Haimovici (1988) de- 
scribed morphological differences between 
these species and commented on their appar- 
ent geographical coexistence on the southern 
Brazilian shelf. Some unusual oocyte struc- 
tures were observed in the ovaries of both 
species. In this paper these structures are de- 
scribed and evidence is given for their in- 
volvement in sperm storage and fertilization. 

MATERIAL AND METHODS 

Reproductive organs of nine females of 
Eledone massyae and four females of E. 



gaucha were examined (Table 1). Octopuses 
were collected with a bottom trawl and fixed in 
10% formalin or seawater Bouin solution. 
Oocytes and oviducal glands were dissected 
and embedded in paraffin (58"C) according to 
standard histological techniques (Gabe, 
1968). Longitudinal and transverse sections 
(5 to 7 |jLm) were stained with Harry's hema- 
toxilin-eosin. 

Terms applied to cephalopods and used in 
this paper: 

Spermatangia: Also called sperm sac. 
Evaginated spermatophores; bladders en- 
closing the sperm mass. 
Spermatophore: Complex sperm package 
used for transfer of sperm from male to fe- 
male. 

Spermatheca: Seminal receptacle. Pouch in 
females in which male gametes are stored at 
mating. 



RESULTS 



Oocytes 



Maturing oocytes remain attached by stalks 
to the inner epithelium of the ovary (Fig. 1A). 
Each oocyte is enveloped by three layers: ex- 
ternally, a stratified epithelium of squamous 
cells; internally, a layer of follicular cells; and 
between these, a layer of connective tissue 



^Biology Department, Dalhousie University, Halifax, Nova Scotia, B3H 4J1, Canada 

^Departamento de Oceanografía, Fundaçâo Universidade do Rio Grande, Cx. P. 474, Rio Grande, 96200 RS, Brasil 

^Departamento de Ciencias IVIoríobiológicas, Fundaçâo Universidade do Rio Grande, Cx. P. 474, Rio Grande, 96200 RS, 

Brasil 



147 



148 



PEREZ, HAIMOVICI & COUSIN 



TABLE I. Females of Eledone massyae Voss. 1964. and Eledone gaucha Haimovici. 1988, used for 
histological analyses. 











DML range 


Oocyte length 




Species 


Number 


Locality 


Date 


(mm) 


range (mm) 


Fixative 


E. massyae 


5 


off Rio Grande 
do Sul state 


Oct. 1988 


25.0-65.0 


0.7-10.0 


Formalin 10% 


E. massyae 


4 


off Rio de 
Janeiro state 


Nov. 1988 


64.0-73.3 


3.8-6.6 


Seawater Bouin 


E. gaucha 


4 


off Rio Grande 
do Sul state 


Apnl 1983 


25.0-39.0 


0.5-5.4 


Formalin 10% 


E. gaucha 


1 




Nov. 1983 


36.0 


5.8 


Formalin 10% 



with cells of different sizes and shapes, fibro- 
blasts and blood vessels. The follicular cells 
proliferate and penetrate the ooplasm, form- 
ing longitudinal folds that give the oocyte a 
striped appearance. At final stages of oogen- 
esis these cells secret the chonon. At the 
stalked end of the oocyte the chorion be- 
comes drawn out into a stalk (Fig. 2d). 

At the animal pole, opposite the stalked 
end, there is a conical filamentous projection 
(Fig. 1 B). Filament sizes range from twice the 
length of a 3 mm long oocyte to one-fourth 
that of the 1 1 mm long oocytes. Microscopi- 
cally, the oocyte-surrounding layers are 
drawn out from the oocyte to form these ex- 
tensions (Fig. 3D). The external epithelium 
and the intermediate connective tissue line 
the filament throughout its length. The follicu- 
lar cells of the inner layer differentiate and 
penetrate the filament, filling it as a compact 
tissue. In the initial stages of maturation (Fig. 
ЗА), follicular cells are dispersed irregularly 
and are fusiform with elongated nuclei. As 
maturation advances, these cells become 
regularly dispersed and cuboidal with large 
oval nuclei (Fig. 3B). Finally, in advanced 
stages of maturity, they have smaller nuclei 
with dense chromatin, which suggests a de- 
generation of the tissue (Fig. 3C). 

Oviducal glands 

In the mid-portion of each oviduct is an 
oviducal gland structurally divided into two 
concentric glands around the oviduct and 
separated by a thin sheet of connective tissue 
(Fig. 3E). Spermatheca are absent (Fig. 4), as 
in E. cirrhosa and E. moschata (Froesch & 
Marthy, 1975). The peripheral gland is formed 
by groups of concentric cells with basal nuclei 
and a central lumen; in females close to ma- 
turity their cytoplasm is densely packed with 
reddish grains. The central gland is com- 



posed of 22-23 ducts around the oviduct, 
each lined by an epithelium with ciliated cells 
having superficial nuclei and glandular cells 
having basal nuclei. Spermatozoa were not 
seen to be associated with the oviducts. 

Sperm storage and fertilization 

In females that had mated, spermatangia 
and free sperm were seen within the ovaries. 
As many as seven spermatangia were inside 
an ovary, often attached to the head of the 
spermatophoric tunic (Fig. ID). Free sperm 
masses occurred around the oocytes and 
were much entangled with the oocyte apical 
filaments (Fig. 1 C, 1 D). Sperm were attached 
to the filament tips, and apparently penetrated 
them. Longitudinal sections of apical fila- 
ments of the oocyte showed hair-like masses 
of spermatozoa, regularly dispersed and sur- 
rounded by flattened differentiated follicular 
cells forming dark purple agglomerates along 
the filament (Fig. 3 G, H; Fig. 5). Duhng de- 
velopment, the evolving layers degenerate, 
thus shortening the filament. The filament-en- 
closed spermatozoa are thus drawn to the 
ooplasm, in which ferlilization occurs, proba- 
bly very shortly before spawning (Fig. 2). 

Oocytes in females of Eledone massyae 
bearing spermatangia, free sperm and traces 
of fertilization ranged from 3.0 to 9.0 mm. In 
E. gaucha they ranged from 0.8 to 5.0 mm. In 
the former, the largest oocyte, although 
striped, was 12.0 mm; in the latter, it was 7.5 
mm (Perez & Haimovici, MS). 



DISCUSSION 

The potential for storage of sperm is known 
for octopods of the subfamilies Octopodinae 
and Bathypolypodinae. The sperm released 
into the oviducts after the "spermatophoric re- 



SPERM STORAGE IN SOUTH AMERICAN ELEDONIDS 



149 




FIG. 1 . Photomicrographs of ovary and maturing oocytes of Brazilian Eledone species. A. Cluster of ma- 
turing oocytes. B. General view of ovary. C. Detail of apical filament attachment to free sperm mass. D. 
Ovary of mated female. Spermatangia, spermatophore tunics and free sperm mass present. Arrow indicates 
apical filament attached to sperm mass. Scale bar= 1 mm. af, apical filament; s, stalk; sf, spermatophore 
tunic; sg, spermatangium; sm, sperm mass. 



150 



PEREZ, HAIMOVICI & COUSIN 






FIG. 2. Schematic diagram of fertilization in Eledone massyae and Eledone gaucha, a. Oocyte in early stage 
of maturation. Apical filament is almost twice oocyte length, b. Maturing oocyte. Apical filament as long as 
oocyte and attached to free sperm mass, с Maturing oocyte. Sperm mass penetrates filament and is 
surrounded by modified follicular cells, d. Oocyte in advanced stage of maturation. At this stage, apical 
filament is very short and sperm mass close to ooplasm. Note position at which it will form chorionic stalk. 
ее, external epithelium: fe, follicular epithelium; mfc, modified folicular cells; o, ooplasm; oe, ovanan epi- 
thelium; sm. sperm mass. 



action" (Mann et al., 1970) enter the oviduca! 
glands, where they remain attached to the ep- 
ithelium of the spermatheca (Froesch & Mar- 
thy, 1975). Spermatozoa can be maintained 
as long as ten months, as observed in Bathy- 
polypus arcticus (O'Dor & Malacaster, 1983); 
thus, mating can occur long before matura- 



tion. Mature eggs are fertilized in the lumina 
of the oviducal glands just before spawning. 
There is evidence of sperm storage in both 
European and South Amencan species of 
Eledone. Aquaria observations of Mediterra- 
nean E. cirrosa suggested that sperm might 
be stored for at least six weeks (Mangold et 



SPERM STORAGE IN SOUTH AMERICAN ELEDONIDS 



151 






y^ 










4* *■• 



FIG. 3. Photomicrographs of cross-sections of apical filaments, oviducal glands and oocytes of Brazilian 
Eledone species. A-C. Differentiation of follicular cells at base of apical filament, at successive stages of 
maturation. A,B, scale bar= 20[лт; С, scale bar= ЮОм-т. D. Oocytes in initial maturation stages showing 
oocyte surrounding layers forming apical filaments. Scale bar= 300(j.m. E. Transverse section of oviducal 
glands. Scale bar= 300[лт. F. Apical filament tip in contact with free sperm mass. Scale bar= 140fim. G. 
Longitudinal sections of apical filament showing surrounding layers and enclosed sperm mass. Scale bar = 
SOOfxm. H. Transverse section of apical filament showing three surrounding layers and central spermatozoa. 
Scale bar= 70|лт. eg, central gland; cl, connective layer; ее, external epithelium, fc, follicular cells; fsm, free 
sperm mass; mfc, modified follicular cells; oo, oocyte; ov, oviduct; pg, peripheral gland; sm, sperm mass. 



152 



PEREZ. HAIMOVICI & COUSIN 




FIG. 4. Schematic diagram of longitudinal cut of 
oviducal gland of South American Eledone. c. cen- 
tral cavity of oviducal gland; eg, central gland; dov, 
distal oviduct; pg, peripheral gland; pov, proximal 
oviduct. 



al,, 1971). A maximal lapse of three months 
between copulation and spawning was esti- 
mated for E. massyae (Perez & Haimovici, 
MS), Maturing females bearing spermatangia 
within their ovaries were observed in South 
American species of Eledone and Mediterra- 
nean E. cirrosa (Mangold-Wirz, 1963), In pop- 
ulations of E. cirrosa in the North Sea (Boyle 
& Knobloch, 1983) and E, moschata in the 
Mediterranean (Mangold, 1983) mating was 
assumed to occur just before spawning be- 
cause females beanng spermatangia within 
their ovanes were scarce and nearly mature. 
The absence of spermathecae in the ovi- 
ducal glands seems to be a consistent char- 
acteristic of the genus Eledone since it was 
observed in four species. Spermatophores 
penetrate the oviducts and oviducal glands 
reaching the ovarian cavity, in which the sper- 
matophoric reaction takes place. Sperm 
masses occur either freely around the 
oocytes or enclosed in bladders known as 
spermatangia (Fori, 1937; Mangold-Wirz, 
1963) or sperm sacs (Boyle, 1983; Mangold, 
1986). In Eledone massyae and Eledone 
gaucha, the apical filaments provide a site for 



sperm storage and a fertilization mechanism. 
Modified follicular cells surrounding the sperm 
mass inside the filament are supposed to 
keep spermatozoa viable until oocytes are 
ripe. Whether the apical-filament mechanism 
is an adaptation of the entire genus is still 
unclear. In E, cirrhosa and E. moschata. sites 
and timing of fertilization, as well as the 
means to keep sperm viable, are not known. 
Early descriptions of the reproductive system 
of E, cirrhosa (Insgrove, 1909; Morales, 1958) 
do not mention structures similar to the apical 
filaments. Photographs of ovarian eggs of the 
same species in Mangold-Wirz (1963; plate II, 
d, e, f) show delicate expansions at the ani- 
mal pole of the eggs, although they are quite 
different from those of the South American 
species (Fig. IB), Boyle (pers, comm,) ob- 
served white masses of sperm attached to the 
apical end of each egg. Whether these struc- 
tures indeed form part of the egg-surrounding 
layers is not known. Comparisons could not 
be made with the three remaining but poorly- 
known described species of the genus, E. ca- 
parti Adam, 1950, E thysanophora Voss, 
1962, and E nigra (Hoyle, 1910) from the 
West Afncan coast. 

In species of Eledone that have been stud- 
ied, as well as in most Octopodinae and 
Bathypolypodinae, males mature earlier and 
remain sexually active for a greater part of life 
than do females. The ability of females to 
store sperm means that mating can occur 
long before spawning. Histological study of E. 
massyae and E gaucha shows that at least in 
the South American eledonids, females can 
copulate a considerable time before matura- 
tion and store sperm until fertilization shortly 
before spawning. The apical filament and the 
oviducal glands' spermatheca both allow stor- 
age of sperm and are important facets of the 
reproductive strategy of these octopods. 

There are some slight differences between 
these adaptations, however. In the Octopodi- 
nae and Bathypolypodinae, each mature 
oocyte is fertilized as it descends through the 
lumen of the oviducal gland (Froesch & Mar- 
thy, 1975), Spermatozoa stored in the outer 
parts of the spermathecae will compete to fer- 
tilize the eggs; the last male to copulate with 
the female will be most likely to sire the off- 
spring. In the South American eledonids the 
sperm of the first male to copulate are likely to 
fertilize most of the eggs. This fact could ex- 
plain the difference in body sizes of the sexes 
of E. massyae at matunty (Perez & Haimovici, 
MS), If they can mate successfully, young 



SPERM STORAGE IN SOUTH AMERICAN ELEDONIDS 



153 

9 




FIG. 5. Longitudinal sections of oocyte apical filament of Eledone massyae sfiowing surrounding layers and 
enclosed sperm mass, cl, connective layer; ее, external epithelium; mfc, modified follicular cells; sm, sperm 
mass. Scale bar= 40 fxm. 



males need not grow so large or live so long 
as females. Indirectly this difference in size 
could be particularly advantageous during the 
spawning and brooding seasons because 
males would not be in the same areas at the 
same time with mature females; this separa- 
tion would not only reduce intraspecific com- 
petition for food but also avoid cannibalism on 
the hatchlings. 



ACKNOWLEDGMENTS 

We gratefully acknowledge Paulo A. S. 
Costa for collection of the Rio de Janeiro 
specimens and José Milton Andriguetto Fo, 
Dr. Albino Sakakibara and colleagues of the 
Laboratory of Genetics of Universidade do 
Paraná for valuable help with the photomi- 
crography. We also thank Dr. Sigurd v. Bo- 
letzky and Dr. Peter Boyle for comments on 
the manuscript and anonymous reviewers for 
valuable suggestions for the discussion. 



LITERATURE CITED 

ADAM, W., 1950, Notes sur les céphalopodes XXII. 
Deux nouvelles espèces de la côte Africaine oc- 
cidentale. Bulletin. Institut Royal des Sciences Na- 
turelles de Belgique 26:1-9. 

BOYLE, P. R., 1983, Eledone cirrhose. Pp. 365- 
386 in: BOYLE, P. R., ed., Cephalopod life cy- 
cles. Vol I. Academic Press London. 

BOYLE, P. R. & D. KNOBLOCH, 1983, The female 
reproductive cycle of Eledone cirrhosa. Journal 
of the Marine Biological Association of the United 
Kingdom. 63:71-83. 

FORT, G., 1937, Le spermatophore des céphalo- 
podes. Etude du spermatophore 6'Eledone cir- 
rhosa (Lamarck, 1799). Bulletin Biologique de la 
France et de la Belgique. 71 :357-373. 

FROESCH, D. & H.-J. MARTHY, 1975, The struc- 
ture and function of the oviducal gland in octo- 
pods (Cephalopoda). Proceedings of the Royal 
Society. Senes В, 188:95-101. 

GABE, N., 1968, Techniques histologiques. Mas- 
son et Cie, Paris, 1 1 13 pp. 

HAIMOVICI, M., 1988, Eledone gaucha, a new spe- 
cies of eledonid octopod (Cephalopoda: Octopo- 
didae) from southern Brazil. Nautilus 102(2):82- 
87. 



154 



PEREZ, HAIMOVICI & COUSIN 



HAIMOVICI. M. & J. M. ANDRIGUETTO Fo, 1986, 
Cefalópodes costeiros capturados na pesca de 
arrasto do litoral Sul do Brasil. Arquivos de Bio- 
logía e Tecnología. 29(3):473-495. 

HOYLE. W. E.. 1910. Mollusca: Cephalopoda. Pp. 
261-268 m SCHULTZE, Zoologische und An- 
thropologische Ergebnisse einer Forschungsre- 
ise im westlichen und zentralen Sudafrika. 1903- 
1905. Band 4. Gustav Fischer, Jena. 

INSGROVE.A.. 1909. Eledone. Liverpool Marine 
Biological Committee Memoirs. 18:105 pp. 

MANN. T.. A. W. MARTIN, jr. & J. B. THIERSCH, 

1970. Male reproductive tract, spermatophores 
and spermatophoric reaction in the giant octopus 
of the North Pacific. Octopus dofleini martini. 
Proceedings of the Royal Society. Series B. 1 75: 
31-61. 

MANGOLD-WIRZ, K.. 1963. Biologie des céph- 
alopodes benthiques et nectoniques de la Mer 
Catalane. Vie et Milieu. 13 (suppl.):1-285. 

MANGOLD, К.. 1983, Eledone moschata. Pp. 387- 
400 m: BOYLE, P. R., ed.. Cephalopod life cy- 
cles. Vol. I. Academic Press. London 

MANGOLD, К., 1986, Reproduction. Pp. 157-200 
in: BOYLE. P. R., ed., Cephalopod life cycles. 
Vol. II. Academic Press. London. 

MANGOLD, K.. S. v. BOLETZKY, D. FROESCH. 

1971. Reproductive biology and embryonic de- 
velopment of Eledone cirrhosa (Cephalopoda: 
Octopoda). Marine Biology. 8:109-117. 



MORALES, E., 1958. Sobre la morfología del 
aparato genital de Eledone aldrovandi (Rafin.) = 
Eledone cirrhosa (Lamarck). Commission infor- 
mational pour l'Exploration Scientifique de la Mer 
Méditerranée. Rapports et Procès-Verbaux des 
Reunions. 14:389-394, 2 figs. 

ODOR. R. K. & E. G. MALACASTER. 1983, 
Bathypolypus arcticus. Pp. 401-410. m: BOYLE, 
P. R., ed.. Cephalopod life cycles. Vol I. Aca- 
demic Press. London. 

PEREZ. A. A. P. & M. HAIMOVICI, 1990, Maturation 
and reproductive cycle of Eledone massyae 
(Cephalopoda: Octopodidae) in southern Brazil. 
Bulletin of Marine Sciences, in press. 

PETERSEN. R. P., 1959, The anatomy and histol- 
ogy of the reproductive systems of Octopus bi- 
maculoides. Journal of Morphology. 104, 61-82. 

VOSS. G., 1962, South Afncan cephalopods. 
Transactions of the Royal Society of South Africa 
36:245-272. 

VOSS. G.. 1964, A note on some cephalopods from 
Brazil with a description of a new species of oc- 
topod: Eledone massyae. Bulletin of Manne Sci- 
ence of the Gulf and Caribbean 1 4:51 1-51 6. 

WELLS, M. J. & J. WELLS, 1977, Cephalopoda: 
Octopoda. Pp. 291-336 in: GIESE, A. С & J. S. 
PEARSE, eds.. Reproduction of manne inverte- 
brates Vol. 4. Academic Press, London. 

Revised Ms. accepted 19 March 1990 



MALACOLOGIA, 1990, 32(1): 155-193 

PREDATORY ECOLOGY OF NATICID GASTROPODS WITH A 
REVIEW OF SHELL BORING PREDATION 

Alan R. Kabat 

Museum of Comparative Zoology, Harvard University, 
Cambridge, Massachusetts 02138 U.S.A. 

ABSTRACT 

This review provides a critical synthesis and analysis of the extensive body of knowledge of 
prédation by the Naticidae, a cosmopolitan family of burrowing marine gastropods. First, the 
diversity of shell boring prédation is reviewed and documented for ten taxa (nine marine, one 
terrestrial), in order to facilitate comparative analyses. These predators are: Naticidae, Muri- 
cidae, Cassidae and Capulidae (Gastropoda, Prosobranchia); Okadaia (Gastropoda, Opistho- 
branchia): Aegopinella (Gastropoda, Pulmonata): Octopus (Cephalopoda): Pseudostylochus 
(Turbellaha); Nematoda: and Asemichthys (Pisces). Second, the proximate mechanisms of 
naticid prédation are explicated. Third, the known prey of naticids are tabulated: over 80 families 
of gastropods and bivalves are subject to naticid prédation which is essentially restricted to 
soft-substrate prey taxa. Fourth, the fossil record of naticid prédation is summarized: this pré- 
dation dates from the Cretaceous, with a possible boring "experiment" in the early Triassic. The 
diagnostic countersunk naticid boreholes are recognizable in fossil and Recent faunas: naticid 
prédation is a readily documented aspect of the otherwise elusive soft-bottom food web. Fifth, 
the studies on physiology and ecology of naticid prédation are integrated into a conceptual 
framework. These aspects of naticid prédation (energy budgets, prey size and species choice, 
unsuccessful prédation) indicate a successful albeit rather stereotyped mode of prédation. The 
macroevolutionary implications (escalation, or "arms races ') suggest generalized predator-prey 
coevolution. 

Key words: Naticidae, prédation, boring. 



DIVERSITY OF BORING PREDATION 

In the Mollusca, many of the post-Paleozoic 
Gastropoda are predators, and an extensive 
body of research has developed around var- 
ious aspects of prédation by mollusks (Kohn, 
1 983). Most of these studies treat Recent mol- 
lusks, including the community ecology, be- 
havior and physiology of prédation. Other, 
more restricted, studies on fossils analyzed 
those elements of prédation revealed by fossil 
shells (boreholes and other signs of shell dam- 
age and repair) (Kohn, 1985). Among the 
predatory gastropods, several families include 
shell borers which excavate a hole in the prey 
shell to provide access to the prey flesh. Ear- 
lier overviews of boring by gastropods by Fis- 
cher (1922, 1966), Carhker (1961), Fatten & 
Roger (1968), Sohl (1969), Bishop (1975), 
Boucot (1981: 200 ff.), Bromley (1981), Ben- 
ton (1986) and Vermeij (1987) have summa- 
rized some of this research. More general re- 
views of gastropod feeding biology were 
provided by Ankel (1938), Fretter & Graham 
(1962: 240-262), Taylor et al. (1980), Kohn 
(1983) and Tsikhon-Lukanina (1987). Inevita- 



bly, numerous previous studies have been 
overlooked by subsequent researchers; this 
paper seeks to provide some unity and a co- 
herent framework to the body of knowledge of 
shell boring prédation by gastropods of the 
family Naticidae. 

The objectives of this paper are: (1 ) to doc- 
ument the diversity of shell boring prédation 
and related phenomena; (2) to summarize the 
mechanical or proximate aspects of naticid 
prey capture and boring; (3) to tabulate the 
known naticid prey taxa in order to indicate 
the prey diversity in relation to the overall di- 
versity of marine mollusks; (4) to review the 
fossil record of naticid prédation in the Meso- 
zoic and Cenozoic; and (5) to integrate and 
synthesize the ecological and evolutionary 
aspects of naticid prédation into a broader 
conceptual framework. 

The diversity of molluscan shell boring 
predators is briefly reviewed, in order to be 
able to distinguish amongst the traces of pré- 
dation left by the various taxonomic groups of 
predators. Based on this review, it is obvious 
that prédation by boring in taxa other than the 
Naticidae and Muricidae is seldom studied. 



155 



156 



KABAT 



Shell breaking predators, particularly crusta- 
ceans and fish, represent an entire field of 
study in themselves; valuable reviews are 
provided by Vermeij (1978, 1983c). Not men- 
tioned herein are the diverse groups of sym- 
biotic (non-predatory) epibionts and endolithic 
shell burrowers, such as certain cyanobacte- 
ria, fungi, algae, sponges, polychaetes, sip- 
unculans, barnacles, lithophagid and pho- 
ladid bivalves, brachiopods and bryozoans 
(reviewed by Boekschoten, 1966, and the 
1969 American Zoologist [vol. 9. #3] sympo- 
sium on calcibiocavitology). Generally speak- 
ing, the latter "bore holes" can be recognized 
by their large number on a single shell, the 
lack of complete penetration, and their obvi- 
ous burrowing aspect. An exception is the 
pedicle attachment scar of brachiopods, 
which may show complete penetration in the 
host shell (often another brachiopod); these 
scars or holes (common in the Paleozoic) 
could be confused with those of other, un- 
known. Paleozoic borers. 

Within the Prosobranchia, there are two 
major groups of shell boring (or drilling) pred- 
ators, the Naticidae (Mesogastropoda) and 
the Muricidae (Neogastropoda). I have sum- 
marized only a small part of the extensive re- 
search on muricid prédation, and have limited 
it to the principal means of distinguishing their 
prédation from naticid prédation. A compre- 
hensive review of muricid prédation will be 
most useful but remains to be wntten. 

An heuristic definition of gastropod bore- 
holes was provided by Carriker & Yochelson 
(1968: 2) as "an excavation of charactehstic 
size and form dnlled by a predatory snail in 
the calcareous exoskeleton of a prey organ- 
ism by means of chemical weakening and 
radular abrasion of the prey shell for the pur- 
poses of obtaining food." Refinements of this 
definition were provided by Chatterton & 
Whitehead (1987: 68). Specifically, naticid 
boreholes are parabolic holes (straight or 
oblique), formally referred to as a "truncated 
spherical paraboloid"; the borehole is coun- 
tersunk (i.e., the enlarged outer margin is 
beveled or tapered, forming a chamfer) (Fig. 
1), and incomplete naticid boreholes are 
charactenzed by a prominent central boss 
(rounded elevation) on the bottom surface 
(Fig. 2). 

The Muricoidea (Neogastropoda) is a di- 
verse group containing a variety of eclectic 
predators, including shell borers, carnon 
feeders, and other specialized predators, as 
well as several herbivores. The majority of 



muricids are shell borers and are distin- 
guished by the presence of the accessory 
boring organ (ABO) in the sole of the foot. The 
muricid borehole is cylindrical, with nearly 
straight edges (Fig. 3); the naticid borehole, in 
contrast, has a more parabolic form and bev- 
eled edges. Much of the research carried out 
on the oyster drill, Urosalpinx cinerea, and 
other shellish pests by Carriker, along with 
research on other muricoideans by Taylor, 
has greatly added to our knowledge of the 
feeding biology of this superfamily (Carriker, 
1981; Taylor et al., 1980). 

The Nassariidae, or mudsnails, are carniv- 
orous or scavenging members of the Neogas- 
tropoda. Fischer (1962a: 75) and Reyment 
(1966: 34) stated in passing that nassahids 
are shell borers. Subsequently, Nina (1987; 
23) also mentioned that they probably are 
shell borers. This appears to be mistaken, as 
no documentation has ever been provided for 
boring by mudsnails. Similarly, Stevanovic 
(1950) thought that the boreholes in mollusks 
from the Serbian Upper Miocene were 
caused by the hydrobiid gastropod Sandria 
[= Pseudamnicola] atava: Nina (1987; 25) re- 
jected this conclusion and attributed the bore- 
holes to the naticid Euspira helicina. 

The Cassidae (Tonnoidea, Mesogas- 
tropoda) are important predators of tropical 
echinoids, using sulfuric acid from their pro- 
boscis gland along with the radula to pene- 
trate the echinoid test (by cutting out a disc, 
rather than dniling a hole) (Fig. 4). Hughes & 
Hughes (1981) provided a comprehensive re- 
view of the biology and ecology of cassid pré- 
dation, and pointed out that other tonnoide- 
ans which feed on mollusks do so without 
boring (i.e., by penetrating between the gas- 
tropod operculum and shell, or between the 
valves of a clam). The numerous unique as- 
pects of cassid prédation clearly suggest an 
independent origin from that of naticids or mu- 
ricids. Tertiary echinoids with cassid holes 
were documented by Sohl (1969: figs. 7-8) 
and Beu et al. (1972). 

The Capulidae (Mesogastropoda) are spe- 
cialized ectoparasitic symbionts of mollusks 
and echinoderms. They are known to drill 
holes into the shell of their mollusk host for 
the purpose of obtaining small amounts of flu- 
ids from the host's feeding current for nutri- 
tion. Matsukuma (1978) reviewed shell boring 
by capulids and recorded several fossil 
records of capulid boreholes: these are 
sharp-sided cylindrical holes, similar to those 
produced by muricids. However, capulid 



NATICID PREDATION 



157 



boreholes can be recognized by the sur- 
rounding attachment scar on the host shell, 
where the edge of the capulid shell had 
slightly worn away the host shell (Figs. 5, 6). 

In the Opisthobranchia, the nudibranch 
Okadaia elegans (Vayssiereidae) is known to 
drill holes into the calcareous tubes of serpulid 
and spirorbid polychaete annelids (Young, 
1 969). These minute bore holes (Figs. 7, 8) are 
similar in shape to those of muricids; however, 
muricids are not known to prey on these poly- 
chaetes, whereas Okadaia does not feed on 
mollusks. 

In the Pulmonata, the terrestrial Ae- 
gopinella (Zonitidae) are known as shell- 
boring predators of other gastropods. Mordan 
(1977: 65) described prédation by A. nitidula. 
in which prey snails (typically other zonitids) 
are first attacked through the aperture (fol- 
lowed by consumption of the head-foot); sub- 
sequently, a quite irregular hole on the umbil- 
ical surface of the last whorl is bored (Fig. 1 1 ), 
allowing the predator access to the rest of the 
prey flesh. Pulmonate shell boring may have 
evolved from simple shell "radulation," or the 
scraping of the outer surface of prey shells 
(Mordan, 1977: 70-1). 

In the Cephalopoda, the octopuses are 
shell boring predators of a variety of marine 
shelled mollusks (Ambrose, 1986; Nixon & 
Maconnachie, 1988). Octopus boreholes can 
be recognized by their distinctly irregular or 
oval (but not circular) outline and their ex- 
tremely small inner borehole diameter, in con- 
trast to the large outer borehole diameter 
(Ambrose et al., 1988) (Fig. 9). Furthermore, 
the purpose of the hole is solely for the injec- 
tion of venom to relax or kill the prey, which is 
then extracted through the aperture or valve 
opening. One problem with the analysis of oc- 
topus prédation is that octopuses frequently 
break open the shell or otherwise capture the 
prey without drilling the shell (Ambrose, 1986: 
table 1). Hence, octopus boreholes represent 
only part of their trophic activities. Probable 
octopus boreholes from the Pliocene were re- 
ported by Robba & Ostinelli (1975: 338-344). 

An unusual polyclad turbellarian flatworm, 
Pseudstylochus ostreophagus, is known to 
bore a hole in the shell of juvenile oysters 
(spat), effecting separation (or relaxation) of 
the prey adductor muscle, which causes the 
shell valves to gape, facilitating entry of the 
predator between the valves leading to prey 
consumption. The irregular oval holes are 
quite small (typically 150 x 190 fxm); further 
details are provided by Woeike (1957). Many 



polyclads are known predators of mollusks, 
but shell boring has not been shown for other 
species (Galleni et al., 1980: table 1). 

Nematode worms are known to prey upon 
the microscopic Foraminifera (Granuloreticu- 
losa), boring one or more holes in the test, 
entering the chamber, and slowly consuming 
the prey. In the past, such holes were thought 
to be produced by juvenile gastropods (Livan, 
1937: 149; Saidova & Beklemishev, 1953; but 
see Fischer, 1962a: 70-1); however, their 
size (less than 60 fxm in diameter) is smaller 
than those produced by newly hatched pred- 
atory gastropods (boreholes 100-160 ц.т in 
diameter). Sliter (1 971 ) found that nematodes 
were responsible for this prédation, and illus- 
trated the various borehole morphologies (ir- 
regular oval to bevelled round). Subse- 
quently, Arnold et al. (1985) deschbed even 
larger boreholes (10-125 \хгт\ in diameter) in 
Foraminifera from the Galápagos hydrother- 
mal vent mounds, and concluded that naticid 
gastropods were probably responsible (de- 
spite the fact that naticids are not known from 
such habitats). These are also likely to be the 
product of nematodes. 

Decapod crustacean prédation on mollusks 
is well known, and typically takes the form of 
shell breaking or cracking followed by extrac- 
tion of the prey. Occasionally, the prey is able 
to escape and repair the broken shell, leaving 
diagnostic shell repair scars (Fig. 10) as a 
sign of unsuccessful prédation (Schäfer, 
1972: 408-411; Vale & Rex, 1988). Usually, 
the shell is fragmented; in a few cases, the 
predator may only effect a smaller, very irreg- 
ular hole in the otherwise intact prey shell. 
Papp et al. (1947), provided an extensive dis- 
cussion of crab prédation; subsequent au- 
thors have documented the presence of shell 
fragments or subsequent shell repair attribut- 
able to prédation attempts (successful and 
unsuccessful, respectively) by crabs and 
other decapods. However, because of frag- 
mentation, one cannot account for all the re- 
mains of such prédation. Shell fragmentation 
may also occur because of wave action; 
Cadée (1968: 87-88) noted that this is usu- 
ally accompanied by signs of abrasion and 
fragmentation in subtidal shells is probably re- 
stricted to prédation. 

A most novel recent discovery is that of 
Norton (1988) who documented holes made 
in gastropod shells by a marine cottid fish, 
Asemichthys taylori. This species has a spe- 
cial set of vomeral teeth that are used to 
punch a hole or series of holes in the prey 



158 



KABAT 





iÉSÉ|^n 



À 



FIGS. 1-6. 



NATICID PREDATION 



159 



shell (Fig. 12). The holes (which are not truly 
"bored") allow the entry of digestive enzymes 
while the shell is in the digestive tract of the 
fish. Shells which are unpunched generally 
pass through undigested and emerge alive 
(except, of course, for limpets which have an 
exposed ventral aspect). Similar rows of 
punctures in Paleozoic brachiopods, conulari- 
ids and nautiloids were attributed to shark 
prédation (Mapes et al., 1989, and references 
therein). 

Shell boring or burrowing is little known in 
the freshwater environment, with a few ex- 
ceptions, such as the endolithic burrowing 
polychaete Caobangia (Jones, 1969). Re- 
cently, the Soviet paleontologist llina (1987) 
found shells of Unio and Viviparus (fresh- 
water mollusks) with regular, round borehgles, 
one to four per shell, with an outer diameter 
up to 2 mm and an inner diameter from 1 .0 to 
1 .5 mm. llina (1987: 29) suggested that these 
holes were made by ". . . ants that for reasons 
not yet known use their formic acid to etch 
perforations in the shells of molluscs . . ."; 
E. O. Wilson {in litt.) stated that "I don't know 
of any documented cases of ants boring mol- 
lusk shells, and I doubt very much if they do 
. . . it's hard to imagine their cutting through a 
clam shell even with the aid of formic acid." In 
any case, since ants are terrestrial, it seems 
unlikely that these freshwater mollusks were 
drilled and consumed in situ; it is more likely 
that empty shells were washed ashore and 
(post-mortem) excavated by some other or- 
ganism, perhaps for a refuge. Further study is 
clearly indicated. 

Finally, there is an extensive and scattered 
literature on shell borings in Paleozoic fossils. 
While providing lengthy descriptions of the 
bore holes and of the prey organisms, these 
studies generally have not elucidated the na- 
ture of the predator (known predatory gastro- 



pods did not evolve until the Mesozoic). Car- 
riker & Yochelson (1968) suggested that 
these holes were made by soft-bodied, ses- 
sile, non-predatory organisms of unknown 
taxonomic affinity (this hypothesis is essen- 
tially non-testable!); Sohl (1969: 728-9) fur- 
ther discussed this problem. More recently. 
Smith et al. (1985) and Chatterton & White- 
head (1987) reviewed the Paleozoic bore- 
holes and suggested that they were, indeed, 
predatory in origin although the identity of the 
predator remains unknown. Vermeij (1987: 
176-7) hypothesized that ectoparasitic pla- 
tyceratid gastropods (ecologically analogous 
to capulids) were the Paleozoic borers. 

The remainder of this paper is restricted to 
analysis of prédation by naticids. The preced- 
ing review of the diversity of shell borers in- 
dicates that prédation by boring has evolved 
independently in a number of taxa; any simi- 
larities are undoubtedly cases of convergent 
evolution. The following section, on the prox- 
imate mechanisms, demonstrates the numer- 
ous unique (dehved) aspects of naticid pré- 
dation, and should be compared with what is 
known for other shell-boring taxa. 



MECHANISMS OF NATICID PREDATION 

For a detailed review and critique of the 
previous morphological studies on naticid 
feeding mechanisms, see Carhker (1981). 
Essentially, early controversies concerning 
naticid boring involved the means of boring: 
i.e., was it solely by mechanical means (rad- 
ular rasping of the prey shell) or did it also 
involve chemical action (acid secretion). It 
was the careful work of Carriker and col- 
leagues (Carhker, 1981) which demonstrated 
that the latter hypothesis is the case for nati- 
cids and muricids. 



FIG. 1. Naticid bore hole (complete) in valve of Dosinia discus (Reeve, 1850) [Cocoa Beach, Florida; MCZ 

145801]. Shell dimensions 52.7 mm ■ 48.8 mm; outer bore hole diameter 5.2 mm; inner borehole diameter 

2.8 mm. 

FIG. 2. Naticid bore hole (incomplete) in valve of Dosinia concéntrica (Born, 1778) [Punta Guanajibo, Puerto 

Rico; MCZ 212607]. Shell dimensions 55.7 mm - 52.3 mm; outer bore hole diameter 2.7 mm. 

FIG. 3. Muricid bore holes [presumably by Urosalpinx or Eupleura] in adjacent valves of Crassostrea 

virginica (Gmelin, 1 791 ) [Stono River, South Carolina; MCZ 226338]. Shell lengths 86 mm and 65 mm; outer 

bore hole diameter 2.5 mm; inner bore hole diameter 2.3 mm. 

FIG. 4. Cassid bore hole in Cassidulus pacificus (A. Agassiz, 1863) [Punta Pescadero, Baja California Sur, 

Mexico; USNM 32907]. Test dimensions 34.9 mm • 28.9 mm, height 1 6.1 mm; bore hole diameter 2.1 mm. 

FIG. 5, 6. Capulus danieli (Crosse, 1858) bore hole in valve of Comptopallium vexillum (Reeve, 1853) 

[Noumea, New Caledonia; ANSP 272383]. Scallop shell dimensions 32.5 mm ^ 29.5 mm; outer bore hole 

diameter 1 .75 mm; capulid shell dimensions 4.9 mm ■ 15.0 mm. 



160 



KABAT 




FIG. 7, 8. Okadaia elegans Baba, 1930 [Nudibranchia] bore hole in tube of spirorbid polychaete [Oahu, 

Hawaii]. Bore hole diameter ca. 115 i^m; worm tube diameter at bore hole ca. 300 |хт. SEM photographs 

courtesy J. D. Taylor. [Magnifications; Figure 7 at 1 10 -- ; Figure 8 at 350 ■ ]. 

FIG. 9, Octopus bimaculatus Verrill, 1883 bore hole in Ventricolaria ford; (Yates, 1890) [Anacapa Island, off 

Ventura, California; MCZ 298337]. Shell dimensions 33.7 mm -- 31 .2 mm; outer bore hole diameter 2.2 mm, 

inner bore hole diameter 0.6 mm. Specimen courtesy R. F. Ambrose. 

FIG. 10. Unsuccessful crustacean prédation: shell repair scars in Architectonica nobilis Röding, 1798 [Puerto 

Plata. Dominican Republic; MCZ 106825]. Shell dimensions 8.8 mm y 17.5 mm. 



A fundamental and little studied problenn 
concerns the methods by which naticids de- 
tect their prey. For many predatory gastro- 
pods, chemoreception (detection of prey 



"chemical odors" by the osphradium) is typi- 
cally the initial mechanism for determining the 
presence and direction of potential prey 
(Kohn, 1961; Croll, 1983). With infaunal nati- 



NATICID PREDATION 



161 



cids, the sediment habitat not only decreases 
the diffusion rate of chemical substances, but 
also may perturb its directionality; hence nat- 
icids may forage with the siphon extending to 
the surface where diffusion is more direct and 
rapid. Kitching & Pearson (1981) found that 
the Australian "Polinices'^ [= Conuber] incei 
responded to artificial sound waves directed 
through the substrate, which presumably 
mimicked the vibration of burrowing prey. 
Mechanoreception may well serve as an ad- 
ditional prey detection mechanism for the nat- 
icids. 

Regardless of how the prey are initially de- 
tected, one can analyze the behavioral per- 
spective: namely, recognition of suitable prey 
serves as a releasing mechanism which elic- 
its a stereotyped sequence of behaviors [ = 
fixed action patterns] (Ansell, 1960). Naticids 
have been little studied with respect to clas- 
sical ethological principles, probably because 
most activity occurs while they are buried. 

Edwards (1969), Schäfer (1972: 242-3), 
Stenzler & Atema (1977) and Hughes (1985) 
discussed the sequence of prey capture 
events: the prey is detected, evaluated, 
seized, covered and immobilized with copious 
pedal mucus, wrapped in the dilated foot of 
the naticid, dragged for some distance, and 
finally carried deep into the sand for com- 
mencement of boring. 

The mechanism of naticid boring involves a 
complex sequence of events. There is alter- 
nate application of the predator's radula and 
accessory boring organ (ABO) to the bore 
hole site on the prey shell. The ABO is found 
on the ventral surface of the proboscis in nat- 
icids (but in the sole of the muricid foot); the 
two ABO types represent a case of conver- 
gent evolution and no homologues in other 
taxa are known. The ABO histology was de- 
scribed by Bernard & Bagshaw (1969), who 
characterized it as a "fungiform papilla" con- 
taining numerous epithelial secretory cells. 
The biochemistry of ABO secretions was dis- 
cussed by Carriker & Williams (1978). The 
ABO secretes a complex mixture of pre- 
sumed enzymes, chelators, and inorganic acid 
(HCl) in a saline, hypertonic solution which 
effects dissolution of the prey shell layers 
(both calcareous and organic matrix). During 
boring, the proboscis becomes engorged, 
everting both the radula and the ABO. The 
radula is protracted and scrapes at the sur- 
face of the bore hole. The proboscis is rotated 
in 90° sectors and the scraping is from the 
outer edge to the center, resulting in the di- 



agnostic boss in the center of incomplete bore 
holes (Ziegelmeier, 1954: fig. 7; Carriker, 
1981: 410). The prey shell fragments are in- 
gested but subsequently excreted without di- 
gestion (Carriker, 1 981 : 41 1 ). The prey tissue 
is ingested by the proboscis through the bore- 
hole; Reid & Gustafson (1989) determined 
that external digestion does not occur. 

Most studies have documented that natic- 
ids capture and consume their prey entirely 
within the sediment. Previous reports of nati- 
cid prédation on the sediment surface were 
usually a result of aquaria studies wherein the 
sediment depth was too shallow and conse- 
quently abnormal behavior patterns were 
manifested. Recently, field observations of 
Natica gualteriana from the Philippines 
(Savazzi & Reyment, 1989) have docu- 
mented that this species was capable of 
searching for and capturing its prey on sand 
bars at low tide (i.e., while exposed to the air). 
Further study is needed to ascertain whether 
other naticid species can also feed on the 
sediment surface (exposed or subtidally). As 
such, this would result in greater competitive 
interactions between those naticids and the 
epifaunal muricids. 

For temperate and boreal naticids, the wa- 
ter temperature can determine the active pe- 
riods of feeding. Hanks (1953) showed that 
the northwest Atlantic Neverita duplicata and 
Euspira heros had a marked temperature- 
dependence, with no feeding at temperatures 
below 5°C and 2°C, respectively. Similarly, 
salinity (brackish or estuarine waters) also af- 
fects feeding rates; these two naticid species 
did not feed at artificial salinities below 10%o 
(normal seawater about 35%o). 

For the calculation of energy budgets, the 
rates of shell boring and of prey tissue inges- 
tion must be determined. Determining the 
time for infaunal prey capture and subjugation 
would be extremely difficult and yields vari- 
able results (here, especially, aquaria studies 
would be of little value). In general, the rela- 
tive sizes of predator and prey (both dimen- 
sional and shell thickness) must be taken into 
account; there will undoubtedly be great inter- 
specific variation in these rates. Ziegelmeier 
(1954) found a boring rate of 0.6 mm/day, or 
0.025 mm/hour by Euspira nitida. Similarly, 
Kitchen et al. (1981: fig. 2) observed that in 
Neverita duplicata preying on various bi- 
valves, the boring rate was a nearly constant 
0.0223 mm/hour, regardless of prey species, 
predator size, or elapsed time. Bayliss (1986) 
noted that for Mya and Spisula prey, Euspira 



162 



KABAT 




FIGS. 11, 12. 



NATICID PREDATION 



163 



alderi bored at an average rate of 0.0097 mm 
per hour; the prey tissue was consumed in 
19.5 hours [M. arenaria), 21.5 hours (S. sub- 
truncata), or 60 hours (S. elliptica). 

For the analysis of naticid boring prédation, 
especially in fossils, the primary source of 
data for the predator is the size of the bore- 
hole. Kitchen et al. (1981: 539, fig. 4) proved 
that the borehole diameter is constant for a 
given predator size, regardless of the prey 
size. Most studies have used the inner bore- 
hole diameter as the basis for analysis, as this 
represents the size of the predator's probos- 
cis. Wiltse (1980a: 189, fig. 1) used the diam- 
eter ". . . at the junction of the prismatic and 
nacreous shell layers"; this does not facilitate 
comparisons with other prey taxa (given that 
the depth of this junction is not constant for all 
taxa). Usually, the outer borehole diameter is 
also directly proportional to the predator size; 
but due to the chamfered borehole edge, it is 
more difficult to measure. However, for cor- 
bulid bivalve prey, there is an exception in 
that the outer borehole is disproportionately 
much larger than the inner borehole; this re- 
flects the conchiolin layer in the prey shell (De 
Cauwer, 1985). Arua & Hoque (1989b), 
based solely on analysis of outer borehole 
sizes, concluded that the opening was more 
oval than circular; regrettably, their data on 
inner borehole sizes was not presented. 

It is unfortunate that a recent paleoecolog- 
ical study (Arua & Hoque, 1989a, 1989c) 
seems to have confused several muricid 
boreholes with those of naticids, and vice 
versa. Their "hole types" A, В and D were 
claimed to be muricid; C, E and F as naticid. The 
authors had stated that naticid boreholes are 
countersunk, with tapering sides, and incom- 
plete ones have a central boss; yet, they 
claimed that their "hole type E," which lacks a 
boss and has vertical sides, was naticid! My 
re-analysis of their descriptions leads to the 
conclusion that their "hole types' E and 
(maybe) A are muricid; whereas B, C, D, and 
F are naticid. This confusion undoubtedly has 
arisen in other studies, and should be consid- 
ered when interpreting community-level anal- 



yses (because the variety of observed bore- 
holes are rarely illustrated therein). 

A more general aspect of naticid prédation 
is the suitability of the substrate for naticid 
locomotion. It is well known that naticids are 
restricted to infaunal sedimentary habitats; 
it is less appreciated that extremely fine 
or smooth grained substrates (silt-mud-clay) 
are precluded because they are too tightly 
packed to burrow through readily, in contrast 
to coarser sand substrates (Yochelson et al., 
1983: 12; Maxwell, 1988: 31). 

Vermeij (1980) and Ansell & Morton (1987) 
discovered that the tropical Polinices "tu- 
midus" [= mammilla], after wrapping its prey 
in a mucus coat within the foot, retained the 
prey until suffocation and gaping occurred. 
Subsequently, the prey was consumed with- 
out boring. Ansell & Morton (1987: 117) sug- 
gested that a "narcotizing toxin" may play a 
role in causing prey gaping, such as by thai- 
dine gastropods preying on barnacles. This 
was questioned by Reid & Gustafson (1989), 
who determined that prey suffocation alone 
caused shell gaping. The ecological and ev- 
olutionary implications of this non-boring pré- 
dation will be discussed below. 

A preposterous view of the evolution of nat- 
icid feeding mechanisms was advanced by 
Stafford (1988), who claimed that naticids 
originated at Ediacaran-Cambhan times (570 
million years ago), as swimming filter feeders, 
and gradually shifted to benthic feeding en- 
tailing eversión of the stomach (as in aster- 
oids) to effect external digestion of the prey. 

To summarize the proximate mechanisms 
of naticid shell boring: (a) Prey are detected 
by chemoreception using the osphradium, 
though mechanoreception may also play a 
role, (b) Suitable prey are seized, covered 
with pedal mucus and wrapped in the foot, (c) 
The proboscoideal acid-enzyme secretory ac- 
cessory boring organ (ABO) together with the 
radula is used to excavate a countersunk 
(bevelled) hole in the prey shell, and the prey 
tissues are extracted through this borehole. 
The size of the borehole (inner diameter) is 



FIG. 11. Aegopinella nitidula (Draparnaud, 1805) [Zonitidae] bore holes in (left) A. pura (Alder, 1830) [bore 
hole 1.5 mm ■ 0.7 mm] and (right) A. nitidula [bore hole 1.6 mm ■ 1.0 mm] [Monks Wood, England]. 
Photographic negative courtesy P. B. Mordan; original in the Biological Journal oí the Linnean Society 
(1977), 9: 65, plate 1A. [Copyright 1977 by The Linnean Society of London]. 

FIG. 12. Asemichthys taylorl Gilbert, 1912 [Pisces], punched holes in Margante sp. [San Juan Island, 
Washington]. Shell width ca. 2 mm. Maximum hole diameters: 165 |лт; 350 |а.т: 380 fj.m. SEM negative 
courtesy S. F. Norton; original in Science (1988), 241(1): cover. [Copyright 1988 by the AAAS]. 



164 



KABAT 



positively correlated with predator size, (d) 
Some tropical Indo-Pacific naticids are able to 
immobilize their bivalve prey until shell gaping 
occurs, allowing direct access to the prey tis- 
sues; thus, no borehole need be made. 



THE PREY OF NATICIDS 

The Appendix tabulates the known prey of 
naticid gastropods (fossil and Recent). The 
genera are arranged alphabetically by family; 
the reference is given in brackets following 
the species name [n.b. this is not the author of 
the taxon!]; some species were reported in 
several studies but only one such is indicated 
herein. This compilation includes an unpub- 
lished data set on Fijian Pleistocene mollusks 
collected by A. J. Kohn. I have corrected for 
obvious changes in generic nomenclature; 
species names were not given for several re- 
ports, as indicated by an asterisk. Many 
records of naticid prédation are purely inci- 
dental or even parenthetical (e.g., "by the 
way, some of the shells of X were bored . . ."), 
which does not facilitate critical comparative 
analyses. 

Generally, the records herein are limited to 
ecological or paleoecological studies empha- 
sizing prédation; it is too time-consuming to 
search through the general systematic and 
faunistic literature for scattered records of 
naticid prédation (which are usually not thor- 
oughly documented in such papers). Need- 
less to say, aquarium studies of naticid feed- 
ing should be based on prey found in the 
same habitat as naticids. Unfortunately, some 
papers (Hayasaka, 1933; Fischer, 1966; 
Sander & Lalli, 1982; and De Cauwer, 1985) 
provided lists of taxa with gastropod bore- 
holes, but without specifying naticid or muhcid 
boreholes. Nonetheless, based on the avail- 
able data, it appears that naticids prey on the 
majonty of benthic, infaunal shelled mollusks. 

A. Class Gastropoda 

Since most archaeogastropods (e.g. Pleu- 
rotomaroidea, Fissurelloidea and Patelloidea) 
are rocky-habitat dwellers, they are not sub- 
ject to naticid prédation. Beebe (1932; 212, 
fig.) made the unusual statement that, in Ber- 
muda, Natica canrena preyed upon the rocky 
intertidal limpet Fissurella barbadensis. leav- 
ing a diagnostic borehole in the limpet shell. 
My subsequent re-analysis of this situation re- 
veals that Beebe had confounded the excur- 



rent slit or foramen ("keyhole") of these lim- 
pets with naticid boreholes and erroneously 
assumed that naticid prédation was responsi- 
ble for the limpet keyholes! 

Many of the soft-substrate taxa in the Me- 
sogastropoda are subject to naticid préda- 
tion. Not included herein are the extensive re- 
ports of confamilial prédation on naticids 
themselves (sometimes referred to as "can- 
nibalism") (Kabat & Kohn, 1986). Reports 
of naticid boreholes in Xenophora [Xeno- 
phoridae] and Lamellaria [Lamellariidae] by 
Adegoke & Tevesz (1974) are questionable, 
given the epifaunal habitat of these taxa. 
While it may appear that neogastropod gen- 
era are more frequent in the list, this could be 
a taxonomic artifact of generic lumping vs. 
splitting. 

Most of the neogastropods are active pred- 
ators themselves; the epifaunal and rocky- 
habitat species generally escape naticid pré- 
dation. It is entirely possible that some of 
these records, especially of Muhcidae, are of 
misidentified muricid boreholes. 

B. Class Bivalvia 

Most infaunal bivalves are subject to naticid 
prédation. In particular, the venerids, tellinids, 
and lucinids (the last two often with relatively 
thin or little-sculptured shells) are frequent 
victims. The infaunal Solemyidae live in re- 
ducing sediments where naticids are not 
found. Bivalve taxa that are in rocky habitats, 
epifaunal byssate or cemented (Dimyoidea, 
Plicatuloidea, Anomioidea, Chamoidea, Lep- 
tonoidea and Cyamioidea) effectively escape 
naticid prédation; the few cases of naticid 
boreholes in the Pterioidea, Limoidea, Os- 
treoidea and Pectinoidea are unusual excep- 
tions. Those that are rock or wood burrowers 
(Lithophagidae, Gastrochaenoidea and Pho- 
ladoidea) are also inaccessible to naticids. 
The Pinnoidea and Tridacnoidea have en- 
crusted and sculptured shells; the Glos- 
soidea, Clavagelloidea and Pholadomyoidea 
are too rare to have been reported in this con- 
text. 

С Class Scaphopoda 

A thorough review of naticid prédation on 
scaphopods by Yochelson et al. (1983) found 
that scaphopods were the occasional prey of 
naticids from the Late Cretaceous to the Re- 
cent. Usually, there is moderate stereotypy of 
borehole siting, with most being laterodorsal 



NATICID PREDATION 



165 



and about midway along the shell axis. It was 
found that coarse-ribbed scaphopods (which 
live in coarse sediments) were much more 
likely to be bored; those with smooth (or no) 
ribs, living in fine sediments, escaped naticid 
prédation by virtue of their habitat which is 
inimical to active naticid burrowing (Yoch- 
elson et al., 1983). 

D. Other Mollusk classes 

Naticid prédation has not been recorded on 
the Aplacophora, Monoplacophora, Polypla- 
cophora, or the Cephalopoda. The shell-less 
Aplacophora would not leave traces of naticid 
prédation. The Monoplacophora (clay-mud 
habitats) and the Polyplacophora (rocky hab- 
itats) are usually not encountered by naticids. 
The epifaunal and pelagic cephalopods, pred- 
ators themselves, are unlikely to be captured 
by the slower naticids. 

E. Polychaetes 

Paine (1963: 69) found one specimen of 
Neverita duplicata from Florida that fed on the 
polychaete Owenia fusiformis; this is the only 
known record of naticids preying on annelids. 
It is not clear whether this represents normal 
behavior or a single, aberrant event. 

F. Crustaceans 

Significantly, Conor (1965: 229) found that 
naticids would not feed on hermit crab occu- 
pied shells. This is of importance as it indi- 
cates that not only can naticids recognize 
such "prey" (of course, the active epibenthal 
hermit crabs may be beyond the range of nat- 
icids), but also that boreholes found in shells 
with recognizable signs of hermit crab occu- 
pancy (worn lips, unrepaired damage, epi- 
bionts) were the cause of the gastropod mor- 
tality, freeing the shell for hermit crab use. 

Ostracods represent a potentially important 
prey source for juvenile naticids. Livan (1 937) 
and Reyment (1966, 1967) attributed numer- 
ous boreholes in ostracods to predatory 
gastropods. Maddocks (1988) reviewed the 
various types of boreholes in ostracods (Cre- 
taceous to Holocene of Texas) and concluded 
that juvenile naticids were responsible for 
most. However, because of the thin ostracod 
test, there is a wide variety of "holes" and it is 
difficult to attribute them to known causes 
(Reyment et al., 1987). 



G. Brachiopods 

Most articulate brachiopods live in rocky 
habitats (rock walls or boulder grounds), 
thereby escaping naticid prédation because 
of habitat incompatibility. However, Witman & 
Cooper (1983: 71, figs. 8c-<l) reported "nati- 
cid" boreholes in values of Terebratulina sep- 
tentrionalis from the Gulf of Maine, which they 
attributed to either Natica clausa or N. pusilla. 
The illustrated boreholes resemble those of 
muricids (albeit with slightly sloping sides); 
further study is recommended. 

H. Pisces 

Perry (1940: 116) reported that the tropical 
western Atlantic Naticarius canrena "preys 
on bivalves and has been seen to devour 
dead fish." This remarkable observation, if 
true, represents the only known record of pi- 
scivory in the Naticidae. However, if it is 
based on aquarium observations, then it may 
simply reflect aberrant behavior by starved in- 
dividuals (see the next paragraph). 

I. Scavenging 

Most studies have shown that naticids will 
only feed on fresh prey; carrion-feeding (as in 
the neogastropod Buccinidae and Nassari- 
idae) is not manifested. A few studies (typi- 
cally in aquaria) have shown that gaping 
(dying) bivalve prey may be consumed di- 
rectly without boring (Ansell & Morton, 1985). 
It is not clear if this laboratory behavior is also 
shown in the field. 

J. Egg Capsules 

Several authors have reported "naticid" 
boreholes in the egg capsules of various 
deep-sea organisms. These observations in- 
clude Thorson (1935: 12-13, figs. 4a-c) in 
egg capsules of the neogastropod buccinid 
Sipho [ = Colus] curtus from East Greenland; 
Jensen (1951, fig. 1) in egg capsules of the 
ray (Rala) from Davis Strait (the boreholes 
ranged from 0.75 to 2.5 mm in diameter; a few 
capsules had multiple boreholes); and Ansell 
(1961) in egg capsules of the dogfish (Scyl- 
liorhinus canícula) with countersunk bore- 
holes. It must be emphasized that naticids 
were not observed boring these holes; these 
authors had merely conjectured that naticids 
were the most likely causative agents. These 
boreholes were clearly effected from the out- 



166 



KABAT 



side (i.e., they are not the hatching-out holes 
of the juveniles within). First, for the buccinid 
egg capsules, it is probable that a muricid 
bored the holes, as is known for some other 
muricids (Abe. 1985). Second, for the elas- 
mobranch egg cases, a more likely predator 
is the unusual deep-sea archaeogastropod 
family Chohstellidae, which are typically as- 
sociated with skate egg capsules upon which 
they feed (Hickman. 1983: 86). 

The pnmary prey sources for naticids are 
infaunal gastropods and bivalves. The data 
[Appendix] document that 47 gastropod fam- 
ilies (out of 129 shelled marine gastropod 
families) and 35 bivalve families (out of 109 
marine bivalve families) are known to be sub- 
ject to naticid prédation. The major gastropod 
prey sources are the Turritellidae and Nati- 
cidae (Mesogastropoda) and the Turhdae 
(Neogastropoda). The major bivalve prey 
sources are the Lucinidae, Tellinidae and 
Veneridae (Heterodonta). 



FOSSIL RECORD OF 
NATICID PREDATION 

This section tabulates the reports of fossil 
naticid prédation and is arranged by geologi- 
cal time period. In general, only bnef summa- 
ries are provided; discussion of any broader 
ecological aspects is deferred to the following 
section in combination with related conclu- 
sions from Recent studies. It must be empha- 
sized that it is difficult to track down all the 
paleoecological studies, especially those that 
are "buried" within lengthy systematic mono- 
graphs (no attempt has been made to search 
through the latter). Indeed, it seems better 
that extensive paleoecological researches 
should be published separately from narrower 
taxonomic studies, in order to bring them to 
wider notice. 

A. Tnassic 

Fürsich & Wendt (1977: 299) mentioned 
"naticid" boreholes from the Cassian Forma- 
tion of northern Italy (Tirol). Subsequently, 
Fürsich & Jablonski (1984) illustrated the 
boreholes, showing the diagnostic counter- 
sunk appearance of incomplete boreholes, 
and discussed the implications thereof. The 
bivalve prey were Cassianella and Palaeonu- 
cula:\he gastropod predators were referred to 
several species of the naticid genus "Ampul- 



Una" Newton (1983; Newton et al., 1987: fig. 
25.2) independently documented "naticid" 
boreholes in the epibyssate limid Mysid- 
ioptera from the Wallowa Terrane of the Hells 
Canyon (Oregon-Idaho); this suggests that 
the Triassic borers were somewhat wide- 
spread, before becoming extinct. However, 
the taxonomy of Triassic "naticids" remains a 
morass, and their familial assignment is still 
uncertain. Further discussion of the evolution- 
ary consequences of Triassic boring préda- 
tion is deferred to the next section. Indeed, if 
these countersunk Triassic boreholes are not 
those of naticids, then it remains uncertain 
whether all the younger occurrences of coun- 
tersunk boreholes are correctly atthbuted to 
naticid prédation. 

Sohl (1969: 726) expressed some doubt as 
to whether the Tnassic forms were true nati- 
cids; in any event, his spindle diagram of nat- 
icid clade diversity (his fig. 1) clearly shows 
that from the Triassic to the mid-Cretaceous, 
there are never more than five genera in any 
epoch; naticid diversification did not com- 
mence until the Upper Cretaceous, with the 
evolution of the bohng habit. Bande! (1988: 
270) claimed that "Thus Tnassic naticids,' to 
a large extent, are neritoideans, some belong 
to other groups, but none appear to be natic- 
ids"; this needs further documentation. 

B. Jurassic 

Sohl (1969: 729) searched through vahous 
paleontological monographs and collections 
of Jurassic mollusks and found no signs of 
molluscan boreholes. Fürsich & Jablonski 
(1 984) also concluded that there were no gas- 
tropod borers in the Jurassic. 

С Cretaceous 

Fischer (1962a) reviewed some reports of 
Cretaceous boreholes and attributed most to 
naticids, as there were relatively few muricids 
at that time. Subsequently, Sohl (1969: 731) 
more carefully analyzed Cretaceous bore- 
holes and found a few from the Cenomanian 
(100 myr) and a much greater abundance 
from the Campanian (75 myr). The Ripley 
Formation (Campanian) was studied in 
greater detail by Vermeij & Dudley (1 982) who 
also found extensive shell repair and a size 
refuge from boring prédation. The oldest Cre- 
taceous records were shifted further back by 
Taylor et al. (1983) who documented naticid 
prédation from the Blackdown Greensand of 



NATICID PREDATION 



167 



England (Alblan, 105 myr). They found that 
the vast majority (92%) of boreholes were 
naticid, with a nearly equal ratio of gastropod 
to bivalve prey (in contrast to the few muricid 
boreholes, found primarily on bivalve prey). 
The diversification of naticids (and other mod- 
ern marine families) at this time represents 
the "Mesozoic marine revolution" of Vermeij 
(1977), and is discussed in the next section. 
Vermeij & Dudley (1982) reported no pré- 
dation on naticids in the Ripley Formation 
(Tennessee); subsequently, Kitchell et al. 
(1986: 293, fig. 1h) found a multiple-bored 
specimen of Euspira rectilabrum. from the 
same outcrops. This is the earliest record of 
confamilial naticid prédation in the fossil 
record. 



*#•'• 



D. Paleocene 



I have not found any paleoecological stud- 
ies from the Paleocene reporting on naticid 
boreholes. Naticids were present then; future 
studies of these faunas would be most worth- 
while. 

E. Eocene 

Fischer (1960, 1962a, 1963) reported on 
naticid prédation in the Lutétien Stage of 
France and found that for the bivalve Petun- 
culus [ = Glycymehs], 4.6% of the specimens 
were bored, primarily the smaller ones. For 
the gastropod Mesalia. 70.9% were bored by 
naticids (of which only 7.7% were incomplete 
holes), and some had multiple complete or 
incomplete boreholes. For Corbula spp., 
there was a rather high rate of boring failure 
(to 26% of the specimens). This fauna was 
also analysed by Taylor (1970) who found nu- 
merous naticid and muricid boreholes and an 
overall confamilial naticid prédation rate of 
1 1 .3%. 

Siler (1965) briefly reported on the Gosport 
Formation of Texas and found both naticid 
and muricid boreholes on the bivalve Lirodis- 
cus tellinoides. A more comprehensive study 
on the Stone City Formation of Texas 
(Stanton & Nelson, 1980; Stanton et al., 
1981 ) recorded a naticid mortality rate of 15% 
and a crustacean mortality rate of 20% for 
molluscan prey. The latter studies entailed 
considerable efforts to reconstruct the food 
web and paleocommunity structure. 

Several studies were carried out on the 
Ameki Formation of Nigeria by Adegoke & 
Tevesz (1974), Arua (1989) and Arua & 



Hoque (1987, 1989a, 1989c). They found that 
turrids and terebrids were the preferred gas- 
tropod prey; the latter authors also found ex- 
tensive prédation on bivalves. However, as 
discussed eartier, some of the boreholes 
seem to have been misidentified (vis á vis 
naticid vs. muricid) by Arua & Hoque. An anal- 
ysis of bivalve prey {Arcopsis and Limopsis) 
from the Pallinup Siltstone in Western Austra- 
lia found that 9.2% of the bivalves had gas- 
tropod boreholes, one fifth naticid and four 
fifths muricid (Darragh & Kendrick, 1980). 

F. Oligocène 

Klähn (1932) analyzed naticid prédation on 
other naticids from the Sternberg Formation 
of Germany and found high prédation rates 
from 53.3% (the second smallest prey size 
class) to 15%-26% (the other classes); the 
documentation provided does not facilitate 
further analysis. 

G. Miocene 

Hoffman et al. (1974) conducted an exten- 
sive study on the Korytnica clays of Poland 
and found a confamilial naticid prédation rate 
of about 10%; unfortunately, their data (table 
1) do not fully partition the boreholes by nat- 
icid or muricid sources. Subsequently, Hoff- 
man (1976a) attributed most of the bivalve 
mortality to sedimentation, rather than préda- 
tion; similarly, abiotic factors accounted for 
much of the gastropod mortality (Hoffman, 
1976b). Other Miocene outcrops from Poland 
were studied by Hoffman & Szubzda (1976), 
primarily with respect to food webs and com- 
munity structure. Kojumdjieva (1974) studied 
the Tortonian and Sarmatian outcrops of Bul- 
garia and found a variety of naticid and muri- 
cid prey taxa; very few unsuccessful or mul- 
tiple boreholes were observed. 

Thomas (1976) analyzed naticid prédation 
on glycymerid bivalves from various Neogene 
(Miocene-Pliocene) outcrops in the eastern 
United States and concluded that prédation 
rates in the Miocene were comparable to those 
on Recent glycymerids; however, the size-se- 
lectivity data seemed questionable. This re- 
search was reanalyzed by Kitchell et al. (1 981 : 
545-548), who determined that the seemingly 
contradictory results of Thomas could be ex- 
plained by the fact that there were actually two 
different naticid predators (of markedly differ- 
ent sizes) in the various fossil faunas; this 
meant that the observed "changes" in preda- 



168 



KABAT 



tion intensity or prey size were merely an ar- 
tifact of wfiich naticid predator was present. 

A series of studies on the Chesapeake 
Group of Maryland was conducted by Kelley 
(1982a-1989b). with an emphasis on bivalve 
prey. Nearly three-fourths of the mortality 
could be attributed to naticid prédation; for 
some prey there was an increase (over geo- 
logical time) of prey size and shell thickness. 
This was hypothesized to be an evolutionary 
response to naticid prédation. Dudley & Dud- 
ley (1980) made a briefer analysis of boring 
prédation on three mollusk species from 
these outcrops, and observed a size refuge 
from prédation for the two bivalves studied. 

Cotbath (1985) reported on the outcrops of 
the Astoria Formation of Oregon and noted 
extensive naticid prédation, primarily of bi- 
valves; other prédation sources were not an- 
alyzed. The Wimer Formation of northern 
California was analyzed by Watkins (1974), 
who found low levels of naticid prédation on 
several bivalves. 

Maxwell provided a thorough systematic 
and paleoecological analysis of the Stillwater 
Mudstone of New Zealand and observed con- 
siderable naticid prédation on various gastro- 
pods and bivalves. The data were used to 
reconstruct food webs (Maxwell, 1 988: 34, fig. 
3) as part of an overall trophic analysis which 
also considered non-fossilized aspects of the 
community. There was extensive confamilial 
naticid prédation, especially of the smaller- 
sized species. This monograph is an excel- 
lent model of integrating systematics with pa- 
leocommunity reconstructions. 

H. Pliocene 

Boekschoten (1967) studied the fauna of 
the Tielrode Sands of Belgium and reported 
some confamilial naticid prédation, although 
crustacean prédation was a far more impor- 
tant source of mortality for the naticids. The 
Emporda of Spain was analyzed by Hoffman 
& Martinen (1984), who observed high selec- 
tivity in prey size and borehole site choices. 
Guerrero & Reyment (1988b) used multivari- 
ate analysis to differentiate between naticid 
and muricid boreholes in Chlamys from the 
Lower Pliocene near Malaga, Spain. Robba & 
Ostinelli (1975) analyzed gastropod, cephalo- 
pod and crustacean prédation in the Albenga 
outcrops of Italy and noted that 13.9% of all 
specimens were bored, nearly all by naticids. 
Hingston (1985) reported on the Muddy 
Creek assemblage from Victoria, Australia, 



and determined that about 75% of the bore- 
holes were naticid and the remainder muricid; 
edge drilling of bivalves was rare, and prey 
shell sculpture resulted in a greater frequency 
of unsuccessful boreholes. 

I. Pleistocene 

Kabat & Kohn (1986) analyzed prédation 
on naticids from the Nakasi Beds of Fiji and 
observed rather high naticid prédation rates 
on Natica spp., but considerably lower con- 
familal prédation on species of Polinices and 
Sinum. Unsuccessful crustacean prédation 
was quite common; successful crustacean 
prédation probably accounted for a greater 
amount of mortality than did confamilial pré- 
dation. Berg & Nishenko (1975) found that 
26% of the shells of Nassarius perpinguis 
from the San Pedro deposits of California 
showed naticid boreholes; stereotypy of bore- 
hole siting was shown, although no data on 
predator or prey sizes were given. A much 
more detailed analysis of the nearly contem- 
poraneous Puerto Libertad deposits of So- 
nora, Mexico, and a thorough trophic web re- 
construction was conducted by Stump (1975: 
fig. 18). 

J. Sub-Holocene 

Yochelson et al. (1983) analyzed naticid 
prédation on scaphopods from the elevated 
"mud lumps," or diapir structures from the 
Mississippi River delta (ca. 15,000 years old), 
and found (in two large samples) that almost 
58% of Dentalium laqueatum had boreholes. 
They noted that other scaphopod assem- 
blages (fossil and Recent) showed far fewer 
naticid boreholes (usually less than 1 0%); this 
assemblage undoubtedly reflected excep- 
tional naticid feeding. 

Since the end of the Early Cretaceous (Al- 
bian), naticid prédation has been documented 
through Holocene faunas (except for the Pa- 
leoceno), although probable naticids are 
known from the Jurassic. Potential "natici- 
form" boreholes from the Thassic are known; 
the evidence is not conclusive as to whether 
or not the Triassic predators actually were 
naticids. The available data do not show any 
clear trends in the rates of gastropod boring 
prédation since the Cretaceous (Vermeij, 
1987: fig. 7.6); however, comparisons be- 
tween assemblages should be based on eco- 
logically analogous taxa, and studies of a sin- 



NATICID PREDATION 



169 



gle prey family need to consider possible 
changes in defense mechanisms (especially 
shell form) over time. 

Another area of interest is the use of bore 
holes in the field of ichnology, or the study of 
trace fossils. Most paleontologists recognize 
animal locomotory tracks as trace fossils; 
however, this field includes any and all re- 
mains of the activities of living organisms. 
Thus, a borehole found in a fossil specimen 
is, per se, a trace fossil, and can be described 
and discussed in the absence of exact knowl- 
edge of the causative agent. Needless to say, 
there has been some controversy over the 
"nomenclature" of trace fossils; the Interna- 
tional Code of Zoological Nomenclature 
(ICZN, 1985; Articles Id, lOd, 42b) currently 
does recognize "ichnotaxon names," as a 
parallel nomenclatural system. Hantzschei 
(1 975), Warme & McHuron (1 978) and Ekdale 
et al. (1984) provided excellent reviews of 
trace fossils. 

Predatory boreholes in fossil specimens 
can be referred to the ichnotaxon "Praedich- 
nia" Ekdale, 1985; those produced specifi- 
cally by mollusks to the ichnotaxon "Oichnus" 
Bromley, 1981; and those identical with nati- 
cid boreholes to the ichnotaxon "Oichnus pa- 
raboloides" Bromley, 1981. Maddocks (1988; 
641-2) "arbitrarily defined" 20 "ichnophena" 
corresponding to different forms of boreholes 
in ostracod tests; this diversity is unrealistic 
and meaningless. These names have no heu- 
ristic value; if they can be attributed to a 
known predator, then they should be referred 

to as "borehole of ", whereas those of 

unknown predators should not be given for- 
mal names. 



ECOLOGICAL ASPECTS OF 
NATICID PREDATION 

This section attempts to integrate and syn- 
thesize, from an ecological perspective, the 
vahed aspects of naticid prédation. It is hoped 
that this will not only indicate what has been 
well documented but also reveal promising 
(or neglected!) areas for future research. I 
have not attempted statistically to re-analyze 
previous studies or to provide detailed criti- 
cisms of previous methodologies, unless it 
seemed directly warranted. Subsequent re- 
searchers would be well advised to re-check 
the relevant previous studies. My section on 



"Mechanisms of naticid prédation" above in- 
cluded the more proximate aspects of naticid 
prey detection, capture and boring; this sec- 
tion covers the broader, ultimate aspects of 
naticid prédation, as well as several topics 
from the "prey's viewpoint." 

A. Prey Size and Species Choice 

The embryos of naticids feed on dissolved 
organic matter (DOM); some species have 
yolk reserves or infertile nurse eggs which 
serve as additional food resources, especially 
for those with direct development. Naticid 
species with planktotrophic larvae feed on the 
phytoplankton while in the swimming stage; 
those with lecithotrophic larvae undoubtedly 
rely on DOM in addition to their yolk reserves 
(Ansell, 1982c). 

The feeding habits of juvenile naticids have 
been much less studied. For example, Ansell 
(1982c) reported that they ate various un- 
specified gastropods or bivalves of small size; 
Berg (1976) was able to feed them Bittium 
and Rissoella, although this was limited to 
aquarium studies. Wiltse (1980a) found that 
juvenile Neverita duplicata at Barnstable Har- 
bor (Massachusetts) consumed the diminu- 
tive venerid Gemma gemma: because of the 
high density of the latter, naticid prédation ac- 
counted for less than 15% of total prey mor- 
tality. Maddocks (1988) concluded that juve- 
nile naticids represented significant predators 
of ostracods; with ontogeny, the naticids shift 
to larger-sized molluscan prey. 

Adegoke & Tevesz (1974; 22) claimed that 
"no direct correlation was found between prey 
size and predator size"; but no statistical data 
were presented to support this statement. 
Other studies, however, have shown that 
there is usually a good correlation between 
predator size (as determined by the inner 
borehole diameter) and the prey size (e.g 
Ansell 1960; Baytiss, 1986; Griffiths, 1981 
Kabat & Kohn, 1986; Kitchell et al., 1981 
Macé, 1978; Martineil & De Porta, 1982 
Robba & Ostinelli, 1975; Selin et al., 1986 
Wiltse, 1980a). Colbath (1985) reported little 
correlation between borehole diameter and 
prey size, except for Katherinella prey. How- 
ever, these results are a consequence of Col- 
bath's use of bivalve shell "width" rather than 
the more conventional length as the dimen- 
sional measure. 

Also of importance is the relative size of the 
prey taxa and the naticid predators. Large 
prey species are often less susceptible to pre- 



170 



KABAT 



dation by naticids than are small prey spe- 
cies. Similarly, within a species, smaller indi- 
viduals usually suffer greater naticid mortality 
(e.g. Franz, 1977; Jackson, 1972). Penney & 
Gnffiths (1984) used three-dimensional pré- 
dation contour diagrams to display the rela- 
tionships between predator size, prey size, 
and quantity of prey consumed. Alternatively, 
Hoffman (1976b: 296) showed no size-selec- 
tivity for some (but not all) gastropod prey 
from the Poland Miocene. However, Green 
(1968) found that mortality from naticid boring 
of the bivalve Notospisula parva actually in- 
creased with prey shell size; similar results 
were shown by Mukai (1973) and Wilson 
(1988). As discussed below, increased prey 
size over geological time may represent an 
evolutionary response to naticid prédation (or 
is of adaptive value to escape prédation) 
(Kelley, 1984, 1989b). 

Prey switching, or prey choice, has been a 
contentious point; the fundamental question 
of "why" a given naticid will pick a certain 
prey species given an equal choice of several 
species can lead to teleological explana- 
tions. Ansell (1983) found that dietary switch- 
ing will not occur and suggested that "pre- 
conditioning" may play a rôle in species 
choice. Broom (1 983) found that younger Nat- 
ica maculosa fed on Pelecyora trígona. 
whereas older predators fed on Anadara gra- 
nosa: ontogenetic dietary switching thus oc- 
curred. 

Several studies, using a variety of prey 
items, have determined a hierarchy of pre- 
ferred prey choices. For Euspira alden. Bayliss 
(1986; 40) found that the preferred bivalve 
prey, in descending order, were; Mya. Spisula. 
Cerastoderma and Parvlcardium: Árctica and 
Corbula were not preyed upon. Similarly, 
George (1 965) found that mortality due to nat- 
icids was most prevalent in Glycymens gly- 
cymeris, and less so in Donax semistnatus 
and D. trunculus (the latter the larger species). 
Kitchen et al. (1981) found that for Neverita 
duplicata, the preferred prey, in descending 
order, were; Mya. Mercenaria. Mytilus and Ne- 
verita. Although Neverita was actually the 
highest in energetic value, the handling costs 
were such that only much smaller conspecific 
prey could be captured by the naticid predator, 
Kelley (1 989a) found that bivalve prey from the 
Maryland Miocene were preferentially bored, 
in descending order, as; Eucrassatella. Ana- 
dara. Astarte (the latter two roughly equiva- 
lent) and Corbula. with slight differences from 
one formation to another. 



The same naticid species, in different local- 
ities, may have markedly different diets. Thus, 
Natica maculosa in Penang (Malaya) feeds 
wholly on gastropod prey, especially the tro- 
chid Umbonium vestianum. whereas this spe- 
cies at Kuala Selangor (Sumatra) feeds on 
bivalve prey, particularly Anadara granosa. In 
this case, it is the relative availability of prey 
taxa which determines (in part) the diet of a 
given naticid species (Broom, 1982; Berry, 
1982). 

A recent series of studies by Kitchell and 
colleagues (Kitchell et al., 1981 ; DeAngelis et 
al., 1984, 1985, 1989) have attempted to 
model the energetic and coevolutionary as- 
pects of naticid ecology. The first study was of 
value in providing a useful model for the test- 
ing of naticid prédation; however, the subse- 
quent papers incorporated multiple assump- 
tions which decreased their representation of 
the real world into a senes of parameters 
couched in advanced equations. This reduc- 
tionist approach cannot account for complex, 
stochastic, and hierarchial ecological commu- 
nities. 

It is worthwhile to elaborate briefly the basic 
pnnciples of the Kitchell models. Essentially, 
the cost;benefit ratio for various prey species 
is determined (costs being the time and en- 
ergy to recognize, capture/subdue, bore, and 
digest the prey; benefits the energetic value 
or gain of prey tissues) and related to both 
prey size and predator size, given that the 
cost of a specific prey will vary according to 
the predator size. From this, one can graphi- 
cally represent the cost-benefit functions with 
prey size as the dependent variable and cost; 
benefit ratios as the independent variable. 
The lowest curve represents the optimal prey 
choice. These curves show that optimum prey 
are of intermediate sizes; too-small prey are 
of low energy value and too-large prey can 
usually escape the predator. Kitchell (1987) 
found that these models lead to the prediction 
that "larger naticid predators should be more 
highly selective than smaller-sized naticids," 
all other factors being equal. Discussion of 
their later models, dealing primarily with pred- 
ator-prey coevolution has been deferred to 
section F, under the evolutionary aspects. 

Kelley (1982b, 1987, 1989a-b) used these 
methods to analyze naticid prédation in the 
Maryland Miocene fauna, and confirmed that 
the models predict prey selection patterns, 
but with some exceptions. She found that 
over time, bivalve prey shell thickness ( = 
cost) increased while there was no overall 



NATICID PREDATION 



171 



trend in shell volume (= benefit). Commito 
(1987) questioned the validity of the Kitchell 
models and noted that their assumptions ne- 
glected several important factors with respect 
to prey defense strategies (or adaptations): 
ignored were the possibilities of depth ref- 
uges, shell ornamentation, chemical de- 
fenses, or behavioral responses, all of which 
could deter naticid prédation. DeAngelis et al. 
(1987) acknowledged these criticisms and 
suggested that yet further modelling would be 
able to incorporate these aspects of prey bi- 
ology. It is difficult to account fully for all the 
parameters or variables that determine or in- 
fluence prédation processes; any model that 
attempts to do so would likely be so unwieldy 
or incomprehensible as to be of little heuristic 
value. 

Interestingly, Ansell (1982b) found that Eu- 
spira alderi would not feed on opened bi- 
valves — only live, closed prey items were 
chosen. These same results were found by 
Kitchell et al. (1986: 297) for Neverita dupli- 
cata. This suggests that the stereotypy of 
prey choice restricts the naticids to fresh prey, 
and rules out scavenging or carrion-feeding. 

Prédation by naticids on other naticids can 
be quite widespread and represents a signif- 
icant source of naticid mortality. Although oc- 
casionally referred to as "cannibalism," that 
term is inappropriate since this prédation 
does not necessarily involve conspecifics. 
Studies from the Nigerian Eocene showed 
that about 15% of naticid shells had naticid 
boreholes (Adegoke & Tevesz, 1974); Col- 
bath (1985) observed only 2.7% such in the 
Oregon Miocene; Hoffman et al. (1974) noted 
10% such in the Poland Miocene. Boekscho- 
ten (1967) found that 7.8% of the naticids 
from the Belgian Pliocene had naticid bore- 
holes. Kabat & Kohn (1986) determined that 
in the Fijian Pleistocene, naticid prédation on 
Natica spp. accounted for 27% of mortality, 
whereas that on Polinices and Sinum spp., for 
only 3% of mortality. The latter genera have 
more globose shells and a larger foot which 
may provide faster locomotion and hence fa- 
cilitate escape from confamilial predators. 
Maxwell (1988) concluded that smaller-sized 
naticids of the New Zealand Miocene had 
much higher naticid prédation rates, confirm- 
ing size-selectivity aspects of naticid préda- 
tion. Several studies on Recent naticids have 
also shown extensive confamilial prédation 
(Burch & Burch, 1 986; Fretter & Manly, 1 979). 
Obviously, there is considerable variation as 
to the extent of confamilial naticid prédation; 



disease and prédation by fish or crustaceans 
may represent more important naticid mortal- 
ity pressures. 

B. Stereotypy of Boring on Prey Shell 

For gastropod prey, there has been some 
confusion among studies with respect to the 
siting of successful boreholes, with some 
"results" actually of no consequence. Thus, 
Arua & Hoque (1989a: 55) emphasized that 
the "preferred drilling site" on the apertural 
side was on the last whorl; however, because 
of whorl overlap, most of the exposed prey 
shell surface is the last whorl, and thus purely 
non-random borehole siting would lead to 
most boreholes located there (their other re- 
sults combine 1 1 prey species into a single 
table which does not facilitate further analy- 
sis). Yet, for some gastropod prey, there is a 
predominance of prédation on the dorsal 
(abapertural) side over the ventral (apertural) 
side; this reflects the increased ability of the 
prey to escape in the latter position (Adegoke 
& Tevesz, 1974). However, other studies 
suggested that prédation on the ventral side 
is preferred since the predator's foot seals off 
the aperture, blocking escape (Berg, 1976: 3; 
Berry, 1982). Some studies have shown that 
certain gastropod prey are preferentially 
bored on the penultimate whorf (rather than 
the last whorl); this, too, reflects prey handling 
factors (Dudley & Dudley, 1980; Hoffman & 
Martinen, 1984). Boreholes that are at either 
extreme end (apical or abapical) may not al- 
low the proboscis to penetrate the entire shell; 
more centrally located boreholes may facili- 
tate complete consumption of the prey tis- 
sues. 

For gastropod prey, it is convenient to an- 
alyze the stereotypy of borehole siting by the 
various geometrical subsets of the shell. Not 
only can one distinguish between the outer 
(body) whorl and the older, apical whorls [i.e. 
the horizontal dimension], but one can also 
partition the prey gastropod shell whorts into 
semicircular sectors, or longitudinal zones 
[i.e. the vertical, or axial dimension]. Thus, 
Berg (1976) and Berg & Nishenko (1975) de- 
veloped two conflicting numbering schemes 
for the latter division. In the 1975 paper (their 
figure lb), the sectors (numbered 1-8) 
started with the apertural plane and pro- 
ceeded counterclockwise (when viewed from 
the apex); thus, their clockwise "pie chart" 
(their figure 1c) of the sectors is actually 
viewed abapically. But, in the 1976 paper (his 



172 



KABAT 



figure 2a) the sectors (also numbered 1-8) 
started with the apertura! plane and pro- 
ceeded clockwise (when viewed from the 
apex); their clockwise "pie chart" (his figure 
2b) is, this time, viewed apically! It is not clear 
what has been done here: my recommenda- 
tion is that future investigators explicitly spec- 
ify which scheme they are using. 

Kabat & Kohn (1986: fig. 4), using the first 
scheme, observed that for naticid prey, bore- 
holes were found in four of the eight shell sec- 
tors, with nearly 90% occurring in two 90° sec- 
tors: however, there was little overall 
evidence for stereotypy of borehole siting. 
Robba & Ostinelli (1975: 327) independently 
depicted an angular measurement system 
which corresponds to the first scheme of 
Berg. Stump (1975: figs. 19-21) devised an 
elaborate "equal-area projections" system to 
show frequency-contours (in percentages) of 
borehole siting on the various prey shells. 
Regrettably, this method is difficult to visualize 
and does not lend itself to comparison with 
the other, more direct schemes: it does not 
seem to have been used by subsequent au- 
thors. 

Some studies have shown that most boring 
occurs near the shell margin of bivalve prey, 
where the shell is thinner and there is no 
sculpture (e.g., Ansell, I960: Ansell & Morton, 
1985). Other studies, however, have shown a 
preference by other naticids for boring near 
the umbones (e.g. Ansell & Morton, 1985: 
Arua & Hoque, 1 989: Bernard. 1 967: Colbath, 
1 985: George, 1 965: Jacobson, 1 968: Kitchell 
et al., 1981: Leidy, 1878: Matsukuma, 1976: 
Negus, 1975: Rieron, 1933; Thomas, 1976; 
Vignali & Galleni, 1986): or in the mid-region 
(Bayliss, 1986: Griffiths, 1981; Vermel) et al., 
1989). The strongly inequilateral Periploma 
margaritaceum was primarily bored on the 
anterior slope, due to its shell form (Rose- 
water, 1980). Some earlier studies had sug- 
gested that naticids preferentially bored near 
the prey gonads or digestive tissues (Pelse- 
neer, 1924; Verlaine, 1936); however, bore- 
hole siting is primarily a function of the ma- 
nipulation of the prey dunng boring and may 
depend on the prey shell morphology. In a 
few cases, little stereotypy is manifested. 
Berg & Porter (1974) found that, for the same 
bivalve prey, there were significant differ- 
ences between naticid species as to the pre- 
ferred borehole position; Berg (1975) sug- 
gested that behavioral differences in prey 
capture and handling influenced species-spe- 
cific patterns. 



Probably of greater importance are (1) the 
size of the prey relative to the predator; (2) the 
shell thickness and presence or absence of 
sculptural elements; (3) the relative convexity 
of the prey shell: (4) other factors relating to 
the predator's manipulation of the prey. 
Based on this review, no one element solely 
determines the locus of borehole siting 
among bivalve prey. 

The majority of studies have shown little 
preference for right vs. left valves of bivalve 
prey, as would be expected given the equiv- 
alve nature of most infaunal bivalves. Some 
studies have shown 10-20% "differences" in 
the frequency of boreholes between valves, 
but no clear trends are apparent. Needless to 
say, for each valve with a borehole, there is a 
matching, unbored valve; hence the naticid 
mortality rate is twice the number of bored 
valves divided into the total number of valves. 
It is incomprehensible as to what Lever et al. 
(1 961 : 341 ) meant when they stated that "the 
percentual mortality may in some cases ex- 
ceed 100 [%]." 

Adegoke & Tevesz (1974) stated that Var- 
icorbula from the Nigerian Eocene was pleu- 
rothetic and invariably bored on the right 
valve which is closer to the surface. However, 
as noted below, the left valve of corbulids has 
a thick periostracum which deters boring pré- 
dation; the position of the corbulid shell in the 
substrate is of less import (De Cauwer, 1985). 
More generally, since naticids usually manip- 
ulate their prey prior to boring, the life position 
may be of little relevance. Newton (1983) 
found that the Thassic limid Mysidioptera was 
always bored through the left valve; this taxa 
is an epibyssate recliner and the left valve is 
adjacent to the substrate (Newton et a!., 
1987: fig. 27). 

С Incomplete and Multiple Boreholes; 
Non-boring Prédation 

Incomplete boreholes are usually inter- 
preted to represent a sign of interruption of 
prédation, whether by prey escape, arrival of 
another predator, or other disturbance. In 
some cases, the same naticid (or another) will 
recapture the prey and commence boring a 
new borehole, elsewhere on the prey shell. 
Sometimes the new hole will coincidentally 
overlap the older hole; but studies have 
shown that naticids cannot recognize their 
own previous borehole and resume drilling 
there (thereby saving considerable time) 
(Kitchell et al., 1981: 539). The related prob- 



NATICID PREDATION 



173 



lern of multiple complete boreholes again 
suggests interruption of prédation after the 
completion of a borehole. Obviously there is 
an evolutionary disadvantage in not recogniz- 
ing previous boreholes (complete or incom- 
plete); the stereotypy of naticid predatory pat- 
terns may not be sufficiently flexible (Vermeij, 
1982: 707; Kitchell et a!., 1986). 

In an analysis of the Miocene Strioterebrum 
monidum from the Caribbean, Kitchell et a!. 
(1986: 294-5) found extremely large num- 
bers of shells with multiple boreholes; one 
such had 15, of which 12 were incomplete 
and three had penetrated the prey shell but 
were not sufficiently wide to allow passage of 
the proboscis. Further studies on living tere- 
brids by these authors confirmed that some 
species of this prey family are highly agile and 
can repeatedly escape naticid prédation dur- 
ing the boring actions. Earlier, Vermeij et al. 
(1980: table 2) showed rather high rates (to 
40%) of incomplete boreholes in various Re- 
cent terebrids; G. J. Vermeij (in litt.) sug- 
gested that the pungent odor of terebrids and 
olivids may represent a chemical defense 
against prédation. 

Fischer (1962b: 97) found that in a large 
sample (n = 1,126) of the Eocene turritellid 
Mesalia, 70.9% had naticid boreholes. Of the 
bored specimens, 84.8% had a single com- 
plete borehole (of which a tenth also had one 
to several incomplete boreholes); 4.2% had 
multiple complete boreholes; 8.7% had a sin- 
gle incomplete borehole; and 2.3% had mul- 
tiple incomplete boreholes. Kitchell et al. 
(1981: 542) observed that the lucinid 
Pseudomiltfia floridana had a ratio of incom- 
plete to complete boreholes of 0.54:1. This 
taxon was stated to be polymorphic for shell 
thickness; the thicker shells were more likely 
to have incomplete boreholes. 

An important recent discovery was that 
some bivalve prey, primarily in the tropics, are 
preferentially bored through the edge of the 
valves (Taylor, 1980: 175; Vermeij, 1980: 
330); not only is the shell thinner there, but 
also the prey shell is unsculptured and easier 
to bore (Ansell & Morton, 1985). The latter 
authors found that some species (i.e. of Po- 
linices) regularly edge-bored Bassina, while 
Glossaulax did not; that genus may preferen- 
tially bore other prey taxa. Some elements of 
"learning" (conditioning) may be involved in 
these responses to shell sculpture. 

The razor clams {Ensis, Solen) have been 
shown to be typically consumed by naticids 
without boring, because when the valves are 



contracted, there are still sizable pedal and 
siphonal gapes through which the naticid pro- 
boscis can be inserted (Turner, 1955; Ed- 
wards, 1975; Schneider, 1981; Frey et al., 
1987); this was also shown for Tresus (Reid & 
Fiesen, 1980: 32). Edwards & Huebner 
(1977) noted that Mya was not consumed di- 
rectly through its large siphonal gape; in- 
stead, naticids always bored through the 
valve; possibly the siphonal tissue deters 
feeding activities. Eartier, Agersborg (1920: 
421) had claimed that Mya and various other 
clams could be suffocated and directly con- 
sumed by Euspira lewisii; this now seems 
doubtful. Vermeij & Veil (1978) found that the 
frequency of gaping bivalves in marine faunas 
decreased from the Arctic to the tropics and 
noted that this was correlated with the in- 
crease in shell boring and other prédation 
sources in warmer habitats. 

Some gastropod prey can be attacked 
through the aperture, as the corneous oper- 
culum is flexible enough for the proboscis to 
be inserted around the margins (Hughes, 
1985). Edwards (1969: 327) found that some 
Olivella prey were consumed without boring, 
and suggested that either the naticid could 
force the operculum, or else the prey "suffo- 
cates while wrapped in the predator's foot and 
relaxes," allowing the predator direct access 
to prey tissues. Interestingly, Yochelson et al. 
(1983: 11) speculated that the stereotypy of 
naticid boring precluded their attacking sca- 
phopods directly through the open apertural 
end; but they suggested that it was more 
likely that once the scaphopod had retracted 
posteriorly, the naticid proboscis would not be 
able to reach the prey tissues. 

As mentioned earlier, the tropical Indo- 
Pacific Polinices mammilla is able to "suf- 
focate" and consume bivalve prey without 
boring. Ansell & Morton (1987) documented 
that this non-boring prédation, in aquarium 
experiments, accounted for 14% to 54% of 
the bivalve mortality (according to prey spe- 
cies). This example, and those in the preced- 
ing two paragraphs, would greatly complicate 
community analyses (especially of fossils!) 
since no "traces" of naticid prédation would 
be left on the post-mortem prey shell. 

It should be noted that the results of several 
studies of naticid prédation were misinter- 
preted as concluding that a significant num- 
ber of the prey were consumed without boring 
(Kitchell et al., 1986: 297). Thus, Edwards 
(1975: 17) found that about 75% of the prey 
were bored and the remainder died of other 



174 



KABAT 



causes; Taylor et al. (1980: 397) erroneously 
took this to mean that the latter 25°o of the 
prey were consumed (by naticids) without be- 
ing bored. Similarly, Medcof & Thurber (1 958) 
misinterpreted their own data to assume that 
all the empty, non-bored bivalve prey shells 
were consumed by naticid predators without 
bohng; this overlooked other mortality 
sources. Another study (Bernard, 1967) 
stated that 'in limited aquarium observation, 
over 60% of Saxidomus consumed showed 
no drill marks" (p. 9); and, again, ". . . in 
aquaria tests 25% of clams [Saxidomus gi- 
ganteus] consumed by Polinices [ = Euspira] 
lewisi bore no marks at all" (p. 10); the dis- 
crepancy in numbers is irreconcilable and all 
bivalve mortality was erroneously attributed to 
naticid prédation. 

D. Prey Defense Mechanisms 

Ansell (1969) and Carter (1968) provided a 
general overview of defense mechanisms in 
various marine mollusks. Many bivalves show 
leaping or rapid burrowing in response to con- 
tact by naticids. Laws & Laws (1972; fig. 1) 
described the escape response of the Austra- 
lian Donacilla angusta. which leaps or pops 
out onto the surface, thereby evading the bur- 
rowing naticid predator; similar responses 
were shown for Ensis directus (Turner, 1955; 
Schneider, 1982) and Ruditapes ptiilippi- 
narum (Rodrigues, 1986). Either rapid or 
deep burrowing (or both), can serve as an 
escape mechanism (Vermeij, 1983a) for bi- 
valve prey. 

Ansell & Morton (1985; 656) found that the 
anomalodesmatan bivalves Lyonsia and Pan- 
dora seemed to escape naticid prédation "by 
coating the postenor edge of the shell with 
mucus to which sand grains adhere"; pre- 
sumably this somehow deterred naticid pré- 
dation. 

Corbulid bivalves have been the object of 
several paleoecological studies; corbulids are 
noteworthy for their well-developed conchiolin 
layer (within the valve) which serves as a 
fairly effective deterrent to gastropod préda- 
tion (Lewy & Samtleben, 1979). Furthermore, 
most successful boreholes are in the right 
valve, since there is well-developed perios- 
tracum on the left valve of corbulids which 
also deters predators. Complete boreholes in 
corbulid valves have a special form, with a 
considerably narrowed inner margin below 
the conchiolin layer (De Cauwer, 1985; figs. 
1d, le). Kelley (1989a: 446-7) also found 



considerably reduced successful prédation 
on Corbula and suggested that the low level 
of selectivity of prey size and borehole siting 
may also account for the high rate of unsuc- 
cessful prédation (60% of boreholes nonfunc- 
tional). Lewy & Samtleben (1979: 350) sug- 
gested that the conchiolin layer serves as a 
compensation for the slow mobility and shal- 
low burrowing of corbulids. 

Alternative "defense" strategies of two bi- 
valves were discussed by Commito (1982): 
Mya arenaria grows rapidly to a large size 
(and deferring reproduction until then), 
thereby escaping naticid prédation [= size 
refuge], whereas Macoma balthica instead 
grows slowly, reproduces early, and escapes 
most naticid prédation by deep burrowing [ = 
spatial refuge]. Of course, Mya is subject to 
naticid prédation while it is still small. The 
former mechanism was used by Hutchings & 
Haedrich (1984) to explain the size structure 
of deep-water nuculanids subject to naticid 
and fish prédation. Actually, these "alter- 
native" life history patterns may represent 
phylogenetic constraints rather than direct ad- 
aptations to naticid prédation, per se. 

Ansell & Morton (1985) discovered that re- 
moval of the sculptural lamellae on the shells 
of the venerid Bassina led to increased boring 
prédation through the shell sides. Otherwise 
the naticids bored through the valve edges 
which do not have sculpture. This experimen- 
tal observation demonstrated the function of 
sculpture as a prey shell defense mechanism 
in addition to stabilizing the bivalve in soft 
sediments. 

Bayliss (1986) found that among bivalve 
prey, the species with the thinnest shell was 
preferentially preyed upon by naticids. Hing- 
ston (1985: table 4) noted that increased prey 
shell sculpture led to increased frequency of 
unsuccessful (incomplete) boreholes. Dudley 
& Vermeij (1978; 439) concluded that strong 
spiral ribs usually deterred boring in turritel- 
lids. Kelley (1982a; 46) reported that uncrenu- 
lated (male) shells of Astarte were more likely 
to be bored than were crenulated (female) 
shells; however this genus is protandrous, 
and the resulting size differences (between 
sexes) may be sufficient to explain differ- 
ences in prédation rate (given that the smaller 
males are less likely to escape prédation). 

Boggs et al. (1984), using Mercenaria mer- 
cenaria prey, artificially ground-down the shell 
surface to half the normal thickness, and 
tested the effects on prédation by Neverita 
duplicata. They found that naticids could not 



NATICID PREDATION 



175 



learn to differentiate between normal and 
thin-shelled prey, although the latter took con- 
siderably less time to bore. The same results 
were found by Rodrigues et al. (1987) for Ne- 
ve ri ta didyma preying on Rudi tapes ptiilippi- 
narum. In some respects, these studies are of 
questionable value since it has not been 
shown that gastropods have any sensory 
mechanism for "determining" shell thickness 
(or shell weight). It is true that preying on thin- 
ner prey freed up additional time for foraging; 
surely the snails are incapable of this realiza- 
tion because they have no method for recog- 
nizing the thinner prey. This is an interesting 
case of a hypothetical coevolutionary re- 
sponse that does not initiate an "arms race." 

E. Food Webs, Energy Flow and 
Physiological Efficiencies 

Food webs are attempts to diagram the 
overall trophic structure of an ecological com- 
munity (predators, herbivores, primary pro- 
ducers, detritivores). Elucidation of the struc- 
ture of a food web and the strength (or 
quantity of interactions) of each link (chain) 
facilitates analyses of community energy flow 
and population dynamics. As infaunal preda- 
tors, naticids (with other infaunal polychaetes, 
crustaceans, and nemerteans) represent an 
often overlooked level of prédation, in addi- 
tion to the more conspicuous epibenthic pred- 
ators (asteroids, fish and crabs) (Commito & 
Ambrose, 1985). An example of the complex- 
ity involved is that both asteroids and naticids 
prey on bivalves, whereas some asteroids 
also prey on naticids (Christenson, 1970: 67); 
the same multiple interactions also occur with 
respect to crabs and fishes. Relatively little 
research has been done on determining the 
complete food webs for soft-bottom commu- 
nities, in contrast to better-known rocky inter- 
tidal communities; this reflects the ease of ac- 
cess and analysis of the latter fauna. 

Several paleocological studies have at- 
tempted to elucidate community structure and 
food webs, based primarily on an analysis of 
shell boring and breaking prédation (Hoffman 
& Szubzda, 1976; Stanton & Nelson, 1980; 
Stanton et al., 1981; Stump, 1975; Taylor et 
al., 1983). While of great heuristic value in 
facilitating comparisons between fossil com- 
munities (as well as with Recent communi- 
ties), these studies are limited by the indeter- 
minate nature of mortality that leaves no 
"traces," as well as shell-removing agents, 



the latter skewing the results towards the re- 
maining predatory agents. 

It is important to realize that naticid préda- 
tion represents only a part of the sum of all 
prédation in soft-bottom communities; several 
authors have carefully reviewed the diversity 
and importance of other predators in these 
habitats (Cadée, 1968; Carter, 1968; Vermeij, 
1978). Thus, Green (1969) found that naticids 
accounted for 9% of the mortality of the tropical 
bivalve Notospisula parva; shell-crushing 
skates were responsible for over 60% of the 
mortality; the remainder was due to other fac- 
tors (disease or abiotic agents). The latter, 
non-predatory sources of mortality are just as 
important but virtually impossible to determine 
precisely from fossil or beach assemblages 
(i.e., an empty, undamaged shell may be the 
outcome of parasitism, other disease, sedi- 
mentation, or other agents) (Hoffman, 1976a). 

A series of excellent physiological studies 
was conducted by Ansell and Macé on the 
European Euspira alderi. Distinct periods of 
shell growth were followed by egg collar pro- 
duction; feeding was considerably greater 
during the latter stage, since over 90% of 
non-respired assimilated energy is used for 
reproduction (Ansell & Macé, 1978; Ansell, 
1982a-b). Prédation rates increased with 
temperature (Macé, 1981a); and oxygen con- 
sumption rates ( = respiration) were affected 
by tfie prey type and quantity (Macé, 1981b; 
Macé & Ansell, 1982). Each week, an adult 
naticid consumed up to its own (dry) weight in 
prey tissue [Tellina tenius] (Ansell, 1982a); 
this is limited by the extensive time spent in 
obtaining suitable prey. Macé (1981c) found 
that energy assimilation efficiency is about 
60% during reproductive periods, and only 
40% at other times. About 50-60% of the 
consumed energy is, however, "lost": not ac- 
counted for by growth, respiration (mainte- 
nance) or reproduction. Ansell (1982b) sug- 
gested that some of this may be accounted 
for by the mucus that is essential for prey cap- 
ture and predator avoidance; much of the re- 
mainder is represented by feces and uncon- 
sumed prey tissue, but Berry (1983) was 
unable to calculate the energetic costs or 
losses due to mucus or feces. Bayliss (1986), 
using t!ie same naticid species, found that 
about 24% of the time was spent drilling, 
11%-18% ingesting prey tissue, and the re- 
maining time in other activities, typically qui- 
escent. 

Related physiological studies on the tem- 
perate Neverita duplicata (in Massachusetts) 



176 



KABAT 



showed that the feeding season was only 
about 35 weeks, during which approximately 
1 .85 prey [Mya arenaria) were consumed per 
week. The naticids consumed about 1% of 
their body weight in prey on a daily basis, and 
the overall growth efficiency rates (snail 
growth in kilojoules per clam tissue consumed 
in kilojoules) declined from almost 50% in 
young snails to 16% in older snails (Edwards 
& Huebner, 1977; Huebner & Edwards, 
1981). 

Another factor of importance in calculating 
energy budgets is whether or not all the prey 
tissue is consumed. Thus, for a high-spired 
gastropod prey, some of the apical tissues 
may not be reached by the proboscis. Ed- 
wards & Huebner (1977) found that when 
feeding on Mya. only about 80% of the prey 
tissues were consumed (i.e. the "energy rich, 
low-ash content tissues"); proboscoidal ac- 
cess is not at issue here and this may reflect 
the less-palatable nature of the mantle edge 
and siphonal tissues of Mya. 

Broom (1982) determined the "consump- 
tion rate" equation of feeding efficiency: this 
represents the mg dry weight of prey con- 
sumed per day, as a function of predator body 
(wet) weight. Thus, for Natica maculosa feed- 
ing on Anadara granosa, the allometric equa- 
tion was CW = 9.13 (W)' °°^^ where W = 
predator wet weight (in grams). Similarly, Gnf- 
fiths (1981) found that the consumption rates 
(of bivalve prey, Choromytilus menodionalis) 
Increased 4.5 fold over a 55% increase in 
predator (Natica tecta) size. 

Many of these studies were based on lab- 
oratory (aquaria) observations. These, of 
course, are a simplification or modification of 
reality (field behavior). Bayliss (1986: 46) co- 
gently noted that "the artificial and enclosed 
environment in an aquarium increases the 
predator's ability to detect and capture a prey 
item as well as reducing the prey's ability to 
avoid and escape from the predator. " Also, 
intertidal naticids are usually quiescent dunng 
low tide; in aquaria where they are continually 
submerged, the duration of activity is more 
extensive. Many laboratory studies (e.g. Rod- 
rigues, 1986) used an aquanum sand depth 
barely greater than the prey or predator size; 
this does not allow for normal burrowing pat- 
terns. Kitchen et al. (1986: 297) noted that in 
their aquaria, the prey frequently "die, gape 
and decompose without the predator taking 
any part in the process"; this suggests that 
their prey were usually moribund or otherwise 
unhealthy, and leads one to question the va- 



lidity of prédation studies on these weakened 
prey. These caveats should be considered 
when calculating feeding rates, energy bud- 
gets, and related trophic measurements 
based on laboratory studies. 

A typical example of the effects of naticid 
prédation on prey population dynamics is that 
of Ansell (1960) who found that of first-year 
Venus [ - Chamelea] stnatula. 40% of the to- 
tal mortality [= 15% of all individuals] was 
due to naticids; for the second-year cohort, 
only 15% of all mortality [= 5% of the cohort] 
was naticid prédation; and for the third-year 
cohort [the last], only about 1% of all mortality 
was due to naticids. Clearly, prédation by Eu- 
spira alderi affects primarily the younger co- 
horts; disease or other predators affect the 
older cohorts. 

.Another interesting taphonomic-ecological 
phenomenon is that of 'ti)each sorting" or the 
differential post-mortem "survival" of valves 
of different bivalves (interspecific and in- 
traspecific analyses), comparing both right vs. 
left valves and bored vs. unbored valves 
(Lever et al., 1961; Lever & Thijssen, 1968; 
Martineil & De Porta, 1980). The critical ques- 
tion is whether or not bored valves are differ- 
entially susceptible to post-mortem damage 
which would affect their representation in the 
fossil (or "beach shell ") assemblage (Dudley 
& Vermeij, 1978: 437). One must also deter- 
mine the extent of other shell-breaking préda- 
tion that wholly removes the shells from the 
assemblage. 

The studies of Lever and colleagues found 
that valves with boreholes (natural or artifi- 
cial) traveled shorter distances but were more 
likely to end up higher on the shore (than non- 
bored valves), because of the biomechanics 
of fluid flow through and around bored valves. 
Thus, the "hole effect" is the upward transport 
of bored valves. The differential transport of 
right and left valves may also occur, resulting 
in greatly distorted ratios thereof in a beach 
assemblage. Indeed, it is possible that some 
paleontological studies showing "differences" 
in boring rates between valves may actually 
be a consequence of this differential sorting. 
A problem with such studies is that the hydro- 
dynamic properties of bivalve shells can vary 
between taxa, and the biomechanical effects 
of one shell morphology may well be the op- 
posite of those of a different morphology. 

F. Enemies and Control of Naticids 

Asteroids (starfishes or seastars) are im- 
portant predators of naticids (Agersborg, 



NATICID PREDATION 



177 



1920; Christenson, 1970); some naticid prey 
will ward off the asteroid by extension of the 
foot over the shell followed by mucus secre- 
tion (Ansell, 1969; Margolin, 1975). The latter 
author documented that Natica stercusmus- 
carum could respond to Astropecten by rasp- 
ing off the spines and consuming the tube 
feet, deterhng the starfish. Clarke (1956) 
noted that Nassarius trivittatus feeds upon the 
egg collars of Euspira heros, serving as a 
means of control. Ironically, this nassariid is, 
in turn, preyed upon by adult naticids! 

Frequently, naticids are "blamed" for ob- 
served declines in populations of commercial 
shellfish (soft shell clams, quahogs, etc.), and 
oyster beds may be disrupted as naticids bur- 
row through them in search of other prey 
items (Agersborg, 1920: 420). Because oys- 
ters are now more commonly cultivated on 
stakes or lines off the substrate, this may now 
be less of a problem. Edwards & Huebner 
(1977: 1231) cogently noted that "bored 
shells . . . are thus an exaggerated indicator of 
[naticid] mortality . . ." because other preda- 
tors (arthropods, fish, birds, humans) remove 
or otherwise destroy bivalve shells. These au- 
thors further stated that naticid predators are 
an easy scapegoat to take the blame for ". . . 
human exploitation patterns, a sensitive 
issue." The various mechanisms and their 
success (or lack thereof) for the control of 
"pests" of shellfish were reviewed by Kor- 
ringa (1952: 347-351); hand collecting is par- 
ticularly ineffective (Turner et al., 1948; Med- 
cof & Thurber, 1958). Carhker (1981: 417) 
suggested that ecological control, involving 
species-specific pheromones or deterrents, 
might be successful. There remains the often 
unacknowledged dilemma that not only is it 
impractical (or even impossible) to eliminate 
these predators, but also the resulting impact 
on the overall community structure and food 
web may actually be more deleterious than 
the effects of the predators themselves on the 
shellfish. 

G. Macroevolutionary Patterns and 
Evolutionary Escalation 

If, as claimed by Fürsich & Jablonski 
(1984), the Triassic boreholes are attributable 
to naticids, then the parallel evolution of the 
naticid boring habit twice (Thassic and Creta- 
ceous) undoubtedly reflected the canalization 
or phylogenetic constraints of shell-boring: 
there are only so many ways a shell can be 
bored, and the underlying mechanisms may 



have remained quiescent in the Naticidae 
during the Jurassic. However, it remains un- 
clear whether the Triassic predators are in- 
deed naticids, or how the Jurassic naticids 
may have fed (possibly as scavengers). 

Taylor et al. (1980: fig. 16) presented a 
hypothetical scenario of the evolutionary radi- 
ation of gastropod prédation. Generalized 
proboscis probing was subsequently supple- 
mented by pedal manipulation, which led var- 
iously to shell boring, wedging, chipping, or 
pedal suffocation. It can be assumed that 
these initial stages represented preadapta- 
tions to shell boring; however, the specific or- 
igins of the complex accessory boring organ 
remain uncertain. The independent evolution 
of shell boring in a number of molluscan taxa 
represents convergent evolution; the struc- 
tures and processes are not necessarily ho- 
mologous. (See "Diversity of Boring Préda- 
tion" above for further comparisons). 

The Cretaceous radiation of naticids is part 
of the Mesozoic marine revolution, involving 
the increase in diversity of many modern ma- 
rine predators as a consequence of the "in- 
crease in shelled food supply resulting from 
the occupation of new adaptive zones by in- 
faunal bivalves and by shell-inhabiting hermit 
crabs" (Vermeij, 1977: 245). Specifically, the 
shift of bivalves from predominantely epifau- 
nal and byssate forms to infaunal, siphonate 
forms served as an escape from the then- 
dominant epifaunal and pelagic predators 
[cephalopods, asteroids, sharks and marine 
reptiles] (see also Taylor, 1 981 : 236) and sub- 
sequently led to selection favoring infaunal 
predators. If the early Mesozoic naticids were 
not burrowers (as suggested by their shell 
morphology), then burrowing in combination 
with shell boring would have opened up a new 
adaptive zone for the Cretaceous naticids. At 
the same time, the diversification of other 
sandy-habitat gastropods (especially turhtel- 
lids, turrids and terebrids) provided further in- 
faunal prey for naticids (Taylor et al., 1980: 
399). 

An important biogeographical phenomenon 
is the pattern of latitudinal diversity (pole- 
equator) of predatory prosobranch gastro- 
pods. For most of these marine families, in- 
cluding the Naticidae, there is a strong 
increase in species diversity from the poles to 
the tropical regions (the two exceptions are 
the Buccinidae and Turhdae) (Taylor & Tay- 
lor, 1977; Taylor et al., 1980: 381-3). Corre- 
lated with this gradient, Dudley & Vermeij 
(1978: 439) showed a marked equatorward 



178 



KABAT 



increase in boring prédation in Turritella. Sub- 
sequently Vermeij et al. (1989), for bivalve 
prey, actually observed an equatorward de- 
crease in the frequency of complete bore- 
holes (and a correlated equatorward increase 
in the frequency of incomplete boreholes): 
they suggested that the turntellids were an 
unexplained exception to this more general 
pattern. 

It appears that since the Cretaceous, the 
general mechanisms and consequences of 
naticid prédation have not greatly changed. 
To be sure, the prey sources have changed, 
not only due to origination and extinction of 
prey taxa, but also because of changes in 
prey defense mechanisms. However, the 
overall "strategy" of naticid prédation has per- 
sisted for the last 100 million years (Kitchell, 
1 987). It is possible that the naticids, following 
their late Cretaceous-early Tertiary adaptive 
radiation, have now reached their maximum 
taxonomic diversity (e.g. Sohl, 1969: fig. 1) 
and are at stasis which may lead to eventual 
decline in the absence of evolutionary inno- 
vations facilitating further expansion. The 
highly stereotyped nature of naticid prédation 
suggests that their canalization may be so 
great as to preclude further breakthroughs 
(but consider the non-boring, suffocation pré- 
dation of Polinices mammilla). 

With the rise of muricids in the later Ter- 
tiary, the naticids may have shifted from gas- 
tropod to bivalve prey, as suggested by Ade- 
goke & Tevesz (1974). Hoffman et al. (1974) 
noted that in a Miocene assemblage, naticid 
boreholes were found mostly in smooth prey 
whereas muricid boreholes were primarily in 
ribbed (sculptured) prey: however the former 
prey are more likely to be infaunal than the 
latter, which may affect these results. Within 
the Maryland Miocene. Kelley (1982a) found 
that naticid prédation shifted from predomi- 
nantly bivalve prey in the Calvert and Chop- 
tank formations to gastropod prey in the St. 
Marys Formation, correlated with the increase 
in diversity of prey gastropods in the latter 
formation. Kelley's results may be a preser- 
vational artifact, as the St. Mary's has a much 
better representation of gastropods than do 
the earlier formations (G. J. Vermeij, in litt.). 
Clearly, one also needs to account for 
changes in the relative abundances of infau- 
nal prey sources; trends as suggested by 
Adegoke & Tevesz (1974) may not be appli- 
cable on a global scale. In addition, the study 
of naticids has been primarily in a few re- 
stricted habitats: more comprehensive analy- 



ses of tropical sub-littoral communities may 
show other naticid prédation patterns. 

Kelley (1982a) suggested that extensive 
naticid and other prédation on bivalves in- 
creased prey species diversity, perhaps by 
reducing competitive interactions. Although 
ecologists recognize several factors that af- 
fect species diversity, prédation is undoubt- 
edly one of the more important, and one that 
can be easily recognized in the fossil record. 
Perturbation experiments involving predator- 
exclusion cages were used by Wiltse (1980b) 
to analyze the role of the western Atlantic Ne- 
venta duplicata in its community structure: 
she found that snail prédation and distur- 
bance (due to burrowing) actually decreased 
the community species diversity by eliminat- 
ing the rare species and blocking strong com- 
petitive interactions. 

Kitchell and colleagues (Kitchell et al., 1 981 : 
Kitchell, 1982, 1983, 1986: DeAngelis et al., 
1 984, 1 985, 1 989) expanded upon their model 
of the energetics of naticid prédation to de- 
velop models of coevolution of naticids and 
their prey. Coevolution, or the reciprocal 
evolutionary interactions of two taxa, is an im- 
portant, albeit difficult to quantify, aspect of 
evolutionary biology. There has been consid- 
erable disagreement as to how tightly or 
broadly coevolution should be defined or re- 
stricted. Indeed, almost any evolutionary trend 
can be "explained" as part of a coevolutionary 
process (Vermeij, 1982: 711-2). Instead of 
recognizing coevolution as "all evolution re- 
sulting from biological interactions," it is much 
more useful to restrict it to "reciprocal adap- 
tation involving the heritable traits of two or 
more species" (Vermeij, 1983b: 311). These 
models of naticid-prey coevolution are sub- 
ject to the same caveats mentioned earlier 
under the discussion of the previous models. 
Nevertheless, I shall attempt to summarize 
their scenarios. 

First, one can hypothesize that some sorts 
of evolutionary "arms races" are involved, 
with the prey evolving various antipredatory 
adaptations, but with the predator also evolv- 
ing new or changed features. One conse- 
quence is that "multiple adaptive tactics pro- 
duce multiple directionality" (Kitchell et al., 
1981 : 550), meaning that diversity may result 
as different prey follow alternative strategies 
and the same is true for different predators. 
This may result in character displacement or 
other isolating mechanisms resulting in spe- 
ciation (Kitchell, 1983). 

A direct test of these coevolutionary pro- 



NATICID PREDATION 



179 



cesses, at least for naticid predators, was con- 
ducted by Kitchell (1982) who analyzed Marin- 
covich's stratigraphie data for the eastern 
Pacific Neogene naticid fauna and concluded 
that predator "efficiency" increased over geo- 
logical time. Specifically, size, globosity and 
streamlining of the shell all increased, as did 
the proportion of apertural area to shell area 
and the general diversification of morphology 
(the latter not fully explained). In some re- 
spects these are all a consequence of general 
phyletic size increase, and may not be directly 
due to coevolutlon. 

Further refinements of their coevolutionary 
models predicted that in the absence of pred- 
ators, prey will reproduce early (i.e., at small 
sizes); whereas in the presence of predators, 
prey will show delayed reproduction at larger 
sizes (DeAngelis et al., 1984). More complex 
age-structured models tested the prey energy- 
allocation functions (growth vs. reproduction) 
as a consequence of prédation levels, and 
resulted in three alternative ecological strate- 
gies for bivalve prey as coevolutionary re- 
sponses: delayed reproduction to large size, 
early reproduction, or increased shell thick- 
ness. Needless to say, the numerous assump- 
tions (DeAngelis et al., 1985: 836) severely 
constrain the value of their coevolutionary 
model. In particular, they assume that no other 
factors affect the population dynamics of the 
naticids or their prey; this overlooks other 
predators, disease and parasitism, and abiotic 
mortality sources, all of which (together and 
severally) are often of greater importance to 
the prey than are naticids, as has been doc- 
umented in the other studies discussed herein. 
Of course, with respect to the evolution of shell 
morphology, the latter factors are not easily 
measured or of great significance. The results 
of their models largely corroborated the con- 
clusions of previous ecological studies. 

Edge-boring of bivalve prey represents 
an escalation in the evolutionary "arms race" 
as an adaptive response to the presence of 
prey sculptural elements and shell-thicken- 
ing. Similarly, non-boring prédation (suffoca- 
tion) also represents an alternative strategy 
(Ansell & Morton, 1987: 117); the selective 
advantages presumably entail a reduction in 
the energetic costs of boring. Further study 
should reveal whether some prey taxa are re- 
sistant to these novel prédation mechanisms. 
The phylogenetic correlations of these two 
traits remain uncertain; at the present time, 
they are only known for a few species from 
the tropical Indo-Pacific. 



To briefly summarize these ecological stud- 
ies: (a) There is a general positive correlation 
between predator and prey size; size selec- 
tivity is shown as larger prey often have a size 
refuge from prédation, (b) Prey defense 
mechanisms not only help prevent prey cap- 
ture, but also may lead to interruptions of 
prédation as shown by incomplete boreholes 
in the prey shell, (c) The successful mode 
of naticid prédation is limited by its seem- 
ing stereotypy (inflexibility), (d) The intrigu- 
ing possibilities of predator-prey coevolutlon 
(arms races) remain unproven for specific 
cases. 



FUTURE DIRECTIONS 

This review has suggested several areas 
needing further research. They are tabulated 
below; readers will undoubtedly recognize yet 
other problems amenable to future studies. 

The detection of prey by naticids remains a 
puzzle: elucidation of the potential interac- 
tions of chemosensory mechanisms (osphra- 
dium) vs. echolocation (Kitching & Pearson, 
1 981 ). A related mechanistic problem is to de- 
termine the precise biochemical constituents 
of the accessory boring organ secretion in 
naticids and the mode of function of shell dis- 
solution. 

More ecologically oriented approaches 
could include sophisticated field analyses of 
prey choice, entailing controlled manipula- 
tions and perturbation experiments (remove 
one species at a time). Further quantification 
of the various links of soft-bottom community 
food webs to determine more precisely the 
quantitative role of naticids in this habitat. De- 
velopment of methods of ecological control of 
naticid predators of shellfish. 

Paleontologists could analyze Paleoceno 
faunas for gastropod boring prédation; and 
conduct more detailed studies of Jurassic and 
Early Cretaceous faunas to supply informa- 
tion on changes in prédation and shell form 
during that time (Vermeij, 1987: 238-9). Fur- 
ther study of the phylogenetic position of the 
Thassic shell borers and the earty fossil 
record of naticids to unravel the complexities 
of the origin(s) of shell boring of the naticid 
type. 

Study of boring prédation from the cold 
temperature southern oceans and the sub- 
Antarctic would be most desirable. The pres- 
ence of several phylogenetically primitive nat- 



180 



KABAT 



¡cid taxa in those faunas would provide further 
clues as to the relationships between naticid 
phylogeny and boring prédation. It remains 
uncertain whether the most primitive subfam- 
ily, the Ampullospirinae [Tnassic? — Recent] 
are shell borers. 

Further research on the geographical and 
phylogenetic extent of epifaunal prédation, 
non-boring suffocation, and edge-boring 
would also add to our knowledge of the phy- 
logenetic correlations of prédation mecha- 
nisms. 



CONCLUSIONS 

(A) Bored or punched holes in prey shells 
are made by nine taxa of manne predators: 
naticid, muncid & capulid snails, octopods, 
Pseudostylochus (Turbellaria) and Asemich- 
thys (Pisces), all in mollusk shells; cassid 
snails in echinoids; Okadaia (Nudibranchia) in 
calcareous polychaete tubes; and nematodes 
in foraminiferal tests. Some terrestnal zonitid 
snails are also shell-borers. Shell-crushing 
predators (sharks, crustaceans) sometimes 
leave holes in othenA/ise intact prey shells. 

(B) Following prey capture, naticid boring is 
accomplished by alternate application to the 
prey shell of the radula and the proboscoideal 
secretory accessory boring organ. The dis- 
tinctive naticid borehole is countersunk, with 
beveled edges. 

(C) The data on naticid prey show that 
many soft-bottom families of bivalves and 
gastropods are subject to naticid prédation. 
Rocky-habitat taxa escape the infaunal natic- 
ids. 

(D) Boring prédation potentially attributable 
to naticids originated in the Triassic but 
shortly became extinct. The naticid boring 
habit definitively evolved in the Late Creta- 
ceous and has been documented through Ho- 
locene faunas, with an unstudied gap in the 
Paleoceno. No clear trends in rates of boring 
prédation since the Cretaceous are obvious. 

(E) Most studies have shown a positive cor- 
relation between predator size and prey size; 
also, smaller prey are usually subject to 
higher rates of naticid prédation. Incomplete 
boreholes reflect interruptions of prédation; 
multiple boreholes demonstrate inflexible ste- 
reotypy of naticid boring. Prey defense can 
take several forms; leaping or burrowing; 
thick or sculptured shells; chemical defenses; 
growth to large size; and the corbulid con- 
chiolin layer. Non-boring prédation, either 



through gaping shells or pedal suffocation, 
greatly confounds ecological studies since no 
signs of prédation are left on the prey shell. 

(F) Naticid prédation is an important and 
easily documented link in the food web of ma- 
rine soft-bottom communities; other predators 
often crush or remove their prey without leav- 
ing recognizable remains. 

(G) The evolution of naticid boring préda- 
tion is part of the Mesozoic marine revolution 
entailing the diversification of infaunal bi- 
valves and other gastropods which greatly in- 
creased naticid prey sources. Evolutionary 
escalation (defenses) on the part of prey taxa 
may have occurred since the Cretaceous; at- 
tempts to prove specific coevolutionary trends 
have been unsuccessful. 



ACKNOWLEDGEMENTS 

Preparation of this paper has been facili- 
tated by numerous colleagues who kindly pro- 
vided reprints of their papers on shell boring 
and related aspects of molluscan prédation. 
For the loan of specimens, photographs, or 
negatives, I thank R. F. Ambrose (U.C. Santa 
Barbara), A. Garback (Academy of Natural 
Sciences, Philadelphia), R. Mooi (National 
Museum of Natural History, Washington, 
D.C.), P. B. Mordan and J. D. Taylor (Bhtish 
Museum (Natural History)) and S. F. Norton 
(Friday Harbor Laboratories). S. P. Kool, G. 
Rosenberg, S. Vogel and E. O. Wilson pro- 
vided helpful comments on several aspects of 
this paper. I greatly appreciate the extensive 
discussion and review of the manuscript by K. 
J. Boss, A. H. Knoll, A. J. Kohn, J. D. Taylor, 
R. D. Turner, G. J. Vermeij and an anony- 
mous reviewer. 



LITERATURE CITED 

ABE. N., 1985, Breeding of Thais clavigera (Küster) 
and prédation of its eggs by Cronia margariticola 
(Broderip). Pp. 381-392, in. Proceedings of the 
Second International Workshop on the Malaco- 
fauna of Hong Kong and Southern China. (Eds. 
B. Morton & D. Dudgeon). Hong Kong: Hong 
Kong University Press. 

ADEGOKE, O. S. & M. J. S. TEVESZ, 1974. Gas- 
tropod prédation patterns in the Eocene of Nige- 
na. Lethaia. 7:17-24. 

AGERSBORG, H. P. K. 1920, The utilization of 
echinoderms and of gasteropod mollusks. The 
American Naturalist. 54: 414-426. 

AITKEN, A. E. & M. J. RISK, 1988, Biotic interac- 



NATICID PREDATION 



181 



tions revealed by macroborings in Arctic bivalve 
molluscs. Lethaia, 21: 339-350. 

AMBROSE, R. F. 1986, Effects of octopus préda- 
tion on motile invertebrates in a rocky subtidal 
community. Marine Ecology — Progress Series, 
30: 261-273. 

AMBROSE, R. F., В. J. LEIGHTON & E. B. 
HARTWICK, 1988, Characterization of boreholes 
by Octopus dofleini in the bivalve Saxidomus gi- 
ganteus. Journal of Zoology, 214: 491-503. 

ANKEL, W, E., 1938, Erwerb und Aufnahme der 
Nahrung bei den Gastropoden. Zoologischer An- 
zeiger, Supplementband 1 1. Verliandlungen der 
Deutschen Zoologischen Gesellschan, 40: 223- 
295. 

ANSELL, A. D., 1960, Observations on prédation of 
Venus striatula (Da Costa) by Natica aldeh 
(Forbes). Proceedings of the Malacological Soci- 
ety of London, 34: 157-164, pi. 9. 

ANSELL, A. D., 1961, Egg capsules of the dogfish 
(Scylliorhinus canícula. Linn.) bored by Natica 
(Gastropoda, Prosobranchia). Proceedings of 
the Malacological Society of London. 34: 248- 
249. 

ANSELL, A. D., 1969, Defensive adaptations to 
prédation in the Mollusca. Proceedings of the 
Symposium on Mollusca. Cochin. 2: 487-512. 
Mandapam: Marine Biological Association of In- 
dia. 

ANSELL, A. D., 1982a, Experimental studies of a 
benthic predator-prey relationship. I. Feeding, 
growth, and egg-collar production in long-term 
cultures of the gastropod drill Polinices aldeh 
(Forbes) feeding on the bivalve Tellina tenuis (da 
Costa). Journal of Experimental Marine Biology 
and Ecology. 56: 235-255. 

ANSELL, A. D., 1982b, Experimental studies of a 
benthic predator-prey relationship. II. Energetics 
of growth and reproduction, and food-conversion 
efficiencies, in long-term cultures of the gastro- 
pod drill Polinices aldeh (Forbes) feeding on the 
bivalve Tellina tenuis (da Costa). Journal of Ex- 
perimental Marine Biology and Ecology. 61: 1- 
29. 

ANSELL, A. D., 1982c, Experimental studies of a 
benthic predator-prey relationship: III. Factors 
affecting rates of prédation and growth in juve- 
niles of the gastropod drill Polinices catena (da 
Costa) in laboratory cultures. Malacologia, 22: 
367-375. 

ANSELL, A. D., 1983, Prey selection by the naticid 
gastropod. Polinices catena. Journal of Mollus- 
can Studies. Supplement. 12A: 1. 

ANSELL, A. D. & A.-M. MACÉ, 1978, Comparative 
studies of the gastropod. Polinices aldeh from 
Mediterranean and North Atlantic populations. 
Haliotis, 9: 65-72. 

ANSELL, A. D. & B. MORTON, 1985, Aspects of 
naticid prédation in Hong Kong with special ref- 
erence to the defensive adaptations of Bassina 
(Callanaitis) calophylla (Bivalvia). Pp. 635-660, 
in. Proceedings of the Second International 
Workshop on the Malacofauna of Hong Kong and 



Southern China. (Eds. B. Morton & D. Dudgeon). 
Hong Kong: Hong Kong University Press. 

ANSELL, A. D. & B. MORTON, 1987, Alternative 
prédation tactics of a tropical naticid gastropod. 
Journal of Experimental Marine Biology and 
Ecology, 111: 109-119. 

ARNOLD, A. J., F. D'ESCRIVAN & W. С PARKER, 
1985, Prédation and avoidance responses in the 
Foraminifera of the Galapagos hydrothermal 
mounds. Journal of Foraminiferal Research, 15: 
38-42. 

ARUA, L, 1989, Gastropod predators and their di- 
etary preference in an Eocene molluscan fauna 
from Nigeria. Palaeogeography, Palaeoclimatol- 
ogy. Palaeoecology. 72: 283-290. 

ARUA, I. & M. HOQUE, 1987, Prédation intensity in 
an Eocene molluscan assemblage from south- 
eastern Nigeria. Geologie en Mijnbouw. 66: 293- 
296. 

ARUA, I. & M. HOQUE, 1989a, Predatory gastro- 
pod boreholes in an Eocene molluscan assem- 
blage from Nigeria. Lethaia. 22: 49-59. 

ARUA, I. & M. HOQUE, 1989b, Study of the shape 
of naticid and muhcid borings in plan view in 
Eocene prey from Southeastern Nigeria. Palae- 
ogeography. Palaeoclimatology. Palaeoecology. 
72: 357-362. 

ARUA, I. & M. HOQUE, 1989c, Fossil predaceous 
gastropod borings from Nigeria. Palaeogeogra- 
phy. Palaeoclimatology. Palaeoecology. 73: 
175-183. 

BÄNDEL, К., 1988, Early ontogenetic shell and 
shell structure as aids to unravel gastropod phy- 
logeny and evolution. Malacological Review. 
Supplement. 4: 267-722. 

BAYLISS, D. E., 1986, Selective feeding on bi- 
valves by Polinices aldeh (Forbes) (Gastropoda). 
Ophelia. 25: 33-47. 

BEEBE, W., 1932, Snail folk. Nature Magazine. 19: 
207-212, 262-263. [Reprinted, pp. 204-223, in 
Nonsuch: land of water. New York: Blue Ribbon 
Books]. 

BENTON, M. J., 1986, Prédation by drilling gastro- 
pods. Nature. 321: 110-111. 

BERG, С J., Jr., 1975, A comparison of adaptive 
strategies of prédation among naticid gastro- 
pods. Biological Bulletin, 149(2): 420-421. 

BERG, С J., Jr., 1976, Ontogeny of predatory be- 
havior in marine snails (Prosobranchia: Nati- 
cidae). The Nautilus, 90: 1-4. 

BERG, С J., Jr. & S. NISHENKO, 1975, Stereotypy 
of predatory boring behavior of Pleistocene nat- 
icid gastropods. Paleobiology, 1 : 258-260. 

BERG, С J., Jr. & M. J. PORTER, 1974, A com- 
parison of predatory behaviors amoung the nat- 
icid gastropods Lunatia heros. Lunatia thseriata 
and Polinices duplicatus. Biological Bulletin. 147: 
469-470. 

BERNARD, F. R., 1967, Studies on the biology of 
the naticid clam drill Polinices lewisi (Gould) 
(Gastropoda Prosobranchiata). Fisheries Re- 
search Board of Canada. Technical Report. 42: 
1-41. 



182 



KABAT 



BERNARD, F. R. & J. W. BAGSHAW, 1969, His- 
tology and fine structure of the accessory boring 
organ of Polinices lewisi (Gastropoda, Proso- 
branchiata). Journal of the Fishenes Research 
Board of Canada. 26: 1451-1457. 

BERRY, A. J., 1982, Prédation by Natica maculosa 
Lamarck (Naticidae: Gastropoda) upon the tro- 
chacean gastropod Umbonium vestianum (L.) on 
a Malaysian shore. Journal of Experimental Ma- 
rine Biology and Ecology. 64: 71-89. 

BERRY, A. J., 1983, Oxygen consumption and as- 
pects of energetics in a Malaysian population of 
Natica maculosa Lamarck (Gastropoda) feeding 
on the trochacean gastropod Umbonium vestia- 
rium (L.). Journal of Experimental f\Aarine Biology 
and Ecology. 66: 93-100. 

BEU, A. G., R. A. HENDERSON & С S. NELSON, 
1972, Notes on the taphonomy and paleoecology 
of New Zealand Tertiary Spatangoida, New 
Zealand Journal of Geology and Geophysics. 1 5: 
275-286. 

BISHOP, G. A., 1975, Traces of Prédation. Pp. 
261-281, in. The study of trace fossils. (Ed. R. W. 
Prey). New York: Springer-Verlag. 

BOEKSCHOTEN, G. J., 1966, Shell bonngs of ses- 
sile epibiontic organisms as palaeoecological 
guides (with examples from the Dutch coast). 
Palaeogeography. Palaeoclimatology. Palaeo- 
ecology. 2: 333-379. 

BOEKSCHOTEN. G. J., 1967, Palaeoecology of 
some Mollusca from the Tielrode Sands 
(Pliocene, Belgium). Palaeogeography. Palaeo- 
climatology. Palaeoecology. 3: 311-362. 

BOGGS, C. H., J. A. RICE, J. A. KITCHELL & J. F. 
KITCHELL, 1984, Prédation at a snail's pace: 
what's time to a gastropod? Oecologia. 62: 13- 
17. 

BOUCOT, A. J., 1981, Principles of benthic manne 
paleoecology. New York: Academic Press, xvi + 
463 pp. 

BROMLEY, R. G., 1981, Concepts in ichnotaxon- 
omy illustrated by small round holes in shells. 
Acta Geológica Hispánica. 16: 55-64. 

BROOM. M. J., 1982, Size-selection, consumption 
rates and growth of the gastropods Natica mac- 
ulosa (Lamarck) and Thais carinifera (Lamarck) 
preying on the bivalve, Anadara granosa (L). 
Journal of Experimental Manne Biology and 
Ecology. 56: 213-233. 

BROOM, M. J., 1983, A preliminary investigation 
into prey species preference by the tropical gas- 
tropods Natica maculosa Lamarck and Thais car- 
inifera (Lamarck). Journal of Molluscan Studies, 
49: 43-52. 

BURCH, B. & T. A. BURCH, 1986, Prédation by 
Naticidae after a devastating storm in Pinna beds 
of Waikiki, Oahu, Hawaii 1980. Western Society 
of Malacologists. Annual Report. 18: 9. 

CADÉE, G. С 1968, Molluscan biocoenoses and 
thanatocoenoses in the Ria de Arosa, Galicia, 
Spain, Zoologische Verhandelingen. 95: 1-121, 
6 pis. 

CARRIKER, M. R., 1961, Comparative functional 



morphology of boring mechanisms in gastro- 
pods. American Zoologist. 1 : 263-266. 

CARRIKER, M. R., 1981, Shell penetration and 
feeding by naticacean and muricacean predatory 
gastropods: a synthesis. Malacologia. 20: 403- 
422. 

CARRIKER, M. R. & L. G. WILLIAMS, 1978, The 
chemical mechanism of shell dissolution by pred- 
atory boring gastropods: a review and an hypoth- 
esis. Malacologia. 17: 143-156. 

CARRIKER. M. R & E. L. YOCHELSON, 1968, 
Recent gastropod boreholes and Ordovician cy- 
lindrical borings. U. S. Geological Survey. Pro- 
fessional Paper. 593-B: 1-26, pis. 1-5. 

CARTER, R. M. 1968, On the biology and palae- 
ontology of some predators of bivalved Mollusca. 
Palaeogeography. Palaeoclimatology. Palaeo- 
ecology. 4: 29-65. 

CHATTERTON, В. D. E. & H. L. WHITEHEAD, 
1987, Predatory borings in the inarticulate bra- 
chiopod Artiotreta from the Silurian of Oklahoma. 
Lethaia. 20: 67-74. 

CHRiSTENSEN, A. M., 1970, Feeding biology of 
the sea-star Astropecten irregulans Pennant, 
Ophelia. 8: 1-134. 

CLARKE. A. H., Jr., 1956, Natural biological control 
of a Mya predator. The Nautilus. 70: 37-38. 

COLBATH, S. L., 1985, Gastropod prédation and 
depositional environments of two molluscan 
communities from the Miocene Astoria Formation 
at Beverly Beach State Park, Oregon. Journal of 
Paleontology. 59: 849-869. 

COMMITO, J, A., 1982, Effects of Lunatia heros 
prédation on the population dynamics of Mya 
arenana and Macoma balthica in Maine, USA. 
Marine Biology. 69: 187-193. 

COMMITO. J. A.. 1987, Polinices prédation pat- 
terns and Mercenaria morphology models. The 
American Naturalist. 129: 449-451. 

COMMITO, J. A. & W. G. AMBROSE, Jr., 1985, 
Predatory infauna and trophic complexity in soft- 
bottom communities. Pp. 323-333, in. Proceed- 
ings of the Nineteenth European Marine Biology 
Symposium. (Ed. P. E. Gibbs). Cambridge: Cam- 
bridge University Press. 

CROLL, R. P., 1983, Gastropod chemoreception. 
Biological Reviews. 58: 293-319. 

DARRAGH, T. A. & G. W. KENDRICK, 1980, 
Eocene bivalves from the Pallinup Siltstone near 
Walpole, Western Australia. Journal of the Royal 
Society of Western Australia. 63: 5-20. 

DEANGELIS, D. L., J. A. KITCHELL & W. M. 
POST, 1985, The influence of naticid prédation 
on evolutionary strategies of bivalve prey: con- 
clusions from a model. The American Naturalist, 
126: 817-842. 

DEANGELIS, D. L., J. A. KITCHELL & W. M. 
POST, 1987, Reply to Commito. The American 
Naturalist. 130: 458-460. 

DEANGELIS, D. L., J. A. KITCHELL & W. M. 
POST, 1989, The nature of stasis and change: 
potential coevolutionary dynamics. In, Coevolu- 



NATICID PREDATION 



183 



tion of Ecosystems. (Ed. N. Stenseth). Cam- 
bridge: Cambridge University Press. 

DEANGELIS, D. L., J. A. KITCHELL, W. M. POST 
& С С. TRAVIS, 1984, A model of naticid gas- 
tropod predator-prey coevolution. Mathematical 
Ecology, 54: 120-135. 

DE CAUWER, G., 1985, Gastropod prédation on 
corbulid bivalves: palaeoecology or taphonomy? 
Annales de la Société Royale Zoologique de Bel- 
gique. 115: 183-196. 

DUDLEY, E. С & E. С DUDLEY, 1980, Drilling 
prédation on some Miocene marine mollusks. 
The Nautilus, 94: 63-66. 

DUDLEY, E. С. & G. J. VERMEIJ, 1978, Prédation 
in time and space: drilling in the gastropod Tur- 
ritella. Paleobiology. 4: 436-441. 

EDWARDS, D. C, 1969, Predators on Olivella bi- 
pllcata. including a species-specific predator 
avoidance response. The Veliger. 1 1 : 326-333, 
pi. 51. 

EDWARDS, D. C, 1975, Preferred prey of Polin- 
ices duplicatus in Cape Cod inlets. Bulletin of the 
American Malacological Union. Inc., 40: 17-20. 

EDWARDS, D. С & J. D. HUEBNER, 1977, Feed- 
ing and growth) rates of Polinices duplicatus prey- 
ing on Mya arenaria at Barnstable Harbor, Mas- 
sachusetts. Ecology. 58: 1218-1236. 

EKDALE, A. A., R. G. BROMLEY & S. G. PEM- 
BERTON, 1984, Ichnology: the use of trace fos- 
sils in sedimentology and stratigraphy. Society of 
Economic Paleontologists and Mineralogists. 
Short Course. 15: 1-317. 

FANKBONER, P. V., 1969, Naticid prédation on 
Dentalium complexum (Dentaliidae, Sca- 
phopoda). Hawaiian Shell News, 17(6): 3. 

FATTON, E. & J. ROGER, 1968, Les organismes 
perforants: vue d'ensemble sur les actuels et les 
fossiles. Université de Paris, Faculté des Sci- 
ences d'Orsay, Travaux du Laboratoire de Pale- 
ontologie, 13-53. 

FISCHER, P.-H., 1922, Sur les gastéropodes per- 
ceurs. Journal de Conchyliologie, 67: 1-56. 

FISCHER, P.-H., 1960, Action des gastéropodes 
perceurs sur un bivalve de l'Étage Lutétien. Jour- 
nal de Conchyliologie, 100: 129-131. 

FISCHER, P.-H., 1962a, Perforations de fossiles 
Pré-Tertiaries attribuées a des gastéropodes pré- 
dateurs. Journal de Conchyliologie, 102: 68-78. 

FISCHER, P.-H., 1962b, Action des gastéropodes 
perceurs sur des Mesalia de l'Étage Lutétien. 
Journal de Conchyliologie, 102: 95-97. 

FISCHER, P.-H., 1963, Corbules fossiles perforées 
par des gastéropodes prédateurs. Journal de 
Conchyliologie, 103: 29-31. 

FISCHER, P.-H., 1966, Perforations de fossiles 
Tertiaires par des gastéropodes prédateurs. 
Journal de Conchyliologie, 105: 66-96. 

FLOWER, F. В., 1954, A new enemy of the oyster 
drill. Science, 120: 231-232. 

FRANZ, D. R., 1977, Size and age-specific préda- 
tion by Lunatia heros (Say, 1822) on the surf 
clam Spisula solidissima (Dillwyn, 1817) off 



Western Long Island, New York. The Veliger, 20: 
144-150. 

FRETTER, V. & A. GRAHAM, 1962, British proso- 
branch molluscs: their functional anatomy and 
ecology. London: The Ray Society, xvi + 755 pp. 

FRETTER, V. & R. MANLY, 1979, Observations on 
the biology of some sublittoral prosobranchs. 
Journal of Molluscan Studies. 45: 209-218. 

FREY, R. W., J. D. HOWARD & J.-S. HONG, 1987 
["1986"], Naticid gastropods may kill solenid bi- 
valves without boring: ichnologic and taphonomic 
consequences. Palaios. 1: 610-612. 

FÜRSICH, F. T., & D. JABLONSKI, 1984, Late Tn- 
assic naticid drillholes: carnivorous gastropods 
gain a major adaptation but fail to radiate. Sci- 
ence, 224: 78-80. 

FÜRSICH, F. T., & J. WENDT, 1977, Biostratinomy 
and palaeoecology of the Cassian Formation 
(Thassic) of the Southern Alps. Palaeogeogra- 
phy, Palaeoclimatology, Palaeoecology, 22: 
257-323. 

GALLENI, L., P. TONGIORGI, E. PERRERO & U. 
SALGHETTI, 1980, Stylochus mediterraneus 
(Turbellaria: Polycladia), predator on the mussel 
Mytilus galloprovincialis. Marine Biology, 55: 
317-326. 

GEORGE, С J., 1965, The use of beached valves 
of the lamellibranch molluscs Glycimeris glycim- 
eris (L.), Donax semistriatus Poli and Donax trun- 
culus L. for the determination of percentage mor- 
tality by Natica spp. Doriana, Supplemente agli 
Annali del Museo Cívico di Storia Naturale "G. 
Doria", 4(164): 1-8. 

CONOR, J. J., 1965, Predator-prey reactions be- 
tween two marine prosobranch gastropods. The 
Veliger, 7: 228-232. 

GREEN, R. H., 1968, Mortality and stability in a low 
diversity subtropical intertidal community. Ecol- 
ogy. 49: 848-854. 

GRIFFITHS, R. J., 1981, Prédation on the bivalve 
Choromytilus meridionalis (Kr.) by the gastropod 
Natica (Tectonatica) tecta Anton. Journal of Mol- 
luscan Studies, 47: 112-120. 

GUERRERO, S. & R. A. REYMENT, 1988a, Pré- 
dation and feeding in the naticid gastropod Nati- 
carius intricatoides (Hidalgo). Palaeogeography, 
Palaeoclimatology, Palaeoecology, 68: 49-52. 

GUERRERO, S. & R. A. REYMENT, 1988b, Differ- 
entiation between the traces of prédation of mu- 
ricids and naticids in Spanish Pliocene Chlamys. 
Estudios Geológicos. Instituto de Investigaciones 
Geológicas "Lucas Mallada". Madrid. 44(3-4): 
317-328. 

HANKS, J. E., 1953, The effect of changes in water 
temperature and salinity on the feeding habits of 
the boring snails, Polinices heros and Polinices 
duplicata. Fifth Report on Investigations of the 
Shellfisheries of Massachusetts, pp. 33-47. 

HÄNTZCHEL, W., 1 975, Trace fossils and problem- 
ática, second edition. In Treatise on Invertebrate 
Paleontology. Part W. Miscellanea, Supplement, 
1: xxii + W1-W269. (Ed. С Teichert). Boulder: 
Geological Society of America. 



184 



KABAT 



HAYASAKA, I.. 1933. Fossil occurrence of pelecy- 
pod shells bored by certain gastropods. Memoirs 
of the Faculty of Science and Agnculture. 
Taihoku Impenal University. 6(4): 65-70. pis. 
1[XVIII]-4[XXI]. 

HICKMAN. С S.. 1983. Radular patterns, system- 
atics. diversity, and ecology of deep-sea limpets. 
The Veliger. 26: 73-92. 

HINGSTON. J. P.. 1985. Prédation patterns among 
molluscs in the Victorian Tertiary. Proceedings of 
the Royal Society of Victona. 97: 49-57. 

HOFFMAN. A.. 1967a. Mortality patterns of some 
bivalves from the Badenian (Miocene) Korytnica 
Clays, Poland. Neues Jahrbuch für Geologie 
und Paläontologie. Monatshefte. 1976(5): 337- 
349. 

HOFFMAN. A.. 1976b. Mortality patterns of some 
gastropods from the Badenian (Miocene) Koryt- 
nica Clays. Poland. Neues Jahrbuch für Geologie 
und Paläontologie. Abhandlungen. 152: 293- 
306. 

HOFFMAN. A. & J. MARTINELL, 1984. Prey selec- 
tion by naticid gastropods in the Pliocene of Em- 
porda (Northeast Spam). Neues Jahrbuch für 
Geologie und Paläontologie. Monatshefte. 
1984(7): 393-399. 

HOFFMAN. A. & В. SZUBZDA. 1976. Paleoecol- 
ogy of some molluscan assemblages from the 
Badenian (Miocene) marine sandy faciès of Po- 
land. Palaeogeography. Palaeoclimatology, 
Palaeoecology. 20: 307-332. 

HOFFMAN. A.. A. PISERA & M. RYSZKIEWICZ, 
1974. Prédation by muricid and naticid gastro- 
pods on the Lower Tortonian mollusks from the 
Korytnica clays. Acta Geológica Polonica. 24: 
249-260. pis. 1-4. 

HUEBNER. J. D. &D.C. EDWARDS. 1981, Energy 
budget of the predatory marine gastropod Polin- 
ices duplicatus. Manne Biology. 61 : 221-226. 

HUGHES. R. N.. 1985. Predatory behaviour of /Vai- 
ica unifasciata feeding intertidally on gastropods. 
Journal of Molluscan Studies. 51 : 331-335. 

HUGHES. R. N. & H. P. I. HUGHES, 1981, Mor- 
phological and behavioural aspects of feeding in 
the Cassidae (Tonnacea, Mesogastropoda). Ma- 
lacologia. 20: 385-402. 

HUTCHINGS. J. A. & R. L. HAEDRICH, 1984. 
Grow/th and population structure in two species of 
bivalves (Nuculanidae) from the deep sea. Ma- 
nne Ecology Progress Senes. 17: 135-142. 

ILINA. L. G., 1987. Evidence of boring in shells of 
brackish-water gastropods. Paleontological Jour- 
nal. 21(3): 23-30. 

(ICZN) INTERNATIONAL COMMISSION ON ZOO- 
LOGICAL NOMENCLATURE, 1985. Interna- 
tional Code of Zoological Nomenclature. Third 
Edition. Berkeley: University of California Press. 
XX ^ 338 pages. 

JACKSON, J. B. C, 1972, The ecology of the mol- 
luscs of Thalassia communities, Jamaica, West 
Indies. II. Molluscan population variability along 
an environmental stress gradient. Marine Biol- 
ogy. 14: 304-337. 



JACOBSON. M. К.. 1965. Double-drilled clams. 
New York Shell Club Notes. 111:5-6. 

JACOBSON. M. К.. 1968. Preliminary results of a 
study on the placement of Polinices holes in the 
shells of Spisula solidissima at Rockaway, N.Y. 
New York Shell Club Notes. 140: 6-7. 

JENSEN. A. S.. 1951. Do the Naticidae (Gas- 
tropoda Prosobranchia) drill by chemical or by 
mechanical means^ Videnskabelige Meddelelser 
fra Dansk Naturhistonsk Forening. Kobenhavn. 
113: 251-261. 

JONES. M. L.. 1969. Bonng of shell by Caobangia 
in freshwater snails of southeast Asia. American 
Zoologist. 9: 829-835. 

KABAT. A. R. & A. J. KOHN. 1986, Prédation on 
Early Pleistocene naticid gastropods in Fiji. 
Palaeogeography. Palaeoclimatology. Palaeo- 
ecology. 53: 255-269. 

KELLEY. P. H., 1982a. The effect of prédation on 
Miocene mollusc populations of the Chesapeake 
Group. Florida Bureau of Geology. Special Pub- 
lication. 25: 35-48. 

KELLEY. P. H., 1982b, Prey preference of naticid 
gastropods of the Chesapeake Group: taxonomic 
and stratigraphie patterns. Geological Society of 
America Abstracts with Programs. 14: 527. 

KELLEY, P. H., 1984, Coevolution in a naticid gas- 
tropod predator-prey system: relation of préda- 
tion intensity to rates of prey evolution. Geologi- 
cal Society of America Abstracts with Programs, 
16: 557 

KELLEY. P. H.. 1987, Naticid gastropod prey pref- 
erence and predator-avoidance strategies of 
Chesapeake Group bivalves. Geological Society 
of America Abstracts with Programs. 19: 724. 

KELLEY. P. H., 1989a ["1988"], Prédation by Mi- 
ocene gastropods of the Chesapeake Group: 
stereotyped and predictable. Palaios. 3: 436- 
448. 

KELLEY. P. H., 1989b, Evolutionary trends within 
bivalve prey of Chesapeake Group naticid gas- 
tropods. Historical Biology. 2: 139-156. 

KITCHELL, J. A., 1982, Coevolution in a predator- 
prey system. Third North American Paleontolog- 
ical Convention. Proceedings. 2: 301-305. 

KITCHELL, J. A., 1983, An evolutionary model of 
predator-mediated divergence and coexistence. 
Geological Society of America Abstracts with 
Programs. 15: 614. 

KITCHELL. J. A., 1986, The evolution of predator- 
prey behavior: naticid gastropods and their mol- 
luscan prey. Pp. 88-110, in Evolution of Animal 
Behavior: Paleontological and Field Approaches. 
(Eds. M. H. Nitecki & J. A. Kitchell). New York: 
Oxford University Press. 

KITCHELL, J. A., 1987, Naticid prédation within 
Cenozoic molluscan communities. American Ma- 
lacological Union Program and Abstracts. 53rd 
Annual Meeting, p. 14. 

KITCHELL, J. A., С H. BOGGS, J. F. KITCHELL & 
J. A. RICE. 1981, Prey selection by naticid gas- 
tropods: experimental tests and application to the 
fossil record. Paleobiology, 7: 533-552. 



NATICID PREDATION 



185 



KITCHELL, J. A., С H. BOGGS, J. A. RICE, J. F. 
KITCHELL. A. HOFFMAN & J. MARTINELL, 
1986, Anomalies in naticid predatory behavior: a 
critique and experimental observations. Malaco- 
logia, 27: 291-298. 

KITCHING, R. L. & J. PEARSON, 1981, Prey loca- 
tion by sound in a predatory intertidal gastropod. 
Marine Biology Letters. 2: 313-321. 

KLÄHN, H., 1932, Der Bohrakt fossiler bohrender 
Schnecken und das Vernichtungsmass durch 
räuberische Gastropoden des Sternberger Oli- 
gozäns. Sitzungsberichte und Ablihiandlungen 
der Naturforsctienden Gesellschiaft zu Rostock. 
3: 89-103. 

KOHN, A. J., 1961, Chemoreception in gastropod 
molluscs. American Zoologist. 1: 291-308. 

KOHN, A. J., 1983, Feeding biology of gastropods. 
Pp. 1-63, in. The Mollusca. 5. Physiology. Pari 2. 
(Eds. A. S. M. Saleuddin & K. M. Wilbur). New 
York: Academic Press. 

KOHN, A. J., 1985, Gastropod paleoecology. Pp. 
174-189, in, Mollusks: Notes for a short course. 
University of Tennessee, Studies in Geology. 13. 
(Eds. D. J. Bottjer et al.). 

KOJUMDJIEVA, е., 1974, Les Gastéropodes per- 
ceurs et leurs victimes du Miocene de Bulgarie 
du Nord-Ouest. Bulgarian Academy of Sciences. 
Bulletin of the Geological Institute. Series Pale- 
ontology. 23: 5-24, pis. 1-4. 

KORNICKER, L. S., С D. WISE & J. M. WISE, 
1963, Factors affecting the distribution of oppos- 
ing mollusk valves. Journal of Sedimentary Pe- 
trology. 33: 703-712. 

KORRINGA, P., 1952, Recent advances in oyster 
biology. Quarterly Review of Biology, 27: 266- 
308, 339-365. 

LAWS, H. M. & D. F. LAWS, 1972, The escape 
response of Donacilla angusta Reeve (Mollusca: 
Bivalvia) in the presence of a naticid predator. 
The Veliger. 14: 289-290. 

LEIDY, J., 1878, Remarks on Mactra. Proceedings 
of the Academy of Natural Sciences of Philadel- 
phia for 1878: 332-333. 

LEVER, J. & R. THIJSSEN, 1968, Sorting phenom- 
ena during the transport of shell valves on sandy 
beaches studied with the use of artificial valves. 
Pp. 259-271, in. Studies in the structure, physi- 
ology, and ecology of molluscs. Symposia of the 
Zoological Society of London, 22. (Ed. V. Fretter). 

LEVER, J., A. KESSLER, P. VAN OVERBEEKE & 
R. THIJSSEN, 1961, Quantitative beach re- 
search. II. The "hole effect": a second mode of 
sorting of lamellibranch valves on sandy 
beaches. Netherlands Journal of Sea Research, 
1: 339-358. 

LEWY, Z. & C. SAMTLEBEN, 1979, Functional 
morphology and palaeontological significance of 
the conchiolin layers in corbulid pelecypods. 
Lethaia, 12: 341-351. 

LIVAN, M., 1937, Über Bohr-Löcher an rezenten 
und fossilen Invertebraten. Senckenbergiana, 
19: 138-150, 314. 

MACÉ, A. -M., 1978, Note préliminaire sur les fac- 



teurs influeçant la sélection des proies chez le 
gastéropode perceur Polinices alderi (Forbes). 
Haliotis, 9: 61-64. 

MACÉ, A. -M., 1981a, Étude expérimentale de l'éc- 
ophysiologie d'un gastéropode perceur. Polin- 
ices aldeh (Forbes). 1. Alimentation, Croissance, 
Reproduction. Téthys. 10: 63-72. 

MACÉ, A. -M., 1981b, Étude expérimentale de l'éc- 
ophysiologie d'un gastéropode perceur. Polin- 
ices alderi (Forbes). 2. Respiration et Excrétion 
azotée. Téthys, 10: 73-82. 

MACÉ, A. -M., 1981c, Étude expérimentale de l'éc- 
ophysiologie d'un gastéropode perceur, Polin- 
ices alderi (Forbes). 3. Bilan énergétique. Téthys. 
10: 117-120. 

MACÉ, A.-M. & A. D. ANSELL, 1982, Respiration 
and nitrogen excretion of Polinices alderi 
(Forbes) and Polinices catena (da Costa) (Gas- 
tropoda: Naticidae). Journal of Experimental Ma- 
rine Biology and Ecology. 60: 275-292. 

MADDOCKS, R. F., 1988, One hundred million 
years of prédation on ostracods: the fossil record 
in Texas. Pp. 637-657, in. Evolutionary Biology 
of Ostracoda. Developments in Palaeontology 
and Stratigraphy. 11. (Eds. T. Hanai et al.). Am- 
sterdam: Elsevier. 

MAPES, R. H., T R. FAHRER & L. E. BABCOCK, 
1989, Sublethal and lethal injuries of Pennsylva- 
nian conulariids from Oklahoma. Journal of Pa- 
leontology. 63: 34-37. 

MARGOLIN, A. S., 1975, Responses to sea stars 
by three naticid gastropods. Ophelia, 14: 85-92. 

MARTINELL, J. & M. J. MARQUINA, 1980 
["1978"], Señales de depredación en los Gas- 
tropoda procedentes de un yacimiento pliocénico 
de Möllns de Rei (Barcelona). — Implicaciones 
paleoecológicas. Acta Geológica Hispánica. 
13(4): 125-128. 

MARTINELL, J. & J. DE PORTA 1980, Observa- 
clones sobre la depredación en Chamelea gall- 
ina (Linné) procedentes de Salou (Tarragona). 
Comunicaciones del Primer Congreso Nacional 
de Malacologia (1979), Madrid: 75-78. 

MARTINELL, J. & J. DE PORTA, 1982, Observa- 
tions on the molluscan thanatocoenoses from 
Platja Llarga (Salou, Spain). Malacologia, 22: 
587-591 . 

MATSUKUMA, A., 1976, Prédation by Recent nat- 
icid snails of two species of Glycymehs (Bi- 
valvia). Report of Fishery Research Laboratory, 
Kyushu University. Fukuoka, 3: 15-31. 

MATSUKUMA, A., 1978, Fossil boreholes made by 
shell-boring predators or commensals. I. Bore- 
holes of capulid gastropods. Venus. 37: 29-45. 

MAXWELL, P. A., 1988, Late Miocene deep-water 
Mollusca from the Stillwater Mudstone at Grey- 
mouth, Westland, New Zealand: paleoecology 
and systematics. New Zealand Geological Sur- 
vey Paleontological Bulletin. 55: 1-120, 13 pis. 

MEDCOF, J. С & L. W. THURBER, 1958, Trial 
control of the greater clam drill (Lunatia heros) by 
manual collection. Journal of the Fisheries Re- 
search Board of Canada, 15: 1355-1369. 



186 



KABAT 



MORDAN, P. В., 1977. Factors affecting the distri- 
bution and abundance of Aegopinella and Neso- 
vitrea (Pulmonata: Zonitidae) at Monks Wood 
National Nature Reserve, Huntingdonshire. Bio- 
logical Journal of the Linnean Society. 9: 59-72. 

MUKAI, H., 1973. Predators of the benthic bivalves 
on the muddy bottom m the Inland Sea of Japan. 
I. The effect of the prédation of Neventa reiniana 
(Dunker) on Veremolpa micra (Pilsbry). Venus. 
31; 147-156. 

NAKAMINE, T. & T. HABE. 1983. Shells of 
Tellinella virgata (Linnaeus) bored by the moon 
shell. Chirobotan. 14: 26. [in Japanese]. 

NEGUS. M., 1975, An analysis of boreholes drilled 
by Natica catena (da Costa) in the valves of 
Donax vittatus (da Costa). Proceedings of thie 
Malacological Society of London. 41 : 353-356. 

NEWTON, С R., 1983. Triassic origin of shell- 
boring gastropods. Geological Society of Amer- 
ica Abstracts with Programs. 15: 652-653. 

NEWTON, С R., M. T. WHALEN, J. B. THOMP- 
SON, N. PRINS & D. DELALLA, 1987, System- 
atics and paleoecology of Norian (Late Triassic) 
bivalves from a tropical island arc: Wallowa Ter- 
rane, Oregon. The Paleontological Society l[/lem- 
oir 22. Journal of Paleontology. 61(4), Supple- 
ment: 1-83. 

NIXON, M. & E. MACONNACHIE, 1988, Drilling by 
Octopus vulgaris (Mollusca: Cephalopoda) in the 
Mediterranean. Journal of Zoology. 216: 687- 
716. 

NORTON. S. P., 1988, Role of the gastropod shell 
and operculum in inhibiting prédation by fishes. 
Science. 241: cover, 92-94. 

PAINE. R. T., 1963, Trophic relationships of 8 sym- 
patric predatory gastropods. Ecology. 44: 63-73. 

PAPP. A., H. ZAPFE, F. BACHMAYER & A. F. 
TAUBER. 1947, Lebensspuren mariner Krebse. 
Akademie der Wissenschaften in Wien. Mathe- 
matisch-naturwissenschaftliche Klasse. Sit- 
zungsberichte. Abteilung I, Mineralogie, Biologie, 
Erdkunde, 155: 281-317. 

PELSENEER. P., 1924, Comment mangent divers 
gastropodes aquatiques. I. — gastropodes marins 
carnivores Natica et Purpura. Annales de la So- 
ciété Royale Zoologique de Belgique. 55: 31-43. 

PENNEY, A. J. & С L. GRIFFITHS, 1984, Prey 
selection and the impact of the starfish Marthas- 
tenas glacialis (L.) and other predators on the 
mussel Choromytilus meridionalis (Krauss). 
Journal of Experimental Manne Biology and 
Ecology. 75: 19-36. 

PERRY, L. M. 1 940, Manne shells of the southwest 
coast of Florida. Bulletins of American Paleontol- 
ogy. 26(95): 1-260, pis. 1-40. 

PETERSON, С H., 1982, The importance of pré- 
dation and intra- and interspecific competition in 
the population biology of two infaunal suspen- 
sion-feeding bivalves, Protothaca staminea and 
Chione undatella. Ecological Monographs. 52: 
437-475. 

PIÉRON, H., 1933, Notes ethologiques sur les gas- 
téropodes perceurs et leur comportement avec 



utilisation de méthodes statistiques. Archives de 
Zoologie Expérimentale et Generale. 75(1): 1- 
20, pi. 1. 

REID, R. G. B. & B. D. GUSTAFSON, 1989, Update 
on feeding and digestion in the moon snail Polin- 
ices lewisii (Gould, 1847). The Veliger. 32: 327. 

reíd, R. G. B. & J. A. FRIESEN, 1980, The diges- 
tive system of the moon snail Polinices lewisii 
(Gould, 1847) with emphasis on the role of the 
oesophageal gland. The Veliger, 23: 25-34. 

REYMENT, R. a., 1966, Preliminary observations 
on gastropod prédation in the western Niger 
delta. Palaeogeography. Palaeoclimatology, 
Palaeoecology, 2: 81-102. 

REYMENT, R. A., 1967, Paleoethology and fossil 
drilling gastropods. Transactions of the Kansas 
Academy of Sciences, 70: 33-50. 

REYMENT, R. A., E. R. REYMENT & A, HONIG- 
STEIN, 1987, Prédation by bonng gastropods on 
Late Cretaceous and Early Palaeocene ostra- 
cods. Cretaceous Research, 8: 189-209. 

RICHTER, G., 1962, Beobachtungen zum Beute- 
fang der marinen Bohrschnecke Lunatia nitida. 
Natur und Museum, 92: 186-192. 

ROBBA, E. & F. OSTINELLI, 1975, Studi paleoeco- 
logici sul Pliocene Ligure. I. Testimonianze di 
predazione sui molluschi Pliocenici di Albenga. 
Rivista Italiana di Paleontología e Stratigrafia. 81 : 
309-372, pis. 41-48. 

RODRIGUES, С. L., 1986, Prédation of the naticid 
gastropod, Neventa didyma (Roding), on the bi- 
valve, Ruditapes philippinarum (Adams & 
Reeve): evidence for a preference linked func- 
tional response. Publications from the Amakusa 
Manne Biological Laboratory. Kyushu University, 
8: 125-141. 

RODRIGUES, С. L., S. NOJIMA & T. KIKUCHI, 
1987, Mechanics of prey size preference in the 
gastropod Neverita didyma preying on the bi- 
valve Ruditapes philippinarum. Marine Ecology 
Progress Series. 40: 87-93. 

ROSEWATER, J., 1980, Predator boreholes in 
Periploma margaritaceum with a brief survey of 
other Periplomatidae (Bivalvia: Anomalodes- 
mata). The Veliger. 22: 248-251, 1 pi. 

SAIDOVA, H. M. & K. V. BEKLEMISHEV, 1953, [On 
the presence in marine sediments of foraminifer- 
ans drilled by young gastropods]. Doklady Aka- 
demia Nauk SSSR, Leningrad, 92: 1061-1063, 
[in Russian]. 

SANDER, F. & С M. LALLI, 1982, A comparative 
study of mollusk communities on the shelf-slope 
margin of Barbados, West Indies. The Veliger, 
24: 309-318. 

SAVAZZI, E. & R. A. REYMENT, 1989, Subaenal 
hunting behaviour in Natica gualteriana (Naticid 
Gastropod). Palaeogeography, Palaeoclimatol- 
ogy, Palaeoecology, 74: 355-364. 

SCHÄFER, W., 1972, Ecology and palaeoecology 
of manne environments, [transi. I. Oertel]. Chi- 
cago: University of Chicago Press, xiv Í 568 pp., 
pis. 1-39. 

SCHNEIDER, D., 1982, Escape response of an in- 



NATICID PREDATION 



187 



faunal claim Ensis directus Conrad, 1843 to a 
predatory snail, Polinices duplicatus Say. 1822. 
The Veliger, 24: 371-372. 

SELIN, N. I., S. K. PONUROVSKY & M. Zh. 
CHERNYAEV, 1986, [The effect of the predatory 
gastropod Cryptonatica ¡anthostoma on popula- 
tion structure of the bivalve Ruditapes philippi- 
narum]. Biologiya Morya. 1986(5): 72-74. [In 
Russian]. 

SILER, W. L., 1965, Feeding habits of some 
Eocene carnivorous gastropods. The Texas 
Journal of Science, 17:213-218. 

SLITER, W. v., 1 971 , Prédation on benthic foramin- 
ifers. Journal of Foraminiferal Research, 1 : 20- 
29, pis. 1-3. 

SMITH, J. E., 1932, The shell gravel deposits, and 
the infauna of the Eddystone grounds. Journal of 
the Mahne Biological Association of the United 
Kingdom, 18: 243-278. 

SMITH, S. A., С W. THAYER & С E. BRETT, 
1985, Prédation in the Paleozoic: gastropod-like 
drillholes in Devonian brachiopods. Science, 
230: 1033-1035. 

SOHL, N. F., 1969, The fossil record of shell bonng 
by snails. American Zoologist, 9: 725-734. 

STAFFORD, W. L., 1988, Polinices through the 
ages. Fossils Quarterly, 7: 18-22. 

STANTON, R. J., Jr., & P. С NELSON, 1980, Re- 
construction of the trophic web in paleontology: 
community structure in the Stone City Formation 
(Middle Eocene, Texas). Journal of Paleontol- 
ogy, 54: 118-135. 

STANTON, R. J., Jr., E. N. POWELL & P. С NEL- 
SON, 1981, The role of carnivorous gastropods 
in the trophic analysis of a fossil community. Ma- 
lacologia, 20: 451-469. 

STENZLER, D. & J. ATEMA, 1977, Alarm response 
of the marine mud snail, Nassarius obsoletus: 
specificity and behavioral priority. Journal of 
Chemical Ecology, 3: 159-171. 

STEVANOVIC, P., 1950, [Sur les coquilles perforés 
des mollusques du Sarmatien, Méotien et de 
Ceux de Bouglowka en Serbie]. Annales Gé- 
ologiques de la Péninsule Balkanique. 18: 129- 
142, pis. 1-4. [in Serbian: Russian and French 
abstracts]. 

STUMP, T. E., 1975, Pleistocene molluscan paleo- 
ecology and community structure of the Puerto 
Libertad Region, Sonora, Mexico. Palaeogeogra- 
phy, Palaeoclimatology, Palaeoecology. 17: 
177-226. 

TAYLOR, J. D., 1970, Feeding habits of predatory 
gastropods in a Tertiary (Eocene) molluscan as- 
semblage from the Paris Basin. Palaeontology, 
13: 254-260, pi. 46. 

TAYLOR, J. D., 1980, Diets and habitats of shallow 
water predatory gastropods around Tolo Chan- 
nel, Hong Kong. Pp. 163-180, in. The malaco- 
fauna of Hong Kong and southern China: Pro- 
ceedings of the first international workshop. (Ed. 
B. Morton). Hong Kong: Hong Kong University 
Press. 

TAYLOR, J. D., 1981, The evolution of predators in 



the late Cretaceous and their ecological signifi- 
cance. Pp. 229-240, in. The evolving biosphere. 
(Ed. P. L. Forey). Cambridge: Cambridge Univer- 
sity Press. 

TAYLOR, J. D. & С N. TAYLOR, 1977, Latitudinal 
distribution of predatory gastropods on the east- 
ern Atlantic shelf. Journal of Biogeography. 4: 
73-81. 

TAYLOR, J. D., R. J. CLEEVELY & N. J. MORRIS, 
1983, Predatory gastropods and their activities in 
the Blackdown Greensand (Albian) of England. 
Palaeontology. 26: 521-553. 

TAYLOR, J. D., N. J. MORRIS & С N. TAYLOR, 
1980, Food specialization and the evolution of 
predatory prosobranch gastropods. Palaeontol- 
ogy, 23: 375-409. 

THOMAS, R. D. K., 1976, Gastropod prédation on 
sympatric Neogene species of Glycymens (Bi- 
valvia) from the Eastern United States. Journal of 
Paleontology, 50: 488-499, pi. 1. 

THORSON G. [A.W.], 1935, Studies on the egg- 
capsules and development of Arctic marine 
prosobranchs. l\/leddelelser am Grönland. 
100(5): 1-71. 

TSIKHON-LUKANINA, E. A., 1987, [Trophobiology 
of aquatic mollusks.] Moscow: Akademiia Nauka 
SSSR. 176 pp. [in Russian]. 

TURNER, H. J., Jr., 1955, How clam dnils capture 
razor clams. The Nautilus, 69: 20-22. 

TURNER, H. J. Jr., J. С AYERS & С L. 
WHEELER, 1948, The horseshoe crab and bor- 
ing snail as factors limiting the abundance of the 
soft-shell clam. Pp. 43-45, in. Report on inves- 
tigations of the propagation of the soft-shell clam, 
Mya arenaria. (Ed. H. J. Turner, Jr.). Woods Hole 
Océanographie Institution, Contribution 462. 

VALE, F. K. & M. A. REX, 1988, Repaired shell 
damage in deep-sea prosobranch gastropods 
from the western North Atlantic. fJlalacologia. 28: 
65-79. 

VERLAINE, L., 1936, L'instinct et l'intelligence chez 
les mollusques. Les gastéropodes perceurs de 
coquilles. Mémoires du Musée Royal d'Histoire 
Naturelle de Belgique (ser. 2) 3: 387-394. 

VERMEIJ, G. J., 1977, The Mesozoic marine rev- 
olution: evidence from snails, predators and 
grazers. Paleobiology. 3: 245-258. 

VERMEIJ, G. J., 1978, Biogeography and adapta- 
tion: patterns of mahne life. Cambridge: Harvard 
University Press, xvi + 322 pp. 

VERMEIJ, G. J., 1980, Dnlling prédation of bivalves 
in Guam: some paleoecological considerations. 
Malacologia. 19: 329-334. 

VERMEIJ, G. J., 1982, Unsuccessful prédation and 
evolution. The American Naturalist, 120: 701- 
720. 

VERMEIJ, G. J., 1983a, Traces and trends of pré- 
dation, with special reference to bivalved ani- 
mals. Palaeontology. 26: 455-465. 

VERMEIJ, G. J., 1983b, Intimate associations and 
coevolution in the sea. Pp. 311-327, in Coevo- 
lution. (Eds. D. J. Futuyma & M. Slatkin). Sun- 
derland. MA: Sinauer Associates. 



188 



KABAT 



VERMEIJ, G. J.. 1983c, Shell-breaking prédation 
through time. Pp. 649-669. in, Biotic interactions 
in recent and fossil benthic communities. (Eds. 
M. J. S. Tevesz & P. L. McCall). New York: Ple- 
num Press. 

VERMEIJ, G. J., 1987, Evolution and escalation: an 
ecological history of life. Princeton: Princeton 
University Press, xv + 527 pp. 

VERMEIJ, G. J. & E. С DUDLEY, 1982, Shell re- 
pair and drilling in some gastropods from the 
Ripley Formation (Upper Cretaceous) of the 
Southeastern U.S.A. Cretaceous Research. 3: 
397-403. 

VERMEIJ, G. J. & J. A. VEIL, 1978, A latitudinal 
pattern in bivalve shell gaping. Malacologia. 17: 
57-61 . 

VERMEIJ. G. J., E. С DUDLEY & E. ZIPSER, 
1989. Successful and unsuccessful drilling pré- 
dation in Recent pelecypods. Ttie Veliger. 32: 
266-273. 

VERMEIJ. G. J., E. ZIPSER & E. C. DUDLEY, 
1980, Prédation in time and space: peeling and 
drilling in terebnd gastropods. Paleobiology. 6: 
352-364. 

VIGNALI, R. & L. GALLENI, 1986, Naticid prédation 
on soft bottom bivalves: a study on a beach shell 
assemblage. Oebalia. 13: 157-177. 

VIGNALI. R. & L. GALLENI, 1987 [■■19861, The 
beach shell assemblage of Pontedoro: a prelim- 
inary list of mollusc species with some notes on 
prédation phenomena. Atti della Società Toscana 
di Scienze Naturali. Метопе. Sene В. 93: 241- 
250. 

WARME, J. E. & E. J. McHURON, 1978, Manne 
borers: trace fossils and geological significance. 
Pp. 77-1 31 , in. Trace Fossil Concepts. Society of 
Economie Paleontologists and t^ineralogists. 
Short Course. 5. (Ed. P. B. Basan). 

WATKINS, R., 1974, Molluscan paleobiology of the 
Miocene Wimer Formation, Del Norte County, 
California. Journal of Paleontology. 48: 1264- 
1282. 

WILSON, J. G., 1988, Resource partitioning and 
prédation as a limit to size in Nucula túrgida (Leck- 
enby & Marshall). Functional Ecology. 2: 63-66. 

WILTSE. W. I., 1980a, Prédation by juvenile Polin- 
ices duplicatus (Say) on Gemma gemma (Tot- 
ten). Journal of Experimental fiarme Biology and 
Ecology. 42: 187-199. 

WILTSE, W. I. 1980b, Effects of Polinices duplica- 
tus (Gastropoda: Naticidae) on infaunal commu- 
nity structure at Barnstable Harbor, Massachu- 
setts, USA. Marine Biology. 56: 301-310. 

WITMAN, J. D. & R. A. COOPER, 1983, Distur- 
bance and contrasting patterns of population 
structure in the brachiopod Terebratulina septen- 
trionalis (Couthouy) from two subtidal habitats. 
Journal of Experimental Marine Biology and 
Ecology. 73: 57-79. 
WOELKE, С е., 1957, The flatworm Pseudostylo- 
chus ostreophagus Hyman, a predator of oys- 
ters. Proceedings of the National Shellfisheries 
Association. 47: 62-67. 



YOCHELSON, E. L., D. DOCKERY & H. WOLF, 
1983, Prédation on sub-Holocene scaphopod 
mollusks from Southern Louisiana. U. S. Geolog- 
ical Survey. Professional Paper. 1282: 1-13, pis. 
1-4. 

YOUNG, D. K., 1969, Okadaia elegans. a tube- 
boring nudibranch mollusc from the central and 
west Pacific. American Zoologist. 9: 903-907. 

ZIEGELMEIER, E., 1954, Beobachtungen über den 
Nahrungserwerb bei der Naticide Lunatia nitida 
Donovan (Gastropoda Prosobranchia). Helgo- 
länder Wissenschaftliche Meeresuntersuchun- 
gen. 5: 1-33. 

Revised Ms. accepted 24 April 1990 



APPENDIX 

An * indicates that no species was given; 
"spp." indicates that more than two species of 
that genus were reported on in one reference. 
I have not included the taxa reported on by 
Arua (1989) or Arua & Hoque (1989a-c) due 
to the questionable nature of their borehole 
determinations. 



A. Class Gastropoda. Subclass Prosobranchia. 

Order Archaeogastropoda. Trochoidea. Trochidae: 

Calliostoma laugiein [Vignali & Galleni, 1987] 

Gibbula vana [Vignali & Galleni, 1987] 

Helicocryptus radiatus [Taylor et al., 1983] 

Jujubinus exasperatus [Vignali & Galleni, 1987] 

Margantes monolifera [Taylor et al., 1983] 

Monilea' [Kohn, unpub.] 

Umbonium vestianum [Berry, 1982] 

Cyclostrematidae: 

Pseudoliotina' [Taylor et al., 1983] 

Turbinidae: 

Turbo' [Kohn, unpub] 

Neritoidea. Neritidae: 

Nenta funiculata [Hughes, 1985] 

N. scabricosta [Hughes, 1985] 

Neritina virgínea [Jackson, 1972] 

Theodoxus luteofasciatus [Stump, 1975] 

Order Mesogastropoda. Littorinoidea. Littorinidae: 
Littonna littorea [Edwards, 1975] 

Rissoidea. Hydrobiidae: 

Hydrobia andrussowi [Kojumdjieva, 1974] 

Rissoidae: 

Alvania alexandrae [Hoffman et al., 1974] 

Ihungia ponden [Maxwell, 1988) 

Mohrensternia angulata [Kojumdjieva, 1974] 

M. inflata [Kojumdjieva, 1974] 

Rissoa inconspicua [Fretter & Manly, 1979] 

Rissoina podolica [Hoffman et al., 1974] 



NATICID PREDATION 



189 



Caecidae: 

Caecum glabrum [Hoffman et al. 

Vitrinellidae: 

Circulus* [Hoffman et al., 1974] 



1974] 



Cerithioidea. Cerithiidae: 
Argyropeza' [Kohn, unpub.] 
Bittium' [Berg, 1976; Taylor, 1970] 
a reticulatum [Hoffman et al., 1974] 
Cerithium europeum [Kojumdjieva, 1974] 
C. vahabile [Jackson, 1972] 

C. vulgatum [Vignali & Galleni, 1987] 
Rhinoclavis* [Kohn, unpub.] 
Procerithiidae: 

Cirsocerithium gracile [Taylor et al., 1983] 

Diastomatidae: 

Sandbergeria perpusilla [Hoffman et al., 1974] 

Fossariidae: 

"Fossarus" granosus [Taylor et al., 1983] 

Turritellidae; 

Archimediella spirata [Robba & Ostinelli, 1975] 

Mesalia spp. [Fischer, 1962] 

M. amekiensis [Adegoke & Tevesz, 1974] 

M. regularis [Taylor, 1 970] 

Turritella spp. [Dudley & Vermeij, 1978] 

T. badensis [Kojumdjieva, 1974] 

T. b/en/asz/ [Kojumdjieva, 1974] 

T. granulata [Taylor et al., 1983] 

7". subangulata [Kojumdjieva, 1974] 

T. tricarinata [Hoffman & Martinell, 1984] 

Stromboidea. Aporrhaidae: 
Aporrhais pespelecani [MartineW & Marquina, 1980] 
A. uttingerianus [Martinell & Marquina, 1980] 
Drepanocheilus calcarata [Taylor et al., 1983] 

D. neglecta [Taylor et al., 1983] 
Strombidae: 

Rimella fissurella [Taylor, 1 970] 
Strombus* [Kohn, unpub.] 
Tibia unidigitata [Adegoke & Tevesz, 1974] 
Hipponicoidea. Hipponicidae: 
HIpponix' [Kohn, unpub.] 
Vanikoriidae: 
'Vanil<oropsis" CÍ. albus [Taylor et al., 1983] 

Tonnoidea. Cassidae: 

Semicassis wannoensis [Hingston, 1985] 

Cymatiidae: 

Cymatium' [Kohn, unpub.] 

Suborder Heteroglossa. Cerithiopsioidea. Cerithi- 

opsidae: 
Cerithiopsis tubercularis [Hoffman et al., 1974] 

Triphoroidea. Triphohdae: 

Triphora perversa [Hoffman et al., 1974] 

Epitonioidea. Epitoniidae: 
Confusiscala fittoni [Taylor et al., 1983] 
Epitonium spinosa [Hoffman et al., 1974] 

Eulimoidea. Eulimidae; 

Eulima subulata [Hoffman et al., 1974] 



Strombiformis glaber [V\gr\a\'\ & Galleni, 1987] 

Rissoelloidea. Rissoellidae: 
Rissoella* [Berg, 1976] 

Order Neogastropoda. Muricoidea. Muhcidae; 

Blackdownea quadrata [Taylor et al., 1983] 

Eupleura caudata [Flower, 1954] 

Hadrlania craticulata [Martinell & Marquina, 1980] 

Hexaplex benedeica [Adegoke & Tevesz, 1974] 

Morula' [Kohn, unpub.] 

Nassa restitutiana [Kojumdjieva, 1974] 

N. dujardini [Hoffman et al., 1974] 

Paramorea lineata [Taylor et a!., 1983] 

Pterynotus' [Adegoke & Tevesz, 1974] 

Terefundus lamelliferus [Maxwell, 1988] 

L/rosa/p/nx [Flower, 1954] 

Buccinidae: 

Cantharus' [Kohn, unpub.] 

Phos* [Kohn, unpub.] 

Siphonalia* [Kohn, unpub.] 

Columbellidae: 

Mitrella' [Adegoke & Tevesz, 1974] 

M. m/nor [Hoffman & Martinell, 1984] 

M. nassoides [Kojumdjieva, 1974] 

Nassariidae: 

Amyclina spp. [Robba & Ostinelli, 1975] 

Cyllene' [Adegoke & Tevesz, 1974] 

Dorsanum duplicatum [Kojumdjieva, 1974] 

Nassarius elatus [Hoffman & Martinell, 1984] 

N. italicus [Martinell & Marquina, 1980] 

N. obsoletus [Edwards, 1975] 

N. perpinguis [Berg & Nishenko, 1975] 

N. pygmaeus [Hoffman & Martinell, 1984] 

N. semistriatus [Hoffman & Martinell, 1984] 

N. tiarula [Stump, 1 795] 

N. trivittatus [Edwards, 1975] 

Niotha crassigranosa [Hingston, 1985] 

Plicarcularia leptospira [Broom, 1983] 

Fasciolariidae: 

Colubraria' [Kohn, unpub.] 

Falslcolus tangituensis [Maxwell, 1988] 

Fusinus* [Kohn, unpub.] 

Granulifusus' [Kohn, unpub.] 

Iscafusus rigidus [Taylor et al., 1983] 

Latirus moorei [Stanton et al., 1981] 

Peristernia' [Kohn, unpub.] 

Turbinellidae [= Vasidae]: 

Exilia tve//man/ [Maxwell, 1988] 

Olividae: 

Alocospira papillata [Hingston, 1985] 

Ancilla buccinoides [Taylor, 1 970] 

Olivella biplicata [Edwards, 1969] 

Marginellidae: 

Marginella spp. [Taylor, 1970] 

Protoglnella bembix [Maxwell, 1988] 

Mitridae: 

Cancilla' [Kohn, unpub.] 

Mitra Orientalis [Kojumdjieva, 1974] 

Scabricola' [Kohn, unpub.] 

Subcancilla* [Kohn, unpub.] 

Volutomitridae: 

Microvoluta nodulata [Maxwell, 1988] 



190 



KABAT 



Costellariidae [ = Vexlllidae]: 
Austromitra' [Hingston, 1985] 
Vexillium' [Kohn. unpub.] 

Cancellarioidea. Cancellanidae: 

Bonellitia amekiensis [Adegoke & Tevesz. 1974] 

B. serrata [Martinen & Marquina, 1980] 

Inglisella parva [Maxwell. 1988] 

/. allophyla [Maxwell, 1988] 

Sydaphera wannonensis [Hingston, 1985] 

Conoidea. Conldae; 

Conus dujardini [Kojumdjieva, 1974] 

С parisiensis [Taylor, 1970] 

Turridae: 

Bela brachystoma [Hoffman & Martinell. 1984] 

B. vulpécula [Hoffman & Martinell. 1984] 

Brachyioma obtusangula [Martinell & Marquina, 

1980] 
Clavatula' [Adegoke & Tevesz, 1974] 
Clavus spp. [Robba & Ostinelli, 1975] 
Comitas nana [Maxwell, 1988] 
Crassispira' [Kohn, unpub.) 
Cythara subcylindrata [Hoffman et al., 1974] 
Eopleurotoma spp. [Adegoke & Tevesz, 1974] 
Gemmula' [Kohn, unpub.] 
Genota ramosa [Kojumdjieva, 1974] 
Hesperiturns nodocannatus [Stanton et al., 1981] 
Heterocithara marwicki [Maxwell. 1988] 
Lophitoma' [Kohn, unpub.] 
Mauidnllia occidentalis [Maxwell, 1988] 
Michela trabeatoides [Stanton et al., 1981] 
Mioawatena personata [Maxwell, 1988] 
Paracomitas beui [Maxwell, 1988] 
Pleurotoma' [Adegoke & Tevesz, 1974] 
Raphitoma hispidula [Hoffman et al., 1974] 
Rugobela' [Maxwell, 1988] 
Splendnllia i/e/Za/ [Maxwell, 1988] 
Tomopleura' [Maxwell. 1988] 
Turricula africana [Adegoke & Tevesz, 1974] 
T. dimidiata [Martinell & Marquina, 1980] 
Viridoturns powelli [Maxwell, 1988] 
Terebridae: 

Gemmaterebra catenifera [Hingston, 1985] 
Strioterebrunn monidum [Kitchell et al., 1986] 
S. pliocenicunn [Martinen & Marquina, 1980] 
Terebra spp. [Vermeij et al., 1980] 
T. dislócala [Kitchell et al., 1986] 
Zeacuminia viapollentia [Maxwell, 1988] 

Subclass Heterobranchia. Superorder Allogas- 

tropoda. 
Architectonicoidea. Architectonicidae: 
Architectonica bendeica [Adegoke & Tevesz, 1974] 
A. olicatum [Taylor, 1970] 
Philippia meditteranea [Vignali & Galleni, 1987] 
Pyramidelloidea. Pyramidellidae; 
Eulimella conulus [Hoffman et al., 1974] 
Evelynella doliella [Maxwell, 1988] 
Odostomia' [Adegoke & Tevesz, 1974] 
G. conoidea [Hoffman & Martinell, 1984] 
Pyramidella digitalis [Hoffman et al., 1974] 
P. plicosa [Hoffman & Martinell, 1984] 



Pyrgulina interstincta [Hoffman et al., 1974] 
Tubonilla rufa [Hoffman & Martinell, 1984] 
T. zesulcata [Maxwell, 1988] 
Waikura elevata [Maxwell, 1988] 

Subclass Opisthobranchia. Order Cephalaspidea, 

Philinoidea. Acteonidae: 

Acteon réussi [Hoffman et al., 1974] 

A. semistnatus [Hoffman & Martinell, 1984] 

A. tornatilis [Vignali & Galleni, 1987] 
Tornatellaea affinis [Taylor et al., 1983] 
T. unlsulcata [Taylor et al., 1983] 
Ringiculidae: 

Avellana incrassata [Taylor, et al., 1983] 

Ringicula auriculata [Hoffman et al., 1974] 

R. buccinea [Hoffman & Martinell, 1984] 

Scaphandridae: 

Acteocina lajonkaireana [Kojumdjieva, 1974] 

Cylichina melitopolitana [Kojumdjieva, 1974] 

С rubignosum [Kojumdjieva, 1974] 

Scaphiander' [Adegoke & Tevesz, 1 974] 

Tornatina heraclitica [Hoffman et al., 1974] 

T. trunculata [Hoffman et al., 1974] 

Hamineidae: 

Atys miliaris [Hoffman et al., 1974] 

Retusidae: 

Retusa kelloggi [Stanton et al., 1981] 

R. truncatula [Hoffman & Martinell, 1984]. 

B. Class Bivalvia. Subclass Protobranchia. Order 
Nucuioida. 

Nuculoidea. Nuculidae: 

Acila conradi [Colbath, 1985] 

Ennucula kalimnae [Hingston, 1985] 

Nucula antiquata [Taylor et al., 1983] 

N. mixta [Taylor, 1970] 

N. nucleus [IHoffman & Szubzda, 1976] 

N. obtusa [Taylor et al., 1983] 

N. túrgida [Wilson, 1988] 

Palaeonucula strigilata [Fürsicn & Jablonski, 1984] 

Nuculanoidea. Nuculanidae: 

f^esosaccella angulata [Taylor et al., 1983] 

/W. Iineata [Taylor et al., 1983] 

Nuculana' [Adegoke & Tevesz, 1974] 

Nuculana spp. [Colbath, 1985] 

N. fragilis [Kojumdjieva, 1974] 

N. pella [Vignali & Galleni, 1987] 

N. pernula [Hutchings & Haedrich, 1984] 

Yoldiidae: 

Yoldia' [Colbath, 1985] 

V. tfiraciaeformis [Hutchings & Haedrich, 1984] 

Malletiidae: 

l^alletia* [Kohn, unpub.] 

Subclass Pteriomorphia. Order Mytiioida. 

Mytiloidea. Mytilidae: 

Choromytilus meriodionalis [Griffiths, 1981] 

Crenella orbicularis [Taylor et al., 1983] 

¡Modiolus auriculatus [Vermeij, 1980] 

/W. reversa [Taylor et al., 1983] 

Mytilus edulis [Edwards, 1 975] 



NATICID PREDATION 



191 



Order Arcoida. Arcoidea. Arcidae: 

Anadara spp. [Kelley, 1989a] 

A. elevata [Dudley & Dudley, 1980] 

A. granosa [Broom, 1982] 

Adevincta [Colbath, 1985] 

A. c/z/uw/ [Kojurлdjieva, 1974] 

A. thisphila [Dudley & Dudley, 1980] 

Barbatia irregularis [Taylor, 1 970] 

Bathyarca' [Maxwell, 1988] 

Noetiidae: 

Arcopsis dissimilis [Darragh & Kendrick, 1980] 

Pachecoa declivis [Kitchell, 1982] 

Cucullaeidae; 

Idonearca glabra [Taylor et al., 1983] 

Limopsoidea. Limopsidae: 

Limopsis chapmani [Darragh & Kendrick, 1980] 

L. beaumarisensis [Hingston, 1985] 

L. minuta [Kojurndjieva, 1974]. 

Giycymerididae: ^■ 

Glycymeris spp. [Thomas, 1976] 

G. albolineata [Matsukuma, 1977] 

G. rta/// [Hingston, 1985] 

G. insubrica [Vignali & Galleni, 1987] 

G. pulvinata [Taylor, 1 970] 

G. vestita [Matsukuma, 1977] 

Glycymerita sublaevis [Taylor et al., 1983] 

G. umbonata [Taylor et al., 1983] 

Ptehoida. Pterioidea. Cassianellidae; 
Cassianella ampezzana [Fürsich & Jablonski, 
1984] 

Order Limoida. Limoidea. Limidae; 
Mysidioptera williamsi [Newton, 1983] 

Order Ostreoida. Ostreoidea. Gryphaeidae: 
Amphidonte obliquata [Taylor et a!., 1983]. 

Pectinoidea. Pectinidae; 
Chlamys radians [Guerrero & Reyment, 1988] 
Pectin opercularis [Boekschoten, 1967] 
Pseudamussium similis [Smith, 1932]. 

Subclass Paleoheterodonta. Order Trigonioida. 

Trigonioidea. Trigoniidae: 

Rutitrigonia eccentrica [Taylor et a!., 1983] 

Subclass Heterodonta. Order Veneroida. 

Lucinoidea. Lucinidae: 

Codakia bella [Vermel], 1980] 

C. orbicularis [Jackson, 1972] 

Ctena decussata [Vignali & Galleni, 1987] 

C. orbiculata [Jackson, 1972] 
Divaricella ornata [Kojumdjieva, 1974] 

D. divaricata [Vignali & Galleni, 1987] 
Epicodakia' [Kohn, unpub.] 

Loripes dentatus [Hoffman et al., 1974] 

L. lacteus [Vignali & Galleni, 1987] 

Lucina anodonta [Kelley, 1989a] 

L. approximata [Stump, 1975] 

L. spinifera [Kojumdjieva, 1974] 

Lucinella divaricata [Hoffman & Martinell, 1984] 



Myrtea papatikiensis [Maxwell, 1988] 

Parvilucina costata [Jackson, 1972] 

Pseudomiltha floridana [Kitchell et al., 1981] 

Wallucina* [Vermeij, 1980] 

Fimbriidae: 

Mutiella canaliculata [Taylor et al., 1983] 

Ungulinidae: 

Diplodonta subquadrata [Vermeij et al., 1989] 

Carditoidea. Carditidae: 
Beguina diversicosta [Kojumdjieva, 1974] 
Cardita spp. [Adegoke & Tevesz, 1974] 
С chamaeformis [Boekschoeten, 1967] 
Cyclocardia subtenta [Colbath, 1985] 
Venericardia greggiana [Kitchell, 1982] 
V. serrulata [Taylor, 1 970] 
Vetericardiella* [Kitchell, 1986] 

Crassatelloidea. Astartidae: 

Astarte spp. [Boekschoten, 1967; Kelley, 1989a] 

Astarte triangularis [Smith, 1932] 

Eriphyla striata [Taylor et al., 1983] 

Lirodiscus tellinoides [Siler, 1965] 

Nicaniella formosa [Taylor et al., 1983] 

Crassatellidae: 

Crassatella spp. [Taylor, 1970] 

C. v/adosa [Sohl, 1969] 

Crassatellites* [Kohn, unpub.] 

Eucrassatella spp. [Kelley, 1982a] 

Cardioidea. Cardiidae: 

Acanthocardia tuberculata [Vignali & Galleni, 1986] 

Cardium spp. [Smith, 1932] 

С politionanei [Kojumdjieva, 1974] 

Cerastoderma edule [Bayliss, 1986] 

Clinocardium nuttallii [Bernard, 1967] 

Dinocardium robustum [Kornicker et al., 1963] 

Fragum fragum [Vermeij, 1980] 

Laevicardium aléñense [Vermeij et al., 1989] 

Loxocardium bouel [Taylor, 1 970] 

Parvicardium scabrum [Bayliss, 1986] 

Protocardia hillana [Taylor et al., 1983] 

Thetis laevigata [Taylor et al., 1983] 

Mactroidea. Mactridae: 

Mactra angulata [Taylor et al., 1983] 

M. australis [Laws & Laws, 1972] 

M. chiinensis [Vermeij et al., 1989] 

M. fragilis [Paine, 1963] 

M. stultorum [Vignali & Galleni, 1987] 

Mactrellona exoleta [Vermeij et al., 1989] 

Notospisula parva [Green, 1968] 

Pseudocardium sachalinense [Vermeij et al., 1989] 

Spisula elliptica [Bayliss, 1986] 

S. solidissima [Franz, 1977] 

S. subtruncata [Bayliss, 1986] 

Tresus nuttallii [Reid & Friesen, 1980] 

Mesodesmatidae: 

Atactodea striata [Ansell & Morton, 1987] 

Coecella chinensis [Ansell & Morton, 1987] 

Donacilla angusta [Laws & Laws, 1972] 

Ervilia ousilla [Hoffman & Szubzda, 1976] 

E. dissita [Kojumdjieva, 1974] 



192 



KABAT 



Solenoidea. Solenidae: 
Ensis directus [Schneider, 1982] 
Solen conradi [Colbath. 1985] 
S. strictus [Frey et al., 1987]. 

Tellinoidea. Donacidae: 

Donax spp. [Vermeij et al.. 1989] 

D. faba [Ansell & Morton, 1987] 

D. semistnata [Vignali & Galleni, 1987] 

D. trunculus [Vignali & Galleni, 1987] 

D. vittatus [Negus, 1975] 

Plebidonax deltoides [Kitching & Pearson, 1981] 

Psammobiidae: 

Gari hamiltonensis [Hingston, 1985] 

Tagelus peruvianas [Vermeij et al., 1989] 

Scrobicularildae; 

Scrobiculana plana [Richter, 1962] 

Solecurtidae: 

Solecurtus antiquatus [Kojumdjieva, 1974] 

Tellinidae: 

Arcopagia robusta [Vermeij, 1980] 

Macoma albana [Colbath, 1985] 

M. arctata [Colbath, 1985] 

M. balthica [Commito, 1982] 

M. calcárea [Aiken & Risk, 1988] 

M. nasuta [Reid & Gustafson, 1989] 

Palaeomoera inaequalis [Taylor et al., 1983] 

Peronidia venulosa [Vermeij et al., 1989] 

Quidnipagus palatam [Vermeij, 1980] 

Scissulina' [Vermeij, 1980] 

Tellina spp. [Vermeij et al., 1989] 

T. donacina [Vignali & Galleni, 1987] 

T. emacerata [Colbath, 1985] 

T. lux [Broom, 1983] 

T. planata [Kojumdjieva, 1974] 

T. púdica [Broom, 1983] 

7. pulchella [Vignali & Galleni, 1987] 

T. tenuis [Ansell, 1982a-c] 

Tellinella virgata [Nakamine & Habe, 1983] 

Temnoconcha cognata [Vermeij et a!., 1989] 



Circomphalus subplicatus [Hoffman & Szubzda, 

1976] 
Costacallista laevigata [Taylor, 1970] 
Dosinia dunken [Vermeij et al., 1989] 
D. lupinas [Vignali & Galleni, 1987] 
Flaventia ovalis [Taylor et al., 1983] 
Gafranum minimum [Smith, 1932] 
G. pectinatum [Vermeij, 1980] 
Gemma gemma [Wiltse, 1980a] 
Gouldia minima [Vignali & Galleni, 1987] 
Katelysia scalanna [Laws & Laws, 1972] 
Kathennella angustifrons [Colbath, 1985] 
Macrocallista nimbosa [Paine, 1963] 
Megapitaria squalida [Vermeij et al., 1989] 
Mercenaria mercenaria [Berg & Porter, 1 974] 
M. campechiensis [Paine, 1963] 
Meretnx lusoria [Vermeij et al., 1989] 
Paraesa faba [Taylor et al., 1983] 
Pelecyora trígona [Broom, 1983] 
Periglypta reticulate [Vermeij, 1980] 
Pitar spp. [Vermeij et al., 1989] 
P. morrhuana [Jacobson, 1965] 
Placamen subroboratum [Hingston, 1985] 
Protothaca spp. [Vermeij et al., 1989] 
P. staminés [Peterson, 1982] 
Ruditapes philippinarum [Rodrigues, 1986] 
Saxidomus giganteus [Bernard, 1967] 
Sunetta gibberula [Hingston, 1985] 
Tapes japónica [\-\amada. 1961] 
T. philippinarum [Ansell & Morton, 1987] 
Timoclea manca [Vermeij, 1980] 
Tivela spp. [Vermeij et al., 1989] 
Venerupis aurea [Vignali & Galleni, 1987] 
V. senegalensis [Vignali & Galleni, 1987] 
Venus multilamella [Kojumdjieva, 1974] 
V. striatula [Ansell, 1960] 
V. verrucosa [Vignali & Galleni, 1987] 
Veremolpa miera [Mukai, 1973] 
Glauconomidae; 
Glauconome chinensis [Ansell & Morton, 1987] 



Arcticoidea. Arcticidae: 
Árctica islándica [Christensen, 1970] 
Epicypnna angulata [Taylor et al., 1983] 
E. subtruncata [Taylor et al., 1983] 
Venilicardia lineolata [Taylor, et al., 1983] 

Veneroidea. Veneridae: 

Anomalocardia squamosa [Ansell & Morton, 1987] 

A. squamosa [Taylor, 1980] 

Aphrodina nitidula [Taylor, 1 970] 

Bassina calophylla [Ansell & Morton, 1985] 

Callistina plana [Taylor et al., 1983] 

Calpitaria distincta [Taylor, 1 970] 

Calva subrotunda [Taylor et al., 1983] 

Chamelea gallina [Guerrero & Reyment, 1988a] 

Chímela caperata [Taylor et al., 1983] 

Chione spp. [Smith, 1932] 

C. bastero^/ [Kojumdjieva, 1974] 

С californensis [Stump, 1975] 

С. cancellata [Paine, 1963] 

C. subrugosa [Vermeij et al., 1989] 

C. undatella [Peterson, 1982] 



Order Myoïda. Myoïdea. Myidae: 

Cryptomya californica [Watkins, 1974] 

Mya arenana [Edwards, 1975] 

Corbulidae: 

Caestocorbula' [Kitchell, 1986] 

Caryocorbula deusseni [Kitchell, 1982] 

Corbula spp. [De Cauwer, 1985] 

Corbula carinata [Kojumdjieva, 1974] 

C. elegans [Taylor et al., 1983] 

C. gibba [Vignali & Galleni, 1987] 

C. idónea [Kelley, 1989a] 

C. rugosa [Taylor, 1 970] 

С. trúncala [Taylor et al., 1983] 

Notocorbula ephamilla [Hingston, 1985] 

N. innerans [Maxwell, 1988] 

Varicorbula amekiensis [Adegoke & Tevesz, 1974] 

Vokesula aidrichi [Kitchell, 1982] 

Hiatelloidea. Hiatellidae: 

Hiatella árctica [Aitken & Risk, 1 988] 

Panopea mandíbula [Taylor et al., 1983] 



NATICID PREDATION 



193 



Subclass Anomalodesmata. Pandoroidea. Periplo- 

matidae: 
Cochlodesma leanum [Rosewater, 1 980] 
Pehploma spp. [Rosewater, 1980] 

Poromyoidea. Cuspidahidae: 

Cuspidaria cuspidata [Hoffman & Martinell, 1984] 

С Scaphopoda. 



Dentaiiidae: 

Dentalium complexum [Fankboner, 1 969] 

D. bedensis [Kojumdjieva, 1974] 

D. spp. [Yochelson et al., 1983] 

Fustiaria miocaenica [Hoffman et a!., 1974] 

Entalinidae: 

Entaliopis brevis [Yochelson et al., 1983] 

Gadilidae: 

Cadulus* [Yochelson et al., 1983] 



MALACOLOGIA, 1990, 32(1): 195-202 



LETTERS TO THE EDITOR 



TOWARDS A PHYLOGENETIC SYSTEM OF GASTROPODA 
PART I: TRADITIONAL METHODOLOGY— A REPLY 

Gerhard Haszprunar 

Institut für Zoologie der Universität Innsbruck 
Technikerstrasse 25, A-6020 Innsbruck. Austria 



ABSTRACT 

<*■ 

Bieler (1990) provides a critique of the methodology of a phylogenetic analysis of the Gas- 
tropoda by Haszprunar (1988). His criticism of an incomplete and inconsistent presentation of 
character-states and methodology is answered by explaining by examples the way in which the 
character analysis and the construction of the cladogram were done. I argue that any maximum 
parsimony analysis with equal weighting of characters will fail to produce the "true" phylogeny 
because of the high degree of parallelism and convergence within the group. The method 
presented applies a priori criteria for estimating the probabilities of homology and apomorphy 
(i.e. significance) of characters and constructs the cladogram according to that significance. In 
the proposed classification, higher taxa are thought to reflect stem-lines of high probability. 

Key words: Gastropoda, systematics, classification, phylogeny, cladistics, critique. 



INTRODUCTION 

Bieler (1990) gives a critique of the meth- 
odology of the recently published phyloge- 
netic analysis of streptoneuran Gastropoda 
by Haszprunar (1988) from a strictly cladistic 
point of view. Here I want, first, to correct cer- 
tain points of the original paper (Haszprunar, 
1988; cf. appendix); second, to explain bhefly 
the reasons that the analysis was not done by 
application of accepted cladistic methodol- 
ogy; and third, to provide significant examples 
of the way in which I weighted the characters 
and did the analysis. Doing the latter, I accept 
the major points of Bieler's (1990) critique — 
no one is perfect. 

Thus, I agree with Bieler (1990) that for any 
"scientific question, it is an integral part of any 
study to present the data unambiguously, to 
employ reproducible methods, and to offer 
testable hypotheses." Maybe I have underes- 
timated the difficulty of following my argu- 
ments. I therefore wish at least to show the 
principles. 

PRESENTATION OF DATA 

Bieler (1990) is correct in assuming that my 
analysis was not done by computer, because 



during the original study adequate hardware 
to run phylogenetic software was unavail- 
able. Since then, adequate hardware has be- 
come available, and I have become familiar 
with the advantages and disadvantages of 
programs like PHYLIP, PAUP and in particu- 
lar HENNIG'86. 

Admittedly there are some mistakes in the 
text, tables and figures, all of which are of 
minor importance, however. Nevertheless, I 
welcome this opportunity to correct those of 
which I am aware. 

Bieler (1990) criticized the fact that I did not 
provide a comprehensive data-matrix. The 
way in which I did the analysis, however, does 
not require a data-matrix (see below), and the 
main results of the character-analysis have 
been presented (Haszprunar, 1988: table 2). 



THEORETICAL CONSIDERATIONS 

Character analysis is the basis of any phy- 
logenetic analysis. Typically plesiomorphic 
versus apomorphic states are estimated by 
application of the rules of Hennig (1966) such 
as outgroup-comparison, data on fossils, on- 
togenetic sequences, and the like. Often, 
however, there is no clear outgroup available 



195 



196 



HASZPRUNAR 



(e.g. Houbrick, 1988: Reid, 1989), and the 
use of fossils and ontogenetic data has been 
criticized (Alberch, 1985). 

The problem of homology, i.e. the problem 
of the frequency of change from plesiomor- 
phic to apomorphic character state during 
phylogeny, seems to be overcome by appli- 
cation of a "maximum-parsimony" analysis, 
whether by hand or by computer. The working 
hypothesis of parsimony minimizes the num- 
ber of analogies (homoplasies) and can pro- 
duce one (or many) "most parsimonious" 
tree(s). Colless (1983) has pointed out that 
the principle of parsimony, which is an oper- 
ational concept rather than an empincal fact 
of evolution, does work with negligible rates of 
failure only if the probability of change in each 
character-state is very low. As outlined by 
Gosliner & Ghiselin (1984) for primitive 
opisthobranchs and identified by Davis (1989) 
in the Hydrobiidae and in my work on the 
streptoneurans, however, there is a very large 
degree of convergence, that is, parallelism, in 
the data. With an increase in the number of 
taxa the degree of homoplasy increases 
(Sanderson & Donoghue, 1989), moreover, 
suggesting a more or less constant rate of 
homoplasy among taxa. Indeed, the neces- 
sity of a parsimony analysis implies that the 
basic data matrix is controversial with respect 
to its proposed synapomorphies. Accordingly, 
the problem of homology cannot be overcome 
by parsimony analysis. 

The recent cladistic study (done with 
PAUP) of the Littonnidae by Reid (1989) also 
shows many cases of homoplasy. Indeed, 
"59.2% of the character state changes could 
be ascribed to homoplasy" (Reid, 1989: 59), 
and this is logically a minimal ("most parsimo- 
nious") estimate. Significantly, a final cla- 
dogram that differs in certain points from the 
consenus tree is preferred because "some 
character-state reconstructions are more 
likely than others" (Reid, 1 989: 63). As stated 
(Haszprunar 1988: 399), the mam problem of 
any phylogenetic study is that of homology. 

As reviewed by Riedl (1975, 1978), Ruppert 
(1982) and Neff (1986), a рл/ол/ entena for in- 
ference of homology have been provided by 
Remane (1952, 1954). More recently, Rieger 
& Tyler (1979, 1985; see also Westheide & 
Rieger, 1987: Tyler, 1988) have formulated 
criteria for the counter-version, i.e. the esti- 
mation of convergence. Both sets of criteria 
should be applied to any analysis (see below). 

I want to stress that both types of character 
analysis (homology versus analogy, apomor- 



phy versus plesiomorphy) must be done prior 
to construction of the tree, and that both are 
principially inductive by application of the cri- 
teria of Remane (1952) and Hennig (1966) 
among others. Accordingly, each proposed 
synapomorphy includes a two-fold degree of 
probability, one with respect to apomorphy 
("-apo- "), and one with respect to homology 
("syn-") (Haszprunar, 1989). In the case of 
character-states, even two analyses of ho- 
mology are necessary: first, whether all the 
states belong to the same (homologous) 
character; secondly, with respect to the ho- 
mology of the advanced state. For example, 
in an analysis of the gills of gastropod groups 
and in particular the (plicate) gill of primitive 
opisthobranchs, two questions of homology 
must be answered (for detailed discussion, 
see Haszprunar, 1985: 20-21; Haszprunar, 
1988: 382): (1) Is the opisthobranch gill a ho- 
mologue of the prosobranch gill (i.e. a ctenid- 
ium)? (2) If so, is the plicatid gill homologous 
in all opisthobranchs? 

I believe that these probabilities must be 
used to "weight" the characters used in the 
analysis. In other words, the "weight" is not a 
feature of the character itself, but the degree 
of likelihood in the present analysis (cf. also 
Bryant, 1989). 

There is no escape from the weighting of 
characters. Also, the usual analysis involving 
maximal parsimony weights characters by se- 
lecting them (characters not selected lack 
weight) and by giving each selected character 
equal weight. Insofar as the degree of ho- 
moplasy is great, differentiated weighting of 
selected characters becomes essential, how- 
ever. Although Remane (1952) has indicated 
the way to infer distinct probabilities for a pro- 
posed homology, there is still no clear proce- 
dure for quantitative a pr/or/ weighting of char- 
acters (e.g. Neff, 1986; Westheide & Rieger, 
1987; Bryant, 1989). In using computer algo- 
rithms, one possibility would be to include in 
the analysis only characters with high signifi- 
cance; another is to establish a system of dif- 
ferential weighting (e.g. 1/3/5 corresponding 
"low/ medium/ high" significance). 

These considerations shed light on data 
presentation as well. For instance, the fre- 
quency and circumstances of transformation 
of coiled shells into asymmetrical limpet-like 
ones are unimportant; the statement "many" 
shows that the significance of this character is 
very low in this phylogenetic analysis. (Its sig- 
nificance might be high in another one, how- 
ever.) 



GASTROPOD SYSTEMATICS: REPLY TO BIELER 



197 



PRACTICAL CONSEQUENCES 
General Remarks 

In this section I wish to show by examples 
the way in which the character analysis and 
the construction of the cladogram were done 
in the original paper. For character analysis, I 
have selected two examples, the number of 
gills and the conditions of the anterior nerve 
ring, the significance of which differ consider- 
ably. These significances are estimated prior 
to construction of any cladogram by applica- 
tion of the rules of Remane (1952), Rieger & 
Tyler (1979, 1985), Neff (1986), Westheide & 
Rieger (1987) and Tyler (1988). Two groups, 
Nehtimorpha and Pyramidelloidea have been 
selected to demonstrate the construction of 
the cladogram. 

It was assumed a priori that the taxa used 
in the study all were holophyletic (i.e. mono- 
phyletic sensu Hennig, 1966), implying that 
changes within a taxon are secondary phe- 
nomena. This approach also concerned the 
Euthyneura the holophyly of which has been 
shown earlier (Haszprunar, 1985a,b). It will 
be shown that in one case (Allogastropoda) 
this assumption did not work and necessi- 
tated the consideration of the subtaxa (see 
below). 

Number of Ctenidia 

The question of ctenidial homology 
throughout the gastropods has been dis- 
cussed at length by Haszprunar (1985a: 
20-22; 1988: 377-383). Whereas the gills of 
Cocculiniformia, of Valvatoidea, of the allo- 
gastropod groups and the Euthyneura were 
considered to be secondary structures, the 
gills of the remaining streptoneuran groups 
were assumed to represent homologues. 

Outgroup comparison (Cephalopoda, Try- 
blidiida) makes it nearly certain that the pres- 
ence of paired palliai organs is the primitive 
condition among gastropods. This conclusion 
is supported by the facts that even gastro- 
pods with two gills often have reduced the 
right one, and that in the Trochoidea and Lep- 
etodhloidea the blood supply of the right gill is 
retained although the gill itself has been lost. 
The probability for the hypothesis "plesio- 
morphy: two ctenidia; apomorphy: one (left) 
ctenidium" is therefore very high. 

Next the probability was considered 
whether the change from two ctenidia to one 
occurred once or often in gastropod evolution. 



Two functional gills are present only in veti- 
gastropod groups ("zeugobranchs"), and 
even within this group two subgroups have 
lost the right ctenidium. Anatomical features 
in Neritimorpha, in which most species have a 
diotocardian heart and certain species even 
have a gill-rudiment (Fretter, 1 965) and Doco- 
glossa — Patellidae (with two osphradia) like- 
wise suggest an original condition of two 
ctenidia in these taxa. 

Functional morphology shows that a 
change from two to one ctenidium results in 
advantages for the animal with respect to wa- 
ter currents in the mantle cavity (Yonge, 
1947). Indeed, the presence of two ctenidia 
necessitates a slit or hole(s) in the shell for 
passage of waste. Finally, because in zeugo- 
branch gastropods, such as Haliotis, the left 
ctenidium is formed first in ontogeny (Crofts, 
1937), a heterochronic process might easily 
result in a loss of the right ctenidium. 

On the whole, I concluded that the change 
from two to one (left) ctenidium probably oc- 
curred several times in gastropod evolution. 
Thus the probability of the respective syn — 
apomorphy, i.e. the probability of the homol- 
ogy of the change from plesiomorphic to apo- 
morphic condition, is low. 

Anterior Nervous System 

The homology of the main ganglia of the 
anterior nerve ring in gastropods is well es- 
tablished by identical relative positions and 
interconnections and by identical fields of in- 
nervation. 

Among the Streptoneura, two conditions of 
the anterior nerve ring with respect to the rel- 
ative position of the ganglia can be distin- 
guished: the pleural ganglion might be close 
to the pedal one (hypoathroid condition) or 
close to the cerebral one (epiathroid condi- 
tion). Outgroup comparison is unsatisfactory, 
because the Tryblidiida lack pleural ganglia 
and the Cephalopoda have a highly concen- 
trated nervous system. Ingroup comparison 
reveals, however, that the hypoathroid condi- 
tion is generally correlated with other plesio- 
morphic characters, such as presence of na- 
cre, paired palliai and excretory organs, or 
external fertilization. The hypothesis "plesio- 
morphic: hypoathroid condition — apomorphic: 
epiathroid condition" therefore appears well 
founded. 

Again it is now necessary to estimate the 
number of changes from the hypo- to the epi- 
athroid condition. There is not a single strep- 



198 



HASZPRUNAR 



toneuran taxon in which a mixture of the two 
conditions occurs (for Viviparidae cf. Hasz- 
prunar. 1988: 395). In addition, the distribu- 
tion of both character states is largely corre- 
lated with the ability to produce planktotrophic 
larvae (exception: certain Neritoidea). On the 
other hand, a selection pressure that could 
force such a change is unknown. Moreover, 
both conditions are uneffected by concentra- 
tion of the nervous system. Summarizing the 
argument, I assumed the syn — apomorphy 
"epiathroid nerve ring" to be of very high sig- 
nificance for streptoneuran phylogeny. 

Position of Nerltimorpha 

Based on the results of the character anal- 
ysis, estimation of the systematic position of 
the Neritimorpha starts with consideration of 
the characters with the highest significance, 
such as the hypoathroid nervous system. 

This step alone reduces drastically the 
number of possible trees. Starting the analy- 
sis with 28 taxa (Haszprunar, 1 988: fig. 5) 1 .6 
X 10^^ trees are possible. Accepting "epia- 
throid nervous system" as a synapomorphy 
leaves 18 taxa and thus 6.3 x 10^^ possible 
trees [x = (2n-3)!2<" 2)(r,_2)!; ¡n which n 
is the number of taxa]. 

Among those "Archaeo-" gastropods, there 
are two sequences of radular types, each of 
them again with high significance (stereoglos- 
sate — flexoglossate: rhipidoglossate — tae- 
nioglossate; Haszprunar, 1988: 390-391). 
This places the Docoglossa (= Patellogas- 
tropoda) and hot-vent group С below and the 
architaenioglossate groups above the Nenti- 
morpha. The number of possible trees involv- 
ing the Neritimorpha is further reduced to 3.2 
X 10^M"I3 taxa remain). Upon consideration 
of the distribution of ctenidial skeletal support 
(Haszprunar, 1988: 377-381), Neomphalus. 
the Vetigastropoda and the Seguenzioidea 
are placed above the neritimorph clade. The 
number of possible trees is now 15 (4 taxa 
remain). The Cocculiniformia share many ple- 
siomorphic characters with the Docoglossa, 
including the pnmary limpet shell (Haszpru- 
nar, 1988: 370-372); thus they are grouped 
below the Neritimorpha. Finally, Melano- 
drymia shares several characters, such as 
radula type and protoconch features, with 
Neomphalus and the Vetigastropoda, and is 
therefore placed above the neritimorph off- 
shoot. 

This solution agrees with several character 
sequences of high significance. The assump- 



tion that the Neritimorpha belong among the 
higher gastropods is based, however, on 
character stages each of which is correlated 
with reproductive biology, namely internal fer- 
tilization and planktotrophic veligers. The 
probability of convergent evolution of the fea- 
tures of neritimorph reproductive biology is 
very high: first, details of the respective char- 
acters differ considerably between Neritimor- 
pha and higher groups (genital system, sperm 
structure, protoconch features); and second, 
there are numerous examples of internal fer- 
tilization within other archaeogastropod 
clades, and larval planktotrophy has been es- 
tablished through parallel evolution among 
the Bivalvia. 



Position of Pyramidelloidea 

Again, the analysis begins with consider- 
ation of neural conditions. Earlier the Pyra- 
midelloidea were placed together with the 
Architectonicoidea in a clade called Allogas- 
tropoda (Haszprunar, 1985a). The epiathroid 
condition of the anterior nerve ring placed the 
Allogastropoda among the 'Apo-"gastropoda, 
the lack of parietal ganglia and the retention 
of (at least osphradial) streptoneury (Hasz- 
prunar, 1988: 394) suggest a grouping of the 
Allogastropoda below the euthyneuran level 
of organization. 

On the other hand, the Pyramidelloidea and 
the Euthyneura share synapomorphies of 
high significance, such as giant nerve cells, a 
rhinophoral and a lateral nerve and charac- 
ters of the sperm (Haszprunar, 1988: 396- 
397; Healy, 1988a,b). Such proposed syn- 
apomorphies were in direct contrast to the 
originally assumed synapomorphies of the Al- 
logastropoda, namely a shared gill position to 
the right of the dorsal ciliary tract, an acrem- 
bolic proboscis of distinct type (shifted posi- 
tion of buccal ganglia) and spermatophores 
(Haszprunar, 1985a). Meanwhile, however, 
pyramidelloids with a different position of the 
gills (Amathinidae; Ponder, 1987) and a 
mathildid with the usual placement of buccal 
ganglia {Geganyia; Haszprunar, 1985b) were 
described. This leaves the spermatophores 
with very low significance. 

As a conclusion, I corrected my earlier 
opinion and regard the 'Allogastropoda* now 
as a distinct grade rather than a clade. Within 
this grade, the Pyramidelloidea are placed 
closest to the Euthyneura, and both taxa rep- 
resent a sister-group relationship. 



GASTROPOD SYSTEMATICS: REPLY TO BIELER 



199 



CLASSIFICATION 

In my approach, the final cladogram is a the- 
orem of probability with very different degrees 
of likelihood in the various stem-lines of taxa. 
It is essential to note that a reconstruction of 
phylogeny should be translated into a clas- 
sification, and not the phylogeny itself. In an 
attempt to base the classification on the same 
principle as the analysis (probabilities), the 
central taxa should reflect the highest degrees 
of certainty in the analysis. A similar point of 
view was made by Wiley (1 979, 1 981 ) in claim- 
ing to retain "important" taxa, which very often 
reflect stem-lines with high probabilities. 

Evolutionary systematists often claimed the 
inclusion of the "anagenetic component" into 
the classification (e.g. Mayr, 1981). Taxa of 
high rank are interpreted as an expression of 
major evolutionary gaps. This array can be 
real if caused by fast adaptive radiations and 
a lack of intermediate forms. I interpret taxa of 
high rank as reflecting stem lines with very 
likely monophyly. This interpretation equals 
the distinction between apomorphy and plesi- 
omorphy and gaps between character states 
(in a reconstruction). In this way, dado- and 
anagenesis are considered by correlating 
each with some probability. 

Whereas many authors prefer Wiley's 
(1979, 1981) sequential method of classifica- 
tion, other cladists still use the dichotomic 
Hennigian way (e.g. Ax, 1984, 1987; Berthold 
& Engeser, 1987). I prefer the former, and 
regard my own proposal as a modification of 
Wiley's (1981) methodology. 

Gauthier (1986) has proposed marking 
so-called "metataxa" (i.e. taxa, the holo- or 
paraphyletic status of which cannot be given 
at present) by an asterisk (taxon*). In combin- 
ing my original mode of marking grades as 
"taxa" (Haszprunar, 1986) with Gauthier's 
(1986) ideas, I have more recently proposed 
to mark grades by asterisks (*taxa*) and 
to mark metataxa (e.g. *Architaenioglossa*, 
*Cerithiimorpha*) by a combination of aster- 
isks and sedis mutabilis in the subtaxa 
(Salvini-Plawen & Haszprunar, 1987; Hasz- 
prunar, 1988). This enables a better con- 
versation of a metataxon into an para- or hol- 
ophyletic taxon upon addition of new data. At 
the time the study was finished, the Cerithio- 
idea was an example of a metataxon. I regard 
Bieler's (1990) solution of omitting the Cer- 
ithioidea from the classification as less ac- 
ceptable than my proposal of marking the 
taxon unequivocally. 



PRESERVATION OF 
TRADITIONAL NAMES 

In the earliest phase of my phylogenetic 
work (Haszprunar, 1985a,b), I frequently cre- 
ated new taxa of high rank. However, 
"nobody can hinder me to become wiser," 
and several of my friends (see Acknowledge- 
ments) have convinced me that preservation 
of traditional names is a better way. As out- 
lined in Haszprunar (1988: 370), certain new 
taxa still appear necessary to present phylog- 
eny unequivocally or to reflect taxa with high 
propabilities. I consider the Archaeogas- 
tropoda in its traditional, paraphyletic (ortho- 
phyletic) sense still useful in systematics, be- 
cause in many cases only shells (and 
radulae) are available, which do not allow for 
a more specific classification. Thiele's Meso- 
gastropoda is a paraphyletic group — it inde- 
pendently gives rise to both the Stenoglossa 
and Euthyneura — and therefore has been 
abandoned. I also regard the Neotaenioglo- 
ssa (again paraphyletic) as a provisional con- 
struct which should be abandoned in the fu- 
ture. 



CONCLUSION 

I have responded Bieler's (1 990) critique on 
the mode of my phylogenetic analysis on 
streptoneuran gastropods as follows: 

(a) I have provided arguments against do- 
ing a maximum-parsimony analysis with 
equal weighting of characters, (b) I have pre- 
sented examples of the character analysis 
and of placement of taxa to demonstrate the 
method used in the analysis. Proposed syn- 
apomorphies are considered as two-fold hy- 
potheses with distinct degrees of likelihood. 
Accordingly the cladogram is regarded as a 
theorem of probability, and taxa of high rank 
are thought to reflect stem-lines of high cer- 
tainty, (c) I have explained the use of certain 
taxa in the proposed classification. 



ACKNOWLEDGEMENTS 

I thank Rüdiger Bieler for the critical review 
of my papers, in particular with respect to de- 
tection of mistakes made, as well as for open- 
ing this discussion. I thank an anonymous ref- 
eree for very helpful criticism of an earlier 
draft of this manuscript. I am indebted to 
David Lindberg (University of California, Ber- 



200 



HASZPRUNAR 



keley), who introduced me in the field of com- 
puter-cladistics. I particularly acknowledge 
the kindness of James H. McLean (Los An- 
geles County Museum of Natural History) 
who detected my nomenclatonal mistakes. I 
am also grateful to Winston F. Ponder (Aus- 
tralia Museum. Sydney) and Anders Waren 
(National Museum of Sweden, Stockholm) for 
our discussions on presentation of taxa. 



REFERENCES 

ALBERCH. P., 1985. Problems with the interpreta- 
tion of developmental sequences. Systematic 
Zoology. 34: 46-58. 

AX. P.. 1984. Das phylogenetische System. Sys- 
tematisierung der lebenden Natur Auf Grund 
ihrer Phylogenese. Gustav Fischer Verlag. Stutt- 
gart. 

AX. P.. 1987, The phylogenetic system. The sys- 
tematization of organisms on the basis of their 
phylogenesis (Translation of Ax, 1 984, by R. P. S. 
Jefferies). John Wiley & Sons, New York. 

BERTHOLD, T. & T. ENGESER, 1987, Phyloge- 
netic analysis and systematization of the Ceph- 
alopoda. Verhandlungen des Naturwissenschaft- 
lichen Vereins Hamburg. (NF) 29: 187-220. 

BIELER. R.. 1988. Phylogenetic relationships in the 
gastropod family Architectonicidae, with notes on 
the family Mathildidae. Pp. 205-240 In: PON- 
DER, W. F., ed.. Prosobranch phytogeny. !\/iala- 
cological Review Supplement. 4. 

BIELER. R. 1990, Haszprunars 'clado- 
evolutionary " classification of the Gastropoda — a 
critique. Malacologia. 31:371-380. 

BIELER, R. & P. M. MIKKELSEN. 1988, Anatomy 
and reproductive biology of two Western Atlantic 
species of Vitnnellidae, with a case of protan- 
drous hermaphroditism in Rissoacea. Nautilus. 
102: 1-29. 

BRYANT, H. N., 1989, An evaluation of cladistic 
and character analyses as hypothetico-deductive 
procedures, and the consequences for character 
weighting. Systematic Zoology. 38: 214-227. 

COLLESS. D. H., 1983, Wagner trees in theory and 
praxis. Pp. 259-278 In: FELSENSTEIN, J., ed.. 
Numerical taxonomy. NATO ASI Series Gl. 
Springer Verlag. Heidelberg. 

CROFTS, D. R., 1937, The development of Haliotis 
tuberculata. with special reference to the organo- 
genesis during torsion. Philosophical Transac- 
tions of the Royal Society of London. B208: 219- 
268. 

DAVIS, G. fvl., 1989. Character analysis, conver- 
gence, taxonomy and evolution of the Hydrobi- 
idae. Abstracts of the 10th International l^alaco- 
logical Congress. Tübingen 1989: 56. 

FRETTER, v., 1 965, Functional studies of the anat- 
omy of some neritid prosobranchs. Journal of Zo- 
ology. London, 147: 46-74. 



GAUTHIER, J., 1986, Saunschian monophyly and 
the ongin of birds. Pp. 1-55 In: PADIAN, K., ed., 
The origin of birds and the evolution of flight. 
t\J\emoirs of the Californian Academy of Sci- 
ences. 8: 98 pp. 

GOSLINER. T. M. & M. T. GHISELIN, 1984, Paral- 
lel evolution in opisthobranch gastropods and its 
implications for phylogenetic methodology. Sys- 
tematic Zoology. 33: 255-274. 

HASZPRUNAR, G,, 1985a, The Heterobranchia— 
a new concept for the phylogeny of the higher 
Gastropoda. Zeitschnft für Zoologische System- 
atik und Evolutionsforschung. 23: 15-37. 

HASZPRUNAR. G.. 1985b, On the anatomy and 
systematic position of the Mathildidae (Mollusca, 
Allogastropoda). Zoológica Scripta. 14: 201-213. 

HASZPRUNAR, G., 1986, Die klado-evolutionäre 
Klassifikation — Versuch einer Synthese. Zeit- 
schnft für Zoologische Systematik und Evolution- 
sforschung. 24: 89-109. 

HASZPRUNAR, G., 1988, On the origins and evo- 
lution of major gastropod groups, with special ref- 
erence to the Streptoneura. Journal of Molluscan 
Studies. 54: 367-441. 

HASZPRUNAR, G.. 1989, Estimation of homology: 
implications for the methodologies of classifica- 
tion. Abstracts of the 2nd European Society of 
Evolutionary Biology Congress. Roma 1989.:3^. 

HEALY, J M., 1988a. The ultrastructure of sperma- 
tozoa and spermiogenesis in pyramidellid gastro- 
pods, and Its systematic importance. Helgo- 
lander f\/leeresuntersuchungen. 42: 303-318. 

HEALY, J. M., 1988b, Sperm morphology and its 
systematic importance in the Gastropoda. Pp. 
251-266 In: PONDER, W. F., ed., Prosobranch 
phylogeny. Malacological Review Supplement. 4. 

HENNIG, W., 1966, Phylogenetic systematics. Uni- 
versity of Illinois Press. Urbana, 263 pp. 

HOUBRICK. R. S., 1988, Cerithioidean phylogeny. 
Pp. 88-128 In: PONDER, W. F., ed., Proso- 
branch phylogeny. Malacological Review Sup- 
plement. 4. 

HOUBRICK, R. S., 1989, Campanile revisited: Im- 
plications for cerithioidean phylogeny. American 
t\/lalacological Bulletin. 7: 1-6. 

LINDBERG, D. R., 1988, The Patellogastropoda. 
Pp. 35-63 In: PONDER, W. F., ed., Prosobranch 
phylogeny. Malacological Review Supplement, 4. 

MACKIE, G. L., 1984, Bivalves. Pp. 351-418 In: 
TOMPA, A. S., N. H. VERDONK & J. A. M. VAN 
DEN BIGGELAAR, eds.. The Mollusca. Vol. 7: 
Reproduction. Academic Press, London. 

MAYR, E., 1981, Biological classification: Towards 
a synthesis of opposing methodologies. Science, 
214: 510-514. 

NEFF, N. A., 1986, A rational basis for a pr/or/ char- 
acter weighting. Systematic Zoology. 35: 11- 
123. 

PONDER, W. F., 1987, The anatomy and relation- 
ships of the pyramidellacean limpet Amathina tri- 
carinata (Mollusca: Gastropoda). Asian Manne 
Biology. 4: 1-34. 

PONDER, W. F., 1988, The truncatelloidean 



GASTROPOD SYSTEMATICS: REPLY TO BIELER 



201 



( = rissoacean) radiation — a preliminary phylog- 
eny. Pp. 129-164 In: PONDER, W. P., ed., 
Prosobranch phylogeny. Malacological Review 
Supplement. 4. 

PONDER, W. F. & A. WAREN, 1988, Appendix. 
Classification of the Caenogastropoda and Het- 
erostropha — a list of the family-group names and 
higher taxa. Pp. 288-326 In: PONDER, W. P., 
ed., Prosobranch phylogeny. Malacological Re- 
view Supplement. 4. 

REID, D. G., 1989, The comparative morphology, 
phylogeny and evolution of the gastropod family 
Littorinidae. Philosophical Transactions of the 
Royal Society of London, B324: 1-110. 

REMANE, A., 1952, Die Grundlagen des 
natürlichen Systems, der vergleichen-den Anat- 
omie und der Phylogenetik. Akademische Ver- 
lagsgesellschaft Geest & Portig, Leipzig. 

REMANE, A., 1954, Morphologie als Homologien- 
forschung. Verhandlungen der Deutschen Zool- 
ogischen Gesellschaft, 1954: 159-183. 

RIEDL, R., 1975 Die Ordnung des Lebendigen. 
Systembedingungen der Evolution. Paul Parey, 
Hamburg & Berlin. 

RIEDL, R., 1978, Order in living organisms. A sys- 
tems analysis of evolution (Translation of Riedl, 
1975, by R. P. D. Jefferies), Wiley Interscience, 
New York. 

RIEGER, R. M. & S. TYLER, 1979, The homology 
theorem in ultrastructure research. American Zo- 
ologist. 19: 655-664. 

RIEGER, R. M. & S. TYLER, 1985, Das Homologi- 
etheorem in der Ultrastruktur Forschung. Pp. 21- 
36 In: OTT, J. A., G. P. WAGNER & F. M. 
WUKETITS, eds.. Evolution. Ordnung und 
Erkenntnis. Parey Verlag, Berlin & Hamburg. 

RUPPERT, E. E., 1982, Homology recognition as a 
basis for comparative biology. Pp. 31-54 In: Pro- 
ceedings of the Third International Symposium 
on the Tardigrada. East Tennessee State Univer- 
sity Press. 

SALVINI-PLAWEN, L. V. & G. HASZPRUNAR, 
1 987, The Vetigastropoda and the systematics of 
streptoneurous gastropods (Mollusca). Journal 
of Zoology. London, A211: 747-770. 

SANDERSON, M. J. & M. J. DONOGHUE, 1989, 
Patterns of variation in levels of homoplasy. Ev- 
olution. 43: 1781-1795. 

TYLER, S., 1988, The role of function in determi- 
nation of homology and convergence — 
examples from invertebrate adhesive organs. 
Fortschritte der Zoologie. 36: 331-347. 

WESTHEIDE, W. & R. M. RIEGER, 1987, System- 
atics of the amphi-Atlantic Microphthalmus- 
listensis-specles group (Polychaeta: Hesion- 
idae). Facts and concepts for reconstruction of 
phylogeny and speciation. Zeitschrift für Zoolo- 
gische Systematik und Evolutionsforschung. 25: 
12-39. 

WILEY, E. O., 1979, An annotated Linnean hierar- 
chy, with comments on natural taxa and compet- 
ing systems. Systematic Zoology. 28: 308-337. 

WILEY, E. O., 1981, Phylogenetics. The theory and 



practise of phylogenetic systematics. John Wiley 
& Sons, New York, 439pp. 
YONGE, С M., 1947, The palliai organs in the as- 
pidobranch gastropods and their evolution 
throughout the Mollusca. Philosophical Transac- 
tions of the Royal Society of London. B232: 443- 
518, 1 pi. 



APPENDIX 
Corrections of Haszprunar (1988) 

(1) p. 377: replace "McLean, 1987" by 
"McLean, 1988". 

(2) p. 378/legend Fig. 1 : Replace "Ponder, 
1987" by "Ponder, 1988a". 

(3) p. 381 : So far as is known Truncatel- 
loidea — Vitrinellidae have monopectinate gills 
(e.g. Bieler & Mikkelsen, 1988). 

(4) p. 389, p. 400/Table 2, p. 416: As re- 
cently outlined by Houbrick (1989), I have 
misinterpreted his earlier data on Campanile 
symbolicum, listing "eggs connected by cha- 
lazae" for this taxon. In fact, campanllid egg- 
mass connections resemble those found in 
the Epitoniidae. True chalazae are present in 
the Valvatidae, however. 

(5) p. 400/Table 2: A loss of teleoconch oc- 
curs also in numerous euthyneuran taxa. 

(6) p. 401 /Table 2: Tubular salivary glands 
with ducts occur in patelloid Docoglossa (Pa- 
tellogastropoda), but not in Lepetelloidea. 

(7) p. 401 /Table 2: A cord-like visceral loop 
throughout its length is restricted to the Patel- 
loidea and Nacelloidea (Neolepetopsidae?). 

(8) p. 401 /Table 2: Eyes with a lens also 
occur in the Fissurellidae and Scissurellidae. 

(9) p. 413: Replace "Haszprunar, 1988" by 
"Haszprunar, 1989". 

(10) p. 420: Bieler (1988) found some 
more diagnostic differences between Archi- 
tectonicidae and Mathildidae. 

(11) p. 424/Fig. 5: Points 41 and 42 should 
be interchanged. 

(12) p. 428/Table 5a: The arrangement 
and subordination of "Superfamily Hot-Vent 
group A {Melanodrymia)" and "Superfamily 
Neomphaloidea" might appear to include 
them in the Neritimorpha. Judged from text 
(pp. 412-414) and phylogram (p. 424/Fig. 5), 
it should be clear that this is not the case, 
however. 

(13) p. 428/Table 5a: Change "Nacel- 
loidea Lindberg, 1988" to "Nacelloidea 
Thiele, 1891"; "Helicinoidea Thompson, 
1980" to "Helicinoidea Férrusac, 1822"; and 



202 



HASZPRUNAR 



"Scissurelloidea McLean & Haszprunar, 
1988" to "Scissurelloidea Gray, 1847." Ac- 
cording to Ponder and Waren (1988) it should 
be ■Ampullarioidea Gray, 1824": "Janthi- 
noidea Lamarck, 1810"; "Littorinoidea Gray, 
1840"; "Tonnoidea Suter, 1913"; and Ptero- 
trachoidea Férrusac, 1821" should be re- 
placed by "Carinarioidea Blainville, 1818." 

(14) p. 430 Table 5d; The wrong (printer's 
error) ranking should be corrected so that N. 
N. ("Helicoida") becomes superior to Neriti- 
morpha and N. N. ("Euhelicoida"). 

(15) p. 436; Mackie (1984) was missing in 
the reference list. 

(16) The symposium-volume, "Proso- 
branch phytogeny." was published in late 
1988, and there are differences between the 
published papers and the manuscripts and ab- 
stracts that were made available to me prior to 



publication (Bieler, 1988; Healy, 1988b; Hou- 
brick, 1988; Lindberg, 1988; Ponder, 1988; 
Ponder & Waren, 1988). For example. Pon- 
ders (1988) "Cingulopsoidea Fretter & Patil, 
1 958" was not included in my classification, for 
nomenclatorial corrections; see (13). 



The editor-in-chief of Malacologia welcomes letters 
that comment on vital issues of general importance 
to the field of Malacology, or that comment on the 
content of the journal. Publication is dependent on 
discretion, space available and, in some cases, re- 
view. Address letters to: Letters to the Editor, Mal- 
acologia, care of the Department of Malacology, 
Academy of Natural Sciences, 19th and the Park- 
way, Philadelphia, PA 19103. 



Publication dates 



Vol 


28, 


No. 


1-2 


19 January 19 


Vol 


29, 


No. 


1 


28 June 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 



WHY NOT SUBSCRIBE TO MALACOLOGIA? 
ORDER FORM 



Your name and address 



Send U.S. $21.00 for a personal subscription (one volume) or U.S. $35.00 for an 
institutional subscription. Make checks payable to "MALACOLOGIA." 

Address: Malacologia, Academy of Natural Sciences 
Nineteenth and the Parkway, Philadelphia 
PA 19103, U.S.A. 



MALACOLOGIA, 1990, 32(1 ; 



32(1 : 



MALACOLOGIA 



1990 



INSTRUCTIONS FOR AUTHORS 



1. MALACOLOGIA publishes original re- 
search on the Mollusca that is of high quality 
and of broad international interest. Papers 
combining synthesis with innovation are par- 
ticularly desired. While publishing symposia 
from time to time, MALACOLOGIA encour- 
ages submission of single manuscripts on 
diverse topics. Papers of local geographical 
or systematic interest should be submitted 
elsewhere, as should papers whose primary 
thrust is physiology or biochemistry. Nearly all 
branches of malacology are represented on 
the pages of MALACOLOGIA. 

2. Manuscripts submitted for publication 
are received with the tacit understanding that 
they have not been submitted or published 
elsewhere in whole or in part. 

3. Manuscripts 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 translations 
(complete with main titles). Include all foreign 
accents. Both American and British spellings 
are allowed. 

4. Unless indicated otherwise below, con- 
tributors should follow the recommendations 
in the Council of Biology Editors (CBE) Style 
Manual (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 consistent. Avoid 
internal page references (which have to be 
added in page proof). 

7. Choose a running title (a shortened 
version of the main title) of fewer than 50 
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. 

10. Use the metric system throughout. Mi- 
cron should be abbreviated fxm. 

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 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 
Vz 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 |xm," not "163 |xm"). 
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 dupli- 
cate information given in the text. Each let- 
tered abbreviation labeling an individual fea- 
ture in a figure must either be explained in 
each caption (listed alphabetically), or be 
grouped in one alphabetic sequence after the 
Methods section. Use the latter method if 
many abbreviations are repeated on different 
figures. 



17. Tables are to be used sparingly, and 
vertical lines not at all. 

18. References cited in the text must be in 
the Literature Cited section and vice versa. 
Follow a recent issue of MALACOLOGIA for 
bibliographic style, noting especially that se- 
rials are cited unabbreviated. Supply pagina- 
tion for books. Supply information 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 Litera- 
ture Cited section. 

20. For systematic papers, all new type- 
specimens must be deposited in museums 
where they may be studied by other scien- 
tists. Likewise MALACOLOGIA requires that 
voucher specimens upon which a paper is 
based be deposited in a museum where they 
may eventually be reidentified. 

21. Submit each manuscript in triplicate. 
The second and third copies can be reproduc- 
tions. 

22. Authors who want illustrations returned 
should request this at the time of ordering re- 
prints. Otherwise, illustrations will be main- 
tained for six months only after publication. 



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 COSTS 

25. For Vol. 32, personal subscriptions are 
U.S. $21.00 and institutional subscriptions 
are U.S. $35.00. Back issues and single vol- 
umes: $35.00 for non-institutional purchaser; 
$40.00 for institutional purchaser. There is a 
one dollar handling charge per volume for all 
purchases of single volumes. Address inqui- 
ries to the Subscription Office. 



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 
primarily 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, 19th and 
the Parkway, Philadelphia, Pennsylvania 19103, U.S.A. 



VOL. 32, NO. 1 MALACOLOGIA 1990 

CONTENTS 

UMITAS MALACOLOGICA 

10th International Malacologlcal Congress Symposium 

BIOLOGY AND EVOLUTION OF TOXOGLOSSAN GASTROPODS 

JOHN D. TAYLOR 

Introduction 1 

YURI I. KANTOR 

Anatomical Basis for the. Origin and Evolution of the Toxoglossan Mode 

of Feeding 3 

JOHN D. TAYLOR 

The Anatomy of the Foregut and Relationships in the Terebridae 19 

JAMES NYBAKKEN 

Ontogenetic Change in the Conus Radula, its Form, Distribution Among 

the Radula Types, and Significance in Systematics and Ecology 35 

ALAN J. KOHN 

Tempo and Mode of Evolution in Conidae 55 

PHILIPPE BOUCHET 

Turrid Genera and Mode of Development: The Use and Abuse of Pro- 
toconch Morphology 69 

E. ALISON KAY 

Turrid Faunas of Pacific Islands 79 

MALACOLOGIA CONTRIBUTED PAPERS 

RITA TRIEBSKORN & С KÜNAST 

Ultrastructural Changes in the Digestive System of Deroceras Reticula- 
tum (Mollusca; Gastropoda) Induced by Lethal and Sublethal Concen- 
trations of the Carbamate Molluscicide Cloethocarb 89 

ROGER R. SEAPY 

The Pelagic Family Atlantidae (Gastropoda: Heteropoda) From Hawai- 
ian Waters: A Faunistic Survey 1 07 

RÜDIGER BIELER & RICHARD E. PETIT 

On the Vahous Editions of Tetsuaki Kira's "Coloured Illustrations of the 
Shells of Japan" and "Shells of the Western Pacific in Color Vol. I," With 
an Annotated List of New Names Introduced 131 

JOSE ANGEL ALVAREZ PEREZ, MANUEL HAIMOVICI & 

JOAO CARLOS BRAHM COUSIN 

Sperm Storage Mechanisms and Fertilization in Females of Two South 
American Eledonids (Cephalopoda:Octopoda) 1 47 

ALAN R. KABAT 

Predatory Ecology of Naticid Gastropods with a Review of Shell Boring 
Prédation 1 55 

Letters to the Editor 

GERHARD HASZPRUNAR 

Towards a Phylogenetic System of Gastropoda Part I: Traditional Meth- 
odology — A Reply 1 95 




L 32, NO, 2 1M^1 



MALACOLOGIA 



UMITAS MALACOLOGICA 
9th International Malacological Congress Symposium 

EVOLUTIONARY BIOLOGY OF OPISTHOBRANCHS 



nternational 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 

Nineteenth Street and the Parkway 

Philadelphia, Pennsylvania 19103, U.S.A. 



EUGENE COAN 

California Academy of Sciences 

San Francisco, CA 



Co-Editors: 



Assistant Managing Editor: 

CARYL HESTERMAN 

Associate Editors: 



CAROL JONES 
Dundalk, MD 



JOHN B. BURCH 
University of f^ichigan 
Ann Arbor 



ANNE GISfVlANN 
Maadi 

Egypt 



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



KENNETH J. BOSS, 

Museum of Comparative Zoology 

Cambndge, Massachusetts 

JOHN BURCH, President-Elect 

MELBOURNE R. CARRIKER 
University of Delaware, Lewes 

GEORGE M. DAVIS 
Secretary and Treasurer 

CAROLE S. HICKMAN. Vice-President 
University of California, Berkeley 



JAMES NYBAKKEN, President 
Moss Landing Marine Laboratory 
California 

CLYDE F. E. ROPER 
Smithsonian Institution 
Washington, D.C. 

W. D. RUSSELL-HUNTER 
Syracuse University, New York 

SHI-KUEI WU 

University of Colorado Museum, Boulder 



Participating Members 

EDMUND GITTENBERGER JACKIE L VAN GOETHEM 

Secretary, UNITAS MALACOLOGICA Treasurer, UNÎTAS MALACOLOGICA 

Rijksmuseum van Natuurlijke Koninklijk Belgisch Instituut 

Historie voor Natuunwetenschappen 

Leiden, Netherlands Brüssel, Belgium 



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

ELMER G. BERRY, 
Germantow" ''"-^r-'i-nd 



Emeritus Members 

ROBERT ROBERTSON 

The Academy of Natural Sciences 

Philadelphia, Pennsylvania 

NORMAN F. SOHL 
U.S. Geological Survey 
Reston, Virginia 



Copyright Ф 1991 by the Institute of Malacology 



1991 
EDITORIAL BOARD 



Mrz 

LíbKM.'^Y 

JUN 1 4 1V91 



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. CALOW 

University of Shieffield 
United Kingdom 

A. H. CLARKE, Jr. 
Portland, Texas, U.S.A. 

B. С CLARKE 
University of Nottingtiam 
United Kingdom 

R. DILLON 

College of Chiarleston 

SC, U.S.A. 

С J. DUNCAN 
University of Liverpool 
United Kingdom 



E. FISCHER-PIETTE 



V. FRETTER 
University of Reading 
United Kingdom 

E. GITTENBERGER 
Rijksmuseum van Natuurlijke Historie 
Leiden, Netherlands 

F. GIUSTI 

Università di Siena, Italy 

A. N. GOLIKOV 
Zoological Institute 
Leningrad, U.S.S.R. 

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



A. V. GROSSU 

Universitatea Bucurestiy f\ рэ r\ 

Romania UNiVERolTY 

T. HABE 

Tokai University 

Shiimizu, Japan 



R. HANLON 

Marine Biomedical Institute 

Galveston, Texas, U.S.A. 

J. A. HENDRICKSON, Jr. 
Academy of Natural Sciences 
Philadelphia, PA, U.S.A. 

K. E. HOAGLAND 

Association of Systematics Collections 

Washington, DC, U.S.A. 

B. HUBENDICK 
Naturhistoriska Museet 
Göteborg, Sweden 

S. HUNT 

University of Lancaster 

United Kingdom 

R. JANSSEN 

Forschungsinstitut Senckenberg, 
Frankfurt am Main, Germany 
(Federal Republic) 

R. N. KILBURN 
Natal Museum 
Pietermaritzburg, South Africa 

M. A. KLAPPENBACH 

Museo Nacional de Historia Natural 

Montevideo, Uruguay 

J. KNUDSEN 

Zoologisk Institut & Museum 

Kebenhavn, Denmark 

A. J. KOHN 

University of Washington 

Seattle, U.S.A. 

A. LUCAS 

Faculté des Sciences 

Brest, France 

C. MEIER-BROOK 
Tropenmedizinisches Institut 
Tübingen, Germany (Federal Republic) 



H. к. MIENIS 

Hebrew University of Jerusalem 

Israel 

J. E. MORTON 
The University 
Auckland, New Zealand 

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

R. NATARAJAN 
Marine Biological Station 
Porto Novo, India 

J. 0KLAND 
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. 

W. F. PONDER 
Australian Museum 
Sydney 

R. D. PURCHON 

Chelsea College of Science & Technology 

London, United Kingdom 

01 Z. Y. 

Academia Sinica 

Oingdao, People's Republic of China 

N. W. RUNHAM 

University College of North Wales 

Bangor, United Kingdom 



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

F. STARMÜHLNER 

Zoologisches Institut der Universität 

Wien, Austria 

Y. I. STAROBOGATOV 
Zoological Institute 
Leningrad, U.S.S.R. 

W. STREIFE 
Université de Caen 
France 

J. STUARDO 
Universidad de Chile 
Valparaiso 

S. TILLIER 

Muséum National d'Histoire Naturelle 

Paris, France 

R. D. TURNER 
Harvard University 
Cambridge, Mass., U.S.A. 



W. S. S. VAN BENTHEM JUTTING 



J. A. VAN EEDEN 
Potchefstroom University 
South Africa 

N. H. VERDONK 
Rijksuniversiteit 
Utrecht, Netherlands 

В. R. WILSON 

Dept. Conservation and Land Management 

Netherlands, Western Australia 

H. ZEISSLER 

Leipzig, Germany (Democratic Republic) 

A. ZILCH 

Forschungsinstitut Senckenberg 
Frankfurt am Main, Germany (Federal 
Republic) 



UMITAS MALACOLOGICA 

Ninth International Malacological Congress Symposium 

Edinburgh, Scotland 1986 



EVOLUTIONARY BIOLOGY OF OPISTHOBRANCHS 



Malcolm Edmunds 

Organizer and Editor 

Department of Applied Biology 

Lancashire Polytechnic 

Preston, United Kingdom 



Malacologia Guest Editor 

Christopher D. Todd 

Gatty Marine Laboratory 

University of St. Andrews 

Scotland 



MALACOLOGIA, 1991, 32(2): 203 

EDITOR'S NOTE 

I accepted these papers for publication on sium has not been delayed. Accordingly, Пгл- 
2 January 1991 . Due to problems beyond the ited attempt has been made to make all pro- 
control of UNITAS this symposium was in cedures for this publication conform exactly to 
danger of not being published. As the result of Malacologia specifications. Our sympathy is 
the concerted efforts of Malcolm Edmunds with the authors and UNITAS. 
and Chris Todd in the fall of 1990, the manu- 
scripts were provided to Malacologia on com- 
puter disc. 

In order to expedite the publication of this 
symposium and to better serve UNITAS in a George M. Davis 

timely fashion, the publication of this sympo- Editor-in-Chief 

Malacologia 



♦F 



203 



MALACOLOGIA, 1991, 32(2): 205-207 



INTRODUCTION 

Malcolm Edmunds 
Department of Applied Biology, Lancashire Polytechnic, Preston 



In suggesting the title for this Symposium 
'Evolutionary Biology of Opisthobranchs' I 
was aware of the wealth of recent studies on 
taxonomy, comparative morphology, neuro- 
physiology, zoogeography, faunistics and 
ecology; but while comparative morphology 
papers usually have an evolutionary theme, 
papers on zoogeography and neurophysiol- 
ogy rarely do. I was also aware that some 
talks (e.g. very specialized or narrowly taxo- 
nomic papers) can be tedious to listen to un- 
less one is just as badly bitten with the same 
enthusiasm bug as the speaker. I therefore 
decided to encourage papers from as wide a 
field of study as possible but united by an 
evolutionary theme. I hoped that this would 
make the Symposium of interest to the non- 
specialist biologist as well as to opistho- 
branch aficionados, and I hoped too that it 
might stimulate new approaches to those 
fields of study that have hitherto lacked an 
evolutionary approach. 

Did this strategy succeed? It is hardly for 
me to judge, but there are certainly some 
gaps in the range of subjects covered. There 
is nothing on neurophysiology, for example, 
but there are papers on comparative morphol- 
ogy, development and ecology. Most papers 
concentrate on the ever popular Nudibranchia 
and Ascoglossa, but there are others which 
deal with the Bullomorpha and the Aplysio- 
morpha, and there is even one which reviews 
current knowledge of a very little-known 
group, the Rhodopidae. 

The names used for the major taxa of 
opisthobranchs are still very far from being 
agreed by all workers. There is still contro- 
versy over the most appropriate names for 
several of the orders: Bullomorpha or Ceph- 
alaspidea, Aplysiomorpha or Anaspidea, and 
Ascoglossa or Sacoglossa. Possible confu- 
sion with chordate class names together with 
the advantage of a name that relates to a typ- 
ical genus in the group suggest that use of 
Bullomorpha and Aplysiomorpha should be 
encouraged. But Ascoglossa versus Sacoglos- 
sa is more of a problem: both names relate to 
the radular sac or ascus. Having used Saco- 



glossa for a quarter of a century the word As- 
coglossa still sticks in my gullet (or perhaps in 
my ascus like a discarded radular tooth), but 
since most of the recognized authorities on 
the group now prefer Ascoglossa, I guess I 
must discard my old preference into my ascus 
and accept the change. 

A century ago comparative morphology 
was the central plank of zoology. It was also 
used as an aid to devising a systematic 
arrangement, but there was rarely any 
discussion of evolution. Today comparative 
morphology is much more functional in its 
approach. It is still important, but primarily for 
the light it sheds on evolution, and only 
secondarily for the help it gives with classifi- 
cation. Classification today is an exercise in 
trying to devise a system that reflects the 
evolution of the group, while recognizing the 
limitations imposed by a rigid hierarchy of 
taxa which can never adequately take ac- 
count of the realities of the different evolu- 
tionary rates of different species. Similarly 
classical zoogeography was really little more 
than putting dots on maps whereas today 
one looks for causes of the geographical 
range of a particular species. These causes 
can be found in the animal's physiology, 
ecology or development (Clark, 1975; Ed- 
munds, 1977), while a more detailed study 
gives insight into the process of speciation 
(Edmunds, 1982). Evolution occurs by 
means of natural selection, and the forces of 
natural selection have been studied exten- 
sively by observing changes in morph fre- 
quency in terrestrial gastropods such as 
Cepaea nemoralis (reviewed by Jones et al., 
1977; Clark et al., 1978; and Cain, 1983), and 
also in some marine gastropods such as 
Littorina spp. (reviewed by Berger, 1983; 
and Raffaelli, 1982). No such studies have 
been made on opisthobranchs. Two reasons 
for this are the lack of a hard shell with easily 
quantifiable characters, and the difficulty of 
monitoring individuals and populations of 
such small animals in the sea. Yet such 
studies are now possible: many opistho- 
branch workers are expert SCUBA divers, 



205 



206 



EDMUNDS 



and several polymorphic species of opistho- 
branch are now known which are ideal for 
such a study, for example the widespread Eu- 
ropean aeolid Eubranchus farrani (Forbes & 
Goodsir) (Edmunds & Kress, 1969), and the 
Indo-Pacific Phestilla minor Rudman (Rud- 
man, 1981). 

This Symposium includes papers that 
cover a variety of different aspects of the 
evolutionary biology of opisthobranchs. First 
there are three papers on food and feeding 
habits. Each major taxon of opisthobranchs is 
associated with a particular type of food: 
sponges for dorids, coelenterates for aeolids 
and dendronotaceans, and green algae for 
sacoglossans (sorry, ascoglossans). While 
some species of opisthobranch are eurypha- 
gous in their choice of food, others are 
stenophagous to the extreme of eating just a 
single species of prey. Jensen's paper is 
essentially a comparative morphology study 
of the evolution of feeding structures in the 
Ascoglossa. While the fine details of feeding 
structures are closely linked to specific foods, 
Jensen is able to tease out those anatomical 
characters that indicate evolutionary relation- 
ships and which can also be used in 
classification. The second paper, by Cat- 
taneo Vietti and Balduzzi, reviews the food 
and radular characteristics of the Mediterra- 
nean genera of dohds. It then applies a 
method of correspondence analysis to the 
data set and attempts to relate food to radular 
structure. A clear correlation is found be- 
tween radula width and diet, but the relation- 
ship between finer details of radular tooth 
shape and diet is not so evident. The third 
paper in this section by Picton examines food 
and feeding habits from yet a third viewpoint. 
It describes the feeding habits of a single 
species, the aberrant aeolid Cumanotus 
beaumonti, and relates this to its behaviour 
and ecology. 

The next group of papers relate to defence. 
The molluscan shell probably evolved as a 
defensive adaptation in a sluggish, benthic 
animal that would otherwise have been vul- 
nerable to faster moving, jawed predators 
(Edmunds, 1974). But the shell also has its 
disadvantages: it is cumbersome to carry 
around, its formation requires a lot of energy 
(which could perhaps be better expended 
in reproduction), it locks up a lot of calcium 
so that the animal can only live where there 
is a supply of this mineral, and it has a regu- 
larity of shape such that it is difficult to cam- 
ouflage from predators. With all these disad- 



vantages it is hardly surphsing that some 
representatives of all three gastropod sub- 
classes have reduced and lost the shell. But it 
is in the opisthobranchs that shell reduction 
has occurred most often and been so out- 
standingly successful as judged by the diver- 
sity of form and number of species that are 
extant. Shell reduction and loss can only oc- 
cur provided that an animal has other means 
of defence including bodily appendages, col- 
oration and glandular secretions (Edmunds, 
1966). The paper by Poulicek, Voss-Foucart 
and Jeuniaux presents the results of a de- 
tailed analysis of the chemical constitution of 
shells from a variety of opisthobranchs, some 
with well-developed shells, and others with 
reduced shells. It attempts to draw functional 
conclusions relating chemical constitution to 
the development (or reduction) of the shell, 
but precisely why one shell should have more 
Chitin or lysine than another remains un- 
known. The second paper in this section by 
García-Gómez, Medina and Coveñas de- 
scribes the histology of the large mantle 
'glands' of chromodorids. These are wide- 
spread in the family and their precise location 
has been used as a taxonomic character 
(Rudman, 1984). But this careful study shows 
that, far from being typical defensive glands 
opening on the dorsal mantle surface, they do 
not open to the exterior at all, hence the au- 
thors call them mantle dermal formations' 
rather than glands. Nevertheless they must 
surely have a defensive function (as evi- 
denced by their location), but whether they 
can only affect a predator that tears open the 
mantle, or whether appropriate stimulation 
causes the contents to burst out in the same 
way that nematocysts burst out of the cni- 
dosac of an aeolid even though it initially has 
no opening, is not known. The final paper in 
this section by Edmunds examines the evi- 
dence for the occurrence of warning colora- 
tion in nudibranchs. It concludes that some 
species probably are aposematic, it predicts 
likely consequences of aposematism which 
could be tested experimentally and it dis- 
cusses how aposematism may have evolved 
in nudibranchs. 

The third section of the Symposium covers 
embryological development and larval ecol- 
ogy. Soliman reviews the patterns of de- 
velopment of opisthobranchs with particular 
reference to those from the Red Sea, and 
compares these with the developmental pat- 
terns of prosobranchs. Todd reviews data on 
the development and larval ecology of Onchi- 



INTRODUCTION 



207 



dons bilamellata, and shows that the meta- 
morphosed juveniles spend several weeks 
feeding on detritus before they are large 
enough to attack the definitive adult prey (bar- 
nacles). The evolution of different modes of 
larval development is reviewed in the light of 
larval and adult feeding habits and prey avail- 
ability. 

The fourth section of the Symposium 
comprises a single ecological faunistic paper 
by Cattaneo Vietti and Chemello on the 
opisthobranch fauna of one particular habitat. 
Besides giving a list of the species found in 
one Mediterranean lagoon, this paper also 
reviews the species recorded from other 
lagoons in the Mediterranean. While the data 
are inevitably very incomplete, this is the sort 
of habitat that would be ideal to select for an 
investigation into the origin of an entire fauna. 
With knowledge of preferred foods, devel- 
opmental pattern, salinity tolerances and 
other physiological factors of the various 
species in the neighbouring sea, it should be 
possible to explain why some species are 
commonly found in lagoons while others are 
absent. 

The final section of the Symposium con- 
tains two papers on comparative morphol- 
ogy. The first, by Salvini-Plawen, is in 
classical mould using every available piece of 
anatomical information to build up a picture of 
the evolutionary relationships of a very 
little-known group of molluscs, the Rhodop- 
idae. A new genus, Helminthope, is de- 
scribed which has several differences from 
Rhodope. It is now possible to argue the 
phylogenetic relationships of this group with 
much more confidence than has been possi- 
ble in the past, but, perhaps surprisingly, their 
systematic position still remains obscure. The 
final paper by Gosliner reviews the numerous 
examples of parallel evolution in opistho- 
branchs but here the comparative approach 
follows the phylogenetic cladistic method 
advocated by Hennig (1966) of uniting 
groups on the basis of shared derived 
(apomorphic) characters. Arguments con- 
cerning the relative merits of traditional 
evolutionary classification and of phylo- 
genetic cladistic methods are summarized 
by Ridley (1986). The conclusions reached 
on the basis of Gosliner's analysis do not 
agree with those of all recent authors, some 
of whom link groups into taxa on the basis 
of shared ancestral (plesiomorphic) charac- 
ters. 



LITERATURE CITED 

BERGER, E. M., 1983, Population genetics of ma- 
rine gastropods and bivalves, pp. 563-596. In: 
RUSSELL-HUNTER, W. D. (ed.), Ecology. The 
Mollusca (ed. К. Wilbur) 6. Academic Press, Lon- 
don & New York. 

CAIN, A. J., 1983, Ecology and ecogenetics of ter- 
restrial molluscan populations, pp. 597-647. In: 
RUSSELL-HUNTER, W. D. (ed.). Ecology. The 
Mollusca (ed. К. Wilbur) 6. Academic Press, Lon- 
don & New York. 

CLARK, K. В., 1975, Nudibranch life cycles in the 
Northwest Atlantic and their relationship to the 
ecology of fouling communities, Helgoländer wis- 
senschaftliche Meeresuntersuchungen, 27: 28- 
69. 

CLARKE, В., ARTHUR, W., HORSLEY, D. T. & 
PARKIN, D. T., 1978, Genetic variation and nat- 
ural selection in pulmonate molluscs, pp. 219- 
270. In: FRETTER, V. & PEAKE, J. (eds.), Sys- 
tematics. Evolution and Ecology. Pulmonates 2A. 
Academic Press, London & New York. 

EDMUNDS, M., 1966, Defensive adaptations of 
Stiliger vanellus Marcus, with a discussion on the 
evolution of 'nudibranch' molluscs. Proceedings 
of the Malacological Society of London, 37: 73- 
81. 

EDMUNDS, M., 1 974, Defence in animals: a survey 
of anti-predator defences. Longman, Harlow, 357 

PP- 

EDMUNDS, M., 1977, Larval development, oceanic 
currents, and origins of the opisthobranch fauna 
of Ghana. Journal of Molluscan Studies, 43: 
301-308. 

EDMUNDS, M., 1982, Speciation in chromodorid 
nudibranchs in Ghana. Malacologia 22: 515- 
522. 

EDMUNDS, M. & KRESS, A., 1969, On the Euro- 
pean species of Eubranchus (Mollusca Opistho- 
branchia). Journal of the Marine Biological Asso- 
ciation of the United Kingdom, 49: 879-912. 

HENNIG, W., 1966, Phylogenetic Systematics. Uni- 
versity of Illinois Press, Urbana. 

JONES, J. S., LEITH, B. H. & RAWLINGS, P., 
1 977, Polymorphism in Cepaea: a problem with 
too many solutions? Annual Reviews of Ecology 
and Systematics, 8: 109-143. 

RAFFAELLI, D., 1982, Recent ecological research 
on some European species of Liftorina. Journal 
of Molluscan Studies, 48: 342-354. 

RIDLEY, M., 1986, Evolution and Classification. 
Longman, London & New York, 201 pp. 

RUDMAN, W. В., 1 981 , Further studies on the anat- 
omy and ecology of opisthobranch molluscs 
feeding on the scleractinian coral Pontes. Zoo- 
logical Journal of the Linnean Society, 71 : 343- 
412. 

RUDMAN, W. В., 1984, The Chromodorididae 
(Opisthobranchia: Mollusca) of the Indo-West 
Pacific: a review of the genera. Zoological Jour- 
nal of the Linnean Society, 81 : 1 15-273. 



MALACOLOGIA, 1991, 32(2): 209 

COMPARISON OF ALIMENTARY SYSTEMS IN SHELLED AND 

NON-SHELLED SACOGLOSSA ( = ASCOGLOSSA) 

(GASTROPODA: OPISTHOBRANCHIA) 

Kathe R. Jensen 
Zoological Museum, Universitetsparken 15, DK-2100 Copenhagen 0, Denmark 

ABSTRACT 

The Sacoglossa comprise a "complete" evolutionary series from species with a large shell into 
which the animal can withdraw completely, over species with a reduced shell covering only the 
visceral mass, to shell-less ("nudibranchiate") forms some of which have lateral wing-like ex- 
tensions, parapodia, others bearing leaf-like or cylindrical dorsal appendages, cerata. Phylogeny 
has been based on the morphology of the central nervous system and, in part, the reproductive 
system. It is surprising that the organ system most characteristic of the Sacoglossa, the alimen- 
tary system, has never been used when attempting to deduce phyiogenetic relationships within 
the group. The Sacoglossa are all specialized suctorial feeders, and almost all are stenopha- 
gous herbivores. Hence, many anatomical adaptations to a particular food occur in the alimen- 
tary system. However, a number of characters seem to reflect phyiogenetic relationships as well. 
The present study compares the alimentary system of 17 species of Sacoglossa. 

The characters which have phyiogenetic importance are: Pharyngeal pouches (presence/ 
absence, size and shape), mode of attachment of the descending limb of the radula and its 
surrounding ascus-muscle, shape of radular teeth, branching pattern of digestive gland, and 
position of anus. Pharyngeal pouches occur in all genera of shelled Sacoglossa. Pharyngeal 
pouches also occur in all genera of the non-shelled Polybranchiidae ( = Caliphyllidae), in some, 
but not all, species of Bosellia and Costasiella, and in Plakobranchus. Very few species of the 
family Elysiidae have pharyngeal pouches, and they are completely absent in the Stiligehdae 
and Hermaeidae. In these families the descending limb of the radula and its surrounding ascus- 
muscle is only attached to the pharynx anteriorly. In other sacoglossans the ascus-muscle is 
attached throughout its length. 

Three basic types of radular teeth occur in the Sacoglossa: Teeth with a ventral concavity and 
lateral denticles, blade-shaped teeth with or without denticles on the median cutting edge, and 
sabot-shaped teeth with a scoop-like cusp and a dorsal keel over which the preceding tooth fits. 
The first type of tooth resembles the central tooth of some cephalaspideans, hence must be 
plesiomorphic. Blade-shaped teeth occur in some shelled species as well as most non-shelled 
families, hence may have evolved more than once. Sabot-shaped teeth occur only in hermaeids 
and stiligerids. 

From the original solid digestive gland found in the shelled Sacoglossa and in the non-shelled 
Cyerce, forms evolved with two lateral main ducts. In the shelled genera Volvatella and Berthe- 
linia some digestive gland tubules extend into the mantle. In the cerata-bearing genera the 
digestive gland is composed of long, wide main tubules, sending lateral branches into the cerata, 
dorsal body surface and head region. In the Elysiidae, Plakobranchus and Bosellia the digestive 
gland consists of short main ducts sending a dense network of narrow branches throughout the 
body, including the head and parapodia. 

In the shelled Sacoglossa the anus is located in the posterior part of the mantle cavity. In the 
majority of unshelled species the anus is located at the anterior right corner of the pericardium. 
In some polybranchiids the anus is located laterally, below the anterior rows of cerata, and many 
hermaeids and stiligerids have an anal spout on top of the pericardium. 



209 



MALACOLOGIA, 1991, 32(2): 211-217 

RELATIONSHIP BETWEEN RADULAR MORPHOLOGY AND FOOD IN THE 
DORIDINA (MOLLUSCA: NUDIBRANCHIA) 

Riccardo Cattaneo Vietti & Andrea Balduzzi 
Istituto di Zoología deH'Università degli Studi di Genova, Via Baibi 5, I- 16126 Genova, Italy 

ABSTRACT 

The diet of nudibranchs has been the subject of numerous reports and it is well known that 
different suborders select different types of prey. In the case of the Doridina, these are usually 
sponges, bryozoans and ascidians. 

The relationship between morphological characteristics of the prey and the functional mor- 
phology of radular structure has been studied in 28 genera of Doridina. These were put into 12 
groups according to different radular models. These groups were identified by the combination 
of three different characteristics: number of teeth per half-row, shape of the teeth and their 
uniformity in each row. Employing literature data and original reports, it was possible to relate 
these 12 radular models with 136 species of prey, re-grouped under 23 different taxa, using 
correspondence analysis. 

First, the analysis discriminated the broad-radula groups, sponge eaters, from the narrow- and 
very narrow-radula groups, which have a more catholic diet. The narrow radular groups feed 
mainly on softer food (Ctenostomata Carnosa and Cheilostomata Añasca), while the groups with 
a very narrow radula feed more on harder organisms (Cheilostomata Ascophora, Cirripedia and 
solitary ascidians). 

The relationship of tooth shape to diet is less clear, while other anatomical characteristics, 
such as the presence of a suctorial apparatus or a caecate gut, are more clearly related to 
specific foods. 



INTRODUCTION 

The predator-prey relationship of dorida- 
ceans has been studied from a variety of view- 
points. Thompson (1958) demonstrated the 
role of food availability for dorid metamorpho- 
sis, while Miller (1961) and Bouchet & Tardy 
(1976) showed the importance of the prey 
in species distribution. Yoshioka (1982) de- 
scribed one prey's response to prédation while 
Harvell (1984) considered nudibranchs to be 
'prudent predators' of bryozoan colonies. Elvin 
(1976) studied the role of Chemotaxis, while 
several authors (Ros, 1 974; Nybakken & East- 
man, 1 977; Bloom, 1 981 ; Chadwick & Thorpe, 
1981) analysed the interspecific competition 
for food among sympatric species. Ros (1979, 
1980) also gave evidence that doridacean 
stenophagy can be related to their K-selected 
ecological strategy. 

The radula may be a species-specific tool 
adapted to the animal's particular prey. Feed- 
ing regimes for most species are quite narrow 
and the shapes of radulae correspond well to 
food preference (Nybakken & McDonald, 
1981). Todd (1981), in his review of nudi- 
branch ecology, distinguished four trophic 
groups: sponge-grazers (dorids), bryozoan- 



grazers (mainly dorids), hydroid-grazers (ae- 
olids) and 'miscellaneous' groups (including 
representatives of all nudibranch suborders). 
Generally the radula of a dorid which eats 
sponges differs markedly from that of a dorid 
feeding on sessile acorn barnacles or en- 
crusting bryozoans, but difficulties arise when 
trying to find the relationships between radu- 
lar morphology and prey within these two 
groups. 

In this paper an attempt is made to point out 
the relationship between morphological char- 
acters of the prey and functional morpholog- 
ical aspects of radular structure in 28 genera 
of Doridina. 

MATERIALS AND METHODS 

Data available in literature were used to- 
gether with original data collected by one of 
us (R.C.V.). We have used mainly the data 
reported by Miller (1961), Swennen (1961), 
Thompson (1964), McBeth (1971), Clark 
(1975), Ros (1975, 1978), Barletta (1976), 
Bloom (1976), Ryland (1976), McDonald & 
Nybakken (1978, 1980), Behrens (1980), 
Chadwick & Thorpe (1981), Todd (1981), 
Garcia & Bobo (1984), Thompson & Brown 



211 



212 



CATTANEO VIETTI & BALDUZZI 



(1984), Milien (1985) and Milien & Gosliner 
(1985). In all, 28 genera of Dorldina which 
occur in the Mediterranean Sea were used in 
the analysis, together with 136 species of 
prey from all over the world. The dorid genera 
have been divided according to their radular 
structure (Table 1). 

We have discriminated, first of all, broad (B), 
narrow (N) and very narrow (V) radulae, with 
>30, 4-30 and <4 teeth per half-row respec- 
tively. It is known that there are intraspecific 
and ontogenetic radular variations (Bertsch, 
1976; Nybakken & McDonald, 1981) and that 
the number of teeth per row is statistically re- 
lated to the specimen's weight (Bloom & 
Bloom, 1977), but in the genera considered 
here the number of teeth per half-row has a 
quite narrow range of variation. Moreover the 
differences in food between adult and young 
specimens were not taken into consideration. 

Secondly we have discriminated the tooth 
shape, considering separately simple hooked 
teeth (S), complex hooked teeth (C) and other 
teeth (O). In this last category we put all types 
of teeth difficult to catalogue as, for example, 
the long serrated ones of Aldisa, the flat ones 
of Ancula and those of Crimora. According to 
Bloom (1 976) the curvature of the hook seems 
to be important in prédation. In this work it is 
not considered because we have often seen 
variation in the hooks in the same row. 

Finally the similarity (A) or dissimilarity (D) 
in the shape of the teeth in the same row was 
considered. 

The prey data for each radular group of 
genera were first studied qualitatively as pres- 
ence or absence of a particular prey in the 
diet of each radular group. These data were 
then quantified to give a weighted prédation 
value (V) for each prey, on the basis of the 
frequency of reports on diet of each nudi- 
branch species in the literature, according to 
the following formula: 



V,.k = 



where Vi, к = weighted prédation value by 
the k-th radular group on the 
i-th prey; 

N| к = number of bibliographic rec- 
ords of prédation by the k-th 
radular group on the i-th prey; 

Nk = total number of bibliographic 
records of prédation by the k-th 
radular group on the whole prey 
set. 



Both qualitative and weighted data were 
used for statistical analysis by a method of 
factorial analysis (correspondence analysis, 
Benzécri et al., 1973) which allows one to 
evaluate at the same time both the diet differ- 
ences between radular groups and the rela- 
tionship between these groups and different 
kinds of prey. 

Following the first analysis which treated all 
the prey species separately, further analyses 
were carried out after re-grouping these spe- 
cies into 23 different taxa (generally orders) 
(Table 2), to reduce the noise of errors or dis- 
agreements in the prey determinations at a 
specific level. 



RESULTS 

Fig. la gives the results of a correspon- 
dence analysis of the presence/absence of 
particular prey in the diet of each radular 
group, keeping separated the data on all 136 
prey species. It shows the expected discrim- 
ination between dorids with broad radulae (B) 
and a sponge diet, and those with narrow (N) 
or very narrow radulae (V) and a more cath- 
olic diet. This discrimination is very evident 
along the first axis, while along the 2nd, 3rd 
and 4th axes (these last two not represented 
in the figure) only three radular groups (NSA, 
VGA and NOD) are dispersed, all with a very 
specialized diet. 

The variances yielded by the first four axes 
are, however, low (11.5% for each axis): in 
fact the heterogeneity of the data and the al- 
ready mentioned problems of specific deter- 
mination of prey make the significance of this 
analysis doubtful. 

The results of the second group of analy- 
ses, carried out on the prey records put into 
23 systematic groups and utilizing the 
weighted data, confirm many of the preceding 
observations. Three well-defined clusters are 
formed (Fig. 1 b): the В cluster, with broad rad- 
ulae feeding nearly exclusively on demo- 
sponges; the N-V cluster, with narrow radulae 
and a non-sponge diet, and the NSA cluster, 
with narrow radulae feeding on calcareous 
sponges. The two radular groups VGA and 
NOD, which in the first analysis remained dis- 
criminate along different axes, in this case fall 
together in the N-V cluster: in fact, their diet is 
very unusual only at a specific level. It is clear 
that in this second analysis there is no dis- 
crimination between the finer differences of 
radula and diet within both the В cluster and 



RADULAR MORPHOLOGY AND FOOD IN DORIDINA 



213 



TABLE 1 . Diet and anatomy in the Doridina. For details of radular models, see text. For each prey group the 
number of species present in the diet of every radular model is reported in parentheses. The other anatomical 
characters are: A = acaecate; С = caecate; NS = non-suctorian; S = suctorian. In the final columns: 
N = total number of single bibliographic records, and V = weighted value for each prey group (see text for 
derivation). 







Other 


Radular 




anatomical 


models 


Dorid genera 


characters 



Prey groups 
(number of species) 



BSA Doris, Archidoris, 

Discodoris ( = Anisodoris), 
Peltodoris, Jorunna, 
Platydoris, Carminodoris 



BCA Cadlina, Chromodoris, 
Hypselodoris 



BOA 

NSA 
NSD 



NOD 
VSA 

VGA 
VCD 



VOD 



C, A 



BCD Rostanga 



Aldisa 



Aegires 



Polycera, Greilada, Palio, 
Limada, Ttiecacera 



NS 

NS 



NCD Acantliodoris, Adataría 



Crímora 
Polycerella 

Trapania 

Onchidoris, Goniodoris, 
Okenia, Diaphorodoris 



NS 
NS 

S 
S 



Arícala 



Porifera Calcárea (1) 

Porifera Choristida (1) 

Porifera Hadromerida (4) 

Porifera Halichondrida (6) 

Porifera Poecilosclerida (6) 

Porifera Spirophorida (1) 

Porifera Haplosclerida (5) 

Porifera Dictyoceratida (1) 

Bryozoa Cheilostomata Ascophora (3) 

Porifera Homosclerophorida (1) 
Porifera Choristida (1) 
Porifera Halichondrida (3) 
Porifera Poecilosclerida (3) 
Porifera Axinellida (1) 
Porifera Haplosclerida (2) 
Porifera Dictyoceratida (8) 
Porifera Dendroceratida (2) 

Porifera Halichondrida (2) 
Porifera Poecilosclerida (11) 
Porifera Haplosclerida (2) 

Porifera Halichondrida (1) 
Porifera Poecilosclerida (5) 

Porifera Calcárea (3) 

Cnidaria Anthozoa Gorgonacea (1) 
Bryozoa Ctenostomata Stolonifera (3) 
Bryozoa Cheilostomata Añasca (12) 
Bryozoa Cheilostomata Gymnocystidea (1) 
Bryozoa Cheilostomata Ascophora (5) 
Bryozoa Cyclostomata (1) 

Bryozoa Ctenostomata Carnosa (6) 
Bryozoa Cheilostomata Añasca (4) 
Bryozoa Cheilostomata Ascophora (4) 

Bryozoa Cheilostomata Añasca (3) 

Bryozoa Ctenostomata Stolonifera (4) 
Bryozoa Cheilostomata Añasca (1) 

Entoprocta (1) 

Bryozoa Ctenostomata Stolonifera (1) 
Bryozoa Ctenostomata Carnosa (4) 
Bryozoa Cheilostomata Añasca (9) 
Bryozoa Cheilostomata Cribrimorpha (1) 
Bryozoa Cheilostomata Gymnocystidea (2) 
Bryozoa Cheilostomata Ascophora (14) 
Bryozoa Cyclostomata (2) 
Crustacea Cirripedia (4) 
Ascidiacea (solitary ascidians) (6) 
Ascidiacea (colonial ascidians) (5) 

Entoprocta (1) 

Bryozoa Ctenostomata Stolonifera (1) 

Ascidiacea (colonial ascidians) (3) 



1 


.01 


1 


.01 


5 


.06 


34 


.47 


11 


.15 


1 


.01 


15 


.20 


1 


.01 


3 


.04 


1 


.02 


4 


.08 


4 


.08 


4 


.08 


2 


.04 


8 


.17 


15 


.33 


7 


.15 


2 


.12 


12 


.75 


2 


.12 


1 


.12 


7 


.87 


4 


1 


1 


.02 


5 


.14 


21 


.60 


1 


.02 


6 


.17 


1 


.02 


15 


.57 


6 


.23 


5 


.19 


3 


1 


4 


.80 


1 


.20 


1 


1 


2 


.02 


5 


.07 


11 


.16 


1 


.01 


2 


.02 


23 


.34 


2 


.02 


5 


.07 


7 


.10 


9 


.13 


2 


.33 


1 


.16 


3 


.50 



214 



CATTANEO VIETTI & BALDUZZI 



TABLE 2. List of systematic groups of prey which 
have been regrouped for some analyses. 





Number of 




Prey groups species 


1 


Porifera Calcárea 


4 


2 


Porifera Homosclerophorida 


1 


3 


Ponfera Choristida 


2 


4 


Porifera Hadromerida 


4 


5 


Porifera Halichondrida 


6 


6 


Porifera Poecilosclerida 


22 


7 


Porifera Axinellida 


1 


8 


Porifera Spirophorida 


1 


9 


Porifera Haplosclerida 


9 


10 


Porifera Dictyoceratida 


8 


11 


Porifera Dendroceratida 


2 


12 


Cnidaria Gorgonacea 


1 


13 


Entoprocta 


2 


14 


Bryozoa Ctenostomata Stolonifera 


6 


15 


Bryozoa Ctenostomata Carnosa 


7 


16 


Bryozoa Cheilostomata Añasca 


21 


17 


Bryozoa Cheilostomata Cribrimorpha 


1 


18 


Bryozoa Cheilostomata Gymnocystidea 


2 


19 


Bryozoa Cheilostomata Ascophora 


18 


20 


Bryozoa Cyclostomata 


3 


21 


Crustacea Cirripedia 


4 


22 


Tunicata (solitary ascidians) 


6 


23 


Tunicata (colonial ascidians) 


5 


Total number of species 


136 



the N-V cluster. Further analyses were there- 
fore carried out for these groups. 

From the analysis carried out on В groups 
only (Fig. 1c), it is now possible to dischmi- 
nate three of the four broad-radular models 
along the first axis (which is the only very sig- 
nificant one in this analysis). BCA radulae, 
feeding mainly on horny demosponges (Dic- 
tyoceratida and Dendroceratida), form a dis- 
tinct cluster at one extreme of the diagram. 
BCD and BOA radulae, linked to Poeciloscler- 
ida, form a second very discrete cluster at the 
other extreme. BSA radulae, with a more gen- 
eralized diet, form an intermediate group dis- 
persed along the second axis. The prey point 
of ascophoran bryozoans (no. 19) falls in this 
last cluster too, as the supposed diet of Platy- 
dohs argo is the bryozoan Sertella. 

Fig. 1d presents the analysis carried out on 
the radular groups of the non-sponge eating 
genera. Radular groups VGA {Trapania) and 
VOD (Ancula) are clearly separated from the 
main group of genera: both include ento- 
procts in their diet. Trapania has no other 



food, while Ancula also feeds on colonial 
ascidians and bryozoans. VSA and NCD 
groups, both eating principally ctenostoma- 
tous bryozoans, are also well separated, 
showing a good relationship with Stolonifera 
and Carnosa respectively. The remaining 
radular groups form two clusters with just two 
intermediate points. The first cluster (VCD) 
shows a close relationship with ascophoran 
bryozoans, barnacles and solitary ascidians, 
and the second (NOD and NSD) with gorgo- 
naceans and añascan bryozoans. 



DISCUSSION 

The study of the relationship between rad- 
ular morphology and food in the Dohdina still 
has many unresolved problems. Literature re- 
ports sometimes indicate as prey the organ- 
ism on which the nudlbranch was crawling 
when it was collected: this makes some of the 
data unreliable. On the other hand widely dis- 
tributed nudibranchs may have different pref- 
erences in different geographical locations 
(McDonald & Nybakken, 1978): for example 
Goniodoris nodosa feeds on the ascidian 
Dendrodoa grossularia along the Atlantic 
French coast (Bouchet & Tardy, 1976), yet In 
England (Thompson & Brown, 1984) it feeds 
on fleshy ctenostomes. 

Other data appear surprising: for example, 
as mentioned above, Platydoris argo in the 
Mediterranean Sea (Ros & Gill, 1984) feeds 
on the bryozoan Sertella, but has a typical 
sponge-eating radula; Polycera atra grazes 
on the gorgonian Lophogorgia chilensis (Lew- 
bel & Lance, 1 975), but other members of this 
genus feed exclusively on bryozoans. More- 
over, according to Ryland (1 976), somedohda- 
ceans associated with bryozoans are not nec- 
essarily predators: sometimes they can feed 
on bacterial films and detritus. 

The investigation of gut contents can also 
create errors: while grazing on a sponge, a 
nudlbranch can ingest other casual food such 
as polychaete larvae, copepods and algal fil- 
aments (Aboul-Ela, 1959). These prey are 
sometimes given a greater importance than 
the sponge itself. 

It is also surprising that some very common 
species, such as bryozoans of the genera 
Flustra, Chartella and Sertella or the sponge 
Petrosia ficiformis, have only one predator: 
this is probably due to scarcity of data. 

In spite of these problems this study con- 
firms the relationship between radula and 



RADULAR MORPHOLOGY AND FOOD IN DORIDINA 



215 



© VCÄ 


12 RADULAR MODELS 
136 PREY SPECIES 

V, = 11.5% 
V, = 11.5% 


1 i I±L 

NOD BOA BCD BCA BSA 
NSA 


Til П 

VCD NSD NOD VSA 
VOD 



NCDI 
VCDÍ^ 


2-11j:« BCA 
S« BCD 

BSA BOA 




12 RADULAR MODELS 
23 PREY GROUPS 

V, = 18.0% 
Vj = 17.8% 


NSUI \ 

nodK \ 

VSA ..^,2.23 

VOD — 5:«'^ " 

VCA 






NèA 



© 



4 RADULAR MODELS 
12 PREY GROUPS 



V, = 62.8% 
V, = 35.9% 



BCD 
BQÎ 



© 




7 RADULAR MODELS 


•15 


12 PREY GROUPS 




NCD 


V| = 34.0% 
V2 = 23.2% 




17 




19. Лг, 

VCD 






18m20 


,3^ 




NOÍnsd 


VOD 




12 






.1* 
VSA 



FIG. 1. Diagrams to show the results of the ordination models along the 1st (x) and 2nd (y) axes, using 
correspondence analysis: a. on all 12 radular models and 136 prey species; b. on all 12 radular models and 
23 prey groups; с on 4 radular models that feed on 12 prey groups (demosponges and bryozoans); d. on 
7 radular models that feed on 1 2 non-sponge prey groups. In each diagram the variance percentages yielded 
by the axes (v1 and v2 respectively) are reported. Radular model points are indicated by the same abbre- 
viations used in Table 1 (see text for explanation). Prey points are indicated by dots: in diagrams b, с and 
d numbers refer to the prey groups listed in Table 2. 



diet: dorids with broad radulae prey exclu- 
sively on sponges while those with narrow 
and very narrow radulae eat mainly soft (Bry- 
ozoa Añasca and Carnosa) and hard prey 
(Cirripedia, Bryozoa Ascophora, solitary as- 
cidians) respectively. Aegires, grazing on Cal- 
carea (Bertsch, 1980), is unusual among 
sponge-eaters in having a narrow radula, but 
with more than 20 teeth per row it is really 
intermediate between the broad- and narrow- 
radula groups. 



The relationship between tooth shape and 
diet remains unclear: in the most complex 
group (VCD) it is possible to find species 
feeding on barnacles {Onchidoris bilamel- 
lata), on a large variety of bryozoans (O. mu- 
ricata) and on tunicates [Goniodoris castanea 
and Okenia elegans). Only in the sponge- 
eaters can we note that complex hooked 
teeth (C) are used mainly on horny sponges, 
while the simple hooked teeth (S) are used on 
a much wider vahety of sponges. This last 



216 



CATTANEO VIETTI & BALDUZZI 



type of tooth can also be found both in spe- 
cialist species (e.g. Peltodoris atromaculata 
which feeds exclusively on Petrosia ficiformis) 
and in more generalist species such as Archi- 
doris pseudoargus. 

In addition to these correlations between 
radular teeth and diet there are other anatom- 
ical and physiological adaptations to different 
foods. Among the anadoridaceans there are 
two types of feeding behaviour, linked to the 
presence or absence of a suctorial pump con- 
nected to the buccal mass. The non-suctorial 
groups (Polycera and allied genera, Crimora 
and Polycerella) are rasping feeders and 
have a diet based mainly on soft bryozoans, 
while the suctorial ones, such as Onchidoris 
or Goniodohs, are sucking feeders, which 
prey principally on harder organisms such as 
barnacles, solitary tunicates and strongly cal- 
cified bryozoans. 

Within the sponge-eating groups the gen- 
era Chromodons, Hypselodoris and Cadlina 
graze mainly on horny sponges and have a 
caecate gut, while Rostanga and Aldisa feed 
mainly on the Poecilosclehda and have an 
acaecate gut. Probably this difference is 
linked to the different nature of the extra-cel- 
lular collagen matrix in the various demo- 
sponges (Bergquist, 1 978) or to differences in 
the skeletal organization (Bloom, 1981). 



ACKNOWLEDGEMENTS 

We offer grateful thanks to Prof. M. Ed- 
munds (Lancashire Polytechnic, Preston) and 
Prof. M. Sara (University of Genoa) for their 
criticisms and to our colleagues F. Boero, M. 
Pansini and G. Pulitzer-Finali for their useful 
suggestions. 



LITERATURE CITED 

ABOUL-ELA, I.A., 1959, On the food of nudi- 
branchs. Biological Bulletin of the Marine Biolog- 
ical Laboratory, Woods Hole, 117: 439-442. 

BARLETTA, G., 1976, Considerazioni sulla biono- 
mia dei Nudibranchi e sulla loro alimentazione. 
Conchiglie, 12: 117-128. 

BEHRENS, D.W., 1980, Pacific Coast Nudi- 
branchs. A Guide to the Opistobranchs of the 
Northeastern Pacific. Sea Challengers, Los 
Osos, California, 1-112 p. 

BENZÉCRI, J.L. et alii [35 authors], 1973, L'ana- 
lyse des correspondences. L'analyse des don- 
nées 2. Dunod, Paris, 1-619 p. 



BERGQUIST, P.R., 1978, Sponges. Hutchinson, 

London, 1-268 p. 
BERTSCH, H., 1976, Intraspecific and ontogenetic 
radular variation in opisthobranch systematics 
(Mollusca: Gastropoda). Systematic Zoology, 25: 
117-122. 

BERTSCH, H., 1980, The nudibranch Aegires al- 
bopunctatus (Polyceratacea: Aegiretidae) preys 
on Leucilla nuttingi (Porifera: Calcárea). The 
Veliger. 22: 222-224. 

BLOOM, S.A., 1976, Morphological correlations be- 
tween dorid nudibranch predators and sponge 
prey. The Veliger, 18: 289-301. 

BLOOM, S.A., 1981, Specialization and noncom- 
petitive resource partitioning among sponge- 
eating dorid nudibranchs. Oecologia, Berlin, 49: 
305-315. 

BLOOM, S.A. & CF. BLOOM, 1977, Radular vari- 
ation in two species of sponge-rasping dorid 
nudibranchs. Journal of Molluscan Studies, 43: 
296-300. 

BOUCHET, P. & J. TARDY, 1976, Faunistique et 
biogéographie des nudibranches des côtes fran- 
çaises de I Atlantique et de la Manche. Annales 
de l'Institut Océanographique, Paris, 52: 205- 
213. 

CHADWICK, S.R. & J.P. THORPE, 1981 , An inves- 
tigation of some aspects of bryozoan prédation 
by dorid nudibranchs (Mollusca: Opisthobran- 
chia). In: LARWOOD, G. P. & С NIELSEN, eds., 
Recent and Fossil Bryozoa. Olsen & Olsen, Fre- 
densborg, Denmark, p. 51-58. 

CLARK, K.B., 1975, Nudibranch life cycles in the 
Northwest Atlantic and their relationship to the 
ecology of fouling communities. Helgolaender 
wissenschaftliche f^eeresuntersuchungen, 27: 
28-69. 

ELVIN, D.W., 1976, Feeding of a dorid nudibranch, 
Diaulula sandiegensis, on the sponge Haliclona 
permollis. The Veliger, 19: 194-198. 

GARCIA, J. С & A. BOBO, 1984, Una nueva espe- 
cie de Polycera Cuvier (Mollusca: Nudibranchia) 
del litoral ibérico. Cahiers de Biologie Marine, 25: 
361-373. 

HARVELL, D., 1984, Why nudibranchs are partial 
predators: intracolonial variation in bryozoan pal- 
atability. Ecology, 65: 716-724. 

LEWBEL, GS., & J.R. LANCE, 1975, Detached 
epidermal sheaths of Lophogorgia chilensis as a 
food source for Polycera atra (Mollusca: Opistho- 
branchia). The Veliger 17: 346. 

McBETH, J.W., 1971, Studies on the food of nudi- 
branchs. The Veliger 14: 158-161. 

McDonald, G.R. & J.W. NYBAKKEN, 1978, Ad- 
ditional notes on the food of some California 
nudibranchs with a summary of known food hab- 
its of California species. The Veliger, 21: 110- 
119. 

Mcdonald, g.r. & j.w. nybakken, i980. 

Guide to the Nudibranchs of California. American 
Malacologists Inc., Florida, 1-72 p. 
MILLEN, S.V., 1985, The nudibranch genera On- 
chidoris and Diaphorodoris (Mollusca, Opistho- 



RADULAR MORPHOLOGY AND FOOD IN DORIDINA 



217 



branchia) in the Northeastern Pacific. The Veli- 
ger, 28: 80-93. 

MILLEN, S.V. & T.M. GOSLINER, 1985, Four new 
species of dorid nudibranchs belonging to the ge- 
nus Aldisa (Mollusca: Opisthobranchia), with a 
revision of the genus. Zoological Journal of the 
Linnaean Society, 84: 195-233. 

MILLER, M.C., 1961, Distribution and food of the 
nudibranchiate Mollusca of the south of the Isle 
of Man. Journal of Animal Ecology, 30: 95-1 16. 

NYBAKKEN, J. & J. EASTMAN, 1977, Food pref- 
erence, food availability and resource partitioning 
in Triopha maculata and Triopha carpenteri. The 
Veliger, 19:279-289. 

NYBAKKEN, J. & G. McDONALD, 1981, Feeding 
mechanisms of West American nudibranchs 
feeding on Bryozoa, Cnidaria and Ascidiacea, 
with special respect to the radula. Malacologia, 
20: 439-449. 

ROS, J.D., 1974, Competencia i evolució en es- 
pecies veines de gastéropodes marins. Colloquis 
de la Societat Catalana de Biología, 7: 101-121 . 

ROS, J.D., 1975, Opistobranquios (Gastropoda: 
Euthyneura) del litoral ibérico. Investigación 
pesquera, 39: 269-372. 

ROS, J.D., 1978, La alimentación y el sustrato en 
los opistobranquios ibéricos. Oecol. aquat., 3: 
153-166. 

ROS, J.D., 1979, Tipos biológicos en los opisto- 
branquios. Actas 1° Simposio Ibérico de Estudios 
del Benthos l\/larino, San Sebastian, p. 413- 
440. 

ROS, J.D., 1980, Estrategias ecológicas en los 



Opistobranquios. Comunicaciones 1° Congreso 
Nacional de Malacologia, Madrid, 85-93. 

ROS, J.D. & J.M. gilí, 1984, Opisthobranches des 
grottes sous-marines de l'ôle de Majorque (Balé- 
ares). Rapports et Procès-Verbaux des Ré- 
unions. Commission Internationale pour l'Ex- 
ploration Scientifique de la Mer Méditerranée 
Monaco, 29: 141-145. 

RYLAND, J.S., 1976, Physiology and ecology of 
marine bryozoans. Advances in Marine Biology, 
14: 285-443. 

SWENNEN, C, 1961, Data on the distribution, re- 
production and ecology of the nudibranchiate 
molluscs occurring in the Netherlands. Nether- 
lands Journal of Sea Research, 1 : 191-240. 

THOMPSON, Т.Е., 1958, The natural history, em- 
bryology, larval biology and post-larval develop- 
ment of Adatarla próxima (Alder & Hancock). 
Philosophical Transactions of the Royal Society 
of London, Series B, 242: 1-58. 

THOMPSON, Т.Е., 1964, Grazing and the life cy- 
cles of British nudibranchs. In: CRISP, D.J., ed.. 
Grazing in Terrestrial and Marine Environments. 
Blackwell, Oxford, p. 275-297. 

THOMPSON, Т.Е. & G. BROWN, 1984, Biology of 
Opisthobranch Molluscs, volume 2. Ray Society, 
London, 1-229 p. 

TODD, CD., 1 981 , The ecology of nudibranch mol- 
luscs. Oceanography and Marine Biology, An- 
nual Review, 19: 141-234. 

YOSHIOKA, P.M., 1982, Predator-induced poly- 
morphism in the bryozoan Membranipora mem- 
branácea. Journal of Experimental hñarine Biol- 
ogy and Ecology, 61 : 233-242. 



MALACOLOGIA, 1991, 32(2): 219-221 

Cumanotus beaumonti (ELIOT, 1906), A NUDIBRANCH 
ADAPTED FOR LIFE IN A SHALLOW SANDY HABITAT? 

Bernard E. Picton 
Ulster Museum, Belfast BT9 5AB, U.K. 

ABSTRACT 

Many of the anatomical peculiarities of the family Cumanotidae are possibly explained by the 
ecology of the species. New observations on the habitat and diet of the species Cumanotus 
beaumonti suggest that the broad foot, long muscular cerata modified for swimming and unusual 
spawn coil are adaptive features which enable Cumanotus to be successful in a shallow sandy 
environment feeding on the hydroid Corymorpha nutans. 



INTRODUCTION 

The aeolid nudibranch Cumanotus beau- 
monti has been rather enigmatic since its al- 
most simultaneous discovery in southern En- 
gland and Norway during the first decade of 
this century. Recent publications describing 
the British fauna have relied on the original 
descriptions of this species, and little was 
known about its habitat and ecology. The dis- 
covery of populations in Northern Ireland dur- 
ing 1985 and direct observations by SCUBA 
diving have provided new information about 
this attractive aeolid. 

Synonymy 

Coiyphella beaumonti Eliot, 1 906 
Cumanotus laticeps Odhner, 1 907 

Description 

The body is 20-25 mm long in mature 
specimens and the foot is broad, about 8 mm 
wide. There are prominent propodial tenta- 
cles at the front corners of the foot and small 
oral tentacles on the anterior corners of the 
head. Two smooth erect rhinophores are 
placed close together on the top of the head; 
they are shorter than the surrounding cerata 
(Fig. 1). The cerata are long and numerous, 
exceeding three-quarters of the body length; 
the anterior ones arise in front of the rhino- 
phores. The cerata are arranged in rows, 6 
rows of up to 9 cerata arise from the anterior 
liver ducts and 6-7 single rows of up to 8 
cerata arise from the posterior liver. Elongate 
cnidosacs can be seen at the tips of the cer- 
ata. The anal papilla is on the right side of the 



body, between the cerata arising from the an- 
terior and posterior liver. The coloration of the 
body is pellucid white, becoming rosy pink in 
the dorsal and head region. The digestive 
gland is purple in colour. The dorsal surface 
and cerata are speckled with gold-coloured 
pigment, concentrated in the head region. 

The reproductive system consists of a 
coiled ampulla, two bursae, and a coiled vas 
deferens leading to a large penial sheath con- 
taining an extensively coiled penis. There are 
two distinctive rosettes of tubercles tipped 
with tiny chitinous hooks alongside the female 
aperture, as noted by previous authors. 

The radula of a 20 mm preserved specimen 
consisted of 20 rows of teeth of formula 1 — 
1 — 1. The central tooth is horseshoe-shaped, 
with a strong central denticle and 1 1-16 small 
denticles on either side. The lateral teeth also 
have strong main denticles and cutting edges 
of 13-20 small denticles. The radula tapers 
rapidly from the oldest row with central tooth 
50 |xm wide to the youngest row, with central 
tooth 200 fxm wide. 

Biology 

Fifteen specimens were collected in June 
1985 in Church Bay and one from Arkill Bay 
on Rathlin Island, Co Antrim, Northern Ire- 
land. The animals were either crawling on a 
sea-bed of medium sand in 15 m of water or 
were at the tops of the stalks of the hydroid 
Corymorpha nutans M Sars, 1 835, which was 
common on the sand. This hydroid consists of 
a solitary stem, 50-100 mm tall, bearing a 
single large polyp which measures 15-20 
mm across the ring of long, undulating tenta- 
cles. Animals on the Corymorpha stems were 



219 



220 



PICTON 




Fig. 1. Cumanotus beaumonti, dorsal view of a living specimen (length 20 mm). 



in the process of devouring the polyps, and 
were inconspicuous, with their long, flowing 
cerata looking very similar to the tentacles of 
the hydroid. Spawn coils of Cumanotus were 
more numerous than the animals, and con- 
sisted of two to four turns of egg-bearing jelly 
attached to the sand by a long, string-like por- 
tion covered with sand grains. Specimens 
were seen to swim vigorously by moving their 



cerata in a co-ordinated back and forth motion 
when collected or disturbed. 



DISCUSSION 

Two other species of the family Cumano- 
tidae have been described: Cumanotus cue- 
noti Pruvot-Fol, 1936, and Cumanotus fer- 



ECOLOGY OF CUMANOTUS BEAUMONTI 



221 



ла /d/ Thompson & Brown, 1984. Cumanotus 
cuenoti has recently been redescribed by 
Tardy & Gantes (1980) and is a smaller ani- 
mal with no trace of oral tentacles or propodial 
tentacles and only 5-9 denticles on the cen- 
tral and lateral teeth. It has no surface pig- 
mentation apart from a small orange mark on 
the head of some individuals. Cumanotus fer- 
naldi was proposed as a new name for the 
species described as Cumanotus beaumonti 
by Hurst (1957) from the Pacific coast of 
North America. This species is illustrated by 
Thompson (1976), Thompson & Brown 
(1976), Behrens (1980) and McDonald & Ny- 
bakken (1980). It differs from С beaumonti in 
colouration, having white apical bands of pig- 
ment on the cerata, yellowish-brown digestive 
gland and none of the gold speckling of С 
beaumonti. Thompson (1984) reports that the 
radula is also different, with a slender central 
cusp flanked by up to 26 denticles on the cen- 
tral tooth and a short cusp on the lateral tooth 
flanked by 28 denticles. 

All known species of Cumanotidae appear 
to feed on athecate hydroids. The present 
species feeds on Corymorptia nutans as re- 
ported above, C. cuenoti feeds on Ectopleura 
dumortieri and Tubularia according to Tardy & 
Gantes (1980), and Behrens (1980) reports 
that С fernaldi feeds on Tubularia. Cumano- 
tus species have a number of unusual fea- 
tures in which they differ from most other ae- 
olid nudibranchs. Several of these features 
could be adaptations to life on unstable sed- 
imentary sea-beds, feeding on transitional 
populations of hydroids. Tardy & Gantes 
(1980) point out the resemblance between 
the ceratal morphology, broad foot and spawn 
coils in Cumanotus and Cerberilla and sug- 
gest that Cumanotus may be capable of bur- 
rowing. The ability to swim up into the water 
column was reported for С fernaldi and С 
cuenoti, and is shared by the С beaumonti 
populations reported here. Tardy & Gantes 
speculate on the possibility that this enables 
Cumanotus populations to follow medusae of 
their reproducing prey to areas down-current 
where new populations are being established. 
There is some evidence from the present ob- 
servations that this may actually happen. On 



Rathlin Island there was a steady current of 
0.5-1 knot and many more spawn coils than 
seemed possible for the observed population 
of Cumanotus. At sites off Kilkeel, Co Down, 
in May 1984 numerous spawn coils and stalks 
of Corymorpha nutans were seen on the 
muddy sand sea-bed, but no adult animals 
could be found despite extensive searching. 
Is it possible that the animals had exhausted 
their food supply at this site and dispersed en 
masse in search of new pastures? 

ACKNOWLEDGEMENTS 

I would like to thank my diving companions 
Christine Howson and Dave Connor for sup- 
port in the field when these observations were 
made. The work was carried out during a sur- 
vey of the Northern Ireland coastline financed 
by the Conservation Branch of the Depart- 
ment of the Environment (N.I.) and directed 
by David Erwin of the Ulster Museum. Thanks 
are also due to Heather White of the Ulster 
Museum for the illustration. 

LITERATURE CITED 

BEHRENS, D. W., 1980, Pacific Coast Nudi- 
branchs. Sea Challengers, Los Osos, California, 
1-112 p. 

ELIOT, С N. е., 1906, Notes on some British nudi- 
branchs. Journal of the Marine Biological Asso- 
ciation, U.K., 7: 333-382. 

ELIOT, С N. е., 1908, On the genus Cumanotus. 
Journal of the Marine Biological Association, 
U.K., 8: 313. 

Mcdonald, g.r. & j.w. nybakken, i980. 

Guide to the nudibranchs of California. American 
Malacologists Inc., Melbourne, Florida, 1-72 p. 

TARDY, J. & H. GANTES, 1980, Un mollusque 
nudibranche peu connue: Cumanotus cuenoti A. 
Pruvot-Fol, 1948; redescription, biologie. Bulletin 
de la Société zoologique Française, 105: 199- 
207. 

THOMPSON, Т.Е., 1976, Nudibranchs. T. F. H. 
Publications, Inc., New Jersey, 1-96 p. 

THOMPSON, Т.Е. & G. H. BROWN, 1976, British 
opisthobranch molluscs. Synopses of the British 
Fauna (New Series), No. 8, 1-203 p. 

THOMPSON, Т.Е. & G.H. BROWN, 1984, Biology 
of Opisthobranch Molluscs, volume 2. Ray Soci- 
ety, London, 1-229 p. 



MALACOLOGIA, 1991, 32(2): 223-232 

REGRESSIVE SHELL EVOLUTION AMONG OPISTHOBRANCH GASTROPODS 

Mathieu РоиНсекЛ^ Marie-Françoise Voss-Foucart^ & Charles Jeuniaux^ 

ABSTRACT 

Many opisthobranch molluscs have reduced or lost their shells during evolution. This paper 
discusses the microstructural and chemical changes associated with shell regression in opistho- 
branchs. All shells of benthic species examined so far (twelve species belonging to Pyramidel- 
lacea, Cephalaspidea and Anaspidea) show the same features: two thick layers of complex 
crossed-lamellar material below one thin external layer of granular or homogeneous structure. 
The shells of thecosomatous pteropods (six species with a planktonic mode of life) show an 
inner helicoidal structure surmounted by a thin granular layer. An intermediate condition is found 
in the most primitive pteropods {Limacina) with a complex crossed-acicular fabric under the 
granular one. The degree of calcification, chitin content and amino acid composition of both 
acid-soluble and insoluble fractions of shells of 20 species belonging to Cephalaspidea, Pyra- 
midellacea, Anaspidea, Notaspidea, Sacoglossa and Pteropoda are presented. The degree of 
calcification shows a marked tendency to decrease and the chitin content to increase from the 
primitive to the more advanced species. The amino acid content of both protein fractions ap- 
pears relatively stable in all species, and very similar to that of prosobranchs, although there are 
some minor variations. All these features can be correlated with the need for suppleness in very 
thin reduced shells that would othenwise be too brittle. These features are polyphyletic and 
convergent with shells of other molluscs showing the same tendency to reduce the shell (ceph- 
alopods, heteropods, pulmonate slugs, Polyplacophora). 



INTRODUCTION 

The Opisthobranchia are not defined by a 
set of features common to all members of the 
subclass but by certain marked tendencies, 
one of which is the tendency to lose or to 
reduce the shell in the course of evolution. 
Some of them still retain the shell and even 
the operculum (e.g. the Pyramidellidae and 
Acteonidae); in others, the shell has become 
greatly reduced, either remaining external, or 
internal and covered by two folds of the man- 
tle that may fuse dorsally. In the majority of 
Opisthobranchia, the shell disappears at the 
end of larval life. "No doubt some of the early 
evolutionary experiments along these lines 
ended in failure, but many more were partially 
or totally successful and we are fortunate in 
that modern oceans contain about three thou- 
sand species of gastropods which show inter- 
mediate stages in the general trend outlined 
above" (Yonge & Thompson, 1976). 

The goal of this paper is to try to describe 
and understand the process of shell regres- 
sion at a microstructural and biochemical 
level. This 'neo-conchological' approach will 
take into account the principles of Florkin 



(1966), i.e. that the molecular aspects of ad- 
aptation and phylogeny must rely upon rela- 
tionships established following 'classical' 
methods (anatomy, embryology, etc.) in order 
to detect molecular convergences. So we 
base this paper on the approach of Ghiselin 
(1966) in which "the comparative and func- 
tional anatomy of the reproductive system 
throughout the subclass is treated critically to 
provide a sounder basis for phylogenetic 
studies". We will attempt to outline the ways 
the shell has changed along some of the main 
evolutionary lines of the phylogenetic tree 
proposed by Ghiselin (1966) (Fig. 1). The 
phylogeny of thecosomatous pteropods 
adopted here is that of Rampai (1973) (Fig. 
2). 

The authors wish to dedicate this paper to 
the memory of Professor С M. Yonge in rec- 
ognition of his long-standing interest in, and 
contribution to, malacology. 



MATERIALS AND METHODS 

All molluscs studied were collected alive. 
After removal of the soft parts, the shells were 



■'Senior Research Assistant of the National Fund for Scientific Research of Belgium (FNRS) 

^Departnnent of Morphology, Systematics and Anirnal Ecology, Zoological Institute, State University of Liège, 22 quai Van 

Beneden, В-4020 Liège, Belgiunn 



223 



224 



PLEUROBRANCHIDAE 



POULICEK, VOSS-FOUCART & JEUNIAUX 
UMBRACULIDAE 



PHILINIDAE 



ATYIDAE 




RETUSIDAE 



HYDATINIDAE 



PYRAMIDELLIDAE 



FIG. 1. Phylogeny of opisthobranch gastropods (simplified from Ghiselin, 1966). 



cleaned from the periostracal layers with a 
rotatory metal brush, washed in distilled water 
and either preserved dried (for SEM) or in 
70% ethanol (for biochemical analyses). 

Scanning electron microscopy (SEM) 

Pieces of shell were cut from the last whorl, 
some distance away from the aperture and 
either broken under the dissecting micro- 
scope (fracture surfaces) or cut perpendicu- 
larly to the growth axis, polished and etched 
with 0.05N HCl. The material was fixed in 4% 
glutaraldehyde in 0.2M cacodylate buffer pH 



7.4 (2 h), washed and postfixed in OSO4 (2%) 
in the same buffer (2 h). After dehydration 
through graded ethanol series, the dried ma- 
tehal was orientated and mounted onto Al- 
stubs with silver or nickel print, coated with 
10nm Au-Pd in a cool-diode sputter coater 
(Balzers SCD 030). The matehal was ob- 
served with a Cambridge Scientific Instru- 
ments Stereoscan or a Siemens ETEC Auto- 
scan electron microscope operated at 20 kV. 

Biochemical analyses 

The shell material was ground and decalci- 
fied with 0.5N HCl at ambient temperature. 



REGRESSIVE SHELL EVOLUTION 225 

CAVOLINIA 



STYIOLA 



LIMACINA 




DIACRIA 



FIG. 2. Phylogeny of thecosomatous pteropods (redrawn from Rampal, 1973). 



The organic rernains were centrifuged 
(18,000 rpm), washed and dried to constant 
weight. The supernatants were dialysed 
against H20 (min. 1/360 V/V) and the acid- 
soluble material recovered by evaporation in 
vacuum in a rotatory evaporator. For amino 
acid determinations, the material was hydrol- 
ysed with 6N HOI for 24 h at 1 05°C in vacuum. 
The amino acid patterns were analysed by 
automatic ion exchange chromatography 
(single column procedure according to Deve- 
nyi (1971)). The method gives reproducible 
results of better than 1% at the 10-8M level 
for identical runs (Ghiselin et al., 1967). Chitin 
was estimated by the specific enzymatic 
method of Jeuniaux (1965). Its accuracy is 
better than 4% for chitin weights above 1 5 |xg. 



RESULTS 
Scanning electron microscopy 

Fractures and etched polished surfaces of 
shells of 18 species belonging to Cephalaspi- 
dea (9), Pyramidellacea (1), Anaspidea (2) 
and Pteropoda (6) were examined with SEM. 

The shells of all the twelve benthic species 
examined {Pyramidella terebelloides (A. Ad- 
ams), Acteon tornatilis (L.), Hydatina physis 
(L.), H. zonata (Gmelin), Bulla ampulla L., B. 



punctulata A. Adams, Scaphander lignarius 
(L.), Atys cylindricum Hinds, Haminoea hyda- 
tis (L.), Philine aperta (L.), Aplysia depilans 
Gmelin, A. punctata Cuvier) are aragonitic in 
nature and mainly of crossed-lamellar fabric, 
typical of more advanced gastropods. The 
shells of Pyramidellacea and Cephalaspidea 
are three-layered, with two thick layers of 
crossed-lamellar structure under a thin layer 
of homogeneous or granular material. The 
first order lamellae of the two crossed- 
lamellar layers are perpendicular to each 
other. The internal shells of Anaspidea (Apl- 
ysia spp.) are much less calcified except in 
the apex area where poorly organized cross- 
lamellar elements are obvious (Fig. 3). 

The shells of the six pelagic species exam- 
ined so far (Limacina inflate (Orbigny), Cre- 
seis acicula Rang, Hyalocylis striata (Rang), 
Euclio pyramidata (L.), Diacria trispinosa 
(Lesueur), Cavolinia longirostris (Lesueur)) 
look quite different. Apart from Limacina in- 
flata, the shells are built following the same 
scheme: under a thin granular layer, there is a 
crossed-acicular layer progressively becom- 
ing helicoidal close to the inner side of the 
shells (Fig. 4). These observations are con- 
sistent with the works of Be et al. (1972) and 
Bändel (1977) on the shells of Cuvierina col- 
umella and Cavolinia tridentata. The shell of 
Limacina inflata occupies an intermediate po- 



226 




FIG. 3. SEM picture of the crossed-lamellar fabric 
at the apex of the shell of Aplysia punctata. Scale: 
10^im. 




FIG. 4. SEM picture of the helicoidal fabric at the 
inner side of the shell of Cavolinia longirostris. 
Scale: 10p.m. 



sition: under a thin granular layer, there are 
two poorly individualized layers of somewhat 
disorganized crossed-acicular fabric. This is 
consistent with the view that the Limacinidae 
are the nnost primitive pteropods (Rampai, 
1973) (Fig. 2). 



POULICEK, VOSS-FOUCART & JEUNIAUX 

Chitin and degree of calcification 



We have previously shown that the degree 
of calcification and the chitin content of mol- 
lusc shells can be discussed in an evolution- 
ary perspective (Goffinet & Jeuniaux, 1979; 
Poulicek, 1982; Poulicek & Kreusch, 1986). 
We have estimated the degree of calcification 
and the chitin content of the shells of 20 spe- 
cies of opisthobranchs, some with external 
shells, others with different degrees of shell 
reduction. The results are given in Table 1 . 

Except in the case of Aplysia, Oxynoe and 
the Notaspidea, the organic content of the 
shells is low (< 1 % of the dry calcified weight) 
as previously shown for other gastropod 
shells with crossed-lamellar fabric (Poulicek, 
1982). The high organic content of Retusa Is 
probably due to contamination since it is par- 
ticularly difficult to get rid of the perlostracum 
in such tiny shells. Nevertheless there Is a 
slight tendency for organic content to In- 
crease with shell regression, particularly In 
the internal shells. 

The chitin content data display the same 
features: low level (<1% of the Insoluble or- 
ganic matter) In the external shells and higher 
levels of the chitin content in species with 
smaller internal shells. This tendency is par- 
ticularly obvious in the internal shells within 
each of the taxa Cephalaspidea, Anaspidea 
and Notaspidea: the chitin content Is always 
>1% of the organic matter, close to 8% In 
Aplysia, Berthella and Berthellina. 

Amino acid composition 

Table 2 shows the amino add composition 
of the proteins of the insoluble fraction of the 
shells of eleven species. The patterns are 
typical of conchlolins' with very high acidic 
amino acid content (Asp and Glu), and much 
Gly, Ala, Ser and Leu. The sum of these last 
six amino adds constitutes more than 60% of 
the total amounts of the residues in the Insol- 
uble protein fraction of the shells. There is 
little Cys in most species and no OH-Pro (ex- 
cept in the case of Akera and Oxynoe (De- 
gens et al., 1967)). 

The variability between the different spe- 
cies is low, probably because they are all 
fairly closely related. The primitive Pyramldel- 
lacea and Cephalaspidea are characterized 
by high levels of Asp, whereas the shells of 
the notaspidean Umbraculum appear some- 
what peculiar with very high Pro content, and 



REGRESSIVE SHELL EVOLUTION 



227 



TABLE 1 . Organic matter and chitin content in the shells of 20 species of Opisthobranchia. 

Calcified Decalcified Organic Weight of Chitin as % 





Type of 


weight 


weight 


content 


chitin 


of organic 


Opisthobranch species 


shell 


(mg) 


(mg) 


(%) 


(^Jьg) 


matter 


PYRAMIDELLACEA 














Pyramidella terebelloides 


external 


3361.1 


12.1 


0.36 


46.33 


0.38 


(A. Adams) 














CEPHALASPIDEA 














Acteon tornatilis (L.) 


external 


2355.2 


17.9 


0.76 


53.09 


0.30 


Hydatina zonata (Lightfoot) 


external 


1658.2 


13.1 


0.79 


21.26 


0.16 


Retusa obtusa (Montagu) 


external 


374.0 


11.4 


3.05 


(4.32) 


(0.04) 


Bulla punctulata A. Adams 


external 


3882.4 


13.2 


0.34 


37.43 


0.28 


Scaphander lignarius (L.) 


external 


1955.9 


11.9 


0.61 


14.57 


0.12 


Atys cy lind he игл Hinds 


external 


2533.3 


11.4 


0.45 


41.16 


0.36 


Haminoea navícula (da Costa) 


external 


1200.0 


10.8 


0.90 


51.19 


0.47 


Philine aperta (L.) 


internal 


493.2 


3.2 


0.65 


42.80 


1.34 


ANASPIDEA 


-r*" 












Akera bullata Müller 


external 


352.9 


2.4 


0.68 


17.52 


0.73 


Aplysia punctata Cuvier 


internal 


796.4 


121.4 


15.24 


10234.02 


8.43 


A. depilans Gmelin{^) 


internal 


224.4 


32.7 


14.57 


2346.00 


7.17 


Dolabella auhculaha (Lightfoot) 


internal 


8386.2 


9.6 


0.11 


369.89 


3.85 


NOTASPIDEA 














Umbraculum mediterraneum 


external 


4583.6 


71.9 


1.57 


582.39 


0.81 


(Lamarck) 














Tylodina citrina Joannis 


external 


1552.5 


29.2 


1.88 


289.08 


0.99 


Berthella plumula (Montagu) 


internal 


1072.7 


125.8 


11.73 


9497.90 


7.55 


Berthellina citrina Rüppell & 


internal 


361.0 


38.0 


10.52 


3005.82 


7.91 


Leuckart 














SACOGLOSSA 














Oxynoe olivácea Rafinesque 


external 


325.1 


5.7 


1.75 


41.04 


0.72 


THECOSOMATA 














Cavolinia longirostris (Lesueur) 


external 


6307.7 


16.4 


0.25 


56.50 


0.34 


С tridentata (Niebuhr) 


external 


1131.8 


5.3 


0.47 


26.07 


0.49 



much Leu whereas Iso and Val are less abun- 
dant than in other shells. 

As far as amino acid composition of shell 
insoluble proteins Is concerned, the evolu- 
tionary trends are not obvious. The ratio Gly/ 
Ala appears to Increase in the course of ev- 
olution: it Is 0.95 and 0.93 respectively in the 
shells of Pyramidella and Acteon, it reaches 
1.31 in Philine and Aplysia and even 1.57 in 
Umbraculum. At the same time, the Leu and 
Iso content diminishes slightly with the reduc- 
tion of the shell, except in Umbraculum. 

Table 3 shows the amino acid composition 
of the proteins of the acid-soluble fraction of 
the shells of nine species. Once again, the 
typical molluscan shell pattern of soluble pro- 
teins is well shown: the sum of Asp, Gly, Ser 
and Glu residues amounts to 60 to 70% of the 



total amino acid residues. There is much less 
basic amino acid (Arg, Leu and His) than in 
the insoluble fraction. In opisthobranch shells, 
the relatively low concentration of Asp is com- 
pensated for by higher levels of Glu and very 
high amounts of Ser, which is much more 
abundant than in the acid-soluble fractions of 
prosobranch shells. 

The variability in the soluble proteins be- 
tween the different species is higher than in 
the case of the insoluble fraction. This vari- 
ability in composition may be related to its 
heterogeneity (Weiner et al., 1977; Samata et 
al., 1980; Poulicek, 1982; Poulicek et al., 
1986), since it is composed of macromolecu- 
lar assemblages of several proteins, peptides 
and glycoproteins which differ from species to 
species (Krampitz, pers. comm.). This vari- 



228 



POULICEK, VOSS-FOUCART & JEUNIAUX 



О 



о 



о О) 

со о 

о о 

Q.T- 

Е à 

о 3 

о -о 



со 

03 






3 
(0 

> 


о 


Ю со -^ 1Л C\J со со '— ;Г^^;г-0 — ;=Г'— — ;ГГ 

^ 1 ^;i^c\j^(>jcDco'3-!p:cN'íTi-^in 


Е 


ю 


1 со 1 1 1 1 1 1 1 1 1 1 1 1 1 I 1 

OiCXDCOCNJCOOiCNTl-TtcpO-'-^Ln-^r^-r^CJ) 






CDC^l¿lhc0¿¿OCÓOC4J^¿O^^OO 


X 

-Si- 


Z 


TrcDinr^i-cooicnoq'í^-r^T^'ícpcorvj 


с 
пз 

Q) 




■r^-^cxiobcoihoöö'^^cor-^-r^cvJCMcbcM 






■О 




çoo-^incococvj'^h-'5tr^t^crD'*a3'--cr) 
<3)Oícr)'^Lnc\ja)oocMt^O'^ir>'^oocoT- 


ся 


^ 


■r^ÖT^^cvicvJT^ÖT^T^T-T-^cb-^ddcvj 


+1 


11 


+1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 


со 


Z 


cûc^Jr^Lnc7)coюcз)CDC\JCJ)t^coc^Jl-cot 
cor^co-^coo-.-in-.-cooc\j-.-c\jLn(?oc^ 




TtihcjicMLn-r^ddcûT-^'iodT^cocNJdco 




■■- 7- ■.- 




^ 


'í 1Л|-^1Л-^Г^Г^т-Ю'^<3)-'--^1ЛСХ)'а- 
СТ)ОЮ<ОСО-^1ЛСМС7)Г--(ГОСОГ^ОГ^СЗ^-'- 




О 


■r^Lricx)C3:)(D'^r^-r^ir)c\Jcoir)->-^(NC\idd 






■^ ■^ 




_; 


ООГ-0>'>-0'>-00 Г-.+-.-ОГ-О---+00 

r-oooi'ícvjcn.'^.incjicqcDqco 

'í-cood^'crJcoaJcDcooó-'-^'^'cri -^ 

■--■.-■,- + + 




Р 




О 


cззcJ)lЛLncr5CГ>c^Jтfc^J•.-c^Jc^JC\JГ^^■^c^J 
c^jt-^or-C7)Ocr)coor^ocr)cocj)-.-cor^ 




О 


crjiricTJi-cocji-^dcûd'^cDi-cro-.-dT^ 




Е 


^О''-Г^СГ)С0СТ>1--^С0С0СГ)С0Ю'*Г^'^ 
r^ODCO'^crvJCJlOO-'-COOJCOpCD-'-Lr)-^'^ 




э 


-r^ihdd-r^crJíxicvicNJtn^ddT^T^dd 






■г- 1— 1— 1— -^ t— 




5 


CO^Tj-C\JQOOCOOOOO-^C\JOOT<3)(MCO'í 
(ХЗСООСОООГ^-^->-Ю1ЛСОЮ(Х)'г^О)СО'^ 




cNihdcü-^dcxJdr^-r^-^cxDdcücNJdr^ 




< 


T— 1— Ч— 










^ 


TíCO-.-CMOvJ'í-LnLnOíaíCNOCOI^COOCJ) 

incoco-r-oqr^O'^oq-r^csj-r-cû-^o-r-cj 




с/) 


c\ico-^c\JdT-^ddcDT^-^o6-r^cr)c\Jdcó 




< 


T- T— Ч— T- 1— 




со 


СЛСХЗЮГ^'-СОЮ ■^+COCX>+OCO+'^ 

inh-cr)(3)c\j(X)qc)o,a>cD,qoq,co 

Lncoc3Íco-^-r^(TJ cDcrir^^ -^-r-' со 

^ ■,- ■•- + + + 




CL 




^ 






ZZ- 


ocoюlnŒ)C^Jf^col^cDcoю■.-<3)^-c\I^o 
■.-r^CMir)OC\Jco-.-r^->-Qor-c\jqc\iqc4j 




с/) 


^~-'Tfc\Jdcbdr^cMCD^cót^->-^d-^dcr) 




cd 


T- 1—1— 1— 




Q. 


CO-*LnO)CMi-0 C4J+CDCDin'*CD+'* 
C^OOC^J■*COOOJ C7)CT>05CDi-CMq 

incOQ¿cocodo6+uS"'~-^oói-^'3^co'^ 




cd 



Осл 

Z Q 
< < 



1^ C\J O) -^ 00 LO O 
со 1- со h-; r^ Ю C\J 
C¿ CD C75 со Ovj к 00 



со со œ CD CM 00 00 



со 1- Ю CD о -^ CvJ 



CÛ +00С0СОЮСОГ^1- 
C0COI^COCOC0-^C\J 

CD ihcrid-^cüd-^ 

+ 



о 



CLa:criDO>< -I ^ n =51^ ty сося ^ 
сяхш-1а:-|_]сл<шиш>-х>.^2а: 

<l-WÜQ.Ü<>->^í^_il-ü--JX< 






;o 



OD ñ 



cd 



со ф 



■ч: g cj) со 

<■ :е L-" ф 

• о ■ '" 

qj о 
"О с WU 

?; >vCO ел 

"5 о °а о 

•Q ■ ел != 

gj о с 5 



Ф о 



ф о 
о. ел 



■Q 2 



^ =; Í- 

тз 3 Е -о 
È с 2ri 

tu СТЗ ■*" Q. 

(ГА 00 о 

?" ê c^£i 



REGRESSIVE SHELL EVOLUTION 



229 



« 


< 


JZ 


X 


*^ 


Ü 


^ 


z 


t: 


< 


5 


CE 




Dû 


0) 


О 


Q. 


СО 
О 


E 


CL 


о 


CL 



О 


< 




X 


С 

о 


Ü 




Z 


й 


< 


А 


ОС 


"*- 


CÛ 


ш 


о 


хз 


X 


3 


н 


о 


сл 


W 


CL 


тЬ 


о 


ü 




га -5^ 




О) 




.У zj 




с тз 




я от 




О) 0) 




о "" 




о 




Q) о 




х: г: 




■^ 0) 




ч_ 3 




о -о 




со 




5 OJ 




о ïi- 




•й< 




я< 





о 


0) 


го 






Q. 


о 


X 


с 


ш 






Е 


го 


< 


le 



^2 

I- CL 



ü 



С/3 



Осл 

ZQ 

< < 



'- 1- о о C\J со -^ 

+1 +1 +1 +1 +1 +1 +1 



О CD CD c\j Г-- 1Л аз 



C\J о 'í -^ ■■- ■■- о 


o 


+1 +1 +1 +1 +1 +1 +1 


+1 


со Ю t^ СП t^ Ю 1- 

1Л о ср о t^ in о 


со 
со 


ai ló Ö ю in ai ai 

C\J •.- T- 


со 



•^ Tj- 00 ■* CD со t^ 



О) СО 1^ О -^ ю 

CT) со ю c\j m ю со 

(7) ■^ со -^ со CD СГ) 



СОСЛСЗ)00С0О)1-СОСО 
СОГ^С\|(3>-*1П1Г)ЮСЗ^ 

оЫооооооо 

+1 +1 +1 +1 +1 +1 +1 +1 +1 

cococoinc\jcor^c\jco 
cococor^t^i-c\jom 

■^CDC\JCDOC\JC\ÍT^C4J 



_1_ооооооо 

_,_ +1 +1 +1 +1 +1 +1 +1 

"•"Г-^ОСЗЗСрСПОСО 
Т-' со о -г-' ^ c\i 1-^ 



СО 4-CDCJ)CDC3)CJ0O 

LT) I -.-CMCJ^CD-^CVJ 



CJ)C0CnC3)i-'^i— озсо 

I ¡r> -^ a) t^ О) со 1- IT) -^ 



■<- c\j о ■■- 1- 1 h 



inr^r-^C\J0Oi-4- (33+Lr)CDCDi-CJ)LOO 

OajLDCVJCDCDCO I COj^ini-CDCOinoOCD 

C\J ■.- CNJ 



C\J 



CvJ о ■>- 1- C\J о 



о •— о 

CD со CD 
C\i со irí -^ CD CD CT) 



•^ CD C33 ■^ о t^ en 
CO 00 00 00 (^ ■^3" 00 


LO 


CD -^ CO CD CD (D r^ 
C\J 1- CM 


C4J 



■^ 1ЛОС0С01^00 + 

CDTj-T-T-T-r^in, 
CVJT-'cvj^-r^-r-'cvi 



CO LO о CO h- CvJ + 
r^ CO CO C4J CO r^ , 
■•-^ CO Ö T-' ^ -r^ 

+ 



ООЮЮЮЮЮОСП 
COr^œr^OOîCVJCOCTl 
CÓOt-^CÓt-^t-^CÜ-t^^ 



т- о о cvj со о 



со ю со t-^ CD со о 



СТЗ+ОГ^ОСО-.-'^СТ) 
со T-T^CJh-OOCOCD 

■^' CJ'^-r^cJcü-r^cO 

+ 



CLCrcr30>-< _| 
СЛХШ-1СГ-1_1С0<^,„^, — . 

<i-cocDcl(D<>->Síí2_i(-cl1ix< 



шОш>:х>г5^сс 






■о 

со N 

00 -¿ 



а. с 

(Л о 
(Л 

О) — : (С 

у < ^ 

о <«" у 

Q- ns i: 

E ^ w 



230 



POULICEK, VOSS-FOUCART & JEUNIAUX 



ability in amino acid composition is associ- 
ated with comparable variability in the pro- 
teins, peptides and glycoproteins of the 
different species. The reason why shells 
should be so variable in composition is prob- 
ably because each species' shell is adapted 
to quite specific environmental conditions 
(temperature, salinity, sand or mud particle 
size, etc.), but the precise significance of this 
variation remains unknown or must await fur- 
ther data (Degens & Spencer, 1966; Degens 
et al., 1967; Ghiselin et al., 1967; Meenakshi 
étal., 1971; Grégoire, 1972; Poulicek, 1982). 
The only evolutionary conclusion that can be 
drawn from Table 3 is that a decrease of the 
Lys content of the fraction correlates with a 
reduction in calcification of the shells (Degens 
et al., 1967). 

Some amino sugars other than chitin are 
present in both fractions. The amount of car- 
bohydrate (mainly glucosamine and galac- 
tosamine) varies considerably throughout the 
phylum (Ghiselin et al., 1967), but in most 
mollusc classes the relative proportion of car- 
bohydrate to protein seems to have de- 
creased progressively with the evolution of 
the shells (Poulicek, 1982). This seems to be 
true also in the Cephalaspidea where those 
species with well-developed shells have high 
hexosamine/amino acid ratios and those with 
reduced shells have low ratios (Table 4). The 
primitive shelled Pyramidella also has a high 
ratio whereas the more advanced Thecoso- 
mata have a low ratio. The Notaspidea, 
Anaspidea and Sacoglossa are linked by their 
very high amino sugar content, even if chitin 
is not taken into account, and irrespective of 
whether the shell is external or internal. Table 
4 also shows that the hexosamine/amino acid 
ratio is always higher in the insoluble fraction 
of the organic matrix than in the acid-soluble 
one. 



DISCUSSION AND CONCLUSIONS 

The Euthyneura are characterized by her- 
maphroditism, a tendency to lose the effects 
of torsion, a distinctive type of spermatozoon, 
a peculiar structure of the palliai complex and 
a heterostrophic larval shell (Ghiselin et al., 
1967). But the most striking external feature 
of the whole group is a tendency towards 
shell regression, affecting all evolutionary 
lines of the opisthobranchs as well as some 
lines of pulmonates quite independently. 

Despite the regressive features altering the 



TABLE 4. Estimate of the ratio (hexosamines/ 
amino acids) x 100 in the acid-soluble and 
insoluble fractions of the shell organic matrix of 
14 species of Opisthobranchia. Chitin was not 
taken into account in computing the hexosamines 
of the insoluble fraction. 







Insoluble 




Acid- 


fraction 




soluble 


(chitin 


Opisthobranch species 


fraction 


excluded) 


PYRAMIDELLACEA 






Pyramidella terebelloides 


1.86 


4.38 


CEPHALASPIDEA 






Acteon tor natu is 


1.88 


4.20 


Hydatina zonata 


1.44 


— 


H. physis (1) 


— 


3.73 


Bulla pu net и lata 


1.38 


3.94 


B. striata (1) 


— 


2.93 


Pliiline aperta 


0.67 


2.81 


ANASPIDEA 






A/cera soluta (1) 


— 


5.40 


Aplysia punctata 


3.48 


— 


A. willcoxi (1) 


— 


4.55 


NOTASPIDEA 






Umbraculum mediterraneum 


— 


3.99 


SACOGLOSSA 






Oxynoe olivácea 


— 


3.04 


THECOSOMATA 






Cavolinia longirostris 


0.22 


— 


С tridentata 


— 


2.12 



(1) Computed from data in Degens et al., 1967. 

(1) Data from Jeuniaux, 1963; remainder original 

(2) Data from Degens & Spencer, 1966. 

(3) Compiled from various sources, see Poulicek, 1982 



shells macroscopically, there is no fundamen- 
tal reworking of shell structure nor of the 
chemical composition of its organic matrix. 
The crossed-lamellar architecture of all shells 
of benthic opisthobranch species examined is 
typical of gastropods with a similar level of 
complexity (i.e. similar to that of mesogastro- 
pod and neogastropod prosobranchs) (Pou- 
licek, 1982). Even the most altered internal 
shells (as in Aplysia) exhibit the same kind of 
crossed-lamellar fabric. The helicoidal micro- 
structure of thecosomatous pteropod shells is 
presumably adaptive, but can be directly de- 
rived from the crossed-lamellar type via some 
kind of crossed-acicular microstructure. The 
most primitive Thecosomata (Limacina) ac- 
tually exhibit such intermediate microstruc- 
tural features. A helicoidal fabric of the same 



REGRESSIVE SHELL EVOLUTION 



231 



type ¡s also found In shells of the phylogenet- 
ically unrelated heteropods (prosobranchs 
with a similar planktonic mode of life) (Batten 
& Dumont, 1976). 

The organic matrix isolated from the shells 
is composed of an insoluble chitin-protein 
complex and an acid-soluble glycoprotein 
fraction whose amino acid patterns are typical 
of 'conchiolins' isolated from shells of 
crossed-lamellar fabric, whatever the origin of 
the shell. In both prosobranchs and opistho- 
branchs it appears that as one goes from 
primitive to more advanced species, the hex- 
osamine content decreases, the Gly/Ala ratio 
increases, and the Lys content, which is 
linked to the degree of calcification, de- 
creases. However, the covahant groups of 
amino acids described by Degens et al., 
(1 967) are not found here, thus confirming the 
close relationship of the species. The chitin 
content and degree of calcification have been 
shown to be linked in the evolution of mollusc 
shells (Poulicek, 1982; Poulicek et al., 1986). 
This relationship is confirmed here: as 
opisthobranch shells get smaller so their 
chitin content increases and the degree of 
calcification decreases. Prosobranchs, by 
contrast, show a tendency to develop lower 
chitin content and higher levels of calcifica- 
tion. 

This variation in chemical composition of 
shells must have some adaptive (functional) 
significance, and where parallel changes in 
composition occur in unrelated groups it is 
reasonable to seek for similar causes. In most 
molluscs, the essential functions of the or- 
ganic matrix of the shells (contributing to its 
strength) are carbonate nucleation, shell min- 
eralization and maintenance of shell integrity 
(Degens et al., 1967; Poulicek, 1982; Pou- 
licek & Voss-Foucart, 1984; Poulicek et al., 
1986). While shells may vary in the organic 
matrix, these key features are retained. In 
opisthobranchs, however, the shells become 
thinner and a further decrease in organic con- 
tent would cause them to become brittle. This 
brittleness has been avoided by opistho- 
branchs in three different ways: 

1 . Incorporation of OH-Pro in the insoluble 
protein matrices of Anaspidea and Sa- 
coglossa. This probably makes the shell 
more flexible (Ghiselin et al., 1 967), par- 
ticularly as these shells generally have a 
high protein content. 

2. Increase of the chitin content of the 
shells in Cephalaspidea, Anaspidea and 



Notaspidea. Chitin can be considered to 
form a skeleton' of the organic matrix 
onto which intercrystalline carrier pro- 
teins are polymerized (Poulicek et al., 
1986). An increase of the chitin content 
could thus thicken the wall between 
crystallites and thus provide some sup- 
pleness to the structure. An increased 
chitin content also occurs in other unre- 
lated species with internal reduced 
shells (Cephalopoda, Polyplacophora, 
Prosobranchia, Pulmonata) (Poulicek, 
1982; Poulicek & Kreusch, 1986; Pou- 
licek et al., 1986). 
3. Development of a peculiar helicoidal mi- 
crostructure in thecosomatous ptero- 
pods. The mechanical characteristics of 
these very light shells allow flexibility 
and reduce their brittleness. A similar 
microstructure has evolved in the unre- 
lated heteropods with similar light shells 
and similar mode of life (Batten & Du- 
mont, 1976). 

Thus the main features characteristic of the 
regressive evolution of opisthobranch shells 
can be considered to be adaptive and corre- 
lated with the need for suppleness in very 
thin, calcified shells that otherwise would be 
too brittle. These features are polyphyletic 
and convergent with shells of other molluscs 
showing similar reduction of the shell or the 
same mode of life. 



ACKNOWLEDGEMENTS 

The authors wish to express their thanks to 
Prof. M. Edmunds for correcting the English 
and critical reading of the manuscript. Most 
analyses were conducted with the technical 
assistance of Miss Claudine Toussaint. This 
work was supported by Research Grant no. 
1.5.622.79 F and FRFC Grant no. 2.4506.83 
of the National Fund for Scientific Research of 
Belgium (FNRS). 



LITERATURE CITED 

BÄNDEL, К., 1977, Die Herausbildung der 
Schraubenschicht der Pteropoden. Biomineral- 
ization Research Reports, 9: 73-85. 

BATTEN, R.L. & M.P. DUMONT, 1976, Shell ultra- 
structure of the Atlantidae (Heteropoda, Meso- 
gastropoda) Oxygyrus and Protatlanta, with com- 
ments on Atlanta inclinata. Bulletin of the 



232 



POULICEK. VOSS-FOUCART & JEUNIAUX 



American Museum of Natural History. 1 57: 263- 
310. 

BE, A.W.H., С MacCLINTOCK & D.C. CURRIE, 
1972, Helical shell structure and growth of the 
pteropod Cuvierina columella (Rang) (Mollusca, 
Gastropoda). Biomineralization Research Re- 
ports. 4: 47-79. 

DEGENS, E.T. & D.W. SPENCER, 1966, Data file 
on aminoacid distribution in calcified and uncal- 
cified tissues of shell-forming organisms. Techni- 
cal Report of the Woods Hole Océanographie In- 
stitution No. 66, 27. 120 tables. 1-32 p. 

DEGENS, E.T., D.W. SPENCER & R.H. PARKER, 
1967, Paleobiochemistry of molluscan shell pro- 
teins. Comparative Biochemistry and Physiology. 
20: 553-579. 

DEVENYI, T., 1971, Single column procedure for 
the automatic analysis of aminoacids. Acta Bio- 
chimica Biophysica (Hungarian Academy of Sci- 
ences). 3: 429-432. 

FLORKIN, M., 1966, Aspects moléculaires de l'ad- 
aptation et de la phylogénie. Massen. Paris, 1- 
258p. 

GHISELIN, M. T., 1966, Reproductive function and 
the phylogeny of opisthobranch gastropods. Ma- 
lacologia. 3: 327-378. 

GHISELIN, M.T., E.T. DEGENS, D.W. SPENCER & 
R.H. PARKER, 1967, A phylogenetic survey of 
molluscan shell matrix proteins. Breviora. 262: 
1-35. 

GOFFINET, G. & С JEUNIAUX, 1979, Distnbution 
et importance quantitative de la chitine dans les 
coquilles de mollusques. Cahiers de Biologie Ma- 
nne. 20: 341-349. 

GREGOIRE, C, 1972, Structure of the molluscan 
Shell. In: FLORKIN, M. & B. T. SCHEER, eds., 
Chemical Zoology, volume 7. Academic Press, 
New York & London, p. 45-102. 

JEUNIAUX, C, 1963, Chitine et chitinolyse, un 
chapitre de la biologie moléculaire. Massen, 
Paris, 1-183 p. 

JEUNIAUX, C, 1965, Chitine et phylogénie: appli- 
cation d'une méthode enzymatique de dosagede 
la chitine. Bulletin de la Société de Chimie Bi- 
ologique, 45: 2267-2278. 



MEENAKSHI, V.R., P.E. HARE, & K.M. WILBUR, 
1971, Aminoacids of the organic matrix of neo- 
gastropod shells. Comparative Biochemistry and 
Physiology. 40B: 1037-1043. 

POULICEK, M., 1982, Coquilles et autres struc- 
tures squelettiques des mollusques — composi- 
tion chimique, biomasse et biodégradation en mi- 
lieu marin. Unpubl. PhD Thesis, University of 
Liege. 1-180 p. 

POULICEK, M. & В. KREUSCH, 1986, Evolution- 
ary trends in skeletal structures of Polyplacoph- 
ora. Proceedings of the Vlllth international mal- 
acological congress (Budapest. 1983). p. 207- 
212. 

POULICEK, M. & M.F. VOSS-FOUCART, 1984, 
Approche fonctionnelle de l'évolution microstruc- 
turale des coquilles de mollusques. Annales de la 
Société Royale Zoologique de Belgique, 115: 
89-90. 

POULICEK, M., M.F. VOSS-FOUCART & C. JEU- 
NIAUX, 1986, Chitinoproteic complexes and min- 
eralization in mollusk skeletal structures. In: 
MUZZARELLI, R. A. A., C. JEUNIAUX & G. 
GOODAY, eds., Chitin in Nature and Technology. 
Plenum, New York, p. 7-12. 

RAMPAL, J., 1973, Phylogénie des ptéropodes 
thécosomes d'après la structure de la coquille et 
la morphologie du manteau. Comptes-rendue 
Hebdomadaires des Séances de l'Académie des 
Sciences de Paris. 277: 1345-1348. 

SAMATA, T., P. SANGUANSRI, C. CAZAUX, M. 
HAMM, J. ENGELS & G. KRAMPITZ, 1980, Bio- 
chemical studies on components of mollusc 
shells. In: OMORI, M. & N. WATABE, eds., The 
Mechanisms of Biomineralization in Animal and 
Plants. Tokai University Press, Tokyo, p. 37- 
48. 

WEINER, S., H.A. LOWENSTAM, & L. HOOD, 
1977, Discrete molecular weight components of 
the organic matrices of mollusk shells. Journal of 
Experimental Marine Biology and Ecology, 30: 
45-51. 

YONGE, CM. & Т.Е. THOMPSON, 1976, Living 
Marine Molluscs. Collins, London, 1-288 p. 



MALACOLOGIA, 1991, 32(2): 233-240 

STUDY OF THE ANATOMY AND HISTOLOGY OF THE MANTLE DERMAL 

FORMATIONS (MDFs) OF CHROMODORIS AND HYPSELODORIS 

(OPISTHOBRANCHIA:CHROMODORIDIDAE) 

José С. García-Gómez\ Antonio Medina^ & Rafael Coveñas^ 

ABSTRACT 

Mantle dermal formations (MDFs) were studied in 12 European species of Chromodons and 
Hypselodoris. In Chromodons the MDFs are small, numerous and irregular in shape, and are 
located in a band around the mantle edge. In Hypselodoris the MDFs are larger, less numerous 
and spherical. They are usually located in the anterior and posterior regions, although the anterior, 
or both anterior and posterior, MDFs may be absent in certain species. Therefore, the presence/ 
absence, location and shape of the MDFs may be of taxonomic importance to separate certain 
European species, in particular of the genus Hypselodoris. The main histological difference 
between the MDFs of Chromodoris and Hypselodoris is the presence of a thick, muscular capsule 
enveloping the MDFs in the latter genus. The strategic location and unpleasant taste of the MDFs 
suggest that they play a defensive role, but they do not appear to open externally. 



INTRODUCTION 

The loss of the shell in the Nudibranchia is 
compensated by the appearance of other de- 
fensive mechanisms which are frequently as- 
sociated with warning coloration. The family 
Chromodorididae includes many colourful 
species and, although recent works (see Dis- 
cussion) show that they possess substances 
of presumed defensive value, the almost uni- 
versal presence of dermal formations located 
in the mantle of the Chromodorididae has re- 
ceived scant attention. Only a few reports deal 
with such structures (Bergh, 1890; Marcus, 
1955; Thompson, 1960, 1972; Edmunds, 
1981; Rudman, 1984). The valuable paper of 
Rudman (1984) thoroughly describes the 
'mantle glands' in chromodorid nudibranchs 
from the Indo-West-Pacific, although from an 
anatomical and taxonomic viewpoint. In the 
present paper we study the anatomy and his- 
tology of the mantle dermal formations (MDFs) 
of 1 2 European species of the nudibranch gen- 
era Chromodoris and Hypselodoris, and com- 
pare them with similar formations in other 
opisthobranchs. 



MATERIALS AND METHODS 

Most of the specimens studied were col- 
lected by SCUBA diving in waters of the 
Straits of Gibraltar. A few specimens were 



collected in the intertidal zone of Cadiz 
(Spain). For the anatomical examination of the 
MDFs the animals were frozen and subse- 
quently fixed and preserved in 4% formalde- 
hyde. 

For the histological study MDFs were re- 
moved from specimens of Chromodoris pur- 
purea, С luteorosea, C. krohni, Hypselodoris 
elegans, H. tricolor and H. cantábrica. They 
were immediately fixed in 2.5% glutaralde- 
hyde in 0.1 M Millonig's buffer (pH 7.3) for 3 h, 
dehydrated through an ascending series of 
alcohols or acetones, and embedded in par- 
affin or Spurr's resin (Spurr, 1969). After fix- 
ation in glutaraldehyde, some MDFs were 
postfixed in 1% osmium tetroxide. Semi-thin 
sections were cut on an LKB III ultramicro- 
tome and stained with toluidine blue. 

Paraffin and semi-thin sections were sub- 
jected to histochemical tests for the demon- 
stration of neutral mucosubstances (Periodic 
Acid-Schiff, PAS), acid mucosubstances (Al- 
dan Blue at pH 2.5, AB) and proteins (Nin- 
hidrin-Schiff, NS) (Pearse, 1968). 

RESULTS 

Anatomy and location of MDFs in 
different species: 

A. Genus Chromodoris 

"Single submarginal row of ramifying man- 
tle glands opening dorsally" (in 'diagnosis' of 
the genus Chromodoris, Rudman, 1984). 



^Laboratorio de Biología, Marina, Facultad de Biología, Universidad de Sevilla; ^Laboratorio de Biología, Facultad de 
Ciencias del tHar, Universidad de Cadiz; ^Departamento de Biología Celular, Facultad de Biología, Universidad de Sala- 
manca. 

233 



234 



GARCIA-GOMEZ, MEDINA & COVEÑAS 



In the European species of Chromodoris 
mantle dermal formations (MDFs) are present 
in all the specimens we have examined. They 
are distributed along the edge of the mantle, 
including the cephalic region in some species 
(Fig. 1A), whereas in others they are absent 
in front of the inter-rhinophoral plane (Fig. 
IB). Their shape and size are quite variable, 
even in animals of the same species. When 
the mantle skin is torn with forceps at the level 
of the MDFs, numerous spherical structures 
(10-40 M-m in diameter) are released. These 
structures (as revealed by histological exam- 
ination) correspond to vacuolar cells. 

In some species, such as C. luteorosea, C. 
luteopunctata and С britoi, the MDFs are 
densely packed, which results in an almost 
uniform disthbution along the periphery of the 
mantle. In others, the MDFs are less dense 
and sometimes quite isolated (e.g. С pur- 
purea, С krohni). In these two species the 
young animals usually have fewer MDFs than 
the adults. 

The MDFs are opaque white, usually 
clearly visible because of the transparency of 
the mantle, and unlike the posterior MDFs in 
Hypselodoris, they hardly distort the edge of 
the mantle. 

C. luteorosea (Rapp, 1827) (5 specimens) 

The MDFs are located along the whole 
edge of the mantle, except in front of the rhi- 
nophores, forming blurred radial bands. Their 
vacuolar cells are 30-40 ^lm in diameter. 

С purpurea (Laurillard, 1831) 
(8 specimens) 

The MDFs are distributed all along the 
edge of the mantle. Even the smallest speci- 
mens (10-13 mm) show MDFs in front of the 
rhinophores . The MDFs are usually rounded 
and the vacuolar cells they contain are nor- 
mally spherical and quite uniform in size (20 
jxm), though they are sometimes egg-shaped 
and measure 30-40 \i.rr\. 

C. krohni (Vérany, 1846) (12 specimens) 

The edge of the mantle is densely packed 
with MDFs though in small specimens (6-7 
mm) they may be absent in front of the rhino- 
phores. The MDFs are usually rounded and 
the vacuolar cells are very small and usually 
spherical (10-20 \xm). 




A 



О 0. 



в 



FIG. 1 . Diagram to show the location of MDFs in A: 
Chromodoris purpurea and С krohni: B: С luteo- 
rosea, С luteopunctata and C. britoi. 

С luteopunctata (Gantes, 1962) 
(4 specimens) 

MDFs are present along the whole edge of 
the mantle, except in front of the rhinophores. 
They form dense white accumulations, which 
are usually long and irregular in shape. Vac- 
uolar cells, however, have not been distin- 
guished. 

C. britoi (Ortea & Pérez, 1983) 
(1 specimen) 

MDFs are found all along the edge of the 
mantle, except in front of the rhinophores. 
They are similar to those in C. luteorosea, 
though the vacuolar cells, also spherical, 
measure 10-20 fxm. 

B. Genus Hypselodoris 

"The mantle glands are single and occur 
along the edge of the mantle opening at the 
edge. Posteriorly the glands are greatly en- 
larged and closely packed and on preserva- 
tion are partly extruded" (in 'diagnosis' of the 
genus Hypselodoris, Rudman, 1984). 

In the European Hypselodoris'^ the location 
of the MDFs varies depending on the species. 
They may be completely absent (Fig. 2E), 
present simultaneously at the rear of the man- 
tle and on both sides of the cephalic region 
(Figs. 2A-C), or confined to the extreme pos- 
terior region of the mantle (Fig. 2D). 

When anterior and posterior MDFs are 
present, the posterior MDFs are always 
larger. However, the smallest posterior MDFs 
may be similar in size to the largest anterior 
MDFs. As a general rule, the posterior MDFs 
increase in size towards the posterior end of 



'The diagnosis' of the mantle glands in Chromodoris and Hypselodoris by Rudman (1984) is based on species from the 
Indo-West Pacific. 



MANTLE DERMAL FORMATIONS OF CHROMODORIDS 



235 



о о I = 




о о 




о о 




о о 




о о 




ABC DE 

FIG. 2. Diagram to show the location of MDFs in species of Hypselodoris. A: H. elegans; B: H. villafranca; 
C: H. bilineata; D: H. cf. tricolor; E: H. cf. messinensis. 



the mantle, causing a deformation of its ven- 
tral surface (Fig. 3). Although the MDFs are 
often close together and may appear to be 
partially fused, each one is a separate, dis- 
crete structure. 

The MDFs are opaque white in colour, and 
easily visible because of the transparency of 
the mantle. The number of MDFs located 
close to each rhinophore varies: for example 
a specimen of H. elegans, 60 mm in length, 
possessed four MDFs on the left and 12 on 
the right. The distribution is more balanced in 
the posterior MDFs. In young and adult ani- 
mals belonging to the same species a similar 
distribution of the MDFs has been observed, 
though the size and number of them tend to 
increase with the size of the animal. Occa- 
sionally, however, it has been observed that 
large specimens have fewer MDFs than 
smaller specimens of the same species. 

H. villafranca (Risso, 1818) (26 specimens) 

Anterior and posterior MDFs are present. In 
the smallest specimens observed (4-5 mm) 
there are one to four MDFs (50 ixm) close to 
each rhinophore, and four (250 |лт) in the 
rear region of the mantle. In the largest spec- 
imens (15-25 mm) there are one to four 
MDFs (300 fxm) on each side of the head and 
four to eight (700 |xm) at the caudal end of the 
mantle (Fig. 2B). 




FIG. 3. Caudal region of Hypselodoris showing the 
position of the posterior MDFs (arrowheads). Note 
the large size of the posterior MDFs which causes 
deformation of the mantle edge. 

H. elegans (Cantraine, 1 835) (8 specimens) 
Anterior and posterior MDFs are present. In 
the smallest specimens examined (60-65 
mm) 2-12 MDFs (450 |хт) are found close to 
each rhinophore, and 7-24 (1300 |xm) in the 
rear region (Fig. 2A). In the largest specimens 
(1 1 0-1 30 mm) there are 1 5-1 8 MDFs (1 000 
|jLm) close to each rhinophore, and 14-20 
(2000 |xm) in the rear region. 



236 



GARCÍA-GÓMEZ, MEDINA & COVEÑAS 



H. cf. ír/co /ол (Cantraine, 1835)^ 
(8 specimens) 

Only posterior MDFs are present. In the 
smallest specimens (10 mm) there are two 
MDFs (500 |j.m), and in the largest ones (14- 
20 mm) 4 (900 цт) (Fig. 2D). 

H. coelestis (Deshayes, 1866) 
(34 specimens) 

Only posterior MDFs are found. In the small- 
est specimens (8-9 mm) 1-5 MDFs (400 |хт) 
are present, while in the largest specimens 
(13-17 mm) there are 2-5 (500 y^vn). 

H. cf. messinensis (Ihering, 1880) 
(7 specimens) 
There are no MDFs (Fig. 2E). 

H. bilineata (Pruvot-Fol, 1953) 
(12 specimens) 

In the smallest specimens (5-1 mm) there 
are no anterior MDFs but two to four (1 80 jxm) 
are present posteriorly. In the largest speci- 
mens (13-20 mm) two or three MDFs (200 
jxm) are present close to each rhinophore, 
and four or five (450 \xm) in the rear region 
(Fig. 2C). 

H. cantábrica (Bouchet & Ortea, 1980) 
(5 specimens) 

Anterior and posterior MDFs are present. In 
the smallest specimens (13-25 mm) there 
are one to three MDFs (300 ixm) close to 
each rhinophore, and three to six (1900 ^xm) 
in the rear region. In the largest specimens 
(40-45 mm) three or four MDFs (600 ixm) are 
located close to each rhinophore, and seven 
(1400 fxm) posteriorly. 

Histology of the MDFs 
A. Genus Chromodons 

For the histological examination of the 
MDFs in Chromodoris, three species have 
been examined; С purpurea, С krohni and 
C. luteorosea. The present description is valid 
for all these species. 

The MDFs in Chromodoris are small and 
irregular structures which are embedded in 
the subepidermal connective tissue. They 
consist of an outer cell layer enclosing an in- 
ner accumulation of vacuolar cells (Fig. 4A). 
The cytoplasm of the cells in the outer layer 
appears to contain neutral mucosubstances, 



since it is strongly stained by the PAS proce- 
dure. Curiously, cells with the same histolog- 
ical and histochemical features are present in 
the epidermis. A thorough microscopic exam- 
ination of the MDFs shows that their outer 
layer is continuous and hence they do not ap- 
pear to discharge into the external medium. 
Consequently, the term gland', which has so 
far been applied to these formations, could 
lead to an erroneous interpretation of their 
functioning, since this terminology suggests 
an active secretion of substances. 

The vacuolar cells show a peripheral cyto- 
plasmic ring surrounding a big central vacuole 
which occupies nearly the whole cell volume. 
The nucleus is displaced towards the periph- 
ery of the cell. The content of the central vac- 
uole is weakly stained by toluidine blue when 
the tissue is post-fixed with osmium tetroxide. 
The histochemical tests used for the demon- 
stration of neutral (PAS) and acid (AB) muco- 
substances, and proteins (NS) gave negative 
results in the vacuolar cells. 

On occasions, the cellular organization in 
the centre of the largest MDFs is lost (Fig. 
4A). When the tissue is post-fixed in osmium 
tetroxide, the central area is weakly stained, 
so it seems likely that it is filled with sub- 
stances from the surrounding vacuolar cells. 

In the subepidermal connective tissue and 
between epidermal cells of the mantle of 
Chromodoris, free vacuolar cells, some of 
which appear to open onto the dorsal surface 
of the mantle, are present (Fig. 4B). 

B. Genus Hypselodoris 

The histology of the MDFs has been stud- 
ied in three species of Hypselodoris: H. ele- 
gans, H. cantábrica and H. tricolor. Since the 
histology of the MDFs in all these species is 
similar, we shall describe in detail the obser- 
vations made on H. elegans, and then draw 
attention to significant differences in the other 
two species. 

The MDFs in H. elegans are spherical 
structures consisting of a thick outer capsule 
which completely surrounds an accumulation 
of vacuolar cells (Fig. 4C). The capsule is 
mainly formed by muscle fibres (Fig. 4D) 
which, as shown by tangential sections, are 
oriented in all directions. In H. cantábrica and 



^he identification of these three species will be discussed in a paper currently in preparation by Ortea, Bouchet and 
García-Gómez. It will show that H. coelestis is a distinct species, and that the other two are hitherto undescribed species 
which resemble H. tricolor and H. messinensis. 



MANTLE DERMAL FORMATIONS OF CHROMODORIDS 



237 







FIG. 4. A: Semi-thin section of MDFs of Chromodohs purpurea fixed in glutaraldehyde and stained with 
toluidine blue. Note the vacuolar cells (VC), the surrounding single cell layer (arrowed), and the central area 
lacking vacuolar cells (asterisk). x190. B: Section of MDFs and mantle epidermis of C. purpurea. Note 
connective tissue (CT), and vacuolar cells in epidermis (arrowheads), one of which opens onto the dorsal 
surface of the mantle (double arrowhead). Fixation and staining as for 1 a (above). x480. C: Semi-thin section 
of a MDF of Hypselodoris elegans fixed in gluteraldehyde, post-fixed in osmium tetroxide, and stained with 
toluidine blue. VC, vacuolar cells; M, muscular capsule; arrowheads, granular cells. x75. D: Portion of MDF 
of H. elegans fixed and stained as for 1С (above). Note vacuolar cells (VC) with contents weakly stained by 
toluidine blue. M, muscular capsule; SL, surrounding cell layer; arrow, granular cell. x480. 



238 



GARCIA-GOMEZ, MEDINA & COVEÑAS 



H. cf. tricolor the capsule is identical in struc- 
ture, but thinner. Beneath the capsule there is 
a continuous cell layer (Fig. 4C,D) which ap- 
pears to be equivalent to the cell layer sur- 
rounding the vacuolar cells in Chromodoris. In 
this layer some nuclei are seen, but no defi- 
nite intercellular limits can be distinguished. 

The vacuolar cells are similar to those de- 
scribed in Chromodoris, but they are more 
closely packed. These cells are always 
clearly identifiable, even in the centre of the 
largest MDFs. In H. elegans and H. tricolor 
some cells containing dense cytoplasmic 
granules (Fig. 4C,D) are located between the 
vacuolar cells. This type of cell is not present 
in H. cantábrica nor in the genus Chromo- 
doris. 

In all the species of Hypselodoris studied, 
free vacuolar cells are located in the subepi- 
dermal connective tissue and in the epider- 
mis. 



DISCUSSION 

The loss of the shell in opisthobranchs rep- 
resents an important defensive disadvantage 
which must be compensated by the acqui- 
sition of other methods of defense. In the 
shell-less opisthobranchs numerous protec- 
tive mechanisms have evolved (Edmunds, 
1966a,b) which vary from defensive behav- 
iour to the presence of protective structures, 
such as glands or spicules, in the skin. 

Many pleurobranchids and dorids secrete 
acid substances (pH 1 or 2) of possible de- 
fensive function when they are molested (Ed- 
munds, 1968; Thompson, 1969,1983; Mar- 
bach & Tsurnamal, 1973). The release of 
these substances has been attributed to epi- 
dermal glandular cells (called 'acid' or 'clear' 
cells) and subepidermal multicellular glands. 
The epidermal glandular cells are prismatic 
and show a clear polarity: their nucleus occu- 
pies the basal region of the cell and the apical 
portion is vacuolar (Thompson, 1969). In con- 
trast, the vacuolar cells of the multicellular 
glands do not show such a polarity, their 
shape is more spherical and most of their cy- 
toplasm appears clear (Edmunds, 1968). The 
multicellular glands of Berthellina (Thompson 
& Colman, 1984), Discodoris (referred as An- 
isodoris by Edmunds) stellifera, D. pusae and 
D. tema (Edmunds, 1968) are similar to the 
MDFs of Hypselodoris in that they are formed 
by an accumulation of vacuolar cells envel- 
oped by a muscular sheath, but both types of 



structures differ in that the multicellular glands 
are scattered all over the mantle, possess a 
central lumen and open onto the dorsal sur- 
face of the mantle. 

When some Hypselodoris (e.g. H. vil- 
lafranca and H. cf. messinensis) are mo- 
lested, the release of an opaque substance 
from the mantle can be observed. This sub- 
stance is probably mixed with mucus and thus 
remains around the body of the animal. This 
phenomenon was previously suggested to 
occur in dorids by Potts (1981). The unpleas- 
ant taste and neutral pH of the substance se- 
creted are similar to those of the MDFs, so 
that it is reasonable to suppose that the con- 
tent of the MDFs and the substance released 
from the mantle are similar in chemical com- 
position. The distasteful substance is proba- 
bly responsible for the fact that predatory fish 
(e.g. some species of Blennius) and opistho- 
branchs (e.g. Pleurobranchaea meckeli) re- 
ject Hypselodoris as food, while other nudi- 
branchs left in the same aquarium are 
immediately devoured (pers. obs.). Since the 
MDFs are internal structures and do not open 
on the surface, they do not appear to be in- 
volved in the discharge of the repulsive sub- 
stance. However, inserted between epider- 
mal cells of Chromodoris and Hypselodoris 
are free vacuolar cells. This cell type could 
thus be responsible for the secretion of the 
repulsive substance. 

Ros (1977) points out that Chromodoris 
and Hypselodoris do not release acid sub- 
stances, but he assumes that they must pro- 
duce repulsive secretions. In this connection, 
recent work has shown the presence in these 
genera of substances which could be ob- 
tained from the diet and utilized as chemical 
defence against potential predators (Hoch- 
lowski & Faulkner, 1981; Hochlowski et al., 
1982; Faulkner & Ghiselin, 1983; Faulkner, 
1984; Okuda & Scheuer, 1985). These stud- 
ies have not demonstrated the precise loca- 
tion of these substances, so we cannot defi- 
nitely conclude that they are present in the 
MDFs. 

The strategic location of the MDFs in the 
cephalic and caudal regions of many 
Hypselodoris species, as well as their un- 
pleasant taste, suggest that they may play a 
defensive role. Such formations could thus 
protect the most important external organs 
(head, rhinophores and gills) from attack by 
other animals. Although the MDFs of Chro- 
modoris are apparently different from those of 
Hypselodoris, the study of both under the mi- 



MANTLE DERMAL FORMATIONS OF CHROMODORIDS 



239 



croscope shows that their cells share similar 
histological features. This does not prove that 
the MDFs perform the same function in both 
genera, but the phylogenetic proximity of 
these genera suggests a similar function for 
the MDFs. 

In Chromodons and Hypselodoris the pre- 
sumed defensive region (the edge of the 
mantle) is associated with a striking coloured 
band which contrasts with the general colour 
of the body. This peripheral band is yellow in 
the adult of all the species, except H. coeles- 
tis in which it is white. The band is also white 
in the young specimens of C. krohni and 
some Hypselodoris. 

The observations noted above (i.e. the re- 
lease of repulsive substances by some Hyp- 
selodoris; the distastefulness of the content of 
the MDFs and their strategic location; the 
contrasting coloration associated with the 
MDFs; the presence of metabolites in numer- 
ous Chromodoris and Hypselodoris species 
which, according to several authors may act 
as deterrent substances) lead us to think that, 
as suggested by Ros (1976)^, in the Euro- 
pean species of Chromodoris and Hypselo- 
doris aposematic circles, corresponding to a 
Müllerian mimicry, occur. 

The 'diagnosis' of the mantle glands ( = 
MDFs) of Chromodoris and Hypselodoris 
given by Rudman (1984) is based on species 
from the Indo-West Pacific. Our observations 
on European species of these genera provide 
additional data which may be used in taxon- 
omy. Thus, in Hypselodoris the MDFs may be 
extruded through the ventral surface of the 
mantle as the animal dies (1); they may be 
present simultaneously on both sides of the 
cephalic region and in the posterior region of 
the mantle (2), only in the posterior region of 
the mantle (3), or completely absent (4). Rud- 
man (1984) also reported that "mantle glands 
appear to be absent around the anterior end", 
which concurs with our own observations. 

The MDFs of Chromodoris and Hypselo- 
doris are isolated and do not open onto the 
surface of the mantle. However, in some pre- 
served specimens of both genera whole 
MDFs appear to be extruded through orifices 
formed in the skin of the mantle. Rudman 
(1984) made the same observation on pre- 
served specimens of Chromodoris and 
Hypselodoris. Unfortunately, in very few 
specimens were we able to investigate the 



possible extrusion of MDFs in living animals, 
and in no case did we find this phenomenon 
to occur. Furthermore, the MDFs were not ex- 
truded when specimens of some Hypselo- 
doris (e.g. H. gracilis, H. bilineata and H. ele- 
gans) were prodded. The extrusion occurs 
only when the MDFs themselves are pressed, 
which also justifies the hypothesis that they 
could be defensive structures. In Hypselo- 
doris extrusion was most often observed in 
the largest MDFs. The MDFs might therefore 
be storage vessels which would slowly accu- 
mulate material and then extrude it when they 
were full. This explanation of their activity 
would imply an excretory rather than a defen- 
sive function. 

Recently, Rudman (1984) discussed the 
phylogeny of the different genera of Chromo- 
dorididae by taking into consideration the dis- 
tribution of the MDFs. The observations of 
Rudman and ourselves show that in Hypselo- 
doris the MDFs may be completely absent (/-/. 
cf. messinensis), or located along the entire 
edge of the mantle (/-/. bennetti), with a range 
of intermediate situations present in other 
species. Following Rudman's hypothesis 
(1984), H. cf. messinensis would represent 
the final stage in the evolutionary loss of 
MDFs in the European species of Hypselo- 
doris. 



ACKNOWLEDGEMENTS 

We thank Prof. M. Edmunds for the helpful 
critical reading of the manuscript. 



LITERATURE CITED 

BERGH, R., 1980, Reports on the results of dredg- 
ing .... in the Gulf of Mexico (1877-78) and in 
the Caribbean Sea (1 879-80), by the U.S. Coast 
Survey Steamer "Blake" .... XXXII. Report on 
the nudibranchs. Bulletin of the Museum of Com- 
parative Zoology, Harvard, 19: 155-181. 

EDMUNDS, M., 1966a, Protective mechanisms in 
the Eolidacea (Mollusca Nudibranchia). Zoologi- 
cal Journal of the Linnaean Society, 46: 27-71 . 

EDMUNDS, M., 1966b, Defensive adaptations of 
Stiliger vanellus Marcus, with a discussion on the 
evolution of nudibranch molluscs. Proceedings 
of the Malacological Society of London, 37: 73- 
81. 



^Ros (1976,1977) referred to species of both Chromodoris and Hypselodoris under the generic name Glossodoris. 



240 



GARCIA-GOMEZ, MEDINA & COVEÑAS 



EDMUNDS, M., 1968, Acid secretion in some spe- 
cies of Doridacea (Mollusca, Nudibranchia). Pro- 
ceedings of the Malacological Society of London, 
38: 121-133. 

EDMUNDS, M., 1981, Opisthobranchiate Mollusca 
from Ghana: Chromodorididae. Zoological Jour- 
nal of the Linnaean Society. 72: 175-201 . 

FAULKNER, D.J., 1984, Marine natural products- 
metabolites of marine invertebrates. Natural 
Products Research, 1: 551-598. 

FAULKNER, D.J. & M.T. GHISELIN, 1983, Chem- 
ical defense and evolutionary ecology of dorid 
nudibranchs and some other opisthobranch gas- 
tropods. Marine Ecology Progress Series, 13: 
295-301. 

HOCHLOWSKI, J.E. & D.J. FAULKNER, 1981, 
Chemical constituents of the nudibranch Chro- 
modoris marislae. Tetrahedron Letters, 22: 271- 
274. 

HOCHLOWSKI, J.E., R.P. WALKER, С IRELAND 
& D.J. FAULKNER, 1982, Metabolites of four 
nudibranchs of the genus Hypselodoris. Journal 
of Organic Chemistry, 47: 88-91. 

MARBACH, A. & M. TSURNAMAL, 1973, On the 
biology of Berthellina citrina (Gastropoda: 
Opisthobranchia) and its defensive acid secre- 
tion. fHarine Biology, Berlin, 21 : 331-339. 

MARCUS, E., 1955, Opisthobranchia from Brazil. 
Boletim da Faculdade de Filosofía, Ciencias e 
Letras da Universidade de Sao Paulo, Zoología, 
20:89-261. 

OKUDA, R.K. & P.J. SCHEUER, 1985, Latrunculin- 
A, ichthyotoxic constituent of the nudibranch 
Chromodohselisabethina. Experientia, 41:1 355- 
1356. 

PEARSE, A.G.E., 1968, Histochemistry: Theoreti- 



cal and Applied, volume 1. 3rd edn. Churchill, 
Edinburgh, 1-759 p. 

POTTS, G.W., 1981, The anatomy of respiratory 
structures in the dorid nudibranchs, Onchidoris 
bilamellata and Archidoris pseudoargus, with de- 
tails of the epidermal glands. Journal of the Ma- 
nne Biological Association, U.K., 61: 959-982. 

ROS, J., 1976, Sistemas de defensa de los opisto- 
branquios. Oecologia Aquatica, 2: 41-77. 

ROS, J., 1977, La defensa en los opistobranquios. 
Investigación y Ciencia, 12: 48-60. 

RUDMAN, W.B., 1984, The Chromodorididae 
(Opisthobranchia: Mollusca) of the Indo-West 
Pacific: a review of the genera. Zoological Jour- 
nal of the Linnaean Society, 81 : 1 15-273. 

SPURR, A.R., 1969, A low-viscosity epoxy resin 
embedding medium for electron microscopy. 
Journal of Ultrastructure Research, 26: 31-43. 

THOMPSON, Т.Е., 1960, Defensive adaptations in 
opisthobranchs. Journal of the Marine Biological 
Association, U.K., 39: 123-134. 

THOMPSON, Т.Е., 1969, Acid secretion in Pacific 
Ocean gastropods. Australian Journal of Zool- 
ogy, 17: 755-764. 

THOMPSON, Т.Е., 1972, Chromodorid nudi- 
branchs from eastern Australia (Gastropoda, 
Opisthobranchia). Journal of Zoology, London, 
166: 391-409. 

THOMPSON, Т.Е., 1983, Detection of epithelial 
acid secretions in marine molluscs: review of 
techniques and new analytical methods. Com- 
parative Biochemistry and Physiology, 74 (A): 
615-621. 

THOMPSON, Т.Е. & J.G. COLMAN, 1984, Histol- 
ogy of acid glands in Pleurobranchomorpha. 
Journal of Molluscan Studies, 50: 65-67. 



MALACOLOGIA, 1991, 32(2): 241-255 

DOES WARNING COLORATION OCCUR IN NUDIBRANCHS? 

Malcolm Edmunds 

Department of Applied Biology, Lancashire Polytechnic, Corporation Street, Preston, 

Lanes PR1 2TQ, UK 

ABSTRACT 

Direct evidence for the occurrence of aposematic (warning) coloration in nudibranch molluscs 
is reviewed and is shown to be inconclusive. Many species are conspicuously coloured and are 
distasteful to possible predators, but there is very little evidence that they are avoided by 
predators because of their colours or that these colours give better protection than would cryptic 
colours. Indirect evidence for aposematism can be obtained by assuming that it occurs, pre- 
dicting the consequences of this assumption, and then testing these predictions. Information to 
test most of these predictions is either not available or is inconclusive, but the strongest support 
for aposematism is the widespread occurrence of Batesian and/or Müllerian mimicry. It is con- 
cluded that warning coloration does occur in nudibranchs but that there is much scope for further 
experimental studies. 



INTRODUCTION 

The occurrence of warning coloration in in- 
sects is well established (Cott, 1940), but the 
possibility of its occurrence in opisthobranch 
molluscs is more controversial (Thompson, 
1960; Harris, 1973; Edmunds, 1974; Todd, 
1981). Nearly one hundred years ago Wal- 
lace (1889), Garstang (1889, 1890), Herdman 
(1890) and Herdman & Clubb (1890) sug- 
gested that many nudibranchs have warning 
colours, and furlher possible examples of 
warningly coloured species are given by 
Hecht (1896), Crossland (1911) and by 
numerous more recent authors (e.g. Harris, 
1973). Warning coloration was first given 
a scientific (Greek) name by Poulton (1890) 
who defined aposematic coloration as: 
". . . an appearance which warns off enemies 
because it denotes something unpleasant or 
dangerous; or which directs the attention of 
an enemy to some specially defended, or 
merely non-vital part; or which warns off other 
individuals of the same species." 

The last two parts of this definition make 
the concept of aposematic coloration very 
wide and today most authors restrict it to the 
first clause. Edmunds (1974), for example, 
gives the following definition: "Animals which 
have dangerous or unpleasant attributes, and 
which advertise this fact by means of charac- 
teristic structures, colours or other signals so 
that some predators avoid attacking them, are 
said to be aposematic, and the phenomenon 
is called aposematism." Despite the fact that 



it is nearly a hundred years since it was sug- 
gested that nudibranchs have warning co- 
lours, there is still considerable uncertainty as 
to whether any nudibranchs are really apose- 
matic. This is due to lack of evidence. This 
paper reviews the current state of knowledge, 
and suggests the type of evidence that needs 
to be sought in order to establish if a nudi- 
branch is aposematic. 

Criteria Necessary for Aposematism 

If the definition given above is accepted 
then in order to demonstrate that a particular 
species is aposematic it is necessary to es- 
tablish that it fulfils four criteria (Edmunds, 
1987): 

1 . it is sufficiently noxious that some pred- 
ators will not eat it; 

2. it is conspicuously coloured, or adver- 
tises itself by means of some other sig- 
nals; 

3. some predators avoid attacking it be- 
cause of its signals; 

4. these conspicuous signals provide bet- 
ter protection to the individual or to its 
genes than would other (e.g. cryptic) 
signals. 

Only if all these criteria are met will there be 
selective advantage to an animal in possess- 
ing warning colours (Edmunds, 1987). If cri- 
terion 4 is not met then an animal would be 
better protected if it were cryptic and apose- 



241 



242 



EDMUNDS 



matic colours could not evolve. Do any nudi- 
branchs fulfil these four criteria? 

Criterion 1 : Are any nudibranchs sufficiently 
noxious that predators will not eat them? 

Many nudibranchs have noxious dermal 
and epidermal glandular secretions and ae- 
olids retain nematocysts from their food and 
store them in their cerata; both have a defen- 
sive function (Thompson, 1960; Edmunds, 
1966). Herdman & Clubb (1890), Crossland 
(1911), Crozier (1916) and Thompson (1 960) 
have all established that some brightly co- 
loured nudibranchs are unpalatable to fish. 
The animals were usually dropped into an 
aquarium or into the sea and were immedi- 
ately attacked by fish as they fell through the 
water column. Almost all the nudibranchs sur- 
vived although they were snapped up and 
spat out several times before reaching the 
substrate after which the fish ignored them. 
Herdman & Clubb (1890) and Thompson 
(1960) also showed that many cryptically co- 
loured nudibranchs were unpalatable to fish. 
Harris (1973) obtained similar results with 
Phestilla melanobranchia Bergh, but he found 
that P. sibogae Bergh ( = P. lugubris Bergh) 
was usually eaten as it fell through the water. 
This was because repeated ingestion and 
spitting out by fish caused it to autotomize all 
of its cerata so that the "naked" nudibranch 
was eventually palatable. These simple ex- 
periments clearly show that many nudi- 
branchs are unpalatable to some species of 
predatory fish. 

Criterion 2: Are any nudibranchs 
conspicuously coloured? 

It has been suggested that the following 
nudibranchs may have warning colours: 
Limada clavigera (Mijiler), Polycera quadrilin- 
eata (Müller), Facelina coronata (Forbes & 
Goodsir) and Eubranchus tricolor Forbes 
from Europe (Hecht, 1896); Chromodoris re- 
ticulata (Pease), C. diardii {Ke\aar\) and Phyl- 
lidia varicosa Lamarck from the Indo-Pacific 
(Crossland, 1911; Harris, 1973); and Triopha 
carpenteri Stearns and Diaulula sandiegensis 
(Cooper) from the north-west Pacific (Harris, 
1973). These are all brightly coloured and so 
fulfil criterion 2. Perusal of recent mono- 
graphs with colour illustrations of nudibranchs 
(Behrens, 1980; Bertsch & Johnson, 1981; 
Schmekel & Portmann, 1982; Thompson & 
Brown, 1984; Willan & Coleman, 1984; Just & 
Edmunds, 1985) indicates that a very large 



number of nudibranchs are brightly coloured 
and conspicuous on unnatural backgrounds. 
Garstang (1890), Hecht (1896) and Thomp- 
son (1960), however, have all pointed out that 
some brightly coloured species are actually 
cryptic in their natural environment. For ex- 
ample Rostanga pulchra McFarland is red but 
usually lives close to the red sponge Ophlita- 
spongia pennata Lambe (Cook, 1962) while 
Catriona gymnota (Couthouy) is also red but 
lives almost exclusively on the red hydroid Tu- 
bularia spp. (Edmunds, 1987). 

It is also well known that as light penetrates 
the sea the red end of the spectrum is ab- 
sorbed much more quickly than the blue 
(Hardy, 1956). In consequence many animals 
that are red actually appear black or brown at 
the depths they normally inhabit. Dr H. 
Bertsch (pers. comm.) informs me that Chro- 
modoris petechialis (Gould) and Hexabran- 
chus sanguineus Rüppell & Leuckart appear 
brown and cryptic at 18 m depth. But why 
should these species be red rather than 
brown or black, since if they were to move into 
well-lit, shallower waters they could be very 
conspicuous? The answer to this question 
may be that it is economically cheaper to 
evolve red pigment, because red carotenoids 
are easily sequestered from the animal's 
food, whereas browns and blacks may have 
to be synthesized de novo. 

Criterion 3: Do any predators avoid 
attacking nudibranchs because of their 
colour (or because of some other signal) ? 

There is evidence that some fish and some 
cephalopods can learn to avoid prey that pro- 
vide aversive stimuli yet still attack other pal- 
atable prey which are of different appearance 
(i.e. have different signals) (summarized in 
Edmunds, 1974), but only very preliminary 
experiments along these lines have been car- 
ried out with nudibranchs as prey. These 
showed that grey mullet (Мид/У labrosus) 
quickly learned to avoid the red aeolid 
Coryphella pellucida (Alder & Hancock). They 
also learned to avoid a papillate model cou- 
pled with an aversive stimulus while continu- 
ing to attack a similar but non-papillate model 
coupled with food (Edmunds, 1974). 

Criterion 4: Do conspicuous colours provide 
better protection than cryptic colours? 

Guilford (1990) has recently reviewed the 
possible ways in which warning colours work. 
For example, they may be more memor- 



WARNING COLORATION IN NUDIBRANCHS 



243 



able than cryptic colours because they are 
conspicuous or because they are unfamiliar; 
they may increase the rate of capture and 
hence of aversive learning; they may reduce 
recognition errors in experienced predators; 
or they may provide frequent reminders of a 
noxious experience. Gittleman & Harvey 
(1980) showed that chicks learn more readily 
to avoid noxious crumbs if these are conspic- 
uous than if they are cryptic, and similar ex- 
periments have been carried out by Schüler & 
Hesse (1985), Sillen-Tullberg (1985) and oth- 
ers (see Guilford, 1990). No comparable ex- 
periments have been undertaken using fish or 
other marine animals as predators. However, 
if some nudibranchs are aposematic to cer- 
tain species of fish, then a knowledge of the 
behaviour of these fish is crucial to our un- 
derstanding of how the aposematism has 
evolved. Since different species of predators 
may respond differently to aposematic prey, 
the evolution of aposematic nudibranchs may 
have followed several routes. 

The direct evidence for aposematism in 
nudibranchs is thus not conclusive. There is 
evidence that some nudibranchs fulfil criteria 
1 and 2, although further field observations 
are also required. Much more experimental 
work needs to be undertaken to see if criteria 
3 and 4, which relate to the behaviour of the 
predator, are applicable. However, it is also 
possible to look for indirect evidence for 
aposematism in nudibranchs. 

Indirect Evidence for Aposematism 

The indirect evidence for aposematism in 
nudibranchs is obtained by assuming that it 
does occur, predicting the consequences of 
this assumption and then testing these pre- 
dictions. 

1 . Evolution of especially noxious qualities 
and of ability to survive attacks 

Because of its bright colours an aposematic 
animal is more likely to be found and attacked 
than is a cryptic animal, and so there will be 
stronger selection pressure on it favouring 
more effective noxious qualities and greater 
resilience to attack. For monarch butterflies 
[Danaus plexippus) this has led to sequester- 
ing cardenolides from their food which are 
emetic to their avian predators, and emesis is 
a particularly effective way of negatively con- 
ditioning a bird (Brower et al., 1968, 1970). 
For nudibranchs I know of no evidence that 



the defences of aposematic species are any 
more effective than those of cryptic ones. 
Many nudibranchs are resilient to attack with 
autotomizable papillae and good powers of 
regeneration, but again I know of no evidence 
that this resilience is more pronounced in 
brightly coloured species. 

2. Evolution of Batesian and 
Müllerian mimicry 

Aposematic animals may suffer injury or 
death while predators learn their colour pat- 
tern, so selection will favour individuals of dif- 
ferent species having the same pattern. Pred- 
ators need then learn only one pattern for 
individuals of both species to be protected, 
and the chances of any one animal being 
killed will be reduced. This is Müllerian mim- 
icry. Selection will also favour the evolution of 
Batesian mimicry, that is of animals with the 
same colour pattern as the aposematic 
model, but which are palatable to those pred- 
ators which avoid the model. 

Mimicry is common among nudibranchs; 
Ros (1976, 1977) lists five mimetic groups 
from the Mediterranean, including blue and 
gold chromodorids and orange and white 
nudibranchs. There are groups of blue and 
gold chromodorids from the Pacific coast of 
North America (Bertsch, 1978a,b,c), and nu- 
merous colour groups of chromodorids have 
been described from the Indo-Pacific by Rud- 
man (summarized by Edmunds, 1987). Each 
mimicry group comprises species belonging 
to more than one genus, so similarity of colour 
is not due simply to recent speciation. Some 
of these mimicry groups are listed in Table 1 . 
Not all mimetic species of a single group oc- 
cur in any one habitat, for example on the reef 
off Tema, Ghana, only Mexichromis tricolor 
and Hypselodoris bilineata have been found 
(Edmunds, 1981). The occurrence of mimicry 
in these species is almost impossible to ex- 
plain unless some of them are aposematic. It 
is not known which species in each group are 
Müllerian and which are Batesian because no 
experiments involving naturally occurring 
predators have been carried out, but all of the 
relevant nudibranchs appear to have glandu- 
lar or nematocyst defences which could make 
them unpalatable. Most are therefore proba- 
bly Mijllerian mimics though some could be 
Batesian, at least towards some predators 
(Edmunds, 1987). There are, however, some 
problems. Eubranchus farrani (Alder & Han- 
cock) is typically orange and white, conspic- 



244 



EDMUNDS 



TABLE 1. Presumed mimetic groups of nudi- 
branchs 

Blue and gold group from Mediterranean and 
east Atlantic (Ros, 1977; Edmunds, 1987) 

Chromodoris krohni (Verany) 
Hypselodoris bilineata (Pruvot-Fol) 
Hypselodoris cantábrica (Bouchet & Ortea) 
Hypselodoris messinensis (ihering) 
Hypselodoris tema (Edmunds) 
Hypselodoris valenciennesi (Cantraine) 
Hypselodoris villaf ranea (Risso) 
Hypselodoris webbi (Orbigny) 
Mexichromis tricolor (Cantraine) 

Orange and white group from Mediterranean and 
northern Europe (Ros, 1977; Edmunds, 1987) 

Ancula gibbosa Risso 
Chromodoris elegantula Philippi 
Crimora papillata Alder & Hancock 
Diaphorodoris papillata Portmann & Sandmeier 
Eubranchus farrani (Alder & Hancock) 
Limada clavigera (Müller) 
Polycera faeroensis Lemche 
Polycera quadrilineata (Müller) 
Trapania maculata Haefelfinger 

Blue chromodorids from western America 
(Bertsch, 1978a,b,c) 

Chromodoris mcfarlandi Cockerell 
Hypselodoris agassizii (Bergh) 
Hypselodoris californiensis (Bergh) 
Hypselodoris ghiselini Bertsch 
Hypselodoris lapislázuli (Bertsch & Ferreira) 
Mexichromis antonii (Bertsch) 
Mexichromis porierae (Cockerell) 
Mexichromis tura (Marcus & Marcus) 

White chromodonds with gold border from 
Indo-Pacific (Rudman, 1985) 

Ardeadoris egretta Rudman 
Cadlina nigrobranchiata Rudman 
Cadlina willani Miller 

Chromodoris aureomarginata Cheeseman 
Glossodoris averni Rudman 
Glossodoris pallida (Rüppell & Leuckart) 
Glossodoris undaurum Rudman 
Hypselodoris ku lomba (Burn) 
Noumea nivalis Baba 
Noumea sudanica Rudman 
Thorunna africana Rudman 
Thorunna furtiva Bergh 



uous, and so presumably aposematic, but 
some populations are polymorphic (Edmunds 
& Kress, 1969). Polymorphism has clear ad- 
vantages for cryptic animals, but it is difficult 



to see any advantage for an aposematic spe- 
cies (Edmunds, 1974, 1987). 

3. Evolution of kin selection 

The individuals who gain from aposema- 
tism can be the ones attacked by the potential 
predator or they can be other individuals in 
the vicinity. Where an aposematic animal is 
always killed during the educational experi- 
ence of the predator, the animals that benefit 
from the aposematism must share genes with 
the individual sachficed: this is kin selection. It 
applies to social Hymenoptera and other gre- 
garious species which live in family groups or 
in colonies of related individuals. By contrast, 
where the aposematic animal survives the ex- 
perience of sampling by a predator, the se- 
lective advantage is gained by this individual 
{individual selection), and there is no neces- 
sary requirement that it should be gregarious. 
The question of whether aposematism can 
only evolve through kin selection or whether it 
can also evolve through individual selection 
has generated a series of papers in recent 
years (e.g. Harvey & Paxton, 1981a, b; 
Harvey et al., 1982; Jarvi et al., 1981 ; Sillen- 
Tullberg & Bryant, 1983; review in Guilford, 
1990). In nudibranchs the defensive glands 
and nematocysts are superficial, some ani- 
mals have survived being taken into the 
mouth and spat out by fish, and most species 
have planktonic larvae so that the probability 
that two individuals that happen to settle on 
the same substrate are genetically related is 
slight. Aposematism is therefore most likely to 
have evolved because of the selective advan- 
tage it gives to the aposematic individuals 
themselves rather than to their kin. Neverthe- 
less, assuming the same level of predator 
sampling, kin selection clearly gives greater 
protection to an individual's genes in terms of 
inclusive fitness (Hamilton, 1984) than indi- 
vidual selection, so we might predict the ex- 
istence of kin selection in aposematic nudi- 
branchs. 

Is there any evidence of kin selection in 
aposematic nudibranchs? Kin selection can 
only occur in aposematic animals if they live 
in groups of related individuals. In nudi- 
branchs it can only occur in species lacking 
planktonic larvae because this will enable the 
young grow up close to the parent. A cryp- 
tic animal, on the other hand, is more likely to 
benefit by having a planktonic larva to ensure 
wide dispersal so that predators will be less 
likely to acquire a searching image for its pat- 



WARNING COLORATION IN NUDIBRANCHS 



245 



tern. I therefore predict a higher incidence of 
non-planktonic development in presumed 
aposematic than in cryptic nudibranchs. 

It should be possible to test this prediction 
since the developmental pattern of more than 
150 species of nudibranch is known. Table 2 
summarizes the results, but there are prob- 
lems with this analysis. First, developmental 
type has adaptive significance in terms of a 
species' life cycle and ecological habit (Todd, 
1983) which may be of much greater impor- 
tance than its possible consequences relating 
to kin selection. Thus the development of two 
species of arminid is known, but since most of 
their life is probably spent burrowing it is un- 
likely that developmental mode is related to 
colouration. The same argument can be ap- 
plied to the burrowing aeolid Cerberilla and 
to intersitial acochlidiaceans. These species 
have been omitted from Table 2. 

Second, while planktotrophic larvae clearly 
have planktonic development and direct de- 
veloping eggs have non-planktonic develop- 
ment, lecithotrophic larvae include some with 
planktonic and others with non-planktonic de- 
velopment. It is possible that for some leci- 
thotrophic species the planktonic stage is 
obligatory while for others it is facultative so 
that the larvae usually settle close to the egg 
ribbon. Such information is rarely published 
because it is trivial in terms of the develop- 
ment although of crucial importance to the 
possible occurrence of kin selection. In Tenel- 
lia fuscata (Gould) the veliger stage may 
never leave the egg or it may swim for up to a 
day (Harris et al., 1980). In Cuthona nana 
(Alder & Hancock) and Tenellia adspersa 
(Nordmann) ( = T. pallida (Alder & Hancock)) 
the occurrence of planktonic or non- 
planktonic larvae varies with population or 
with environmental conditions (Harris et al., 
1975; Rivest, 1976; Roginskaya, 1970; Ey- 
ster, 1979). These three species have also 
been omitted from the analysis. Two other 
species reported by one worker to have 
planktonic larvae and by another to have 
non-planktonic development are Doriopsilla 
miniata (Alder & Hancock) and Cuthona pus- 
tulata (Alder & Hancock) (Shyamasundari & 
Najbuddin, 1976; Thompson, 1975; Rogin- 
skaya, 1962; Gosliner & Milien, 1984). These 
conflicting reports may be explained by vari- 
ation in developmental mode, or they may im- 
ply that the different workers were actually 
studying different species. These two species 
have also been omitted from Table 2, but sim- 
ilar plasticity of developmental mode may oc- 



cur in some of the other species included in 
this table. Moridilla brockii Bergh and Favori- 
nus argentimaculatus Rao have also been 
omitted from the table because although they 
have veliger larvae, these metamorphose a 
few hours after liberation (K.P. Rao, 1965; 
K.V. Rao, 1970); thus it is not clear if the de- 
velopmental mode is effectively planktonic or 
non-plankton. 

Finally, while the colour of these animals is 
known, it is not always easy to decide 
whether they are cryptic or aposematic. In this 
table I have assumed that all chromodorids 
are aposematic while most other dohdaceans 
are cryptic. The decision is even more difficult 
for some aeolids: red Coryphella and Flabel- 
lina spp. could be cryptic on Tubularia or in 
deep water, or they could be conspicuous and 
aposematic; and Aeolidiella and Spurilla spp. 
could be cryptic or mimetic among sea anem- 
ones or they could be conspicuous and 
aposematic. I have left these species with a 
'?' in Table 2. 

What conclusions can be drawn from Table 
2? In the dorids (including chromodorids) 
there are 45 cryptic and 22 conspicuous spe- 
cies with planktonic development compared 
with 8 and 7 with non-planktonic develop- 
ment. These figures are not significant 
(x^ = 0.51). However, if the chromodorids are 
all considered to be conspicuous and are 
compared with the other dohdaceans (most of 
which are cryptic), we get the following: 53 
dorids and 14 chromodorids have planktonic 
development compared with 8 and 7 with 
non-planktonic development. This gives a 
X^ = 3.03 which is still not significant, but is 
close to the 5% level. For aeolids 41 cryptic 
and 9 aposematic species have planktonic 
development compared with 4 cryptic and 
aposematic species without planktonic devel- 
opment. This difference is obviously not sig- 
nificant . Whether we assume that the species 
whose colour is entered with a '?' are cryptic 
or conspicuous makes little difference: the fig- 
ures are still not significantly different. There 
is therefore no evidence from a study of de- 
velopmental mode that kin selection occurs in 
nudibranchs. 

Kin selection should also favour apose- 
matic species with non-planktonic develop- 
ment living longer post-reproductively than 
cryptic species. If a predator learns the colour 
pattern by killing a senile animal, this will re- 
duce the chances of that individual's offspring 
or siblings being taken by a predator, and this 
will increase the chances of the individual's 



246 



EDMUNDS 



TABLE 2. Type of development and colour of nudibranchs. Development is classed as with or without 
planktonic larvae. Colour is assessed as cryptic (C), conspicuous (i.e. aposematic, A), or uncertain (?). 
Only one reference has been given for each species to economize on space. 

WITH PLANKTONIC LARVAE 



Species 



Colour 



Reference 



Doridacea minus Chromodorididae 

Acanthodohs brunnea MacFarland 

Acanthodoris nanaimoensis O'Donoghue 

Acanthodohs pilosa (Müller) 

Adalaha próxima (Alder & Hancock) 

Aegires sublaevis Odhner 

Aegires punctilucens (Orbigny) 

Aldisa coopeh Robilliard & Baba 

Aldisa tara Milien 

Ancula evelinae Marcus 

Ancula gibbosa (Risso) 

Anisodohs prea Marcus & Marcus 

Archidohs montereyensis (Cooper) 

Archidoris odhneh (MacFarland) 

Archidoris pseudoargus (Rapp) 

Asteronotus caespitosus (van Hasselt) 

Crimora papillata Aider & Hancock 

Dendrodohs fumata (Rüppell & Leuckart) 

Dendrodoris krebst (Mörch) 

Diaphorodohs lirulatocauda Milien 

Diaulula sandiegensis (Cooper) 

Discodoris concinna (Aider & Hancock) 

Discodoris erythraeensis Vayssière 

Dohdella obscura Verrill 

Dohdella steinbergae (Lance) 

Doriopsis aurantiaca (Eliot) 

Dohopsis viridis Pease 

Doris ocelligera Berg h 

Goniodohs castanea Aider & Hancock 

Goniodoris nodosa (Montagu) 

Goniodohs sugashimae Baba 

Gymnodoris bicolor (Alder & Hancock) 

Gymnodohs citrina (Berg h) 

Haigerda rubicunda Baba 

Hexabranchus sanguineus Rijppell & Leuckart 

Homoiodohs japónica Bergh 

Jorunna tomentosa (Cuvier) 

Nembrotha limaciformis Eliot 

Okenia ascidicola Morse 

Okenia impexa Marcus 

Onchldohs bilamellata (Linnaeus) 

Onchidons muhcata (Müller) 

Onchidohs neapolitana (Chi aje) 

Peltodohs hummelincki Marcus 

Phyllidia varicosa Lamarck 

Platydoris scabra (Cuvier) 

Polycera quadhlineata (Müller) 

Polycerella emertoni Verrill 

Rostanga pulchra MacFarland 

Sebadohs crosslandi (Eliot) 

Taringa telopia Marcus 

Thordisa filix Pruvot-Fol 

Triopha catalinae (Cooper) 

Thppa areolata (Alder & Hancock) 



С 
С 
С 

с 
с 
с 
с 
с 
с 

А 
С 
С 
С 
С 

с 

А 
С 
С 
С 
С 
С 
С 

с 
с 
с 
с 
с 
с 
с 
с 

А 
А 
С 
С 
С 
С 
А 
С 
С 
С 
С 
С 
С 
А 
С 
А 
С 
С 
С 
С 
С 
А 
С 



Hurst, 1967 

Hurst, 1967 

Thompson, 1967 

Thompson, 1958 

Schmekel & Portmann, 1982 

Thiriot-Quièvreux, 1972 

Milien & Gosliner, 1985 

Milien & Gosliner, 1985 

Eyster, 1980 

Thompson & Brown, 1984 

Eyster, 1980 

McGowan & Pratt, 1954 

Hurst, 1967 

Thompson, 1967 

Gohar & Soliman, 1967e 

Schmekel & Portmann, 1982 

Gohar & Soliman, 1967a 

Bändel, 1976 

Goddard, 1984 

Hurst, 1967 

Gohar & Soliman, 1967f 

Gohar & Aboul-Ela, 1959 

Perron & Turner, 1977 

Bickell & Chia, 1979 

Hamatani, 1961b 

Hamatani, 1961b 

Schmekel & Portmann, 1982 

Schmekel & Portmann, 1982 

Thompson, 1967 

Hamatani, 1961a 

Hamatani, 1960a 

Young, 1967 

Hamatani, 1960b 

Gohar & Soliman, 1963 

Hamatani, 1962 

Thompson, 1967 

Soliman, 1991 

Morse, 1972 

Eyster, 1980 

Thompson, 1967 

Thompson, 1967 

Schmekel & Portmann, 1982 

Bändel, 1976 

Soliman, 1986 

Soliman, 1978 

Thompson, 1967 

Franz & Clark, 1972 

Chia & Koss, 1978 

Soliman, 1980 

Bändel, 1976 

Schmekel & Portmann, 1982 

Hurst, 1967 

Gohar & Soliman, 1967g 



WARNING COLORATION IN NUDIBRANCHS 



247 



TABLE 2. (Continued) 



Species 



Colour 



Reference 



Doridacea СЬгоглоаопс11аае 

Chromodoris africana Eliot 
Chromodoris amoena Chesseman 
Chromodoris annulata Eliot 
Chromodoris clenchi (Russell) 
Chromodoris inornata Pease 
Chromodoris luteopunctata (Gantes) 
Chromodoris perola Marcus 
Chromodoris pulchella (Rüppell & Leuckart) 
Chromodoris tinctoria (Rüppell & Leuckart) 
Glossodoris atromarginata (Cuvier) 
Glossodoris pallida (Rüppell & Leuckart) 
Hypselodoris bilineata Pruvot-Fol 
Hypselodoris elegans (Cantraine) 
Hypselodoris kayae Young 

Aeolidiacea 



Gohar& Aboul-Ela, 1959 

Thompson, 1972a 

Gohar & Aboul-Ela, 1957b 

Bändel, 1976 

Gohar & Soliman, 1967b 

Gantes, 1962 

Bändel, 1976 

Gohar & Aboul-Ela, 1957b 

Gohar & Soliman, 1967c 

Gohar & Aboul-Ela, 1959 

Soliman, 1991 

Gantes, 1962 

Rho, 1888 

Young, 1967 



Aeolidia papulosa (Linnaeus) 

Aeolidiella glauca (Alder & Hancock) 

Aeolidiella mannarensis (Rao & Alagarswami) 

Aeolidiella sanguínea (Norman) 

Antonietta luteorufa Schmekel 

Austraeolis calina (Marcus & Marcus) 

Berghia benteva (Marcus) 

Berghia coerulescens (Laurillard) 

Berghia verrucicornis (Costa) 

Catriona gymnota (Couthouy) 

Coryphella browni Picton 

Coryphella fusca O'Donoghue 

Coryphella gracilis (Alder & Hancock) 

Coryphella lineata (Lovén) 

Coryphella nobilis Verrill 

Coryphella parva Hadfield 

Coryphella pedata (Montagu) 

Coryphella pellucida (Alder & Hancock) 

Coryphella trilineata O'Donoghue 

Coryphella verrucosa (Sars) 

Cratena peregrina (Gmelin) 

Cratena pilata (Gould) 

Cumanotus beaumonti (Eliot) 

Cuthona adyarensis Rao 

Cuthona albocrusta (MacFarland) 

Cuthona albopunctata (Schmekel) 

Cuthona caerulea (Montagu) 

Cuthona futairo Baba 

Cuthona genovae (O'Donoghue) 

Cuthona ilonae (Schmekel) 

Cuthona miniostriata (Schmekel) 

Cuthona ocellata (Schmekel) 

Cuthona ornata Baba 

Cuthona pinnifera (Baba) 

Dicata odhneri Schmekel 

Dondice occidentalis (Engel) 

Dondice paguerensis Brandon & Cutress 

Embletonia pulchra (Alder & Hancock) 

Eubranchus cingulatus (Alder & Hancock) 



Williams, 1980 

Hadfield, 1963a 

Rao & Alagarswami, 1960 

Tardy, 1969a 

Schmekel & Portmann, 1982 

Clark & Goetzfried, 1978 

Eyster, 1980 

Tardy, 1962c 

Tardy, 1962c 

Clark, 1975 

Thompson & Brown, 1984 

Roginskaya, 1969 

Kuzihan, 1979 

Thompson, 1967 

Kuzihan, 1977 

Hadfield, 1963b 

Schmekel & Portmann, 1982 

Kuzihan, 1979 

Bridges & Blake, 1972 

Kuzihan, 1979 

Schmekel & Portmann, 1982 

Eyster, 1980 

Hurst, 1967 

Rao, 1962 

Hurst, 1967 

Schmekel & Portmann, 1982 

Schmekel & Portmann, 1982 

Hamatani, 1960b 

Schmekel & Portmann, 1982 

Schmekel & Portmann, 1982 

Schmekel & Portmann, 1982 

Schmekel, 1966 

Hamatani, 1960b 

Hamatani, 1960b 

Schmekel & Portmann, 1982 

Eyster, 1980 

Brandon & Cutress, 1 985 

Schmekel & Portmann, 1982 

Tardy, 1970 

(continued) 



248 



EDMUNDS 



TABLE 2. (Continued) 



Species 



Colour 



Reference 



Eubranchus doriae (Trinchase) 

Eubranchus exiguus (Alder & Hancock) 

Eubranchus farrani (Alder & Hancock) 

Eubranchus misakiensis Baba 

Eubranchus olivaceus (O'Donoghue) 

Eubranchus pallidus (Alder & Hancock) 

Facelina annulicornis (Chamisso & Eysenhardt) 

Facelina coronata (Forbes & Goodsir) 

Facelina dubia Pruvot-Fol 

Facelina fusca Schmekel 

Favorinus auritulus Marcus 

Favorinus branchialis (Rathke) 

Fiona pinnata (Eschscholtz) 

Flabellina afflnis (Gmeiin) 

Flabellina engeli Marcus 

Hermissenda crassicornis (Eschscholtz) 

Limenandra nodosa Haefelfinger & Stamm 

Phestilla lugubris Bergh 

Phestilla melanobranchia Bergh 

Phidiana lynceus Bergh 

Phyllodesmium xenia Gohar & Aboul-Ela 

Piseinotecus sphaeriferus (Schmekel) 

Pruvotfolia pselliotes (Labbé) 

Spurilla japónica (Eliot) 

Spurilla neapolitana (Chiaje) 



Tardy, 1962a 

Hadfield, 1963a 

Thompson, 1967 

Hamatani, 1961b 

Hurst, 1967 

Hadfield, 1963a 

Thompson & Brown, 1984 

Tardy, 1970 

Schmekel & Portmann, 1982 

Schmekel & Portmann, 1982 

Clark & Goetzfhed, 1978 

Haefelfinger, 1962 

Holleman, 1972 

Schmekel & Portmann, 1982 

Bändel, 1976 

Harrigan & Alkon, 1978 

Schmekel & Portmann, 1982 

Harris, 1975 

Harris, 1975 

Clark & Goetzfried, 1978 

Gohar & Aboul-Ela, 1957a 

Schmekel & Portmann, 1982 

Tardy, 1969b 

Hamatani, 1967 

Clark & Goetzfried, 1978 



Dendronotacea 



Dendronotus albopunctatus Robilliard 

Dendronotus frondosus (Ascanius) 

Dendronotus iris Cooper 

Dendronotus ruf us O'Donoghue 

Dendronotus subramosus MacFarland 

Doto coronata (Gmeiin) 

Doto doerga Marcus & Marcus 

Doto fragilis (Forbes) 

Doto japónica Odhner 

Doto paulinae Trínchese 

Doto pinnatifida (Montagu) 

Doto rosea Trínchese 

Doto yongei Thompson 

Hancockia burni Thompson 

Hancockia uncinata (Hesse) 

Lomanotus stauberi Clark & Goetzfried 

Melibe fimbriata Alder & Hancock 

!\flelibe leonina (Gould) 

Tritonia cincta Pruvot-Fol 

Tritonia diomedea Bergh 

Tritonia hombergi Cuvier 

Tritonia plebeia Johnston 



Robilliard, 1972 
Miller, 1958 
Robilliard, 1970 
Robilliard, 1970 
Robilliard, 1970 
Thompson, 1967 
Schmekel & Portmann, 1982 
Kress, 1975 
Hamatani, 1963 
Schmekel & Portmann, 1982 
Kress, 1975 

Schmekel & Portmann, 1982 
Thompson,- 1972b 
Thompson, 1972b 
Schmekel & Portmann, 1982 
Clark & Goetzfried, 1978 
Thompson & Crampton, 1984 
Bickell & Kempf, 1983 
Schmekel & Portmann, 1982 
Kempf & Willows, 1977 
Thompson, 1962 
Thompson, 1967 



Arminacea 



Dirona albollneata Cockerell & Eliot 
Dirona aurantia Hurst 
Hero formosa (Lovén) 
Janolus cristatus (Chiaje) 



Hurst, 1967 
Hurst, 1967 
Thompson, 1967 
Thompson & Brown, 1984 



WARNING COLORATION IN NUDIBRANCHS 



249 



TABLE 2. (Continued) 



Species 



WITHOUT PLANKTONIC LARVAE 
Colour 



Reference 



Doridacea minus Chromodorididae 

Austrodohs macmurdensis Odhner 
Dendrodoris krebsi (Mörch) 
Dendrodoris limbata (Cuvier) 
Discodoris rosi Ortea 
Doriopsilla pharpa Marcus 
Okadaia elegans Baba 
Trippa spongiosa (Kelaart) 
Vayssieria caledonica (Risbec) 

Doridacea Chromodorididae 

Cadlina laevis (Linnaeus) 
Chromodohs loringi (Angas) 
Glossodoris obsoleta (Rüppell & Leuckart) 
Glossodoris sibogae (Bergh) 
Hypselodoris bennetti (Angas) 
Hypselodoris villaf ranea (Risso) 
Mexichromis tricolor (Cantraine) 

Aeolidacea 

Aeolidiella alder (Cocks) 
Coryphella salmonacea (Couthouy) 
Cuthona granosa Schmekel 
Cuthona poritophages Rudman 
Embletonia gracilis Risbec 
Herviella mietta Marcus & Burch 



С 


Gibson et al. 1970 


с 


Clark & Goetzfried, 1978 


с 


Si, 1931 


с 


Ortea, 1979 


с 


Clark & Goetzfried, 1978 


с 


Baba, 1937 


с 


Gohar & Soliman, 1967g 


с 


Risbec, 1953 


А 


Thompson, 1967 


А 


Thompson, 1972a 


А 


Gohar & Soliman, 1967d 


А 


Usuki, 1967 


А 


Thompson, 1972a 


А 


Gantes, 1962 


А 


Haefelfinger, 1969 


? 


Tardy, 1962b 


? 


Morse, 1971 


С 


Schmekel, 1966 


С 


Rudman, 1979 


с 


Gosliner & Griffiths, 1981 


с 


Young, 1967 



own genes contributing to the next genera- 
tion. Conversely if a cryptic animal dies soon 
after reproducing this will reduce the chances 
of a predator finding it, acquiring a searching 
image for its patttern, and then hunting out 
similarly coloured prey in the vicinity. This ar- 
gument was first used by Blest (1963) and 
was supported by evidence from saturniid 
moths. It should be possible to test this pre- 
diction by comparing the times survived after 
the last oviposition by cryptic and by apose- 
matic nudibranchs with non-planktonic devel- 
opment. Unfortunately although this informa- 
tion is probably available in the files of the 
many workers who have studied oviposition 
and development in nudibranchs it has never 
been published. 

4. Evolution of innate responses 
of predators 

Until recently it was assumed that the pred- 
ators which avoid aposematic prey had to 
learn through experience to associate unpal- 



atability with colour or with other specific sig- 
nals, but several predators are now known 
which have innate aversion responses to spe- 
cific signals (see Edmunds, 1974). These 
predators include certain species of birds 
(Smith, 1975,1977; Caldwell & Rubinoff, 
1983) and fish (Rubinoff & Kropach, 1970), 
but other species of predator which do not 
experience the specific aposematic signals in 
their normal environment lack innate aver- 
sions (Smith, 1980). An animal is most un- 
likely to evolve an innate aversion to a stim- 
ulus unless there is clear advantage in 
avoiding that stimulus, otherwise several Ba- 
tesian mimics would quickly evolve to exploit 
the situation (Guildford, 1990). Such aversive 
responses evolved when the predators origi- 
nally developed learned aversions, but where 
there was selective advantage in minimizing 
the time spent, or the danger, in having to 
learn. It follows from this argument that innate 
aversive responses imply a long evolutionary 
history of behaviour in response to a specific 
aposematic signal. 



250 



EDMUNDS 



It is known that some fish have innate aver- 
sive responses to sea snakes (Rubinoff & 
Kropach, 1970), but no such response is 
known towards any nudibranch, and the ex- 
periments of feeding nudibranchs to fish indi- 
cate that the aversive responses have to be 
learned. However, no critical experiments 
have been undertaken, and where fish ap- 
pear to ignore nudibranchs crawling around in 
aquaria it is possible that they actually see 
and then deliberately avoid them innately. 

The Evolution of Aposematism 
in Nudibranchs 

Assuming that a particular nudibranch is 
aposematic the question arises of which 
evolved first, unpalatability or warning co- 
lours? The first possibility is supported by the 
fact that many cryptic nudibranchs are unpal- 
atable to fish (Thompson, 1960). In addition, 
conspicuous colours will only evolve in a pal- 
atable animal, or one that is only slightly un- 
palatable, if they confer some selective ad- 
vantage which is greater than the liability of 
attracting predators. Thus they might evolve 
because they confer an advantage in intraspe- 
cific encounters (e.g. sexual, territorial, etc.). 
However, there is no evidence that colour has 
such a function in nudibranchs. Colour can 
also be of value in interspecific contexts 
(Batesian mimicry, deimatic and flash co- 
lours), but these are comparatively rare in 
nudibranchs (Edmunds, 1987), so I consider it 
highly improbable that bright colours evolved 
before unpalatability. I therefore conclude that 
if aposematic nudibranchs occur, they have 
evolved through individual selection from ini- 
tially cryptic but relatively unpalatable species. 



CONCLUSIONS 

I conclude from this review that aposema- 
tism does occur in some nudibranchs and that 
the evidence for it is particularly strong in 
those species which are conspicuously co- 
loured in their natural habitat, are unpalatable 
to some predators, and are part of a mimetic 
group of species. Clearly there is plenty of 
scope for more detailed experimental studies 
of aposematism in nudibranch molluscs, and 
the purpose of this paper is to raise some of 
the issues that might usefully be addressed in 
such an investigation. There are also many 
anomalies that need further study; one exam- 
ple which I gave recently is Polycera elegans 



(Bergh) which is a brilliantly coloured species 
that is rare (Edmunds, 1987). Unless an ani- 
mal is exceptionally noxious and resilient to 
sampling by a predator, it is difficult to see the 
selective advantage for a rare species of be- 
ing conspicuous instead of being cryptic. I 
hope that this review will stimulate further ex- 
perimental work on the significance of brilliant 
colours in nudibranchs. 



ACKNOWLEDGEMENTS 

I wish to thank Professor Arthur Cain and Dr. 
Janet Edmunds for critically reading the paper, 
Dr. Hans Bertsch for permission to publish 
some of his observations, and numerous other 
nudibranch enthusiasts for discussions on the 
functions of colours in these animals. 



LITERATURE CITED 

BABA, К., 1937, Contribution to the knowledge of a 
nudibranch Okadaia elegans Baba. Japanese 
Journal of Zoology. 7: 147-190. 

BÄNDEL, К., 1976, Egg masses of 27 Caribbean 
opisthobranchs from Santa Marta, Columbia. 
Studies on Neotropical Fauna and Environment, 
11: 87-118. 

BEHRENS, D.W., 1980, Pacific Coast Nudi- 
branchs. Sea Challengers, Los Osos, California, 
112 pp. 

BERTSCH, H., 1978a, The Chromodoridinae nudi- 
branchs from the Pacific coast of America. Part II. 
The genus Chromodoris, Veliger, 20: 307-327. 

BERTSCH, H., 1978b, The Chromodoridinae nudi- 
branchs from the Pacific coast of America. Part 

III. The genera Chromolaichma and Mexi- 
chiromis. Veliger, 21 : 70-86. 

BERTSCH, H., 1978c, The Chromodoridinae nudi- 
branchs from the Pacific coast of America. Part 

IV. The genus Hypselodoris. Veliger, 21: 236- 
250. 

BERTSCH, H. & S. JOHNSON, 1981, Hawaiian 
nudibranchs. Oriental Publishing Co., Honolulu, 
Hawaii. 112 pp. 

BICKELL, L.R. & F.S. CHIA, 1979, Organogenesis 
and histogenesis in the planktonic veliger of 
Doridella steinbergae (Opisthobranchia: Nudi- 
branchia). Marine Biology Berlin 52: 291-313. 

BICKELL, L.R. & S.C. KEMPF, 1983, Larval and 
metamorphic morphogenesis in the nudibranch 
Melibe leonina (Mollusca: Opisthobranchia). Bio- 
logical Bulletin. Marine Biological Laboratory, 
Woods Hole. 165: 119-138. 

BLEST, A.D., 1963, Longevity, palatability and nat- 
ural selection in five species of New World sat- 
urniid moth. Nature, 197: 1183-1186. 

BRANDON, M. & CE. CUTRESS, 1985, A new 



WARNING COLORATION IN NUDIBRANCHS 



251 



Dondice (Opisthobranchia: Favorinidae), preda- 
tor of Cassiopea in southwest Puerto Rico. Bul- 
letin of Marine Science, 36: 139-144. 

BRIDGES, С & J.A. BLAKE, 1972, Embryology 
and larval development of Coryphella trilineata 
O'Donoghue, 1921 (Gastropoda: Nudibranchia). 
Veliger, 14: 293-297. 

BROWER, L.P., F.H. ROUGH & H.R. MECK, 1970, 
Theoretical investigations of automimicry, 1 . Sin- 
gle trial learning. Proceedings of the National 
Academy of Sciences of tfie United States of 
America, 66: 1059-1066. 

BROWER, L.R, W.N. RYERSON, LL. COP- 
PINGER & S.C. GLAZIER., 1968, Ecological 
chemistry and the palatability spectrum. Science, 
161: 1349-1351. 

CALDWELL, G.S. & R.W. RUBINOFF, 1983, 
Avoidance of venomous sea snakes by naive 
herons and egrets. Auk, 100: 195-198. 

CHIA, F.S. & R. KOSS, 1978, Development and 
metamorphosis of Rostanga pulchira (Mollusca: 
Nudibranchia). Marine Biology, Berlin 46: 109- 
120. 

CLARK, K.B., 1975, Nudibranch life cycles in the 
northwest Atlantic and their relationship to the 
ecology of fouling communities. Helgoländer wis- 
senscfiaftliche Meeresuntersucfiungen, 27: 28- 
69. 

CLARK, K.B. & A. GOETZFRIED, 1978, Zoogeo- 
graphie influences on development patterns of 
north Atlantic Ascoglossa and Nudibranchia, with 
a discussion of factors affecting egg size and 
number. Journal of Molluscan Studies, 44: 283- 
294. 

COOK, E.F., 1962, A study of food choices of two 
opisthobranchs, Rostanga pulchra McFarland 
and Archidoris montereyensis (Cooper). Veliger, 
4: 194-196. 

COTT, H.B., 1940. Adaptive coloration in animals. 
Methuen, London. 508 pp. 

CROSSLAND, C, 1911, Warning coloration in a 
nudibranch mollusc and in a chamaeleon. Pro- 
ceedings of the Zoological Society of London, 79: 
1062-1067. 

CROZIER, W.J., 1916, On the immunity coloration 
of some nudibranchs. Proceedings of the Na- 
tional Academy of Sciences of the United States 
of America, 2: 672-675. 

EDMUNDS, M., 1966, Protective mechanisms in 
the Eolidacea (Mollusca Nudibranchia). Journal 
of the Linnean Society (Zoology), 46: 27-71 . 

EDMUNDS, M., 1974, Defence in Animals. Long- 
man, Harlow. 357 pp. 

EDMUNDS, M., 1981, Opisthobranchiate Mollusca 
from Ghana: Chromodorididae. Zoological Jour- 
nal of the Linnean Society, 72: 175-201 . 

EDMUNDS, M., 1987, Color in opisthobranchs. 
American Malacological Bulletin, 5: 185-196. 

EDMUNDS, M. & A. KRESS, 1969, On the Euro- 
pean species of Eubranchus (Mollusca Opis- 
thobranchia). Journal of the Marine Biological 
Association of the United Kingdom, 49: 879-91 2. 

EYSTER, L.S., 1979, Reproduction and develop- 



mental variability in the opisthobranch Tenellia 
pallida. Marine Biology, Berlin 51: 133-140. 

EYSTER, L.S., 1980, Distribution and reproduction 
of shell-less opisthobranchs from South Carolina. 
Bulletin of Marine Science, 30: 580-599. 

FRANZ, D.R. & K.B. CLARK, 1972, A discussion of 
the systematics, reproductive biology, and zoo- 
geography of Polycerella emertoni and related 
species (Gastropoda: Nudibranchia). Veliger, 14: 
265-270. 

GANTES, H., 1962, Recherches sur quelques 
larves de Glossodorididae (Mollusques Opistho- 
branches). Bulletin de la Société des Sciences 
naturelles et physiques du Maroc, 42: 267-277. 

GARSTANG, W., 1889, Report on the nudibranchi- 
ate Mollusca of Plymouth Sound. Journal of the 
Marine Biological Association of the United King- 
dom, 1: 173-198. 

GARSTANG, W., 1890, A complete list of the 
opisthobranchiate Mollusca found at Plymouth. 
Journal of the Marine Biological Association of 
the United Kingdom, 1 : 399-457. 

GIBSON, R., Т.Е. THOMPSON & G.A. ROBIL- 
LIARD, 1970, Structure of the spawn of an Ant- 
arctic dorid nudibranch Austrodoris macmurden- 
sis Odhner. Proceedings of the Malacological 
Society of London, 39: 221-225. 

GITTLEMAN, J.L. & RH. HARVEY, 1980, Why are 
distasteful prey not cryptic? Nature, 286: 149- 
150. 

GODDARD, J.H.R., 1984, The opisthobranchs of 
Cape Arago, Oregon, with notes on their biology 
and a summary of benthic opisthobranchs known 
from Oregon. Veliger, 27: 143-163. 

GOHAR, H.A.F. & I.A. ABOUL-ELA, 1957a, On a 
new nudibranch "Phyllodesmium xenia" (from 
the Red Sea, with a description of its develop- 
ment). Publications of the Marine Biological Sta- 
tion, Ghardaqa, 9: 131-144. 

GOHAR, H.A.F., 1975b, The development of three 
chromodorids (with the description of a new spe- 
cies). Publications of the Marine Biological Sta- 
tion, Ghardaqa, 9: 203-228. 

GOHAR, H.A.F. & I.A. ABOUL-ELA, 1959, On the 
biology and development of three nudibranchs 
(from the Red Sea). Publications of the Marine 
Biological Station, Ghardaqa, 10:41-62. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1963, The biol- 
ogy and development of Hexabranchus san- 
guineus (Rupp. and Leuck.) (Gastropoda, Nudi- 
branchiata). Publications of the Marine Biological 
Station, Ghardaqa, 12: 219-248. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967a, The 
biology and development of Dendrodoris 
( = Doridopsis) fumata (Rupp. and Leuck.) (Gas- 
tropoda, Nudibranchia). Publications of the Ma- 
rine Biological Station, Ghardaqa, 14: 31-54. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967b, The bi- 
ology and development of Chromodoris inornata 
Pease (Gastropoda, Nudibranchia). Publications 
of the Marine Biological Station, Ghardaqa, 14: 
77-94. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967c, The bi- 



252 



EDMUNDS 



ology and development of the nudibranch Chro- 
modoris tinctoria (Rupp. and Leuck.) (with refer- 
ence to the taxonomic value of spawning 
characters). Publications of the Marine Biological 
Station, Ghardaqa. 14: 95-108. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967d, The di- 
rect development of the nudibranch Casella ob- 
soleta (Rupp. and Leuck.) (with remarks on the 
metamorphosis). Publications of tfie Marine Bio- 
logical Station. Gtiardaqa. 14: 149-166. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967e, The bi- 
ology and development of Asteronotus cespito- 
sus (van Hasselt) (Gastropoda, Nudibranchia). 
Publications of the Mahne Biological Station, 
Ghardaqa, 14: 177-195. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967f, The bi- 
ology and development of Discodoris concinna 
(Alder and Hancock) (Gastropoda, Nudibran- 
chia). Publications of the Mahne Biological Sta- 
tion, Ghardaqa. 14: 197-214. 

GOHAR, H.A.F. & G.N. SOLIMAN, 1967g, On two 
rare nudibranchs of the genus Thppa Bergh (of 
different developmental behaviour). Publications 
of the Mahne Biological Station, Ghardaqa, 14: 
269-293. 

GOSLINER, T.M. & R.J. GRIFFITHS, 1981, De- 
scription and revision of some South African ae- 
olidacean Nudibranchia (Mollusca, Gastropoda). 
Annals of the South African Museum, 84: 105- 
150. 

GOSLINER, T.M. & S.V. MILLEN, 1984, Records of 
Cuthona pustulata (Alder & Hancock, 1 854) from 
the Canadian Pacific. Veliger, 26: 183-187. 

GUILDFORD, T, 1990. The evolution of aposema- 
tism. In: Insect Defenses, ed. D.L. Evans & J.O. 
Schmidt, SUNY Press, New York, pp 23-61. 

HADFIELD, M.G. 1 963a, The biology of nudibranch 
larvae. Oikos, 14: 85-95. 

HADFIELD, M.G., 1963b, Coryphella pan/a n. sp., 
a new nudibranchiate mollusc from the Oresund. 
Videnskabelige Meddelelser fra Dansk naturhis- 
torisk Forening, 125: 371-376. 

HAEFÈLFINGER, H.R., 1962, Quelques faits con- 
cernant la nutrition chez Favohnus branchialis 
(Rathke 1806) et Stiliger vesiculosus (Deshayes 
1864), deux mollusques opisthobranches. Revue 
suisse de zoologie, 69: 312-316. 

HAEFELFINGER, H.R., 1969, Contnbution a la 
systématique des glossodoridiens méditerra- 
néens (Gastropoda, Opisthobranchia). Revue 
suisse de zoologie, 76: 703-710. 

HAMATANI, I., 1960a, Notes on veligers of Japa- 
nese opisthobranchs (1 ). Publications of the Seto 
Mahne Biological Laboratory, 8: 55-70. 

HAMATANI, I., 1960b, Notes on veligers of Japa- 
nese opisthobranchs (2). Publications of the Seto 
Mahne Biological Laboratory, 8: 307-315. 

HAMATANI, I., 1961a, notes on veligers of Japa- 
nese opisthobranchs (3). Publications of the Seto 
Manne Biological Laboratory, 9: 67-79. 

HAMATANI, I., 1961b, Notes on veligers of Japa- 
nese opisthobranchs (4). Publications of the Seto 
Mahne Biological Laboratory, 9: 353-361. 



HAMATANI, I., 1962, Notes on veligers of Japa- 
nese opisthobranchs (5). Publications of the Seto 
Mahne Biological Laboratory, 10: 283-292. 

HAMATANI, I., 1963, Notes on veligers of Japa- 
nese opisthobranchs (6) Publications of the Seto 
Mahne Biological Laboratory, 1 1 : 125-130. 

HAMATANI, 1967., Notes on veligers of Japanese 
opisthobranchs (7). Publications of the Seto Ma- 
hne Biological Laboratory, 15: 121-131. 

HAMILTON, W.D., 1964, The genetical evolution of 
social behaviour. I-II. Journal of Theoretical Biol- 
ogy, 7: 1-52. 

HARDY, A.C., 1956, The open sea. Its natural his- 
tory: the world of plankton. Collins, London, 335 
pp. (New Naturalist 34). 

HARRIGAN, J. F. & D.L. ALKON, 1978, Larval rear- 
ing, metamorphosis, growth and reproduction of 
the eolid nudibranch Hermissenda crassicornis 
(Eschscholtz, 1831) (Gastropoda: Opisthobran- 
chia). Biological Bulletin. Mahne Biological Lab- 
oratory, Woods Hole, 154: 430-439. 

HARRIS, L.G., 1973, Nudibranch associations. 
Current Topics in Comparative Pathobiology, 2: 
213-315. 

HARRIS, L.G., 1975, Studies on the life history of 
two coral-eating nudibranchs of the genus Phe- 
stilla. Biological Bulletin. Mahne Biological Labo- 
ratory. Woods hole, 149: 539-550. 

HARRIS, L.G., M. POWERS & J. RYAN, 1980, Life 
history studies of the estuarine nudibranch Tenel- 
lia fuscata (Gould, 1870). Veliger, 23: 70-74. 

HARRIS, L.G., L.W. WRIGHT & B.R. RIVEST, 
1975, Observations on the occurrence and biol- 
ogy of the aeolid nudibranch Cuthona nana in 
New England waters. Veliger, 17: 264-268. 

HARVEY, P.H., J.J. BULL., M. PEMBERTON & 
R.J. PAXTON, 1982, The evolution of apose- 
matic coloration in distateful prey: a family model. 
American Naturalist, 119: 710-719. 

HARVEY, P.H. & R.J. PAXTON, 1981a, The evolu- 
tion of aposematic coloration. Oikos, 37: 391- 
393. 

HARVEY, P.H. & R.J. PAXTON, 1981b, On apose- 
matic coloration: a rejoinder. Oikos, 37: 395- 
396. 

HECHT, е., 1896, Contribution a l'étude des nudi- 
branches. Mémoires de la Société de Zoologique 
de France, 8: 537-711. 

HERDMAN, W.A. 1890, On the structure and func- 
tion of the cerata or dorsal papillae in some nudi- 
branchiate Mollusca. Quarterly Journal of Micro- 
scopical Science, 31 : 41-63. 

HERDMAN, W.A. & J.A. CLUBB, 1 890, Third report 
on the Nudibranchiata of the L.M.B.C. district. 
Proceedings and Transactions of the Liverpool 
Biological Society, 4: 131-169. 

HOLLEMAN, J.J., 1972, Observations on growth, 
feeding, reproduction and development in the 
opisthobranch Fiona pinnata (Eschscholtz). 
Veliger, 15: 142-146. 

HURST, A., 1967, The egg masses and veligers of 
thirty northeast Pacific opisthobranchs. Veliger, 
9: 255-288. 



WARNING COLORATION IN NUDIBRANCHS 



253 



JARVI, T., В. SILLEN-TULLBERG & С. WIKLUND, 
1981, The cost of being aposematic. An experi- 
mental study of prédation on larvae of Papilio 
machaon by the great tit Parus major. Oikos, 36: 
267-272. 

JUST, H. & M. EDMUNDS, 1985, North Atlantic 
nudibranchs (Mollusca) seen by Henning Lem- 
che. Ophelia supplement 2, 170 pp. 

KEMPF, S.C. & A.O.D. WILLOWS, 1977, Labora- 
tory culture of the nudibranch Tritonia diomedea 
Bergh (Tritoniidae: Opisthobranchia) and some 
aspects of its behavioral development. Journal of 
Experimental Marine Biology and Ecology, 30: 
261-276. 

KRESS, A., 1975, Observations during embryonic 
development in the genus Doto (Gastropoda, 
Opisthobranchia). Journal of the Marine Biologi- 
cal Association of the United Kingdom, 55: 691- 
701. 

KUZIRIAN, A.M., 1977, The rediscovery and biol- 
ogy of Coryphella nobilis Verrill, 1880 in New En- 
gland (Gastropoda: Opisthobranchia). Journal of 
Molluscan Studies, 43: 230-240. 

KUZIRIAN, A.M., 1979, Taxonomy and biology of 
four New England coryphellid nudibranchs (Gas- 
tropoda: Opisthobranchia). Journal of Molluscan 
Studies, 45: 239-261. 

McGOWAN, J.A. & I. PRATT, 1954, The reproduc- 
tive system and early embryology of the nudi- 
branch Archidoris montereyensis (Cooper). Bul- 
letin of the Museum of Comparative Zoology at 
Harvard, 111: 261-276. 

MILLEN, S.V. & TM. GOSLINER, 1985, Four new 
species of dorid nudibranchs belonging to the ge- 
nus Aldisa (Mollusca: Opisthobranchia), with a 
revision of the genus. Zoological Journal of the 
Linnean Society 84: 195-233. 

MORSE, M.P., 1971, Biology and life history of the 
nudibranch mollusc, Coryphella stimpsoni (Verrill 
1879). Biological Bulletin. Marine Biological Lab- 
oratory, Woods Hole, 140: 84-94. 

MORSE, M.P., 1972, Biology of OI<enia ascidicola 
spec. nov. (Gastropoda: Nudibranchia). Veliger, 
15: 97-101. 

ORTEA, J.A., 1979, Deux nouveaux doridiens 
(Mollusca, Nudibranchiata) de la côte nord 
d'Espagne. Bulletin du Muséum national d'his- 
toire naturelle, Paris (4) 1 (A): 575-583. 

PERRON, F.E. & R.D. TURNER, 1977, Develop- 
ment, metamorphosis, and natural history of the 
nudibranch Doridella obscura Verrill (Coram- 
bidae: Opisthobranchia). Journal of Experimental 
Marine Biology and Ecology, 27: 171-185. 

POULTON, E.B., 1890, The Colours of Animals. 
Their Meaning and Use, especially considered in 
the Case of Insects. Kegan Paul, Trench, Trüb- 
ner, London. 360 pp (International Scientific Se- 
ries 68). 

RAO, K.P., 1965, Moridella brockii Bergh 1888, re- 
described with notes on anatomy and early de- 
velopment. Journal of the Mahne Biological As- 
sociation of India, 7: 61-68. 

RAO, K.V., 1962, Development and life history of a 



nudibranchiate gastropod Cuthona adyarensis 
Rao. Journal of Marine Biological Association of 
India, 3: 186-197. 

RAO, K.V., 1 970, On the structure and life-history of 
a new aeolid, Favorinus argentimaculatus from 
Palk Bay. pp 1009-1016. in: Proceedings of the 
Symposium on Mollusca (Cochin, 1968). Sympo- 
sium Ser. Manne Biological Association of India, 
3. 

RAO, K.V. & K. ALAGARSWAMI, 1960, An account 
of the structure and early development of a new 
species of a nudibranchiate gastropod, Eolidina 
(Eolidina) mannarensis. Journal of the Marine Bi- 
ological Association of India, 2: 6-16. 

RHO, F., 1888, Studii sullo sviluppo della Chromo- 
doris elegans. Atti dell' Accademia delle scienze 
fisiche e matematiche, Napoli (2) 1 App. 3: 1-7. 

RISBEC, J., 1953, Mollusques nudibranches de la 
Nouvelle-Calédonie. Faune de l'Union français 
15: 1-189. 

RIVEST, B.R., 1978, Development of the eolid 
nudibranch Cuthona nana (Alder and Hancock, 
1842), and its relationship with a hydreid and a 
hermit crab. Biological Bulletin. Mahne Biological 
Laboratory, Woods Hole, 154: 157-175. 

ROBILLIARD, G.A., 1970, The systematics and 
some aspects of the ecology of the genus Den- 
dronotus (Gastropoda: Nudibranchia). Veliger, 
12: 433-479. 

ROBILLIARD, G.A, 1972, A new species of Den- 
dronotus from the northeastern Pacific with notes 
on Dendronotus nanus and Dendronotus robus- 
tus (Mollusca: Opisthobranchia). Canadian Jour- 
nal of Zoology, 50: 421-432. 

ROGINSKAYA, I.S., 1962, The egg-masses of 
nudibranchs of the White Sea. Reports of the 
White Sea Biological Station of the State Univer- 
sity of Moscow, 1 : 201-214. 

ROGINSKAYA, I.S., 1969, Taxonomy and ecology 
of the nudibranch mollusc Coryphella fusca. Zoo- 
logicheskii zhurnal, 48: 1614-1617. 

ROGINSKAYA, 1970, Tenellia adspersa, a nudi- 
branch new to the Azov Sea, with notes on its 
taxonomy and ecology. Malacological Review, 3: 
167-174. 

ROS, J., 1976, Sistemas de defensa en los opisto- 
branquios. Oecologia aquatica, 2: 41-77. 

ROS, J., 1977, La defensa en los opistobranquios. 
Investigación y Ciencia, 12: 48-60. 

RUBINOFF, I. & С. KROPACH, 1970, Differential 
reactions of Atlantic and Pacific predators to sea 
snakes. Nature, Lond. 228: 1288-1290. 

RUDMAN, W.B., 1979, The ecology and anatomy 
of a new species of aeolid opisthobranch mol- 
lusc; a predator of the scleractinian coral Pontes. 
Zoological Journal of the Linnean Society, 65: 
339-350. 

RUDMAN, W.B., 1985, The Chromodorididae 
(Opisthobranchia: Mollusca) of the Indo-West 
Pacific: Chromodoris aureomarginata, С verrieh 
and С fidelis colour groups. Zoological Journal 
of the Linnean Society, 83: 241-299. 

SCHMEKEL, L., 1966, Zwei neue Arten der Familie 



254 



EDMUNDS 



Cuthonidae aus dem Golf von Neapel: Trinchesia 
granosa n. sp. und Trinchesia oceilata n. sp. 
(Gastr. Opisthobranchia). Pubblicazioni della 
Stazione zoológica di Napoli, 35: 13-28. 

SCHMEKEL, L. & A. PORTMANN, 1982, Opistho- 
branchia des Mittelmeeres. Springer- Verlag, Ber- 
lin. 410 pp. 

SCHULER, W. & E. HESSE, 1985, On the function 
of warning coloration: a black and yellow pattern 
inhibits prey-attack by naive domestic chicks. Be- 
havioral Ecology and Sociobiology. 1 6: 249-255. 

SHYAMASUNDARI, K. & NAJBUDDIN, M., 1976, 
Experimental investigations of salinity and tem- 
perature effects on early developmental stages in 
Dendrodoris (Doriopsilla) minlata (Alder & Han- 
cock) (Gastropoda Opisthobranchia). Monitore 
Zoológico Italiano. 10: 93-94. 

SI, T., 1931, Un nouveau cas de condensation em- 
bryogénique chez un nudibranche {Doridopsis 
limbata Cuvier). Compte rendu hebdomadaire 
des séances de l'Académie des sciences. Paris 
192: 302-304. 

SILLEN-TULLBERG, В., 1985, Higher survival of 
an aposematic than of a cryptic form of a distate- 
ful bug. Oecologia. 67: 411-415. 

SILLEN-TULLBERG, B. & E.H. BRYANT, 1983, 
The evolution of aposematic coloration in distate- 
ful prey: an individual selection model. Evolution. 
37: 993-1000. 

SMITH, S.M., 1975, Innate recognition of coral 
snake pattern by a possible avian predator. Sci- 
ence. N.Y., 187: 759-760. 

SMITH, S.M., 1977, Coral-snake pattern recogni- 
tion and stimulus generalisation by naive great 
kiskadees (Aves: Tyrannidae). Nature, Lond. 
265: 535-536. 

SMITH, S.M., 1980, Responses of naive temperate 
birds to warning coloration. American Midland 
Naturalist. 103: 346-352. 

SOLIMAN, G.N., 1978, The redescription, repro- 
duction and development of the dorid nudibranch 
Platydoris scabra (Cuvier) from the northwestern 
Red Sea. Journal of Molluscan Studies, 44: 1 51- 
165. 

SOLIMAN, G.N., 1980, On the dorid nudibranch 
Sebadoris crosslandi (Eliot) from the northwest- 
ern Red Sea. Journal of Molluscan Studies. 46: 
227-238. 

SOLIMAN, G.N., 1986, The reproduction and de- 
velopment of the dorid nudibranch Phyllldia var- 
icosa Lamarck 1801 from the northern Red Sea. 
Proceedings of the 8th International Malacologi- 
cal Congress,. Budapest. 1983. 255-260. 

SOLIMAN, G.N., 1991, A comparative review of 
the spawning, development and metamorphosis 
of prosobranch and opisthobranch gastropods 
with special reference to those from the north- 
western Red Sea. In: Edmunds, M (ed.). 
Evolutionary biology of opisthobranchs. Proceed- 
ings of the 9th International Malacological Con- 
gress. Edinburgh. 1986, Malacologia, 32(2): 
257-271 . 

TARDY, J., 1962a, Observations et expériences 



sur la métamorphose et la croissance de Capel- 
linia exigua. (Aid. et H.) (Mollusque Nudi- 
branche). Compte rendu hebdomadaire des sé- 
ances de l'Académie des sciences, Paris 254: 
2242-2244. 

TARDY, J., 1962b, Cycle biologique et métamor- 
phose d'Aaeolidina alderi (Gastéropode, Nudi- 
branche). Compte rendu hebdomadaire des sé- 
ances de l'Académie des sciences, Paris 255: 
3250-3252 

TARDY, J., 1962c, A propos des espèces de 
Berghia (Gastéropodes Nudibranches) des cotes 
de France et de leur biologie. Bulletin de l'Institut 
océanographique, Monaco 59 (1255): 1-20. 

TARDY, J., 1969a, Etude systématique et bi- 
ologique sur trois espèces d'Aeolidielles des côtes 
européenes (Gastéropodes Nudibranches). Bul- 
letin de l'Institut océanographique. Monaco 68 
(1389): 1-40. 

TARDY, J., 1969b, Un nouveau genre de nudi- 
branche méconnu des côtes atlantique et de la 
Manche: Pruvotfolla (nov. g.) pselliotes (Labbé), 
1923. Vie et Milieu, 20: 327-346. 

TARDY, J., 1970, Contribution à la connaissance 
de la biologie chez les nudibranches: développ- 
ement et métamorphose; vie prédatrice: 1. 
Facelina coronata (Forbes) et Aeolis sp. Bulletin 
de la Société zoologique de France, 95: 765- 
772. 

THIRIOT-QUIEVREUX, C, 1977, Véligère plancto- 
trophe du doridien Aegires punctilucens (d'Or- 
bigny) (Mollusca: Nudibranchia; Notodorididae): 
description et métamorphose. Journal of Experi- 
mental Marine Biology and Ecology, 26: 177- 
190. 

THOMPSON, Т.Е., 1958, The natural history, em- 
bryology, larval biology and post-larval develop- 
ment of Adalaria próxima (Alder and Hancock) 
(Gastropoda Opisthobranchia). Philosophical 
Transactions of the Royal Society, series B, 242: 
1-58. 

THOMPSON, Т.Е., 1960, Defensive adaptations in 
opisthobranchs. Journal of the Marine Biological 
Association of the United Kingdom, 39: 123-134. 

THOMPSON, Т.Е., 1962, Studies on the ontogeny 
of Trifonía hombergi Cuvier (Gastropoda 
Opisthobranchia). Philosophical Transactions of 
the Royal Society series S, 245: 171-218. 

THOMPSON, Т.Е., 1967, Direct development in a 
nudibranch, Cadlina laevis, with a discussion of 
developmental processes in Opisthobranchia. 
Journal of the Manne Biological Association of 
the United Kingdom. 47: 1-22. 

THOMPSON, Т.Е., 1972a, Chromodorid nudi- 
branchs from eastern Australia (Gastropoda, 
Opisthobranchia). Journal of Zoology, London, 
166: 391-409. 

THOMPSON, Т.Е., 1972b, Eastern Australian Den- 
dronotoidea (Gastropoda: Opisthobranchia). 
Zoological Journal of the Linnean Society, 51 : 
63-77. 

THOMPSON, Т.Е., 1975, Dorid nudibranchs from 
eastern Australia (Gastropoda, Opisthobran- 



WARNING COLORATION IN NUDIBRANCHS 



255 



chia). Journal of Zoology, London, 176: 477- 
517. 

THOMPSON, Т.Е. & G. H. BROWN, 1984, Biology 
of Opisthobranch Molluscs 2. Ray Society, Lon- 
don, 229 pp. 

THOMPSON, TE. & D.M. CRAMPTON, 1984, Bi- 
ology of Melibe fimbriata, a conspicuous opistho- 
branch mollusc of the Indian Ocean, which has 
now invaded the Mediterranean Sea. Journal of 
f^olluscan Studies, 50: 113-121. 

TODD, CD., 1 981 , The ecology of nudibranch гло1- 
luscs. Oceanography and Marine Biology Annual 
Reviews, 19: 141-234. 

TODD, CD., 1983, Reproductive and trophic ecol- 
ogy of nudibranch molluscs, pp. 225-259. In: 
Russell-Hunter, W.D. (ed.). Ecology. The Mol- 
lusca (ed. К. Wilbur) 6, Academic Press, London 
& New York. 



USUKI, I., 1967, The direct development and the 
single cup-shaped larval shell of a nudibranch 
Glossodoris sibogae (Bergh). Scientific Reports 
of Niigata University, Series D (Biology), 4:. 75- 
85. 

WALLACE, A.R., 1889, Darwinism. Macmillan, 
London. 494 pp. 

WILLAN, R.C & N. COLEMAN, 1984, Nudibranchs 
of Australia. Australasian Marine Photographic 
Index, Sydney. 56 pp. 

WILLIAMS, L.G., 1980, Development and feeding 
of larvae of the nudibranch gastropods Hermis- 
senda crassicornis and Aeolidia papulosa. Mala- 
cologia, 20: 99-116. 

YOUNG, D.K., 1967, New records of Nudibranchia 
(Gastropoda: Opisthobranchia) from the central 
and west-central Pacific with a description of a 
new species. Veliger, 10: 159-173. 



MALACOLOGIA, 1991,32(2): 257-271 

A COMPARATIVE REVIEW OF THE SPAWNING, DEVELOPMENT 

AND METAMORPHOSIS OF PROSOBRANCH AND OPISTHOBRANCH 

GASTROPODS WITH SPECIAL REFERENCE TO THOSE FROM THE 

NORTHWESTERN RED SEA 

Garni! N. Soliman 

Department of Zoology, Faculty of Science, University of Cairo, Egypt 

ABSTRACT 

Aspects of spawning, development and metamorphosis of 50 prosobranch and opisthobranch 
gastropods from the northwestern Red Sea are reviewed. For almost every species, data are 
given of the breeding season, size and number of eggs laid, and period and type of develop- 
ment. The early embryology, larval structure and behaviour, and post-lan/al development are 
summarized. Only the nudibranch Casella obsoleta has direct development. In agreement with 
Thorson's rule, most species have pelagic development, although prosobranchs (neogastro- 
pods in particular) show a tendency towards lecithotrophy and rapid metamorphosis. The in- 
trinsic and extrinsic factors affecting the type of development are discussed. 



INTRODUCTION 

In an attempt to review reproduction in 
prosobranchs and opisthobranchs, a study of 
their egg masses has recently been made 
(Soliman, 1987). The present paper aims to 
extend the comparative study to other as- 
pects of reproduction, namely egg size and 
number, early embryology, larval structure 
and behaviour, and the type of reproduction, 
and to review the factors which affect devel- 
opment and metamorphosis. This paper also 
aims to find out to what extent the patterns of 
molluscan development in the northwestern 
Red Sea agree with Thorson's rule (1950) 
and to compare our results with those re- 
ported from other areas lying more or less 
within the same latitudes. 

The present study, like the former, is based 
mainly on new data and on studies made on 
Red Sea gastropods during the last 30 years 
(Gohar & Aboul-Ela, 1957, 1959; Gohar & 
Eisawy, 1963,1967; Gohar & Soliman, 1963, 
1967; Eisawy & Serial, 1968, 1974, 1976; 
Eisawy, 1970; Soliman, 1977, 1978, 1980, 
1983, 1986) together with data from other 
sources. 

SPAWNING, EMBRYOLOGY AND 
LARVAL DEVELOPMENT 

Egg Capsules 

Primitively, gastropod eggs are laid singly, 
uncovered and they are externally fertilized. 



In most gastropods, however, eggs are en 
closed either singly or in groups (up to hun 
dreds) in transparent thin-walled cases, ir 
leathery sacs or hard capsules. Cases ma^ 
be dispersed freely becoming planktonic 
Most often they are embedded in a gelatinoui 
matrix and moulded in thin sheets, ovoid oi 
globular jelly masses (or without a definite 
shape), cords or ribbons (Soliman, 1987). 

In opisthobranchs, egg cases may lie di- 
rectly in the spawn jelly or, as in many nudi- 
branch egg ribbons, may be primarily en- 
closed in tubes of thick mucus winding in the 
spawn matrix in variable fashion. Thus in small 
ribbons the tubes run in a closely parallel man- 
ner (Fig. 1A), but they radiate peripherally in 
massive undulating ones (Fig. IB). In some 
species, each egg case within the tube may be 
isolated in a thin-walled compartment of var- 
ious shapes (Fig. 1C,D). The egg string of 
Strombus tricornis (Eisawy & Serial, 1 968) has 
a similar construction with the egg cases each 
enclosed in a gelatinous compartment that is 
finally coated by the egg string (Fig. 1 E). 

The size limits of opisthobranch egg cases 
(and the numbers of eggs enclosed) are gen- 
erally lower than those recorded for proso- 
branch capsules (Table 1). In the former, 
cases measure 0.1-0.3 mm on average, but 
cases up to 0.6 mm across are not uncom- 
mon. The largest cases encountered in some 
masses of the nudibranch Hexabranchius 
sanguineus (2 x 0.6 mm, with more than 100 
eggs) (Gohar & Soliman, 1963b) are still far 
smaller than what exists in certain proso- 



257 



258 



SOLIMAN 




100 pm 



1 mm 



FIG. 1. Patterns of egg case arrangement in the jelly matrix in Red Sea gastropods. A. Trippa spongiosa: 
part of two parallel egg cords in jelly with no compartments (after Gohar & Soliman, 1967g). В. Discodoris 
concinna: enlarged portion of spawn ribbon with cords radiating and winding distally (modified after Gohar 
& Soliman, 1967f). С Asteronotus cespitosus: segmented egg cord with one case in each segment (after 
Gohar & Soliman, 1967e). D. Chromodohs inornata: egg cases with each enclosed in a separate compart- 
ment (after Gohar & Soliman, 1 967b). E. Strombus tricornis: similar arrangement to D, but eggs are in cords, 
not in a common jelly (after Eisawy & Sorial, 1968). 



branchs (19-35 x 5-9 mm in Pleuroploca tra- 
pezium with up to 400 eggs, and 8-9 x 3.5 
mm in Nassa francolina with up to 1678 eggs) 
(Gohar & Eisawy, 1967b). 

Size and number of eggs 

The number of eggs laid by a gastropod is 
inversely proportional to egg size. According 



to the available data, the maximal size at- 
tained in prosobranchs markedly exceeds 
that in opisthobranchs (up to 0.75 mm in Co- 
nus (Natarajan, 1957), 0.44 mm in Strombus 
tricornis (Eisawy & Sorial, 1 968) and 0.42 mm 
in Pleuroploca trapezium (Gohar & Eisawy, 
1967b), against only 0.39 mm in Cadlina lae- 
vis (Thompson, 1967) and 0.33 mm In Ca- 
sella obsoleta (Gohar & Soliman, 1967d)). As 



DEVELOPMENT OF RED SEA GASTROPODS 



259 



anticipated, the number of eggs in a spawn 
deposited by an opisthobranch substantially 
surpasses what is recorded for prosobranchs 
(apart from certain archaeogastropods). 148 
million eggs were recorded laid by Aplysia 
californica (MacGinitie, 1934) and a little less 
than 5 million by the nudibranch Asteronotus 
cespitosas (Gohar & Soliman, 1967e). 

However, in many prosobranchs (princi- 
pally neogastropods) the majority of eggs laid 
act as nurse cells subserving as food for the 
very small fraction of viable eggs which pro- 
ceed to full development (15% in Pleuroploca 
trapezium, < 6% in Ctiicoreus ramosus (Go- 
har & Eisawy, 1967b), 3% in Fusinus tuber- 
culata (Eisawy & Serial, 1976a), 2% in 
Chicoreus virgineus (Gohar & Eisawy, 
1967b), and even less in certain other spe- 
cies). According to Thorson (1940) 50,000- 
100,000 nurse eggs may exist per embryo in 
Volutopsius norwegicus. Nurse cells are not 
reported to exist in opisthobranchs. 

Cleavage and early embryology 

In the majohty of opisthobranchs, the two 
initial divisions of the egg invariably bring 
about the formation of nearly equal mac- 
romeres which proceed thereafter in typical 
spiral cleavage (Fig. 2A). Unequal division is, 
however, typical of large, yolky prosobranch 
eggs. Actually such division is mainly depen- 
dent on the amount of yolk in the egg irre- 
spective of its size. Thus while the initial divi- 
sion of Tonna olearium eggs (0.25 mm in 
diameter) gives rise to nearly equal mac- 
romeres (Gohar & Eisawy, 1967a), similar or 
even smaller eggs of other species (0.18- 
0.22 mm, Ctiicoreus virgineus; 0.25 mm, C. 
ramosus (Gohar & Eisawy, 1967b)) exhibit 
markedly unequal division with the mac- 
romeres A-C appearing as if budding from the 
giant D cell (Fig. 2B). The subsequent divi- 
sions do not much affect the relative disparity 
in the size of the macromeres. 

In contrast to the opacity of most gastropod 
embryos and larvae developing from leci- 
thotrophic eggs, planktonic larvae are rela- 
tively transparent. It is not uncommon never- 
theless to come across opaque embryos 
arising from comparatively small eggs (70 
|jLm, Sebadoris crosslandi (Soliman, 1980)) 
or, conversely, lecithotrophic larvae {Trippa 
spongiosa, egg 0.2 mm across (Gohar & Soli- 
man, 1967g)) or veliger and metamorphosing 
stages of directly developed species (Acte- 
ocina atrata (Mikkelsen & Mikkelsen, 1984)) 



with markedly clear structure. Generally, how- 
ever, as far as the planktotrophic Red Sea 
gastropods are concerned, the developmen- 
tal stages of opisthobranchs are much more 
transparent than are those of prosobranchs. 
Particularly in Hexabranctius sanguineus 
(Gohar & Soliman, 1963b), with intensely red 
yolk globules mostly condensed at the vege- 
tative side of the egg, it is possible to follow, in 
whole live embryos, the formation of the 4d 
mesoblast and its division into two cells (Fig. 
20), which remain visible even after their mi- 
gration inwards between the future ectoderm 
and endoderm. In fact, H. sanguineus, be- 
cause of its abundance, the huge number of 
eggs it lays, and the clarity of its cells, is ideal 
material for live study of spiral cleavage, cell 
lineage and germ layer formation in molluscs. 

Among the structures to appear early in 
opisthobranch embryos are the anal cells. 
These are initially posteroventral and slightly 
to the left. Their gradual shifting to the anterior 
right side (with the future proctodeal invagi- 
nation and anus) is the only ontogenetic evi- 
dence of torsion, which is thus much less than 
180°. Casella obsoleta exhibits detorsión (Fig. 
5D) (Gohar & Soliman, 1967d), with the anal 
cells and associated organs migrating poste- 
riorly (with the hind gut) during metamorpho- 
sis, until they reach their final position in the 
middle line just in front of the secondary larval 
kidney. They persist for 7-10 days after 
hatching and eventually vanish. This differs 
from Bonar's (1976) report that the anal cells 
disappear by the time the secondary kidneys 
develop. 

The mouth in some cases forms shortly be- 
fore the complete closure of the blastopore. In 
a number of species {Dendrodoris fumata 
(Fig. 4A), Chiromodoris inornata) the second- 
ary larval kidney develops as early as the anal 
cells (Gohar & Soliman, 1967a,b). It attains its 
maximal development structurally and func- 
tionally during larval life. It is still encountered 
in the hatching juvenile of Casella obsoleta, 
but gradually diminishes in size and disap- 
pears in three weeks, i.e. after the disappear- 
ance of the anal cells. 



Hatching and larval behaviour 

At hatching, the whole upper wall of the 
neritid capsule detaches, thus liberating the 
larvae (Soliman, 1987). In neogastropods in 
particular (and certain mesogastropods, e.g. 
Tonna olearium (Gohar & Eisawy, 1967a)), 



260 



SOLIMAN 



TABLE 1. Breeding season, size, number and type of development of eggs of Red Sea prosobranchs 
and opisthobranchs. 









Number of 


Max. 






Dimension(s) of 


eggs per 


number 




Breeding 


egg-case or 


case or 


of eggs 


Species 


season 


capsule (mm) 


capsule 


deposited 


PROSOBRANCHS 










Trochus erythraeus 


May-Aug 


0.124-0.15 

X 0.148-0.17 


1-2 


1 1 ,200 


Trochas dentatus 


Apr-Jui 


0.4,0.48 X 0.43 av. 




— 


Turbo radiatus 


Feb-IVlay 


— 




— 


Nerita forskali 


Jan-Oct 


2.2-2.5 


60-210 


— 


Lambis trúncala' 


Apr-Jul 


0.3 




21,750 


Strombus tricornis 


May-Aug 


1.2-1.3 




2,800 


Strombus gibberulus 


— 


— 




— 


St rom bus fasciatus 


— 


— 




157,000 


Polinices mammilla ' 


Aug-Sep 


0.135 




580,000 


Polinices melanostoma' 


— 


— 




61 ,000 


Tonna olearium' 


Aug-Sep 


1.4-1.9 X 1-1.5 


18-35 


— 


Chicoreus virgineus' 


Apr-Jul 


12-20x10-12 


1036-2511 


625,000 


Ctiicoreus ramosus' 


Apr-May 


15-21 x5-7 


300-346 


13,450 


Nassa francolina' 


Jul-Dec 


8-9x3.5 


1390-1723 


1,720 


Thais savignyi 


Aug-Nov 


3.5-3.7 X 2.2-3 


250-500 


15,000 


Leptoconchus cumingii 


Feb-Nov 


6-9 X 3.7-6.3 


600-1600 


— 


Leptoconchus globosus 


Feb-Nov 


7-8 X 5-6 


700-1800 


1 1 ,250 


Magilopsis lamarckii 


Feb-Nov 


5-7x3.5-4.5 


500-1400 


— 


Pleuroploca trapezium' 


Apr-May 


19-35x5-9 


70-400 


70,000 


Fusinus tuberculatus' 


Feb-May 


11x7 


— 


— 


Conus sp. 


— 


— 


— 


— 


Conus sp. 


— 


— 


— 


— 


OPISTHOBRANCHS 










Aplysia dactylomela 


Apr-Oct 


— 


4-7 


> 1,000,000 


Dolabella auhcularia 


Apr-Oct 


— 


1 


>5,000,000 


Berthellina citrina 


Annual 


0.38-0.44 


1 


23,760 


Phyllobranchillus 


Jun-? 


0.09 


1 


117,000 


orientalis 










Elysia olivaceus 


May-Aug 


— 


— 


— 


Nembrotha limaciformis 


Jun-Aug 


0.09 X 0.075 


1 


75,600 


Gymnodoris sp. 


Jun-Sep 


0.19 X 0.22 av. 


1 


61,500 


Hexabranchus 


Annual 


0.3-0.7, 


4-30 


4,063,500 


sanguineus 




0.6-1 X 0.2-0.4 
up to 2 xO.6 


100 




Chromodoris quadricolor 


Mar-Sep 


0.1-0.135 




68,400 


Chromodoris pulchella 


Mar-Apr 


— 




48,000 


Chromodoris annulate 


Jul-? 


— 




108,000 


Chromodoris ghardaqana 


— 


— 




— 


Chromodoris i nor nata 


May-Nov 


0.1 




161,000 


Chromodoris tinctoria 


Jun-Jul 


0.15 


1-2 


62,000 


Chromodoris pallida 


Jun-? 


— 


— 


— 


Casella atromarginata 


Jun-Aug 


0.25 




188,000 


Casella obsoleta 


May-Sep 


0.52-0.58 




1,900 


Asteronotus cespitosus 


May-Sep 


0.18-0.24 


7-10 


4,885,650 


Platydoris scabra 


May-Sep 


0.16x0.23 




1,507,520 


Discodoris erythraeensis 


Feb-Sep 


0.24-0.25 




56,100 


Discodoris concinna 


May-Sep 


0.11-0.126 




3,796,700 


Discodoris sp. 


Jul-? 


— 




21,760 


Trippa areolata 


May-Nov 


0.11 




4,113,500 


Trippa spongiosa 


Jun-? 


0.26-0.3 




103,600 


Sebadoris crosslandi 


Jun-Jul 


0.12 




902,000 


Dendrodoris fumata 


Annual 


0.11-0.135 




165,000 


Phyllidia varicosa 


Jul-Oct 


0.2 




67,000 


Phyllodesmium xeniae 


May-Oct 


0.13-0.14 




1 1 ,200 



'Nomenclature updated. 



DEVELOPMENT OF RED SEA GASTROPODS 



261 





Period to 








Egg 


veliger 


Temperature 


Type of 




diameter 


forrлation 


of culture 


develop- 




(f^m) 


(d) 


ГС) 


ment''' 


Reference 


75 


3-4 


28 


P 


Gohar & Eisawy, 1963 


200-225 


24 h 


27.5 


L 


Eisawy, 1970 


200 


8-9 


26 


L 


Eisawy & Serial, 1974a 


120 


— 


— 


P 


Pers. obs. 


210-260 


7 


26 


L 


Gohar & Eisawy, 1967a 


410-440 


10-11 


28 


L 


Eisawy & Sohal, 1968 


90 


3-4 


28 


P 


Eisawy & Serial, 1976b 


130 


— 


— 


P 


Eisawy & Serial, 1976b 


— 


8 


— 


P 


Gohar & Eisawy, 1967a 


— 


— 


— 


P 


Gohar & Eisawy, 1967a 


240-250 


25 


25 


L 


Gohar & Eisawy, 1967a 


180-200 


40-45 


26.5 


L 


Gohar & Eisawy, 1967b 


250 


35-38 


25 


L 


Gohar & Eisawy, 1967b 


180 


30-32 


20 


P 


Gohar & Eisawy, 1967b 


185-190 


30-60 


28-21 


L 


Eisawy & Serial, 1974b 


200 


— 


— 


P 


Gohar & Soliman, 1963a 


— 


— 


— 


P 


Gohar & Soliman, 1963a 


— 


— 


— 


P 


Gohar & Soliman, 1963a 


400-420 


45-47 


24 


L 


Gohar & Eisawy, 1967b 


180-200 


30-50 


22-17 


L 


Eisawy & Serial, 1976a 


— 


— 


— 


P 


Gohar & Eisawy, 1967b 


— 


— 


— 


L 


Gohar & Eisawy, 1967b 


80 


7-10 


22-30 


P 


Pers. obs. 


92 


8-11 


22-30 


P 


Pers. obs. 


200-250 


7 


28.5 


L 


Gohar & Aboul-Ela, 1957a 


60 


7V2 


27 


P 


Pers. obs. 











P 


Pers. obs. 


65 


4V2 


28 


P 


Pers. obs. 


140 


8 


26 


P 


Pers. obs. 


110-120 


6-12 


30-16 


P 


Gohar & Soliman, 1963b 


70-90 


12 


18 


P 


Gohar & Aboul-Ela, 1957c 


120-170 


6 


20.5 


P 


Gohar & Aboul-Ela, 1957c 


120-160 


бУг 


29 


P 


Gohar & Aboul-Ela, 1957c 


— 


— 


— 


— 


Gohar & Aboul-Ela, 1957c 


80 


7-9 Уг 


30-22 


P 


Gohar & Soliman, 1967b 


100 


6-6У2 


29-27.4 


P 


Gohar & Soliman, 1967c 


— 


— 


— 


P 


Pers. obs. 


130-140 


10 


25 


P 


Gohar & Aboul-Ela, 1959 


300-330 


13-14 


25 


D 


Gohar & Soliman, 1967d 


65-70 


4-5У2 


27.8-24.8 


P 


Gohar & Soliman, 1967e 


90-100 


5-5У2 


27.6 


P 


Soliman, 1978 


140 


11Уг 


19 


L 


Gohar & Aboul-Ela, 1959 


75 


4У2 


27.5 


P 


Gohar & Soliman, 1967f 


93 


7У2 


29 


P 


Pers. obs. 


100 


4-5У4 


27.4-23.9 


P 


Gohar & Soliman, 1967g 


200 


7 


27 


L 


Gohar & Soliman, 1967g 


70 


7 


27.6 


P 


Soliman, 1980 


100 


5У^17 


30-16 


P 


Gohar & Soliman, 1967a 


100 


10 


27.5 


P 


Soliman, 1986 


95 


4 


28.5 


P 


Gohar & Aboul-Ela, 1957b 



^P, planktotrophic; L, lecithotrophic; D, direct development. 



262 



SOLIMAN 



В 




50 \im 



FIG. 2. Early embryology of Red Sea gastropods. A. Hexaoranchus sanguineus: spiral cleavage, first 
quartette of micromeres formed (after Gohar & Soliman, 1963b). B. Chicoreus virgineus: unequal cleavage 
of egg in a lecithotrophic prosobranch (after Gohar & Eisawy, 1967b). С Hexabranchus sanguineus: for- 
mation of mesoderm mother cell (the mesoblast 4d) (after Gohar & Soliman, 1963b). 



the egg capsule has an exit hole with fixed 
shape and position. The exit hole remains 
closed throughout development but opens at 
hatching to permit the release of larvae or 
succeeding stages. Empty capsules remain 



intact with firm walls. No comparable exit 
holes are encountered in opisthobranchs. In 
the Red Sea opisthobranch species studied, 
as development proceeds, the embryonic 
capsules gradually increase in size becoming 



DEVELOPMENT OF RED SEA GASTROPODS 



263 



turgid with extremely thin walls (apparently 
due to increased intracapsular osmotic pres- 
sure). This allows for an easy penetration by 
the hatching stages. In only a few cases did 
the perforated capsules retain their contour. 
Generally, however, they become collapsed, 
deformed or entirely ruptured and so can be 
barely detected in the jelly matrix. This latter 
In turn may either remain intact, become wrin- 
kled, dissociate into fragments or be con- 
verted into thick mucus. 

The wide temperature range (16-30°C) 
and high salinity (around 40%o) of Red Sea 
waters directly affect the development and 
larval behaviour of the gastropods studied. In 
species with extended breeding, the develop- 
mental time varies much with temperature 
(e.g. Thais savignyi, 30 d at 28°C, 38 d at 
26°C and 45 d at 24.2°C (Eisawy & Serial, 
1974b); Fusinus tuberculatus, 30-32 d at 
27°C and 45-50 d at 22°C (Eisawy & Serial, 
1976a); Hexabranchus sanguineus, 6 d at 
27°C and 10 d at 23.5°C (Gohar & Soliman, 
1963b); Dendrodoris fumata, бУг d at 28°C 
and 17 d at 17°C (Gohar & Soliman, 1967a)). 
The latter species is interesting since the 
length of the developmental period varies with 
the slight thermal changes within the same 
month: 152 h at 26.2°C, 156 h at 25.6°C and 
164 h at 25°C. Being a shore species subject 
to substantial fluctuations in temperature and 
salinity, its larvae display remarkable toler- 
ance to salinity changes (surviving for several 
days in 30%o and for 40 h in 50%o). 

At hatching, planktotrophic larvae develop 
for some time marked positive phototaxis, 
pursuing phytoplankton for food, and negative 
geotaxis to effect larval dispersal. Thereafter 
they become positively geotactic, invariably 
moving near the bottom but without displaying 
any tendency to metamorphose for reasons 
discussed later. Lecithotrophic larvae may re- 
main planktonic for days or hours; they even- 
tually settle and metamorphose. The newly 
emerging juveniles of Casella obsoleta, like 
the adults, are nocturnal in habit (Gohar & 
Soliman, 1967d). 

Light has a decisive effect on the degree of 
pigmentation of the veliger shell, and in turn 
on the general colour of the spawn mass. 
Even in the same ribbon of D. fumata, parts 
exposed to more light appear darker (Gohar & 
Soliman, 1967a). Light affects the whole pro- 
cess of development: in the total absence of 
light it has been shown experimentally that 
development is retarded or completely inhib- 
ited. 



Larval Structure 

The velum, foot and shell are among the 
most conspicuous gastropod larval organs 
which can be of outstanding taxonomic value 
in prosobranchs. 

The enlargement and subdivision of the ve- 
lum into 4, 6 or more lobes is a common char- 
acter of large prosobranch larvae (of ad- 
vanced mesogastropods and neogastropods) 
which helps them meet their needs for buoy- 
ancy and food (Fig. ЗА). A large or subdivided 
velum is not an indication of a long planktonic 
existence as is sometimes stated (Gohar & 
Eisawy, 1967a, in the case of Polinices mel- 
anostoma). Many such larvae have only a 
short pelagic life, becoming benthic one or 
two days after hatching; the velar lobes are 
eventually resorbed (e.g. Strombus tricornis 
(Eisawy & Serial, 1968), Chicoreus ramosus 
(Fig. 3B, C), Pleuroploca trapezium (Gohar & 
Eisawy, 1967b), Fusinus tuberculata (Eisawy 
& Serial, 1976a)). A multilobed velum is re- 
ported to exist in the larvae of only one 
opisthobranch, Philine denticulata (Horikoshi, 
1967). In lecithotrophic and directly-devel- 
oping opisthobranchs in general, the velum is 
relatively reduced in size and mobility. During 
metamorphosis, it may take part in the forma- 
tion of the juvenile rhinophores {Casella ob- 
soleta, Gohar & Soliman, 1967d). 

A pedal operculum does not form in di- 
rectly-developing opisthobranchs. It is lost in 
early juvenile development in aplysiids (Swit- 
zer-Dunlap & Hadfield, 1977), or during meta- 
morphosis in all other opisthobranchs, but in 
prosobranchs it is only lost in a few non-oper- 
culate species. 

With only a few exceptions, the larval shell 
is dextral in prosobranchs, with IV2 to 3 
whorls {Chicoreus ramosus, Gohar & Eisawy, 
1967b). In opisthobranch larvae, it may be 
cup-shaped, inflated or incipiently sinistrally 
coiled (hyperstrophic) with 3/4-1У2 whorls. 
Anomalous larvae possessing large tubular 
uncoiled shells are commonly observed in 
those opisthobranchs laying millions of eggs 
{H. sanguineus, Asteronotus cespitosus (Go- 
har & Soliman, 1963b, 1967e), Platydoris 
scabra (Soliman, 1978)). While the larval 
shell is retained in shelled opisthobranchs, it 
is cast off during metamorphosis in the re- 
maining opisthobranch groups. 

Except for Casella obsoleta and Phyllodes- 
mium xeniae (Gohar & Aboul-Ela, 1 957b), the 
veliger shells of all other Red Sea opistho- 
branchs studied belong to type В of Vester- 



264 



SOLIMAN 




1 mm 

FIG. 3. Larval structure and development of Red Sea prosobranchs. A. Lambis truncata: veliger larva with 
6-lobed velum (after Gohar & Eisawy, 1967a). B. Chicoreus ramosus: newly hatched veliger with 4-lobed 
velum (after Gohar & Eisawy, 1967b). С Chicoreus ramosus: postlarva showing degeneration of velum 
during metamorphosis (after Gohar & Eisawy, 1967b). VI, velum. 



gaard & Thorson (1938) and Thorson (1946). 
While the larval shell type of P. xeniae was not 
reported, that of С obsoleta is of type A. The 
validity of the latter type was a nnatter of 
controversy (Soliman, 1977). It has been re- 
jected by Thompson (1961) on the basis of 
its possession only by premature abnormal 
larvae. Its occurrence in the directly develop- 
ing species С obsoleta and Glossodoris si- 
bogae (Usuki, 1967) was considered as evi- 
dence for the view that such larval shells are 
vestigial, pertaining only to capsular devel- 
opment (Hadfield & Switzer-Dunlap, 1984). 
However, cup-shaped larval shells have been 



recently reported from planktonic veligers of 
two lecithotrophlc gymnodorid nudibranchs 
(Boucher, 1986). The still rare occurrence of 
this type of larval shell and its primitive con- 
struction do not preclude its recognition as a 
valid type. 

The variable sculpture and shape of proso- 
branch larval shells can provide a basis for 
their identification, but this is not possible with 
opisthobranch larval shells. Among these, 
only a few have roughened surfaces and they 
rarely have characteristic patterns (Hurst, 
1967). Colour and exact measurements can 
nevertheless be reliable characters in certain 



DEVELOPMENT OF RED SEA GASTROPODS 



265 




50 pm 



FIG. 4. Embryology and larval development of Red Sea opisthobranchs. A. Dendrodoris fumata: early 
formation of secondary (larval) kidney in gastrula; anal cells in pretorsional position (after Gohar & Soliman, 
1967a). B. Dendrodoris fumata: 7Уг d old embryo. Note excretory structures: nephrocyst, secondary (larval) 
kidney, and excretory vesicles (after Gohar & Soliman, 1967a). C. Hexabranchus sanguineus: newly 
hatched larva. Note secondary (larval) kidney discharging a large droplet of fluid (modified after Gohar & 
Soliman, 1963b). D. Chromodohs inornata: 7 d old embryo. Note larval kidney and excretory vesicles with 
attenuated blunt ends (after Gohar & Soliman, 1967b). E. Chromodoris inornata: newly hatched veliger with 
separated excretory vesicles discharging droplets of excretory fluid (after Gohar & Soliman, 1967b). AC, 
anal cells; H, heart; K, secondary (larval) kidney; M, midgut; MD, midgut diverticula; N, nephrocyst; Vc, 
excretory vesicle(s). 



266 



SOLIMAN 



cases (Gohar & Soliman, 1967g; Soliman, 
1978). 

As with the early embryological stages, live 
opisthobranch veligers are ideal material for 
studying the internal structure of gastropod 
larvae, e.g. gut, midgut diverticula, retractor 
muscle, heart, excretory and nervous ele- 
ments. A heart is said to exist only occasion- 
ally in opisthobranch larvae and to have been 
reported among nudibranchs only for Adalaria 
próxima (Bonar, 1978). In the present mate- 
rial, a pulsatile heart has been described in 
the nudibranchs Hexabranchus sanguineus, 
Dendrodoris fumata, Chromodons inornata 
and Casella obsoleta (with 20-21 beats. min- 
1 in the latter) (Gohar & Soliman, 1963b, 
1967a,b,d) (Figs. 4, 5D). 

Among the conspicuous larval excretory 
structures in many nudibranch species are 
the nephrocysts (symmetrically placed on the 
anterolateral aspect), the secondary larval 
kidney, and the large excretory vesicles (lo- 
cated on the right side in the close neighbour- 
hood of the kidney (Fig. 4)). The larval kidney 
of H. sanguineus is highly distinctive by its 
deep red colour, and clearly has a neck and 
aperture through which fluid drops are dis- 
charged (Fig. 4C). The larval kidney seems to 
function not only during embryonic and larval 
life, but also for some time after the juvenile 
stage is attained (Casella obsoleta, Fig. 5D; 
Philine denticulata, Horikoshi, 1967). Very lit- 
tle is known about the excretory vesicles, but 
the extrusion of hyaline droplets in certain 
cases (Chromodoris inornata. Fig. 4E) sug- 
gests they may have an excretory function. 

Eyes and tentacles are typical of proso- 
branch larvae. Their presence in newly 
hatched opisthobranch planktotrophic velig- 
ers is very unusual (Thorson, 1946). They de- 
velop 6 days after hatching in Phyllodesmium 
xeniae (pers. obs.), and some time during the 
larval phase in aplysiids (Switzer-Dunlap & 



Hadfield, 1977). Some cephalaspids hatch 
with only the right eye present {Acteocina 
canaliculata (Franz, 1971)), the left eye de- 
veloping a few days later. Eyes are, however, 
discernible in the veliger stages of leci- 
thotrophic and directly developing species 
{Berttiellina citrina, Discodoris erythraeensis 
(Gohar & Aboul-Ela, 1957a, 1959), Trippa 
spongiosa, Casella obsoleta (Gohar & Soli- 
man, 1967g,d)) (Fig. 5). 

The statocysts develop earlier than the 
eyes. They are virtually the eartiest embryonic 
nervous elements to develop during gastro- 
pod ontogeny and are retained in adult life. 



DISCUSSION 

Based on the studies of Thorson (1946, 
1950), Thompson (1967), Mileikovsky (1971) 
and Todd (1981) and data of the present 
study (Table 1), the main types of develop- 
mental patterns among gastropods (applica- 
ble also to other molluscs) are: 

1 . Planktotrophic development, with typical 
veliger larvae feeding during their short 
or long pelagic existence; 

2. Lecithotrophic development, which may 
be pelagic or non-pelagic; and 

3. Direct or capsular development. 

Each developmental type is correlated with 
a specific egg size range. In the present ma- 
terial the egg diameter range for the three 
types was 60-80 ^JLm, 140-440 |хт and 
300-330 |jLm, respectively. This last figure for 
direct development is, however, based on in- 
adequate data (just a single species, Casella 
obsoleta). Certain factors may, however, in- 
tervene allowing relatively small eggs to go 
through lecithotrophic or direct development, 
e.g. rich yolk content, rich albumen content of 



FIG. 5. Metamorphosis of Red Sea non-planktotrophic opisthobranchs with selected stages of development. 
A. Discodoris erythraeensis: pelagic lecithotrophy in a dorid nudibranch (modified after Gohar & Aboul-Ela, 
1959). From left to right: 7 d old embryo: metamorphosing postlarva with reflected mantle fold and no shell; 
juvenile. B. Berthellina citrina: non-pelagic lecithotrophy in a notaspidean (modified after Gohar & Aboul-Ela, 
1957a). From left to right: intracapsular veliger stage: newly hatched swimming-crawling stage; pediveliger 
with absorbed velum and enlarged foot; juvenile with internal shell. С Trippa spongiosa: non-pelagic leci- 
thotrophy in a dorid nudibranch (after Gohar & Soliman, 1967g). From left to right: intracapsular veliger 
stage; metamorphosing stage; hatching stage deserting its shell; juvenile. D. Casella obsoleta: direct de- 
velopment in a dorid nudibranch (after Gohar & Soliman, 1967d). From left to right: intracapsular veliger 
stage: metamorphosing stage with enlarged foot, subvelar ridge, and reflected mantle fold; embryo, 2 d 
before hatching, without velum or shell, with anal cells and larval kidney reaching their final detorsional 
position; juvenile. AC, anal cells: F, foot; H, heart; K, larval kidney; P, reflected mantle fold; R, rhinophore; 
S, shell; T, tentacle; Vc, excretory vesicle(s); VI, velum; Vs, subvelum. 



DEVELOPMENT OF RED SEA GASTROPODS 

СЛ1 



267 




268 



SOLIMAN 



exceptionally large egg capsules, extracapsu- 
lar yolk, or presence of nurse cells. Clark & 
Goetzfried (1978) report even smaller egg di- 
ameters of 91 .9 |a.m and 97.7 |j.m, for the as- 
coglossans Elysia papulosa and Costasiella 
lilianae which develop lecithotrophically and 
directly, respectively, being provided with ex- 
trazygotic food reserves. Extracapsular yolk 
has been previously described in the spawn 
ribbons of Chromodons tinctoria (Gohar & 
Soliman, 1967c). Here, although the egg is 
100 |j.m across, and extracapsular yolk has 
been shown to be almost depleted before 
hatching, yet lecithotrophic development was 
not encountered. Since egg masses were laid 
in the laboratory only during June and July, 
animals may proceed to lecithotrophic devel- 
opment at other periods of the year. The de- 
velopment of Elysia cauze (Clark & Goetz- 
fried, 1978) is seasonally variable and is 
apparently controlled by variable utilization of 
the extracapsular yolk. It is noticeable, how- 
ever, that in С tinctoria, the newly hatched 
larvae attained a relatively large size com- 
pared with larvae of other species developing 
from eggs of the same size but having no 
extrazygotic yolk. 

Lecithotrophic development involves a typ- 
ical veliger stage (which may be the hatching 
stage) that remains pelagic for a variable pe- 
riod of time, usually not exceeding two weeks 
(in Lambis trúncala (Gohar & Eisawy, 1967a) 
(Fig. ЗА); 7 d in Discodoris erythraeensis (Go- 
har & Aboul-Ela, 1959) (Fig. 5A); 4-6 d in 
Chicoreus ramosus (Fig. 3B); 2-3 d in C. vir- 
gineus (Gohar & Eisawy, 1967b); 2 d in 
Strombus tricornis and 1-2 d in Fusinus tu- 
berculatus (Eisawy & Serial, 1968, 1976a)). 
During their planktonic existence, which pri- 
marily effects their dispersal, the larvae may, 
but do not necessarily have to, feed. In non- 
pelagic lecithotrophic development, the 
veliger stage is passed intracapsularly, and 
on hatching already metamorphosing swim- 
crawling or crawling pediveligers are liberated 
which shortly attain the young stage {Berthell- 
ina citrina, Trippa spongiosa (Fig. 5B,C); Cu- 
thona nana (Rivest, 1978)). 

It is not uncommon nevertheless to have 
pelagic and non-pelagic lecithotrophy occur- 
ring in the same species (e.g. Chicoreus vir- 
gineus (Gohar & Eisawy, 1967b); Fusinus tu- 
berculatus (Eisawy & Serial, 1976a)). In such 
cases, while the majority of embryos hatch as 
proper planktonic larvae which start to meta- 
morphose 1-2 d later, a few, having their 
hatching delayed (possibly due to culture con- 



ditions), proceed in development intracapsu- 
larly emerging as creeping stages. 

In the third type, the whole development 
and metamorphosis takes place in the embry- 
onic capsule. The veliger stage is either nor- 
mal, although the velum is not well developed 
{Retusa obtusa (Smith, 1967); Phyllaplysia 
taylori (Bridges, 1975)), or is suppressed to 
varying degrees (Casella obsoleta (Fig. 5D); 
Cadlina laevis (Thompson, 1967)). 

Bonar (1978) designates the direct type of 
development with no proper veliger stage as 
ametamorphic, exemplified by the dorid 
Okadaia elegans described as having no 
trace of shell or velum during development 
(Baba, 1937). This is, however, different from 
the case of Casella obsoleta (included by Bo- 
nar among species with ametamorphic devel- 
opment). In this species the veliger stage pos- 
sesses a reduced but distinct velum, bearing 
short cilia and a subvelar ridge, and a cup- 
shaped shell (Fig. 5D). Because metamor- 
phosis does not only affect the locomotory 
and other external organs, but also (particu- 
larly in opisthobranchs) several internal or- 
gans including the gut and nervous elements, 
the use of the term 'ametamorphic' in this 
context is misleading as it implies that there is 
no process of metamorphosis. It should ap- 
propriately be replaced by 'incomplete' or 're- 
duced' metamorphosis (i.e. heterometamor- 
phic). Veliger stages with reduced velar lobes, 
meanwhile, are not restricted to directly de- 
veloping species, but have also been re- 
ported in lecithotrophic species {Cuthona 
nana (Rivest, 1978)) the ontogeny of which 
could equally be described as involving re- 
duced metamorphosis. 

From the above review, the major factors 
affecting metamorphosis are, in chronological 
sequence: food conditions, acquiring compe- 
tence for metamorphosis, and suitable sub- 
strata for settlement and metamorphosis. 
Planktotrophs and pelagic lecithotrophs pass 
through an obligatory planktonic (precompe- 
tent) phase for dispersal and feeding (essen- 
tial for the former category). Therefore, in lab- 
oratory cultures, such larvae should be 
supplied with suitable food to maintain their 
survival until after becoming competent to 
metamorphose. In Acteocina canaliculata, 
only the fed larvae normally metamorphose in 
culture (Franz, 1971; Mikkelsen & Mikkelsen, 
1984). Death of larvae, however, is not only a 
result of starvation but also of infection by 
bacteria and cillâtes. This has been success- 
fully controlled in laboratories by the use of 



DEVELOPMENT OF RED SEA GASTROPODS 



269 



selected antibiotics (Bonar & Hadfield, 1974; 
Hadfield, 1984), ultrafiltration, and/or boiling 
of sea water before use. Finally, a suitable 
substratum is essential for metamorphosis in 
some species. The proximal cue to settlement 
is probably the presence of specific chemicals 
which trigger the onset of metamorphosis 
(Bonar, 1976; Hadfield, 1984), but the advan- 
tage of this behaviour is that the settled mol- 
lusc has an assured supply of food. Prey or- 
ganisms of the adult have been frequently 
reported to be necessary to elicit meta- 
morphosis, while a particular alga must be 
provided to stimulate settlement and meta- 
morphosis in aplysiids (Switzer-Dunlap & 
Hadfield, 1977). The specific substratum 
could also be associated with certain individ- 
uals or may provide substances essential for 
adult life, beside affording optimal conditions 
for the species. In many species, however, 
planktotrophic and lecithotrophic larvae, after 
a period of pelagic existence, normally settle 
and metamorphose in the absence of a spe- 
cific substratum (e.g. Discodoris eryttiraeen- 
sis (Gohar & Aboul-Ela, 1959); Acteocina 
canaliculata (Franz, 1971); Pleuroploca tra- 
pezium (Gohar & Eisawy, 1967a), among 
others). 

Other defects in laboratory conditions can 
possibly also prohibit metamorphosis directly 
or indirectly. In the present study, rendering 
the adult's prey or substratum available to the 
postlarvae of several planktotrophic species 
(of which many had already become posi- 
tively geotactic) was unsuccessful in inducing 
metamorphosis. This involved the use of the 
definitive coral species bored by the adult in 
the case of coralliophilids, dead coral pieces 
for many dorids, the alcyonarian Sarcophy- 
tum in the case of Hexabranciius sanguineus 
(on which the adults feed, at least in part), and 
the alcyonarian ¡Heteroxenia among whose 
polyps the aeolid Ptiyllodesmium xeniae 
lives. Improving laboratory conditions can in- 
duce metamorphosis of such larvae develop- 
ing from large yolked eggs (e.g. Tonna olear- 
ium (Gohar & Eisawy, 1967a); Tliais savignyi 
(Eisawy & Serial, 1974b)) which othenwise die 
a few days (6-12) after hatching, and it can 
help metamorphosing larvae to complete this 
process successfully (e.g. Lambis trúncala, 
whose postlarvae often perish before attain- 
ing the young stage (Gohar & Eisawy, 
1967a)). 

Non-pelagic lecithotrophy and direct devel- 
opment are thus successful modes of mollus- 
can development having advantages over 



planktotrophy and pelagic lecithotrophy. For 
the former the dangers of free planktonic ex- 
istence (e.g. mortality due to prédation, star- 
vation, and drifting far from any suitable sub- 
stratum) are minimized, and no external 
source of food or a specific substratum for 
settling and metamorphosing is required. 
While pelagic lecithotrophic larvae have over- 
come the food crisis often faced by plank- 
totrophic larvae, they still share with them 
these other problems. The danger of failing to 
find a suitable settling ground is even more 
critical for them than for planktotrophic larvae, 
because their length of planktonic life is de- 
pendent upon their limited yolk supply (Smith, 
1967). Non-pelagic lecithotrophs and directly 
developing species nonetheless have the dis- 
advantages of limited distribution, possible 
overcrowding and genetic isolation. 

The present data agree in general with 
Thorson's rule (1950) that among benthic in- 
vertebrates there is an increase in species 
with pelagic larvae from the poles towards the 
tropics and equator. Accordingly, among the 
50 species of Red Sea gastropods whose on- 
togeny has been studied, 35 species (70%) 
have planktotrophic development, 14 are lec- 
ithotrophic and only one (a nudibranch) has 
direct development. The percentage of pe- 
lagic species would appear substantially 
higher if pelagic lecithotrophic species are 
taken into consideration. However, within the 
prosobranchs, the non-planktotrophic species 
represent a relatively high percentage, i.e. 
50% (previously also recorded in the Baha- 
mas (D'Asaro, 1970)), against only 14.3% for 
opisthobranchs. This may indicate a tendency 
to planktotrophy among Red Sea opistho- 
branchs, and to suppress planktonic in favour 
of non-pelagic development among neogas- 
tropods. There are, however, counter views 
which suggest that ecological conditions in 
the tropics (Florida, 17-32°C) favour direct 
development in nudibranchs and Ascoglossa 
(and probably all opisthobranchs) rather than 
planktotrophy (Clark & Goetzfried, 1978). 
While 87% of the nudibranchs studied from 
our area are planktotrophic, the limited num- 
ber of species and higher taxa examined do 
not permit arriving at firm conclusions on this 
point. 

LITERATURE CITED 

BABA, К., 1937, Contribution to the knowledge of a 
nudibranch, Okadaia elegans Baba. Japanese 
Journal of Zoology, 7: 147-190. 



270 



SOLIMAN 



BONAR, D. В., 1976, Molluscan metamorphosis: a 
study in tissue transformation. American Zoolo- 
gist. 16: 573-591. 

BONAR, D. В., 1978, Morphogenesis at meta- 
morphosis in opisthobranch molluscs. In: США, 
F. S. & M. E. RICE, eds., Settlement and meta- 
morphosis of marine invertebrate larvae. 
Elsevier North Holland Biomedical Press, New 
York, p. 177-1 96. 

BONAR, D. B. & M. G. HADFIELD, 1974, Metamor- 
phosis of the marine gastropod Phestilla sibogae 
Bergh (Nudibranchia: Aeolidacea). I. Light and 
electron microscopic analysis of larval and meta- 
morphic stages. Journal of Experimental tsarine 
Biology and Ecology. 16: 227-255. 

BOUCHER, L. M., 1986, Vestigial larval shells in 
planktonic veligers of two gymnodorid nudi- 
branchs. Journal of Molluscan Studies, 52: 30- 
34. 

BRIDGES, C, 1975, Larval development of Phyl- 
laplysia taylori Dall, with a discussion of develop- 
ment in the Anaspidea (Opisthobranchiata: 
Anaspidea). Ophelia, 14: 161-184. 

CLARK, K. B. & A. GOETZFRIED, 1978, Zoogeo- 
graphie influences on development patterns of 
North Atlantic Ascoglossa and Nudibranchia, 
with a discussion of factors affecting egg size and 
number. Journal of f^olluscan Studies, 44: 283- 
294. 

DASARO. С N., 1970, Egg capsules of proso- 
branch molluscs from South Florida and the Ba- 
hamas and notes on spawning in the laboratory. 
Bulletin of f^arine Science, 20: 414-440. 

EiSAWY, A. M., 1970, Spawning, development and 
metamorphosis of Trochus dentatus Forskal. 
Bulletin of the Institute of Oceanography and 
Fisheries. Egypt, 1 : 379-393. 

EISAWY, A. M. & A. E. SORIAL, 1968, The egg 
masses, development and metamorphosis of 
Strombus tricornis Lamarck. Proceedings of the 
l^/lalacological Society of London, 38: 13-26. 

EISAWY, A. M. & A. E. SORIAL, 1974a, The egg 
masses and development of Turbo radiatus 
Gmelin. Bulletin of the Institute of Oceanography 
and Fisheries, Egypt. 4: 221-235. 

EISAWY, A. M. & A. E. SORIAL, 1974b, Egg cap- 
sules and development of Thais savignyi De- 
shayes. Bulletin of the Institute of Oceanography 
and Fisheries. Egypt, 4: 237-258. 

EISAWY. A. M. & A. E. SORIAL, 1976a, Egg cap- 
sules and development of Fusus tuberculatus 
Chemnitz. Bulletin of the Institute of Oceanogra- 
phy and Fisheries, Egypt, 6: 215-237. 

EISAWY, A. M. & A. E. SORIAL, 1976b, Studies on 
the development of two species of Strombidae 
from the Red Sea. Bulletin of the Institute of 
Oceanography and Fisheries, Egypt, 6: 257-274. 

FRANZ, D. R., 1971, Development and metamor- 
phosis of the gastropod Acteocina canaliculata 
(Say). Transactions of the American Microscopi- 
cal Society, 90: 174- 182. 

GOHAR, H. A. F. & I. A. ABOUL-ELA, 1957a, The 
development of Berthellina citrina. Publications 



of the Marine Biological Station, Al-Ghardaqa, 
Egypt. 9: 69-84. 

GOHAR, H. A. F. & I. A. ABOUL-ELA, 1957b, On a 
new nudibranch, "Phyllodesmium xeniae". Pub- 
lications of the Marine Biological Station, Al- 
Ghardaqa, Egypt, 9: 131-144. 

GOHAR, H. A. F. & I. A. ABOUL-ELA, 1957c, The 
development of three chromodorids. Publications 
of the Marine Biological Station, Al-Ghardaqa, 
Egypt, 9: 203-228. 

GOHAR, H. A. F. & I. A. ABOUL-ELA, 1959, On the 
biology and development of three nudibranchs 
(from the Red Sea). Publications of the Mahne 
Biological Station, Al-Ghardaqa, Egypt, 10: 41- 
62. 

GOHAR, H. A. F. & A. M. EISAWY, 1963, The egg 
masses and development of Trochus erythraeus 
Brocchi. Publications of the Mahne Biological 
Station, Al-Ghardaqa, Egypt, 12: 191-204. 

GOHAR, H. A. F. & A. M. EISAWY, 1967a, The egg 
masses of four taenioglossan prosobranchs from 
the Red Sea. Publications of the Mahne Biolog- 
ical Station, Al-Ghardaqa, Egypt, 14: 109-148. 

GOHAR, H. A. F. & A. M. EISAWY, 1967b, The egg 
masses and development of five rachiglossan 
prosobranchs from the Red Sea. Publications of 
the Mahne Biological Station, Al-Ghardaqa, 
Egypt, 14: 215-268. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1963a, On the 
biology of three coralliophilids boring in living cor- 
als. Publications of the Manne Biological Station, 
Al-Ghardaqa, Egypt, 12: 99-126. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1 963b, On the 
biology and development of Hexabranchus san- 
guineus (Ruppell & Leuckart). Publications of the 
Mahne Biological Station, Al-Ghardaqa, Egypt, 
12: 219-248. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1967a, The 
biology and development of Dendrodoris fumata 
(Ruppell & Leuckart). Publications of the Marine 
Biological Station, Al-Ghardaqa, Egypt, 14: 31- 
54. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1967b, The 
biology and development of Chromodoris inor- 
nata Pease. Publications of the Mahne Biological 
Station, Al-Ghardaqa. Egypt, 14: 77-94. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1967c, The 
biology and development of Chromodoris tincto- 
ria (Ruppell & Leuckart). Publications of the Ma- 
nne Biological Station, Al-Ghardaqa, Egypt, 14: 
95-108. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1967d, The 
direct development of the nudibranch Casella ob- 
soleta (Ruppell & Leuckart). Publications of the 
Mahne Biological Station, Al-Ghardaqa, Egypt, 
14: 149-166. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1967e, The 
biology and development of Asteronotus cespito- 
sus (Van Hasselt). Publications of the Marine Bi- 
ological Station, Al-Ghardaqa, Egypt, 14: 177- 
196. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1967f, The 
biology and development of Discodoris concinna 



DEVELOPMENT OF RED SEA GASTROPODS 



271 



(Alder & Hancock). Publications of the Marine Bi- 
ological Station, Al-Ghardaqa, Egypt, 14: 197- 
214. 

GOHAR, H. A. F. & G. N. SOLIMAN, 1 967g, On two 
rare nudibranchs of the genus Trippa Bergh of 
different developmental behaviour. Publications 
of the Marine Biological Station, Al-Ghardaqa, 
Egypt, 14: 269- 293. 

HADFIELD, M. G., 1984, Settlement requirements 
of molluscan larvae: new data on chemical and 
genetic roles. Aquaculture, 39: 283-298. 

HADFIELD, M. G. & M. SWITZER-DUNLAP, 
Opisthobranchs. In: TOMPA, A. S., N. H. VER- 
DONK & J. A. M. van den BIGGELAAR, eds., 
Reproduction. The Mollusca (WILBUR, K. ed.) 
volume 7. Academic Press, Orlando & London, p. 
209-350. 

HORIKOSHI, M., 1967, Reproduction, larval fea- 
tures and life history of Philine denticulata (J Ad- 
ams) (Mollusca — Tectibranchia). Ophelia, 4: 43- 
84. 

HURST, A., 1967, The egg masses and veligers of 
thirty Northeast Pacific opisthobranchs. The 
Veliger, 9: 255-288. 

MacGINITIE, G. E., 1934, The egg laying activities 
of the sea hare Tethys calif ornicus (Cooper). Bi- 
ological Bulletin of the Marine Biological Labora- 
tory, Woods Hole, 67: 300-303. 

MIKKELSEN, P. S. & P. M. MIKKELSEN, 1984, A 
comparison of Acteocina canaliculata (Say, 
1826), A. cande/ (d'Orbigny, 1841) and A. atrata 
spec. nov. (Gastropoda: Cephalaspidea). The 
Veliger, 27: 164-192. 

MILEIKOVSKY, S. A., 1971, Types of lan/al devel- 
opment in marine bottom invertebrates, their dis- 
tribution and ecological significance: a re-evalu- 
ation. Marine Biology, Berlin, 10: 193-213. 

NATARAJAN, A. V., 1957, Studies on the egg 
masses and larval development of some proso- 
branchs from the Gulf of Mannar and the Palk 
Bay. Proceedings of the Indian Academy of Sci- 
ence, 46(B): 170-228. 

RIVEST, B. R., 1978, Development of the eolid 
nudibranch Cuthona nana (Alder & Hancock, 
1842) and its relationship with a hydroid and a 
hermit crab. Biological Bulletin of the Marine 
Biological Laboratory, Woods Hole, 154: 157- 
175. 

SMITH, S. T., 1967, The development of Retusa 
obtusa (Montagu) (Gastropoda, Opisthobran- 
chia). Canadian Journal of Zoology, 45: 737- 
764. 

SOLIMAN, G. N., 1977, A discussion of the sys- 
tems of classification of dorid nudibranch veliger 
shells and their taxonomic significance. Journal 
of Molluscan Studies, 43: 313-318. 

SOLIMAN, G. N., 1978, The redescription, repro- 
duction and development of the dorid nudibranch 
Platydoris scabra. Journal of Molluscan Studies, 
44: 151-165. 



SOLIMAN, G. N., 1980, On the dorid nudibranch 
Sebadoris crosslandi. Journal of Molluscan Stud- 
ies, 46: 227-238. 

SOLIMAN, G. N., 1983, The discodorid nudi- 
branchs of the northwestern Red Sea. Journal 
of Molluscan Studies, Supplement, 12a: 179- 
184. 

SOLIMAN, G. N., 1986, The reproduction and de- 
velopment of the dorid nudibranch Phyllidia var- 
icosa from the northwestern Red Sea. Proceed- 
ings of the 8th International Malacological 
Congress (Budapest, 1983): 255-260. 

SOLIMAN, G. N., 1987, A scheme for classifying 
gastropod egg masses with special reference to 
those from the northwestern Red Sea. Journal of 
Molluscan Studies, 53: 1-12. 

SWITZER-DUNLAP, M. & M. G. HADFIELD, 1977, 
Observations on development, larval growth and 
metamorphosis of four species of Aplysiidae 
(Gastropoda: Opisthobranchia) in laboratory cul- 
ture. Journal of Experimental fi/larine Biology and 
Ecology, 29: 245-261. 

THOMPSON, T E., 1961, The importance of the 
larval shell in the classification of the Sacoglossa 
and the Acoela (Gastropoda, Opisthobranchia). 
Proceedings of the Malacological Society of Lon- 
don, 34: 233-238. 

THOMPSON, T E., 1967, Direct development in a 
nudibranch, Cadlina laevis, with a discussion of 
developmental processes in Opisthobranchia. 
Journal of the Marine Biological Association, 
U.K., 47: 1-22. 

THORSON, G., 1940, Notes on the egg capsules of 
some North Atlantic prosobranchs of tfie genus 
Troschelia, Chrysodomus, Volutopsis, Sipho and 
Trophon. Videnskabelige Meddelelser fra dansk 
naturhistorisk Forening, 104: 251-265. 

THORSON, G. 1946, Reproduction and larval de- 
velopment of Danish marine bottom inverte- 
brates with special reference to the planktonic 
larvae in the Sound (Oresund). Meddelelser fra 
Kommissionen for Danmarks Fiskeri-og Ha- 
vundersegelser, Serie Plankton, 4: 1-523. 

THORSON, G., 1950, Reproduction and larval 
ecology of marine bottom invertebrates. Biologi- 
cal Reviews, 25: 1-45. 

TODD, С D., 1981, The ecology of nudibranch 
molluscs. Oceanography and Marine Biology, an 
Annual Review, 19: 141-234. 

USUKI, I., 1967, The direct development and the 
single cup-shaped larval shell of a nudibranch, 
Glossodoris sibogae (Bergh). Scientific Reports 
of Niigata University, (D) 4: 75-85. 

VESTERGAARD, K. & G. THORSON, 1938, Über 
den Laich und die Larven von Duvaucelia ple- 
beja, Polycera quadhlineata, Eubranchus palli- 
dus und Limapontia capitata (Gastropoda, 
Opisthobranchiata). Zoologischer Anzeiger, 124: 
129-138. 



MALACOLOGIA, 1991,32(2): 273-289 

LARVAL STRATEGIES OF NUDIBRANCH MOLLUSCS: 
SIMILAR MEANS TO THE SAME END? 

Christopher D. Todd 
University of St Andrews, Gatty Marine Laboratory, St Andrews, Fife KY16 8LB, U.K. 

ABSTRACT 

Growth, development and metamorphosis of the planktotrophic larvae of Onchidoris bilamel- 
lata (L.) were followed in the laboratory at a range of culture temperatures and on a variety of 
microalgal diets. Successful culture of the nudibranch larvae at 5°C (approximating field tem- 
peratures) indicated developmental periods in agreement with previous estimations. Thus, pre- 
dictions of the coincident timing of settlement and metamorphosis of O. bilamellata and the prey 
(Semibalanus balanoides (L.)) barnacles in the field are upheld. In this respect Todd & Doyle's 
(1981) "settlement-timing" hypothesis — as an explanation for the observed larval strategy of O. 
bilamellata — appears tenable. However, juveniles were observed to subsist on detritus for sev- 
eral weeks prior to their ability to prey upon S. balanoides. This precludes the validity of inferring 
a close matching of predator and prey settlement in explaining the specific larval strategy. In the 
light of other published data on larval/post-larval development, growth and feeding an alternative 
hypothesis, concerned with selective opportunities' in the evolution of nudibranch larval strat- 
egies, is outlined. 



INTRODUCTION 

It is apparent, for a wide variety of marine 
invertebrate phyla (Strathnnann, 1978,1985), 
that reproduction by means of small eggs giv- 
ing rise to planktotrophic larvae is the primi- 
tive (or ancestral) mode of development, and 
that pelagic, and thence non-pelagic, leci- 
thotrophy are more advanced evolutionary 
derivatives. Evolutionary re-acquisition of lost 
larval feeding structures appears to be com- 
paratively infrequent and we are thus con- 
fronted with an essentially uni-directional se- 
quence of events. While there may remain 
debate over this generalization with respect to 
particular groups (even the Mollusca as a 
phylum), there can be little doubt (see Had- 
field & Switzer-Dunlap, 1984) that such a 
trend pertains to the subclass Opisthobran- 
chia. What remain to be ascertained, how- 
ever, are the selective factors that have dic- 
tated a shift away from planktotrophy (which 
requires a more or less extended period of 
feeding, growth and development of the pe- 
lagic larva prior to metamorphosis), toward 
lecithotrophy (in which all reserves necessary 
for the completion of development of the 
benthic juvenile are provided within the egg 
by the parent). At present a consensus ap- 
pears lacking (see Jablonski & Lutz, 1983; 
Day & McEdward, 1984; Hadfield & Switzer- 
Dunlap, 1984; Grahame & Branch, 1985; 



Todd, 1985a for reviews). Indeed, even if it Is 
accepted that non-feeding larval forms have 
evolved from planktotrophic counterparts, it is 
likely that the selective regimes for such evo- 
lutionary shifts will comprise 'special cases' 
for many particular species. 

In a previous study concerning the repro- 
ductive strategy of the Boreo-Arctic dorld 
nudibranch Onchiidoris bilamellata (L.) (Todd 
& Doyle, 1981), emphasis was placed on the 
possible significance of the differing egg-to- 
benthic juvenile periods conferred by the 
three fundamental larval strategies of plank- 
totrophy, pelagic lecithotrophy and non- 
pelagic lecithotrophy (or 'direct' develop- 
ment). Spawning of this specialist barnacle 
predator occurs at the coldest time of year 
and our hypothesis highlighted the approxi- 
mately 3V2-month time gap between peak 
spawning of the adult nudibranchs (January) 
and cyprid settlement of Semibalanus bal- 
anoides (L.) (May), the major prey species. 
Having a strictly annual and semelparous life- 
history, these nudibranchs have only the one 
reproductive opportunity: embryonic and/or 
larval failure is, therefore, absolute. Of central 
importance here is the presumably crucial co- 
incidence of newly-metamorphosed molluscs 
with the establishment of the smallest phase 
of the barnacle life-cycle, bearing in mind that 
settlement of barnacles is restricted (but vari- 
able) both spatially and temporally. Data then 



273 



274 



TODD 



available suggested that only long-term pe- 
lagic planktotrophy could bridge the gap be- 
tween the empirically predicted optimal time 
for adults to spawn and the observed optimal 
time for the veligers to settle and metamor- 
phose. Development data for other British 
dorids indicated that both the pelagic and 
non-pelagic lecithotrophic strategies would 
result in juvenile nudibranchs establishing on 
the shore some weeks in advance of the 
availability of post-metamorph barnacle prey. 
It was perhaps surprising to note that pelagic 
lecithotrophy resulted in by far the shortest 
egg-to-juvenile interval, and that non-pelagic 
development was intermediate in duration. 

Our settlement-timing hypothesis was 
based on a number of assumptions, foremost 
of which were the prediction of larval devel- 
opmental periods in the field (from artificial 
laboratory culture observations at higher tem- 
peratures), and the immediate dependence of 
juvenile O. bilamellata upon the smallest 
post-metamorphic barnacles as prey. Criti- 
cisms of the original report (Grant & William- 
son, 1985) were defended (Todd, 1985b) on 
the strength of available information while, in- 
dependently, yet further specific reservations 
as to the validity of the principle were also 
being expressed (e.g. Hadfield & Switzer- 
Dunlap, 1984; Strathmann et al., 1984). Prior 
to this, more detailed evaluations of the lar- 
val and post-larval development of O. bila- 
mellata had been initiated, and it is upon 
these that I report here. My primary objec- 
tives were to determine the duration of the 
pelagic phase at ambient field temperatures 
(= 5°C), and thereby to obtain further obser- 
vations of post-larval dependence on barna- 
cle spat. (Previous culturing had resulted in 
post-metamorph dorids in advance of S. bal- 
anoides settlement in the field.). I also exam- 
ined the developmental effects of particular 
algal dietary species on growth and meta- 
morphosis success in the laboratory. Culture 
conditions are by definition artificial (espe- 
cially in terms of algal species and concentra- 
tions thereof) and the pitfalls of extrapolation 
to the field situation are self-evident. Never- 
theless, by rearing larvae on a range of mono- 
cultures and combinations of phytoplankters 
ecologically realistic predictions may be 
made. 

I show here that the original hypothesis 
does not provide an adequate descriptor of 
the behaviour of the nudibranch, but offer an 
alternative model for nudibranchs which cen- 
tres upon larval and post-larval nutrition. It 



should be emphasized that this hypothesis 
explains only certain adaptive features of 
nudibranch reproductive strategies. It does 
not in itself provide an all-embracing frame- 
work of selection for particular larval types, 
which (if tenable) would also require an ap- 
praisal of the bioenergetic constraints and 
genetic implications of particular life-cycle, 
life-history and larval strategies (Todd, 1985a, 
1987; Havenhand & Todd, 1988a,b,c; Todd & 
Havenhand, 1988, 1990; Todd et al., 1989). 



MATERIALS AND METHODS 

Larval culture of Onchidoris bilamellata has 
been undertaken from spawn masses depos- 
ited in the laboratory by field-collected adults. 
Throughout spawning, the nudibranchs were 
maintained on Semibalanus balanoides at 
ambient field temperatures. Culture methods 
are outlined elsewhere (Todd & Havenhand, 
1984), but the salient features are summa- 
rized here. All flagellate larval diets were 
raised from inocula obtained from the Cam- 
bridge Collection of Algae and Protozoa, with 
the exception of Isochrysis galbana Parke, 
which was supplied by SMBA, Oban. Other 
algae employed were Rhodomonas sp., Pav- 
lova lutheri (Droop) and Tetraselmis sp. Lar- 
vae were reared in glass beakers of 250- 
1000 ml volume, according to the number of 
veligers present, at an approximate concen- 
tration of 3 larvae. ml-1 and a total algal con- 
centration of 50 cells ^JLl \ Where mixtures of 
algal species were provided as the larval diet 
these were presented in equal numerical pro- 
portions. This is particularly pertinent for 
Rhodomonas which, at approximately 15 ^JLm 
diameter, is much the largest of the species 
used. Larvae were cultured in 0.22 \xm filtered 
seawater with the antibiotics Streptomycin 
sulphate and Penicillin G (50 and 60 jxg.ml ^ 
respectively) added to control any bacterial 
and ciliate infestations. Cultures were 
changed every 5 days by concentrating the 
veligers in a mesh-bottomed filter and pipet- 
ting them into freshly prepared beakers. All 
items of glassware were washed in hot fresh 
water only, or autoclaved, and at no time were 
detergents or disinfectants used. 

Temperatures of the larval cultures (5, 10, 
15 and 18°C) were controlled to within 0.2°C 
by immersing the vessels in thermostatically 
controlled water-baths and all were subjected 
to constant illumination in order to preclude 
possible complications of variable photope- 



LARVAL STRATEGIES OF NUDIBRANCHS 



275 




5 15 25 35 45 55 65 75 

DAYS POST- HATCHING 

FIG. 1 . Growth and development of the planktotrophic larvae of Onchidoris bilamellata in laboratory culture 
on a 1 :1 :1 mixture of the flagellates Rhodomonas, Isochrysis and Pavlova at 5°C and 15°C. Mean (with 95% 
confidence limits), maximum and minimum shell lengths are shown at each change of medium. Occurrence 
of first metamorphosis is indicated by arrows. 



riod. The flagellate diets were, however, all 
raised at room temperature (== 20°C), a factor 
which may be of some importance to the 
colder larval cultures (see below). 

In view of our previous lack of success in 
precluding 'rafting' of the larvae In the surface 
film (Todd, 1981; Todd & Havenhand, 1984), 
no such preventive steps were taken. Instead, 
entrapped larvae were resuspended by pipet- 
ting small drops of water onto the culture sur- 
faces; this was undertaken at least once daily. 
'Rafting' was usually only a problem in the 
early periods of development, when upward 
swimming is particularly marked. Periodic 
measurements of larvae were made to the 
nearest 3 |xm with between 10 and 20 
veligers sampled for this purpose. All mea- 
surements are of the maximum shell dimen- 
sion taken from the aperture lip. 'Compe- 
tence' to metamorphose is visibly assessable 
from the overall size of the veliger, the pres- 
ence of eyes, the larval 'heart', a well-formed 
propodium, and the ability of the larva to crawl 
across a substratum. Once pediveligers were 
detected within a culture, subsamples were 
removed to glass dishes containing live adult 
Semibalanus balanoides (or S. crenatus (Bru- 
guière)). If competent, the pediveligers gen- 



erally commenced metamorphosis within a 
few hours. Further observations of post- 
larvae were made by maintaining these at 8°C 
(12 h light, 12 h dark) in plastic peth-dishes, 
within which field-collected cyprids of S. bal- 
anoides had previously been triggered to 
metamorphose. 

The SEM preparations of S. balanoides 
plates were critical-point dried and gold sput- 
ter-coated in the standard manner, and were 
photographed on a JEOL JSM-35CF scan- 
ning electron microscope. 



RESULTS 

The initial success in rearing the larvae of 
this nudibranch species (Todd, 1981) was ob- 
tained in cultures containing a 1:1:1:1 mix of 
Isochrysis, Pavlova, Rhodomonas and Tetra- 
selmis. Subsequently only the first three 
flagellates have been used, either as mono- 
cultures or equal mixtures (total concentration 
50 cells |хГ^). The selected algal concen- 
trations appear to promote maximal growth 
and metamorphosis success in a range of 
opisthobranch species (e.g. Chia & Koss, 
1978; Bickell & Kempf, 1983). Fig. 1 shows 



276 



TODD 



the growth and development of O. bilamellata 
larvae maintained on 1:1:1 mixtures of Iso- 
chrysis. Pavlova and Rhodomonas at 5°C and 
15°C: several striking similarities and con- 
trasts are apparent. First, larvae maintained 
at both temperatures showed qualitatively 
similar sigmoid growth curves (a feature prob- 
ably characteristic of opisthobranchs [see e.g. 
Perron & Turner, 1977; Bickell & Chia, 1979; 
Bickell & Kempf, 1983], bearing in mind the 
geometry of shell growth), and attained simi- 
lar sizes at metamorphosis. (A cessation of 
shell, but not tissue, growth is generally noted 
some days before development of the propo- 
dium, and competence to metamorphose.) 
Second, larvae at the lower temperature (ap- 
proximating to ambient field conditions (see 
Todd, 1985b)) developed at a very much 
slower rate: 79 days at 5°C versus 31 days at 
15°C. Third, in both cases there is consider- 
able variation in size at a given age, and 
hence individual growth rates, within each 
culture. 

Between-culture variation in growth and 
successful completion of development is in- 
evitable (see e.g. Pechenik & Lima, 1 984, and 
references therein), particularly at lower tem- 
peratures where development is so pro- 
longed. But of greater concern, particularly 
when evaluating the efficacy of differing di- 
etary regimes, is the invariably high within- 
culture variation. This will be due, in part, to 
inherent differences amongst the larvae, but 
the major source of the variance is undoubt- 
edly experimental. Rafting is almost certainly 
responsible for much of the observed reduc- 
tion in growth for many larvae. For example, 
larvae reared on a mixture of Isochrysis and 
Rhodomonas show, at any age, marked di- 
vergences in overall size and in the colour of 
the left digestive diverticulum: small slow- 
growing larvae are invariably green, while the 
larger fast-growing larvae have dark-red di- 
gestive glands. The former are undoubtedly 
veligers which have persistently become en- 
trapped and which encounter difficulty in ob- 
taining sufficient food. Moreover, Rhodomo- 
nas, which is a larger, less motile flagellate, 
tends to precipitate to the bottom of still cul- 
tures and is of markedly reduced availability 
to rafted veligers. Nevertheless Rhodomonas 
alone can promote growth and development 
equal, or even superior, to that in mixtures 
(see below). Because development rates are 
usually expressed in terms of time to first 
metamorphosis, it is perhaps ecologically 
valid to compare growth in terms of the fastest 



growing individuals, rather than the notional 
'average individual'. Certainly, it is usual to 
note a rapid increase in minimum sizes, in the 
later stages of development, due to the de- 
mise of slower-growing individuals and/or 
those that persistently became entrapped. 

Fig. 2 summarizes the successful culture, 
through to metamorphosis, of O. bilamellata 
at a range of temperatures and on a variety of 
dietary regimes. It should be emphasized that 
the majority of cultures were attempted at 5°C 
and the presented data relate only to those 
cultures in which growth and development 
were seen to proceed normally'. Data for 
many other cultures in which survivorship, 
growth and metamorphosis success were 
considered unsatisfactory (or were not at- 
tained), have not been included. Attention 
should also be drawn to the extent of the pe- 
lagic phase: even assuming no inherent mor- 
tality, the fact that only = 90% of larvae can 
be successfully transferred at each change of 
culture medium results in an 80% loss of 
veligers over, say, 1 1 weeks of rearing. De- 
spite the incompleteness of the data sufficient 
observations are available to make some 
comment on the effects of both diet and tem- 
perature. 

Effects of algal diet 

In general, mixtures of flagellates, even of 
only Rhodomonas and Isochrysis, promoted 
the highest growth rates and greatest meta- 
morphosis success. Nevertheless, based 
upon these (and other) cultures it is apparent 
that Rhodomonas alone is almost equally ef- 
ficacious and, indeed, routine culture of meta- 
morphs is now undertaken on Rhodomonas 
monocultures. Even in mixtures, larvae evi- 
dently ingest and digest Rhodomonas more 
than other algae — veligers frequently regurgi- 
tate this flagellate when being measured on 
glass slides, and it colours the digestive di- 
verticula dark red. Whether the above obser- 
vations ahse from differential selection or 
availability remains unclear, although the 
former appears more likely. Two other obser- 
vations are perhaps substantive; first, larvae 
reared on monocultures of Pavlova (at all 
temperatures) never achieved competence, 
and second, larvae reared on Isochrysis 
alone were only successfully raised to meta- 
morphosis on one occasion (at 5°C). Larvae 
on Pavlova invariably grew well, but devel- 
oped very dark concretions in the left diges- 
tive diverticulum some time prior to death at a 



LARVAL STRATEGIES OF NUDIBRANCHS 



277 



1 RHODOMONAS, PAVLOVA, lEIRASELMlS, ISOCHRYSIS 
3 2 RHOOOMONAS, ISOCHRYSIS 
11 10 9 6 4 RHODOMONAS ONLY 

12 8 5 RHODOMONAS, ISOCHRYSIS, PAVLOVA 
7 ISOCHRYSIS ONLY 



90 



о 

z 

5 60 

< 

X 

о 

Û. 

СЛ 
< 30 

û 



Об 

7 



8 



09 



Оз 



02 



^ 



10 



15 



18 



О. 



TEMPERATURE, с 



FIG. 2. Time to first metamorphosis of Onchidoris bilamellata larvae in culture at a range of temperatures and 
dietary regimes. 



pre-connpetent stage. Such concretions were 
also observed in larvae from 'гл1хес1' cultures 
containing Pavlova, and for this reason use of 
this flagellate was discontinued. The general 
inadequacy of Isochrysis was manifest in sur- 
vivorship, growth and metamorphosis suc- 
cess. For Isochrysis monocultures at 15°C 
growth and development appeared to pro- 



ceed normally. Eyed veligers were noted after 
25 days of culture, cessation of shell growth 
between 25 and 30 days, and propodial de- 
velopment after 30 days. Larvae were reared 
for a further 31 days, during which time they 
were never observed to crawl and would 
not metamorphose. Moreover, many larvae 
(which had completed shell growth) continued 



278 



TODD 



-1.3 



■1.5 



"О 



О) 



-1.9 - 




Q,o= 2.40 



345 



V* 



355 



350 
FIG. 3. Arrhenius plot of the data from Fig. 2. The 0,0 is derived from the fitted regression equation. 



360 



to grow somatically, with the tissues finally 
bulging from the shell. At 10°C, on Isochrysis 
alone, the veligers grew well and rapidly, de- 
veloped eyes within 26 days, propodia within 
29 days and commenced crawling within 36 
days. Nevertheless, metamorphosis was 
never achieved and larvae either died or 
evacuated the shell spontaneously. It there- 
fore remains possible that Isochrysis is an ad- 
equate (if not ideal) diet at low temperature, 
but at higher temperatures its biochemical 
composition and nutritive value are radically 
or critically altered. 

Effects of temperature 

Fig. 3 shows the conventional Arrhenius 
plot for the data in Fig. 2. Several points re- 
quire emphasis here: first, no culture at a tem- 
perature below б^С proved successful. Sec- 
ond, only the one culture was attempted at 
18°C (a temperature probably at the physio- 
logical limit for this species (Todd, 1979a)), 
and third, there is considerable between-cul- 
ture variation — even at the same temperature 



and with the same diet (see, for example, Cul- 
tures 9 and 11, Fig. 2). Nevertheless, least 
squares regression analysis for these data 
(comprising a range of dietary regimes) 
shows a very significant relationship (P < 
.001), from which the Q^ is derived as 2.40. 
Undoubtedly a more complete data set is nec- 
essary to properly distinguish this quotient, 
but it is perhaps relevant that the Q^ br em- 
bryonic development for O. bilamellata (upon 
which basis the pelagic phase duration at field 
temperatures was previously predicted) was 
found to be 2.34 (Todd & Doyle, 1981). De- 
spite possible dietary inadequacy, it is clear 
that at field temperatures (= 5°C) the embry- 
onic and larval developmental phases would 
concur with the predictions made initially 
(Todd & Doyle, 1981), and that thus far the 
settlement-timing hypothesis appears at least 
tenable. That is, with peak spawning in mid- 
January, embryonic (pre-hatching) and pe- 
lagic larval development would require a total 
of (39 + 73 =) 112 days, resulting in peak 
larval settlement by early May. Our initial es- 
timation of the larval period in the field (based 



LARVAL STRATEGIES OF NUDIBRANCHS 



279 



on a cultured pelagic phase of 32 days at 
15°C, corrected by an observed embryonic 
Qio of 2.34) was 69 days. 

Observations of post-larval development 

Precise timetables for post-larval ontogeny 
and development toward the functional pred- 
atory juvenile cannot be ascertained; the fol- 
lowing is a composite summary of observa- 
tions of post-larvae reared on various larval 
diets. At the completion of metamorphosis 
(when the mantle is fused anteriorly and pos- 
teriorly, and the anus has adopted its defini- 
tive medial posterior position) the dorsum is 
both tuberculate and spiculate, but the rhino- 
phores and gills are lacking. Mantle length is 
~ 0.6 mm. The rhinophores are differentiated 
at 0.85-0.90 mm mantle length, but the gills 
only become evident at =1.7 mm length (ap- 
proximately one month post-metamorphosis). 

Throughout this first month of growth and 
continued development no prédation on bar- 
nacle cyprids or spat has been noted. Never- 
theless, juveniles were seen to be consis- 
tently (if not exclusively) associated with the 
lateral calcareous plates of juvenile S. bal- 
anoides, in a manner suggestive of thigmot- 
actic 'refuging' behaviour. Pigmentation of the 
digev^cive gland (attributable to the larval di- 
etary flagellates) is progressively lost over the 
first few weeks of benthic life and is suc- 
ceeded by a uniform green-brown coloration. 
This, in turn, becomes less visible as the 
mantle tissues thicken. Close inspection of 
the dorids showed that the buccal mass is, in 
fact, functional from very early on, and that 
rather than simply refuging against the barna- 
cles the juvenile nudibranchs were actually 
grazing on diatoms and/or detritus from the 
plate surfaces. Similar feeding was subse- 
quently observed on field-collected plates of 
dead adult S. balanoides, which are invariably 
colonized by a considerable microflora. How- 
ever, metamorphosis will only occur following 
contact with live barnacles. Figs. 4 and 5 il- 
lustrate a portion of the calcareous plates of a 
laboratory-metamorphosed S. balanoides, 
and here can be seen a radula tooth presum- 
ably lost by a juvenile O. bilamellata which 
had been consistently associated with that 
barnacle. 

The dorid Cadlina laevis undergoes non- 
pelagic lecithotrophic development and 
hatches as a fully-formed juvenile at = 0.8 
mm length. 'Growth', by up to 0.2 mm over the 
first 10 days of benthic life in the absence of 




FIG. 4. SEM of the lateral calcareous plates of a 
metarnorphosed Semibalanus balanoides (aper- 
ture plates to right) with which a juvenile Onchidoris 
bilamellata had been associated. Scale: 100 p.m. 



feeding, was reported by Thompson (1967). 
However, Cadlina juveniles continue to sub- 
sist on stored yolk which is externally visible 
by its opacity. Indeed, feeding on the defini- 
tive poriferan diet {Halisarca dujardini (John- 
son)) will only proceed after the complete ex- 
haustion of these reserves, perhaps up to 3- 
4 weeks post-hatching (pers. obs.). 

The smallest O. bilamellata to be seen to 
penetrate the valve-plates of S. balanoides 
and consume the barnacle tissues measured 
1.9 mm (approximately 4-6 weeks post- 
metamorphosis), and there is no doubt that 
the juvenile dorids are incapable of tackling 
live prey up to this time, but choose to graze 
the surfaces of the calcareous plates. Clearly, 
therefore, this extended obligatory period of 



280 



TODD 




FIG. 5. SEM detail of radula tooth of Onchidoris 
bilamellata arrowed in Fig. 4. Scale; 10 ¡i-m. 



detrital grazing — preparatory to stenopha- 
gous prédation on the definitive barnacle 
diet — precludes the necessity to infer a close 
matching of settlement-tinning between pred- 
ator and prey in accounting for the observed 
ecological association and the predator repro- 
ductive strategy. Thus, the settlement-timing 
hypothesis is inappropriate. 



DISCUSSION 

Successful laboratory culture of plank- 
totrophic nudibranch veligers has been de- 
tailed for 1 1 species (see Hadfield & Switzer- 
Dunlap, 1984), but perhaps the most elegant 
and comprehensive studies of opisthobranch 



larval biology concern the tropical aeolid Phe- 
stilla sibogae (Harris, 1973, 1975; Bonar & 
Hadfield, 1974; Hadfield, 1977, 1978, 1984; 
Bonar, 1978a,b; Hadfield & Scheuer, 1985; 
Kempf & Hadfield, 1985; Hirata & Hadfield, 
1986; Miller & Hadfield. 1986; Yool et al., 
1986). This species hatches as a pelagic lec- 
ithotrophic veliger which may metamorphose 
without feeding within 1-2 days of release 
from the benthic capsule. Furthermore, larvae 
of Phestilla will ingest and digest flagellates, if 
available, and this led Kempf & Hadfield 
(1985) to use the term "facultative planktotro- 
phy" to distinguish this form of larval behav- 
iour from what would be conventionally un- 
derstood as a truly pelagic (obligatory) 
lecithotrophic strategy. In this respect Phes- 
tilla sibogae and the British dorid Adalaria 
próxima are very similar. Both hatch from in- 
termediate-sized eggs, undergo an obligatory 
1-2 day Cpre-competent') period, will meta- 
morphose (with or without prior feeding) on 
contact with the live adult prey, and can delay 
metamorphosis in the absence of this cue 
(Thompson, 1958; Kempf & Hadfield, 1985; 
Kempf & Todd, 1987; Todd et al., 1991). 

A survey of the literature shows that re- 
markably few phytoplankters have been used 
in the culture of opisthobranch veligers, and 
that Isochrysis galbana (often in combination 
with another flagellate or a diatom) features 
particularly prominently. It is clear that differ- 
ent larval diets yield strikingly different devel- 
opmental outcomes (see e.g. Pilkington & 
Fretter, 1970; Lucas & Costlow, 1979 for 
prosobranchs; Switzer-Dunlap & Hadfield, 
1977; Chia & Koss, 1978 for opisthobranchs). 
Interestingly, Switzer-Dunlap & Hadfield 
(1977) found Pavlova lutheri to confer the 
highest growth and survival in sub-tropical 
anaspideans, but also noted Isochrysis gal- 
bana to be quite suitable. Similarly, Chia & 
Koss (1978) also found Pavlova and Isochry- 
sis to provide the best results, and encoun- 
tered success with other (unspecified) organ- 
isms. It is relevant, therefore, that for O. 
bilamellata, Pavlova was found to be totally 
unsuitable and Isochrysis of only limited effi- 
cacy, with Rhodomonas so markedly supe- 
rior. Care should, therefore, be taken in ex- 
trapolating such observations of growth, 
survivorship or development to the field. 

Notwithstanding the above, the present de- 
velopmental data uphold the original predic- 
tions with respect to the duration of the pe- 
lagic phase of O. bilamellata. While not 
overstating the case it is perhaps noteworthy 



LARVAL STRATEGIES OF NUDIBRANCHS 



281 



that the presently derived 0^ br the larval 
stage is remarkably close to that previously 
determined (Todd & Doyle, 1 981 ) for the intra- 
capsular embryonic phase. This may not be a 
thvial result, especially if it proved to be char- 
acteristic of opisthobranch development. Em- 
bryonic development data are not difficult to 
obtain and, on the strength of only a single 
successful larval culture on a mixture of 
flagellates and at a realistic temperature, it 
may be possible to confidently predict the nat- 
ural egg-to-juvenile period. 

The extended 4-6 week obligatory period 
of post-metamorphic detrital grazing in O. bi- 
lamellata clearly confounds the criteria for 
support of the settlement-timing hypothesis. 
Indeed, intermediate detrital feeding appears 
to be widespread — if not actually predom- 
inant — amongst those species for which 
developmental data are available. Thus, 
Doridella obscura obligatorily grazes detritus 
for up to 5 days, before the juveniles (~ 1 
mm) handle Membranipora crustulenta, the 
definitive añascan bryozoan diet (Perron & 
Turner, 1977). D. obscura attains some 230 
|jLm in length at metamorphosis (Table 1) and 
it is thus curious that its congener D. steinber- 
gae (210 ixm) does not apparently feed at all 
for the first three days post-settling, but there- 
after handles the adult diet Membranipora 
spp. (Bickell & Chia, 1979; Bickell et al., 
1981). For the small, short-lived aeolid Tenel- 
lia pallida, Eyster (1979) recorded feeding of 
juveniles on "debris" from hydroid surfaces, 
but adults preying directly upon Eudendrium. 

My own observations at field ambient tem- 
peratures include obligatory feeding of post- 
larval Onchidoris muricata (Müller) and Ad- 
alaria próxima (Alder & Hancock) (Todd & 
Havenhand, unpublished; cf. Thompson, 
1958) on detritus for perhaps 1-3 weeks prior 
to their being capable of handling Electra 
pilosa zooids. Furthermore, Archidoris pseu- 
doargus (Rapp) post-metamorphs also graze 
detritus and microalgae for an as yet unspec- 
ified but certainly prolonged period. Although 
competent larvae of Archidoris may metamor- 
phose (Todd & Havenhand, 1984) in the pres- 
ence of the adult prey sponge, Halichondria 
panicea (Pallas), recent observations (Todd & 
Davies, unpublished) show that this sponge is 
not the metamorphosis trigger. Pediveligers 
are most reluctant to crawl on, or othenwise 
associate with, this heavily-spiculate sponge 
and may well subsist on detritus for an ex- 
tended period, not dissimilar to O. bilamellata, 
before taking the definitive prey species. 



Despite the above findings there are nu- 
merous records of post-metamorphs not tak- 
ing any intermediate dietary material before 
handling the definitive adult prey (Table 1). 
Relative sizes at metamorphosis are almost 
impossible to compare inter-specifically, due 
to the lack of mass measurements. Length 
data are, however, available which, albeit im- 
precisely, permit some contrasts to be drawn. 
Table 1 shows that there is remarkable simi- 
larity of post-larval sizes, but three species 
are exceptional: Tritonia diomedea attains 
perhaps 440 (xm and O. bilamellata 470 ixm 
at metamorphosis, while Cadlina laevis (non- 
pelagic lecithotrophic) measures approxi- 
mately 800 (xm at hatching. Moreover, corre- 
lations between metamorph size, whether or 
not intermediate 'detritus' grazing is under- 
taken, and adult prey type are also apparent. 
Thus, species in which grazing does not oc- 
cur prior to specialist prédation include cni- 
darian (7. diomedea, T. hombergi, P. melano- 
branchia, P. sibogae, С salmonacea), and 
spiculate (Rostanga) or slime {Cadlina) 
sponge associates. Melibe leonina juveniles 
attack ciliates and subsequently microcrusta- 
cea (as do adults), but this is a highly special- 
ized species. Those which clearly do graze 
detritus as an intermediate diet include the 
bryozoan grazers (D. obscura, A. próxima, О. 
muricata) and the one barnacle predator (O. 
bilamellata). Data for Archidoris pseudoargus 
are clearly at variance with Rostanga, as are 
data for Tenellia with Phestilla spp. Neverthe- 
less, Tenellia is very small (~ 150 (xm) at 
metamorphosis and Eudendrium is probably 
a well-defended prey item to this small aeolid. 

The conclusion from these data is that with 
the exception of D. steinbergae, those spe- 
cies which encounter considerable (perhaps 
overwhelming) prey-size constraints appear 
to undertake a more or less extended period 
of post-larval feeding (and growth) on detritus 
prior to switching to the definitive prey. Cer- 
tainly bryozoan zooids and barnacles present 
few problems to the adult forms, but to the 
meiofaunal-sized juveniles the size differen- 
tial is enormous (see e.g. Bickell et al., 1981 : 
their Plate 10, 'inset'). 

On the strength of the foregoing, and in ac- 
knowledging the redundancy of the settle- 
ment-timing hypothesis, I offer the following 
argument to account for the evolutionary 'op- 
portunities' and constraints for selection away 
from the ancestral state amongst nudi- 
branchs. Fig. 6 provides a schematic sum- 
mary. 



282 



TODD 



О 



с Ф 

i ^ 



ш 


п 


_J 


я 


m 


га 


< 


> 



го о "О ^ 
л °- Д Ф 



8 ° 

с ill 



со 


г^ 


^ 






^ 


CD 


Ф 






с 


CJ> 


с 




СО 


ф 










> 




3 




£ 


(0 

X 


Г-' й "^ 
m ^^ '^ 


об 




О 

0Í3 


об ■Ч- 


homp: 
pers. 
lorse, 


С 


1^ 


_ Oi 


О 


О 
Ф 




Ф fc 
j£ CD 


1- 


\- ^ 


CL 




GO 



О 

ф ^ ■— ni (— m г- 



О 



ф -2 2 ф i5 о i5 

ü Ф -ÍÉ ö Ф -С Ф . 

<" Рта g '?-5 ■?■■ 

2i5 "5. m S ^ S 



СП 

со 

о. " с 



со >. ее 



s п Р <=' I Ф 

го о i- m -f- (< 

О Ф tj 2 О .у 

N Œ m Ф -=: с 

О СЛ г^ :^ 49 ''J 

¿^ Ф Lu - -^ - 

ш о. 






о ™ ел 
I- ^ -^ 



о 



.5 



о 



•о 
-У 



о. о. 





JO 




















о 










о. 


со 






ся 

с 


= s 


Î 


ел 

с 


>> 
го 


го 


Ü -Q 


э 


го 


о 


о 


ф g 


'со 


о 


ф 


N 


ф 5 


N 


Q. 


о 


о 


ел 


£^ 


о 


^ 


ф 


m 






ш 





Q г« 



■о 

со 
см 


5 со 




« 



ф Ï ■^^ 



с 

О ^ О г £ ^ 

>< Ф -о О О ю 

^ ^~ >< ?S го ,„' 



О ш О г 

(*) л' _сг ^bí _fr ел 



О 



о 



со 




Ч) 




О 




СО 




С 




о 




Е 








"со 


^^ 


ел 


с 


■S 


о 




ел 


Ф 


Q. 


■§. 


1 


b 


Ж 



О 






LARVAL STRATEGIES OF NUDIBRANCHS 



283 



00 

= <^ 
CO 



CO 

•I— '^ 
.X 

■ >> 00 

5 S §> 

о 



X О 



ч_ r^ Q. 
E ^? О 



ta ï? о 

■cd (Л о 

■~ О ^ 



.О 




Q. 










с 






со 








ел 


(Л 

аЗ 

О) 


О 

Е 


о 








.2 


со 


0) 

о- 








Xi 


О) 


аЗ 
Е 








О 






Э 








.та 










_>> 












с 
со 



Ü 






1 
1 








о 

Q. 


со 


2 








со 


со 


э 


та 






с 
о 


и 

3 


<ü 


с 
-2 


"¡С 

■о 
о 


i 




2 

с 

со 

о. 


о 
о 
о 


(Л 
0) 

о 

со 

с 


5 
1 


■о 


о. 




о 

Г>1 


Е 


m 


00 



Q. i5 



■S 
та 


со 


■о 


CJ 


Ч 


о. 




о 


о 


H 



л Е 

3 3 

с/) о 

со ^ 

О 



N CD 






-^ -а 



ел о. 



ел 









- "О ' 
= W (Л с <" 

й- .Ä 10 ^ л 



m О 



о о 
О о. 



с о 

о Cl 

. ел 

с та 

со S 

^S 



о 

Е 
о 

т 

О) 



о 

Е 
m 



Е § 
^ < 



со 3 
1| 

Lu 



> ел 

> ^ 

< г 



со с t- 

со с ^ 

3 го g 

° 1° Р^ 

го °-S¿ -У 

> < -ч: 











СО 


со 




ф 


> 




е- 
О 






ф 










о.^ 






ел 












■о 


CD 




го 


ф 

ф 




го 

00 


С 

ф 


СП 


со 

3 


■D 




го 
oÓ 


го 


со 


1 


ел 


оЗ 


5 


го 


го *- 


■>< 


с 


■■^ 


О 


тэ 




го 
о 


1 
о 


со" 

0) 


CD 


о. 
со" 

0) 


со 


та 
•о 

о 


11 

о о 


ф 

ел' 
Ф 


ф 
ф 


ф 

"О 

с 


■О 


¡£2, 

о 


о 

¿^ 

о 


Z 


Z 


> 




> 




Z 


Z Z 


>- 




о 




Z 


Z 


Е 


Е 


Е 




Е 




Е 


Е Е 


Е 








Е 


Е 


i 


i 


i 




± 




± 


i i 


i 








i 


i 


ю 


ю 


о 




о 




ю 


о о 


о 








о 


о 


■^ 


со 


1^ 




о 




см 


о со 


LO 








t 


t 


см 


со 


■* 




со 




см 


со см 


т— 








'S- 


со 


« 


« 


« 




й 




й 


« й 


й 








« 


« 


Ü 

C\J 


о ■ 

т 


О 




О 

о 




о 

ю 


^5" 

см 1 










ü 

CVJ 


о 

о 


т 


см 


со 




^ 




см 


1 о 










й 


т 



см ~-' "О 





1 

-Q 


05 5 
2. -с 




СЗ) U 

It 
ел 




О 


^ та 


-S 


С 


iÇ СЗ) 




та 


?5 с: 


оЗ 


CD 


W Я 


а. 


Е 


qj со 


^ 


-С О 


0. 




0. ОС 



284 



TODD 



VELIGER 



No ^ 
prey-size - 
-constraint 



P Rostanga pulchra 

L Tritonia hombergi 

D Cadlina laevis 



/POST LARVA' 



EMBRYO 



Prey-size 
constraint 



-P ^ DG Onctiidoris bilarлellata 
-L ^ DG Adalaria próxima 




FIG. 6. Schematic summary of the relationship between post-larval prey-size constraints and reproductive 
strategy opportunities' for evolution away from the pnmitive, or ancestral, status of planktotrophy amongst 
nudibranchs. It is assumed that the potential for detritus-grazing by post-larval planktotrophs has always 
remained, but may not be expressed where it is not required. 

(P; planktotrophy; L: (pelagic) lecithotrophy; D; direct (non-pelagic) development; D [struck through]: direct 
development not supportable; -i-DG: plus detrital grazing). 



Assumptions 

1. Ttie production of small eggs hatching 
as pelagic planktotrophic larvae is the 
primitive or ancestral condition, as is 
dethtivory by the immediate post-meta- 
morph. 

2. Pelagic lecithotrophy is a consequential 
derivative from 1 (above) and is attained 
essentially by increasing the nutritive 
(yolk) reserves and protection of the em- 
bryo. Pelagic lecithotrophic larvae ap- 
pear othenA^ise very similar to compe- 
tent planktotrophs. 

3. Non-pelagic lecithotrophy — whether as 
capsular metamorphosis (e.g. AeoW- 
diella alderi, Tardy, 1970; Tenellia pall- 
ida, Eyster, 1979), or truly vestigial, in- 
tra-capsular development (e.g. Cadlina 
laevis, Thompson, 1967) — is a deriva- 
tive of 2., and C. laevis represents the 
extreme of embryogenesis, i.e. opposite 
to 1. (See Todd, 1981; Hadfield & Swit- 
zer-Dunlap, 1984 for qualifying reviews 
and discussion of the above.) 

4. Planktotrophy is a highly conservative 
mode of development. It is displayed 
by the overwhelming majority of nudi- 
branch species. 

5. There is considerable intra-specific vari- 
ation in egg-size (see Todd, 1987) and 
this provides scope for at least the initial 



stage on the path to establishing leci- 
thotrophy. 

6. The primary function of pelagic larvae is 
not necessarily dispersal (Strathmann, 
1978,1985; Todd, 1985a; Todd et al., 
1988): rather dispersal, which may or 
may not be advantageous to particular 
organisms, should be viewed as an un- 
avoidable consequence of the ancestral 
larval strategy. 

7. Opisthobranch larvae, as opposed to 
prosobranchs, apparently cease growth 
on attaining competence (see Kempf, 
1981; Pechenik & Lima, 1984) [possibly 
related to biophysical size constraints in 
detorting?]; nudibranch metamorphic 
size potential therefore seems more or 
less specifically fixed. 

8. Nudibranchs of all developmental strat- 
egies are characteristically of broadly 
similar post-larval sizes; even Cadlina 
laevis at 800 |xm is small by contrast to, 
for example, many of the larger muricid 
prosobranch hatchlings. Almost without 
exception, nudibranch embryos are pro- 
vided only with zygotic yolk (but see 
Boucher, 1983). 

Conjecture 

A. If no post-metamorphic prey-size con- 
straints apply, the ancestral planktotrophic 



LARVAL STRATEGIES OF NUDIBRANCHS 



285 



pattern may well suffice for establishment of 
the benthic phase (e.g. Tritonia diomedea, 
Phestilla melanobranchia). 

B. If such prey-size constraints do apply, 
some increase in post-larval size may derive 
from extended planktotrophy (as shown by O. 
bilamellata), but this alone may be inade- 
quate: in such cases (e.g. O. muricata) an 
intermediate diet is expedient. Nudibranchs 
are typically specialist predators. If such a pu- 
tative intermediate diet were another inverte- 
brate this scenario would require two suitable 
prey species to be consistently and reliably 
sympatric and adjacent, because post-larvae 
lack both the reserves and motility to under- 
take extensive searching. Furthermore, which 
species (intermediate or definitive?) should 
comprise the metamorphosis stimulus? Se- 
lection ought to favour specific cueing to the 
definitive prey, and thus detritus perfectly fits 
the intermediate requirements if only due to 
its ubiquity and non-specificity. 

C. Neither detritivory, nor microalgal graz- 
ing, demand morphological or physiological 
specializations — the larval gut is, after all, 
adapted for phytoplanktivorous microphagy 
(Bickell & Chia, 1979; Bickell et al., 1981; 
Bickell & Kempf, 1983) and, with the present 
exception of Melibe, post-larval stages bear a 
radula. In essence, post-larval detritivory by 
planktotrophic forms necessitates only an on- 
togenetic delay in the reorganization to ac- 
commodate the carnivorous status. 

Evidence 

Here, I confine my argument to those spe- 
cies from the British Isles with which I have 
previous expehence: these embrace the full 
spectrum of fundamental larval strategies. 

1. Onchidoris muricata. Following meta- 
morphosis, this species encounters size 
constraints in handling Electa pilosa. 
Planktotrophy plus post-larval detrital 
feeding appears to suffice, and the an- 
cestral larval form is retained. 

2. Onchidoris bilamellata. As outlined and 
discussed above this species has criti- 
cal juvenile prey-size constraints. Direct 
development would not yield hatchlings 
sufficiently large to handle barnacle spat 
and, furthermore, that strategy pre- 
cludes detritivory because a larval gut is 
not differentiated (see Cadlina below). 
With the possible exception of 'faculta- 
tive' planktotrophy (see Adatarla below) 



only long-term planktotrophy plus detri- 
tivory appears to provide the necessary 
growth potential. Of all cultured plank- 
totrophic nudibranchs, O. bilamellata 
has the largest post-metamorphic size. 
In the absence of putative genetic or en- 
ergetic constraints there appears to be 
no obvious adaptive advantage to be 
gained from, or which demands, a shift 
from the ancestral planktotrophic condi- 
tion in this species. 

3. Adalaria próxima. Like O. muricata, this 
dorid is a specialist bryozoan predator 
which preferentially takes Electra pilosa. 
We have shown that O. muricata and A. 
próxima display an extraordinarily high 
degree of genetic similarity, and that they 
undoubtedly share a recent common ev- 
olutionary ancestry (Havenhand et al., 
1986). Adalaria is presumed to be the 
more advanced derivative because of its 
pelagic lecithotrophy. Selection away 
from planktotrophy appears to have 
been dictated by the unpredictability of 
energy flux divertible toward reproduc- 
tion by individual adults (see Todd, 
1979b, 1987; Todd & Havenhand, 1983; 
1988, 1990; Havenhand & Todd, 
1 988a,b,c). Here, lecithotrophy is viewed 
as enhancing individual fitness, com- 
pared with long-term planktotrophy, by 
reducing reproductive variance as a re- 
sult of the higher probabilities of larval 
survival and metamorphisis. Nonethe- 
less post-larval prey-size constraints 
persist, as for O. muricata, since these 
two dorids both metamorphose at similar 
sizes (^ 300 Jim). Adalaria larvae can 
feed (Thompson, 1958), but despite di- 
gestion of flagellates in culture it is evi- 
dent that somatic degrowth occurs, just 
as it does on starvation (Kempf & Todd, 
1989). The question as to why Adalaria 
retains a functional larval gut therefore 
appears to relate to this species' require- 
ment to undertake post-larval particulate 
or detrital feeding prior to definitive bry- 
ozoan grazing. The retention of an ap- 
parently functionless, explicitly larval, 
structure to perform a strictly post-larval 
activity is, I believe, a deduction of fun- 
damental importance which supports the 
hypothesis. 

4. Tritonia hombergi. This species has lec- 
ithotrophic larvae which apparently lack 
a functional gut (Thompson, 1962), al- 
though some larvae in culture clearly in- 



286 



TODD 



gest flagellates (Kempf & Todd, 1989). 
Current investigations of the biennial 
life-cycle and reproductive energetics of 
this dendronotid are not yet complete so 
it is premature to speculate on why it 
has become lecithotrophic. However, in 
contrast to A. próxima, no prey-size con- 
straints are encountered by post-larvae, 
and grazing immediately ensues on the 
alcyonarian prey ectoderm once the ju- 
venile gut structures become organized. 
5. Cadlina laevis. Despite undergoing ves- 
tigial 'larval' development, embryos still 
transiently differentiate typical larval 
gastropod features such as a shell and 
velum (Thompson, 1967). Juveniles 
hatch and complete development to- 
ward the adult form, but subsist entirely 
on stored yolk for a few weeks before 
preying only upon the slime sponge Hal- 
isarca dujardini (Johnson). Here a func- 
tional larval gut is not differentiated, but 
it is not required because prey-size 
constraints do not apply. Similarly, Cory- 
phella salmonacea commences preying 
upon hydroids immediately on hatching 
(Morse, 1971). 

If ones perspective of the evolution of nudi- 
branch larval types were confined to the pre- 
juvenile period it would appear intuitively sen- 
sible to suggest that the sequence is one of 
planktotrophy -^ (non-feeding) lecithotrophy 
-* non-pelagic lecithotrophy. Thus, the reten- 
tion of a functional larval gut by Adataría próx- 
ima (and Phestilla sibogae) would be sugges- 
tive of only an intermediate step along the 
path to true lecithotrophy (see Kempf & Had- 
field, 1985; Kempf & Todd, 1989). Alterna- 
tively, one has to infer an adaptive advantage 
to such feeding because of the resource de- 
mands in differentiation of the larval gut; but 
this confounds the a priori assumption of se- 
lection to circumvent the (redundant) larval di- 
gestive system in the shift from planktotrophy 
to true lecithotrophy. The detritus hypothesis 
obviates this non sequitur. 

Nevertheless, Ptiestilla sibogae presents 
an as yet intractable obstacle: fed larvae (in 
contrast to starved larvae) at least maintain 
somatic tissues during the facultative pelagic 
phase, in addition to better retaining compe- 
tence to metamorphose (Kempf & Hadfield, 
1985). Here, some adaptive advantage is de- 
ducible, but despite this the undoubtedly high 
levels of planktonic mortality may still render 
the (smaller) earlier-settling P. sibogae larvae 



of higher mean fitness. In outlining the detri- 
tus-feeding hypothesis I emphasized the im- 
probability of a sympatric intermediate prey 
organism, but one aeolid species appears to 
show just such an adaptation. The leci- 
thotrophic veligers of Eubranclius farrani 
would not metamorphose in response to the 
adult prey hydroid Aglaophenia pluma, but did 
so on encountering Obelia geniculata (Todd, 
1981). Obelia, by contrast to Aglaophenia, 
has a wide aperture to the hydratheca which 
presents no size constraint to post- 
metamorphs gaining access to the tissues of 
individual polyps. 

With regard to the reproductive strategy of 
Onchidoris bilamellata, it has proven that 
Hadfield (1963) showed remarkable foresight 
in predicting that "small adults may depend 
entirely on grazing of algae and sessile cili- 
ates until they reach sufficient size to feed on 
barnacles". In withdrawing the settlement- 
timing hypothesis I have presented an alter- 
native argument predicting selection in favour 
of particular larval strategies amongst nudi- 
branchs. But this should not be interpreted as 
an adaptive explanation of evolutionary shifts 
along the axis highlighted in Assumptions 1- 
3. Rather, it defines which larval types are 
possible under given circumstances. Thus, 
for example, because there appear to be con- 
strained upper (cf. muricid prosobranchs with 
nurse eggs) and lower limits to nudibranch 
post-metamorph/hatchling sizes, this hypoth- 
esis would not predict truly direct develop- 
ment in a species which encounters major 
prey-size constraints following metamorpho- 
sis. However, either of planktotrophy or pe- 
lagic lecithotrophy would be supportable if de- 
trital grazing potential were retained. It has 
long been my contention (Todd, 1979a,b, 
1981, 1985a, 1986; Todd & Havenhand, 
1983, 1988; Havenhand & Todd, 1988a,b,c) 
that absolute energetic allocations and parti- 
tioning within the individual's budget may be 
important, if not actually uppermost, in deter- 
mining the scope for such evolutionary shifts. 
Selection acts on the differential production of 
offspring among genotypes and this is inevi- 
tably some function of energetic capacity and 
the manner in which this is partitioned. With- 
out wishing specifically to resurrect the settle- 
ment-timing hypothesis, one final point should 
also be stressed. Although planktotrophy and 
lecithotrophy appear equivalent, in terms of 
size and level of development at metamor- 
phosis, they do differ markedly in their con- 
ferred egg-to-benthic juvenile periods. Both 



LARVAL STRATEGIES OF NUDIBRANCHS 



287 



the temperate and tropical lecithotrophic spe- 
cies discussed above attain their benthic sta- 
tus more rapidly, and so the three fundamen- 
tal larval strategies cannot be viewed as 
essentially similar means to the same end. 
Questions relating to the adaptive signifi- 
cance of this intriguing feature of larval biol- 
ogy remain very much open. 



ACKNOWLEDGEMENTS 

This work was supported, in part, by grant 
no. GR3/4487 from the Natural Environment 
Research Council. For many hours of 
thoughtful discussion I thank in particular 
Steve Hall, Jon Havenhand, Steve Kempf and 
Jon Davies. To all I am grateful. 



LITERATURE CITED 

BICKELL, L.R. & F.S. CHIA, 1979, Organogenesis 
and histogenesis in the planktotrophic veliger of 
Doridella steinbergae (Opisthobranchia: Nudi- 
branchia). Marine Biology, Berlin, 52: 291-313. 

BICKELL, LR., F.S. CHIA, & B.J. CRAWFORD, 
1981, Morphogenesis of the digestive system 
during metamorphosis of the nudibranch 
Doridella steinbergae (Gastropoda); conversion 
from phytoplanktivore to carnivore. Marine Biol- 
ogy, Berlin, 62: 1-16. 

BICKELL, L.R. & S.C. KEMPF, 1983, Larval and 
metamorphic morphogenesis in the nudibranch 
Melibe leonina (Mollusca: Opisthobranchia). Bio- 
logical Bulletin of the Marine Biological Labora- 
tory, Woods Hole, 165: 119-138. 

BONAR, D.B., 1978a, Fine structure of muscle in- 
sertions on the larval shell and operculum of the 
nudibranch Phestilla sibogae (Mollusca: Gas- 
tropoda) before and during metamorphosis. Tis- 
sue and Cell, 10: 143-152. 

BONAR, D. В., 1978b, Ultrastructure of a cephalic 
sensory organ of the gastropod Phestilla sibogae 
(Aeolidacea, Nudibranchia). Tissue and Cell, 10: 
153-265. 

BONAR, D.B., 1978c, Morphogenesis at metamor- 
phosis in opisthobranch molluscs. In: CHIA, F.S. 
& M.E. RICE, eds.. Settlement and Metamorpho- 
sis of Marine Invertebrate Larvae. Elsevier/North- 
Holland Biomedical Press, New York, p. 177- 
196. 

BONAR, D.B. & M.G. HADFIELD, 1974, Metamor- 
phosis of the marine gastropod Phestilla sibogae 
Bergh (Nudibranchia: Aeolidacea). I. Light and 
electron microscopic analysis of larval and meta- 
morphic stages. Journal of Experimental Marine 
Biology and Ecology, 16: 227-255. 

BOUCHER, L.M., 1983, Extra-capsular yolk bodies 



in the egg masses of some tropical Opisthobran- 
chia. Journal of Molluscan Studies, 49: 232-241 . 

CHIA, F.S. & R. KOSS, 1978, Development and 
metamorphosis of the planktotrophic larvae of 
Rostanga pulchra (Mollusca: Nudibranchia). Ma- 
rine Biology, Berlin, 46: 109-119. 

DAY, R. & L. McEDWARD, 1984, Aspects of the 
physiology and ecology of pelagic larvae of ma- 
rine benthic invertebrates. In: STEIDINGER, K.A. 
& L.M. WALKER, eds.. Marine Plankton Life Cycle 
Strategies. CRC Press, Boca Raton, Florida, p. 
93-120. 

EYSTER, L.S., 1979, Reproduction and develop- 
mental variability in the opisthobranch Tenellia 
pallida. Marine Biology, Berlin, 51: 133-140. 

GRANT, A. & P. WILLIAMSON, 1985, The settle- 
ment-timing hypothesis: a critique. Marine Ecol- 
ogy-Progress Series, 23: 75-83. 

HADFIELD, M.G., 1963, The biology of nudibranch 
larvae. Oikos, 14: 85-95. 

HADFIELD, M.G., 1977, Chemical interactions in 
larval settling of a marine gastropod. In: 
FAULKNER, D.J. & W.H. FENICAL, eds.. Marine 
Natural Products Chemistry. Plenum, New York, 
p. 403-413. 

HADFIELD, M.G., 1978, Metamorphosis in marine 
molluscan larvae: and analysis of stimulus and 
response. In: CHIA, F.S. & M.E. RICE, eds., Set- 
tlement and Metamorphosis of Marine Inverte- 
brate Larvae. Elsevier/North-Holland Biomedical 
Press, New York, p. 165-175. 

HADFIELD, M.G., 1984, Settlement requirements 
of molluscan larvae: new data on chemical and 
genetic roles. Aquaculture, 39: 283-298. 

HADFIELD, M.G.& D. SCHEUER, 1985, Evidence 
for a soluble metamorphic inducer in Phestilla: 
ecological, chemical and biological data. Bulletin 
of Marine Science, 37: 556-566. 

HADFIELD, M.G.& M. SWITZER-DUNLAP, 1984, 
Opisthobranchs. In: TOMPA, A.S., N.H. VER- 
DONK & J.A.M. van den BIGGELAAR, eds.. The 
Mollusca, volume 7, Reproduction. Academic 
Press, Orlando & London, p.209-350. 

HARRIS, L.G., 1973, Nudibranch associations. 
Current Topics in Comparative Pathobiology, 2: 
213-315. 

HARRIS, L.G., 1975, Studies on the life history of 
two coral-eating nudibranchs of the genus Phe- 
stilla. Biological Bulletin of the Marine Biological 
Laboratory, Woods Hole, 149: 539-550. 

HAVENHAND, J.N. & CD. TODD, 1988a, The 
physiological ecology of Adalaria próxima (Alder 
& Hancock) and Onchidoris muricata (Müller) 
(Gastropoda: Nudibranchia). I. Feeding, growth 
and respiration. Journal of Experimental Marine 
Biology and Ecology, 118: 1 51 -1 72. 

HAVENHAND, J.N. & CD. TODD, 1988b, The 
physiological ecology of Adalaria próxima (Alder 
& Hancock) and Onchidoris muricata (Müller) 
(Gastropoda: Nudibranchia). II. Reproduction. 
Journal of Experimental Marine Biology and 
Ecology, 118: 173-189. 

HAVENHAND, J.N. & CD. TODD, 1988c. Repro- 



288 



TODD 



ductive effort of the nudibranch molluscs Adataría 
próxima (Alder & Hancock) and Onchidoris mu- 
ricata (Müller): an evaluation of techniques. 
Functional Ecology. 3: 153-163. 

HAVENHAND, J.N., J. P. THORPE, & CD. TODD, 
1986, Estimates of biochemical genetic diversity 
within and between the nudibranch molluscs Ad- 
atarla próxima (Alder & Hancock) and Onchidoris 
muricata (Müller) (Doridacea: Onchidorididae). 
Journal of Experimental Marine Biology and 
Ecology. 95: 105-111. 

HIRATA, K.Y. & M.G. HADFIELD, 1986, The role of 
choline in metamorphic induction of Phestilla 
(Gastropoda: Nudibranchia). Comparative Bio- 
chemistry and Physiology, 84C: 15-21. 

JABLONSKI, D. & R.A. LUTZ, 1983, Larval ecology 
of marine benthic invertebrates: paleobiological 
implications. Biological Reviews, 58: 21-89. 

KEMPF, S.C, 1981, Long-lived larvae of the gas- 
tropod Aplysia juliana: do they disperse and 
metamorphose or just slowly fade away? Marine 
Ecology-Progress Series. 6: 61-65. 

KEMPF, S.C.& A.O.D. WILLOWS, 1977, Labora- 
tory culture of the nudibranch Tritonia diomedea 
Bergh (Tritoniidae: Opisthobranchia) and some 
aspects of its behavioural development. Journal 
of Experimental Marine Biology and Ecology, 30: 
261-276. 

KEMPF, S.C.& M.G. HADFIELD, 1985, Planktotro- 
phy by the lecithotrophic larvae of a nudibranch, 
Phestilla sibogae (Gastropoda). Biological Bulle- 
tin of the Marine Biological Laboratory, Woods 
Hole, 169: 119-130. 

KEMPF, S.C. & CD. TODD, 1989, Feeding poten- 
tial in the lecithotrophic larvae of Adatarla próx- 
ima and Tritonia hombergi: an evolutionary per- 
spective. Journal of the Marine Biological 
Association. U. K., 69: 659-682. 

LUCAS, J.S. & J.D. COSTLOW, 1979, Effects of 
various temperature cycles on the larval devel- 
opment of the gastropod mollusc Crepiduta for- 
nicata. Marine Biology, Berlin, 51: 111-117. 

MILLER, S.E. & M.G. HADFIELD, 1986, Ontogeny 
of phototaxis and metamorphic competence in 
larvae of the nudibranch Phestilla sibogae Bergh 
(Gastropoda: Opisthobranchia). Journal of Ex- 
perimental Marine Biology and Ecology, 97: 95- 
112. 

MORSE, M.P., 1971 , Biology and life history of the 
nudibranch mollusc, Coryphella stimpsoni (Ver- 
rill, 1879). Biological Bulletin of the Marine Bio- 
logical Laboratory, Woods Hole, 140: 84-94. 

PECHENIK, J. & G.M. LIMA, 1984, Relationship 
between growth, differentiation, and length of lar- 
val life for individually reared larvae of the marine 
gastropod, Crepidula fornicata. Journal of Exper- 
imental Mahne Biology and Ecology, 166: 537- 
549. 

PERRON, F.B. & R.D. TURNER, 1977, Develop- 
ment, metamorphosis, and natural history of the 
nudibranch Doridella obscura Verrill (Coram- 
bidae: Opisthobranchia). Journal of Experimental 
Marine Biology and Ecology, 27: 171-185. 



PILKINGTON, M.C & V. FRETTER, 1970, Some 
factors affecting the growth of prosobranch 
veligers. Helgolander wissenschaftliche Meere- 
suntersuchungden, 20: 576-593. 

STRATHMANN, R.R., 1 978, The evolution and loss 
of feeding larval stages of marine invertebrates. 
Evolution, 32: 894-906. 

STRATHMANN, R.R., 1985, Feeding and non- 
feeding larval development and life-history evo- 
lution in marine invertebrates. Annual Review of 
Ecology and Systematics, 16: 339-361 . 

STRATHMANN, R.R., M.F. STRATHMANN & R.H. 
EMSON, 1984, Does limited brood capacity link 
adult size, brooding, and simultaneous hermaph- 
roditism? A test with the starfish Asterina phylac- 
tica. American Naturalist, 123: 796-818. 

SWITZER-DUNLAP, M. & M.G. HADFIELD, 1977, 
Observations on development, larval growth and 
metamorphosis of four species of Aplysiidae 
(Gastropoda: Opisthobranchia) in laboratory cul- 
ture. Journal of Experimental Marine Biology and 
Ecology. 47: 245-261. 

TARDY, J., 1970, Contribution à l'étude des méta- 
morphoses chez les nudibranches. Annales des 
Sciences Naturelle (Zoologie et Biologie Ani- 
male), 12: 299-370. 

THOMPSON, Т.Е., 1958, The natural history, em- 
bryology, larval biology and post-larval develop- 
ment of Adatarla próxima (Alder & Hancock) 
(Gastropoda Opisthobranchia). Philosophical 
Transactions of the Royal Society of London, Se- 
nes В, 242: 1-58. 

THOMPSON, Т.Е., 1962, Studies on the ontogeny 
of Tritonia hombergi Cuvier (Gastropoda, Opis- 
thobranchia). Philosophical Transactions of the 
Royal Society of London, Series B, 245: 171- 
218. 

THOMPSON, Т.Е., 1967, Direct development in a 
nudibranch, Cadlina laevis, with a discussion of 
developmental processes in Opisthobranchia. 
Journal of the Marine Biological Association, U. 
K,, 47: 1-22. 

TODD, CD., 1979a, The population ecology of On- 
chidoris bilamellata (L.). Journal of Experimental 
Marine Biology and Ecology, 41 : 213-255. 

TODD, CD., 1979b, Reproductive energetics of 
two species of dorid nudibranchs with plank- 
totrophic and lecithotrophic larval strategies. Ma- 
nne Biology, Berlin, 53: 57-68. 

TODD, CD., 1981, Ecology of nudibranch mol- 
luscs. Oceanography and Marine Biology, an An- 
nual Review. 19: 141-234. 

TODD, CD., 1985a, Reproductive strategies of 
north temperate rocky shore invertebrates. In: 
MOORE, P.G. & R. SEED, eds.. Ecology of Rocky 
Coasts. Hodder & Stoughton, Sevenoaks, p. 
203-219. 

TODD, CD., 1985b, Settlement-timing hypothesis: 
reply to Grant & Williamson. Marine Ecology- 
Progress Series, 23: 197-202. 

TODD, CD., 1987, Reproductive energetics and 
larval strategies of nudibranch molluscs: effects 
of ration level during the spawning period in On- 



LARVAL STRATEGIES OF NUDIBRANCHS 



289 



chidoris muricata (Müller) and Adalaria próxima 
(Alder and Hancock). American Malacological 
Bulletin, 5: 293-301 . 

TODD, C.D.& R.W. DOYLE, 1981, Reproductive 
strategies of marine benthic invertebrates: a set- 
tlement-timing hypothesis. Marine Ecology- 
Progress Series, 4: 75-83. 

TODD, CD. & J.N. HAVENHAND, 1983, Repro- 
ductive effort: its definition, measurement and in- 
terpretation in relation to molluscan life history 
strategies. Journal of Molluscan Studies, Supple- 
ment, 12A: 203-208. 

TODD, CD & J.N. HAVENHAND, 1984, Prelimi- 
nary observations on the embryonic and larval 
development of three dorid nudibranchs. Journal 
of Molluscan Studies, 51 : 97-99. 

TODD, CD. & J.N. HAVENHAND, 1 988, The phys- 
iological ecology of Adalaria próxima (Alder & 
Hancock) and Onchidoris muricata (Müller) (Gas- 
tropoda: Nudibranchia).lll. Energy budgets. Jour- 
nal of Experimental Marine Biology and Ecology, 
118: 191-205. 



TODD, CD. & J.N. HAVENHAND, 1990, Repro- 
ductive effort and larval strategies of benthic ma- 
rine invertebrates: is there any relationship? 
Functional Ecology, 4: 132-133. 

TODD, CD., J.N. HAVENHAND & J. P. THORPE, 
1989, Genetic differentiation, pelagic larval trans- 
port and gene flow between local populations of 
the intertidal marine mollusc Adalaria próxima 
(Alder & Hancock). Functional Ecology, 2: 441- 
451 

TODD, CD., M.G. BENTLEY & J.N. HAVENHAND, 
1991, Larval metamorphosis of the nudibranch 
Adalaria próxima (Gastropoda: Nudibranchia): 
the effects of choline and elevated potassium ion 
concentration. Journal of the Marine Biological 
Association, U.K. 71:53-72. 

YOOL, A.J., S.M. GRAU, M.G. HADFIELD, R.A. 
JENSEN, D.A. MARKELL & D.E. MORSE, 1986, 
Excess potassium induces larval metamorphosis 
in four marine invertebrate species. Biological 
Bulletin of the Marine Biological Laboratory, 
Woods Hole, 170: 255-266. 



MALACOLOGIA, 1991, 32(2): 291-299 

THE OPISTHOBRANCH FAUNA OF A MEDITERRANEAN LAGOON 
(STAGNONE Dl MARSALA, WESTERN SICILY) 

Riccardo Cattaneo Vietti^ & Renato Chemello^ 

ABSTRACT 

The opisthobranch fauna present in a lagoon (Stagnone di Marsala) near Marsala (western 
Sicily) is described. There is a rich opisthobranch fauna, with more than 20 species, some of 
which are very abundant. Bullomorpha, with several species often very common in this particular 
ecosystem, are well represented, but the species of Sacoglossa and Nudibranchia are quite 
different from those in other Mediterranean lagoons. Many species, which are usually common 
in similar environments, are rare or absent, e.g. several aeolids and anadorids, but a rich sponge 
population supports the presence of several eudoridaceans, including the little-known Paradons 
granúlala and Dohopsilla rahspina. Elysia tímida, Hypselodoris villafranca and Dendrodohs 
limbata can, perhaps, be considered euryhaiine species as they were frequently collected in 
brackish or lightly polluted waters. Finally, the opisthobranch fauna present in Mediterranean 
lagoon waters is reviewed. 



INTRODUCTION 

Brackish water lagoons in the Mediterra- 
nean have considerable annual variations in 
temperature and salinity (Colombo, 1977; 
SacchI, 1979; Barnes, 1980; Guelorget & 
Perthuisot, 1983) and this therefore poses 
major physiological problems for animals liv- 
ing there. Opisthobranchs from brackish wa- 
ters in the Mediterranean are known from 
several studies (see Table 2) but there have 
been few detailed systematic investigations of 
opisthobranchs in this habitat. 

The purposes of this paper are to present 
the results of a 3-year study of opistho- 
branchs in the Marsala Lagoon and to review 
the opisthobranchs living in this habitat 
throughout the Mediterranean. 

The Marsala Lagoon, which is called 'Stag- 
none', extends for 2000 ha in western Sicily 
and has been extensively studied (Cavallaro 
et al., 1977; Calvo et al. 1982). Information on 
its malacological fauna has been reported by 
Cavallaro et al. (1977) and Sparia (1985). 

The Marsala Lagoon is morphologically di- 
vided into two basins (Fig. 1). The southern 
basin is connected with the open sea by a 
large channel, between Punta d'Alga and the 
Isola Grande. The northern basin, the true 
'Stagnone', has markedly more lagoon char- 
acteristics such as shallow waters, irregular 
water movements and variable salinity and 
temperature. There is, moreover, a progres- 



sive silting up because the Birgi River, re- 
cently deviated southward, was canalized 
near the 'Tramontana' mouth, and so dis- 
charges its abundant waste into the lagoon. 

Sampling Stations 

The average depth of the northern basin of 
the Marsala lagoon is around 0.5-1 .0 m, with 
a maximum of 3 m near Isola Grande. Most of 
the specimens were collected by snorkeling in 
different periods of the years 1984-86, near 
sparse shoots of Posidonia oceánica (Sta- 
tions A,E), in Cymodocea nodosa prairies, 
under or on small hard objects or animals 
(stones, anchor logs, sponges) and on 
Rytiphloea tinctoria aegagropyla forms near 
Mozia (Stns A,E,G), Punta Grassellino (Stn 
C), Punta Palermo (Stns B,F), Saline (Stn D), 
between Mozia and Punta Palermo (Stns H,l) 
and near Punta d'Alga (Stn L). The average 
depth of the samples taken was at 0.5-1 .5 m. 
The species collected and the numbers of 
specimens are reported in Table 1 . 

The opisthobranch fauna of the 
Marsala Lagoon 

The opisthobranch fauna in the Marsala la- 
goon is quite rich (Table 1 ), with more than 20 
species, compared with about 60 species re- 
corded from all other Mediterranean lagoons. 



Mstituto di Zoología dell'Universita degli Studi di Genova, via Baibi 5, 1-16126 Genova, Italy 
^Istituto di Zoología dell'Universita degli Studi di Palermo, via Archirafi 18, 1-90123 Palermo, Italy. 



291 



292 



CATTANEO VIETTi & CHEMELLO 



F. BIRGI 




SALINE 



C-f PTA 
IP ■ \ GRASSELLINO 




FIG. 1. The Marsala Lagoon (western Sicily). Sampling areas: A, E, G: Mozia Isle; B, F: Punta Palermo; C: 
Punta Grassellino; D: Saline; H, I: between Mozia Isle and Punta Palermo; L: Punta d'Alga. 



This richness could be due to the presence of 
at least five habitats: Posidonia beds, Cymo- 
docea prairies, sandy bottoms, the peculiar 
'nnicrohabitat' of aegagropyla forms of Rytiph- 
loea tinctoria (Sparia & Riggio, 1984; Riggio & 
Sparia, 1985), and hard substrates made up, 
above all, of tufaceous outcrops and calcare- 



ous red algae {Lithothamnium fruticulosum, L. 
calcareum, and pleustophytic forms of Meso- 
phyllum lichenoides). 

On the mud beds, Bullomorpha are com- 
mon with well-known euryhaline species, 
such as l\/1amilloretusa mammillata, Bulla stri- 
ata and Haminoea hydatis, while Haminoea 



OPISTHOBRANCHS OF A MEDITERRANEAN LAGOON 



293 



TABLE 1. Opisthobranch molluscs present in the Marsala Lagoon (western Sicily). For the sampling 
areas see the legend to Fig. 1 . 

Sites sampled 

Species A В С DEFGHIL Total 

Retusa semisulcata^ — — — — — — — — — — — 

Retusa truncatula^ — — — — — — — — — — — 

Mamilloretusa mammillata^ 2 — 2 — — 1 — — — — 5 

Bulla striata^ 3 — 1 — 1 i____ q 

Haminoea hydatis^^ 2 1 3 _з_____ g 

Haminoea cymoelium — — 1 — — — — — — — 1 

Elysia tímida — — 6+ — — — — — — — 6 + 

Aplysia fasciata^ — — — — — — — — — — — 

Berthella aurantiaca 1 — — 2 2 23 2 4— 16 + 

Berthella stellata ____1__2— — 3 

Doris verrucosa^ — — — — — — — — — — — 

Don's sp.^ — — — — — — — — — — — 

Glossodoris sp.^ — — — — — — — — — — — 

Chromodoris sp. 5 — — — — — — — — — 5 

Hypselodoris villafranca 10+ — — — 2 — — — — — 12 + 

Hypselodoris elegans 1 — 2 — — — — — — — 3 

Hypselodoris messinensis — — 10+ — — 4 — — — — 14 + 

Paradoris granulata 11+ — 4 — — — — — 2 — 17 + 

Plalydoris argo — — — — — — — — — 1 1 

Dendrodoris limbata 2 — — — — 4 — 1 — — 7 

Dendrodoris grandiflora^ 1 — — — — — — — — — 1 

Doriopsilla rarispina 13+ — 2 — 1 — — 2 — — 18 + 

Spurilla neapolitana 1 — — — — — — — — — 1 

Total number of 52+ 1 31 + 2 10 12 3 7 6 1 125 + 
specimens 

+ = more specimens were observed than collected 

1 = recorded by Cavallaro et al., 1977 

2 = recorded by Sparia, 1985 (unpublished data, thesis) 



cymoelium should probably be considered a 
young H. hydatis. 

The only common notaspidean in the 
Marsala lagoon is Berthella aurantiaca. 

The species composition of Sacoglossa 
and Nudibranchia in this lagoon is quite dif- 
ferent from that known from other Mediterra- 
nean lagoons. Among the Sacoglossa, Elysia 
tímida is very common on hard artificial bot- 
toms at 0.2 m depth, near Acetabularia ace- 
tabulum, on which it feeds (Ros & Rodriguez, 
1985). This easily-recognized white species 
with red spots (Bouchet, 1 984) seems to pre- 
fer shallow and euryhaline waters. It has also 
been found in other brackish waters such as 
Strea Lagoon (Ionian Sea), Oristano (Sardi- 
nia) and S. Marco Cape near Sciacca (Sicily). 

A rich sponge population (Corriere, 1984) 
supports the presence of several eudor- 
idaceans, some of them also of considerable 
scientific interest. Doriopsilla rarispina, a very 
rare species recently re-described by Perrone 
(1986) from the Ionian Sea, has been found 



and there is a large population of Paradoris 
granulata. This beautifully-camouflaged spe- 
cies is easily found inside the sponge Ircinia 
variabilis or in its oscula. There is also what is 
probably a new, undescribed species of chro- 
modorid in the Marsala Lagoon. It is a few 
centimetres long and presents a translucid 
pale white colour with characteristic pale 
azure-white ocelli surrounded by an opaque 
white ring. It lives at the base of Cystoseira 
barbata brown algae, on the sponge Tedania 
anelans. 

Many young specimens of Hypselodoris vil- 
lafranca and H. messinensis were collected 
among Rytiphloea tinctoria aegagropyla 
forms which seem to be 'nurseries' for these 
nudibranchs. 

On the other hand, some common eury- 
haline or widely-distributed species {Polycera 
quadrilineata, Favorinus branchialis, Cory- 
phella pedata and Cratena peregrina) are ab- 
sent, probably due to the sparseness of hard 
bottoms with few hydroids and bryozoans. 



294 



CATTANEO VIETTI & CHEMELLO 




FIG. 2. Mediterranean Sea: brackish waters in which opisthobranchs were found (see Table 2) — SPAIN: 1 . 
Mar Menor; 2. Salinas de Calblanque. FRANCE: 3. Seises; 4. Sigean; 5. Thau (or Sète); 6. Berre; 7. Brusc; 
ITALY: 8. Orbetello; 9. Oristano; 1 0. Lungo; 1 1 . Caprolace; 1 2. Fogliano; 1 3. Patria; 1 4. Fusaro; 1 5. Faro and 
Ganzirri; 16. Tindari; 17. Stagnone of Marsala; 18. Sciacca; 19. Vendicari; 20. Mar Piccolo of Taranto; 21. 
Strea of Porto Cesáreo; 22. Venezia; 23. Grado and Maraño; USSR: 24. Azov Sea; ISRAEL: 25. Dor; 26. 
Mikhmoret. EGYPT: 27. Bardawil. TUNISIA: 28. Biban; 29. Bizerte; 30. Tunis; MOROCCO: 31. Nador. 



The rarity of Spurilla neapolitana and the ab- 
sence of the common euryhaline Aeolidiella 
spp. are surprising since their food, Para- 
stephanauge pauxii is very common on Cy- 
modocea leaves. 

Finally, the common Mediterranean spe- 
cies Elysia tímida, Hypselodoris villafranca 
and Dendrodoris limbata can be considered 
euryhaline, because they were frequently 
seen also in other brackish habitats such as 
the Strea' of Porto Cesáreo (Taranto) and 
Oristano lagoon. According to Perrone 
(1984), Dendrodoris limbata can live in pol- 
luted waters and this fact confirms its adapt- 
ability to environmental changes. 

Review of opisthobranchs in Mediterranean 
brackish waters 

The Opisthobranch fauna present in Medi- 
terranean brackish waters is still not well 
known (Fig. 2). There are useful reports by 
Barletta (1980) and Torelli (1982), while Gue- 
lorget (1985), in his review of the parhalic do- 
main, considered Akera bullata an exclusively 
brackish species, while Aplysia depilans and 
Philine aperta are the most common opistho- 
branchs present in Mediterranean lagoons. 
Additional information is available from gen- 
eral studies carried out on the malacological 
fauna of lagoons, even though nudibranchs 



were often ignored. A summary of opistho- 
branch records in Mediterranean lagoons is 
given in Table 2. 

Many opisthobranchs can live in this pecu- 
liar habitat which is subject to considerable 
variation in the main physico-chemical condi- 
tions such as salinity, temperature, oxygen 
saturation, pH, etc. The Bullomorpha, living 
on soft bottoms, are widespread and include 
euryhaline species such as Retusa trunca- 
tula, Haminoea navícula, Bulla striata, Philine 
aperta and Akera bullata. 

Many Sacoglossa are present in the lagoon 
ecosystem but they only inhabit areas where 
food plants occur. Oxynoe and Lobiger, for 
example, are confined to Caulerpa prairies 
(sometimes living in outer zones of the la- 
goons). The majority of other species are 
linked to the distribution of the green algae 
Bryopsis and Cladophora. Only Limapontia 
capitata, found in the Fusaro Lake (Naples), 
seems to occur exclusively in brackish waters 
(Jensen, 1977). The occurrence of Alderia 
modesta, an Atlantic euryhaline species 
(Adam & Leioup, 1939; Engel et al., 1940), is 
still uncertain in the Mediterranean Sea. 

Algal availability is also important for aplysi- 
omorphs: they are sometimes abundant in the 
zones where Ulvales flourish while Bursatella 
leachii, in the Mediterranean Sea, was always 
recorded in still-water areas like lagoons. 



OPISTHOBRANCHS OF A MEDITERRANEAN LAGOON 



295 



TABLE 2. Opisthobranch molluscs present in Mediterranean lagoons. 



Species 



Locality (Reference) 



BULLOMORPHA 



Acteon tornatilis (L.) 

Cylichnina girardi (Audouin) 
Retusa semisulcata (Philippi) 

R. perstriata (Cerulli Irelli) 
R. truncatula (Bruguière) 



R. umbilicata (Montagu) 
Retusa spp. 
Mamilloretusa mammillata 

(Philippi) 
Ringicula auriculata 

(Menard de la Groye) 
R. conformis (Monterosato) 
Bulla striata Bruguière 



Atys blainvilliana (Récluz) 
Haminoea hydatis (L.) 



H. navícula (Da Costa) 



H. orbignyana (Férussac) 
Akera bullata Müller 

Philine aperta (L.) 

Philine cf. scabra (Müller) 
APLYSIOMORPHA 



France: Berre (Mars, 1966) 

Italy: Venice (Coen, 1928); Grado, Maraño (Zucchi Stolfa, 1977) 

Egypt: Bardawil (Mienis, 1976; Barash & Danin, 1982) 

Italy: Marsala (Cavallaro et al., 1977); Grado, Maraño (Zucchi Stolfa, 

1977) 
Italy: Grado, Maraño (Zucchi Stolfa, 1977) 
France: Salses, Sigean, Thau, Berre (Mars, 1966) 
Italy: Orbetello (Mari, 1976); Marsala (Cavallaro et al., 1977) 
Tunisia: Tunis (Zaouali, 1981) 
Morocco: Nador (Saubade, 1979) 
France: Berre (Mars, 1966) 
Egypt: Bardawil (Barash & Danin, 1982) 
Italy: Marsala (Cavallaro et al., 1977); Mar Piccolo (Tortonici & Panetta, 

1977) 
Italian Lagoons (Torelli, 1982) 

Italy: Porto Cesáreo (Parenzan, 1970) 

France: Berre (Mars, 1966) 

Italy: Orbetello (Mari, 1976); Caprolace (Ardizzone, 1985); 

Faro, Ganzirri (Giudice, pers. comnn.); Marsala (Sparia, 
1985 — thesis); Vendicari (Chemello, pers. obs.); Mar Piccolo 

(Parenzan, 1969; Tortoricj & Panetta, 1977) 
Egypt: Bardawil (Barash & Danin, 1982) 
Tunisia: Bizerte (Zaouali, 1979) 
Morocco: Nador (Saubade, 1979) 
Egypt: Bardawil (Barash & Danin, 1982) 
Spain: Mar Menor (Olmo & Ros, 1984) 
Italy: Lungo, Caprolace, Fogliano (Ardizzone, 1985); Fusaro (Ferro & 

Russo, 1981); Faro, Ganzirri (Scordia, 1927; Parenzan, 1979); 

Tindari (Chemello, pers. obs.); Marsala (Cavallaro et al., 1977; 

Sparia, 1985— thesis); Venice (Coen, 1933,1938; Vatova, 1940); 

Grado, Maraño (Zucchi Stolfa, 1977) 
Egypt: Bardawil (Barash & Danin, 1982) 
Spain: Mar Menor (Olmo & Ros, 1984); Salinas de Calblanque 

(Templado et al., 1983) 
France: Sigean, Thau, Berre (Mars, 1966) 
Italy: Orbetello (Mari, 1976); Faro, Ganzirri (Parenzan, 1979); Mar 

Piccolo (Tortorici & Panetta, 1977); Venice (Coen, 1933;1938) 
Tunisia: Biban (Zaouali & Baeten, 1985); Bizerte (Zaouali, 1979); 

Tunis (Zaouali, 1974,1981) 
Morocco: Nador (Saubade, 1979) 
Spain: Mar Menor (Murillo & Talavera, 1983; Olmo & Ros, 1984); 

Salinas de Calblanque (Templado et al., 1983) 
Spain: Salinas de Calblanque (Templado et al., 1983) 
France: Thau, Berre (Mars, 1966) 
Italy: Venice (Coen, 1933) 
Spain: Mar Menor (Olmo & Ros, 1984) 
France: Thau, Berre (Mars, 1966) 

Italy: Mar Piccolo (Parenzan, 1969; Tortorici & Panetta, 1977) 
Tunisia: Bizerte (Zaouali, 1979) 
Italy: Grado, Maraño (Zucchi Stolfa, 1977) 



Aplysia depilans Gmelin 



France: Thau (Mars, 1966) 

Italy: Venice (Coen, 1938) 

Tunisia: Bizerte (Zaouali, 1979); Tunis (Zaouali, 1974) 



(continued) 



296 

TABLE 2. (Continued) 



CATTANEO VIETTI & CHEMELLO 



Species 



Locality (Reference) 



A. fasciata Poiret 



A. punctata Cuvier 

Bursatella leachii leachii 
Blainville 

B. I. savignyi Audouin 

Notarchus punctatus Philippi 



Spain: Salinas de Calblanque (Templado et al., 1983) 

France: Thau, Berre (Mars, 1966) 

Italy: Orbetello (Mari, 1976); Patria (Sacchi, 1961); Faro, Ganzirri 

(SI. M. comm.); Marsala (Cavallaro et al., 1977); Porto Cesáreo 

(Cattaneo Vietti, pers. obs.) 
France: Thau, Berre (Mars, 1966) 
Italy: Venice (Coen, 1933) 
Italy: Venice (Cesan et al., 1986) 

Italy: Mar Piccolo (Tortorici & Panetta, 1977) 
Israel: Dor, Mikhmoret (Barash & Danin, 1972) 
Italy: Mar Piccolo (Parenzan, 1969) 



NOTASPIDEA 

Pleurobranchaea meckelii 

Leue 
Berthella aurantiaca (Risso) 



Italy: Mar Piccolo (Parenzan, 1969) 
Tunisia: Bizerte (Zaouali, 1979) 



SACOGLOSSA 

Oxynoe olivácea Rafinesque 
Lobiger serradifalci (Calcara) 

Elysia viridis (Montagu) 



E. tímida (Risso) 



Calliopaea bellula d'Orbigny 

Placida viridis (Trinchese) 
P. dendritica 

(Alder & Hancock) 
Ercolania funérea (Costa) 
Limapontia capitata (Müller) 
Alderia modesta (Lovén) 



Itaiy: Mar Piccolo (Parenzan, 1969,1970; Tortorici & Panetta, 1977) 
Italy: Orbetello (Mari, 1976); Mar Piccolo (Parenzan, 1969; Tortorici & 

Panetta, 1977) 
Spain: Mar Menor (Olmo & Ros, 1984) 
France: Thau (Mars, 1966) 
Italy: Fusaro (Schmekel, 1968) 
Spain: Mar Menor (Ros & Rodriguez, 1985) 
Italy: Oristano (Cattaneo Vietti, pers. obs.); Sciacca (Chemello, pers. 

obs.); Porto Cesáreo (Cattaneo Vietti, pers. obs.) 
Spain: Salinas de Calblanque (Templado et al., 1983) 



Italy 
Italy 
Italy 



Fusaro (Schmekel, 1968) 
Fusaro (Schmekel, 1968) 
Fusaro (Schmekel, 1968) 



Italy: Fusaro (Schmekel, 1968) 
Italy: Fusaro (Schmekel, 1968) 
— (Pruvot-Fol, 1954) 



NUDIBRANCHIA Doridina 

Okenia elegans (Leuckart) 
Doris verrucosa L. 



D. bicolor (Bergh) 
Hypselodoris vi I la franca 

(Risso) 
Chromodoris krohnii 

(Verany) 
Polycera quadrilineata 

(Müller) 

P. dubia Sars 
Polycerella emertoni Verrill 
Limada clavigera (Müller) 
Dendrodoris limbata (Cuvier) 



Italy: Fusaro (Toscano, pers. comm.) 

France: Thau (Mars, 1966) 

Italy: Orbetello (Mari, 1976); Marsala (Cavallaro et al., 1977) 

Tunisia: Bizerte (Zaouali, 1979) 

Italy: Venice (Coen, 1938) 

Italy: Orbetello (Mari, 1976); Porto Cesáreo (Cattaneo Vietti, pers. obs.) 

Italy: Orbetello (Mari, 1976) 

France: Canaux de Sète (Mars, 1966) 

Italy: Orbetello (Mari, 1976); Fusaro (Schmekel, 1968; Toscano, pers. 

comm.) 
Italy: Fusaro (Schmekel, 1968) 
Italy: Fusaro (Schmekel, 1968) 
Italy: Fusaro (Toscano, pers. comm.) 
Italy: Faro, Ganzirri (Giudice, pers. comm.); Porto Cesáreo (Cattaneo 

Vietti, pers. obs.) 



OPISTHOBRANCHS OF A MEDITERRANEAN LAGOON 



297 



TABLE 2. (Continued) 



Species 



Locality (Reference) 



NUDIBRANCHIA Arminina 

Janolus cristatus 
(Delle Chiaje) 

NUDIBRANCHIA Aeolidina 

Coryphella pedata (Montagu) 

C. lineata (Lovén) 
Calmella cavolinii (Verany) 
Facelina corónala 

(Forbes & Goodsir) 
F. annulicornis (Chamisso & 

Eysenhardt) 
Cratena peregrina (Gmelin) 

Favorinus branchialis 

(Rathke) 
Eubranchus exiguus 

(Alder & Hancock) 
Cuthona caerulea (Montagu) 
Tenellia adspersa (Nordmann) 
Calma glaucoides 

(Alder & Hancock) 
Aeolidiella alderi (Cocks) 



A. rubra (Cantraine) 
Baeolidia nodosa 

(Haefelfinger & Stamm) 
Spurilla neapolitana 

(Delle Chiaje) 

Berghia verrucicornis (Costa) 



Italy: Orbetello (Mari, 1976); Fusaro (Schmekel, 1968; Toscano, pers. 
comm.) 



France: Canaux de Sète (Mars, 1966) 
Italy: Fusaro (Toscano, pers. comm.) 
France: Canaux de Sète (Mars, 1966) 
Italy: Orbetello (Mari, 1976) 
France: Thau (Mars, 1966) 

France: Thau (Mars, 1966) 

France: Canaux de Sète (Mars, 1966) 
Italy: Orbetello (Mari, 1976) 
France: Brusc (Riva & Vicente, 1976) 
Italy: Fusaro (Schmekel, 1968) 
Italy: Fusaro (Schmekel, 1968) 

France: Canaux de Sète (Mars, 1966) 
USSR: Azov Sea (Roginskaya, 1970) 
Tunisia: Tunis (Zaouali, 1974) 

France: Brusc (Riva & Vicente, 1976) 

Italy: Fusaro (Schmekel & Portmann, 1982); Porto Cesáreo (Cattaneo 

Vietti, pers. obs.) 
France: Thau, Berre (Mars, 1966) 
Spain: Salinas de Calblanque (Templado et al., 1983) 

Spain: Salinas de Calblanque (Templado et al., 1983) 
France: Thau (Mars, 1966); Brusc (Riva & Vicente, 1976) 
Italy: Caprolace (Ardizzone, 1985); Fusaro (Schmekel, 1968) 
Italy: Orbetello (Mari, 1976) 



Many species of nudibranchs occur in the 
lagoons. Species which are characteristic of 
shallow water and tide-pools (e.g. Polycera 
quadrilineata, Polycerella emertoni, Calmella 
cavolinii, Doris verrucosa) can also thrive in 
brackish waters. When hydroids settle on 
hard bottoms, aeolids are connmonly found on 
them (e.g. Coryphella spp., Facelina spp., 
Cratena peregrina and Cuthona coerulea). 
Typical euryhaline species are Spurilla nea- 
politana, Aeolidiella spp., Favorinus branchi- 
alis and Tenellia adspersa. Finally, Tergipes 
tergipes and Embletonia pulchra, which are 
euryhaline species along the Atlantic coasts 
(Pruvot-Fol, 1954; Thompson & Brown, 
1984), have rarely been recorded in the Med- 
iterranean Sea and do not appear to be as- 
sociated with any particular ecological condi- 
tions. 

Mediterranean lagoons vary in their spe- 



cies composition of hydroids (Morri & Bianchi, 
1 983), serpulids (Bianchi et al., 1 984; Bianchi, 
1985) and prosobranchs (Torelli, 1983), and 
doubtless similar variation will be found to oc- 
cur in opisthobranchs. One might expect the 
North Atlantic type lagoons of the north Adri- 
atic to have a very different fauna from the 
xero-Mediterranean lagoons of Sicily and 
north Africa, but data even for the better 
known Bullomorpha are too poor to enable 
any such conclusions to be made. 



ACKNOWLEDGEMENTS 

We offer grateful thanks to Prof. M. Ed- 
munds (Lancashire Polytechnic, Preston) for 
his criticism and to Prof. S. Riggio, M.P. 
Sparia, G. Corriere, A. Giudice and F. Tos- 
cano for their help. 



298 



CATTANEO VIETTI & CHEMELLO 



LITERATURE CITED 

ADAM, W. & E. LELOUP, 1939. Sur la presence 
dAldeha modesta (Lovén, 1844) en Belgique. 
Bulletin Museum Royal Histoire naturelle du Bel- 
gique. 15 (64): 1-13. 

ARDIZZONE, G. 1985, Zoología. In: Seminario in- 
formativo sui risultati del progetto 'Laghi costieri' 
(Terracina, 1985). Amministrazione Provinciale 
di Latina e Université degli Studi La Sapienza', 
Rome, 1-77 p. 

BARASH, A. & Z. DANIN, 1971, Opisthobranchs 
from Mediterranean waters of Israel. israelJour- 
nal of Zoology. 20: 151-200. 

BARASH, A. & Z. DANIN, 1972, The Indo-Pacific 
species of Mollusca in the Mediterranean and 
notes on a collection from the Suez Canal. Israel 
Journal of Zoology, 21 : 301- 374. 

BARASH, A. & Z. DANIN, 1977, Additions to the 
knowledge of Indo-Pacific Mollusca in the Medi- 
terranean. Conchiglie. 13: 85-116. 

BARASH, A. & Z. DANIN, 1982, Contribution to the 
knowledge of Mollusca in the Bardawil lagoon. 
Bollettino malacologico. Milano, 18: 107-128. 

BARLETTA, G., 1980, Gasteropodi marini nudi. 
Guide per il hconoscimento delle specie animali 
del le acque lag и nah e costiere italiane 3. CNR, 
Rome, 1-124 p. 

BARNES, R.S.K., 1980, Coastal Lagoons. Cam- 
bridge University Press, Cambridge, 1-106 p. 

BIANCHI, C.N., 1985, Biogéographie des lagunes 
méditerranéenes d'après la distribution des Poly- 
chétes Serpuloidea. Rapport de la Commission 
International pour l'Etude de la Mer Méditerra- 
née. 29 (4): 39-40. 

BIANCHI, C.N., L.A. CHESSA & С MORRI, 1984, 
Serpuloidea (Annelida, Polychaeta) délia Sarde- 
gna, con particolare riguardo aile lagune cos- 
tiere. Rendiconti del Seminario délia Facoltà di 
Scienze dell 'Université di Cagliari, 54 (Suppl.): 
49-58. 

BOUCHET, P., 1984, Les Elysiidae de Méditer- 
ranée (Gastropoda, Opisthobranchiata). Annales 
de l'Institut Océanographique de Paris, 60: 19- 
28. 

CALVO, S., G. GIACCONE & S. RAGONESE, 
1982, Tipología délia vegetazione sommersa 
dello stagnone di Marsala (TP). Naturalista Sicil- 
iano, 6: 187-196. 

CAVALLARO, G. et al., 1978, Studio di un ambi- 
ente lagunare: lo Stagnone di Marsala. In: Ambi- 
enti umidi costieh. Atti II' Convegno Siciliano di 
Ecología (Noto), p. 47-69. 

CESARI, P., L. MIZZAN & E. MOTTA, 1986, 
Rinvenimento di Bursatella leachi leachi Blain- 
ville 1 81 7 in Laguna di Venezia. Phma segnalazi- 
one adriatica. Lavoh Società Veneziana di Sci- 
enze Naturali, 1 1 : 5-16. 

COEN, G., 1933, Saggio di una sylloge mollusco- 
rum adriaticum. Memorie Regio Comitate talas- 
sografico italia. 192. 1-186 p. 

COEN, G., 1938, Nota sui molluschi della Laguna 



véneta. Atti della Società Italiana per il Progressa 
delle Scienze. 26: 114-123. 

COLOMBO, G., 1977, Lagoons. In: BARNES, R. 8. 
K., ed.. The Coastline, Wiley, London & New 
York, p. 63-81. 

CORRIERO, G., 1984, Note sul popolamento di 
poriferi dello Stagnone di Marsala (Sicilia). Nova 
Thalassia. 6 (Suppl.): 213-223. 

ENGEL, H., S.J. GEERTS, & CO. VAN REG- 
TEREN ALTERA, 1940, Alderia modesta (Lovén) 
and Limapontia depressa Alder & Hancock in the 
brackish waters of the Dutch coast. Basteria, 5: 
6-34. 

FERRO, R. & G. F. RUSSO, 1981, Aspetti di parti- 
colare interesse nella malacofauna del lago 
Fusaro. Bollettino malacologico, Milano, 17: 
191-198. 

GUELORGET, O., 1985, Entre mer et continent. 
Contribution à l'étude du domaine paralique. Un- 
published. Thèse, Université de Sciences et 
Techniques de Languedoc, Montpellier, 1-721 p. 

GUELORGET, O. & J.P. PERTHUISOT, 1983, Le 
domaine paralique. Travaux du Laboratoire de 
Geologie de Paris, 16: 1-136. 

JENSEN, К., 1977, Optimal salinity and tempera- 
ture intervals of Limapontia capitata (Opistho- 
branchia, Sacoglossa) determined by growth and 
heart rate measurements. Ophelia, 16: 175-185. 

MARI, M., 1976, Osservazioni sulla malacofauna 
delle lagune di Orbetello. Memorie della Società 
Toscana di Scienze naturali, (B) 83: 190-204. 

MARS, P., 1966, Recherches sur quelques étangs 
du littoral méditerranéen français et sur leurs 
faunes malacologiques. Vie et Milieu, 20 
(Suppl.): 1-359. 

MIENIS, H.K., 1976, Ventomnestia girardi (Au- 
douin, 1827) from the Mediterranean. Conchiglie, 
12: 209-210. 

MORRI, С. & C.N. BIANCHI, 1983, Contributo alla 
conoscenza degli idrozoi lagunari italiani: idropo- 
lipi del Delta del Po (nord Adriático). Atti del Mu- 
seo civico di Storia naturale di Trieste, 35: 185- 
205. 

MURILLO, L. & P. TALAVERA, 1983, Contribución 
al conocimiento de la malacofauna do Mar Menor 
(Murcia). Iberus, 3: 15-28. 

OLMO, R. & J.D. ROS, 1984, Las malacocenosis 
del Mar Menor. Estudio y comparación con co- 
munidades de medios lagunares semejantes. 
Actas 6" Simposio Ibérico del Estudio del 
Benthos Marino, Lisboa, 1 : 253-260. 

PARENZAN, P., 1969, II Mar Piccolo e il Mar 
Grande di Taranto. Thalassia salentina, 3: 19- 
34. 

PARENZAN, P., 1970, Gasteropodi. Carta d'iden- 
tità delle conchiglie del Mediterráneo. 1. Bios- 
Taras, Taranto. 1-283 p. 

PARENZAN, P., 1979. Fauna malacologica del 
laghi di Ganzirri e del Faro (Messina). Thalassia 
salentina, 9: 67-78. 

PERRONE, A., 1983, Opistobranchi (Aplysiomor- 
pha, Pleurobranchomorpha, Sacoglossa, Nudi- 
branchia) del litorale salentino (Mar Ionio) 



OPISTHOBRANCHS OF A MEDITERRANEAN LAGOON 



299 



(Elenco-Contributo primo). Thalassia salentina, 
13: 118-144. 

PERRONE, A., 1986, II genere Doriopsilla Bergh 
1880 in Mediterráneo: descrizione di Doriopsilla 
rarispina Pruvot-Fol 1951 (Opisthobranchia: 
Nudibranchia). Bollettino malacologico, Milano, 
22:97-112. 

PRUVOT-FOL, A., 1954, Mollusques opistho- 
branches. Faune de France, 58. 1-460 p. 

RIGGIO, S. & M.P. SPARLA, 1985, A survey of the 
invertebrate populations inhabiting Rytiphloea 
tinctoria (Clem.) С Ag. aegagropyla in the Stag- 
none Sound (Western Sicily). Rapport de la 
Commission International pour l'Etude de la Mer 
Méditerranée, 29 (4): 143-144. 

RIVA, A. & N. VICENTE, 1976, Influence des fac- 
teurs du milieu sur la biologie de trois espèces de 
nudibranches: Aeolidiella alderi, Spurilla neapo- 
litana et Favorinus branchialis. I. Haliotis, 7: 39- 
43. 

ROGINSKAYA, I.S., 1970, Tenellia adspersà, a 
nudibranch new to the Azov Sea, with notes on 
its taxonomy and ecology. Malacological Review, 
3: 167- 174. 

ROS, J.D. & J. RODRIGUEZ, 1985, La simbiosis 
algal en Elysia timida Risso 1818. Primeros re- 
sultados. Anales de Biología, Universidad de 
Murcia, 4: 37-47. 

SACCHI, CF., 1961, Vivificazione marina e mala- 
cofauna nel lago salmastro litoraneo di Patria 
(Napoli-Caserta). Annuario dell'lstituto e Museo 
Zoológico deír Université di Napoli, 13 (6): 1-37. 

SAUBADE, A.M., 1979, La malacofaune actuelle 
(Bivalves et Gastéropodes) de la lagune de Na- 
dor (côte méditerranéene du Maroc). Bulletin de 
l'Institut Géologique du Bassin d'Aquitaine, 26: 
69-90. 

SCHMEKEL, L., 1968, Ascoglossa, Notaspidea 
und Nudibranchia im Litoral des Golfes von Ne- 
apel. Revue Suisse de Zoologie, 75: 103-155. 

SCHMEKEL, L. & A. PORTMANN, 1982, Opistho- 
branchia des Mittelmeeres. Springer- Verlag, Ber- 
lin, 1-410 p. 

SCORDIA, C, 1927, La eurialinità di un mollusco 
opistobranchio allô stato giovanile. Rivista di Bio- 
logía, Milano, 9 (1): 3-42. 



SPARLA, M.P. & S. RIGGIO, 1984, Notes on the 
invertebrate fauna associated to the red alga 
Rytiphloea tinctoria (Clem.) С Ag. aegagropyla 
in the Stagnone sound. Nova Thalassia, 6 
(Suppl.): 105-111. 

TEMPLADO, J., P. TALAVERA & L MURILLO, 
1983, Adiciones a la fauna de Opistobranquios 
del Cabo de Palos (Murcia). I. Iberus, 3: 47-50. 

THOMPSON, Т.Е. & G. H. BROWN, 1984, Biology 
of Opisthobranch Molluscs, 2: Ray Society, Lon- 
don, 1-229 p. 

TORELLI, A., 1982, Gasteropodi conchigliati. 
Guide per il riconoscimento delle specie animali 
del le acque lag и nah e costiere italiane. 8. CNR, 
Rome, 1-235 p. 

TORELLI, A., 1983, Notes malacologiques sur les 
étangs sales de la Sardaigne méridionale. Rap- 
port de la Commission International pour l'Etude 
de la Mer Méditerranée, 28 (6): 229-230. 

TORTORICI, R. & P. PANETTA, 1977, Notizie eco- 
logiche su alcuni Opistobranchi raccolti nel Golfo 
di Taranto. Atti Società Italiana di Scienze Nat- 
urali, 118 (2): 249-257. 

VATOVA, A., 1940, Le zoocenosi délia Laguna 
véneta. Thalassia, 3 (10): 1-25. 

ZAOUALI, J., 1 974, La faune malacologique du Lac 
de Tunis (parties nord et sud) et de ses canaux 
(Canal Central, Canal Nord et Canal Sud). Hali- 
otis, 4: 179-186. 

ZAOUALI, J., 1979, Étude écologique du Lac de 
Bizerte. Bulletin de l'Office National de la Pèche, 
Tunisie, 3 (2): 107-140. 

ZAOUALI, J., 1981, Étude malacologique de la 
frange côtière sud du Lac de Tunis nord. Haliotis, 
11:241-250. 

ZAOUALI, J., & S. BAETEN, 1985, Étude des pe- 
uplements animaux macro-benthiques de la 
zone centrale et du bassin oriental de la Mer du 
Bibans (Tunisie méridionale) par la méthode de 
l'analyse factorielle des correspondances. Rap- 
port de la Commission International pour l'Etude 
de la Mer Méditerranée, 29 (4): 199-203. 

ZUCCHI STOLFA, M.L., 1977, Gasteropodi recenti 
delle lagune di Grado e Maraño. Atti Società Ital- 
iana di Scienze Naturali, 118 (2): 144-164. 



MALACOLOGIA, 1991, 32(2): 301-311 

THE STATUS OF THE RHODOPIDAE (GASTROPODA: EUTHYNEURA) 

L. von Salvini-Plawen 
Institut für Zoologie, Universität Wien, A-1090 Wien IX, Althanstraße 14, Austria 

ABSTRACT 

Based on investigations of Rhodope veranil, R. transtrosa sp. nov. and Helminthope psam- 
mobionta gen. et sp. nov., the organization of the Rhodopidae is reconsidered. Helminthope is 
characterized by a slender body, by typical verrucose rods, by lack of a radula, stomach and 
dorso-rostral caecum, and especially by five free ganglia on the visceral loop. The animals 
inhabit the interstitia of subtidal sands. The number of ganglia confirms earlier developmental 
investigations in R. veranil with respect to the pentaganglionate ( = euthyneurous) level. The shift 
of the visceral ganglion to the left side, as well as the lack of special vacuolar bodies in the 
epidermal cells, argue against a classification of the Rhodopidae within the Anthobranchia ( = 
Doridacea) and the Nudibranchia. The lack of both a paired procerebrum and cerebral gland 
excludes a direct relationship of the Rhodopidae with the Gymnomorpha (Onchidiacea, Sole- 
olifera) and Pulmonata. Furthermore, the free visceral ganglion in Helminthope and the monauly 
preclude a classification of the Rhodopidae amidst higher tectibranch groups ( = Paratectibran- 
chia). Consequently, the Rhodopidae, (including Helminthope and five presumed species of 
Rhodope) remain as a taxon Rhodopemorpha, of uncertain systematic rank and affinity, as a 
specialized off-shoot from the lower opisthobranchs. 



INTRODUCTION 

At the present time, the Rhodopidae are 
scientifically known only by the Mediterra- 
nean Rhodope veranil Kölliker and by the 
southwest Atlantic R. marcusi (see p. 308). 
Since the original description of R. veranil 
(Kölliker, 1847), few additional specimens 
have been found. Due to the investigations of 
Graff (1883), Böhmig (1983), and Riedl 
(1959,1960), however, we are fairly well In- 
formed about the anatomy, histology, biology, 
and development of this species; Marcus & 
Marcus (1952) supplemented this knowledge 
by the description of a closely related form 
(see p. 308). The central question about 
Rhodope concerns Its phylogenetic affinities. 
After the definitive classification of the spe- 
cies as an euthyneurous gastropod (RIedl, 
1960), its affinities within that subclass still 
remain uncertain (cf. Oberzeller, 1969; Sal- 
vini-Plawen, 1970; Tilller, 1984: 359). Further 
recent findings of Rhodope veranil, R. tran- 
strosa, and Helminthope psammobionta en- 
large our knowledge of the Rhodopidae and 
permit a re-evaluation of its systematic rela- 
tionships. 

Rhodope veranii Kölliker 

Fifteen Rhodope veranii were recently 
found In one of the marine aquaria of the Zoo- 



logical Institute (Universität Wien) filled with 
sediment and secondary hard-bottom mate- 
rial from the Northern Adriatic Sea and the 
Gulf of Naples. In nature, R. veranii appears 
to Inhabit shallow subtidal areas with stones 
and Ulva growth (Graff, 1883: p.74; F Star- 
mühlner, pers. comm., for Rovigno/lstna; Sal- 
vinl-Plawen in Arnaud et al., 1986: p 158). All 
specimens beyond 1 mm In length are char- 
acterised by the more or less distinctly T- 
shaped dorsal orange-red pigmentation 
(RIedl, 1960). In contrast to previously found 
animals with a maximum length of 4 mm 
(Graff, 1883: p.74; Riedl, 1960: p.297), the 
present individuals were distinctly larger, rang- 
ing up to 8 mm in length. The subepithelial 
spicules and the inconspicuous eyes are typ- 
ical. However, there is remarkable variation in 
the location of the genital opening: Graff 
(1883: p. 79) confused the protonephridiopore 
with the male gonopore and the anus with the 
female opening (both located at the right pos- 
terior border of the transverse pigment bar, i.e. 
anterior to the middle of the body). RIedl (1959: 
his Fig. 2) located the genital opening at the 
right anterior border of the transverse pigment 
bar. Irrespective of the state of contraction of 
the animals. Apparently the location of the go- 
nopore varies In different individuals. The ex- 
amination of six serially sectioned specimens 
(Riedl's and the present material) revealed 
that only one specimen possessed the genital 



301 



302 



SALVINI-PLAWEN 




FIG. 1. Rhodope veranii: Two successive cross sections through the cerebral nervous ring in a specimen 
with the foregut (fg) outside the pedal commissure (pc). oc eye, sta statocyst. 



opening in the location indicated by Riedl; two 
aninnals show the gonopore laterally at the 
level of the perioesophageal central nervous 
mass (with embedded eyes visible, see Fig. 
1), while three individuals show the genital 
opening distinctly anterior to the ganglia com- 
plex, viz. anterior to the eyes (in one the go- 
nopore is even located at the level of the 
mouth). 

In the concentrated nervous system the 
closely adjoining cerebro-pleuro-paheto- 
intestinal ganglia (cf. Riedl, 1960; Oberzeller, 
1969) have a short cerebral commissure and 
the eyes as well as optical ganglia incorpo- 
rated (Fig. 1 ); the optical connective itself has 
its origin in the pleuropedal connective. Be- 
sides the buccal connectives, there are three 
pairs of rostral nerves, the two medial ones 
with a common (?) root running to the oral 
region (labial nerves). The most lateral one at 
each side corresponds to the Hancock's or 
rhinophoral nerve in other opisthobranchs 
and fias a basal swelling which shows a dou- 
ble root in the cerebral ganglion; there is no 
head-shield-tentacle nerve (Huber, 1987). In 
addition, a strong lateral nerve, with bifur- 
cated root in both the (cerebro-) pleural gan- 
glion and the pedal ganglion, runs antehor- 
laterally to the body flanks; at the right it also 
innervates the copulatory organ. The two ab- 
dominal 'nerves' (right-visceral and left-gen- 
ital) running ventrally close to the body end 
are regularly provided with nuclei, thus as- 
suming the aspect of weak medullary cords. A 
peculiarity was noticed in one of the speci- 
mens: instead of being surrounded by the 
mass of the concentrated ganglia, the oe- 



sophagus runs outside (i.e. below) the pedal 
comissure (Fig. 1). 

The midgut shows the usual, somewhat 
winding, rostral caecum or right midgut gland 
(Riedl, 1960: p. 284). Close to the junction of 
the short intestine and the voluminous midgut 
there is a narrow pouch or small diverticulum. 
In both Riedl' s and the present material, this 
pouch is lined with a low, ciliated epithelium 
that is histologically continuous with the intes- 
tinal epithelium. In contrast to Böhmig (1893: 
p. 56 & Fig. 13), however, this pouch is well 
separated from the adjacent intestine, and, in 
agreement with Riedl (1960: pp.284-285), it 
corresponds to the remnant of the true stom- 
ach. 

The chromosome number of R. veranii is 
2n = 32 (pers. comm. Claudia R. Schweizer, 
Wien). The spermatozoa, with a spiralated 
head, have a characteristic shape and fine 
structure; in some aspects they appear to be 
fairly pnmitive and similar to prosobranch 
sperm (pers. comm. F. Giusti di Massa, 
Siena). 

Rhodope transtrosa sp. nov. 

A single specimen (Fig. 2A) was collected 
from an aquarium (Ehrmann Zoo, Wien XII) 
filled with phytal material from the tropical 
Indo-Pacific (Ceylon/Sri Lanka ?). The living 
animal measured 1.65 mm x 160 jxm maxi- 
mum. The anterior third of its whitish body is 
provided with a characteristic dorsal trans- 
verse bar (transtrum) of orange-reddish pig- 
mentation (about 160 \xvn in length). The an- 
teriormost section is markedly elongated and 




// » 



¡■í> 



^•.. 



■'••• i/ 



STATUS OF THE RHODOPIDAE 

В 



303 








200 Mm \i; '/V 






FIG. 2. Rhodope transtrosa: A: living animal (1.65 mm); В: semi-preserved animal; С: contracted animal; D: 
arrangement of organs as seen from the right side in preserved animal; E: spicules, a anus, ag adhesive 
gland, e eye, fg foregut, gg genital gland, go genital opening, о outlet of oral glands, ov ovarial sac, p 
protonephridiopore, sc spermatocyst, so hermaphroditic duct, t testicle, vg visceral ganglion. 



acts as a highly-bendable snout with subfron- 
tal mouth opening; in the contracted state this 
snout may be retracted far into the subse- 
quent, still pre-ocular section. The eyes are 



clearly visible in life. The body is somewhat 
truncated terminally due to the distinct adhe- 
sive organ. The spicules are fairly densely ar- 
ranged and are slightly curved with a faintly 



304 



SALVINI-PLAWEN 



verrucose surface. They measure 150-170 
цт X 14-17 цт (Fig. 2E). 

The internal organization closely resem- 
bles that of R. veranil, but there are some 
distinct differences. Ventral to the subterminal 
mouth, a short median sac bearing the pe- 
ripherally arranged oral glands opens (the 
glands are arranged as a paired cluster in R. 
veranil and R. marcusi). The foregut shows a 
precerebral enlargement with taller epithe- 
lium, but no pharyngeal bulb. This section re- 
ceives the salivary glands in paired arrange- 
ment. Immediately behind the central nervous 
complex the foregut leads dorsally into the 
sac-like midgut (midgut-gland). The latter 
forms an elongated organ extending from 
above the foregut to the posterior end of the 
body; in the preserved animal, there are sev- 
eral contraction-folds along its course, but 
no actual winding. Somewhat behind the 
mid-length of the body, the short intestine 
emerges dorsally from the sac-like midgut 
and runs directly, without any winding, to the 
nght. It opens laterodorsally closely behind 
the laterodorsal protonephhdiopore, both be- 
ing located (in contrast to R. veranii and R. 
marcusi) posterior to the middle of the body 
(preserved animal). Immediately adjacent, 
and to the left of the intestine, a small but 
distinct posteriorly directed pouch or divertic- 
ulum is present; this corresponds to the rem- 
nant of the true stomach in R. veranii. 

The nervous system largely resembles that 
of R. veranii with respect to the general ar- 
rangement of the ganglia and, on each side, 
the two proximally joined labial nerves, the 
double root of the rhinophoral nerve, the bi- 
furcated (pleural and pedal) root of the lateral 
nerve, and the optical ganglion emerging with 
Its connective from the pleuropedal con- 
nective. Differences are evident in the less 
concentrated state of the ganglia with the dis- 
crete statocysts between the cerebro-pleuro- 
parieto-intestinal ganglia and the pedal gan- 
glia, the discrete optic ganglia, the strong 
parapedal commissure, the fairly free and 
median visceral ganglion, as well as the sym- 
metrical origin of the right visceral and left 
genital medullary nerves (cf. Huber, 1987). 

In the hermaphroditic genital system there 
is a 55 X 45 ^JLm terminal testicle and a much 
larger testicle (70 x 50 ixm) more anteriorly on 
the left. The median hermaphroditic duct then 
connects two ovarial sacs on the left, one on 
the hght, and two more on the left (these be- 
ing located anterior to the anal region of the 
body). Approximately half way between the 



intestine and the central ganglia complex the 
spermoviduct turns to the right and continues 
in the form of a narrow connection with an 
enlarged portion filled with sperm. In contrast 
to R. veranii and more similar to R. marcusi, 
this sac represents a distinct elaboration 
(spermatheca) rather than a simple enlarge- 
ment of the spermoviduct (as in R. veranii). It 
opens anteriorly into a three-lobed glandular 
complex (albumen and mucus glands) from 
which the genital duct runs antero-laterally to 
open on the right at the level of the cerebral 
ganglia and eyes. In contrast to R. veranii and 
R. marcusi, no copulatory organ is developed 
in the present specimen. 

Helminthope psammobionta 
gen. et sp. nov. 

This mesopsammic species comes from 
the western North Atlantic. Specimens were 
collected by R. Rieger (Innsbruck) and W. 
Sterrer (Bermuda) from Bermuda (North Rock 
reef and Tobacco Bay, at 8-10 m depth), 
North Carolina (30 m depth) and Georgia (2 m 
depth). They inhabit fairly clean, coarse sub- 
tidal sands (cf. Rieger & Sterrer, 1975: pp. 
263-264 & their Figs. 34-35). The present 
animals ranged between 1 mm and about 2.5 
mm in length (Figs. 3-4) and are circular in 
cross section (diameter 60-150 ixm), but are 
able to contract by 30-50%. They are whitish 
with black eyes; in transmitted light they ap- 
pear transparent-colourless with a darker, 
somewhat greenish tinge to the midgut. The 
body openings are almost invisible as two cil- 
iated patches arranged one close behind the 
other on the right anterior side. These 
patches indicate the sites of the protonephhd- 
iopore and the anus. The genital opening 
could not be seen. Because of its internal or- 
ganization (below), the present specimens 
are defined as Helminthope psammobionta 
gen. et sp. nov. (Figs. 3-4). 

All three specimens sectioned were unfor- 
tunately poorly preserved for histological ex- 
amination, so only an outline of the body or- 
ganization can be given. The entire body is 
covered by ciliated epidermal cells among 
which gland cells are interspersed. In no ani- 
mal could a definite terminal gland be seen (in 
contrast to Rhodope veranil, R. marcusi and 
R. transtrosa; see Fig. 2D). The loosely and 
irregularly arranged spicules measure be- 
tween 45 X 5.5 ^Jlm and 70 x 7 \xm, and they 
are weakly curved to slightly angled or geni- 
culate in shape (Fig. 4C). Towards their tips 
the spicular surface is generally roughened. 



STATUS OF THE RHODOPIDAE 



305 




FIG. 3. Helminthope psammobionta: living specimen (about 1.5 mm In size) from Bermuda. 



and some spicules seem to be hollow (Rieger 
& Sterrer, 1975: Fig. 34). The spicules are 
situated in the fibrous connective tissue di- 
rectly below the epidermis and are sur- 
rounded by a spicule-forming cell (Rieger & 
Sterrer, 1975: Fig. 35). In addition to the elon- 
gate spicules (similar to those in R. veranii 
(Graff, 1883: p.75-76) or Я. transtrosa (Fig. 
2E)) very small and platelet-like elements are 
embedded subepithelially. As indicated by the 



high contractility of the body, there is a well- 
defined longitudinal musculature; no regular 
circular fibres could be seen. 

The alimentary canal begins with a subter- 
minal mouth. This leads into a narrow foregut 
which soon widens and is lined with a tall 
ciliated epithelium surrounded by circular 
muscle fibres (pharynx). A pair of ill-defined 
salivary glands accompany the foregut dor- 
solaterally. In the region of the cerebral ner- 



306 



SALVINI-PLAWEN 





50 Mm 



FIG. 4. Helminthope psammobionta: specimen from North Carolina gliding (A) and in contracted state below 
cover glass (B); C: spicules; D: main internal organization, ace accessory ganglia, anc right abdominal nerve 
cord (= visceral nerve), buc buccal ganglion, cer cerebro-pleural ganglion, gl epidermal glands, gt genital 
tube (spermoviduct), mc area of (reduced) mantle cavity, mg midgut, mo mouth opening, opt optic ganglion, 
par parietal ganglion, ped pedal ganglion, ph pharynx, sbi sub-intestinal ganglion, spi supra-intestinal gan- 
glion, sta statocyst, vis visceral ganglion. 



STATUS OF THE RHODOPIDAE 



307 



vous ring the foregut narrows again. Behind 
the pedal ganglia the oesophagus connects 
to the midgut. There is no antero-dorsal cae- 
cum (right digestive gland). The midgut rep- 
resents a homogeneous tube-like organ with 
high glandular epithelium extending the 
length of the body. There is no histological 
break during the course of the midgut except 
for the almost total disappearance of the 
lumen in the terminal portion (representing 
the posterior = left midgut gland?). The lu- 
men is also restricted to a somewhat nar- 
rower space posteriorly due to the genital or- 
gans. The intestine emerges from the midgut 
dorsolaterally (approximately 100 |xm behind 
the visceral ganglion) and runs obliquely di- 
rect to the lateral anus. Only in one specimen 
could an organ closely associated with the 
anus and extending a short distance anteri- 
orly be discerned, probably the protonephrid- 
ium. 

In the nervous system apart from the cere- 
bro-pleural complex all ganglia are separate 
(Fig. 4D). The fused cerebro-pleural ganglia 
still show a mid-dorsal incisure and a distinct 
cerebral commissure; the pleural portion is 
separated by an accumulation of nuclei. An- 
teriorly, at least two pairs of cerebral nerves 
(the labial and the Hancock's/rhinophoral) 
leave the ganglia. Very characteristic is the 
presence at each side of two complexes of 
accessory ganglia. These appear to be incor- 
porated in the course of the cerebral nerves, 
each complex assuming two to three swel- 
lings. 

A short connective runs from each cerebral 
ganglion to the respective pedal ganglion and 
a strong, terminal cerebropleural connective 
splits to connect with the optic, the parietal, 
and the pedal ganglion, as well as the buccal 
ganglion at each side. The buccal ganglia 
themselves are well separated and lie behind 
the statocysts. The small optic ganglia with 
the embedded eyes are separate from the ce- 
rebral ganglia and are located above the 
pedal ganglia (Figs. 4A,D). Each of the latter 
is very prominent and shows an anterior lobe. 
A lateral nerve of the pedal ganglion emerges 
at each side close to the pleuropedal connec- 
tive. The statocysts, each with a single stato- 
lith, are close by, but they are separate from 
the ganglia and main nerves. The parietal 
ganglia are widely separated from the cere- 
bral complex. On the right a strong connec- 
tion exists to the supra-intestinal ganglion la- 
teroventrally to the midgut, and to the left a 
much longer connective leads to the (ventre) 



lateral sub-intestinal ganglion. Both connec- 
tives unite in the mid-ventral visceral or ab- 
dominal ganglion, which appears to be lo- 
cated at a fairly constant distance of about 
135 jjim from the beginning of the cerebral 
ganglia. In histological continuation of the 
right (supra-intestinal) connective the strong 
genital 'nerve' emerges from the left portion of 
the ganglion and runs posteriorly as the left 
cord, and the left connective continues into 
the right cord (visceral nerve). Thus, there is a 
chiasma of fibres in the visceral/abdominal 
ganglion reflecting the only remaining trace of 
streptoneury. Furthermore, both abdominal 
'nerves' terminate together with the midgut- 
sac and exhibit a regular coat of nuclei. They 
thus assume the aspect of medullary cords 
and each gives off a strong anterior branch 
(Fig. 4D). 

The genital system is absent in one speci- 
men, is represented only by the rudiment of a 
simple tube extending below the midgut in a 
second specimen, and is not fully differenti- 
ated in the third sectioned individual. In the 
latter two individuals, the anteriormost part of 
the genital system is represented by a narrow 
tube extending posteriorly from between the 
buccal ganglia; the outer portion and genital 
opening could not be discerned and still ap- 
pear to be absent. The genital tube (sperm- 
oviduct) gradually enlarges posteriorly where 
it is lined with a high, glandular epithelium 
without a well-defined lumen (prostate 
gland?). The tube then continues posteriorly 
between the midgut and nervous system, its 
epithelium decreasing again in size toward 
the anal region. More posteriorly, approxi- 
mately 850 ixm from the anterior tip of the 
body (about 600 ixm behind the visceral gan- 
glion or 500 |jLm behind the anal region), the 
simple tube enlarges again to become a 
weakly ciliated vesicle filled with spermatozoa 
(spermatheca). This vesicle opens dorsally 
with its terminal narrowed portion into a volu- 
minous (albumen and mucus) gland, the an- 
terior region of which is lined by densely 
granulated and ciliated cells which are then 
replaced posteriorly by large slime cells. A 
narrow, ciliated duct continues from this gland 
and appears to become a ramified germ 
gland. This latter condition could not, how- 
ever, be ascertained in detail, and only some 
accumulation of sperm in vesicles (testicles) 
were observed. Thus, as is Rhodope veranii 
(Riedl, 1960:p.299), the present new species 
also appears to be protandric (gonochorism is 
also possible). 



308 



SALVINI-PLAWEN 



Systematic Discussion of the Rhodopldae 

In contrast to Rhodope veranii, R. tran- 
strosa and R. marcusi (see pp. 300, 302), the 
new type Helminthope psammobionta is cha- 
racterized and defined by the wide nervous 
system with free ganglia and the differentia- 
tion of precerebral ( = accessory) ganglia, by 
the axial connection of the foregut and nnidgut 
without anterior caecum, and by the lack of a 
ventroterminal adhesive gland. These cha- 
racteristics indicate that it belongs to a distinct 
genus. It is also characterized by the elongate 
body without pigmentation, the less verrucose 
and smaller spicules, the far postehor location 
of the spermatheca and the presence of the 
albumen/mucus genital gland behind it, as 
well as the interstitial habitat. There is, how- 
ever, no doubt that Helminthope is a rhodopid 
characterized by a pentaganglionate visceral 
loop with medullary visceral and genital 
nerves, subepidermal spicules, reduced mant- 
le cavity on the right side (anus and