Full text of "Malacologia"

See other formats

ans
PTR ET ET DIE BELLE EL кв ment

ее Peery kee TEEN AAA a

OS A A

so O

me AT

КЛАЙ
rs
D
PR PARDON
AIRE . СИИ
mass “
bad ube

pate ant eto И
PARTOUT CHOICE
RAA АКК CORTE
ALT vine u.
EL Hiatt nk ий EA » A LEURS
A A a anna

TEE ИИ

PANIERS TO
PA AS PARANA

var . » pe mt

na.

A
PAUSE

PAPA TOR à
AVENIR
ari
AN
A RT
РАО RAS
22

rt ste

ATEN
CPI
APTE
es re ;
CREER
= . . ose hate
DE
RTE
DO

RINA
TE

PRIRENT
А

PAPERS
REC

AA
A LES
CAR
dune tenet
nn RO Oe
a tern PERLES
ptas

ee
A еее

арт rain An
пря ха: TELE

o da мини.

Pagas

DICH

LIVE)
ETUI EEE

ne
АЛЯ. ENTER

CRT ET SUR

ama
nn NE Tan nein авы,
nn eh

MET Son EL LATE Darm
na CLP A ee

COR UT
men

man
A A

de IA

A Petit à

A mn SF
etn no Mea etme

MERS
Ту а ая

in ge мель
RES CE

HARVARD UNIVERSITY
€
Library of the

Museum of

Comparative Zoology

см 4
a
= | — y A
==
>
Y
г ' al
Л
=, ¡Le La e re р x 4 el 2

val Kidman: уф CITE AAN
o
ey Kuno lin
Y 2
aay 12)
| man sale

№ Matar alto a боль,

VOL. 30 1989

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 1988
Vol. 29, No. 1 28 June 1988
Vol. 29, No. 2 16 Dec. 1988

NO o A LIBRARY 1989

AUG 1 0 1989

Y { HARVARD

I.
In

ernational Journal of Malacolog y

148
я

_ Revista Internacional de Malacologia
Journal International de Malacologie
Международный Журнал Малакологии

Les |

Internationale Malakologische Ze itschrift

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.

Co-Editors:
EUGENE COAN CAROL JONES
California Academy of Sciences Vasser College
San Francisco, CA Poughkeepsie, NY

Assistant Managing Editor:
CARYL HESTERMAN

Associate Editors:

о ANNE GISMANN
University of Michigan Maadi

Ann Arbor Egypt

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

KENNETH J. BOSS, President JAMES NYBAKKEN, President-Elect =.)

Museum of Comparative Zoology Moss Landing Marine Laboratory

Cambridge, Massachusetts California

JOHN BURCH, Vice-President CLYDE F. E. ROPER р
Smithsonian Institution

MELBOURNE R. CARRIKER À

University of Delaware, Lewes RSI DEEE

GEORGE M. DAVIS W..D. RUSSELL-HUNTER

д 3 3
Secretary and Treasurer Syracuse University, New York

SHI-KUEI WU
CAROLE S. HICKMAN
University of California, Berkeley 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 Brussel, Belgium

Emeritus Members

J. FRANCIS ALLEN, Emerita ROBERT ROBERTSON
Environmental Protection Agency The Academy of Natural Sciences
Washington, D.C. Philadelphia, Pennsylvania
ELMER G. BERRY, NORMAN F. SOHL

Germantown, Maryland U.S. Geological Survey

Reston, Virginia

1989

EDITORIAL BOARD

J. À. 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 Sheffield
United Kingdom

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

B. C. CLARKE
University of Nottingham
United Kingdom

R. DILLON
College of Charleston
SC, U.S.A.

C. J. DUNCAN
University of Liverpool
United Kingdom

E. FISCHER-PIETTE
Muséum National d'Histoire Naturelle
Paris, France

VIRRENTER
University of Reading
United Kingdom

E. GITTENBERGER
Rijksmuseum van Natuurlijke Historie
Leiden, Netherlands

Е GIUSTI
Universita di Siena, ltaly

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
Тока! University
Shimizu, Japan

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
Bernice P. Bishop Museum
Honolulu, Hawaii, U.S.A.

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 Oswaldo Cruz, Rio de Janeiro
Brazil

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

W. Е. PONDER
Australian Museum
Sydney

R. D. PURCHON
Chelsea College of Science & Technology
London, United Kingdom

OZ
Academia Sinica
Qingdao, People's Republic of China

N. W. RUNHAM
University College of North Wales
Bangor, United Kingdom

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

G. A. SOLEM
Field Museum of Natural History
Chicago, U.S.A.

Е. STARMÜHLNER
Zoologisches Institut der Universitát
Wien, Austria

У. |. STAROBOGATOV
Zoological Institute
Leningrad, U.S.S.R.

W. STREIFF
Université de Caen
France

J. STUARDO
Universidad de Chile
Valparaiso

Т. E. THOMPSON
University of Bristol
United Kingdom

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
Domburg, Netherlands

. J. A. VAN EEDEN

Potchefstroom University
South Africa

М. Н. VERDONK
Rijksuniversiteit
Utrecht, Netherlands

В. В. 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)

MALACOLOGIA, 1989, 30(1-2): 1-303

COMPARATIVE MORPHOLOGY, PHYLOGENY AND CLASSIFICATION OF LAND
SNAILS AND SLUGS
(GASTROPODA: PULMONATA: STYLOMMATOPHORA)

Simon Tillier

Muséum national d'Histoire naturelle

Laboratoire de Biologie des Invertébrés marins et Malacologie

(CNRS UA 699)

55, rue Buffon, 75005 Paris, France

CONTENTS
Abstract Limacization
Introduction Discussion
Methods Central nervous system

General morphology
Size and shape
Size
Shape
External morphology of foot
Proportions
Interpretation of aulacopod condition
Internal anatomy
Description
Relative positions of organ systems
Proportions in visceral mass
Pulmonary complex
Morphological characters
General description, function
Statistical analyses
Non-orthurethran pallial complexes
Factor maps
Discussion
Orthurethran pallial complexes
Plesiomorphy and apomorphy in pulmo-
nary complex
Digestive tract
Anterior digestive tract
Buccal mass, salivary glands
Oesophagus, oesophageal crop
Gastric region, intestine
Gastric region
Intestinal loops, rectum

Cerebral ganglia, cerebral commissure
Lateral connectives, pedal ganglia
Visceral chain

Position and length

Position of ganglia in visceral chain
Discussion

Characters of stylommatophoran families

Orthurethra

Achatinellidae (+ Tornatellinidae)

Valloniidae

Pupillidae

Pyramidulidae

Chondrinidae

Cochlicopidae

Amastridae

Vertiginidae

Orculidae

Partulidae

Enidae
Non-orthurethran families

Zonitoidea: Zonitidae

Zonitoidea: Trochomorphidae

Zonitoidea: Euconulidae

Zonitoidea: Discidae

Zonitoidea: Arionidae

(+ Phylomycidae)
Zonitoidea: Parmacellidae
Zonitoidea: Limacidae

Zonitoidea:
Helicoidea:
Helicoidea:
Helicoidea:
Helicoidea:
Helicoidea:
Helicoidea:
Helicoidea:
Helicoidea:
Helicoidea:

Milacidae

Helicidae
Helminthoglyptidae
Bradybaenidae
Polygyridae
Camaenidae
Sagdidae
Haplotrematidae
Helicarionidae
Vitrinidae

Achatinoidea: Succineidae
Achatinoidea: Ferussaciidae
Achatinoidea: Subulinidae
Achatinoidea: Achatinidae
Achatinoidea: Streptaxidae
Achatinoidea: Oleacinidae

(+ Spiraxidae + Testacellidae)
Endodontoidea: Charopidae
Endodontoidea: Punctidae
Endodontoidea: Athoracophoridae
Endodontoidea: Endodontidae
Endodontoidea: Systrophiidae
Clausilioidea: Clausiliidae
Clausilioidea: Cerionidae
Clausilioidea: Urocoptidae
Clausilioidea: Bulimulidae

Acavoidea:

Oreohelicidae

(+ Ammonitellidae)

Acavoidea:
Acavoidea:
Acavoidea:

Corillidae
Acavidae
Rhytididae

(+ Chlamydephoridae)

Phylogeny and classification
Principles and methods of classification
of Stylommatophora

Phenetic nature of classical classifi-

cations

Principles of phylogenetic classifica-

tion

TILLIER

Method of phylogenetic classification

Paleobiogeography
Phylogeny and classification of Orthure-
thra
Phylogeny
Classification
Speculations on history of Orthure-
thra
Phylogeny and classification of non-
orthurethran Stylommatophora
Zonitoidea
Helicoidea
Achatinoidea
Clausilioidea
Endodontoidea
Acavoidea
Relationships of superfamilies
Classification
Conclusions
Acknowldgments
References
Figures
Appendix A. Materials studied, abbrevia-
tions used in Text-figures
Appendix B. Data (general morphology)
Appendix C. Upper limits of classes used
in factor analyses
Appendix D. Limits of classes used in
phenetic and phylogenetic analyses
(character states)
Appendix E. Character states used in
phylogenetic reconstructions
Appendix F. Character states of CCAs of
stylommatophoran families (nodes in
Text-figs. 20-22
Appendix G. Character states of nodes of
Text-fig. 28A

ABSTRACT

The morphologies observed in the pallial complex, digestive tract and
central nervous system of numerous stylommatophoran pulmonate gas-
tropod species are described and depicted. It is then possible both to
recognize numerous homoplasies in various characters and to arrange
the character states in morphoclines, of which the polarity is proposed on
the basis of outgroup comparison. Most of the homoplasies are probably
convergences related to size, shape, diet or limacization. Some are inter-
preted as indicating close phyletic relationships. The comparison of a
phenetic classification based upon morphological characters suggests
that most current, empirical classifications of the Stylommatophora are
based upon overall morphological similarity. In the new phylogenetic clas-
sification proposed, the order is divided into three supposedly monophyl-

suborders: Orthurethra Pilsbry, 1900; Brachynephra subordo nov.;
| Dolichonephra subordo nov. These suborders were probably already
ated in the Carboniferous.

STYLOMMATOPHORAN SYSTEMATICS 3

INTRODUCTION

Among the Gastropoda, the subclass Pul-
monata is characterized by the development
of a pulmonary cavity that fuses secondarily
with the pallial cavity (Fretter, 1975). Three
radiations may be recognized: the Archaeo-
pulmonata, which live in marine, littoral envi-
ronments but include a few terrestrial repre-
sentatives; the Basommatophora, which live
in freshwater; all of the Stylommatophora, un-
der study here, which are terrestrial (Tillier,
1984b). The latter include about 20,500 spe-
cies (Solem, 1978), and form the richest of
the three radiations. Land snails and slugs
occur from subpolar to tropical regions, and
their local abundance and diversity are a
function of climatic stability, at least at lower
taxonomic levels (Solem, 1984).

The classification of the suborder Stylom-
matophora is far from stable. Species are cur-
rently grouped into more than 1100 genera
that constitute nearly 60 families (1105 gen-
era and 60 families recognized by Zilch, 1960;
56 families recognized by Solem, 1978). Al-
though a general agreement may be found on
definition and contents of genera, the rank of
suprageneric taxa and the definition of supra-
familial taxa are still under discussion (е.д.
Solem, 1978; Boss, 1982; Schileyko, 1978a).
This relative confusion makes classification of
the Stylommatophora a mystery for non-
specialists. At least two reasons may explain,
in my opinion, why no systematist of the
Stylommatophora can explain how groups
should be formed: one is the unrepresentative
nature of samples that have been used in at-
tempts to find diagnostic characters; another
reason is the nearly total absence of discus-
sions of morphoclines (Maslin, 1952) that link
the various character states together in an ev-
olutionary perspective.

Solem (1978) has analyzed the enormous
gaps in sampling for characters; in the syn-
thetic volume edited by Fretter and Peake
(1975) on functional morphology and physiol-
ogy of pulmonates, only 25 of the 60 families
recognized by Zilch (1960), 53 (4.9%) of the
1105 genera and 47 (0.2%) of the 20500 spe-
cies are mentioned. Data on transformation
sequences in characters are either obviously
insufficient (nervous system morphology, his-
tology of genital apparatus and pulmonary
complex, shell and radula microstructure), or
show so much convergence that the use of
corresponding characters has, in most cases,
no practical value for recognition of higher

taxonomic groups (shell morphology, genital
and pulmonary morphology, pallial lobes,
etc.). The morphology of the digestive tract,
which occupies most of the general cavity, is
not even mentioned in any classificatory sys-
tem (Tillier, 1984a). In the cladistic terminol-
ogy, one can say that few or no synapomor-
phies have been recognized: although Pilsbry
(e.g. 1900a) explicitly used outgroup compar-
ison to define suborders, groups are currently
based on overall similarity of subgroups, and
the hierarchy of characters is variable as far
as it is defined.

The aim of the present work is to analyze
morphological patterns of macroevolution in
the Stylommatophora, in order to construct a
classification that may be rationally dis-
cussed. The work advances in three steps.
The first is character analysis and definition of
morphoclines. Such an analysis must be
based on a sample as representative as pos-
sible of all the Stylommatophora, and of their
morphological characters. In the earlier parts
of this paper, | try to discuss the functional
significance of observed character states and
to determine the polarity of morphoclines. Ev-
idence for generality of appearance of similar
derived character states in different groups is
given. The second step is description of fam-
ilies. Character states observed in each fam-
ily and basic biogeographic data are given in
the fifth section. The results of the first sec-
tions are used to construct combinations of
the most plesiomorphic character states ob-
served in each family, which are thought to
represent the closest possible common an-
cestor of each family. The final step is recon-
struction of phylogeny and classification. A
phylogenetic reconstruction, taking the fre-
quency of parallel evolution into account, is
attempted in the last section. A new classifi-
cation, based on phylogenetic hypotheses
that have not been rejected, is proposed.

METHODS

In order to analyze morphoclines at the su-
prageneric level, it is necessary to study a
sample representative ofthe various groups of
genera. Such a sample can be chosen only
within the frame of a preexisting classification.
| used as a point of departure Zilch's classifi-
cation (1960), as partly modified by Solem
(Helicarionidae: 1966a; Endodontoidea: 1976,
1982), Breure (Bulimulidae: 1979) and Tillier
(Systrophiidae: 1980). | admitted as a postu-

4 TILLIER

late that subfamilies recognized by these au-
thors form holophyletic groups. Obviously this
monophyly is based on similarity of genera
that constitute the various subfamilies; | do not
see how one could escape from this approx-
imation, which appeared to be false in a few
cases, without restudying representatives of
all of the more than 1100 genera. Within this
classification, | tried to examine at least one
representative of every suprageneric taxon.
Only in the beginning of the dissection were
several conspecific specimens dissected; af-
terwards | dissected several conspecific spec-
imens only when the first dissected specimen
had some feature looking odd to me. When
several species belonging to a single supra-
generic taxon were to be examined, | tried to
choose them as different as possible on prior
data, in order to approach the limits of variation
inthe taxon under consideration. The choice of
species under study was limited by the extinc-
tion of a few families and by the absence of
material in collections to which | had access.

A total of 243 species, representing 217
genera, 111 subfamilies (of 142) and 51 fam-
ilies (of 60), have been examined (Appendix
A). The nine families recognized by Zilch
(1960) that are not represented total 34 gen-
era, i.e. about 3% of all genera. Three (rep-
resenting 12 genera) are extinct and three
others are monogeneric (Pleurodiscidae, Oto-
conchidae, Thyrophorellidae). | considered
the Orthalicidae (seven genera) as Bulimul-
idae, the Trigonochlamydidae (seven genera)
as Limacidae, the Aillyidae (one species; Van
Mol, 1978) as Helicarionidae, and the Oto-
conchidae (one genus) as Charopidae, which
is an accepted position (Solem, 1978). To my
eyes the most important lacuna in the sample
is the absence of any of the Megaspiridae:
although the family is formed by only five gen-
era, it has an interesting distribution and its
position is still problematic (?European Upper
Cretaceous-Recent in Brazil and Australia;
Pilsbry, 1904).

In the last 20 years, there has been a trend
to define and use microscopical characters
for classificatory purposes: this has been
done by Van Mol (1967) for cerebral ganglia,
by Delhaye and Bouillon (1972a, b, c) for the
excretory apparatus, and by Visser (1973,
1977, 1981a, b) for the genital apparatus. In
my opinion this approach can be successful
only if very long-term, because histological
techniques are so laborious and time-
consuming that the size of the sample exam-
ined within a reasonable time is necessarily

reduced. Therefore the authors who use mi-
croscopical characters are obliged to sup-
pose that characters observed in a very few
species are present in large groups, a point
which none of them discusses: the use of
these characters relies on the underlying as-
sumption that characters are more general
when smaller, which seems very question-
able. Given the size of the sample under
study here, characters used are either mac-
roscopic or visible at magnifications up to
100x. In this order of magnitude accessible
characters are those of the morphology of the
animal and of its main organ systems: diges-
tive tract, genital apparatus, central nervous
system, pallial complex, arterial system. Only
characters whose morphoclines seemed well
defined at taxonomical levels under study
(from the genus to the suborder) were re-
tained: most characters of the external mor-
phology, of the arterial system and of the
genital apparatus were eliminated. Another
reason for eliminating almost totally external
and genital morphologies is that very many
more data may be found in the literature on
these morphologies than on morphologies of
other organ systems, and | preferred to de-
scribe and discuss the latter first.

All drawings were made under a camera
lucida. In practice the shell of the animals,
when present, was first observed and mea-
sured (height, diameter, number of whorls)
and then cracked or dissolved in acid. The
animal without shell was drawn (slugs in dor-
sal view; axis of flat visceral masses perpen-

dicular to the drawing plane; axis of elongate

visceral masses parallel to the drawing plane:
Tillier, 1984a). Then the lung roof was re-
moved, spread, and drawn after the lower
border of the pneumostome had been cut to
allow observation of the anus and ureter
opening. After this the rectal and visceral bor-
ders of the kidney were cut, the pulmonary
kidney wall was folded back upon the pericar-
dium, and the internal kidney morphology
was drawn. The animal was then placed
again in the same position as in the initial
drawing, and the digestive tract was progres-
sively revealed by dissection and drawn be-
fore its position within the general cavity was
altered by further dissection. Its internal mor-
phology was diagramed in the same drawing.
Finally the central nervous system was re-
moved after the nerves and the oesophagus
had been cut, and drawn twice in dorsal view:
once before, and once after the cerebral com-
missure had been cut and the cerebral gan-

STYLOMMATOPHORAN SYSTEMATICS 5

glia had been spread on each side of the vis-
ceral chain.

Measurements and observations were
treated by factor analysis. For each analysis
and each variable, the sample was divided
into classes of equal effectives, and each mo-
dality so defined was treated as an indepen-
dent variable coded O or 1 (matrix written in
complete disjunctive form). First, partial anal-
yses were performed by a version of the ANA-
FAC program (Jambu & Lebeaux, 1979)
adapted by Michel Roux and Raymond Bau-
doin for Apple IIE or ИС microcomputers.
Then overall analyses were performed by the
ADDAD programs in the CIRCE through the
Centre Informatique du Muséum. There is no
methodological justification for partial analy-
ses, but only a practical one: the data set is
too large to be handled by a microcomputer.
Histograms were drawn by a program, written
by Raymond Baudoin, which allows the loca-
tion of every individual, and thus the visual-
ization of the range of every taxonomical
group, in the histogram; simplified versions of
these histograms are presented below.

The algorithms used for phylogenetic re-
constructions are derived from an unpub-
lished work by the late Pierre Delattre, and
are described in the final section. The pro-
gram was written by Raymond Baudoin.
Felsenstein's PHYLIP program, version 2.4,
was also used.

GENERAL MORPHOLOGY

In all land snails, one can recognize the
conical visceral mass, which is coiled into a
spiral helix (Thompson, 1917), the epithelium
of which secretes the shell; and the foot,
which can be retracted into the distal portion
of the visceral mass. The coil of the visceral
mass is generally dextral, but may be sinis-
tral: where the words right and left are em-
ployed below, they must be interchanged to
apply to sinistral animals. The pallial border
forms the limit between the foot and the vis-
ceral mass, and is generally the only part not
protected by the shell when the animal is re-
tracted (Solem, Tillier 4 Mordan, 1984). The
pneumostome opens on the right side of the
pallial border, in the parieto-palatal angle of
the shell aperture (Text-fig. 1). The inferior
surface of the foot is the pedal sole, which is
used for crawling by means of mucus se-
creted by the pedal gland, whose opening is
situated above the anterior extremity of the

pedal sole (mechanism analyzed by Jones,
1973, 1975). The part of the foot anterior to
the pallial border is the cephalic region, or
head, which bears the invaginable ocular ten-
tacles and rhinophores. The part of the foot
posterior to the pallial border is the tail, whose
posterior extremity sometimes includes a
caudal mucous gland (discussed by Pilsbry,
1896).

In the course of limacization, the number of
whorls described by the visceral mass is re-
duced, the contents of which are incorporated
into the foot (Van Mol, 1970; Tillier, 1984a). In
semislugs, the distal part of the visceral mass
is too reduced to include the whole of the re-
tracted foot, but the stomach is included in the
visceral mass, above the pallial border. In full
slugs, the visceral mass is still more reduced
or absent, and the stomach is included in the
pedal cavity. The pallial border extends over
the surface of the reduced shell and may
cover it totally, forming a dorsal shield
(Solem, 1966a); the shell might be absent in
full slugs. In the most limacized slugs, the
shield itself can hardly be distinguished from
the dorsal surface of the foot (Philomycinae,
Athoracophoridae, Chlamydephorinae), ap-
proaching the structure of an opisthobranch
notum.

Size and shape

Size: The largest linear dimension of ani-
mals studied here varies between 1.2 and
about 140mm, which in first approximation
corresponds to a ratio of 1 to 100,000 in vol-
ume. One might expect that such a difference
in size would be correlated with differences in
anatomy, in accord with Galileo's principle of
similitude (Thompson, 1917; Lambert & Teis-
sier, 1927). If such correlative variations oc-
cur, detecting them in animals partly coiled
into a spiral, torted around this spiral and lack-
ing any internal skeleton will be particularly
difficult. Insofar as it is admitted that charac-
ters used in phylogenetic reconstructions
should be functionally independant (Diels,
1921, in Hennig, 1966; Cain, 1982), it is im-
portant to detect the extent to which the prin-
ciple of similitude applies to the morphoclines
that are recognized.

Although | did not make statistical tests to
check this statement, it seems to me that the
relative size of the foot is larger when animals
are large; the foot is relatively smaller in small
species than in large ones. The principle of
similitude may explain this observation: since

6 TILLIER

it is constituted to a large extent by the mus-
cles that move the animal, one can expect
that the foot is relatively larger in large ani-
mals (problem abundantly illustrated by
D'Arcy Thompson, 1917). The same feature
may be related to a surface/volume ratio
problem (ibid.): evaporation from the surface
of the foot can be stopped only by retraction
of the animal within its shell (Machin, 1975)
and is proportional to this surface when the
animal protrudes from the shell. Therefore
evaporation is more important, relative to the
volume of the animal, in a small animal than in
a large one, and the only mechanical means
to keep the ratio evaporation/volume constant
when size diminishes is to reduce the relative
size of the foot.

The necessity of avoiding relative increase
inthe evaporation surface as size decreases
also allows one to expect the aperture of
small shells to be relatively smaller than that
of large shells, because when the animal is
retracted into the shell, evaporation occurs
from the surface of the animal that remains
exposed at the aperture (mantle border or
some part of the foot: Solem, Tillier & Mordan,
1984). It seems to me that this is the case,
and that the same hypothesis might explain,
in possibly a more satisfying way than does
the hypothesis of their function as a barrier to
predators (Solem, 1966b), why apertural bar-
riers are particularly common in small shells.
In Polynesian Partula, the surface area of the
aperture is correlated with mean rainfall (Em-
berton, 1982); even in large snails such as
New Caledonian Placostylus, there is a pos-
itive correlation between aridity of environ-
ment and development of apertural barriers
(Cherel-Mora, 1983).

Slugs have no shell to protect themselves
from desiccation, and can use only physiolog-
ical means and contraction of their foot to limit
evaporation from the pedal surface. An indi-
cation that these solutions are less efficient
for small animals than retracting into a shell
may be found in the observation that, in the

sample under study here, the smallest slugs
are much larger than the smallest snails (ratio
of 1 to 5 in maximal length, i.e. about 1 to 125
in volume). However, one should not con-
clude that slugs can not live in dry environ-
ments: a large slug may have a smaller sur-
face/volume ratio than a small snail, and may
burrow to escape desiccation and perhaps
more easily than a snail, whose shell cannot
be deformed. In fact, southern European
large parmacellids are found in xeric environ-
ments where no snail lives, and it is not ab-
surd to consider that the limiting factor for
limacization in xeric environments is not arid-
ity, but rather food supply sufficient for ani-
mals large enough to have a favorable surface/
volume ratio.

In his discussion of the definition of size
parameters of snails, Gould (1969) has shown
that at least two linear measurements, height
and diameter of the shell, are necessary. He
used their sum whereas other authors used
their product (Tillier, 1981; Cherel-Mora,
1983). Inthese cases, the shape was reason-
ably constant within each group under study.
Here we deal with snails, semislugs and slugs
exhibiting a large range of shapes. Conse-
quently the various snails and slugs under
study were arranged following their apparent
size from 1 to 236 (snails without shell; spec-
imens belonging to seven species were no
longer available when this was done), and
their rank was used as a statistical parameter
(Appendix B). Approximate measurements of
the volume of 40 specimens without shell
showed that the rank so determined is an in-
creasing function ofthe volume ofthe animals.

Shape: It is equally important to detect cor-
relations between the shape of the animals
and character states: | showed elsewhere
(Tillier, 1983, 1984a, 1984b) that some char-
acter states used as synapomorphies of or-
ders are only the most apomorphic states of
morphoclines related to limacization. Like-
wise, the important variations occurring in the

TEXT-FIG. 1. Plan of organization of the Stylommatophora. 1A, sagittal section; 1B, dorsal view; 1C, shell
and terminology of the various regions of the apertural border, of which the growth generates the corre-

sponding areas of the shell.

The visceral mass has been drawn uncoiled, which is biological nonsense, for asymmetry, torsion and
coiling result from a single cause, differential growth of the visceral mass.
AD, opening of anterior duct of digestive gland; AO, aorta; APG, hermaphrodite portion of genital apparatus;

AU, auricle; BM, bu
са Y: СУГ VIS(

kidney: PI

free retractor muscle:

cal mass; BP, pallial border; CGE, general cavity; CPE, pedal cavity; CPU, pulmonary
eral cavity; D, diaphragm; HD, hermaphrodite duct; HG, hermaphrodite gland; I, intestine; К,
al retractor muscle; PD, opening of posterior duct of digestive gland; PS, gastric pouch; RM,
C, gastric crop; SN, nervous system; V, ventricle.

STYLOMMATOPHORAN SYSTEMATICS

BASAL

8 TILLIER

0 1 2

N 189

OMEAN 1.18

SD 0.706

—. ALL

-- ORTHURETHRA
— DOLICHONEPHRA
BRACHYNEPHRA

á 5
SHELL HEIGHT/ SHELL DIAMETER

TEXT-FIG. 2. Shell shape of snail and semislug species studied (ratio shell height/shell diameter; data,
Appendix B). Suborders Brachynephra and Dolichonephra defined in last section.

shape of visceral mass of snails can be cor-
related with anatomical character states that
could be interpreted as bringing additional in-
formation to bear on phylogenetic relation-
ships. Furthermore, recent works by Cain
(1977, 1978a, b, 1980, 1981), Chérel-Mora
(1983), Emberton (1982), and Tillier (1981)
tend to show that shell shape has an impor-
tant adaptive value at the level of the guild of
terrestrial gastropods.

As noticed by D'Arcy Thompson himself
(1917: 515), his work on modelling of gastro-
pod shell shapes is useless for analyzing the
internal variations related to variations in ex-
ternal shape: his models, like those of Raup
(1966), describe only surfaces. To analyze
the internal variations related to variations in
shape, we should define a system of coordi-
nates within the cone representing the vis-
ceral mass, and study the deformations of this
system when this cone is coiled into a spiral
helix whose height and diameter vary. Being
unable to build such a model, | used the ratio,
height of shell/diameter, to define the shape
of the visceral mass, as done by Solem
(1966b), Cain (op. cit.) and others (Appendix
B). Shapes can be compared precisely by this
ratio only if they are similar, which is not the

case here where such a gross approximation
must be compensated by wide classes. In Ap-
pendix B, the shape of semislugs is repre-
sented by the number 1000 and the shape of
slugs is represented by the number 2000.

_ The histogram of snail shell shapes (189
species; Text-fig. 2) exhibits for all Stylom-
matophora the bimodality already discussed
by Cain (op. cit.) for various faunas: average
shapes (H/D = 1.18) are less represented
than elongate and flat shapes.

The number of whorls of the visceral mass
within the shell is another important shape pa-
rameter, which is independent of size. When
low, it constitutes an index of the degree of
limacization. Generally the viscera do nat fill
the shell cavity up to its top. In most cases,
the apical portion of the shell cavity is filled
with mucus-like matter in alcohol-preserved
specimens. Comparison of the histograms
(Text-figs. 3,4) shows that there is seemingly
no simple relationship between the number of
whorls of the shell and the number of whorls
of the visceral mass (Appendix B). Although
in general the length of the visceral mass in-
creases with the length of the shell (in whorls),
the number of whorls in a shell does not al-
low one to predict the number of whorls in

STYLOMMATOPHORAN SYSTEMATICS ©

N 185

OMEAN 5.36

SD 0.14

= ALE

-- ORTHURETHRA

— DOLICHONEPHRA
BRACHYNEPHRA

8.259 210, 910,12
SHELL WHORLS

TEXT-FIG. 3. Whorl number of shells of the snail and semislug species studied (data, Appendix B).

the corresponding visceral mass. Intraspecific
variation in visceral mass whorl number is rel-
atively important (studied in only one genus,
the New Caledonian charopid genus Para-
rhytida: Tillier & Mordan, 1986). | must sup-
pose that the size of the sample divided into
broad classes eliminates the effects of this
variability in the statistical analyses of mor-
phological characters.

External morphology of foot

Proportions: Observation of stylommato-
phorans alive shows that the proportions of
the foot do not vary by chance more than rel-
ative foot size. In order to discuss these vari-
ations, it is necessary to distinguish between
aulacopod feet, surrounded by a well-defined
suprapedal groove, and holopod feet, which
lack such a groove (Pilsbry, 1896). As stated
by Solem (1978), “the basic distinction seems

a real one”; but in my opinion it is not possible
to use aulacopody as a synapomorphy before
its biological meaning and other characters of
the group which its presence defines (Aula-
copoda Pilsbry, 1896; Solem, 1978) have
been discussed.

In the Stylommatophora, width of the pedal
sole is generally greater in proportion to sole
length in large animals than in small ones.
This feature might correspond to the maximi-
zation of the surface used for creeping, which
allows the animal to keep the ratio of the sur-
face of the pedal sole to the weight of the
animal nearly constant as size increases: this
feature is reminiscent of the relatively large
diameter of the bones in the legs of large te-
trapods, discussed by Galileo long ago. For
the same size, holopod feet are relatively
wider than aulacopod feet. This difference in
pedal sole shape seems to be related to dif-
ferences in body shape of animals that exploit

10 TILLIER

N 188

OMEAN 3.83

SD 0.86

— ALL

-- ORTHURETHRA

— DOLICHONEPHRA
BRACHYNEPHRA

WHORLS VISCERAL MASS

TEXT-FIG. 4. Length in whorls of visceral mass of snails and semislugs studied (data, Appendix B).

vertical or oblique surfaces. lt has been
noticed that there is a correlation between
elongation of the shell and exploitation of
such surfaces (Cain & Cowie, 1978; Cam-
eron, 1978, 1981; Cook 8 Jaffar, 1984); but
no author noticed that all the animals seen to
have such a correlation also have a holopod
foot, probably because in the Northern Hem-
isphere there are very few animals with an
aulacopod foot, which exploit vertical or
oblique surfaces. In Africa, Southeast Asia
and the Pacific Islands, many arboreal or
semi-arboreal aulacopod snails are helicari-
onids and endodontoids. The shell of these is
not elongate as in most holopod species that
exploit vertical or oblique surfaces, but the
proportions of the foot are different: in most
arboreal aulacopod snails that | have ob-
served, the head is clearly shorter than the
tail and the foot is especially narrow, whereas
the foot of related snails that live in the same
forests at ground level has more usual
proportions, 1.е. it is wider and the head is
about as long as the tail.

Snails whose head is clearly longer than
the tail are not common. As noted by Watson
(1915), the most evident functional explana-
tion for such proportions is aptitude for car-
nivory, which involves a large buccal mass
and consequently a large pedal cavity and a
long head (Rhytididae and Oleacinidae, but
no observed Streptaxidae, in which the great
length of the lung allows housing of the buccal
mass in the lower part of the visceral cavity).

Interpretation of aulacopod condition: Al-
though the aulacopod condition is used as a
synapomorphy in the Pilsbry-Baker system,
no interpretation of its origin has been pro-
posed (Watson, 1920; Wächtler, 1935; Baker,
1955; Solem, 1978). The distinction between
aulacopod and holopod morphologies is often
not very clear-cut (Solem, 1978). With a few
exceptions (Pilsbry, 1946), the distinction is
easy in groups placed by Solem (1978) either
in the Holopoda or in the Aulacopoda. But, as
noticed by Solem himself, an aulacopod or
quasi-aulacopod condition might occur not

STYLOMMATOPHORAN SYSTEMATICS 11

only in some representatives of most groups
of the Holopodopes sensu Solem (Ferussaci-
idae, Subulinidae, Spiraxinae, Streptaxidae,
Systrophiidae), but even in some Mesure-
thra sensu Solem (some Clausiliidae).

In nearly all species with an aulacopod foot,
whether belonging to the Aulacopoda sensu
Solem (endodontoids excepted), shell growth
is indefinite: passage to the adult stage, indi-
cated by maturation of the genital apparatus,
is not marked by any change in the growth of
the shell, the border of which remains sharp
and unexpanded. On the other hand, shell
growth is definite in nearly all animals having
a holopod foot, including the Holopoda sensu
Solem: as the genital apparatus reaches
maturity, the orientation in growth of the last
whorl is often modified and the apertural
border is thickened or expanded. Finally in
the few cases (about ten) in which | could ob-
serve well-developed embryos or newly born
juveniles of orthurethran and holopod species
(sensu Solem, 1978), their foot was distinctly
aulacopod. From these three sets of ob-
servations, | propose the following hypothe-
ses: first, the aulacopod morphology is a
character of juvenile stylommatophorans;
second, in groups in which adult animals have
an aulacopod foot, the aulacopody is a pae-
domorphosis that might be regarded as
neoteny (retardation) in the sense defined by
Gould (1977): the absence of an adult mor-
phology in the shell of aulacopod animals in-
dicates that everything seems as if most aul-
acopod snails continue in the juvenile mode
of growth until after sexual maturity has been
attained.

However, this paedomorphosis in foot and
shell morphology occurs to a variable extent:
in some groups, an aulacopod foot is as-
sociated with a shell with definite growth
(e.g. some Trochomorpha with an expanded
aperture, endodontoids) and in some other
groups, a holopod foot is associated with
indefinite shell growth (e.g. Achatina fulica).
This shows, at least as far as the hypothesis
of the paedomorphic origin of aulacopody is
true, that paedomorphosis may be affected
by mosaic evolution and is not necessarily
effective for all organs of an organism; a point
that is confirmed by the observation of a
paedomorphic morphology in the digestive
tract of snails whose other characters exhibit
adult morphology (Vallonia, Zonitoides de-
scribed below). As stated by Delsol and
Flatin (1979), paedomorphosis should be
discussed character by character.

Internal anatomy

Description: Two principal cavities, the pul-
monary and the general, may be distin-
guished in land snails. During ontogeny, the
pulmonary cavity forms by invagination of the
mantle border, which secondarily fuses with
the pallial cavity, which is homologous with
the prosobranch pallial cavity, the latter form-
ing the pneumostome region (Regondaud,
1964; Fretter, 1975; Tillier, 1984b). The heart
and kidney form the proximal part of the lung
roof, and the whole system is called the pallial
complex or pulmonary complex. The general
cavity contains the digestive tract, central ner-
vous system, arterial system and free retrac-
tor muscles (Text-fig. 1). For convenience in
description, | shall distinguish within the gen-
eral cavity the pedal cavity, extending from
the anterior extremity of the foot to the level of
the pallial border, and the visceral cavity,
which occupies all the visceral mass except
the lung cavity. This distinction is purely for-
mal, for the same organ systems extend with-
out any discontinuity from the pedal cavity
into the visceral cavity. The pedal cavity does
not extend into the tail of snails as it does in
slugs, of which the visceral mass is included
in the foot. In such slugs the visceral cavity,
pedal cavity and general cavity are indistin-
guishable (Van Mol, 1970, 1971; Тег,
1984a).

The arrangement of the digestive tract, the
object of incredibly few descriptions, has
been discussed elsewhere (Tillier, 1984a).
The oesophagus (OE, Text-fig. 1, Figs.)
opens dorsally from the anterior buccal mass
(BM). lts internal ornamentation, if present,
consists in longitudinal ridges. The oesopha-
gus might be partly differentiated into an in-
flated oesophageal crop (OC, Figs.); but, con-
trary to the statement of most treatises of
zoology (e.g. Franc, 1968), many stylom-
matophoran species do not have an oesoph-
ageal crop (Figs.). Two salivary glands (SG)
are appressed to the oesophagus. They com-
prise numerous acini whose ducts converge
into two main salivary ducts, which open into
the buccal cavity on each side of the oesoph-
ageal opening. The oesophagus runs back-
ward along the parietal side of the visceral
cavity, and most generally expands into a
gastric crop (SC) a short distance above the
top of the lung. The most common internal
ornamentation of the gastric crop, if present,
consists of two ventral ridges that delimit a
groove; more longitudinal ridges might be

12 TILLIER

present. The gastric crop is prolonged by the
caecum of the gastric pouch (PS), from which
the intestine (l) runs forward and ventrally.
The anterior duct of the digestive gland (AD)
opens into the concavity of the gastric pouch,
between the openings of the gastric crop and
proximal intestine. The posterior duct of the
digestive gland (PD) generally opens through
the parietal wall of the gastric pouch. When
present, the ventral groove of the gastric crop
leads to the opening of the anterior duct, from
which one usually short typhlosole emerges
into the proximal intestine. A second, longer
typhlosole, issuing from the opening of the
posterior duct, runs parallel to the first one
into the proximal intestine and reaches at
most the beginning of the periaortic intesti-
nal loop. A ridge might join the openings of
the ducts of the digestive gland and with the
proximal portions of the typhlosoles delimit
a curvilinear triangle on the parietal wall of
the gastric pouch. The arrangement, length
and thickness of these ridges are variable
even within a single family. The intestine runs
forward along the columellar side of the vis-
ceral mass, turns to the left under the anterior
gastric crop or the posterior oesophagus,
turns around the aorta clockwise in dorsal
view (periaortic bend) before turning forward
again (prerectal bend). The rectum runs along
the suture from the summit of the lung to the
roof of the pneumostome.

The genital apparatus will be much less dis-
cussed here than other systems (Duncan,
1975). The common genital orifice opens on
the right side of the head. The hermaphrodite
branch runs along the parietal side of the vis-
ceral mass to a point above the stomach,
where the hermaphrodite gland, formed by
one or several clumps of acini, is embedded
in the upper lobe of the digestive gland (HG,
Text-fig. 1). Above the lung, the albumen
gland either spreads between intestinal loops
distal to the gastric crop (Orthurethra, Aula-
copoda sensu Solem, 1978), or spreads ap-
pressed along the gastric crop (most Hol-
opoda sensu Solem, 1978). When very long,
the spermathecal stalk turns to the left along
the periaortic intestinal bend at the level of the
summit of the lung, and the spermathecal
head lies appressed beneath the pericardium
on the left side of the aorta. In large species,
the vagina is attached to the pedal wall by
very short muscles. Exceptionally, one
branch of the free retractors might insert on
the atrium or close to the base of the sper-
matheca (Tillier & Mordan, 1986). The penis

connects the hermaphrodite branch of the
genital apparatus into a genital atrium close to
the genital orifice, and is generally attached
by a penial retractor muscle (PR) to either the
lung floor (= diaphragm) or the stem of the
free retractor muscles.

The central nervous system (SN, Text-fig.
1) is always entirely contained in the pedal
cavity in the Stylommatophora. It is formed by
two dorsal cerebral ganglia (CG, Figs.) con-
nected by cerebral commissure (CC) above
the digestive tract, two latero-ventral pleural
ganglia (PLD, PLG) connected by a visceral
chain of ganglia beneath the digestive tract,
and two ventral pedal ganglia (PG). The vis-
ceral chain comprises two parietal ganglia
(PAG, PAD) and one median visceral gan-
glion (VG). The cerebro-buccal connectives
join the lower surface of the cerebral ganglia
to the buccal ganglia, which are appressed to
the buccal mass on each side and below the
origin of the oesophagus.

The free retractor muscular system (RM,
Text-fig. 1, Figs.) is made of branches of the
columellar retractor. The latter is inserted on
the inner shell surface along the columella;
this insertion is the only fixed point in the sys-
tem formed by the animal and its shell. As
may be expected from the principle of simili-
tude, the free retractor system is more devel-
oped in large animals; the lateral pedal retrac-
tors in particular are very important in large
snails such as Helix pomatia (Jones, 1975:
Fig. 1) and are absent from small species.
Other free retractors are the buccal retractor
muscle, distally divided into at least two
branches inserted under the buccal mass,
and the ocular retractors, distally divided into
ocular and rhinophoral branches. In all cases
the free retractor system passes along the
right side and beneath other organ systems in
the pedal cavity, the genital apparatus ex-
cepted. The latter, penis excepted, lies on the
right side and beneath the free retractor sys-
tem, and the right ocular retractor generally
passes between the penis and vagina. This
general pattern varies, particularly with taxo-
nomic position and degree of limacization
(Solem, 1978; Tillier, 1984a). My impression,
based on dissection of several hundred taxa,
is that the free retractor system might be use-
ful for comparison of closely related genera
(Wurtz, 1955), but that convergence is far too
common for it to provide useful data in an
analysis of morphoclines at familial or supra-
familial levels.

The principal branches of the arterial sys-

STYLOMMATOPHORAN SYSTEMATICS 13

tem are remarkably constant in their arrange-
ment (Duval & Runham, 1981). The cephalic
aorta is often appressed to the underside of
the diaphragm in large snails, but never in
slugs.

Relative positions of organ systems: The
relative positions of the various organ sys-
tems and of the pulmonary complex must be
kept in mind, in order to understand some of
the constraints on anatomical transformations
and relative movements during contraction
within the shell. There are some anatomical
constraints (Text-fig. 1). First, the central ner-
vous system surrounds the oesophagus,
which limits nervous concentration: the latter
limits food size, which seems surprising to
vertebrates like us. Second, the aorta must
cross the periaortic bend of the intestine:
some a priori bizarre arrangements in the di-
gestive or arterial systems of some slugs and
semislugs may result directly from this con-
straint (Tillier, 1984a; Testacellinae: Lacaze-
Duthiers, 1887; Chlamydephorinae: Watson,
1915). Third, the position of every organ sys-
tem relative to the others is generally fixed by
two points on the surface of the general cavity
(genital orifice and penial retractor insertion
for genital apparatus, free muscle insertions,
mouth and anus for digestive tract, etc.). To-
pological transformations of the organs are
possible only to the extent that two portions of
organs between two fixed points do not over-
lap. For example, the relative position of the
penis is quite constant when a penial retractor
is present (succineids and snails having a
free right ocular retractor excepted); but the
relative position of the rest of the genital ap-
paratus, which is fixed only by the genital or-
Шсе and the genital branch of the aorta, may
vary greatly in slugs: on the right side of the
digestive tract in the Helicarionidae, on the
left and underneath in Arion, above in Omal-
onyx, etc.

It is equally important to distinguish fixed
from mobile parts in the movements of the
animals. Fixed parts are all those that are prox-
imal to the tops of the lung and the columellar
retractor. Indeed, there is no muscle that
would allow movements of the proximal part of
the visceral mass within the shell; if such
movements occur, they are provoked only by
a violent retraction of the foot or by variations
in hydrostatic pressure, but they do not affect
the relative positions of the organs. This fact
allows us to use measurements of the various
organs above the lung, at least for a variation

that was analyzed in only one genus (Para-
rhytida: Tillier 8 Mordan, 1986): distance from
the top of the lung to the top of the stomach,
length occupied by intestinal loops, length of
the part of the visceral mass above stomach
(Appendix B). These lengths have been mea-
sured in whorls, in order to have measure-
ments that are independent of size.

When a snail retracts within its shell, the
ocular tentacles and rhinophores are first in-
vaginated. Then the buccal mass is pulled
backward by the buccal retractors, whose
contraction invaginates the anterior part of the
head. Next, pedal retractors, when present,
contract progressively backwards. When re-
traction is achieved, the head is totally invagi-
nated and the tail is generally proximal to the
pallial border (Solem, Tillier & Mordan, 1984).
All organs between the mouth and the top of
the lung are folded, or translated into a space
formerly occupied by the lung cavity, by
means of the expansion of the diaphragm.
The lung roof may be folded (by retraction of
the columellar muscle?), but seemingly not
contracted (no intrinsic muscles), such that
the pallial border is translated as far as pos-
sible from the shell aperture. When the nerve
ring is wide enough, as is usual, the buccal
mass slides backwards through the ring;
when the nerve ring is too narrow, the buccal
mass is pulled back above the nerve ring, by
means of sliding of the oesophagus forward
through the ring (observed in the Partulidae).

Proportions in visceral mass: Only snails
have been treated because it is not possible
to take the same characters into account in
snails and slugs. The data set being incom-
plete for all species, 159 species have been
treated. For each species, available data are:
size rank (TA); shape (HD = shell height/
shell diameter); number of whorls of the shell
(WH) and visceral mass measurements. The
latter are lengths in whorls: length of the lung
from pneumostome to top of the lung along
the rectum (LP); length between the top ofthe
lung and the top of the stomach (LS); length
of the upper lobe of the digestive gland be-
tween the top of the stomach and the top of
the spire (SS); and length of the stomach plus
gastric crop (ST), the foremost part of the lat-
ter being in a few cases at the level of the lung
cavity (Appendix B).

For each parameter the sample has been
divided into five classes including equal num-
bers of species, and the table has been writ-
ten in complete disjunctive form (limits of the

14 TILLIER

ST4
LV4
LS4
WH4
LP4
554 553
HD4
т
|
LV3 |
LS3 TAY ST3 |
LP3 |
HD3 |
WH3 TAS H
|
WH2 |
|
EP2 |
ST2 |
LV2 |
|
|
|
|
|
|
|

LV5
WH5
LPS
HD5ST5
LS5
555
TA2 ТАЗ
=== =---------------------------- 2
LS2
552
551
EPA
HDI STI
LS1 WH 1
LVI

TEXT-FIG. 5. Factor map of correspondences among characters of visceral mass. Plane (1,2). Data,

Appendix B, limits of the classes, Appendix C.

classes, Appendix C). Each class is num-
bered from 1 to 5 in increasing value of the
variable considered. The results of the anal-
ysis are represented Text-fig. 5 (modalities,
plane 2,1), Text-fig. 6 (individuals, plane 2,1)
and Text-fig. 7 (individuals, plane 3,1).
Text-figure 5 clearly shows an overall cor-
relation among lengths of the lung, stomach
and upper part of the visceral mass, and
shape of the shell (= shape of the visceral

mass): value-clusters of various characters
follow one another along the first axis, and
axis 2 separates very flat and very elongate
shells (on the right) from average-shaped
shells (on the left). This general correlation is
independent of the systematic position of the
animals (Text-fig. 6). Indeed, species are ar-
ranged in Text-figs. 6 and 7 by their shapes,
not their taxonomical suprageneric affinities.
This means that the proportions of the various

STYLOMMATOPHORAN SYSTEMATICS 15
Gr 37Alb
1 13Zeb
1 44Mac 43Ce2
! ITek
! 58Col
! 13Cho 9Sol 37Nen
1 41Bos33Sub
1 13Imp
1
1 37Ita 24Gas
1 13Ena 44Ber
! 33Rum 8Pag
! 10Lau 51Sag49Pty
4Lep ! BOrc
! 43Cer
|
33Psel
5Cocl
38Spi 29Tro
16Lib 10Pup 50Mes
I
|
24Ven 41Plg !
33Boc lLam 13Cer !
32Cae 17Pa2 !
SET !
!
2Ach 1
5BMoc !
34Ach ITor lAur
4 Ama30Co2 45Sys 49Ede
17Pa3 53501 31Tro 1 38Var
30Co3 17Tra 124Zon 57Hel
11Spe 51Lac 1
57Cep 13Rac 40Cla !
54Amm 31Eve 1
lO gue 0Sty===9/MOn====5= 4 Mla= => 38 ROV= === == Ss? SS SS ===
56Hel 7Ste INSitir 11Aca
17Par р lEla
5BHyg 18Phr 58Cea ! 6Pyr
53Lab 7Bot 3Par !
31Kal 50A11 ! 24Mes
53Amp 11Kle40Pyg 51Pro
53Ple 56Bra !
19Ang 41Dis 40Hed 17Ste !
41Coc 50Trd 5BHec ! 17And
53Plc 50Tri 7Ave 47Nat
58Cen 30Di2 40Car 50Pol 240xy
5BHal 58Hel SEG) 1 45Sy2 52Ple
5B8Art 7Son ! 19Dis 52Cra
52Cor ! 30Dis 47Dip 57Ce2
40Amp | ВНер 45 Тат
53Rha ! 30Cop 52Scu 31Hel 14Suc
58The 153Sin 4DAca 41Sim 58Can
58Sph | 5BElo 24Zot 11Pty 58Heg 14Su2
54Gly ! 4OPed 40Tri 24Vit 58He3
! 46Hap 40Str 47Pri 50Thy
1 40Pad!9Hel 11Val 57Ps2 550re
| 57Epi yODor 24Aeg
| 4OHel 47Rhy
| 9Gy1
1

TEXT-FIG. 6. Factor map of correspondences among individuals for characters of visceral mass. Plane
(1,2). Abbreviations, Appendix A; data, Appendix B; limits of classes, Appendix C.

parts of the visceral mass of stylommatopho-
rans are fairly constant. Indeed, if one
Stretches a flat visceral mass into an elongate
one, the length of every part measured in
whorls would increase. This is exactly what is
shown in Text-fig. 5, where modalities repre-
senting the lengths in whorls and those rep-
resenting shapes correlatively follow one an-
other.

The proportions in the visceral mass that
correspond to this overall correlation have
probably no taxonomical or phylogenetic
value within the Stylommatophora, although

they might represent a synapomorphy of the
Stylommatophora within gastropods (not
checked here). On the other hand, character
states that significantly modify the position of
a taxon with respect to shape should be ex-
amined and discussed. In such cases the
problem is to determine whether these char-
acter states are synapomorphic at the familial
level, or at some lower level not considered
here. One example is the position of Trocho-
morpha (29Tro) in Text-figs. 6 and 7, not as-
sociated with the corresponding shape; since
only one trochomorphid species was exam-

16 TILLIER
у 13Zeb
! 43Ce2
yyMac 37A1b
1 58Col ITek
!
! 13Cho 9Sol
113Imp37Nen 33Sub
1 41Bos
1
37Ita ! 24Gas
13Ena ! 44Ber
149Pty 8Pag 33Rum
! BOrc 51Sag
4Lep ! 10Lau
43Cer !
|
33Pse 1
5Coc !
29Tro ! 50Mes
16Lib 1 10Pup
38Spi |
1
! 41Plg 24Ven
33Boc 135er lLam |
1Str 32Cae 17Pa2
!
I
2Ach
1 58Moc
45Sys 34Ach ITor lAur
30Co2 ЧАта 49Ede
17Pa3 30Co3 31Тго 1 38Var
17Tra 5350124Zon 57Hel
51Lac 1150 |
31Eve 1 40Cla 57Cep
54Amm 13Rac !
-5вНед--------=-==----=------- ое HOSEV=-=3B8Por-=- = IMAN A See 3
156Hell1Str
1Ela ! llAca
31Kal 53Lab 17Par 58Cea
58Hyg 7Bot! 40Pyg 18Phr 3Par
24Mes ! 6Pyr 50A11 41015
Sie rio | 53Amp 56Bra
! 41Coc
17Ste 40Hed ! 53Ple 19Ang
5BHec 17And 50Trd ! 11Kle
57Ave 47Nat | 50Tri
240xy 53 Ple I 31Elg 50Pol58Hal 40Car 58Cen
52Ple 455у2 1 58Hel
52Cra 19Dis 1 30Di2 58Art57Son
57Ce2 1 52Сог 47Dip
45Tam | 58Hep 40 Amp 53Rha
30Dis 52Scu 1 40Aca 58The
24Vit 145ис 31Hel41Sim | 11Pty 30Cop 54Gly 58Elo
14Su2 58Heg 58Can ! 24Zot 53Sin 58Sph
24Aeg47Pri 58He3! 4OPed 4OTri
11Val 50Thy 40Pad 46Hap 9Gyl
550re 57Ps2 | 19Hel 40Str40Dor
! 40Hel
47Rhy 1 57Epi

TEXT-FIG. 7. Factor map of correspondences among individuals for characters of visceral mass. Plane
(1,3). Abbreviations, Appendix A; data, Apoendix B; limits of classes, Appendix C.

ined, no use can be made of the resulting
proportions.

In some other cases, the seemingly aber-
rant position of some taxa is probably related
to the fact that the H/D ratio does not take into
account all shape parameters. For example,
the position of succineid (14Su1, 14Su2) and
oleacinid (38) snails in Text-figs. 6 and 7,
close to shapes much flatter than theirs, may
be related to the rapid increase of the whorls:
their last whorls occupy relatively much more
of the total volume of the animal than in other
snails with the same H/D ratio and whorl num-
ber, and their organs are relatively shorter in

whorl length. To find such taxa aggregating
with those of similar shape, | should probably
have introduced a variable for increase in di-
ameter of the cone formed by the visceral
mass (e.g. the ratio of shell diameters from
one whorl to the next).

Unlike shapes, sizes show very weak cor-
relations, if any, with other modalities taken
into account in the analysis (Text-fig. 5). Very
small sizes (TA1) do not contribute at all to
the first five axes. In other terms, miniaturiza-
tion (or size increase, if the ancestral snails
were small) has no influence on proportions
of the visceral mass. This is true not only

STYLOMMATOPHORAN SYSTEMATICS 17

overall, but often also locally, as may be seen
by comparing the drawings of the visceral
masses of two species of Helicopsis (Figs.
665, 671) of which shell diameter varies from
1.5to 7mm. The contribution of other size mo-
dalities is also extremely weak, and their ab-
solute contribution to the first axis is nil.

It might be concluded that, to a first approx-
imation, the principle of similitude does not
apply to the proportions of the visceral mass
of snails. This is surprising particularly regard-
ing the lung, whose length should increase
with size at constant shape in order to main-
tain the ratio of respiratory surface/total vol-
ume constant. It might be supposed that dim-
inution of this ratio as size increases is
compensated by physiological mechanisms,
but I do not believe so because congeneric
species of various sizes, whose physiological
mechanisms are probably very similar, have
the same lung length (e.g. Helicopsis). It
might be concluded that the respiratory sur-
face is not the functional limiting factor for
lung surface decrease. This point is con-
firmed by the important decrease of the pul-
monary surface, without any other apparent
change, observed in the Helicarionidae (Figs.
296-340) and other semislugs. Such a limit-
ing factor may be either the necessity for re-
tracting the foot into the volume of the lung
cavity, if retraction is the only way to limit
evaporation from the pedal surface; or the ne-
cessity for a long ureteric groove, ifthere is no
closed ureter (v. infra).

The absence of variations in the propor-
tions of the visceral parts of the digestive
tract, intestine excepted (v. infra), is less sur-
prising. The relative diminution in size of the
stomach and gastric crop as size increases
might be counterbalanced by the develop-
ment of an oesophageal crop (v. infra). The
relative diminution of the digestive surface
might be counterbalanced by an increase in
the number of lobules of the digestive gland,
as in the case of the alveoli of vertebrate
lungs (not checked here).

PULMONARY COMPLEX

Following principally the work of Semper
and Simroth (1894), Pilsbry (1900a, b) pro-
posed to distinguish suborders in the Stylom-
matophora by the morphology of their excre-
tory system, and chiefly by the degree of
closure of the ureter. He defined the orthure-
thran kidney, long and without any ureteric

tube, as the diagnostic character of the Orth-
urethra; the heterurethran kidney, transverse
and prolonged by a ureteric tube, as diagnos-
tic of the Heterurethra; and the sigmurethran
kidney, shorter than the orthurethran one and
prolonged by a ureteric tube, as diagnostic of
the Sigmurethra. The formerly-introduced
Aulacopoda (Pilsbry, 1896), whose morphol-
ogy is discussed above, became a division of
the last group. In 1955, Baker defined the
mesurethran kidney as non-orthurethran but
lacking a ureteric tube, and characteristic of
the suborder Mesurethra. These characters
have been discussed by Watson (1920) and
the corresponding suborders have been ac-
cepted by Baker (1955), Solem (1978), Zilch
(1959-1960) and Boss (1982). Wächtler
(1934) insisted on the occurrence of interme-
diate morphologies and recommended the re-
jection of the taxonomic use of these charac-
ters, whereas Schileyko (1978a, b) accepted
only the first of the divisions proposed by Pils-
bry between the Orthurethra and the non-
Orthurethra.

The microscopic structure and ultrastruc-
ture of the kidney and ureter have been stud-
ied by Bouillon (1960), and Delhaye and
Bouillon (1972a, b, c), but in a sample obvi-
ously too small: the ultrastructure was dis-
cussed only in helicids and the microscopical
structure was described in only seven stylom-
matophoran species belonging to seven of
the 60 families recognized by Zilch (1960).
The function of the kidney and lung have
been surveyed by Machin (1975) and Ghiretti
and Ghiretti-Magaldi (1975).

The absence of any retrograde ureteric
tube is probably plesiomorphic, as implicitly or
explicitly expressed by all authors since Pils-
bry (1900a) who have discussed kidney orga-
nization: no prosobranchs, no opisthobranchs
and few archaeopulmonates and basom-
matophorans exhibit such a structure (Fretter
& Graham, 1962; Gosliner, 1981; Hubendick,
1978).

Morphological characters

General description, function: The pulmo-
nary, or pallial, complex forms the parietal
and basal parts of the visceral mass of snails,
from the pallial border to the periaortic intes-
tinal bend (Text-fig. 1). It is composed of the
walls of the lung cavity, kidney and heart. The
lung cavity is generally considered not homol-
ogous with the pallial cavity of other gastro-

18 TILLIER

pods, which forms only the pneumostome re-
gion as noted above (Régondaud, 1964;
Fretter, 1975).

The lung roof is applied to the parietal and
basal walls of the shell. The rectum, which
runs along the suture, forms its right border,
and a branch of the columellar muscle forms
its left border. The lung cavity is distally lim-
ited by the pallial border, and opens to the
outside through the pneumostome, which is
adjacent to the anus. The kidney is proximal,
on the left side of the lung roof. Its shape is
approximately triangular: its visceral side runs
along the periaortic intestinal bend; its rectal
side runs parallel to the rectum; and the peri-
cardium is applied to its pericardial side,
opposite the rectum. The pericardium is pro-
longed toward the pallial border by the pul-
monary venous system, and toward the
visceral cavity, outside the lung cavity, by
the aorta, which crosses the intestinal bend in
the concavity of the latter before dividing into
an anterior and a posterior branch. The ureter
(ureteric zone, ureteric groove or ureteric
tube) runs backwards from the renal pore,
which is usually at the recto-pericardial sum-
mit of the kidney, along the rectal side of the
kidney to the top of lung cavity, and turns for-
ward to the pneumostome along the rectum.
The opening of a ureteric tube, which always
originates from the kidney pore, can be at
any distance from the kidney pore, and on the
dorsal side of the pneumostome when sig-
murethry is achieved. The lung cavity is sep-
arated from the visceral cavity by the lung
floor, or diaphragm, which generally has
transverse muscular fibres. In large species,
the anterior branch of the aorta is partly fused
to the diaphragm, distal to its circuit of the
intestine. The pallial border includes a glan-
dular formation that spreads either between
the rectum and ureter (Zonites: Turchini 8
Broussy, 1935; Spiraxis, Priodiscus); or along
the columellar muscle (Labyrinthus: Tillier,
1980); or along the lung roof (Endodontidae:
Solem, 1976; Oreohelix; various Orthurethra:
Leptachatina, Cochlicopa, Chondrula, Ena,
Imparietula, Klemmia, etc.).

Respiratory mechanisms are wholly simple,
although their physiology is imperfectly un-
derstood; whether inspiration is active or pas-
sive is still a matter of discussion (Ghiretti &
Ghiretti-Magaldi, 1975). The lung cavity fills
with air as the pneumostome opens and the
diaphragm contracts to dilate the lung. After a
while the pneumostome closes and the dia-
phragm relaxes, which produces a slight ex-

cess of pressure inside the lung cavity. Gas-
eous exchange occurs through the walls of
the pulmonary venous system, and the cycle
starts again. The lung cavity may also serve
as a water reservoir (Blinn, 1964), a feature
considered by Solem (1978) as essential for
the success of land snails in colonizing ter-
restrial habitats.

The inner walls of the kidney sac form folds
that resorb the wastes from the liquid coming
from the pericardium vía the reno-pericardial
pore and canalize the residues to the renal
pore and the ureter or ureteric zone. The mor-
phology of these folds, or lamellae, has not
been described previously and 1$ here figured
in the anatomical drawings. Their histology
has been discussed by Bouillon (1960) and
Delhaye and Bouillon (1972 b, c) in a few spe-
cies. The inner wall of the kidney sac of the
Stylommatophora might also have small
caruncles, more frequent in the distal region
of the kidney sac. In small animals, these
caruncles seem equivalent to islets of lamel-
lae (e.g. Figs. 246, 611), whereas caruncles
are outgrowths on the surface of the lamellae
in large species (e.g. Figs. 257, 691). | have
no idea of the histology and function of such
caruncles, and do not know whether the two
types of caruncles are homologous.

The inner wall of the ureteric tube, when
present, may also have folds. From their sam-
ple, Delhaye and Bouillon (1972b, c) con-
cluded that these ureteric folds are present
in the primary ureter of sigmurethran land
snails only (i.e. in the portion of the ureter

between the kidney pore and the top of the

lung). | have shown that such folds occur also
in the secondary ureter of sigmurethran snails
and slugs (i.e. in the portion of the ureter be-
tween the top of the lung and the pneumo-
stome; Tillier, 1983), and will argue further
that the presence of ureteric folds is an effect
of size.

We do not know how the system of cavities
between the inner and outer walls of the kid-
ney and ureter functions. | suppose from their
arrangement that these lacunae allow the liq-
uid resorbed from the kidney and ureter con-
tents to flow back into the pulmonary venous
system: the waste is probably more concen-
trated in the kidney-ureter system as it ap-
proaches the orifice opening into the lung
cavity or pneumostome.

The presence or absence of lamellae or
caruncles frequently allows us to recognize
two or three anatomical regions in the kidney
sac, from its proximal region (close to the

STYLOMMATOPHORAN SYSTEMATICS 19

NA PE

-- ORTHURETHRA
— DOLICHONEPHRA
BRACHYNEPHRA

2 2.5 3
LUNG WHORLS

TEXT-FIG. 8. Length in whorls of lung in snails and semislugs studied (data, Appendix B).

reno-pericardial pore) to its distal extremity
(usually close to the kidney pore). This mor-
phological differentiation might correspond to
a functional differentiation. In at least two
orthurethran species (Achatinella fulgens and
Rachistia braunsi), Delhaye and Bouillon
(1972b) have recognized two histologically
distinct regions that seemingly correspond
largely to the morphological regions here
recognized. In their interpretation, the proxi-
mal region is the kidney proper while the
distal region functions as a ureter and will be
called henceforth the ureteric region of the
kidney.

Statistical analyses: External characters
whose correlations with size and shape mo-
dalities (Appendix B) are studied are the de-
gree of development of a ureteric tube, the
relative length of the kidney and the lung
length. The sample has been split into four or
five classes of equal sizes for measurable
data (limits of the classes, Appendix C).

The length of the lung used in the analyses

is relative to the length of the visceral mass.
LP being the length of the lung in whorls (Ap-
pendix B, Text-fig. 8) and LV the length of the
visceral mass in whorls (Appendix B), the rel-
ative length of the lung is equal to LP/LV. It is
strongly correlated with both LP and LV, and
the result of the analysis is similar when LP is
used.

The relative length of the kidney (LR) has
been calculated as the ratio of kidney length
to lung length, both being measured in mm
from dissections of the pulmonary complex
(Text-fig. 9). In general the points of reference
for measurements are the viscero-rectal apex
of the kidney (S1), the recto-pericardial apex
of the kidney (S2) and the limit between lung
and pneumostome roofs (S3). The relative
length of the kidney is S1S2 / (S1S2 +
S2S3). In heterurethran, transverse kidneys,
S1 is the origin of the aorta because the
lengthening of the visceral and rectal sides in
such a kidney modifies the significance of
measurements made with the viscero-rectal
summit as a point of reference (Succineidae,

20 TILLIER

N 180

OMEAN 0.52

SD 0.025

ALL
ORTHURETHRA
DOLICHONEPHRA
+ BRACHYNEPHRA

0.6 0.7 0.8 0.9 1
KIDNEY LENGTH/ LUNG LENGTH

TEXT-FIG. 9. Relative length of kidney in snails and semislugs studied.

Ferussaciidae, some Oleacinidae). In U-
shaped kidneys, S1S2 is the length of the
pericardial arm of the kidney, measured from
the apex of the U formed by the ureter (end-
odontoids, some discids). The relative length
of the kidney has not been taken into account
for semislugs and slugs whose kidney is lon-
gitudinal, but whose pneumostome is lateral
(i.e. where the axis S1S2 diverges signifi-
cantly from the axis S2S3).

It is not necessary to insist on the impor-
tance of the length of the closed part of the
ureter, which is acharacter of subordinal value
in the Pilsbry-Baker system. Four degrees of
closure of the ureter (Appendix E) have been
distinguished: no closed retrograde ureter
(UR1), which corresponds to orthurethran and
mesurethran morphologies (mesurethran as
originally defined by Baker in 1955; since then,
snails with a partly closed ureter have been
included in the Mesurethra because of general
similarity); closed ureter reaching at most lung
top (UR2); ureteric tube reaching a point be-
tween lung top and pneumostome (UR3); ure-
teric tube reaching the pneumostome (UR4 =
full sigmurethry). Attempts to use classes
of lesser amplitude showed rather weaker
correlations than did such modalities, but
this might be the effect of too small a sample
size.

Numerous internal morphological charac-
ters of the kidney obviously occur in taxonom-
ically restricted groups. Such characters,
mentioned below, are not used in the general
statistical analysis. The characters used are
the number and density of the kidney lamellae
and the density of their anastomoses, the in-
ternal division of the kidney into morphologi-

‘cally distinct regions, and the presence of

caruncles.

Four modalities represent the morphologi-
cal pattern of kidney lamellae (LI): LI1—few
lamellae, not anastomose; LI2— few lamel-
lae, anastomose distally; LI8—numerous
lamellae with few anastomoses; Ll4—numer-
ous lamellae with many anastomoses.

The morphological division of the kidney
into several regions is represented by three
modalities (RR): RR1—kidney homogeneous
in internal morphology, with lamellae reach-
ing the distal region and the level of the
kidney pore; RR2—two distinct regions, the
distal one usually lacking lamellae; RR3—
three distinct regions, the median one either
lacking lamellae or with lamellae different in
appearance from those in the proximal
region.

The presence and absence of caruncles
have been respectively formalized as CA2
and CA.

STYLOMMATOPHORAN SYSTEMATICS

LIU ТАЗ |
LI3 1

RP3 1
UR3 !

UR2 |
L12 |
|

HD3 1

LR2 |

LR3 TA3 RP2 |
|

HD2 CA2 |
|

RR2I

LR4Y |
|

21
1 URI
КРУ HD5
RRI
LRI
HD4
СА!
2
HD1
TA2
URY
(EE,
TAI
RPI

TEXT-FIG. 10. Factor map of correspondences among characters of pulmonary complex. Plane (1,2). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

The data set (156 species) has been ana-
lyzed in completely disjunctive form (each
modality coded 0 or 1). Although not uninter-
pretable, the first analyses were very com-
plex, involving a great number of axes. This
complexity arose from the weight of the
orthurethran snails that, being relatively nu-
merous (31 of 156) and homogeneous in mor-
phology, were discriminated by the first axes.
This is why two analyses were performed:

one taking only the non-orthurethran snails
into account, the orthurethran snails being
treated as supplementary lines; the second
treating only the orthurethran snails. In the
former (Text-figs. 10, 11, 12), the three first-
axes represent only 31.3% of the total vari-
ance. This is why only correlations which are
corroborated along other axes will be dis-
cussed; in some cases, correlations which
appear only along the latter will be discussed,

22 TILLIER

LI4 TAY
LI3

UR2

CA2
HD1

TAI

TEXT-FIG. 11. Factor map of correspondences among characters of pulmonary complex. Plane (1,3). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

whenever their contribution is strong enough
(Text-fig. 12).

Non-orthurethran pallial complexes

In this section lungs, when the kidney is ei-
ther mesurethran or sigmurethran in the sense
of Solem (1978), are discussed: the occur-
rence of partial sigmurethry in some of the
MesurethrasensuSolem(Acavidae,Figs.406—
469), and of mesurethry in some Sigmurethra
sensu Solem (Corillidae, Figs. 470—491);
Oreohelicidae, Fig. 493; Urocoptidae, Fig.
528) makes combined treatment necessary.

Factor maps: In summary, the plane (1,2)
shows the relationships between extremes
(Text-fig. 10), whereas the planes (1,3) (Text-
fig. 11) and (4,5) (Text-fig. 12) show the rela-
tionships between mean modalities. The mo-
dality CA (caruncles) contributes to axis 3
only.

If one joins the modalities of every single
variable in order in the plane (1,2) (Text-fig.
10), two directions appear: one, SE-NW, cor-
responds to size increase (TA1 to TA4); the
other one, SW-NE, corresponds to kidney
length decrease (LR4 to LR1). Axis 1 op-
poses complete mesurethry (UR1) to com

STYLOMMATOPHORAN SYSTEMATICS 23

UR3 14
!
|
!
|
!
|
HDI l
az |
|
ТАЗ RRI |
|
LR2 | RP4
|
LEI
HD4 1
| СА!
TAI Г
CA2
ERIN
Г
RP2 Г
Г
UR4Y
|
L13

|
|
I
|
I
!
I
HD3 |
|
|
|
I
|
Г
1

RR2

LI4

UR2 RPI

TA2
LR3

HD5

URI

TEXT-FIG. 12. Factor map of correspondences among characters of pulmonary complex. Plane (3,4). Data,
Appendix В, Figs. 1-704; limits of classes, Appendix С.

plete sigmurethry (UR4); large size (TA4) to
small size (TA1, TA2); short kidney (LR1) to
long kidney (LR4); homogeneous kidney
(RR1) to kidney with two internal regions
(RR2); few renal lamellae (LI1) to numerous
and anastomose lamellae (LI3, LI4). Axis 2
opposes large size (TA3, TA4) to small size
(TA1, TA2); very short kidney (LR1) to all
longer ones (LR2, LR3, LR4); few renal lamel-
lae (LI1, LI2) to numerous renal lamellae (LI3,
LI4); and mesurethran kidney (UR1) to any
kidney having a portion of the ureter closed
(UR2, UR3, UR4). The most important contri-
butions to axis 3 are those of partly closed

ureters (UR2 opposed to UR3), of flat shapes
(HD1 opposed to HD2), of caruncles (CA1 op-
posed to CA2) and of few renal lamellae dis-
tally anastomose (LI2) (Text-fig. 11). Modali-
ties whose contribution is stronger along axes
4 and 5 than along the first three axes are
LR2, LR3, HD1, HD2, HD3, HD4, TA2, ТАЗ,
LI3, RP2 and RP3 (Text-fig. 12).

Discussion: With regard to degree of clo-
sure of the ureter, in the plane (1,2), LR1,
RR1, RP4 and UR1 are opposed to LR4,
RR2, RP1 and UR4 (Text-fig. 10). In other
words, the absence of a ureteric tube is as-

24 TILLIER

sociated with a short kidney lacking internal
differentiation, in a long lung. It is opposed to
the presence of a ureteric tube closed as far
as the pneumostome, which is asssociated
with a short lung and often with the internal
differentiation of the kidney into two distinct
regions. These associations are totally inde-
pendent of size, which varies in a direction
normal to plane (1,2). Absence of a ureteric
tube is also associated with elongate shapes
(HD5), whereas all shapes are associated
with not only complete, but also partial sig-
murethry (UR4, Text-fig. 10; UR2 and UR3,
axis 3, Text-fig. 11).

The correlation between elongate shape,
short kidney and absence of a ureteric tube
may have a functional cause. In very young
snails, the lung is relatively shorter than in
adults (growth allometry): to a first approxima-
tion, the lung is relatively longer when whorl
number is increased. On the other hand, elon-
gate shells have more whorls than flat shells
(within the general correlation discussed
above, correlation is better between large
numbers of whorls and very elongate shape
than within other groups of modalities in Text-
fig. 5). If it is easier for land snails to develop
a relatively long lung when shape is elongate
(Text-fig. 5), then snails which functionally
need a long lung because of the absence of a
ureter, as discussed below, would frequently
have an elongate shape.

The proposition that a ureteric tube is
present whenever the lung is short appears
as the corollary of Solem's remark (1978) that
there are no slugs, ¡.e. no forms with reduced
lungs, without complete sigmurethry. In all
mesurethran lungs, a narrow zone, different
in aspect from the lung roof, runs from the
kidney pore along the rectal side of the kidney
and along the rectum to the pneumostome;
when sigmurethry is incomplete, such a zone
prolongs the ureteric tube to the pneumos-
tome. According to Van Mol (in Delhaye 8
Bouillon, 1972b), this zone is ciliated in the
Clausiliidae as it probably is in all non-
orthurethran lungs in which sigmurethry is not
achieved (Schileyko, 1978b). It is often delim-
ited along the kidney and rectum by one or
two ridges, forming a ureteric groove. From its
position, this ureteric zone or groove is ho-
mologous with a sigmurethran ureteric tube. If
its function is to resorb water from the wastes,
itis in first approximation half as efficient as a
ureteric tube, length being equal, because its
surface is about halved. Therefore, all other
things being equal, such a groove should be

twice longer than a ureteric tube, and this fact
may explain why snails lacking a ureteric tube
have a long lung.

Morphologies intermediate between full
mesurethry and full sigmurethry are much
more frequent than Wáchtler (1934) thought,
although the degree of variation is generally
limited within each particular family. The fam-
ilies Corillidae and Urocoptidae, in which
snails without a ureteric tube and fully sigmur-
ethran snails occur, appear as two noticeable
exceptions. The Clausiliidae and Cerionidae
have no ureteric tube. In the Endodontidae,
Oreohelicidae (+ Ammonitellidae), and Aca-
vidae, of which some have no ureteric tube,
the closed portion of the ureter does not run
farther than the recto-visceral angle of the kid-
ney. п the Bulimulidae (Odontostominae in-
cluded), Rhytididae, Helminthoglyptidae, Hel-
icidae and Solaropsis, the opening of the
ureteric tube is between the top ofthe lung and
the pneumostome, and often at the level of the
latter. In all other non-orthurethran families,
the ureter is closed as far as the pneumo-
stome.

Delhaye and Bouillon (1972b) have insisted
on the taxonomical importance of the pres-
ence of folds in the ureteric internal wall, ob-
served by them only in the primary ureter of
the Sigmurethra sensu Solem. Such folds
may be found in the secondary ureter (Tillier,
1983). | did not note their presence during all
dissections, and could not take them into ac-
count in factor analyses. However, | observed
internal folds in the ureter of large Stylom-

- matophora only, and it seems clear to me that

the development of ureteric folds is related to
large size: the ureter functioning as a water
recouping system and its efficiency being pro-
portional to its internal surface, any size in-
crease requires a larger increase of the ure-
teric surface (area / volume). As lung length,
and ureter length which depends on it, are not
correlated with size, the necessary increase
in ureter surface is obtained by the formation
of folds. On the other hand, for a given size
range, the frequency of the presence of ure-
teric folds is in proportion to the relative short-
ening of the lung and ureter.

However, ureteric folds are not the only
compensatory structures developed in the
process of shortening of sigmurethran lungs.
In Limax, the primary ureter forms a flat pouch
above the kidney (Simroth 8 Hoffman, 1908—
1928); in the Parmacellidae and Athoraco-
phoridae, the ureter ramifies into diverticulae
interdigitating with the respiratory structures

STYLOMMATOPHORAN SYSTEMATICS 25

(Tillier, 1984a); in Succinea (Succineidae, Fig.
224) and in various Limacidae, the ureter is
prolonged by a caecum beyond the pneumos-
tome along the lung roof border; in Nata and
Rhytida (Rhytididae), the primary ureter de-
scribes one loop between the kidney and peri-
cardium before running along the rectal side of
the kidney (Fig. 382; Watson, 1934).

Such morphologies related to lung shorten-
ing aside, the arrangement of ureteric tubes is
quite constant, except in some Helicellinae
(Helicidae) in which a remarkable morpho-
cline may be observed. In Candidula (Fig.
673), the ureter opens into a groove along the
rectum, close to the top of the lung. The gen-
eral arrangement is similar in Leucochroa, but
here the ureteric tube is prolonged beyond
the ureteric pore toward the pneumostome by
a long caecum that runs between the rectum
and the ureteric groove (Fig. 681); the wall
between the rectum and this caecum is
brownish and spongy (in alcohol). Finally, in
Helicella, ureteric pore and ureteric groove
seem absent: along the rectum the ureter
seems to be reduced to a long, blind divertic-
ulum. If there is really no ureteric pore, the
wastes can be evacuated only through the
anus, after they have been concentrated
while passing through the spongy tissue be-
tween ureter and rectum. This arrangement is
unique among pulmonates, and may be re-
lated to the capacity of Leucochroa and Heli-
cella to live in xeric environments.

Like the development of ureteric folds, the
density of renal folds and of their anastomoses
is clearly related to size (Text-fig. 10) and il-
lustrates the principle of similitude. The high-
est density of anastomoses (114) is also cor-
related with the presence of an incompletely
closed ureter, reaching at most the recto-vis-
ceral angle of the kidney (Text-fig. 11). This
morphological modality has been observed
only in Malagasian, Ceylonese and Australian
acavids, and its cause could be considered
phylogenetic as well as functional (Figs. 416,
427, 466). Renal lamellae are generally more
developed on the kidney roof (shell side) than
on the kidney floor (lung side). However, in
Cecilioides (Ferussaciidae) and in all Ameri-
can Oleacinidae, kidney lamellae have been
observed only on the kidney floor (Figs. 358,
367). The same trend, although less devel-
oped, appears in snails here considered as
Corillidae (Sculptaria, Craterodiscus, Plecto-
pylis; Figs. 472, 480, 483).

It has been sometimes hypothesized that
the absence of a ureteric tube should be com-

pensated by the presence of an ureter within
the kidney (Wächtler, 1934). Although there is
doubtless a balance between the absence of
a ureteric tube and the density of renal lamel-
lae (Text-figs. 10, 11), | identified surely a dif-
ferentiated internal ureter only in Corilla hum-
berti (Fig. 489). In this species, the kidney
pore opens in the middle of the rectal side of
the kidney (Pilsbry, 1905). The kidney sac is
divided into two regions by a transverse ridge.
The distal region is itself divided into two re-
gions by a longitudinal ridge, more developed
than the other kidney lamellae, which isolates
an internal duct running from the distal ex-
tremity of the kidney to the kidney pore. In two
acavid genera lacking an ureteric tube, Clav-
ator and Strophocheilus, renal lamellae are
absent from a zone that runs parallel to the
rectal side of the kidney and reaches the kid-
ney pore (Figs. 446, 463). This zone might
function as an internal ureter. However, in the
Clausiliidae, Macroceramus (Urocoptidae)
and other fully mesurethran snails, kidney
lamellae converge on the kidney pore with a
change in their orientation, but without any
other morphological differentiation that could
be interpreted as an internal ureter (Figs. 507,
9515, 518, 920):

In factor analysis the presence of renal
caruncles can be interpreted only along axis
3 (Text-fig. 11), where it is correlated with flat
shapes (HD1). One possible interpretation,
which is not very satisfying, is that the devel-
opment of caruncles increases kidney sur-
face, compensating for the relatively smaller
size of the kidney when shape is flat; this de-
crease of absolute kidney size is correlated
with decrease in lung length.

The internal differentiation of the kidney
sac into two regions is strongly correlated with
a fully closed ureter (UR4, Text-fig. 10). | can
see no functional interpretation of this corre-
lation, and the cause of this differentiation
might be phylogenetic since a similar differ-
entiation occurs in the Basommatophora and
Orthurethra (Delhaye & Bouillon, 1972a, b; v.
infra).

Lung length and kidney morphology are
closely related, but the modalities of this re-
lation also depend on other parameters in
such a way that several types of kidney mor-
phologies occur with each lung length. These
other parameters are the presence or ab-
sence of a ureteric tube, and the taxonomic
position. The kidney is always approximately
triangular when associated with a long lung,
and variations in absolute length of the lung

26 TILLIER

are correlated with variations in length of the
visceral mass (v. supra). We have seen (Text-
fig. 10) that first, a relatively long lung is usu-
ally associated with a relatively short kidney
and no ureteric tube; and second, a relatively
short lung (RP1), even when disregarding
slugs, is usually associated with a relatively
long kidney (LR4), a fully closed ureter (UR4)
and internal differentiation of the kidney sac
into two regions (RR2). These correlations
are independent of size and shape.

In snails having no ureteric tube, the kidney
pore is never apical, but opens through the
rectal wall of the kidney; in a few extreme
cases it opens midway between lung top and
kidney apex. The absence of any closure of
the ureter is general only in the Ciausiliidae
and Cerionidae. In other families, some mem-
bers of which have no ureteric tube at all
(Acavidae, Corillidae, Urocoptidae), partly
closed or fully sigmurethran ureters also oc-
cur. In true mesurethran snails, the lung has
either a long absolute length (>1 whorl) in the
Clausiliidae, Cerionidae and Urocoptidae, or
a medium absolute length (about 0.7 to 1
whorl) but a long relative length (RP4, Text-
fig. 10) in the Acavidae and Urocoptidae. As
discussed above, the considerable length of
the lung in mesurethran snails is probably a
functional necessity.

In the family Urocoptidae, to which both
mesurethran and sigmurethran snails belong,
the shape of the kidney is nearly constant
(Figs. 528, 532, 533). In the Acavidae, the
ureter is never closed farther than the lung top,
and the morphology of the pulmonary complex
is often remarkable (Figs. 415-466): the kid-
ney top (proximal end of the kidney) may be
distal to the lung top, and the visceral wall of
the kidney may be very oblique, nearly paral-
lel to the suture, whereas it is usually subper-
pendicular to the latter. Correlatively the peri-
aortic intestinal bend is distal to the lung and
the kidney top, instead of proximal, and the
first secondary pulmonary vein on the left is
often much more developed than it is in any
other family. Knowing that the visceral mass
is shorter in most Acavidae than in other fam-
ilies including mesurethran snails (Corillidae
excepted, Appendix B), | interpret this mor-
phology to be related to shortening of the vis-
ceral mass: in snails having no ureteric tube or
only a short portion of the ureter closed, the
lung cannot be shortened below a limit that
depends on the functional importance of the
ureteric groove. Thus, as the length of the
visceral mass decreases, the periaortic intes-

tinal bend extends distally along the left side
of the kidney. The pericardium and pericardial
wall of the kidney are displaced distally be-
cause of the position of the aorta, but the recto-
visercal summit of the kidney is less displaced
distally; as a result the kidney tends to be
elongate parallel to the suture, following an
axis passing from its recto-visceral summit to
the origin of the aorta. If deformation of the
kidney is slight and if the last whorl has a large
enough section, the kidney pore remains in
approximately the same position (e.g. Dorca-
sia, Fig. 431); if deformation of the kidney is
important and if the last whorl is narrow, the
kidney pore appears displaced toward the
recto-visceral summit of the kidney as in He-
licophanta (Figs. 465, 466), sometimes as far
as the level of the origin of the aorta, as in
Strophocheilus oblongus (Hylton Scott, 1939)
or in Stylodon (Fig. 450). In the case of Styl-
odon, the visceral mass might be secondarily
lengthened because it is longer than in other
acavids. The development of the proximal left
secondary vein is probably related to change
in shape of the lung roof, in which diminution
of the proximal surface (between kidney and
rectum) might be compensated by improving
the irrigation of the distal surface. There is
seemingly no strict correlation between this
pattern of deformation and the partial closure
of the ureter, but there is a partly closed ureter
in Pandofella, which has the shortest lung
seen in the acavids.

Like acavids, corillids have a relatively
short visceral mass, but their lung is generally

longer (0.8—1 whorl). Only in Corilla (Fig. 489;

Pilsbry, 1905), which has a shorter lung than
those of other observed corillids, are the kid-
ney and lung top deformed in the same man-
ner as in many acavids, although the first left
secondary vein is not so developed as in
some of the latter. All acavids are large snails,
and Corilla is the largest of mesurethran coril-
lids: adjusting lung surface and ureteric
groove length to short length of the visceral
mass is probably more critical in large snails.
Perhaps the space for retracting the foot re-
quires higher values of these parameters than
is critical in relatively small snails, whereas
the smaller area/volume ratio requires that
lung surface and ureteric groove length be
larger than is needed for foot retraction in
large mesurethran snails.

In non-acavid Stylommatophora whose
ureter is at least partly closed the length of the
closed portion of the ureter increases as lung
length decreases (Text-fig. 10), and there is

STYLOMMATOPHORAN SYSTEMATICS 27

no stylommatorphoran having a lung less
than 0.5 whorl long whose ureter is not closed
down as far as the pneumostome. Relative
length of the kidney varies inversely with lung
and visceral mass lengths (Text-figs. 10, 11,
12) because the absolute size of the kidney is
nearly constant for a given body size, even in
the first steps in limacization. Snails with an
elongate shape associated with a short lung
(Ferussaciidae, some Oleacinidae, Succinei-
dae) are exceptional in having a very short
kidney. In fact, the rectal and visceral sides of
the kidney are much longer than the cardiac
side in these snails, whose kidney appears
transversally elongated (Figs. 224, 358, 361).
This is the condition Pilsbry (1900a) called
heterurethran, which justified for a long time
the classification of the Succineidae in a sep-
arate suborder, although Watson had de-
scribed heterurethry in ferussaciids as early
as 1928. The arrangement of lamellae in
heterurethran kidneys (Fig. 225) clearly
shows that heterurethry is the result of elon-
gation subperpendicular to the rectum on the
visceral and rectal sides of a relatively short
kidney. Because it is found only in relatively
short lungs, heterurethry might be the result
of lung shortening without shortening of the
upper visceral mass in elongate snails having
a short kidney.

The U-shaped kidney of some Charopidae,
Punctidae and Discidae (Figs. 182, 186)
somewhat reminds one of a heterurethran
kidney. However, deformation in this case is
due to the elongation of the rectal and visceral
sides of the kidney along the rectum toward
the pneumostome instead of perpendicular to
the rectum; this is shown by the arrangement
of the internal lamellae (Figs. 182, 183, 186).
There is no simple, clear correlation between
U-shape of the kidney and shortness of the
lung. Charopid snails with a relatively short
lung have a U-shaped kidney (Paryphantop-
sis: Solem, 1970; Mystivagor), but the kidney
of endodontoid semislugs (Ranfurlya, Oto-
concha) is not U-shaped.

Because kidney size can be modified only
to a limited extent and limacization implies
lung reduction, kidney morphology is neces-
sarily modified in limacization. Modification in
kidney morphology is probably necessary for
preventing the kidney from occupying too
much of the respiratory surface of the lung
roof. Modifications in the lung proper were
discussed elsewhere (development of the
venation, formation of alveoli, development of
air sacs: Tillier, 1983). In the first steps of

limacization and in some more-advanced
semislugs, the axis of the kidney, passing
through the reno-pericardiac pore and the kid-
ney pore, is directed toward the pneumo-
stome, as it is in snails (e.g. Vitrinidae, Gym-
narioninae, Figs. 325, 326). In further steps in
limacization, there are three main patterns,
morphologically and taxonomically well de-
fined, in the modification of the kidney mor-
phology related to limacization: compaction
and rotation clockwise (in dorsal view), folding
and rotation counterclockwise (in dorsal view)
and formation of an annular kidney.

Kidney compaction is generally associated
with rotation of the kidney axis clockwise in
dorsal view, such that the kidney axis be-
comes approximately parallel to the longitudi-
nal axis of the foot, while the pneumostome
remains lateral. The kidney is shortened and
thickened, and the density of kidney lamellae
is increased. In the most advanced slugs (e.g.
Atoxon, Fig. 336), the kidney is ovoid. Com-
paction occurs in zonitoid slugs (Daudebardi-
inae, Milacidae, Limacidae, ?Trigonochlamy-
didae), in Vitrinidae (Plutonia), in western
helicarionids (Urocyclinae: Van Goethem,
1977), and in Rhytididae (Schizoglossa). As
far as can be observed, the kidney is also
compact in a similar way in the Athoraco-
phoridae. In limacization of snails with a long
head (and with an elongate shape? Olea-
cinidae: Strebelia, Testacellinae; Rhytididae:
Chlamydephorinae), the pneumostome mi-
grates backwards and the kidney is corre-
spondingly rotated clockwise in dorsal view:
the kidney axis remains approximately di-
rected toward pneumostome (Testacellinae:
Lacaze-Duthiers, 1887; Chlamydephorinae:
Watson, 1915).

Kidney rotation and folding occur in Orien-
tal helicarionids. Rotation is counterclock-
wise, and brings the kidney pore farther back
than the pneumostome. The kidney axis
tends to be perpendicular to the longitudinal
axis of the foot. The kidney is not compact; on
the contrary, and probably in order to avoid
the kidney's covering the lung surface, its
proximal part is first enlarged (Figs. 294, 297),
and forms a flat lobe or lobes along the lung
floor in further steps in limacization (Solem,
1966a). In these kidney lobes the arrange-
ment of the kidney lamellae, converging to-
ward the kidney pore, is not modified.

An annular kidney, surrounding the pericar-
dium, occurs in the Arionidae (Philomycinae
included). The structure of this extraordinary
configuration is more easily understood and

28 TILLIER

described in the North American semislug
Hemphillia (Figs. 195-199) than in other ari-
onids whose kidney is more compact in rela-
tion to more advanced limacization. In Hemp-
hillia the renal ring does not entirely sur-
round the pericardium, but is interrupted at
the level of the origin of the aorta. Careful
examination of the kidney in other arionids
shows that the seemingly complete ring
formed by the kidney around the pericardium
results in fact from contact between the ex-
tremities of the horseshoe formed by the kid-
ney sac: the annular kidney sac is interrupted
by a wall below the origin of the aorta. Kidney
lamellae converge toward the kidney pore,
which is near the right side of the pericardium:
one might suppose that the annular kidney is
simply the result of flattening and expansion
above and beneath the pericardium. The
problem comes from the contiguity of the
reno-pericardial and kidney pores (Fig. 198).
This arrangement seems functionally aber-
rant, because the products of cardiac ultrafil-
tration should flow along the kidney sac sur-
face before reaching the kidney pore; the
usual arrangement, with reno-pericardial and
kidney pores opposed at the two kidney ex-
tremities, seems much more efficient for this
purpose than contiguity (however, in vaginulid
archaeopulmonates, the reno-pericardial and
kidney pores are also contiguous: Tillier,
1984b). This arrangement makes reconstruc-
tion of intermediate steps difficult, although
not impossible.

The ureter of arionids starts at a point on
the right side of the pericardium, at the level of
the auricle (Figs. 197, 211). It first runs to the
right towards the pneumostome; this section
is internally smooth or ornamented with trans-
verse folds of the inner wall (Figs. 197, 212).
Then it turns back and to the left (Hemphillia,
Figs. 197, 199) or backwards (slugs, Fig. 212)
before turning again toward the pneumo-
stome (Figs. 196, 199, 211). These two last
sections are internally ornamented either with
longitudinal folds and caruncles (Hemphillia,
Fig. 199) or with longitudinal folds only (Phi-
lomycus, Figs. 211, 212). This internal mor-
phology of the two distal sections of the ureter
is unique among the Stylommatophora, and
strikingly similar to the internal morphology of
the kidney in endodontids, charopids and dis-
cids, whose internal renal morphology is
characterized by the great development of
two longitudinal folds and, frequently, carun-
cles (Figs. 169, 183, 186, 188). One possible
hypothesis explaining this set of unique mor-

phologies in kidney and ureter is that the di-
stal part of the arionid ureter is homologous to
a snail kidney, and that the arionid kidney sac
results from secondary development of the
reno-pericardial pore. The idea of the hyper-
trophy of the region of the kidney pore in slugs
related to discids is not totally eccentric, for
one abnormal specimen of the discid An-
guispira alternata dissected had a hypertro-
phied kidney pore, internally ornamented with
folds. This hypothesis would explain why the
arionid ureter is similar in morphology to
some snails' kidneys, and why the reno-
pericardial and kidney pores are contiguous
in the Arionidae. Unfortunately, comparative
data on the ontogenetic development of the
kidney that would allow the testing of this hy-
pothesis are lacking.

Orthurethran pallial complexes

Contrary to what has been written, even re-
cently (Boss, 1982), the kidney pore is not
apical in orthurethran kidneys and does not
open towards the pneumostome. lt opens
from the right side of the distal region of the
kidney, and the frequent occurrence of a ridge
delimitating a groove along the kidney toward
lung top (e.g. Fig. 5) shows that wastes are
evacuated from the kidney in this direction, as
in all Stylommatophora. The mistake arises
from the original description of the orthure-
thran morphology by Pilsbry himself (1900a),
and may reflect the very poor quality of Pils-
bry’s microscope, emphasized by Baker

- (1958).

From the preceding paragraphs it is clear
that orthurethran kidneys are not defined by
the absence of a closed ureter as proposed
by Pilsbry in 1900. Furthermore, Watson de-
scribed a ureteric tube in Acanthinula as early
as 1920; a similar structure was described by
Solem in 1964 in Amimopina and it also oc-
curs at least in Rachistia (Fig. 137). This tube
does not extend farther than the recto-
visceral summit of the kidney, but seemingly
does not differ in any character from incom-
pletely closed ureters found in some other
Stylommatophora. Orthurethran pulmonary
complexes are defined by great relative
length of the kidney (80% to 99% of lung
length, Text-fig. 9); by great absolute length of
the lung (more than 0.7 whorls, Text-fig. 8,
Appendix B); and by internal morphological
differentiation of the kidney sac into two, often
three distinct regions (e.g. Figs. 5, 6). The
association of these three character states

STYLOMMATOPHORAN SYSTEMATICS 29

defines a group homogeneous for other mor-
phological characters, ¡.e. the Orthurethra.

The results of partial factor analyses treat-
ing 31 orthurethran species are obscure. The
only clear correlation is that of very small
sizes to flat shapes, which may indicate an
origin of most small orthurethrans by progen-
esis, for the very first whorls of most larger,
elongate orthurethran snails are flatter than
the adult shell. This confusion, contrasting
with the relative clarity of the results for non-
orthurethrans using the same modalities,
shows how homogeneous the Orthurethra
are in morphology. To obtain significant re-
sults, the sample should be larger to allow the
use of modalities representing classes of
lower amplitude. However, the results ob-
tained here, as well as examination of the fig-
ures, admit of a few interpretations.

As in other snails, the increase in density of
kidney lamellae and of their anastomoses is
related to large size, and probably represents
the same solution to the area/volume ratio
problem. Kidney lamellae are few, not anas-
tomose, and present only in the proximal re-
gion of the kidney (at the level of the pericar-
dium) in small species, whereas they are
numerous and reach the distal part of the kid-
ney, where they anastomose, in large species
(Figs. 5-142).

In species with either a ridge along the kid-
ney and rectum (e.g. Cerastua, Fig. 134) ora
ureteric tube along the kidney and a groove
along the rectum (e.g. Rachistia, Fig. 137;
Acanthinula, Fig. 87), the lung is shorter than
in species without such structures which be-
long to the same groups: the lung is shorter in
Acanthinula than in other valloniids, and in the
Cerastuinae than in the Eninae and in the
Chondrulinae among Enidae. It may be sup-
posed that, as in other Stylommatophora, the
formation of a ureter external to the kidney is
related to lung shortening. The Cerastuinae
show not only several steps in ureter forma-
tion, but also a strong and prominent pulmo-
nary venation (Mordan, 1984) and particularly
dense and anastomosed kidney lamellae. All
these characters are related to lung shorten-
ing in other Stylommatophora, and there is no
reason to suppose that it is not the case for
the Cerastuinae. Therefore, they cannot be
used individually to distinguish the Cerastu-
inae from other Enidae, as suggested by Mor-
dan (1984): visceral mass shortening, includ-
ing lung shortening, development of lung
venation and formation of a ureteric groove
and ureteric tube constitute a single synapo-

morphy of the Cerastuinae because the co-
occurrence of these character states is a
functional necessity.

The function of the distal region of the or-
thurethran kidney is ureteric (Delhaye &
Bouillon, 1972b, c). It explains the great rela-
tive length of the kidney, its macroscopic in-
ternal differentiation and possibly the ab-
sence of limacization (reflecting the kidney
sac would eliminate the ureteric zone along
the kidney, and possibly the zone for storing
wastes if such structures occur). A ureteric
groove reaching the pneumostome occurred
in only a few of the observed Orthurethra
(Orcula, Solatopupa, Chondrula, Imparietula,
Cerastua, Rachistia, Amimopina), whereas
the kidney opening is always directed toward
the top of the lung and a ureteric zone usually
occurs along the kidriey. Generally the ure-
teric zone seems to reach at most the recto-
visceral summit of the kidney. The apparent
absence of a ureteric zone along the rectum
of most Orthurethra might result from poor ob-
servation, for most of them are small; but it is
likely that, at least in large Orthurethra (e.g.
Eninae), there is really no ureteric zone join-
ing the top of the lung to the pneumostome. In
the model proposed by Solem (1978), wastes
are flushed from the lung cavity with pallial
water; if this really occurs, presumably wastes
are stored in the proximal region of the lung
between flushes. This model implies a lack of
water conservation by the snails, some of
which live in dry environments. On the other
hand, the presence of a zone of brownish tis-
sue along the rectum of some enids, similar to
that in Helicella, suggests that in some cases
the wastes might be evacuated through the
rectum and anus. Such a mechanism would
prevent limacization, because in its course
the rectum becomes separated from the right
border of the lung roof.

Plesiomorphy and apomorphy in pulmonary
complex

Plesiomorphy and apomorphy in the pul-
monary complex of the Stylommatophora can
be recognized by outgroup comparison with
the pallial complex of the Basommatophora,
Archaeopulmonates (Hubendick, 1978; Tillier,
1984b), plesiomorphic opisthobranchs (Gos-
liner, 1981) and prosobranchs (Fretter & Gra-
ham, 1962).

The occurrence of a lung ontogenetically
distinct from the pallial cavity is a synapomor-
phy of pulmonates. The pulmonary complex

30 TILLIER

of the Stylommatophora differs from that of
most other pulmonates in two major charac-
ters: great absolute and relative length of the
lung; and position of the kidney, which shares
a wall with the visceral cavity, whereas in all
other pulmonates, Otinidae (Archaeopulmo-
nata) excepted, the lung roof surrounds the
kidney. Because the lung is a neoformation, a
lung extending around the kidney is apomor-
phic and the position of the kidney in the
Otinidae and Stylommatophora is plesiomor-
phic. In the Archaeopulmonata and Basom-
matophora, the lung is never longer than
about 0.5 whorl. In the Stylommatophora,
such a short length is possible only if associ-
ated with a ureteric tube, which is apomor-
phic. Hence the long lung is a synapomorphy
of the Stylommatophora, but is plesiomorphic
within this group.

In all Archaeopulmonata and Basommato-
phora, the kidney is about 70% of lung length,
¡.e. shorter than in the Orthurethra. As men-
tioned above, the absence of a ureteric tube
is plesiomorphic. The internal differentiation
of the kidney sac into renal and ureteric re-
gions is probably plesiomorphic, because this
configuration occurs in the few archaeopul-
monates and basommatophorans studied by
Delhaye and Bouillon (1972a, c).

It may be concluded that the primitive ar-
rangement of the stylommatophoran pulmo-
nary complex should include: no ureteric
tube; lung length more than 0.5 whorl, but as
close to this length as possible; length of the
kidney about 70% of lung length; probably a
ureteric groove, because this structure is
present in some orthurethrans and in all non-
orthurethrans lacking a ureteric tube; kidney
sac internally differentiated into two regions,
supposedly renal and ureteric. No stylom-
matophoran snail examined has the whole set
of character states; this implies that the mor-
phologies observed do not represent a single
morphocline (a primitive-like condition is ap-
proached in Pagodulina (Orculidae, Figs. 56,
57), but its lung length is too great to be ple-
siomorphic). However, it is essential to dis-
cuss morphoclines in which a short, otinid-like
pulmonary complex is transformed into the
various types of pulmonary complexes found
in the Stylommatophora. The transformations
proposed here entail two criteria, parsimony
and homology in the distal region of the kid-
ney, when differentiated. Both criteria may be
contested: presence of an internal ureteric re-
gion, because there is no histo-physiological
data showing that the differentiated distal re-

gion of non-orthurethran kidneys functions as
a ureter; parsimony because we do not know
whether evolution is parsimonious. Given
these criteria, several patterns appear func-
tionally acceptable. The first is the lengthen-
ing of the ureteric region of the kidney. This
solution probably improves efficiency, but im-
plies that the kidney occupies a larger portion
of the lung roof and therefore reduces the res-
piratory area. This diminution may be coun-
terbalanced either by lung lengthening or by a
change in shape that increases the palatal
and basal areas of the last whorl. Another so-
lution, a change in vascularization, seems to
occur only in relation to lung shortening.
These are exactly the arrangements ob-
served in the Orthurethra. The ureteric groove
may have been lost in most of these snails, as
discussed above. The second is dedifferenti-
ation of the ureteric region of the kidney with-
out closure of the ureteric zone. Such a mod-
ification allows kidney shortening and lung
surface increase, but must be counterbal-
anced by lengthening of the ureteric zone of
the lung roof, which itself is not possible with-
out lung lengthening. Such a pattern occurs in
the Clausiliidae, Cerionidae, Urocoptidae and
Corillidae. If size is large and a ureteric tube is
absent, a secondary internal ureter may be
differentiated (Corilla, some Acavidae: Clava-
tor, Strophocheilus). In such pulmonary com-
plexes a ureteric tube may be secondarily
formed, as in the Urocoptidae and Corillidae.
A third pattern is the modification of the ar-
rangement of the pulmonary complex. № the

visceral mass becomes shorter before a suf-

ficiently long ureteric tube forms, the pulmo-
nary complex changes shape to preserve the
length of the ureteric groove, as described
above for the Acavidae.

The fourth is the formation of a ureteric tube
before internal dedifferentiation of the kidney
and kidney shortening: this solution allows
change in shape and size without important
change in the pulmonary complex arrange-
ment, and is apparently the most successful
in terms of radiation. A ureteric tube is asso-
ciated with internal differentiation of a long
kidney among all Aulacopoda sensu Solem,
Achatinidae, Subulinidae, Systrophiidae,
Haplotrematidae, Helminthoglyptidae, Helici-
dae, Camaenidae, Bradybaenidae, Sag-
didae, Polygyridae and Ammonitellinae (Oreo-
helicidae). Formation of a ureteric tube might
obviate lung lengthening: among all dissected
genera belonging to these families, in only
Rumina, Sagda and Solaropsis is the lung

STYLOMMATOPHORAN SYSTEMATICS 31

longer than one whorl (Text-fig. 8). If it is long
enough or includes structures compensatory
of low length (internal folds, caeca), a ureteric
tube allows internal dedifferentiation of the
kidney, which occurs in at least some genera
of most of these families.

The fifth pattern is kidney shortening. This
implies the formation of a ureteric tube; but
dedifferentiation of the kidney, occuring after
formation of a ureteric tube in a long kidney,
might allow relative shortening of the kidney if
the tube is long enough, and the resulting
morphology is indistinguishable from that re-
sulting from formation of a ureteric tube dur-
ing the course of kidney shortening if no
member of the family under consideration has
either a kidney without any ureter, or a long
kidney with a ureteric tube. This pattern oc-
curs in the Rhytididae, Bulimulidae (including
Odontostominae), Streptaxidae, Ferussaci-
idae, Oleacinidae and Succineidae. When
lung shortening occurs, it is related to visceral
mass shortening in the first three of these
families, whereas it is associated with a pe-
culiar shape in the three last ones, as dis-
cussed above.

These facts indicate convergence or parallel
evolution (homoiology). It seems unlikely that
closure of a ureteric tube, kidney shortening,
and kidney internal dedifferentiation under-
went reversal in polarity; but it has been shown
above that high relative kidney length may re-
sult from shortening of a lung associated with
a short kidney because the kidney tends to
keep the same relative size, and therefore
converge on the plesiomorphic condition. It is
therefore extremely important to use the most
plesiomorphic condition found in each family
considered in discussing phylogenetic rela-
tionships of families. If the representatives of
this condition are extinct or not studied, there
is little hope of proposing relationships approx-
imating real cladogenetic events.

In any case taxonomic use of the degree
of closure of the ureteric tube may be mean-
ingful only if related to other characters the
functional value of which can be evaluated:
on this point the criticisms aimed by Wáchtler
(1934) at the Pilsbry-Baker classification are
just, although his scheme of morphoclines in
the stylommatophoran pulmonary complex is
much too simple. Criticisms made by Schil-
eyko (1978a, b) are also proper only when
taking this character alone into consideration:
combination of this character with others, the
evolutionary polarity of which can be deter-
mined, makes the construction of phyloge-

netic trees for the Stylommatophora both pos-
sible and justifiable.

DIGESTIVE TRACT

The arrangement of the digestive tract in the
Stylommatophora has been described above.
| have emphasized and discussed elsewhere
how the nearly complete absence of prior data
leads to erroneous generalizations and con-
clusions (Tillier, 1984a); data and references
on digestive tract anatomy in the Basommato-
phora and Archaeopulmonata may also be
found elsewhere (Tillier, 1984b). The function-
ing of the digestive tract of pulmonates has
been surveyed by Runham (1975).

In the factor analysis used in this chapter,
the following variables and modalities have
been used (Appendices B, C; Text-figs. 13,
14, 15):

1. Size: five classes with equal sizes (TA1
to ТА5);

2. Shape: five classes with equal sizes
(HD1 to HD5);

3. Buccal mass shape: spheroidal (BM1) or
cylindrical (BM2); the latter shape corre-
sponds to carnivory;

4. Oesophageal crop: absent (OC1); sepa-
rated from gastric crop by a distinct portion of
the oesophagus (OC2); separated from gas-
tric crop by a simple constriction (OC3); as in
OC3 but extending forward to the nerve ring
(OCA);

5. Gastric crop: funnelform, widening from
oesophagus to stomach (SC1); cylindrical
(SC2); median portion inflated (SC3); funnel-
form, decreasing in diameter from oesopha-
gus to stomach (SC4);

6. Gastric pouch: differentiated, separated
from gastric crop by a constriction (PS1); dif-
ferentiated, prolonging gastric crop without
any constriction (PS2); reduced (PS3);

7. Intestine length: intestinal loops long,
reaching proximally at least the level of the
distal limit of gastric pouch (IL1); intestinal
loops reaching a level between the distal limit
of gastric pouch and the middle of gastric crop
(IL2); intestine shorter, but intestinal loops
distinct (IL3); intestinal loops reduced to an
almost flat sigmoid (IL4).

In the factor maps (Text-figs. 13, 14, 15),
large sizes (TA4, TA5) are opposed to small
sizes (TA1, TA2) along the first axis: size is
the principal factor of variation in the morpho-
logical characters of the digestive tract con-
sidered. Shapes load much less than do sizes

32 TILLIER

PS3 5С4 |
|
|
|
|
|
|

BM2 |
TA2 |

IL3 |

OC1

HD1 |

Horse Fi CD a ee DM

HD2

SC3 |
0C2

RS] |
|

ET |

TAS |

TAI
TA3

PS2

------------- HD3--------------------2
502
HDYIL2
TAY
oc4

TEXT-FIG. 13. Factor map of correspondences among characters of digestive tract. Plane (1,2). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

on the first axes: HD1 (flat shapes) contrib-
utes very weakly to axis 1; HD3, HD4 and
HD5 contribute weakly to axis 2 but more
strongly to axis 3; HD1, НО? and HD5 con-
tribute to axis 4 and HD3 contributes to axis 5.
Overall correlations with size are principally in
the planes (1,2) and (1,3) (Text-figs. 13, 14),
and correlations with shape are in the plane
(3,4) (Text-fig. 15).

Anterior digestive tract

The anterior digestive tract includes the
buccal mass, salivary glands and oesopha-
gus, part of which may be inflated to form an
oesophageal crop.

Buccal mass, salivary glands: The buccal
mass comprises a complex arrangement of
muscles for protracting and retracting the rad-
ula and jaw. When the animal is not feeding,
the radula forms a dorsally open tube cover-
ing the internal surface of the buccal cavity,
and is posteriorly prolonged into the radular
sheath, where teeth are formed. The jaw is
above and forward of the buccal cavity. When
the animal feeds, the radula is evaginated,
and then invaginated as the jaw cuts the food,
to bring the food back into the buccal cavity
and oesophagus (Runham, 1975).

In carnivorous snails, radular teeth are
dagger-shaped and the jaw might be reduced

STYLOMMATOPHORAN SYSTEMATICS 33

I
!
!
|
|
|
|
|
[WAZ
|
|
I
1
|
|
l
|
|

ТА!
IL3
ос!
HD1
TA3
PS2
HD5----SC1------------- BM|------------
SC2
HD2
HD4 IL2
503
002
PS1

PS3 sC4
BM2
----------------------------- HD3-----3
TAY
IL1
oc4
003

TEXT-FIG. 14. Factor map of correspondences among characters of digestive tract. Plane (1,3). Data,
Appendix В, Figs. 1-704; limits of classes, Appendix С.

or absent. The buccal mass tends to be cy-
lindrical and its outer surface has only trans-
verse muscular fibers, instead of a complex
arrangement of oblique muscular fibers.
These modifications allow substitution for
abrasion and cutting of vegetal matter by pre-
hension, with subsequent shredding and
swallowing of animal prey. Correlatively, the
head is often longer than in non-carnivorous
snails, the distal part of the genital apparatus
is reduced in size and the genital orifice might
be displaced backwards (in the Rhytididae),
in order to increase space for housing the
larger buccal mass (Watson, 1915, 1934). In

some carnivorous snails, the buccal mass is
so big that it cannot be contained in the pedal
cavity and extends into the visceral cavity, un-
der the diaphragm (Oleacinidae, Rhytididae,
Streptaxidae; Figs. 371, 380, 397). In Text-
figs. 13-15, modality BM2 indicates the posi-
tion of the center of gravity of carnivorous
species: carnivory is clearly correlated with
well-defined morphological modalities of
other parts of the digestive tract, whereas the
contribution of BM1 is nil, indicating the asso-
ciation of this modality with all other character
states of the digestive tract.

The position of the salivary glands along

34 TILLIER

oc4

ly
|
|
HD3 |
|
|
PS3 |
|
|
|
003 BM2 |
TAY |
|
|
|
|
|
!
|
]
TA2
IBMI
1502
|
oc! |
|
PSI |
TAI |
IL2
503 |
|
|
HDS HD4 |
|

TAS

HD2

sc!

TEXT-FIG. 15. Factor map of correspondences among characters of digestive tract. Plane (3,4). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

the oesophagus is variable among taxa, but
| do not know the factors that determine
this position. When an oesophageal crop is
present, the salivary glands are appressed
to each side of its anterior portion. Generally
they are compact and made of numerous
agglomerated small lobes. Their apparent
size is relatively larger in some carnivorous
Stylommatophora, on one hand (Nata, Schiz-
oglossa, Ptychotrema; Figs. 381, 393, 394);
and in large non-aulacopod species, on the
other hand (e.g. Figs. 430, 630). In carnivo-
rous stylommatophorans, the salivary glands
are very thick and they seem to show a

real increase in relative size. In large
non-aulacopod species, the apparent in-
crease in size of the salivary glands is rather
a change in shape: the salivary glands are
relatively thin, formed by relatively large flat
lobes lying loose in a matrix of conjunctive
tissue. The apparent fusion of the salivary
glands around the oesophagus, formerly
used in taxonomy (e.g. by Fischer, 1880—
1887), is not, in my opinion, an important
feature. If such fusion occurs, only the con-
junctive matrix is involved and the duct of
each lobe runs either to the left or right
principal salivary duct.

STYLOMMATOPHORAN SYSTEMATICS 35

Oesophagus, oesophageal crop: The inter-
nal surface of the oesophagus usually has a
system of longitudinal ribs and grooves, more
apparent when the animal is large and when
the oesophageal wall is thick. The length of
the oesophagus is variable, but this variation
is probably related to variations in size of
functionally more important organs rather
than to a definite change in aptitude of the
oesophagus proper.

Unlike oesophageal length, the develop-
ment of an oesophageal crop, the internal or-
namentation of which usually differs from that
of the oesophagus sensu stricto, is probably
functionally important. The oesophageal crop
has been considered a general feature of the
Stylommatophora (e.g. Franc, 1968). This is
not true: about half of the species here exam-
ined do not have an oesophageal crop, and
all intermediate stages between complete ab-
sence and occurrence of an oesophageal
crop extending from the gastric crop to the
nerve ring occur in the Stylommatophora.

Factor analysis shows a strong correlation
between size and degree of development of
an oesophageal crop (Text-figs. 13, 14). In
general, small snails (TA1, TA2) lack an oe-
sophageal crop (OC1); a small oesophageal
crop (OC2) may be present in medium-sized
snails (TA3) and is very frequent in large
snails (TA4); very large snails (TA5) nearly
always have a well-developed oesophageal
crop (OC3, OC4). Among the latter, extension
of the oesophageal crop forward to the nerve
ring (OC4) is related to medium shapes (HD3,
Text-fig. 15), in which the relative volume of
the last whorl is more important than in other
shapes. As discussed above, the proportions
of the visceral mass are not related to size.
The relationship between the degree of de-
velopment of an oesophageal crop and size
increase can be interpreted in terms of the
principle of similitude, taking time into account
(Lambert & Teissier, 1927): the development
of a storage volume allows the increase in
the time for digestion without increasing no-
ticeably the time for feeding; correlatively the
proportions in the visceral mass need not be
altered.

Gastric region, intestine

The gastric region is composed of three
parts: the gastric crop, which prolongs the oe-
sophagus, from which it differs by its larger
diameter and generally distinct internal orna-
mentation; the gastric pouch, which consti-

tutes the proximal part of the digestive tract,
and which is often separated by a constriction
from the gastric crop; and the proximal intes-
tine, which runs from the level of the orifices
of the ducts of the digestive gland to the be-
ginning of the periaortic intestinal loop. There
is often a constriction between the proximal
intestine and the distal intestine, by which the
proximal intestinal loops and the distal rectum
may be distinguished.

Gastric region: Commonly the differentia-
tion in internal morphology of the gastric com-
plex is relatively more important in large ani-
mals, and may allow the distinction between
the various (functional?) regions.

All observed internal morphologies can be
discussed in terms of a general plan of orga-
nization, in which two ventral ridges, delimit-
ing a groove, would run from the upper end of
the oesophagus into the openings of the
ducts of the digestive gland, and from these
Openings into the proximal intestine, were the
animals not torted. Because of the torsion and
rotation of the visceral mass, these ridges,
when present, run along the right side (under-
side) of the gastric crop, and along the left
side (upper side) of the proximal intestine.
The lower (left) ridge enters the anterior (left)
duct of the digestive gland, from which a gen-
erally short typhlosole extends. The upper
(right) ridge runs above the opening of the
anterior duct of the digestive gland and enters
the posterior (right) duct, from which a gener-
ally long typhlosole extends. The latter usu-
ally is not distinguishable farther than the con-
striction separating the proximal intestine
from the beginning of the periaortic intestinal
loop, when present. The groove between
these two ridges is ciliated at least in the gas-
tric crop (Rigby, 1963, 1965) and probably
functions as a conveyor belt that carries food
particles into and out of the digestive gland.
The most frequent modifications of this gen-
eral plan of organization are: absence of the
portion of the right ridge between the two
openings of the digestive gland; absence of
the two crop ridges; and reduction or absence
of the left typhlosole. Exceptionally the left
typhlosole, issuing from the anterior duct of
the digestive gland, is more developed than
the right one, as in Solaropsis and Laby-
rinthus (Camaenidae?) (Figs. 556, 559).

Although unusual, the occurence of internal
ornamentation on the dorsal side of the gas-
tric crop is not exceptional. It is composed of
longitudinal ridges, prolonging those in the

36 TILLIER

oesophagus. In such cases, two more devel-
oped ridges delimiting the ventral groove are
usually present, but this is not always the
case and sometimes only the larger diameter
of the gastric crop marks the end of the oe-
sophagus.

When present, the internal ornamentation
of the gastric pouch consists in ridges that
prolong those in the gastric crop and disap-
pear at the level of the entry into the proximal
intestine. These ridges are generally more or
less wrinkled and look different from those in
the crop, especially when the latter is sep-
arated from the gastric pouch by a constric-
tion. In a few acavids (Stylodon, Acavus, Figs.
447, 467) and in bulimulids s./. with a
well-differentiated gastric pouch (Plagiodon-
tes, Discoleus, Figs. 537, 545), the gastric
pouch is not regularly spheroid, but divided
by a constriction in a plane tangent to both
the gastric crop and the proximal intestine.
Internally this constriction is marked by either
a transverse ridge (Stylodon), or a change in
the ornamentation of the stomach wall. A
somewhat similar arrangement occurs in Eua
(Partulidae), in which the part of the gastric
pouch that prolongs the proximal intestine
protrudes and is itself divided transversally by
a ridge (Figs. 25, 26). In Partula, a similar
protrusion of the lower half of the gastric
pouch is less developed than in Eua and
lacks any internal transverse ridge (Fig. 29). |
do not know the function of such internal
differentiations of the gastric pouch, which
might be understood by observing the move-
ments of the particles in it.

The typhlosoles are the only internal orna-
mentation of the proximal intestine, except in
American oleacinids and in a few acavids in
which the proximal intestine has longitudinal
ridges (Figs. 365, 371, 372, 412, 430, 433,
440). Similar internal ridges occur at the level
of the constriction between the proximal in-
testine and the periaortic intestinal loop, when
present (Ena, Fig. 126; Oxychilus, Fig. 240).

The most common of the four shapes of the
gastric crop here distinguished is the nearly
cylindrical one (diameter nearly constant be-
tween the increase in diameter indicating the
upper end of the oesophagus and the gastric
pouch; SC2, Text-figs. 13-15; e.g. Fig. 280).
It is not clearly associated with any body size
or shape, and its corresponding modality,
SC2, is located near the origin along the axes
1, 3, 4 and 5 (Text-figs. 14, 15). A funnelform
gastric crop, regularly widening from oesoph-
agus to gastric pouch, has been seen only

in the family Discidae, including Helicodiscus
(Figs. 184, 187, 190; SC1, Text-figs. 13-15).

The presence of an oesophageal crop in-
flated in its median portion (SC3) is correlated
with strong differentiation of the gastric pouch
(PS1), and with elongated shapes (Text-figs.
14, 15). The former of these correlations is
obvious, for the posterior part of the gastric
crop usually must become narrower when
separated from the gastric pouch by a con-
striction. The correlation with elongate shapes
of the visceral mass is more interesting, and
probably indicates the usual pattern of change
in the morphology of the gastric region as body
shape changes from flat to elongate; this can
be checked by examining variation in gastric
region morphology in relation to shape varia-
tion in a group homogeneous in other ana-
tomical characters, such as the Orthurethra
(Figs. 1-146).

However, a few snails have a flat shape
associated with an inflated gastric crop, as
does Zonitoides arboreus (Gastrodontinae,
Fig. 277) or Vallonia albula (Valloniidae, Fig.
84). The morphology of the gastric region of
these snails forcibly reminds one of that in
newly hatched snails (Ghose, 1963, and per-
sonal observations). Because flat snails hav-
ing such a gastric morphology are very small,
| interpret the apparent discordance between
their gastric morphology and their shape as
paedomorphosis resulting from progenesis.

The occurrence of a gastric crop of which
the anterior portion is inflated (SC4) is corre-
lated with the occurrence of an undifferenti-

` ated gastric pouch (PS3), small sizes (TA1,

TA2), a short intestine (IL3, IL4) and the ab-
sence of an oesophageal crop (OC1). The
correlation of this gastric crop shape with the
absence of a differentiated gastric pouch is
obvious: if, in any other gastric morphology,
the diameter of the digestive tract decreases
at the level of the gastric pouch, the anterior
part of the gastric crop appears inflated (v.
infra). Correlation with the absence of an oe-
sophageal crop here might indicate identity of
function of the two types of crops. From this
interpretation it follows that, in the absence of
dedifferentiation of the gastric pouch, a stor-
age volume for food is acquired through т-
crease in the volume of the anterior part of the
gastric crop in small snails, and through dif-
ferentiation of an oesophageal crop in big
snails. The presence of such an arrangement
is nearly the rule in the Aulacopoda group A
sensu Solem (1978), and in endodontoid
snails. However, the occurrence of this mor-

STYLOMMATOPHORAN SYSTEMATICS 37

phology in other taxonomic groups suggests
that its interpretation should be more func-
tional than phylogenetic (Orthurethra: Auri-
culella, Fig. 1; carnivorous families; Corill-
idae: Sculptaria, Fig. 470).

In American oleacinids (Figs. 364—374)
and in Priodiscus (Rhytididae?, Fig. 375), the
anterior part of the gastric crop extends to
form a rostrum above the upper end of the
oesophagus. Functionally, the formation of
such a rostrum seems to correspond to the
exaggeration of the trend in increase in vol-
ume of the anterior part of the gastric crop
observed in all carnivorous snails. In such
snails an oesophageal crop can hardly be de-
veloped without further anatomical modifica-
tions, because the space required for its de-
velopment is occupied by the buccal mass.

The gastric crop is not distinct from the oe-
sophagus in Eua (Partulidae, Fig. 26), and
can be recognized only from internal morpho-
logical details in Bocageia (Subulinidae, Fig.
343). In both cases, this apparent regression
in the development of the gastric crop is cor-
related (counterbalanced?) with the hypertro-
phy of the gastric pouch. This pouch exhibits
a particularly important internal differentiation
in Eua (Fig. 25).

In Diplomphalus (carnivorous, Rhytididae,
Fig. 388), the digestive tract is a simple tube,
in which there is hardly any internal or exter-
nal morphological differentiation. This mor-
phology might result from the association of
carnivory (dedifferentiation of the stomach, у.
infra) with whorls narrower than those in any
other observed snails, and with specialization
in diet, which consists at least in part of other
snails' eggs (young and adult Diplomphalus
megei were found in egg clutches of Placo-
stylus fibratus).

The occurrence of a well-developed gastric
pouch that prolongs the gastric crop without
being separated from it by a constriction
(PS2) is a general character of the Stylom-
matophora that occurs in all groups, associ-
ated with all shapes and sizes, although less
common in large animals (Text-figs. 13, 14).
The increase in the differentiation of the gas-
tric pouch, indicated by the presence of a con-
striction between pouch and crop (PS1), is
associated principally with large sizes (TA4,
TA5, Text-figs. 13, 14), and secondarily with
elongate shapes (HD3, HD4, HD5, Text-fig.
15). In the latter case, differentiation of the
gastric pouch is associated with change in the
shape of the gastric crop, as discussed
above. Dedifferentiation of the gastric pouch

(PS3) is associated with carnivory (BM2,
Text-figs. 13-15), and (or) with small sizes
(TA1, TA2). Because the gastric pouch is not
differentiated in newly hatched snails, ab-
sence of differentiation of the gastric pouch
might result from paedomorphosis through
progenesis when associated with small sizes
and few whorls.

The palatal surface of the gastric pouch has
a fleshy appendix in Eua (Partulidae, Fig. 26)
and in both Euglandina species | dissected
(Oleacinidae, Fig. 371). The internal surface
of the gastric pouch lacks any visible differ-
entiation that might correspond to the devel-
opment of this outer appendix in Euglandina;
in Eua the wall of the stomach seems covered
internally by a cuticle in the place to which the
appendix is attached. | am unable to interpret
the presence of such an appendix, which can
hardly be considered homologous with cutic-
ular and muscular plates in the stomach of
other pulmonates (Hubendick, 1978) because
of its arrangement.

In the Stylommatophora the shape of the
proximal intestine varies from cylindrical to
funnelform. This pattern of variation in shape
is difficult to interpret. A funnelform proximal
intestine is frequent in the family Zonitidae
(Figs. 241, 248, 254), and in the family Olea-
cinidae (Fig. 371) in which the diameter of the
proximal limit of the intestine approaches or
equals the diameter of the opening of the gas-
tric crop into the gastric pouch; but it occurs
also in other families, e.g. the Discidae (Fig.
187) and the Acavidae (Fig. 440).

Intestinal loops, rectum: When present, the
internal ornamentation of the intestine con-
sists of longitudinal ridges and grooves hav-
ing a radial symmetry. The principal typhlo-
sole might sometimes extend into the
periaortic loop, but | never observed it beyond
the crossing of the intestine and the aorta.
Although the intestine was not opened in all
species dissected, the following general prop-
ositions seem reasonable: first, ridges and
grooves are more frequent close to the anus
than more proximally; second, they are also
more frequent when the visceral mass is
short; and third, they are particularly frequent
among the Zonitoidea, whatever the degree
of reduction of their visceral mass.

The internal intestinal ornamentation prob-
ably improves the absorption (of water?)
through the intestinal surface by increasing
this surface, and improves the efficiency of
evacuating the feces. If this is true, occur-

38 TILLIER

rence of ornamentation allows the shortening
of the intestine without major functional prob-
lems: such an advantage might explain in part
the success of zonitoid slugs in their geo-
graphic area (Palearctic Parmacellidae,
Limacidae, Milacidae, and Trigonochlamy-
didae).

With only one exception, and disregarding
limacization, the intestine varies much more
in length than in arrangement. Variations in
relative intestinal length are approximately
proportional to variations in size of the ani-
mals: increase in size requires a larger in-
crease in intestinal surface (principle of simil-
itude). The latter is most generally attained
through increase in relative length of the in-
testine (111, 12, ТАЗ, TA4, TAS, Text-figs. 14,
15). The intestine of Helicophanta (Acavidae,
Fig. 464) has one loop more than does any
other stylommatophoran snail observed. This
odd arrangement might be related to the con-
junction of large size with visceral mass short-
ening. Very short intestines are associated
with small sizes (principle of similitude, TA,
TA2) and with carnivory (BM2, Text-figs. 14
and 15). Most usually the intestine is nearly
constant in diameter from its proximal end to
the rectum, except in Achatinella and in
zonitid snails, whose intestinal loops are
wider than the other portions of the intestine
(Figs. 16, 240, 241, 248).

In addition to the rectal wall, a second par-
tition often separates the lumen of the rectum
from the pulmonary cavity (Solem, 1966a). As
far as could be observed, it seems that this
arrangement results from the prolongation of
the visceral cavity forward between the lung
cavity and the rectum.

Limacization

The morphoclines and patterns of the di-
gestive tract in the course of limacization
have been described and discussed else-
where (Tillier, 1984a). All the slugs here stud-
ied can be placed in one of these already de-
scribed patterns. Three points should be
borne in mind. First, in most cases and at
least in its early stages, limacization involves
dedifferentiation of the gastric pouch and de-
velopment of a crop, the oesophageal or gas-
tric origin of which can hardly be determined.
Second, limacization also most usually in-
volves at least in its early stages the shorten-
ing of the intestine; this shortening may be
balanced by the development of intestinal
caeca or of ducts of the digestive gland. Third,

in advanced slugs, lengthening of the intes-
tine may cause secondary coiling or hypertor-
sion of the digestive tract into patterns that
are certainly not related to the torsion and
coiling of the ancestral snails.

In view of the diversity of taxa showing
limacization, the number of morphoclines in
the patterns of the digestive tract in limaciza-
tion is too low for their use in familial or su-
prafamilial classification. However, they might
be useful within families to determine whether
the patterns in two slugs whose close rela-
tionship is suspected are compatible, ¡.e. to
determine whether two observed patterns
might belong to a single morphocline. Com-
patibility of the digestive tract arrangements in
a group of slugs and semislugs does not
prove monophyly, but incompatibility obliges
one to reconsider the monophyly of the group.

Discussion

The morphology of the digestive tract is
generally more complex in non-stylom-
matophoran pulmonates than in the Stylom-
matophora, the family Otinidae excepted
(Morton, 1955; Hubendick, 1978; Тег,
1984b). Consequently, one might be tempted
to suppose that the digestive tract of the
Stylommatophora has been simplified in the
course of evolution. However, the morphol-
ogy of the digestive tract of the Stylom-
matophora is close to that of plesiomorphic
opisthobranchs and of the Otinidae (mor-
phology of the gastric pouch, contiguity of the

‚ Openings of the ducts of the digestive gland:

Gosliner, 1981; Tillier, 1984b). Consequently,
using the Otinidae as an outgroup for diges-
tive tract morphology, as is done for lung mor-
phology, is more appropriate than using other
pulmonate groups.

It then becomes clear that the most wide-
spread character states are plesiomorphic.
These are the spheroidal shape of the buccal
mass (when diet is not carnivory), the cyl-
indrical shape of the gastric crop, the differ-
entiation of the gastric pouch without an an-
terior constriction separating it from the gas-
tric crop. In plesiomorphic opisthobranchs
and in Otina the intestine is relatively long (IL1
in the factor analysis). However, intestinal
length is so often correlated with size in the
Stylommatophora it seems likely that, if the
primitive Stylommatophora were small, they
had a relatively short intestine (IL3) that was
secondarily either lengthened or shortened.
From such an archetypal pattern five patterns

STYLOMMATOPHORAN SYSTEMATICS 39

emerge. First, increase of intestinal length,
development of an oesophageal crop, and in-
creased differentiation of the gastric pouch
(constriction) are generally related to increase
in size. Second, inflation of the median por-
tion of the gastric crop associated with a gas-
tric crop separated from the gastric pouch by
a constriction is generally associated with
elongate shape. Third, increase in the diam-
eter of the anterior portion of the gastric crop
associated with dedifferentiation of the gastric
crop, lengthening of the buccal mass, and in-
testine shortening are generally related to
carnivory. Fourth, a poorly differentiated gas-
tric pouch, associated with a gastric crop hav-
ing the median portion inflated, might result
from paedomorphosis. Fifth, greater volume
for food storage might be gained by the de-
velopment of the anterior part of the gastric
crop if size is small or if the animal is carniv-
orous, or by the development of an oesoph-
ageal crop (generally when size is increased).
The taxonomic distribution of the trans-
formed character states implies parallel evo-
lution, as is to be expected from the functional
and statistical analysis presented above.

CENTRAL NERVOUS SYSTEM

The general arrangement of the stylom-
matophoran central nervous system has been
described above. The use of the variations in
this arrangement for taxonomy and phyloge-
netic reconstruction has been attempted by
Bargmann (1930) and Van Mol (1967). A pre-
cise nomenclature of the nerves issuing from
the central nervous system may be found in
the latter work.

Bargmann (1930) described principally the
arrangement of the visceral chain, constituted
by the visceral, parietal and pleural ganglia.
She stressed the importance of the apparent
fusion of the ganglia and recognized eight
types of arrangements, seven of which occur
in the Stylommatophora and can be grouped
into four patterns: in the orthurethran type
(Bargmann's type Ill), the visceral ganglion
appears fused with the right parietal (as in
Fig. 3); in the helicoid type (Bargmann's types
| апа V), the visceral ganglion appears fused
with the left parietal (as in Fig. 698); in the
other types the ganglia are either all distinct
(Bargmann's types |, Il) or fused with both
sides of the visceral ganglion (Bargmann's
types VI, VIII). The possibility of ordering

these arrangements into morphoclines has
been discussed by Bishop (1978).

Van Mol (1967) described principally the
microscopic anatomy of the cerebral ganglia.
With regard to phylogenetic patterns, he at-
tributes the greatest importance to the degree
to which the procerebrum is integrated with
the metacerebrum (persistence of two proce-
rebral commissures), and to the position of
the origin of the peritentacular nerve (meso-
or metacerebrum).

However, the taxonomic conclusions of
both these remarkable works can hardly be
accepted, mainly because of the small size of
the samples examined: in both cases whole
superfamilies are not represented, and obser-
vations in a single species are generalized to
family level, or even to superfamily or subor-
der levels. A caricatural example is provided
by Van Mol's use of the absence of a posterior
procerebral commissure as a synapomorphy
separating all Stylommatophora from the suc-
cineids, which justifies the suborder Heteru-
rethra. In fact, Watson (1928) described a
posterior procerebral commissure in ferus-
saciids; this single observation suffices to
modify Van Mol’s conclusions to a large ex-
tent. Bargmann's observations lack precision.
On one hand, she seems not to have dis-
sected the conjunctive sheath that envelops
the ganglia and makes the discontinuities be-
tween them inconspicuous, particularly in rel-
atively large animals (a few of Bargmann's
dissections were seen by me in the BMNH).
On the other hand, she seems not to have
realized that the contact of two ganglia does
not imply their fusion, even when no connec-
tive is visible: in general, two appressed gan-
glia are separated by a conjunctive wall (for
example, compare the descriptions of the vis-
ceral chain in Helix aspersa by Bargmann and
by Kerkut and Walker, 1975).

Having neither technical capacities nor
time to examine microscopically the central
nervous system of several hundred species,
| did not use the microscopic characters
shown and used by Van Mol (1967), ¡.e. the
degree of integration of the procerebrum and
the actual origin of the nerves. Similarly, |
did not use the number and arrangement of
the nerves originating from the central ner-
vous system. Indeed, the conjunctive sheath
around the central nervous system becomes
opaque and hard when preserved in alcohol
for a while, and nerve roots may easily be
torn in dissection. This is why, although they
have been figured, the number and arrange-

40 TILLIER

PEDZ2
PAG2TA1
IPAGI

! VG1

CPD2CPR2

HD1--TA2-------------------------------

HD2
CPDI ССЗ
CPR3PAG3
FGI! HD4
PLGI FG4
VG2
PLDIPAD2
PAG4 ТАЗ

HD5

TEXT-FIG. 16. Factor map of correspondences among characters of nervous system. Plane (1,2). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

ment of the nerves depicted in the figures
must be viewed with caution. Furthermore, it
is likely that such characters vary at subfamil-
ial levels, i.e. at a level lower than that con-
sidered here (see variations in the origin of
the penial nerve as described by Baker, 1938,
1940, 1941).

As a consequence the factor analysis here
(Text-figs. 16, 17, 18) is limited to the ar-
rangement of the ganglia in the central ner-
vous system, in relation to size and shape.
Semislugs and slugs are included. Four char-
acters and 15 character states represent the
arrangement of the anterior nerve ring, com-

posed of the cerebral, pleural and pedal gan-
glia. The first character is the length of the
cerebral commissure. The width of the right
cerebral ganglion was measured from the or-
igin of the optic nerve to the most posterior
point, at which the cerebro-pedal connective
joins the cerebral ganglion, giving three mo-
dalities: commissure distinctly shorter than
cerebral ganglia width (CC3), about as long
as the latter (CC2) or distinctly longer (CC3).
The second is the length of the right cerebro-
pedal connective. This length is an approxi-
mate index of the total length of the nerve ring
without the visceral chain. Three modalities

STYLOMMATOPHORAN SYSTEMATICS 41

PLG2

HD3

PAG2
ВЕ!
СРОЗ
ТА? CPR2
CPDI
FG1
HD4 VG2
TAY

teal
|
|
|
I
| PLD2 РАВ!
| TAI
|
|
|
|
№ Е02

PADI
| VG1
I
|
|
!
|
|
| CPR]
|
| CC2
|
|
| СРО2

=== === HD1------------------PLG3--3

PAG3 HD2
ССЗ CPR3
PLG1 PLD3
|

FG4
| РАО?

PLDI

PAG4Y TA3

|
I VG3
| HD5
|
| FG3
l

TEXT-FIG. 17. Factor map of correspondences among characters of nervous system. Plane (1,3). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

were found: shorter than width of the right ce-
rebral ganglion (CPD3), between one and two
times right cerebral ganglion width (CPD2),
more than twice longer than right cerebral
ganglion width (CPD3). The third character is
ratio between the lengths of the cerebro-
pedal connectives (left/right). This ratio is an
index of the asymmetry in the nerve ring, ven-
tral chain excepted. There are three modali-
ties: less than 0.9 (CPR1), between 0.9 and
1.2 (CPR2), from 1.2 to 2.5 (CPR3). The
fourth is the position of the pleural ganglia on
either side (PLG, the position of the left pleu-
ral ganglion; PLD, the position of the right

one). The positions of the pleural ganglia are
often not symmetrical, and the nerve ring
may be nearly epiathroid on one side and hy-
poathroid on the other. Because the pleural
ganglia form the extremities of the visceral
chain, their positions also represent its rela-
tive position. Three modalities occur on each
side: pleural ganglion closer to the pedal gan-
glion than to the cerebral ganglion (PLD1,
PLG1), closer to the cerebral ganglion than to
the pedal ganglion (PLD2, PLG2), very close
to both the pedal and the cerebral ganglion
(Р9З: 2863).

Four characters and 13 character states

42 TILLIER

PAG3
fey
РАО!
НО5
PLG3
PAG2 FG4
--------------------- PLD3-------------
СРВ!
СРО!
HD1
TAY

HD3

|
|
|
|
l
|
|
|
|
|
|
ТАЗ

|
| CPR3
| TA2
| CC2
| VG2
|
|

PLDI
IPLGICPD2
| FG2

CPD3

VG] PAG
CPR2PLD2
| PAG]
| HD2

HD4

TEXT-FIG. 18. Factor map of correspondences among characters of nervous system. Plane (4,5). Data,
Appendix B, Figs. 1-704; limits of classes, Appendix C.

represent the arrangement of the ventral
chain. The first is the position of the visceral
ganglion. lts modalities are: median plane of
the visceral ganglion on the right side of the
median plane of the pedal ganglia (VG1), me-
dian (VG2), on the left side of the median
plane of the pedal ganglia (VG3). The second
15 the position of the right parietal ganglion; it
has two modalities: in contact with the vis-
ceral ganglion only (PAD1), in contact with
both the visceral and right pleural ganglia
(PAD2). The third character is position of the
left parietal ganglion. Its modalities are: closer
to the left pleural ganglion than to the visceral

ganglion (PAG1), closer to the visceral gan-
glion than to the left parietal ganglion (PAG2),
in contact with the visceral ganglion only
(PAG3), in contact with both the left pleural
and visceral ganglia (PAG4). A compact vis-
ceral chain is represented by PAD2 and
PAG4. The last is the apparent fusion of the
visceral ganglion. The modalities are: none
(FG1), with the right parietal ganglion (FG2),
with the left parietal ganglion (FG3), with both
parietal ganglia (FG4).

The size of the animals is represented by
four modalities, /.e. four classes of equal size
(TA1 to TA4). Their shape is represented by

STYLOMMATOPHORAN SYSTEMATICS 43

five modalities: for HD1, the H/D ratio is less
than 0.6 (flat shells); for HD2, H/D is between
0.6 and 1.5 (helicoid shells); for HD3, H/D is
more than 1.5 (elongate shells). HD4 repre-
sents semislugs and HD5 represents slugs.

In the factor maps (Text-figs. 16-18), the
effect of variations in size appears along axis
1 (ТА1 opposed to ТАЗ and TA4), axis 3 (TA2
opposed to TA3) and axis 4 (TA4 opposed to
TA3 and TA2). The effect of variations in
shape appears along axis 2 (HD4 and HD5,
semislugs and slugs, opposed to snails) and
in the plane (4,5) (HD1, HD2, HD3, shapes of
snails).

Cerebral ganglia, cerebral commissure

The shape of the cerebral ganglia, which
varies from clearly elongate to spheroidal,
seems fairly constant within each family.
Comparison of their degrees of elongation
with Van Mol's data (1967) suggests that
there is no direct relationship between short-
ening of the cerebral ganglia and the integra-
tion of the procerebrum with the metacere-
brum.

The angle formed by the cerebral ganglia in
a plane parallel to the pedal sole is variable.
This angle seems to be generally larger when
the visceral mass is shortened and the cere-
bral commissure is short, as shown by the
figures: the longitudinal axis of the cerebral
ganglia is generally subperpendicular to the
cerebral commissure when the latter is long in
snails with a long visceral mass; it is generally
subparallel to the cerebral commissure in ad-
vanced slugs, and all intermediate arrange-
ments occur in relation to cerebral commis-
sure and visceral mass shortening.

Apparent shortening of the cerebral com-
missure is often due more to the expansion of
the metacerebrum towards the median plane
than to the actual shortening of the commis-
sure proper. However, this is not so when the
apparent difference in length is important, as
when comparing, for example, the oleacinids
(Figs. 362, 368, 373) with the clausiliids (Figs.
509, 511, 520). This case apart, correlations
of the cerebral commissure length with other
modalities appear along axis 1 (Text-figs. 16,
17), and particularly in the plane (4,5) (Text-
fig. 18). Occurrence of a short cerebral com-
missure is partly related to large size (Text-fig.
16): no large snail has a long cerebral com-
missure, but some large slugs like Parmacella
(Parmacellidae, Fig. 262) and Oopelta (Arion-
idae, Fig. 208) do. The converse is not true,

for very small snails like Varicella (Olea-
cinidae, Fig. 368) have a very short cerebral
commissure.

In snails a long cerebral commissure is
usually associated with an elongate shell
(CC1, CC2, HD3, axis 4, Text-fig. 18), and its
shortening is associated with flat or globular
shapes (CC2, HD1, HD2). This overall corre-
lation, which often is not true in particular
cases, might have a functional origin but
might equally well be explained by phyloge-
netic relationships if both an elongate shape
and a long cerebral commissure are plesio-
morphic. At least in some corillids (Fig. 484)
and in some oreohelicids (Ammonitellinae in-
cluded) (Figs. 495, 498), a flat shape is asso-
ciated with a long cerebral commissure,
whereas a short cerebral commissure is often
associated with an elongate shape in the
Orthurethra, Urocoptidae (Figs. 530, 534) and
Bulimulidae (Fig. 538).

Lateral connectives, pedal ganglia

As stated above, length of the cerebro-
pedal connectives relative to cerebral ganglia
width is an index of the total relative length of
the nerve ring, the visceral chain excepted.
On each side the length of the cerebro-pedal
connective is slightly shorter than the sum of
the cerebro-pleural connective length, pleural
ganglion width and pleuro-pedal connective
length. The lengths of the right and left
cerebro-pedal connectives are subequal in
about half of the observed species; in the
other half the left cerebro-pedal connective is
more commonly the shorter.

In the factor maps (Text-figs. 16-18), oc-
currence of a short right cerebro-pedal con-
nective (CPD3) contributes mainly to axes 2
and 5; medium lengths (CPD2) contribute
mainly to axis 3 and less to axis 2; long
lengths (CPD1) contribute to axis 2 and more
to axis 5. The ratio of the lengths of the left
and right cerebro-pedal connectives (CPR1,
CPR2, CPR3) contributes mainly to axes 3
and 4, but the contribution of the lowest val-
ues of this ratio (CPR1) to axis 3 is nil.

The shortest cerebro-pedal connectives
are clearly related to limacization (СРОЗ,
HD4, HD5; axis 2, Text-fig. 16). As might be
expected, short length or apparent absence
of the cerebro-pleural and pleuro-pedal con-
nectives (PLG3, PLD3) is also related to lima-
cization, which therefore is generally associ-
ated with shortening of the anterior nerve ring.
Exceptions are one parmacellid (Fig. 261),

44 TILLIER

and carnivorous slugs (Plutonia, Fig. 233;
Daudebardia, Fig. 250; Strebelia, Fig. 374). In
the case of carnivorous slugs, the retention of
a long anterior nerve ring can be explained
easily by the functional necessity for extend-
ing around the voluminous buccal mass as-
sociated with carnivory. On the contrary, me-
dium and long cerebro-pedal connectives are
clearly related to neither size nor shape, but
long cerebro-pedal connectives are associ-
ated with a long cerebral commissure (CPD1,
CC1; axis 5, Text-fig. 18). The latter associa-
tion is probably the plesiomorphic condition of
the anterior nerve ring, as will be discussed
below.

Although varying less within families than
between them, the ratio of cerebro-pedal con-
nective lengths seems to vary independently
of the other modalities to a large extent. How-
ever, relative shortening of the left cerebro-
pedal connective is weakly correlated with
flat shapes (CPR1, HD1; Text-figs. 17, 18);
whereas its relative lengthening is slightly bet-
ter correlated with elongate shapes (CPR3,
HD3; Text-fig. 18). This pattern of variation can
hardly be associated with hypertorsion in the
visceral mass in relation to change in shape,
which should cause the opposite effect. Vari-
ation in the asymmetry of the cerebro-pedal
connectives might be related to variation in the
angle between the longitudinal axes of the
columella and the foot when the animals are
active, but this hypothesis is purely intuitive
and cannot be tested at present.

The cerebro-pedal and pleuro-pedal con-
nectives insert on the posterior external sur-
face of the pedal ganglia. The position of the
pedal ganglia is related to connective length,
the ganglia being farther back in the pedal
cavity when the lateral connectives are short.
In most cases, there are two commissures
uniting the pedal ganglia. | do not know
whether two commissures are present also
when pedal ganglia are in close contact.

The pedal ganglia might be elongated par-
allel to the foot longitudinal axis (not seen in
any large animal), subcircular, or elongated
perpendicular to the foot longitudinal axis.
The last shape seems to be related to short-
ening of the visceral mass, and no observed
semislug or slug has pedal ganglia longitudi-
nally elongate.

The statocysts are on the upper side of the
pedal ganglia, behind the insertion of the cere-
bro-pedal connectives and usually on the inner
side of the latter. In small snails with the pedal
ganglia longitudinally elongate, the statocysts

often form their posterior extremity. In large
snails and slugs, especially when the pedal
ganglia are transversally elongate, the stato-
cysts are generally farther forward. In very
large snails and slugs, they are generally
deeply embedded in the pedal ganglia and are
invisible without dissection.

Visceral chain

Position and length: The position of the vis-
ceral chain relative to the anterior nervous
ring is determined by the ratio of the lengths
of the cerebro-pleural and pleuro-pedal con-
nectives on each side (PLG1, PLG2; PLD1,
PLD2). As discussed above, extreme short-
ening of both connectives on each side
(PLG3, PLD3) is usually related to limaciza-
tion. In the factor maps (Text-figs. 16-18)
these modalities contribute significantly to
axis 1, and weakly to axis 4. Relatively impor-
tant lengths of the visceral chain are repre-
sented by PAG1, PAG2 and PAD1, and short
visceral chains by PAG4 and PAD2, which
contribute mainly to axis 1.

A short visceral chain (PAG4, PAD2) is
closely related to large size (ТАЗ, TA4),
whereas a long visceral chain (PAG1, PAG2,
PAD1) is closely associated with small size
(particularly TA1). This correlation is impor-
tant for phylogenetic analysis, inasmuch as a
long visceral chain is plesiomorphic: shorten-
ing of the visceral chain might have a different
significance in small than in large animals,
and cannot be used if not related to size. Like
compaction of the anterior nerve ring, com-
paction of the visceral chain is also closely
related to limacization (HD5); consequently
close relationship of a slug having a very
compact central nervous system with a snail
having a very long central nervous system
does not imply the former existence of a snail
ancestor of the slug with a short central ner-
vous system.

For a visceral chain of a given length, the
diameter of the perioesophageal ring is
smaller when the cerebro-pleural connectives
are shorter than the pleuro-pedal connectives
(PLG2, PLD2), than when the pleural ganglia
are closer to the pedal ganglia (PLG1, PLD1).
This explains why (Text-fig. 16) the occur-
rence of a short visceral chain is closely re-
lated to proximity of pleural and pedal ganglia
(PAG4, PAD2, PLG1, PLD1, along axis 1): a
short visceral chain is in general functionally
impossible unless the pleural ganglia are
close to the pedal ganglia, because it would

STYLOMMATOPHORAN SYSTEMATICS 45

strangle the oesophagus. Correlatively, the
proximity of the pleural to the cerebral ganglia
is unlikely unless the visceral chain is rela-
tively long.

It can be seen in Text-fig. 16, along axis 1,
that shortening of the cerebro-pedal connec-
tives, when associated with shortening of the
visceral chain, is not symmetrical. For a given
length of the visceral chain, the pleural gan-
glion is closer to the cerebral ganglion on the
right side than to that on the left (e.g. in some
endodontoids, Figs. 147-183). This general
asymmetry might be the direct result of gas-
tropod torsion; it does not seem necessary to
seek a further explanation.

Position of ganglia in visceral chain: In
nearly all Stylommatophora, the right parietal
ganglion closely touches the visceral gan-
glion. In contrast, the left parietal ganglion
might be closer to the left pleural ganglion
than to the visceral ganglion, closer to the vis-
ceral ganglion, or in close contact with the
latter even in a long visceral chain. In very
short visceral chains, the parietal ganglia
touch the pleural ganglia.

The location of the visceral ganglion on the
right side of the median plane is clearly re-
lated to a long visceral chain (VG1, Text-fig.
16). Itis probably plesiomorphic, inasmuch as
it might result directly from gastropod torsion.
The displacement of the visceral ganglion to-
ward the median plane and to the left side of
the latter is related to shortening of the vis-
ceral chain, but only to the extent that limaci-
zation is not involved. This relation appears in
the plane (1,2) of Text-fig. 16, where VG1 ap-
parently is not associated with any shape mo-
dality, whereas VG3 (visceral ganglion on the
left) is opposed to HD4 and HD5, which гер-
resent semislugs and slugs along axis 2.
Whereas visceral chain shortening results in
displacing the visceral ganglion to the left
when related to size increase, it might be that
shortening of the visceral chain related to
limacization preserves to a large extent the
arrangement of the ganglia in the ancestral
snails. When related to visceral chain com-
paction associated with size increase, the dis-
placement of the visceral ganglion to the left
might be a simple mechanical consequence
of the disappearance of the connectives, be-
cause the right parietal ganglion is generally
larger than the left one.

Absence of fusion of the visceral ganglion
with both parietal ganglia (FG1), or its appar-
ent fusion with the right parietal ganglion

(FG2), is correlated with small size and a rel-
atively long visceral chain (Text-figs. 16, 17).
Apparent fusion of the visceral ganglion with
the left parietal ganglion is correlated with dis-
placement of the former to the left, ¡.e. with
shortening of the visceral chain in snails
(VG3, FG3; axis 3, Text-fig. 17). Apparent fu-
sion of the visceral ganglion with both parietal
ganglia is strongly correlated with visceral
chain compaction when associated with lima-
cization (FG4, HD5; axis 2, Text-fig. 16).

The position of the left parietal ganglion
closer to the left pleural ganglion than to the
visceral ganglion (PAG1) might be plesiomor-
phic, because it occurs in the Otinidae and is
probably plesiomorphic for all pulmonate
groups (Tillier, 1984b). lts displacement to-
ward the visceral ganglion seems indepen-
dent of variations in the relative position of
other ganglia, and therefore might be impor-
tant for phylogenetic analysis (PAG2, Text-
fig. 16). Unfortunately the distances between
this ganglion and the visceral ganglion are
comparable only if the visceral chain is not
compact, and in this respect snails and slugs
with a compact visceral chain provide no in-
formation; in other words, PAG2 seems cor-
related with a long visceral chain and small
sizes (Text-fig. 16) only because large size is
correlated with compaction of the visceral
chain.

In carnivorous snails, the visceral chain is
necessarily either relatively long or inserted
close to the pedal ganglia, in order to encircle
the large buccal mass. If lateral connectives
are relatively short, lengthening of the visceral
chain seems to have occurred, but the visceral
connectives seem to have been unequally
lengthened. Presumably only the connectives
that were the longest before lengthening have
been lengthened, which results in odd ar-
rangements. The very originality of these ar-
rangements makes them particularly interest-
ing for phylogenetic reconstruction. In the
families Rhytididae (Figs. 376, 385, 391) and
Streptaxidae (Fig. 396) the visceral and pari-
etal ganglia are compacted into a single mass
that is united to the pleural ganglia by long
connectives. These connectives are excep-
tionally long in the Streptaxidae (Fig. 396). In
the family Systrophiidae (Fig. 401), only the
left parieto-pleural connective is visible and is
exceedingly long; the left pleural ganglion is
appressed to the left cerebral ganglion, and
the visceral and right parietal ganglia are
seemingly fused together and appressed to
the right cerebral ganglion.

46 TIELIER

Discussion

A polarity can easily be proposed for the
morphoclines of most characters of the cen-
tral nervous system, by outgroup comparison
with plesiomorphic opisthobranchs and ar-
chaeopulmonates, in particular otinids (Gos-
liner, 1981; Hubendick, 1978; Tillier, 1984b).
The following modalities are plesiomorphic: a
long cerebral commissure; proximity of the
left parietal and left pleural ganglia; probably
rather long cerebro-pedal connectives; char-
acter states that might result directly from
gastropod torsion, i.e. the position of the vis-
ceral ganglion on the right side of the median
plane, a left pleuro-pedal connective shorter
than the right one and a left cerebro-pedal
connective shorter than the right one.

On the other hand, proximity of pleural and
cerebral ganglia is probably apomorphic with
respect to proximity of pleural and pedal
ganglia. Comparison of the various patterns
suggests that it results primarily from length-
ening of the pleuro-pedal (and cerebro-pedal)
connectives, followed by shortening of the
cerebro-pleural connectives. The latter might
disappear, to the extent that the visceral
chain is long enough to allow retraction of the
anterior digestive tract through the perioe-
sophageal ring.

Proximity of the right parietal and visceral
ganglia seems to be a synapomorphy of all
Stylommatophora. Within the Stylommato-
phora, any long connective is apomorphic with
respect to a shorter homologous connective,
and the apparent fusion of two ganglia is apo-
morphic with respect to simple contact of the
same ganglia, of which the limits remain dis-
tinct.

Although compaction in the central nervous
system might be irreversible in most cases,
the extreme length of some connectives, such
as the lateral and visceral connectives of
some carnivorous snails (Streptaxidae, Sys-
trophiidae), can hardly be plesiomorphic. It
seems that some connectives might lengthen
in the course of evolution insofar as they are
not very short, but the secondary develop-
ment of a connective between two appressed
ganglia is unlikely.

CHARACTERS OF
STYLOMMATOPHORAN FAMILIES

Ir thi 5 tion
genetic pattern

before discussion of phylo-
in the Stylommatophora, an

account of temporal and geographical distri-
bution, and of the amplitude of the variations
seen, is given for each family. Transfers at
infrafamilial levels are tentatively justified.
The order of presentation of the families pre-
supposes the next section, a better approach
than following here an order that will be re-
jected a few pages further. In the course of
this systematic description, some groupings
and transfers are discussed and justified. The
most plesiomorphic character states seen in
each family are summarized, in order to allow
the phylogenetic reconstructions attempted
below: the data are too numerous to be easily
handled for building phylogenetic trees, and
the most reasonable way to reduce their num-
ber seems to be to use only the most plesio-
morphic states in each group for which mono-
phyly is accepted. This approach, of course,
implies that the direction of evolutionary
change was correctly stated and that reversal
can be neglected at familial and suprafamilial
levels. This operation is equivalent to replac-
ing members of a family by their hypothetical
closest common ancestor (CCA).

In the course of the discussion, the results
of the preceding sections are used, in part as
a table of character states ordered from 1 (=
plesiomorphic) to n (= apomorphic), together
with two phenetic classifications, one for the
species dissected and one for the families
(Appendix E, Text-figs. 19, 20). These phe-
netic classifications are not at all definitive,
but provide indications of similarity among
taxa that can be used in further discussions.

Both phenetic classifications are ascending
hierarchical classifications (Jambu 8 Le-
beaux, 1979), built by the CAHCAR program
in the ADDAD package (CIRCE) from the
data of Appendices B, C. The limits of the
classes, which are not the same as in the
preceding partial factor analyses, are given in
Appendix D. All the data were recoded 0 or 1,
but no other change was made; missing data
were coded 0. The data table for families was
obtained by summing the lines representing
the members of each family column by col-
umn, and recoding.

The conclusions of the preceding discus-
sions have been used to determine the value
of each character state in the table of ordered
character states (Appendix E). The limits for
each character state are the same as in Ap-
pendix D and as in the phenetic classifica-
tions; only the coding is different. Characters
for which no direction of evolutionary change
can be proposed have been eliminated.

STYLOMMATOPHORAN SYSTEMATICS 47

When two paths are possible from a single
plesiomorphic state, one has been noted 1,2,
3, 4, etc. and the other 1, 2’, 3’, 4’, etc. For
further analyses such data were recoded in
two columns, as explained below.

Orthurethra

In nearly all the Orthurethra, the kidney is
very long (LR2’), internally divided into at
least two morphologically distinct regions
(RR1), and without a ureteric tube (UR1).
The few exceptions will be mentioned in the
descriptions of the families to which they be-
long. No orthurethran is carnivorous (BM1).
The oesophageal crop is most usually absent
(OC1), sometimes little developed (OC2).
The gastric crop is cylindrical (SC1) or in-
flated in its median portion (SC2). The gastric
pouch is always differentiated (PS1, PS2).
The intestine is generally rather long (IL1).
Contrary to Bargmann’s opinion (1930), the
visceral ganglion is not always seemingly
fused with the right parietal ganglion (FG2).
The right cerebro-pedal connective is usually
shorter than the left one. The lateral con-
nectives are always distinct (PLG and PLD 1
and 2).

Far more precise descriptions than those
presented here, but dealing with fewer taxa,
may be found in the remarkable work of
Steenberg (1925).

Achatinellidae (+ Tornatellinidae) (Figs. 1-
24): The grouping of the families Achatinell-
idae and Tornatellinidae has been proposed
by Cooke and Kondo (1960). Although not
discussed by them, this grouping clearly
seems based upon overall similarity and geo-
graphical distribution. Although | do not know
of a synapomorphy of the two groups, uniting
them in a single family seems justified from
the data presented here on three grounds.
First, all characters retained (Appendix E) ei-
ther have the same state in both groups, or
are more apomorphic in Achatinella than in
non-Achatinellinae achatinellids; second, in
the phenetic classification of the species, the
three species the closest to Achatinella are
tornatellinids (Text.-fig. 19); and third, in the
phenetic classification of the families, the two
groups are more similar to each other than to
any other family (Text-fig. 20).

The Recent Achatinellidae are endemic to
the Pacific Islands and along the western
margin of the Pacific Ocean, although a few
species have been introduced to the islands

of the Indian Ocean (Cooke & Kondo, 1960).
The Pitysinae, Tubuaia excepted, are en-
demic to the Austral Islands and the Achati-
nellinae are endemic to the Hawaiian Islands,
whereas the Tornatellininae occur throughout
the range of the family. One Carboniferous
genus from Europe and North America has
been attributed to the Achatinellidae (Solem &
Yochelson, 1979). This attribution can be jus-
tified only on conchological characters, of
which the evolutionary state is either dubious
(shape, sculpture) or probably plesiomorphic
(apertural teeth). The same arguments cast
doubt on the affinities of Protornatellina, de-
scribed from the Danian of North America.
These genera are probably primitive, but their
primitiveness does not at all imply monophyly
with Recent achatinellids.

The digestive tract of the Achatinellidae is
plesiomorphic for all characters considered in
the various analyses. However, the important
length of the intestine of Partulina was noticed
by Pilsbry (1900a), and one can see (Fig. 16)
that the intestine of Achatinella lorata is par-
ticularly broad. From the size of achatinel-
lines, which are larger than in the other sub-
families, it might be supposed that these
modifications in intestinal morphology are re-
lated to size increase, as previously dis-
cussed for all the Stylommatophora; but the
increase in intestinal surface through an in-
crease in intestinal diameter has been ob-
served in Achatinella only, whereas the in-
crease in surface through an increase in
length seen in Partulina is more usual.

In achatinellids the cerebral commissure is
never very long (CC1 absent). The right
cerebro-pedal connective is never longer than
the left (CPR1 absent). The right pleural gan-
glion is always closer to the right cerebral
ganglion than to the right pedal ganglion
(PLD2), whereas the position of the left pleu-
ral ganglion varies. The visceral ganglion is to
the right of the median plane (VG1), and ap-
parently fused with the right parietal ganglion
(FG2).

Association of the most plesiomorphic
characters seen in the Achatinellidae pro-
vides the following hypothetical closest com-
mon ancestor (CCA): BM1 OC1 SC1 PS1 IL1
ER2ZIZURIFRRIEGE2ZERBIZEPR2ZPLD2
PLG1 VG1 PAD1 РАС1 FG2.

Valloniidae (+ Strobilopsidae) (Figs. 84—
107, 110, 111): The Valloniidae are prin-
cipally Holarctic; Pupisoma $.1. occurs in trop-
ical regions, where probably it has been

SISdO 1! Sth
SIS4O4YI0SES
VINYdLWIAdOh

SNIT3HI0K40YLSOKh
У ЗАЗ 10 4HOb
VLNVH 4991 13H0h
v1 13-9GNYdOh

XV 11492
vIN91713d[h

YLdA 190NYOh

XWWI 71/2

EV 11399439456
1139 Уч 456
1139 1398$
V113891048£

VNO9O 11406

YNI IALNS0364

SNF 10ISIQTh
SMIALSOIV IdTh

XI 13HONW 170 1VH8S
V111211348S
$1$400013105
VILL УЗЛА 9$
VINNIdY3SOYdTS
Y1N0(09173H8S
VHIVNOW8S
SILINOZhZ

soal IVXTS

УМП 103129115
NOTYVNWAITE

МО IYYHLNYIVTE
NOTYVITYSVSINTS
VNILVHIVRE
SMWOYA IHANVES

SN ALSO 13H96
VI49V204SS
3019491908
WINSSI190ANISAEE
NOA0S3W0S
VN1VaAGVH4 9

VI dAT9OHIN IN 13H/S
Y113NN4 3985

ZW 3810 14H 9h

SI 71043945

SNHINI MARY IES
YNO 1385
Va 3HL8S
v3Vd3

TILLIER

TEXT-FIG. 19. Phenetic ascending hierarchical classification of stylommatophoran species studied. Data,

Appendix B, Figs. 1-704; limits of classes, Appendix D.

STYLOMMATOPHORAN SYSTEMATICS

VINO TIVATI
VINLVdOHIALATT
)0)S 100€

VIaVLAINISZS
SI 1AdO1931d2S
$$1$909113Н85

$1$909113Н85
VIVN/h
SN IVHANO 141(14h
CVANIDINSAT
VINIDINMSHT
VNO91911148S
43095415
VINGIONVI8S
651$409113Н85
VYOHdONVSAHLOS

¿S1 1043945

SISdONIYLIAOE

VTIAVIYVNTS
V38L1AhZ
CWIHdOYLSASSH
VW3410 14VH9h
X113H0 340SS
VaILAHUZh

V113100Nvd0h
NOTYYWUVAOHIIWTE
NOTYYWUVATE

VSSO 1907 IHIS/h
XWWI10SI 141s
NOXOLYTE

NOT HV INNVY9TE
VNYOWVS£O
VINLYVdEO

vN3£0
$31000 199 14Th
$15400013105
11130905

YNIYO 1HIOITH
VWOLSO1 dA19hS
VITHIONFILINIHdS8S
vVavovHuss

NO 13WNNISES
SI0AYVI0h

Е 113300515
VLI 13dWv0h

SI 194309 VON
VYAION IG 1d0t
VHOdOW9VHHdI 4345
NOGO 141509
SNYHA3NO918L0H
VISVIYOCOh
SNAVIVORh
SId0Y101931d£5

TILLIER

O ANA
RA
sIYv8ldS8£

#1) 1:41 INO IOS
SMS 1104 3193205
SN): :0014d/h
SNJJ!MHINJANC

$3. INGZU)HIOALTE
Va IUNINVYZT

Yul dS iNONVET
YCQUHHdILSZT
СУ IICIVNIHIEZ
Wd/.9A 10405
WTNO dd IL THIS
¥113N09 191$
SId0#1019314£S
V1931dI3HIS

VI 11343AV/S
VNLSVH IST

WwOV 18T
VILSIHIVAST

WI dOWINVST
VIONVNYTAVAOST
V113d01SAI8T
ЗУа 1 YOHdO VEOH LYST
CSNIAWO TIHdTZ
VI111HdW3H02
XANO 1VWOHT

NO [3902

Хим 1 10 13906

NO I YY 345 ЗНОС
NOVS 9 403406
XVW IL IN IHdEZ
§N3109¥Z02

ZW! 1140002

NO LYVI “HAVOZ
SNIY IVO 3902
SNIAWO TIHdTZ
VIQYVa 3UNVONZ
VIiTIN! 409 3vhZ
YOOWATLSAWLT
XIHAWOS IWhZ
SATIHJAXONE

¥1 1340002

Y1131 INOWWVHNS
X113H8S
V13H2V410 1! 14960
S30 101 1NOZhZ
S$NIS1G091 13H6T
2¿5n3S1061
SNSIGGT

STYLOMMATOPHORAN SYSTEMATICS

[
203165
HLOIZO
[dNdOL
any 1OT
124080
VN11N009V480
YNI INOXILTO
VINGINYYAd9O
$19$19039134$ ТТ
Yd0911H90950
VIXY31S/0
VINNIHINVIVIT
Y301 1134710
VNILVHIV1d31h0
VAL SVWVHO
SVIWSYIITO
SN114081STO
V113NILVHIVZO
VNITII3LYNYOLTO
ATEN,
SI1d0204Nth
S30101 119492
VW 34 10H9A1d6h
VILON3938hh
VOIVSTS

V11191 1420285
XABLSOATH
VINYANOHIET
VN3£T

NOIY39€h
VINLAIYVANIST
VNI443ZST
YdNd0L¥10S60
VNIWNYEE

VNI INENSES
SMWVYSIONIVWhh
VIVLIZE

VININZE

VIAVNIA 1VZE
VIWWITATT

XI 13SONNVZT
NOQOLVWNVHLIT
SNHLYNIX I HHdBT
CSNINNOIOIS 140$
VIINVYSSIYONYZT
SILSAIAHIVALZT
VdOUVHIZT
VIHdOYLSASSh
Va4 1191

SNA NODOS
VIOYVHIOANISAZT
EVLIIIININOIOE

52 TILLIER

53 CAMAENIDAE

57 HELMINTHOGLYPTIDAE
50 POLYGYRIDAE

56 BRADYBAENTDAE
34 ACHATINIDAE

58 HELICIDAE

41 BULIMULIDAE

40 ACAVIDAE

20 ARIONIDAE

21 PHILOMYCINAE

23 VITRINIDAE

15 ATHORACOPHORIDAE
25 PARMACELL IDAE
27 LIMACIDAE

26 MILACIDAE

31 HELICARIONIDAE
38 OLEACINIDAE RR
49 STREPTAXIDAE

47 RHYTIDIDAE

52 CORILLIDAE

44 UROCOPTIDAE

37 CLAUSILIIDAE

33 SUBULINIDAE

32 FERUSSACIIDAE
51 SAGDIDAE

45 SYSTROPHIIDAE
17 CHAROPIDAE

18 PUNCTIDAE

16 ENDODONTIDAE

30 EUCONULIDAE

24 ZONITIDAE

29 TROCHOMORPHIDAE
19 DISCIDAE

14 SUCCINEIDAE

48 HAPLOTREMATIDAE
55 OREOHELICIDAE
54 AMMONITELLINAE
13 ENIDAE

03 PARTULIDAE

04 AMASTRIDAE

43 CERIONIDAE

08 ORCULIDAE

09 CHONDRINIDAE

10 PUPILLIDAE

06 PYRAMIDULIDAE
11 VALLONIIDAE

07 VERTIGINIDAE

05 COCHLICOPIDAE
02 ACHATINELLIDAE
01 TORNATELLININAE

TEXT-FIG. 20. Phenetic ascending hierarchical classification of stylommatophoran families. Data were
obtained by summing indices representing each state, column by column, in each family, in data appendix

used for classification shown in Text-fig. 19.

spread by man (Boss, 1982), and at least one
“Strobilopsidae” occurs in New Guinea
(Solem, 1968). The family ranges from the
Paleocene to the Recent in Europe.

In the phenetic classification of the species,
representatives of this family are dispersed as

are all the Orthurethra (Text-fig. 19). The fam-
ily is generally similar to the Vertiginidae and
the Cochlicopidae, but the meaning of this
similarity is doubtful in view of the great intra-
familial variability and the method used to
construct the classification (Text-fig. 20).

STYLOMMATOPHORAN SYSTEMATICS 53

Pilsbry (1948) proposed a monogeneric
family Strobilopsidae based on the similarity
of some conchological characters of Stro-
bilops to those of the Pupillidae and Achati-
nellidae, i.e. upon purely phenetic concholog-
ical arguments. These characters might be
plesiomorphic (apertural teeth); and because
no anatomical character other than the plesi-
omorphic absence of fusion of the right pari-
etal with the visceral ganglia allows distinction
between the “Strobilopsidae” and other Val-
loniidae, | see no reason to maintain two dis-
tinct families.

Characters of the digestive tract are gener-
ally plesiomorphic, with the exception of a fre-
quently short intestine and some gastric crop
morphologies (gastric crop inflated in its me-
dian portion in Vallonia, possibly through
paedomorphosis; and funnelform in Ptycho-
patula). A ureteric tube occurs in Acanthinula
(UR2). The cerebral commissure and cere-
bro-pedal connectives are never long (CC2,
CC3, CPD2, CPD3). The left cerebro-pedal
connective is longer than the right, except in
Spermodea (CPR1, CPR2, CPR3). The right
pleural ganglion is always closer to the cere-
bral ganglion than to the pedal ganglion
(PLD2), whereas the left pleural ganglion is in
the same position only in Strobilops. As noted
above, this genus differs from all other genera
in its lack of fusion of the visceral and right
parietal ganglia (FG1). The visceral ganglion
is always on the right side (VG1).

Character states of the CCA: BM1 OC1
sCi Psi Iki LR2’ URI ARI CC2 CPD2
CPR1 PLD2 PLG1 VG1 PAD1 PAG1 FG1.

Pupillidae (Figs. 74—83): The family name
is used here in its restricted sense, as defined
by Zilch (1960; = Pupillinae + Lauriinae).
Solem (1978) and Boss (1982) consider that
the Orculidae, Chondrinidae and Vertiginidae
should also be included. This position is com-
prehensible in view of the proximity of these
families or of some of their members in the
phenetic classifications (Text-figs. 19, 20).

Some Pupillinae are’ known from the
Eocene in North America, and from the Oli-
gocene in Europe. Recent taxa have been de-
scribed from nearly everywhere in the world,
but it will be impossible to evaluate the influ-
ence of accidental introductions as long as
these very small and very variable animals
remain unrevised.

In both genera studied here the median
portion of the gastric crop is inflated (SC2).
The cerebral commissure is not long (ab-

sence of CC1). The cerebro-pedal connec-
tives are subequal in length (CPR2) and rel-
atively long (CPD1, CPD2). The pleural
ganglia are closer to the cerebral ganglia than
to the pedal ganglia (CPD2, CPG2). The vis-
ceral ganglion is median (VG2). The left pari-
etal ganglion is closer to the visceral ganglion
than to the left pleural ganglion (PAG2), the
former being apparently fused with the right
parietal (FG2).

Character states of the CCA: BM1 OC1
362 РНЕ В 2 URI RRIACC2ICPDA
CPR2 PLD2 PLG2 VG2 PAD1 PAG2 FG2.

Pyramidulidae (Figs. 49-51): The family is
monogeneric or contains two genera if Pleu-
rodiscus, the only genus of the “Pleurodis-
cidae,” is included. No specimen of the latter
could be found, and oniy one Pyramidula was
dissected. The family is known from the Eu-
ropean Eocene. In both phenetic classifica-
tions and in phylogenetic reconstructions (v.
infra), the only pyramidulid seen is closer to
the Pupillidae than to members of any other
family: if my sampling were not so slight, |
would not hesitate to consider the pyramid-
ulids a subfamily of the Pupillidae.

The gastric crop is inflated in its median
portion (SC2). The cerebral commissure is
short (CC3), but the lateral connectives are
long (CPD1) and subequal in length (CPR2).
The pleural ganglia are closer to the cerebral
ganglia than to the pedal ganglia (PLD2,
PLG2). The visceral ganglion is on the right
side (VG1), and the right parietal ganglion is
closer to the right pleural ganglion than to the
visceral ganglion (PAG1). The visceral gan-
glion seems fused with the right parietal gan-
glion (FG2).

Character states of Pyramidula (used as
the CCA): BM1 OC1 SC2 PS1 IL1 LR2’ UR1
RR1 ССЗ CPD1 CPR2 PLD2 PLG2 VG1 PAD
1 PAG1 FG2.

Chondrinidae (Figs. 65-73): The family
ranges from the Eocene to the Recent in Eu-
горе. It is now mainly Holarctic and Oriental,
but contains a few taxa of which the shell mor-
phology is peculiar to the family, and that ex-
hibit an interesting disjunct geographical dis-
tribution. This small group is composed of the
African and Madagascan Fauxulus, the South
American Gibbulina and Ulpia, and the Orien-
tal, Australian and South African subfamily
Hypselostomatinae. The great morphological
difference between the two genera dissected
in the family, Gyliotrachela (Hypselostomati-

54 TILBIER

nae) and Solatopupa (Chondrininae), makes
the hypothesis of the phylogenetic unity of
this small group worth testing (Text-fig. 19).
Unfortunately, not enough material was avail-
able to me.

Taken together the two species dissected
are close to the Cerionidae (probably not sig-
nificant, v. infra), the Orculidae, Pupillidae
and Pyramidulidae (Text-fig. 20).

The gastric crop is inflated in its median
portion (SC2). The gastric pouch is separated
from the gastric crop by a constriction (ES2’).
The cerebral commissure is long to medium
in length (CC1, CC2). The cerebro-pedal con-
nectives are medium in length (CPD2); they
are asymmetrical but their asymmetry is re-
versed in the two genera (CPR1 in Solato-
pupa, CPR3 in Gyliotrachela). The pleural
ganglia are close to the cerebral ganglia in
Gyliotrachela (PLD2, PLG2), and close to the
pedal ganglia in Solatopupa (PLD1, PLG1).
The left parietal ganglion is closer to the left
pleural than to the visceral ganglion (PAG1).
The latter seems fused with the right parietal
ganglion (FG2).

Characters of the CCA: BM1 OC1 SC2
PS2NIEMER2AURTRRI:CCINCPD2 GPRI
PLD1 PLG1 VG2 PAD1 PAG1 FG2.

Cochlicopidae (Figs. 43-45, 47, 48): The
Cochlicopidae are Holarctic and include only
four genera, of which only the most common
species, Cochlicopa lubrica, was dissected.
The family is known from the Paleocene in
Europe and North America.

In both phenetic classifications, Cochlicopa |

is close to the Vertiginidae and Valloniidae
(Text-figs. 19, 20). It is relatively close to the
Amastridae, as in the commonly accepted
classification and in the phylogenetic trees
presented below, in the phenetic classifica-
tion of the species but not in the phenetic
classification of the families (sampling? Coch-
licopidae + Amastridae = Cionellacea =
Cochlicopacea in Boss, 1982).

In Cochlicopa lubrica the cerebral commis-
sure and the lateral connectives are particu-
larly long (CC1, CPD1). The asymmetry in the
nerve ring is well marked (CPR3). The left
parietal ganglion is closer to the visceral than
to the left pleural ganglion (PAG2). The vis-
ceral ganglion, which lies on the right side
(VG1), is appressed to the right parietal gan-
glion (FG2).

Considering that the characters used as
characters of a CCA are here those of a spe-
cies actually studied, and not of a recon-

structed taxon, the number of plesiomorphic
states is particularly high: BM1 OC1 SC1 PS1
11 LR2’ UR1 RR1 СС1 CPD2 CPR3 PLD1
PLG1 VG1 PAD1 PAG2 FG2.

Amastridae (Figs. 34-42): The Amastridae
are endemic to the Hawaiian Islands. The
family includes four to 12 genera depending
on the authors (Boss, 1982; Zilch, 1959—
1960), and is known from the Pleistocene of
its modern distribution area only.

In the phenetic classification of the species,
the Amastridae are close to Cochlicopa (the
generally accepted position). They are farther
away in the phenetic classification of the fam-
ilies, mainly owing to their larger size (Text-
fig. 20).

All characters of the digestive tract are
plesiomorphic, as in Cochlicopa. The cere-
bral commissure and lateral connectives are
medium in length (CC2, CPD2). The left
cerebro-pedal connective is longer than the
right one (CPR3, CPR4). The pleural ganglia
are near the pedal ganglia (PLG1, PLD1).
The position of the visceral ganglion varies
(VG1, VG3). The left parietal ganglion is
appressed to the visceral ganglion (PAG3),
and also touches the left pleural ganglion in
Amastra (PAG4). The visceral ganglion
seems fused with the right parietal ganglion
(FE2):

Character states of the CCA: BM1 OC1
SC1 PS1 IL1 LR2 URI RAI CC24EPD2
СРАЗ PLD1 PLG1 VG1 PAD1 PAG31EG2

Vertiginidae (Figs. 52-55): The Vertigi-
nidae range in Europe from the Paleocene to
the Recent, and are now nearly ubiquitous.
The two representatives of the family here
dissected are closer to each other than to any
other species in the phenetic classification of
the species (Text-fig. 19). Together they are
phenetically close to the Cochlicopidae and
Valloniidae (Text-fig. 20).

All characters of the digestive tract are ple-
siomorphic. The cerebral commissure is short
(CC3). The cerebro-pedal connectives are
medium in length (CPD2), the right one being
longer than the left (CPR3). The pleural gan-
glia are closer to the cerebral ganglia than to
the pedal ganglia (PLD2, PLG2). The visceral
ganglion is on the right side of the median
plane (VG1). The left parietal ganglion is
closer to the visceral ganglion than to the left
pleural ganglion (PAG2), and the right parietal
ganglion seems fused with the visceral gan-
glion (FG2).

STYLOMMATOPHORAN SYSTEMATICS 55

Character states of the CCA: BM1 OC1
SC1 PS1 11 LR2’ UR1 RR1 ССЗ CPD2
CPR3 PLD2 PLG2 VG1 PAD1 PAG2 FG2.

Orculidae (Figs. 56-66): The family is
known in Europe from the Paleocene, and
now occurs eastward and southward to the
Caucasus, Asia Minor and Ethiopia. In the
phenetic classification of the families, it is cu-
riously close to the Cerionidae, but also, and
more expectedly, to the Chondrinidae, Pupil-
lidae and Pyramidulidae (Text-fig. 20).

Pagodulina has an oesophageal crop asso-
ciated with a cylindrical gastric crop, whereas
Orcula has no oesophageal crop but a gastric
crop inflated in its median portion. The kidney
is relatively shorter in Pagodulina than in any
other orthurethran dissected; but, as previ-
ously discussed, this relatively short length
probably results from secondary lengthening
of the distal region of the lung, rather than
from plesiomorphy. In the two genera dis-
sected the cerebral commissure is short
(CC3). The cerebro-pedal connectives are
medium in length (CPD2), the right one being
longer than the left (CPR3). The pleural gan-
glia are close to the pedal ganglia in Orcula
(PLD1, PLG1), but close to the cerebral gan-
glia in Pagodulina (PLD2, PLG2). The vis-
ceral ganglion is median (VG2). The left pari-
etal ganglion is appressed to the visceral
ganglion (PAG3), which seems fused with the
right parietal ganglion (FG2).

Character states of the CCA: BM1 OC1
ЭРУ ЕТ LR2 ВТ АВТ ‘ССЗ. PLD
PLG1 VG2 PAD1 PAG3 FG2.

Partulidae (Figs. 25-33): The family Partu-
lidae is endemic to Polynesia and Micronesia;
no fossil is known. Following Pilsbry (1900a),
authors have insisted on the primitiveness of
the Partulidae, some even doubting their
orthurethran affinities (Boss, 1982). Such
Opinions are not supported by any character
here used or, as far as | know, by any genital
character. However, one can understand the
doubts raised by these authors in view of the
phenetic classification of the species (Text-
fig. 19), which shows both the homogeneity
and the distinctness of the family. However,
the Partulidae should be considered Orthure-
thra because: (1) their pallial complex has ev-
ery orthurethran character state, including the
great length of the kidney, which is here in-
terpreted as a synapomorphy of the Orthure-
thra; (2) in the phenetic classification of the
families (Text-fig. 20), they are closer to the

Amastridae and Enidae than to any other sty-
lommatophoran family.

In two of the three genera studied, Eua and
Partula, the digestive tract is unique in having
a gastric caecum. Eua also has a fleshy gas-
tric appendix. As discussed above, | do not
know whether these characters are apomor-
phic or plesiomorphic. The gastric crop is in-
flated in its median portion (SC2). The cere-
bral commissure is not very long in any of the
three genera. The cerebro-pedal connectives
are never very long (CPD1 absent), and the
right one is always shorter than the left one
(absence of CPR1). The pleural ganglia are
close to the pedal ganglia (PLG1, PLD1). The
visceral ganglion is median (VG2). The left
parietal ganglion touches either the visceral
ganglion only, or the visceral and the left pleu-
ral ganglia (PAG3, PAG4). The visceral gan-
glion seems fused with the right parietal
(FG2):

Character states of the CCA: BM1 OC1
SC2 PS1 IL1 LR2’ UR1 RR1 СС2 CPD2 CPR
2PEDINPEGIIMG2\PADINPAGS'EC?:

Enidae (Figs. 108-146): The Enidae are
mainly Palearctic. The subfamily Cerastuinae
is principally East African, ranging from Ara-
bia to South Africa; but it includes a few Ori-
ental genera, and Amimopina in Queensland
and Draparnaudia in New Caledonia and
Vanuatu. The cerastuine affinities of the old-
est genus known, Procerastus from the Euro-
pean Paleocene, are considered dubious by
Zilch (1960). No other enid genus is known to
be older than Miocene.

In the phenetic classification of the families
(Text-fig. 20), the Enidae are closer to the
Amastridae and Partulidae than to any other
orthurethran family. This similarity may be
attributed largely to the relatively large size of
the animals in these three families. In the
phenetic classification of the species (Text-
fig. 19), the heterogeneity of the family as a
whole and the homogeneity of the Cerastu-
inae on the one hand, and of the other
Enidae on the other, is evident. Mordan
(1984) cast doubt on the monophyly of the
group. All characters here studied have either
the same state in both subgroups, or are
more apomorphic in the Cerastuinae. The
pulmonary characters interpreted by Mordan
as possible indications of paraphyly are those
that were interpreted above as related to
visceral mass shortening, which is common
among the Stylommatophora although ex-
ceptional in the Orthurethra. These charac-

56 TILLIER

ters might constitute a synapomorphy of the
subfamily, but cannot be used to reject the
monophyly of the family.

An oesophageal crop is present in Rachis-
tia, where it may related to the obvious short-
ening of the visceral mass (OC2). The gastric
crop is inflated in its median portion in Chon-
drula, and separated from the gastric pouch
by a constriction in this genus and in Zebrina
(elongate shape). The intestine is short (IL1,
IL2).

The Cerastuinae are unique in their general
tendency to the formation of a ureteric tube,
produced in part in Rachistia and Amimopina
(UR2).

The cerebral commissure and the lateral
connectives are never very long (CC1 and
CPD1 absent). The left cerebro-pedal con-
nective is always longer than the right one
(CPR2, CPR3). The left pleural ganglion is
closer to the left pedal ganglion (PLG1) in all
the genera studied, whereas the right one is
closer to the right cerebral ganglion in Chon-
drula and Ena. The left parietal ganglion al-
ways touches both the left pleural and the vis-
ceral ganglia (PAG4). In some genera the
visceral ganglion is distinct from the right pa-
rietal ganglion (FG1).

Character states of the CCA: BM1 OC1
SEMHPSTAIETTER2TZ ARI CC2 CPD2 CPR2
PLD1 PLG1 VG1 PAD1 PAG4 FG1.

Non-orthurethran families

The non-orthurethran families show no syn-
apomorphy among the characters described
above: their hypothetical common closest an-
cestor (CCA) would be plesiomorphic for all
characters. For convenience in presentation,
families are here grouped in superfamilies
that cannot be justified from familial descrip-
tions, but only from the phylogenetic analysis
presented below and resulting in the recogni-
tion of two suborders, the Dolichonephra and
the Brachynephra.

Zonitoidea: Zonitidae (Figs. 235-258, 269-
277): Some fossil shells from the Cretaceous
of North America are generally thought prob-
ably to belong to the Zonitidae (Zilch, 1959—
1960). In Europe the Zonitidae range from the
Paleocene to the Recent. The family is now
principally North American and circum-Medi-
terranean. The subfamily Godwininae occurs
in the Hawaiian Islands (Baker, 1941), and
the genus Zonitoides occurs throughout the
Northern Hemisphere (introduced?).

The affinities of the Gastrodontinae with the
Zonitidae are problematic; | have already un-
derlined their similarity to the Systrophiidae in
genital anatomy (Tillier, 1980). Furthermore,
the gastrodontine Ventridens has a character
found nowhere else but in the Systrophiidae,
i.e. the position of the anterior duct of the di-
gestive gland distinctly in front of the concave
angle of the gastric pouch. The Gastrodon-
tinae have the left parietal ganglion closer to
the left pleural than to the visceral ganglion,
unlike all other Zonitidae (PAG1, plesiomor-
phic); in Gastrodonta the right pleural gan-
glion is close to the right cerebral ganglion, as
in systrophiids but unlike any Zonitidae exam-
ined (PLD2, apomorphic). With doubt | retain
the Gastrodontinae in the Zonitidae because
their CCA is closer to the CCA of the Zoniti-
dae than to the CCA of the Systrophiidae.

In the phenetic classification of the families,
the Zonitidae are close to the other unlima-
cized aulacopod families (Text-fig. 20).

The Zonitidae tend to be carnivorous, but
not to such an extent as to have a cylindrical
buccal mass, except in the daudebardiine
slugs (BM1, BM2). There is no oesophageal
crop (OC1). Gastric crop shape (SC1, SC2,
SC3, SC2’), stomach shape (PS1, PS2,
PS2') and intestinal length (IL1 in the Gas-
trodontinae, IL2, IL3) are much more variable.
The kidney is rather long (LR1, LR2; LR3 in
the Gastrodontinae). The ureteric tube always
extends as far as the pneumostome (UR4).
The cerebral commissure is short (CC3), ex-
cept in the Daudebardiinae, in which the oc-
currence of a long cerebral commissure
(CC1) might be related to limacization and
carnivory. The cerebro-pedal connectives are
always long (CPD1), but the ratio of their

‘lengths varies (CPR1, CPR2, CPR3; CPR4 in

Gastrodonta). The visceral ganglion is on the
right side (VG1), or median (VG2). The posi-
tion of the right parietal ganglion varies
(PAD1, PAD2). The left parietal ganglion is
close to the left pleural ganglion in the Gas-
trodontinae (PAG1), but close to the visceral
ganglion in the other Zonitidae (PAG2). The
visceral ganglion might seem distinct (FG1) or
fused with the right parietal ganglion (FG2).

When the Gastrodontinae are included in
the Zonitidae, all the characters of the CCA
but the occurrence of a ureteric tube are ple-
siomorphic: BM1 OC1 SC1 PS1 IL1 LR1 UR4
RRI CCi CRDi CPR PEDI-PEGI МЕ
PAD? РАСТ ЕЕТ.

Such a number of symplesiomorphies might
result from para- or polyphyly of the Zoni-

STYLOMMATOPHORAN SYSTEMATICS 57

tidae, which really need a revision embracing
the other zonitoid families. | suspect that, in its
present definition, the family comprises sev-
eral monophyletic groups, of which the sister
groups might be among the other zonitoid
families. This is why such small groups as the
Limacidae, Milacidae, and Parmacellidae are
here retained at the familial level.

Zonitoidea: Trochomorphidae (Figs. 278—
282): Representatives of the family occur in
Southeast Asia and the Western Pacific (from
Japan to Vanuatu), and eastward to the So-
ciety Islands (Solem, 1959). Only one species,
Trochomorpha, was dissected. It is pheneti-
cally close tothe Zonitidae, which corresponds
to the position assigned by Baker (1941), who
considered the Trochomorphinae a subfamily
of the Zonitidae (Text-fig. 20).

Trochomorpha is not carnivorous (BM1)
and lacks an oesophageal crop (OC1). The
gastric crop widens from oesophagus to
stomach (SC2’) and the gastric pouch is dif-
ferentiated (PS1). The intestine is short (IL2).
The kidney is not very long (LR2), and is in-
ternally divided into two distinct regions
(RR1). The ureteric tube is closed as far as
the pneumostome (UR4). The cerebral com-
missure and the lateral connectives are short
(CC3, CPD3), the latter being subequal in
length (CPR2). Both pleural ganglia are
closer to the cerebral ganglia than to the
pedal ganglia (PLD2, PLG2). The visceral
chain is compact (PAD2, PAG4), with the vis-
ceral ganglion in a median position (VG2) and
distinct from the parietal ganglia (FG1).

Character states of the CCA (= character
states of Trochomorpha): BM1 OC1 SC2'
PS1 IL2 LR2 UR4 RR1 ССЗ CPD3 CPR2
PLD2 PLG2 VG2 PAD2 PAG4 FG1.

Zonitoidea: Euconulidae (Figs. 280, 283—
295): The genus Euconulus is reported with
doubt from the Upper Cretaceous, and with
more certainty from the European and North
American Paleocene (Zilch, 1959-1960).
The present distribution of the Euconulidae is
nearly world-wide, with maximum diversity in
the Central and West Pacific and quasi-
absence in the Australian region (Baker,
1938-1941).

In the phenetic classification of the families,
the Euconulidae are close to the endodon-
toids sensu Solem, and to the Zonitidae (Text-
fig. 20).

The Euconulidae are not carnivorous
(BM1), and lack an oesophageal crop (OC1).

The gastric crop is cylindrical (SC1), or nar-
rows from oesophagus to stomach (SC3), the
latter being differentiated (PS1) or not (PS2).
The intestine is short to very short (IL2, IL3).
The kidney is rather short (LR2, LR3), but a
relatively long length can occur in relation to
visceral mass shortening (LR1 in semislugs).
A big renal lamella bearing transverse folds
runs along and inside the rectal side of the
kidney. The ureteric tube is closed as far as
the pneumostome (UR4). The cerebral com-
missure is short (CC3). The lateral connec-
tives are rather short (CPD2, CPD3) and
shorter on the left side than on the right when
asymmetric (CPR1). The right pleural gan-
glion is near the right cerebral ganglion
(PLD2), whereas the position of the left pleu-
ral ganglion varies (PLD1, PLD2). The vis-
ceral ganglion is on the right side of the me-
dian plane (VG1). The left parietal ganglion is
close to the left pleural ganglion (PAG1),
whereas the right parietal ganglion seems
fused with the viscera! ganglion (PAD2, FG2).
Character states of the CCA: BM1 OC1
SCIMPSIMIL2RER2AURAVRRIACCS CPD2
CPR1 PLD2 PLG1 VG1 PAD2 PAG1 FG2.

Zonitoidea: Discidae (Figs. 184-194): To
the extent that one can accept as criteria for
familial attribution shell characters of which
the direction of evolution is not known, the
Discidae are known in the North American
Carboniferous (Solem 4 Yochelson, 1979);
other fossils attributable to the family are not
known older than Paleocene in the same re-
gion and in Europe, where they still live.

Helicodiscus is here considered a discid,
because: first, there are fewer differences be-
tween Discus and Helicodiscus than between
some genera considered as Discidae sensu
stricto, e.g. Discus and Anguispira; and sec-
ond, Helicodiscus is closer to Discus than to
any other genus of the Endodontoidea sensu
Solem in the phenetic classification of the
species dissected (Text-fig. 19). The Oriental
Stenopylis, placed by Solem (1984) in his He-
licodiscidae, is not necessarily involved: in my
opinion, Stenopylis is an endodontoid by my
definition (i.e. Solem's Endodontoidea, with-
out the Discidae but with the Systrophiidae
and Athoracophoridae), because its kidney is
short and its ureter is partly open (UR2).

The Discidae, including Helicodiscus, are
here considered zonitoids because they are
much closer to the zonitids than to any endo-
dontoid family in both the phenetic and phy-
logenetic classifications (Text-fig. 20, and

58 TILLIER

below). This position might be wrong in vari-
ous ways: the Helicodiscidae might be an en-
dodontoid family, distinct from the Discidae,
which are zonitoids; all the Discidae might
be endodontoids. The North American An-
guispira (Discidae) looks much more like the
South American Stephanoda, which belongs
to the Charopidae according to Solem (1982),
than like any other of the Discidae or Zoniti-
dae (Text-fig. 19).

The Discidae are not carnivorous (BM1)
and lack an oesophageal crop (OC1). The
gastric crop either is cylindrical (SC1) or wid-
ens from the oesophagus to the stomach
(SC2'), the latter being differentiated (PS1).
The intestine is rather short (IL2), except in
Anguispira, in which its greater length (IL1)
might be an effect of size increase. The kid-
ney is also rather short (LR2, LR3). The inter-
nal kidney lamellae are little developed on the
palatal surface of the kidney, whereas a very
large lamella runs along the rectal border of
the kidney. The ureter is closed as far as the
pneumostome (UR4). The cerebral commis-
sure is short (CC3) and the lateral connec-
tives are long (CPD1). The left cerebro-pedal
connective is shorter than the right one
(CPR1). The pleural ganglia are both closer to
the pedal ganglia than to the cerebral ganglia
(PLD1, PLG1), but the right one is farther
from the right pedal ganglion than the left one
is from the left pedal ganglion. The visceral
ganglion is on the right side of the median
plane (VG1). The right parietal ganglion
touches the visceral ganglion, both seeming
fused in Anguispira. The left parietal ganglion
is closer to the left pleural than to the visceral
ganglion (PAG1).

Character states of the CCA: BM1 OC1
SC1 PS1 IL2 LR2 UR4 RR1 CC3 CPD1
CPRIPEDIPLGI1 VGTPAD2 PAG1FGi:

Zonitoidea: Апотаае (+ Philomycidae)
(Figs. 195-215, 217): The Arionidae and the
Philomycinae share a synapomorphy, the an-
nular kidney. The Philomycinae are distinct
only in their more advanced slug condition,
which results in a more compact nervous sys-
tem among the characters used here. | can
see no reason for separating them from the
Arionidae at the familial level. They are North
American and Oriental, whereas the other Ari-
onidae are Holarctic and South African
(Oopelta)

The South African genus Oopelta must be

icluded in the Arionidae because it has a ring

idney, and resembles more the Arionidae

than any other family even when this charac-
ter state is disregarded (Text-fig. 19). This po-
sition is not so aberrant as one might suppose
from a biogeographical point of view: several
families (Enidae, Cerastuinae, Helicidae)
seem to have invaded Africa from its north-
east extremity, probably during the Miocene
(v. infra), the epoch during which the family
Arionidae appeared in the fossil record in Eu-
rope and in North America (Zilch, 1959—
1960). The Arionidae have probably followed
the same track, but became extinct in East
Africa. Phenetically the Arionidae are close to
the other aulacopod slugs (Text-fig. 20).

The Arionidae are not specialized in car-
nivory (BM1). The gastric crop is variable in
shape (SC1, SC2, SC2’, SC3). The gastric
pouch is differentiated (PS1), except in the
semislug Hemphillia (PS2). The intestine is
variable in length, but never very short (IL1,
IL2, 112’). The kidney is annular, with a ure-
teric tube reaching the pneumostome (UR4);
its internal morphology has been described
above. The cerebral commissure is long in
Oopelta (CC1), and shorter in the Northern
Hemisphere taxa (CC2). The cerebro-pedal
connectives are medium to short in length
(CPD2, CPD3). They are subequal in length,
or the right one is longer than the left (CPR1,
CPR2). The pleural ganglia are closer to the
pedal ganglia than to the cerebral ganglia
(PLD1, PLG1). The visceral ganglion is on the
right side of the median plane (VG1) and
seems fused either with the left parietal gan-
glion only (FG2', Oopelta) or with both pari-
etal ganglia (FG3).

Character states of the CCA: BM1 OC1
SC1 PS1 11 LR5 UR4 RR2 (RRIAACOA
CPD2 CPR1 PLD1 PLG1 VG1 РАО? PAG3

`ЕЕ2..

Zonitoidea: Parmacellidae (Figs. 259-
262): The oldest fossils of this circum-Med-
iterranean family of slugs occur in the Euro-
pean Eocene. The species dissected here are
phenetically close to the Limacidae and Mi-
lacidae (Text-figs. 19, 20).

The Parmacellidae are not carnivorous
(BM1). The gastric crop is cylindrical (SC1)
and the gastric pouch is dedifferentiated
(PS2). The intestine is rather short (IL2). The
kidney is compact, not internally differentiated
(RR2), and the ureteric tube reaches the
pneumostome (UR4). The cerebral commis-
sure and the lateral connectives are long
(CC1, CPD1), the latter being subequal in
length (CPR2). The pleural ganglia are close

STYLOMMATOPHORAN SYSTEMATICS 59

to the pedal ganglia (PLD1, PLG1). The vis-
ceral chain is compact (PAD2, PAGA), but the
visceral ganglion is distinct from the parietal
ganglia (FG1) and median (VG2). Although
the visceral chain is compact, its arrangement
suggests derivation from a visceral chain in
which the left parietal ganglion was closer to
the left pleural than to the visceral ganglion.
Characters of the CCA: BM1 OC1 SC1 PS2
IL2 LR5 UR4 RR2 CC1 CPD1 CPR2 PLD1
PLG1 VG2 PAD2 PAG4 (РАС1?) FG1.

Zonitoidea: Limacidae (Figs. 263, 265-—
266): Limacid shells occur in the European
Oligocene (Zilch, 1959—1960), and the family
is now circum-Mediterranean (if one admits
that Agriolimax = Deroceras has been intro-
duced in North America as it was elsewhere).

The only species dissected, Limax maxi-
mus, differs from Parmacella in the shape of
its gastric crop (SC2’), its very long intestine
(IL2’: size effect?) and its shorter cerebral
commissure and lateral connectives (CC3,
CPD2).

Character states of the CCA (= character
states of Limax maximus): BM1 OC1 SC2'
PS1 IL2’ LR5 UR4 RR2 ССЗ CPD2 CPR2
PLD1 PLG1 VG2 PAD2 PAG4 FG1.

Zonitoidea: Milacidae (Figs. 264, 267-
268): Milacid shells occur in the European
Upper Eocene (Zilch, 1959-1960). The family
includes only four genera, two of which are
fossil, and is circum-Mediterranean.

The only significant difference allowing
separation of Milax from Limax and Parma-
cella is among characters in the visceral
chain, in which the visceral ganglion is on the
left side of the median plane (VG3), and
seems fused with the left parietal ganglion
(FG2'). These differences might indicate dif-
ferent ancestral snails, but might also be due
to the larger size of the semislug ancestors of
Milax.

Character states of the CCA (= character
states of Milax): BM1 ОС1 SC2 PS2 IL2’ LR5
UR4 RR2 CC3 CPD2 CPR1 PLD1 PLG1 VG3
PAD2 PAG4 FG2”.

Helicoidea: Helicidae (Figs. 640-704): The
Helicidae are known with certainty from the
European Oligocene. Zilch (1959-1960) at-
tributes with doubt a few Lower Eocene gen-
era to the family. At present the Helicidae are
principally European and circum-Mediter-
ranean. The genus Halolimnohelix, from east-
ern Africa, belongs in the Helicidae, not in the

Bradybaenidae: Halolimnohelix (H.) sericata
lejeuneihas finger-shaped multifid glands sim-
ilar to those classically used as a synapomor-
phy of the Helicidae. Seemingly no known Af-
rican helicoid has an amatorial organ formed
by a muscular trunk terminated by a sagittiform
glandular formation, as occurs in the Brady-
baenidae (e.g. Pilsbry, 1919).

As might be expected, the Helicidae are
phenetically close to the Camaenidae, Brady-
baenidae, Helminthoglyptidae and Polygy-
ridae (classical position); but less expectedly,
they are also close to the Achatinidae and
Bulimulidae (Text-figs. 19, 20).

Among the character states used here only
the absence of internal differentiation of the
kidney, the position of the contact of the left
parietal ganglion with the visceral ganglion
and the closure of the ureteric tube at least as
far as the recto-renal angle are synapomor-
phous for all helicids (RR2, PAG4, UR3). The
helicids are never carnivorous (BM1), and of-
ten have an oesophageal crop (OC2, ОСЗ,
OC4). The kidney is never very short (LR1,
LR2, LR3). The pleural ganglia are near the
pedal ganglia (PLD1, PLG1). The visceral
ganglion is generally either median or on the
left side (VG2, VG3).

Character states of the CCA: BM1 OC1
SCAIMPSTMESERIRURSIRR2ICGCINEPDI
CPR1 PLD1 PLG1 VG1 PAD1 PAG3 FG1.

Although classically considered “the most
highly organized Helices upon the globe,”
probably because they are the dominant fam-
ily in Western Europe (Taylor, 1900), the
Helicidae have a greater number of plesio-
morphic characters than nearly any other
non-orthurethran family. As in the case of the
Zonitidae, this might result from para- or poly-
phyly. Among those genera dissected here
two, Helicodonta and Cochlicella, show such
a morphology of the amatorial organ that no
one would hesitate to classify them among
the helminthoglyptids, if they occured in North
America (Germain, 1930; Pilsbry, 1940).
Within the family classification is at best con-
fused. The only contribution possible here is
the proposal to restrict the Helicellinae, of
which the monophyly has been long con-
tested (Watson, 1922; Shileyko, 1978c), to
genera having a ureteric appendage along
the rectum, as described above.

Helicoidea: Helminthoglyptidae (Figs. 607—
639): The oldest genera that Zilch (1959—
1960) attributes to the helminthoglyptids oc-
cur in the Cretaceous (Mesoglypterpes) and

60 TILLIER

the Upper Paleocene (Glypterpes) of North
America. Representatives of the family live
principally in western North America, Mexico
and Central America. The genus Epiphrag-
mopora extends southward in western South
America to Argentina.

The Central American and Caribbean Thy-
sanophorinae are here included in the Hel-
minthoglyptidae, not in the Polygyridae, for
two reasons. First, Thysanophora is pheneti-
cally closer to some Helicidae than to any Poly-
gyridae (the distinction between Helicidae
and Helminthoglyptidae is often impossible
without geographical data, as discussed
above and below). Second, the ureteric tube
is not closed as far as the pneumostome in
Thysanophora, as seen in about half of the
dissected Helminthoglyptidae but in none of
the Polygyridae (Fig. 610).

The tropical South American genus Psa-
dara is included in the Cepolinae (! formerly
supported its classical position in the Ca-
maenidae, Tillier, 1980) for two reasons. First,
Psadara is phenetically closer to Thysano-
phora and to some Helicidae than to any of
the Camaenidae (Text-fig. 19). The arrange-
ment of organs in Cepolis maynardi looks
very much like that in Psadara (Figs. 621,
622). Second, both Psadara and Cepolis
have a kidney longer than that in any other
non-limacized non-orthurethran snail (syn-
apomorphy; Figs. 616, 618).

The Cepolinae and Thysanophorinae might
form a single subfamily (or even a family dis-
tinct from the Helminthoglyptidae, the inclu-
sion of the Cepolinae in the latter being not
obvious: Pilsbry, 1940). However, | consider
that better knowledge of these groups than
mine is necessary to support this unity.

Sonorella is closer to Epiphragmopora than
to any other helicoid in the phenetic classifi-
cation of species (Text-fig. 19). This similarity
might justify the inclusion of the Sonorellinae
in the Helminthoglyptidae, about which Pils-
bry himself (1940) was unsure. On the other
hand, the Helminthoglyptidae are closer to
the Polygyridae than to any other family ofthe
Helicacoidea sensu stricto in the phenetic
classification of the families (Text-fig. 20), and
it would be worth testing the monophyly of the
Polygyridae and of the Sonorellinae (Helica-
coidea sensu stricto = Helicidae, Helmintho-
glyptidae, Bradybaenidae).

The Helminthoglyptidae and Helicidae are

nguished by remarkably few character
states: in the former the cerebral commissure
always very short (not always so in the He-

licidae); the kidney is internally differentiated
in at least a few genera; the left parietal gan-
glion always touches the visceral ganglion (it
does not in a single dissected helicid genus,
Sphincterochila). Perhaps these differences,
of which none is absolute (no synapomorphy
of either of the families), would not resist fur-
ther sampling, which might possibly reveal
amphi-Atlantic monophyletic groups.
Character states of the CCA: BM1 OC1
SC1 PS1 111 ERI URS ВАТ CE3-CRDA
CPR1 PLD1 PLG1 VG2"PAD2PAG4:FGif

Helicoidea: Bradybaenidae (Figs. 587-
594): The Bradybaenidae can be defined by
the morphology of their amatorial organ,
formed by a muscular trunk, issuing from the
vagina, on which numerous acini forming a
sagittiform apical mass are inserted. The
same structure occurs in the Dyakiinae (He-
licarionidae?); this point will be discussed fur-
ther. This definition excludes the African
Halolimnohelix from the family, as discussed
above.

When so defined the Bradybaenidae are
principally Oriental and rarefy westward, only
the genus Bradybaena reaching Western Eu-
rope. Fossils are known from the Chinese
Pliocene (Cathaica); the purely anatomical
definition of this family makes its earlier oc-
currence possible, and some of the fossil he-
licoids might belong to the Bradybaenidae.

In the phenetic classifications of the spe-
cies and families, the Bradybaenidae are
close to other helicoids (Text-figs. 19, 20).

No bradybaenid is carnivorous (BM1). The
two genera dissected have a large oesoph-
ageal crop (ОСЗ, OC4), which is either cylin-
drical or inflated in its median portion (SC1,

- SC2; the latter state is an effect of shape in

Helicostylus). The intestine is long (IL1). The
kidney is long to very long (LR1, LR2’), such
considerable relative lengths being possibly
an effect of visceral mass and lung shorten-
ing. The ureteric tube is closed as far as the
pneumostome (UR4). The kidney is internally
differentiated (RR1). The cerebral commis-
sure is very short (CC3). The cerebro-pedal
connectives are rather long (CPD1, CPD2)
and subequal in length (CPR2). The pleural
ganglia are close to the pedal ganglia (PLD1,
PLG1). The visceral ganglion is either median
or on the left side (VG2, VG3). The right pa-
rietal ganglion touches the visceral ganglion
(PAD2), and the left parietal ganglion is in
contact with both the visceral and the left
pleural ganglia (PAG4). Apparent fusion of

STYLOMMATOPHORAN SYSTEMATICS 61

the left parietal and visceral ganglia occurs in
one case.

Character states of the CCA: BM1 OC3
561 Р51 JAERI) UR4 ВАТ ССЗ CPD1
CPR2 PLD1 PLG1 VG2 РАО? PAG4 FG1.

Helicoidea: Polygyridae (Figs. 577-586):

The Polygyridae occur in the Miocene of
North America, where most genera are still
endemic; the Cretaceous fossil attributed to
the Polygyridae by Zilch (1959-1960) be-
longs to the Ammonitellinae (Solem, т litt.).
The North American genus Polygyra extends
to Cuba in the West Indies (Boss, 1982).

The species dissected here are dispersed,
as are the other helicoids, in the phenetic
classification of the species (Text-fig. 19), and
the family is phenetically close to the other
helicoid families (Text-fig. 20).

No polygyrid is carnivorous (BM1). An oe-
sophageal crop might be present (OC1, OC2,
OC3). The shape of the gastric crop and gas-
tric pouch is variable, but the gastric pouch is
never reduced (PS2 absent). The lengths of
the intestine and of the kidney are variable
(IL2, 12’, LR1, LR2, LR3). The ureteric tube
is always closed as far as the pneumostome
(UR4). There are two morphological regions
in the kidney (RR1). The cerebral commis-
sure is short (CC3). The cerebro-pedal con-
nectives vary in length, but the right one is
longer than the left one in the species dis-
sected (CPR3). The pleural ganglia are close
to the pedal ganglia (PLD1, PLG1). The vis-
ceral ganglion is median or on the left side
(VG2, VG3). The right parietal ganglion
touches the visceral ganglion (PAD2). The left
parietal ganglion seems fused with the vis-
ceral ganglion, and touches the left pleural
ganglion (PAGA).

Character states of the CCA: BM1 OC1
ЭР ВТ UR4. ВАТ ССЗ. СРВ
CPR2 PLD1 PLG1 VG2 РАО? PAG4 FG2’.

Helicoidea: Camaenidae (Figs. 556-576):

The Camaenidae are classically defined
among the helicoids s.s. by the absence of an
amatorial organ. Even if this absence is apo-
morphic, which 1$ likely, it is also shown by the
Polygyridae and Sonorellinae in the same
group. The definition and classification of the
Camaenidae are thus far from being secure.
In its present definition, the family seems to
occur in the Cretaceous of North America
(Hodopeus) to the extent that shell characters
are reliable in the helicoids. It is now West
Indian and Oriental, with representatives

reaching Australia on one side and possibly
South America on the other side (v. infra).

The Oriental genus Plectotropis is here
placed among the Camaenidae, as sug-
gested by Solem (1966a) on a conchological
basis: Plectotropis goniocheila has no amato-
rial organ.

The position of the American camaenids,
Pleurodonte and the related Antillean genera
on one hand, and the South American genera
Solaropsis, Labyrinthus and Isomeria on the
other hand, is far more problematic. First, So-
laropsis and Labyrinthus are very similar in
the general arrangement of their organs
(Figs. 556, 559), and in the aspect of living
animals (Tillier, 1980). Both genera have a
character unique among the Stylommato-
phora dissected, ¡.e. the typhlosole that is-
sues from the anterior duct of the digestive
gland is more developed than the one issuing
from the posterior duct. Both genera live in
the Amazonian region, but occupy well-
defined and different niches (Tillier, 1980),
and convergence is unlikely. Secondly, the
lung and kidney of Solaropsis have a peculiar
arrangement, which reminds one more of the
Acavidae than of the Camaenidae (Tillier,
1980). The same organs are more camaenid-
like in Labyrinthus, although the kidney is
longer than in any camaenid, as in Solarop-
sis. Thirdly, Labyrinthus is very similar to
Pleurodonte, although the latter does not
show the same morphology of the typhlo-
soles, and Pleurodonte is itself very similar to
the Australian camaenid Sinumelon (not de-
picted here); the genital anatomy and its pat-
tern of variation are similar in these two gen-
era (Solem, pers. comm.). Even if we sup-
pose that the group of Solaropsis does not
belong to the Camaenidae, the position of the
Pleurodonte group remains problematic ow-
ing to its similarity to Labyrinthus. This is why
the family is here preserved in its larger def-
inition, which can easily be justified by the
absence of synapomorphy whatever solution
is adopted. However, the hypothesis that the
Pleurodonte group belongs to, or is directly
related to the Cepolinae, Thysanophorinae or
Sonorellinae should be carefully evaluated.

The Camaenidae are not carnivorous
(BM1). All species dissected have an oesoph-
ageal crop (OC2, ОСЗ, OC4). The gastric
crop is cylindrical (SC1), except in Sinumelon
where its widening toward the stomach
(SC2') might be related to its very short length.
The intestine varies in length (IL1, IL2, IL2”),
but is never very short. The kidney is rather

62 TILLIER

long (LR1, LR2). The ureteric tube is closed
as far as the pneumostome (UR4), except in
Solaropsis (UR3). The kidney has two inter-
nal regions in all the Oriental camaenids dis-
sected (RR1), but in no Neotropical one
(RR2). The cerebral commissure is short
(CC3). The cerebro-pedal connectives are
medium in length (CPD2). The pleural ganglia
are close to the pedal ganglia (PLD1, PLG1).
The position of the visceral ganglion varies
(median or on the left in all Oriental ca-
maenids: VG2, VG3). The visceral chain is
compact, and the visceral ganglion might look
fused with the left parietal ganglion (PAD2,
РАСА, FG1 or FG2”).

Character states of the CCA: BM1 OC2
SC1 PS1 IL1 LR1 ЦВЗ (UR4 if Solaropsis is
omitted) RR1 (RR2 for the American ca-
maenids) ССЗ CPD2 CPR1 PLD1 PLG1 VG1
(VG2 in Oriental camaenids) FG1 (FG3 for
American camaenids).

Helicoidea: Sagdidae (Figs. 600-606):
Baker (1962) moved the Sagdidae, which oc-
cur in the Greater Antilles and in southeastern
North America, from the Polygyracea to the
Achatinacea. His arguments are based on the
geographical distribution, the carnivorous diet
and the incompletely closed ureteric tube,
features which unite, in his opinion, the Sag-
didae and the Spiraxidae, which he considers
as belonging to the achatinoids. The Spiraxi-
dae are here considered oleacinids (v. infra).
The Sagdidae are closer to the helicoids than
to the oleacinids in the phenetic classification
of the species (Text-fig. 19), whereas they are
closer to the oleacinids in the phenetic clas-
sification of the families (Text-fig. 20). Three
further pieces of evidence bear on these
points. First, the Oleacinidae and the Spirax-
inae have a short kidney (apomorphic) and a
visceral ganglion distinct from the parietal
ganglia (plesiomorphic). Second, the Sag-
didae have a long kidney (plesiomorphic; ex-
cept in Sagda, in which the relatively short
length of the kidney is related to the length-
ening of all other organs, including the lung
roof) and a visceral ganglion that seems
fused with the left parietal ganglion (apomor-
phic). Third, the lateral penial appendix of the
Sagdidae closely resembles the amatorial or-
gan of the Helminthoglyptidae, and | suspect
that it represents the same organ, but shifted
from the vagina to the penis.

Even if the amatorial organ is plesiomor-
Mic in the Stylommatophora, or if the penial
appendix of the Sagdidae is not homologous

with the amatorial organ of the Helmintho-
glyptidae, there is one synapomorphy with the
latter and no synapomorphy with the olea-
cinids. Consequently, it might be more rea-
sonable to include the Sagdidae in the heli-
coids beside the Polygyridae and the
Helminthoglyptidae (v. infra) rather than in the
achatinoids, which include the oleacinids in
the classification presented here, beside the
Spiraxinae as proposed by Baker.

Although the Sagdidae are carnivorous,
their buccal mass is not particularly length-
ened (BM1). They do not have an oesoph-
ageal crop (OC1). The gastric crop is inflated
in its median portion (SC2) or widens toward
the stomach (SC3). The intestine is short
(IL2) and the kidney is long (LR1), except in
Sagda where a long intestine and a short kid-
ney result from visceral mass lengthening
(lung three whorls long). The ureteric tube is
closed as far as the pneumostome (UR4). In
Sagda it is prolonged by a transverse caecum
above the pneumostome, as in Succinea (Fig.
224). The kidney is not differentiated inter-
nally (RR2). The cerebral commissure is short
(CC3). The cerebro-pedal connectives are
medium in length, and subequal (CPD2,
CPR2). The pleural ganglia are close to the
pedal ganglia (PLD1, PLG1). The visceral
ganglion is median or on the left side (VG2,
VG3), and seems fused with the left parietal
ganglion (FG2’), which touches the left pleu-
ral ganglion (PAG4). The right parietal дап-
glion is close to the right pleural ganglion
(PAD2).

Character states of the CCA: BM1 OC1
SC2 PS1 IL2 LR1 UR4 RR2 ССЗ CPD2
CPR2 PLD1 PLG1 VG2 РАО? PAG4 F@2'.

Helicoidea: Haplotrematidae (Figs. 595—

| 599): The family Haplotrematidae is endemic

to western North America, Central America
and the northern Caribbean region. They are
close to the Oreohelicinae in the phenetic
classifications (Text-figs. 19, 20).

The Haplotrematidae are carnivorous, but
their buccal mass is not particularly devel-
oped (BM1). They do not have an oesoph-
ageal crop (OC1). The gastric crop is cylindri-
cal and the gastric pouch is differentiated
(SC1, PS1). The intestine is rather long (IL1).
The kidney is rather short (LR2) and internally
differentiated into two distinct regions (RR1).
The ureteric tube is closed as far as the pneu-
mostome (UR4). The cerebral commissure is
short (CC3). The cerebro-pedal connectives
are medium in length (CPD2) and the right

STYLOMMATOPHORAN SYSTEMATICS 63

one is longer than the left (CPR3). The pleural
ganglia are close to the pedal ganglia (PLD1,
PLG1). The visceral ganglion is median
(VG2) and seems fused with the left parietal
ganglion (FG2’). The visceral chain is short,
but the left parietal ganglion does not touch
the left pleural ganglion (PAD2, PAG3).

Character states of the CCA: BM1 OC1
SC1 IL1 LR2 UR4 RR1 CC3 CPD2 CPR3
PED PEGI VG2 PAD2 PAGS FG2’.

Helicoidea: Helicarionidae (Figs. 296-340):
The contents of this taxon are those indicated
by Zilch (1959-1960), i.e. with the Euconul-
idae excluded. The Euconulidae and the He-
licarionidae have an aulacopod foot and sig-
moid radular teeth tending to be very
numerous (Baker, 1938, 1940, 1941), but the
aulacopody might be convergent through
paedomorphosis, and the shape and number
of the radular teeth might be convergent
through diet: at least all the Helicarionidae
seen feeding ate macromycetes (observed by
C. Blanc in Madagascar, by E. Binder in West
Africa and by me in New Caledonia).

The Helicarionidae occur in tropical re-
gions, except South America and the Central
Pacific Islands. The groups defined by Solem
(1966a) within the Helicarionidae might be
monophyletic, but | doubt that the whole
family forms a monophyletic group although |
cannot justify rigorously this opinion. The
representatives of the family are very dis-
persed in the phenetic classification of the
species (Text-fig. 19), and the family is close
to other aulacopod groups in the phenetic
classification of the families (сопуегдепсе?;
Text-fig. 20).

The subfamily Dyakiinae is peculiar in its
general and genital morphology (Solem,
1966a). In this subfamily the amatorial appa-
ratus is similar to that in the Bradybaenidae;
the Dyakiinae and Bradybaenidae possibly
form a monophyletic group. Although raising
further questions, this remark supports the
placement of the helicarionids among the He-
licoidea, already proposed by Schileyko
(1978a). However, it is unlikely that the ama-
torial organ of the Urocyclinae and Gym-
narioninae is homologous with that of the
Helicoidea, because the hypothesis of its sec-
ondary differentiation seems well documented
(Van Mol, 1970; Binder & Tillier, 1985).

All the Helicarionidae have an aulacopod
foot. Their buccal mass is not elongated
(BM1), and the snails lack an oesophageal
crop (OC1). The homology of the crop of

semislugs and slugs (OC3, OC4) is dubious
(v. supra and Tillier, 1984a). The intestine is
never very long (IL1, IL2). The kidney is long
(LR1). Most internal renal lamellae are not
very developed, but a large, transversely
folded lamella runs along the rectal side of the
kidney. The ureteric tube is closed as far as
the pneumostome (UR4). The cerebral com-
missure is short (CC3). The cerebro-pedal
connectives are medium to short in length
(CPD2, CPD3), subequal in length (CPR2), or
the right one is longer than the left (CPR3).
The pleural ganglia are close to the pedal
ganglia (PLD1, PLG1), except in the West Af-
rican semislug Granularion (PLD2, PLG2).
The position of the visceral ganglion varies
(VG1, VG2, VG3). The right parietal ganglion
is always Closer to the visceral ganglion than
to the right pleural ganglion (PAD2). The left
parietal ganglion touches either the visceral
ganglion only (PAG3) or both the visceral and
left pleural ganglia (PAG4). Generally the vis-
ceral ganglion is distinct from the parietal gan-
glia (FG1), but it seems fused with the right
parietal ganglion in the ariophantine Hemi-
plecta (FG2), and seems fused with the left
parietal ganglion in the parmarionine slug
Parmarion (FG3).

Character states of the CCA: BM1 OC1
ЭС. Р51 A LR URA4 АВТ. СС63, CPD2
CPR2 PLD1 PLG1 VG1 РАО? PAG3 ЕСТ.

Helicoidea: Vitrinidae (Figs. 228-234):
The earliest genus attributed to the Vitrin-
idae is Provitrina, described from the Euro-
pean Paleocene. Shells assigned to Recent
genera occur in the European Oligocene
(Zilch, 1959—1960). At present the family is
principally Holarctic, and includes a few rep-
resentatives in east and northeast Africa and
in the Central Pacific islands.

All the Vitrinidae have a visceral mass
shortened or even incorporated in the pedal
cavity. Their similarity to the Helicarionidae is
obvious, and their position close to some He-
licarionidae in the phenetic classification of
the species (Text-fig. 19) reminds one of Schi-
leyko's assessment (1978a) of their affinities.
They are near the families of aulacopod slugs
in the phenetic classification of the families
(Text-fig. 20).

These animals are carnivorous, but the
buccal mass is elongated in the slug Plutonia
only (BM1, BM2). There is no oesophageal
crop (OC1). The gastric crop varies in shape
(SC1, SC2, SC2'). The gastric pouch is dif-
ferentiated in both the snails and semislugs

64 TILLIER

(PS1), but dedifferentiated in Plutonia (PS2,
carnivory). The intestine is short (IL2). The
ureteric tube is closed as far as the pneumos-
tome (UR4). The kidney is not differentiated
internally (RR2), and has a thick internal
lamella along its rectal side. The cerebral
commissure is short (CC3). The cerebro-
pedal connectives are short (CPD3), except
in Plutonia where their greater length might
be related to the development of the buccal
mass (carnivory, CPD2). The right cerebro-
pedal connective is longer than the left
(CPR3). The pleural ganglia are close to the
pedal ganglia (PLD1, PLG1). The visceral
ganglion is median (VG2) or on the left side
(VG3, Plutonia). The parietal ganglia are
closer to the visceral ganglion than to the
pleural ganglia (PAD2, PAG3). The visceral
ganglion seems fused with the right parietal
ganglion in Phenacolimax (FG2), and with the
left parietal ganglion in Plutonia (FG3).

Character states of the CCA: BM1 OC1
PS1 IL2 LR5 UR4 RR2 ССЗ CPD2 CPR3
PLD1 PLG1 VG2 PAD2 PAG3 FG1.

Plutonia is very similar to Milax not only in
its external morphology (tail carinate), but
also in the arrangement of its nervous sys-
tem. However, | do not think that close affinity
must be asserted, principally because the kid-
ney of Milax does not have a thick internal
lamella running along its rectal side.

Achatinoidea: Succineidae (Figs. 221-
227): The Succineidae are world-wide, and
some occur in the European Paleocene
(Zilch, 1959-1960). Phenetically, they are
close to the Discidae, Haplotrematidae and
Oreohelicidae (Ammonitellinae included) de-
spite their elongate shape (Text-fig. 20). |
have already discussed and rejected their
placement with the Athoracophoridae in a
suborder Heterurethra (Tillier, 1984a).

The Succineidae are not carnivorous
(BM1), and lack an oesophageal crop (OC1).
The gastric crop is cylindrical (SC1), except in
the slug Omalonyx (SC2'). The stomach
might be dedifferentiated (PS1, PS2), a cae-
cum being developed in compensation
(Tillier, 1984a). The intestine is rather long in
snails (11), and shorter in slugs (IL2). The
kidney is short (LR3) but is transversely elon-
gate (heterurethry). The ureteric tube is
closed as far as the pneumostome, and is
prolonged by a caecum across the pneumos-
tome roof in some species of Succinea (Fig.
224). The kidney is not differentiated inter-
nally (RR2). The cerebral commissure is me-

dium in length (CC2). The cerebro-pedal con-
nectives are short (CPD3), the left one being
shorter than the right (CPR1). The pleural
ganglia are near the pedal ganglia (PLD1,
PLG1). The visceral ganglion is median
(VG2) and seems fused with both the parietal
ganglia (FG3). The visceral chain is compact
(PAD2, PAGA).

Character states of the CCA: BM1 OC1
SC1 PSi IET АЗ: UR4 RR2 CGE2=ECRD3
CPR1 PLD1 PLG1 VG2 РАО? PAG4 FG3.

Achatinoidea: Ferussaclidae (Figs. 356,
358-360): The Recent genus Coilostele oc-
curs in the European Lower Eocene (Zilch,
1959-1960). The present distribution of the
family is circum-Mediterranean, African, Ori-
ental and Neotropical.

Only one species, Cecilioides acicula, was
dissected. It is close to the Subulinidae, but
also to the Urocoptidae and Clausiliidae in the
phenetic classification of the families. As
mentioned above, the Ferussaciidae and the
Succineidae both exhibit heterurethry (apo-
morphic) and aulacopody (apomorphic), but
also two procerebral commissures (plesio-
morphic; Watson, 1928). Furthermore the pe-
nial retractor muscle in Cecilioides is a branch
of the columellar retractor, which probably
corresponds to the ancestral condition in the
Succineidae (Tillier, 1984a).

Cecilioides is not carnivorous (BM1) and
does not have an oesophageal crop (OC1).
The median portion of the gastric crop is in-
flated (SC2) and a constriction separates the
gastric crop from the gastric pouch (PS2’).
The intestine is short (IL2). The kidney is very
short (LR4) but transversely elongate such
that its primitively visceral side runs along the

_rectum. The internal renal lamellae are re-

duced to a series of lamellae perpendicular to
the rectal border and present only on the pul-
monary surface of the kidney; their arrange-
ment reminds one of that in the Succineidae
(small lamellae perpendicular to the rectal
border) and in the Oleacinidae (lamellae ab-
sent from the palatal surface of the kidney)
(Figs. 358, 225, 367). The cerebral commis-
sure is short (CC3). The cerebro-pedal con-
nectives are medium and subequal in length
(CPD2, CPR2). The pleural ganglia are close
to the pedal ganglia (PLD1, PLG1). The pari-
etal ganglia touch the visceral ganglion
(PAD2, PAG3) but are distinct from them
(FG1). The visceral ganglion is on the right
side (VG1).

Character states of the CCA (= character

STYLOMMATOPHORAN SYSTEMATICS 65

states observed in Cecilioides): BM1 OC1
SCPRPS2NIID I ERS UR4 RR2 CC3 CPD2
CPR2 PLD1 PLG1 VG1 PAD1 PAGS FG1.

Achatinoidea: Subulinidae (Figs. 341-
352): A few genera from the European and
North American Paleocene are assigned to
the Subulinidae (Zilch, 1959-1960). The fam-
ily is now pantropical, and includes the prima-
rily circum-Mediterranean genus Rumina.

The Subulinidae are not carnivorous
(BM1). An oesophageal crop occurs in only
one of the four genera dissected (OC1, OC2).
The gastric crop is inflated in its median por-
tion (SC2, except in Bocageia, in which it is
dedifferentiated: SC2') and separated from
the gastric pouch by a constriction (PS2’).
The intestine is rather short (IL1, IL2). The
kidney varies in length (LR1, LR2, LR3) and is
generally differentiated internally into two dis-
tinct regions (RR1). The ureteric tube is
closed as far as the pneumostome (UR4).
The cerebral commissure is short (CC3). The
cerebro-pedal connectives are medium in
length or long (CPD1, CPD2), the right one
being longer than the left (CPR3, CPR4). The
pleural ganglia are close to the pedal ganglia
(PLD1, PLG1). The visceral ganglion is either
on the right side or in the middle (VG1, VG2).
The right parietal ganglion touches and
seems fused with the visceral ganglion in Bo-
cageia and Pseudoglessula (PAD2, FG2); it is
close to the right pleural ganglion and sepa-
rated from the visceral ganglion in Rumina
(PAD1, FG1).

Character states of the CCA: BM1 OC1
562.252’ 11 LRi, UR4 RRIMCCS CPD1
ОРАЗРЕВИРЕСТ МЕТРА PAG3 FG1.

Achatinoidea: Achatinidae (Figs. 353-
355): The Recent Achatinidae occur in trop-
ical Africa only, and none is earlier than the
Pleistocene (Zilch, 1959-1960). The Mada-
gascan genus Leucotaenius belongs to the
Acavidae (Mead, 1986). The Achatinidae are
phenetically close to helicoids (Text-figs. 19,
20), but this might be the effect of their large
size. | think that the Achatinidae might be sim-
ply giant subulinids, but the classification of
the latter is so confused that | am unable to
discuss the monophyly of the achatinids with
one of the groups probably lumped together
in the subulinids.

The sole species here dissected is distinct
from the subulinids only in apomorphic char-
acter states that might result from large size:
the oesophageal crop is very large (OC3),

and the visceral chain is compact (PAD2,
PAGA).

Character states of the CCA (= character
states of Achatina са): BM1 ОСЗ SC2
PS2’ IL1 LR1 UR4 RR2 ССЗ CPD2 CPR3
PLD1 PLG1 VG2 РАО? PAG4 FG1.

Achatinoidea: Streptaxidae (Figs. 394-
398): The Recent Streptaxidae are tropical
African, South American, Indian and Oriental.
A few Upper Cretaceous and Paleocene Eu-
ropean genera have been assigned to the
Streptaxidae (Strophostomella, Lychnopsis,
Enneopsis, Anastomopsis; Zilch, 1959-
1960); but | do not understand which charac-
ters other than size distinguish them from the
Anadromidae, which occur in the same for-
mations. The Recent Canarian genus Gibbu-
linella occurs in the south European Upper
Cretaceous. Representatives of the genus
Rillya, which is very similar to the Recent ge-
nus Edentulina in shell morphology, are
known from the Paleocene, and the genus
Paracraticula, which looks like a Recent en-
neine, occurs in the European Eocene.
Brasilennea has been described from the
Brasilian Miocene (Zilch, 1959-1960).

The Streptaxidae are phenetically close to
the other carnivorous families, as in classical
classifications (Zilch, 1959-1960; text-figs.
19, 20).

The buccal mass is very long, in associa-
tion with carnivory (BM2). Edentulina differs
from the other genera dissected in its posses-
sion of a large oesophageal crop (OC3, OC1).
The gastric crop is cylindrical (SC1). The gas-
tric pouch might be dedifferentiated (PS1,
PS2), and the intestine is short, as a function
of carnivory (IL2). The kidney is very short
and internally homogeneous in morphology
(LR4, RR2). The ureteric tube is closed as far
as the pneumostome (UR4). The cerebral
commissure is short (CC3). The cerebro-
pedal connectives are medium and subequal
in length (CPD2, CPR2). The pleural ganglia
are close to the pedal ganglia (PLD1, PLG1).
The visceral ganglion is median (VG2), in
contact with both the parietal ganglia, from
which it is, however, distinct (PAD2, РАСА,
FG1). The pleuro-parietal connectives are
longer than in any other family, as described
above (carnivory).

Character states of the CCA: BM2 OC1
SC1 PS1 IL2 LR4 UR4 RR2 CC3 CPD2
CPR2 PLD1 PLG1 VG2 РАО? PAG3 FG1.

This CCA differs from that of the ferussaci-
ids only in lesser differentiation of the gastric

66 TILLIER

pouch, which may be due to carnivory, and in
a shorter kidney; but the relatively greater
length of the kidney in ferussaciids might be
related to lung roof shortening.

Achatinoidea: Oleacinidae (+ Spiraxidae
+ Testacellidae) (Figs. 357, 361-374): The
oleacinids range from the Eocene in Europe,
and from the Miocene in Florida (Zilch, 1959—
1960) to the Recent. The Recent species are
South European, North African, Central Amer-
ican and Caribbean.

The Testacellinae are here included in the
Oleacinidae, rather than recognized as a dis-
tinct family. This idea is not new (Watson,
1915), and reflects the refusal to base family-
level divisions on autapomorphies related to
limacization solely: the remarkable anatomi-
cal description of Testacella by Lacaze-
Duthiers (1887) indicates that this genus
might be derived easily from an ancestor sim-
ilar to Strebelia (Figs. 372-374), except for
internal kidney morphology.

Baker (1962) removed the genera Euglan-
dina, Spiraxis and Streptostyla from the Ole-
acinidae to form the family Spiraxidae, which
he considers close to the Sagdidae and Acha-
tinidae (v. supra; Boss, 1982). This position is
unacceptable. Varicella (Oleacinidae sensu
Baker) is more similar to Euglandina and
Spiraxis than to any other genus dissected
here; furthermore the crop is anteriorly ex-
tended over the oesophagus as a rostrum in
these three genera, not as in any other genus
dissected but therhytidid Priodiscus (v. infra).
Use of this character as a synapomorphy
would produce a division different from that
adopted by Baker: the Neotropical Varicella,
Spiraxis and Euglandina (+ Streptostyla? +
Priodiscus ??) would form the sister group of
the Neotropical and west Mediterranean
group including Poiretia, Strebelia and Testa-
cella. On the other hand, all the Neotropical
genera dissected have a similar internal renal
morphology, characterized by the absence of
lamellae on the palatal surface of the kidney,
as in ferussaciids. The west Mediterranean
Poiretia and Testacella have lamellae on both
the palatal and pulmonary surfaces of the kid-
ney. The use of this character as a synapo-
morphy would lead again to a division differ-
ent from that adopted by Baker.

All the Oleacinidae are carnivorous (BM2).
Euglandina, Varicella and Spiraxis lack an oe-
sophageal crop, but their gastric crop extends
far forward as a rostrum. In Poiretia, the oe-
sophagus is inflated and the gastric crop is

not large. The morphology in the slugs
Strebelia and Testacella might be derived
from such an arrangement. The gastric pouch
might be dedifferentiated (PS1, PS2; car-
nivory). The intestine might be very short (IL1,
IL3), in relation to carnivory associated either
with limacization (Strebelia), or with very
small size (Spiraxis). The kidney is either
short (LR2) and equilaterally triangular, or
very short (LR4) and transversely elongate
(heterurethry). It is not differentiated internally
(RR2). Spiraxis futilis has a ureteric groove
along the rectum (UR3). In the other genera
dissected the ureter is closed as far as the
pneumostome (UR4). The cerebral commis-
sure is short (CC3). The cerebro-pedal con-
nectives vary in length (CPD1, CPD2, CPD3);
they are either subequal, or the right one is
longer than the left (CPR2, CPR3, СРВА).
The pleural ganglia are close to the pedal
ganglia (PLD1, PLG1). The visceral ganglion
is on the left side of the median plane (VG3),
and seems either distinct from the parietal
ganglia or fused with the right one (FG1,
FG2). The position of the right parietal gan-
glion varies (PAD1, PAD2). The left parietal
ganglion touches both the left pleural and vis-
ceral ganglion (PLG3, PLG4).

Character states of the CCA: BM2 OC1
SCi PS? 111 LR2 УАЗ RR2 CCIACENA
CPR2 PLD1 PLG1 VG3 PAD1 PAG3 FG1.

Endodontoidea: Charopidae (Figs. 153-
174): The Charopidae are nearly restricted to
the Southern Hemisphere, although a few
members of the family occur in North Amer-
ica. Their diversity is maximal in the Pacific
Islands, particularly New Caledonia and New
Zealand, but less in Australia, South Africa,
Madagascar and southernmost South Amer-
ica (Solem, 1982).

The Charopidae are close together in the
phenetic classification of the species, with the
exception of the slug Ranfurlya (Text-fig. 19).
They are close to the Systrophiidae, Punc-
tidae and Endodontidae in the phenetic clas-
sification of the families (Text-fig. 20).

The Charopidae are not carnivorous
(BM1), and lack an oesophageal crop (OC1).
The anterior portion of the gastric crop is in-
flated, narrowing toward the gastric pouch
(SC3), except in Stephanoda, in which the
gastric crop widens toward the stomach as in
the Discidae (SC2’). The gastric pouch is dif-
ferentiated in most species dissected (PS1).
The intestine is short (IL2). The kidney is
short to very short (LR2 to LR4), and usually

STYLOMMATOPHORAN SYSTEMATICS 67

differentiated internally into two distinct re-
gions (RR1). The ureteric tube is generally
closed as far as the pneumostome (UR4), ex-
cept in a few undescribed Australian genera
(UR2; Solem, pers. comm.). The length of the
cerebral commissure is medium to short
(CC2, CC3). The cerebro-pedal connectives
are medium in length (CPD2), except in the
slug Ranfurlya, in which they are shorter
(CPD3, limacization). The right pleural gan-
glion is closer to the right cerebral ganglion
than to the pedal ganglion (PLD2, except in
Ranturlya), whereas the left pleural ganglion
is nearer the left pedal ganglion (PLG1). The
visceral ganglion is on the right side (VG1).
The parietal ganglia are closer to the pleural
ganglia than to the visceral ganglion (PAD1,
PAG1), except in Ranfurlya (PAD2, limaciza-
tion) and in Mystivagor (PAG2, elongate
shape).

Character states of the CCA: BM1 OC1
563 PS1 IL2 LR2 UR2 ВАТ CC2 CPD2
CRRATWPED2PEG1 VG1 PAD PAGA ЕСТ.

Endodontoidea: Punctidae (Figs. 175-
183, 216): The genus Punctum, which 1$ prin-
cipally Palearctic, south and east African in
distribution, occurs in the European Oli-
gocene (Zilch, 1959-1960). Most Recent
genera occur in New Zealand, the Pacific Is-
lands, Australia, South Africa and South
America (Solem, 1982). The Australian semi-
slug Cystopelta is here placed among the
Punctidae because it is similar to slugs be-
longing to the Endodontoidea sensu Solem,
and to the athoracophorid slugs here consid-
ered endodontoids as well as to the Punc-
tidae in its radular characters (Solem, pers.
comm.).

The Punctidae are not carnivorous (BM1),
and lack an oesophageal crop (OC1). The an-
terior portion of the gastric crop is inflated
(SC3), except in Cystopelta, in which the cy-
lindrical shape of the organ might result from
limacization. The gastric pouch is differenti-
ated (PS1), and the intestine is short (IL2).
The kidney is short (LR3) and internally dif-
ferentiated into two distinct regions (RR1).
The ureteric tube is closed as far as the pneu-
mostome (UR4). The cerebral commissure is
short (CC3). The cerebro-pedal connectives
are medium in length (CPD2; short in Cys-
topelta: limacization?), the right one being
longer than the left (CPR3, except in Cys-
topelta). The right pleural ganglion is close to
the right cerebral ganglion (PLD2), whereas
the left pleural ganglion is closer to the left

pedal ganglion (PLG1) (compaction occurs in
Cystopelta). The visceral ganglion is on the
right side of the median plane, and seems
fused with the right parietal ganglion (VG1,
FG2). The left parietal ganglion is closer to
the left pleural than to the visceral ganglion
(PAG1).

Character states of the CCA: BM1 OC1
SC3 PS1 IL2 LR3 UR4 RR1 CPD2 CPR3
PLD2 PLG1 VG1 PAD2 PAG1 FG2.

Endodontoidea: Athoracophoridae (Figs.
218-220): The Athoracophoridae are all
slugs and are Australasian. Phenetically, the
non-athoracophorid species closest to the
Athoracophoridae is the Australian endodon-
toid slug Cystopelta (Text-fig. 19). At the fa-
milial level, the Athoracophoridae are more
similar to the Arionidae (Philomycinae in-
cluded) and Vitrinidae than to any other family
(Text-fig. 20).

The diet is not carnivory (BM1), and the
crop is cylindrical (SC1). The gastric pouch is
reduced (PS2). The intestine is very long, as
a function of advanced limacization (IL2';
Tillier, 1984a). The kidney is compact (LR5)
and filled with anastomose lamellae (RR2).
The ureteric tube is closed to the pneumos-
tome, and forms numerous convolutions
(Plate, 1898; Tillier, 1983). The cerebral com-
missure is short (CC3). Connectives could
not be distinguished owing to compaction of
the central nervous system (CPD3, PLD3,
PLG3); the asymmetry in the arrangement of
the parietal and pleural ganglia, which seem
fused (Fig. 220), suggests that in the athora-
cophorid ancestors the left pleural ganglion
was close to the left pedal ganglion, and the
right pleural ganglion was close to the right
cerebral ganglion (PLD2, PLG1). The visceral
ganglion lies to the right of the median plane
(VG1), and seems fused with both parietal
ganglia (PAD2, PAG4, FG3).

Character states of the CCA: BM1 OC1
SC1 PS2 12’ LR5 UR4 RR2 ССЗ CPD3
CPR2 PLD2 PLG1 VG1 РАО? PAG4 FG3.

Endodontoidea: Endodontidae (Figs. 147-
152): The family Endodontidae is endemic
to the Central Pacific Islands (Solem, 1976).
Both species here dissected are close to the
Charopidae, Punctidae and Systrophiidae in
the phenetic classifications (Text-figs. 19, 20).

Genital and radular characters of this family
have been discussed by Solem (1976). In
characters used here, the Endodontidae differ
from the Charopidae in the dedifferentiation of

68 TILLIER

their gastric pouch (PS2), in the position of
the left pleural ganglion close to the cerebral
ganglion (PLG2), and in a ureteric tube not
closed farther than kidney-rectum angle
(UR2, found also in a few Australian Charop-
idae).

Character states of the CCA: BM1 OC1
563 PS2 IL2 ERS UR2 АВТ CC2 CPD2
CPR1 PLD2 PLG2 VG1 PAD1 PAG1 FG2.

The CCA of the Endodontidae has either
the same character states as the CCA of the
Charopidae, or more apomorphic states.
Consequently the CCA of the Charopidae is a
CCA of the Endodontidae, but the converse is
not true. This finding contradicts the opinion
of Solem (1982), who estimates that the En-
dodontidae are closer to the common ances-
tor of both families than are the Charopidae,
or even represent the stem group of the En-
dodontoidea.

Endodontoidea: Systrophiidae (Figs. 399—
405): The Systrophiidae are Neotropical, but
do not occur in the Greater Antilles and in
Central America. They share a synapomor-
phy, contiguity between both pleural ganglia
with the cerebral ganglia associated with the
grouping of the right pleural, right parietal and
visceral ganglia beneath the right cerebral
ganglion (exaptation or adaptation to car-
nivory). In the phenetic classification of the
families they are close to the Charopidae,
Punctidae, Endodontidae and Euconulidae
(Text-fig. 20).

The Systrophiidae are carnivorous, but
their buccal mass is not always elongate
(BM1, BM2). They do not have an oesoph-
ageal crop (OC1). The gastric crop becomes
narrower toward the gastric pouch (SC3),
which might be dedifferentiated (PS2, car-
nivory). In Systrophia the anterior duct of the
digestive gland opens into the gastric crop
farther forward than in any other stylommato-
phoran dissected (Figs. 399, 405), with the
exception of Ventridens (Gastrodontinae, Fig.
269), which has the same arrangement. The
intestine is short (IL2). The kidney is relatively
long (LR1) in all species dissected, Systro-
phia eudiscus excepted (LR2); this relatively
great length may result from shortening of the
visceral mass and lung. The ureteric tube is
only partly closed in Systrophia eudiscus
(UR2), and closed as far as the pneumo-

tome in the other species dissected. In the

atter, it runs Aihara the pneumostome along
the lung roof and opens on the side of the
pneumostome opposite the anus (Tillier,

1980). There are two regions inside the kid-
ney pouch (RR1); a thick internal lamella runs
along the rectal border of this pouch. The ce-
rebral commissure is short (CC3), the lateral
connectives are medium and subequal in
length (CPD2, CPR2), and the visceral chain
has the peculiar arrangement described
above.

Character states of the CCA: BM1 OC1
SC3 PS1 IL2 LR2 "UR22RRIMCCSLCED?2
CPR2 PLD2 PLG2 VG1 PAD2 PAG1 FG2.

Clausilioidea: Clausiliidae (Figs. 505-520):
Recent Clausiliidae occur from Indonesia to
Europe, in the Greater Antilles and in north-
ern South America. The subfamily Neniinae
includes the Neotropical taxa, as well as two
doubtfully attributed Oriental genera (Zilch,
1959-1960). The subfamily Phaedusiinae oc-
curs in the European Danian. Zilch (1959—
1960) considers the European Eocene and
Oligocene Filholiidae related to the Clau-
silidae. The family Megaspiridae, which in-
cludes a few European Eocene taxa and Re-
cent genera in Brazil, New Guinea and
Queensland, has been interpreted by Pilsbry
(1904) as related to this group.

In the phenetic classifications of the spe-
cies (Text-fig. 19) and of the families (Text-fig.
20), the Clausiliidae are close to the Urocop-
tidae on the one hand and to the Ferussaci-
idae and Succineidae on the other.

The diet of the Clausiliidae is never car-
nivory (BM1), and no oesophageal crop has
been seen (OC1). The median portion of the
gastric crop is inflated (SC2), and the crop is
separated from the well-differentiated gastric
pouch by a constriction (PS2'). The intestine

_ Is short (112). The kidney is very short (LR3,

LR4), and is not differentiated internally
(RR2). There is no ureteric tube (UR1). The
cerebral commissure is very long (CC1). The
cerebro-pedal connectives are medium to
short in length (CPD2, CPD3), the right one
being shorter than the left (CPR1, CPR2; the
observed asymmetry is actually reversed
because the Clausiliidae are sinistral). The
pleural ganglia are usually closer to the pedal
ganglia than to the cerebral ganglia (PLD1,
PLG1). The visceral ganglion is either on the
right side, or median (VG1, VG2) and does
not seem fused with either of the parietal gan-
glia (FG1). The parietal ganglia are closer to
the visceral than to the pleural ganglia, and
may touch the former (PAD1, PAD2, PAG2,
PAGS).

Character states of the CCA: BM1 OC1

STYLOMMATOPHORAN SYSTEMATICS 69

SC2 PS2" IL2 LR3 UR1 RR2 CC1 CPD2
CPR1 PLD1 PLG1 VG1 PAD1 PAG2 FG1.

Clausilioidea: Cerionidae (Figs. 521-527):
The Cerionidae are endemic to the Antillean
region, from Florida to the Venezuelan coast.
The earliest genus is Eostrophia, from the Mi-
ocene of Florida (Zilch, 1959-1960).

Their proximity in the phenetic classification
of species to Urocoptis (Text-fig. 19) is more
representative of their obvious similarity to the
Urocoptidae than is their position in the phe-
netic classification of families (Text-fig. 20)
among the Orthurethra, which reflects the
great length of the kidney in one of the spe-
cies dissected.

Dissection of a series of Cerion belonging
to several species recognized by Gould and
Woodruff (C. casablancae, C. glans, C. rubi-
cundum, C. agassizi, C. copium; Woodruff,
1978; Chung, 1979) shows that intraspecific
variation in proportions of the lengths of the
organ systems is more marked here than in
any other group studied. This variability, to-
gether with the similarity of cerionids to uro-
coptids, leads me to consider the great length
of the kidney seen in some species of Cerion
secondary, not plesiomorphic.

The cerionids are not carnivorous (BM1),
and lack an oesophageal crop (OC1). The
gastric crop is cylindrical (SC1). The gastric
pouch is differentiated (PS1) and might be
separated from the gastric crop by a constric-
tion (PS2’). The intestine is medium in length
(IL1). The kidney is short (LR3, LR4), some-
times secondarily longer (LR2, LR1), and not
differentiated internally (RR2). There is no
ureteric tube (UR1). The cerebro-pedal con-
nectives and the cerebral commissure are
medium in length (CC2). The left cerebro-
pedal connective is longer than the right one
(CPR3). The pleural ganglia are close to the
pedal ganglia (PLD1, PLG1). The visceral
ganglion is median (VG2), and in contact with
both parietal ganglia (PAD1, PAG3).

Character states of the CCA: BM1 OC1
SGINPSIMEMLERS URI RR2-Ce2 CPD2
CPR3 PLD1 PLG1 VG2 PAD1 PAG3 FG1.

Clausilioidea: Urocoptidae (Figs. 528-
535): Recent Urocoptidae are endemic to
Central America and to the Antillean region.
The Recent Mexican genus Holospira also
occurs in the North American Paleocene
(Zilch, 1959-1960). The Urocoptidae are
close to the Clausiliidae in the phenetic clas-
sification of the families (Text-fig. 20).

Urocoptids are not carnivorous (BM1), and
they lack an oesophageal crop (OC1). The
median portion of the gastric crop is inflated
(SC2). The morphology of the gastric pouch is
variable (PS1, PS2, PS2'). The intestine is
short (IL2). The kidney is short (LR3, LR4),
and not differentiated internally (RR2). The
length of the ureteric tube is very variable:
Macroceramus has no ureteric tube (UR1);
the ureteric tube reaches the recto-visceral
angle of the kidney in Urocoptis (UR2), and
nearly reaches the pneumostome region in
Berendtia (UR3). The cerebral commissure is
short (CC3). The cerebro-pedal connectives
are long (CPD1) and subequal in length
(CPR2). The visceral chain has the same ar-
rangement seen in the Cerionidae (PLD1,
PLG1, VG2, PAG3), but is more compact
through apparent fusion of the left parietal
and visceral ganglia (FG2’) and contact of the
right parietal with the right pleural ganglia
(PAD2).

Character states of the CCA: BM1 OC1
SC21PSIMILEZ2 ГАЗОВ АВ" GESZERDI
CPR2 PLD1 PLG1 VG2 РАО? PAG3 FG2’.

Clausilioidea: Bulimulidae (Figs. 536-555):
The Bulimulidae are Neotropical and Aus-
tralasian. The oldest known representative is
Paleobulimulus, from the Patagonian Eocene
(Parodiz, 1969).

In the phenetic classifications of both spe-
cies and families, the Bulimulidae are close to
helicoids (Text-figs. 19, 20).

The Bulimulidae are not carnivorous (BM1)
and have an oesophageal crop (OC2, OC3).
The gastric crop is either cylindrical (SC1), or
inflated in its median portion (SC2). The gas-
tric pouch is variable in morphology (PS1,
PS2, PS2') and frequently divided internally
into two chambers by a fold in the plane of
separation of the crop and intestine. The in-
testine is medium to long in length (IL1, IL2”).
The kidney is short (LR3), but may be sec-
ondarily longer in taxa with a shortened vis-
ceral mass and lung (Simpulopsis, Pellicula),
and is not differentiated internally (RR2). The
ureteric tube reaches a point between the up-
per extremity of the lung and the pneumo-
stome (UR3, UR4). The cerebral commissure
is variable in length (CC1, CC2, CC3). The
cerebro-pedal connectives are medium to
long and subequal in length (CPD1, CPD2,
CPR2). The pleural ganglia are close to the
pedal ganglia (PLG1, PLD1). The visceral
ganglion is on the left side of the median
plane (VG3) and seems fused with the left

70 TILLIER

parietal ganglion (FG2') or with both parietal
ganglia (FG3). The visceral chain is compact
(PAD2, PAG3, РАСА).

Character states of the CCA: BM1 OC2
SC1 PS1 Ш1 LR3 UR3 RR2 CC1 CPD1
CPR2 PLD1 PLG1 VG3 PAD2 PAG3 FG2’.

Acavoidea: Oreohelicidae (+ Ammonitelli-
dae) (Figs. 492-504): The Ammonitellidae
auctt. are here considered to belong to the
Oreohelicidae, because the Oreohelicidae
auctt. differ from them only in more apomor-
phic character states, and because both
groups are endemic to North America. The
two groups are close together and close to
the Haplotrematidae in the phenetic classifi-
cation of the families (Text-fig. 20). The spe-
cies dissected are close to helicoids in the
phenetic classification of the species (Text-
fig. 19).

The Ammonitellinae (Figs. 497—504) соп-
sists of three Recent genera and is endemic
to North America, where at least one species
occurred in the Upper Cretaceous (Solem, in
litt.). These animals are not carnivorous
(BM1), and they lack an oesophageal crop
(OC1). The gastric crop is cylindrical, or in-
flated in its median portion (SC1, SC2). The
gastric pouch is differentiated and is sepa-
rated from the gastric crop by a constriction in
Glyptostoma (PS1, PS2’). The intestine is
medium to long in length (IL1, IL2’). The kid-
ney is short (LR3, LR4), and is not differenti-
ated internally (RR2). The ureteric tube is not
closed farther than the recto-visceral extrem-
ity ofthe kidney (UR2). The cerebral commis-
sure varies in length (CC1, CC3). The
cerebro-pedal connectives are long and,
when they are unequal in length, the right one
is the longer (CPD1, CPR1, CPR2). The pleu-
ral ganglia are close to the pedal ganglia in
Glyptostoma (PLD1, PLG1), but close to the
cerebral ganglia in Ammonitella (PLD2,
PLG2). The visceral ganglion is on the right
side of the median plane (VG1), and seems
fused with the right parietal ganglion in Am-
monitella only (FG1, FG2). The left parietal
ganglion touches the visceral ganglion
(PAG3).

The Oreohelicinae (Figs. 492-496), now
endemic to western North America, occur in
the Upper Cretaceous of the same region
(Zilch, 1959—1960). The only species dis-
sected, Oreohelix barbata, differs from the
Ammonitellinae dissected in its shorter ure-
2 length and asymmetry of its

connectives (CPD3, CPR3),

if t

cere eda

and the asymmetric arrangement of its pleural
ganglia (PLD2, PLG1).

Character states of the CCA: BM1 OC1
SCH РЭ Iki ERS UR2 RR21CCIACEDS
CPR3 PLD2 PLG1 VG1 PAD2)PAG3 ЕСТ.

Acavoidea: Corillidae (Figs. 470—491): The
family Corillidae, unknown as fossils, includes
only five genera: Sculptaria in South Africa,
Corilla in Sri Lanka and southern India, Plec-
topylis in Indochina, Amphicoelina in China
and Craterodiscus in Australia. Craterodiscus
is here removed from the Camaenidae be-
cause the morphology of its pulmonary com-
plex is similar to that of Sculptaria and Plec-
topylis. Furthermore, it lacks an oesophageal
crop, as do all corillids; all dissected ca-
maenids have such a crop. A ureteric tube
reaching the pneumostome cannot be used to
exclude Craterodiscus from the Corillidae, be-
cause several species of Plectopylis exhibit
the same character state (Solem, 1966a;
Solem & Tillier, unpublished observations).
On the other hand, Craterodiscus differs from
all the other Corillidae dissected in the ex-
treme shortness of its intestine; however, this
character state is not seen in any camaenid,
and might be related to the very small size of
Craterodiscus.

In the phenetic classification of the families,
the Corillidae are close to the groups that in-
clude the Urocoptidae, Clausiliidae and Rhy-
tididae (Text-fig. 20).

The buccal mass is spheroidal (BM1), and
an oesophageal crop is absent (OC1). The
gastric crop is cylindrical or inflated in its an-
terior portion (SC1, SC3). The gastric pouch
varies in shape and degree of differentiation

‚ (PS1: medium size; PS2’: large size; PS2:

small size). The intestine is very long (IL2’),
except in Craterodiscus (IL2: very small size).
The kidney is very short (LR4), and is not dif-
ferentiated internally (RR2). There is usually
no ureteric tube (UR1), but a few species
have a ureteric tube reaching the pneumo-
stome (UR4). The cerebral commissure is vari-
able in length (CC1, CC2, CC3). The cerebro-
pedal connectives are medium to long and
subequal in length (CPD1, CPD2, CPR2).
The right pleural ganglion is close to the right
cerebral ganglion (PLD2), except in Corilla, in
which it is closer to the right pedal ganglion
(PLD1; large size?). The left pleural ganglion
is close to the pedal ganglion in all species
dissected (PLG1). The visceral ganglion is ei-
ther median (VG2) or on the left side (VG3).
Apparent fusion of the visceral and parietal

STYLOMMATOPHORAN SYSTEMATICS 71

ganglia does not occur (FG1) except in Cra-
terodiscus, in which the three ganglia form a
single mass (FG3). Whenever the visceral
chain is not compact, the left parietal ganglion
is closer to the left pleural than to the visceral
ganglion (PAG1), and the right parietal gan-
glion is distinct from the visceral ganglion
(PAD1).

Character states of the CCA: BM1 OC1
SGierol 2” ERS URI RR2 CCl CPD1
CPR2 PLD1 PLG1 VG2 PAD1 PAG1 FG1.

Acavoidea: Acavidae (Figs. 406-469):
The Acavidae occur in South Africa, South
America, Madagascar, the Seychelles, Sri
Lankaand Australia. Theliving genera Stropho-
cheilus and Dorcasia range respectively in
South America and in South Africa from the
Paleocene to the Recent. | reject the arrange-
ment adopted by Boss (1982), who divides
the Acavidae into several families placed in
different superfamilies and suborders. Al-
though it is likely that the Acavidae include
several monophyletic groups, | do not know of
any synapomorphy which might define these
groups and unite some of them with other
families. The Acavidae are united by overall
similarity and by their general trend of short-
ening of the visceral mass with deformation of
the kidney, as described above. Solaropsis
might be included in the family on this basis,
but in this case Labyrinthus and Isomeria
should be included too (v. supra).

Phenetically the Acavidae are close to the
groups that include the Helicoidea, Bulimul-
idae and Achatinidae (Text-fig. 20).

The Acavidae are generally not carnivo-
rous and have a spheroidal buccal mass
(BM1), with the exception of the Seychellian
Stylodon and South American Macrocyclis
(BM2). An oesophageal crop might be pres-
ent and very large (OC1, OC2, ОСЗ, OCA).
The gastric crop varies in shape (SC1, SC2,
SC2'). The gastric pouch is differentiated
(PS1, PS2’). The intestine is very long (IL2’),
except in the semislug Pandofella (IL1). The
kidney varies in length in relation to lung
shortening, as shown above (LR1, LR2, LR3,
LR4). The plesiomorphic state is probably the
state related to minimum deformation in the
arrangement of the lung roof and kidney, i.e.
LR3 (kidney short). A ureteric tube might be
absent or present, although never reaching
the pneumostome (UR1, UR2, UR3), and the
kidney might be divided internally into two dis-
tinct regions (RR1, RR2). The cerebral com-
missure is medium to short in length (CC2,

CC3). The cerebro-pedal connectives are
generally long, but can be very short (CPD1,
CPD2, CPD3). When they are unequal, the
right one is shorter than the left (CPR2,
CPR3, CPR4). The pleural ganglia are close
to the pedal ganglia (PLD1, PLG1). The vis-
ceral ganglion varies in position (VG1, VG2,
VG3), and is generally distinct from the pari-
etal ganglia (FG1); apparent fusion occurs
with the right parietal ganglion, if present
(FG2). The visceral chain is generally com-
pact; however, Caryodes has a very long vis-
ceral chain despite its large size, and its pa-
rietal ganglia touch its visceral ganglion
(PAD1, PAGS).

Character states of the CCA: BM1 OC1
sch PSi №2” ВАЗ URI RR СС? СРВ]
CPR2 PED? PEGI МСТ РАБОТ PAGS ЕСТ.

Acavoidea: Rhytididae (+ Chlamydephori-
dae) (Figs. 375-393): According to their
current definition, adopted by Zilch (1959—
1960), the Rhytididae occur in South Africa,
the Seychelles and the Australasian region.
The Rhytididae are phenetically close to the
Oleacinidae, Streptaxidae (carnivory) and
Corillidae.

Some Acavidae being carnivorous (Styl-
odon, Macrocyclis), it is difficult to define a
limit between them and the Rhytididae. The
Rhytididae might be defined by a synapomor-
phy, the presence of loops of the ureteric tube
between the renal pore and the rectal side of
the kidney: such loops are either between the
pericardium and the kidney (Nata, Rhytida;
incipient in Ouagapia and Schizoglossa), or
along the pulmonary vein distal to the renal
роге (Diplomphalus). If accepted, this solution
would exclude Natalina (Watson, 1934) and
Priodiscus from the Rhytididae. If transferring
Natalina to the Acavidae is not particularly
problematic, such is not the case with Prio-
discus, which has several features that might
indicate other relationships. First, the nervous
system has the arrangement found elsewhere
only in the Streptaxidae, ¡.e. very long parieto-
pleural connectives (compare Figs. 376, 396).
Second, the anterior portion of the gastric
crop is developed as a rostrum, occurring
elsewhere only in some American Olea-
cinidae (v. supra). Third, its kidney has inter-
nal lamellae only on its pulmonary surface, as
in American Oleacinidae and some Corillidae.
Fourth, its shell (Zilch, 1959-1960, Fig. 1921)
reminds one of the shells of the Rhytididae
only because it is rather flat, but does not
resemble more closely the shells of the

72 TILLIER

Streptaxidae or of the Oleacinidae. Fifth, in
terms of overall similarity, Priodiscus resem-
bles more closely Craterodiscus (Corillidae)
than it does any member of the three carni-
vorous families (Text-fig. 19).

My knowledge of the genital anatomy in
these families is not sufficient for me to judge
the affinities of Priodiscus, of which the gen-
ital apparatus is shown in Fig. 379.

The case of the Chlamydephorinae is far
simpler, and | do not hesitate to include
Chlamydephorus in the Rhytididae because
the New Zealand rhytidid slug Schizoglossa is
intermediate in morphology between rhytidid
snails and these very advanced slugs (Wat-
son, 1915; Figs. 392,393).

These animals are carnivorous, and their
buccal mass is always cylindrical (BM2). In
Nata, Schizoglossa and Chlamydephorus
(OC2) the anterior portion of the oesophagus
instead ofthe median portion is inflated to form
the oesophageal crop. The other genera lack
an oesophageal crop (OC1). The gastric crop
either is cylindrical (SC1), or becomes nar-
rower toward the stomach (SC3). The gastric
pouch is dedifferentiated (PS2), except in Pri-
odiscus (PS1). The intestine is usually very
long (IL2'), except in the slug Schizoglossa
(IL2, limacization), and in Priodiscus, in which
its very short length might be related to the
association of carnivory with small size (IL3).
The kidney is short (LR3), except in taxa hav-
ing the visceral mass and lung secondarily
shortened (LR2, LR1); the first branch of the
pulmonary vein is particularly developed in
Rhytida, which has such a short visceral mass
and lung, as п the Acavidae. The kidney is not
differentiated internally (RR2). The ureteric
tube is not closed farther than the upper ex-
tremity of the kidney in three genera of five
dissected. It is closed as far as the pneumos-
tome in the slug Schizoglossa (limacization),
and in Priodiscus. The cerebral commissure is
short (CC3). The cerebro-pedal connectives
are subequal and vary in length (CPR2, CPD1,
CPD2, CPD3). The pleural ganglia are close to
the pedal ganglia (PLD1, PLG1). The visceral
ganglion lies in the median plane (VG2) or, in
Priodiscus, on the left side (VG3). The parietal
and visceral ganglia form a single mass sep-
arated from the pleural ganglia by distinct con-
nectives, as in the acavid Caryodes and in the
Streptaxidae (PAD1, PAG2). However, except
in Priodiscus, the parieto-pleural connectives

ire much shorter than those in the Streptaxi-
Ihe three ganglia might seem fused to-
gether (FG1, FG3).

Character states of the CCA (excluding Pri-
odiscus because of its problematic affinities):
BM2 OC1 SC1 PS2 IL2’ LR3 UR2 RR2 ССЗ
CPD1 CPR2 PLD1 PLG1 VG2 PAD1 PAG3
FG:

PHYLOGENY AND CLASSIFICATION

Principles and methods of classification of
Stylommatophora

Phenetic nature of classical classifications:
Although not identical to any current classifi-
cation, the phenetic classification of the
stylommatophoran families presented in Text-
fig. 20 is familiar to a systematist of land
snails. Comparison with Appendix A, in which
the sequence of families follows Zilch's
classification (1959—1960), shows that the
members of the Holopoda, Aulacopoda and
Orthurethra cluster together in a somewhat
similar order. Most families of which the po-
sition seems aberrant in Text-fig. 20 belong
either to the Mesurethra, as might be ex-
pected when not overweighting the absence
of a ureteric tube, or to the never-defined
trash-can group of the Holopodopes formed
by Baker by lumping together those families
not clearly belonging to the Aulacopoda,
Holopoda or Orthurethra. Although | have not
done so, | have little doubt that any classifi-
cation produced by the Pilsbry-Baker school
could be obtained easily by the method used
to obtain Text-fig. 20 by weighting various
characters. The similarity of current classifi-
cations to this phenetic classification might

‘be surprising, inasmuch as the latter was

built with 31 characters of which only one, the
degree of closure of a ureteric tube, has been
explicitely used before.

On the other hand, it seems impossible to
formulate any clear, precise diagnosis of the
suprafamilial groups proposed by the Pilsbry-
Baker school. When by chance a character
might seem diagnostic, its mention should
include a list of exceptions. The relatively
great length and the internal structure of the
orthurethran kidney, without any mention of
the absence of a ureteric tube which is not
general within the group, constitutes an ex-
ception, but its use was unfortunately aban-
doned after Pilbry (1900a). The synthetic
work of Boss (1982) is illustrative of this
situation, in which it is impossible to find the

STYLOMMATOPHORAN SYSTEMATICS 73

position of a given snail or slug by using his
diagnoses, of which each element is repeated
in several diagnoses. The reason for this ap-
parent confusion lies not in the quality of the
work, but in the empiricism of earlier authors
who never defined the rationale upon which
they based their classifications and never de-
fined clearly the suprafamilial groups that they
proposed (e.g. Baker, 1955; Solem, 1978).

It may be deduced from these two remarks
that first, the classifications elaborated by the
Pilsbry-Baker school are basically phenetic,
although originally based on outgroup com-
parison (Pilsbry, 1896, 1900a) and secondly,
the phenetic nature of his classification was
not clearly understood and controlled by
Baker himself, who implicitly used characters
that he had seen but never mentioned and
possibly never explicitly considered. This ap-
proach resulted in a classification that is rea-
sonably stable and resistant to the inclusion
of new taxa, but which hardly can be under-
stood and handled without personal experi-
ence in land snail anatomy.

The principle upon which current classifica-
tions are built being identified, it should of
course be possible to improve the method
and, by studying more taxa and more charac-
ters than was done here, to obtain a classifi-
cation that would be reasonably stable what-
ever the phenetic distance used. Analysis of
the nodes of such a tree would allow the ex-
plicit identification of the characters that dis-
criminate the groups in a more satisfying
manner than presently. This way seems safe,
but not very attractive: even without consider-
ation of the arguments tending to show that a
phylogenetic classification is more stable than
a phenetic classification (Wiley, 1981), | pre-
fer to try to build a phylogenetic classification
of land snails that reflects as far as possible
the evolutionary history of the group.

Principles of phylogenetic classification:
Hypotheses about the characters useful for
phylogenetic purposes, about the direction of
evolution of these characters and about their
ancestral states having been accepted in the
preceding discussion, the phylogenetic rela-
tionships of the hypothetical common an-
cestors (CCA) of the various families can now
be discussed. However, the members of only
one group, the Orthurethra, share a syn-
apomorphy sensu stricto, 1.е. а seemingly
uniquely derived character state, in kidney
morphology; in later discussion, synapomor-
phy is used sensu lato for derived character

states shared by members of a supposedly
monophyletic group, but not necessarily un-
known in other groups. This synapomorphy
s.s. is confirmed by their morphological ho-
mogeneity, which appears even when char-
acters that are not correlated in the other sty-
lommatophoran groups are used (functionally
independent characters?). Within the Or-
thurethra and among the other stylommato-
phoran groups, there is virtually no character
state which can be used as a synapomorphy
sensu stricto: nearly every apomorphic state
occurring in two families also occurs else-
where, without giving by itself any indication
of common ancestry, and the various trees
that can be constructed with the various char-
acters are so incongruent that hypotheses
about monophyly can hardly be proposed on
such a basis. This result suggests that, at
least at the morphological levels discussed
here, parallel evolution is the rule rather than
the exception, and illustrates Eldredge's
(1979:170, fn) statement, “As a result of the
relatively recent production of a plethora of
cladograms, parallelism turns out to be a far
more common evolutionary phenomenon
than even most of its more ardent aficionados
had thought.” In theory the classical Henni-
gian method avoids the use of distance; in
practice it reveals the generality of parallel ev-
olution and, as discussed below, the need to
minimize the length of the phylogenetic trees,
not because the shortest tree is necessarily
the best, but because nobody has yet found
another solution.

A first approach to this problem was ex-
plored by the late Pierre Delattre (unpub-
lished). In the data set, every object is repre-
sented by a sequence of indices as in the
matrices commonly used in numerical phylo-
genetics. Each index represents a character of
which the states are symbolized by values
ranging from 1 (plesiomorphic) to n (apomor-
phic). An object having all its indices at most
equal to those of another object, and having at
least one index lower than the equivalent index
of this other object is a possible ancestor of this
other. It is therefore possible to reconstruct all
the possible ancestors of any object, as long
as reversal is not allowed. For example, an
object denoted 121 has six possible ances-
tors, 111,011,001, 101, 100 and 000. Itis then
possible to construct the possible closest com-
mon ancestor (= CCA) of any pair of objects
by taking the lowest value of each index found
in both, and all the possible ancestors of this
CCA are also common possible ancestors of

74 TILLIER

this pair. For example, the objects noted 123
and 321 have 121 as CCA, and the six objects
listed above are also possible common an-
cestors of this pair. All possible phyletic rela-
tionships in any data set can be determined in
this way and, if the set is not too large, a
network diagram can be constructed to show
for discussion any possible path and eliminate
most of the paths until a tree is obtained (if
hybridization is not allowed).

A few attempts with real data showed that
the number of possible relationships is lower
than one might have feared, and that this
method is convenient for small data sets. Un-
fortunately, even if most of the possible trees
are eliminated the method cannot be applied
directly here, where, with 50 taxa and 17
characters, the number of possible phyletic
paths is too large to allow discussion of each.
However, it can be applied to a data set cor-
responding to a tree of which the length has
already been minimized.

Minimizing the length of a diagram of phy-
logenetic relationships is not necessarily par-
simony, contrary to what many systematists
seem to think: parsimony is reduction of the
number of ad hoc hypotheses necessary to
justify the choice of a given path, but not the
reduction of the length of this path, which is
minimization. The minimization of the length
of phylogenetic trees is generally justified on
the grounds that each independent appear-
ance of an apomorphic state requires an ad
hoc hypothesis; and that consequently, re-
ducing the number of independent appear-
ances of derived character states, equivalent
to reducing the length of the tree, is parsi-
mony (Wiley, 1981).

The problem raised by this interpretation is
that the only coherent and available theory of
evolution is the synthetic theory, which at-
tributes the differences between taxa to a
combination of chance and natural selection
(adaptation), but does not establish any direct
relationship between evolution and phylogeny
(Tillier, 1986); it does not justify the minimiza-
tion of the number of independent derivations
of a given character state but, on the contrary,
provides a single general ad hoc hypothesis
that explains all convergences by adaptation
and therefore justifies disregarding the num-
ber of independent derivations of a given
character state in constructing a phylogenetic

liagram. There is, therefore, a contradiction
tween the current theory of evolutionary
hanisms and the practice of numerous
genetic 'stematists, which perhaps

has more to do with blindness than with par-
simony: a contradiction can hardly be re-
solved by ignoring one of its terms.

Two attitudes seem to me consistent under
such conditions. First, one can accept the
synthetic theory only, and construct phyloge-
netic diagrams by using minimization. The re-
sulting diagram represents a synthesis of
chance and environmental conditions, but
there is no reason to believe that it reflects the
phylogenetic relationships unless a complete
series of fossils is available (Utopia). The taxa
constituted in this manner are ecological
groups that are probably para- or polyphy-
letic; at least the theory does not provide any
argument in favor of their monophyly. Sec-
ondly, one can introduce an ad hoc hypothe-
sis that, without refuting the current results as
to the mechanisms of microevolution, justifies
minimization. | propose the following hypoth-
esis: the closer the phylogenetic relationships
of two taxa are (1.е., the greater the ratio of the
number of their common ancestors and of the
total number of their ancestors is), the greater
the probability for the appearance of the same
derived character states in the two taxa.

This postulate reminds one of Vavilov's law
of homologous series in variation (1922). Cain
(1982:393) asserts it at least at infrageneric
level (*. . . it is precisely in those stocks which
are already closely allied that convergence is
most likely to occur . . .”). As discussed by
Gould (1983), Vavilov's law is not Darwinian,
or at least not neo-Darwinian: in the fifth chap-
ter of “Origin of Species”, Darwin explicitly
used analogous morphoclines as an argument
for a common origin, and therefore implicitly
considered their states as homoiologous as
construed by Sneath and Sokal, 1973); fur-

‚ thermore, Vavilov's law does not allow inno-

vation. The postulate presented here is no
more neo-Darwinian, and supposes that all
variations are not equiprobable, but that the
members of every lineage vary within a well-
defined potential for variation, which is modi-
fied to only a limited extent by each speciation
event. Such an ad hoc hypothesis justifies the
use of phenetic classifications in the search for
phylogenetic relationships, and explains the
general similarity of phenetic and phylogenetic
classifications of the same group.

Method of phylogenetic classification: The
method employed here consists in:

1. Construction of a short tree of phyletic
relationships among the CCAs of the various
families;

STYLOMMATOPHORAN SYSTEMATICS 1
19 MAS VER TIGIMD AE ORCULIDAE PARTULIDAE
> En р
| г | it
| |
18 | | | |
| | | |
17 | PYRAMIDULIDAE AAA IEA 1 :
| | A AMASTRIDAE |
ie 3 ea Г pa ENIDAE
| ye р |
15 | CHONDRINIDAE COCHLICOPIDAE 4 3
| | Van
| ; | | |
14 ACHATINELLIDAE | |
| | |
| |
18 | VALLONIDAE [Meri A 5
о
(a ee |
NF De |
7 Е. |
о С OO Е
ca J oe |
Inte A pS En OR et LE ee |
| E cs 6
11 =
BEE Were asus а ВЕ 1
| |
10 A A EE ee ee e ze]

TEXT-FIG. 21. Network of possible phyletic relationships among orthurethran families. Indices of nodes,
Appendix F. Numbers to the left correspond to sums of indices of nodes:

Possible ancestor-descendant relationships between nodes:

Achatinellidae: Vertiginidae, Pyramidulidae, Pupillidae, =3=.

Amastridae: Orculidae.

Cochlicopidae: Amastridae, Vertiginidae, Orculidae, =1=.

Valloniidae: Vertiginidae.
=1 = : Vertiginidae, Orculidae.

=2= : Orculidae, Partulidae, Amastridae.

=3= : Pyramidulidae, Pupillidae.

=4= : Orculidae, Partulidae, Amastridae, Enidae, =2=.

=5= : Partulidae, Amastridae, Cochlicopidae, Vertiginidae, Orculidae, Enidae, =1 2 4

=6= : Partulidae, Amastridae, Cochlicopidae, Vertiginidae, Orculidae, Chondrinidae, 'Valloniidae, Enidae,
ls, ]2=,) =] 5. =5

=7= : Achatinellidae, Pyramidulidae, Vertiginidae, Pupillidae, Valloniidae, =3=.

2. Determination of the other possible kin-
ship relationships among the nodes without
calculating additional nodes;

3. Choice of paths;

4. Possibly, return to step one using the
CCAs of the various branches retained after
step three as the data set; the necessity for
this step results from the algorithm used, in
which the stability of the nodes is less closer
to the root of the tree;

5. Construction of a tree representing the
phyletic relationships that have been retained.

Before the last step, some objects are in
the position of ancestors: This point does not
seem very important, because first, all the ob-
jects actually handled are abstractions; and
second, the only really important point is the
relative position of the actual taxa. The topo-
logical relationship of three points A, B, C is
the same if the points are located along a line

ABC or at the extremities of the branches of a
tree (A(BC)) or ((AB)C), i.e. B is between A
and C (the position of the root is here deter-
mined independently, because reversal is not
allowed).

The data set is constituted by the CCAs of
the various families, as defined above (Ap-
pendix F). Characters of which the plesiomor-
phic state is not an extremity of a morpho-
cline, denoted 1, 2, 3 or 1, 2’, 3’ in Appendix
F, were recoded in two columns: the plesio-
morphic state was recoded 11; one part of the
morphocline was coded 12, 13, etc., and the
other part of the morphocline was coded 21,
31, etc. The distance is the Manhattan dis-
tance (Wiley, 1981).

Two algorithms have been tested (and the
PENNY program of the PHYLIP package: v.
infra). Only the one used further is here de-
scribed in detail. The values of each index

76 TILLIER

VALL AE HONDR INIDAE
= / Wi à
3 m— a 3 ya N
ae / Va N
y / A
вл я Wi
EA | Sat
х
N A

\ ENIDAE

ARTULIDAE COCr
PAGI
N
N \
\ \
\
\ mo:

iLICOPIDAE AMASTRIDAE VERTIGINIDAE JRCUL IDAE

\
Ne N

TEXT-FIG. 22. Phylogenetic tree of the Orthurethra. Branches correspond to paths of Text-Fig. 21 that were

retained.

were determined by outgroup comparison,
and every CCA was defined by the lowest
values of the indices in all members of the
group of which the CCA was sought. The se-
quence of steps is:

1. Find the object of which the sum of the
indices is the largest;

2. Find a second object such that the CCA
of both objects is close to the first one;

3. When several objects fit the criteria of
step two, choose the object such that the sum
of the distances between the two objects to
be retained and their CCA is the least;

4. Replace the two objects retained by their
CCA;

5. Go back to step one.

In cases in which equal distances remain af-
ter step three, both solutions are indicated to
allow exploration of both of them (this did not
happen; see Appendices).

The principal inconvenience of this algo-
rithm is that reversal is not allowed. On the
other hand, it is fast, and permits the handling
of very large tables with a very small micro-
computer. The branches farthest from the
root are relatively stable, and well resist
changes in characters or in objects; but the
lowest nodes, near the root, do not. This is
why two runs are necessary for large tables:
the first one allows choice of the main
branches after the next step described below,
and the second takes the CCAs of the groups
retained after the first run.

The second algorithm tried uses the dis-
tances between pairs of objects and their
CCAs: it starts with two objects such that the
sum of the distances between them and their
CCA is minimal, replaces the two objects by
their CCA and starts again. lt uses much more
computer time than the first algorithm because
all the distances must be recalculated every
time, and tends to root those objects that are
a little farther from the others than the average
distance directly to the basal node of the tree.
As a result, the tree is generally longer than
that built with the former algorithm. However,
the final result is not different after the next
step, but requires much more time.

To find non-minimum paths, the table in-
cluding the objects and nodes of the minimum
tree is then scanned, to determine all possible
filiation relationships among the nodes, such
a relation being possible when one of the two
objects or nodes under consideration has all
its indices at most equal to those of the other
object or node, and at least one index lower.
This procedure allows one to draw a network
of possible phylogenetic relationships, consti-
tuted by the tree obtained in the former step
as well as the new paths so determined.

In order of preference, the criteria used to
justify the choices among several possible or-
igins are these:

1. Occurrence of a synapomorphy $.5.;

2. Suppression of a node representing an
object not included in the primitive data set,

STYLOMMATOPHORAN SYSTEMATICS ИИ

without increase in the number of multiple
nodes;

3. Occurrence of a rare homoplasy;

4. Similarity in characters of which the func-
tional value has not been demonstrated;

5. Overall similarity;

6. Geographical distribution.

In practice the choice is more difficult when-
ever the sum of the indices of the objects is
close to either the theoretical maximum (ad-
vanced apomorphy in many characters), or
the theoretical minimum, which is the CCA of
all the objects and the root of the tree (many
plesiomorphic states). This difficulty is fore-
seeable in theory: the late Pierre Delattre (un-
published) showed that, when one makes a
system in which all possible homoplasies are
admitted, including all the possible evolution-
ary paths from plesiomorphic to apomorphic
state for all characters, the taxonomic objects
represented along the most numerous paths
are those that are relatively close to the ob-
jects of which the sum of the indices is the
theoretical maximum or the theoretical mini-
mum. If synapomorphies cannot be found, the
solution might be to seek characters of which
the states are plesiomorphic in the objects
having many apomorphies, and apomorphic
in the objects having many plesiomorphies, to
bring the sum of the indices close to the me-
dian between the theoretical maximum and
minimum, and to provide the best possible
discrimination between the possible paths.

The data sets of Appendix F were treated
with the program described, and with the
PENNY program of the PHYLIP package,
version 2.5 (1984) (Felsenstein, 1982). The
latter builds all the possible trees and picks
out the shortest one(s); it was run under the
same assumptions, ¡.e. reversal not being al-
lowed, on an HB Mini6 computer.

For the Orthurethra, the shortest tree of
60,000 (3.34% of all possible trees) is 32
steps long and has only one node that was
not identified in Text-fig. 21, i.e. the Chondrin-
idae were placed as the sister group of the
Pupillidae- Pyramidulidae- Achatinellidae-
Valloniidae group. The minimum tree found
after the process described above (and be-
fore discussing non-minimum paths, dashed
lines in Text-fig. 21), is 33 steps long but re-
quired one-tenth of the computer time on a
much smaller computer (ApplellE 64K vs.
Mini6). The tree retained here after discus-
sion (Text-fig. 22) is 35 steps long.

Congruence among the shortest trees
found for the non-Orthurethra (five trees 178

steps long, one 163 steps long among 90,000
trees) is worse, but it is no better with any of
the classifications of the Pilsbry-Baker school.
Among the families of which the position might
seem the most dubious in the classification
presented below, the Helicarionidae are the
sister group of the Camaenidae in the six
shortest trees found; the Sagdidae and Hap-
lotrematidae are grouped with the Helmintho-
glyptidae and Polygyridae in the six trees; the
Corillidae are grouped with the Acavidae and
Rhytididae in the six trees; the Oreohelicinae
are grouped with the Acavidae, Rhytididae
and Corillidae in three trees and with some
clausilioid families in three trees; the Ammo-
nitellinae are never far from clausilioid fami-
lies, but form a monophyletic group with some
of them in only two trees and never form a
monophyletic group with any helicoid family;
the Succineidae are included in a monophy-
letic group lumping some families of aulaco-
pod slugs and semislugs, as in the minimum
tree (dashed lines, Text-fig. 23) in the six
trees.

Paleobiogeography: As far as possible, the
phylogenetic trees presented below have
been confronted with paleogeographical data
(which in some cases have been used as a
test) extracted from the following works: for
the Paleozoic: Morel and Irving, 1978; Sco-
tese et al., 1979; for the Mesozoic and Cen-
ozoic: Smith and Briden, 1977; Owen, 1983;
for the possibile opening of the Pacific Ocean:
Nur and Ben Avraham, 1981.

In every case the paleobiogeographical hy-
potheses presented below must be used with
caution, because: first, some stylommatopho-
ran groups are probably very ancient (Solem
8 Yochelson, 1979; Solem, 1981), but only
very few fossils older than Eocene are avail-
able; secondly, even newer fossils are not
common, and are virtually unknown outside
Europe and North America; thirdly, we know
nothing of the anatomy of fossil land snails,
and convergence in shell characters is likely;
and finally, Paleozoic paleogeographical data
are not precise, and to a large extent still un-
der discussion.

Phylogeny and classification of Orthurethra

The occurrence of a synapomorphy, ¡.e. a
kidney both long and divided internally into
two distinct regions, allows the independent
analysis of the Orthurethra.

TILLIER

|
|
|
|
3va11409 |
р
QISNVTO PT фу,
|
|
|
|
р

901835

N19113H03H0 01190909 П

vin

OIHOHdOIVHOHLY

(OH
etz Al
| Е RE Е КС A Е -
| CA =
| ny |
1 E: м
A en A Nez RS ur ve |
' | | |
| | |
| | | '
| |
| р
| |
р | | лее
| | |
| |
| | | | |
a QZ AlLINOZ OINTSH
eek | |
| SI
A г
r 02 NITIBLINONNY {| | гг
| | | | |
1 ===
| GiAvOY | | GidOHVHO ike a
| | | { |
| Г | | Г \ | |
A ape Ll al ea el |
| | 1: | | | N |
|
= | | t fl | | 777 офнаммтэн OINIVWWO
St | 1 | | | 1 | | р ;
т | | | | | | | 1 1 BR
| ее at 6771 ansia | | | OINOIHV9113H G | a
| | | La | | me | \
N | | za: | annoon3 | | GINANENS E | J, SHADANOA
| \ E N
| | | р ! | et |
= aiNoIHv GILNOGOGN3 / 7777 "О
bt = | | | ŒIHdOHLS AS 1" я | |
Г
| в 1 р | |
| GIGILAHE | a | | | | | OINIVBAOV88
IAS | | | | |
| | | | | | | H fae GILVN310714VH
anne | | | | | | | у
| en, een,
aNv310 | | r G aiaovs
| | 1 |
| | | |
aianioons E 2 139vWHvd i $ GIHAHONOHDOHL GLONNd a'IVSSNH34 т |
| | | axvLd3HLS
(est | |
| | | | |
(Ir: | |
Е] | | OINILWHOW
ll] | |
| | | | НИ.
от | GINIALIA | el
| |
| |
| |
|
|
|
|
|
|

OQISAWOTHd

ус

6£

+
+

STYLOMMATOPHORAN SYSTEMATICS 79

TEXT-FIG. 23. Network of possible phyletic relationships among non-orthurethran families. Indices of nodes, Appendix F.
Numbers to the left correspond to sums of indices of nodes.

Possible ancestor-descendant relationships between nodes:
Charopidae: Endodontidae, Punctidae, Systrophiidae, =7=.

Arionidae: Philomycinae, Athoracophoridae, Vitrinidae, Milacidae, Limacidae, =1=, =2=, =3=.
Zonitidae: Athoracophoridae, Punctidae, Discidae, Philomycinae, Vitrinidae, Milacidae, Limacidae, Trocho morphidae, Euconulidae, Heli-
carionidae, Ferussaciidae, Subulinidae, Achatinidae, Oleacinidae, Hap lotrematidae, Streptaxidae, Sagdidae, Bradybaenidae, =1=, =2=

=3=, =4=, =5=, =9=, =10=, =16=, =18=, =19=.

Euconulidae: Punctidae.

Helicarionidae: Athoracophoridae, Philomycinae, Vitrinidae, Limacidae, Achatinidae, Haplotrematidae, Sag didae, =1=, =3=, =4=.
Subulinidae: Achatinidae.

Acavidae: Athoracophoridae, Rhytididae.

Cerionidae: Vitrinidae.

Polygyridae: Sagdidae.

Camaenidae: Achatinidae.

Ammonitellinae: Succineidae, Athoracophoridae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Mi lacidae, Limacidae, Ferussaciidae,
Achatinidae, Oleacinidae, Bulimulidae, Streptaxidae, Sagdidae, =1=, =2=, =3=, =5=, =6=, =8=, =11=, =12=, =17=.
Bradybaenidae: Achatinidae.

Helminthoglyptidae: Milacidae, Limacidae, Achatinidae, Polygyridae, Sagdidae, Bradybaenidae, =2=, =4=, =15=.

Helicidae: Succineidae, Athoracophoridae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Milacidae, Limacidae, Ferussaciidae, Acha-
tinidae, Oleacinidae, Bulimulidae, Streptaxidae, Sagdidae, =1=, =2=, =3=, =4=, =5=, =6=, =8=, =10=, =12=, =17=.
= 1= :Athoracophoridae, Philomycinae.

= 2= : Milacidae, Limacidae.

= 3= : Athoracophoridae, Philomycinae, Vitrinidae, =1 =.

= 4= : Achatinidae, Sagdidae.

= 5= :Ferussaciidae, Streptaxidae.

= 6= : Succineidae, Milacidae, Limacidae, =2=.

= 7= :Punctidae, Systrophiidae.

= 8= : Athoracophoridae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Milacidae, Limacidae, =1=, =2=, =3=.

= 9= :Punctidae, Trochomorphidae, Euconulidae.

=10= : Ferussaciidae, Achatinidae, Streptaxidae, Sagdidae, =4=, =5=.

=11= : Oleacinidae, Rhytididae, Streptaxidae.

=12= : Succineidae, Philomycinae, Vitrinidae, Milacidae, Limacidae, Bulimulidae, =2=, =6=.

=13= : Philomycinae, Vitrinidae, Limacidae, Ferussaciidae, Oleacinidae, Urocoptidae, Rhytididae, Streptaxidae, =5=, =11=

=14= : Vitrinidae, Cerionidae, Oreohelicinae.

=15= : Limacidae, Achatinidae, Polygyridae, Sagdidae, Bradybaenidae, =4=.

=16= : Athoracophoridae, Philomycinae, Vitrinidae, Limacidae, Helicarionidae, Ferussaciidae, Achatinidae, Haplotrematidae, Streptaxidae,
Sagdidae, =1=, =3=, =4=, =5=, =10=.

=17= : Succineidae, Athoracophoridae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Milacidae, Limacidae, Bulimulidae, =1=, =2=,
=3=, =6=, =8=, =12=.

=18= : Athoracophoridae, Philomycinae, Vitrinidae, Limacidae, Helicarionidae, Ferussaciidae, Subulinidae, Achatinidae, Oleacinidae, Hap-
lotrematidae, Streptaxidae, Sagdidae, Bradybaenidae, =1=, =3=, =4=, =5=, =10=, =16=.

=19= : Punctidae, Discidae, Vitrinidae, Limacidae, Trochomorphidae, Euconulidae, =9=.

=20= : Athoracophoridae, Philomycinae, Vitrinidae, Limacidae, Ferussaciidae, Oleacinidae, Acavidae, Cerionidae, Urocoptidae, Rhytididae,

Streptaxidae, =1=, =3=, =5=, =11=, =13=.

=21 = : Athoracophoridae, Philomycinae, Vitrinidae, Milacidae, Limacidae, Helicarionidae, Achatinidae, Haplotrematidae, Polygyridae, Sag-
didae, Camaenidae, Bradybaenidae, Helminthoglyptidae, =1=, =2=, =3=, =4=, =15=.

=22= : Athoracophoridae, Philomycinae, Vitrinidae, Milacidae, Limacidae, Helicarionidae, Ferussaciidae, Subulinidae, Achatinidae, Olea-
cinidae, Haplotrematidae, Streptaxidae, Polygyridae, Sagdidae, Camaenidae, Bradybaenidae, Helminthoglyptidae, =1=, =2=,

=3=, =4=, =5=, =10=, =]5=, =16=, =18=, =21=.

=23= : Athoracophoridae, Philomycinae, Vitrinidae, Parmacellidae, Limacidae, Ferussaciidae, Clausiliidae, Oleacinidae, Bulimulidae, Cerion-
idae, Urocoptidae, Rhytididae, Streptaxidae, Oreohelicinae, =1=, =3=, =5=, =11=, =13=, =14=.

=24 = : Succineidae, Athoracophoridae, Endodontidae, Charopidae, Punctidae, Discidae, Philomycinae, Vitrinidae, Zonitidae, Milacidae,
Limacidae, Trochomorphidae, Euconulidae, Helicarionidae, Ferussaciidae, Subulinidae, Achatinidae, Oleacinidae, Systrophiidae, Hap-
lotrematidae, Rhytididae, Streptaxidae, Polygyridae, Sagdidae, Camaenidae, Bradybaenidae, Helminthoglyptidae, =1=,=2=,=3=,

=4=, =5=, =6=, =7=, =9=, =10=, =11=, =15=, =16=, =18=, =19=, =21=, =22=.

=25 = : Athoracophoridae, Philomycinae, Vitrinidae, Parmacellidae, Limacidae, Ferussaciidae, Clausiliidae, Oleacinidae, Bulimulidae, Cerion-
idae, Urocoptidae, Rhytididae, Streptaxidae, Corillidae, Oreohelicinae, =1=, =3=, =5=, =11=, =13=, =14=, =23=.

=26 = : Succineidae, Athoracophoridae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Milacidae, Limacidae, Ferussaciidae, Oleacinidae,
Acavidae, Bulimulidae, Cerionidae, Urocoptidae, Rhytididae, Streptaxidae, Ammonitellinae, Oreohelicinae, =1=, =2=, =3=, =5=,
=6=, =8=, =11=, =12=, =13=, =14=, =17=, =20=.

=27= : Succineidae, Athoracophoridae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Milacidae, Limacidae, Helicarionidae, Ferussaci-
idae, Subulinidae, Achatinidae, Oleacinidae, Bulimulidae, Haplotrematidae, Streptaxidae, Polygyridae, Sagdidae, Camaenidae, Brady-
baenidae, Helminthoglyptidae, Helicidae, =1=, =2=, =3=, =4=, =5=, =6=, =8=, =10=, =12=, =15=, =16=, =17=,
=18=, =21=, =22=.

=28= : Succineidae, Athoracophoridae, Endodontidae, Punctidae, Arionidae, Philomycinae, Vitrinidae, Parmacellidae, Milacidae, Limacidae,
Ferussaciidae, Clausiliidae, Oleacinidae, Acavidae, Bulimulidae, Cerionidae, Urocoptidae, Rhytididae, Streptaxidae, Corillidae, Am-
monitellinae, Oreohelicinae, =1=, =2=, =3=, =5=, =6=, =8=, =11=, =12=, =13=, =14=, =17=, =20=, =23=,
=25=, =26=.

=29= : Succineidae, Athoracophoridae, Endodontidae, Charopidae, Punctidae, Discidae, Arionidae, Philomycinae, Vitrinidae, Zonitidae, Par-
macellidae, Milacidae, Limacidae, Trochomorphidae, Euconulidae, Helicarionidae, Ferussaciidae, Subulinidae, Achatinidae, Olea-
cinidae, Bulimulidae, Systrophiidae, Haplotrematidae, Rhytididae, Streptaxidae, Polygyridae, Sagdidae, Camaenidae, Ammonitellinae,
Oreohelicinae, Bradybaenidae, Helminthoglyptidae, Helicidae, =1=, =2=,=3=,=4=,=5=,=6=,=7=,=8=,-=9=,=10=,
==, =12=, =15=, =16=, -17=, =18=, =19=, =21=, =22=, =24=, =27=.

80 TILLIER

Phylogeny: As already emphasized, my
sampling of the Orthurethra is probably not
sufficient and | am consequently reluctant to
trust my results. Taking more species belong-
ing to more genera into account would possi-
bly allow classes (and therefore character
states) of lesser amplitude and increase the
reliability of the results.

The 11 family-level taxa analyzed share
one synapomorphy and several symplesio-
morphies that have been eliminated from the
data set: synapomorphic length of the kidney
(LR2'), symplesiomorphic spheroid buccal
mass (BM1), absence of an oesophageal crop
(OC1), relatively short intestine (IL1), absence
of ureteric tube (UR1), position of the right
parietal ganglion (PAD1). The characters re-
maining are: gastric crop shape (SC), degree
of differentiation of the gastric pouch (PS), and
the remaining characters of the nerve ring
(CG, СРО, CPR; PED, PEG, VG; PAG; FG).

Possible phyletic relationships are shown in
Text-fig. 21, the data set and the coordinates
of the calculated nodes are given in Appendix
F, and the list of all the possible filiations be-
tween the nodes is given in the caption of
Text-fig. 21.

A first remark, which might seem surprising
because the Achatinellidae and Partulidae
have often been considered the most “prim-
itive’ of the Stylommatophora (Boss, 1982), is
that the Valloniidae are the most plesiomor-
phic for the largest number of characters and
might have been isolated very early. They
also exhibit a plesiomorphy unique among the
Stylommatophora, i.e. the position of the gen-
ital opening relatively far back. As discussed
by Watson (1915), the same arrangement is
probably related to carnivory and therefore
apomorphic in some rhytidids, which does not
seem to be the case in the valloniids. These
numerous plesiomorphies of the valloniidae
might result from their trend to paedomorpho-
sis; however this is probably not the case,
because of the position of the genital pore,
which is plesiomorphic (Tillier, 1984b) but not
paedomorphic, the genital apparatus devel-
oping very late in ontogeny without any de-
scribed displacement of the genital opening.

The early differentiation ofthe Chondrinidae
is also likely if, as suggested above, the family
includes two groups of which one is Holarctic
and the other exclusively Gondwanian.

Two paths can be chosen without too much
hesitation among the possibilities shown in
Text-fig. 21. The first is the common origin of

the Cochlicopidae and Amastridae. This op-

tion allows the suppression of node 2, and
groups together not only the Cochlicopidae
and Amastridae (classical group), but also the
Partulidae and Enidae. The latter group cor-
responds to the classification by Solem
(1978), and can be accepted because the
trend to visceral mass shortening occurs in
these two families alone among the Orth-
urethra; furthermore, it is biogeographically
coherent inasmuch as the Central and South
Pacific Partulidae might be derived from some
Oriental or western Pacific Cerastuinae. The
second is the common origin of the Pupillidae
and Pyramidulidae. lt seems clear to me that
these two groups differ only in their shell
shape.

Monophyly of the Valloniidae and Vertig-
inidae can be rejected inasmuch as, within
the Orthurethra the position of the right pleu-
ral ganglion close to the right cerebral gan-
glion (PLD2) appears to be a synapomorphy
of the group that includes the Valloniidae but
not the Vertiginidae.

My knowledge of the Orthurethra is insuffi-
cient to allow me to discuss the other possible
relationships. In particular | have no support
other than previous classifications to reject
the monophyly of the Orculidae, Amastridae
and Cochlicopidae. The tree constructed on
the basis of these accepted hypotheses is
shown in Text-figure 22.

Classification: The tree retained divides
the Orthurethra into two groups, which are de-
fined and subdivided on the basis of charac-
ters of the nervous system: digestive tract
morphology appears only once as a synapo-
morphy, not unique, uniting the Pupillidae and
the Pyramidulidae. | do not trust this tree

enough to give names to these two groups,

and | shall call Group | the group including the
Pupillidae, Pyramidulidae, Achatinellidae and
Valloniidae; and Group Il the group including
the Orculidae, Vertiginidae, Amastridae, Co-
chlicopidae, Partulidae, Enidae and Chon-
drinidae (Text-fig. 22).

Group | is based on the synapomorphic
proximity of the right pleural ganglion to the
right cerebral ganglion and on the primary
shortening of the cerebral commissure.
Group Il is based on the synapomorphic
shortening of the cerebro-pedal connectives.
Apart from these character states, the distinc-
tion between the two groups represents two
different patterns in nerve ring shortening.
Group | 15 characterized by shortening of the
nerve ring primarily through shortening of the

STYLOMMATOPHORAN SYSTEMATICS 81

cerebral commissure and of the cerebro-
pleural connectives, the cerebro-pedal con-
nectives keeping their plesiomorphic length
and asymmetry (except in the Valloniidae):
shortening is primarily in the dorsal portion of
the nerve ring. Group Il is characterized by
primary shortening of the lateral connectives,
whereas the upper portion of the nerve ring
primarily keeps its plesiomorphic arrange-
ment. Most individual apomorphic states are
found in both groups: it is their succession
and the resulting evolutionary sequence in
the arrangement of the nervous system that
define the groups, which would ultimately
converge in a single arrangement if evolution
were pursued further in compaction of the
nervous system.

Speculations on history of Orthurethra: If
the taxonomic distinction established by
Solem and Yochelson (1979) between the
Carboniferous Dendropupinae and Anthra-
copupinae, attributed by them respectively to
the Enidae and Achatinellidae, is reliable,
then by implication groups | and Il were al-
ready differentiated. Their differentiation
through isolation in Gondwana and Laurasia,
before the formation of Pangea, is then pos-
sible. If it occurred, group II is the best can-
didate for differentiation in Gondwana, be-
cause it contains the only orthurethran group
that is exclusively Gondwanian (Hypselosto-
matinae, Fauxulus, Gibbulina, Ulpia). This hy-
pothesis coincides with the early differentia-
tion of the Chondrinidae, shown in Text-fig.
22. The Chondrinidae appeared in Europe
during the Eocene, that is, later than most of
the other orthurethran families, which are at
least as old as Paleocene; this might be due
to late migration northward through the Pan-
amanian region, whereas the other families
belonging to group Il would have been differ-
entiated in the Laurasian region after the
opening of the Tethys Sea.

The other families of group Il, Amastridae
and Partulidae excepted, did occur in Europe
during the Paleocene. One can therefore sup-
pose that they occurred also in North America,
where they all became extinct (unless one ad-
mits that the only Recent species now occur-
ring in North America, Cochlicopa lubrica, al-
ready occurred). Some Enidae, close in shell
morphology to Recent Cerastuinae, did ap-
pear first in Laurasia, possibly during the Car-
boniferous (unless the Dendropupinae repre-
sent the stem-group of the non-chon-
drinid members of group II). During the Mi-

ocene this stock could colonize East Africa
through Arabia, and the Indo-Malayan Archi-
pelago down to Queensland, New Caledonia
and possibly Vanuatu (ancestors of the Re-
cent Cerastuinae). Some members of this
group and some Cochlicopidae colonized the
Central Pacific islands at a date difficult to
determine, where their descendants form re-
spectively the Partulidae and Amastridae. At
some time between the Eocene and the Mi-
ocene, the primitive stock of the Enidae had
been replaced in Eurasia by the Chondrulinae
and Eninae.

No element indicates the occurrence of any
family belonging to group | outside the Laur-
asian region and the Pacific earlier than re-
cently (dispersal of the Pupillidae is very easy
owing to their very small size, and might be
quite recent). No element allows the dating of
the differentiation of the Valloniidae (between
the Carboniferous and the Paleocene), un-
less one can find a synapomorphy of Anthra-
copupa with the Achatinellidae; in which case
the Valloniidae would be as old as Carboni-
ferous. The Recent forms of the Pupillidae-
Pyramidulidae group seem to have appeared
during the Eocene, but the actual date of the
separation between the ancestors of this
group and those of the Achatinellidae, and the
date of colonization of the Pacific Islands by
the latter remains a mystery.

Phylogeny and classification of
non-orthurethran Stylommatophora

Possible relationships of families and su-
perfamilies are represented in Text-fig. 23.
Only the paths to be discussed are shown, to
avoid confusion in the diagram (non-minimum
paths that result in multiple nodes generally
are without interest). The complete list of pos-
sible paths between the nodes of the primary
tree is given in the caption of Text-fig. 23.

Zonitoidea: The Zonitoidea include the
families Zonitidae, Trochomorphidae, Parma-
cellidae, Limacidae, Milacidae, Euconulidae,
Discidae and Arionidae (including the Philo-
mycinae).

Although the group did not occur in any of
the shortest trees produced either here (Text-
fig. 23, dashed lines) or by the PENNY algo-
rithm, it is probably not worth discussing the
grouping of the first five of these families,
which has long been admitted, and against
which | cannot at present think of any argu-
ment (Zonitidae, Trochomorphidae, Parma-

ZONITIDAE

TROCHOMORPHIDAE

DISCIDAE

+PHILOMYCINAE

TILLIER

ARIONIDAE PARMACELLIDAE MILACIDAE LIMACIDAE

LR5 (LIMACIZATION)

TEXT-FIG. 24. Minimum tree of possible phyletic relationships among zonitoid families (retained; ? =

position doubtful).

cellidae, Limacidae, Milacidae). | have no ar-
guments other than nomenclatural stabiilty
and geographical distribution in favor of this
group.

Inclusion of the families Euconulidae and
Discidae in the Zonitoidea is justified not only
by the scheme shown in Text-fig. 23, but also
by overall similarity (Text-fig. 20). Close rela-
tionships between the Euconulidae and
Zonitidae have already been suggested (Van
Mol 8 Van Bruggen, 1971), based on similar-
ity in some genital characters (glandular area
occurring along the vagina, and not along the
free oviduct as in the Helicarionidae).

Because the Philomycinae differ from the
Arionidae only in having more apomorphic
character states, and because both groups
share a synapomorphy in kidney morphology,
grouping them into a single family seems jus-
tified. Although members of this group can be
easily identified, analyzing their relationships
with other groups is far less easy. The Arion-
idae are grouped here with the Discidae
within the Zonitoidea because the Arionidae
and Discidae differ very little in characters un-
related to limacization.

Character states of the CCA of the Zoni-
toidea are plesiomorphic for all characters ex-

cept the closure of the ureteric tube, which is
always complete (UR4).

The tree built from the diagram of the pos-
sible relationships of the zonitoid families has
two groups (Text-fig. 24). One includes the
Zonitidae, Euconulidae and Trochomor-
phidae; the other includes the Discidae, the
Arionidae and the three other slug families.
The latter group is characterized by kidney
shortening and internal dedifferentiation

` (LR2, RR2), displacement of the right parietal

ganglion toward the visceral ganglion (PAD2),
and perhaps loss of the amatorial organ. The
synapomorphies uniting the four slug families
are only character states that are probably
related to limacization (LR5, PAG3); it is also
possible that the Discidae and Arionidae form
a monophyletic group. The families Parma-
cellidae, Limacidae and Milacidae share the
median position of the visceral ganglion
(VG2), and compaction in the left portion of
the visceral chain (PAG4). These three fami-
lies are placed there on the tree probably only
because limacization results in a shorter dis-
tance from the arionid slugs than from any
other zonitoid family; it is also possible that
there are two groups on this side of the tree,
i.e. the Апопасеа (Arionidae + Discidae) and

STYLOMMATOPHORAN SYSTEMATICS

HELICIDAE POLYGYRIDAE SAGDIDAE BRADYBAENIDAE

FG2'

RR2

HELMINTHOGL

TIC

\E CAMAENIDAE HEL ICARIONIDAE HAPLOTREMAT IDAE VITRINIDAE

IL2

PD3

TEXT-FIG. 25. Minimum tree of possible phyletic relationships among helicoid families (doubtful: see text).

the Limacacea (Limacidae + Milacidae +
Parmacellidae + ?Trigonochlamydidae); or,
possibly more consistently, the Discidae +
Arionidae constitute the sister group of all
other zonitoid families.

The other group, which might be called the
Zonitacea, includes the family Zonitidae,
which exhibits plesiomorphy in nearly all its
observed character states, and of which the
monophyly may hardly be justified. However,
cerebral commissure shortening and occur-
rence of a glandular zone along the vagina
might constitute synapomorphies of the group
composed of the Zonitidae, Trochomorphidae
and Euconulidae within the Zonitoidea. The
Euconulidae and Trochomorphidae share
apomorphies in kidney length (LR2), nerve
ring (CPD2, PLD2, PAD2), and intestinal
length (IL2). Absence of a vaginal glandular
zone might be a synapomorphy of the Trocho-
morphidae within this group, whereas the Eu-
conulidae are characterized by the position of
the spermathecal insertion (Baker, 1941).

The only clear points in the history of the
Zonitoidea are that the group is probably Lau-

rasian in origin, and that the families were
probably already differentiated in the Pa-
leocene in Europe and North America. The
Euconulidae (Afroconulus) and the Arionidae
(Oopelta) might have colonized eastern Af-
rica from Arabia during the Miocene, when
the Tethys Sea was closed in this region (like
the Cerastuinae, Helicidae and possibly the
Achatinoidea). The tree presented in Text-fig.
24 implies that the Carboniferous genus Pro-
todiscus is probably not a member of the Re-
cent family Discidae, but at most a disciform
primitive zonitoid: the characters upon which
Solem and Yochelson (1979) base their famil-
ial attribution of Protodiscus to the Discidae
might be at best synapomorphies of the zoni-
toids, and at worst symplesiomorphies of the
latter (= synapomorphies of the non-orth-
urethran Stylommatophora?).

As explained below, | have placed the He-
licarionidae and Vitrinidae in the Helicoidea
for the sake of further discussion. However,
their inclusion in the Zonitoidea would modify
very little the tree of Text-fig. 24: the Helicar-
ionidae would originate from a triple node at

84 TILLIER

which the Zonitidae and Euconulidae + Tro-
chomorphidae originate; the Vitrinidae, if not
derived from the Helicarionidae (Text-fig. 23),
would originate between the Parmacellidae
and Milacidae (Text-fig. 24).

Helicoidea: The Helicoidea include the
families Helicidae, Helminthoglyptidae, Brady-
baenidae, Camaenidae, Polygyridae, Hap-
lotrematidae, Sagdidae, Helicarionidae and
Vitrinidae. Grouping together the first four of
these families is classical (Zilch, 1959—
1960), although rejected by Solem (1978)
and Boss (1982), who prefer to group the
Polygyridae with the Corillidae, and the
Oreohelicidae with the Ammonitellidae into a
distinct superfamily. These last three families
being considered here acavoid (v. infra), and
nobody doubting the relative proximity of the
Polygyridae, Camaenidae and remaining He-
licoidea, the shortest tree in Text-fig. 23 is
accepted here.

Placing the Sagdidae among the Heli-
coidea has been discussed above and seems
better to me than placing them among the
Achatinoidea, which, however, is more ac-
ceptable in terms of phyletic distance (Text-
fig. 23): Sagda has an apomorphy common
among the Helicoidea, apparent fusion of the
visceral and left parietal ganglia, and does not
have a synapomorphy of the Achatinoidea,
the absence of an amatorial organ. In this in-
terpretation the penial appendix of Sagda is
homologous with the amatorial organ of the
helminthoglyptids from which it seems to dif-
fer only in position, not in morphology.

Placement of the Helicarionidae among the
Helicoidea is more a challenge than a certi-
tude. Their position in either the Helicoidea or
Zonitoidea does not modify the relative posi-
tion of the superfamilies discussed below,
and their placement among the Helicoidea
provides shorter trees than placement among
the Zonitoidea, whatever the computer pro-
gram used. The position of the Vitrinidae,
which are much closer to the Helicarionidae
than to the Zonitidae in terms of both phyletic
distance (Text-fig. 23) and phenetic distance
(Text-fig. 20), is the consequence of this
choice; diversity of the Vitrinidae is relatively
high in East Africa, where one part of the He-
licarionidae perhaps originated (у. infra).
However, the Vitrinidae have surprisingly
many apomorphic character states even
hough their degree of limacization is slight,
facts that make the discussion of their affini-
ties more uncertain.

Placing the Haplotrematidae among the
Helicoidea is also questionable, for they re-
semble much less any helicoid family than
they do the Oreohelicidae, which are here
considered members of the Acavoidea (Text-
fig. 20); they might be the North American
representatives of the carnivorous trend in
the acavoids. They are placed here among
the Helicoidea not only because this solution
provides shorter trees, but also because Hap-
lotrema has the apparent fusion of the vis-
ceral and left parietal ganglia seen in many
helicoids but in no oreohelicid.

The Oreohelicidae are here removed from
the Helicoidea, because: first, in all the short-
est trees obtained, they are grouped with the
Acavoidea; and second, this position makes
sense in view of the way in which the very
short kidney associated with the very short
ureteric tube commonly found in the family
(Figs. 493, 494, 500, 501) can be integrated
into a macroevolutionary pattern at suprafa-
milial levels (v. infra).

The synapomorphies of the Helicoidea so
defined are the occurrence of a ureteric tube
closed at least to the recto-renal angle of the
lung roof (UR3), and the contact between the
left parietal and visceral ganglia (PAG3).

The tree shown in Text-fig. 25 must be
viewed with caution, because the monophyly
of the Helicidae and Helicarionidae is far from
certain, and the inclusion of some families is
perhaps dubious. However, removing the He-
licarionidae and Vitrinidae from the data set
modifies only the relative position of the Hap-
lotrematidae, which become the sister group
ofthe Polygridae + Sagdidae. Introducing the
Oreohelicidae results only in their placement
in the tree as the sister group of all the other

` families.

The Helicoidea are primarily Laurasian: the
South American helminthoglyptids (Parodiz,
1969), African helicids, and Australasian ca-
maenids are probably Late Tertiary immi-
grants. If the Vitrinidae and Helicarionidae
form a monophyletic group, this group might
have originated before the Late Cretaceous
(European Provitrina), and dispersed into Af-
rica and Australasia before the extinction of
the Helicarionidae in Europe; the absence of
the Helicarionidae from America is problem-
atic, however, and it might be that the Urocy-
clinae and Gymnarioninae in East Africa and
the other subfamilies in the Orient were de-
rived independently from various helicoids. If
monophyletic, the Helicarioninae might have
been differentiated in Eurasia before the other

STYLOMMATOPHORAN SYSTEMATICS 85

SUBUL INIDAE ACHAT INIDAE OLEACINIDAE

SUCCINEIDAE

STREPTAXIDAE

FERUSSACIIDAE

TEXT-FIG. 26. Phyletic trees of achatinoid families. 26A) minimum tree. 26B) tree retained (see text).

subfamilies and during the Miocene colonized
East Africa and Madagascar on the one hand,
and Australasia on the other hand before be-
coming extinct farther north; their monophyly
cannot be argued seriously before the family is
revised.

Achatinoidea: The Achatinoidea include
the families Achatinidae, Subulinidae, Ferus-
saciidae, Oleacinidae, Succineidae and
Streptaxidae (Text-fig. 23). Although they re-
semble more some aulacopod families (Text-
fig. 20), the Succineidae are tentatively in-
cluded here on the basis of their special
similarities with the Ferussaciidae, i.e. heter-
urethry (apomorphic) and occurrence of two
procerebral commissures (symplesiomorphy
unique among the Stylommatophora, as far
as known). The Streptaxidae differ from the
Ferussaciidae only in their very long lung
(shortened in the Ferussaciidae), their inter-
nal kidney morphology, and character states
related to carnivory.

Synapomorphic character states in the
Achatinoidea are the closure of the ureteric

tube (UR4), short cerebral commissure
(CC2), the contiguity of the left parietal gan-
glion with the visceral ganglion (PAG3), and
possibly the symmetry of the cerebro-pedal
connectives (CPR2; only exception occurs in
the Succineidae). To distinguish the Achati-
noidea from the Helicoidea, the absence of an
amatorial organ (and the shell shape?) must
be added.

By considering characters not included in
the data set used to build the tree shown in
Text-fig. 26A, the tree shown in Text-fig. 26B
is obtained. Here the Oleacinidae are the sis-
ter group of the Ferussaciidae because they
share derived states in internal and external
morphology of the kidney, and the Succinei-
dae are the sister group of this group because
all three families share their trend to heteru-
rethry. The Streptaxidae are the next sister
group, without lung shortening, and the Acha-
tinidae and Subulinidae are the sister group of
all the others, having no kidney shortening at
all.

Given that the achatinoid families, the fam-
ily Achatinidae excepted, were differentiated

86 TILLIER

SUBULINIDAE ACHATINIDAE STREPTAXIDAE

OLEACINIDAE FERUSSACIIDAE

SUCCINEIDAE

heterurethran trend

Fig. 26b

in the Eocene in Europe, two scenarios are
possible. If the superfamily is Laurasian in or-
igin, these families might have invaded South
America through North America. The same,
Oleacinidae excepted, might have invaded
the Orient and, during the Miocene, Africa,
where the Achatinidae might have been de-
rived from the Subulinidae. The alternative
solution is to consider the Achatinoidea as
Gondwanian in origin: some Streptaxidae,
Subulinidae and the common ancestor of the
Oleacinidae, Ferussaciidae and Succineidae
would have invaded Europe before the
Eocene via the Panamanian region and North
America; the Subulinidae and Streptaxidae
would be vicariant in South America, Africa
and the Orient, and the Achatinidae would
have appeared in Africa later than the open-
ing of the South Atlantic and than the drift of
the Indian plate away from Africa, i.e. later
than the end of the Cretaceous. This solution
does not explain the quasi-total absence of
the Achatinoidea from the Australasian re-
gion; the hypothesis of a Laurasian origin,
which is also supported by the tree presented
below (Text-fig. 29), seems preferable.

Clausilioidea: The Clausilioidea include the
families Clausiliidae, Cerionidae, Urocoptidae
and Bulimulidae. In the absence of any new
element, | follow Pilsbry (1904) and include

also the Megaspiridae. Choosing the non-
minimum paths that allow this grouping (Text-
fig. 23) leads to a multiple node (node 23).
However, this group might be formed because
at least some members of the various families
composing the superfamily form an almost
perfect morphocline in many characters: the
Cerionidae look exactly like Clausiliidae with a
shortened nerve ring (and without a clausil-
ium); within the Urocoptidae Urocoptis and
Macroceramus have the general morphology
seen in Cerion, whereas Berendtia is similar to
bulimulids in general anatomy. Furthermore,
the Urocoptidae and Bulimulidae have a pos-
sible synapomorphy, the apparent fusion of
the visceral and left parietal ganglia. The sy-
napomorphic character states of the super-
family are: LR3 RR2 CPR2 PAG2.

In the minimized diagram of possible rela-
tionships within the superfamily (Text-fig.
27A), the Cerionidae are the sister group of
the Bulimulidae on the basis of the shortening
of the cerebro-pedal connectives in both fam-
ilies. | prefer to consider the Urocoptidae the
sister group of the Bulimulidae, because both
have a tendency to fusion of the visceral and
left parietal ganglia, and because of the strik-
ing similarity among some members of both
families. Two scenarios can be inferred from
the tree so constructed (Text-fig. 27B), given
that the Clausiliidae occurred in Europe dur-

STYLOMMATOPHORAN SYSTEMATICS 87

MEGASPIRIDAE CLAUSILIIDAE

UROCOPTIDAE

CERIONIDAE BULIMULIDAE

TEXT-FIG. 27. Phyletic trees of clausilioid families.
doubtful; see text).

ing the Late Cretaceous, the Urocoptidae oc-
curred in North America during the Pa-
leocene, and the Bulimulidae occurred in
South America during the Paleocene (Par-
odiz, 1969). The first scenario involves a
Laurasian origin. The Urocoptidae and Ceri-
onidae would have been differentiated re-
spectively in North America and in the Pana-
manian region during the Paleocene, and the
present distribution of these families and of
the South American and Australasian Bulim-
ulidae represents a Paleocene South Ameri-
can and trans-Antarctic track from North
America to Tasmania, part of Australia, New
Zealand, New Caledonia, the Solomon ls-
lands and Vanuatu. It seems that the clausiliid
Neniinae arrived in the Andes in the Late
Tertiary (Parodiz, 1969); consequently they
would have become extinct in North America
more recently. The Megaspiridae might have
appeared in Laurasia and have migrated
along the same track as the Bulimulidae (Pils-
bry, 1904); but if the European Danian Paleo-
stoa is closer to the clausiliids than to the me-

27A) minimum tree. 27B) tree retained ( ? = position

gaspirids, the latter might be the Gondwanian
sister group of all the other clausilioids. This
would imply that the superfamily was differ-
entiated before the break-up of Pangea. The
other scenario involves a Gondwanian origin.
The Megaspiridae and the ancestral stock of
all the other families would have been differ-
entiated within Gondwana, and the Clausili-
idae, Cerionidae and Urocoptidae would rep-
resent successive immigration waves that
issued from this ancestral stock through the
Panamanian region. This model does not ex-
plain the absence of the Bulimulidae and Me-
gaspiridae from the Ethiopian and Indian re-
gions.

Endodontoidea: The group including the
families Charopidae, Endodontidae, Punc-
tidae, and Systrophiidae is formed equally
well using phenetic (Text-fig. 20) and phyletic
(Text-fig. 23 and PENNY algorithm) dis-
tances. The inclusion of the Systrophiidae п
the Endodontoidea, which might seem sur-

88 TILLIER

MEGASPIRIDAE CLAUSILIIDAE

CERIONIDAE

UROCOPTIDAE BUL IMUL IDAE

overall similarity
FG2'

Fig. 27b

prising, has already been proposed by Schi-
leyko (1978a) on a mainly conchological ba-
sis. The unique morphology of the nerve ring
of the systrophiids might result from the ex-
aggeration of the trends observed in the en-
dodontids (in particular, displacement of the
pleural ganglia toward the cerebral ganglia).
Weak as the reasons to include the Athora-
cophoridae in the Endodontoidea might ap-
pear, they seem stronger than reasons for
placing the family elsewhere. First, of all the
taxa studied, the semislug Cystopelta, which
is here considered to belong to the Punctidae,
is closest in anatomy to the Athoracophoridae
(Text-fig. 19); this similarity is particularly re-
markable because these two taxa represent
very different degrees of limacization. Sec-
ond, this hypothesis is biogeographically co-
herent: diversity of the Endodontoidea is max-
imal in the Australasian region.

The synapomorphies of the Endodontoidea
so defined are shortening in intestinal length,
kidney length, cerebral commissure length
and cerebro-pedal connective length (IL2,
LR2, CC2, CPD2), and occurrence of a ure-
teric tube reaching at least the recto-renal an-
gle of the lung (UR2). In the tree constructed
from the data set including the CCAs of the
endodontoid families (Text-fig. 28), the
Charopidae are the sister group of the other

amilies (and possibly the stem group, if one

accepts that the shape of the gastric crop of
the Athoracophoridae is secondarily plesio-
morphic, which is not unlikely). According to
this tree, the Charopidae were differentiated
from the common ancestor of the other fam-
ilies, which was probably northeast Gondwa-
nian, before the break-up of Gondwana,
which resulted in their present distribution.
The Charopidae might have spread into the
In-donesian Archipelago, via New Guinea,
during the Miocene; and North America, via
the Panamanian region, during the Creta-
ceous (unless the superfamily was differenti-

` ated before the break-up of Pangea, in which

case the North American Charopidae might
be a relict of a former Pangean distribution).
The Punctidae might have the same history,
whereas the Athoracophoridae stayed in the
Australasian region. The present distribution
of the Punctidae, Athoracophoridae, Endo-
dontidae and Systrophiidae produces a trans-
Pacific track when compared with the tree of
Text-fig. 28. This track might of course result
from dispersal from the west eastward across
the Pacific Ocean; but vicariance resulting
from the possible opening of this ocean
(during the Triassic? Melville, 1981) is also
possible. If accepted, this hypothesis leads
to the possibility of attributing the present
distribution of North and South American
charopids to the same event, rather than to

STYLOMMATOPHORAN SYSTEMATICS 89

CHAROPIDAE ENDODONT IDAE SYSTROPHIIDAE PUNCTIDAE ATHORACOPHOR IDAE
FG3
PAG4
CPR2 CPR3 PS2
PS2 CC3 503 LR5
LR3 PAD2 CPD3
Farin
7 CPR2
2 1
TE?
SC3

TEXT-FIG. 28. Minimum tree of phyletic relationships among endodontoid families (retained).

Gondwanian vicariance with trans-Pan-

amanian dispersal.

Acavoidea: The superfamily Acavoidea
should include at least the families Acavidae
and Rhytididae (suppression of nodes 11 and
13, Text-fig. 23). The position of the Oreohe-
licidae (including the Ammonitellinae) and
Corillidae cannot be justified from Text-fig. 23,
because they are taken up by the algorithm
only at the end of the process of construction
of the diagram, and their position is estab-
lished in relation to nodes that might be mod-
ified greatly by the adoption of paths not in-
cluded in the first short tree. Consequently,
their position is discussed and established
further from a data set that includes the CCAs
of the superfamilies already retained, and the
CCAs of the acavoid families. First, however,

it should be noted that the four families here
considered acavoid have the association of
an apomorphic short kidney with a plesiomor-
phic little- or not-developed ureteric tube, as
do the Clausilioidea and Endodontoidea, but
not any other superfamily. Including the Oreo-
helicidae in the Helicoidea does not modify
the pattern of relationships between the su-
perfamilies established below.

The synapomorphic character states are:
IL2’, LR3 CC2 CPR2 PAGS for the Acavidae
and Rhytididae; IL2’ LR3 RR2 CPR2 PLD2
PAG2 for the Corillidae; and LR3 UR2 RR2
РАСЗ for the Oreohelicidae.

The phylogenetic relationships of these
families established below (Text-fig. 29) can-
not be considered very reliable, principally be-
cause of the impossibility of resolving in detail
the relationships within the Acavidae-Rhy-

90 TILEIER

ZONITOIDES

ENDODONTOIDES

ACAVIDAE

CLAUSILIOIDES OREOHELICIDAE RHYTIDIDAE CORILLIDAE

TEXT-FIG. 29. Phyletic trees of non-orthurethran stylommatophoran superfamilies. 29A) minimum tree
obtained by introducing acavoid families separately in data set. 29B) tree retained (see text).

tididae. However, all acavoid families but the
Oreohelicidae are exclusively Gondwanian,
and, vicariance occurring within families, per-
haps they originated before the break-up of
Gondwana. The Corillidae, which are absent
from South America, might have originated
after the opening of the South Atlantic Ocean
but before the drift of the Indian plate, on
which they still occur, i.e. between the earliest
Cretaceous and the Cenomanian. Ifthe Oreo-
helicidae really belong to the Acavoidea, their
occurrence in North America might result
from either dispersal from South America be-
fore or during the Cretaceous; or from trans-
Pacific vicariance ifthe Pacific Ocean opened
during the Mesozoic; or from their origin by
vicariance in Laurasia, which would imply that
the Acavoidea originated before the break-up
of Pangea.

Relationships of superfamilies: | discuss
next the phyletic relationships of the super-
families, and describe two new suborders, the
Brachynephra and Dolichonephra.

The tree presented in Text-fig. 29A was
constructed with the CCAs of the various
groups retained above, after the symplesio-
morphies and the autapomorphies, which

modify the distances without adducing any in-
ormation about relationships, were elimi-
ated from the data set. This tree divides the

Stylommatophora into two groups based on
relative chronology in kidney shortening and
ureteric tube formation. In the first group,
which includes the Endodontoidea, Clausilio-
idea and Acavoidea, kidney shortening oc-
curs before formation of a ureteric tube. In the
second group, which includes the Zonitoidea,
Helicoidea and Achatinoidea, the formation of
a ureteric tube precedes kidney shortening, if
any. This subdivision fits two patterns previ-
ously and independently recognized as prob-
able on the basis of functional hypotheses.
Furthermore, those superfamilies whose ori-
gin was probably Gondwanian are all in the
first group, whereas nearly all those whose
origin was probably Laurasian are in the sec-
ond group. The conjunction of this division
with hypotheses already entertained indepen-
dently seems to allow the proposal of their
recognition as likely monophyletic suborders:
| propose the name Brachynephra for the first
suborder, including the superfamilies Endo-
dontoidea, Clausilioidea and Acavoidea; and
the name Dolichonephra for the suborder in-
cluding the superfamilies Zonitoidea, Heli-
coidea and Achatinoidea. These names take
only the plesiomorphic state in each suborder
into account, and do not express fully the fun-
damental criterion of division, which is the rel-
ative chronology of kidney shortening and
ureteric tube formation: the kidney is second-

STYLOMMATOPHORAN SYSTEMATICS 91

HELICOIDES ACHATINOIDES ZONITOIDES

DOL ICHONEPHRA

CLAUSILIOIDES ENDODONTOIDES ACAVOIDES

BRACHYNEPHRA

Fig.29b

arily long in the Acavidae and secondarily
short in some Achatinoidea if the hypotheses
here adopted are true.

The superfamilies included in the Brachy-
nephra are all partly or wholly Gondwanian,
as seen above. The apparent monophyly of
the Clausilioidea and Oreohelicidae shown in
Text-fig. 29A is problematic. One can imagine
two explanations of this apparent monophyly.
The first is that the tree presented in Text-fig.
29A reflects correctly the history of the sub-
order: the common ancestor of the Clausilio-
idea and Oreohelicidae migrated into North
America, and was the sister group of the Me-
gaspiridae, which remained Gondwanian but
became extinct in Africa and the Indian re-
gion. After divergence of the Clausilioidea
and Oreohelicidae, the remaining clausilioid
families first used their present track south-
ward. If the tree shown Text-fig. 27 is false,
the Bulimulidae first occurred in Gondwana
after the appearance of the Megaspiridae and
before the migration of the ancestor of the
other clausilioids and oreohelicids. The sec-
ond explanation is that the tree in Text-fig.
29A falsely suggests apparent monophyly of
the Clausilioidea and Oreohelicidae, which
belong in fact to the Acavoidea although still
having some more plesiomorphic characters
than the other families belonging to the latter.

The Oreohelicidae are then Gondwanian in
origin, and reached North America before the
end of the Mesozoic. The clausilioid families
have either followed the same track to the
northern continents and back to the Gondwa-
nian regions, or represent a Pangean group
of which the Megaspiridae are the Gondwa-
nian relict whereas the other families repre-
sent a radiation issuing from the Northern
Hemisphere. This last solution involves one
migration fewer than the others, and results in
the tree shown Text-fig. 29B. The conse-
quence of this tree is acceptance of the idea
that the two non-orthurethran suborders were
already differentiated during the Carbonifer-
ous, even though no element allows construc-
tion of a scenario for this divergence.

The date of the divergence of the Endodon-
tidea and Acavoidea can hardly be estimated
because the superfamilies are not vicariant;
however, it is doubtless anterior to the break-
up of Gondwana. The patterns of the various
families have been discussed above.

The Dolichonephra contains the Helico-
idea, Achatinoidea and Zonitoidea. The simi-
larity of the shell of the Carboniferous genus
Protodiscus to the shell of the zonitoid
Discidae is consistent with both the position of
the Zonitoidea on the tree shown in Text-
fig. 29B, and the hypothesis of the differenti-

92 TILLIER

ation of the suborders before the break-up of
Pangea.

The position of the Achatinoidea as the sis-
ter group of the Helicoidea fits the hypothesis
of their Laurasian origin better than the hy-
pothesis of their Gondwanian origin dis-
cussed above (the synapomorphy that allows
the distinction of the Achatinoidea from the
Helicoidea is the primary absence of an am-
atorial organ, not taken into account in Ap-
pendix G).

There are surprisingly few elements indi-
cating an origin of the familial and suprafamil-
ial groups of the Stylommatophora through vi-
cariance events. The suborders might be
originally vicariant only if: either all the Clau-
silioidea are of Gondwanian origin, a hypoth-
esis that contradicts Pilsbry's hypotheses
(1904) and implies more migrations and ex-
tinctions than does the hypothesis of the dif-
ferentiation of suborders anterior to the break-
up of Pangea; or the suborders originated
before the formation of Pangea. In addition,
there is no pair of superfamilies of which the
vicariant origin can be recognized. Finally,
within the superfamilies there are few pairs of
families or family groups whose origin is prob-
ably vicariant: clausilioid groups of families;
Partulidae—Enidae; Endodontidae—Systro-
phiidae—Punctidae + Athoracophoridae;
possibly Helicidae—Helminthoglyptidae, in-
sofar as both are monophyletic.

On the contrary, there are many indications
of migrations, as shown in the discussions of
the history of the various superfamilies: the
general impression is that most usually the
sister groups diverged within a biogeographic
region, and that the present geographic dis-
tributions result from subsequent dispersals
(e.g. Subulinidae and Streptaxidae). However,
this general impression should not be consid-
ered definitive for two reasons. First is the
scarcity of fossils and the difficulties of assign-
ing them to a family (convergence and sym-
plesiomorphy in shell characters); in particular
the quasi-absence of fossils older than Danian
is dramatic, because most Recent superfam-
ilies seem to be older. Second is the impreci-
sion of paleogeographical data for older peri-
ods, during which most families probably
appeared: the incertitudes aboutthe place and
date of the appearance of the various groups
forbids us to exclude their vicariant origin in
various regions of single continental units. It
might be more reliable to revise the families
and try to propose phylogenies for genera.
Unfortunately, both collections (it seems, for

example, that no museum has preserved Me-
gaspiridae) and systematists are lacking.

CLASSIFICATION

In the classification which follows, the su-
perfamilies seem large compared with those
proposed by Solem (1978) or Boss (1982). |
do not believe it very important, the only real
problem being to recognize monophyletic
units; furthermore, nothing keeps one from
using more restricted groups at suprafamilial
rank. Fifty-one families being classified in
three suborders, the information would be
maximal if every suborder included 17 fami-
lies and four superfamilies of about four fam-
ilies each. The classification presented below
is not very far from these numbers, although
not deliberately so. The two superfamilies that
might seem very large compared with the
other ones, i.e. the Helicoidea and Zoni-
toidea, are precisely those whose phylogeny |
consider the most poorly resolved. The clas-
sification of the Orthurethra is not very satis-
fying, as noted above, more especially as the
Chondrinoidea are polyphyletic according to
my hypotheses; however, my feeling is that it
is worth recognizing the Partuloidea, whereas
| am not very confident of the position of the
Chondrinidae in the tree of Text-fig. 22.

Order Stylommatophora

Suborder Orthurethra

Superfamily Pupilloidea
Valloniidae (+ Strobilopsidae)
Achatinellidae (+ Tornatellinidae)
Pupillidae
Pyramidulidae

Superfamily Chondrinoidea
Chondrinidae
Vertiginidae
Orculidae
Cochlicopidae
Amastridae

Superfamily Partuloidea
Partulidae
Enidae

Suborder Dolichonephra

Superfamily Zonitoidea
Zonitidae
Trochomorphidae
Euconulidae
Discidae
Arionidae (+ Philomycidae)
Parmacellidae
Limacidae
Milacidae

STYLOMMATOPHORAN SYSTEMATICS 93

Trigonochlamydidae
Superfamily Helicoidea
Helicidae
Helminthoglyptidae
Bradybaenidae
Polygyridae
Camaenidae
Sagdidae
Haplotrematidae
Helicarionidae (+ Aillyidae)
Vitrinidae
Superfamily Achatinoidea
Achatinidae
Subulinidae
Streptaxidae
Oleacinidae( + Spiraxidae, Testacell-
idae)
Ferussaciidae
Succineidae
Suborder Brachynephra
Superfamily Clausilioidea
Megaspiridae
Clausiliidae
Cerionidae
Urocoptidae
Bulimulidae
Superfamily Endodontoidea
Charopidae
Endodontidae
Systrophiidae
Punctidae
Athoracophoridae
Superfamily Acavoidea
Oreohelicidae (+ Ammonitellidae)
Corillidae
Acavidae
Rhytididae (+ Chlamydephoridae)

CONCLUSIONS

The goal of the first part of this work is pri-
marily to describe hitherto unknown morphol-
ogies and their variations within the stylom-
matophoran plan of organization in relation to
classification; it is also an attempt to identify
and eliminate the characters obviously func-
tional or redundantly variable. Although the
practical result is original, the principle is not:
as noted by Cain (1982), nearly every sys-
tematist (but not he) tries to use non-adaptive
characters to build classifications. If reached,
this objective implies that taxonomic diversity
does not result from adaptation: if applied to
groups that are defined by non-adaptive char-
acters, the concept of adaptive radiation is at
least ambiguous. As discussed elsewhere

(Tillier, 1986), the contradiction between the
adaptationist program and elaboration of phy-
logenetic classification can be shown as fol-
lows: first, characters are adaptive; second, if
adaptive characters are used to build classi-
fications, selection pressures will be classified
instead of kinships; third, therefore non-
adaptive characters should be used to build
phylogenetic classifications; fourth, go to step
one. The obligation to use non-adaptive char-
acters to build phylogenetic classifications im-
plies that macroevolution is neutral regarding
adaptation. The tenet that all evolution results
from adaptation, defended by the neo-
Darwinians, should result in renunciation of
phylogenetic reconstruction and therefore in a
modification of taxonomic principles and of
the theories of biological classifications, at
present based on the principle of a relation-
ship between kinship and classification.

There is no doubt that the classification pre-
sented above is very far from perfect. How-
ever, | believe it at leasi as coherent as those
presented by Solem (1978) and Boss (1982).
Insofar as it represents progress, this lies in
the set purpose of elaborating a phylogenetic
classification, and hopefully in the possibility
for the reader of finding the reasons for taxo-
nomic decisions without having to reexamine
all the materials | have dissected.

Of course the necessity of attempting to
build a phylogenetic classification of land
snails and slugs can be questioned, inasmuch
as the basically phenetic classifications of the
Pilsbry-Baker type have recently raised no
other protest than Schileyko's (1978a, b),
whose results are, in my opinion, far from con-
vincing (problem of translation?). The principal
inconvenience of classifications built on a phe-
netic basis is that their control is not possible
without wide personal experience; but there is
no doubt that to some extent they allow gen-
eralizations and predictions on morphological
characters. Imperfect as it might appear, a
phylogenetic classification seems better to me
for two reasons. First, it is better to try to pro-
vide an image of the history of the group that
is unique and independent of the observer,
than an image of similarities which necessarily
depend on the observer. Secondly, if evolution
is the efficient cause of the apparent order of
the living, kinship is the formal cause. If one is
interested mainly in evolutionary mechanisms,
one will give precedence to the efficient
causes in building classifications and, as | tried
to discuss, will favor morphological similarity in
relation to the current theory of evolution. Fa-

94 TILLIER

voring formal causes, as attempted by the phy-
logenetic systematicist, seems better to me,
inasmuch as a classification should be a plan
of the natural order, not an account of the
realization of this plan. In both cases it is easy
to pass to the final causes, the evocation of
which provokes repugnance in contemporary
biologists, and it should be kept in mind that
the word adaptation is finalist not only in ety-
mology, but also in usage: talking about ad-
aptation to an environment, instead of talking
about adaptation to a biological property under
the action of the environment, is a rather fi-
nalist expression of evolution (Gould &
Lewontin, 1979; Gould & Vrba, 1982). How-
ever, the synthetic theory doubtless better pre-
vents finalism than the absence of macroev-
olutionary theory.

The dilemma between phenetics and phy-
logenetics was avoidable as long as it was
believed that homoplasy is exceptional, and
as the absence of phylogenetic significance
of symplesiomorphy was not emphasized
(and as gradualism was not questioned;
Tillier, 1986). Unfortunately, for this approach
had the immense merit of being simple, the
application of the methods and principles of
Hennig (1966) revealed more homoplasies
than anyone expected, and the importance of
this problem seems to be de facto avoided by
most phylogeneticists (Cain, 1982, 1983).
The abundance of parallel evolution evident
in the Stylommatophora, associated with the
larger number of taxa under study, obliged
me to use minimization of phyletic distances
as a criterion of choice for phyletic relation-
ships. | can justify this approach only by a
postulate following which the appearance of a
given character state in two taxa is more
probable when these taxa are phylogeneti-
cally closer. This postulate is not justified in
the synthetic theory of evolution, but it seems
to me that it is implicit in all methods of phy-
logenetic reconstruction in which minimiza-
tion is used. In addition, accepting the gener-
ality of homoplasy implies a shift in the
meaning of the word “synapomorphy” from its
restricted sense, ¡.e. “uniquely derived char-
acter state,” to its etymological sense, i.e.
“derived character state uniting taxa.”

The classification so constructed and pro-
posed above should be both tested and
improved. Three areas in particular require at-
tention. First, the sampling should be im-
proved. The method used here necessitates
using the most plesiomorphic state occurring
in each taxon classified. In most cases, my

sampling is such that correspondence be-
tween what has been done and this intention
would be pure chance; in particular the sample
should be extended in the Orthurethra and
most families of which few species have been
examined, not to mention the Megaspiridae,
the only important family of which no specimen
was available. Secondly, characters of the
hard parts might be added. Although it might
be expected that shell and radular morphology
converge too easily to be really useful at high
taxonomical levels, Solem (1983; Solem &
Yochelson, 1979) showed at least similarities
in microscopical structures that might have
some taxonomic value; the problem is to de-
termine at which taxonomic level these char-
acters vary, and to determine the direction of
their evolution. Thirdly, characters of the gen-
ital system might be included. At present it
does not seem possible to determine which
genital characters can be used to determine
relationships among families and at higher tax-
onomic levels, a few cases excepted. A pleth-
ora of opinions can be found, but, as far as |
know, no verification based upon comparable
and numerous enough elements. Cain (1982,
1983) has proposed the use of Secondary sex-
ual characters in phylogenetic reconstruction,
in particular in pulmonate gastropods, for the-
oretical reasons. This approach has not been
attempted here because | preferred to de-
scribe and compare first other organ systems
far less Known in order to make them usable,
but it is worth being tried.

In general conclusion, one cannot insist
strongly enough on the necessity for a real
theory of macroevolution, which does not now
exist. | hope to have shown here and else-
where (Tillier, 1986), perhaps too briefly, that
the approach adopted above, although clas-
sical in principle, requires hypotheses that
current theory about evolutionary mecha-
nisms does not justify. The synthetic theory
has allowed immense progress in species
and population taxonomy, but we still await
the theory that will allow equivalent progress
at the generic level and above.

ACKNOWLEDGMENTS

This work was initiated in 1981 in the Field
Museum of Natural History, Chicago, where |
benefitted from Dr. Alan Solem’s extensive
knowledge of the Stylommatophora, thanks to
grants from the Thomas Dee Fund and the
French Ministere des Affaires étrangeres. It is

STYLOMMATOPHORAN SYSTEMATICS 95

translated, with a few modifications, from a
French Thèse de Doctorat d'Etat prepared
under the direction of Pr. Claude Lévi, and
defended in the Muséum national d'Histoire
naturelle and in the Université Paris VI in
1985. The English version was corrected by
Prof. A.J. Cain, to whom | am deeply indebted
for his help and constant encouragement; all
remaining errors are mine.

No museum in the world having liquid-
preserved collections representative of all the
Stylommatophora, this work would not have
been possible without the loan and authoriza-
tion to dissect material from the following cu-
rators and museums: Dr. A. Solem (Field Mu-
seum of Natural History, Chicago); Drs. G.
Davis and R. Robertson (Academy of Natural
Sciences of Philadelphia); Dr. E. Gittenberger
(Rijksmuseum van Natuurlijke Historie, Lei-
den); Dr. P. Mordan (British Museum, Natural
History); Dr. Puylaert (Musée Royal de l'Af-
rique centrale, Tervuren); Dr. M. Bishop (Uni-
versity Museum, Cambridge); and Dr. A.
Coomans (Zoological Museum, Amsterdam).

In the course of the work, the late Pierre
Delattre (Commissariat à l'Energie atomique,
Saclay), brought a decisive theoretical contri-
bution to my attempts to find solutions to the
problems raised by phylogenetic reconstruc-
tions in relation to parallel evolution. The as-
sistance of Raymond Baudoin (Muséum na-
tional d'Histoire naturelle) was decisive for
designing, writing and running computer pro-
grams.

Text-figs. 1, 21 and 23 were drawn by
Claude Marcillaud, technicien à l'ORSTOM-
Nouméa. The other 730 figures have been
mounted and labelled by Annie Tillier, tech-
nicienne au CNRS.

REFERENCES

BAKER, H.B., 1938, Zonitid snails from Pacific
islands.1. Southern genera of Microcystinae.
Bernice P. Bishop Museum Bulletin, 158: 1-102.

BAKER, H.B., 1940, Zonitid snails from Pacific
islands.2. Hawaiian genera of Microcystinae. Ber-
nice P. Bishop Museum Bulletin, 165: 102-201.

BAKER, H.B., 1941, Zonitid snails from Pacific
islands.3. Genera other than Microcystinae.4.
Distribution and indexes. Bernice P. Bishop Mu-
seum Bulletin, 166: 205-346.

BAKER, H.B., 1955, Heterurethrous and aulaco-
pod. Nautilus, 68(4): 109-112.

BAKER, H.B., 1958, H.A. Pilsbry, 1862-1957. Nau-
tilus, 71(3): 73-115.

BAKER, Н.В., 1962, Puerto Rican oleacinoids.
Nautilus, 75(4): 142-145.

BARGMANN, H.E., 1930, The morphology of the
central nervous system in the Gastropoda Pul-
monata. Journal of the Linnean Society (Zool-
09y), 37:1-58.

BINDER, E. 8 TILLIER, S., 1985, Acantharion
browni, un nouveau Gymnarioninae d'Ethiopie
(Gastéropode Pulmoné Stylommatophore Lim-
acacea). Revue suisse de Zoologie, 92(1): 177-
187.

BISHOP, M.J., 1978, The value of the pulmonate
nervous system in phylogenetic hypothesis.
Journal of Molluscan Studies, 44: 116-119.

BLINN, W.C., 1964, Water in the mantle cavity of
land snails. Physiological Zoology, 37: 329-337.

BOSS, K.J., 1982, Mollusca. Pp. 945-1166 in:
PARKER, S.P., ed., Synopsis and Classification
of Living Organisms Vol. 2. McGraw-Hill, New
York.

BOUILLON, J., 1960, Ultrastructure des cellules ré-
nales des mollusques. |. Gastéropodes pulmonés
terrestres (Helix pomatia). Annales des Sciences
naturelles, Zoologie, 12eme série: 719-749.

BREURE, A.S.H., 1979, Systematics, phylogeny
and zoogeography of Bulimulinae (Mollusca). Zo-
ologische Verhandelingen, 168: 215pp.

CAIN, A.J., 1977, Variation in the spire index of
some coiled gastropod shells, and its evolution-
ary significance. Philosophical Transactions of
the Royal Society of London, B277: 377-428.

CAIN, A.J., 1978a, Variation of terrestrial gastro-
pods in the Philippines in relation to shell shape
and size. Journal of Conchology, 29: 239-245.

CAIN, A.J., 1978b, The deployment of operculate
land snails in relation to shape and size of shell.
Malacologia, 17(2): 207-221.

CAIN, A.J., 1980, Whorl number, shape and size of
shell in some pulmonate faunas. Journal of Con-
chology, 30: 209-221.

CAIN, A.J., 1981, Variation in shell shape and size
of helicid snails in relation to other pulmonates in
faunas of the Palearctic region. Malacologia,
21(1-2): 149-176.

CAIN. A.J., 1982, On homology and convergence.
Pp. 1-19 in: JOYSEY, К.А. 8 FRIDAY, A.E., eds.,
Problems of Phylogenetic Reconstruction. Sys-
tematics Association Special Volume, 21. Aca-
demic Press, London.

CAIN, A.J., 1983, Concluding remarks. Pp. 391—
397 in: OXFORD, G.S. & ROLLINSON, D., eds.,
Protein Polymorphism: Adaptive and Taxonomic
Significance. Systematics Association Special
Volume, 24: Academic Press, London.

CAIN, A.J. 8 COWIE, R.H., 1978, Activity of different
species of land-snail on surfaces of different in-
clinations. Journal of Conchology, 29: 267-272.

CAMERON, R.A.D., 1978, Differences in the sites
of activity of coexisting species of land mollusc.
Journal of Conchology, 29: 273-278.

CAMERON, R.A.D., 1981, Functional aspects of
shell geometry in some British snails. Biological
Journal of the Linnean Society, 16: 157-167.

CHÉREL-MORA, C., 1983, Variation géographique
et taxonomie des Placostylus (Gastéropodes

96 TILLIER

Pulmonés Stylommatophores) de Nouvelle Calé-
donie. Thèse de 3eme cycle, Université Paris VII:
103pp.

CHUNG, D.J., 1979, Aspects of the anatomy of Ce-
rion (Mollusca: Pulmonata). Thesis MS, Purdue
University: 158pp.

COOK, L.M. 8 JAFFAR, W.N., 1984, Spire index
and preferred surface orientation in some land
snails. Biological Journal of the Linnean Society,
21(3): 307-318.

COOKE, С. 8 KONDO, Y., 1960, Revision of Tor-
natellinidae and Achatinellidae. Bernice P.
Bishop Museum Bulletin, 221: 303pp.

DELHAYE, W. 8 BOUILLON, J., 1972a, L'évolution
et l'adaptation de l'organe excréteur chez les
mollusques gasteropodes pulmones. |. Introduc-
tion générale et histophysiologie comparée du
rein chez les basommatophores. Bulletin Bi-
ologique, 106(1): 45-77.

DELHAYE, W. & BOUILLON, J., 1972b, L'évolution
et l'adaptation de l'organe excréteur chez les mol-
lusques gastéropodes pulmonés. Il. Histophysi-
ologie comparée du rein chez les stylommato-
phores. Bulletin Biologique, 106(2): 123-142.

DELHAYE, W. 8 BOUILLON, J., 1972c, L'évolution
et l'adaptation de Гогдапе excréteur chez les mol-
lusques gastéropodes pulmonés. Ш. Histophysi-
ologie comparée du rein chez les soléoliferes et
conclusions générales pour tous les stylommato-
phores. Bulletin Biologique, 106(4): 295-314.

DELSOL, M. 8 FLATIN, J., 1979, Embryologie et
évolution. Aspects néozoologiques. In Aspects
modernes de recherches sur l'évolution, Comp-
te-rendus du Colloque de Montpellier 12-16 sep-
tembre 1977. Mémoires et Travaux de l'Institut
de Montpellier, Ecole pratique des Hautes
Etudes, 10: 1-70.

DUNCAN, C.J., 1975, Reproduction. Pp. 309-366
IN ЕВЕТТЕВ, \. & РЕАКЕ, JE 1975.

DUVAL, А. & RUNHAM, N.W., 1981, The arterial
system of six species of terrestrial slugs. Journal
of Molluscan Studies, 47: 43-52.

ELDREDGE, N., 1979, Cladism and common

sense. Pp. 165-198 т: CRACRAFT, J. 8 EL- .

DREDGE, N., eds., Phylogenetic Analysis and
Paleontology, Columbia University Press, New
York.

EMBERTON, K., 1982, Environment and shell
shape in the Tahitian land snail Partula otaheit-
ana. Malacologia, 23(1): 23-25.

FELSENSTEIN, I., 1982, Numerical methods for in-
ferring phylogenetic trees. The Quarterly Review
of Biology, 57(4): 379-404.

FISCHER, P., 1880-1887, Manuel de conchyliolo-
gie et de paléontologie conchyliologique. Librai-
rie F. Savy, Paris: 1369pp.

FRANC, A., 1968, Sous-classe des pulmonés. Pp.
325-607 in: GRASSE, P.P., ed., Traité de Zool-
ogie, 5(3), Masson, Paris.

FRETTER, V., 1975, Introduction. Pp. I-XXIX in:
FRETTER, V. 8 PEAKE, J., 1975.

FRETTER, V. 8 GRAHAM, A., 1962, British Proso-
branch Molluscs. Ray Society, London, 755pp.

FRETTER, V. 8 PEAKE, J., eds., 1975, Pulmo-
nates. Vol. 1, Functional Anatomy and Physiol-
ogy. Academic Press, London, 417pp.

FRETTER, V. 8 PEAKE, J., eds., 1978, Pulmo-
nates. Vol. 2A, Systematics, Evolution and Ecol-
ogy. Academic Press, London, 540pp.

GERMAIN, L., 1930, Mollusques terrestres et fluvi-
atiles. Faune de France, Lechevalier, Paris, vol.
21, 477pp.

GHIRETTI, Е. & GHIRETTI-MAGALDI, A., 1975,
Respiration. Pp. 33-52 in: FRETTER, V. 4
PEAKE, J., 1975.

GHOSE, K.C., 1963, Embryogenesis and larval or-
gans of the giant land snail Achatina fulica Bowd-
ich. Proceedings of the Royal Society of Edin-
burgh, section B, 68: 237-260.

GOSLINER, T.M., 1981, Origins and relationships
of primitive members of the Opisthobranchia
(Mollusca: Gastropoda). Biological Journal of the
Linnean Society, 16(3): 197-225.

GOULD, S.J., 1969, An evolutionary microcosm:
Pleistocene and Recent history of the land snail P.
(Poecilozonites) in Bermuda. Bulletin of the Mu-
seum of Comparative Zoology, 138(7): 406-531.

GOULD, S.J., 1977, Ontogeny and Phylogeny.
Harvard University Press, Cambridge (Mass.),
501pp.

GOULD, S.J., 1984, Hen’s Teeth and Horse's Toes.
Further Reflections in Natural History. W.W.
Norton & Co, New York, 413pp.

GOULD, S.J. & LEWONTIN, R.C., 1979, The span-
drels of San Marco and the Panglossian para-
digm: a critique of the adaptationist programme.
Pp. 581-598 in: MAYNARD-SMITH, J. 4 HOLL-
IDAY, R., eds., The Evolution of Adaptation by
Means of Natural Selection: Royal Society, Lon-
don.

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

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

HUBENDICK, B., 1978, Systematics and compar-
ative morphology of the Basommatophora. Pp.
1-47 in: FRETTER, V. 8 РЕАКЕ, J., 1978.

HYLTON SCOTT, 1939, Estudio anatomico del
Borus “Strophocheilus lorentzianus” (Doer.). Re-
vista del Museo de La Plata (Nueva serie), Sec-
cion Zoologia, 1: 217-278.

JAMBU, M. & LEBEAUX, M.O., 1979, Classification
automatique pour l'analyse des données. Il.
Logiciels. Dunod, Paris, 400pp.

JONES, H.D., 1973, The mechanism of locomotion
of Agriolimax reticulatus (Mollusca: Gastropoda).
Journal of Zoology, 171: 489—498.

JONES, H.D., 1975, Locomotion. Pp. 1-32 in:
РВЕТТЕВ, \. & PEAKE JE 1975:

КЕВКЧТ, С.А. 8 WALKER, R.J., 1978, Nervous
system, eye and statocyst. Pp. 165-244 in:
FREMIERSVAS РЕАКЕ, 25978:

LACAZE-DUTHIERS, H. de, 1887, Histoire de la
testacelle. Archives de Zoologie expérimentale et
générale, 2eme série, 5: 459-496.

STYLOMMATOPHORAN SYSTEMATICS 9

LAMBERT, R. & TEISSIER, G., 1927, Théorie de la
similitude biologique. Annales de Physique et de
Physico-chimie Biologique, 3: 212-246.

MACHIN, J., 1975, Water relationships. Pp. 105—
163 in: FRETTER, V. 8 PEAKE, J., 1975.

MASLIN, T.P., 1952, Morphological criteria of phy-
letic relationships. Systematic Zoology, 1(2): 49—
70.

MEAD, A.R., 1986, Anatomical studies transfer
Leucotaenius from Achatinidae to Acavidae (Pul-
monata: Sigmurethra). Archiv fúr Mollusken-
kunde, 116 (4-6) (1985): 137-155.

MELVILLE, R., 1981, Vicariant plant distributions
and palaeogeography of the Pacific region. Pp.
238-274 in: NELSON, G. & ROSEN, D., 1981.

MORDAN, P., 1984, Taxonomy and biogeography
of southern Arabian Enidae (sensu lato) (Pulmo-
nata: Pupillacea). Pp. 124-133 in: SOLEM A. &
VAN BRUGGEN, A.C., eds., 1984, World-wide
snails. Brill, Leiden.

MOREL, P. & IRVING, E., 1978, Tentative paleo-
continental maps for the Early Phanerozoic and
Proterozoic. Journal of Geology, 86(5): 535-
561.

MORTON, J.E., 1955, The functional morphology
of Otina otis, a primitive marine pulmonate. Jour-
nal of the Marine Biological Association of the
United Kingdom, 34: 113-150.

NELSON, G. & ROSEN, D., eds., 1981, Vicariance
Biogeography. A Critique. Columbia University
Press, New York, 593pp.

NUR, A. & BEN-AVRAHAM, Z., 1981, Lost Pacifica
continent: À mobilistic speculation. Pp. 341-358
in: NELSON, G. & ROSEN, D., 1981.

OWEN, H.G., 1983, Atlas of continental displace-
ment, 200 million years to the present. Cam-
bridge University Press, 159pp.

PARODIZ, J.J., 1969, The Tertiary non-marine Mol-
lusca of South America. Annals of Carnegie Mu-
seum, 40: 242pp.

PILSBRY, H.A., 1896, The Aulacopoda: a primary
division of the monotremate land Pulmonata.
Nautilus, 9(10): 109-111.

PILSBRY, H.A., 1900a, On the zoological position
of Partula and Achatinella. Proceedings of the
Academy of Natural Sciences of Philadelphia, 3:
561-567.

PILSBRY, H.A., 1900b, The genesis of Mid-Pacific
faunas. Proceedings of the Academy of Natural
Sciences of Philadelphia, 3: 568-581.

PILSBRY, H.A., 1904, Manual of Conchology. Vol.
XVI, Urocoptidae, Achatinidae: Academy of Nat-
ural Sciences, Philadelphia, 329pp.

PILSBRY, H.A., 1905, Anatomical and systematic
notes on Dorcasia, Trigonephrus n.gen., Corilla,
Thersites and Chloritis. Proceedings of the Mal-
acological Society of London, 6(5): 286-291.

PILSBRY, H.A., 1919, A review of the land mollusks
of the Belgian Congo chiefly based on the col-
lection of the American Museum Congo Expedi-
tion, 1909-1915. Bulletin of the American Mu-
seum of Natural History, 40(1): 1-370.

PILSBRY, H.A., 1939, Land Mollusca of North

America (North of Mexico).l,1. Academy of Маг
ural Sciences of Philadelphia, Monographs, 3:
1-573.

PILSBRY, H.A., 1940, Land Mollusca of North
America (North of Mexico).1,2. Academy of Nat-
ural Sciences of Philadelphia, Monographs, 3:
574-994.

PILSBRY, Н.А., 1946, Land Mollusca of North
America (North of Mexico).Il,1. Academy of Nat-
ural Sciences of Philadelphia, Monographs, 3: 1—
520.

PILSBRY, H.A., 1948, Land Mollusca of North
America (North of Mexico). 11,2. Academy of Nat-
ural Sciences of Philadelphia, Monographs, 3:
521-1113.

PLATE, L.H., 1898, Beitráge zur Anatomie und
Systematic der Janelliden. Zoologische Jahr-
bücher, Abteilung für Anatomie und Ontogenie
der Tiere, 11: 193-280.

RAUP, D.M., 1966, Geometric analysis of shell coil-
ing: general problems. Journal of Paleontology,
40:1178—1190.

REGONDAUD, J., 1964, Origine embryonnaire de
la cavité pulmonaire de Lymnaea stagnalis L.
Considérations particulières sur la morpho-
génèse de la commissure viscérale. Bulletin bi-
ologique de la France et de la Belgique, 98: 433—
471.

RIGBY, J.E., 1963, Alimentary and reproductive
systems of Oxychilus cellarius (Múller) (Stylom-
matophora). Proceedings of the Zoological Soci-
ety of London, 141: 311-359.

RIGBY, J.E., 1965, Succinea putris: a terrestrial
opisthobranch mollusc. Proceedings of the Zoo-
logical Society of London, 144: 445-486.

RUNHAM, N.W., 1975, Alimentary canal. Pp. 53—
104 in: FRETTER, V. 8 PEAKE, J., 1975.

SCHILEYKO, A.A., 1978a, The system of the order
Geophila (= Helicida) (Gastropoda Pulmonata).
In Morphology, Systematics and Phylogeny of
Molluscs. Trudy zoologicheskogo Instituta, Aka-
demija nauk SSSR, 80 (1979): 44—69 (in Rus-
sian).

SCHILEYKO, A.A., 1978b, Peculiarities of the
structure of the excretory system of the Pulmo-
nata in relation to subclass classification. Mala-
cological Review, 11: 68—70.

SCHILEYKO, A.A., 1978c, Nazemnye Molljuski
Nadsemejstva Helicoidea. Fauna SSSR, Moll-
juski, 3(6). Zoologicheskij Institut, Akademija
nauk SSSR, novaja serija, 117: 360pp. (in Rus-
sian).

SCOTESE, C., BAMBACH, R., BARTON, C., VAN
DER VOO, R. & ZIEGLER, A., 1979, Paleozoic
base maps. Journal of Geology, 87(3): 217-277.

SEMPER, C. 8 SIMROTH, H., 1894, Zur ver-
gleichenden Morphologie der Pulmonaten Niere.
Pp. 45-91 in: SEMPER, C., Reisen im Archipel
der Philippinen, 3(2).

SIMROTH, H. 8 HOFFMANN, H., 1908—1928, Pul-
monata. Bronns Klassen und Ordnungen des
Tier-Reichs, 3, 2, 2: 1354pp.

SMITH, A.G. 8 BRIDEN, J.C., 1977, Mesozoic and

98 MIEBIER

Cenozoic Paleocontinental Maps. Cambridge
University Press, Cambridge, 63pp.

SNEATH, P. & SOKAL, R., 1973, Numerical Tax-
onomy. W.H. Freeman and Company, San Fran-
cisco, 573pp.

SOLEM, A., 1959, Systematics and zoogeography
of the land and freshwater Mollusca of the New
Hebrides. Fieldiana: Zoology, 43: 1-359.

SOLEM, A., 1964, Amimopima, an Australian enid
land snail. Veliger, 6(3): 115-120.

SOLEM, A., 1966a, Some non-marine mollusks
from Thailand, with notes on classification of the
Helicarionidae. Spolia Zoologica Musei Haunien-
sis, 24: 110pp.

SOLEM, A., 1966b, The Neotropical land snail
genera Labyrinthus and Isomeria (Pulmonata,
Camaenidae). Fieldiana: Zoology, 50: 226pp.

SOLEM, A., 1968, “Ptychodon” misoolensis Adam
& van Benthem Jutting, 1939, a New Guinea
strobilopsid land snaii and review of the genus
Enteroplax. Veliger, 11(1): 24-30.

SOLEM, A., 1970, The endodontid land snail genera
Pilsbrycharopa and Paryphantopsis (Mollusca:
Pulmonata). Veliger, 12(3): 239-264.

SOLEM, A., 1976, Endodontoid Land Snails from
Pacific islands (Mollusca: Pulmonata: Sigmure-
thra). Part 1. Family Endodontidae. Field Museum
of Natural History, Chicago, 508pp.

SOLEM, A., 1978, Classification of the land mol-
lusca. Pp. 49-98 in: FRETTER, V. & РЕАКЕ, J.,
1978.

SOLEM, A., 1981, Land-snail biogeography: A true
snail's pace of change. Pp. 197-237 т: NEL-
SON, G. & ROSEN, D., 1981.

SOLEM, A., 1982, Endodontoid Land Snails from
Pacific islands (Mollusca: Pulmonata: Sigmure-
thra). Part Il. Families Punctidae and Charop-
idae, Zoogeography. Field Museum of Natural
History, Chicago, 336pp.

SOLEM, A., 1983, Lost or kept internal whorls: or-
dinal differences in land snails. Journal of Mol-
luscan Studies, supplement 12A: 172- 178.

SOLEM, A., 1984, A world model of land snail di-
versity and abundance. Pp. 6-22 in: SOLEM, A.
& VAN BRUGGEN, A.C., eds., 1984, World-wide
snails. Brill, Leiden.

SOLEM, A., TILLIER, S. & MORDAN, P., 1984,
Pseudo-operculate pulmonate land snails from
New Caledonia. Veliger, 27(2): 193-199.

SOLEM, A. & YOCHELSON, E.L., 1979, North
American Palezoic land snails, with a summary of
other Paleozoic nonmarine snails. United States
Geological Survey Professional Paper 1072:
42pp.

STEENBERG, C., 1925, Etudes sur l'Anatomie et
la Systématique des Maillots (fam. Pupillidae
s.lat.). C.A. Reitzel, Copenhagen, 211pp.

TAYLOR, J.W., 1894-1900, Monograph of the land
and freshwater Mollusca of the British Isles,
Structural and general volume. Taylor Brothers,
Leeds, 434pp.

THOMPSON, D'A.W., 1917, On Growth and Form.
Cambridge University Press, 793pp.

TILLIER, S., 1980, Gastéropodes terrestres et flu-
viatiles de Guyane francaise. Mémoires du Mu-
séum national d'Histoire naturelle, série A (Zool-
ogie), 118: 188pp.

TILLIER, S., 1981, Clines, convergence and char-
acter displacement in New Caledonian diplom-
matinids (land prosobranchs). Malacologia, 21(1—
2): 177-208.

TILLIER, S., 1983, Structures respiratoires et ex-
crétrices secondaires des limaces. Bulletin de la
Société Zoologique de France, 108(1): 9-19.

TILLIER, S., 1984a, Patterns of digestive tract mor-
phology in the limacisation of helicarionid, succi-
neid and athoracophorid snails and slugs (Mol-
lusca: Pulmonata). Malacologia, 25(1): 173-192.

TILLIER, S., 1984b, Relationships of gymnomorph
gastropods. Zoological Journal of the Linnean
Society, 82(4): 345-362.

TILLIER, S., 1986, Reconstruction des phylogénies
et théorie de l'évolution. Pp. 205-213 in: TASSY,
P., ed., L'Ordre et la Diversité du Vivant, Fayard,
Paris.

TILLIER, S. & MORDAN, P.M., 1986, New Cale-
donian charopid land snails.l. A revision of the
genus Pararhytida Ancey, 1882 (Mollusca: Pul-
monata: Charopidae). Malacologia, 27(2): 1-39.

TURCHINI, J. 4 BROUSSY, J., 1934, Contribution
à l'étude de la glande pulmonaire de Zonites al-
girus L. Archives de Zoologie expérimentale et
générale, second volume jubilaire, 75: 639—
650.

VAN GOETHEM, J.J., 1977, Recherches systéma-
tiques sur les Urocyclinae (Mollusca, Gas-
tropoda, Urocyclidae). Annales du Musée Royal
de l'Afrique Centrale, serie in-8°, sciences
zoologiques, 218: 355pp.

VAN MOL, J.J., 1967, Etude morphologique et phy-
logénétique du ganglion cérébroide des gas-
téropodes pulmonés (Mollusques). Mémoires de
l'Académie Royale de Belgique, classes des sci-
ences, collection in-8°, 37: 168pp.

VAN MOL, J.J., 1970, Révision des Urocyclidae
(Mollusca, Gastropoda, Pulmonata). Premiere
partie. Annales du Musée Royal de l'Afrique Cen-
trale, série in-8°, sciences zoologiques, 180:
234pp.

VAN MOL, J.J., 1971, Notes anatomiques sur les
Bulimulidae (mollusques, gasteropodes, pul-
monés). Annales de la Société Royale Zoo-
logique de Belgique, 101: 183-226.

VAN MOL, J.J., 1978, Etude de la morphologie et
des affinités systématiques de Alllya camerunen-
sis. Revue de Zoologie et de Botanique Africaine,
92(4): 1003-1021.

VAN MOL, J.J. & VAN BRUGGEN, A.C., 1971, De-
scription d'un nouveau genre d'Euconulidae d'Af-
rique et discussion sur les affinités systéma-
tiques de cette famille (Mollusca, Gastropoda
Pulmonata). Revue de Zoologie et de Botanique
Africaine, 84(3—4): 285-296.

VAVILOV, М.1., 1922, The law of homologous series
in variation. Journal of Genetics, 12: 47-89.

VISSER, M.H.C., 1973, The ontogeny of the repro-

STYLOMMATOPHORAN SYSTEMATICS 99

ductive system of Gonaxis gwandaensis (Pres-
ton) with special reference to the phylogeny of the
spermatic conduits of the Pulmonata. Annale Uni-
versiteit van Stellenbosch, serie A, 48(4): 79pp.

VISSER, M.H.C., 1977, The morphology and sig-
nificance of the spermoviduct and prostate in the
evolution of the reproductive system of the Pul-
monata. Zoologica Scripta, 6: 43-54.

VISSER, M.H.C., 1981a, Monauly versus diauly as
the original condition ofthe reproductive system of
Pulmonata and its bearing on the interpretation of
the terminal ducts. Zeitschrift fúr zoologische Sys-
tematik und Evolutionforschung, 19(1): 59-68.

VISSER, M.H.C., 1981b, The reproductive system
of Oopelta polypunctata Collinge and O. nigro-
punctata Mörch (Pulmonata, Arionidae?), with re-
flections upon the systematic position of the sub-
family Oopeltinae. Zoologica Scripta, 10: 293-
314.

WÂCHTLER, W., 1934, Der Nierenapparat der Sty-
lommatophoren Lungenschnecken, vergleichend
anatomisch betrachtet. Zoologischer Anzelger,
105: 161-172;

WACHTLER, W., 1935, Zur äusseren Morphologie
des Fusses der monotrematen Land-Lun-
genschnecken. Jahrbücher der Akademie ge-
meinnútz Wissenschaften zu Erfurt, 52: 107-
1135:

WATSON, H., 1915, Studies on the carnivorous
slugs of South Africa, including a monograph of
the genus Apera, and a discussion on the phy-
logeny of the Aperidae, Testacellidae and other

ABBREVIATIONS

A anus

AD anterior duct of digestive gland
AG albumen gland

AO — aorta

AS stomachal appendix

AU auricle

BM _ buccal mass

BP pallial border

C crop

СС cerebral commissure

CG cerebral ganglion

CJ conjunctive tissue

DC diverticulum of duct of digestive gland
DS diverticulum of gastric pouch
GP pallial gland

GU ureteric groove

HD hermaphrodite duct

HG hermaphrodite gland

| intestine

K kidney

KO kidney роге

LOP раша! lobe

NO optic nerve

ОС oesophageal crop

OE oesophagus

OR ocular retractor muscle

2 penis

agnathous Pulmonata. Annals of the Natal Mu-
seum, 3: 107-267.

WATSON, H., 1920, The affinities of Pyramidula,
Patulastra, Acanthinula, and Vallonia. Proceed-
ings of the Malacological Society of London, 14:
6-30.

WATSON, H., 1922. Notes on the nomenclature of
Hygromia, Helicella, etc . . . Journal of Conchol-
ogy, 16(9): 277-285.

WATSON, H., 1928, The affinities of Cecilioides
and Ferussacia, illustrating adaptive evolution.
Journal of Conchology, 18(8): 217-243.

WATSON, H., 1934, Natalina and other South Af-
rican snails. Proceedings of the Malacological
Society of London, 21: 150-196.

WILEY, E.O., 1981, Phylogenetics. The Theory and
Practice of Phylogenetic Systematics. John
Wiley and Sons, New York, 439pp.

WOODRUFF, D.S., 1978, Evolution and adaptive
radiation of Cerion: A remarkably diverse group
of West Indian land snails. Malacologia, 17(2):
223-239.

WURTZ, C.B., 1955, The American Camaenidae
(Mollusca: Pulmonata). Proceedings of The
Academy of Natural Sciences of Philadelphia,
107: 99-143.

ZILCH, A., 1959-1960, Euthyneura. In WENZ, W.,
Handbuch der Paläozoologie, 6(2). Gebrüder
Borntraeger, Berlin, 833pp.

Ms. accepted 23 March, 1987

PAD right parietal ganglion

PAG left parietal ganglion

PD posterior duct of digestive gland
PG pedal ganglion

PGE pedal gland

PLD right pleural ganglion

PLG left pleural ganglion

PN pneumostome

PO reno-pericardial pore
PR penial retractor muscle
PS gastric pouch

R rectum

RM retractor muscle

RS radular sheath
SC gastric crop
SG salivary gland

SO spermoviduct

SP spermatheca
STA statocyst

TD digestive tract
TG genital apparatus
TRY. typhlosole

U ureter

UA ureteric appendix
UO ureteric pore

VA vagina

VD vas deferens

V ventricle

VG visceral ganglion

100 TILLIER

FIGS. 1-10: ACHATINELLIDAE. Nota bene: the scale line in each plate varies in value; the length stated in
an individual caption is the length of the scale line with respect to that figure.

1) Auriculella auricula, digestive tract. 4mm.

2) Auriculella auricula, nervous system in dorsal view. 2mm.

3) Idem, cerebral commissure cut. 2mm.

4) Elasmias sp., nervous system in dorsal view, cerebral commissure cut. 1mm.
5) Elasmias sp., pulmonary complex. 2mm.

6) Elasmias sp., kidney internal morphology. 2mm.

7) Lamellidea cf. pusilla, digestive tract. 2mm.

8) Lamellidea cf. pusilla, nervous system in dorsal view. 1mm.

9) Idem, cerebral commissure cut. 1mm.

10) Elasmias sp., digestive tract. 2mm.

101

STYLOMMATOPHORAN SYSTEMATICS

102 TILLIER

FIGS. 11-22: ACHATINELLIDAE

11) Tornatellides oblongus, digestive tract. 2mm.

12) Strobilus plicosa, digestive tract. 2mm.

13) Tekoulina pricei, digestive tract. 4mm.

14) Tornatellides oblongus, nervous system in dorsal view. 1mm.
15) Idem, cerebral commissure cut. 1mm.

16) Achatinella lorata, digestive tract. 8mm.

17) Strobilus plicosa, nervous system in dorsal view. 1mm.
18) Idem, cerebral commissure cut. 1mm.

19) Tekoulina pricei, nervous system in dorsal view. 1mm.
20) Idem, cerebral commissure cut. 1mm.

21) Achatinella lorata, nervous system in dorsal view. 2mm.
22) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

103

104 TILLIER

FIGS. 23-28: ACHATINELLIDAE, PARTULIDAE

23) Achatinella lorata, pulmonary complex. 8mm.

24) Achatinella lorata, kidney internal morphology. 4mm.

25) Eua expansa, internal morphology of gastric pouch. 2mm.
26) Eva expansa, digestive tract. 8mm.

‚ Eua expansa, nervous system in dorsal view. 2mm.

28) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

105

106 TILLIER

FIGS. 29-35: PARTULIDAE, AMASTRIDAE

29) Partula caledonica, digestive tract. 7mm.

30) Partula caledonica, kidney internal morphology. 4mm.

11) Idem, external morphology. 7mm.

3°) Partula caledonica, nervous system in dorsal view. 2mm.
33) Idem, cerebral commissure cut. 2mm.

34) Amastra pullata, nervous system in dorsal view. 2mm.
35) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS 107

108 TILLIER

FIGS. 36-42: AMASTRIDAE

36) Amastra pullata, digestive tract. 4mm.

37) Amastra pullata, morphology of gastric region. 3.8mm.

38) Leptachatina balteata, digestive tract. 3.8mm.

39) Leptachatina balteata, pulmonary complex. 7.6mm.

40) Leptachatina balteata, kidney internal morphology. 4mm.
41) Leptachatina balteata, nervous system in dorsal view. 1mm.
42) Idem, cerebral commissure cut. 1mm.

STYLOMMATOPHORAN SYSTEMATICS 109

110 TILLIER

FIGS. 43-51: COCHLICOPIDAE, PYRAMIDULIDAE
43) Cochlicopa lubrica, digestive tract. 4mm.
44) Cochlicopa lubrica, pulmonary complex. 4mm.
45) Cochlicopa lubrica, kidney internal morphology. 2mm.
46) Pyramidula rupestris, nervous system in dorsal view. 0.5mm.
7) Cochlicopa lubrica, nervous system in dorsal view. 1mm.
idem, cerebral commissure cut. 1mm.
'yramidula rupestris, digestive tract. 0.9mm.
ramidula rupestris, pulmonary complex. 1.9mm.
ила rupestris, kidney internal morphology. 1mm.

STYLOMMATOPHORAN SYSTEMATICS

111

112 TILLIER

FIGS. 52-64: VERTIGINIDAE, ORCULIDAE
52) Bothriopupa breviconus, digestive tract. 1mm.
53) Bothriopupa breviconus, nervous system in dorsal view. 1mm.
54) Idem, cerebral commissure cut. 1mm.
55) Sterkia eyriesii, digestive tract. 1mm.
56) Pagodulina serveri, digestive tract. 2mm.
57) Pagodulina serveri, pulmonary complex. 3mm.
58) Orcula dolium, pulmonary complex. 4.3mm.
}\ Orcula dolium, kidney internal morphology. 2.5mm.
y Orcula dolium, digestive tract. 2.9mm.
| Pagodulina serveri, nervous system in dorsal view. 1.3mm.
Idem, cerebral commissure cut. 1.3mm.
63) Orcula dolium, nervous system in dorsal view. 1.25mm.
64) Idem, cerebral commissure cut. 1.25mm.

113

STYLOMMATOPHORAN SYSTEMATICS

114 TILLIER

FIGS. 65-73: CHONDRINIDAE
65) Solatopupa similis, digestive tract. 2.6mm.
66) Solatopupa similis, pulmonary complex. 2.8mm.
57) Solatopupa similis, Kidney internal morphology. 2mm.
68) Solatopupa similis, morphology of gastric region. 2mm.
69) Solatopupa similis, nervous system in dorsal view. 1mm.
Idem, cerebral commissure cut. 1mm.
Svliotrachela depressispira, digestive tract. 1mm.
Svliotrachela depressispira, nervous system in dorsal view. 0.5mm.
m, cerebral commissure cut. 0.5mm.

11,5

STYLOMMATOPHORAN SYSTEMATICS

116 TILLIER

FIGS. 74-85: PUPILLIDAE, VALLONIIDAE

74) Pupilla muscorum, digestive tract. 2mm.

75) Pupilla muscorum, pulmonary complex. 2mm.

76) Pupilla muscorum, kidney internal morphology. 2mm.

77) Lauria cylindracea, nervous system in dorsal view. 1.2mm.

78) Idem, cerebral commissure cut. 1.2mm.

79) Pupilla muscorum, nervous system in dorsal view. 1mm.

80) Idem, cerebral commissure cut. 1mm.

81) Lauria cylindracea, digestive tract. 2mm.
') Pupilla muscorum, morphology of gastric region. 2mm.
wria cylindracea, morphology of gastric region. 2mm.

34) Vallonia albula, digestive tract. 1mm.

25) Vallonia albula, nervous system in dorsal view, cerebral commissure cut. 1mm.

STYLOMMATOPHORAN SYSTEMATICS

118 TILLIER

FIGS. 86-94: VALLONIIDAE

86) Acanthinula aculeata, digestive tract. 1mm.

87) Acanthinula aculeata, pulmonary complex. 1mm.

88) Acanthinula aculeata, nervous system in dorsal view. 1mm.

89) Idem, cerebral commissure cut. 1mm.

90) Ptychopatula dioscoricola, digestive tract. 1mm.

91) Ptychopatula dioscoricola, pulmonary complex. 1.8mm.

92) Ptychopatula dioscoricola, kidney internal morphology. 1.125mm.

93) Ptychopatula dioscoricola, nervous system in dorsal view, cerebral commissure cut. 0.6mm.
94) Ptychopatula dioscoricola, nervous system in dorsal view. 0.6mm.

STYLOMMATOPHORAN SYSTEMATICS 119

86 87

120 TILLIER

FIGS. 95-104: VALLONIIDAE

95) Spermodea lamellata, digestive tract. 1mm.

96) Spermodea lamellata, nervous system in right lateral view. 1mm.
97) Spermodea lamellata, nervous system in dorsal view, cerebral commissure cut. 1mm.
98) Strobilops aenea, pulmonary complex. 1mm.

99) Strobilops aenea, kidney internal morphology. 1mm.

100) Spermodea lamellata, pulmonary complex. 1.9mm.

101) Spermodea lamellata, kidney internal morphology. 1mm.

102) Strobilops aenea, nervous system in dorsal view. 0.54mm.

103) Idem, cerebral commissure cut. 0.54mm.

104) Strobilops aenea, digestive tract. 1mm.

121

STYLOMMATOPHORAN SYSTEMATICS

122 TILLIER

FIGS. 105-115: VALLONIIDAE, ENIDAE

105) Klemmia magnicosta, digestive tract. 1.6mm.

106) Klemmia magnicosta, pulmonary complex. 2.7mm.

107) Klemmia magnicosta, kidney internal morphology. 1.3mm.
108) Chondrula tridens, digestive tract. 4mm.

109) Chondrula tridens, morphology of gastric region. 3mm.
110) Klemmia magnicosta, nervous system in dorsal view. 1.3mm.
111) Idem, cerebral commissure cut. 1.3mm.

112) Chondrula tridens, pulmonary complex. 6.5mm.

113) Chondrula tridens, kidney internal morphology. 3.3mm.
114) Chondrula tridens, nervous system in dorsal view. 1.4mm.
115) Idem, cerebral commissure cut. 1.4mm.

123

STYLOMMATOPHORAN SYSTEMATICS

124 TILLIER

FIGS. 116-125: ENIDAE

116) Imparietula jousseaumei, digestive tract. 3.6mm.

117) Imparietula jousseaumei, pulmonary complex. 7.2mm.

118) Imparietula jousseaumei, kidney internal morphology. 3.6mm.
119) Imparietula jousseaumei, nervous system in dorsal view. 1.25mm.
120) Idem, cerebral commissure cut. 1.25mm.

121) Zebrina detrita, digestive tract. 7.2mm.

122) Zebrina detrita, pulmonary complex. 5mm.

123) Zebrina detrita, kidney internal morphology. 5.6mm.

124) Zebrina detrita, nervous system in dorsal view. 1.8mm.

125) Idem, cerebral commissure cut. 1.8mm.

125

STYLOMMATOPHORAN SYSTEMATICS

126 TILLIER

FIGS. 126-134: ENIDAE

126) Ena montana, digestive tract. 4.4mm.

127) Ena montana, pulmonary complex. 8.8mm.

128) Ena montana, kidney internal morphology. 8.8mm.

129) Cerastua somaliensis, digestive tract. 5.8mm.

130) Ena montana, nervous system in dorsal view. 1mm.

131) Idem, cerebral commissure cut. 1mm.

132) Cerastua somaliensis, nervous system in dorsal view. 2mm.
123) /dem, cerebral commissure cut. 2mm.

1 Cerastua somaliensis, pulmonary complex. 5.8mm.

127

STYLOMMATOPHORAN SYSTEMATICS

128 TILLIER

FIGS. 135-146: ENIDAE

135) Rachistia histrio, digestive tract. 9mm.

136) Rachistia histrio, morphology of gastric region. 9mm.

137) Rachistia histrio, pulmonary complex. 9mm.

138) Rachistia histrio, kidney internal morphology. 4.5mm.

139) Rachistia histrio, nervous system in dorsal view. 2mm.

140) Idem, cerebral commissure cut. 2mm.

141) Draparnaudia michaudi, pulmonary complex. 9mm.

142) Draparnaudia michaudi, kidney internal morphology. 4.5mm.
143) Draparnaudia michaudi, nervous system in dorsal view. 2mm.
144) Idem, cerebral commissure cut. 2mm.

145) Amimopina macleayi, nervous system in dorsal view. 2mm.
146) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

129

130 TILLIER

FIGS. 147-155: ENDODONTIDAE, CHAROPIDAE

147) Thaumatodon hystricelloides, digestive tract. 2mm.

148) Thaumatodon hystricelloides, nervous system in dorsal view. 1mm.
149) Idem, cerebral commissure cut. 1mm.

150) Libera fratercula, kidney internal morphology. 2mm.

151) Libera fratercula, pulmonary complex. 2mm.

152) Libera fratercula, digestive tract. 2.4mm.

153) Andrefrancia sp., nervous system in dorsal view. 1mm.

154) Idem, cerebral commissure cut. 1mm.

155) Andrefrancia sp., digestive tract. 4mm.

131

STYLOMMATOPHORAN SYSTEMATICS

Se 7

>
ACTE EST
LANCER

132 TILLIER

FIGS. 156-166: CHAROPIDAE

156) Charopidae sp., digestive tract. 2.25mm.

157) Charopidae sp., pulmonary complex. 4.5mm.

158) Charopidae sp., kidney internal morphology. 4.5mm.

159) Annoselix dolosa, digestive tract. 6.5mm.

160) Annoselix dolosa, nervous system in dorsal view. 1mm.

161) Idem, cerebral commissure cut. 1mm.

162) Charopidae sp., nervous system in dorsal view, cerebral commissure cut. 1mm.
163) Charopidae sp., nervous system in dorsal view. 1mm.

164) Trachycystis capensis, digestive tract. 2mm.

165) Trachycystis capensis, nervous system in dorsal view, cerebral commissure cut. 1.1mm.
166) Trachycystis capensis, nervous system in dorsal view. 1.1mm.

133

STYLOMMATOPHORAN SYSTEMATICS

134 TILLIER

FIGS. 167-177: CHAROPIDAE, PUNCTIDAE

167) Stephanoda binneyana, digestive tract. 9mm.

168) Stephanoda binneyana, pulmonary complex. 9mm.

169) Stephanoda binneyana, kidney internal morphology. 4.5mm.

170) Stephanoda binneyana, nervous system in dorsal view. 2mm.

171) Idem, cerebral commissure cut. 2mm.

172) Ranfurlya constanceae, nervous system in dorsal view. 2mm.

173) idem, cerebral commissure cut. 2mm.

174) Ranfurlya constanceae, digestive tract. 4mm.

1/5) Cystopelta purpurea, digestive tract. 0.8mm.

176) Сузюрейа purpurea, nervous system in dorsal view. 2mm.
7) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

135

136 TILLIER

FIGS. 178-186: PUNCTIDAE, DISCIDAE

178) Phrixgnathus erigone, digestive tract. 2mm.

179) Phrixgnathus erigone, nervous system in dorsal view. 1mm.

180) Idem, cerebral commissure cut. 1mm.

181) Paralaoma lateumbilicata, digestive tract. 1mm.

182) Phrixgnathus erigone, internal morphology of cardiac arm of kidney. 4mm.
183) Phrixgnathus erigone, internal morphology of rectal arm of kidney. 4mm.
184) Helicodiscus parallelus, digestive tract. 1.3mm.

185) Discus patulus, pulmonary complex. 4.5mm.

186) Discus patulus, kidney internal morphology. 1mm.

STYLOMMATOPHORAN SYSTEMATICS 137

178

PAG VG PAD

138 TILLIER

FIGS. 187-194: DISCIDAE

187) Discus rotundatus, digestive tract. 2.5mm.

188) Discus rotundatus, kidney internal morphology. 1mm.

189) Discus rotundatus, pulmonary complex. 4.5mm.

190) Anguispira alternata, digestive tract. 6mm.

191) Anguispira alternata, nervous system in dorsal view. 2mm.
192) Idem, cerebral commissure cut. 2mm.

193) Anguispira alternata, kidney internal morphology. 4.5mm.
194) Anguispira alternata, pulmonary complex. 9mm.

STYLOMMATOPHORAN SYSTEMATICS 139

140 TILLIER

FIGS. 195-203: ARIONIDAE

195) Hemphillia camelus, pulmonary complex. 4mm.

196) Idem, pneumostome and secondary ureter opened. 4mm.

197) Idem, pericardium and primary ureter opened. 4mm.

198) Idem, detail: positions of kidney pore (KO) and reno-perocardial pore (PO). 4mm.
199) Idem, internal morphology of secondary ureter. 1.7mm.

200) Hemphillia camelus, digestive tract. 8mm.

201) Hemphillia camelus, morphology of gastric region. 4mm.

202) Hemphillia camelus, nervous system in dorsal view. 1.7mm.

203) Idem, cerebral commissure cut: internal side of left cerebral ganglion. 1.7mm.

141

STYLOMMATOPHORAN SYSTEMATICS

200

203

142 TILLIER

FIGS. 204-212: ARIONIDAE (including PHILOMYCINAE)

204) Prophysaon humile, digestive tract. 9mm.

205) Arion rufus, digestive tract. 13.8mm.

206) Oopelta granulosa, digestive tract. 13.8mm.

207) Oopelta nigropunctata, nervous system in dorsal view, cerebral commissure cut. 6mm.

208) Oopelta nigropunctata, nervous system in dorsal view. 6mm.

209) Philomycus carolinianus, ventral side of left cerebral ganglion. 2mm.

210) Philomycus carolinianus, nervous system in dorsal view. 2mm.

211) Philomycus carolinianus, pulmonary complex, secondary ureter opened, showing internal folds. 4mm.
212) Philomycus carolinianus, pulmonary complex, kidney cut and primary ureter opened. 4mm.

143

STYLOMMATOPHORAN SYSTEMATICS

144 TILLIER

FIGS. 213-217: ARIONIDAE (including PHILOMYCINAE), PUNCTIDAE
213) Philomycus bilineatus, digestive tract. 8.6mm.

214) Philomycus carolinianus, diagram of relative positions of organs and of internal partitions in general
cavity. 6mm.

215) Philomycus carolinianus, digestive tract. 6mm.

216) Cystopelta purpurea, diagram of relative positions of organs and of internal partitions in general cavity.
0.57mm.

217) Hemphillia camelus, diagram of relative positions of organs and of internal partitions in general cavity.
6mm.

STYLOMMATOPHORAN SYSTEMATICS

Vax
ei

145

146 TILLIER

FIGS. 218-220: ATHORACOPHORIDAE
218) Aneitea simrothi, digestive tract. 7.6mm.
219) Athoracophoridae sp., digestive tract. 8mm.

220) Athoracophoridae sp., nervous system in dorsal view, cerebral commissure cut. 1mm.
(from Tillier, 1984a)

147

STYLOMMATOPHORAN SYSTEMATICS

220

148 MEBIER

FIGS. 221-227: SUCCINEIDAE

221) Succinea putris, digestive tract. 6.4mm.

222) Succinea putris, nervous system in dorsal view. 2mm.
223) Idem, cerebral commissure cut. 2mm.

224) Succinea putris, pulmonary complex. 8mm.

225) Succinea putris, kidney internal morphology. 4.5mm.
226) Succinea propinqua, digestive tract. 4.5mm.

227) Omalonyx matheroni, digestive tract. 10mm.

(Figs. 221, 226, 227 from Tillier, 1984a)

STYLOMMATOPHORAN SYSTEMATICS

149

150 TILLIER

FIGS. 228-234: VITRINIDAE

228) Phenacolimax major, digestive tract. 3.8mm.

229) Phenacolimax? ugandensis, digestive tract. 5mm.

230) Phenacolimax major, nervous system in dorsal view. 2.2mm.

231) Idem, cerebral commissure cut. 2.2mm.

232) Plutonia atlantica, digestive tract. 7mm.

233) Plutonia atlantica, nervous system in dorsal view, cerebral commissure cut. 0.75mm.
234) Plutonia atlantica, nervous system in dorsal view. 0.75mm.

STYLOMMATOPHORAN SYSTEMATICS

151

152 TILLIER

FIGS. 235-244: ZONITIDAE

235) Vitrea crystallina, digestive tract. 1.6mm.

236) Vitrea crystallina, kidney internal morphology. 1.4mm.

237) Vitrea crystallina, pulmonary complex. 1.6mm.

238) Oxychilus draparnaudi, nervous system in dorsal view. 2.25mm.
239) Idem, cerebral commissure cut. 2.25mm.

240) Oxychilus draparnaudi, digestive tract. 10.6mm.

241) Aegopinella nitidula, digestive tract. 4.3mm.

242) Aegopinella nititula, morphology of proximal intestine. 4.3mm.
243) Aegopinella nitidula, nervous system in dorsal view, cerebral commissure cut. 2mm.
244) Aegopinella nitidula, nervous system in dorsal view. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

153

154 TILLIER

FIGS. 245-253: ZONITIDAE

245) Aegopinella nitidula, pulmonary complex. 4.7mm.

246) Aegopinella nitidula, kidney internal morphology. 3.3mm.

247) Daudebardia lederi, digestive tract. 4.3mm.

248) Mesomphix inornatus, digestive tract. 5mm.

249) Mesomphix inornatus, pulmonary complex. 7mm.

250) Daudebardia sp., nervous system in dorsal view, cerebral commissure cut. 3.3mm.
251) Daudebardia sp., nervous system in dorsal view. 3.3mm.

252) Mesomphix inornatus, nervous system in dorsal view. 2mm.

253) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

155

156 TIEBIER

FIGS. 254-258: ZONITIDAE

254) Zonites algirus, digestive tract. 16mm.

255) Zonites algirus, pulmonary complex. 11.6mm.

256) Zonites algirus, kidney internal morphology. 6.6mm.
257) Zonites algirus, nervous system in dorsal view. 6mm.
258) Idem, cerebral commissure cut. 6mm.

STYLOMMATOPHORAN SYSTEMATICS 157

158 TILLIER

FIGS. 259-268: PARMACELLIDAE, LIMACIDAE, MILACIDAE

259) Parmacella valenciennesi, digestive tract. 11mm.

260) Parmacella deshayesi, nervous system in dorsal view, cerebral commissure cut. 3.15mm.
261) Parmacella valenciennesi, nervous system in dorsal view, cerebral commissure cut. 3.15mm.
262) Parmacella valenciennesi, nervous system in dorsal view. 3.15mm.

263) Limax maximus, digestive tract. 20.6mm.

264) Milax gagates, digestive tract. 12.7mm.

265) Limax maximus, nervous system in dorsal view. 3mm.

266) Idem, cerebral commissure cut. 3mm.

267) Milax gagates, nervous system in dorsal view. 2.5mm.

268) Idem, cerebral commissure cut. 2.5mm.

STYLOMMATOPHORAN SYSTEMATICS

159

160 TILLIER

FIGS. 269-277: ZONITIDAE (GASTRODONTINAE)

269) Ventridens acera, digestive tract. 9mm.

270) Ventridens acera, nervous system in dorsal view. 4.5mm.
271) Idem, cerebral commissure cut. 4.5mm.

272) Gastrodonta interna, nervous system in dorsal view. 1mm.
273) Idem, cerebral commissure cut. 1mm.

274) Ventridens acera, pulmonary complex. 13.8mm.

275) Ventridens acera, kidney internal morphology. 2mm.

276) Gastrodonta interna, digestive tract. 4.5mm.

277) Zonitoides arboreus, digestive tract. 2mm.

STYLOMMATOPHORAN SYSTEMATICS 161

269

272

275

162 TILLIER

FIGS. 278-287: TROCHOMORPHIDAE, EUCONULIDAE
278) Trochomorpha sp., digestive tract. 3.8mm.

279) Trochomorpha sp., pulmonary complex. 8.5mm.

280) Conibycus cf. dahli, digestive tract. 4mm.

281) Trochomorpha sp., nervous system in dorsal view. 2mm.
282) Idem, cerebral commissure cut. 2mm.

283) Discoconulus sp. 1, nervous system in dorsal view. 1.4mm.
284) Idem, cerebral commissure cut. 1.4mm.

285) Discoconulus sp.1, pulmonary complex. 2.7mm.

286) Discoconulus sp.1, digestive tract. 4.7mm.

287) Discoconulus sp.2, pulmonary complex. 2.7mm.

STYLOMMATOPHORAN SYSTEMATICS

163

164 TILLIER

FIGS. 288-295: EUCONULIDAE

288) Coneuplecta sp.3, digestive tract. 2.5mm.

289) Coneuplecta sp.3, pulmonary complex. 3.7mm.

290) Coneuplecta sp.3, nervous system in dorsal view. 1.3mm.
291) Idem, cerebral commissure cut. 1.3mm.

292) Coneuplecta sp.1, digestive tract. 3mm.

293) Coneuplecta sp.2, pulmonary complex. 3.7mm.

294) Vitrinopsis sp.1, pulmonary complex. 2.5mm.

295) Vitrinopsis sp.1, digestive tract. 3mm.

294

STYLOMMATOPHORAN SYSTEMATICS

165

166 TILLIER

FIGS. 296-302: HELICARIONIDAE (HELICARIONINAE)

296) Kalidos oleatus, digestive tract. 19.5mm.

297) Kalidos oleatus, pulmonary complex. 19.5mm.

298) Helicarioninae sp., digestive tract. 4.8mm.

299) Helicarioninae sp., nervous system in dorsal view. 3.9mm.
300) Idem, cerebral commissure cut. 3.9mm.

301) Helicarioninae sp., pulmonary complex. 4.8mm.

302) Helicarioninae sp., kidney internal morphology. 2.3mm.
(Figs. 296, 298 from Tillier, 1984a)

STYLOMMATOPHORAN SYSTEMATICS

167

168 TILLIER

FIGS. 303-309: HELICARIONIDAE (DYAKIINAE, PARMARIONINAE)
303) Everettia corrugata, digestive tract. 9.2mm.

304) Everettia corrugata, pulmonary complex. 9.2mm.

305) Everettia corrugata, nervous system in dorsal view. 2.5mm.

306) Idem, cerebral commissure cut. 2.5mm.

307) Parmarion martensi, digestive tract. 3.1mm.

308) Microparmarion pollonerai, nervous system in dorsal view. 2.6mm.
309) Idem, cerebral commissure cut. 2.6mm.

(Fig. 307 from Tillier, 1984a)

169

STYLOMMATOPHORAN SYSTEMATICS

170 TILLIER

FIG. 310-317: HELICARIONIDAE (ARIOPHANTINAE, TROCHOZONITINAE)
310) Hemiplecta humphreysiana, digestive tract. 2cm.

311) Mariaella dussumieri, digestive tract. 10.6mm.

312) Hemiplecta humphreysiana, nervous system in dorsal view. 5.1mm.
313) Idem, cerebral commissure cut. 5.1mm.

314) Trochozonites percarinatus, nervous system in dorsal view. 2mm.

315) Idem, cerebral commissure cut. 2mm.

316) Trochozonites percarinatus, digestive tract. 5.2mm.

317) Trochozonites percarinatus, pulmonary complex. 8mm.

STYLOMMATOPHORAN SYSTEMATICS ral

Wg TILLIER

FIGS. 318-324: HELICARIONIDAE (UROCYCLINAE, GYMNARIONINAE)
318) Trochonanina simulans, digestive tract. 6.2mm.

319) Trochonanina simulans, nervous system in dorsal view. 2.25mm.
320) Idem, cerebral commissure cut. 2.25mm.

321) Acantharion browni, nervous system in dorsal view. 3.1mm.

322) Idem, cerebral commissure cut. 3.1mm.

323) Acantharion browni, digestive tract. 10mm.

324) Gymnarion sowerbyanus, digestive tract. 10mm.

(Figs. 321-324 from Binder & Tillier, 1986)

STYLOMMATOPHORAN SYSTEMATICS 173

318

319

174 TILLIER

FIGS. 325-333: HELICARIONIDAE (GYMNARIONINAE, UROCYCLINAE)
325) Acantharion browni, pulmonary complex. 5.6mm.

326) Gymnarion sowerbyanus, pulmonary complex. 8.5mm.

327) Acantharion browni, kidney internal morphology. 5.6mm.

328) Granularion lamottei, digestive tract. 7.6mm.

329) Mesafricarion maculifer, digestive tract. 6mm.

330) Granularion lamottei, cerebral ganglia in a dorsal view. 3.8mm.

331) Idem., nervous system in dorsal view, cerebral commissure cut. 3.8mm.
332) Mesafricarion maculifer, nervous system in dorsal view. 2.5mm.

333) Idem, cerebral commissure cut. 2.5mm.

(Figs. 328, 329 from Tillier, 1984a)

STYLOMMATOPHORAN SYSTEMATICS

175

176 TILLIER

FIGS. 334-340: HELICARIONIDAE (UROCYCLINAE)
334) Tresia parva, digestive tract. 2.5mm.

335) Estria? sp., digestive tract. 5mm.

336) Atoxon pallens, kidney internal morphology. 2.5mm.
337) Elisolimax madagascariensis, digestive tract. 5mm.
338) Atoxon pallens, digestive tract. 5mm.

339) Atoxon pallens, nervous system in dorsal view. 1mm.
340) Idem, cerebral commissure cut. 1mm.

(Fias. 334, 335, 337, 338 from Tillier, 1984a)

17474

STYLOMMATOPHORAN SYSTEMATICS

178 TILLIER

FIGS. 341-348: SUBULINIDAE

341) Bocageia carpenteri, nervous system in dorsal view. 2.5mm.
342) Idem, cerebral commissure cut. 2.5mm.

343) Bocageia carpenteri, digestive tract. 11.3mm.

344) Pseudoglessula hessei, digestive tract. 4.5mm.

345) Pseudoglessula hessei, nervous system in dorsal view. 2mm.
346) Idem, cerebral commissure cut. 2mm.

347) Pseudoglessula hessei, pulmonary complex. 5mm.

348) Pseudoglessula hessei, kidney internal morphology. 2.8mm.

179

STYLOMMATOPHORAN SYSTEMATICS

D
[\

SRE

348

180 TILLIER

FIGS. 349-355: SUBULINIDAE, ACHATINIDAE

349) Rumina decollata, pulmonary complex. 7.9mm.

350) Rumina decollata, digestive tract. 7.9mm.

351) Rumina decollata, nervous system in dorsal view. 3.25mm.
352) Idem, cerebral commissure cut. 3.25mm.

353) Achatina fulica, digestive tract. 1.9mm.

354) Achatina fulica, nervous system in dorsal view. 5mm.

355) Idem, cerebral commissure cut. 5mm.

181

STYLOMMATOPHORAN SYSTEMATICS

182 TIELIER

FIGS. 356-363: FERUSSACIIDAE, OLEACINIDAE
356) Cecilioides acicula, digestive tract. 2mm.
357) Poiretia dilatata, digestive tract. 5.8mm.
358) Cecilioides acicula, pulmonary complex. 2mm.
359) Cecilioides acicula, nervous system in dorsal view. 1.3mm.
360) Idem, cerebral commissure cut. 1.3mm.
361) Poiretia dilatata, pulmonary complex seen from outside. 5.8mm.
552) Poiretia dilatata, nervous system in dorsal view. 3.6mm.
363) Idem, cerebral commissure cut. 3.6mm.

STYLOMMATOPHORAN SYSTEMATICS

183

184 TILLIER

FIGS. 364-371: OLEACINIDAE

364) Spiraxis futilis, digestive tract. 1.4mm.

365) Varicella biplicata, digestive tract. 9mm.

366) Varicella biplicata, pulmonary complex. 4.4mm.

367) Varicella biplicata, kidney internal morphology. 4.4mm.

368) Varicella biplicata, nervous system in dorsal view. 2mm.

369) Varicella biplicata, right lateral view of nervous system. 2mm.

370) Varicella biplicata, nervous system in dorsal view, cerebral commissure cut. 1mm.
371) Euglandina carminensis, digestive tract. ca. 14mm.

185

STYLOMMATOPHORAN SYSTEMATICS

186

FIGS. 372-379: OLEAC

TILLIER

INIDAE, RHYTIDIDAE

372) Strebelia berendti, digestive tract. 4.5mm.
373) Strebelia berendti, nervous system in dorsal view. 2mm.

375) Priodiscus serratus
376) Priodiscus serratus
377) Priodiscus serratus
378) Priodiscus serratus
379) Priodiscus serratus

)

374) Idem, cerebral com
)
)

missure cut. 2mm.

, digestive tract. 2.3mm.

, nervous system in dorsal view, cerebral commissure cut. 1mm.
, nervous system in dorsal view. 1mm.

, pulmonary complex. 2.3mm.

, genital apparatus. 1.9mm.

STYLOMMATOPHORAN SYSTEMATICS

187

188 TEMER

FIGS. 380-387: RHYTIDIDAE

380) Rhytida inaequalis, digestive tract. 11.9mm.

381) Nata vernicosa, digestive tract. 5mm.

382) Nata vernicosa, pulmonary complex. 7mm.

383) Nata vernicosa, kidney internal morphology. 4.1mm.
384) Nata vernicosa, nervous system in dorsal view. 2.1mm.
385) Idem, cerebral commissure cut. 2.1mm.

386) Ouagapia raynali, nervous system in dorsal view. 5mm.
387) Idem, cerebral commissure cut. 5mm.

STYLOMMATOPHORAN SYSTEMATICS 189

190 TILLIER

FIGS. 388-393: RHYTIDIDAE

388) Diplomphalus megei, digestive tract. 6.5mm.

389) Diplomphalus megei, pulmonary complex. 6.5mm.

390) Diplomphalus megei, nervous system in dorsal view. 2mm.
391) Idem, cerebral commissure cut. 2mm.

392) Chlamydephorus gibbonsi, digestive tract. 1.5mm.

393) Schizoglossa novoseelandica, digestive tract. 10mm.

STYLOMMATOPHORAN SYSTEMATICS

192 TILLIER

FIGS. 394-398: STREPTAXIDAE

394) Ptychotrema sp., digestive tract. 5.9mm.

395) Ptychotrema sp., pulmonary complex. 10mm.

396) Ptychotrema sp., nervous system in dorsal view. 1.9mm.
397) Edentulina sp., digestive tract. 10mm.

398) Edentulina sp., pulmonary complex. 10mm.

193

STYLOMMATOPHORAN SYSTEMATICS

194 TILLIER

FIGS. 399-405: SYSTROPHIIDAE

399) Systrophia eudiscus, digestive tract. 6.8mm.

400) Systrophia eudiscus, nervous system in dorsal view. 2mm.
401) Idem, cerebral commissure cut. 2mm.

402) Systrophia eudiscus, pulmonary complex. 11.3mm.

403) Systrophia eudiscus, kidney internal morphology. 4mm.
404) Tamayoa decolorata, digestive tract. 3.1mm.

405) Systrophia (Wayampia) cayennensis, digestive tract. 4mm.

STYLOMMATOPHORAN SYSTEMATICS

ur Deck HERR
и O ps =

a Peete Pt Dh
= LA

=
Di Cp AA ER WR

[a
Siem
ya

196 MLLIER

FIGS. 406-411: ACAVIDAE (Australia)

406) Caryodes sp., digestive tract. 10mm.

407) Caryodes sp., nervous system in dorsal view. 3mm.
408) /dem, cerebral commissure cut. 3mm.

409) Panda larreyi, digestive tract. 1.5mm.

410) Panda larreyi, nervous system in dorsal view. 5mm.
411) Idem, cerebral commissure cut. 5mm.

197

STYLOMMATOPHORAN SYSTEMATICS

198 TILLIER

FIGS. 412-416: ACAVIDAE (Australia)

412) Pygmipanda kershawi, digestive tract. 2cm.

413) Pygmipanda kershawi, nervous system in dorsal view. 2.3mm.
414) Idem, cerebral commissure cut. 2.3mm.

415) Pygmipanda kershawi, pulmonary complex. 2cm.

416) Pygmipanda kershawi, kidney internal morphology. 7mm.

STYLOMMATOPHORAN SYSTEMATICS

199

200 TILLIER

FIGS. 417-422: ACAVIDAE (Australia)

417) Hedleyella falconeri, digestive tract. 2cm.

418) Pandofella whitei, digestive tract. 1cm.

419) Pandofella whitei, nervous system in dorsal view. 3.25mm.
420) Idem, cerebral commissure cut. 3.25mm.

421) Hedleyella falconeri, nervous system in dorsal view. 4.2mm.
422) Idem, cerebral commissure cut. 4.2mm.

STYLOMMATOPHORAN SYSTEMATICS

202 TILEIER

FIGS. 423-427: ACAVIDAE (Australia)

423) Pedinogyra sp., nervous system in dorsal view. 4mm.
424) Idem, cerebral commissure cut. 4mm.

425) Pedinogyra sp., digestive tract. 1.7cm.

426) Pedinogyra sp., kidney internal morphology. 8mm.
427) Pedinogyra sp., pulmonary complex. 2.9cm.

STYLOMMATOPHORAN SYSTEMATICS 203

204 TILLIER

FIGS. 428-432: ACAVIDAE (South Africa)

428) Dorcasia alexandri, nervous system in dorsal view. 2mm.
429) Idem, cerebral commissure cut. 2mm.

430) Dorcasia alexandri, digestive tract. 7.8mm.

431) Dorcasia alexandri, pulmonary complex. 22.5mm.

432) Dorcasia alexandri, kidney internal morphology. 4.2mm.

205

STYLOMMATOPHORAN SYSTEMATICS

206 TWELIER

FIGS. 433-439: ACAVIDAE (South Africa, South America)

433) Trigonephrus rosaceus, digestive tract. 12mm.

434) Trigonephrus rosaceus, nervous system in dorsal view. 3.25mm.
435) Idem, cerebral commissure cut. 3.25mm.

436) Macrocyclis laxata, digestive tract. 2cm.

437) Macrocyclis laxata, nervous system in dorsal view. 6mm.

438) Idem, cerebral commissure cut. 6mm.

439) Macrocyclis laxata, kidney internal morphology. 6mm.

STYLOMMATOPHORAN SYSTEMATICS

207

208 MIELIER

FIGS. 440-449: ACAVIDAE (South America, Seychelles)
440) Strophocheilus chilensis, digestive tract. 6mm.
441) Strophocheilus chilensis, cerebral ganglia in dorsal view, sheath dissected. 3mm.
442) Strophocheilus chilensis, nervous system in dorsal view, sheath intact. 3mm.
443) Idem, cerebral commissure cut. 3mm.
444) Idem, cerebral commissure cut, conjunctive sheath dissected. 3mm.
445) Strophocheilus chilensis, pulmonary complex. 8.6mm.
446) Strophocheilus chilensis, kidney internal morphology. 3mm.
447) Stylodon studerianus, digestive tract. 17mm.
448) Stylodon studerianus, nervous system in dorsal view. 3.3mm.
)

449) Idem, cerebral commissure cut. 3.3mm.

STYLOMMATOPHORAN SYSTEMATICS 209

210 MEIER

FIGS. 450-456: ACAVIDAE (Seychelles, Madagascar)

450) Stylodon studerianus, pulmonary complex. 4.5cm.

451) Stylodon studerianus, kidney internal morphology. 2cm.

452) Ampelita petiti, pulmonary complex. 2.25cm.

453) Ampelita petiti, kidney internal morphology. 6.4mm.

454) Ampelita petiti, digestive tract. 10.6mm.

455) Ampelita petiti, nervous system in dorsal view, cerebral commissure cut. 3.5mm.
456) Ampelita petiti, nervous system in dorsal view. 3.5mm.

STYLOMMATOPHORAN SYSTEMATICS

211

212 TILLIER

FIGS. 457-464: ACAVIDAE (Madagascar)

457) Clavator eximius, digestive tract. 1.6cm.

458) Clavator eximius, nervous system in dorsal view. 4mm.

459) Idem, cerebral commissure cut. 4mm.

460) Helicophanta vesicalis, nervous system in dorsal view. 4mm.
461) Idem, cerebral commissure cut. 4mm.

462) Clavator eximius, pulmonary complex. 2.5cm.

463) Clavator eximius, kidney internal morphology. 8mm.

464) Helicophanta vesicalis, digestive tract. 2.3cm.

STYLOMMATOPHORAN SYSTEMATICS 213

AIN
/ má A
at AN
KEG RS
77 EN AN SEN ENS
. > Se N

214 TILLIER

FIGS. 465-469: ACAVIDAE (Madagascar, Ceylon)

465) Helicophanta vesicalis, pulmonary complex. 1.6cm.

466) Helicophanta vesicalis, kidney internal morphology. 8mm.
467) Acavus superbus, digestive tract. 1.6cm.

468) Acavus superbus, nervous system in dorsal view. 6.1mm.
469) Idem, cerebral commissure cut. 6.1mm.

215

STYLOMMATOPHORAN SYSTEMATICS

466

465

468

469

216 TILLIER

FIGS. 470-475: CORILLIDAE

470) Sculptaria collaris, digestive tract. 2mm.

471) Sculptaria collaris, pulmonary complex. 4.75mm.

472) Sculptaria collaris, kidney internal morphology. 1.9mm.
473) Sculptaria collaris, nervous system in dorsal view. 0.8mm.
474) Idem, cerebral commissure cut. 0.8mm.

475) Sculptaria collaris, genital apparatus. 3mm.

STYLOMMATOPHORAN SYSTEMATICS

217

218 MIELIER

FIGS. 476-485: CORILLIDAE

476) Craterodiscus pricei, digestive tract. 2mm.

477) Craterodiscus pricei, right lateral view of central nervous system. 1mm.
478) Craterodiscus pricei, nervous system in dorsal view. 1mm.
479) Idem, cerebral commissure cut. 1mm.

480) Craterodiscus pricei, kidney internal morphology. 1mm.
481) Plectopylis sp., digestive tract. 5mm.

482) Plectopylis Sp., pulmonary complex. 8mm.

483) Plectopylis sp., nervous system in dorsal view. 2mm.

484) Idem, cerebral commissure cut. 2mm.

485) Plectopylis sp., kidney internal morphology. 3.1mm.

STYLOMMATOPHORAN SYSTEMATICS 219

220 TILLIER

FIGS. 486-491: CORILLIDAE

486) Corilla humberti, digestive tract. 6.4mm.

487) Corilla humberti, nervous system in dorsal view. 1.4mm.
488) Idem, cerebral commissure cut. 2mm.

489) Corilla humberti, pulmonary complex. 10mm.

490) Corilla humberti, kidney internal morphology. 3.8mm.
491) Corilla humberti, genital apparatus. 10mm.

STYLOMMATOPHORAN SYSTEMATICS 221

222 TILLIER

FIGS. 492-499: OREOHELICIDAE (OREOHELICINAE, AMMONITELLINAE)
492) Oreohelix barbata, digestive tract. 5mm.

493) Oreohelix barbata, kidney internal morphology. 3.7mm.

494) Oreohelix barbata, pulmonary complex. 8.5mm.

495) Oreohelix barbata, nervous system in dorsal view. 2mm.

496) Idem, cerebral commissure cut. 2mm.

497) Ammonitella yatesi, digestive tract. 3.7mm.

498) Ammonitella yatesi, nervous system in dorsal view. 1.5mm.

499) Idem, cerebral commissure cut. 1.5mm.

223

STYLOMMATOPHORAN SYSTEMATICS

224 TILLIER

FIGS. 500-504: OREOHELICIDAE (AMMONITELLINAE)

500) Ammonitella yatesi, pulmonary complex. 6.3mm.

501) Ammonitella yatesi, kidney internal morphology. 2.5mm.

502) Glyptostoma gabrielense, digestive tract. 10.3mm.

503) Glyptostoma gabrielense, nervous system in dorsal view. 2.5mm.
504) Idem, cerebral commissure cut. 2.5mm.

STYLOMMATOPHORAN SYSTEMATICS

225

226 TILLIER

FIGS. 505-515: CLAUSILIIDAE

505) ltala itala, digestive tract. 4mm.

506) Itala itala, morphology of gastric region. 3.4mm.
507) Itala itala, kidney internal morphology. 1.9mm.
508) Itala itala, pulmonary complex. 5.6mm.

509) Itala itala, nervous system in dorsal view. 1.4mm.
510) Idem, cerebral commissure cut. 1.4mm.

511) Albinaria olivieri, nervous system in dorsal view. 2mm.
512) ldem, cerebral commissure cut. 2mm.

513) Albinaria olivieri, digestive tract. 4.25mm.

514) Albinaria olivieri, pulmonary complex. 8.5mm.

515) Albinaria olivieri, kidney internal morphology. 2mm.

227

STYLOMMATOPHORAN SYSTEMATICS

228 TILLIER

FIGS. 516-523: CLAUSILIIDAE, CERIONIDAE

516) Nenia tridens, digestive tract. 6mm.

517) Nenia tridens, pulmonary complex. 10.6mm.

518) Nenia tridens, kidney internal morphology. 3.4mm.

519) Nenia tridens, nervous system in dorsal view. 2mm.

520) Idem, cerebral commissure cut. 2mm.

521) Cerion casablancae, digestive tract. 10mm.

522) Cerion casablancae, nervous system in dorsal view. 2mm.
523) Idem, cerebral commissure cut. 2mm.

229

STYLOMMATOPHORAN SYSTEMATICS

230 TILLIER

FIGS. 524-531: CERIONIDAE, UROCOPTIDAE

524) Cerion copium, digestive tract. 10mm.

525) Cerion copium, morphology of gastric region. 5mm.

526) Cerion copium, kidney internal morphology. 5mm.

527) Cerion copium, pulmonary complex. 10mm.

528) Macroceramus signatus, pulmonary complex. 5mm.

529) Macroceramus signatus, digestive tract. 5mm.

530) Berendtia taylori, nervous system in dorsal view. 2.6mm.
)

531) Idem, cerebral commissure cut. 2.6mm.

STYLOMMATOPHORAN SYSTEMATICS 231

232 TILLIER

FIGS. 532-539: UROCOPTIDAE, BULIMULIDAE (ODONTOSTOMINAE)
532) Urocoptis procera, pulmonary complex. 10mm.

533) Berendtia taylori, pulmonary complex. 10mm.

534) Urocoptis procera, nervous system in dorsal view. 2mm.

535) Idem, cerebral commissure cut. 2mm.

536) Plagiodontes daedaleus, pulmonary complex. 10.6mm.

537) Plagiodontes daedaleus, digestive tract. 9.4mm.

538) Plagiodontes daedaleus, nervous system in dorsal view. 2.4mm.
539) Idem, cerebral commissure cut. 2.4mm.

STYLOMMATOPHORAN SYSTEMATICS 233

234 TILLIER

FIGS. 540-547: BULIMULIDAE (BULIMULINAE)
540) Bostryx bermudezae, digestive tract. 3.7mm.
541) Bostryx bermudezae, pulmonary complex. 5mm.
542) Bostryx bermudezae, kidney internal morphology. 2mm.
543) Bostryx bermudezae, nervous system in dorsal view. 1.5mm.
544) Idem, cerebral commissure cut. 1.5mm.
545) Discoleus azulensis, digestive tract. 3.9mm.
546) Discoleus azulensis, nervous system in dorsal view. 2mm.
)

547) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

235

236 TEMER

FIGS. 548-555: BULIMULIDAE (BULIMULINAE, AMPHIBULIMINAE)
548) Pellicula depressa, digestive tract. 0.8mm.

549) Simpulopsis miersi, nervous system in dorsal view. 2.5mm.
550) Idem, cerebral commissure cut. 2.5mm.

551) Pellicula depressa, nervous system in dorsal view. 2.5mm.

552) Idem, cerebral commissure cut. 2.5mm.

553) Simpulopsis miersi, pulmonary complex. 8.4mm.

554) Simpulopsis miersi, kidney internal morphology. 4.5mm.

555) Simpulopsis miersi, digestive tract. 5.8mm.

237

STYLOMMATOPHORAN SYSTEMATICS

238 TILLIER

FIGS. 556-561: CAMAENIDAE (?)

556) Solaropsis undata, digestive tract. 11mm.

557) Labyrinthus leprieurii, nervous system in dorsal view. 2mm.
558) Idem, cerebral commissure cut. 2mm.

559) Labyrinthus leprieurii, digestive tract. 5.9mm.

560) Labyrinthus leprieurii, pulmonary complex. 13mm.

561) Labyrinthus leprieurii, kidney internal morphology. 3.7mm.

239

STYLOMMATOPHORAN SYSTEMATICS

CASS RSS

240 TILLIER

FIGS. 562-568: CAMAENIDAE (?)

562) Pleurodonte lychnuchus, digestive tract. 8.4mm.

563) Pleurodonte lychnuchus, pulmonary complex. 13.5mm.

564) Pleurodonte lychnuchus, kidney internal morphology. 4.5mm.

565) Amphidromus cognatus, digestive tract. 10mm.

566) Pleurodonte lychnuchus, cerebral ganglia in dorsal view, sheath dissected. 2.5mm.
567) Pleurodonte lychnuchus, nervous system in dorsal view. 2.5mm.

568) Idem, cerebral commissure cut. 2.5mm.

241

STYLOMMATOPHORAN SYSTEMATICS

OWS
5 NE

AS => er NZ Dean

242 MÉÈPIER

FIGS. 569-576: CAMAENIDAE

569) Plectotropis goniocheila, digestive tract. этт.

570) Plectotropis goniocheila, nervous system in dorsal view. 1.6mm.
571) Idem, cerebral commissure cut. 1.8mm.

572) Amplirhagada burnerensis, nervous system in dorsal view. 4mm.
573) Idem, cerebral commissure cut. 4mm.

574) Amplirhagada burnerensis, digestive tract. 10mm.

575) Amplirhagada burnerensis, pulmonary complex. 20mm.

576) Amplirhagada burnerensis, kidney internal morphology. 6mm.

243

STYLOMMATOPHORAN SYSTEMATICS

244 TILLIER

FIGS. 577-583: POLYGYRIDAE (POLYGYRINAE)

577) Polygyra matermontana, digestive tract. 4.4mm.

578) Polygyra matermontana, kidney internal morphology. 3.6mm.
579) Polygyra matermontana, pulmonary complex. 5.4mm.

580) Trilobopsis trachypepla, digestive tract. 1.2mm.

581) Mesodon andrewsae, digestive tract. 10.6mm.

582) Polygyra matermontana, nervous system in dorsal view. 2mm.
583) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

246 TILLIER

FIGS. 584-591: POLYGYRIDAE (TRIODOPSINAE), BRADYBAENIDAE
584) Triodopsis fraudulenta, digestive tract. 8mm.

585) Triodopsis fraudulenta, nervous system in dorsal view. 2mm.

586) Idem, cerebral commissure cut. 2mm.

587) Bradybaena tourannensis, digestive tract. 6.8mm.

588) Bradybaena tourannensis, pulmonary complex. 6mm.

589) Bradybaena tourannensis, kidney internal morphology. 10.6mm.
590) Bradybaena tourannensis, nervous system in dorsal view. 2mm.
591) Idem, cerebral commissure cut. 2mm.

STYLOMMATOPHORAN SYSTEMATICS

248 TILLIER

FIGURES 592-599: BRADYBAENIDAE, HAPLOTREMATIDAE

592) Helicostylus palawanensis, digestive tract. 13mm.

593) Helicostylus palawanensis, nervous system in dorsal view. 5mm.
594) Idem, cerebral commissure cut. 5mm.

595) Haplotrema minimum, nervous system in dorsal view. 1.8mm.
596) Idem, cerebral commissure cut. 1.8mm.

597) Haplotrema concavum, digestive tract. 6.3mm.

598) Haplotrema concavum, pulmonary complex. 13mm.

599) Haplotrema concavum, kidney internal morphology. 6.3mm.

STYLOMMATOPHORAN SYSTEMATICS 249

250 TIELIER

FIGS. 600-606: SAGDIDAE

600) Sagda grandis, digestive tract. 6.7mm.

601) Sagda grandis, nervous system in dorsal view. 2.5mm.
602) /dem, cerebral commissure cut. 2.5mm.

603) Proserpinula opalina, nervous system in dorsal view. 2mm.
604) Idem, cerebral commissure cut. 2mm.

605) Proserpinula opalina, pulmonary complex. 2.7mm.

606) Proserpinula opalina, digestive tract. 2.5mm.

STYLOMMATOPHORAN SYSTEMATICS

251

252 TILLIER

FIGS. 607-614: HELMINTHOGLYPTIDAE (THYSANOPHORINAE, CEPOLINAE)
607) Thysanophora horni, digestive tract. 1.5mm.

608) Thysanophora horni, nervous system in dorsal view. 1mm.

609) Idem, cerebral commissure cut. 1mm.

610) Thysanophora horni, pulmonary complex. 0.9mm.

611) Thysanophora horni, kidney internal morphology. 2.2mm.

612) Cepolis varians, nervous system in dorsal view. 6.8mm.

613) Idem, cerebral commissure cut. 2.8mm.

614) Cepolis varians, digestive tract. 2.8mm.

253

STYLOMMATOPHORAN SYSTEMATICS

254 TILLIER

FIGS. 615-622: HELMINTHOGLYPTIDAE (CEPOLINAE)

615) Cepolis varians, kidney internal morphology. 5mm.

616) Cepolis varians, pulmonary complex. 8.75mm.

617) Psadara nubeculata, kidney internal morphology. 4.2mm.
618) Psadara nubeculata, pulmonary complex. 7.7mm.

619) Psadara nubeculata, nervous system in dorsal view. 1.6mm.
620) Idem, cerebral commissure cut. 1.6mm.

621) Cepolis maynardi, digestive tract. 5mm.

622) Psadara marmatensis, digestive tract. 3.3mm.

STYLOMMATOPHORAN SYSTEMATICS

255

256 MLCIER

FIGS. 623-629: HELMINTHOGLYPTIDAE (XANTHONYCINAE, SONORELLINAE)
623) Averellia coactiliata, digestive tract. 3.9mm.

624) Averellia coactiliata, nervous system in dorsal view. 2mm.

625) Idem, cerebral commissure cut. 2mm.

626) Sonorella walkeri, nervous system in dorsal view. 2mm.

627) Idem, cerebral commissure cut. 2mm.

628) Sonorella walkeri, pulmonary complex. 8.5mm.

629) Sonorella walkeri, digestive tract. 8.5mm.

STYLOMMATOPHORAN SYSTEMATICS 257

624 623

258 TILLIER

FIGS. 630-634: HELMINTHOGLYPTIDAE (HELMINTHOGLYPTINAE)
630) Helminthoglypta sequoicola, digestive tract, 10mm.

631) Helminthoglypta sequoicola, nervous system in dorsal view. 4.4mm.
632) Idem, cerebral commissure cut. 4.4mm.

633) Helminthoglypta sequoicola, kidney internal morphology. 10mm.
634) Helminthoglypta sequoicola, pulmonary complex. 16mm.

STYLOMMATOPHORAN SYSTEMATICS 259

260 TIEEIER

FIGS. 635-639: HELMINTHOGLYPTIDAE (HELMINTHOGLYPTINAE)

635) Epiphragmopora claromphalos, digestive tract. 10mm.

636) Epiphragmopora claromphalos, pulmonary complex. 1.7mm.

637) Epiphragmopora claromphalos, nervous system in dorsal view. 4.7mm.
638) Idem, cerebral commissure cut. 4.7mm.

639) Monadenia fidelis, digestive tract. 12.5mm.

261

STYLOMMATOPHORAN SYSTEMATICS

262 MIELIER

FIGS. 640-647: HELICIDAE (SPHINCTEROCHILINAE)

640) Sphincterochila zonata, digestive tract. 6.1mm.

641) Sphincterochila zonata, pulmonary complex. 10.3mm.

642) Sphincterochila zonata, kidney internal morphology. 5.9mm.

643) Sphincterochila zonata, nervous system in dorsal view, cerebral commissure cut. 2.5mm.
644) Sphincterochila zonata, nervous system in dorsal view. 2.5mm.
645) Halolimnohelix sericata, nervous system in dorsal view. 2mm.
646) Idem, cerebral commissure cut. 2mm.
647) Halolimnohelix sericata, digestive tract. 7.8mm.

STYLOMMATOPHORAN SYSTEMATICS

263

264 TILLIER

FIGS. 648-657: HELICIDAE (SPHINCTEROCHILINAE, HELICODONTINAE)
648) Halolimnohelix sericata, pulmonary complex. 10mm.

649) Halolimnohelix sericata, kidney internal morphology. 7.4mm.
650) Cochlicella acuta, digestive tract. 6.25mm.

651) Cochlicella acuta, nervous system in dorsal view. 2mm.

652) Idem, cerebral commissure cut. 2mm.

653) Helicodonta obvoluta, nervous system in dorsal view. 2.3mm.
654) Idem, cerebral commissure cut. 2.3mm.

655) Cochlicella acuta, pulmonary complex. 5.7mm.

656) Helicodonta obvoluta, pulmonary complex. 6.6mm.

657) Helicodonta obvoluta, digestive tract. 8mm.

STYLOMMATOPHORAN SYSTEMATICS 265

266 TILLIER

FIGS. 658-664: HELICIDAE (HYGROMIINAE)

658) Hygromia limbata, digestive tract. 7mm.

659) Trichia hispida, nervous system in dorsal view. 2mm.
660) Idem, cerebral commissure cut. 2mm.

661) Hygromia limbata, nervous system in dorsal view. 2mm.
662) Idem, cerebral commissure cut. 2mm.

663) Trichia hispida, pulmonary complex. 6.3mm.

664) Trichia hispida, kidney internal morphology. 3.1mm.

STYLOMMATOPHORAN SYSTEMATICS 267

268 TILLIER

FIGS. 665-671: HELICIDAE (HYGROMIINAE)

665) Helicopsis striata, digestive tract. 4.4mm.

666) Helicopsis striata, pulmonary complex. 4.8mm.

667) Helicopsis sp., pulmonary complex. 1mm.

668) Helicopsis sp., kidney internal morphology. 1mm.
669) Helicopsis sp., nervous system in dorsal view. 1.4mm.
670) Idem, cerebral commissure cut. 1.4mm.
671) Helicopsis sp., digestive tract. 2.2mm.

STYLOMMATOPHORAN SYSTEMATICS 269

270 MIELIER

FIGS. 672-679: HELICIDAE (HYGROMIINAE, HELICELLINAE)
672) Candidula unifasciata, digestive tract. 3.5mm.

673) Candidula unifasciata, pulmonary complex. 1.8mm.

674) Candidula unifasciata, nervous system in dorsal view. 1.7mm.
675) Idem, cerebral commissure cut. 1.7mm.

676) Monacha cartusiana, nervous system in dorsal view. 2mm.
677) Idem, cerebral commissure cut. 2mm.

678) Monacha cartusiana, digestive tract. 6.25mm.

679) Monacha cartusiana, pulmonary complex. 6.8mm.

STYLOMMATOPHORAN SYSTEMATICS 271

OM TILLIER

FIGS. 680-686: HELICIDAE (HELICELLINAE, CAMPYLEINAE)
680) Helicella itala, digestive tract. 6mm.

681) Leucochroa explanata, pulmonary complex. 5mm.

682) Helicella itala, nervous system in dorsal view. 2mm.

683) Idem, cerebral commissure cut. 2mm.

684) Helicigona lapicida, pulmonary complex. 7.4mm.

685) Helicigona lapicida, nervous system in dorsal view. 2mm.
686) Idem, cerebral commissure cut. 2mm.

273

STYLOMMATOPHORAN SYSTEMATICS

274 TILLIER

FIGS. 687-692: HELICIDAE (CAMPYLEINAE)

687) Elona quimperiana, digestive tract. 7.6mm.

688) Elona quimperiana, nervous system in dorsal view. 2.5mm.
689) Idem, cerebral commissure cut. 2.5mm.

690) Elona quimperiana, pulmonary complex. 10.4mm.

691) Elona quimperiana, kidney internal morphology. 6.2mm.
692) Theba pisana, pulmonary complex. 8.1mm.

275

STYLOMMATOPHORAN SYSTEMATICS

276 TILLIER

FIGS. 693-698: HELICIDAE (CAMPYLEINAE, HELICINAE)
693) Arianta arbustorum, digestive tract. 5.9mm.

694) Arianta arbustorum, nervous system in dorsal view. 2.5mm.
695) Idem, cerebral commissure cut. 2.5mm.

696) Arianta arbustorum, pulmonary complex. 10.8mm.

697) Helix aspersa, nervous system in dorsal view. 4.6mm.

698) Idem, cerebral commissure cut. 4.6mm.

STYLOMMATOPHORAN SYSTEMATICS 277

278 TILLIER

FIGS. 699-704: HELICIDAE (HELICINAE)

699) Cepaea nemoralis, digestive tract. 9.7mm.

700) Cepaea nemoralis, nervous system in dorsal view. 2.5mm.
701) /dem, cerebral commissure cut. 2.5mm.

702) Cepaea nemoralis, pulmonary complex. 19.4mm.

703) Cepaea nemoralis, kidney internal morphology. 6.7mm.
704) Helix aspersa, pulmonary complex. 19.4mm.

STYLOMMATOPHORAN SYSTEMATICS

279

280 TILLIER

APPENDIX A. MATERIAL STUDIED, ABBREVIATIONS USED IN
TEXT-FIGURES

The order of presentation follows Zilch (1959—
1960). Each family and some subfamilies are
numbered in this order; fossil families, and a
few Recent families are not represented.
Each species is abbreviated by three or four
signs, the first of them being the number of
the family or subfamily (Text-figs. 6, 7). The
systematic position adopted in the present
work is indicated in parentheses.

Acronyms are: AMS = Australian Museum,
Sydney; ANSP = Academy of Natural Sci-
ences, Philadelphia; BMNH = British Mu-
seum (Natural History), London; ЕММН =
Field Museum of Natural History, Chicago;
MNHN = Muséum national d'Histoire na-
turelle, Paris; MRAC = Musée Royal de l'Af-
rique centrale, Tervuren; RMNH = Rijksmu-
seum van natuurlijke Historie, Leiden; ZMA =
Zoological Museum, Amsterdam

1-2 ACHATINELLIDAE (+ Zilch's
Tornatellinidae; Orthurethra, Pupilloidea)

Auriculellinae

1Aur: Auriculella auricula (Férussac).
Mt. Tantalus, Manoa Cliff Trail 2, Oahu,
Hawaii. Solem! X11.16.1961. FMNH
166368.

1Aup: Auriculella pulchra (Pease). Ku-
lau Crest, 500-600m, Kalihi Valley, Oahu,
Hawaii. Solem 8 Kondo! ХН.1961. FMNH
155391.

1Ela: Elasmias sp. Fautaua valley, Ta-
hiti. Solem! 1.9.1962. FMNH 155306.

1Lam: Lamellidea cf. pusilla (Gould).
Fautaua valley. Tahiti. Solem! 1.7.1962.
FMNH 155283.

Tornatellininae

1Tor: Tornatellides oblongus oblongus
(Anton). Musée Gauguin, Papeare, Tahiti.
Solem! 1V.18.1974. FMNH 182057.

1Str: Strobilus plicosa (Ohdner). Portua-
zela, Masatierra, Juan Fernandez.
Malkyn! 7.1V.1962. FMNH 167976.

1Tek: Tekoulina pricei Solem. Summit
of Mt. Te Kou, 630m, Rarotonga, Cook
Islands. Price! 9.X11.1965. FMNH 153414.

2 Achatinellinae
2Ach: Achatinella lorata (Férussac).
Slopes of Mt. Tantalus, Oahu, Hawaii.
Solem! FMNH.

3 PARTULIDAE (Orthurethra, Partuloidea)

3Par: Partula caledonica Pfeiffer.
Forest along coast, 20km SE of Port Vila,
Efate, Vanuatu. Price! 2.X.1972. FMNH.

3Eua: Eua expansa (Pease). Mt. Solau,
Upolu, Western Samoa. Solem 4 Price!
8.11.1965. FMNH 152889.

3Sam: Samoana conica (Gould).
Pago-Pago, Fagasa Pass, Tutuila,
Samoa. Price! 13.111.1975. FMNH
181061.

4 AMASTRIDAE (Orthurethra, Chon-
drinoidea)

Leptachatininae
4Lep: Leptachatina balteata
(Pease). Kalalau trail, Pali-Kona, NE
Kauai, Hawaii. Price! 2.1V.1975. FMNH
181168.

Amastrinae
4Ama: Amastra pullata umbrata Bald-
win. Puukolekole. ANSP 108656.

5 COCHLICOPIDAE (Orthurethra, Chon-
drinoidea)

5Coc: Cochlicopa lubrica (Müller). Lou-
isville, Kentucky, USA. Tillier! 12.1X.1981.
MNHN.

6 PYRAMIDULIDAE (Orthurethra, Pupil-
loidea)

6Pyr: Pyramidula rupestris
(Draparnaud). Marigna-sur-Valouse, Jura,
France. Bouchet! 3.X.1976. MNHN.

7 VERTIGINIDAE (Orthurethra, Chon-
drinoidea)

Truncatellininae
7 Bot: Bothriopupa breviconus Pilsbry.
St-Georges de ГОуароск, French Guy-
ana. Geay! 1900. MNHN.

Nesopupinae
7Ste: Sterkia eyriesii (Drouét). Régina,
French Guyana. Tillier! 25.1V.1977.
MNHN.

8 ORCULIDAE (Orthurethra, Chondrinoidea)

8Огс: Огсша dolium dolium (Drapar-
naud). Georges du Moutier, 250m, canton

STYLOMMATOPHORAN SYSTEMATICS 281

de Berne, Switzerland. Gittenberger!
22.V.1971. RMNH 2314.

8Pag: Pagodulina serveri (Zilch). St-
Cézaire sur Siagne, 240m, Alpes-
maritimes, France. Gittenberger! V.1976.
RMNH.

9 CHONDRINIDAE (Orthurethra, Chon-
drinoidea)

Chondrininae
9Sol: Solatopupa similis (Bruguiere).
Sommières, Gard, France. Boulinier!
11.1972. MNHN.

Hypselostomatinae
9Gyl: Gyliotrachela depressispira
Van Benthem Jutting. Bukit Chintamani,
Pahang, Malaysia. Berry! 1.11.1960. ZMA.

10 PUPILLIDAE (Orthurethra, Pupilloidea)

Pupillinae
10 Pup: Pupilla muscorum (L.). Berck,
Pas-de-Calais, France. Chevallier!
21.X.1970. MNHN.

Lauriinae
10Lau: Lauria cylindracea (Da Costa).
Box Hill, Surrey, England. BMNH.

11 VALLONIIDAE (Orthurethra, Pupilloidea)

Valloniinae
11Val: Vallonia albula Sterki. Bethle-
hem, South Dakota, USA. Tillier!
6.1X.1981. MNHN.

Acanthinulinae

11Pty: Ptychopatula dioscoricola
insigne (Pilsbry). Confluence of rivers
Courouaïe and Oyapock, French Guyana.
Tillier! 2.V.1977. MNHN.

11Aca: Acanthinula aculeata (Müller).
Bladafval, Azores. Backhuys! 13.V1.1969.
RMNH 2339.

11Spe: Spermodea lamellata (Jeffreys).
Burnham Beeches, Buckinghamshire, Eng-
land. 20.V.1962. BMNH.

Spelaeodiscinae
11Klem: Klemmia magnicosta
Gittenberger. Спа Gora, Comarro,
Yugoslavia. Gittenberger! V.1975. RMNH
2426.

Strobilopsinae
11Str: Strobilops aenea Pilsbry.1-
Rockville, Parke Co., Indiana, USA. Dy-
bas 8 Kethley! 16.V1.1972. FMNH

170108. 2- Pine Hills, Union Co., Illinois,
USA. Solem & Kethley! 14.X.1975. FMNH
171612

12 PLEURODISCIDAE (not seen)

13 ENIDAE (Orthurethra, Partuloidea)

Chondrulinae
13Cho: Chondrula tridens (Múller).
Meximieux, Ain, France. Chevallier!
3.V.1971. MNHN.

Jamininae
13lmp: /mparietula jousseaumei
(Smith). Jabal Akhdar, Oman. 22.V.1981.
BMNH.

Eninae
13 Ena: Ena montana (Draparnaud).
Cirque du Fer, 995m, Cheval, Haute-
Savoie, France. Chevallier! 7.V11.1969.
MNHN.

Zebrina eburnea (Pfeiffer). Tarsus, “Iran”
(Turkey?). 28.V1.1961. BMNH.

Cerastuinae

13Rach: Rachistia histrio (Pfeiffer). La
Roche, Mare, Loyalty Islands, New Cale-
donia. 7.X.1958. FMNH 109435.

13Cer: Cerastua somaliensis (Smith).
Salan Gudun, Somalia. BMNH.

13Ami: Amimopina macleayi (Brazier,
1876). Cape York Peninsula, Australia.
Wassell! 15.11.1959. FMNH.

13Dra: Draparnaudia michaudi (Mon-
trouzier). Hienghene, New Caledonia.
Price! 13.X.1967. FMNH 159299.

14 SUCCINEIDAE (Dolichonephra,
Achatinoidea)

Catinellinae

Succineinae

14Suc: Succinea putris (L.). Le Hade,
Seine-maritime, France. Chevallier!
29.1X.1967. MNHN.

14Su2: Succinea propinqua Drouëêt. Пе
le Pere, French Guyana. Geay! MNHN.

14Hya: Hyalimax perlucidus (Quoy 8
Gaimard). Forét de Saint-Philippe, La
Réunion. Lantz! MNHN.

140ma: Omalonyx matheroni (Potiez &
Michaud). 17km SE of Kourou, French
Guyana. Tillier! 1978. ММНМ.

15 ATHORACOPHORIDAE (Brachynephra,
Endodontoidea)

15Ane: Aneitea simrothi Grimpe 8 Hoff-

282 TILLIER

mann. Thiem, 10-50m. New Caledonia.
Bouchet! 25.X11.1978. MNHN.

15Ath: Athoracophoridae sp. Vallée d'-
Amoa, 20m, New Caledonia. Tillier, Tillier
& Mordan! 18.1.1981. MNHN.

16 ENDODONTIDAE (Brachynephra,
Endodontoidea)

16Tha: Thaumatodon hystricelloides
(Mousson). Mt. Lanuto, 830m, Upolu,
Western Samoa. Price! 12.1.1965. FMNH.

16Lib: Libera fratercula rarotongensis
Solem. Paratype. E of Avarua, Rarotonga,
Cook Islands. Price! 7.X11.1965. ЕММН.

17 CHAROPIDAE (Brachynephra,
Endodontoidea)

17 Tra: Trachycystis (Xerocystis)
capensis (Pfeiffer). Mainland River, Cape
Province, South Africa. Van Bruggen!
2.1V.1961. RMNH.

17Cha: Charopidae sp. Mt. Toolbrunup,
N of Albany, Western Australia. Solem!
19.1V.1980. FMNH 204670.

17Ann: Annoselix dolosa (lredale). Oak-
ley Dam, Darling Range, Western Austra-
lia. Solem & Price! 10.11.1974. FMNH
182229.

17Lui: Luinodiscus sp. Yallingup Caves,
W of Busselton, Western Australia. Solem
8 Price! 10.11.1974. FMNH 182179.

17Pse: Pseudocharopa lidgbirdi
Etheridge. Mt. Lidgbird, 500m, Lord Howe
Island. Price! IX.1963. ЕММН 127977.

17Mys: Mystivagor mastersi (Brazier).
Mt. Lidgbird, 400m, Lord Howe Island,
Price! IX.1963. FMNH 127963.

17Ran: Ranfurlya constanceae Suter.
Musgrave Peninsula, Auckland, New
Zealand. 15.X1.1943. Auckland Museum.

17Par: Pararhytida dictyodes (Pfeiffer).
Numerous localities, New Caledonia. Cf.
Tillier & Mordan, 1986.

17Pa2: Pararhytida marteli (Dautzen-
berg). Numerous localities, New Cale-
donia. Cf. Tillier & Mordan, 1986.

17Pa3: Pararhytida mouensis (Crosse).
Numerous localities, New Caledonia. Cf.
Tillier & Mordan, 1986.

17And: Andrefrancia sp. Mt. Mou,
450m, New Caledonia. Mordan, Tillier &
Тег! 5.1.1981. MNHN.

17Ste: Stephanoda binneyana (Pfeiffer).
Rio Cisnes, Aysen, Chile. Pena! 11.1961.
FMNH 135428.

18 PUNCTIDAE (Brachynephra,
Endodontoidea)

18Cys: Cystopelta purpurea (Davies).
Mt. Donna Buang, 1350m, Great Dividing
Range, Victoria, Australia. Solem! FMNH
182252.

18Par: Paralaoma lateumbilicata
(Suter). South West Island, Three Kings
Islands, New Zealand. Price! 1.1963.
FMNH.

18Phr: Phrixgnathus erigone (Gray).
Waitakere Range, North Island, New
Zealand. Price! FMNH 135477.

18Lao: Laoma leiomonas (Gray).
Herekina Gorge, $ of Kaitaia, North Is-
land, New Zealand. Price! X1.1962. FMNH
135401.

19 DISCIDAE (Dolichonephra, Zonitoidea)

Discinae

19Dis: Discus rotundatus (Müller). Hen-
daye, Basses-Pyrenees, France. Cheval-
lier! 27.V11.1968. MNHN.

19Di2: Discus patulus (Deshayes).
Bixby State Park, Clayton Co., lowa,
USA. Solem! 18.V1.1974. FMNH 171131.

19Ang: Anguispira alternata (Say).
Grundy Co., Illinois, USA. Burke!
31.11.1965. FMNH 153221.

19An2: Anguispira macneilli Clapp.
Tombigbee River, Leroy, Washington Co.,
Alabama, USA. Hubricht! 3.V11.1960.
FMNH 170671.

Helicodiscinae
19Hel: Helicodiscus parallelus Say.
Raymond Cave, Fulton Co., Arkansas,
USA. Barnett! 10.V.1969. FMNH 173668.

20-21 ARIONIDAE (+ Zilch's Philomy-
cidae; Dolichonephra, Zonitoidea)

Binneyinae
20Hem: Hemphillia camelus Pilsbry &
Vanatta. Rye Patch Creek, 700m, Lochsa
River, Idaho Co., Idaho, USA. FMNH
97995.

Ariolimacinae

20Ari: Ariolimax columbianus (Gould).
Myrtle Grove State Park, Curry Co., Ore-
gon, USA. Dybas! 22.V.1957. FMNH
193553.

20Aph: Aphallarion buttoni Pilsbry &
Vanatta (= Ariolimax columbianus?: Pils-
bry, 1948). Oakland, California, USA. But-
ton! 1.V111.1896. ANSP A2864.

STYLOMMATOPHORAN SYSTEMATICS 283

20Zac: Zacoleus idahoensis Pilsbry.
Kootenai Falls, Montana, USA. Forrester!
17.1X.1959. FMNH 117624.

20Hes: Hesperarion niger (Cooper).
Brown's Ravine, Folsom Lake State Park,
Sacramento Co., California, USA. Smith &
Solem! 9.1V.1960. FMNH 98045.

Anadeninae
20Pro: Prophysaon humile Cockerell.
Rye Patch Creek, 700m, Lochsa River,
Lowell, Idaho Co., Idaho, USA. Walton &
Solem! 24.1V.1960. FMNH 98095.

Arioninae
20Ari: Arion rufus (L.). Confranc,
Huesca, Spain. Alvarez! VII.1971. MNHN.
20Geo: Geomalacus maculosus Allman.
Glengariff, W of Cork, Ireland. Quick!
V111.1931. RMNH 4125.

Oopeltinae

2000p: Oopelta granulosa Collinge.
Nieuwoudtville, Willems River, Cape Prov-
ince, South Africa. Visser 8 Kotzé!
2.1X.1980. ANSP A9361.

20002: Oopelta nigropunctata Mórch.
Capetown, South Africa. Collinge!
23.1V.1909. University Museum, Cam-
bridge.

21 Philomycinae

21Phi: Philomycus carolinianus (Bosc).
Cumberland Falls, McCreary Co., Ken-
tucky, USA. Solem! 16.V1.1954. FMNH
198648.

21Ph2: Incillaria (= Philomycus?) bilin-
eata Benson. Isushima (?). BMNH
1903.5.22.33.

22 THYROPHORELLIDAE (not seen)

23 VITRINIDAE (Dolichonephra, Zonitoidea)

23Phe: Phenacolimax major (Ferussac).

Les Egletons, Corrèze, France. Cheval-
lier! 22.1V.1967. MNHN.

23Ph2: Phenacolimax? ugandensis
(Thiele). Kimakia, 2700m, Kenya. Verd-
court! 14.1V.1963. BMNH.

23Plu: Plutonia atlantica (Morelet). Ro-
salas, Sao Jorge Island, Azores, Back-
huys! RMNH 2339.

24 ZONITIDAE (Dolichonephra, Zonitoidea)

Vitreinae
24Vit: Vitrea crystallina (Müller). Box

Hill, Surrey, England. Peake! V.1960.
BMNH.

Zonitinae

24Mes: Mesomphix inornatus (Say).
Elkton, NNW of Charlottesville, Virginia,
USA. Hopman! 29.V.1982. FMNH.

24Aeg: Aegopinella nitidula (Drapar-
naud). Nogent-sur-Seine, Aube, France.
Cayet! 1980. MNHN.

24Zon: Zonites algirus (L.). Nimes,
Gard, France. Testud! 7.1V.1976. MNHN.

24Оху: Oxychilus draparnaudi (Beck).
Gournay-sur-Marne, France. Chevallier!
3.X1.1974. MNHN.

Daudebardiinae
24Dau: Daudebardia lederi Boettger.
Caucasus. BMNH 86.3.11.114.
24Da2: Daudebardia sp. Hamsikoï,
1900m, 45km. SSW of Trabzon, Turkey.
Vader! 5.V1.1959. RMNH 1583.

Gastrodontinae

24Ven: Ventridens acera (Lewis), Buck
Creek Cove, Sherwood, Franklin Co.,
Tennessee, USA. Goodfriend! 6.1X.1974.
FMNH 171144.

24Zon: Zonitoides arboreus (Say).
Hertfordshire, England. BMNH.

Gastrodonta interna (Say). Hand-

pole Brook, Nantahala Gorge, Macon Co.,
North Carolina, USA. FMNH 171325.

25 PARMACELLIDAE (Dolichonephra,
Zonitoidea)

25Par: Parmacella valenciennesi Webb
8 Berthelot. Loulé, Algarve, Portugal.
Coiffait! 1970. MNHN.

25Pa2: Parmacella deshayesi Moquin-
Tandon. Algeria, Marés! 1876. MNHN.

26 MILACIDAE (Dolichonephra, Zonitoidea)

26Mil: Milax gagates (Draparnaud).
France, laboratory grown. Chevallier!
MNHN.

27 LIMACIDAE (Dolichonephra, Zonitoidea)

27Lim: Limax maximus L. St. Martin-
de-Vesubie, 960m, Alpes-maritimes,
France. Chevallier! 31.V111.1979. MNHN.

28 TRIGONOCHLAMYDIDAE (not seen; =
Limacidae, Zonitoidea, Dolichonephra)

Trigonochlamydinae

284 TILLIER

Parmacellillinae

29 TROCHOMORPHIDAE (Dolichonephra,
Zonitoidea)

29Tro: Trochomorpha sp. Mt. Kinabalu,
3100-3300m, Sabah, Borneo, Malaysia.
Bouchet! XII. 1980. MNHN.

30 EUCONULIDAE (Dolichonephra,
Zonitoidea)

30Con: Conibycus dahli Thiele. RÓ,
Maré, Loyalty Islands, New Caledonia.
Bouchet! 8.V1.1978. MNHN.

30Dis: Discoconulus sp.1. Mt. Kinabalu,

3100-3300m, Sabah, Borneo, Malaysia.
Bouchet! X11.1980. MNHN.

30Di2: Discoconulus sp.2. same lo-
cality.

30Cop: Coneuplecta sp.1. same lo-
cality.

30Со2: Coneuplecta sp.2. same lo-
cality.

30C03: Coneuplecta sp.3. same lo-
cality.

30Vit: Vitrinopsis sp.1. same locality.

30Vi2: Vitrinopsis sp.2. same locality.

31 HELICARIONIDAE (+ Zilch's
Urocyclidae; Dolichonephra, Helicoidea)

Helicarioninae

31Kal: Kalidos oleatus (Ancey).
Marojezy, 1300m, W of Sambava, Mada-
gascar. Blanc! 1.X11.1972. MNHN.

31Mal: Malagarion paenelimax Tillier.
Holotype, Marojezy, 600m, Madagascar.
Blanc! 12.X11.1972. MNHN.

31Hel: Helicarionidae sp. Poindimie,
New Caledonia. Bouchet! 29.X11.1978.
MNHN.

Dyakiinae

31Eve: Everettia corrugata Laidlaw. Mt.

Kinabalu, 3100-3300m, Sabah, Borneo,
Malaysia. Bouchet! X11.1980. MNHN.

Ariophantinae

31Par: Parmarion martensi Simroth.
Cambodia. Harmand! 1875. MNHN.

31Mic: Microparmarion pollonerai Col-
linge & Godwin-Austen. Mt. Kinabalu,
3100 m, Sabah, Borneo, Malaysia.
Bouchet! 13.1.1981. MNHN.

31Hem: Hemiplecta humphreysiana

(Lea). Sengel Kumbang, 1600m, Korinchi,

Sumatra, Malaysia. BMNH.

31Mar: Mariaella dussumieri Gray.
Mahe, India. Dussumier! 1835. MNHN.

Trochozonitinae
31Tro: Trochozonites percarinatus Von
Martens. 10km from Buea, road to Tiko,
Zaire. Van Mol! 2.1X.1968. MRAC
795610.

Gymnarioninae

31 Gym: Gymnarion sowerbyanus (Pfeif-
fer). Assinie, Ivory Coast. Chaper! 1882.
MNHN.

31Aca: Acantharion browni Binder &
Tillier. Paratypes, Tississat Falls, Blue
Nile, Ethiopia. Brown & Posser!
22.V111.1965. MNHN.

Urocyclinae

31Elg: Trochonanina (Montanobloyetia)
simulans meruensis Verdcourt. Mt. Meru,
1700m, Tanzania. Ross & Leech!
28.X.1957. FMNH 106125.

31Mes: Mesafricarion maculifer Pilsbry.
Lodjo, Mongbwalu, Zaire. Lepersonne!
V111.1939. MRAC.

31Gra: Granularion lamottei Van Mol.
Mt. Nimba, Guinea, Lamotte! 1956.
MNHN.

31Est: Estria? sp.A (cf. Van Goethem,
1977: 87). Gopoupleu, Ivory Coast, Con-
damin & Roy! 1959. MNHN.

31Tre: Tresia parva Van Goethem.
Paratypes, Mt. Nimba, Guinea. Lamotte!
1956-1957. MNHN.

31Eli: Elisolimax madagascariensis
(Poirier). Montagne d'Ambre, Diego
Suarez, Madagascar. Salvat & Blanc!
1970. MNHN.

31Ato: Atoxon pallens Simroth. Virunga,
Zaire. Vanschuytbroek! 1954. MNHN.

32 FERUSSACIIDAE (Dolichonephra,
Achatinoidea)

32Cae: Cecilioides acicula (Múller). Los
Blanos, 350m, La Palma, Canary Islands.
V.R. Altena! 27.1V.1947. RMNH 705.

33 SUBULINIDAE (Dolichonephra,
Achatinoidea)

Subulininae
33Sub: Subulina octona (Bruguiere).
Poum, New Caledonia. Bouchet!
16.V.1979. MNHN.
ЗЗВос: Bocageia carpenteri Connolly.
Kitala, 2500m. Mt. Elgon, Zaire. Bouillon!
5.X11.1953. MRAC 6118117.

STYLOMMATOPHORAN SYSTEMATICS 285

33Pse: Pseudoglessula hessei (Boett-
ger). Banryville, Zaire. Wanson! IX.1949.
MRAC 315164.

Rumininae
33Rum: Rumina decollata (L.). Aleria,
Corsica, France. Monniot! IV.1980.
MNHN.

34 ACHATINIDAE (Dolichonephra,
Achatinoidea)

34Ach: Achatina fulica Bowdich. No-
loina, Tamatave, Madagascar. Pointel!
1964. MNHN.

35 MEGASPIRIDAE (not seen;
Brachynephra, Clausilioidea?)

36 FILHOLIIDAE (fossil; Brachynephra,
Clausilioidea?)

37 CLAUSILIIDAE (Brachynephra,
Clausilioidea)

Neniinae
37Nen: Nenia tridens (Schweigger).
Catano, Puerto Rico. Baker! ANSP
A1929.

Alopiinae
37Alb: Albinaria olivieri (Roth). Fileri-
mos, Rhodes, Greece. Riedel! 23.1V.1980.
FMNH.
37lta: Itala ¡tala (von Martens). Som-
mieres, Gard, France. Boulinier! 111.1972.
MNHN.

38 OLEACINIDAE (+ Zilch's Testacellidae;
Dolichonephra, Achatinoidea)

Spiraxinae
38Spi: Spiraxis futilis Baker. Nexaca,
Mexico. ANSP A1908.

Oleacininae

38Var: Varicella biplicata dissimilis Pils-
bry. J20, SW VII.25.23 (?). ANSP A1860.

38Poi: Poiretia dilatata (Philippi). Sicily.
Caron! 1836. MNHN.

38 Str: Streptostyla streptostyla (Pfeif-
fer). Sumidero, Mexico. 25.V1.1926. ANSP
A1855.

38Eug: Euglandina carminensis (More-
let). Starr Creek Valley, Honduras, Sand-
erson! 17.1.1940. BMNH.

38Stb: Strebelia berendti (Pfeiffer). Tex-
olo, Vera Cruz, Mexico. Rhoades! 1899.
ANSP A1924A.

39 Testacellinae (= Oleacinidae,
Achatinoidea, Dolichonephra)

39Tes: Testacella haliotidea Drapar-
naud. Montlhéry, Essonne, France.
MNHN.

40 ACAVIDAE (Brachynephra, Acavoidea)

40Tri: Trigonephrus rosaceus minor
Connolly. Namaqualand, South Africa.
Connolly! V1.1924. University Museum,
Cambridge.

40Dor: Dorcasia alexandri Gray. Vaal-
grass Farm, Khomas Hochland, Namibia.
Lavranos! MNHN.

40Str: Strophocheilus chilensis (Sow-
erby). Manquehua, Coquimbo, Chile.
Pena! X.1961. FMNH 121481.

40Ped: Pedinogyra sp. Mt. Dryander,
16km NE of Proserpina, North Queens-
land, Australia. Price! 18.X1.1971. FMNH.

40Hed: Hedleyella falconeri (Reeve).
Hortons Creek, S. of Nymboida, New
South Wales, Australia. Price! 20.X.1971.
FMNH 173554.

40Pyg: Hedleyella (Pygmipanda) ker-
shawi (Brazier). 11km N of the lakes, Vic-
toria, Australia. Price! 19.X.1968. FMNH
160038.

40Ano: Anoglypta launcestonensis
(Reeve). Myrtle Bank, NE Tasmania.
Dartnell! X1.1972. FMNH 170413.

40Car: Caryodes sp. Ouse River
bridge near the Great Lake, central
Tasmania. Price! 18.X.1968. FMNH
154857.

40Pan: Pandofella whitei (Hedley). Eun-
gella State Park, 800m, 80km W of
Mackay, North Queensland, Australia.
Price! 15.X1.1971.

40Pad: Panda larreyi (Brazier). Near
Dorrigo, 900m, NE New South Wales,
Australia. Price! 23.X.1971.

40Mac: Macrocyclis laxata (Ferussac).
W. of Ancud, Chiloe island, Chile. Engel!
2.11.1976. FMNH 206301.

40Amp: Ampelita petiti Fischer-Piette.
W of the Ibory, Madagascar. Blanc!
15.X11.1970. MNHN.

40Hel: Helicophanta vesicalis Lamarck.
Ste Luce forest, Madagascar. Blanc!
9.X11.1971. MNHN.

40Cla: Clavator eximius Shuttleworth.
Anjavidilava, Madagascar. Blanc!
8.1.1971. MNHN.

40Sty: Stylodon studerianus (Férussac).

286 TILLIER

Vallée de Mai, Mahé, Seychelle Islands.
Hunon! 1979. MNHN.

40Aca: Acavus superbus (Pfeiffer).
Ceylon. Dobell! 11.X1.1909. University
Museum, Cambridge.

41 BULIMULIDAE (+ Zilch's
Odontostomidae and Orthalicidae;
Brachynephra, Clausilioidea)

Bulimulinae

41Bos: Bostryx bermudezae Weyrauch.
Quichero, Valley of Rio Canete, Province
of Lima, Peru. Breure! 6.11.1975. RMNH
2605.

41Coc: Cochlorina aurisleporis (Bru-
guière). Linhares, Espiritu Santo, Brazil.
Elias! X11.1972. RMNH 2629.

41Pla: Placostylus fibratus (Martyn).
Kouto, lle des Pins, New Caledonia.
Bouchet! 15.V111.1978. MNHN.

41Dis: Discoleus azulensis (Döering).
Sierra de Lihuelcalal, La Pampa, Argen-
tina. Hylton Scott! RMNH.

41Sim: Simpulopsis miersi Pfeiffer. Lin-
hares, Espiritu Santo, Brazil. Domingos!
1.1972. RMNH 9036.

Amphibuliminae
41Pel: Pellicula depressa (Rang). La
Soufriere, 1100m, La Guadeloupe. Tillier!
26.111.1983. MNHN.

Odontostominae
41Plg: Plagiodontes daedaleus
(Deshayes). Cerro del Moro, San Luis,
Argentina. Castellanos! 1949. RMNH
3698.
41Ano: Anostoma depressa Lamarck.
Baixa Verde, Brazil. ANSP 109325.

42 ANADROMIDAE (fossil)

43 CERIONIDAE (Brachynephra,
Clausilioidea)

43Cer: Cerion copium Maynard. Cay-
man Brac. Hummelinck! 1.V1.1973.
MNHN.

43Ce2: Cerion casablancae Bartsch.
Andros, Bahamas. Gould & Woodruff!
MNHN.

44 UROCOPTIDAE (Brachynephra,
Clausilioidea)

Eucalodiinae
44Ber: Berendtia taylori (Pfeiffer). Mis-
sion of San Javier, 1300m, Baja California

Sur, Mexico. Christensen! 25.X.1972.
RMNH.

Urocoptinae

44Mac: Macroceramus signatus (Guild-
ing). Anegada, Virgin Islands. FMNH
151675;

44Uro: Urocoptis procera (Adams).
Near Windsor Cave, Windsor, Trelawny,
Jamaica. Goodfriend! 28.V11.1974. FMNH
196515.

45 SYSTROPHIIDAE (Brachynephra,
Endodontoidea)

45Sys: Systrophia (Systrophiella) eudis-
cus Baker. Sierra de Perija, El Roncon, E
of Becerril, Cesar, Colombia. 16.1X.1969.
FMNH 167951.

45Sy2: Systrophia (Wayampia) cayen-
nensis (Pfeiffer). Kaw, French Guyana.
Tillier! 29.1V.1977.

45Tam: Tamayoa decolorata (Drouët).
Saut Sabbat, French Guyana. Tillier!
13.V.1977. MNHN.

46 HAPLOTREMATIDAE (Dolichonephra,
Helicoidea)

46Hap: Haplotrema concavum (Say).
Louisville, Kentucky, USA. Tillier!
12.1X.1981. MNHN.

46Ha2: Haplotrema minimum (Ancey).
Alameda Co., California, USA. 9.1V.1960.
FMNH 98105.

47 RHYTIDIDAE (+ Zilch’s Chlamydeph-
oridae; Brachynephra, Acavoidea)

47Pri: Priodiscus serratus Adams.
Mahé, Seychelle Islands. Benoît 8 Van
Mol! 8.V111.1972. MRAC 798907.

470ua: Ouagapia raynali (Gassies). Col
de Petchékara, 400-450m, New Cale-
donia. Bouchet! 18.X1.1978.

47Dip: Diplomphalus megei
(Lambert). Forêt Nord, New Caledonia.
Mordan, Тег & Tillier! 21.1.1981. MNHN.

47Rhy: Rhytida inaequalis (Pfeiffer).
Monéo, New Caledonia. Bouchet!
15.V.1978. MNHN.

47Nat: Nata cf. vernicosa (Krauss). Pie-
termaritzburg, Natal, South Africa. Kellsall!
9.1V.1963. MNHN.

47Sch: Schizoglossa novoseelandica
Pfeiffer. Egmont, New Zealand. Murdoch!
BMNH 94.10.30.1.

STYLOMMATOPHORAN SYSTEMATICS 287

48 Chlamydephorinae (— Rhytididae,
Acavoidea, Brachynephra)

48Chl: Chlamydephorus gibbonsi (Bin-
ney). Cape Province, South Africa. Neale!
BMNH 79.6.25.1.

49 STREPTAXIDAE (Dolichonephra,
Achatinoidea)

49Ede: Edentulina sp. Esten Province,
Kenya. Charles! 22.V.1971. MRAC
140766.

49Gon: Gonaxis enneoides
(von Martens). Marangu, 1550m,
Kilimandjaro, Tanzania. Basilensky 8
Leloup! 27.11.1956. MRAC 7890055.

49Pty: Ptychotrema sp. Soyo Matadi,
Zaire. Dartevelle! 11.1937. МВАС 211492.

50 POLYGYRIDAE (Dolichonephra,
Helicoidea)

Polygyrinae

50Pol: Polygyra matermontana jalis-
coensis (Pilsbry). Casa Helena, 5000m,
Chapala, Jalisco, Mexico. deVry! X1.1963.
FMNH 145705.

50Mes: Mesodon andrewsae normalis
(Pilsbry). Near Ocoll, 300m, Polk Co.,
Tennessee, USA. Goodfriend! 2.1X.1974.
FMNH.

50Tri: Trilobopsis trachypepla Berry. Ее!
River, Humboldt Co., California, USA.
Solem! 12.1V.1960. FMNH 98097.

Triodopsinae

50Trd: Triodopsis fraudulenta vulgata
Pilsbry. McGee Hill, Pine Hills, N of Wolf
Lake, Union Co., Illinois, USA. Solem 4
Kandhley! 14.X.1975. FMNH 171598.

50All: Allogona ptychophora Brown.
John Day Creek, N of Lucile, Idaho Co.,
Idaho, USA. Solem & Walton! 21.1V.1960.
FMNH.

51 SAGDIDAE (Helicoidea, Dolichonephra)

51Sag: Sagda cf. grandis. Albert Town,
Trelawny, Jamaica. Goodfriend!
12.V111.1974. FMNH 196152.

51Lac: Lacteoluna turbiniformis (Pfeif-
fer). Corner Shop, Clarendon, Jamaica.
Goodfriend! 28.V1.1974. FMNH 196079.

51Pro: Proserpinula opalina (Adams).

Flint River, Hanover, Jamaica.
Goodfriend! 30.V1.1974. FMNH 196104.

52 CORILLIDAE (Brachynephra, Acavoidea)

52Cor: Corilla humberti Brot. Ceylon,
2000m, Collett! ANSP 87433.

52Scu: Sculptaria collaris (Pfeiffer).
Southern Angola. Gofas! 1982. MNHN.

52Ple: Plectopylis sp.1. Damsang Peak
or Rissom Peak, 2000m, Malaysia, Rob-
ert! BMNH.

52PI2: Plectopylis sp.2. Same sample.

52Cra: Craterodiscus pricei
MacMichael. Hypipamee Crater, Atherton
Tableland, Queensland, Australia. Price!
V111.1964. FMNH 135141.

53 CAMAENIDAE (Dolichonephra, Heli-
coidea)

53Ple: Pleurodonte lychnuchus (Müller).
Maison de la Nature, Parc naturel, La
Guadeloupe. Tillier! 1V.1978. MNHN.

53Lab: Labyrinthus leprieurii (Petit).
Trois Sauts, French Guyana. Tillier!
8.V.1978. MNHN.

53Sol: Solaropsis undata (Lightfoot).
Trois Sauts, French Guyana. Tillier!
V.1978.

53Rha: Rhagada sp. Mt. Hart Home-
stead, King Leopold Range, Western Aus-
tralia. Price & Christensen! 13.X11.1976.
FMNH 200050.

53Sin: Sinumelon lennum lredale. Bal-
ladonia Hotel, NE of Esperance, Western
Australia. Solem & Price! 21.11.1974.
FMNH 182477.

53Amp: Amplirhagada burnerensis
(Smith). Winjana Gorge, Napier Range,
Western Australia, Price & Christensen!
21.11.1977. FMNH.

53Amd: Amphidromus cognatus Fulton.
Milikapiti Bay, Melville Island, Northern
Territories, Australia. AMS 126706.

53Plc: Plectotropis goniocheila (Pfeif-
fer). Kao Pra Put, Lop Buri, Thailand.
Brandt! 24.X.1966. FMNH 155087.

54-55 OREOHELICIDAE (Brachynephra,
Acavoidea)

54 Ammonitellinae

54Amm: Ammonitella yatesi Cooper.
San Domingo Creek, Murphys, El Dorado
Co., California, USA. Solem & Smith!
10.1V.1960. FMNH 98073.

54Gly: Glyptostoma gabrielense Pilsbry.
Monrovia Canyon, San Gabriel Mts., Los
Angeles Co., California, USA. Solem,
Miller 4 Gregg! 3.1V.1960. FMNH 98041.

288 TILLIER

55 Oreohelicinae
550re: Oreohelix barbata Pilsbry. Cave
Creek Canyon, Chiricahua Mts., Arizona,
USA. Solem & Walton! 24.1.1960. FMNH
98103.

56 BRADYBAENIDAE (Dolichonephra,
Helicoidea)

Helicostylinae

56Hel: Helicostylus palawanensis (Pfeif-
fer). Puerto Princesa, Palawan, Philip-
pines. Hoogstraal! 19.1V.1947. FMNH
54961.

56Coc: Cochlostyla melanocheila (Pfeif-
fer). W. slope of Mt. Halcon, 400—600m,
Mindoro oriental, Philippines. Bouchet!
X11.1980. MNHN.

Bradybaeninae
56Bra: Bradybaena tourannensis (Ey-
doux & Souleyet). Lan Yu island, S of
Hung Tsu Tsun, SE of Taiwan. Kuntz!
111.1959. FMNH 82155.

57 HELMINTHOGLYPTIDAE (+ Zilch's
Thysanophorinae; Dolichonephra,
Helicoidea)

Xanthonycinae
57Ave: Averellia coactiliata (Deshayes).
El Panden, Guatemala. McCarthy!
6.V111.1976. FMNH 192826.
57Lep: Leptarionta trigonostoma (Pfeif-
fer). Montana del Mico, Isabal, Guate-
mala. Steyermark! 16.111.1940. FMNH.

Helminthoglyptinae

57Mon: Monadenia fidelis (Gray). Sal-
yer, Trinity Co., California, USA.
Talmadge! 13.1V.1960. FMNH.

57Hel: Helminthoglypta sequoicola
(Cooper). Batchelder Ranch, Santa Cruz,
California, USA. Strohbeen! 16.1.1960.
FMNH 111545.

57Epi: Epiphragmopora claromphalos
(Deville & Hupé). Quillabamba, 200m,
Urubamba valley, Peru. Weyrauch! 1949.
FMNH 30697.

Sonorellinae
57Son: Sonorella walkeri (Pilsbry 8 Fer-
riss). Gardner Canyon, Santa Rita Mts.,
Arizona, USA. Walton & Solem!
18.11.1960. FMNH.

Cepolinae
57Cep: Cepolis varians (Menke).
French Leave, Eleuthera, Bahamas.
Burke! 7.V11.1965. FMNH 147768.
57Ce2: Cepolis maynardi Pilsbry. South

Bimini, Bahamas. Haas! 1.VIII.1956.
FMNH 57104.

57Psa: Psadara nubeculata
(Deshayes). Mt. St Marcel, 500m, French
Guyana. Tillier! V.1978. MNHN.

57Ps2: Psadara marmatensis (Pfeiffer).
Trois Sauts, French Guyana. Lescure!
1.1V.1976. MNHN.

Thysanophorinae

50Thy: Thysanophora horni (Gabb).
Casa Helena, Chapala, Jalisco, Mexico.
deVry! X1.1963. FMNH 145669.

58 HELICIDAE (Dolichonephra, Helicoidea)
Sphincterochilinae

58Sph: Sphincterochila zonata
(Bourguignat). Near Massada, W of
Dead Sea, Israel. Holthuis! 21.1V.1973.
RMNH.

58 Hal: Halolimnohelix sericata lejeunei
Verdcourt. Paratype. Butembo, Kivu,
Zaire. Lejeune! 1968. MRAC 794977.

Helicellinae

58Cer: Cernuella virgata (Da Costa).
Claye- Souillis, Seine-et-Marne, France.
Chevallier! 21.1X.1967. MNHN.

58Leu: Leucochroa explanata (Múller).
La Grande Motte, Hérault, France. Tes-
tud! IV.1976. MNHN.

58Hec: Helicella itala (L.). S of
Lincheux, Somme, France. Chevallier!
22.X.1970. MNHN.

58Tro: Trochoidea elegans (Gmelin) (=
Hygromiinae?). Without locality, MNHN.

58Tr2: Trochoidea geyeri (Soos) (=
Hygromiinae?). Langeray, Geneve, Swit-
zerland. Falkner! 1.111.1968. MNHN.

58Can: Candidula unifasciata (Poiret)
(= Hygromiinae?). Between Ste-Affrique
and St-Servin, Aveyron, France. Cheval-
lier! 28.1V.1967. MNHN.

58Ca2: Candidula intersecta (Poiret) (=
Hygromiinae?). Les Traverses, NE of Em-
brun, Hautes -Alpes, France. Chevallier!
25.V11.1969. MNHN.

58Hep: Helicopsis striata (Müller) (=
Hygromiinae). Podersdorf, Burgenland,
Austria. Clerx! VI11.1973. RMNH.

58He2: Helicopsis n.sp. (= Hygromii-
nae). Spain. RMNH.

58He3: Helicopsis apicina (Lamarck)
(= Hygromiinae). Les Gondes, Cap
Croisandte, Bouches-du-Rhöne, France.
Gittenberger! RMNH.

58Col: Cochlicella acuta (Müller) (=

STYLOMMATOPHORAN SYSTEMATICS 289

Helicodontinae?). Deauville, Calvados,
France, Chevallier! 1.X.1967. MNHN.

58Mon: Monacha cartusiana (Müller)
(— Hygromiinae). Forêt de Sénart, Seine-
et-Oise, France. Testud! 25.1X.1977.
MNHN.

Hygromiinae
58Hyg: Hygromia limbata (Draparnaud).
Chambre d'Amour, Pyrénées-Atlantiques,
France. Chevallier & Deguirmenci!

Ampuero, 50m, Santander, Spain.
Bouchet! 21.V111.1970. MNHN.

98Art: Arianta arbustorum (L.). Val du
Giffre, 700m, Samoëns, Haute- Savoie,
France. Chevallier! 5.V11.1969. MNHN.

58Heg: Helicigona lapicida (L.). Val du
Herisson, Jura, France. Chevallier!
1.V11.1969. MNHN.

Helicinae

6.1V.1966. MNHN.

58Tri: Trichia hispida (L.). Jardin des
Plantes, Paris, France. Chevallier!
21.V.1969. MNHN.

Helicodontinae
58Hed: Helicodonta obvoluta (Müller).
Georges de la Fou, Pyren&es-orientales,
France. Tillier! 18.V11.1980. MNHN.

Campyleinae
58Elo: Elona quimperiana (Ferussac).

58The: Theba pisana (Müller). Giber-
ville, Calvados, France. Delaunay!
14.V111.1969. MNHN.

58Cea: Cepaea nemoralis (L.). Toruy,
Seine-Saint-Denis, France. Kassem!
IX.1982. MNHN.

58Hel: Helix aspersa Müller.
Sommieres, Gard, France. Boulinier!
11.1972. MNHN.

290 TMEBIER

APPENDIX B. DATA (GENERAL MORPHOLOGY)

H, shell height; D, shell diameter; H/D, ratio; WH, whorl number of shell; TA, size rank; LV, whorl number
of visceral mass; LP, lung length in whorls; ST, length in whorls of stomach above summit of lung; SS,
length in whorls of visceral mass above summit of stomach; LS, total length in whorls of gastric crop and
gastric pouch. Semislugs = 6000; slugs = 8000; missing data = 9999.

H D H/D МН TA LV ER ST SS LS
Auriculella auricula 4 2.4 1.66 4.5 22 3.75 1 1.25 1.5 155)
Auriculella pulchra 9999 9999 9999 9999 99999999 9999 9999 9999 9999
Elasmias sp. 4.2 3 1.42 3:5 27 3:25 8 We 1.25 ee
Lamellidea sp. 3.4 1.9 1.83 6 20 4.75 1.1 1.6 2.05 1
Tornatellides oblongus 2.9 1.8 1.62 5 12 4.25 1525 1 2 les
Strobilus plicosa 4.8 2.6 1.86 7 28 4.75 3 .8 2.65 .8
Tekoulina pricei 7.9 2.8 2279 ЛИ 33 6.75 3.1 1.6 2.05 1.6
Achatinella lorata 15 9.5 1.58 5 120 4 Ue. 1.25 1.65 143
Partula caledonica 282.5 1.75 5 168 3.75 1.3 .6 1.85 .6
Eua expansa 18 14 1.28 3 166 2.75 aif 5 1255 22
Samoana conica 20 13.2 1185229999 167 3.25 1 .8 1.45 29
Leptachatina balteata 10.4 4.9 2.13 6.5 75 4.75 al 1.25 2.4 leat
Amastra pullata 19.5 129 1:51 5.25 150 4.25 1.2 1 2.05 lc
Cochlicopa lubrica il 2.4 2.93 6 35 4.5 15 1.25 175 .9
Pyramidula rupestris 1.6 2.4 .66 4 9 3.75 1.8 olf 2.25 .6
Bothriopupa breviconus 1.6 1.3 1225 4.3 2 4 9 9 2.2 6
Sterkia eyriesii 1.8 ileal 1.62 4.25 3 4 EL Ue 2.8 8
Orcula dolium 7.3 3:5 2.09 8.5 40 57 2.5 1 2.2 .6
Pagodulina serveri 3:2 2 1.6 7.25 (19 6 1.5 ÉS 3:2 6
Solatopupa similis 10.6 3.6 2.9 8.4 53 6.5 1ES 1.4 3.6 vA
Gyliotrachela depressispira 17 32 -53 3 13 3 ath 5 1.8 5
Pupilla muscorum 4 2 2 6:25) 18 5:25 2.1 19 2:25 .66
Lauria cylindracea 4.4 2.2 2 6:5 25 5.4 2.25 li 2.15 .6
Vallonia albula 1.3 2.7 .48 3.3 5 2 .8 .6 .6 .5
Ptychopatula dioscoricola 2 1.92 1.04 3 8 2:5 5 7% 1.05 5
Acanthinula aculeata 3.2 2.3 1:39 4252210 4.1 afl 1.4 2 05]
Spermodea lamellata STA 2:2 Zul, 5 7 4.3 1 1 2.3 119
Klemmia magnicosta 1.8 3:2 .56 5 24 3.75 15 1 2 .8
Strobilops aenea 1:5 2.4 .62 5 6 375 1 1.4 1.6 ol
Chondrula tridens 10 4 2.5 6.8 55 5.8 1.4 1 3.4 1
Imparietula jousseaumei 10 4.6 217 6.3 67 6.2 1.75 1:3 3.15 1
Ena montana 14.8 8.5 1.74 7 83 6 123 Uist 3.6 1.1
Zebrina eburnea 25.1 7.8 3.22 9 89 TES 1.9 1.3 4.3 1225
Rachistia histrio 13 8.3 1.57 5.9% 112 3.75 1 BAS) 2 1
Cerastua somaliensis 13.7 8.3 1.65 6.2 113 4.5 15 1625 2.5 1.5
Amimopima macleayi 12.5 8 1:56. 5:5 82 9999’ 9999, 99991 899998889999
Draparnaudia michaudi 10 8.5 1.18 6.5 929999 9999 9999 9999 29
Succinea putris 17.4 9.6 1.8 S 131 215 ES .5 197 .6
Succinea propinqua 9 6.7 1:35 2 78 15 .25 .25 1 25
Hyalimax perlucidus 8000 8000 8000 8000 128 8000 8000 8000 8000 8000
Omalonyx matheroni 8000 8000 8000 8000 152 8000 8000 8000 8000 8000
Aneitea simrothi 8000 8000 8000 8000 130 8000 8000 8000 8000 8000
Athoracophoridae sp. 8000 8000 8000 8000 127 8000 8000 8000 8000 8000
Thaumatodon hystricelloides 2.7 4 .67 5:5 26 3.65 AS .8 2.1 .6
Libera fratercula 4.5 5:5 .82 17% 49 5 .8 162 3 We
Trachycystis capensis 4.7 2.9 1.62 5.25) 2.36 4.3 6 1 ZU, 1
Charopidae sp. 9999 3.9 9999 6 9999 3:25 19 dl 1.6 .9
Annoselix dolosa 3.3 57 .59 4.3 41 3:5 LD 8 1.95 st
Luinodiscus sp. 9999 9999 9999 9999 99999999 9999 9999 9999 if
Pseudocharopa lidgbirdi 9999 9999 9999 9999 99999999 9999 9999 9999 9999
Mystivagor mastersi 6.9 4.7 1.47 3 519999 9999 9999 9999 6
Ranfurlya constanceae 8000 8000 8000 8000 80 1 5 3 230) 9999
Pararhytida dictyodes 16 21 .76 6.2 192 3:7 19 We 1.95 1
Pararhytida marteli 9.8. 17.9 .55 5.9 141 5:5 .9 (Ej 3.5 1

Pararhytida mouensis
Andrefrancia sp.
Stephanoda binneyana
Cystopelta purpurea
Paralaoma lateumbilicata
Phrixgnathus erigone
Laoma leiomonas
Discus rotundatus
Discus patulus
Anguispira alternata
Helicodiscus parallelus
Hemphillia camelus
Ariolimax columbianus
Aphallarion buttoni
Zacoleus idahoensis
Hesperarion niger
Prophysaon humile
Arion rufus

Geomalacus maculosus
Oopelta granulosa
Oopelta nigropunctata
Philomycus carolinianus
Philomycus bilineatus
Phenacolimax major
Phenacolimax? ugandensis
Plutonia atlantica

Vitrea crystallina
Mesomphix inornatus
Aegopinella nitidula
Zonites algirus
Oxychilus draparnaudi
Daudebardia lederi
Daudebardia sp.
Ventridens acera
Zonitoides arboreus
Gastrodonta interna
Parmacella valenciennesi
Parmacella deshayesi
Milax gagates

Limax maximus
Trochomorpha sp.
Conibycus cf. dahli
Discoconulus sp.A
Discoconulus sp.C
Coneuplecta sp.B
Coneuplecta sp.D
Coneuplecta sp.E
Vitrinopsis sp.1
Vitrinopsis sp.2

Kalidos oleatus
Malagarion paenelimax
Helicarionidae sp.
Everettia corrugata
Parmarion martensi
Microparmarion pollonerai
Hemiplecta humphreysiana
Mariaella dussumieri
Trochozonites percarinatus
Gymnarion sowerbyanus
Acantharion browni
Mesafricarion maculifer

STYLOMMATOPHORAN SYSTEMATICS 291
H D H/D WH WAY У EP ST SS LS
1262115 :59 6.17 175 4.5 E) 1.1 2.5 1
3.6 7.4 .49 45 45 215 25 1 1625 1
65 14.5 .45 5 109 3.7 .6 .8 2.3 .6
6000 6000 6000 6000 126 6000 6000 6000 6000 6000
.8 1.22 .66 3.1 1 3 5 ath 1.7 ÉS
3.5 4.1 .86 5.5 382 3.7 75 1 1895 .8
2.9 2.1 1.38 TE NE 4 IS 1.4 2 1:25
DT: 5:5 .49 5.25 38 3.6 .6 :5 2.5 15
3.3 8 41 5.25 43 9999 9999 9999 9999 15
13 23.9 54 5 149 3.75 US SUS 1.75 ¿9
1.8 4 .45 A 21 3.2 6 6 2 :5
6000 6000 6000 6000 114 6000 6000 6000 6000 6000
8000 8000 8000 8000 226 8000 8000 8000 8000 8000
8000 8000 8000 8000 225 8000 8000 8000 8000 8000
8000 8000 8000 8000 95 8000 8000 8000 8000 8000
8000 8000 8000 8000 129 8000 8000 8000 8000 8000
8000 8000 8000 8000 125 8000 8000 8000 8000 8000
8000 8000 8000 8000 207 8000 8000 8000 8000 8000
8000 8000 8000 8000 172 8000 8000 8000 8000 8000
8000 8000 8000 8000 196 8000 8000 8000 8000 8000
8000 8000 8000 8000 198 8000 8000 8000 8000 8000
8000 8000 8000 8000 132 8000 8000 8000 8000 8000
8000 8000 8000 8000 182 8000 8000 8000 8000 8000
6000 6000 6000 6000 64 2.5 .2 25 1.8 9999
6000 6000 6000 6000 97 ES 115 .35 1 9999
8000 8000 8000 8000 144 8000 8000 8000 8000 8000
1.4 3.1 .45 42 Al 225 .5 9 US .6
6 11.6 .52 4.25 84 4.5 ÉS) T 3.3 MT.
3.5 Teil .49 3:0 50 3 .4 .6 2 .6
22.4 38 109 6:5: 21 3.6 .6 1 2 1.3
7 15 .47 6 106 2.25 3 8 1815 ohh
8000 8000 8000 8000 63 8000 8000 8000 8000 8000
8000 8000 8000 8000 96 8000 8000 8000 8000 8000
12 17.2 af Y 151 5 TS 1.5 205 1.4
1.8 3.4 152 3.7 16 3.2 75 .6 1.85 .6
4.7 Tel .66 #7 65 6 lei 1.3 3.6 1.3
8000 8000 8000 8000 201 8000 8000 8000 8000 8000
8000 8000 8000 8000 205 8000 8000 8000 8000 8000
8000 8000 8000 8000 169 8000 8000 8000 8000 8000
8000 8000 8000 8000 224 8000 8000 8000 8000 8000
6.2 9.9 .62 58 19 5.3 .6 125 3.2 1.4
6.5 TT .84 4.1 56 3 ES £) ee 6
5.4 6.8 79 So 58 2.3 ‚25 .8 1625 .8
3.3 4 .825 5:2. “Sil 3.5 LS 15 2 8
3.8 4.5 84 5.1 39 2.2 if 5 .6 :5
5.9 6.3 .94 9.922159 4.7 29 1 2.8 .8
2.9 3.4 .85 3:9) «23 4.6 5 1.1 2.75 x)
6000 6000 6000 6000 60 27 3 8 12 15
6000 6000 6000 6000 61 2.25 .25 .8 122 "5
23:5: 36 .65 4.5 221 J 5 1 2.5 ПЕ
6000 6000 6000 6000 100 125 25 tf 589999
4 6.6 .61 3.25 86 3 .35 .65 2 .6
10 15.3 .65 4.4 123 4.25 .6 9 2.75 1.2
6000 6000 6000 6000 111 6000 6000 6000 6000 6000
6000 6000 6000 6000 122 6000 6000 6000 6000 6000
39 68 97 4.9 233 3.25 LAS) A) 2 8
8000 8000 8000 8000 176 8000 8000 8000 8000 8000
Ua) Wild .675 7 107 3.25 1.1 1.1 1.05 led
6000 6000 6000 6000 206 2.5 .25 5 12519999
6000 6000 6000 6000 11777 3 .25 aif 21519999
6000 6000 6000 6000 115 2 lo 3 1.55,9999

292 TILLIER

H D H/D WH TA LV ЕР ST SS LS

Trochonanina simulans 8.5 14 .6 4.6 124 3.2 .6 .8 1.8 .8
Granularion lamottei 6000 6000 6000 6000 170 2 115 25 1.6 9999
Estria sp.A 8000 8000 8000 8000 200 8000 8000 8000 8000 8000
Tresia parva 8000 8000 8000 8000 99 8000 8000 8000 8000 8000
Elisolimax

madagascariensis 8000 8000 8000 8000 208 8000 8000 8000 8000 8000
Atoxon pallens 8000 8000 8000 8000 171 8000 8000 8000 8000 8000
Cecilioides acicula 4.6 1.3 3.54 5.3 14 5 29 1 Sail 1
Subulina octona 21 4.3 4.88 9.5 54 5:25 1:2 1.8 2.25 157
Bocageia carpenteri 35 15:5 2.26 5.6 193 4.75 1 1 2315 :9
Pseudoglessula hessel 11 5.6 1.96 6.2 74 5 1 1.25 3.75 1425
Rumina decollata 25 10 2.5 45 148 5:25 2 1.25 3 1
Achatina fulica 71 41 1:73 7 228 4.5 1.2 .8 2:5 Sí
Мета tridens 24.3 5 4.86 7 93 5.25 215 1:5 1625 1.3
Albinaria olivieri 18.2 4.2 dS6 12 76 es 2:5 1:5 3.5 1625
Itala ¡tala 13.8 2.6 5.3 10 52 4.75 15 ical 3.15 1
Spiraxis futilis 3 9 3:39 6.3 4 4.6 1.5 1.4 17 dt
Varicella biplicata 4.6 ie 3.54 5:3 15 5:5 75 3 4.45 3
Poiretia dilatata 32 Ми 2.73 6 174 4 4 3 3.3 3
Streptostyla streptostyla 20 9 2.2 9999 1549999 9999 9999 9999 9999
Euglandina carminensis 48 19 2.53 9999 217 4.5 4 1 3.1 1
Strebelia berendti 8000 8000 8000 8000 104 8000 8000 8000 8000 8000
Testacella europaea 8000 8000 8000 8000 202 8000 8000 8000 8000 8000
Trigonephrus rosaceus 24.6 26.8 92 4.1 214 2.19 T4 2 1.85 .5
Dorcasia alexandri 15.9 217.5 .58 4.1 194 3 Se) 5 1275 5
Strophocheilus chilensis 19.2 14.2 1:35 4 190 3 ah 3 2 5
Strophocheilus oblongus 9999 9999999929399 236 9999 9999 9999 9999 9999
Pedinogyra sp. 24 60.5 4 Sal 229 2.4 7 6 1.1 .6
Hedleyella falconeri 57 60 2195 3.9.. 235 3.75 .6 .75 2.4 SUS
Pygmipanda kershawi 22 45 .49 5 218 4.66 all 8 3.16 75
Anoglypta launcestonensis 19.5 33.5 .58 9999 203 9999 9999 9999 9999 9999
Caryodes sp. 24 14.6 1.64 4.5 160 3:5 7 6 2.2 6
Pandofella whitei 6000 6000 6000 6000 183 1:5 2 3 1 9999
Macrocyclis laxata 35 73 .48 4.3 232 3 A ES 2 13
Panda larreyi ИИ 35 .51 4 213 94 15 .6 2 .6
Ampelita petiti 154 26.9 .56 5.25 180 3.3 7 .3 2.3 .3
Helicophanta vesicalis 40 74 .54 3.5 234 3 7. 3 2 4
Clavator eximius 83 32 2:59 6 223 3:75 1 6 2.15 6
Stylodon studerianus 39.8 60.5 .66 5 231 4.6 1:5 ‘5 2.6 .8
Acavus superbus 47 57 .82 4.2 227 3 T4 .6 197 19
Bostryx bermudezae WS 4.7 3.64 6.75 73 5.5 Wee 1.2 3.1 1:2
Cochlorina aurisleporis 33 21.5 1.53: 5 179 3.5 1.1 .8 1.6 all
Placostylus fibratus 84 38.5 2.18 6.7 230 3.15 65) 9 165 aff
Discoleus azulensis 21.8 ala 1.98 5 146 3.5 >19 it 2.05 af
Plagiodontes daedaleus 25.6 13 1:97 6.75 145 4.5 1.3 a 2.5 6
Anostoma depressa 15 31 .48 43 181 9999 999 9999 9999 9999
Simpulopsis miersi 18.3 18 1.02 3.25 153 2.5 .6 .6 1.3 .65
Pellicula depressa 8000 8000 8000 8000 195 8000 8000 8000 8000 8000
Cerion copium 19.1 10.9 1.74 9.25 133 6 3 lee 2.8 9
Cerion casablancae 29 12.7 2.28 10.75 134 6.75 2.3 1.25 3.2 162
Вегепайа taylori 47 14 3.36 10.5 165 9999 2 1.2 9999 151
Macroceramus signatus 14.7 6.9 2.13 10,22 105 5.3 1.3 1.7 2.3 1.6
Urocoptis procera 25 8 3.12 8 138 6.75 2.25 2.25 2.25 2
Systrophia eudiscus 4.2 10 .42 5:8 101 4.7 1 1 2.7 9
Systrophia cayennensis 4.7 8.6 .55 4:29 A 0 15 .65 9999 .6
Tamayoa decolorata 3.25 6.5 :5 4.2 48 0 9 .6 9999 .6
Haplotrema сопсауит 9 21 .43 4.3 156 239 :35 .65 1:75 5
Haplotrema minimum 9 19.5 .46 4.3 155 9999 9999 9999 9999 9999
Priodiscus serratus 3.3 6.5 ‘Di 5.25 46 2 .4 .4 142 3
Ouagapia raynali 9999 9999 9999 9999 99999999 9999 9999 9999 9999

Diplomphalus megei 5:1 7.8 .65 3.8 72 3:25 1 3 1.95 3

STYLOMMATOPHORAN SYSTEMATICS 293

H D H/D WH TA LV EP ST SS LS

Rhytida inaequalis 12 24 5 4.25 163 2 5 35 1515 8)
Nata cf. vernicosa 6.2 UU .56 4.25 87 4.2 15 Sil 3.6 6
Schizoglossa

novoseelandica 8000 8000 8000 8000 140 8000 8000 8000 8000 8000
Chlamydephorus gibbonsi 8000 8000 8000 8000 212 8000 8000 8000 8000 8000
Edentulina sp. 33 15 2.2 6.15 188 5 1.25 25 3.5 .4
Gonaxis enneoides 9999 9999 9999 9999 9999 9999 9999 9999 9999 9999
Ptychotrema sp. 14.7 The 2.04 7 110 6.8 2 8 3.9 .6
Polygyra matermontana 4.6 8.5 .54 4.6 68 32 ay AS) 6 1.85 19
Mesodon andrewsae 25.5, 30.5 .84 57 210 5.5 1.5 BV As) 3.25 12
Trilobopsis trachypepla 4.8 7.9 .61 4.75 69 3:5 1.1 .9 1.5 5
Triodopsis fraudulenta Zeil 14.3 5 5.1 119 3:75 .75 7 2.3 .6
Allogona ptychophora 14 19.3 .72 ISE, 5 9 5 3.6 5
Thysanophora horni 2.8 4.2 .66 3:7 29 22 6 55 1.05 15
Lacteoluna turbiniformis 5.4 7:5 we 5.4 66 4.2 6 1 2.6 1
Proserpinula opalina 3 5:7 .53 AO 3 3.75 .6 .9 2.25 9
Sagda cf. grandis 22.1 23.7 .93 8.8 189 6.5 2:5 2 2 2.2
Corilla humberti 19.8 24.5 .81 5 187 1.75 8 .65 -3 .65
Sculptaria collaris 2 6.4 roi 4.3 34 2.6 9 6 Val .6
Plectopylis sp.1 4.2 10.4 4 6.1 dl 8 1 6 1.4 a5
Plectopylis sp.2 5.6 13.9 4 7.25 94 5 8 9 3.3 .8
Craterodiscus pricei Pal 5 .38 5:5 30 222 .9 6 4 .6
Pleurodonte lychnuchus 16 26.8 .6 4.5 191 3.75 la 8 1.85 "7
Labyrinthus leprieurii dot 22:5 .49 4.75 158 4 Tal 1 1.9 1
Solaropsis undata 29 48 .6 55 222 4 2 1 2 Weil
Psadara nubeculata 8 13 .62 3.8 103 9999 9999 9999 9999 9999
Psadara marmatensis 6.4 10.4 .62 3.5 98 2.75 6 5 1.65 15
Rhagada sp. 10 15.6 .64 4.8 135 3:25 55 8 17 5
Sinumelon lennum 21.5 24 .9 4.75 197 32 ый 6 1575 15
Amplir hagada burnerensis 13 18.7 .69 4.3 157 4 1 .66 2.34 .6
Amphidromus cognatus 22 16 1.37 9999 173 4 5 .6 2.9 .65
Plectotropis goniocheila 5.6 9.7 .58 5 90 3.1 .8 8 1:5 .8
Ammonitella yatesi 3.5 725 47 6.3 44 4.25 1 9 2.4 6
Glyptostoma gabrielense 11.67 23.6 .49 5 185 3.25 715 75 1.75 .45
Oreohelix barbata ih 13.6 51 4.25 102 2:5 .6 6 1,8 4
Helicostyla palawanensis 29 48 6 5:9 062419 4 1 Ye 2.25 .8
Cochlostyla melanocheila 31:8 43.7 7:3 4.1 220 9999 9999 9999 9999 9999
Bradybaena tourannensis 11.6 13 .89 49 137 5 ET AS) .65 3.6 .4
Averellia coactiliata 4.1 10.6 .39 4 91 3:25 4 TAS) 21 .9
Leptarionta trigonostoma 19 24 .79 4.2 186 3 .45 .6 1:95 .8
Monadenia fidelis 18.9 31.3 .6 6.4 204 3.75 .95 255 2.25 .75
Helminthoglypta sequoicola 19:87 29:9 .66 6.224.215 51 1.2 25 3.4 .45
Epiphragmopora

claromphalus 14.5 28 52 41 199 3:2 7 ‘5 2 5
Sonorella walkeri 14 23 61 43 162 3.3 nS 5 2.05 .6
Cepolis varians 13.6 15.2 .89 4.75 139 4.2 12 1 2 We
Cepolis maynardi 7.5 127 .59 42 117 23 45 fh 1:15 le
Sphincterochila zonata 17.8 22.4 79 4.75 164 3:25 it 55 2 5
Halolimnohelix sericata 10.4 15.2 .68 542 3.5 :55 off 2:25 sth
Candidula unifasciata 5.2 7 74 4.1 62 2.9 5 6 1.8 6
Candidula intersecta 3.7 6 .62 3.8 42 3:25 75 7, 1275 75
Cernuella virgata 10.5 14.4 18 5:30 147 3:5 .66 st 2:2 “ff
Leucochroa explanata 5.2 12.4 .42 4.5 108 21 .6 4 1.75 9999
Helicella itala NT 13.8 152 Selle) 3.65 255 .8 2.3 .6
Trochoidea elegans 6 9.5 .63 6.3 88 4.25 1 N; 2.55 9999
Trochoidea geyeri 4 6.3 .63 4.3 47 27 Té .6 1.4 9999
Helicopsis striata Sf 1.2 .79 4.5 70 3.1 .9 .6 1.6 .45
Helicopsis sp. 9999 9999 9999 9999 9999 2 5 155 .95 3
Helicopsis apicina 4.6 Y .66 4:25: 57 3 15 .6 1.9 135
Cochlicella acuta 14.2 5.3 2.68 9.1 81 Y 1525 1le7/ 4.05 1.25
Monacha cartusiana 8.9 14.2 .63 6 143 5.4 U) 1 3.65 7

294

Hygromia limbata
Trichia hispida
Helicodonta obvoluta
Elona quimperiana
Arianta arbustorum
Helicigona lapicida
Theba pisana
Cepaea nemoralis
Helix aspersa

TILLIER

H D HD WH TA
11 16 69 6 136

58 86 67 51 85

56 11.6 48 6 116
13 262 5 45 209
‘em 220 85 5.25 178

74 15.5 48 as 12
141 19.1 74 5 161
18.4 24 77 4.75 184
327 376 87 4 216

APPENDIX C. UPPER LIMITS OF CLASSES USED IN FACTOR ANALYSES
The size of each class is in parentheses.

TEXT-FIGURES 5, 6, 7 (PROPORTIONS OF VISCERAL MASS, APPENDIX B)

HD1 HD2 HD3 HD4 HD5 TA1 TA2
0.52(32) .66(37) .89(28) 1.96(31) 5.3(31) 35(32) 77(32)
WH1 WH2 WH3 WH4 WH5 LP1 LP2
4.25(39) 4.9(25) 5.3(32) 6.3(32) 12(31) .6(39) .75(40)
LS1 LS2 LS3 LS4 LS5 $$1 SS2
0.5(38) .6(28) .8(34) 1.1(34) 3.8(25) 1.6(32) 2(43)
ST ST2 ST3 ST4 ST5

0.6(51) 7(17) .9(18) 1.2(34) 2(26)

TEXT-FIGURES 10, 11, 12 (PULMONARY COMPLEX, APPENDIX B AND FIGS. 1-704)
HD1 HD2 HD3 HD4 HD5 TA1 TA2
0.55(37) .75(40) 1(23) 2(29) 5.3(27) 46(39) 107(39)
LR1 LR2 LR3 LR4 UR1 UR2 UR3
0.33(43) .45(37) .63(39) .99(39) 0(43) 2(17) 4(27)
LI2 LI3 LI4 RR1 RR2 RR3 CA‘
2(26) 3(33) 4(9) 1(78) 2(52) 3(26) 0(107)
RP2 RP3 RP4

0.23(47) .28(32) .5(36)

TEXT-FIGURES 13, 14, 15 (DIGESTIVE TRACT, APPENDIX B AND FIGS. 1-704)
HD1 HD2 HD3 HD4 HD5 TA1 TA2
0.52(32) .66(37) .89(28) 1.96(31) 5.3(31) 35(32) 77(32)
BM1 BM2 OCA OC2 OC3 OC4 SCi
1(149) 2(10) 0(103) 1(31) 2(5) 3(22) 1(14)
IL1 IL2 IL3 IL4 PS1 PS2 PS3
1(35) 2(59) 3(59) 4(6) 1(35) 2(107) 3(17)
TEXT-FIGURES 16, 17, 18 (NERVOUS SYSTEM, APPENDIX B AND FIGS. 1-704)
HD1 HD2 HD3 HD4 HD5 TA! TA2
0.6(45) 1.5(46) 8(45) semislugs slugs 55(39) 115(39)
CCI CC2 CC3 CPD1 CPD2 CPD3 CPR1
1(12) 2(25) 3(119) 1(38) 2(81) 3(37) .9(20)
PLD1 PLD2 PLD3 PLG1 PLG2 PLG3 VG1
2(39) 3(5) 1(137) 2(15) 3(4) 1(51) 2(71)
PAD1 РАО? PAG1 PAG2 PAG3 PAG4 FG1
1(53) 2(103) 1(26) 2(13) 3(31) 4(86) 1(52)
FG4

4(16)

TA3
134(32)
LP3
.9(18)
$53
2.25(21)

ТАЗ
164(39)
UR4
5(69)
CA2
1(49)

TA3
134(32)
SC2
2(80)

TA3
169(39)
CPR2
1.2(86)
vG2
3(34)
FG2
2(53)

TA4
184(32)
LP4
1.25(33)
SS4
2.8(32)

TA4
235(39)
LI
1(88)
RP1
.18(41)

TA4
184(32)
SC3
3(43)

TA4
235(39)
CPR3
2.5(50)
VG3
1(112)
FG3
3(35)

ТАБ
235(31)
LP5
3.1(29)
SS5
4.45(31)

TA5
235(31)
SC4
4(22)

STYLOMMATOPHORAN SYSTEMATICS 295
APPENDIX D. LIMITS OF CLASSES USED
(CHARACTER STATES)

(TEXT-FIGURES 19-29, APPENDICES E,F,G)
The character states retained for phylogenetic reconstructions are denoted from 1 to п or from 1 to п’,
from plesiomorphic to apomorphic. The data set is extracted from Appendix C and from the data set
used in previous factor analyses. Some of the data missing from the latter have been found and added

IN PHENETIC AND PHYLOGENETIC ANALYSES

at this step to avoid the absence of some families (Appendix E).

GENERAL MORPHOLOGY

HD1 HD2 HD3 HD4 HD5 HD6 HD7 HD8
0.55 0.75 le 2 10 semislugs slugs missing
WH1 WH2 WH3 WH4 WH5 WH6 WH7 WH8 WH9
4.1 4.5 5 5.5 6.4 20 semislugs slugs missing
TA1 TA2 TA3 TA4 TAS
48 96 144 192 243
LV1 LV2 LV3 LV4 LV5 LV6 LV7 LV8
2:9 3.3 3.9 4.8 20 semislugs slugs missing
LS1 LS2 LS3 LS4 LS5 LS6 LS7
0.6 0.8 1:2 10 semislugs slugs missing
SS1 SS2 SS3 SS4 SS5 SS6 SS7
1.65 2 2:5 10 semislugs slugs missing
ST1 ST2 ST3 ST4 ST5 ST6 ST7
0.55 0.75 1.1 5 semislugs slugs missing
PULMONARY COMPLEX
EP LP2 LP3 LP4 LP5 LP6 LP7
0.6 0.75 152 10 semislugs slugs missing
LR1 LR2 LR2' LR3 LR4 LR5 LR6
0.45,0.7 0.36,0.45 0.7,1 0.25,0.36 0,0.25 semislugs, missing
slugs
UR1 UR2 UR3 UR4
0 2 4 5
LI LI2 LI3 LI4 RR1 RR2 CA1 CA2
1 2 3 4 12 0,1 0 1
DIGESTIVE TRACT
BM1 BM2 OC1 OC2 OC3 OC4
1 2 0 1 2 3
$С1 SC2 $С2' SC3 PS1 PS2 PS2’
We 2,3 1 3,4 11,2 2,3 1
IL1 IL2 IL2’ IL4
12 2,3 0,1 3,4
NERVOUS SYSTEM
СС1 CC2 CC3 CPD1 CPD2 CPD3 CPR1 CPR2 CPR3 CPR4
1 2 3 1 2 3 0.9 1.1 155 5
PLD1 PLD2 PLD3 PLG1 PLG2 PLG3 VG1 VG2 VG3
1 2 3 1 2 3 1 2 3
PAD1 PAD2 PAG1 PAG2 PAG3 PAG4 FG1 FG2 FG2' FG3
1 2 1 2 3 4 1 2 & 4

296

TILLIER

APPENDIX E. CHARACTER STATES USED IN PHYLOGENETIC RECONSTRUCTIONS This data set
is extracted from Appendices C and D, and from the data set used in previous factor analy-
ses. Some of the data missing in the latter have been found and added at this step to avoid the absence
absence of some families. Each line represents a species, in the same order as in Appendix A. The
numbers preceding each generic name are those of the families or subfamilies, as in Appendix A.

BM OC SC PS
O1AURICULELLA 1 1
O1ELASMIAS
O1LAMELLIDEA
01 TORNATELLINA
01STROBILUS
01 TEKOULINA
02ACHATINELLA

w
O
O
a
U

O3PARTULA
ОЗЕЧА
ОЗЗАМОАМА

w
O
O
wm
>

O4LEPTACHATINA
04AMASTRA

w
Q
O
Ф
av)

05СОСНИСОРА

w
Q
O
Ф
>

O6PYRAMIDULA

w
Q
O
10)
er

07BOTHRIOPUPA
07STERKIA

w
O
O
10)
U

080RCULA
08PAGODULINA

ee)
Q

O9SOLATOPUPA
O9GYLIOTRACHELA

w
Q
Ф
TU

10PUPILLA
10LAURIA

(ee)
Q
0
U

11VALLONIA
11PTYCHOPATULA
11ACANTHINULA
11SPERMODEA
11KLEMMIA
11STROBILOPS

a a
U

13CHONDRULA
13IMPARIETULA
13ENA

13ZEBRINA
13RACHISTIA
13CERASTUA
13AMIMOPIMA
13DRAPARNAUDIA

œ
Q
0
D

14SUCCINEA
14SUCCINEA2
140MALONYX

N a =p VNNanNaNnNa a NP MP MH —]— —\H NM TH nr — — = — _ — — m = г _ m A A Ap ыы — nd

2
De
2!

D & & D NN N NN N NN D NN NN NN D

(= €

(= E E =

D D

D : D

O O
ОО ОО où D D (À où uw & & ND ND

O
O
U

Q Q
lw

O
O
©

| №

O
TU

O
aU)

O
TU

®
TU
O О DO мо WHNTD OC = © © D w

| №

O O O
TU TU TU

O
ne)

| w
O
ae

O
©

®
©

O
TU
OD LO UI oh MoD £m AO A A

O
©
D ID

O
©
D

(9)
<
UN

O
<

D D D D © D D & D TD GO D — © CO D

| —
>
=

ae)
r

TU
r

TU
=

>
г
Ono 20 20 ---D D D D D D ©

TU TU TU
[== f= [=
| №

D
r

>
r

D
ES
вю раю DO D D D D D DO nb DO D = © D = ©

| —
>
=

>
r

a
r

>
г

¡ES
<

<
AS E

< < <

=
O © © D D D = D DO ---2-2-2-09 ЮО ЮО ЮО

IL LR UR АА СС СРО CPR PLD PLG VG PAD PAG

T
>

TU
>

TU TU
> >
© = 0.209 еее a a. ola

TU qu) TU
> > | >

U
>

x
>

TU
>
ON NWN == Ем ON ON де

>
In

о
>

TU T TU
> > >
QS (9) ING) NS@) ISIN ONDES

v
>
Im

U U
> >

D
>

w
>

>
QD Sf PR Sp BAR @) DARIN DD =:20 000

TU
>
la

O о wo OD nr pb DD PDP D DDR

an an
| nw

=-NONNNO OND ww мо

O m=umnmn=m-0

m
| co

BM OC
15ATHORACOPHORID 1 1

BM OC
16THAUMATODON 1 1
16LIBERA 1 1

17TRACHYCYSTIS
17RANFURLYA
17ANNOSELIX
17STEPHANODA
17CHAROPID
17ANDREFRANCIA
17MYSTIVAGOR
17PARARHYTIDA1
17PARARHYTIDA2
17PARARHYTIDA3

18CYSTOPELTA
18PHRIXGNATHUS
18LAOMA

19ANGUISPIRA
19DISCUS1
19DISCUS2
19HELICODISCUS

20HEMPHILLIA
200OPELTA
20ARIOLIMAX
20APHALLARION
20ZACOLEUS
20HESPERARION
20PROPHYSAON
20ARION
20GEOMALACUS
2000PELTA2
21PHILOMYCUS1
21PHILOMYCUS2

ies]

23PHENACOLIMAX1
23PLUTONIA
23PHENACOLIMAX2

24MESOMPHIX
24AEGOPINELLA
24ZONITES
240XYCHILUS

24 VENTRIDENS
24GASTRODONTA
24VITREA
24DAUDEBARDIA
24ZONITOIDES

B
25PARMACELLA
25PARMACELLA2

u
Q

26MILAX
B

1
2
1

M
1
1
1
2
1
1
1
2
4

M
1
1

M
1

27LIMAX

STYLOMMATOPHORAN SYSTEMATICS

SC
4

nN
O © © © © © © ND ©

10)

O = = D DO WHO

10)

PAE

DO DO 220 ND ND & x + 2 D D DO = D DO = © + + ND D

10)

(0)

PS
2

"TU "TU

(AS

IL
9!

TT D D D D © D D DDR

| лю-ёл7 YOYNAF DORE

Sa ean

UR RR CC CPD CPR PLD PLG VG PAD
4 =. ours a poe’ бе ho: ae
UR RR CC CPD CPR PLD PLG VG PAD
Di el OS A ae | ©: Я 1
Dim eek A
UR RR CC CPD CPR PLD PLG VG PAD
EE O O OS 1
AD O O, A 1 2
A e ar dal ed 1
Aa a Re ne 1
De a PERO o Ч 1
APS 3 A na. A 1
ДЕ UM ip) ST yy 1
A Whee aia! A e IM
At a SS Se
A DP aed. ESS CE e AAA
UR RR CC CPD CPR PLD PLG VG PAD
ar eB. AB OO Ae
Ay a. 092. 28). e * ail, dee
Ae) Е A —
UR RR CC CPD CPR PLD PLG VG PAD
4 2 3 4 г 1. “40 ge
E Se E ye ee
de Bie El eS es, a AAN
O OR o E
UR RR CC CPD CPR PLD PLG VG PAD
4 2.2 oe | RS > >
42 4 2..1. 4 jy Si 52
Ae |. A i
A Bea, = ae
AO ee ee ra
er —
Aa =
A =

LS — =
ПЕ —
а 8 Se о 2% 2
72 E E E: IR o
UR RR CC CPD CPR PLD PLG VG PAD
E E 35 it <2, e
Ay, fo 9. Ga 9.1 ie “Bs #2
oS en —
UR RR CC CPD CPR PLD PLG VG PAD
ao 3 © 1 4 11 a м
A sh tH oe Y 14 1
AL Ok oe ME 2 1 1. Bere
AB. ae O: 3 4 р 2
AO |e NI Ss à TRES
ae SO Sr (An oi DE
AND Е оо M
aS | 1 во 1 ni
ae ee Ц eee
UR RR СС СРО CPR PLD PLG VG PAD
Am Dal 1 DEN 1 32) 2
D о = — nu —
UR RR CC CPD CPR PLD PLG VG PAD
APE Er be dl I <3" 2
UR RR CC CPD CPR PLD PLG VG PAD
A (egos e

U U
> >
Laso le |

[SIE ©

>
(MET)
=
Eo)

TU
>

>
RO noO

[amic om main 0

| vu 0

| №0

т т
| ююо | w |

IEEE ES ho: @)

7 m
20 DO

298

BM OC SC PS

29TROCHOMORPHA 1

BM OC SC PS

30DISCOCONULUS1 1
30DISCOCONULUS2
30CONIBYCUS
SOCONEUPLECTA1
SOCONEUPLECTA2
30CONEUPLECTA3
30VITRINOPSIS

4
4
1
1
1
1

31HEMIPLECTA 1

31EVERETTIA 1

31MICROPARMARION 1

31TROCHOZONITES 1

31ACANTHARION 1

31 TROCHONANINA 1

31ATOXON 1

31KALIDOS 1

31MARIAELLA 1

31PARMARION 1

31HELICARION 1

31GYMNARION 1

31MESAFRICARION 1

31GRANULARION 1

31ELISOLIMAX 1

M
1
M
4
1

B
32CECILIOIDES

B
33BOCAGEIA
33PSEUDOGLESSULA
33RUMINA
33SUBULINA

BM
34ACHATINA

w
=
Q

37NENIA
37ALBINARIA
37ITALA

38VARICELLA
38POIRETIA
38STREBELIA
38SPIRAXIS

40TRIGONEPHRUS
40DORCASIA
40STROPHOCHEILUS
40PEDINOGYRA
40HEDLEYELLA
40PYGMIPANDA
40CARYODES
40PANDOFELLA
40PANDA
40AMPELITA
40HELICOPHANTA
40CLAVATOR
40STYLODON
40ACAVUS
40ANOGLYPTA

ies}

O
AMO ооо оо TERRA TSE NE AN ET

O
©

D 20 =NNNNO =

10)

(0)

(ep)
NNA]ANNNNONNNAANNOD 2020 NO NO NNNNOD DO © © & & ND © D NO = 0 + © 1-10 200000

—

юг

NNNNNNNANNNNNANNF WWE YNNF AF YANN DT NNAAANNANNANNANF NN GG NN

BE = =
ю | In- | = ù0 0 © = 0 7 N № Go Co | ФА my

=
D фо D =D nn ©

r
[Sa Eee: | ha

ЕЕ

UR АВ СС СРО CPR PLD PLG VG PAD РАС FG
ee! 3 2 2 ee 2 Zul

ОВ RR: СС СРО’ ERR- RLDIPEGVGTRADZPAGEIES

À и 3. 3. 2 O Рю
ar з 2 № 2 © о ще
4 Е
и —
а = Jaa
ASS DS — — —
UR RR CC CPD CPR PLD PLG VG PAD PAG FG
4 18 о + IC
4, 2 8 2 2 т MC
дз 3 зо Me Moss
аз 3. 91 CLONES"
413 52, 2 я le ще
4 42 2 1 Tom A
48>) BA E A A 1
a4 —
pr
DO — —
Ae rit: B¥y Eee ee ae ae a 2
| Tr — —
SE ae or ee eee ee. :
== 2 3 2 № Co п
E E EA EA RÁ
UR RR CC CPD CPR PLD PLG VG PAD PAG FG
4 оз. о а 34° 2 A
UR ВВ СС СРО СРВ PLD PLG VG PAD PAG FG
42 3 12 ear 4)
4.1.3 2 8 À т 2 MMS
4.13 1 3 1 1 5
A = о OO —
UR ВВ СС СРО СРВ PLD PLG VG PAD PAG FG
4.2 8 2. 8 wf. па ра
UR RR CC CPD CPR PLD PLG VG PAD PAG FG
eee at 2 2 1 +1 2:
т РИ Pe о Вы ПЖ
1 2 4 8. 4° ила
UR RR CC CPD CPR PLD PLG VG PAD PAG FG
449 09 By 4 1 1 За
4 2 31 8 та
4-2. 3 "2 ма: ar
ве ug + —
UR RR СС СРО СРВ PLD PLG VG PAD PAG FG
1023 1 3.1 %1 а
| 2 GSN A! Эра FO 1a
{ 4 2 1:08 4° то м
22 2 чот 1 м eee
АЕ RCE 2 4 м 3 Ba
2-2 8 “2 Mo 1 м 2 И
9 2 8 1° 3 ot C1 2 а
382; 8, Di чае, El Gao
2 2 8 A. 0 1 12 Sour
2.1 3 204. 8 2.2 И
2 2 43. 1191 Ш 2 CR
Ом о О О
1,1 3 ооо ИЕ ee |:
E: E WOME AE AS
E eee Se ee |

STYLOMMATOPHORAN SYSTEMATICS 299

40MACROCYCLIS А ее 2 Ok eS 1 2 1 1 2 2 Ae
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
41BOSTRYX 1 CR a RE Om ae 72 2 2 1 1 3 2 4 2
41DISCOLEUS 1 See see Po) So! 2 2 4 1 1 3 2 JT
41COCHLORINA 1 ES SN 8 2 —= — == => te == E E
41PLACOSTYLUS 1 ЭР У" Sas CA 127 == — = == — — — er VE
41PLAGIODONTES 1 UL 2 BS 2 - 1 1 3 2 4 2
41SIMPULOPSIS 1 92 42 4 ‘23 1 2 1 3 2 4 3
41PELLICULA 1 Se eee! ede Oe 3 2 4 1 1 3 2 4 2
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
43CERION1 1 1 lel Ral) di cl on 2 2 3 1 1 2 1 JO
43CERION2 1 1 ao e il 2 — — — —
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
44BERENDTIA 1 IS 3 3 82h 3 1 2 1 12 2 4 -2
44MACROCERAMUS 1 Mara ete Sk oi 2 =>. = == - —— — — —
44UROCOPTIS — me I Zr EI EZ 1 2 1 1 2 2 Se 2
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
45SYSTROPHIA1 1 if Sem er 3 2 2 2 2 A 2 1 72
45SYSTROPHIA2 1 132 2 1 4 — = — == — — — — —
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
46HAPLOTREMA1 1 1 Wale 2272 4, 7 — = — — — — — —
46HAPLOTREMA2 SSS SS SS SS SS 2 3 1 1 2 2 З р 2
ВМ OC SC PS IE ER UR RR CG CPD CPR PLD (PEG VG PAD PAG FG
47PRIODISCUS 2 4 i 1 43 SF 4 2 33 2 2 1 1 3 1 SIMS
47DIPLOMPHALUS 2 oil m2 227 ES 12h 10. 8 3 2 1 1 2 1 3.3
47МАТА 2 25 © Spat 2: 826 Al 2025 3 1 2 1 Inn? 1 Steel
47RHYTIDA DANS A oe er — == AS — — —
47SCHIZOGLOSSA A A № 4 2 = = == - > — — —
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG F
49PTYCHOTREMA AAA a2 2 4 1 E 2 2 1 1 2 1 si
49EDENTULINA ES GN 2. фа = = = — — — — = —
BM OG SC PS Ш ER UR RR CC :CPD' CPR PIED РЕ VG PAD PAG ЕС
50POLYGYRA i 120 ele 2 cilia ds ail 3 3 3 1 1 2 2 4 $2
50MESODON 1 3 Hl «they oh ga at 3 1 3 1 1 2 2 4 2
50TRIODOPSIS 1 2 ere 2.4 il 3 2 3 1 1 3 2 4 2
50TRILOBOPSIS 1 ee IL А | — = SSS SS = — = =
50ALLOGONA 1 Se OP EE a gta И — — о SS SS
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
51PROSERPINULA 1 3214 23 2 2 1 1 3 2 4 2
51SAGDA 1 Wi 2) a D 4 4 2 83 2 2 1 1 2 2 A *2
51LACTEOLUNA 1 I AA Ge re = ANNE = AR —
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
52CORILLA 1 1 ¡ARES 281 2 12 2 2 1 1 2 2 4 1
525 СУЕРТАН!А 1 i oe 4, 5 2 3 2 2 2 1 3 1 1 1
52PLECTOPYLIS 1 i dl 2,72 | Du 1 2 2 1 3 1 1 1
52CRATERODISCUS 1 MARS LAS 4, 025 58 2 2 2 1 3 1 4 3
ВМ OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
53PLEURODONTE 1 2, Gili egg 2, 2) a ча 3 2 3 1 1 2 2 4 3
53LABYRINTHUS 1 2 poe din le il, AZ 53 2 4 1 1 1 2 3 2
53AMPLIRHAGADA 1 Ape pile di id à mit 3 2 3 1 1 2 2 Ay ail
53PLECTOTROPIS 1 A A ea E: 2 1 1 3 2 As 92%
53SOLAROPSIS 1 Al IZ À il 2 == O а See
5ЗАНАСАОА 1 ЗЕ ATA TRE EN 6. — —
53SINUMELON 1 CT A VE LS РЕ A SS ==
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
54AMMONITELLA 1 1 LIRA A DIR 1 1 2 Due 2 3 12
54GLYPTOSTOMA 1 de ALA A LS 1 2 1 1 1 1 3 1
550REOHELIX 1 1 lv CAT 2) NO nl 3 3 2 1 1 2 30
BM OC SC PS IL LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG

300

56HELICOSTYLUS 1

56BRADYBAENA 1
BM

57THYSANOPHORA 1
57AVERELLIA 1
57HELMINTHOGLYPTA 1
57EPIPHRAGMOPORA 1

57 SONORELLA 1
57CEPOLIS1 1
57MONADENIA 1
57CEPOLIS2 1
57PSADARA 1

58SPHINCTEROCHILA 1
58HALOLIMNOHELIX 1
58CANDIDULA
58HELICELLA
58COCHLICELLA
58MONACHA
58HELICODONTA
58HELICOGONA
58CERNUELLA
58TROCHOIDEA
58HELICOPSIS1
58HELICOPSIS2
58HELICOPSIS3
58HYGROMIA
58ELONA
58ARIANTA
58THEBA
58HELIX
58CEPAEA

Ser ee IN ee ml

O
OR

D © CO © CO D BR D —

VUORLAOAD-NN==NN=a=nNN=0=0

NN 2222NNN22NNNN==2=23NF 2222NNN==NF =

TILLIER

€

ORWOWAWWWWWWHWHWHWWWHWWATD PR BR À BR À BR OC CG OO D ap

D © D © D D © © © D D © D D © D D D D TD 1 1 1 à ND ND ND = AD A=

| WWWWWWOD Ww

| WWWWWWWWO о |

O
TU
O
TU

IES

O
©

[Гомо D D

O
©

| OWNWWNH x © ID w |

>
г
eh Y Y een lA

A A la 1

>
г
елеем м (6) — —

Ак АА А Е er

| D D & © © & 9 mu

À 1 & © & © À © G ю |

D
>

| © © 1 © 1 © O D D

TU
>

[Si NN №№ = а ю |

D
>

MS SO) ha

=
>

SS ESTO) |

Innnnnn® +o

|=-nnnnn-0@ - |

STYLOMMATOPHORAN SYSTEMATICS 301

APPENDIX F. CHARACTER STATES OF CCAs OF STYLOMMATOPHORAN FAMILIES (NODES IN
TEXT—FIGS. 20-22)

The character states of the families correspond to the lowest values in each familial group in Appendix
E. For the Orthurethra, the symplesiomorphies and synapomorphies of all members of the group, and
the autapomorphies of one member only have been eliminated: the former contribute no information at
all, and the latter modify the distances without contributing any information about relationships. For taxa
not belonging to the Orthurethra, the columns that have been doubled (SC, PS, IL, FG) correspond to
characters of which the plesiomorphic state is in the middle of the morphocline, denoted either from 1 to
п or from 1 to п’ in Appendix E. The numbers on the left, denoted “=п=” correspond to the
hypothetical nodes in Text—Figs. 20-22 (constructed by the algorithm).

ORTHURETHRA (Text-Fig. 20) SC PS CC CPD CPR PLD PLG VG PAG FG
O1ACHATINELLIDAE 1 22 ei 1
O3PARTULIDAE
04AMASTRIDAE
O5COCHLICOPIDAE
O6PYRAMYDULIDAE
O7VERTIGINIDAE
O80RCULIDAE
O9CHONDRINIDAE
10PUPILLIDAE
11VALLONIIDAE
13ENIDAE

юзаю UN =p =
= eût iA ed el ml) ke ar, Е IN) E E E = el dt eh
= D = = D ND D & ND ND ND — OU GO GO + ND N D
= D D D = D D D ND = D D ND = D D ND
= D ND ND D & N = D = & © DW © N
=o2 240424 = IN = DIN = 4
ara аа
à NON NO = № +

2 — — ND OO — OO ND fH DM H WNH — ND CO W
= = |= nm NN NN PY = = D D D D ND D D D D

LR UR RR CC CPD CPR PLD PLG VG PAD PAG FG
31
31

14SUCCINEIDAE
15ATHORACOPHORIDAE
16ENIDAE
17CHAROPIDAE
18PUNCTIDAE
19DISCIDAE
20ARIONIDAE
21PHILOMYCINAE
23VITRINIDAE
24ZONITIDAE
25PARMACELLIDAE
26MILACIDAE
27LIMACIDAE
29TROCHOMORPHIDAE
SOEUCONULIDAE
31HELICARIONIDAE
32FERUSSACIIDAE
33SUBULINIDAE
34ACHATINIDAE
37CLAUSILIIDAE
38OLEACINIDAE
40ACAVIDAE
41BULIMULIDAE
43CERIONIDAE
44UROCOPTIDAE
45SYSTROPHIIDAE
46HAPLOTREMATIDAE
47RHYTIDIDAE

pe Ga Ge Gee errs ee Cae E TE = = = № № = А = № SO EN d p)

Na = = = |S D = = = ld ld ld dl = ND A = = dl ld ld dl — — py —
=— = PM = = | | fP— = = NM = NN D NN = ND = NM = = ND NN D юм = —
OO D D © © CG OO R CO — о - ND ND O1 O1 O1 — O1 O1 O1 D D G D © O1 CO
NOANHHHWHWHHAHAAHAAAAAAAAHANMNAA
ND = = ND ND D = D ND D = D = HHA D D D = D D D D = = HAND
D N = D ND = = = D = D ND DO D ND = = © © ND = ND N D WwW
A = ND = = SH See = = dl dl A POD = +... Ww HSH = ND = NN NN =
жим At mn = = dl ld = NN = — = QD = Où = py = et
# D ND = ND = = = D = = NN D D D D = D ND ND D ND = = D N

= D D OO — CO — ND D — ND D — — ND ND OC ND D — ND ND HHH à 4 4 PO
CO CO — CO CO CO CO CO D BR CO CO OO — D BB BR BR = À CO On HH KH PH

D © D D OO D D D ND OO GO ND ND D — ND ND D — ND — CD ND HAH HWH HN

2
3
2
2
3
3
1
3
3
3
1
3
3
3
3
3
3
3
3
1
3
2
1
2
3
3
3
3

Dis 2 Se SH A Pp ad dl ld dl dl dl dl ld = dl dl = dl = dl dl = —
=> A A A A py a (QQ ld dl dl dl dl dl dl dl dl dl dl = dl — —
A A» > A A oo = = = = = = A DD = = = a dl = = ld dl = Ц
CON NO NO NON ею ооо
Na 2 2 2 A ANONONA mm m5 5 5 5 coc“

A AN Ata dl dl dl dl dl — ND A = — dl dl Ц
=— MO ANDAN pq. <<

TILLIER

302

11
LZ
12

We ot)
WA
doit

49STREPTAXIDAE
50POLYGYRIDAE

51SAGDIDAE

52CORILLIDAE

53CAMAENIDAE

54AMMONITELLINAE

550REOHELICIDAE

56BRADYBAENIDAE

57HELMINTHOGLYPTIDAE 1

58HELICIDAE

eal
1
el
vi
Veal

leat
Val
1m
da
11
11

ila
lial

|
|

||
||

le
Vz

13
11
11

Wed

|
|

det

1,1

181

142
11
lil
11
15

Wet
151
el
1551
uel
1a,
11
leat
iby
11

IS
== ON CD Sr i) OO 0 =

—

1:1
Wa
Ця

ll
N
==

==
=A

15
==
==
=18=

=19
=20=

ile

11
1
11

lea

1
11

22=
—23 =

151

ИА

24=

==
==
==
99
a0

11
17
ll
wat

15
We
11

STYLOMMATOPHORAN SYSTEMATICS 303

APPENDIX G. CHARACTER STATES OF NODES OF TEXT—FIG. 28A

IL LR UR RR CE CPR PAG
ENDODONTOIDEA 1 2 2 1 2 1 1
ZONITOIDEA 1 1 4 1 1 1 1
CLAUSILIOIDEA 1 3 1 2 1 2 2
HELICOIDEA 1 1 3 1 1 1 3
ACHATINOIDEA 1 1 3 1 2 1 3
ACAVIDAE + RHYTIDIDAE 2 3 1 1 2 2 3
CORILLIDAE 2 3 1 2 1 2 1
OREOHELICIDAE 1 3 2 2 1 1 3
=1= 2 3 1 1 1 2 1
=2= 1 3 1 2 1 1 2
=3= 1 3 1 1 1 1 1
=4= 1 1 3 1 1 1 1
=5= 1 2 1 1 1 1 1
== 1 1 1 1 1 1 1

MALACOLOGIA, 1989, 30(1-2): 305-315

CEPAEA NEMORALIS FROM LEXINGTON, VIRGINIA: THE ISOLATION AND
CHARACTERIZATION OF THEIR MITOCHONDRIAL DNA, THE IMPLICATIONS

FOR THEIR ORIGIN AND CLIMATIC SELECTION

O. Colin Stine

Department of Biology
University of Virginia
Charlottesville, Virginia, U.S.A. 22901

ABSTRACT

The introduced population of Cepaea nemoralis in Lexington, Virginia, provides a natural
experiment in which selection may be investigated. This colony of snails, polymorphic for shell
color, has a well-described history. However, the geographic origin of these snails is uncertain.
In this study, the origin of these snails has been inferred from measures of genetic relatedness
obtained from the restriction enzyme analysis of mitochondrial DNA. The analysis is performed
using Southern blotting and low-stringency hybridization with a radioactive probe made from
cloned Xenopus mitochondrial DNA. The Xenopus probe hybridizes to most of the Cepaea
mitochondrial genome, demonstrating the highly conservative nature of the major part of the
mitochondrial DNA sequence. Based on this restriction analysis, snails from Lexington were
more closely related to ones from England than to ones from Italy. Hence, a British origin has
been inferred for this population.

This conclusion is consistent with 19th century observations that brown-shelled individuals,
which are characteristic of northern Europe, could be found in the Lexington population. How-
ever, this dominant allele is no longer present in Lexington. The temporal loss of the brown
morph in this introduced population is attributed to natural selection mediated by climate, be-
cause the mean July temperature in Lexington is comparable to that in the southern extremes
of the native European range of Cepaea, where brown individuals are absent. The selection

coefficient against browns in Lexington is estimated to be 0.35.

INTRODUCTION

Directional selection in a polymorphic pop-
ulation produces a temporal change in geno-
typic frequencies resulting from the action of
some causal agent. Although selection is of-
ten difficult to demonstrate in natural popula-
tions, it may be more easily detected in intro-
duced populations exposed to new and
different environmental conditions. Cepaea
nemoralis in Lexington, Virginia, is an intro-
duced population with a well-described his-
tory since its introduction. These snails are
thought to have been inadvertently imported
from Europe in 1883 (Howe, 1898). However,
the geographic origin is uncertain. Once the
origin is ascertained, they will be a natural
population in which climatic selection on shell
morphs can be demonstrated. Cepaea nem-

oralis is polymorphic for shell color. The locus
for this trait has alleles for brown, pink, and
yellow, in order of decreasing dominance
(Murray, 1975). The frequency of the brown
morph in the colony of snails at Lexington has
decreased since the introduction in 1883.
Cockerell (1889, 1894, & 1897), Brooke
(1897) and Howe (1898) reported the exis-
tence of brown snails in the population. At
present, however, there are no brown-shelled
individuals in Lexington.

The geographic origin of the population in
Lexington may be either Italy or England
(Howe, 1898). These two possible origins
suggest different hypotheses for explaining
the frequencies of shell colors in the present-
day population. An Italian origin would sup-
port the hypothesis that the Lexington snails
simply reflect the frequencies of the founding

Present address: 933 Traylor Bldg., 720 Rutland Ave., The Johns Hopkins Medical Institutes, Baltimore, Maryland, U.S.A.

21205.

(305)

306 STINE

population (Brussard, 1975), since there are
predominantly yellow snails, a few pinks, and
no browns in Italy (Taylor, 1914; Sacchi 8
Valli, 1975). Alternatively, a British origin
would support the hypothesis that brown-
shelled morphs have been eliminated by cli-
matic selection in Lexington.

The origin of the Cepaea in Lexington has
been investigated in two studies by measur-
ing the genetic relatedness of the population
to those in Europe by protein electrophoresis.
The conclusions are at least partially contra-
dictory. Brussard (1975) surveyed 30 popula-
tions from Virginia, New York, Massachu-
setts, Ontario, and Wales. The populations
were separated into two groups based on
gene frequencies at two variable loci. Despite
the small number of phylogenetically informa-
tive loci (two) and the failure to include any
snails from Italy, Brussard (1975) concluded
that the two groups represent two phyloge-
netic lineages. One was thought to be north-
ern European; and the other, including the
Lexington samples, Italian. Johnson et al.
(1984) surveyed populations from four loca-
tions: Lexington, Virginia; Berrow, Somerset,
England; Florence, Tuscany, Italy; and Santa
Croce, Lombardy, Italy. Based on 16 informa-
tive loci, the genetic distance (Nei, 1972) be-
tween the Berrow and Lexington populations
was 1/2 to 1/3 of the distance between the
Lexington and Italian populations. These ob-
servations are consistent with a British origin
for the Lexington snails. Despite this result,
the authors concluded that the colony could
be of hybrid origin because the Lexington col-
ony shared some alleles with each possible
progenitor population which were found in
only one of the populations. However, this
conclusion should be interpreted with caution
since the Lexington population contained
an allele not found in any of the possible pro-
genitors. Furthermore, other electrophoretic
surveys of protein differences between popu-
lations of Cepaea nemoralis have demon-
strated that large genetic distances and often
fixed allelic differences may be found be-
tween populations separated by only a few
kilometers (Johnson, 1976, 1979; Ochman et
al., 1983).

Restriction analysis of mitochondrial DNA
provides information on the genetic related-
ness between populations within a species
(see Avise, 1986, for a review). Although
some nucleotides in the mitochondrial ge-
nome vary within and between species, direct
sequencing has shown that approximately

70% of the bases are identical in representa-
tives of artiodactyls (Anderson et al., 1982),
rodents (Bibb et al., 1981), and primates
(Anderson et al., 1981). If these nucleotides in
the mitochondrial genome are conserved
throughout the animal kingdom, then the pos-
sibility exists of rapidly characterizing the mi-
tochondrial DNA from other organisms by us-
ing Southern blotting and low-stringency
hybridizations (Howley et al., 1979).

MATERIALS AND METHODS
Collection localities

Cepaea were collected from several sites in
and around Lexington, Virginia: the vetch-
covered embankment of Virginia Route 39
just west of its intersection with U.S. Route 11
north of town, the southern embankment of
U.S. Route 60 across from Reeds Pond just
west of town, and the embankment across the
street from the maintenance shops of Virginia
Military Institute. Other North American sites
were the slope around the primary sewage
treatment plant along Virginia Route 39
0.8 km west of Warm Springs, Virginia, and
the weedy piles of rubble near the leaf com-
post pile near the State Highway Department
east of East Avenue and south of E. Jefferson
Street near Pittsford in Monroe County, New
York. The English samples were provided by
Dr. B. C. Clarke of the University of Notting-
ham and Dr. J. J. Murray of the University of
Virginia. These were collected near Gamston,
Nottinghamshire, on the towpath 46 m south
of a bridge (U.K. Ordnance Survey, SK598
378); and near Kirk Dale, Derbyshire, on the
east side of the road along a limestone wall
(SK182 686). The Italian snails were the gift
of Drs. C. F. Sacchi, S. D. Jayakar, C. Violani,
and S. Malcevski of the University of Pavia.
These snails are a subset of those used by
Johnson et al. (1984). An additional Italian
sample was collected in rough herbage next
to a field uphill from the stream near Monte
Morello near Florence, Italy.

Isolation of mitochondrial DNA

The foot, genitalia, and hepatopancreas of
a single individual were minced. The tissue
was homogenized in 5 ml of ice-cold 0.25 M
sucrose in TEK (50 mM Tris-HCI, pH 7.5, 10
mM EDTA, 1.5% KCl) containing 140 p.g/ml of
ethidium bromide in a Dounce homogenizer.

CEPAEA IN LEXINGTON 307

The ethidium must be added to inhibit nu-
clease activity. The mucopolysaccharides
were removed by centrifuging the sample
through a layer of 1.1 M sucrose in TEK at 13
000 xg for 50 min at 4”C. The pellet contain-
ing the mitochondria was resuspended in
0.25 М sucrose in ТЕК and the centrifugation
through the dense sucrose was repeated.
The two sucrose step gradients are required
to remove contaminating mucopolysaccha-
rides which copurify with the DNA and which
inhibit subsequent endonuclease digestion.
Other methods including cesium-chloride gra-
dients and Proteinase K treatment did not re-
move the mucopolysaccharides (for details,
see Stine, 1986).

The mitochondria were disrupted by resus-
pending the pellet in 1.4 ml of ice-cold 2%
NP40 (Sigma N-6507) in TEK (NP40 lyses all
membranes in the cell except the nuclear
membrane) (Chapman & Powers, 1984). The
sample was placed on ice for 10 min, and
then centrifuged at 13 000 xg for 10 min at
4”C. The supernatant was extracted first with
phenol, and then with chloroform. The DNA
was recovered by ethanol precipitation.

The mitochondrial DNA used as a hybrid-
ization probe came from either Cepaea or Xe-
nopus. lf Cepaea DNA was to be used, 12 to
20 DNA pellets were combined and closed-
circular DNA was isolated on a cesium chlo-
ride (Kawecki Berylco Industries) equilibrium
gradient containing ethidium bromide. De-
spite this purification step, the Cepaea probe
could not be used to identify restriction frag-
ments of mitochondrial DNA isolated from fro-
zen snails. A second source of mitochondrial
DNA that was used for a probe was the plas-
mid pXIm31, which contains the entire mito-
chondrial genome of Xenopus laevis and was
isolated by Igor Dawid. The plasmid was iso-
lated from E. coli strain HB 101 by standard
techniques (Maniatis et al., 1982).

Restriction analysis of mitochondrial DNA

The DNA from a single individual snail
could be aliquoted and used in as many as
five separate restriction digests. The DNA
samples were restricted to completion with 10
units of the appropriate enzyme(s) in core
buffer (Bethesda Research Laboratories
(BRL)) containing 2.5 mM spermidine for 6
hours at 37°C. The samples were subjected
to electrophoresis on agarose gels and
stained using standard techniques (Maniatis
et al., 1982). The gel was photographed using

a light box, a UV-340 filter (Hoya Optics), and
a 30-minute exposure time. The gel was
treated and transferred to Gene Screen Plus
nylon membrane (New England Nuclear) ac-
cording to the manufacturer's directions. The
oven-dried membrane was placed in a “seal-
a-meal” bag (Sears) with approximately
0.1 ml of Stark's hybridization buffer [0.75 M
NaCl, 0.075 M Na citrate, pH 7.0, 0.02 M
NaPO,, pH 7.0, 20% formamide, 10% dex-
tran sulfate (Sigma D-6001), 0.1% Na pyro-
phosphate, 0.1% SDS, 0.1% Ficoll (Sigma
F-4375), 0.1% polyvinylpyrrolidone (Sigma
PVP360), 0.1% bovine serum albumin and
100 g/ml depurinated herring sperm DNA]
рег ст? of membrane. The blot was prehy-
bridized for at least 4 hours at 42°C.

The appropriate radioactive probe (Xeno-
pus, Cepaea, or lambda DNA (BRL)) was pre-
pared using a nick translation kit (BRL); 10° to
10’ cpm of probe was denatured, quenched
and added to the bag containing one 15 x 10
cm? blot for overnight incubation at 42°C. Ex-
cess radioactivity was washed off in repeated
washes of 0.3 M NaCl, 0.03 M Na citrate and
0.1% SDS. The resulting mitochondrial DNA
bands were displayed by autoradiography.

The DNA was mapped using the restriction
enzymes Bam HI, Hind Ill, Pst |, and Xho 1,
either singly or in pairs. The relatedness of
the populations was estimated by computing
the proportion of shared fragments, F, for
each pair of populations (Lansman et al.,
1981).

RESULTS
Detection of Restriction Fragments

Restriction fragments of mitochondrial DNA
from Cepaea were detected on Southern
blots using two hybridization probes: (1) Ce-
paea DNA recovered from closed-circular
band on a cesium chloride gradient contain-
ing ethidium bromide, and (2) Xenopus mito-
chondrial DNA isolated from the cloned plas-
mid pXim31. Radioactive probes were
necessary because the total amount of DNA
from a single snail could only occasionally be
observed by ethidium bromide staining
(Figure 1A, lanes 1, 2, and 4). Both the Ce-
paea and the Xenopus probes detected re-
striction fragments of Cepaea mitochondrial
DNA, but each probe had limitations. The Ce-
paea probe detected all of the restriction frag-
ments of purified mitochondrial DNA from

308 STINE

PHOTOGRAPH

AUTORADIOGRAM

FIG. 1. À comparison between a photograph of an
ethidium stained gel and its autoradiogram using a
Xenopus probe. Lanes 1 and 7 contain lambda
DNA digested with Hind Ill. Each of the lanes 2
through 6 is the mitochondrial DNA from an individ-
ual snail from Lexington digested by Hind Ill. The
arrows indicate the heteroplasmic bands in lane 3.

fresh snails (Figure 2). The signal from each
band was in proportion to its molecular
weight. The maps of the mitochondrial ge-
nomes for some populations of Cepaea are
shown in Figure 3. However, this probe could
not be used on samples isolated from frozen
snails because of background hybridization in
the lane (data not shown), presumably
caused by contamination of both the probe
and the sample by nuclear DNA.

Restriction fragments from Cepaea mito-
chondrial DNA hybridized to a probe made
from cloned Xenopus mitochondrial DNA un-
der conditions of reduced stringency, i.e. in

Stark's buffer containing 20% formamide at .

42°C. In Figure 1, every Hind Ill restriction
fragment that can be visualized by ethidium
bromide staining is detected by the Xenopus
probe on the subsequent Southern blot of the
same gel.

Additionally, the restriction fragments de-
tected by the Xenopus probe were compared
to those detected by the Cepaea probe. All
but one of the fragments were detected by
both probes. Figure 4 shows the 3.0 kb Pst |
fragment being detected by the Cepaea
probe but not the Xenopus probe. The posi-
tion of this fragment is located on the map in
Figure 3. Although the cloned Xenopus probe
does not detect the entire mitochondrial ge-
nome of Cepaea, it does provide interpretable
results when hybridized to a DNA sample
from a frozen snail (Figure 6, Lanes 1 and 6).

го

= 5:0 kD

— 1.5 ВБ

FIG. 2. Autoradiogram of the restriction fragments
of the mitochondrial DNA from an individual snail
from Pittsford, New York. The Southern blot was
probed with Cepaea mitochondrial DNA. The DNA
in lane 1 was digested with Xho I; that in lane 2, with
Xho | and Hind Ill; that in lane 3, with Hind Ill; that
in lane 4, with Hind Ш and Pst I; and that in lane 5,
with Pst I.

Comparison of restriction patterns

Restriction patterns of mitochondrial DNA
were obtained from a total of 129 snails from
7 populations: Lexington, Virginia; Warm
Springs, Virginia; Pittsford, New York; Kirk
Dale, Derbyshire, England; Gamston, Not-
tinghamshire, England; Santa Croce, Lom-
bardy, Italy; and Florence, Tuscany, Italy. The
patterns generated by four restriction en-
zymes, Bam HI, Hind Ill, Xho |, and Pst | dif-
fered in the position of their cleavage sites
between populations but not within popula-
tions. This was true even in the Lexington
population, which includes snails from several
collecting sites.

The restriction patterns in the Lexington

CEPAEA IN LEXINGTON 309

Warm Springs

and Pittsford x H H Р.В EP X PH H HX
о у
Kirk Dale x H HB Р HP H _H HX
esa ааа р
CO) A ВН | и A

ow similarity
with Xenopus —
| kb

FIG. 3. Maps of the mitochondrial genome of Cepaea nemoralis. The maps were constructed by comparing
the fragment patterns obtained from two enzymes together with those obtained from each enzyme individ-
ually. Each site is labelled by the first letter of the enzymes Bam HI, Hind Ill, Pst I, and Xho I, cleaving at that
site. The asterisks mark the fragment in which the length polymorphism occurs. The region with low se-

quence similarity to Xenopus mitochondrial DNA is identified by the arrows.

snails appeared heterogeneous because of a
length polymorphism. Indeed, 50% of the in-
dividuals were heteroplasmic, i.e. carried two
or more size classes of DNA as exemplified
by the individual in lane 3 of Figure 1A. The
heteroplasmic fragments can be seen as the
less intensely staining bands at 4.0 and 4.2 kb
(arrows). When the DNA from several individ-
uals is combined, as in Figure 5, the different
size classes can be seen. The mitochondrial
DNA in Lane 2 of this photograph was iso-
lated from 6 snails and digested with Hind III
and Pst |. As can be seen, there are major
bands at 2.3 and 2.7 kb and a series of 12
bands intermediate in intensity forming a lad-
der ranging in size from 3.1 kb to 6.0 kb. This
pattern suggests that bands differ by the
number of copies of a 240 base pair repeat.
The point of insertion can be mapped to the
fragment marked with asterisks in Figure 3..

The restriction fragments detected by the
Xenopus probe were used for comparing the
various populations (Table I) because only
frozen Italian snails were available for this
study. Fragments were compared instead of
restriction sites, because the sites could not
be mapped for the Italian snails due to the
lack of fragments shared with the mapped En-
glish mitochondrial DNA and an insufficient
number of samples for independent mapping.
Figure 6 shows some of the data on which
Table | is based. As can be seen, the Italian
snails do not share any fragments with Amer-
ican and English snails. In contrast, the Amer-
ican and English snails share a total of three
fragments (one in Lanes 2 and 3, and two in

-epoea Probe

Xenopus Probe

FIG. 4. Autoradiogram comparing Cepaea and Xe-
nopus probes. The probe on the left was prepared
from Cepaea while that on the right was from Xe-
nopus mitochondrial DNA. Lane 1 is a standard
consisting of lambda DNA restricted with Hind Ill.
Lanes 2 and 3 contain Cepaea mitochondrial DNA
restricted by the endonucleases Pst | and Hind Ill,
respectively. The arrow indicates the band not hy-
bridized by the Xenopus probe.

Lanes 7 and 8). Indeed for each enzyme in-
dividually (see Table I), the Lexington popu-
lation shares fragments with other American
and English populations but not with either
Italian population. The Warm Springs and
Pittsford populations have identical fragment

310 STINE

— © КБ

= 3.0 kb

FIG. 5. The polymorphism in the length of mito-
chondrial DNA from Lexington. This gel was
stained with ethidium bromide. Lane 2 contains mi-
tochondrial DNA isolated from six snails digested
with Hind Ill and Pst I. The ladder that is formed by
the polymorphic fragments extends from 3.1 to 6.0
kb. Lane 1 contains DNA digested with Hind Ill.

patterns for all of the enzymes tested. These
two populations are probably the result of a
single introduction independent from the one
at Lexington.

The relatedness of the populations was es-
timated by computing the proportion of
shared fragments, F, for each pair of popula-
tions (Table Il). The Lexington population was
most closely related to the populations from
Pittsford, Warm Springs, and Kirk Dale; and
more distantly related to the Gamston popu-
lation. The Lexington population was not
closely related to either Italian population.
Thus, based on this restriction analysis of mi-
tochondrial DNA, the Cepaea nemoralis that
were introduced to Lexington, Virginia, were
more closely related genetically to snails from
England than they were to those from Italy.

таза 6 789
® - 23.7 kb
| -94 kb
heme —67 kb

> + -4.3kb

St Bh
æ —23kb
# —-20 kb

FIG. 6. An autoradiogram comparing Cepaea mito-
chondrial DNA from Lexington, Kirk Dale, and
Santa Croce. Lane 1 contains DNA isolated from a
snail from Santa Croce; lane 6, another snail from
Santa Croce; lanes 2 and 7, a snail from Lexington;
lanes 3 and 8, a snail from Kirk Dale; lanes 4 and 9
from phage lambda. The DNA in lanes 1, 2, and 3
was digested with Xho | and Pst | combined; that in
lanes 4, 6, 7, 8, and 9 with Hind Ill.

DISCUSSION
Molluscan mitochondrial DNA

The mitochondrial DNA of Cepaea nemor-
alis typifies that of animals. Although some
nucleotides vary within and between species,
substantial portions of the sequence are
strictly conserved between different phyla, as
is clearly demonstrated by the cross-hybrid-
ization between Cepaea and Xenopus mito-
chondrial DNAs. The sequence similarity of
two DNAs can be estimated from the condi-
tions in which hybridization occurs. DNAs that
have 70% or greater sequence identity rean-
neal in Stark’s hybridization buffer with 20%
formamide at 42°C, while at least 77% simi-
larity is required if the formamide is increased
to 30% (Howley et al., 1979). In 20% forma-
mide, most Xenopus and Cepaea mitochon-

CEPAEA IN LEXINGTON 311

TABLE 1. The restriction fragments in selected populations of Cepaea nemoralis.

Warm

Lexington, Springs,
Virginia Virginia

n= 96 7 5

Pittsford,
New York

Kirk Dale, | Gamston,
Derby- Nottingham- Croce, Florence,
shire shire Lombardy Tuscany

9 5 4 3

Santa

Hind Ш 4.1 Sh +
fragments 4.0

+ + + +

Xho | 14
fragments 10.5 Te т +

Pst | 11.8 + +
fragments 7.9

Bam HI 16 ie + +
fragments 7.0

Pstl/Xhol 11
fragments 10.0

“polymorphic for length

drial restriction fragments hybridize with each
other; but in the more stringent conditions of
30% formamide, none of these same frag-
ments cross-hybridize. These observations
are consistent with the hypothesis that at
least five distinct and widely spaced portions
of the two mitochondrial DNAs share at least

70%, but not more than 77% of their base
sequence. This study, in combination with
previous work, therefore establishes that mol-
luscs, of which Cepaea is a representative,
share substantial mitochondrial sequence
identity with vertebrates (Anderson et al.,
1982), arthropods (Clary & Wolstenholme,

312 STINE

TABLE Il. The proportion of shared fragments (above the diagonal) and the variance (below the

diagonal) between populations of Cepaea nemoralis.

Lexington Warm Springs Pittsford Kirk Dale Gamston Santa Croce Florence

п= 96 п=7 n=5 n=9 n=5 n=4 n=3
Lexington, Virginia = 0.53 0.53 0.59 0.13 0 0
Warm Springs, Virginia 0.02 = 1.00 0.63 0.29 0 0
Pittsford, New York 0.02 0 = 0.63 0.29 0 0
Kirk Dale, England 0.03 0.03 0.03 — 0.29 0 0
Gamston, England 0.01 0.03 0.03 0.03 — 0.17 0
Santa Croce, Italy 0 0 0 0 0.02 = 0.5
Florence, Italy 0 0 0 0 0 0.06 —

1985), annelids (Dawid & Brown, 1970), nem-
atodes (Wolstenholme et al., 1987), and
echinoderms (Roberts et al., 1983). The ob-
servation of a highly conserved sequence
among these six phyla strongly supports the
mathematical arguments suggesting that
strong selective pressures must be operating
on the mitochondrial genome (Adams 4 Roth-
man, 1982; Aquadro et al., 1984).

Many but not all restriction fragments from
the mitochondrial genome of Cepaea are rec-
ognized by the Xenopus probe. Those frag-
ments which do hybridize probably contain
the genes for large and small subunits of ri-
bosomal RNA, the three cytochrome oxidase
subunits, cytochrome b, and ATPase 8 be-
cause these genes have a high sequence
similarity in another comparison between a
protostome and a deuterostome (Drosophila
and Mus, Clary 8 Wolstenholme, 1985). The
fragment which failed to hybridize must con-
tain only dissimilar sequences. These genes
may include the NADH dehydrogenase sub-
units 4b, 5 and 6; and the D-loop all of which
share less than 43% of the base sequence
between Drosophila and Mus (Clary 8 Wol-
stenholme, 1985), a sequence similarity far
less than that necessary for DNAs to hybrid-
ize in 20% formamide.

The portion of the mitochondrial genome
detected by the Xenopus probe can be used
for comparing the genetic relatedness of indi-
viduals. The genome can be mapped. Re-
striction fragments can be compared. Two
classes of fragments will be undetected:
those that are too small to remain on the gel
and those that only contain DNA from within
the region of low similarity.

Cepaea in Lexington

The only detected variation in the mito-
chondrial DNA of the Cepaea from Lexington
is a length polymorphism. There are 12 differ-

ent size classes of mitochondrial DNA that
appear to have different numbers of copies of
a 240 base pair repeat. Approximately 50% of
the individuals in the Lexington population are
heteroplasmic. Exactly what this putative re-
petitive sequence might be is unknown. How-
ever, similar inserts have been observed in
mitochondrial DNA from Drosophila mauriti-
ana (500 or 200 bp, Solignac et al., 1984),
Gryllus firmus and G. pennsylvanicus (330
bp, Harrison et al., 1985), and Morone saxa-
tilis (110 bp, Chapman, 1987).

Two other snails had mitochondrial DNA
patterns that were consistent with the pres-
ence of a length polymorphism. One individ-
ual from Kirk Dale was heteroplasmic in the
same fragment as the Lexington snails. And
in an individual from Florence, the total length
of Bam HI fragments is consistent with het-
eroplasmy. However, in this latter case, the
presence of the length polymorphism cannot
be confirmed until the genome is mapped for
that population. Whether any phylogenetic in-
formation is present in these observations will
remain unclear until the inheritance charac-
teristics of this trait are understood.

The Cepaea in Lexington are more closely
related to snails from England than to ones
from Italy based on restriction analysis of mi-
tochondrial DNA, the quantitative data for
isozymes (Johnson et al., 1984) and the early
records of brown-shelled individuals in Lex-
ington (Cockerell, 1889, 1894, & 1897;
Brooke, 1897; Howe, 1898). The most closely
related populations share a smaller proportion
of restriction fragments than do populations
within other species (see Avise, 1986), even
though the Cepaea populations maybe geo-
graphically close, e.g. Gamston and Kirk Dale,
which are 96 km apart. This parallels obser-
vations from isozyme analysis where smaller
genetic identities (Nei, 1982) are seen be-
tween populations of Cepaea compared to

CEPAEA IN LEXINGTON 313

other species whether the populations of Ce-
paea are geographically widely separated
(Johnson et al., 1984) or geographically close
(Johnson, 1976, 1979; Ochman et al., 1983).
The dominant brown allele is found only in the
cool climates of Northern Europe and, except
for Como in the Swiss Alps, never in Italy (Tay-
lor, 1914; Sacchi 4 Valli, 1975; pers. obs.).

All 96 snails from Lexington had the same
pattern of restriction sites, consistent with a
single origin. While a hybrid origin cannot be
completely ruled out, several observations
make it less likely. Cepaea are obligate out-
crossing hermaphrodites which exhibit simul-
taneous intromission. Breeding experiments
have shown that crosses between English
and Italian snails are fertile (Johnson et al.,
1984). Thus, all individuals would be ex-
pected to contribute mitochondrial DNA to
subsequent generations.

Additionally, the population parameters of
the Lexington colony are nearly ideal for re-
taining all of the genetic information present in
the founders in the absence of selection. First,
these snails have a very large neighborhood
size, so that drift due to population subdivision
is unlikely. The sample density in 1897 was
100 per m? (Howe, 1898). Presently it is over
50 per m?. These data when combined with
data on the movement of Cepaea nemoralis
(Murray, 1964) produce an estimated neigh-
borhood size of 2,500 to 5,000 snails. Second,
the colony expanded rapidly, so that every
adult would be expected to contribute to the
next generation. From a few introduced indi-
viduals, the colony expanded to cover an area
estimated to be over 130 ha in 1897 (Howe,
1898), to have doubled its area by 1935 (Mc-
Connell, 1935), and now, to cover at least 780
ha. Third, Murray (1964) showed that the mat-
ing system, which includes multiple mating
and sperm storage, minimizes any effect of a
temporary decrease in population numbers.

The current absence of brown-shelled Ce-
paea in Lexington (Murray, unpub; Brussard,
1976; per. obs.) must be attributed to selec-
tion because the large number brown-shelled
individuals and the population parameters ef-
fectively rule out random drift. (The average
effect of drift per generation is less than 10 °
(see equation 3.1, Falconer, 1976)). Howe
(1898) collected 2,134 snails of which at least
52 (2.4%) were brown (vars. olivacea, hepat-
ica, purpureo-tincta, or petivera; see Taylor,
1914) in an area less than half the panmictic
area for this species. His observation is con-
sistent with the interpretation that the high

rate of successful reproduction led to a nu-
merical increase in the number of browns
even though there was a proportional de-
crease due to selection. Assuming that
Howe's collections provide a good estimate of
the gene frequency (0.125) in 1897, that
brown morphs are now present only at the
rate of mutation (10 ©), and that a generation
is three years long (Clarke & Murray, 1962), a
coefficient of selection can be calculated us-
ing Bulmer's method (Clarke 8 Murray, 1962).
The coefficient of selection is 0.35. (The cal-
culated coefficient increases by about 0.1, if
the present frequency is assumed to be 10 ?
or if the generation time is assumed to be four
years long, and decreases by 0.1, if the
present frequency is assumed to be 10 °).

The frequency of the dominant brown
morph at the time of introduction can be cal-
culated because the introduction was five
generations earlier and the selective coeffi-
cient is known. The calculated frequency in
the founding population is likely to have been
about 0.11, well within the range observed in
English populations (Jones et al., 1977), as
expected for a population with a mating sys-
tem that minimizes the effect of bottlenecks
(Murray, 1964). Although the selective coeffi-
cient seems high, the numbers are consistent
with the observation by Brooke (1897) that
the brown Cepaea were initially common, but
by 1897 were uncommon.

The calculated selective coefficient is also
consistent with the observed 7% increase in
the frequency of the brown morph in caged
populations in a single summer (Bantock,
1974), where browns would be expected to
be favored by climatic selection. Although a
selective coefficient cannot be calculated be-
cause the change in gene frequency was not
measured over a generation, the observation
is consistent with the hypothesis that climate
may substantially alter the gene frequency in
a single generation.

The calculated coefficient can be ascribed
to climatic selection for three reasons. First,
brown shells absorb more heat than yellow
shells when exposed to sunlight (Heath,
1975). Second, in Europe the frequency of
yellow shells is positively correlated with the
mean July temperature (p < 0.001, Jones et
al., 1977): the frequency of yellow-shelled in-
dividuals approaches 100% at 22°C. Third,
the morph frequencies in the sampled North
American populations are consistent with
those in Europe. In Lexington, the proportion
of yellow-shelled individuals is greater than

314 STINE

95% and the mean July temperature is 2.6°C.
In cooler localities, like Pittsford and Warm
Springs, the proportion of yellow-shelled indi-
viduals is reduced and the brown morph is
present (pers. obs.).

An alternative, selection by predators, is
unlikely to cause the loss of the allele for
brown. Visual selection by song thrushes
(Turdus ericetorum) has been demonstrated
in Oxfordshire and other places (see Jones et
al., 1977, for a review). There are no song
thrushes in Lexington, but there are avian
predators such as grackles (Quiscalus quis-
cula, pers. obs.), chickens (Howe, 1898),
crows (Sacchi 8 Valli, 1975), cardinals
(Webb, 1952), and possibly robins and black-
birds (Bent, 1949). Visual selection probably
occurs in Lexington (Howe, 1898; Richards 8
Murray, 1975), but it should favor the brown
allele in shady habitats, thus maintaining it in
the population.

A third alternative, microclimatic selection
should maintain the brown morph in the pop-
ulation because it favors the brown allele in
dark habitats which have been and are
present in Lexington. Howe (1898) states, “in
one locality, quite protected from the light
there seems to be a preponderance of darker
shells and fused bands.” Currently, Cepaea
can be collected on the steep wooded banks
of Woods Creek and along the dark north fac-
ing cliffs of the Maury River.

The physiological mechanism of climatic
selection remains obscure. It is unlikely to be
a simple effect, such as mortality from over-
heating in the sunlight. Visual selection favors
brown-shelled snails in shaded areas where
they would be protected from direct sunlight.

Although Richardson (1974) thought he had .

demonstrated a direct effect of sunlight on vi-
ability resulting from shell color, several flaws
seriously compromise this conclusion. First,
he assumed that the frequencies in random
quadrats are the same as those over a wide
area, which is not the case in Cepaea (Cain,
1968; Murray 8 Clarke, 1978). In addition, the
author used the identical proportions of ex-
pected (live) morphs in two years. If his claims
that selection occurred in the first year are
correct, the proportions of live morphs must
have changed for the second.

In conclusion, mitochondrial DNA has been
isolated from Cepaea nemoralis. This DNA
has been hybridized to Xenopus mitochon-
drial DNA, demonstrating the highly con-
served nature of a large portion of the se-
quence of the mitochondrial genome. Based

on a restriction analysis of mitochondrial
DNA, the Cepaea nemoralis from Lexington,
Virginia, are more closely related to snails
from England than to those from Italy. The
major implication of the English origin for the
Cepaea in Lexington is that it supports the
early report of brown-shelled snails in this
population and suggests a clear temporal
change caused by climatic selection.

LITERATURE CITED

ADAMS, J. & ROTHMAN, Е. D., 1982, Estimation
of phylogenetic relationships from DNA restric-
tion patterns and selection of endonuclease
cleavage sites. Proceedings of the National
Academy of Sciences, U.S.A., 79:3560-3564.

ANDERSON, S., BANKIER, A. T., BARRELL, B. G.,
de BRUIN, М. H. L., COULSON, A. R., DROUIN,
J); ЕРЕВОМ, I. 'С., NIERLICH DP AROEABTAS
SANGER, F., SCHREIER, Р. Н., SMITH, А. J.H.,
STADEN, В. & YOUNG, 1. G., 1981, Sequence
and organization of the human mitochondrial ge-
nome. Nature, 290:457—465.

ANDERSON, S., de BRUIJN, М. H. L., COULSON,
А. R., EPERON, |. C., SANGER, Е. & YOUNG, I.
G., 1982, Complete sequence of bovine mito-
chondrial DNA: Conserved features of the mam-
malian mitochondrial genome. Journal of Molec-
ular Biology, 156:683-717.

AQUADRO, C. F., KAPLAN, N. & RISKO, K. J.,
1984, An analysis of the dynamics of mammalian
mitochondrial DNA sequence evolution. Molecu-
lar Biology and Evolution, 3:423—434.

AVISE, J. C., 1986, Mitochondrial DNA and the ev-
olutionary genetics of higher animals. Philosoph-
ical Transactions of the Royal Society of London,
Ser. B, 312:325-342.

BANTOCK, С. R., 1974, Experimental evidence for
non-visual selection in Capaea nemoralis. He-
redity, 33:409—437.

ВЕМТ, А. С., 1949, Life histories of North American
thrushes, Kinglets, and their allies. Smithsonian
Institution, Washington, D.C., United States Na-
tional Museum Bulletin, 196.

BIBB, M. J., VAN ETTEN, R. A., WRIGHT, C. T.,
WALBERG, M. W. & CLAYTON, D. A., 1981, Se-
quence and gene organization of mouse mito-
chondrial DNA. Cell, 26:167-180.

BROOKE, J. M., 1897, The colony of Helix nemor-
alis at Lexington, Virginia. The Nautilus, 10:142—
143.

BRUSSARD, P. F., 1975, Geographic variation in
North American colonies of Cepaea nemoralis.
Evolution, 29:402-410.

CAIN, A. J., 1968, Studies on Cepaea V. Sand-
dune populations of Cepaea nemoralis (L.).
Philosophical Transactions of the Royal Society
of London, Ser. В, 253:499-519.

CHAPMAN, R. W., 1987, Changes in the popula-
tion structure of male striped bass, Morone sax-
atilis spawning in three areas of the Chesapeake

CEPAEA IN LEXINGTON 315

Bay from 1984 to 1986. Fishery Bulletin, 85:167—
170.

CHAPMAN, В. W. & POWERS, D. A., 1984, A
method for the rapid isolation of mitochondrial
DNA from fishes. Maryland Sea Grant Technical
Report, Publication Number UM-SG-TS-84—05.

CLARKE, B. & MURRAY, J., 1962, Changes of
Gene Frequency in Cepaea nemoralis (L.): the
estimation of selective values. Heredity, 17:467—
476.

CLARY, D. O. & WOLSTENHOLME, D. R., 1985,
The mitochondrial DNA molecule of Drosophila
yakuba: Nucleotide sequence, gene organization
and genetic code. Journal of Molecular Evolu-
tion, 22:252-271.

COCKERELL, T. D. A., 1889, The Virginia colony of
Helix nemoralis. The Nautilus, 3:73—77.

COCKERELL, T. D. A., 1894, The Virginia colony of
Helix nemoralis. The Nautilus, 8:92-95.

COCKERELL, T. D. A., 1897, The Virginia colony of
Helix nemoralis. Science, 5:985—986.

DAWID, 1. В. 8 BROWN, D. D., 1970, The mito-
chondrial and ribosomal DNA components of
oocytes of Urechis caupo. Developmental Biol-
ogy, 22:1-14.

FALCONER, D. S., 1976, Quantitative Genetics,
The Ronald Press, New York.

HARRISON, R. G., RAND, D. M. & WHEELER, W.
C., 1985, Mitochondrial DNA size variation within
individual crickets. Science, 228:1446-1447.

HEATH, D. J., 1975, Colour, sunlight and internal
temperatures in the land-snail Cepaea nemoralis
(L.). Oecologia, 19:29-38.

HOWE, J. L., 1898, Variation in the shell of the
Helix nemoralis in the Lexington, Va. colony.
American Naturalist, 32:913-923.

HOWLEY, P. M., ISRAEL, M. A., LAW, M. F. &
MARTIN, M. A., 1979. A rapid method for detect-
ing and mapping homology between heterolo-
gous DNAs. Journal of Biological Chemistry, 254:
4876—4883.

JOHNSON, М. S., 1976, Allozymes and area ef-
fects in Cepaea nemoralis on the western Berk-
shire Downs. Heredity, 36:105-121.

JOHNSON, M. S., 1979, Inheritance and geo-
graphic variation of allozymes in Cepaea nemor-
alis. Heredity, 43:137-141.

JOHNSON, M. S., STINE, O. C. & MURRAY, J.,
1984, Reproductive compatibility despite large-
scale genetic divergence in Cepaea nemoralis
(L.). Heredity, 53:655-665.

JONES, J. S., LEITH, B. H. & RAWLINGS, P.,
1977, Polymorphism in Cepaea: A problem with
too many solutions. Annual Review of Ecology
and Systematics, 8:109-143.

LANSMAN, R. A., SHADE, В. O., SHAPIRA, J. Е. &
AVISE, J. C., 1981, The use of restriction endo-
nucleases to measure mitochondrial DNA se-
quence relatedness in natural populations Ill.
Techniques and potential applications. Journal of
Molecular Evolution, 17:214-226.

MANIATIS, T., FRITSCH, E. Е. 8 SAMBROOK, J.,
1982, Molecular Cloning: A Laboratory Manual.
Coldspring Harbor Press, New York.

MCCONNELL, D., 1935, Changes in the frequen-
cies of the variations of Cepaea nemoralis
(Linne). American Naturalist, 64:614-620.

MURRAY, J., 1964, Multiple mating and effective
population size in Cepaea nemoralis. Evolution,
18:283-289.

MURRAY, J., 1975, The genetics of the mollusca.
Pp 3:3-31, in: R.C. King, ed., Handbook of Ge-
netics. Plenum Press, New York.

MURRAY, J. & CLARKE, B., 1978, Changes of
gene frequency in Cepaea nemoralis over 50
years. Malacologia, 17:317-330.

NEI, M., 1972, Genetic distance between popula-
tions. American Naturalist, 106:283-292.

ОСНМАМ, H., JONES, J. S. & SELANDER, В. K.,
1983, Molecular area effects in Cepaea. Pro-
ceedings of the National Academy of Sciences,
U.S.A., 80:4189-4193.

RICHARDS, A. V. & MURRAY, J., 1975, The rela-
tion of phenotype to habitat in an introduced col-
ony of Cepaea nemoralis. Heredity, 34:128-131.

RICHARDSON, A. M. M., 1974, Differential climatic
selection in natural populations of the land snail
Cepaea nemoralis. Nature, 247:572-573.

ROBERTS, J. W., GRULA, J. W., POSANKONY, J.
W., HUDSPETH, R., DAVIDSON, E. H. 8 BRIT-
TEN, R. J., 1983, Comparison of sea urchin and
human mitochondrial DNA: Evolutionary rear-
rangement. Proceedings of the National Acad-
emy of Sciences, U.S.A., 80:4614-4618.

SACCHI, С. Е. & VALLI, G., 1975, Recherches sur
lécologie des populations naturelles de Cepaea
nemoralis (L.) (Gastr., Pulmonata) en Lombardie
meridionale. Archives de Zoologie Experimentale
et Generale, 116:549-578.

SOLIGNAC, M., GENERMONT, J., MONNEROT,
М. & MOUNOLOU, J.-C., 1984. Genetics of mi-
tochondria in Drosophila: Mitochondrial DNA
inheritance in heteroplasmic strains of D. mauri-
папа. Molecular and General Genetics, 197:183-—
188.

STINE, O. C., 1986, The Origin of Cepaea nemor-
alis in Lexington, Virginia: Mitochondrial DNA
and Climatic Selection. Ph.D. Dissertation; Uni-
versity of Virginia.

TAYLOR, J. W., 1914, Monograph of the Land and
Freshwater Mollusca of the British Isles. Taylor
Brothers, Leeds, England.

WEBB, С. R., 1952, Beneficial—Cepaea nemoralis
8 cardinals. Gastropodia, 1:7-8.

WOLSTENHOLME, D. R., MACFARLANE, J.L.,
OKIMOTO, R., CLARY, D. O. & WAHLEITHNER,
J. A., 1987, Bizarre tRNAs inferred from DNA
sequences of mitochondrial genomes of nema-
tode worms. Proceedings of the National Acad-
emy of Sciences, U.S.A., 84:1324-1328.

Revised Ms. accepted 1 Dec. 1988

nis

u FF en

MALACOLOGIA, 1989, 30(1-2): 317-324

BIOLUMINESCENCE IN THE TERRESTRIAL SNAIL Dyakia (Quantula) striata

Jonathan Copeland'

Maryellen Maneri Daston?

ABSTRACT

The occurrence and function of bioluminescence in the terrestrial snail Dyakia (Quantula)
striata was studied. Luminescence was produced by intermittent flashes (glows) from a discrete
luminescent organ located in the anterior head-foot. A time-budget ethogram showed that the
snails flashed most during locomotion, about half of the time when feeding, and never while at
rest. When tested with a conspecific flashing snail, flash rate increased. When tested with a
counterfeit flash, flash rate again increased. These observations are consistent with а commu-
nicative function for flashing in D. striata. We suggest that flashing promotes congregation, a

behavior that requires locomotion.
Key Words:

INTRODUCTION

Bioluminescence occurs throughout the an-
imal kingdom, more commonly in marine than
in terrestrial environments (reviewed in Her-
ring, 1978; Johnson & Haneda, 1966; Harvey,
1952). Regardless of where it is found, how it
is produced, or how it is controlled, the adap-
tive significance of bioluminescence is usually
unknown (Buck, 1978).

Buck (1978) proposed that the adaptive
significance of bioluminescence could be
reduced to four functional categories: (1) emit-
ter-limited behavior (food getting and defense,
including jamming, concealment, repelling,
sacrifice lure, camouflage, and warning); (2)
illumination; (3) intraspecific communication
for congregation or mate attraction; and (4)
interspecific adaptation, e.g. aggressive mim-
icry.

In mollusks, bioluminescence occurs in
cephalopods, bivalves, and gastropods (re-
viewed in Tsuji, 1983). The world's only
known terrestrial bioluminescent mollusk is
the pulmonate snail Dyakia (Quantula) striata
(Godwin-Austen, 1891), found throughout
tropical Southeast Asia (Haneda, 1981).° The
flash of D. striata is not bright: it is yellow-
green in color, 0.5-6.0 seconds in duration,
with a variable interval between flashes, and

terrestrial snail, behavior, bioluminescence, bioluminescent communication.

some flashes with multiple peaks of intensity
(Copeland & Maneri, 1984). Flashes are emit-
ted from a luminescent organ embedded in
the head ventral to the buccal orifice.

The occurrence of bioluminescence in D.
Striata was first reported in 1942 by Dr. Yata
Haneda (Haneda, 1946), then head of the Zo-
ology Section at the Raffles Museum in Sin-
gapore (Corner, 1981). Since 1942, Haneda
and others have reported on different aspects
of D. striata's natural history (Haneda, 1955,
1963, 1981; Haneda & Tsuji, 1969; Parmen-
tier & Barnes, 1975) and the light microscopy
of the luminescent organ (Bassot & Martoja,
1968; Martoja & Bassot, 1970). More re-
cently, Counsilman, Copeland, and their stu-
dents have carried out a number of different
studies on the behavior, physiology, and
anatomy of bioluminescence in this unique
snail (summarized in Counsilman & Cope-
land, 1986).

Despite many casual observations of the
circumstances under which flashing occurs in
D. striata, its function, be it behavioral or
physiological, has remained largely enig-
matic.

To learn more about the function of lumi-
nescence in D. striata, we made collections,
observations, and carried out experiments for
two months in various locations in the Repub-

‘Department of Biology, Swarthmore College, Swarthmore, Pennsylvania, U.S.A. 19081

“College of Medicine, University of Cincinnati, Cincinnati, Ohio, U.S.A. 45267

However, according to an anonymous reviewer, bioluminescence has been observed in the Tasmanian terrestrial pulmo-
nate Cystopelta by Ron Kershaw of the Queen Victoria Museum, Launceston.

(317)

318 COPELAND & DASTON

lic of Singapore and at National University of
Singapore. Here, we present the results of the
first long-term laboratory observations of the
bioluminescence of D. striata. A time budget
ethogram is presented, as is a description of
the response of flashing snails to flashing
conspecific snails and to counterfeit (artificial)
flashes.

METHODS

Observations were made in a darkroom
which we built within a temperature-controlled
laboratory at the Department of Zoology, Na-
tional University of Singapore.

Snails were maintained on a reversed light
cycle with darkness occurring at 8 AM and
lights-on occurring at 8 PM. We observed
consistent flash behavior after animals had
been in the lab for 3 days.

Animals were maintained individually or in
groups in small petri dishes or vials. They
were fed fresh lettuce, carrots, cucumber, or
apple and provided with water-impregnated
filter paper and a block of calcium carbonate.
Temperature was 26°C and relative humidity
was 95%.

Not all D. striata possess a luminescent or-
gan (Haneda, 1963, 1981; Haneda & Tsuji,
1969). We screened snails for the presence of
a luminescent organ by allowing them to
crawl on a plate of glass. When the anterior
foot was illuminated by a fluorescent-bulb
blacklight (NIS FL4BLB) (a technique sug-
gested to us by Dr. Ivan Polunin, a Singa-
porean naturalist who was aware that luciferin
produced a luminescence when illuminated

with uv light), autofluorescence could be in- |

duced from an oval patch of tissue embedded
in the animal’s anterior foot. This oval patch
was the luminescent organ, and it could be
seen through the animal’s translucent white
skin. All snails that possessed this inducible
autofluorescent patch also produced self-
generated flashes.

Flashes were recorded using a tripod-
mounted end window photomultiplier tube
(RCA 6655-A) which modulated a voltage
controlled oscillator (VCO) (A.R. Vetter, Inc.).
This signal was recorded on one channel of a
two channel portable tape recorder (Sony
WM-DG). Voice recorded behavioral obser-
vations were entered on the second tape re-
corder channel. The VCO signal was later
transcribed by playback through a demodula-
tor circuit into separate channels of a Grass

model 79B polygraph. Written tallies of flash-
es were also made simultaneously, as were
written notes. An illuminated digital stopwatch
provided the time base.

Behavioral Observations (Locomotion,
Feeding, Flashing) in a Simulated Natural
Environment

A time budget was constructed for eight in-
dividuals (four individuals with 16 mm aver-
age shell diameter (medium) and four larger
individuals with 23 mm average shell diame-
ter (large)). A focal individual sampling tech-
nique (Lehner, 1979) was used. Thus, the ac-
tivities of a single individual were recorded
during each five-minute sampling period. In-
termittent dim red illumination was provided
by a hand held electric torch covered with a
red filter (Edmunds Scientific, medium red).

These studies were carried out in a 20 gal-
lon glass aquarium covered with cheese
cloth. The aquarium was filled with sterilized
potting soil and had a single brick (5 cm x 5
cm x 20 cm) added as a shelter. Water and
food were changed daily and were present
during all tests. Individuals were marked on
the shell using colored nail polish.

To determine the behavioral context of the
D. striata's flashing, 30 hours of laboratory
observation were made during the dark por-
tion of the light cycle. During each sample
period such parameters as position of the ten-
tacles, degree of retraction, and the occur-
rence of locomotion or feeding were noted.
Encounters (defined below) with other ani-
mals and the number of flashes were also
recorded. Then, a time budget, based on per-
cent time spent in the various activities, was
constructed (Fig. 1). Since the medium snails
flashed more frequently and seemed to be
more active, a separate time budget was con-
structed for medium and large snails. How-
ever, when compared, there was no signifi-
cant difference (x? = 5.14, 3 d.f., ns) in the
amount of time devoted to each activity cate-
gory between large and medium snails.

Response to Stimulation with a
Conspecific Flashing Snail: Number of
Flashes and Movement

A small rectangular glass cage (30 cm x
20 cm x 10 cm) was placed on top of a num-
bered grid so that the position of the snail(s) in
the cage at the beginning of each five-minute
sample period could be noted and the number

BIOLUMINESCENCE IN DYAKIA (QUANTULA) STRIATA 319

MEDIUM SNAILS
(x shell diameter =16mm)

LARGE SNAILS
(x shell diameter =23mm)

0
227 22%
RESTING

no
f /
lash,

locomotion fully retracted

\
\ fully retracted no flashing

21%
LOCOMOTION

partially
retracted

26%

RESTING
> with
flashing

partially
retracted

Locomotion
with flashing

32%

TOTAL # FLASHES-1194 TOTAL + FLASHES=340

FIG. 1. Time budget ethogram for medium size (X shell diameter = 16 + 0.64 mm) and large snails (X shell
diameter = 23 + 1.8 mm). Each number represents the portion of the total time devoted to the different
behaviors of 4 medium and 4 large snails. Total observation time = 30 hours. Flashing almost always occurs

during locomotion.

of flashes recorded. A single flashing snail
(animal 1) was placed in the chamber and
observed for 30—40 minutes (pre-test). Then
a second flashing snail (animal 2) was added
(test) and the movements and number of
flashes of both snails were recorded for the
next 30-40 minutes. The second snail was
added to the opposite side of the chamber,
approximately 30 cm from where the first snail
was added. The actual distance between the
two snails varied, depending upon how much
the first snail moved during the pre-test.

Response to Stimulation with a Counterfeit
(Artificial) Flash: Number of Flashes

Single animals were placed in the plastic
vials and allowed to climb continuously during
each ten-minute trial. Each trial consisted of
five minutes control followed by five minutes
during which one counterfeit flash (0.2-1.0
second duration) was presented every 30
seconds. The number of D. striata flashes
was counted during each period. Time be-
tween trials was 8-15 minutes. Counterfeit
flashes were delivered via a small white in-
candescent bulb mounted on a small hand-
held, hand-activated torch.

RESULTS
Behavioral Observations: Time Budget

The observed behavior fell into three cate-
gories: resting, locomotion, and feeding (Fig.
1). These three categories occurred with or
without flashing.

Animals considered at rest had the tenta-
cles retracted and foot either partially or fully
retracted into the shell. The snails in this
study spent most of the time (41% medium,
61% large) at rest (Fig. 1). Animals consid-
ered in locomotion always had the foot fully
extended, the tentacles extended, and were
moving. Much of the remainder of the obser-
vation time (33% medium, 27% large) was
spent in locomotion. Animals were consid-
ered to be feeding when the mouth was in
contact with food. Here, only a small amount
of time was spent in feeding (12%).

Flashing almost always occurred with loco-
motion (95% medium, 77% large), about half
of the time with feeding (46% medium, 41%
large), and almost never occurred while at
rest (7% medium, 0% large) (Fig. 1). Only me-
dium sized D. striata would flash with the
body stationary but with head and tentacles

320 COPELAND & DASTON

moving. Stationary large D. striata never
flashed.

During the 30 hours of behavioral observa-
tion there were ten instances in which a snail
came into physical contact with another snail.
We called this contact an encounter. The sig-
nificance of these encounters is unknown.
Encounters consisted of physical contact be-
tween animals and, prior to physical contact,
orientation and locomotion towards another
animal. Whether or not luminescence plays a
role in stimulating or prolonging an encounter
is not known. However, during these behav-
ioral observations, flashing occurred in all but
one of the ten encounters. The distance be-
tween snails when an encounter began was
not recorded, nor could we determine, by ob-
servations alone, that the eyes were involved.
A typical encounter started with an approach,
as both snails crawled toward each other or
one remained stationary while the other
crawled toward it. On meeting, they would
touch tentacles and then crawl past or over
each other. In one case of unusually long du-
ration, a snail followed another for 30 minutes
with its head touching the posterior tip of the
other snail's foot. Afterward, they turned fac-
ing each other and they remained in contact
for six additional minutes. Both snails flashed
throughout the entire 36-minute encounter.
Encounters always occurred between two an-
imals of the same size class because obser-
vations of the two size classes were made
separately.

In general, as flashing individuals crawled
toward each other, flashes became more reg-
ular in frequency. Both alternation of flashes
between individuals and simultaneous flash-
ing occurred.

Response to Flashing Conspecific Snail or
Counterfeit (Artificial) Flash

Because we observed changes in flashing
during encounters, and because we also ob-
served that in a cage with no flashing occur-
ring, the flash of one animal was often fol-
lowed by the flashes of several other animals,
we thought that flash alone might be sufficient
to influence the number of flashes and move-
ment of D. striata. Thus, to more systemati-
cally view encounters, we simplified the con-
ditions. Hence, in one type of test we used a
flashing conspecific snail as a stimulus and in
a second type of test we used a counterfeit
(artificial) flash as a stimulus.

1. Effects of Conspecific Flashing D. striata
on Flashing and Movement

In toto, eight experiments were done, but
animals remained active throughout the test
period in only five cases. A representative ex-
ample is presented (Fig. 2).

The number of flashes produced by the test
animal during the 35-minute pre-test period
was fairly constant (ranging from 5 to 11
flashes per 5 minute period). At the beginning
of sample period eight (SP8), the second
flashing snail was added to the test chamber.
Between SP8 and SP9 the distance between
the two animals decreased and the number of
flashes produced by both animals increased
approximately 2.5 times as they crawled to-
ward each other (but did not meet). Between
SP10 and SP11, and SP11 and SP12, the
distance between the animals decreased and
the number of flashes produced by animal 1
again increased from 7 to 18 flashes per five-
minute period. During SP12 through SP15,
the animals were close together, one moving
after the other, and the number of flashes pro-
duced by animal 1 decreased from 18 to 7
flashes per five-minute period.

When the number of flashes produced by
animal 1 and the distance traveled were com-
pared during the pre-test and the test period,
the number of flashes showed a significant
increase during the test period (t = 2.04, 7
d.f., p<0.05). However, the distance traveled,
although oriented toward animal 2, showed
no significant change (t = 1.5, 7 d.f., n.s.).
(The results in the other four experiments
were consistent with these findings. Addition-
ally, in other experiments (not reported here)
D. striata consistently oriented or moved to-
ward a [counterfeit] flashing light.)

2. Counterfeit Flash: Number of Flashes

Exposure to counterfeit flashes significantly
elevated the number of flashes produced by
D. striata (Randomization Test for Matched
Pairs, p<0.05) (Fig. 3 А-С). No consistent
delay occurred between a counterfeit flash
and the next snail flash, as would have been
the case if increased number of flashes were
caused by simple stimulus-response cou-

pling.
DISCUSSION

The terrestrial mollusk Umwelt is thought to
be mediated primarily by chemo- and mech-

BIOLUMINESCENCE IN DYAKIA (QUANTULA) STRIATA 321

Pre-Trial

Number
of
Flashes

ELE Mom 16:17" Bin +9

10% 195125183: 14. 15

Sample Period Number

@—e Animal 1

o---e Animal 2

Х. = mean # of flashes/5 min. period

FIG. 2. Number of flashes produced by animal 1 during pre-trial and animals 1 and 2 during trial period. Each
sample was 5 minutes. Animal 2 was introduced into test cage at beginning of sample period 8. X = number
of flashes produced per five-minute period of animal 1 (Xz) and animal 2 (X;). During the trial period, the

number of flashes increased.

anosensory cues and, to a lesser extent, by
photosensory cues (Fretter & Peake, 1975).
Furthermore, in terms of animal communica-
tion, gastropods are thought to use only
chemical and mechanical senses (Fretter €
Peake, 1975). Recently, however, the possi-
bility of form vision in snails has been ex-
plored (Hamilton & Winter, 1982, 1984). Fur-
thermore, Frings 8 Frings (1968) speculated
that, given the simple molluscan eye and the
occurrence of bioluminescence in gastro-
pods, if visual communication were ever to be
found in non-cephalopod mollusks, biolumi-
nescent communication would be a likely
form.

Animal communication has been defined
as occurring when the behavior of one animal
influences the behavior of a second animal
and the outcome is adaptive to both (Wilson,
1975).

All of our observations and experiments to
date are consistent with the notion that the
flash of D. striata is involved in animal com-
munication, i.e. the flashing of one D. striata
influences the flashing and possibly the
movement of a second snail. For example,
bioluminescence usually occurred during lo-
comotion (Fig. 1). During an encounter, which
also occurred with locomotion, animals
moved toward each other and increased their
flash rate (Fig. 2). When a counterfeit flash
was substituted for a flashing conspecific
snail, the flash rate increased (Fig. 3). Thus,
we suspect that the flash alone is sufficient to
affect behavior (flashing and movement) and,
furthermore, speculate that the flash is in-
volved in intraspecific communication.

The intent of our study in Singapore was to
examine on the enigmatic function of biolumi-
nescence in D. striata. Except for the possi-

322 COPELAND & DASTON

28 28
20 20
12 12
4 2

19)
Lu
5
< AE IR EE
3 =". - ar
LL
LL
O
cc С D
= 28 28
= 20 20
< 12 12
4 4
ео AA 003

TRIAL NUMBER

E No Counterfeit Flash Present

[] Counterfeit Flash Present

FIG. 3. Effect of counterfeit flashes on number of flashes produced by four different snails (A-D). Animal В
was tested on two consecutive days. Counterfeit flashes elevated the number of flashes produced by а snail.
A, p < 0.001; B, p < 0.01; C, p < 0.05; D, p < 0.07, but sample size was small; Fisher Randomization

matched pair t-test.

bility of intraspecific communication, we could
find no evidence to support any other of
Buck's (1978) four behavioral functions pos-
ited for bioluminescence. Prey attraction
seems an unlikely function because D. striata

feeds on vegetation and already dead ani- .

mals. Defense against predation was ruled
out because D. striata does not respond to
mechanical disturbance by flashing, but
rather by withdrawing into its shell. Also, D.
striata's flash is weak and intermittent and
therefore would not be a good source of illu-
mination.

Thus, we are left with intraspecific commu-
nication as the most likely function for biolu-
minescence in D. striata and this raises the
further question of, if this is flash communica-
tion, then what is being communicated?

Bioluminescence occurs in individuals of all
size classes. Since we often found snails in
the field lying close together (within a few cm),
and flashing snails often moved toward each
other in our behavioral studies, we suggest
that bioluminescent communication in D. stri-

ata might facilitate congregation (aggrega-
tion), something that involves locomotion.
Congregation might serve many different
functions, such as an aid to finding a mate,
food, or protection from desiccation.

The fact that bioluminescence occurs in
snails of all sizes may provide a significant
clue as to its function. Bioluminescence in D.
striata has previously been considered to oc-
cur only in juvenile snails (reviewed in
Haneda, 1981). However, Haneda reached
this conclusion on the basis of his own work,
where snails were considered to be juvenile
due to their small size (Haneda, 1963;
Haneda & Tsuji, 1969), and from the work of
Bassot and Martoja (Bassot & Martoja, 1968;
Martoja & Bassot, 1970). This latter work
used the techniques of light microscopy to de-
scribe how the luminescent organ disap-
peared and was replaced by a resorption cyst
before the (histological) maturation of the go-
nads.

More recent evidence suggests that the lu-
minescent organ occurs in sexually mature

BIOLUMINESCENCE IN DYAKIA (QUANTULA) STRIATA 323

snails (Maneri, 1985; Counsilman et al., 1987;
Counsilman, personal communication). Fur-
thermore, snails of all sizes (small, medium,
and large) all possess the luminescent organ
(Copeland & Maneri, 1984; Counsilman et al.
1986; Counsilman, personal communication).
To date no study has occurred that corre-
lates size of the snail to size of the gonads
and the presence of mature gametes. Thus,
bioluminescence in D. striata could be asso-
ciated with mating behavior, as it is in numer-
ous other terrestrial forms (Buck, 1978).

The assumption underlying our work is that
all bioluminescence has an extant behavioral
function. However, this need not be the case.
For example, Seliger (1975) has argued that
some forms of bioluminescence may actually
be vestigial. He hypothesized that, in an
anaerobic environment, oxygen could have
been toxic and might have been de-toxified
via bioluminescence. Furthermore, Seliger
(1975) also argued that some forms of biolu-
minescence may presently be purely physio-
logical in function, with no behavioral mean-
ing whatsoever. Indeed, not all traits, be they
structural, physiological, or behavioral, need
have an extant adaptive function (Gould & Le-
owontin, 1979). Thus, the flashing of D. stri-
ata could be the result of some earlier adap-
tation in a precursor organism to which there
is, now, no selective pressure.

Still, our results show that flashing is co-
temporal with locomotion and that flashing,
be it from a conspecific snail or a counterfeit
source, can affect a second animal’s behav-
ior. These results make us suspect that a be-
havioral function is likely.

Thus, we think that the flashes of D. striata
are likely to be involved in an intraspecific bio-
luminescent communication system. Exactly
what is being communicated is now unclear,
but photic cues seem to be able to influence
the flashing of the snail.

ACKNOWLEDGEMENTS

We thank the National Geographic Society
(Grant 2716-83) for the support of this re-
search. We also thank Dr. George Daston, Dr.
Robin Curtin, Cynthia Vernon, and Don Dov-
ala for technical support, and Dr. Ivan and
Mrs. Siu Yin Polunin, Dr. T. J. Lam, J. J.
Counsilman, D. H. Murphy, and J. Sigurdson
for providing various facilities, good col-
leagueship, and good food during our stay
both at National University of Singapore and

at a “dream of a dream of a mountain
stream.”

LITERATURE CITED

BASSOT, J. M. & MARTOJA, M., 1968. Presence
d'un organe lumineux transitoire chez le gastero-
pode pulmone, Hemiplecta weinkauffiana
(Crosse et Fischer). Comptes Rendus des Seance
de l'Academi des Sciences, Paris, 266:1045—
1047.

BUCK, J. B., 1978. Functions and evolutions of bio-
luminescence. /n: Herring, P.J., ed. Biolumines-
cence in action, 419-460, Academic Press, New
York.

COPELAND, J. & MANERI, M., 1984. Biolumines-
cence and communication in the Terrestrial Snail
Dyakia (Quantula) striata. Society for Neuro-
science Abstracts, 10:396.

CORNER, E. J. H., 1981. The Marquis. A Tale of
Syonan-to, Heinemann, Singapore.

COUNSILMAN, J. J. & COPELAND, J., 1986. Neu-
roethological studies of behavior in the land snail
Dyakia (Quantula) striata. International Society
for Neurosciences Abstracts, 1.

COUNSILMAN, J. J., LOH, D., CHAN, S. Y., TAN,
W. H., COPELAND, J., & MANERI, M., 1987.
Factors affecting the rate of flashing and loss of
luminescence in an Asian land snail, Dyakia stri-
ata. Veliger, 29:394-399.

FRETTER, V. & PEAKE, J. The Pulmonates, Vol.
1., Academic Press, New York.

FRINGS, H. & FRINGS, M., 1968. Other inverte-
brates. In: Sebeok, T. A., ed., Animal communi-
cation: techniques of study and results of re-
search, 244-270, Indiana Univ. Press,
Bloomington.

GOULD, S. J. & LEOWANTIN, R. C., 1979. The
spandrals of San Marco and the Panglossian
paradigm: a critique of the adaptationist pro-
gramme. Proceedings of the Royal Society of
London, 205:581-598.

HAMILTON, P. V. & WINTER, M. A., 1982. Behav-
ioral responses to visual stimuli by the snail Lit-
torina irrorata. Animal Behavior, 30:752—760.

HAMILTON, Р. V. & WINTER, M. A., 1984. Behav-
¡oral responses to visual stimuli by the snails Tec-
tarius muricatus, Turbo castanea, and Helix as-
persa. Animal Behavior, 32:51-57.

HANEDA, Y., 1946. A luminous land snail, Dyakia
striata, found in Malaya. Seibutsu, 1:294—298.
HANEDA, Y., 1955. Luminous organisms of Japan
and the Far East. In: Johnson, F. H., ed. Lumi-
nescence of Biological Systems, 253-254.

AASS, Washington, D.C.

HANEDA, Y., 1963. Further studies on a luminous
land snail, Quantula striata, in Malaya. Science
Report of the Yokosuka City Museum, 8:1-9.

HANEDA, Y., 1981. Luminous activity of the land
snail Quantula striata. In: DeLuca, М. A. 4 McEl-
roy, W. D., Bioluminescence and Chemilumines-
cence, 257-265, Academic Press. New York.

324 COPELAND & DASTON

HANEDA, Y. & TSUJI, Е. 1., 1969. Observations on
the luminescence of the land snail Quantula stri-
ata, and its life history. Science Report of the
Yokosuka City Museum, 15:10-13.

HARVEY, Е. М., 1952. Bioluminescence. Academic
Press, New York.

HERRING, P. J., ed., 1978. Bioluminescence in Ac-
tion. Academic Press, New York.

JOHNSON, Е. H. 8 HANEDA, Y., eds., 1966. Bi-
oluminescence in Progress, Princeton University
Press, Princeton.

LEHNER, P. N., 1979. Handbook of Ethological
Methods, Garland, New York. :

MARTOJA, M. & BASSOT, J. M., 1970. Etude his-
tologique du complexe glandulaire pedieux de
Dyakia striata, Godwin et Austen, gasteropode
pulmone données sur l'organe lumineux, Vie et
milieu, Serie A: Biologie Maine Tome XXI-Fasc.
2-A:395-452.

MANERI, М., 1985. Bioluminescence and sexual
maturity in the terresterial snail Dyakia striata.
Masters thesis, University of Wisconsin, Milwau-
kee.

PARMENTIER, J. 8 BARNES, A., 1975. Observa-
tions on the luminescence production by the Ma-
layan gastropod Dyakia striata. Malayan Nature
Journal, 28:173-180.

SELIGER, H.H., 1975. The origin of bioluminesce.
Photochemistry and Photobiology, 21:355-361.

TSUJI, F.l., 1983. Molluscan bioluminescence. In:
Saleuddin, A.S.M. & Wilbur, K.M., eds. The Mol-
lusca, Vol. 2, 257-279, Academic Press, New
York.

WILSON, E.O., 1975. Sociobiology. Harvard Uni-
versity Press, Cambridge.

Revised Ms. accepted 2 December 1988

MALACOLOGIA, 1989, 30(1-2): 325-339

A MECHANISM FOR THE CREATION OF CONJOINED TWINNING IN
LEHMANNIA VALENTIANA PRESENT IN THE PRIMARY OOCYTE
(GASTROPODA, PULMONATA)

Jeanine Mason! & Jonathan Copeland?

ABSTRACT

Lehmannia valentiana, a terrestrial mollusk, naturally produces viable conjoined twins at a
greater than 1.0% rate; but the closely related species Limax maximus and Deroceras reticu-
latum do not.

Using electron microscopy, we examined the fertilized, uncleaved zygote of all three species.
In addition, we used both light and electron microscopy to determine at what point, from primary
oocyte to oviposited fertilized zygote, conjoined twinning occurred.

Joined zygotes/ova were found in the oviduct, albumen gland and fertilization chamber in L.
valentiana. (They were not found in the hermaphroditic duct.) The junction involved cytoplasmic
continuity.

Others have suggested that conjoined twinning could be caused by a defective or missing
vitelline membrane which facilitated the fusion of zygotes or ova. However, no differences could
be found after oviposition in the vitelline membrane of L. valentiana (conjoined twinning occurs)
and L. maximus or D. reticulatum (conjoined twinning does not occur).

Of the possibilities examined, two viable mechanisms may underly conjoined twinning in L.
valentiana. One is random fusion of two or more zygotes in one egg capsule. A second is the
central constriction of primary oocytes. Since we found centrally constricted primary oocytes,
oocyte cell clusters and amoeboid-like primary oocytes, we suspect that central constriction is
the principal mechanism underlying conjoined twinning in L. valentiana.

(Key Words: Lehmannia valentiana; Limax maximus; Deroceras reticulatum; conjoined twins;

terata; reproduction; oocyte/ovum; vitelline membrane)

INTRODUCTION

Studies of the possible causal mechanisms
underlying conjoined twinning in mollusks and
other animals have been hampered by the
relatively infrequent occurrence of this terat-
ism. We found viable conjoined twinning oc-
curring in the terrestrial mollusk Lehmannia
valentiana (Férussac) at a rate greater than
1% (Mason & Copeland, 1988). Conjoined
twinning did not occur in the closely related
species Limax maximus Linnaeus and Dero-
ceras reticulatum Múller during the study.

Others (Crabb, 1931; George, 1958; Bigus,
1981) have attributed conjoined twinning in
mollusks to a fusion of multiple zygotes or
embryos in the egg capsule after oviposition.
For example, N. H. Verdonk (personal com-
munication, 1982, 1983) suggested that if fu-
sion were the cause of mollusk conjoined
twinning, it might be possible that the vitelline
membrane was defective or missing.

However, in addition to fusion after ovipo-
sition, incomplete fission of egg cells could
also produce conjoined twins. By examining
the egg cell at several locations in the repro-
ductive tract in all three species, we thought a
possible mechanism for conjoined twinning
might be found.

Based on the frequent asymetrical mor-
phology ofthe conjoined twins (Mason 8 Cope-
land, 1988) and the data presented here, we
think that the mechanism is incomplete fission
rather than random fusion.

MATERIAL AND METHODS

All animals were maintained according to
Reingold & Gelperin (1980) except that Te-
gosept M (a mold inhibitor) was deleted from
the diet and Gerber baby food green beans
and fresh lettuce were added to the diet.

Oviposited zygotes

Egg capsules were collected immediately
after oviposition and examined using a dis-

'W297 №3020 Oakwood Grove Road Pewaukee, Wisconsin, U.S.A. 53072. Correspondence should be directed to this

address.

“Department of Biology Swarthmore College Swarthmore, Pennsylvania, U.S.A. 19081
| (325)

326 MASON & COPELAND

secting microscope (25-30X) to ensure that
cleavage had not begun. Capsules containing
zygotes to be studied with the electron micro-
scope were immersed in 3.0% glutaraldehyde
buffered with Millonig's phosphate or Sym-
Collidine at pH 7.6. Excess albumen sur-
rounding the zygotes was removed manually
under a dissecting microscope after two
hours, and the zygotes were re-immersed in
glutaraldehyde solution overnight at 5°C. This
procedure was used to maintain the integrity
of any cortex membranes, especially the vi-
telline membrane.?

The method of Berg (1967), which removes
zygotes/ova from the surrounding egg cap-
sule and albumen prior to fixation, was not
used because of the possibility of disrupting
the vitelline membrane and of separating
joined zygotes.

Following a buffered rinse, zygotes were
post-fixed in 1.0% osmium tetroxide for one
hour at room temperature, dehydrated in a
graded ethanol series and infiltrated and poly-
merized in Spurr Low-Viscosity Embedding
Media. Thin sectioned specimens were
stained with 5.0% uranyl acetate and Rey-
nolds' lead citrate (Reynolds, 1963). Sections
were examined and photomicrographs taken
with a Hitachi HU 11B transmission electron
microscope.

At separate intervals, 11 samples were car-
ried through the fixation process. In total, the
number of egg capsules were as follows:

п Species
43 L. valentiana

Morphology

Each egg capsule contained
two zygotes; no observable
gap between zygotes when
viewed at 30X*

28 L. valentiana Each egg capsule contained

one zygote

20 L. maximus Each egg capsule contained
one zygote

20 D. reticulatum Each egg capsule contained
one zygote

Ova/oocytes in the reproductive tract

To obtain ova/oocytes from the reproduc-
tive tract, we fixed animals that had just be-

gun ovipositing. We injected ovipositing
slugs, using a hypodermic needle containing
10 ml of glutaraldehyde (for EM) directly into
the open pneumostome, or 10 ml of Bouin-
Duboscgq (alcoholic) fixative (Humason, 1979)
for light microscopy. After injection, additional
glutaraldehyde or Bouin-Duboscq fixative
was injected through the body wall in the vi-
cinity of the fertilization chamber. The repro-
ductive tract was then dissected out of the
animal. Four L. valentiana ova-filled fertiliza-
tion chambers and four ovotestes were pre-
pared for electron microscopy using the meth-
ods previously detailed.

For light microscopy, part of the oviduct and
albumen gland proximal to the fertilization
chamber were retained, as well as the fertili-
zation chamber, hermaphroditic duct and
ovotestis. These structures were held in
Bouin-Duboscq fixative overnight followed by
a graded ethanol dehydration. Infiltration and
embedding was in Paraplast. Sections (10
um) were stained with Weigert lron Hematox-
ylin (Lillie Modification 1968) and counter-
stained with eosin (Humason, 1979).

Reproductive structures for light microscopy
were obtained from the following animals:

п Species Reproductive Status
33 L. valentiana ovipositing
20 L. valentiana mature, non-ovipositing
3 D. reticulatum ovipositing
3 D. reticulatum mature, non-ovipositing
3 L. maximus ovipositing
3 L. maximus mature, non-ovipositing

Every section of the fertilization chambers,
hermaphroditic ducts and ovotestis was ex-
amined microscopically to look for ovum/
oocyte fusion or fission. Each tenth section
was viewed at 400X magnification.

RESULTS

The reproductive tract of terrestrial mol-
lusks consists of the ovotestis (ovt), hermaph-

“Egg cells, like other animal somatic cells, are bounded externally by a cytoplasmic (plasma) membrane. This basic cell
membrane is an integral part of the cell and cannot be removed from the rest of the cell without causing immediate cytolysis.

In addition to the plasma membrane, egg cells may have primary, secondary and/or tertiary membranes enveloping the
cortex. The primary membrane, formed in the ovary by the egg cell, is generally known as the vitelline membrane (Raven,

1961), and 1$ the one which we investigated.

“Many egg capsules contain zygotes that are close, but not all become conjoined twins; many just become two separate
embryos in one capsule. We know, by observation, that unless zygotes are very close (no observable separation viewed
microscopically) at the time of oviposition, they will not become conjoined twins. No single egg divides into two zygotes after
oviposition. No two zygotes with an observable gap between them become conjoined twins.

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 327

JEC

Pr,
u.
er,
ere
| e

FIG. 1: Simplified portion of reproductive system of terrestrial mollusks L. valentiana, L. maximus, D.
reticulatum and the sites where joined egg cells (JEC) were found in L. valentiana. ovt = ovotestis (gonad);
hd = hermaphroditic duct; fc = fertilization chamber: ag = albumen gland; od = oviduct.

328 MASON & COPELAND

FIGS. 2-4: 2: Outer cortex of Limax maximus uncleaved fertilized zygote after oviposition. o = zygote;
arrow = membrane fragments. 3: Outer cortex of Deroceras reticulatum uncleaved fertilized zygote after
oviposition. o = zygote; arrow = membrane fragments. 4: Outer cortex of uncleaved fertilized zygote after
oviposition of one of a pair of Lehmannia valentiana; two zygotes were close in one egg capsule. o = zygote;
arrows = membrane fragments being extruded.

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 329

FIGS. 5-8: 5: L. valentiana ovum in fertilization chamber. o — ovum; arrow = dark-staining cortex beads.
6: Higher magnification of L. valentiana ovum in fertilization chamber. о = ovum; arrow = dark-staining
cortex beads. 7: Protrusion of dark-staining bead (arrow) on cortex of L. valentiana ovum in fertilization
chamber. o = ovum. 8: Vacuolation of L. valentiana ovum cortex (boxed arrows) in fertilization chamber

surrounded by cross-sectioned spermatozoa. o = ovum.

330 MASON & COPELAND

FIGS. 9-12: 9: Extrusion of dark-staining beads at L. valentiana cortex indicative of decidual reaction.
boxed arrows = extrusion of beads; o = ovum. 10: Saggital sectioned spermatozoan juxtaposed to L.
valentiana cortex in fertilization chamber. arrow = apposition of spermatozoan membrane; о = ovum.
11: Saggital section of L. valentiana fertilization chamber with multiple ova in one chamber. ag = albumen
gland; fc = fertilization chamber; h = hermaphroditic duct; o = ova. 12: Cross section of L. valentiana
fertilization chamber. ag = albumen gland; fcl = fertilization chamber lumen; upper arrow = joined ova.

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 331

—10p m

E
= |
À

+ и a ыы $3

Е
=

FIGS. 13-18: 13: L. valentiana ovum/ova in fertilization chamber joined by cytoplasmic strand (arrow).
14: Higher magnification of joined ova/ovum shown in Figure 12. lower arrow = point of cytoplasmic
continuity; upper arrow = spindle apparatus of first maturation division forming. 15: L. valentiana ova/ovum
in fertilization chamber joined by cytoplasmic strand (arrow). 16: Vacuolated closely apposed Deroceras
reticulatum ova in fertilization chamber. v = vacuole. 17: Joined L. valentiana ova/ovum found in albumen
gland. asterisk = junction; left arrow = centrioles. 18: Fertilized uncleaved L. valentiana ova/ovum in
oviduct (at arrows).

332 MASON & COPELAND

FIGS. 19-22: 19: Cluster of oocytes in L. valentiana ovotestis. 20: Cluster of oocytes in L. valentiana
ovotestis. 21: Serial sections of L. valentiana constricted oocyte in ovotestis. 22: Serial sections of L.
valentiana constricted oocyte in ovotestis.

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 333

roditic duct (hd), fertilization chamber (fc), al-
bumen gland (ag) and oviduct (od) as
generalized in Figure 1. Both oocytes and
spermatozoa develop in the ovotestis and are
then released into the hermaphroditic duct.
Oocytes are fertilized in the fertilization cham-
ber, then move into the oviduct where they
are surrounded by albumen from the albumen
gland and encased in gelatinous capsules
prior to oviposition. The mature oocyte in the
ovotestis is in the germinal vesicle stage. In
the fertilization chamber, spindles and asters
of the first maturation are present, and the
ovum is surrounded by spermatozoa. The first
polar body is not extruded until after oviposi-
tion. Joined egg cells (JEC) were found in
each of the structures as indicated.

One possible mechanism suggested for
conjoined twinning was that a defective or
missing vitelline membrane might allow fusion
of two or more zygotes in the same egg cap-
sule. We found that all zygotes, L. maximus
(Fig. 2), D. reticulatum (Fig. 3), and L. valen-
tiana (Fig. 4), whether they were single or two
close zygotes in an egg capsule, had an outer
layer of vitelline membrane fragments sur-
rounding the plasma membrane when they
were examined after oviposition. The darker
stained granular substance into which the
fragments were released is the albumen or
perivitelline fluid. The micrograph of L. valen-
tiana (Fig. 4) shows one ovum of a pair of
close uncleaved zygotes.

Since no difference between the vitelline
membrane was seen between the three spe-
cies, we proceded to examine the egg cell
step-by-step upstream through the reproduc-
tive system.

Oviduct

Dissection of the oviduct in ovipositing L.
valentiana showed that the zygotes were
close or joined at this stage (Fig. 18). There-
fore, the mechanism would have to be further
up the reproductive tract.

Fertilization chamber

Electron micrographs of L. valentiana ova in
the fertilization chamber show that the vitelline
membrane has vacuolated and begun to lift off
the cortex at this stage (Figs. 5-10). Bead-like

(dark staining) structures appear along the
cortex of the ovum (Fig. 5). On closer exam-
ination (Fig. 6), the dark granules appear to be
enclosed in an outer layer of fibrous material.
Acondensation and protrusion of the bead-like
granules (Fig. 7) seems to precede a decidual
reaction (Fig. 8). Surrounded by cross-sec-
tioned spermatozoa, the granules are re-
leased as the outer envelope becomes vacu-
olated. A higher magnification of the cortex
(Fig. 9) confirms the evidence of granule
extrusion.° Multiple vacuoles appear on and
beneath the cortex. The axoneme of a sper-
matozoan and its encasing membrane lie jux-
taposed to the ovum (Fig. 10).

At light microscope magnification, a saggi-
tal section of a L. valentiana fertilization
chamber shows two distinct chambers (Fig.
11). The upper chamber contains close ova in
the first maturation division. A cross-section
of the L. valentiana shows what appears to be
joined ova (Fig. 12). At higher magnification
(Fig. 14), the ova are joined by a strand of
cytoplasm. This cytoplasmic continuity is also
apparent between other L. valentiana ova
(Figs. 13, 15) in the fertilization chamber.
Since there is cytoplasmic continuity between
these ova, no plasma membrane would be
present at the junction. Fusion of two oocytes
seems highly unlikely unless it were possible
for the plasma membrane to be disrupted
without cytolisis occurring.

We have not seen any other micrographs in
the literature of ova in the fertilization cham-
ber. The micrographs presented here demon-
strate the exocytosis of granules, vacuolation
and fragmentation of the vitelline membrane
in addition to showing that conjoined twins are
already determined at this stage.

D. reticulatum ova were occasionally found
close together in the fertilization chamber
(Fig. 16). The plasma membrane between the
ova was intact. The highly vacuolated condi-
tion of the cytoplasm indicates the ova are
degenerating. No close ova were found in L.
maximus.

Ova were found within the albumen gland
in all the species studied. Even here, we
found L. valentiana joined ova (Fig. 17).

Hermaphroditic duct

In both L. maximus and D. reticulatum,
oocytes were consistently found in the her-

5Whether a substance is moving in or out of a cell is always a difficult question to answer in EM. However, the granules can
be assumed to be moving outward because we see them enveloped in a fibrous material (Fig. 5—7) initially. These granules
are not present in electronmicrographs taken at oviposition, but are part of the remnants seen surrounding the egg cortex.
We also have no evidence of their being anywhere but in the egg cortex in the earliest micrographs.

334 MASON & COPELAND

de
2

ee, |

FIGS. 23-28: 23: Serial sections of L. valentiana constricted oocyte in ovotestis. 24: Serial sections of L.
valentiana oocyte cluster in ovotestis. arrow in fourth section = dividing line in largest oocyte. 25: L.
valentiana oocyte in ovotestis with bud-like protrusion (arrow). 26: L. valentiana oocyte in ovotestis with
constricted germinal vesicle (arrow). 27: L. valentiana partially constricted, amoeboid-like oocyte in ovotes-
tis. arrows = apposition to ovotestis membrane. 28: Most irregularly shaped oocyte found, in ovotestis
lumen of L. valentiana.

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 335

maphroditic duct in ovipositing animals. Her-
maphroditic ducts from 53 L. valentiana were
serially sectioned and examined, but oocytes
were found in the duct in only one specimen.
The number of ova in that specimen was less
than 10. Of the 53 specimens, 20 were from
non-ovipositing animals in which we would
not expect to find oocytes in the hermaphro-
dictic duct. Ova are not released until just be-
fore egg laying in /ncilaria (or Meghimatium)
bilineata (Ikeda, 1937) and in D. reticulatum
(Runham & Hunter, 1970). Of the 20 non-
ovipositing animals, 13 were from a colony of
slugs (all the same age) that was producing
more than 150 egg capsules per week.

Ovotestis

An abundance of oocyte clusters were
seen in the L. valentiana ovotestis, mostly in
reproductively mature but non-ovipositing an-
imals. In many cases, none of the individual
oocytes within the cluster had a visible germi-
nal vesicle (ascertained by tracing serial sec-
tions) (Figs. 19, 20). The size of the oocyte
and texture of the cytoplasm confirm that it is
an oocyte and not an accessory cell or sper-
matocyte. (Spermatids and spermatocytes
can be seen near the clusters.) The acini of
the ovotestis in D. reticulatum was similar to
that of L. valentiana; but similar clusters of
oocytes were not seen. (Individual oocytes
were not as close to each other.) L. maximus
acini were much smaller than either those of
L. valentiana or D. reticulatum and usually
contained only one mature oocyte.

We looked for two oocytes in the ovotestis
of L. valentiana that seemed to be less than 5
um apart microscopically, which might indi-
cate conjoined twinning. When two such
oocytes were identified (Figs. 21, 22, 23), we
traced them in the serial sections. We found
that only one germinal vesicle was present
and that in reality there was only one centrally
constricted oocyte present. These dumbell-
shaped, centrally constricted oocytes were
not found in L. maximus or D. reticulatum.
This L. valentiana structure could explain the
cytoplasmic continuity seen in the fertilization
chamber ova as well as the asymetrical mor-
phology of conjoined twins.

A similarity between the centrally con-
stricted oocytes and constituents of a cell
cluster were occasionally noted. The largest
part of the cluster in Figure 24 seems to be
split longitudinally (at arrow in fourth micro-
graph of the series); but no germinal vesicle is

visible. The majority of cell clusters show no
connection between constituents.

Isolated instances of two other peculiarities
were noted. An oocyte with a bud-like struc-
ture (at arrow) (Fig. 25) and an oocyte with an
indentation of a germinal vesicle (at arrow)
(Fig. 26) are shown.

Irregularly shaped oocytes (Figs. 27, 28)
were present in the L. valentiana ovotestis.
One oocyte resembles the centrally con-
stricted oocytes except that the constriction is
seen only on the side apposed to the acinar
membrane (Fig. 27). The most irregularly
shaped oocyte found (Fig. 28) was in the ac-
inar lumen, indicative of ovulation.

DISCUSSION
Role of the vitelline membrane

М. Н. Verdonk's suggestion (personal com-
munication, 1982, 1983) that a defective or
missing vitelline membrane might facilitate fu-
sion of independent egg cells and result in the
formation of conjoined twins was followed by
examining the vitelline membrane of L. valen-
tiana, D. reticulatum and L. maximus. In all
three species, the vitelline membrane was
fragmented and elevated above the plasma
membrane of the uncleaved zygote at the
time of oviposition (Figs. 2-4).

In Lymnaea (Lymnaea) stagnalis (Lin-
naeus, 1758) the embryo “. . . leaves the vi-
telline membrane at about 32 hrs after first
cleavage (25°C) . . .” and then begins to ro-
tate within the albumen filled capsule (N. H.
Verdonk, personal communication, 1983). We
found uncleaved zygotes collected after ovi-
position were rotating (turning around) at the
time the polar bodies were extruded (Mason
& Copeland, 1988). Our micrographs demon-
strate that vitelline membrane vacuolation
and fragmentation begins in the fertilization
chamber (Figs. 5-10). The dense granules
seem to be important in this procedure. In L.
stagnalis, the granules were still present in
the cortex at the time of oviposition (Luchtel,
1976). Longo (1976) stated that the function
of cortical granules in fertilization of spiralian
eggs was unknown.

The timing of vitelline membrane release in
our three species differs from that in L. stag-
nalis. We do not know what the significance of
this finding might be. But, if the vitelline mem-
brane prevents fusion prior to encapsulation
and rotation prevents fusion thereafter, then,
in our study, all three species are subject to
the same temporal events and should be sub-

336 MASON & COPELAND

ject to the same probability of fusion. But nei-
ther L. maximus nor D. reticulatum produced
conjoined twins.

Role of cytoplasmic continuity

In the fertilization chamber, light micros-
copy showed cytoplasmic continuity between
close ova (Figs. 12, 13, 14, 15). Fusion be-
tween two ova in the fertilization chamber
seems unlikely because (1) any fusion be-
tween the cortex of two ova would be dis-
rupted when the vitelline membrane vacuo-
lates and lifts off, and (2) cytoplasmic
continuity between close ova means no
plasma membrane or vitelline membrane is
present at that junction. If the plasma mem-
brane were absent prior to fusion, cytolysis of
the cell would occur. Cytolysis would be seen
microscopically as varying degrees of cyto-
plasmic granulation and vacuolization and nu-
clear pyknosis (Luchtel, 1972).

Although there is cytoplasmic continuity be-
tween two ova (Fig. 15), two sets of meiotic
structures are clearly visible. The egg cell
may be considered as one unit, arguable be-
cause there is cellular communication via the
cytoplasm or as two units, arguable because
of the presence of separate meiotic spindles
and asters.

The presence of cytoplasmic continuity
also explains the asymetry and irregularity
seen in the morphology of conjoined twins
(Mason & Copeland, 1988). In conjoined
twins and in separate twins developing from
very close zygotes, one is usually smaller
than the other—this difference can be noted
prior to cleavage.

The finding of ova in the albumen gland
simply demonstrates that some ova are
joined at this stage and that there is some
way for zygotes to enter the gland proper.
Whether this is a normal procedure or an ar-
tifact of fixation is unknown.

Hermaphroditic duct

Due to the small number of L. valentiana
oocytes found in the hermaphroditic duct, we
were unable to determine if oocytes were
joined at this stage.

Role of the ovotestis

We think the dumbell-shaped, centrally
constricted oocytes (Figs. 21, 22, 23) are
good candidates for conjoined twins. They
were not seen in L. maximus or D. reticula-
tum. They could explain the origin of the ova-
rian cytoplasmic continuity (Figs. 13, 14, 15)
seen in the fertilization chamber as the result
of incomplete fission rather than fusion. Addi-
tionally, since the two halves would have the
same genetics, there would be no problem
with incompatability during development.®

Centrally constricted ova might originate as
amoeboid oocytes in the process of ovulating
and freeing themselves from the acinar wall.
For some unknown reason, they fail to return
to a spherical form; for example, the two
halves may rotate around the central constric-
tion. In some cases, the constriction might
sever after fertilization resulting in the devel-
opment of a single slug with an amorphous
incomplete living tissue mass (sometimes
with eyes) as detailed previously (Mason &
Copeland, 1988).

Another possibility is that centrally con-
stricted oocytes are being viewed during a
temporal event that generally results in divi-
sion. We asked if an irregular-shaped, amoe-
boid oocyte (Fig. 26) could actually be a lon-
gitudinally sectioned cell cluster. Cell clusters
were predominantly seen in mature non-
ovipositing animals in which spherical mature
oocytes were also present (Figs. 19, 20, 24).

We visualized the result of cross-sectioning
an irregular-shaped oocyte (Fig. 29a). But
none of the cross-sections resembled a cell
cluster. If we hypothesize that the irregular-
shaped oocyte is seen in a temporal stage
leading to division (Fig. 29b) and then cross-
section it, we find that the cross-sections look
somewhat like a cell cluster. However, if we
were to longitudinally section a hypothetically
divided irregular-shaped oocyte, it would look
very similar to a cell cluster.

Factors that favor the premise that ir-
regular-shaped oocytes are temporal events
leading to division are (1) many more cell
clusters were found than irregular-shaped
oocytes (since sectioning was non-oriented,
the numbers should be equal), (2) there usu-
ally seemed to be complete separation be-

‘From our observations, all egg cells oviposited by one animal in a single clutch have identical body markings when they
develop, but different clutches vary. We suspect the genetics of a clutch may be identical, and therefore, compatible in any

case

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 337

FIG. 29: Hypothetical drawing of the results of cross-sectioning an irregular-shaped oocyte (e.g., Figure 28)

at (a); and cross-sectioning or saggitally-sectioning a divided oocyte at (b). Arrows indicate the results of
sections taken at six different intervals.

338 MASON & COPELAND

tween cell cluster constituents when traced
serially, and (3) frequently, no germinal vesi-
cle was present in the cell cluster constitu-
ents, whereas, it was prominent in the
irregular-shaped oocytes.

Ancel (1903) found oocytes in groups in the
acini of Helix (Helix) pomatia Linnaeus, 1758.
He claimed that one persisted and the others
degenerated or were utilized as nurse cells.
Stears (1974) stated that oocytes were al-
ways found singly in the L. valentiana ovotes-
tis. Cell clusters could be the terminal stage of
a temporal event in which irregular-shaped
oocytes divide. Such a mechanism might be
an economical method of conserving oocytes
that had matured but not been oviposited.

The premise that irregular-shaped oocytes
are amoeboid is supported by others.
Bretschneider & Raven (1948) found that pri-
mary sex cells and oogonia stages divide;
and that oocytes had “amoeboid motility” only
when they were 170 um?, a time preceding
the growth stage. Neither of these observa-
tions seem to apply, since we are concerned
with mature primary oocytes.

The presence of actin filaments in mature
oocytes has been shown (Saleuddin & Khan,
1981). Actin may be involved in amoeboid
movement as well as in cytokinesis and
cleavage furrow formation. Saleuddin & Khan
(1981) suggest that a brain extract ovulation
factor causes mature oocytes to become
amoeboid for the purpose of freeing them-
selves from the acinar wall. Their micro-
graphs look very similar to zygotes we ob-
served during the extrusion of polar bodies.
Wilson (1904) also noted that fertilized ova
become “... almost amoeboid...” at the time
of polar body extrusion. Other micrographs
(Saleuddin et al., 1983) resemble abnormal
zygotes with elongated processes frequently
seen in egg capsules containing multiple zy-
gotes. Carrick (1938) suggested that the ab-
normality may be an attempted “... incipient
division by budding off an elongate process.”
We know, therefore, that some oocytes do not
return to a spherical form, but they are usually
non-viable. Motility and the presence of actin
filaments in oocytes may enable an oocyte to
free itself from the acinar wall and/or it may
enable an oocyte to divide. We do not know
which of these activities centrally constricted
oocytes were involved in. But whether they
originate due to incomplete division or amoe-
boid movement, they seem to be the most
likely form of oocyte to develop into conjoined
twins.

ACKNOWLEDGMENTS

Sincere thanks to Dr. Eldon Warner, Profes-
sor Emeritus, of the University of Wisconsin-
Milwaukee for his advice on embryology and
reproduction; and to Marilyn Schaller of the
electron microscope laboratory at the Univer-
sity of Wisconsin-Milwaukee for her advice as
well as the provision of material and methods.
We thank Professor A. J. Cain for help with
nomenclatural problems.

The support of Sigma Xi, The Scientific Re-
search Society, is gratefully acknowledged.

We are grateful to N. H. Verdonk for his
advice and suggestions.

LITERATURE CITED

ANCEL, P., 1903, Histogenese et structure de la
glande hermaphrodite d'Helix pomatia (Linn.).
Archives de Biologie (Leige), 19: 389-632.

BERG, W. E., 1967, Some experimental techniques
for eggs and embryos of marine invertebrates. In:
WILT, Е. H. 8 WESSELLS, N. K., eds., Methods
in developmental biology, pp. 767-776. Thomas
Y. Crowell Co., New York.

BIGUS, L. von, 1981, Polyvitellinity and fusion of
germs in Physa acuta (Pulmonata, Basommato-
phora). Zoologische Jahrbúcher; Abteilung fúr
Anatomie und Отодете der Tiere, 105: 526—
550.

BRETSCHNEIDER, L. H. & RAVEN, C. P., 1948,
Structural and topochemical changes in the egg
cells of Limnaea stagnalis L. during oogenesis.
Archives Neelandaises de Zoologie, 10: 1-31.

CARRICK, R., 1938, The life-history and develop-
ment of Agriolimax agrestis L., the gray field slug.
Transactions of the Royal Society of Edinburgh,
59: 563—597.

CRABB, Е. D., 1931, The origin of independent and
of conjoined twins in fresh-water snails. Wilhelm
Roux Archiv für Entwicklungsmechanik der Or-
ganismen, 24: 332-356.

GEORGE, J. D., 1958, Experimental fusion of em-
bryos of Limnaea stagnalis L. Proceedings Akad-
emie van Wetenschappen, Amsterdam, Biologi-
cal and Medical Sciences, Afdeling Natuurkunde,
61: 595-597.

HUMASON, С. L., 1979, Animal tissue techniques.
W. H. Freeman and Co., San Francisco.

IKEDA, K., 1937, Cytogenetic studies on the self-
fertilization of Philomycus bilineatus Benson.
(Studies of hermaphroditism in Pulmonata Il.).
Journal of Science of the Hiroshima University B:
5(5): 1-58. [/ncilaria (or Meghimatium) bilineata
(Genson 1842)]

LONGO, Е. J., 1976, Ultrastructural aspects of fer-
tilization in spiralian eggs. American Zoologist,
16: 375-394.

LUCHTEL, D., 1972, Gonadal development and

LEHMANNIA VALENTIANA CONJOINED TWINNING MECHANISM 339

sex determination in pulmonate molluscs; Il.
Arion ater rufus and Deroceras reticulatum.
Zeitschrift fur Zellforschung, 130: 302-311.

LUCHTEL, D. L., 1976, An ultrastructural study of
the egg and early cleavage stages of Lymnaea
stagnalis, a pulmonate mollusc. American Zool-
ogist, 16: 405—419.

MASON, J. & COPELAND, J., 1988, The incidence
and variety of Lehmannia valentiana conjoined
twins: Related breeding experiments (Gas-
tropoda, Pulmonata). Malacologia, 28(1-2): 17—
27:

RAVEN, C. P., 1961, Oogenesis: the storage of de-
velopmental information, pp. 39, 149-150. Per-
gamon Press, New York.

REINGOLD, S. D. & GELPERIN, A., 1980, Feeding
motor programme in Limax. Il. Modulation by
sensory inputs in intact animals and isolated cen-
tral nervous systems. Journal of Experimental Bi-
ology, 85: 1-19.

REYNOLDS, E. S., 1963, The use of lead citrate at
high pH as an electron opaque stain in electron

microscopy. Journal of Cell Biology, 17: 208-
212.

RUNHAM, N. W. & HUNTER, P. J., 1970, Terres-
trial slugs. Hutchinson & Co., London.

SALEUDDIN, A. S. M. & KAHN, H. R., 1981, Mo-
tility of the oocyte of Helisoma (Mollusca). Euro-
pean Journal of Cell Biology, 26: 5-10.

SALEUDDIN, A. S. M., FARRELL, C. L. 8 GOMOT,
L., 1983, Brain extract causes amoeboid move-
ment in vitro in oocytes in Helix aspersa (Mol-
lusca). International Journal of Invertebrate Re-
production, 6: 31-34.

STEARS, M., 1974, Contributions to the morphology
and histology of the genital system of Limax-Val-
entianus Pulmonata Limacidae. Annale Univer-
sity v. Stellenbosch, South Africa, 49(A3): 1-46.

WILSON, E. B., 1904, Experimental studies on ger-
minal localization. |. The germ-regions in the egg
of Dentalium. Journal of Experimental Biology, 1:
1-72.

Revised Ms. accepted 8 February 1989

MALACOLOGIA, 1989, 30(1-2): 341—364

SPERMATOPHORES OF AQUATIC NON-STYLOMMATOPHORAN
GASTROPODS: À REVIEW WITH NEW DATA ON HELIACUS
(ARCHITECTONICIDAE)

Robert Robertson

Department of Malacology, Academy of Natural Sciences
Nineteenth and the Parkway, Philadelphia, Pennsylvania 19103, U.S.A.

ABSTRACT

At least 55 genera in 27 non-stylommatophoran families have spermatophores. The higher
taxa involved are the Neritimorpha, Cerithioidea, Vermetoidea, Heteropoda, Triphoroidea, Ar-
chitectonicoidea, Pyramidelloidea, Bullomorpha, Thecosomata, Acochlidiomorpha, Nudibran-
chia, Siphonarioidea, and possibly Fissurelloidea and Lymnaeoidea. These are in all four gas-
tropod subclasses. Higher taxa with the most known spermatophore-bearing genera are:
Cerithioidea (18 genera), Neritimorpha (9), Acochlidiomorpha (6), and Heteropoda (5). The
animals and their spermatophores are very diverse morphologically.

Spermatophores are described and illustrated in the Architectonicidae for the first time. The
genus Heliacus is protandric to simultaneously hermaphroditic, and aphallic. Two western At-
lantic species of Heliacus were studied: H. cylindricus (Gmelin, 1791) and H. perrieri (Roche-
brune, 1881). Their spermatophores are long, coiled tubes. The coiling axes are of two lengths
in both species, a possible outcome of reciprocal spermatophore transfer. A previously unde-
scribed spermatophoric groove, extending onto part of the proboscis, handles and possibly
molds spermatophores.

Compared with the more concordant taxonomic distributions of pigmented mantle organs,
ciliated strips, chalazae, and heterostrophy (Robertson, 1985), occurrences of spermatophores
are sporadic.

It is concluded that the mere presence or absence of spermatophores cannot be used to
resolve relationships or to revise the higher classification of gastropods. A possible exception is
the presence of spermatophores in the Architectonicoidea, Pyramidelloidea, and Bullomorpha,
for which there is independent evidence of phyletic relationships. In these and other cases,
homologies remain to be determined. In some taxa, though, spermatophores are useful in
lower-category systematics.

Little is Known about spermatophore-forming organs in prosobranchs and opisthobranchs.
There is no strong evidence that the “prostates” and other such organs are homologous from
higher taxon to higher taxon, and possibly have evolved separately.

Other topics briefly treated in the Discussion are: the early history of spermatophore studies,
and surveys of the occurrences of penes, spermatozeugmata, and spermatophores among non-
stylommatophorans.

Key words: spermatophores; reproduction; fertilization; Heliacus; Architectonicidae

INTRODUCTION

The higher classification of gastropods
since Thiele (1929-1931) has been and still
is in ferment (e.g. Wenz, 1938-1944; Knight,
1952; Boettger, 1955; Risbec, 1955; Zilch,
1959—1960; Cox, 1960; Moore, 1960; Fretter
8 Graham, 1962; Taylor & Sohl, 1962; Ghis-
elin, 1965; Kosuge, 1966; Golikov 8 Star-
obogatov, 1975; Minichev 8 Starobogatov,
1979;Fretter, 1980;Salvini-Plawen, 1980;Gos-
liner, 1981; Fretter & Graham, 1982; Tillier,
1984; Haszprunar, 1985d, 1988a,b; Ponder &

Warén, 1988). The relationships of certain
families have come under special scrutiny be-
cause they do not conveniently fit into such
categories as the Prosobranchia and Opistho-
branchia. These families include the Pyra-
midellidae (Thorson, 1946; Fretter & Graham,
1949; Franzén, 1955; Haszprunar, 1985a;
Healy, 1987, 1988b), Architectonicidae (Rob-
ertson, 1973; Climo, 1975; Healy, 1982;
Haszprunar, 1985a, 1985b), and Mathildidae
(Risbec, 1955; Climo, 1975; Haszprunar,
1985c). Increasingly, new tools and new ap-
proaches are yielding new characters and

(341)

342 ROBERTSON

character sets, e.g. from the ultrastructure of
spermatozoa (Healy, 1987) and of osphradia
(Haszprunar, 1985a). New names for higher
taxa are proliferating at a rapid rate.

This study of spermatophores was under-
taken with the belief that it might be relevant
to the higher category systematics of gastro-
pods. As will be seen, the conclusion is neg-
ative, with one possible exception.

Observing living Pyramidellidae, Architec-
tonicidae, and Epitoniidae, using equipment
no more complex than a light microscope, |
have studied four characters that | believe
have bearing on the systematic relationships
of these families, and, indeed, on the higher
category systematics of gastropods (Robert-
son, 1985): “pigmented mantle organs,” cili-
ated (pallial) strips, chalazae (threads con-
necting egg cocoons), and heterostrophy
(larval hyperstrophic shell coiling). Each of
the three families has one to four of these
characters. A literature search revealed that
these four characters recur somewhat con-
cordantly among some other mesogastro-
pods, some lower opisthobranchs, and some
basommatophoran pulmonates (Table 1).
This pattern of taxonomic occurrence sug-
gests that these four characters are homolo-
gous among higher taxa, and that these three
groupings (straddling all four gastropod sub-
classes!) are somewhat and somehow re-
lated (Robertson, 1985).

Spermatophores are usually detachable,
non-living, secreted structures containing and
transferring sperm. Having discovered them
in the American Pyramidellidae (Robertson,
1966, 1968, 1978), and in the Architectoni-
cidae (Robertson, 1973), it was natural for me

to consider them in light of the previously `

studied four characters.

The capsule wall of Runcina spermato-
phores was said by Kress (1985a) to be “al-
buminous,” but there seem to be no chemical
data supporting this statement for this or other
non-stylommatophoran gastropods, relevant
as such data might be. Spermatophore walls
are somewhat hardened, deformable, and re-
sume their shape upon removal of pressure
(but the crook of Fargoa bartschi is glass-
like). Sperm balls, and sperm surrounded by
prostatic secretion or mucus (as in some littor-
inids), not being encapsulated, are here con-
sidered not to be spermatophores (following
Mann, 1984).

If spermatophores (or any other organically
produced structures) have bearing on a
higher category classification, they must be

homologous from taxon to taxon. Spermato-
phores are acellular (extracellular?) secre-
tions, and the homologies of the anatomical
structures producing them are also important.

The main question being asked here is:

(1) Does the pattern of taxonomic occur-
rence of spermatophores match or approxi-
mate those of the four other characters
treated by Robertson (1985) and believed by
him to indicate some homologies and taxo-
nomic relationships? To this end, the scat-
tered original literature on aquatic, non-sty-
lommatophoran gastropod spermatophores is
reviewed here for the first time.

Subsidiary purposes of the present paper
are:

(2) To describe and illustrate for the first
time architectonicid (Heliacus) spermato-
phores and a curious spermatophoric groove
on the right side of the head and part of the
proboscis.

(3) To explain why there are two types of
coiling in the spermatophores of both species
studied (Heliacus cylindricus and H. perrieri).

(4) To try to determine the anatomical site
of production of Heliacus spermatophores
from previously published anatomical data.

LITERATURE REVIEW

An attempt is made in the Appendix to list in
systematic order all the genera and higher
taxa of non-stylommatophoran gastropods in
which spermatophores are known, plus ancil-
lary information on sex. There are no doubt
other published reports that | have over-
looked, and gastropods still to be studied that
have spermatophores.

Spermatophores can be present for only a
brief part of the reproductive cycle, as has
been reported in a runcinid (Ghiselin, 1963,
1965) and a lymnaeid (Jackiewicz, 1986).
Thus, negative reports can be inconclusive.

The stylommatophorans, for which the
spermatophore literature is extensive, are
omitted because their spermatophores are
produced in structures not occurring among
the prosobranchs or opisthobranchs and
therefore are demonstrably not homologous.
The basommatophorans are included be-
cause they are related to basal opistho-
branchs (Morton, 1955).

There frequently is no mention of the ana-
tomical site of production of spermatophores
when these are reported. When it is stated,

SPERMATOPHORES OF AQUATIC GASTROPODS 343

this information is quoted separately in the B
Notes.

Prosobranch spermatophores are still little
known. As recently as 1953, Fretter, in a pa-
per on ‘the transference of sperm from male
to female prosobranch, with reference, also,
to the pyramidellids,” spermatophores were
not mentioned.

MATERIALS AND METHODS: HELIACUS

Two tropical, shallow water, western Atlan-
tic architectonicid species were studied: Heli-
acus cylindricus (Gmelin, 1791), the type spe-
cies of Heliacus Orbigny, 1842, and H.
perrieri (Rochebrune, 1881), the type species
of the subgenus Teretropoma Rochebrune,
1881. Classification and nomenclature here
follow Bieler (1985), who illustrated the shells
of both species; H. perrieri has a more de-
pressed, usually larger shell than H. cylindri-
cus. Haszprunar (1985b) studied the anatomy
of H. perrieri and a few other architectonicids.

Spermatophores of H. cylindricus were first
found and studied by me at the Bermuda
Biological Station in late March and early
April 1970. They were collected at Hungry
Bay, Paget Parish, on the southeast coast,
where the snails were in a feeding associa-
tion with the colonial zoanthiniarian sea
anemone Palythoa grandiflora Verrill, 1901.
More H. cylindricus and H. perrieri were
studied at various Bahama and British Virgin
islands, and both species were also sent to
me from southeast Florida. As expected, all
these other animals were with zoanthid hosts
(Robertson, 1967).

The animals were kept alive in finger bowls
of frequently changed sea water and were
studied with the aid of a Wild M5 dissecting
microscope.

The reproductive systems were not dis-
sected or studied histologically by me; Hasz-
prunar (1985b) has done this.

The term paraspermatozoa is used rather
than atypical spermatozoa, following Melone
et al. (1980) and Healy 8 Jamieson (1981).

RESULTS: LIVING HELIACUS
Sex
Architectonicids have long been known to

be aphallic (Bouvier, 1886; Thiele, 1929; Mer-
rill, unpublished; Haszprunar, 1985b). The im-

plicit question—how is sperm transferred?—
was not addressed by them but is answered
here. It should be noted that at least Heliacus
is protandric to simultaneously hermaphro-
ditic (Robertson, 1973; Haszprunar, 1985b).
(On two errors in the old literature see Appen-
dix, Note A23.)

Spermatophores

The elongate spermatophores of the two
species appear indistinguishable, being long,
narrow, openly coiled tubes slightly tapering
towards each end (Fig. 1), and up to about 25
mm long uncoiled. Near the end where the
sperm emerges, coiling is tighter (Fig. 1B).
These spermatophores are smooth, faintly
greyish white, translucent, and easily broken.
When fresh, the tubes are full of eusperma-
tozoa (and paraspermatozoa?). Both species
have the two coiling types shown in Fig. 1.
These differ in total length, in tightness of coil-
ing, and in the length of their coiling axes.

Spermatophoric groove

А cream-white, ciliated spermatophoric
groove (Fig. 2, spg) arises in the mantle cavity
and extends onto the right side of the neck
and head and of the partly everted acrembolic
proboscis, ending where grey-black (ectoder-
mal?) speckling ends. The groove is perma-
nent and may help to mold the spermato-
phores, and definitely manipulates them.
Several spermatophores could be seen pro-
jecting from the mantle cavity of one animal.
During mating, one spermatophore frequently
is in the groove, but transfer can occur also
over the head, medially, without use of the
groove (Fig. 3).

DISCUSSION
History

Gastropod spermatophores have long
been known. Lister (1694) discovered, de-
scribed, and illustrated them in Helix, calling
the structure a capreolus (Greek for a tendril).
Quoy 8 Gaimard (1833-1834) were the first
to report a spermatophore (“corps en massue
allongée”) in a non-terrestrial gastropod (an
unspecified neritid).

344 ROBERTSON

FIG. 1. A. Spermatophore of Heliacus perrieri; about 10 mm long uncoiled. B. Spermatophore of H. cylin-
dricus; about 25 mm long uncoiled. N.B.: Fig. A may not be a complete spermatophore. Also note: both
species have the same dimorphic spermatophore coiling types. No interspecific differences were observed.
Abbreviation: eus = clump of euspermatozoa. Both Florida.

SPERMATOPHORES OF AQUATIC GASTROPODS 345

FIG. 2. Heliacus cylindricus. Right side of the head of an extended living animal, showing the ciliated and
unpigmented spermatophoric groove (spg) extending from inside the mantle cavity onto the right side of the
slightly everted acrembolic proboscis (acp). Other abbreviations: dft — dorsal surfaces of the foot; eye =
eye in the slight lateral swelling; gr = subsurface white granules; me = mantle edge; vch = ventral ciliated
channel on the left tentacle; vft = ventral surface of the foot. British Virgin Islands.

Spermatophores in the four subclasses

Pulmonate spermatophores appear not to
be homologous with those of prosobranchs
and opisthobranchs because they are formed
by the epiphallus and flagellum (early au-
thors, and Lind, 1973; Breure 8 Eskens,
1977), organs not present in the other two
subclasses (but see Note A19). In certain
textbooks and treatises these subclasses
have been said to form spermatophores in
“prostate” glands (e.g. Hyman, 1967; Bee-
man, 1977). The confused primary literature
makes such a generalization premature
(Notes B1 to B23).

Patterns of taxonomic occurrence
At least 55 non-stylommatophoran gastro-

pod genera and 27 families have spermato-
phores (Appendix, excluding questioned

taxa). The higher taxa involved are the Neri-
timorpha, Cerithioidea, Vermetoidea, Het-
eropoda, Triphoroidea, Architectonicoidea,
Pyramidelloidea, Bullomorpha, Thecosomata,
Acochlidiomorpha, Nudibranchia, Siphonario-
idea, and possibly the Fissurelloidea and Lym-
naeoidea. These taxa are in all four gastropod
subclasses.

The pattern of taxonomic occurrence of
spermatophores at the familial and suprafa-
milial levels is shown in Table 1, where com-
parisons can be made with the patterns of the
four other characters studied by Robertson
(1985).

The most striking concordance is between
the related superfamilies Architectonicoidea,
Pyramidelloidea, and Bullomorpha, all of
which have all five characters. Table 1 is, how-
ever, somewhat misleading because rare, iso-
lated occurrences are equated with common
ones. The Bullomorpha get a plus for sper-

346

1 cm

ROBERTSON

ten

FIG. 3. Heliacus perrieri. Two animals in copulo: medial spermatophore (sp) transfer (more usually, the
spermatophoric groove is used). It is uncertain which animal was the donor or if both were. Other abbrevi-
ations: ft = foot; op = operculum; sh = shell; ten = tentacle. Florida.

matophores, but these are reported in only
four out of a great many genera. Spermato-
phores are also rare in the Nudibranchia. The
Cerithioidea would have none of the four other
characters were it not for an equivocal report
of chalazae in Campanile, which may not be a
member of the superfamily (Note A9). The
higher categories with the most known sper-
matophore-bearing genera are the Cerithio-
idea (18 genera), Neritoidea (9), Acochlidio-
morpha (6), and Heteropoda (5).

The reported taxonomic occurrences of
spermatophores are sporadic and labile, be-
ing found in such disparate taxa as, for exam-
ple, the Neritidae, Cerithioidea, Pyramidell-
idae, and Siphonaria. They are in the
Neritidae and Phenacolepadidae, but are ap-
parently lacking in the Helicinidae. (Why?
Perhaps they have not been looked for in ter-
restrial prosobranchs?). Spermatophores oc-
cur in the Hedylopsidae and Microhedylidae,
so why are they apparently not present also in

SPERMATOPHORES OF AQUATIC GASTROPODS 347

the Acochlidiidae? In the Basommatophora,
spermatophores are known only in Sipho-
naria (where they are commonly observed),
and Chilina and Stagnicola (in which they ap-
pear to have been observed only once or
twice each [Note A38]).

Diversity

Spermatophores occur in aquatic gastro-
pods of diverse forms, modes of life, and hab-
itats, and themselves vary in form and place-
ment; e:g;-

Gastropods:
Limpets (Phenacole- vs.
pas, Siphonaria)

Slugs (Runcin-
idae, Polycera,

and snails (most Aeolidiidae)
taxa)

Marine to estuarine vs. Fresh-water to

(most taxa) slightly estuarine
(Stagnicola,
Chilina)

Benthic (most taxa) vs. Holoplanktonic (Het-
eropoda, Limac-
ina)

Non-interstitial vs. Interstitial (Hedy-

(most taxa) lopsidae, Micro-
hedylidae)

Non-sessile vs. Sessile (Vermet-

(most taxa) idae)
Gonochoric vs. Protandric or simul-

(many taxa) taneously herma-

phroditic (many

taxa)
Penis-bearing vs. Aphallic
(many taxa) (many taxa)
Spermatophores:
Unattached vs. Shell-attached (Far-
(most taxa) goa, Parthenina,
Atlantidae?)
Non-planktonic vs. Planktonic (Ver-
(most taxa) metidae)

There would seem to be no correlations be-
tween habitats or modes of life and the pres-
ence or kinds of spermatophores.

“Prostates”

The “prostates” and ‘other anatomical
structures forming spermatophores (Notes B1
to B23) are not necessarily homologous
among higher taxa. Ghiselin (1965) men-
tioned the “so-called prostate” of some bullo-
morphs. Diverse terminology in the B notes
does not, of course, in itself disprove homol-
ogy. But even in one neritid animal the

“prostate” can be a “complex” of glands
(Fretter, 1946; Berry et al., 1973).

Consistency of occurrence
within higher taxa

Spermatophores may be present in all spe-
cies in some suprageneric taxa, such as the
Cerithiidae or Cerithioidea (as Houbrick
[1980, 1988] suggested.) However, in a few
genera (such as Limacina [Note A29] and
probably Hedylopsis and Siphonaria [Notes
A31, A36, A37], spermatophores are present
in one or some but not all included species.
Predictions about the presence or absence of
spermatophores in a given taxon are hazard-
ous. Such characters make poor taxonomic
characters, but a bad one in one taxon can be
a good one in another taxon.

Lower category systematics

The spermatophores of many stylommato-
phorans, neritids (Andrews, 1937), and Si-
phonaria (Hubendick, 1945, 1946, 1955; Jen-
kins, 1981, 1983, 1984) have useful generic
and specific characters. So also do those of
six eastern North American and Gulf of Mex-
ico pyramidellid species in the genera Boo-
nea and Fargoa (Robertson, 1966, 1978). For
distinguishing genera and species, their sper-
matophores are more useful than the shells,
which have undergone convergent and diver-
gent evolution.

Penes and spermatophores

Most gastropod taxa transfer sperm with
penes rather than spermatophores. One
might expect an internally fertilizing gastro-
pod to have one or the other, not both. How-
ever, the following penis-bearing taxa also
have spermatophores (Appendix, pen): most
Neritimorpha, perhaps some Pyramidel-
loidea, all Heteropoda, and a few Bullomor-
pha and Nudibranchia. In some of these,
though, spermatophores and penes may not
occur in the same species, with one or the
other being present (Notes A29, A31, A36). In
Runcina, the penis may form the outer layer
of the spermatophore (Kress, 1985b), as in
some pulmonates (Mann, 1984). In whole or
in part, gastropod penes, like spermato-
phores, may not be homologous from one su-
prageneric taxon to another.

The role that penes and spermatophores
may play in the courtship of non-stylom-

348 ROBERTSON

TABLE 1. Higher taxa of gastropods and the occurrences of five characters, four from Robertson (1985)
and spermatophores from the Appendix. Observe that the ranking of taxa is various (it is appreciated
that the lower the rank used the lower the probable concordance). Pigmented mantle organ (PMO) data
are given for veligers and postlarvae combined. CSs = ciliated strips. CHae = chalazae. HET =
heterostrophy. Blanks mean data lacking or characters absent (an important difference, but difficult to
separate in practice). The footnotes are supplementary to Robertson (1985).

Taxa PMO CSs CHae HET Spermatophores

Fissurelloidea ?
Patellogastropoda

Neritimorpha

Cerithioidea ?

Heteropoda

Triphoroidea

Janthinoidea pe ?
Other Mesogastropoda

Neogastropoda

Valvatoidea 2
Architectonicoidea aes
Rissoelloidea ?
Omalogyroidea

Pyramidelloidea .
Bullomorpha +4
Thecosomata

Gymnosomata

Aplysiomorpha ? +
Pleurobranchomorpha +6 2
Acochlidiomorpha

Ascoglossa ++ +

Nudibranchia

Ellobiidae +7
Otinidae

Amphibolidae ? + +
Siphonarioidea ee +
Chilinidae Ir

Lymnaeidae ?
Onchidiidae +

Veronicellidae +

Other Pulmonata 2

+ + ++

+3 +

[A]
+
[0]

+ +90 +
+
+
+

+
o
+ +
+ +

++ ++++++++
+

‘| agree with Haszprunar (1985d) and Bieler (1988) that in 1985 | accorded too much systematic significance to the PMO
in the Janthinoidea (Epitonioidea).

2Haszprunar (19854, 1988) has argued that the PMOs of the Janthinoidea are not homologous with those in the
Architectonicoidea, Pyramidelloidea and lower Euthyneura, citing Richter & Thorson (1975) and two papers on nudibranchs
(Phestilla and Doridella). Richter & Thorson did not determine the function of PMOs, and the two nudibranchs lack PMOs.

SHaszprunar, 1985a, 1985b, 1985c.

4BULLIDAE: Bulla; veliger; “nephrocyst”; bright red or garnet; Farfan & Buckle Ramirez, 1988.
PHILINIDAE: Philine; veliger; [gland]; [blackish or black]; Franc, 1948.
GASTROPTERIDAE: Gastropteron; veliger; [gland]; black; Franc, 1948.

°BULLIDAE: Bulla; chalazae; Fartan & Buckle Ramirez, 1988.

®The only pleurobranchomorph known to have a PMO past metamorphosis is Pleurobranchaea (PLEURO-
BRANCHIDAE); Franc, 1948; Thiriot-Quiévreux, 1967 [black spot].

7ELLOBIIDAE: Cassidula; [chalazae]; Berry, 1977.

£SIPHONARIIDAE: Siphonaria; Wimperband; Köhler, 1893.

°In 1985 | doubted that PMOs occur in the Pulmonata (see my note 7). | still have doubts. The five papers on pulmonates
cited by Haszprunar (1988a: 390) do not report a variously colored structure containing fine particles in a paste that can be
put into suspension, characteristics of PMOs.

matophorans is virtually unknown. Houbrick Spermatozeugmata

(1973) described the behavior of Cerithium

muscarum during spermatophore transfer, Spermatophores and spermatozeugmata
but was unable to make interspecific compar- are sometimes confused, having similar func-
isons. tions (transferring sperm) but different cyto-

SPERMATOPHORES OF AQUATIC GASTROPODS 349

logical origins. Spermatozeugmata are en-
larged and modified paraspermatozoa with
the euspermatozoa attached to the “tail.”
Spermatophores are never contained in true
spermatozeugmata. The Epitoniidae, one of
the groups | have studied alive, have sperma-
tozeugmata (Robertson, 1983, and earlier pa-
pers by others cited therein; Melone et al.,
1978, 1980) but no spermatophores.

Spermatophores or spermatozeugmata
might be expected to evolve in unrelated
aphallic gastropods ifthey are without “current
fertilization” (transferring sperm in the current
of water entering the mantle cavity).

Four superfamilies and families in the Meso-
gastropoda are aphallic and have partly to fully
open pallial gonoducts in both sexes (Fretter,
1946, 1951; Johansson, 1946, 1947, 1953,
1956; Graham, 1954; Houbrick, many papers;
Houston, 1985): (1) Cerithioidea, (2) Tri-
phoridae (irregularly?), (3) Cerithiopsidae, and
(4) Janthinoidea. Of these, the Cerithioidea
has spermatophores (Appendix), and the Cer-
ithiopsidae and Janthinoidea have spermatoz-
eugmata (Cerithiopsis: Fretter & Graham,
1962; Houston, 1985; Seila: Houston, 1985).
The Triphoridae has both structures, but per-
haps not in the same genus (Appendix and
Note A22).

Site of Heliacus spermatophore formation

Although he had two species of Heliacus
alive, Haszprunar (1985b) was not fortunate
enough to see spermatophores or the sper-
matophoric groove. Though he knew of their
existence, he did not determine where the
spermatophores are formed and molded.
However, it is possible from Haszprunar's ex-
cellent anatomical account to make a tenta-
tive determination. There is a “prostate” be-
tween the vesicula seminalis and vas
deferens. But the very long spermatophores
(up to 25 mm uncoiled) cannot come from the
“prostate” alone. In the Heliacus studied by
Haszprunar, this structure was elliptical, and
only about 0.5 mm long and 0.3 mm wide
(outer dimensions). The spermatophores
possibly are formed and molded in the
“prostate” and the voluminous and glandular
vas deferens. Molding may also occur in the
spermatophoric groove.

Heliacus spermatophore dimorphism

While they are simultaneous hermaphro-
dites, the animals can be expected to transfer
spermatophores reciprocally. Indeed, this

may partly have been seen in H. perrieri (Fig.
3). Heliacus has an extremely deep mantle
cavity, and the entering and perhaps still pli-
ant spermatophores no doubt have to con-
form to its coiling. The two forms of spermato-
phore may arise from temporarily “donor” and
“recipient” animals.

Spermatophoric and similar grooves

The spermatophoric groove of Heliacus is
comparable to the “omniphoric” (all-carrying)
and other ciliated grooves, such as the open
genital one coming from inside the mantle cav-
ity (if present) and extending onto the right side
of the foot or head of some other aquatic
prosobranchs and opisthobranchs (e.g. Davis,
1967; Beeman, 1977; Houbrick, 1987b). The
extension of the groove onto the proximal end
of the partially everted acrembolic proboscis
and its use in manipulating spermatophores
are, as far as is known, unique to architec-
tonicids. The groove may also function as a
cleansing tract or ovipositor. | was unable to
determine whether the groove handles incom-
ing as well as outgoing spermatophores.

Neritoidean head structures

In taxa other than Heliacus there are other
external structures that apparently aid in the
transfer of spermatophores. Besides the
cephalic “penes” of most male neritids and
phenacolepadids, males and some females
of these taxa have other species-specific
head ridges or projections (Hedley, 1916; An-
drews, 1936, 1937; Starmühlner, 1969; Fret-
ter, 1984).

Architectonicoidea and Pyramidelloidea

The presence of spermatophores in both
the Architectonicidae and Pyramidellidae
might seem to support Haszprunar's (1985a)
placement of these two families in his new
order Allogastropoda. However, nothing is yet
known about the site of formation of pyra-
midellid spermatophores, and there is only
speculation about architectonicids. There
does, though, seem to be a higher category
relationship between the Bullomorpha and
these two families. It is not yet known whether
mathildids (architectonicoideans) have sper-
matophores (Risbec, 1955; Climo, 1975;
Haszprunar, 1985c).

350 ROBERTSON

CONCLUSIONS

The degrees of concordance between the
five characters in Table 1 could be quantified,
but this would imply unwarranted precision as
to homologies. lt can only be said than that
the pattern of taxonomic occurrence of sper-
matophores, when tempered by the caveats
in the text, do not accord well with the four
other characters. In the absence of any defi-
nite information about spermatophore homol-
ogies (or the lack of them), the data in the
Appendix and Table 1 do not shed much if
any new light on the higher classification of
non-stylommatophoran gastropods. The
mere presence or absence of spermato-
phores, while of great interest functionally,
would not have much phylogenetic signifi-
cance without knowledge of relevant homol-
ogies. Many spermatophores may have
evolved independently.

More studies, such as those of Kress
(1985a,b; 1986) on Runcina, are needed, and
taxa should be compared. Information is par-
ticularly desirable on the anatomical formative
sites of spermatophores and their homolo-
gies, on the role of spermatophores in sperm
transfer, and on their utility in higher- and
lower-category systematics.

Spermatophores are no doubt important in
the life histories of individual species, but it is
puzzling why they should occur in some taxa
and not others. Species-specific spermato-
phore characters in some pyramidellids pre-
sumably serve at least in part for intra- and/or
interspecific recognition. It is therefore curi-
ous to find no differences between the sper-
matophores of the two species of Heliacus,

which can co-exist on the same colony of :

zoanthid polyps.

The sources and reliability of data on which
the various classifications cited in the Intro-
duction are based naturally are important.
One wonders if closer scrutiny of other char-
acters would show similar degrees of conver-
gence.

ACKNOWLEDGMENTS

Lt. Col. Corinne E. Edwards, USAF (retired),
deserves belated thanks for sending live He-
liacus from Florida. Jean M. Crabtree pre-
pared the figures for publication. Zenona Hol-
szanska translated Jackiewicz (1986). The
following kindly read and criticized various
drafts of the MS: Rudiger Bieler, Arthur J. Cain,

Kenneth C. Emberton, Vera Fretter, Alastair
Graham, Carole S. Hickman, Richard S. Hou-
brick, Winston F. Ponder, and Gary Rosen-
berg. Bieler and Houbrick also provided un-
published data on three genera in the
Appendix. Early on, (United States) National
Science Foundation Grant DEB 76-18835
supported this work.

LITERATURE CITED

ABE, N., 1940, The homing, spawning and other
habits of a limpet, Siphonaria japonica Donovan.
Science Reports Tohoku Imperial University, ser.
4 (Biology), 15: 59-95.

ALLANSON, B. R., 1958, On the systematics and
distribution of the molluscan genus Siphonaria in
South Africa. Hydrobiologia, 12: 149-180.

ANDREWS, E. A., 1936, Spermatophores of the
snail Neritina reclivata. Journal of Morphology,
60: 191-209.

ANDREWS, E. А., 1937, Certain reproductive or-
gans in the Neritidae. Journal of Morphology, 61:
525—561.

ANKEL, W. E., 1936, Prosobranchia. In: GRIMPE,
С. & WAGLER, E., ed., Die Tierwelt der Nord-
und Ostsee. Leipzig, Teil IX.b,, Lief. 29, 240 pp.

BABA, K., 1967, Supplementary notes on the anat-
omy of Metaruncina setoensis (Baba, 1954),
(N.G.) (Opisthobranchia-Cephalaspidea). Publi-
cations of the Seto Marine Biological Laboratory,
15: 185-197, pl. 2-5.

BEEMAN, R. D., 1977. Gastropoda: Opisthobran-
chia. In: GIESE, А. С. & PEARSE, J. 5., eds.
Reproduction of marine invertebrates . . . mol-
luscs: gastropods and cephalopods, 4: 115-179.
Academic Press.

BERGH, R., 1890, Die Titiscanien, eine Familie der
rhipidoglossen Gasteropoden. Morphologisches
Jahrbuch, 16: 1-26, 3 pl.

BERRY, A. J., 1977, Gastropoda: Pulmonata. In:
GIESE, A. C. 8 PEARSE, J. S., ed., Reproduc-
tion of marine invertebrates. Vol. IV. Molluscs:
Gastropods and Cephalopods. Academic Press,
pp. 181-226.

BERRY, A. J., LIM, R. & SASE KUMAR, A., 1973,
Reproductive systems and breeding condition in
Мета birmanica (Archaeogastropoda: Nerita-
cea) from Malayan mangrove swamps. Journal
of Zoology, London, 170: 189-200.

BIELER, R., 1985, Die Gattungen der Architectoni-
cidae (Gastropoda: Allogastropoda). Teil 3:
Pseudotorinia, Nipteraxis, Heliacus, Eosolarium.
Archiv für Molluskenkunde, 116: 89-117.

BIELER, R., 1988, Phylogenetic relationships in the
gastropod family Architectonicidae, with notes on
the family Mathildidae. In PONDER, W. F., EER-
NISSE, О. J. & WATERHOUSE, J. H., eds., Proso-
branch phylogeny, Proceedings of a Symposium
held at the 9th International Malacological Con-
gress, Edinburgh, Scotland, 31 August—6 Sep-

SPERMATOPHORES OF AQUATIC GASTROPODS 351

tember, 1986. Malacological Review Supplement
4: 205-240.

BOETTGER, C. R., 1955, Die Systematik der eu-
thyneuren Schnecken. Verhandlungen der Deut-
schen Zoologischen Gesellschaft; Zoologischer
Anzeiger Suppl. bd. 18: 253-280.

BOURNE, G. C., 1908, Contributions to the mor-
phology of the group Neritacea of aspidobranch
gastropods.—Part |. The Neritidae. Proceedings
of the . . . Zoological Society of London, 1908:
810-887, pl. 46-66.

BOUVIER, E. L., 1886, Contributions a l'étude des
Prosobranches Pténoglosses. Bulletins de la So-
ciété Malacologique de France, 3: 77-130, pls.
325;

BREURE, A. S.H. & ESKENS, A. A. C., 1977, Ob-
servations on the formation of spermatophores in
a bulimulid land snail, Drymaeus canaliculatus
(Pfeiffer, 1845) (Mollusca, Gastropoda, Pulmo-
nata). Netherlands Journal of Zoology, 27: 271—
276, 2 pls.

CHALLIS, D. A., 1968, A new genus and species
of the order Acochlidiacea (Mollusca: Opistho-
branchia) from Melanesia. Transactions of the
Royal Society of New Zealand, Zoology, 10:
191-197. :

CHALLIS, D. A., 1969, Philinoglossa marcusi п. sp.
(Mollusca: Opisthobranchia: Philinoglossacea)
from the British Solomon Islands Protectorate.
Transactions of the Royal Society of New
Zealand, Biological Sciences, 11: 169-175.

CHAMBERS, L. A., 1934, Studies on the organs of
reproduction in the nudibranchiate mollusks, with
special reference to Embletonia fuscata Gould.
Bulletin of the American Museum of Natural His-
tory, 66: 599-641, pl. 28-31.

CLIMO, Е. M., 1975, The anatomy of Gegania
valkyrie Powell (Mollusca: Het[e]rogastropoda:
Mathildidae) with notes on other heterogastro-
pods. Journal of the Royal Society of New
Zealand, 5: 275-288.

COX, L. R., 1960, Thoughts on the classification of
the Gastropoda. Proceedings of the Malacologi-
cal Society of London, 33: 239-261.

DAVIS, G. M., 1967, The systematic relationship of
Pomatiopsis lapidaria and Oncomelania hupen-
sis formosana (Prosobranchia: Hydrobiidae).
Malacologia, 6: 1-143.

DAZO, B. C., 1965, The morphology and natural
history of Pleurocera acuta and Goniobasis
livescens (Gastropoda: Cerithiacea: Pleurocer-
idae). Malacologia, 3: 1-80.

DIEUZEIDE, R., 1935, Contributions a l'étude de
deux types de Gastéropodes Pulmonés marins:
Siphonaria algesirae Quoy et Gaimard; Gadinia
garnoti Payraudeau. Gouvernement Général de
l'Algérie. Station d’Aquiculture et de Pêche de
Castiglione, “1934”: 1-196, 13 pl.

DOE, D. A., 1974, A new species of the order Aco-
chlidiacea (Opisthobranchia: Microhedylidae)
from New England. Transactions of the American
Microscopical Society, 93: 241-247.

FARFAN, B. C. 8 BUCKLE RAMIREZ, L. F., 1988,

Spawning and ontogeny of Bulla gouldiana (Gas-
tropoda: Opisthobranchia: Cephalaspidea). Veli-
ger, 31: 114-119.

FORSTER, H., 1934, Beitráge zur Histologie und
Anatomie von Philine aperta L. Dissert. Univ.
Kiel. Dresden (Risse), 67 pp. [Not seen. Refer-
ence from Marcus (1974).]

FRANC, A., 1948, Véligeres et Mollusques Gas-
téropodes des Baies d'Alger et de Banyuls. Jour-
nal de Conchyliologie, 88: 13-35.

FRANZEN, Á., 1955, Comparative morphological
investigations into the spermiogenesis among
Mollusca. Zoologiska Bidrag frán Uppsala, 30:
399-456, 2 pl.

FRETTER, V., 1946, The genital ducts of Theo-
doxus, Lamellaria and Trivia, and a discussion on
their evolution in the prosobranchs. Journal of
the Marine Biological Association of the United
Kingdom, 26: 312-351.

FRETTER, V., 1951, Observations on the life his-
tory and functional morphology of Cerithiopsis tu-
bercularis (Montagu) and Triphora perversa (L.).
Journal of the Marine Biological Association of
the United Kingdom, 29: 567-586.

FRETTER, V., 1953, The transference of sperm
from male to female prosobranch, with reference,
also, to the pyramidellids. Proceedings of the Lin-
nean Society of London, 164: 217-224.

FRETTER, V., 1980, The evolution of some higher
taxa in gastropods. In: Symposium on the biology
and evolution of Mollusca. Journal of the Mala-
cological Society of Australia, 4: 226-227.

FRETTER, V., 1984, The functional anatomy of the
neritacean limpet Phenacolepas omanensis
Biggs and some comparison with Septaria. Jour-
nal of Molluscan Studies, 50: 8-18.

FRETTER, V. & GRAHAM, A., 1949, The structure
and mode of life of the Pyramidellidae, parasitic
opisthobranchs. Journal of the Marine Biological
Association of the United Kingdom, 28: 493—532.

FRETTER, V. 8 GRAHAM, A., 1962, British proso-
branch molluscs; their functional anatomy and
ecology. Ray Soc., London, xvi + 755 p.

FRETTER, V. & GRAHAM, A., 1978, The proso-
branch molluscs of Britain and Denmark. Part
3—Neritacea, Viviparacea, Valvatacea, terres-
trial and freshwater Littorinacea and Rissoacea.
Journal of Molluscan Studies Supplement 5: 101—
152.

FRETTER, V. & GRAHAM, A., 1981, The proso-
branch molluscs of Britain and Denmark. Part
6—Cerithiacea, Strombacea, Hipponicacea, Ca-
lyptraeacea, Lamellariacea, Cypraeacea, Natica-
cea, Tonnacea, Heteropoda. Journal of Mollus-
can Studies Supplement 9: 285-363.

FRETTER, V. & GRAHAM, A., 1982, The proso-
branch molluscs of Britain and Denmark. Part
7—Heterogastropoda' (Cerithiopsacea, Trifora-
cea [sic], Epitoniacea, Eulimacea). Journal of
Molluscan Studies Supplement 11: 363—434.

GABE, M., 1965, Données morphologiques et his-
tologiques sur l'appareil génital male des Hét-
éropodes (Gastéropodes Prosobranches). Zeit-

352 ROBERT SON

schrift für Morphologie und Okologie der Tiere,
55: 1024-1079.

GHISELIN, М. T., 1963, On the functional and com-
parative anatomy of Runcina setoensis Baba, an
opisthobranch gastropod. Publications of the
Seto Marine Biological Laboratory, 11: 389-398,
pl. 16.

GHISELIN, M. T., 1965, Reproductive function and
the phylogeny of opisthobranch gastropods. Ma-
lacologia, 3: 327-378.

GOLIKOV, А. М. & STAROBOGATOV, Y. 1., 1975,
Systematics of prosobranch gastropods. Malaco-
logia, 15: 185-232.

GOSLINER, T. M., 1981, Origins and relationships
of primitive members of the Opisthobranchia
(Mollusca: Gastropoda). Biological Journal of the
Linnean Society, 16: 197-225.

GRAHAM, A., 1954, Some observations on the re-
productive tract of lanthina [sic] janthina (L.). Pro-
ceedings of the Malacological Society of London,
31: 1-6.

HADFIELD, M. G., unpublished (1966), The repro-
ductive biology of the California vermetid gastro-
pods Serpulorbis squamigerus (Carpenter, 1857)
and Petaloconchus montereyensis Dall, 1919.
Ph.D. dissertation, Stanford University, 174 pp.
[Not seen. Reference from Hadfield & Hopper,
1980.]

HADFIELD, M. G., 1969, Nurse eggs and giant
sperm in the Vermetidae (Gastropoda). American
Zoologist, 9: 1141-1142. Abstract 520.

HADFIELD; М. ©. & HOPPER, ©. №, 1980, Eco-
logical and evolutionary significance of pelagic
spermatophores of vermetid gastropods. Marine
Biology, 57: 315-325.

HADFIELD, М. G. 8 SWITZER-DUNLAP, M., 1984,
Opisthobranchs. In: TOMPA, A. S., VERDONK,
М. H. 8 BIGGELAAR, J. А. M. van den, ed., Re-
production. In: WILBUR, K. M., ed.-in-chief, The
Mollusca, vol. 7, pp. 209-350.

HARRY, H. W., 1964, The anatomy of Chilina fluc-
tuosa Gray reexamined, with prolegomena on
the phylogeny of the higher limnic Basommato-

phora (Gastropoda: Pulmonata). Malacologia, 1:

355-385.

HASZPRUNAR, G., 1985a, The fine morphology of
the osphradial sense organs of the Mollusca. II.
Allogastropoda (Architectonicidae, Pyramidell-
idae). Philosophical Transactions of the Royal
Society of London, ser. В, 307: 497-505, pl. 1.

HASZPRUNAR, G., 1985b, Zur Anatomie und sys-
tematischen Stellung der Architectonicidae (Mol-
lusca, Allogastropoda). Zoologica Scripta, 14: 25—
43.

HASZPRUNAR, G., 1985c, On the anatomy and
systematic position of the Mathildidae (Mollusca,
Allogastropoda). Zoologica Scripta, 14: 201-213.

HASZPRUNAR, G., 1985d, The Heterobran-
chia—a new concept of the phylogeny of the
higher Gastropoda. Zeitschrift fúr zoologische
Systematik und Evolutionsforschung, 23: 15-37.

HASZPRUNAR, G., 1988a, On the origin and evo-
lution of major gastropod groups, with special ref-

erence to the Streptoneura. Journal of Molluscan
Studies, 54: 367-441.

HASZPRUNAR, G., 1988b, A preliminary phyloge-
netic analysis of the streptoneurous gastropods.
In PONDER, W. F., EERNISSE, D. J. & WATER-
HOUSE, J. H., eds. Prosobranch phylogeny, Pro-
ceedings of a Symposium held at the 9th Inter-
national Malacological Congress, Edinburgh,
Scotland, 31 August—6 September, 1986. Mal-
acological Review Supplement 4: 7-16.

HEALY, J. M., 1982, Ultrastructure of spermiogen-
esis of Philippia (Psilaxis) oxytropis, with special
reference to the taxonomic position of the Archi-
tectonicidae (Gastropoda). Zoomorphology, 101:
197-214.

HEALY, J. M., 1986, Euspermatozoa and parasper-
matozoa of the relict cerithiacean gastropod,
Campanile symbolicum (Prosobranchia, Meso-
gastropoda). Helgolánder Meeresuntersuchun-
gen, 40: 201-218.

HEALY, J. M., 1987, Spermatozoan ultrastructure
and its bearing on gastropod classification and
evolution. Australian Zoologist, 24: 108—113.

HEALY, J. M., 1988a, Sperm morphology in Serpu-
lorbis and Dendropoma and its relevance to the
systematic position of the Vermetidae (Gas-
tropoda). Journal of Molluscan Studies, 54: 295—
308.

HEALY, J. M., 1988b, The ultrastructure of sperma-
tozoa and spermiogenesis in pyramidellid gastro-
pods, and its systematic importance. Helgo-
lánder Meeresuntersuchungen, 42: 303-318.

HEALY, J. M. & JAMIESON, B. G. M., 1981, An
ultrastructural examination of developing and ma-
ture paraspermatozoa in Pyrazus ebeninus (Mol-
lusca, Gastropoda, Potamididae). Zoomorphol-
ogy, 98: 101-119.

HEDLEY, C., 1916, Studies on Australian Mollusca.
Part XIII. Proceedings of the Linnean Society of
New South Wales, 41: 680—719, pl. 46-52.

HERTLING, H., 1930, Uber eine Hedylide von Hel-
goland und Bemerkungen zur Systematik der
Hedyliden. Wissenschaftliche Meeresuntersu-
chungen ... Kommission zur wissenschaftlichen
Untersuchung der deutschen Meere in Kiel und
der Biologischen Anstalt auf Helgoland, neue
Folge, Abt. Helgoland, 18, Heft 1, Abhandl. 5:
1-10, 1 pl.

HOISAETER, T., 1965, Spermatophores in Chry-
sallida obtusa (Brown) (Opisthobranchia, Pyra-
midellidae). Sarsia, 18: 63-68.

HOUBRICK, В. S., 1970, Reproduction and devel-
opment in Florida Cerithium. American Mala-
cological Union Annual Reports, “1970”: 74. Ab-
stract.

HOUBRICK, В. 5., 1973, Studies on the reproduc-
tive biology of the genus Cerithium (Gastropoda:
Prosobranchia) in the western Atlantic. Bulletin of
Marine Science, 23: 875-904.

HOUBRICK. R. S., 1974, The genus Cerithium in
the western Atlantic (Cerithiidae: Prosobran-
chia). Johnsonia, 5: 33-84.

HOUBRICK, R. S., 1977, Reevaluation and new

SPERMATOPHORES OF AQUATIC GASTROPODS 353

description of the genus Bittium (Cerithiidae).
Veliger, 20: 101-106, 5 figs.

HOUBRICK, В. S., 1978, The family Cerithiidae in
the Indo-Pacific. Part 1: The genera Rhinoclavis,
Pseudovertagus and Clavocerithium. Mono-
graphs of Marine Mollusca, 1: 1-130.

HOUBRICK, В. S., 1980, Observations on the anat-
omy and life history of Modulus modulus (Proso-
branchia: Modulidae). Malacologia, 20: 117-142.

HOUBRICK, R. S., 1981a, Anatomy and systemat-
ics of Gourmya доигту! (Prosobranchia: Cerithi-
idae), a Tethyan relict from the southwest Pacific.
Nautilus, 95: 2-11.

HOUBRICK, R. S., 1981b, Anatomy of Diastoma
melanioides (Reeve, 1849) with remarks on the
systematic position of the family Diastomatidae
(Prosobranchia: Gastropoda). Proceedings of
the Biological Society of Washington, 94: 598-
621.

HOUBRICK, R. S., 1981c, Anatomy, biology and
systematics of Campanile symbolicum with ref-
erence to adaptive radiation of the Cerithiacea
(Gastropoda: Prosobranchia). Malacologia, 21:
263-289.

HOUBRICK, R. S., 1984, Revision of higher taxa in
[the] genus Cerithidea (Mesogastropoda: Pota-
mididae) based on comparative morphology and
biological data. American Malacological Bulletin,
2: 1-20.

HOUBRICK, R. S., 1985, Genus Clypeomorus
Jousseaume (Cerithiidae: Prosobranchia).
Smithsonian Contributions to Zoology, no. 403:
1-131.

HOUBRICK, R. S., 1987a, Anatomy of Alaba and
Litiopa (Prosobranchia: Litiopidae): systematic
implications. Nautilus, 101: 9-18.

HOUBRICK, R. S., 1987b, Anatomy, reproductive
biology, and phylogeny of the Planaxidae (Cer-
ithiacea: Prosobranchia). Smithsonian Contribu-
tions to Zoology, (445): Ш + 57 pp., 27 figs.

HOUBRICK, R. S., 1988, Cerithioidean phylogeny.
In PONDER, W. F., EERNISSE, D. J. & WATER-
HOUSE, J. H., eds., Prosobranch phylogeny,
Proceedings of a Symposium held at the 9th In-
ternational Malacological Congress, Edinburgh,
Scotland, 31 August—6 September, 1986. Mal-
acological Review Supplement 4: 88-128.

HOUBRICK, R. S., ROBERTSON, R. & ABBOTT,
R. T., 1987, Anatomy and systematic position of
Fastigiella carinata Reeve (Cerithiidae: Proso-
branchia). Nautilus, 101: 101-110.

HOUSTON, R. S., 1985, Genital ducts of the Cer-
ithiacea (Gastropoda: Mesogastropoda) from the
Gulf of California. Journal of Molluscan Studies,
51: 183-189.

HUBENDICK, B., 1945 [19477], Phylogenie und
Tiergeographie der Siphonariidae. Zur Kenntnis
der Phylogenie in der Ordnung Basommatophora
und des Ursprungs der Pulmonatengruppe. Zo-
ologiska Bidrag fran Uppsala, 24: 1-216.

HUBENDICK, B., 1946, Systematic monograph of
the Patelliformia. Kungliga Svenska Vetenskap-
sakademiens Handlingar, (3) 23(5): 1-93, 6 pl.

HUBENDICK, B., 1955, On a small quantity of Si-
phonaria material from Queensland. Memoirs of
the National Museum of Victoria [Melbourne], 19:
126-136.

HUBENDICK, B., 1964, Studies on Ancylidae. The
subgroups. Góteborgs Kungl. Vetenskapsoch
Vitterhets-Samhálles Handlingar, ser. 7, B, 9(6):
1-72 (Meddelanden frán Góteborgs Musei Zool-
ogiska Avdelning, no. 137).

HUBENDICK, B., 1978, Systematics and compar-
ative morphology of the Basommatophora. In:
FRETTER, V. 8 PEAKE, J., eds., Pulmonates . . .
systematics, evolution and ecology, 2A: 1-47.
Academic Press.

HULINGS, N. C., 1986, Aspects of the reproduction
of rocky intertidal mollusks from the Jordan Gulf
of Agaba (Red Sea). Veliger, 28: 318-327.

HUTTON, F. W., 1882, Notes on the structure and
development of Siphonaria australis, Quoy 8
Gaimard. Annals and Magazine of Natural His-
tory, (5)9: 341-344, pl. 15.

HYMAN, L. H., 1967, The invertebrates, 6 (Mol-
lusca 1); Aplacophora, Polyplacophora, Mono-
placophora, Gastropoda; the coelomate Bilateria.
McGraw-Hill, New York, etc., vii + 792 pp.

IHERING, H. von, 1886, Beitráge zur Kenntniss der
Nudibranchien des Mittelmeeres. Il. 4. Polycera-
den. Malakozoologische Blátter, (2)8: 12-48, 2 pl.

IRIKI, S., NISHIWAKI, $. 8 TOCHIMOTO, T., 1963,
On the peculiar mode of the spermatophore
transfer in Nerita albicilla L. (Prosobranchia,
Neritidae)(a prelliminary [sic] note). Venus; Jap-
anese Journal of Malacology, 22: 290-292.

JACKIEWICZ, M., 1986, Pakiet plemników Lym-
naea corvus (Gmelin)(Gastropoda, Basommato-
phora) [A packet of spermatozoa of. . . . ]. Prze-
glad Zoologiczny [Warsaw], 30: 121-122.

JENKINS, В. W., 1981, Siphonaria funiculata
Reeve (Siphonariidae, Pulmonata): a redescrip-
tion making S. virgulata Hedley a geographical
variant of S. funiculata. Journal of the Malacolog-
ical Society of Australia, 5: 1-15.

JENKINS, B. W., 1983, Redescriptions and rela-
tionship of Siphonaria zelandica Quoy and Gaim-
ard to S. australis Quoy and Gaimard with a de-
scription of S. propria sp. nov. (Mollusca:
Pulmonata: Siphonariidae). Journal of the Mala-
cological Society of Australia, 6: 1-35.

JENKINS, B. W., 1984, A new siphonariid (Mol-
lusca: Pulmonata) from southwestern Australia.
Journal of the Malacological Society of Australia,
6: 113-123.

JEWELL, D. D., 1931, Observations on reproduc-
tion in the snail Goniobasis. Nautilus, 44: 115—
119.

JOHANSSON, J., 1946, Von den Geschlechtsorga-
nen bei Turritella communis nebst Bemerkungen
über die diaulen Geschlechtsorgane der Nerita-
ceen. Arkiv fór Zoologi, 38A(12): 1-11.

JOHANSSON, J., 1947, Uber den offenen Uterus
bei einigen Monotocardiern ohne Kopulationsor-
gan. Zoologiska Bidrag fran Uppsala, 25: 102-
110.

354 ROBERTSON

JOHANSSON, J., 1953, On the genital organs of
some mesogastropods: Cerithium vulgatum
Brug., Triphora perversa (L.) and Melanella (Eu-
lima) intermedia (Cantr.). Contributions to the
phylogeny of the pallial gonoducts of the Proso-
branchia. Zoologiska Bidrag fran Uppsala, 30: 1—
23, 1 pl.

JOHANSSON, J., 1956, On the anatomy of Tym-
panotonus fuscatus (L.), including a survey of the
open pallial oviducts of the Cerithiacea. Atlantide
Report 4: 149-166, 1 pl.

KNIGHT, J. B., 1952, Primitive fossil gastropods
and their bearing on gastropod classification.
Smithsonian Miscellaneous Collections, 117(13):
1-56, 2 pl.

KOHLER, A., 1893, Beitráge zur Anatomie der Gat-
tung Siphonaria. Zoologische Jahrbücher.
Abtheilung für Anatomie und Ontogenie der
Thiere, 7: 1-92, pl. 1-6.

KOSUGE, S., 1966, The family Triphoridae and its
systematic position. Malacologia, 4: 297-324.
KRESS, A., 1985a, A structural analysis of the
spermatophore of Runcina ferruginea Kress
(Opisthobranchia: Cephalaspidea). Journal of
the Marine Biological Association of the United

Kingdom, 65: 337-342.

KRESS, A., 1985b, The male copulatory apparatus
in an opisthobranch mollusc, Runcina. Tissue &
Cell, 17: 215-226.

KRESS, A., 1986. Male copulatory apparatus and
formation of spermatophores in a marine mollusc
[Runcina]. 18th Annual Meeting of the Union of
Swiss Societies of Experimental Biology, Basel,
Switzerland, Mar. 20-21, 1986. Experientia, 42:
645—646.

LACAZE-DUTHIERS, H. de, 1899, Des organes de
la reproduction de l'Ancylus fluviatilis. Archives
de Zoologie Expérimentale et Générale, ser. 3, 7:
33-120, pl. 3-8.

LALLI, С. М. & WELLS, Е. E., 1978, Reproduction
in the genus Limacina (Opisthobranchia: Theco-
somata). Journal of Zoology, London, 186: 95—
108.

LENDERKING, R. E., 1954, Some recent observa-
tions on the biology of Littorina angulifera Lam. of
Biscayne and Virginia Keys, Florida. Bulletin of
Marine Science of the Gulf and Caribbean, 3:
273-296.

LEVINGS, S. C. & GARRITY, S. D., 1986, Notes
on reproduction of a tropical pulmonate limpet,
Siphonaria gigas (Sowerby). Veliger, 29: 86-
90.

LIND, H., 1973, The functional significance of the
spermatophore and the fate of spermatozoa in
the genital tract of Helix pomatia (Gastropoda:
Stylommatophora). Journal of Zoology, London,
169: 39-64.

LISTER, M., 1694, Exercitatio anatomica. In qua de
cochleis, maxime terrestribus £ Limacibus, agi-
tur. Omnium dissectiones . . . illustrantur. . . .
London, pp. xi + 208, 8 pl.

LUPU, D., 1979, Tanzaniella gen. n. (Gastropoda,
Neritidae) from the coast of Indian Ocean.

Travaux du Muséum d'Histoire Naturelle
“Grigore Antipa” [Bucharest], 20(1): 15-19, 5 pl.

MANN, T., 1984, Spermatophores; development,
structure, biochemical attributes and role in the
transfer of spermatozoa. Zoophysiology 15.
Springer-Verlag, xii + 217 pp.

MARCUS, E., 1953, Three Brazilian sand-
Opisthobranchia. Boletim da Faculdade de
Filosofia, Ciéncias e Letras da Universidade de
Sao Paulo, 165 (Zoologia 18): 165-203.

MARCUS, E., & MARCUS, E. [du B.-R.], 1963,
Mesogastropoden von der Kúste Sáo Paulos.
Akademie der Wissenschaften und der Literatur
[п Mainz]; Abhandlungen der mathematisch-
naturwissenschaftlichen Klasse, 1963(1): 1-105.

MARCUS, E. du B.-R., 1974. On some Cephalaspi-
dea (Gastropoda: Opisthobranchia) from the
western and middle Atlantic warm waters. Bulle-
tin of Marine Science, 24: 300-371. :

MARCUS, E. [du B.-R.] & MARCUS, E., 1954, Uber
Philinoglossacea und Acochlidiacea. Kieler
Meeresforschungen, 10: 215-223, pl. 26.

MARCUS, E. [du B.-R.] & MARCUS, E., 1960a,
Opisthobranchs from American Atlantic warm
waters. Bulletin of Marine Science ofthe Gulf and
Caribbean, 10: 129-203.

MARCUS, E. [du B.-R.] & MARCUS, E., 1960b, On
Siphonaria hispida. Universidade de Sao Paulo
Faculdade de Filosofia, Ciéncias e Letras, Bole-
tim 260 (Zoología 23): 107-130, 4 pl.

MEDEM, F. von, 1945, Untersuchungen úber die El
und Spermawirkstoffe bei marinen Mollusken.
Zoologische Jahrbúcher. Abteilung fúr allge-
meine Zoologie und Physiologie der Tiere, 61:
1-44, 2 pl.

MELONE, G., DONIN; С. 1. №. & COMELLIAES
1978, Aspetti ultrastrutturali degli spermatozoi at-
ipici in Scalidae (Gastropoda, Prosobranchia).
Bollettino di Zoologia, 45: 261-268.

MELONE, G., BONIN, С. 1. E & СОТЕН Е,
1980, The paraspermatic се! (atypical sperma-
tozoon) of Prosobranchia: a comparative ultra-
structural study. Acta Zoologica, 61: 191-201.

MERRILL, A. S., unpublished (1970), The family
Architectonicidae (Gastropoda: Mollusca) in the
western and eastern Atlantic. Ph.D. dissertation,
University of Delaware, [xi +] 338 pp., 42 pl.

MINICHEV, Yu. $. & STAROBOGATOV, Ya. I.
1979, The subclasses of Gastropoda and their
phylogenetic relations. Zoologicheskii Zhurnal,
58: 293-305. In Russian.

MOORE, R. C., ed., 1960 [authors include J. B.
Knight, L. R. Cox, and А. M. Keen], Mollusca |, In:
Treatise on invertebrate paleontology, part I, xxiii
+ 351 pp. Geol. Soc. America and Univ. Kansas
Press.

MOQUIN-TANDON, A. 1852, Recherches
anatomico-physiologiques sur ГАпсуе fluviatile
(Ancylus fluviatilis, Mull.) (Suite et fin.). Journal
de Conchyliologie, 3: 337-357.

MOQUIN-TANDON, A., 1855, Histoire naturelle
des Mollusques terrestres et fluviatiles de
France... . 2 vol., Atlas. Paris.

SPERMATOPHORES OF AQUATIC GASTROPODS 355

MORSE, M. P., 1976a, Observations on interstitial
opisthobranchs along the New England coast.
Journal of Molluscan Studies, 42: 302.

MORSE, M. P., 1976b, Hedylopsis riseri sp. n., a
new interstitial mollusc from the New England
coast (Opisthobranchia, Acochlidiacea). Zoolog-
ica Scripta, 5: 221-229.

MORTON, y. E., 1955, The evolution of the Ellobi-
idae with a discussion on the origin of the Pul-
monata. Proceedings of the Zoological Society of
London, 125: 127-168.

MORTON, J. E. & YONGE, C. M., 1964, Classifi-
cation and structure of the Mollusca. In: WILBUR,
К. М. 8 YONGE, С. M., eds., Physiology of Mol-
lusca, 1: 1-58. Academic Press.

NEWMAN, L. J., 1986, Snails with clear ‘plastic’
shells. Australian Shell News, no. 55: 1-2.

ODHNER, N. H., 1937, Hedylopsis suecica n. sp.
und die Nacktschneckengruppe Acochlidiacea
(Hedylacea). Zoologischer Anzeiger, 120: 51-64.

PACE, G. L., 1973, The freshwater snails of Taiwan
(Formosa). Malacological Review Suppl. 1: 1—
118.

PERRIER, R. & FISCHER, H., 1914, Sur l’exis-
tence de spermatophores chez quelques
Opisthobranches. Comptes Rendus ... de ГАса-
démie des Sciences [Paris], 158: 1366-1369.

PILSBRY, H. A., 1895, Manual of Conchology,
(1)15(60): 181—436, pl. 43-50, 59-61. Academy
of Natural Sciences of Philadelphia.

POHL, H., 1905, Uber den feinern Bau des Geni-
talsystems von Polycera quadrilineata. Zoologis-
che Jahrbúcher, Abt. Anatomie und Ontogenie
der Tiere, 21: 427-452, pl. 25-26.

POIZAT, C., 1981, Gastéropodes mesopsam-
miques de la Mer du Nord (Robin Hood's Bay,
U.K.); écologie et distribution. Journal of Mollus-
can Studies, 47: 1-10.

POIZAT, C., 1987, Interstitial opisthobranch gastro-
pods: ecological and biological account. Western
Society of Malacologists Annual Report, 19: 19—
22.

POIZAT, C., “in press,” Life history observations
on a few interstitial opisthobranch gastropods
from the Gulf of Marseilles, Bouches-du-Rhone,
France. Boletim de Zoologia, Sáo Paulo, Brazil.
[Not seen. Reference from Poizat (1987).]

PONDER, W. F., 1988, Bioluminescence in Hinea
braziliana (Lamarck) (Gastropoda: Planaxidae).
Journal of Molluscan Studies, 54: 361.

PONDER, W. Е. & WAREN, A., 1988, Classifica-
tion of the Caenogastropoda and Heterostro-
pha—A list of the family-group names and higher
taxa. Appendix in PONDER, W. F., EERNISSE,
D. J. & WATERHOUSE, J. H., eds., Proso-
branch phylogeny, Proceedings of a Symposium
held at the 9th International Malacological
Congress, Edinburgh, Scotland, 31 August—6
September, 1986. Malacological Review Supple-
ment 4: 288-328.

QUOY, [J. R. C.] & GAIMARD, [J. P.], 1833-1834,
Voyage de... l'Astrolabe. . . . Zoologie, 3, Mol-
lusques, Atlas, 93 pl.; 366 pp. Paris.

RANKIN, J. J., 1979, A freshwater shell-less mol-
lusc from the Caribbean: structure, biotics, and
contribution to a new understanding of the Aco-
chlidioidea. Life Sciences Contributions Royal
Ontario Museum, 116: 1-123.

REID, D. G., 1986, The littorinid molluscs of man-
grove forests in the Indo-Pacific region; the ge-
nus Littoraria. British Museum (Natural History),
London, xv + 228 pp.

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

RISBEC, J., 1942, Recherches anatomiques sur
les prosobranches de Nouvelle-Calédonie (part
3). Annales des Sciences Naturelles, Zoologie,
(11) 4: 21-84.

RISBEC, J., 1955, Considérations sur l'anatomie
comparée et la classification des Gastéropodes
Prosobranches. Journal de Conchyliologie, 95:
45-82.

ROBERTSON, R., 1966, The life history of Odos-
tomia bisuturalis, and Odostomia spermato-
phores (Gastropoda: Pyramidellidae). Year Book
of The American Philosophical Society, “1966”:
368—370.

ROBERTSON, R., 1967, Heliacus (Gastropoda: Аг-
chitectonicidae) symbiotic with Zoanthiniaria (Co-
elenterata). Science, 156: 246-248.

ROBERTSON, R., 1968, Hosts, spermatophores,
and the systematics of five east American spe-
cies of Odostomia, 5.1. (Pyramidellidae). Ameri-
can Malacological Union Annual Reports “1967”
Bulletin 34: 12-13. Abstract.

ROBERTSON, R., 1973, The biology of the Archi-
tectonicidae, gastropods combining prosobranch
and opisthobranch traits. Proceedings of the
Fourth European Malacological Congress. Mala-
cologia, 14: 215-220.

ROBERTSON, R., 1978, Spermatophores of six
eastern North American pyramidellid gastropods
and their systematic significance (with the new
genus Boonea). Biological Bulletin, 155:
360-382.

ROSERTSON, R., 1983, Observations on the life
history of the wentletrap Epitonium albidum in the
West Indies. American Malacological Bulletin, 1:
1-11.

ROBERTSON, R., 1985, Four characters and the
higher category systematics of gastropods. In:
Perspectives in malacology, honoring Professor
Melbourne R. Carriker. American Malacological
Bulletin, special ed. 1: 1-22.

SALVINI-PLAWEN, L. von, 1980, A reconsidera-
tion of systematics in the Mollusca (phylogeny
and higher classification). Malacologia, 19: 249—
278.

SCHEUWIMMER, A., 1979, Sperm transfer in the
sessile gastropod Serpulorbis (Prosobranchia:
Vermetidae). Marine Ecology—Progress Series,
1: 65-70.

SCHEUWIMMER, A. & NISHIWAKI, S., 1982,
Comparative studies on three Japanese species
of Serpulorbis (Prosobranchia: Vermetidae) with

356 ROBERTSON

description of a new species. Venus; Japanese
Journal of Malacology, 41: 85-101.

SCHMEKEL, L. & РОВТМАММ, A., 1982, Opistho-
branchia des Mittelmeeres; Nudibranchia und
Saccoglossa. Springer-Verlag, Berlin, Heidel-
berg, New York, x + 410 pp.

SPOEL, $. van der, 1972, Notes on the identifica-
tion and speciation of Heteropoda (Gastropoda).
Zoologische Mededelingen (Rijksmuseum van
Natuurlijke Historie te Leiden), 47: 545-560.

SPOEL, S. van der, 1976, Pseudothecosomata,
Gymnosomata and Heteropoda (Gastropoda).
Bohn, Scheltema & Holkema, Utrecht, 484 pp.

STARMÜHLNER, F., 1969, Ergebnisse der Öster-
reichischen Madagaskar-Expedition 1958. Die
Gastropoden der madagassischen Binnenge-
wässer. Schale, Makro- und Mikroanatomie des
Weichkôrpers, Lebensraum und geographische
Verbreitung. Malacologia, 8: 1-434.

SUMIKAWA, S. & ONIZUKA, S., 1973, On the
spermatophore of the siphone snail, Siphonaria
sirius. Seikatsu Kagaku = Science of Human
Life (Fukuoka [Japan] Women's University), 9:
97-100. In Japanese; English figure captions
and summary.

SWEDMARK, B., 1964, The interstitial fauna of ma-
rine sand. Biological Reviews, 39: 1-42.

SWEDMARK, B., 1968a, Deux especes nouvelles
d'Acochlidiacées (Mollusques Opisthobranches)
de la faune interstitielle marine. Cahiers de Biol-
ogie Marine, 9: 175-186, 2 pls.

SWEDMARK, B., 1968b, The biology of interstitial
Mollusca. In: FRETTER, V., ed., Studies in the
structure, physiology and ecology of molluscs.
Symposia of the Zoological Society of London,
22: 135-149.

TARDY, J., 1965, Spermatophores chez quelques
espèces d'Aeolididae (Mollusques Nudi-
branches). Comptes Rendus des Séances de la
Société de Biologie, 160: 369-371.

TAYLOR, D. W. & SOHL, N. F., 1962, An outline of
gastropod classification. Malacologia, 1: 7-32.
TESCH, J. J., 1949, Heteropoda. Dana-Report, 34:

1-53, 5 pls.

THIELE, J., 1929, Handbuch der systematischen
Weichtierkunde, Teil 1: 1-376. Fischer, Jena.
THIELE, J., 1931, Handbuch der systematischen

Weichtierkunde, Teil 2: 377-778. Fischer, Jena.

THIRIOT-QUIEVREUX, C., 1967, Descriptions de
quelques véligeres planctoniques de Gastérop-
odes. Vie et Milieu, ser. A (Biologie Marine), 18:
303-315.

THIRIOT-QUIÉVREUX, C. 8 MARTOJA, M., 1974,
Appareil génital femelle des Atlantidae (Mollusca
Heteropoda). |. —Etude anatomique. Vie et Mi-
lieu, ser. A (Biologie Marine), 24: 389-411.

THIRIOT-QUIEVREUX, C. & MARTOJA, M., 1976,
Appareil génital femelle des Atlandidae [sic]
(Mollusca, Heteropoda). Il. Etude histologique
des structures larvaires, juvéniles et adultes.
Données sur la fécondation et la ponte. Vie et
Milieu, ser. A (Biologie Marine), 26: 201-233.

THOMPSON, T. E., 1976, Biology of opisthobranch

molluscs, vol. 1, Ray Soc., London, 206 pp., 8 pl.
(color).

THOMPSON, T. E. 8 BROWN, G. H., 1984, Biology
of opisthobranch molluscs, vol. 2, Ray Soc., Lon-
don, 229 pp., 41 pl. (most color).

THORSON, G., 1946, Reproduction and larval de-
velopment of Danish marine bottom inverte-
brates, with special reference to the planktonic
larvae in the Sound (@resund). Meddelelser fra
Kommissionen for Danmarks Fiskeri- og Ha-
vundersogelser, ser. Plankton, 4: 1-523.

TILLIER, S., 1984, Relationships of gymnomorph
gastropods (Mollusca: Gastropoda). Zoological
Journal of the Linnean Society, 82: 345-362.

TROSCHEL, F. H., 1861, Ueber die systematische
Stellung der Gattung Solarium. Archiv für Naturg-
eschichte, Jahrgang 27, 1: 91-99, pl. 5.

VOSS, N. A., 1959, Studies on the pulmonate gas-
tropod Siphonaria pectinata (Linnaeus) from the
southeast coast of Florida. Bulletin of Marine Sci-
ence of the Gulf and Caribbean, 9: 84-99.

WAWRA, E., 1986, Geschlechtsdimorphismus bei
Pontohedyle milachewitchii (Kowalevevsky,
1901) (Gastropoda: Opisthobranchia). Annalen
des Naturhistorischen Museums in Wien, 87B:
325-329.

WELLS, F. E., 1978, Subgeneric relationships in
the euthecosomatous pteropod genus Limacina
Bosc, 1817. Journal of the Malacological Society
of Australia, 4: 1-5.

WENZ, W., 1938-1944. Gastropoda. In: SCHINDE-
WOLF, O. H., ed., Handbuch der Paläozoologie,
6(1-7), 1639 pp. Berlin.

WESTHEIDE, W. & WAWRA, E., 1974, Organisa-
tion, Systematik und Biologie von Microhedyle
cryptophthalma nov. spec. (Gastropoda, Opistho-
branchia) aus dem Brandungsstrand des Mittel-
meeres. Helgoländer Wissenschaftliche Meere-
suntersuchungen, 26: 27-41.

WOODARD, Т. M. [Yr.], 1934, Anatomy of the re-
productive system of Goniobasis laqueata (Say).
Journal of the Tennessee Academy of Science,
9: 243-259.

WOODARD, T. M. [Jr.], 1935, Spermic dimorphism
in Goniobasis laqueata (Say). Journal of Mor-
phology, 57: 1-29.

WOODARD, T. M., Jr., 1940, The function of the
apyrene spermatozoa of Goniobasis laqueata
(Say). |. The behavior of the apyrene and
eupyrene spermatozoa under natural and artifi-
cial conditions. Journal of Experimental Zoology,
85: 103-125.

WOODARD, T.M., Jr., 1942, Behavior of the nurse-
cells of Littorina irrorata (Say). Biological Bulletin,
82: 461-466.

YONGE, С. M., 1946, On the habits of Turritella
communis Risso. Journal of the Marine Biological
Association of the United Kingdom, 26: 377-380.

ZILCH, A., 1959-1960, Gastropoda. In: SCHINDE-
WOLF, O. Н., ed., Handbuch der Paläozoologie,
6, Teil 2 Euthyneura, xii + 834 pp. Berlin.

Revised Ms. accepted 24 January 1989

SPERMATOPHORES OF AQUATIC GASTROPODS 357

APPENDIX

Systematic list of the genera and higher
taxa of non-stylommatophoran gastropods
known to have spermatophores.

The classification, sequencing of taxa, and
nomenclature here and in the text are from
such orthodox sources as Thiele (1929-
1931) and Taylor & Sohl (1962), plus some
innovative recent sources, including Morton &
Yonge (1964), Houbrick (1977 on), Huben-
dick (1978), and especially Ponder & Warén
(1988) for the Caenogastropoda and Thomp
son (1976) for the Opisthobranchia. Taxa are
arranged alphabetically within superfamilies
and families. Here and elsewhere in this pa-

per, taxonomic nomenclature has been up-
dated where necessary.
Abbreviations:

aph = aphallic

c. pen = cephalic “penis”-bearing
gon = gonochoric (unisexual)
herm = hermaphroditic

pen = penis-bearing

Notes A1 to A39 explain dubious or erro-
neous reports and other observations. Some-
what doubtful cases are listed in the table but
with “(?)” after the taxonomic name. Notes B1
to B23 quote the authors concerning the an-
atomical sites of formation of the spermato-
phores they studied.

358 ROBERTSON

Taxa Sex Penis Notes References
PROSOBRANCHIA gon aph
Archaeogastropoda

Fiurelloidea (?)
Fissurellidae (?)

Fissurella (?) А1 В1 Medem, 1945
Neritimorpha gon c.pen,
rarely
aph
Neritidae B2 Quoy & Gaimard, 1832
Clithon Andrews, 1937;
Starmühiner, 1969;
Pace, 1973
Fluvinerita Andrews, 1937
Nerita A2 B3 Bergh, 1890;

Bourne, 1908;
Andrews, 1937;
Risbec, 1942;
Iriki et al., 1963;
Berry et al., 1973;
Hulings, 1986
Neritilia aph Andrews, 1937
Neritina B4 Bergh, 1890;
Bourne, 1908;
Andrews, 1936,
1937; Starmühlner,
1969; Pace, 1973

Puperita Andrews, 1937
A3
Smaragdia aph A4 B5 Andrews, 1937; Lupu,
1979
B5
Theodoxus A5 Andrews, 1937
Phenacolepadidae
Phenacolepas B6 Thiele, 1929;
Fretter, 1984
Caenogastropoda gon,
rarely
herm
Neotaenioglossa
Discopoda
A6
Cerithioidea gon aph . A7
Batillariidae A8
[Batillaria] Houbrick, 1988
Campanilidae (?)
Campanile (?) A9 Houbrick, 1981c
Cerithideidae A8
Cerithidea B7 Houbrick, 1984
Cerithiidae
Bittium B8 Marcus & Marcus,
1963;
Houbrick, 1977
Cerithium A10 B9 Houbrick, 1970, 1973,
1974
Clypeomorus Houbrick, 1985
Fastigiella (?) A11 Houbrick et al., 1987
Gourmya Houbrick, 1981a
Rhinoclavis A12 Houbrick, 1980, 1985

Diastomatidae

Taxa

Diastoma
Litiopidae

Alaba

Litiopa

Modulidae
Modulus
Pachychilidae

(“Pleuroceridae”)

Goniobasis

Pleurocera (?)
Planaxidae

Fissilabia
Hinea
Planaxis
Supplanaxis
Potamididae
Terebralia

Turritellidae
Turritella

Vermetoidea
Vermetidae
Dendropoma

Petaloconchus

Serpulorbis

Vermetus
Heteropoda
(“Atlantoidea”)
Carinarioidea

Atlantidae

Atlanta

Oxygyrus

Carinariidae
Carinaria
Firolidae

(“Pterotracheidae”)

Firoloida
Pterotrachea

Sex

gon

Penis

pen

Notes

SPERMATOPHORES OF AQUATIC GASTROPODS

References

A13

A14

A15

A16
A17

A18

A19

A20
A21

A21

B10

B11

B12

B13

B15

Houbrick, 1981b

Houbrick, 1987a
Houbrick, 1987a

Houbrick, 1980

Jewell, 1931;
Woodard, 1934,
1935, 1940;
Dazo, 1965
Dazo, 1965

Houbrick, 1987b
Houbrick, 1987b
Houbrick, 1987b

Houbrick, in litt.

Johansson, 1946;
Houbrick, 1988

Hadfield &
Hopper, 1980
Hadfield,
unpublished, and
1969; Hadfield &
Hopper, 1980
Hadfield,
unpublished, and
1969; Scheuwimmer,
1979; Hadfield &
Hopper, 1980;
Scheuwimmer &
Nishiwaki, 1982
Bieler, т litt.

Spoel, 1972

Tesch, 1949;
Spoel, 1972, 1976;
Thiriot-Quiévreux 8
Martoja, 1974, 1976
Tesch, 1949;
Newman, 1986

Spoel, 1972

Spoel, 1972
Spoel, 1976

359

360

Taxa

Sex

Ptenoglossa
Triphoroidea
Triphoridae
Mastonia (?)
Triphora, $.1.

HETEROBRANCHIA

Heterostropha
Architectonicoidea

Architectonicidae
Heliacus
Pyramidelloidea

Pyramidellidae
Boonea

Fargoa
Parthenina

OPISTHOBRANCHIA

Bullomorpha
Retusidae (7?)
Rhizorus (?)

Haminoeidae
Haminoea

Scaphandridae
Cylichna

Philinidae (?)
Philine (?)
Runcinidae
Metaruncina
Runcina
Thecosomata

Limacinidae
Limacina

Acochlidiomorpha

Hedylopsidae

Hedylopsis

Microhedylidae
Ganitus

gon

herm
or
gon

herm

herm,

rarely
gon

herm

herm
or
gon
herm

gon

ROBERTSON

Penis

aph

pen or
aph

pen
aph?

pen or
aph

aph or
pen

aph

Notes

A22

A23

A24

A26

A28

A29

A30

A31

B14

B16

A25

A25
A27

B17

B18

B19

B20

B21

References

Kosuge, 1966
Marcus € Marcus,
1963

Robertson,
1973,this paper

Robertson, 1966,
1968, 1978
Robertson, 1966,
1968, 1978
Höisaeter, 1965

Marcus & Marcus,
1960a

Perrier 8
Fischer, 1914

Perrier €
Fischer, 1914

Forster, 1934

Ghiselin, 1963,
1965; Baba, 1967
Kress, 1985a,
1985b, 1986

Lalli &
Wells, 1978

Swedmark, 1968a,
1968b; Morse, 1976a,
1976b

Marcus, 1953;
& Marcus |
Marcus, 1954 |

SPERMATOPHORES OF AQUATIC GASTROPODS 361

Taxa Sex Penis

Microhedyle

Paraganitus
Pontohedyle

Unela

Nudibranchia herm pen

Aeolidiidae
Aeolidia
Aeolidiella

PULMONATA herm
Basommatophora
Siphonarioidea
Siphonariidae
Siphonaria “pen”

Lymnaeoidea (?) pen
Chilinidae (?)
Chilina (?)
Lymnaeidae (?)
Stagnicola (?)

Notes References

A32 Hertling,
1930;
Swedmark, 1964,
1968a, 1968b;
Westheide &
Wawra, 1974; Poizat,
1981, in press;
Challis, 1968

A33 Wawra, 1986;
Poizat, 1987,
in press
Marcus &
Marcus, 1954;
Doe, 1974; Hadfield
& Switzer-Dunlap,
1984; Poizat, 1987,
in press

A34

B22 Tardy, 1965
B22 Tardy, 1965
A35

A36 B23 Hutton, 1882;
A37 Köhler, 1893;

Abe, 1940;
Hubendick,
1945, 1946,
1955; Allanson, 1958;
Sumikawa &
Onizuka, 1973;
Jenkins, 1981,
1983, 1984;
Levings &
Garrity, 1986

A38 Harry, 1964
A38 Jackiewicz,

1986
A39

NOTES A1 TO A39

Miscellaneous observations.

A1. These Fissurella “Spermatophoren”
were wrongly considered “Spermiozeugmen”
by Marcus & Marcus (1963). The bounding
structures are remnants of the (multicellular)
testicular epithelium (Medem, 1945), which is
odd for a spermatophore.

A2. In Nerita albicilla, a gelatinous “transfer

tube,” manipulated by the cephalic penis,
transfers a spermatophore once per “copu-
lation” (Iriki et al., 1963).

A3. Spermatophores were not observed in
Septaria by Bourne (1908), Andrews (1937),
or Starmühlner (1969).

A4. Smaragdia (as its synonym Tanzaniella)
lacks a cephalic penis (Lupu, 1979). The latter

362 ROBERTSON

genus was erected solely on the basis of (to
me unconvincing) radular differences.

A5. Curiously, Fretter (1946) and Fretter &
Graham (1962, 1978) did not observe or men-
tion a spermatophore in Theodoxus fluviatilis
of Europe, yet one was reported by Ankel
(1936) and Andrews (1937).

A6. LITTORINOIDEA: LITTORINIDAE: Lit-
toraria: Spermatophores have been reported
wrongly, first by Woodard (1942), then by
Lenderking (1954), and lastly by Reid (1986).
Woodard, who wrote of “rudimentary” sper-
matophores, observed aggregated eusperma-
tozoa with no bounding structures. Lenderk-
ing saw and drew a “nurse cell,” which is the
kind of paraspermatozoon present in littorin-
ids. “Nährzellen” (nurse cells) were equated
with “Spermiozeugmen” by Marcus & Marcus
(1963). Nurse cells with attached eupyrene
spermatozoa were called spermatozeug-
mates by Reid, which “are transferred to the
female in a mucous spermatophore.” Mucus
seems too flimsy a bounding substance to
qualify as forming a spermatophore. The ex-
tensive literature on littorinid nurse cells has
been reviewed by Reid (1986).

A7. Mention of a “spermatophore bursa”
(or “chamber”) in the open pallial oviduct of a
female cerithioidean is here taken to indicate
that the taxon has spermatophores.

A8. Houbrick (1988) erected these two fam-
ilies. He introduced the Batillariidae without
mentioning the presumably sole genus.

A9. In Campanile, “remains of what ap-
peared to be a disintegrating spermatophore
were found in the female oviducal groove;
thus, the male pallial gonoduct may also se-
crete spermatophores, but this needs confir-
mation” (Houbrick, 1981c). On the phyloge-
netic relationships of Campanile see also
Healy (1986) and Houbrick (1988).

A10. In 1973, Houbrick thought that of the
six western Atlantic species of Cerithium that
he studied alive, only C. muscarum produces
spermatophores. He stated that “C. [erithium]
variabile lacks spermatophore-forming glands
in its gonadal ducts and does not produce
spermatophores as does C. muscarum.” This
was before Houbrick discovered many more
cerithioidean spermatophores, and the state-
ment now seems questionable.

A11. Spermatophores and a spermato-
phore bursa were not seen in the single Fas-
tigiella available for study by Houbrick et al.
(1987), “but [a bursa’s] function may be taken
over by the large closed portion of the poste-
rior pallial oviduct.”

A12. In his paper on Modulus, Houbrick
(1980) mentioned that he had also seen sper-
matophores in Rhinoclavis. He had not men-
tioned these in his 1978 monograph of Rhin-
oclavis. He observed spermatophores in R.
fasciata and R. vertagus when he was at the
Lizard Island Research Station, Great Barrier
Reef, Australia, in 1979 (Houbrick in litt.).

A13. Houbrick (1988) mentioned the MEL-
ANOPSIDAE as having spermatophores, but
gave no details.

A14. Dazo (1965) appears not to have seen
spermatophores in Pleurocera, but he did see
the “spermatophore organ.”

A15. Angiola is synonymized with Hinea
following Ponder (1988).

A16. According to Fretter & Graham
(1981), “some thiarids” (THIARIDAE) have
spermatophores, but maybe they were includ-
ing the Pachychilidae in the Thiaridae.

A17. TURRITELLIDAE: Turritella was be-
lieved by Fretter (1946, 1953) to have “current
fertilization” (like that of nonneritoidean ar-
chaeogastropods). Since Turritella is such a
sedentary mud-dweller (Yonge, 1946; Fretter
8 Graham, 1962, 1981), and since all cerithio-
ideans may have spermatophores, sperm
transfer in Turritella needs further investiga-
tion. However, in a soon-to-be-published
study of the reproduction of Vermicularia
spirata (also Turritellidae), Bieler & Hadfield
did not find spermatophores (Bieler, in /itt.).

A18. On the basis of sperm morphology and
published data on other characters, Healy
(1988a) advocated removal of the VER-
METIDAE from the Cerithioidea and place-
ment in its own superfamily Vermetoidea
(erected by Rafinesque in 1815). Vermetid
sperm morphology may in some way be con-

‘nected with the unique planktonic spermato-

phores of this family. Ponder & Warén (1988)
ranked the Vermetidae in the Vermetoidea but
Houbrick (1988) kept it in the Cerithioidea.

A19. A “flagellum” in prosobranchs can be
any whip-like structure, e.g. the one in hetero-
pods (Gabe, 1965).

A20. Gabe (1965) disbelieved Tesch's
(1949) report of spermatophores in atlantids,
believing that there were no structures bound-
ing the spermatozoa. Subsequently, Tesch
was shown to be right.

A21. In Atlanta and Oxygyrus the sper-
matophore is attached to the outside of the
shell (Tesch, 1949).

A22. “Since [in Mastonia] a penis is miss-
ing, the spermatheca are conveyed directly to
the female mantle cavity” (Kosuge, 1966, my

SPERMATOPHORES OF AQUATIC GASTROPODS 363

italics). Since both spermatophores and sper-
matozeugmata are known in the TRI-
PHORIDAE (the latter in a “Triphora” and a
Viriola: Houston, 1985; Healy, 1987), Kosuge
may have meant either. Viriola spermato-
zeugmata appear formless.

A23. Troschel (1861) made two erroneous
statements: that Architectonica has separate
sexes, and that it has a large penis that
projects from the mantle cavity on the right
side. His fig. 1 does not show this.

A24. In Rhizorus “it seems that a spermato-
phore is produced” (Marcus & Marcus,
1960a, my italics).

A25. BULLIDAE and PHILINOGLOSSIDAE
both have spermatophores according to Ghis-
elin (1965), but without documentation.

A26. Haminea [Haminoea] hydatis L. and
H. arachis Quoy 8 Gaimard of Perrier 8 Fis-
cher (1914) are not congeneric or even con-
familial. The former is the type species of
Haminoea (HAMINOEIDAE) and the latter a
Cylichna (SCAPHANDRIDAE) (Pilsbry, 1895;
Thompson, 1976).

A27. PHILINOGLOSSIDAE: Philinoglossa:

. the lack of a penis suggests that
autosperm are transmitted during copulation
in a spermatophore.” (Challis, 1969, my ital-
ICS).

A28. According to Marcus (1974), Förster
(1934) considered that in Philine aperta “the
noncoiled but sausage-shaped prostatic
gland [15]... a spermatophore-forming gland
... in which [are] . . . found masses of sperm
in the glandular lumen included in a loose
capsule secreted by the gland cells of the
'prostate.'” Seven malacologists in eight ear-
lier and later publications did not mention
spermatophores in P. aperta.

A29. In Limacina, five species have penes
and no spermatophores; the sixth, L. inflata,
is aphallic and has spermatophores (Lalli 8
Wells, 1978). Wells (1978) considered the dif-
ference subgeneric.

A30. Rankin's (1979) systematics of the
Acochlidioidea is not accepted here for the
reasons given by Thompson & Brown (1984).

A31. In Hedylopsis, H. suecica has a penis
(Odhner, 1937) and no spermatophores.
Other known species in this genus do not
have a penis but at least some have sper-
matophores (Morse, 1976).

A32. Hertling (1930) was the first to de-
scribe and illustrate an acochlidiomorph sper-
matophore (in Microhedyle), but he did not
recognize it as such. He thought he saw an
injury, and that the last, thin part of the vas

deferens could evert and function as a penis
(Sperma-Ubertrager).

A33. Pontohedyle: see also under Micro-
hedyle above.

A34. Ihering (1886) reported spermato-
phores in Polycera quadrilineata (POLYCER-
IDAE). Pohl (1905), who studied the genital
system of the same species in greater detail
and who did not cite Ihering (1886), wrote of
elastic “Spermapatrone” (sperm cartridges)
formed of sperm surrounded by a thin layer of
prostatic secretion. Schmekel & Portmann
(1982) called them sperm balls. These do not
qualify as spermatophores.

A385. EMBLETONIIDAE: Embletonia: About
eight times, Chambers (1934) mentioned a
“sperm mass” and once a “sperminal mass”
or “spermatophore” [his internal quotes] (p.
634). These masses seem not to have been
contained in any structures, in which case
they would not have been spermatophores. In
an earlier place, Chambers had described
“the sperm mass [as] a tangled bundle of
spermatozoa embedded in the prostatic se-
cretion.”

A36. At least one species of Siphonaria ap-
pears to lack spermatophores. It seems un-
likely that Kohler (1893), Dieuzeide (1935),
Hubendick (1946), and Voss (1959) all
missed them in S. pectinata. Kohler studied
this species in most detail, but reported sper-
matophores in three other species. Also, Mar-
cus & Marcus (1960b) did not find them in S.
hispida.

A37. Co-occurrence of a spermatophore
and a penis in Siphonaria was reported by
Hutton (1882), Köhler (1893), and Abe
(1940), but according to Hubendick (1945)
there is no “true” penis.

A38. The reports of a “spermatophore” in a
Chilina and a “packet” of spermatozoa in a
Stagnicola are both based on one or two ob-
servations of a structure in the reproductive
tract. Confirmation is needed that they func-
tion as spermatophores.

A39. ANCYLIDAE: Ancylus: Moquin-
Tandon (1852, 1855) described and illus-
trated a “capréolus” (spermatophore). Lacaze-
Duthiers (1899) showed that Moquin-Tandon
illustrated the projecting tip of the penis. (See
also Hubendick, 1964.)

NOTES B1 TO B23

Genera in which the anatomical site of
spermatophore formation is recorded, how-
ever vaguely, misleadingly, or tentatively.

364 ROBERTSON

B1. Fissurella: testis (Medem, 1945, trans-
lated).

B2. NERITIDAE: “The cavity of the defer-
ent duct enlarges as the ‘thalamus’ and then
ends as the ‘terminal chamber.’ The sperm
that enters the thalamus meets the secretion
of the very large prostate gland and... this
secretion makes the inner tube of the
spermatophore. . . . Farther along, the basal
gland may make the outer case of the sper-
matophore, while other glands lining the ter-
minal chamber secrete granules that make
much of the bulk of the spermatophore” (An-
drews, 1937).

B3. Nerita: prostate complex: in “a channel
lined by mucoid-secreting and tall ciliated
cells pass[ing] anteriorly from the right limb of
the prostatic lumen . . .” the “horny” inner
coat [of the spermatophore is secreted in] the
annex gland” [part of the “prostate”] (Berry et
al., 1973).

B4. Neritina: “one part arising from the
basal gland and another from the thalamus,
and both joined together by the body more or
less imperfectly filled with the granules se-
creted by the knob of the ridge and adjacent
walls” (Andrews, 1936).

B5. Smaragdia: “deferent canal” (Lupu,
1979).

B6. Phenacolepas: presumably “trough-
shaped ventral channel. . . [in the] . . . pros-
tate . . . [and] leading forwards to the genital
papilla opening to the mantle cavity” (Fretter,
1984).

B7. Cerithidea: “proximal third [of pallial
gonoduct] white, thick and... . glandular [and
is] probably the prostate-spermatophore
forming gland” (Houbrick, 1984).

B8. Bittium: ‘male pallial gonoduct”
(Marcus & Marcus, 1963, translated).

B9. Cerithium: “the upper portion of the me-
dial lamina [of the pallial genital groove,
which] is very thick, glandular, and opaque,
and functions as a spermatophore organ”
(Houbrick, 1974).

B10. Diastoma: “proximal end of the gon-
oduct where the outer lamina becomes thick-
ened and white along its base. This area is

probably the prostate and spermatophore-
forming organ” (Houbrick, 1981b).

B11. Modulus: probably “inner surface of
the lateral lamina [of the distal pallial gono-
duct, extending] to the distal end [of the duct]
(Houbrick, 1980).

B12. Goniobasis, Pleurocera (?): “sper-
matophore organ” (Woodard, 1935; Dazo,
1965).

B13. Hinea and similar statements for other
planaxids: “free portions of the medial lamina
. . . composed of thick, white, glandular tissue
proximally and medially. .. . These thickened
portions of the gonoduct appear to delineate
the prostate-spermatophore gland” (Hou-
brick, 1987b).

B14. Heliacus: “possibly formed and
molded in the “prostate” and the glandular
vas deferens, and perhaps too the spermato-
phoric groove” (this paper).

B15. Atlanta, Oxygyrus: “glandular portion
of the male copulatory apparatus” (Tesch,
1949); “accessory sexual glands’ (Spoel,
1972); “gland annexed to copulatory organ
... flagellum” (Thiriot-Quiévreux & Martoja,
1974, translated).

B16. Rhizorus (?): “outer tubular part [of
male duct]... .” (Marcus & Marcus, 1960a).

B17. Metaruncina: “prostate” (Ghiselin,
1963).

B18. Runcina: “prostate” (Kress, 1985b).

B19. Limacina: “prostate gland” (Lalli &
Wells, 1978).

B20. Hedylopsis: “a series of narrower
glandular areas of the anterior sperm duct”
(Morse, 1976).

B21. Ganitus: “terminal ciliated ejaculatory
duct” (Marcus, 1953).

B22. Aeolidia, Aeolidiella: “distal portion of

‘sperm duct” (Tardy, 1965, translated).

B23. Siphonaria: “The penis ...has a large
pale yellow gland for the secretion of the sper-
matophore.” (Hutton, 1882). Epiphallus
gland. (Hubendick, 1945, translated). Possi-
bly “epithelium of... very small portion of...
wall... of the accessory organ” (Hubendick,
1955).

MALACOLOGIA, 1989, 30(1-2): 365-372

MORPHOMETRY, LENGTH-WEIGHT RELATIONSHIPS AND LENGTH
DISTRIBUTIONS OF FIVE POPULATIONS OF THE FRESHWATER BIVALVE
ASPATHARIA SINUATA (UNIONACEA, MUTELIDAE) IN NIGERIA

John Blay, Jr.'

Department of Biological Sciences,
University of llorin, llorin,
Nigeria

ABSTRACT

This study examines the shell morphometrics, length-weight relationships, and length distri-
butions of some lotic and lentic populations of the mutelid bivalve Aspatharia sinuata occurring
in Nigeria. The shell dimension ratios suggest that the populations belong to a common specific
unit, and the low variabilities (< 10%) in intrapopulation shell dimension ratios indicate the
relative stability of shell form at all sizes and in both sexes. The shell dry weight, tissue dry
weight, and live weight increased exponentially with increasing shell length, and in the latter the
exponents ranged from 2.8691 to 3.5653, thus suggesting a general isometric growth in the
populations. The variations observed in the length frequency distributions of the populations
were possibly ecologically determined.

Key words: Aspatharia sinuata, populations, shell dimension ratios, length-weight relation-

ships, Mutelidae, unionaceans.

INTRODUCTION

The mutelid bivalve Aspatharia sinuata
(von Martens, 1883) occurs in many lotic and
lentic freshwater habitats in Nigeria, and like
many of its relatives in other parts of Africa
there is a dearth of information on its biology.
Reports on African unionaceans have often
dwelt on their distribution, taxonomy and anat-
omy (see Pilsbry 8 Bequaert, 1927; Bloomer,
1932; Daget, 1962; Pain 8 Woodward, 1962;
Crowley, 1964; Odei, 1974; Lévéque, 1980).
In spite of the abundant information on their
taxonomy, there still remains some contro-
versy on this aspect of the mutelids and the
related unionids because of the superfluous
reliance on shell morphology in describing
new taxonomic forms (Kat, 1983a, b).

Although some workers (e.g. Pilsbry & Be-
quaert, 1927; Pain & Woodward, 1962; Crow-
ley, 1964) recognize mutelids to exhibit fre-
quent variations in shell form, this assertion
has seldom been confirmed statistically. Ac-
cording to Lévéque (1980), statistical analy-
ses are indispensable for any meaningful tax-
onomic work on African unionaceans. The
only known biometric information on mutelids

is that provided by Crowley et al. (1973) on
Aspatharia complanata occurring in Bornu
State, Nigeria. Morphometry has been stud-
ied in a few members of the Unionacea either
for establishing the existence of different spe-
cies complexes or sexual dimorphism (e.g.
Tudorancea, 1972; Badino, 1982; Dudgeon 4
Morton, 1983).

This work aims to determine the character-
istic shell dimension ratios of A. sinuata by
investigating some fluviatile and lacustrine
populations in Nigeria. It also examines the
length-weight relationships and the length
distributions of these populations.

METHODS

The collection sites of Aspatharia sinuata in
Nigeria lie between latitudes 7°30’ and 8°15'
N, and longitudes 4°30’ and 10°00’ E. Clams
were collected from three water bodies in
Kwara State, namely Asa Reservoir at llorin
(the state capital), Oyun Reservoir at Offa, a
town about 67 km south-east of Ilorin, and
River Oyun downstream the latter lake at the
University of llorin Main Campus. Samples

“Present address: Department of Zoology, University of Cape Coast, Cape Coast, Ghana.

(365)

366 BLAY

were also taken from Odo-Otin River at
Okuku in the north of Oyo State and approx-
imately 20 km south of Offa, and Agbuur
River (a tributary of River Benue) at Uga-
Mbagwa in Benue State.

The animals were handpicked, and loose
littoral substrates were scooped with a sieve
(35 meshes/cm?) in search of spat.

Three shell dimensions (length, height and
width) were measured to the nearest 0.01 cm
for the determination of their ratios. Length (L)
is the longest antero-posterior shell distance;
height (H) is the dorso-ventral distance from
the umbo to the ventral shell margin perpen-
dicular to the shell length; and width (T) refers
to the greatest transverse distance perpen-
dicular to the length and height.

The whole wet weight (live weight), shell
dry weight and tissue dry weight of animals
whose shell dimensions had been recorded
were also determined to the nearest 0.01 g.
The soft parts were removed after weighing
the live animals; tissues and their correspond-
ing shells were oven-dried at 60°C for 48
hours before weighing. The relationships be-
tween shell length and the three weight pa-
rameters were ascertained using the linear
regression analysis following the logarithmic
transformations of weights and lengths.

RESULTS

Figure 1 shows the mean ratios (percent) of
height to length (H/L) and width to length (T/L)
in the Aspatharia sinuata populations. The
means of H/L (%, + standard deviation) were
48.23 + 3.39, 46.90 + 1.48, 46.48 + 1.45,
45.66 + 1.82, and 44.36 + 1.87% in Oyun
Reservoir, Agbuur River, Odo-Otin River,
Oyun River and Asa Reservoir, respectively.
The corresponding mean values of T/L were
32:97 = 2.55, 2842 = 159, 30:39 = 173,
28.87 = 2.25, and 27.00 = 2.14%.

The above presentation follows that sug-
gested by Hubbs & Perlmutter (1943) for in-
vestigating racial differences in species pop-
ulations. The means of the ratios in males and
females did not differ appreciably from that
computed for all individuals in each popula-
tion, thus suggesting the existence of similar
shell proportions and the absence of sexual
dimorphism with regard to shell shape. The
coefficient of variability (CV) of H/L(%) ranged
from 3.11 in Odo-Otin River to 7.04% in Oyun
Reservoir, and for T/L(%) from 5.59 in Agbuur
River to 8.93% in Asa Reservoir. Hence,

there was a generally low variability (< 10%)
in shell dimension ratios within the popula-
tions (cf. Tudorancea, 1972).

The relationships between shell length and
the three weight parameters (Figure 2 a — с)
are described by the hyperbolic equation,
Y = aXP, where Y is the weight in grams, X is
shell length in centimeters, and a and b are
constants (Table 1). The values of the expo-
nent (b) in the length-live weight relationships
indicate that the clams were heaviest for
length in Agbuur River, while Odo-Otin River
recorded the least weights. However, the dry
tissue weight increased faster in Oyun Res-
ervoir and was slowest in River Oyun. For
shell dry weight, the rate of increase was
highest in River Oyun and lowest in Odo-Otin
River. It is also observed that in all popula-
tions shell weight increased faster than tissue
weight.

Figure 3 illustrates the length frequency
distributions of the bivalve populations at 0.5
cm class intervals. The Agbuur River samples
showed two modes (6.5 and 7.5 cm; length
range 3.26-8.94 cm), but the remaining
groups were characterised by unimodal
length distributions. The respective length
ranges of clams in Asa Reservoir, Oyun Res-
ervoir, River Oyun and Odo-Otin River were
3.15-10.15, 3.20-11.50, 1.57—7.51 and 4.57—
8.33 cm; their corresponding modal lengths
were 6.0, 8.5, 5.0 and 6.5 cm. It is evident
from the above that spat were scarce in the
samples.

Table 2 gives the mean lengths of clams in
the five populations. These and the modal
lengths show that there was a general ten-
dency for bigger clams to occur in Oyun Res-

_ervoir, and River Oyun to support smaller in-

dividuals. Except for the latter population,
females tended to be bigger than males.

DISCUSSION

A comparison of the mean shell dimension
ratios in the five populations of Aspatharia
sinuata indicates that they have similar shell
forms. Also, the limited coefficient of variabil-
ity (< 10%) in the ratios within the populations
predicts the relative stability of shell form at all
sizes, and in males and females. These com-
plement the observation that unionids of the
same species possess similar relative shell
proportions regardless of size, sex or age (Tu-
dorancea, 1972; Golightly 8 Kosinski, 1981).
However, the slight variations occurring in in-

MORPHOMETRY OF ASPATHARIA SINUATA 367

All sexes

E
вон |=:
Z

> 40
N
<=
20
0
ES
Е
= 20 5
>
E E
=
2

ASA RESERVDLR __
OYUN RESERVOIR, __

z
5
:
:
2
A
si
|

900-0TIN RIVER __
AGRUUR RIVER __ _

FIG. 1. Shell dimension ratios of five populations of A. sinuata. Vertical bars denote standard deviations.

trapopulation shell ratios may be attributed to
environmental influences (Pain & Woodward,
1962; Tudorancea, 1972; Chalmer, 1980).
The values of b for the shell length-live
weight relationships fell within the range 2.5—
3.5 found to be typical in most animals (Win-
berg, 1971). The closeness of these values to
3.0 indicates that the clams portrayed isomet-
ric growth which was probably sustained
throughout life. The clams also increased

shell dry weights more rapidly than tissue dry
weights, an observation that is consistent with
reports by Cameron et al. (1979), and
Golightly 8 Kosinski (1981) in some union-
aceans occurring in southern USA and Can-
ada.

The relatively higher rate of increase in
shell dry weight in the River Oyun population
is noteworthy. This population recorded the
least dry tissue weights for lengths, and their

368 BLAY

220 я
о Asa Reservoir

x Oyun Reservoir
eOyun River
"Odo-Otin River

O Agbuur River

200

(
—
>
©

120

100

(-:]
>

г
>

2 4 6 8 10 12 h 6 8 10 12
Shell length (cm) Shell length (cm)

140

120

Dry shell weight (g)

i TENTE
Shell length (cm)

FIG. 2. Relationship between shell length and (a) whole wet weight, (b) dry tissue weight, and (c) dry shell
weight in five populations of A. sinuata. The curves (fitted by the calculated regressions) are defined by the

equation Y = aX? (see Table 1).

369

MORPHOMETRY OF ASPATHARIA SINUATA

¿6 = М,

9966 0 6/rL'0
6766 0 ee91'0

6cE9 € OLLO'O 6978`0 591<`0 + 9S19'€ с100`0 8766`0 ГУРО`О + ES9S'E 86c0 0 лем лппабу
89155 $5<0`0 79820 £06€'0

+ + 128l'€ 62000 14660 1yZc'0 + 1698" 6LLL'O 18AIH UNO-0PO

SE86 0 5505`0 + 9086 € G900'0 12980 cllyO + 99832 vZLO'O 21/60 6700 + /8Ic'E€ 91S0°0 18A1H UNAO

186 0 6SLL'O + SCHE 6900 0 €86/ 0 £601'0 + 6zb9'€ SL00'0 £066'0 80tL'0 + 0661 '€ 04S0'0 11I0M3S3y UNÁD

21860 Zv60'0 + 2109 E 8,100 ve0Z'0 c80c 0 + 69/7 € 22000 7716`0 LICL'O = €9ZE'E 0550`0 JOASSSH ESY

úl "SL + q e 1 "SS! += q e J "SSI + q e uoyendog
lybiam Jays Ала 1ybiam anssi Ала JUBiam jam ajoym

А ЕЕ: Е Е АЕ РЕ ANA AE IL
‘(U8194809 иоце!элл0э = 1:001 = 9215 ajdwes) jueuodxe = q pue juejsuoo = в “siajauuao ul щбиа|
= x ‘swes6 и! yybiam = À элэцм ‘хе = À 0} рай вер jy 'suoneindod Bens ‘y ul UIBUe] |ец$ o) лэ}эшелед juBIem Jo diysuonejoy “| эла\

370 BLAY

FREQUENCY (%)

1.0 2.0 30 40 5.0 6.0 70 80 90 100 10 12.0
Shell length (ст)

FIG. 3. Length-frequency distribution of A. sinuata in (a) Asa Reservoir, (b) Oyun Reservoir, (c) Oyun River,
(d) Odo-Otin River, and (e) Agbuur River.

MORPHOMETRY OF ASPATHARIA SINUATA 371

TABLE 2. Mean shell length in A. sinuata populations.

ál_NMA——a——oeoeo—_e— ee ee

Mean length (cm) + S.D.

Population All specimens

Males Females

Asa Reservoir
Oyun Reservoir
Oyun River
Odo-Otin River
Agbuur River

“Number of clams in parentheses.

heavier shells might therefore be a compen-
sation for their lesser tissue dry weights.
Thicker and heavier shells would doubtless
aid the bivalves to stabilize and maintain their
positions under harsh fluviatile conditions.
Perhaps much energy was diverted into shell
growth than tissue growth.

The variations in the size distributions of
the population were probably due to environ-
mental influences resulting in different growth
rates (Blay, unpubl. data).

The general scarcity of spat in the A. sinu-
ata populations is similar to the observation in
Aspatharia complanata (Crowley et al., 1973).
Nonetheless, this is reported to be a common
feature т unionaceans (see Isely, 1911;
Crowley, 1957; Negus, 1966; Fisher &
Tevesz, 1976). The discovery of spat measur-
ing 0.50-1.29 cm shell length in a sandy bed
of a dry season pool in the Oyun River basin
was therefore unique. The smaller size of the
pool may have contributed to their ready
availability. Because the larvae of union-
aceans parasitize fish (Arey, 1921; Fryer,
1961, 1970; Trdan, 1981; Kat, 1984), it might
be a lot easier for A. sinuata larvae to encoun-
ter their fish hosts more readily in smaller
pools than they would in larger habitats.

It is apparent from this account that while
the proportions of shell dimensions tended to
be stable in the A. sinuata populations, some
differences occurred in the length-weight re-
lationships, and size distributions. The rela-
tive stability in the shell dimension ratios sug-
gests that the populations belong to a
common specific unit, i.e. Aspatharia sinuata
(von Martens, 1883) ЗЕЕ бу Pilsbry &
Bequaert (1927).

ACKNOWLEDGEMENTS

| thank Mrs. Solene Morris of the British
Museum (Natural History), U.K., for identify-

( : (
8 05 + + 0.78 (242) 8.22 + 0.97 (188)
5.46 + 0.69 (112) 5.26 + 0.83 (103)
6.12 + 0.64 on 6.23 + 0.66 (48)
6.15 + 1.31 (41) 6.31 + 1.04 (51)

ing the bivalves used in this research and an
anonymous reviewer for his useful sugges-
tions on the manuscript.

This study was jointly funded by the Senate
Research Grants and the Department of Bio-
logical Sciences, University of llorin (Nigeria).

LITERATURE CITED

AREY, L. B., 1921, An experimental study on
glochidia and the factors underlying encystment.
Journal of Experimental Zoology, 33: 463-499.

BADINO, G., 1982, Unio elongatulus Pfeiffer (Bi-
valvia): Variabilité biométrique et génétique des
populations du Piémont (Italie du Nord). Malaco-
logia, 22: 673-677.

BLOOMER, H. H., 1932, Notes on the anatomy of
some African Naiades. Proceedings of the Mala-
cological Society of London, 20: 166-173.

CAMERON, С. J., CAMERON, I. Е. 8 PATERSON,
C. G., 1979, Contribution of organic shell matter
to biomass estimates of unionid bivalves. Cana-
dian Journal of Zoology, 57: 1666-1669.

CHALMER, P. N., 1980, A comparison of variability
in size dimensions of intertidal and subtidal mus-
sels (Mytilus edulis) (Mollusca: Bivalvia). Journal
of Zoology, London, 191: 241-246.

CROWLEY, T. E., 1957, Age determination in An-
odonta. Journal of Conchology, 25: 263-265.
CROWLEY, T. E., 1964, Aspatharia (Spathopsis)
bourguignati (Bourguignat). Journal of Conchol-

ogy, 25: 263-265.

CROWLEY, T. E., GREEN, J. R. & McMILLAN, N.
F., 1973, Some mollusca from Bornu Province,
Northern Nigeria; with appendix: Statistical anal-
yses of two species (Pila wernei Phillipi and
Aspatharia complanata Jousseaume). Journal of
Conchology, 28: 81-94.

ОАСЕТ, J., 1962, Note sur les Aspatharia (Mutel-
idae) de l'Ouest Africain. Journal de Conchyliol-
ogie, 102: 63-77.

DUDGEON, D. & MORTON, B., 1983, The popu-
lation dynamics and sexual strategy of Anodonta
woodiana (Bivalvia: Unionacea) in Plover Cove
Reservoir, Hong Kong. Journal of Zoology, Lon-
don, 201: 161-183.

372 BLAY

FISHER, J. В. & TEVESZ, M. J. B., 1976, Distribu-
tion and population density of Elliptio complanata
in Lake Pocotopang, Connecticut. Veliger, 18:
332-338.

FRYER, G., 1961, The developmental history of
Mutela bourguignati (Ancey) Bourguignat (Mol-
lusca: Bivalvia). Philosophical Transactions of
the Royal Society of London, (В), 244: 259—
298.

FRYER, G., 1970, Biological aspects of parasitism
of freshwater fishes by crustaceans and mol-
luscs. In: Aspects of Fish Parasitology. Symposia
of the British Society for Parasitology, Volume 8.
TAYLOR, A. E. R., and MULLER, R., eds. Black-
well Scientific Publications, Oxford and Edin-
burgh. pp. 103-118.

GOLIGHTLY, C. G. 8 KOSINSKI, R. J., 1981, Es-
timating the biomass of freshwater mussels (Bi-
valvia: Unionidae) from shell dimensions. Hydro-
biologia, 80: 263-267.

HUBBS, C. L. 8 PERLMUTTER, A., 1943, Biomet-
ric comparison of several samples, with particular
reference to racial investigations. The American
Naturalist, 76: 582-592.

ISELY, F. B., 1911, A preliminary note on the ecol-
ogy of the early juvenile life of the Unionidae.
Biological Bulletin of the Marine Biology Labora-
tory, Woods Hole, 20: 77-80.

KAT, P., 1983 a, Genetic and morphologic diver-
gence among nominal species of North American
Anodonta (Bivalvia: Unionidae). Malacologia, 23:
361-374.

KAT, P., 1983 b, Sexual selection and simultaneous
hermaphroditism among the Unionidae (Bivalvia;
Mollusca). Journal of Zoology, London, 201: 395—
416.

KAT, P., 1984, Parasitism in the Unionacea (Bi-
valvia). Biological Reviews, 59: 189-207.

LEVEQUE, C., 1980, Mollusques. /n: Flore et faune
aquatiques de l'Afrique Sahelo-Soudanienne.
Initiations et document techniques, Volume 2,
Number 44. DURAND, J. R. and LEVEQUE, C.,
eds. O.R.S.T.O.M. Paris. pp. 283-305.

NEGUS, C. L., 1966, A quantitative study of growth
and production of unionid mussels in the River
Thames at Reading. Journal of Animal Ecology,
35: 513-532.

ODEI, М. A., 1974, A new species, Mutela voltae
from the Volta Lake in Ghana. Revue Zoologique
Africain, 88: 445-449.

PAIN, T. & WOODWARD, Е. R., 1962, The African
freshwater bivalve Aspatharia (Spathopsis)
rubens (Lamarck), its synonymy and distribution.
Journal of Conchology, 25: 73-78.

PILSBRY, H. A. & BEQUAERT, J., 1927, The
aquatic Mollusca of the Belgian Congo, with a
geographical and ecological account of Congo
malacology. Bulletin of the American Museum of
Natural History, 53: 417-418.

TRDAN, В. J., 1981, Reproductive biology of
Lampsilis radiata siliquoidea (Pelecypoda:
Unionidae). American Midland Naturalist, 106:
243-248.

TUDORANCEA, C., 1972, Studies on Unionidae
populations from the Crapina-Jijila complex of
pools (Danube zone liable to inundation). Hydro-
biologia, 39: 527-561.

WINBERG, G. G., 1971, Methods for the estimation
of production of aquatic animals. Academic
Press, New York, 175 pp.

Revised Ms. accepted 1 November 1988

MALACOLOGIA, 1989, 30(1-2): 373-395

PREY VALUE TO THE CARNIVOROUS GASTROPODS MORULA MUSIVA
(KIENER) AND THE TWO FORMS OF THAIS CLAVIGERA (KÜSTER):
EFFECT OF FORAGING DURATION AND ABANDONMENT OF PREY

Naoya Abe!
Seto Marine Biological Laboratory, Kyoto University, Shirahama, Wakayama 649-22, Japan

ABSTRACT

Intertidal carnivorous gastropods Morula musiva (Kiener) and the two forms of Thais clavigera
(Küster), Form C and Form P, showed preference for smaller items of most of the prey species
investigated than the maximum, though the E/H ratios increased monotonically with size in all
prey species. To explain the devaluation of larger items, two models are proposed. The first
model assumes that larger items cannot be entirely consumed when they take handling time
longer than the available foraging duration. This model is applicable to the two forms of Thais
whose foraging duration was limited to about 11 h. The prey selection by them, particularly Form
C, agreed well with the model. The second model, in which it is assumed that a predator
abandons its prey item at a constant rate, also predicted that the food value of larger prey would
decrease with the probability of abandonment. The second model is suitable for the prey se-
lection by Morula which foraged for several successive days. The relations of these models to
feeding efficiency, risks and foraging pattern are also discussed.

Key words: muricid; Thais; Morula; feeding; prey choice; optimal foraging; activity pattern

INTRODUCTION

The problem of food type selection by ani-
mals has been theoretically analyzed by
many authors (e.g. Emlen, 1966a; Schoener,
1971; Pulliam, 1974). The classical prey
choice model assumed that animals should
maximize their net gain of energy (reviewed
by Hughes, 1980, 1986; Pyke, 1984;
Stephens 8 Krebs, 1986). This model pre-
dicts (1) animals should select the highest
ranking food with the largest energy gain per
handling time, and successively add lower
ranking food types according to the rate of
encounter with higher ranking ones, and (2) a
food type is either always included in the diet
or not. These predictions have been tested in
many animals and are generally supported,
though most animals investigated showed
partial preference unlike the second predic-
tion (Pyke, 1984).

In contrast, through some applications ofthe
classical model to intertidal muricids (Emlen,
1966b; Menge, 1974; Hughes 8 Dunkin, 1984;
Dunkin 4 Hughes, 1984), a critical disagree-
ment with the model has been clarified. The
dry weight of flesh ingested per unit handling
time (W/H), which has been generally used as
the measure of prey value, was calculated in-

directly in Nucella emarginata feeding on Bal-
anus glandula (Emlen, 1966b), and directly in
Acanthina punctulata feeding on Littorina
planaxis and L. scutulata (Menge, 1974), and
in Nucella lapillus feeding on Mytilus edulis
(Hughes & Dunkin, 1984) and Semibalanus
balanoides (Dunkin 4 Hughes, 1984). In all
these cases, the W/H ratios monotonically in-
crease with the prey size (with the single ex-
ception of Acanthina feeding on L. scutulata
(Menge, 1974) when we recalculate W/H from
the two relations between dry weight and prey
length, and handling time and prey length,
though Menge herself presented the mono-
tonically increasing curve). However, many in-
tertidal muricids prefer smaller prey items in
spite of an abundance of larger items (Ton-
giorgi et al., 1981; Broom, 1982; Hughes &
Dunkin, 1984; Dunkin 8 Hughes, 1984;
Palmer, 1984; McQuaid, 1985).

Palmer (1983, 1984) got around the contra-
diction between prey selection and W/H
curves by measuring the prey value in terms of
growth rates of Nucella lamellosa, N. canali-
culata and N. emarginata maintained on pure
diets in field cages. These snails behaved as
if they recognized the “food value” of encoun-
tered prey, thus maximizing their expected
growth rate. Palmer's work suggests that the

Present address: Department of Zoology, Faculty of Science, Kyoto University, Sakyo-ku, Kyoto 606, Japan.

(373)

374

prey preference of muricids can be predicted
by the optimal theory if the food value is mea-
sured properly.

We should note that most of the calcula-
tions of W/H were based on data obtained
from laboratory experiments where conditions
surrounding the snails were constant. Pat-
terns of feeding activity have been paid atten-
tion recently (Menge, 1974; Menge, 1978;
Spight, 1982; McQuaid, 1985; Moran, 1985;
Hughes & Drewett, 1985). The information on
feeding patterns is important for a consider-
ation of the problem of prey selection since
foraging duration should affect the prey value
to animals with long handling time such as
intertidal muricids. In this paper | propose two
models in which prey value reaches a peak at
a certain prey size. In the first model it is as-
sumed that larger items cannot be entirely
consumed when the handling time is longer
than foraging duration. In the second model, it
is assumed that a predator abandons its prey
item with a constant rate.

Morula musiva (Kiener) and Thais clavigera
(Kúster) are common whelks on the intertidal
rocky shore of East Asia. Because Thais clav-
igera is an enemy of cultured oysters, many
studies related to their control have been
done (Arakawa et al., 1977; Lin & Hsu, 1979),
including some experimental studies on feed-
ing rate (Kinoshita & Kinoshita, 1931; Tanaka,
1949; Koganezawa, 1963; Lin 4 Hsu, 1979).
In nature, both Morula musiva and Thais clav-
igera mainly feed on such sessile animals as
barnacles and bivalves (Luckens, 1970; Abe,
1980; Taylor, 1980; Tong, 1986).

Abe (1985) reported that 7. clavigera con-
sists of two forms, Form C and Form P, which
differ in nodule shape and color pattern of the
shell, and that cross-matings between them
were rarely observed. Ecological features also
differ between the two forms: Form С grows
larger than Form P (Abe, 1985) and Form С is
more abundant on exposed rocks than Form
P, though their distributions largely overlap.

In this paper, prey selection, handling time
and foraging pattern of Morula musiva and
the two forms of Thais clavigera are de-
scribed, and two models are proposed to ex-
plain the results.

MATERIALS AND METHODS

General methods

For convenience sake the species treated
in this paper are referred to as follows: Morula
musiva as Morula, Thais clavigera Form C as

Thais-C, Thais clavigera Form P as Thais-P,
the small barnacle Chthamalus challengeri as
Chthamalus, the large barnacle Tetraclita
squamosa japonica as Tetraclita, the stalked
barnacle Pollicipes mitella as Pollicipes, the
tubeworm Pomatoleios krausii as Poma-
toleios, and the mussel Septifer virgatus as
Septifer.

The field study was conducted in a rocky
shore of Banshozaki in Shirahama on the
west coast of Kii Peninsula, Japan (33°42'N,
135°21’E). The tidal range is approximately
from —20 to +200 cm above chart datum at
spring tides. Chthamalus predominated over
the large platform which occupied the central
part of the study area, and Tetraclita, Pollici-
pes, Septifer and Pomatoleios as well as
Chthamalus were interlacingly distributed on
the peripheral slopes.

Shell length of all the predatory snails
(Morula, Thais-C and Thais-P) and Septifer,
apertural diameter of Chthamalus, Tetraclita
and Pomatoleios, and rostral-carinal length of
Pollicipes were measured using vernier cali-
pers to the nearest 0.1 mm.

All the laboratory experiments were con-
ducted in running sea water, water tempera-
ture not being regulated. Snails and prey
were collected from the study area.

Field observations

An observation site was selected in the
study area at each time to cover the area
where many individuals of Morula, Thais-C
and Thais-P were observed attacking prey.
Since Morula drills a hole in most cases when
attacking prey (see Table 7), the prey item

With a hole at the position to which Morula

was attached was regarded as being at-
tacked. Thais-C and -P do not always drill a
hole but often insert the proboscis through the
shell margin of the prey item without a hole.
Thus, when they attached to the margin of a
shell or to the aperture of barnacles, they
were regarded as attacking the prey. Shell
length of the snail showing feeding behaviour
was measured, and also its prey size mea-
sured if it was either Tetraclita, Septifer or Pol-
licipes. Field observations were carried out 36
times from April 1982 to October 1984 (about
1 hour/observation), most of which were dur-
ing daytime low tides but in a few cases dur-
ing high tides by scuba diving.

Since the probability of observing a snail
feeding on a prey item becomes higher in pro-
portion to handling time, the proportion of prey

PREY CHOICE BY CARNIVOROUS GASTROPODS 375

observed in the field does not represent the
actual diet nor preference of predators (Pe-
terson & Bradley, 1978; Fairweather & Under-
wood, 1983), and small items which took
shorter handling time would be observed less
frequently. To estimate true composition of di-
ets the proportion of predators attacking a
given type of prey was divided by the mean
handling time for that prey (Peterson 4 Brad-
ley, 1978). Similarly, arithmetic mean of prey
size measured in length is inadequate to repre-
sent preferred prey size, and a corrected mean
of attacked prey sizes was calculated from
1

le
2 a

where |; = the size of ¡th prey measured in
length and H = the function of handling time
to L.

Prey selection and prey size selection

In August 1985, about 60 individuals of var-
ious sizes of each of Morula, Thais-C and
Thais-P were placed in a plastic aquarium (35
x 25 x 10 cm?) with four prey species—
Chthamalus, Tetraclita, Pomatoleios and
Septifer for Morula, and Chthamalus, Tetra-
clita, Pollicipes and Septifer for Thais-C and
-P. Chthamalus and Tetraclita were collected
with small stones to which they adhered
(about 10 x 10 cm?). The other prey items
were clusters of similar size. Each cluster of
prey was placed near one of the four corners
of the aquarium. Observations were done
once an hour for the first four hours and sev-
eral times a day afterwards. The snail re-
garded as attacking prey was measured in
shell length, and was eliminated together with
the prey item attacked. This experiment con-
sisted of two replicates: the first was done
immediately after collecting the snails, and
the second after a week's starvation. Obser-
vations terminated when most snails were ob-
served attacking prey and were eliminated.
Since observations were done intensively,
most of the snails which began to feed during
the experiment should have been observed
and the handling time for each prey would not
affect observed frequenties of attacks.

The experiment on prey size selection of
Septifer was carried out in August-September
1984 for nine groups of the predators, S (10—
15 mm), М (15-20 mm), L (20-25 mm) and LL
(25—mm) size groups of Thais-C; $, M and L
size groups of Thais-P; and $ and ML (15-25

mm) size groups of Morula. To make the en-
counter rate identical for each size class of the
prey, the number of prey mussels given was
adjusted to equalize the total circumference
length in each size class: 22 of 5-10 mm, 13
of 10-15 тт, 9 of 15-20 mm, 7 of 20-25 mm,
6 of 25-30 mm and 5 of 30-35 mm. Five
snails of each group were placed in a plastic
aquarium (30 x 5 x 7 cm?) with mussels. The
experiment started on the day following the
collection of snails. The aquaria were checked
once a day: eaten (dead and vacant) mussels
were eliminated and fresh ones of the same
size class were added. The experiment was
carried out for 16 days for M and larger size
groups of Thais-C and -P, and was conducted
over 29 days for the other groups because of
small number of eaten mussels.

The experiment on prey size selection of
Chthamalus by Thais-C and -P were carried
out in September 1984 for 7 groups, S, M, L
and LL size groups of Thais-C, and 5, M, and
L size groups of Thaıs-P. Four individuals of
each group were placed in a plastic aquarium
(10 x 6.5 x 6 cm’), and were fed with bar-
nacles on a stone (about 3 x 3 cm?). After 2
or 3 days when about 10 or more barnacles
were eaten, sizes of eaten and uneaten bar-
nacles were measured.

Dry flesh weight and caloric content of prey

Specimens of Chthamalus, Tetraclita, Pol-
licipes, Pomatoleios, and Septifer were col-
lected from the study area in October 1984.
Flesh dissected from the specimens, exclud-
ing cirri of barnacles and byssal threads of
mussels, was dried at 80°C for 4 hours and
weighed. The carbon and nitrogen content for
flesh of these prey species was determined
using the C.H.N. corder (Yanagimoto MFG.
Co. LTD.), and the caloric content (cal) was
estimated from 10C + 1.9N, where C = car-
bon content (mg) and N = nitrogen content
(mg), assuming that crude protein content =
6.25N, C/N of crude protein = 3.25, caloric
content/crude protein = 5.5 cal/mg, and av-
erage caloric content/carbon derived from lip-
ids and carbohydrates = 10 cal/mg.

Handling time

Handling time consists of inspection time,
penetration time, and ingestion time (Hughes
8 Dunkin, 1984). The experiment to estimate
handling time of Morula, Thais-C and Thais-P
was carried out from November 1982 to Oc-
tober 1984 except for cold seasons, using a
video camera and a video tape recorder (VTR

376

ABE

Table 1. The number of attacks by different-sized predators observed in the field. Figures in parentheses
represent actual diets in percentage estimated using mean handling times. See text for further expla-

nations.
Size class
Predator (mm) Chthamalus Tetraclita Pollicipes Pornatoleios Steptiter Others
—15 2) (13) 6 (13) 0 12 (57) WE (1) 1
Morula 15-20 0 (0) 18 (9) 2 16 (24) 162 (67) 11
20-25 0 (0) 12 (21) 0 0 (0) 60 (79) 4
—15 36 (99) 5 (<) 0 (0) 0 0 (0) 0
Thais-C 15-20 70 (89) 64 (7) и (al) 0 31) 117
20-25 15 (45) 110 (35) 18 (3) 0 53 (18) 18
25- O0) 49 (50) 16 (35) 0 18 (16)
—15 53 (100) 0 (0) 0 (0) 0 1 (0) 1
Thais-P 15-20 108 (99) 4 (1) i (0) 0 3 (0) 3
20- 22 (97) se (2) 1 (0) 0 a (Gl) 0

experiment). The plastic aquarium (24 x 17
x 3.5 cm?) was divided into nine small com-
partments by plastic boards and was covered
with a glass plate. A watch was placed above
one compartment, and eight compartments,
lit by fluorescent lights all the time, were si-
multaneously filmed. In this experiment the
film was made for 3 seconds at intervals of 5
minutes using a timer. Each compartment,
filled with running sea water, had one or two
snails with a few prey items. After finishing
ingestion of the prey item, the prey size was
measured and the prey was inspected for
presence of a drilled hole.

To estimate ingestion time, Septifer, with
shells that were broken, were measured in
wet weight and given to Morula and Thais-C.
After two days' filming by the video camera,
the wet weight of the shells together with the
flesh left in them was measured.

Foraging pattern

Observations on the feeding activity of pred-
ators were carried out in aquaria (35 x 25 x
10 cm?) set in open air. The aquaria had three
different conditions, ¡.e. water filled, sprinkled
and exposed to air. The duration of each of
these conditions was regulated to coincide
with that at the mid-tide level of the shore. To
avoid overheating of the aquaria during day-
time, they were supplied with a little sea water
so as to wet the bottoms even in an exposed
condition. Experiments were carried out two
times. The first experiment was done during a
day of spring tide in July 1983 after one day
acclimation of the snails in the aquaria. The
snails used consisted of seven groups, S and

ML size groups of Morula, M, L, and LL size
groups of Thais-C, and M and L size groups of
Thais-P, each group consisting of 20 individ-
uals. Prey species were Septifer and Chtha-
malus for Morula, and Chthamalus and Tet-
raclita for Thais-C and -P. In the second
experiment carried out for 10 days in July-
August 1983, the size groups of snails and
prey species used were the same as those in
the first experiment, but each size group con-
sisted of about 50 individuals. In both of the
experiments, the number of snails showing
feeding behaviour, which was judged by the
position of snails in relation to the prey shells
to avoid disturbances by manipulation, were
recorded several times a day.

Feeding activities in the field were ob-
served in July 1983. The study site, selected
inside the breakwater at Banshozaki for ease
of the observation, was a small intertidal rock

. flat of 32 m°. About 50 individuals of each of

Morula, Thais-C and Thais-P were marked by
india ink covered with a cyanoacrylate glue,
and released at the study site. For two days
after release the behaviour of the marked
snails found within the site were observed
several times.

RESULTS
Prey selection and prey size selection

Table 1 shows the diets of Morula, Thais-C
and Thais-P observed in the field. As the size
of Morula increased, the diets changed: small
individuals (<15 mm in shell length) mainly
fed on Pomatoleios, larger ones (>15 mm)
mostly on Septifer. Tetraclita was attacked by
every size class of snails, but preyed on less

PREY CHOICE BY CARNIVOROUS GASTROPODS

377

Table 2. The number of attacks by different-sized predators observed in the aquaria. Four prey species
were given to predators. The results from two replicates were pooled. Figures in parentheses are

percentages.
Size class

Predator (mm) Chthamalus Tetraclita Pollicipes Pomatoleios Septifer
10-15 6 (86) 1 (14) — 0 (0) 0 (0)

Morula 15-20 14 (22) 4 (6) — 1 (2) 46 (71)
20-25 2 (10) O (0) = 1 (5) 18 (86)
10-15 7 (100) 0 (0) 0 (0) — 0 (0)

Thais-C 15-20 34 (92) le) 158) — IAS)
20-25 20 (59) 4 (12) 9 (26) — IAS)
25- 6 (27) 0 (0) 14 (64) — 2 (9)
10-15 14 (100) On (0) 0 (0) — 0 (0)

Thais-P 15-20 70 (100) 0 (0) 0 (0) — 0 (0)
20-25 20 (100) 0 (0) 0 (0) — 0 (0)

frequently than Septifer. Their minor diets
were Chthamalus, Pollicipes, Patelloida sac-
charina lanx (Reeve), Siphonaria acmaeoides
Pilsbry, Hormomya mutabilis (Gould) and
Saccostrea kegaki Torigoe 8 Inaba. Thais-C
mainly fed on Tetraclita, Pollicipes, Chtha-
malus and Septifer, and in rare cases on Bal-
anus tintinnabulum volcano Pilsbry, Acan-
thopleura japonica (Lischke), Acanthocithon
sp., Cellana toreuma (Reeve), Patelloida pyg-
maea (Dunker), Patelloida saccharina lanx,
Siphonaria japonica (Donovan), Hormomya
mutabilis, Mytilus corsucus Gould, Septifer
bilocularis (L.), Saccostrea mordax (Gould)
and S. kegaki. Thais-C also showed diet
change with increasing snail size: small indi-
viduals (<20 mm) mostly fed on Chthamalus,
large ones (20-25 mm) on Tetraclita and
Septifer as well as Chthamalus, and the
largest ones (>25 mm) on the former two
species and Pollicipes but not on Chtha-
malus. Thais-P mostly fed on Chthamalus,
and Tetraclita, Pollicipes and Septifer were
rarely attacked. Other minor diets were Col-
lisella langfordi Habe, Siphonaria japonica
and Saccostrea kegaki. Thais-P did not ap-
pear to change its diet with size.

The data from the two replicates of the ex-
periment on prey selection in aquaria were
pooled because no differences between them
were detected (Table 2). Prey selection in the
laboratory showed similar tendencies to the
diets in the field: large 'individuals of Morula
mostly attacked Septifer, and small individuals
of Thais-C and all individuals of Thais-P ex-
clusively attacked Chthamalus. Some differ-
ences were that small Morula (10-15 mm) did
not attack Pomatoleios but selected Chtha-
malus more frequently in the laboratory, and

that the largest individuals of Thais-C (>25
mm) selected no Tetraclita in the laboratory.

Table 3 shows the relationship between the
sizes of the snails and of the prey items at-
tacked in the field. In every combination of
predator and prey species, there was a weak
but statistically significant positive correlation.
This demonstrates that larger snails preferred
larger items of each prey. The corrected
means are smaller than the arithmetic means
(Table 3), but the difference is small. Though
Thais-C preferred a little larger Tetraclita than
Morula, the mean size of Septifer attacked by
Thais-C was smaller than that by Morula.

The mean size of Septifer attacked by both
Morula and Thais-C was a little larger in the
laboratory than that in the field (Table 4).
Morula consumed every size of mussels,
though the number of small mussels con-
sumed (<30 mm) was greater than that of
large ones. On the other hand, no individuals
of Thais-C attacked larger mussels (>30 mm)
except the largest Thais-C. All size classes of
Thais-P, which seldom attacked Septifer in
the field, preferred small mussels.

When feeding on Chthamalus the mean
sizes of barnacles eaten by each of Thais-C
and -P were larger than that of uneaten bar-
nacles irrespective of the sizes of snails and
barnacles given (Table 5). lijima (1974) also
reported that Reishia (= Thais) clavigera ate
larger individuals of Chthamalus when vari-
ous-sized barnacles were given in the labo-
ratory.

Flesh weight and caloric content of prey

The dry flesh weight of five prey species
was related to the prey size by the following
equations.

378

ABE

Table 3. The mean size of prey items attacked by different-sized predators in the field. r, correlation coefficient
calculated from the predator and prey sizes measured in shell length. See text for further explanations.

Predator Prey
Morula Tetraclita
Thais-C Tetraclita
Thais-C Pollicipes
Morula Septifer
Thais-C Septifer

Mean prey size (mm)

size class of AA A

predator (mm) N Arithmetic + SE Corrected
—15 5 2:1! == 0.16 2.0
15-20 9 4.2 + 0.42 ST
20-25 8 4.4 + 0.79 2.2
-15 2 4.1 + 0.50 4.0
15-20 39 4.2 = 0.28 37
20-25 84 4.4 = 0.20 3.9
25— 37 5.9 + 1.49 4.6
—20 5 HOME TO 9.4
20-25 23 181051201 10.2
25- 16 15.4 + 0.92 13.9
—15 8 10.2 =n 52 6.9
15-20 114 16.3 = 0:61 122
20-25 41 1179210776 15.2
—20 28 8.8 + 0.66 7.8
20-25 49 10.410.583 9.3
25- 16 13:8 == 1.70 11.4

Table 4. The number of Septifer consumed by different-sized predators in the laboratory. Prey items with
the same size-composition were given to each size group of predators. See text for further explanations.

Prey size (mm)

Size class
Predator of predator N
(mm) 5— 10- = 20— PAS 30— 35 Mean+SE
Morula 10-15 19 7 5 3 1 1 2 15.4 = 1.82
15=25 19 3 5 4 4 1 2 1740141878
10-15 32 29 2 1 0 0 0 1.6==10:53
Thais-C 15-20 57 43 11 3 0 0 0 8:93 0133
20-25 44 19 20 4 1 0 0 11.320,57
25— 32 11 12 1 4 3 1 leer se 1.14
10-15 27 23 4 0 0 0 0 8.0 + 0.46
Thais-P 15-20 28 27 4 0 0 0 0 71:60:24
20-25 21 16 4 1 0 0 0 8.5 + 0.68
Chthamalus: т DW = -2.684 + 2.475 In L тп = 32, rf 0.848) (1)
Tetraclita: п DW = -1485 + 2666 In L п = 23, г = 095 (2)
Pollicipes: п DW = -3.961 + 2.830 In L (n = 25, r = 0.958) (3)
Pomatoleios: In DW = -0.777 + 3313 In L (n = 15, ré = 0.916) (4)
Septifer: In DW = -4351 + 2.670 In. L (n = 31, Г = 0.980) (5)

where DW = dry weight of flesh (mg) and
L = size of the prey (mm).

The caloric content and C/N of five species
were estimated from the carbon and nitrogen
content (Table 6). There are significant dif-
ferences in values of the caloric content

and C/N among the prey species (d.f. = 38,
F = 32.17, "p<0:005 ‘for calorie “content
ЧЁ "= 38, F = 36:02, p<0/005<for G/N):
The C/N ratio of Chthamalus was largest and
those of the others were similar to each
other.

PREY CHOICE BY CARNIVOROUS GASTROPODS 379

Table 5. The mean size of eaten and uneaten Chthamalus when barnacles were given to different-sized
predators. The values of t represent the difference between the mean sizes of eaten and uneaten
barnacles.

À EEE

| Eaten Uneaten
size class EE —
of predator Mean size Mean Size
Predator (mm) N (mm) N (mm) t
10-15 12 1:9 42 1.3 4.60 p< 0.001
Thais-C 15-20 11 2.1 34 1.5 3.89 p< 0.001
20-25 10 2.6 38 les 6.00 p< 0.001
25- 11 17 44 1.4 2.73 p< 0.01
10-15 26 1.9 24 alas 3.59 p< 0.001
Thais-P 15-20 12 2 52 1.0 9.72 p< 0.001
20—25 9 2.5 42 1.4 4.71 p< 0.001

Table 6. Caloric content and C/N of prey species.

Caloric content + SE

Prey N C/N + SE (cal/mg)

Chthamalus 7 5.74 + 0.26 3.34. 0411
Tetraclita 9 3.99 + 0.06 3.90 + 0.14
Pollicipes 9 3.79 + 0.06 4.46 + 0.12
Pomatoleios 9 4.26 + 0.10 502018
Septifer 9 4.11 = 0:14 4.44 + 0.05

Table 7. The number of attacks by different methods in the VTR experiments. Figures in parentheses are
percentages.

Predator Prey Drilling Edge-drilling Non-drilling

Chthamalus 17 (26) 45 (68) 4 (6)
Morula Tetraclita 26 (96) 0 (0) 1 (4)
Pomatoleios 13 (100) —| 0 (0)
Septifer 60 (100) 0 (0) 0 (0)
Chthamalus 0 (0) 15 (26) 43 (74)
Thais-C Tetraclita 1 (2) 49 (82) 10 (17)
Pollicipes 14 (44) @ (0) 18 (56)
Septifer 1 (2) 48 (91) 4 (8)
Chthamalus 0 (0) то Ga) 83 (89)
Thais-P Tetraclita 1 (2) 27 (60) 17. (38)
Pollicipes 3 (16) 3 (16) 13 (68)
Septifer 3 (14) 17 (77) 2 (9)

(1)Edge-drilling cannot be defined for Pomatoleios

Methods of attack the shells of almost all prey items except
Chthamalus, which was mainly attacked by
Three methods of ‘attack were distin- edge-drilling (Table 7). The method of attack

guished by the feeding marks left on the of the two forms of Thais was quite different
shells attacked by snails: (1) drilling through from that of Morula. They attacked Tetraclita
the shell surface (drilling), (2) drilling through and Septifer mainly by edge-drilling, and
the valve edge (edge-drilling), and (3) insert- Chthamalus mainly by non-drilling. When at-
ing proboscis between two valves without tacking Tetraclita by drilling, Thais-C and -P
drilling (non-drilling). Morula drilled through drilled a hole on the scuta while Morula drilled

380

ABE

Table 8. Range of values of prey size, predator size and water temperature in the VTR experiments.

Prey Predator Water
size size temperature
Predator Prey No. (mm) (mm) (°C)
Chthamalus 42 1.4- 4.2 11.2-21.0 22-25
Morula Tetraclita 21 2.3- 6.7 14.8-21.1 23-27
Pomatoleios 13 1.2— 1.9 19.0-21.1 25-26
Septifer 22 6.7-20.5 11.9-21.4 21-29
Chthamalus 42 2.0- 5.0 15.4-27.0 21-25
Thais-C Tetraclita 34 2.0- 5.9 14.8-25.8 20-30
Pollicipes 20 6.2-19.6 14.8-27.0 19-29
Septifer 29 4.2-23.9 12.0-31.5 25-29
Chthamalus 121 0.8- 4.2 14.0-23.7 20-29
Thais-P Tetraclita 34 1.8- 7.4 13.0-23.7 19-30
Pollicipes 16 4.1-17.7 15.1-21.2 19-29
Septifer 17 4.3-13.4 19.8—23.7 25-28

on the parietal plate. Most Pollicipes were at-
tacked by Thais by a non-drilling method, but
some were also drilled through the shell, and
in the case of Thais-P also by edge-drilling.
Though the two forms of Thais used a similar
method of attack, Thais-P attacked prey by
non-drilling more frequently than Thais-C.
When attacked by edge-drilling, shapes and
sizes of drill holes at valve edges varied from
a large complete hole to only a scratch: holes
drilled by Morula were mostly the former type
and those by Thais clavigera were rather the
latter type.

Handling time

The handling time may be affected by many
factors, including prey size, water tempera-
ture, and predator size. Since all these factors
were not controlled in the VTR experiment
(Table 8), the following simple equation was
employed to relate them to the handling time:

ПН = A + b1 In Lp + b2 In Ls + b3 In T

where, H = handling time (h), Lp = size of
the prey (mm), Ls = shell length of the pred-
ator (mm), and T = water temperature (°C).

To estimate A, b1, b2 and b3, multiple re-
gression analysis was employed using the
data in Table 8 (Table 9). All the values of b1
were positive, and in 9 of 12 groups they were
significant (Table 9). Some of the values of b2
and b3 were significant, and whenever these
values were significant, they were negative,
i.e. handling time was negatively correlated
with water temperature and with predator
size. Most of the values of standard deviation

for Thais-C and -P were larger than those for
Morula.

From Table 9 regression lines of handling
time to prey size are presented in Fig. 1. The
water temperature was adjusted to 25°C and
the snail size to 20 mm, values within the
ranges investigated (Table 8). Comparison of
the handling times was difficult because the
variances or slopes were usually significantly
different between Morula and the two forms of
Thais (Table 10). However, most differences
in handling time between them were remark-
able: Morula had much longer handling time
than both Thais-C and -P (Fig. 1). The only
exception was that when feeding on Septifer
the handling time of Morula was slightly
shorter than that of Thais-P. The difference in
handling time between Thais-C and -P was
slight when feeding on Chthamalus and
Tetraclita. When feeding on Pollicipes and
Septifer, Thais-P had a comparatively longer
handling time than Thais-C.

The ingestion time, which was estimated as
the time taken to ingest Septifer with broken
shells, was related to snail size and wet
weight of the flesh ingested, but not to water
temperature because of the constant temper-
ature about 28°C during the experiment. The
calculated regression lines were

In! = 4.764 — 1.904 In Ls + 0.577 In WW

for Morula (d.f. = 2,10, F = 7.88, p<0.01),
and

п | = 5.037 — 3.295 пт + 1.099 In. WWW

for Thais-C (494. = 2,13, Е = 23:55,
p<0.005), where | = ingestion time (п), Ls =

PREY CHOICE BY CARNIVOROUS GASTROPODS 381

Table 9. Results of multiple regression analysis to relate handling time to prey size, predator size and
water temperature. See text for further explanations. * p<0.05; ** p<0.01; *** p<0.005.

Predator Prey A b1 b2 b3 SD F
Chthamalus 6.357 DIME: 0.267 =1.719 0.2747 7.81
Morula Tetraclita — 6.838 11655577 — 1.084 3.561 0.3628 13.66”
Pomatoleios — 29.482 1.184 0.926 8.945 0.4096 1.00
Septifer 3.938 1.653277 95 — 0.376 0.2234 42.14
Chthamalus 6.869 0.572 0.893 — 3.055 0.8517 0.83
Thais-C Tetraclita 10.508 12.4037 —0.800* —2:373*** 033387 13.85*
Pollicipes 10.997 1.860* — 0358 —3.548* 0.6255 Sana
Septifer —1.945 Bois — 9492 2.379 0.5078 9.00**
Chthamalus 13.740 0.888* — 1.031 — 34972 0.8620 11.91*
Thais-P Tetraclita 6.939 1.098** =05185 —1.767 0.7254 3.19*
Pollicipes 7.477 0.782 0.161 — 1.986 0.5176 1.83
Septifer — 6.296 142437 — 2.763 4.596 0.3527 11.98***

size of the snail (mm), and WW = wet weight
of flesh of Septifer ingested (mg). The inges-
tion time was significantly correlated with the
snaiksize (dí = 12, t = 2.43, р- 0.05 for
Могла; d.f. = 15, 1 = 4.21, p<0.005 for
Thais-C) and the wet weight of flesh (4.1. =
12,t = 3.67, p<0.005 for Morula; d.f. = 15,
t = 6.11, p<0.005 for Thais-C). These equa-
tions are converted by the equation of the wet
weight of flesh of Septifer to the dry weight,

In WW = 1.663 + 1.027 In DW
(n = 26, r = 0.978),

into the equations of the ingestion time to the
dry weight of flesh when Ls = 20:

In| = 0.020 + 0.593 In DW for Morula (6)

and

In| = —3.006 + 1.129 In DW
for Thais-C (7).

The functions of the ingestion time to mussel
size, which were estimated from equations
(5), (6) and (7), are shown in Fig. 1D. The
ingestion time of Thais-C was significantly
shorter than that of Morula (d.f. = 1,25, F =
18.82, p<0.005).

Foraging pattern

The foraging patterns of Morula, Thais-C
and Thais-P in the aquaria set in the open air
are shown in Figs. 2 and 3. The proportion of
prey-attacking Morula was large during a few
days around neap tide. As the spring tide was
drawing on, individuals attacking prey be-

came scarce and occurred only during sub-
mergence. This tendency was found in both
size groups of Moru/a. The proportion of
Thais-C attacking prey was large during sub-
mergence and small during exposure to air,
especially around spring tide when no individ-
uals foraged during exposure. The pattern of
foraging activity of Thais-P was similar to that
of Thais-C, except that around neap tide
Thais-P actively attacked prey without rest,
and that the percentage of feeding individuals
of Thais-P was larger than those of Thais-C
irrespective of the time and the snail size.
There were no clear differences in the forag-
ing pattern among different size groups of
Thais-C and -P, though the proportion of at-
tacking individuals of Thais-C were larger in
M size group than in L and LL size groups.

The difference in the foraging pattern be-
tween Morula and Thais was greatest on a
few days before and after a neap tide when
exposure occurred around dusk or dawn: few
individuals of Thais-C and -P attacked prey,
whereas many Morula attacked. The activity
pattern observed in the field also showed sim-
ilar differences (Table 11). Morula attacked
prey in high proportion during exposure to air
as well as submergence, while most of Thais-
C and -P attacked prey in the middle of sub-
mergence (22:00), but few did during expo-
sure.

During the spring tide, Thais-C and -P ini-
tiated foraging just after submergence, and
stopped feeding and found refuge a few hours
before the next mid-day exposure (Fig. 3).
The average foraging duration of Thais-C and
-P around spring tide was calculated as the

382

Handling time (h)

0 10

ABE

100

20.70 10 20

Prey size (mm)

FIG. 1A-D. The handling time of Morula (4, M), Thais-C (0, С) and Thais-P (e, P), plotted against size of
the prey. Solid lines represent the regression lines calculated from Table 11. Size of the snail and water
temperature were adjusted to 20 mm and 25°C, respectively. Prey species are Chthamalus (A), Tetraclita
(B), Pomatoleios (C, for Morula), Pollicipes (C, for Thais-C and -P), and Septifer (D). Broken lines represent
the ingestion time of Morula (M) and Thais-C (C) when Septifer with broken shells were given.

length of time when the number of feeding
snails exceeded a half of the maximum пит-
ber of feeding snails recorded during each
foraging bout, using the data of June 25-26
(Fig. 3) and August 4-8 (Fig. 2). This duration
approximately represents the period from the

onset of first attack to the end of last attack.
Since the average foraging duration was not
significantly different among different size
groups (df. = 2,12, Е = 0.82, p>0'1*tor
Thais-G; du. = 1,8, F = 00.08,"p=0sl Mion
Thais-P), all the data were pooled and the

PREY CHOICE BY CARNIVOROUS GASTROPODS 383

Table 10. Comparison of multiple regressions of handling time on prey size, predator size and
temperature between Morula and the two forms of Thais. Figures represent the values of F. * p<0.05;

FT) <0) =p <01009:

ms ee ee

Prey Morula vs. Thais-C Morula vs. Thais-P Thais-C vs. Thais-P
Variance Conte MOD 1.02
Chthamalus Slope 0.24 lus 1.42
Elevation 108.04*** 123739 0.93
Variance 1.02 4.00*** 3.94***
Tetraclita Slope 4.04 1.08 0.10
Elevation 17229072 228, = 1.45
Variance — — leon
Pollicipes Slope = = 0.87
Elevation — u 2.45
Variance ADE 2.49* РТ
Septifer Slope 2:25 3.98* 0.32
Elevation 1.24 1.91 9322

Table 11. Percentage of active individuals to the total marked snails found during each observation time

in the field.
Morula Thais-C Thais-P
Tidal
Date time condition Attacking Moving Attacking Moving Attacking Moving
July 23 10:00 exposure 37 0 11 0 Uf 0
15:30 submergence 29 7 5 50 29 21
22:00 submergence 39 27 50 10 50 21
July 24 8:30 awash 40 20 7£ 7 4 0
10:15 exposure 45 6 0 0 0 0

averaged duration, 10.55 h for Thais-C and
11.34 for Thais-P, was estimated. The differ-
ence in the average duration between Thais-
С and -P was not significant (d.f. = 1,23, F =
0.99, p>0.1).

DISCUSSION
Method of attack and feeding efficiency

The method of attack was quite different
between Morula and the two forms of Thais.
Morula mainly attacked prey by drilling or
edge-drilling, which is the typical feeding
method of muricid gastropods (Carriker,
1981). Though Thais-C and -P also drill holes
at valve edges, the drill holes were usually too
small to insert proboscises in, and especially
when attacking Chthamalus valves were usu-
ally opened without drilling. These observa-
tions suggest that Thais may produce a toxin

to facilitate prey consumption, as supposed in
American thaidids (Palmer, 1980).

Thais, especially Thais-C, is superior in
feeding efficiency to Morula. The handling
time of Thais-C and -P feeding on barnacles,
Chthamalus and Tetraclita, was much shorter
than that of Morula (Fig. 1), which may be due
to the different methods of attack. Similarly, it
is reported for the snail Nucella lapillus attack-
ing the barnacle Semibalanus balanoides that
the handling time by “prizing” is 0.53 (2 mm
sized barnacle) to 0.64 (6 mm sized barnacle)
of that by drilling (Dunkin & Hughes, 1984).
On the other hand, when feeding on Septifer
there was not a remarkable difference in the
handling time between Morula and the two
forms of Thais though the method of attack
differed between them. Edge-drilling or toxin
may not be so effective for the mussels. This
suggestion is also supported by the fact that
many thaidids, which can attack barnacles
by edge-drilling or non-drilling, usually drill

ABE

| Thais-C(LL)

Thais-P(M)

Date

FIG. 2. Temporal fluctuations of the percentage of snails attacking prey in the aquaria set in the open air from
July 30 to August 8, 1983. Neap tide was on August 2. Column at the top of the figure shows the aquarium
conditions: open, water filled; closed, exposed to air; shaded, sprinkled. Size range of snails is given in
parenthesis: $, 10-15 mm; M, 15-20 mm; L, 20-25 mm; LL, 25— mm; ML, 15-25 mm in shell length.

through the shell when feeding on mussels
(Palmer, 1980; Hughes & Dunkin, 1984).

Prey value

The energy intake per unit handling time
(E/H) is generally used as the index of prey
value in the classical model of diet choice.
Present study showed that E/H increased
monotonically with prey size increases in all
of the combinations of the predators (Morula,
Thais-C and Thais-P) and four prey species,
because the energy intake was proportional
to the 2.5-3.3 power of the prey size (Equa-
tions 1—5) and the handling time proportional
to the 0.5—1.9 power of the prey size (Table
9). According to the classical model, snails
should select the largest items. However,
there is some evidence that indicated the
preference of Morula and Thais-C and -P for

smaller items. First, the sizes of attacked prey
were positively correlated with the sizes of

‚ predators in Tetraclita and Septifer by Morula

and Tetraclita, Pollicipes and Septifer by
Thais-C (Table 3). This means that at least
small predators prefer prey items smaller than
the maximum. Second, Thais-C and -P
mostly attacked small Septifer (<10-15 тт)
in the laboratory experiment (Table 4) though
larger mussels were equally available. Then,
what factors devaluate larger prey?

Some hypotheses

Satiation may reduce the food value of
larger prey. Hughes & Dunkin (1984) reported
that the percentage of flesh of Mytilus edulis
consumed by Nucella lapillus decreased as
the mussel length increased, though the dry
weight of ingested flesh to the handling time
was still monotonically increasing function of

PREY CHOICE BY CARNIVOROUS GASTROPODS 385

Thais-C (L)
4
0

Thais-
р hais-C (LL)
0
Thais-P (M)
60 Thais-P (L)

Time of day (h)

FIG. 3. Temporal fluctuations of the percentage of active snails in the aquaria set in the open air on July
25—26, 1983. Closed bars, feeding individuals; open ones, moving individuals; O, no active individuals. Other

explanations are as in Fig. 2.

mussel length. Morula, Thais-C and Thais-P
showed complete consumption over the size
ranges of prey items given in the present
study. Thus, satiation cannot explain their
aversion to large prey items in these cases.
An interloper that steals flesh of the prey
item opened by another predator will reduce
the value of the item. Emlen (1966b) reported
that two or more Thais emarginata were com-
monly found feeding on the same large Bal-
anus cariosus and suggested that such inter-
lopers would depress food value of the item.
Two or more individuals of Thais-C often at-
tacked a single Tetraclita and Pollicipes in the
field (Table 12). Since the proportion of the
communal attack increased with prey size
x? = 26.68, p<0.01 for Tetraclita; x? = 3.90,
p = 0.14 for Pollicipes), the food value of
larger prey will be more depressed. This effect,
however, seems to be small. For example,
when Thais-C attacks Tetraclita, the caloric
content of flesh consumed per unit handling
time was 3.5 cal/h for a preferred barnacle of
4.3 mm in aperture diameter and 5.5 cal/h for
one of 6 mm. If 20% of the larger barnacles
were attacked by an interloper and the yield
were reduced by about half, the expected
value of E/H would be 5.0 cal/h and still be
larger than the smaller barnacles. Moreover,
no communal attacks were observed in Thais-
C feeding on Septifer and Morula on Septifer

Table 12. The number of attacks by two or more
individuals of Thais-C (communal feeding) and
attacks by one individual (solitary feeding) in the
field. Figures in parentheses are percentages.

Size class Communal Solitary

Prey (mm) feeding feeding
2-5 E) 105 (97)

Tetraclita 5-10 4 (17) 20: (83)
10-15 6 (43) 8 (57)

5-10 0 (0) 10 (100)

Pollicipes 10-15 2 (10) 18 (90)
15-24 5 (25) 15 (75)

and Tetraclita. Devaluation by interlopers can-
not explain the preference for small items of
these prey species.

Interlopers may affect the prey value also
by displacing the occupants. When several
snails were placed in the same aquarium in
the VTR experiment, the snail initiating attack
was sometimes displaced by an interloper.
Such displacement in the laboratory was also
observed in Nucella lapillus feeding on
Mytilus edulis (Hughes & Dunkin, 1984) and
Semibalanus balanoides (Dunkin & Hughes,
1984). If displacement by interlopers oc-
curs in the field, this should affect the prey
value. The effect can be treated in Model 2
(see below) if the rate of encountering a prey
item being attacked by another predator is

386

constant irrespective of the phase of the at-
tack.

Intertidal carnivorous gastropods are ex-
posed to risks from predation, dislodgment by
waves, desiccation and high temperature.
These risks must be greater during foraging
than during resting, because resting snails
usually seek refuge in crevices (Menge, 1974;
Hughes & Drewett, 1985) or form aggrega-
tions (Feare, 1971; Moran, 1985). Some au-
thors have emphasized these risks as a factor
influencing the food value of larger prey with
long handling time (Hughes & Dunkin, 1984;
Dunkin & Hughes, 1984; Palmer, 1984). The
effect of risks on the prey value can be treated
by also considering the foraging time. To
maximize future reproductive output, animals
shorten the foraging time to reduce mortality
during foraging, as well as lengthen it to in-
crease energy gain (Schoener, 1971). When
constant risks during foraging bring snails
death, the optimal length of the total foraging
time (top) becomes short, with the mortality
rate increasing as shown in the Appendix. If
the reproductive output depends on energy
gain, to, decreases with energy gain per unit
foraging time increasing. In both of the cases,
predators should select prey items with larger
value of E/H, as the classical model of prey
choice predicts, even if these prey take a long
handling time, because they can adjust the
total foraging time by lengthening attack
intervals. Thus, prey value is not affected by
constant risks of mortality (cf. lwasa et al.
1984).

None of the above hypotheses can entirely
explain why large prey are avoided by pred-
atory snails. Note that the degree of devalu-
ation of large prey seems to be much larger in
Thais than in Morula: for example, Morula
preferred larger specimens of Septifer than
Thais-C although Morula took a longer time to
consume them. This suggests that different
causes may affect the prey value in these
predators. Here | propose two different mod-
els, Model 1 for Thais and Model 2 for Morula,
and discuss their validity.

Model 1: maximum energy intake during
limited foraging duration

The foraging duration of the two forms of
Thais was limited to about 11 h (Figs. 2 and
3). Under such a condition large prey items
that take longer handling time than the feed-
ing duration cannot be entirely consumed. |
consider what type of prey items should be

selected to maximize energy intake during the
limited foraging duration using a simple
model.

Assumptions are:
(1) A predator forages during a limited time of

(2) Searching time is short enough to be ne-
glected, and the interval between attacks
is nearly zero.

(3) The predator selects only one type of
prey during foraging.

(4) The predator spends some time to open
the prey valves.

(5) After opening the prey valves, the preda-
tor ingests flesh of the item with a con-
stant rate.

Under these assumptions, the curve of the
energy intake of a predator during T to the
prey size becomes polymodal with the high-
est peak at the prey size that can be just con-
sumed during T (Fig. 4), since the larger prey
item has a higher E/H ratio. Actually, the han-
dling time of a predator feeding on prey of the
same size has some dispersion, and thus
lower peaks become smoothed and a gradu-
ally changing monomodal curve will be de-
rived (Fig. 4). The prey size with the highest
energy intake becomes large as the standard
deviation increases.

Energy intakes of the two forms of Thais
feeding on different-sized prey during a lim-
ited foraging duration were estimated by com-
puter simulation. The ingestion time of Thais-
C for different prey was estimated from the
function of the ingestion time to dry weight of
flesh of Septifer (Equation 7), and that of
Thais-P was assumed to be the same with
that of Thais-C. Fig. 5 shows the results of the
simulations when T = 11h and shell length of

. predators = 20 mm. In Thais-C all the curves

except that for Chthamalus became peaked
ones: the peaks were at 4 mm for Tetraclita, at
8-10 mm for Pollicipes and at 10-17 тт for
Septifer. These sizes were comparable to the
mean sizes of prey attacked by Thais-C of 20
mm in shell length (Tables 3, 4). Only the
curve for Chthamalus increased monotoni-
cally with prey size, which does not conflict
with the fact that Thais-C selected larger
Chthamalus in the laboratory. Moreover, the
order of the maximum energy intake, Chtha-
malus > Tetraclita > Pollicipes > Septifer,
approximately agreed with that of the prey
preference, except that Pollicipes was pre-
ferred to Tetraclita in the laboratory (Tables 1,
2):
The estimated energy intake of Thais-P

PREY CHOICE BY CARNIVOROUS GASTROPODS 387

Energy intake

SD= 0

prey size

FIG. 4. Energy intake during limited foraging duration under model 1 as a function of prey size. Figures
represent the number of prey items eaten during the foraging period.

was much smaller for Pollicipes and Septifer
than for Chthamalus (Fig. 5B), which agrees
with the fact that Thais-P rarely fed on the
former two prey species in the field or labora-
tory. The energy intake for Tetraclita in-
creased monotonically, and the maximum
value became larger than that for Chtha-
malus, though Tetraclita was also rarely at-
tacked. This disagreement may be due to the
statistical error in the experiment. Since the
handling time of Thais-P feeding on Tetraclita
approximated to the foraging duration of 11
hours (Fig. 1), the energy intake estimated by
the simulation is very sensitive to slight biases
of the estimation of the snail size-handling
time relation and its standard deviation.
Some predictions can be derived from the
model. First, as the foraging duration be-
comes long, the prey size that maximizes the
energy intake increases, and thus larger prey
should be more profitable. Second, as the
feeding efficiency increases, the most profit-
able size of the prey increases. If the hand-
ling time correlates negatively with the snail
size, which seems to be true, the larger snail
should prefer the larger prey. Fig. 6 shows

the energy intake per day when the handling
time is half of that of a snail of 20 mm. The
energy intake for Tetraclita, Pollicipes and
Septifer becomes large relative to Chtha-
malus. On the other hand, when the handling
time is twice that of a snail of 20 mm, the
profitability of Chthamalus becomes high
relative to that of other prey (Fig. 6). These
predictions agree with the fact that small
individuals of Thais-C preferred Chthamalus
and large ones did not.

The second and third assumptions may not
be realized in the field. When searching time
is relatively long and/or a predator can select
any type of prey during a foraging bout, the
prey value will change with the remaining for-
aging duration. It is difficult to solve this prob-
lem analytically, but small prey requiring short
handling times tend to be more valuable with
the time progressing during the foraging bout
(cf. Menge, 1974). However, the prey value,
estimated under these assumptions, agrees
well with the preference of Thais. This agree-
ment should not be ostensible, but would be
explained by the following reasons. First, this
model takes account of searching time before

ABE

388

1andes ‘e ‘594151104 “7 :eupena]L y :Sn¡eueyiyo ‘о ‘зиоцепашоэ 000! WO эпел иеэш sjuasaldas juiod yoey ‘azis Agıd
jsureBe payojd y | | о} раэйши sí poued BuiBe10} иэцм | ¡apoyy 18pun yyBuaj |э4$ ul WW OZ yo (9) d-s/ey/ pue (y) 9-s/ey/ jo Aep sad ayejul Аблэиз ‘6 HI

(ww) 9ZIS Á9 y

PREY CHOICE BY CARNIVOROUS GASTROPODS 389

100

NO
о

0 10

- - - - Energy intake (cal/day)

Prey size (mm)

FIG. 6. Energy intake per day of Thais-C under Model 1 when handling time is half of that of 20 mm snails
(solid lines), and when it is twice that of 20 mm snails (broken lines). Other explanations are as in Fig. 5.

the first attack in fact, since the foraging du-
ration of 11 hours, which was used in the sim-
ulation, was estimated as the period from the
onset of first attack to the end of last attack.
Second, each prey species is patchily distrib-
uted in the field, and is abundant within each
patch. Thus, after reaching a suitable patch, a
predator can attack prey items successively.
As shown in Table 11, the proportion of mov-
ing individuals was small in the middle of the
foraging duration, 22:00 (16.6% of active in-
dividuals of Thais-C; 29.6% of Thais-P),
though at the beginning of foraging (15:30),
relatively many individuals moved. Third, the
preferable sizes of Tetraclita, Pollicipes and
Septifer were too large for Thais to consume
more than one item during a foraging bout
(Fig. 5). Thus, predators will rarely attack a
second item in that they select these items at
the beginning of a foraging bout. However, if
the first attack takes shorter time than the for-
aging duration or the searching time is long,
the remaining foraging duration will be limited.
In this case, the food value of Chthamalus,

which is the most valuable prey under the
second and third assumptions, will increase
still more because of its short handling time,
and the order of prey value will not change as
a result.

Model 2: prey value when abandoning prey
with a constant rate

Morula continued to feed on prey during
several successive days around neap tides
(Fig. 2), and thus the foraging duration itself
would not affect the prey value as much. For
example, an individual of 20 mm in shell
length spends about 3 days to consume Tetra-
clita of 10 mm in apertural diameter, while it
can forage during almost 5 days. Neverthe-
less, Morula preferred barnacles smaller than
4 mm (Table 3) though prey value measured
by E/H increased with prey size.

Table 13 shows the proportion of Septifer
having incomplete drill-holes. Most of these
incomplete holes seem to have been drilled
by Morula because Thais rarely attacked bar-

390

Table 13. Number of Septifer with incomplete drill-
holes. Figures in parentheses are percentages.

Size class Individuals with
(mm) N incomplete holes
0—5 58 0 (0)
5—10 153 la)
10-15 56 5 (9)
15-20 21 6 (29)
20-25 22 4 (18)
25-30 25 8 (32)
30- 28 9 (32)

nacles by drilling. Morula must have failed in
completely drilling these prey and abandoned
them. The abandonment of prey would be
caused by dislodgement by waves and inter-
ference by other animals, such as pecking by
fish. Snails are likely to be dislodged by strong
waves after they are exposed to air during the
hottest part о the day and their adhesion to the
substratum becomes weak. Interlopers would
also cause the abandonment of prey though
the prey should be drilled completely in this
case.

Effect on prey value of such abandonment
of prey in course of attacking is considered
using a model. Assumptions are:

(1) A predator gives up a prey item with the
instantaneous abandoning rate of c.

(2) The predator takes H-I hours to drill
through the prey shell, where H = han-
dling time and | = ingestion time.

(3) After drilling through the prey shell, the
predator ingests flesh of the prey item
with a constant rate E/l, where E = en-
ergy intake when fully ingesting the item.

The probability of continuing attacking for t h

after initiating attack is derived from

dpit)

FT = —cp(t)
as

p(t) = e *

Thus, the probability of completely consuming
a prey item is

p(H) ie".

The average handling time, i.e. expected han-
dling time per attack including the cases that
prey items are abandoned, is

"H / dptt)\ 1
—— |dt + Hp(H) = (1 —е ).
jai dt | a с

Since the predator gets the energy with the
rate of E/l after completing the drilling, the
expected energy intake per attack is

Thus, the prey value is

Ее)
Aie (еб! — 1) 2

Fig. 7 shows the prey value curve of Morula
of 20 mm in shell length estimated from (8) as
c = 0.025, the value selected to adjust the
preferred size of Septifer. The curve agrees
with the prey preference of Morula in some
respects: Septifer was the most profitable
prey among the four prey species, and Tetra-
clita of 4 mm was most profitable of different-
sized barnacles.

Some predictions are derived from the
model. First, as the feeding efficiency in-
creases, E/H of large prey items relatively in-
creases, which is the same prediction as from
Model 1. The change of prey selection by
Morula with the snail size may be explained
by this prediction. Second, as the abandon-
ment rate increases, the profitable prey size
decreases.

Whether Model 2 is applicable to Morula
depends on the frequency of abandonment of
prey in the field. When the instantaneous
abandoning rate, c = 0.025 as in Fig. 7, the
probability of completely consuming Septifer

‘of 17 mm, the preferred-sized mussel is

0.305. Moran (1980) reported that when three
American thaidids were fed with Balanus car-
iosus, the percentage of success in drilling at
a suture was 43% and that of success in drill-
ing through a lateral plate was 18%. There-
fore it is highly probable that Morula aban-
dons prey items frequently enough to depress
the profitability of large prey items.

Relations of the models to feeding
efficiency, risks and foraging pattern

The prey value to Morula is probably af-
fected by the abandonment of prey as shown
in Model 2. Similarly, Thais should also aban-
don prey. However, since the foraging dura-
tion of Thais is relatively short, the effect of

PREY CHOICE BY CARNIVOROUS GASTROPODS

1-0 F Pomatoleios

E/H (cal/h)

0-5

Tetraclita

Chthamalus

391

Septifer

Prey size (mm)

FIG. 7. Expected energy intake per handling time of Morula under Model 2 when they
abandon prey items with constant instantaneous rate с = 0.025.

abandonment of prey is small. For example,
when it selects an item that takes 11 h to
handle and с = 0.025, the probability of
abandoning prey is only 0.24. Thus, prey se-
lection by Thais would be affected more
strongly by the limitation of the foraging dura-
tion as shown in Model 1. In general, these
two models can be realized simultaneously,
but the importance of each model varies with
the amount of the foraging duration: predators
with short foraging duration tend to obey
Model 1 and those with long foraging duration
tend to obey Model 2.

Then, why does the foraging duration differ
between Morula and Thais? To answer this,

we also consider risks of mortality because
they affect the length of foraging time (see
Appendix). Some of the risks, which include
predation, dislodgment by waves, desicca-
tion, and high temperature, are unlikely to be
constant over the time but will show periodic
changes. Faced by such risks, animals
should decide when to forage as well as how
long to forage. The simplest solution is to for-
age during low-risk periods and to rest during
high-risk periods. Morula foraged for several
days around a neap tide and the two forms of
Thais during night and submergence. Both of
them avoided the mid-day low tides, during
which snails seem to be exposed to heavy

ABE

392

ибн '5861 6 Isnbny 0} AINF $

зэмепь ısej 7 !uoow п} ‘4 “WO OS! ‘(g) (еле эрирны —) шо 86 ‘(Y) :wnyep Heu, anoqe

г шод sjana] 1ерцааи! ом} ye ха} ayy и! рэицар хэри! U
3}0q

| ony
Ge Vom Pesce . ail AR оо

оцелодела au} и! зчоцетоп е.одше| `9'\8 ‘Sls

IM
67 87 15 90 SC

O
хэри! UONDJOIDAZ

PREY CHOICE BY CARNIVOROUS GASTROPODS 393

risks caused by high temperature and severe
desiccation. The degree of effects of these
risks on snails would increase as the duration
of exposure becomes longer and the time of
exposure draws closer to noon. An index is
employed to indicate the degree of risks from
desiccation and high temperature (Moran,
1985). This evaporation index is defined as

2
| (t) fat,
t1

where t = time of the day (h), f(t) = sin(t-6)/
12 when 6<t<18, and f(t) = O when t<6 or
t>18, И = the time (h) when the tidal level
begins to be exposed to air, and t2 = the time
(h) when the level begins to be submerged.
Fig. 8A shows the index at the mid-tide level,
which was calculated from a tide table assum-
ing that the falling and rising tides follow a
sine-curve and waves reach to 10 cm higher
level than that calculated from a tide table.
There are two time periods with small risks
from desiccation: high tide and night which
occur every day, and a few days around neap
tide. Thais exploited the former period, and
Morula did the latter. Morula marginalba
shows similar foraging patterns to those of
Morula musiva (Moran, 1985). At the midtide
level, M. marginalba foraged when low tides
were at dawn and dusk, and sheltered when
low tides were around midday. Moran (1985)
also believed stressful conditions caused by
long exposure to air particularly during the
hottest part of the day prompt M. marginalba
to seek shelter.

The foraging duration of Morula included
some low tides around dawn and dusk when
they would be exposed to little risk from des-
iccation. Exploitation of low risk periods by
Morula may be due to that they cannot gain
sufficient food during the short periods be-
cause of their low feeding efficiency. Morula
of 20 mm in shell length can gain about 150
cal during five days under Model 2 when it
selects the most profitable size of Septifer
and 90 cal under Model 1 over a tidal cycle
from a spring tide to the next one when it
forages for 11 hours every day like Thais
though the total feeding time is longer in the
latter case. If the increase of the expected
future reproductive output from longer forag-
ing is larger than the loss from the mortality
during the foraging duration, they should
choose longer foraging. On the other hand,
Thais-C of the same size can gain about 260
cal, and Thais-P about 210 cal under Model 1

when they attack 3 mm of Chthamalus. These
energy gains are much larger than that of
Morula, and may be sufficient for the two
forms of Thais.

Thais have another advantage in foraging
during the limited time between daytime low
tides. In the upper intertidal, there are not long
periods with low risks around neap tide (Fig.
8B). The two forms of Thais can exploit this
area, but Morula cannot because they forage
for several successive days. In fact, the range
of vertical distribution of Thais extended more
to the upper region than Morula, especially
during the season from September to Decem-
ber when almost all individuals of Thais
stayed at the high zone of 100-160 cm above
chart datum, the level dominated by Chtha-
malus (Abe, in prep.).

Both of Morula and the two forms of Thais
seem to avoid strong risks from desiccation
and high temperature by their specific forag-
ing patterns. Thus, during foraging they are
exposed to relatively low risks. Such risks are
unlikely to cause snails death, but may disturb
their feeding and affect food value of large
prey, as shown in Model 2. We should note
that risks of mortality must not affect prey
value, but risks of failure in attack devaluate
larger prey items. The decision of prey choice
would be also affected by the limited foraging
duration, as Model 1 shows. Since the
amount of foraging duration seems to result
from the avoidance of risks, it can be said that
risks indirectly affect prey choice also in
Model 1.

ACKNOWLEDGMENTS

| wish to thank Professors E. Harada and Н.
Kawanabe, Drs. К. Wada, Y. Iwasa, T. Kuwa-
mura, A.D. Ansell, and M. Yuma for critically
reading the manuscript. Thanks are due to Dr.
G. Coan and an anonymous reviewer for their
critical review of this paper. | also thank Dr. К.
Iwata for his advise on the calculation of ca-
loric contents. My thanks are also due to Dr.
T. Narita and the stuff of Otsu Hydrobiological
station of Kyoto University for providing facil-
ity to use the C.H.N. corder. Special thanks
are due to Dr. M. Imafuku for devising a timer
to control a video tape recorder.

LITERATURE CITED

ABE, N., 1980, Food and feeding habit of some
carnivorous gastropods (preliminary report).
Benthos Research, 19/20: 39-47 (in Japanese).

394 ABE

ABE, N., 1985, Two forms of Thais clavigera
(Küster, 1858). Venus, 44: 15-26.

ABE, N., Interactions between carnivorous gastro-
pods and their sessile animal at a rocky intertidal
shore. (in prep.)

ARAKAWA, K. Y., HOSHINA, T., & MATSUZATO,
T., 1977, Abibliography of enemies of oyster with
annotations. Bulletin of the Hiroshima Fisheries
Experimental Station, 8: 27-52 (in Japanese).

BROOM, M. J., 1982, Size-selection, consumption
rates and growth of the gastropods Natica mac-
ulosa (Larmarck) and Thais carinifera (Larmarck)
preying on the bivalve, Anadara granosa (L.).
Journal of Experimental Marine Biology and
Ecology, 56: 213-233.

CARRIKER, M. R., 1981, Shell penetration and
feeding by Naticacean and Muricacean predatory
gastropods: a synthesis. Malacologia, 20: 403—
422.

DUNKIN, S. de В. & HUGHES, В. N., 1984, Behav-
ioural components of prey-selection by dog-
whelks, Nucella lapillus (L.), feeding on barna-
cles, Semibalanus balanoides (L.), in the
laboratory. Journal of Experimental Marine Biol-
ogy and Ecology, 79: 91-103.

EMLEN, J. M., 1966a, The role of time and energy
in food preference. American Naturalist, 100:611—
617.

EMLEN, J. M., 1966b, Time, energy and risk in two
species of carnivorous gastropods. Ph.D. thesis,
University of Washington, Seattle, 138 pp.

FAIRWEATHER, Р. С. & UNDERWOOD, A. J.,
1983, The apparent diet of predators and biases
due to different handling times of their prey.
Oecologia (Berlin), 56: 169—179.

FEARE, C. J., 1971, The adaptive significance of
aggregation behaviour in the dogwhelk Nucella
lapillus (L.). Oecologia (Berlin), 7: 117-126.

HUGHES, В. N., 1980, Optimal foraging theory in
the marine context. Oceanography and Marine
Biology. Annual Review, 18: 423-481.

HUGHES, R. N., 1986, A functional biology of ma-
rine gastropods. Croom Helm, 245 pp.

HUGHES, В. М. 8 DREWETT, D., 1985, А compar-
ison of the foraging behaviour of dogwhelks, Nu-
cella lapillus (L.), feeding on barnacles or mus-
sels on the shore. Journal of Molluscan Study,
SEIS 16

HUGHES, В. N. 8 DUNKIN, $. de B., 1984, Behav-
ioural components of prey selection by dog-
whelks, Nucella lapillus (L.), feeding on mussels,
Mytilus edulis (L.), in the laboratory. Journal of
Experimental Marine Biology and Ecology, 77:
45-68.

НМА, K., 1974, The predation of Reishia clavigera
on Chthamalus challengeri with special reference
to its size preference. Benthos Research, 7/8:
43-46.

IWASA, Y., SUZUKI, Y. 8 MATSUDA, H., 1984,
Theory of oviposition strategy of parasitoids. 1.
Effect of mortality and limited egg number. The-
oretical Population Biology, 26: 205-227.

KINOSHITA, T. 4 KONOSHITA, S., 1931, On some

muricid shells as enemies to the oyster. Venus, 2:
190-198 (in Japanese).

KOGANEZAWA, A., 1963, Ecological studies of
drills in oyster beds. Bulletin of Miyagi Prefectural
Fisheries Experimental Station, 3: 12-22 (in Jap-
anese).

LIN, Y. S. & HSU, C. J., 1979, Feeding, reproduc-
tion and distribution of oyster drill Purpura clav-
igera (Küster). Bulletin of the Institution of Zool-
ogy, Academia Sinica, 18: 21-27.

LUCKENS, P., 1970, Predation and intertidal zona-
tion at Asamushi. Bulletin of the Marine Biological
Station of Asamushi, 14: 33-51.

McQUAID, С. D., 1985, Differential effects of pre-
dation by the intertidal whelk Nucella dubia (Kr.)
on Littorina africana knysnaensis (Phillipi) and
the barnacle Tetraclita serrata Darwin. Journal of
Experimental Marine Biology and Ecology, 89:
97-107.

МЕМСЕ, В. A., 1978, Predation intensity in a rocky
intertidal community. Relation between predator
foraging activity and environmental harshness.
Oecologia (Berlin), 34: 1-16.

MENGE, J. L., 1974, Prey selection and foraging
period of the predaceous rocky intertidal snail,
Acanthina punctulata. Oecologia (Berlin), 17:
293-316.

MORAN, M. J., 1985, The timing and significance
of sheltering and foraging behaviour of the pred-
atory intertidal gastropod Morula marginalba
Blainville (Muricidae). Journal of Experimental
Marine Biology and Ecology, 93: 103-114.

PALMER, A. R., 1980, A comparative and experi-
mental study of feeding and growth in thaidid
gastropods. Ph.D. thesis University of Washing-
ton, Seattle, 320 pp.

PALMER, А. R., 1983, Growth rate as a measure of
food value in thaidid gastropods: assumptions
and implications for prey morphology and distri-
bution. Journal of Experimental Marine Biology
and Ecology, 73: 95-124.

PALMER, A. R., 1984, Prey selection by thaidid
gastropods: some observational and experimen-
tal field tests of foraging models. Oecologia (Ber-
lin), 62: 162-172.

PETERSON, C. H. 8 BRADLEY, B. P., 1978, Esti-
mating the diet of a sluggish predator from field
observations. Journal of the Fisheries Research
Board of Canada, 35: 136-141.

PULLIAM, H. R., 1974, On the theory of optimal
diets. American Naturalist, 108: 59-74.

PYKE, G. H., 1984, Optimal foraging theory: a crit-
ical review. Annual Review of Ecology and Sys-
tematics, 15: 523-575.

SCHOENER, T. W., 1971, Theory of feeding strat-
egies. Annual Review of Ecology and Systemat-
ics, 2: 369—404.

SPIGHT, T. M., 1982, Risk, reward, and the dura-
tion of feeding excursions by a marine snail.
Veliger, 24: 302-308.

STEPHENS, D. W. 8 KREBS, J. R., 1986, Foraging
theory. Princeton University Press, 248 pp.

TANAKA, Y., 1949, Injuring mechanisms of oyster

PREY CHOICE BY CARNIVOROUS GASTROPODS 395

drill (Purpura clavigera) to Young oyster (Ostrea
gigas). Bulletin of the Japanese Society of Sci-
entific Fisheries, 15: 447-457.

TAYLOR, J. D., 1980, Diets and habitats of shallow
water predatory gastropods around Tolo Chan-
nel, Hong Kong. Proceedings of First Interna-
tional Workshop on the Malacofauna of Hong
Kong and Southern China, р. 163-180.

TONG, L. K. Y., 1986, The feeding ecology of Thais
clavigera and Morula musiva (Gastropoda: Muri-
cidae) in Hong Kong. Asian Marine Biology, 3:
163-178.

TONGIORGI, P., NARDI, P., GALLENI, L., NIGRO,
M. & SALGHETTI, U., 1981, Feeding habits of
Ocinebrina edwardsi (Mollusca; Prosobranchia)
a common mussel drill of the Italian coasts. Ma-
rine Ecology, 2: 169-180.

Revised Ms. accepted 6 February 1989

APPENDIX

To consider the effect of risks of mortality
during foraging on the total foraging time, we
assume:

1. The mortality rate of a snail is non-zero

(=m) during foraging and zero during
other activities including resting.
2. The net energy intake, E is a linear func-
tion of foraging time (t), and thus E = rt.
3. A snail reproduces after P and the repro-
ductive output, G is a linear fundtion of E,
and thus G = KE.
The probability of a snail surviving after t days’
foraging is | = e ™. Animals should maximize
the expected reproductive output Е (=Gl).
Since we can obtain the optimal length of total
foraging time from dF/dt = 0 as 1/m, Fmax =
kr/me when 1/m < Р, and Fmax = kre ™? when
1/m > P. Therefore, as the mortality becomes
higher, animals should shorten the foraging
time when 1/m < P. Note that Fax is propor-
tional to r, which means that animals should
choose the prey that maximize the net energy
intake per unit foraging time. If G is an as-
ymptotic function of E, the optimal length of the
total foraging time is related to r as well as m.
For example, from G = K(1-e E) the optimal
foraging time is derived as 1/rIn(1 +r/m) which
decreases with r increasing.

my
|
|

MALACOLOGIA, 1989, 30(1-2): 397-405

INDEX

Acanthina 373

Acantharion 48, 172, 174, 284, 291, 298

Acanthinula 28, 29, 51, 53, 118, 281, 290,
296

Acanthinulinae 281

Acanthocithon 377

Acanthopleura 377

Асамаае 2724726730, 37238. 52; 65,71.
72, 77-79, 90, 91, 93, 196, 198, 200,
202, 204, 206, 208, 210, 212, 214, 285,
301, 303

Acavoidea 2, 70-72, 84, 90, 91, 93, 286,
287

Acavus 36, 49, 214, 286, 292, 298

acera, Ventridens 160, 283, 291

Achatina 11, 48, 65, 180, 285, 292, 298

Achatinacea 62

Achatinella 19, 38, 47, 51, 102, 104, 280,
290, 296

Achatinellidae 1, 47, 52, 53, 75-77, 80, 81,
92, 100, 102, 104, 280, 301

Achatinellinae 47, 280

Achatinidae 2, 30, 52, 59, 65, 66, 71, 78-
79, 85, 86, 93, 180, 285, 301

Achatinoidea 2, 64-66, 83, 84-86, 90-93,
281, 284, 285, 287, 303

acicula, Cecilioides 64, 182, 284, 292

acmaeoides, Siphonaria 377

Acochlidiidae 347

Acochlidioidea 363

Acochlidiomorpha 341, 345, 346, 348, 360

aculeata, Acanthinula 118, 281, 290

acuta, Cochlicella 264, 288, 293

Aegopinella 50, 152, 154, 283, 291, 297

aenea, Strobilops 120, 281, 290

Aeolidia 361, 364

Aeolidiella 361, 364

Aeolidiidae 347, 361

Afroconulus 83

agassizi, Cerion 69

Agriolimax 59

Aillyidae 4

Alaba 359

albicilla, Nerita 361

Albinaria 51, 226, 285, 292, 298

albula, Vallonia 36, 116, 281, 290

alexandri, Dorcasia 204, 285, 292

algirus, Zonites 283, 291

Allogastropoda 349

Allogona 48, 287, 293, 299

Alopiinae 285

alternata, Anguispira 28, 138, 282, 291

Amastra 51, 54, 106, 108, 280, 290, 296

Amastridae 1, 52, 54, 55, 75, 76, 80, 81,
92, 106, 108, 280, 301

Amastrinae 280

Amimopima 290, 296

Amimopina 28, 29, 55, 56, 128, 281

Ammonitella 50, 70, 222, 224, 287, 293,
299

Ammonitellidae (= Oreohelicidae) 2, 24,

70, 84, 93

Ammonitellinae 30, 43, 52, 61, 64, 70, 77-
79, 222, 224, 287, 302

Ampelita 49, 210, 285, 298

Amphibolidae 348

Amphibuliminae 236, 286

Amphicoelina 70

Amphidromus 48, 240, 287, 293

Amplir 293

Amplirhagada 242, 287, 299

Anadeninae 283

Anadromidae 65, 286

Anastoma 292

Anastomopsis 65

Ancylidae 363

Ancylus 363

Andrefrancia 51, 130, 282, 291, 297

andrewsae, Mesodon 244, 293

Aneitea 146, 281, 290, 291

Angiola 362

Anguispira 28, 50, 57, 58, 138, 282, 291,
297

Animopima 50

Annoselix 51, 132, 282, 297

Anoglypta 48, 285, 292, 298

Anostoma 286

Anthracopupa 81

Anthracopupinae 81

aperta, Philine 363

Aphallarion 50, 282, 291, 297

apicina, Helicopsis 288, 293

Aplysiomorpha 348

arachis, Haminoea 363

arboreus, Zonitoides 36, 160, 283, 291

arbustorum, Arianta 276, 289, 294

Archaeogastropoda 357, 358

Archaeopulmonata 3, 29-31

Architectonica 363

Architectonicidae 341-372

Architectonicoidea 341, 345, 348, 349, 360

Arianta 48, 276, 289, 294, 300

Ariolimacinae 282

Ariolimax 50, 282, 291, 297

Arion 13,50, 142, 283, 291, 297

Arionacea 82

Arionidae 1, 27, 28, 43, 52, 58, 67, 78-79,
81-83, 92, 140, 142, 144, 282, 301

Arioninae 283

Ariophantinae 170, 284

Ascoglossa 348

Aspatharia 365-372

aspersa, Helix 39, 276, 278, 289, 294

Athoracophoridae 2, 5, 24, 27, 52, 57, 67,
78-79, 88, 89, 92, 93, 146, 281, 282,
290, 297, 301

Atlanta 359, 362, 364

atlantica, Plutonia 150, 283, 291

Atlantidae 347, 359

"Atlantoidea" 359

Atoxon 27, 49, 176, 284, 292, 298

Aulacopoda 9-12, 17, 30, 36, 72

398 INDEX

auricula, Auriculella 100, 280, 290
Auriculella 37, 51, 100, 280, 290, 296
Auriculellinae 280

aurisleporis, Cochlorina 286, 292
Averellia 50, 256, 288, 293, 300
azulensis, Discoleus 234, 286, 292

balanoides, Semibalanus 373, 383, 385

Balanus 373, 377, 385, 390

balteata, Leptachatina 108, 280, 290

barbata, Oreohelix 70, 222, 288, 293

bartschi, Fargoa 342

Basommatophora 3, 25, 29, 30, 31, 347,
361

Batillaria 358

Batillariidae 358, 362

berendti, Strebelia 186, 285, 292

Berendtia 51, 69, 230, 232, 286, 292, 299

bermudezae, Bostryx 234, 286, 292

bilineata, Incillaria 283

bilineata, Meghimatum 335

bilineatus, Philomycus 144, 291

bilocularis, Septifer 377

binneyiana, Stephanoda 134, 282, 291

Binneyinae 282

biplicata, Varicella 184, 292

Bittium 358, 364

Bocageia 37, 48, 65, 178, 284, 292, 298

Boonea 347, 360

Bostryx 51, 234, 286, 292, 299

Bothriopupa 51, 112, 280, 290, 296

Brachynephra* 2, 8, 9, 10, 19, 20, 56, 90°,
91, 93, 281, 282, 285-287

Bradybaena 48, 60, 246, 288, 293, 300

Bradybaenidae 2, 30, 52, 59, 60-61, 63,
78-79, 83, 84, 93, 246, 248, 288, 302

Bradybaeninae 288

Brasilennea 65

braunsi, Rachistia 19

breviconus, Bothriopupa 112, 280, 290

browni, Acantharion 172, 174, 284, 291

Bulimulidae 2-4, 24, 31, 43, 52, 59, 69-71,
78-79, 86-88, 91, 93, 232, 234, 236,
286, 301

Bulimulinae 234, 236, 286

Bulla 348

Bullidae 348, 363

Bullomorpha 341, 345, 347-349, 360

burnerensis, Amplirhagada 242, 287, 293

buttoni, Aphallarion 282, 291

Caenogastropoda 357, 358

caledonica, Partula 106, 280, 290

Camaenidae 2, 30, 35, 52, 60-62, 70, 77-
79, 83, 84, 93, 238, 240, 242, 287, 302

camelus, Hemphillia 140, 144, 282, 291

Campanile 346, 358, 362

Campanilidae 358

Campyleinae 272, 274, 276, 289

canaliculata, Nucella 373

Candidula 25, 49, 270, 288, 293, 300

capensis, Trachycystis 132, 282, 290

Carinaria 359

Carinariidae 359

Carinariodea 359

cariosus, Balanus 385, 390

carminensis, Euglandina 184, 285, 292

carolinianus, Philomycus 142, 144, 283,
291

carpenteri, Bocageia 178, 284, 292

cartusiana, Monacha 270, 289, 293

Caryodes 49, 71, 72, 196, 285, 292, 298

casablancae, Cerion 69, 228, 286, 292

Cassidula 348

Cathaica 60

cayennensis, Systrophia 194, 286, 292

Cecilioides 25, 51, 64, 65, 182, 284, 292,
298

Cellana 377

Cepaea 48, 278, 289, 294, 300, 305-315

Cepolinae 60, 61, 252, 254, 288

Cepolis 48, 49, 60, 252, 254, 288, 293,
300

Cerastua 29, 50, 126, 281, 290, 296

Cerastuinae 29, 55, 56, 58, 80, 81, 83, 281

Cerion 51, 69, 86, 228, 230, 286, 292, 299

Cerionidae 2, 24, 26, 30, 52, 54, 55, 69,
78-79, 86-88, 93, 228, 230, 286, 301

Cerithidea 358, 364

Cerithideidae 358

Cerithiidae 347, 358

Cerithioidea 341, 345-349, 358, 362

Cerithiopsidae 349

Cerithiopsis 349

Cerithium 348, 358, 362, 364

Cernuella 48, 288, 293, 300

challengeri, Chthamalus 374

Charopa 51

Charopidae 2, 4, 27, 52, 58, 66-68, 78-79,
87-89, 93, 130, 132, 134, 282, 290, 297,
301

chilensis, Strophocheilus 208, 285, 292

Chilina 347, 361, 363

Chilinidae 348, 361

Chlamydephoridae (= Rhytididae) 2, 71, 93,
286

Chlamydephorinae 5, 13, 27, 72, 287

Chlamydephorus 72, 190, 287, 293

Chondrinidae 1, 52-55, 76, 77, 80, 81, 92,
114, 281, 301

Chondrininae 54, 281

Chondrinoidea 92, 280, 281

Chondrula 18, 29, 51, 56, 122, 281, 290,
296

Chondrulinae 29, 81, 281

Chthamalus 374-383, 386-389, 391, 393

Cionellacea 54

claromphalos, Epiphragmopora 260, 288,
293

Clausiliidae 2, 11, 24-26, 30, 52, 64, 68-70,
78-79, 86-88, 93, 226, 228, 285, 301

Clausilioidea 2, 69-70, 86-87, 90, 91-93,
285, 286, 303

Clavator 25, 30, 48, 212, 285, 292, 298

clavigera, Thais 373-395

clavigeria, Reishia 377

Clithon 358

Clypeomorus 358

coactiliata, Averellia 256, 288, 293

Cochlicella 51, 59, 264, 293, 300

Cochlicopa 18, 51, 54, 81, 110, 280, 290,
296

INDEX

Cochlicopacea 54

Cochlicopidae 1, 52, 54, 75, 76, 80, 81,
92, 110, 280, 301

Cochlorina 49, 286, 292, 299

Cochlostyla 288, 293

cognatus, Amphidromus 240, 287, 293

Coilostele 64

coli, Escherichia 307

collaris, Sculptaria 216, 287, 293

Collisella 377

columbianus, Ariolimax 282, 291

complanata, Aspatharia 365, 371

concavum, Haplotrema 248, 286, 292

Coneuplecta 50, 51, 164, 284, 291, 298

Conibycus 51, 162, 284, 298

conica, Samoana 280, 290

constanceae, Ranfurlya 134, 290

copium, Cerion 69, 230, 286, 292

Corilla 25, 26, 30, 49, 70, 220, 287, 293,
299

Corillidae 2, 24-26, 30, 37, 52, 70-72, 77-
79, 84, 90, 93, 216, 218, 220, 287, 302,
303

corrugata, Everettia 168, 284, 291

corsucus, Mytilus 377

Craterodiscus 25, 50, 70-72, 218, 287,
293, 299

crystallina, Vitrea 152, 283, 291

Cylichna 360, 363

cylindracea, Lauria 116, 281, 290

cylindricus, Heliacus 341-345

ystopelta 50, 67, 88, 134, 144, 282, 291,

297

daedaleus, Plagiodontes 232, 286, 292

dahli, Conibycus 162, 284, 291

Daudebardia 44, 50, 154, 283, 291, 297

Daudebardiinae 27, 56, 283

decollata, Rumina 180, 285, 292

decolorata, Tamayoa 194, 286, 292

Dendropoma 359

Dendropupinae 81

depressa, Anostoma 286, 292

depressa, Pellicula 236, 286, 292

depressispira, Gyliotrachela 114, 281, 290

Deroceras 59, 325-328, 331, 333, 335,
336

deshayesi, Parmacella 158, 283, 291

detrita, Zebrina 124

Diastoma 359, 364

Diastomatidae 358

dictyodes, Pararhytida 290

dilatata, Poiretia 182, 285, 292

dioscoricola, Ptychopatula 118, 290

Diplomphalus 37, 49, 190, 286, 292, 299

Discidae 1, 27, 36, 37, 52, 57-58, 64, 78-
79, 81-83, 91, 92, 136, 138, 282, 301

Discinae 282

Discoconulus 49, 51, 162, 284, 291, 298

Discoleus 36, 48, 234, 286, 292, 299

Discopoda 358

Discus 50, 57, 136, 138, 282, 291, 297

dissimilis, Varicella biplicata 285

Dolichonephra* 2, 8-10, 19, 20, 56, 90*,
91, 92, 281, 282-285, 287, 288

dolium, Orcula dolium 280

399

dolium, Orcula 112, 280, 290

dolosa, Annoselix 132, 282

Dorcasia 26, 49, 71, 205, 285, 292, 298
Doridella 348

draparnaudi, Oxychilus 152, 283, 291
Draparnaudia 50, 55, 128, 281, 290, 296
Drosophila 312

dussumieri, Mariaella 170, 284, 291
Dyakia (Quantula) 317-324

Dyakiinae 60, 63, 168, 284

ebernea, Zebrina 281, 290

Edentulina 48, 65, 192, 287, 293, 299

Edulis, Mytilus 373, 384, 385

Elasmias 51, 100, 280, 290, 296

elegans, Trochoidea 288, 293

Elgonella 50

Elisolimax 48, 176, 284, 292, 298

Ellobiidae 348

Elona 48, 274, 289, 294, 300

emarginata, Nucella 373

emarginata, Thais 385

Embletonia 363

Embletoniidae 363

Ena 18, 51, 56, 126, 281, 290, 296

Endodontidae 2, 18, 24, 52, 66-68, 78-79,
87-89, 91-93, 130, 282

Endodontoidea 2, 3, 57, 66-68, 87-90, 93,
281, 282, 286, 303

Enidae 1, 29, 52, 55-56, 58, 75, 76, 80,
81, 92, 122. 124: 126, 128:281, 301

Eninae 29, 81, 281

enneoides, Gonaxis 287, 293

Enneopsis 65

Eostrophia 69

Epitoniidae 342, 349

Epiphragmopora 49, 60, 260, 288, 293,
300

Epitonioidea 348

ericetorum, Turdus 314

erigone, Phrixgnathus 136, 282, 291

Escherichia 307

Estria 176, 284, 292

Eua 36, 37.49, 55, 104, 280, 290; 296

Eucalodiinae 286

Euconulidae 1, 52, 57, 62, 68, 78-79, 81-
83, 92, 162, 164, 284, 301

Euconulus 57

eudiscus, Зузтор Ма 68, 194, 286, 292

Euglandina 37, 66, 184, 285, 292

europaea, Testacella 292

Euthyneura 348

Everettia 48, 168, 284, 291, 298

eximius, Clavator 212, 285, 292

expansa, Eua 104, 280, 290

explanata, Leucochroa 272, 288, 293

eyriesii, Sterkia 112, 280, 290

falconeri, Hedleyella 200, 285, 292

Fargoa 342, 347, 360

fasciata, Rhinoclavis 362

Fastigiella 358, 362

Fauxulus 53, 81

Ferussaciidae 2, 11, 20, 25, 27, 31, 52,
64-65, 68, 78-79, 85, 86, 93, 182, 284,
301

400 INDEX

fibratus, Placostylus 37, 286, 292
fidelis, Monadenia 260, 288, 293
Filholiidae 68, 285

firmus, Gryllus 312

Firoloida 359

Firolidae 359

Fissilabia 359

Fissurella 358, 361, 364
Fissurellidae 358

Fissurelloidea 341, 345, 348, 358
fluviatilis, Theodoxus 362
Fluvinerita 358

fradulenta, Triodopsis 246, 293
fratercula, Libera 130, 290
fulgens, Achatinella 19

fulica, Achatina 11, 65, 180, 285, 292
futilis, Spiraxis 66, 184, 292

gabrielense, Glyptostoma 224, 287, 293

gagates, Milax 158, 283, 291

Ganitus 360, 364

Gastrodonta 50, 56, 160, 283, 291, 297

Gastrodontinae 36, 56, 68, 160, 283

Gastropoda 3, 325-339

Gastropteridae 348

Gastropteron 348

Geomalacus 50, 283, 291, 297

geyeri, Trochoidea 288, 293

gibbonsi, Chlamydephorus 190, 287, 293

Gibbulina 53, 81

Gibbulinella 65

glandula, Balanus 373

glans, Cerion 69

Glypterpes 60

Glyptostoma 49, 70, 224, 287, 293, 299

Godwininae 56

Gonaxis 287, 293

goniocheila, Plectotropis 61, 242, 287, 293

Goniobasis 359, 364

Gourmya 358

grandiflora, Palythoa 343

grandis, Sagda 250, 287, 293

Granularion 49, 62, 174, 284, 292, 298

granulosa, Oopelta 142, 283, 291

Gryllus 312

Gyliotrachela 50, 53, 54, 114, 281, 290,
296

Gymnarion 48, 172, 174, 284, 291, 298

Gymnarioninae 27, 62, 84, 172, 174, 284

Gymnosomata 348

hagada, Amplir 293

haliotidea, Testacella 285

Halolimnohelix 48, 59, 60, 262, 264, 288,
293, 300

Haminea (Haminoea) 363

Haminoea 360, 363

Haminoeidae 360, 363

Haplotrema 48, 49, 84, 248, 286, 292, 299

Haplotrematidae 2, 30, 52, 62-64, 70, 77-
79, 83, 84, 93, 248, 286, 301

Hedleyella 48, 200, 285, 292, 298

Hedylopsidae 346, 347, 360

Hedylopsis 347, 360, 363, 364

Heliacus 341-372

Helicarion 298

Helicarionidae 2-4, 13, 17, 50, 52, 60, 63,
77-79, 82-84, 93, 166, 168, 170, 172,
174, 176, 284, 291, 301

Helicarioninae 84, 166, 284

Helicella 25, 29, 48, 272, 288, 293, 300

Helicellinae 25, 59, 270, 272, 288

Helicidae 2, 24, 25, 30, 52, 58-60, 78-79,
83, 84, 92, 93, 262, 264, 266, 268, 270,
272, 274, 276, 278, 288, 289, 302

Helicigona 49, 272, 289, 294

Helicinae 276, 278

Helicinidae 346

Helicodiscidae (= Discidae) 58

Helicodiscinae 282

Helicodiscus 36, 50, 57, 136, 282, 291,
297

Helicodonta 48, 59, 264, 289, 294, 300

Helicodontinae 264, 289

Helicogona 300

Helicoidea 2, 59-64, 71, 83-85, 89, 90, 92,
93, 284, 286-288, 303

Helicophanta 26, 38, 48, 212, 214, 285,
292, 298

Helicopsis 17, 49, 268, 288, 293, 300

Helicostyla 293

Helicostylinae 288

Helicostylus 48, 60, 248, 288, 300

Helix 12, 39, 50, 276, 289, 294, 300, 338,
343

Helminthoglypta 48, 258, 288, 293, 300

Helminthoglyptidae 2, 24, 30, 52, 59-60,
62, 771-79, 83, 84, 92, 93,252,204
256, 258, 260, 288, 302

Helminthoglyptinae 260, 288

Hemiplecta 50, 62, 170, 284, 291, 298

Hemphillia 28, 50, 58, 140, 144, 282, 291,
297

hepatica, Cepaea nemoralis var. 313

Hesperarion 50, 283, 291, 297

hessei, Pseudoglessula 178, 285, 292

Heterobranchia 360

Heteropoda 341, 345-348, 359

Heterostropha 360

Heterurethra 17, 39, 64

Hinea 359, 362, 364

hispida, Siphonaria 363

hispida, Trichia 266, 289, 294

histrio, Rachistia 128, 281, 290

Hodopeus 61

Holopoda 10-12, 72

Holopodopes 11, 72

Holospira 69

Hormomya 377

horni, Thysanophora 252, 288, 293

humberti, Corilla 25, 220, 287, 293

humile, Prophysaon 142, 283, 291

humphreysiana, Hemiplecta 170, 284, 291

Hyalimax 281, 290

hydatis, Haminea 363

Hygromia 48, 266, 289, 294, 300

Hygromiinae 266, 268, 270, 289

Hypselostomatinae 53, 81, 281

hystricelloides, Thaumatodon 130, 282, 290

idahoensis, Zacoleus 283, 291
Imparietula 18, 29, 51, 124, 281, 290, 296

INDEX

inaequalis, Rhytida 188, 286, 293
Incillaria 283, 335

inflata, Limacina 363

inornatus, Mesomphix 154, 283, 291
insigne, Ptychopatula dioscoricola 281
interna, Gastrodonta 160, 283, 291
intersecta, Candidula 288

Isomeria 71

ltala 51, 226, 285, 292, 298

itala, Helicella 272, 288, 293

itala, ltala 226, 285, 292

jaliscoenis, Polygyra matermontana 287
Jamininae 281

Janthinoidea 348, 349

japonica, Acanthopleura 377

japonica, Siphonaria 377

japonica, Tetraclita squamosa 374
jousseaumei, Imparietula 124, 281, 290

Kalidos 48, 166, 284, 291, 298
kegaki, Saccostrea 377

kershawi, Pygmipanda 198, 285, 292
Klemmia 18, 51, 122, 281, 290, 296
krausii, Pomatoleios 374

Labyrinthus 18, 35, 48, 71, 238, 287, 293,
299

Lacteoluna 48, 287, 293, 299

laevis, Xenopus 307

lamellata, Spermodea 120, 281, 290

Lamellidea 51, 100, 280, 290, 296

lamellosa, Nucella 373

lamottei, Granularion 174, 284, 292

langfordi, Collisella 377

lanx, Patelloida saccharina 377

Laoma 50, 282, 291, 297

lapicida, Helicigona 272, 289, 294

lapillus, Nucella 373, 383-385

larreyi, Panda 196, 285, 292

lateumbilicata, Paralaoma 136, 282, 291

launcestonensis, Anoglypta 285, 292

Lauria 51, 116, 281, 290, 296

Lauriinae 53, 281

laxata, Macrocyclis 206, 285, 292

lederi, Daudebardia 154, 283, 291

Lehmannia 325-339

leiomonas, Laoma 282, 291

lejeunei, Halolimnohelix sericata 59, 288

lennum, Sinumelon 287, 293

leprieurii, Labyrinthus 238, 287, 293

Leptachatina 18, 51, 108, 280, 290, 296

Leptachatininae 280

Leptarionta 288, 293

Leucochroa 25, 272, 288, 293

Leucotaenius 65

Libera 51, 130, 282, 290, 297

lidgbirdi, Pseudocharopa 282, 290

Limacidae 1, 4, 27, 38, 52, 57-59, 78-79,
81-83, 92, 158, 283, 301

Limacina 347, 360, 363, 364

Limacinidae 360

Limax 24, 48, 59, 158, 283, 291, 297,
325-328, 331, 333, 335, 336

limbata, Hygromia 266, 289, 294

[Пора 359

401

Litiopidae 359

Littoraria 362

Littorina 373

Littorinidae 362

Littorinoidea 362

lorata, Achatinella 47, 102, 104, 280, 290
lubrica, Cochliocopa 54, 81, 110, 280, 290
Luinodiscus 282, 290

Lychnopsis 65

lychnuchus, Pleurodonte 240, 287, 293
ymnaea (Lymnaea) 335

Lymnaeidae 348, 361

Lymnaeoidea 341, 345, 361

macleayi, Amimopina 128, 281, 290

macneilli, Anguispira 282

Macroceramus 25, 51, 69, 86, 230, 286,
292, 299

Macrocyclis 49

Macrocyclis 71, 206, 285, 292, 299

maculifer, Mesafricarion 174, 284, 291

maculosus, Geomalacus 283, 291

madagascariensis, Elisolimax 176, 284, 292

magnicosta, Klemmia 122, 281, 290

major, Phenacolimax 150, 283

Malagarion 284, 291

Malonyx 50

marginalba, Morula 393

Мапае!а 49, 170, 284, 291, 298

marmatensis, Psadara 254, 293

manteli, Pararhytida 282, 290

martensi, Parmarion 168, 284, 291

mastersi, Mystivagor 282, 290

Mastonia 360, 362

matermontana, Polygyra 244, 293

matheroni, Omalonyx 148, 281, 290

Mathildidae 341

mauritiana, Drosophila 312

maximus, Limax 59, 158, 283, 291, 325-
328. 331: 333.335, 336

maynardi, Cepolis 60, 254, 288, 293

Megaspiridae 4, 68, 86-88, 91-94, 285

megei, Diplomphalus 37, 190, 286, 292

Meghimatium 335

НЕЕ, Cochlostyla 288, 293

Melanopsidae 362

meruensis, Trochonanina simulans 284

Mesafricarion 48, 174, 284, 291, 298

Mesodon 48, 244, 287, 293, 299

Mesogastropoda 348, 349

Mesoglypterpes 59

Mesomphix 50, 154, 283, 291, 297

Mesurethra 17, 20, 22, 72

Metaruncina 360, 364

michaudi, Draparnaudia 128, 281, 290

Microhedyle 361, 363

Microhedylidae 346, 347, 360

Microparmarion 49,168, 284, 291, 298

miersi, Simpulopsis 236, 286, 292

Milacidae 2, 27, 38, 52, 57-59, 78-79, 81-
83, 92, 158, 283, 301

Milax 48, 59, 64, 283, 291

miminum, Haplotrema 286, 248, 292

minor, Trigonephrus rosaceus 285

mitella, Pollicipes 374

Modulidae 359

402 INDEX

Modulus 359, 362, 364

Monacha 48, 270, 289, 293, 300

Monadenia 48, 260, 288, 293, 300

montana, Ena 126, 281, 290

Montanobloyetia 284

mordax, Saccostrea 377

Morone 312

Morula 373-395

mouensis, Pararhytida 282, 291

Mus 312

muscarum, Cerithium 348, 362

muscorum, Pupilla 116, 281, 290

musiva, Morula 373-395

mutabilis, Hormomya 377

Mutelidae 365-372

ges 27, 50, 67. 28292907297
ytilus 373, 377, 384, 385

Nata 25, 34, 49, 72, 188, 286, 293, 299

nemoralis, Cepaea 278, 289, 294, 305-315

Nenia 51, 228, 285, 292, 298

Neniinae 68, 87, 285

Nesopupinae 280

Neogastropoda 348

Neotaenioglossa 358

Nerita 358, 361, 364

Neritidae 346, 358, 364

Neritilia 358

Neritimorpha 341, 345, 347, 348, 358

Neritina 358, 364

Neritoidea 346

niger, Hesperarion 283, 291

nigropunctata, Oopelta 283, 291

nitidula, Aegopinella 152, 154, 283, 291

normalis, Mesodon andrewsae 287

novoseelandica, Schizoglossa 190, 286,
293

nubeculata, Psadara 254, 288, 293

Nucella 373, 383-385

Nudibranchia 341, 345-348, 361

oblongus, Tornatellides oblongus 280

oblongus, Strophocheilus 26, 292

oblongus, Tornatellides 102, 280, 290

obvoluta, Helicodonta 264, 289, 294

octona, Subulina 284, 292

Odontostomidae 286

Odontostominae 24, 31, 232, 286

Oleacinidae 2, 10, 20, 25, 27, 31, 33, 37,
43, 52, 62, 64, 66, 71, 72, 78-79, 85,
86, 93, 182, 184, 186, 285, 301

oleatus, Kalidos 166, 284, 291

olivacea, Cepaea nemoralis var. 313

olivieri, Albinaria 226, 285, 292

Omalogyroidea 348

Omalonyx 13, 64, 148, 281, 290, 296

Onchidiidae 348

Oopelta 43, 50, 58, 83, 142, 283, 291, 297

Oopeltinae 283

opalina, Proserpinula 250, 287, 293

Opisthobranchia 341, 357, 360

Orcula 29, 51, 55, 112, 280, 290, 296

Orculidae 1, 30, 52, 53-55, 75, 76, 80, 92,
112, 280, 301

Orehelicidae 2, 22, 24, 30, 52, 64, 70, 84,
9091, 93, 222, 224, 287, 302 303

Oreohelicinae 62, 70, 77-79, 222, 288

Oreohelix 18, 49, 70, 222, 288, 293, 299

Orthalicidae 4, 286

Orthurethra 2, 8-10, 12, 17-20, 25, 29, 30,
36, 37, 43, 47-56, 169) 72 Snore
80, 81, 92, 94, 280, 281, 301

Otina 38

Otinidae 29, 30, 38, 45, 348

Otoconcha 27

Otoconchidae 4

Ouagapia 188, 286, 292

Oxychilus 36, 50, 152, 283, 297

Oxygyrus 359, 362, 364

Pachychilidae 359, 362

paenelimax, Malagarion 284, 291

Pagodulina 30, 51, 55, 112, 281, 290, 296

palawanensis, Helicostylus 248, 288, 293

Paleobulimulus 69

Paleostoa 87

pallens, Atoxon 176, 284, 292

Palythoa 343

Panda 196, 285, 292, 298

Pandofella 26, 48, 49, 71, 200, 285, 292

Paracraticula 65

Paraganitus 361

Paralaoma 136, 282, 291

parallelus, Helicodiscus 136, 282, 291

Pararhytida 9, 13, 50, 282, 290, 291, 297

Parmacella 43, 48, 59, 158, 283, 291, 297

Parmacellidae 1, 24, 38, 43, 52, 57-59, 78-
79, 81, 82, 83, 92, 158, 283, 301

Parmacellinae 284

Parmarion 49, 62, 168, 284, 291, 298

Parmarioninae 168

Parthenina 347, 360

Partula 6, 36, 49, 55, 106, 280, 290, 296

рРамо|дае 1) 13.36: 37. 52155 1756!
80, 81, 92, 104, 106, 280, 301

Partulina 47

Partuloidea 92, 280, 281

parva, Tresia 176, 284, 292

Paryphantopsis 27

Patellogastropoda 348

Patelloida 377

patulus, Discus 136, 282, 291

pectinata, Siphonaria 363

Pedinogyra 49, 202, 285, 292, 298

Pellicula 48, 69, 236, 286, 292, 299

pennsylvanicus, Gryllus 312

percarinatus, Trochozonites 170, 284, 291

perlucidus, Hyalimax 281, 290

perrieri, Heliacus 341-344, 346, 349

Petaloconchus 359

petit, Ampelita 210, 285

petivera, Cepaea nemoralis var. 313

Phaedusiinae 68

Phenacolepadidae 346, 358

Phenacolepas 347, 358, 364

Phenacolimax 50, 64, 150, 283, 291, 297

Phestilla 348

Philine 348, 360, 363

Philinidae 348, 360

Philinoglossa 363

Philinoglossidae 363

Philomycidae (= Arionidae) 58, 78-79, 92,

INDEX

282

Philomycinae 5, 27, 52, 58, 67, 81, 82,
142, 144, 283, 301

Philomycus 28, 50, 142, 144, 283, 291,
297

Phrixgnathus 51, 136, 282, 291, 297

pisana, Theba 274, 289, 294

Pitysinae 47

Placostylus 6, 37, 48, 286, 292, 299

Plagiodontes 36, 49, 232, 286, 292, 299

Planaxidae 359

Planaxis 359

planaxis, Littorina 373

Plectopylis 25, 49, 70, 218, 287, 293, 299

Plectotropis 49, 50, 61, 242, 287, 293, 299

Pleurobranchaea 348

Pleurobranchidae 348

Pleurobranchomorpha 348

Pleurocera 359, 362, 364

"Pleuroceridae" 359

Pleurodiscidae 4, 53, 281

Pleurodiscus 53

Pleurodonte 61, 240, 287, 293, 299

plicosa, Strobilus 102, 280, 290

Plutonia 27, 44, 50, 63, 64, 150, 283, 291,
297

Poiretia 48, 66, 182, 285, 292, 298

Pollicipes 374-389

pollonerai, Microparmarion 168, 284, 291

Polycera 347, 363

Polyceridae 363

Polygyra 50, 61, 244, 287, 293, 299

Polygyridae 2, 30, 52, 59-62, 77-79, 83,
84, 93, 244, 246, 287, 302

Polygyrinae 244, 287

pomatia, Helix 12, 338

Pomatoleios 374-382, 391

Pontohedyle 361, 363

Potamididae 359

pricei, Craterodiscus 218, 287, 293

pricei, Tekoulina 102, 280, 290

Priodiscus 18, 37, 50, 66, 72, 186, 286,
299

Priscodiscus 292

procera, Urocoptis 232, 286, 292

Procerastus 55

Prophysaon 50, 142, 283, 291, 297

propinqua, Succinea 148, 281

Proserpinula 48, 250, 287, 293, 299

Prosobranchia 341

Protodiscus 83, 91

Protornatellina 47

Provitrina 63, 84

Psadara 49, 60, 254, 293, 300

Pseudocharopa 51, 282, 290

Pseudoglessula 48, 65, 178, 285, 292, 298

Ptenoglossa 360

Pterotrachea 359 |

"Pterotracheidae" 359

Ptychopatula 49, 53, 118, 281, 290, 296

ptychophora, Allogonia 287, 293

Ptychotrema 34, 51, 192, 287, 293, 299

pulchra, Auriculella 280, 290

pullata, Amastra 106, 108, 290

Pulmonata 3, 325-339, 348, 361

Punctidae 2, 27, 52, 66-68, 78-79, 87-89,

403
92, 93, 136, 144, 282, 301
punctulata, Acanthina 373
Punctum 67
Puperita 358
purpureo-tincta, Cepaea nemoralis var. 313

Pupilla 51, 116, 281, 290, 296

Pupillidae 1, 52-55, 75-77, 80, 81, 92, 116,
281, 301

Pupillinae 53, 281

Pupilloidea 92, 280, 281

Pupisoma 47

purpurea, Cystopelta 134, 144, 282, 291

pusilla, Lamellidea 100, 280

putris, Succinea 148, 281, 290

pygmaea, Patelloida 377

Pygmipanda 48, 198, 285, 292, 298

Pyramidellidae 341, 342, 346, 349, 360

Pyramidelloidea 341, 345-349, 360

Pyramidula 51, 53, 110, 280, 290, 296

Pyramidulidae 1, 52-55, 75-77, 80, 81, 92,
110, 280, 301

quadrilineata, Polycera 363
Quantula 317-324

quimperiana, Elona 274, 289, 294
Quiscalus 314

quiscula, Quiscalus 314

Rachistia 19, 28, 29, 50, 56, 128, 281,
290, 296

Ranfurlya 27, 50, 66, 67, 134, 290, 297

rarotongensis, Libera fratercula 282

raynali, Ouagapia 188, 286, 292

Reishia 377

reticulatum, Deroceras 325-328, 331, 333,
335, 336

Retusidae 360

Rhagada 49, 287, 293, 299

Rhinoclavis 358, 362

Rhizorus 360, 363, 364

Rhytida 25, 49, 72, 188, 286, 293, 299

Rhytididae 2, 10, 24 25, 27, 31, 33, 37, 45,
52, 70-72, 77-79, 89, 90, 93, 186, 188,
190, 286, 287, 301, 303

ribicundum, Cerion 69

Rillya 65

rosaceus, Trigonephrus 206, 292

rotundatus, Discus 138, 282, 291

rufus, Arion 142, 283, 291

Rumina 30, 51, 65, 180, 285, 292, 298

Rumininae 285

Runcina 342, 347, 350, 360, 364

Runcinidae 347, 360

rupestris, Pyramidula 110, 280, 290

saccharina, Patelloida 377

Saccostrea 377

Sagda 30, 51, 62, 84, 250, 287, 293, 299

Sagdidae 2, 30, 52, 62, 66, 77, 83, 84, 93,
250, 287, 302

Samoana 49, 280, 290, 296

saxatilis, Morone 312

Scaphandridae 360, 363

Schizoglossa 27, 34, 49, 72, 190, 286,
292, 299

Sculptaria 25, 37, 49, 70, 216, 287, 293,

404 INDEX

299
scutulata, Littorina 373
Seila 349
Semibalanus 373, 383, 385
Septaria 361
Septifer 374-391, 393
sequoicola, Helminthoglypta 258, 288, 293
sericata, Halolimnohelix 59, 262, 264, 293
Serpulorbis 359
serratus, Priodiscus 186, 286, 292
serveri, Pagodulina 112, 281, 290
Sigmurethra 17, 22, 24
signatus, Macroceramus 230, 286, 292
similis, Solatopupa 114, 281, 290
Simpulopsis 48, 69, 236, 286, 292, 299
simrothi, Aneitea 146, 281, 290
simulans, Trochonanina 172, 292
sinuata, Aspatharia 365-372
Sinumelon 49, 61, 287, 293, 299
Siphonaria 346-348, 361, 363, 364, 377
Siphonariidae 348, 361
Siphonarioidea 341, 345, 348, 361
Smaragdia 358, 361, 364
Solaropsis 24, 30, 35, 48, 62, 71, 238,
287, 293, 299
Solatopupa 29, 51, 54, 114, 281, 290, 296
somaliensis, Cerastua 126, 281, 290
Sonorella 49,60, 256, 293, 300
Sonorellinae 60, 61, 256, 288
sowerbyanus, Gymnarion 172, 174, 284,
291

Spelaeodiscinae 281

Spelaeodiscus 51

Spermodea 53, 120, 281, 290, 296

Sphincterochila 49, 60, 262, 288, 293, 300

Sphincterochilinae 262, 264, 288

spirata, Vermicularia 362

Spiraxidae (= Oleacinidae) 2, 62, 66, 93

Spiraxinae 11, 62, 285

Spiraxis 18, 50, 66, 184, 292, 298

squamosa, Tetraclita 374

stagnalis, Lymnaea (Lymnaea) 335

Stagnicola 347, 361, 363

Stebplia 50

Stenopylis 57

Stephanoda 50, 58, 66, 134, 282, 291,
297

Sterkia 51, 112, 280, 290, 296

Strebelia 27, 44, 66, 186, 285, 292, 298

Streptaxidae 2, 10, 11, 31, 33, 45, 46, 52,
65-66, 71, 72, 78-79; 85, 86, 92, 93;
192, 287, 302

Streptostyla 66, 285, 292

streptostyla, Streptostyla 285, 292

striata, oe (Quantula) 317-324

Striata, Helicopsis 268, 288, 293

Strobilops 51, 53, 120, 281, 290, 296

Strobilopsidae (= Vallonidae) 47, 52, 53,
92

Strobilopsinae 281

Strobilus 51, 102, 280, 2909, 296

Strophocheilus 25, 26, 30, 48, 71, 208,
285, 292, 298

Strophostomella 65

studerianus, Stylodon 208, 210, 285, 292

Stylodon 26, 36, 49, 71, 208, 210, 285,

292, 298

Stylommatophora 3, 6, 8, 9, 12, 15, 17, 24,
26, 28-31, 37, 92

Subulina 51, 284, 292, 298

Subulinidae 2, 11, 30, 37, 52, 64, 65, 78-
79, 85, 86, 92, 93, 178, 180, 284, 301

Subulininae 284

Succinea 25, 49, 62, 64, 148, 281, 290,
296

Succineidae 2, 19, 25, 27, 31, 52, 64, 68,
77-79, 85, 86, 93, 148, 281, 301

Succineinae 281

suecica, Hedylopsis 363

superbus, Acavus 214, 286, 292

Systrophia 49, 51, 68, 194, 286, 292, 299

Systrophiella 286

Systrophiidae 2, 3, 11, 30, 45, 46, 52, 56,
57, 66-68, 78-79, 87, 88, 92, 93, 194,
286, 301

Tamayoa 194, 286, 292

Tanzaniella 361

taylori, Berendtia 230, 232, 286, 292

Tekoulina 51, 102, 280, 290, 296

Terebralia 359

Teretropoma 343

Testacella 66, 285, 292

Testacellidae (= Oleacinidae) 2, 66, 93,
285

Testacellinae 13, 27, 66, 285

Tetraclita 374-391

Thais 373-395

Thaumatodon 51, 130, 282, 290, 297

Theba 48, 274, 289, 294, 300

Thecosomata 341, 345, 360

Theodoxus 358, 362

Thiaridae 362

Thoracophoridae 50

Thyrophorellidae 4, 283

Thysanophora 49, 60, 252, 293, 300

Thysanophorinae 60, 61, 252, 288

tintinnabulum, Balanus 377

toreuma, Cellana 377

Tornatellides 102, 280, 290

Tornatellina 51, 296

Tornatellinidae (= Achatinellidae) 47, 92,
280

Tornatellininae 47, 52, 280

tourannensis, Bradybaena 246, 288, 293

Trachycystis 51, 132, 282, 290, 297

trachypepla, Trilobopsis 244, 287, 293

Tresia 176, 284, 292

Trichia 48, 266, 289, 294

tridens, Chondrula 122, 281, 290

tridens, Nenia 228, 285, 292

Trigonephrus 49, 206, 285, 292, 298

Trigonochlamydidae 4, 27, 38, 83, 93, 283

trigonostoma, Leptarionta 288, 293

Trilobopsis 244, 287, 293, 299

Triodopsinae 246, 287

Triodopsis 48, 49, 246, 287, 293, 299

Triphora 360, 363

Triphoridae 349, 360, 363

Triphoroidea 341, 345, 348, 360

Trochoidea 288, 293

Trochomorpha 11, 15, 51, 57, 162, 284,

INDEX

291, 298
Trochomorphidae 1, 52, 57, 78-79, 81-83,
92, 162, 284, 301
Trochomorphinae 57
Trochonanina 172, 284, 292, 298
Trochozonites 50, 170, 284, 291, 298
Trochozonitinae 170, 284, 280
Tubuaia 47
turbiniformis, Lacteoluna 287, 293
Turdus 314
Turritella 359, 362
Turritellidae 359, 362

ugandensis, Phenacolimax 150, 283, 291
Ipia 53, 81

umbrata, Amastra pullata 280

Umwelt 320

undata, Solaropsis 238, 287, 293

Unela 361

unifasciata, Candidula 270, 288, 293

Unionacea 365-372

Urocoptidae 2, 22, 24-26, 30, 43, 52, 64,
68-70, 78-79, 86-88, 93, 230, 232, 286,
301

Urocoptinae 286

Urocoptis 51, 69, 86, 232, 286, 292, 299

Urocyclidae 284

Urocyclinae 27, 62, 84, 172, 174, 176, 284

valenciennesi, Parmacella 158, 283, 291

valentiana, Lehmannia 325-339

Vallonia 11, 36, 49, 53, 116, 281, 290,
296

Valloniidae 1, 36, 47, 52-54, 75-77, 80, 81,
92, 116, 118, 120, 122, 281, 301

Valloniinae 281

valvatoidea 348

variabile, Cerithium 362

varians, Cepolis 252, 254

varians, Cepolis 288

varians, Cepolis 293

Varicella 43, 48, 66, 184, 285, 292, 298

Ventridens 50, 56, 68, 160, 283, 291, 297

Vermetidae 347, 359, 362

Vermetoidea 341, 345, 359, 362

Vermetus 359

Vermicularia 362

vernicosa, Nata 188, 286, 293

Veronicellidae 348

Vertiginidae 1, 52, 53-55, 75, 76, 80, 92,
112, 280, 301

vesicalis, Helicophanta 212, 214, 285, 292

virgata, Cernuella 288, 293

virgatus, Septifer 374

Viriola 363

Vitrea 49, 152, 283, 291, 297

Vitreinae 283

Vitrinidae 2, 27, 52, 63-64, 67, 78-79, 83,
84, 93, 150, 283, 301

Vitrinopsis 49, 164, 284, 291, 298

volcano, Balanus tintinnabulum 377

vulgata, Triodopsis fraudulenta 287

walkeri, Sonorella 256, 288, 293
Wayampia 194, 286
whitei, Pandofella 200, 285, 292

405

Xanthonycinae 256, 288
Xenopus 305, 307-310, 312, 314
Xerocystis 282

yatesi, Ammonitella 222, 224, 287, 293

Zacoleus 50, 283, 291, 297

Zebrina 51, 56, 124, 281, 290, 296

zonata, Sphincterochila 262, 288, 293

Zonites 18, 48, 156, 283, 291, 297

Zonitidae 1, 37, 52, 58, 56-57, 59, 78-79,
81-83, 92, 152, 154, 156, 160, 283, 301

Zonitinae 283

Zonitoidea 1,37, 56-59, 81-84, 90-92, 282-
284

Zonitoides 11, 36, 50, 56, 160, 283, 291,
297

MALACOLOGIA, VOL. 30

CONTENTS
N. ABE
Prey value to the carnivorous gastropods Morula musiva (Kiener) and the two
forms of Thais clavigera (Küster): Effect of foraging duration and abandonment
Ге o ара E Aa
J. BLAY, JR.

Morphometry, length-weight relationships and length distributions of five popu-
lations of the freshwater bivalve Aspatharia sinuata (Unionacea, Mutelidae) in
Ме

J. COPELAND & M. М. DASTON
Bioluminescence in the terrestrial snail Dyakia (Quantula) striata .............

J. MASON & J. COPELAND
A mechanism for the creation of conjoined twinning in Lehmannia valentiana
present in the primary oocyte (Gastropoda, Pulmonata) .....................

R. ROBERTSON
Spermatophores of aquatic non-stylommatophoran gastropods: A review with
new data on Heliacus (Architectonicidae) ...................................

O. C. STINE
Cepaea nemoralis from Lexington, Virginia: The isolation and characterization of
their mitochondrial DNA, the implications for their origin and climatic selection .

S. TILLIER
Comparative morphology, phylogeny, and classification of land snails and slugs
(Gastropoda: Pulmonata: Stylommatophora) .......-. 2-22... 6. eee eee nese

373

365

317

325

341

305

AWARDS FOR STUDY AT
The Academy of Natural Sciences of Philadelphia

The Academy of Natural Sciences of Philadelphia, through its Jessup and
МсНепгу 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.

WHY NOT SUBSCRIBE TO MALACOLOGIA?

ORDER FORM

Your name and address

Send U.S. $17.00 for a personal subscription (one volume) or U.S. $27.00 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.

30(1-2)

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.

MALACOLOGIA

1989

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 um.

11. Illustrations are printed either in one
column or the full width of a page of the
journal, so plan accordingly. The maximum
size of a printed figure is 13.5 x 20.0 cm
(preferably not as tall as this so that the cap-
tion does not have to be on the opposite
page).

12. Drawings and lettering must be dark
black on white, blue tracing, or blue-lined
paper. Lines, stippling, letters and numbers
should be thick enough to allow reduction by
Y. ог Уз. Letters and numbers should be at
least 3mm high after reduction. Several
drawings or photographs may be grouped
together to fit apage. 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 pm,” not “163 um”).
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 printed
pages long, MALACOLOGIA requests that an
author pay part of the publication costs.

SUBSCRIPTION COSTS

25. For Vol. 31, personal subscriptions are
U.S. $17.00 and institutional subscriptions
are U.S. $27.00. Address inquiries to the
Subscription Office.

VOL. 30, NO. 1-2 MALACOLOGIA
CONTENTS

S. TILLIER

Comparative morphology, phylogeny, and classification of land snails and slugs =

(Gastropoda: Pulmonata: Stylommatophora) ................................

O. C. STINE
© Cepaea nemoralis from Lexington, Virginia: The isolation and characterization of
their mitochondrial DNA, the implications for their origin and climatic selection .

J. COPELAND & M. M. DASTON |
Bioluminescence т the terrestrial зпай Dyakia (Quantula) striata ........ aid A

J. MASON & J. COPELAND
. A mechanism for the creation of conjoined twinning in Lehmannia valentiana
present in the primary oocyte (Gastropoda, Pulmonata) .....................

R. ROBERTSON
Spermatophores of aquatic non-stylommatophoran gastropods: A review is
new data on Heliacus (Architectonicidae) ...................................

J. BLAY, JR.

lations of the freshwater bivalve Aspatharia sinuata (Unionacea, Mutelidae) i in. A
о E A A A RE RTS DRM ESPN SIRET SE

N. ABE te”
Prey value to the carnivorous gastropods Morula musiva (Kiener) and the two 4
forms of Thais clavigera (Küster): Effect of foraging duration and abandonment
OPPS A Sei ates SR net a A TR

Morphometry, length- weight relationships and length distributions of five popu- A

Malacologia:

inside back co
in table of co

Volume

1(1)
1(2)
1(3)

2 (1)

2 (2)
21630

3 (1)

3 (2)
3218)
4(1)
4(2)
5(1)
5(2)
5(3)
6(1-2)
6(3)
74 3):
7(2-3)
8 (1-2)
O44)
9(2)
10(1)
10(2)
PEE)
11(2)
12(1)
1212)

dates of publication,
Ver SEMI 1 (2) 972]:

ntents or following i
(of 55) are wrongly dated as to the

Actual date

14 November 1962
7 August 1963

1 June 1964

22 September 1964
25 February 1965
29 April 1965

31 August 1965

9 December 1965
31 May 1966

18 August 1966

31 August 1966

31 December 1966
23 June 1967

30 September 1967
31 December 1967
30 June 1968

31 March 1969

13 October 1969
11 November 1969
16 June 1970

20 July 1970
14 November 1970

20 July 1971

8 October 1971
June 1972

25 July 1973
11 March 1974

as given on the
later dates serially

Six numbers

year.

"date" on work

October 1962
July 1963

June 1964
September 1964
February 1965
April 1965
August 1965
November 1965
May 1966

July 1966
August 1966
December 1966
June 1967
September 1967
December 1967
June 1968
"December 1968"
July 1969
October 1969
"November 1969"
"December 1969"
May 1970
"December 1970"
September 1971
May 1972

1973

"19734

13(1-2)
14(1-2)
15(1)
15(2)
16(1)
16(2)
1701)
172)
18(1-2)
19(1)
19(2)
20(1)
20(2)
21(1-2)
22 (1-2)
23(1)
2312)
24 (1-2)
25(1)
25(2)
26(1-2)
272)
27 (2)
28 (1-2)
29(1)
29 (2)
30(1-2)

February 1973

23 January 1974

18
15
12
17.
17
27
18
19
14
22
17

December 1975
July 1976
August 1977
September 1977
February 1978
July 1978

May 1979
September 1979
April 1980
August 1980
June 1981

8 December 1981

24
18
28
29
29
29

June 1982
August 1982
February 1983
September 1983
March 1984
August 1984

9. July 1985
7 March 1986

17
19
28

December 1986
January 1988
June 1988

16 December 1988
1 August 1989

1973
19739
1975
1976
1977
2979
1978
1978
1979
1979
1980
1980
1981
1981
1982
1982
1983
1983
1984
1984
1985
1986
1986
1988
1988
1988
1989

ey
oP |:
4
| 7
|
>
%
6
О : : 7 |
LAS : AÑ q

Wi A

3 2044 072 160 526

Date Due

COR ие RE

LL Bate Ne ERP ОНИ (US x = b Sun y : 2 k ue is sy 1 y г E LE TE LME mer
DANONE us ми FAN HS НИИ ы u en “es ния => - A с =.
! pe 2 a a Ny MoM м > ge 4 WE nues

O ree"

„я + sn

de ra
PET

A Ai ==.

CT elas eal

wenn

LE ei

en una en N

eu
men gens mare

PPT C вероятнее =
IR A m м

ps a e ic
deg BOERNE Hy > ke
; : rg LES
“rene ee vn г
ne nates
a ot
room

e OS эм Deere
pS on MVP RATER AE Е ete es

ar sn ip
PR TENNIS

er AR Sp

Ver
CURE
mv

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Malacologia"

See other formats