Malacologia

Mens fine
ATA SAS MMs tin
ch

eur ca A ds две AER)

КАН > M D A 5 Deven

nas
A

eat

Satin Pa Ns pnt ewe Maran aha AGA Ann Maha SEHEN

oran ine mb atmet ah Gods magn caste Vel Palla ados La de da Pate Pr

cars Ho mdr à bn E O A
A Bev ated Fer,

4 PAPAS 2.
A

HARVARD UNIVERSITY
13
Library of the

Museum of

Comparative Zoology

=)
И | au
DA UN A |
В И | PU й И К 1 LA I LUN
Pr LA 4 LM QUI y
Ñ
|

AS | }
| a: ye в ALI ] We | J
и AUS A O AA
Ba.
| yl le y IM eN ve y
a ie у м р | À Pry Eu
т Я Po i | o | у
| я & SY
A i)

VOL. 34, NO. 1-2 1992

MALACOLOGIA

nternational Journal of Malacology
Revista Internacional de Malacologia
Journal International de Malacologie
Международный Журнал Малакологии

Internationale Malakologische Zeitschrift

Vol.
Vol.
Vol.
Vol.
Vol.
Vol.
Vol.
Vol.

Publication dates
1-2

28, No.
29, No.
29, No.
30, No.
31, No.
31, No.
32, No.
33, No.

19 January 1988
28 June 1988
16 Dec. 1988

1 Aug. 1989
29 Dec. 1989
28 May 1990

7 June 1991

6 Sep. 1991

1992

MCZ
LIBRARY

SEP 115, 1992

HARVARD

y UNIVERSITY
N MALACOLOGIA
ug

yr

BE visto Internacional de Malacologia

JL.
Pt, =
ne?
|
EN
AU

у ieee

1 Journal International de Malacologie
__ MexayHapoanuä Журнал Малакологии

Internationale Malakologische Zeitschrift

MALACOLOGIA

Editor-in-Chief:

GEORGE M. DAVIS

Editorial and Subscription Offices:

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

EUGENE COAN
California Academy of Sciences
San Francisco, CA

Co-Editors:

CAROL JONES
Denver, CO

Assistant Managing Editor:
CARYL HESTERMAN

Associate Editors:

JOHN B. BURCH
University of Michigan
Ann Arbor

ANNE GISMANN
Maadi

Egypt

MALACOLOGIA is published by the INSTITUTE OF MALACOLOGY, the Sponsor Members

of which (also serving as editors) are:

KENNETH J. BOSS
Museum of Comparative Zoology
Cambridge, Massachusetts

JOHN BURCH, President-Elect

MELBOURNE R. CARRIKER
University of Delaware, Lewes

GEORGE M. DAVIS
Secretary and Treasurer

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

JAMES NYBAKKEN, President
Moss Landing Marine Laboratory
California

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

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

SHI-KUEI WU
University of Colorado Museum, Boulder

Participating Members

EDMUND GITTENBERGER
Secretary, UNITAS MALACOLOGICA
Rijksmuseum van Natuurlijke

Historie

Leiden, Netherlands |

JACKIE L. VAN GOETHEM
Treasurer, UNITAS MALACOLOGICA
Koninklijk Belgisch Instituut

voor Natuurwetenschappen

Brussel, Belgium

Emeritus Members

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

ELMER G. BERRY,
Germantown, Maryland

ROBERT ROBERTSON
The Academy of Natural Sciences
Philadelphia, Pennsylvania

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

Copyright © 1992 by the Institute of Malacology

|
я
у

J. A. ALLEN
Marine Biological Station
Millport, United Kingdom

R. BIELER
Field Museum
Chicago, U.S.A.

E. E. BINDER
Muséum d'Histoire Naturelle
Genève, Switzerland

A. J. CAIN
University of Liverpool
United Kingdom

P. CALOW
University of Sheffield
United Kingdom

J. G. CARTER
University of North Carolina
Chapel Hill, U.S.A.

R. COWIE
Bishop Museum
Honolulu, HI., U.S.A.

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

B. C. CLARKE
University of Nottingham
United Kingdom

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

C. J. DUNCAN
University of Liverpool
United Kingdom

D. J. EERNISSE
University of Michigan
Ann Arbor, U.S.A.

V. FRETTER
University of Reading
United Kingdom

1992
EDITORIAL BOARD

E. GITTENBERGER
Rijksmuseum van Natuurlijke Historie
Leiden, Netherlands

F. GIUSTI
Universita di Siena, Italy

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

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

A. V. GROSSU
Universitatea Bucuresti
Romania

T. HABE
Tokai University
Shimizu, Japan

R. HANLON
Marine Biomedical Institute
Galveston, Texas, U.S.A.

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

D. M. HILLIS
University of Texas
Austin, U.S.A.

K. E. HOAGLAND
Association of Systematics Collections
Washington, DC, U.S.A.

B. HUBENDICK
Naturhistoriska Museet
Góteborg, Sweden

S. HUNT
Lancashire
United Kingdom

R. JANSSEN

Forschungsinstitut Senckenberg,
Frankfurt am Main, Germany

R. N. KILBURN
Natal Museum
Pietermaritzburg, South Africa

M. A. KLAPPENBACH
Museo Nacional de Historia Natural
Montevideo, Uruguay

J. KNUDSEN
Zoologisk Institut & Museum
Kobenhavn, Denmark

A. J. KOHN
University of Washington
Seattle, U.S.A.

A. LUCAS
Faculté des Sciences
Brest, France

C. MEIER-BROOK
Tropenmedizinisches Institut
Túbingen, Germany

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.

J. P. POINTER
Ecole Pratique des Hautes Etudes
Perpignan Cedex, France

W. F. PONDER

Australian Museum

Sydney

R. D. PURCHON

Chelsea College of Science & Technology
London, United Kingdom

QUEZ YE
Academia Sinica
Qingdao, People's Republic of China

D. G. REID
The Natural History Museum
London, United Kingdom

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

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

A. STANCZYKOWSKA
Siedlce, Poland

F. STARMÜHLNER
Zoologisches Institut der Universität
Wien, Austria

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

W. STREIFF
Université de Caen
France

J. STUARDO
Universidad de Chile
Valparaiso

S. TILLIER
Muséum National d’Histoire Naturelle
Paris, France

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

J.A.M. VAN DEN BIGGELAAR
University of Utrecht
The Netherlands

J. A. VAN EEDEN
Potchefstroom University
South Africa

N. H. VERDONK
Rijksuniversiteit
Utrecht, Netherlands

B. R. WILSON
Dept. Conservation and Land Management
Netherlands, Western Australia

H. ZEISSLER
Leipzig, Germany

A. ZILCH
Forschungsinstitut Senckenberg
Frankfurt am Main, Germany

MALACOLOGIA, 1992, 34(1-2): 1-24

BIOLOGY AND COMPARATIVE ANATOMY OF THREE NEW SPECIES OF
COMMENSAL GALEOMMATIDAE, WITH A POSSIBLE CASE OF MATING
BEHAVIOR IN BIVALVES

Paula M. Mikkelsen' & Rüdiger Bieler?

ABSTRACT

Three new galeommatid bivalves, Divariscintilla octotentaculata, D. luteocrinita, and D. cordi-
formis, are described as commensals occupying the burrows of the mantis shrimp Lysiosquilla
scabricauda from central eastern Florida. Morphological comparisons are made with all other
known members of the genus, comprising two previously described species from the same
burrow system (D. yoyo, D. troglodytes) and the type species, D. maoria, from New Zealand.
Key characters defining this genus (hinge morphology, flower-like organs, “hanging foot” struc-
ture) are discussed, especially with regard to their presence in other galeommatoidean genera.
Intraspecific interaction resembling mating behavior is noted and discussed as one of the few
possible examples in the Bivalvia.

Key words: Divariscintilla, Galeommatoidea, Bivalvia, systematics, anatomy, mating behavior,

commensalism, Stomatopoda.

INTRODUCTION

Investigation of the organisms associated
with the sand-burrowing mantis shrimp
Lysiosquilla scabricauda (Lamarck, 1818)
(Crustacea: Stomatopoda: Lysiosquillidae) in
shallow waters of eastern Florida has re-
vealed a community of seven molluscan spe-
cies that appear highly dependent on this
specialized habitat. Remarkably, all of these
species were found to be either poorly known
or undescribed. Accounts of the two species
of vitrinellid gastropods—Cyclostremiscus
beauii (Fischer, 1857) and Circulus texanus
(Moore, 1965); Bieler & Mikkelsen, 1988—
and two of the five species of bivalves—Di-
variscintilla yoyo Mikkelsen & Bieler, 1989,
and D. troglodytes Mikkelsen & Bieler,
1989—have appeared elsewhere. This re-
port deals with the remaining three species of
galeommatoidean bivalves. Although superfi-
cially different from the two species previously
described, and preliminarily treated as mem-
bers of another genus (as Scintilla spp.; Mik-
kelsen & Bieler, 1989: 192; Eckelbarger et al.,
1990), detailed study has revealed their
proper placement in Divariscintilla, and has
shown them to be in fact more similar to the
New Zealand type species, D. maoria Powell,
1932, than were the two species previously
described from Florida stomatopod burrows.

MATERIAL AND METHODS

Lysiosquilla burrows in shallow-water sand
flats at several locations on the central eastern
Florida coast were sampled using a stainless
steel bait pump (“yabby pump”) and sieves of
1-2 mm mesh. Sampling depths during ex-
treme low water ranged from less than 0.5 m
to supratidal, when the water level lay several
centimeters below the level of the sand.

Living clams were maintained in finger
bowls of seawater at ambient laboratory con-
ditions (22-25°C), with variable lighting. Water
was changed every 1-2 days, and an irregu-
larly-supplied, unmeasured diet of mixed uni-
cellular algae (e.g., /sochrysis, Chlorella,
Chaetoceras) was provided. Behavioral stud-
ies were aided by video recordings taken of
the living animals in aquaria using a standard
commercial 1/2-inch-format video camera
equipped with a macro lens.

Transfer of specimens between laboratory
bowls was best accomplished using small
spoons. The spoon could be applied against
the glass from below the specimen to gently
break the byssus threads while also cradling
the clam. Handling these kinds of animals
with forceps is awkward and frequently
causes damage to fragile shells and tissue.

Relaxation prior to dissection or preserva-
tion was most effectively accomplished with

‘Harbor Branch Oceanographic Museum, Harbor Branch Oceanographic Institution, 5600 Old Dixie Highway, Ft. Pierce,
Florida 34946 U.S.A., and Department of Biological Sciences, Florida Institute of Technology, Melbourne, Florida 32901

U.S.A.

“Department of Zoology, Field Museum of Natural History, Roosevelt Road at Lake Shore Drive, Chicago, Illinois 60605

U.S.A.

2 MIKKELSEN 8 BIELER

crystalline magnesium sulfate, added directly
to the finger-bowl water in unquantified small
amounts. Methylene-blue/basic-fuchsin and
neutral red were used to delineate tissues and
organs in gross dissections.

For histological serial sections, animals
were fixed in 5% buffered formalin (Humason,
1962: 14). Shells were decalcified using either
dilute (approximately 0.5%) hydrochloric acid
(complete decalcification within minutes) or a
1% solution of ethylenediamene tetraacetic
acid (EDTA, adjusted to pH 7.2; decalcification
complete in 5-6 days). Specimens were em-
bedded in paraplast, sectioned at 8 um and
stained with Gomori's green trichrome (mod-
ified from Vacca, 1985). Staining reactions de-
scribed in the text refer to this method. Colors
referred to in the text are supplied for future
use, that is, to infer homologies of the various
glands. Photomicrographs of sections were
taken with an Olympus BH-2 stereomicro-
scope fitted with an Olympus OM-2 camera
with Kodak Panatomic-X (ASA 32) film.

For scanning electron microscopy (SEM),
preserved specimens were passed through
an ethanol-to-acetone series and critical-point
dried. These and air-dried shells were coated
with gold/palladium and examined using a
Zeiss Novascan-30 scanning electron micro-
scope.

All cited anatomical measurements were
taken from specimens of average size (see
under descriptions). Throughout the text,
“relaxed” refers to the condition of an animal
in normal crawling posture and does not refer
to any chemical treatment.

Cited institutions are (* indicates location of
type and other voucher material):

AMNH— American Museum of Natural His-
tory, New York, New York

DMNH— Delaware Museum of Natural His-
tory, Wilmington

*FMNH— Field Museum of Natural History,

Chicago, Illinois

Harbor Branch Oceanographic In-

stitution, Ft. Pierce, Florida

*HBOM— Harbor Branch Oceanographic

Museum [formerly Indian River

Coastal Zone Museum], HBOI

Museum of Comparative Zoology,

Harvard University, Cambridge,

Massachusetts

*SBMNH—Santa Barbara Museum of Natural
History, California

SMSLP— Smithsonian Marine Station at
Link Port, Ft. Pierce, Florida

HBOI—

*MCZ—

*USNM— National Museum of Natural
History, Smithsonian Institution,
Washington, D. C.

TAXONOMIC DESCRIPTIONS
Family GALEOMMATIDAE Gray, 1840
Genus Divariscintilla Powell, 1932

Type species: Divariscintilla maoria Powell,
1932: 66; by original designation. Recent,
New Zealand.

Remarks: Redescribed by Mikkelsen &
Bieler (1989: 193). See remarks concerning
generic placement under Discussion.

Divariscintilla octotentaculata n. sp.
(Figs. 1, 4, 7-13, 23, 33)

Material examined

Holotype: 5.3 mm [shell length], FMNH
223401. Paratypes (31): 4.2, 4.2 mm, FMNH
223402; 3.2, 3.2, 3.2, 3.3, 3:3, 3.6, 380m:
HBOM 064:01866; 4.2, 4.2, 3.6, 2.9 mm (pre-
served soft-bodies + shells coated for SEM),
HBOM 064:01867; 3.9, 4.3, 4.3 mm, HBOM
064:01865; 3.0, 3.1, 3.2, 3.4, 3.6 mm, USNM
859443; 3.2, 3.2, 3.2, 3.4, 4.1 mm, MCZ
302510; 1.7, 3.4, 3.4, 3.8, 3.8 mm, SBMNH
35167. Total material: 262 specimens: FLOR-
IDA: Ft. Pierce Inlet: 10 March 1987, 3; 2-3
May 1987, 73; 24 June 1987, 91; 03 August
1987, 2; 14 August 1987, 10; 31 August 1987,
1; 28 December 1987, 7; 11 March 1988, 7;
12 April 1988, 38; 16 October 1990, 2; 03
February 1991; 13. -Sebastian Inlet: 30 De-
cember 1987, 26. -St. Lucie Inlet: 18 February
1982, 2.

Type locality

Ft. Pierce Inlet, Indian River Lagoon, St.
Lucie County, eastern Florida, 27°28.3'N,
8017.9'W, occupying Lysiosquilla scabri-
cauda burrows on intertidal sand flats with
patches of the seagrass Halodule wrightii As-
cherson. Paratypes all from type locality.

Diagnosis

Animal translucent white. Mantle thin, with
retractable, papillose folds covering anterior
and posterior thirds of shell. Tentacles origi-
nating at dorsal shell margin, two pairs ante-
riorly, two pairs posteriorly. Posterior foot-
extension relatively short. Shell roundly

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 3

FIGS. 1-3. Photographs of living animals. 1. Divariscintilla octotentaculata. 2. D. luteocrinita. 3. D. cordi-
formis. Scale bars = 1.0 mm.

+ MIKKELSEN & BIELER

FIGS. 4-7. External appearance and internal shell morphology. Tentacle pairs numbered from anterior to
posterior (for text reference only; identical numbers do not imply homology). 4. Divariscintilla octotentaculata,
in crawling position, from left side. 5. D. luteocrinita, same as Fig 4. 6. D. cordiformis, same as Fig. 4. 7.D.
octotentaculata, internal surface of right valve, showing approximate locations of muscle insertions. Scale
bars = 1.0 mm. (aam, anterior adductor muscle scar; apr, anterior pedal retractor muscle scar; bag, byssus
adhesive gland; by, location of byssus gland; c, cowl; cst, club-shaped tentacles; exc, excurrent “siphon”;
ft, foot; inc, incurrent “siphon”; ind, shell indentation; m, mantle fold; mf, point of mantle fusion; mp, mantle
pouch; mpt, median pallial tentacle; pam, posterior adductor muscle scar; pdc, prodissoconch; pfe, posterior
foot-extension; ppm, pedal protractor muscle scar; ppr, posterior pedal retractor muscle scar; sh, shell; vgr,

ventral pedal groove).

triangular, with umbo slightly anterior, smooth
exteriorly, with weak radial ribs interiorly;
length approximately 70% of extended mantle
length. No “flower-like organ” on anterior sur-
face of visceral mass.

Description

External Features and Mantle: Living ex-
tended animal (Figs. 1, 4) generally 5-7 mm in
length, translucent white except for dark di-
gestive gland within visceral mass. Shell
largely external, with only anterior and poste-
rior thirds of shell covered in life by mantle
folds (Fig. 4, m), not meeting at lateral mid-
line. Mantle folds fully retractable, finely pap-
illose, with scattered larger papillae. Mantle
thin, extending beyond shell edges; entire
surface finely papillose, with fringe of longer

papillae along ventral margin of shell. Two
pallial openings: (1) anteropedal opening,
from a point anterior to umbones to a point
posterior to foot (Fig. 4, mf), forming exten-
sive anterior cowl (Fig. 4, c), the edges of
which are held together to form an effective
incurrent “siphon” (Fig. 4, inc); and (2) pos-
terodorsal excurrent opening (Fig. 4, exc) be-
tween posterior tentacle pairs, either as sim-
ple opening or on rounded protuberance
forming distinct siphon; appearance depen-
dent on degree of pallial expansion. Posterior
mantle fusion forming a rounded, protruding
pouch (Fig. 4, mp) containing the gills. Two
pairs (one long, one short) retractable, ante-
rior tentacles arising from near shell edge;
dorsalmost pair longest. Two pairs (one long,
one short) retractable tentacles in vicinity of
excurrent siphon, also arising from near shell
edge; posteriormost pair longest. A single,

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 5

short, mid-dorsal tentacle (Fig. 4, mpt), just
posterior to umbo. An additional 1-2 pairs of
short tentacles (=enlarged papillae?) in
larger specimens (approximately 5 mm) at
ventral shell edge at anterior and posterior
ends of ventral papillose fringe. Each tentacle
with papillose surface and central core of lon-
gitudinal muscle and nerve fibers, visible as
an inner “thread” under low magnification. In-
ner pallial fold of non-shell areas (e.g., cowl,
ventral mantle margin, posterior pouch) highly
muscular.

Preserved specimens completely (or nearly
so) retracted into shell, however, shell usually
gaping, with tentacles (especially anterodor-
sal) slightly protruding beyond shell edges.

Shell (Figs. 7-11, 13): Shell generally 3-5 mm
in length, roundly triangular, longer posteri-
orly, equivalve, rather compressed, glossy, ir-
idescent, transparent, smooth except for fine
concentric growth lines and weak radial ribs
most evident interiorly at ventral margin (Fig.
9). Size large relative to mantle, comprising
approximately 70% of extended mantle
length. Valves held open at a 50-60x angle
while crawling, capable of complete closure
ventrally but gaping slightly anteriorly and
posteriorly. Adductor muscle scars faint,
subequal (Figs. 7, 9). Pallial line entire, indis-
tinct. Periostracum colorless, most evident at
ventral shell edge (Fig. 9, per).

Hinge line short (Fig. 10). One small,
rounded cardinal tooth (Fig. 10, car) in each
valve, that of left valve sometimes slightly bi-
fid; lateral teeth absent. Cardinal teeth abut-
ting, not interlocking. External ligament (Fig.
10, lig) weak, amphidetic, supported by
nymph. Internal ligament (resilium; Fig. 10,
res) stronger, opisthodetic.

Prodissoconch (Fig. 11) brownish-yellow,
approximately 360 fm in length. Prodisso-
conch | corresponding in size to shell of newly
released larva, approximately 32% of length
of prodissoconch II; sculpture not discernible
(surface abraded in adult shell). Prodisso-
conch II sculptured only with coarse and fine
concentric growth lines. Prodissoconch | and
| stages distinct, demarcated more by
change in convexity than by sculpture or
growth discontinuity (Fig. 11, single arrow).
Demarcation between prodissoconch II and
dissoconch abrupt (Figs. 8; 11, double arrow).

Shell microstructure (Fig. 13) cross-
lamellar centrally, with thin prismatic layer
covering each side, as in Divariscintilla yoyo
(see Mikkelsen & Bieler, 1989).

Organs of the Pallial Cavity: Foot (Fig. 4, ft) as
previously described for Divariscintilla yoyo
(see Mikkelsen & Bieler, 1989), including
hatchet-shaped anterior portion, narrowed
posterior extension (Fig. 4, pfe), anterior bys-
sus gland (Fig. 4, by), ciliated ventral groove
(Fig. 4, vgr) supplied with numerous mucous
glands, and terminal byssus adhesive gland
(Fig. 4, bag). Byssus gland of undefined
structure, staining turquoise in histological
sections; byssus adhesive gland of branching
lamellar folds, staining purplish-red. One to
four byssus threads produced, emanating
from extreme posterior tip of foot. Opaque
white pigment band (of unknown function)
along anterodorsal tip of foot, staining dark
purplish-red in sections.

Anterior and posterior adductor muscles
subequal, of moderate diameter. Anterior and
posterior pedal retractor muscles smaller in
diameter, inserting on shell just dorsal and
medial to their respective adductor muscle
scars. Very small pedal protractor muscle
merging with anteroventral edge of anterior
adductor muscle just before both attach to
shell; inserting into anterior visceral mass just
dorsal of labial palps. Muscles leaving very
faint attachment scars on shell (Figs. 7, 9).

Overall morphologies of visceral mass, la-
bial palps and ctenidia as in Divariscintilla
yoyo (see Mikkelsen 8 Bieler, 1989). Palps
with 6-8 lamellae each side. Ctenidia smooth,
unpleated (appearing posteriorly loosely
pleated in preserved specimens due to con-
traction). Outer demibranch approximately
50% smaller than inner; both demibranchs
ventrally rounded and with both interfilamen-
tal and interlamellar connectives. Cilial cur-
rents on palps and ctenidia not verified.

Flower-like organ absent (see Discussion
below).

Digestive System: Structure of digestive sys-
tem (mouth, esophagus, stomach, midgut,
hindgut, rectum) of same organization as that
in Divariscintilla yoyo (see Mikkelsen & Bieler,
1989: fig. 28).

Suprabranchial Chamber. Arrangement of
Openings and presence of glandular patches
(?hypobranchial glands; Fig. 23, hyp) adja-
cent to rectum as in Divariscintilla yoyo (see
Mikkelsen & Bieler, 1989: fig. 29).

Nervous System: Arrangement of ganglia,
statocysts, and major nerves as described for
Divariscintilla yoyo (see Mikkelsen & Bieler,
1989: fig. 31). Left and right posterior tenta-

6 MIKKELSEN 8 BIELER

cles innervated by branches from the pallial
nerve, adjacent to its junction with the visceral
ganglion. Anterior tentacles similarly inner-
vated but both from a common branch of the
pallial nerve, adjacent to its junction with the
cerebro-pleural ganglion.

Reproductive System: Simultaneous her-
maphrodite. Ovotestis white, encompassing
most of volume of visceral mass, as in Di-
variscintilla yoyo (see Mikkelsen & Bieler,
1989).

Mature spermatozoa morphologically indis-
tinguishable from that of D. yoyo (see Eckel-
barger et al., 1990: figs. 28-29, table 1; D.
octotentaculata as Scintilla sp.), except
smaller in relative size. Spermatogenesis fully
described by Eckelbarger et al. (1990; as
Scintilla sp.).

Brooding large number of small larvae for
variable period; brooding time for individuals
collected with larvae, 9-15 days (n = 4); total
brooding time from set to release in labora-
tory, 9, 10, and 12 days (n = 3). Larvae held
within both demibranchs and in suprabran-
chial chamber, where they are circulated via
pallial expansions and contractions. During
brooding, excurrent siphonal opening con-
stricted by sphincter-like muscle, sometimes
noted around free end of rectum, allowing di-
gestive processes to continue. Larvae initially
white, turning pink with shell development on
day 5-7 (n = 2); released as straight-hinged
“D” larvae with apical flagella, 115-123 um in
shell length (X = 119 um, п = 40; Fig. 12).
Larvae expelled through excurrent siphon via
strong contractions of shell and pallial mus-
cles. Adults brooding larvae collected in May,
June, December 1987, and March, April
1988; additional adults setting larvae in labo-
ratory in February, May, June 1982 and No-
vember 1990; one specimen setting two
broods in laboratory, four months apart, with
second brood approximately 20% quantity of
first. No apparent seasonality.

Circulatory and Excretory Systems: As de-
scribed for Divariscintilla yoyo (see Mikkelsen
& Bieler, 1989).

Distribution and Abundance

Known from three locations, all on intertidal
and shallow subtidal sand flats within the In-
dian River Lagoon, eastern Florida: the type
locality, Ft. Pierce Inlet (St. Lucie County,
27°28.3’М, 80°17.9’W), just north of St. Lucie
Inlet (Martin County, 27°11.4’N, 80°11.1’W),

and Sebastian Inlet (Brevard County,
27°51.6’М, 80°27.0’М/). May be quite numer-
ous; largest number per burrow sample = 74.

Etymology

An adjective, octotentaculatus, -a, -um,
from the Latin octo (eight) and the late Latin
tentaculum (a “feeler”), referring to the eight
long mantle tentacles, a diagnostic feature.

Remarks

This is the most common of the five Divaris-
cintilla species in the Lysiosquilla burrows
(see Ecology and Behavior).

Prior to the beginning of this study in March
1987, two individuals of this species were en-
countered, and are presently the only known
specimens of any of the Floridian burrow ga-
leommatoideans from previous collections.
These were collected in a shovel-and-sieve
sample from a sand bar in the Indian River
Lagoon, north of St. Lucie Inlet, Martin
County, Florida, 27°11.4’N, 80°11.1’W, on 18
February 1982. These individuals were main-
tained in the laboratory for approximately four
months, providing material for notes and pho-
tographs on behavior, reproduction, and de-
velopment.

Divariscintilla luteocrinita n. sp.
(Figs. 2, 5, 14-22, 24-25)

Material examined

Holotype: 5.0 mm (shell length), FMNH
223403. Paratypes (12): 4.6 mm, FMNH
223407; 4.4 mm (shell only), FMNH 223404;
3.2, 3.9 mm, HBOM 064:01863; 4.9, 4.5, 4.1,
2.9 mm (shells only, coated for SEM), HBOM
064:01864; 4.8 mm, USNM 860195; 5.5 mm
(shell only), USNM 859444; 3.4 mm, MCZ
302517; 3.5 mm, SBMNH 35168. Total mate-
rial: 16 specimens: FLORIDA: Ft. Pierce Inlet:
10 March 1987, 1; 2-3 May 1987, 5; 24 June
1987, 1; 03 August 1987, 1; 14 August 1987,
2; 31 August 1987, 1; 12 April 1988, 1; 03
February 1990; 4.

Type locality

Ft. Pierce Inlet, Indian River Lagoon, St.
Lucie County, eastern Florida, 27°28.3'N,
80°17.9'W, occupying Lysiosquilla scabri-
cauda burrows on intertidal sand flats with

NEW SPECIES OF COMMENSAL GALEOMMATIDAE

lig car

Figs. 8-13. Shell of Divariscintilla octotentaculata (SEM). 8. Left valve, external view, 2.9 mm length, para-
type, HBOM 064:01867. 9. Right valve, internal view, 3.6 mm length, paratype, HBOM 064:01867. 10.Hinge,
anterior to left, paratype, HBOM 064:01867. Scale bar = 100 рт. 11. Prodissoconch, 360 „m length. Single
arrow = prodissoconch |-!! boundary. Double arrow = prodissoconch Il-dissoconch boundary. 12. Newly
released larval shell, 119 um length. 13. Microstructure, with internal surface at top. Scale bar = 5 um. (car,
cardinal tooth; lig, external ligament; nym, nymph; per, periostracum; res, resilium).

MIKKELSEN & BIELER
17 E

Figs. 14-20. Shell of Divariscintilla luteocrinita (SEM). 14. Left valve, external view, 2.9 mm length, paratype,
HBOM 064:01864. 15. Right valve, internal view, 4.9 mm length, paratype, HBOM 064:01864. 16. Hinge,
anterior to left, paratype, HBOM 064:01864. Scale bar = 100 „m. 17. Prodissoconch, 390 am length. Single
arrow = prodissoconch I-II boundary. Double arrow = prodissoconch Il-dissoconch boundary. 18. Internal
surface, showing smooth adductor muscle scar (left) and adjacent region of opaque thickening (right). Scale
bar = 10 um. 19. Microstructure in region of adductor muscle scar; internal surface at top. Scale bar = 5
um. 20. Microstructure in region of opaque thickening, showing additional layer covering internal prismatic
layer (pr); internal surface at top. Scale bar = 5 wm. (pr, internal prismatic layer).

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 9

patches of the seagrass Halodule wrightii As-
cherson. Paratypes all from type locality.

Diagnosis

Animal translucent yellow. Mantle thin, with
extensive, retractable, papillose folds com-
pletely covering shell, meeting at mid-line.
Tentacles originating at shell edge, three
pairs anteriorly, two singles and 9-11 pairs
posteriorly, plus one pair club-shaped tenta-
cles adjacent to excurrent opening. Posterior
foot-extension relatively short. Shell roundly
triangular, with umbo slightly anterior, smooth
exteriorly, with opaque thickenings interiorly;
length approximately 70% of extended mantle
length. Single “flower-like organ” on anterior
surface of visceral mass.

Description

External Features and Mantle: Living ex-
tended animal (Figs. 2, 5) generally 6-7 mm in
length. Mantle and tentacles translucent pale
yellow; foot white. Upper portion of digestive
gland showing through mantle and shell as
dark elongate-oval spot. Shell entirely cov-
ered in life by anterior and posterior mantle
folds (Fig. 5, m), meeting at lateral dorso-ven-
tral midline on each side. Mantle folds thin,
incompletely retractable, entirely finely papil-
lose, with scattered, elongated papillae espe-
cially posteroventrally. Mantle edge extending
widely beyond shell edges; entire surface
finely papillose. Pallial openings, muscula-
ture, and posterior pouch as in Divariscintilla
octotentaculata. Numerous, long, retractile
tentacles originating at shell edge: three pairs
anterior to umbo; two singles plus five pairs
posterior to umbo (= two singles on midline
+ two pairs + [excurrent siphon] + three
pairs). Shorter accessory pairs posterior to
umbo originating from mantle fold near, but
ventrad of, shell edge: one to three pairs an-
terior, two to three pairs posterior to excurrent
siphon. Structure of these tentacles as in D.
octotentaculata; those on shell edge adjacent
to cowl and posterior pouch (third and sixth,
see numbers, Fig. 5) longest. One prominent
pair of thicker, whitish, club-shaped tentacles
(Fig. 5, cst) immediately posterior to excurrent
siphon originating just inside shell edge; inter-
nal structure differing slightly from that of
other tentacles (see Remarks).

Preserved specimens incompletely re-
tracted into gaping shell; mantle folds con-
tracted to narrow rim along shell edge, expos-

ing most of shell surface; tentacles, cowl,
posterior pouch, and foot contracted but still
usually extending beyond shell edges.

Shell (Figs. 14-20): Shell generally 4-5 mm in
length, roundly triangular to oval, longer pos-
teriorly, equivalve, rather inflated, glossy,
transparent to translucent white, smooth ex-
cept for fine concentric growth lines and irreg-
ular opaque thickening interiorly, imparting a
white-blotched pattern (Fig. 15). Size large rel-
ative to mantle, comprising approximately
70% of extended mantle length. Valves held
open at approximately 40° angle while crawl-
ing, incapable of complete closure. Adductor
muscle scars subequal, distinct due to pres-
ence of surrounding shell thickening (Fig. 15).
Periostracum as in Divariscintilla octotenta-
culata.

Hinge line short (Fig. 16), similar to that of
Divariscintilla octotentaculata. Both cardinal
teeth rounded.

Prodissoconch (Fig. 17) brownish-yellow,
approximately 390 ym in length. Prodisso-
conch | approximately 145 um in length, ap-
proximately 37% of length of prodissoconch
Il; sculpture not discernible (surface abraded
in adult shell). Prodissoconch И sculptured
with coarse concentric growth lines. Demar-
cation between prodissoconch | and Il stages
(Fig. 17, single arrow), and between prodis-
soconch Il and dissoconch (Figs. 14; 17, dou-
ble arrow) as in Divariscintilla octotentaculata.

Shell microstructure as in Divariscintilla oc-
totentaculata, except with additional layer of
parallel crystals covering internal perpendic-
ular prismatic layer, forming regions of
opaque thickening (Figs. 18-20).

Organs of the Pallial Cavity: Foot (Fig. 5, ft) as
in Divariscintilla octotentaculata. Adductor,
pedal retractor, and pedal protractor muscles
(including relative positions) as in D. octoten-
taculata. Muscles leaving distinct attachment
scars on shell (Fig. 15).

Visceral mass, labial palps and ctenidia as
in Divariscintilla octotentaculata. Palps with
approximately 8 lamellae each side. Outer
demibranch approximately 40% smaller than
inner. Cilial currents on palps and ctenidia un-
known.

A single “flower-like organ” (Figs. 21-22;
see Mikkelsen & Bieler, 1989, for explanation
of term) on anterior surface of visceral mass
just ventral to labial palps. Size variable, not
correlated with shell length.

Digestive System: As in Divariscintilla oc-
totentaculata.

10 MIKKELSEN & BIELER

Suprabranchial Chamber. As in Divariscintilla
octotentaculata, except that the whitish glan-
dular patches (?hypobranchial glands) are
much more extensive (Fig. 24).

Nervous System: Arrangement of ganglia,
statocysts, and major nerves as described for
Divariscintilla yoyo (see Mikkelsen & Bieler,
1989: fig. 31). Additional tentacular nerves
arising (independently) from pallial nerve.

Reproductive System: Overall gross morphol-
ogy as in Divariscintilla octotentaculata. Re-
productive mode could not be determined
from sectioned specimen, which showed no
recognizable developed gametes. Adults
brooding larvae have not been collected.

Circulatory and Excretory Systems: As in Di-
variscintilla octotentaculata.

Distribution and Abundance

Known only from the type locality, Ft.
Pierce Inlet, St. Lucie County, Florida,
27°28.3’М, 80°17.9’W, on intertidal and shal-
low subtidal sand flats. Uncommon; only 16
specimens known.

Etymology

An adjective, luteocrinitus, -a, -um, from the
Latin /uteus (yellow) and the Latin crinitus
(hairy), referring to the numerous, long, yel-
lowish tentacles, imparting an overall “hairy”
appearance.

Remarks

The single posterior pair of club-shaped
tentacles is distinct and consistent in both liv-
ing and preserved specimens. Normal tenta-
cles (Fig. 25, nt; similar morphology in all five
Floridian Divariscintilla spp.) show usually
four haemocoelic compartments in cross sec-
tion, each of these supplied with a longitudinal
muscle bundle and nerve fiber, and separated
by connective tissue septa. The muscle and
nerve fibers form a more-or-less concentrated
central core, appearing as a central thread
under low magnification. Judd (1971: fig. 6)
figured a similar internal morphology for ten-
tacles of Divariscintilla maoria, but the individ-
ual compartmental muscle-plus-nerve bun-
dles are not as concentrated at the core of the
tentacle; this difference may not be real, but
rather implied by Judd's interpretation of the
histological sections. An additional difference

is found in the outer surfaces of the tentacles
of D. luteocrinita, which are highly papillose
and convoluted rather than smooth as in D.
maoria as shown by Judd (1971: fig. 6). Inter-
nally, the club-shaped tentacles of D. luteo-
crinita (Figs. 5, 25, cst) are identical in struc-
ture to normal tentacles except that the
muscle fibers (Fig. 25, arrow) are dispersed
over the surfaces of the septa instead of be-
ing concentrated at the core of the tentacle.
This difference may allow this particular pair
of tentacles to undergo stronger contraction,
resulting in the club-like shape. Under full ex-
tension, the club shape and whitish coloration
of these ientacles disappears, indicating that
these features are products of the normally
contracted state.

The function of the club-shaped tentacles is
unknown. They are probably not homologous
to the posterior “defensive appendages” of D.
maoria (see Judd, 1971: fig. 7) or Galeomma
takii (Kuroda, 1945) (see B. Morton, 1973a:
fig. 5), which can be autotomized and pos-
sess only a single, central haemocoelic tube.
No tentacles in D. luteocrinita have this mor-
phology nor were they ever seen to autoto-
mize. However, these tentacles were often
observed to hyperelongate when the animals
were disturbed, for example during specimen
transfer between laboratory dishes. While
most other tentacles contracted, the club-
shaped tentacles elongated immediately to 2-
3 times the shell length just as the transfer
spoon made contact. This bears resemblance
to the dymantic (“threatening”) display of ten-
tacles in Galeomma polita Deshayes, 1856
(see B. Morton, 1975), and Ephippodonta oe-
dipus Morton, 1976 (see B. Morton, 1976),
when disturbed, and suggests defensive
function. Unlike the dymantic tentacles of the
latter two species, however, the club-shaped
tentacles of D. luteocrinita do not fully retract
into the mantle at rest.

The large “?hypobranchial glands” of this
species are located at sites similar to those of
the smaller glands of other Divariscintilla spe-
cies, i.e., adjacent to the rectum and the bran-
chial nerves as they join the visceral ganglia.
In histological sections, they bear striking re-
semblance to “seminal receptacles” de-
scribed and figured for Aligena elevata
(Stimpson, 1851) by Fox (1979: 101, 103, fig.
32), although in the single sectioned speci-
men of D. luteocrinita, they contained no
sperm. The montacutid Aligena elevata is a
protandrous hermaphrodite, and its seminal
receptacles are present only during the fe-

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 11

Figs. 21-22. Flower-like organ of Divariscintilla luteocrinita (SEM). Anterior tip of foot has been severed to
enhance visibility. 21. Scale bar = 100 рт. 22. Scale bar = 50 um. (ct, ctenidium; fl, flower-like organ; ft,

foot; Ip, labial palp; m, mantle).

male stage. The occurrence of such struc-
tures in Galeommatoidea was summarized by
Fox (1979: table 8, as Leptonacea, 15 spe-
cies), who noted that sperm storage organs
were unknown outside of Montacutidae. How-
ever, protandry is known in members of other
families (e.g. Lasaeidae: Arthritica crassi-
formis Powell, 1933 (see B. Morton, 1973b);
see also Fox, 1979: table 11). The reproduc-
tive mode of D. luteocrinita is presently un-
known, but seminal receptacles and/or
protandry would be no surprise within the
context of this reproductively complex super-
family.

Divariscintilla cordiformis n. sp.
(Figs. 3, 6, 26-32)

Material Examined

Holotype: 5.6 mm (shell length), FMNH
223405. Paratypes (4): 6.4 mm (sectioned on
29 microslides), FMNH 223406; 5.4 mm,
HBOM 064:1861; 4.9 mm (partially dissected;
prodissoconch coated for SEM), HBOM 064:
01862; 6.4 mm (shell only, coated for SEM),
USNM 859445. Total material: 6 specimens:
FLORIDA: Ft. Pierce Inlet: 24 June 1987, 2;
-Peanut Island: 10 August 1987, 4.

Type locality
Peanut Island, near Lake Worth Inlet, Palm

Beach County, eastern Florida, 26°46.6’N,
80%02.7'W, occupying Lysiosquilla scabri-

cauda burrows on intertidal sand flats with
patches of the seagrass Halodule wrightii As-
cherson. Paratypes from type locality (FMNH,
USNM, HBOM 064:01861) or Ft. Pierce Inlet
(HBOM 064:01862).

Diagnosis

Animal translucent white. Mantle thin, with
extensive, retractable papillose folds covering
shell completely, meeting at midline. Tenta-
cles originating at shell edge, six pairs ante-
riorly, eight pairs posteriorly. Posterior foot-
extension relatively short. Shell oval, with
umbo slightly anterior; length approximately
65% of extended mantle length. Small, ven-
tral, anteriorly directed indentation in each
valve. Coarse growth lines and slightly
beaded radial ribs restricted to edges of oth-
erwise-smooth shell. Single “flower-like
organ” on anterior surface of visceral mass.

Description

External Features and Mantle: Living ex-
tended animal (Figs. 3, 6) generally 7-10 mm
in length, translucent white except for dark
digestive gland within visceral mass. Shell en-
tirely covered in life by anterior and posterior
mantle folds (Fig. 6, m) meeting at lateral
dorso-ventral mid-line. Mantle folds thin, in-
completely retractable, entirely finely papil-
lose, with scattered larger papillae, which are
longest at posteroventral section. Mantle
edge extending widely beyond shell edges;

12 MIKKELSEN 8 BIELER

23

mus

Figs. 23-25. Histological sections. 23. ?Hypobranchial gland adjacent to rectum in D. octotentaculata. 24.
?Hypobranchial gland adjacent to rectum in D. luteocrinita. 25. Cross-sections of normal and club-shaped
tentacles in D. luteocrinita. Arrow = muscularized septum. Scale bars = 100 um. (cst, club-shaped tentacle;
ct, ctenidium; hyp, ?hypobranchial gland; mus, muscle fibers; n, nerve fiber; nt, normal tentacle; pam,
posterior adductor muscle; r, rectum; sbc, suprabranchial chamber; sep, connective tissue septum; sh, shell;

ten, tentacle).

entire surface finely papillose, with additional
enlarged papillae near shell edge. Ventral
mantle edge entire, not cleft in vicinity of ven-
tral shell indentation. Pallial openings, mus-
culature, and posterior pouch as in Divariscin-
Ша octotentaculata. Paired, retractable,
pallial tentacles numerous, originating near
shell edge. Six anterior pairs, those adjacent
to cowl area (third and fourth, see numbers,
Fig. 6) longest. Eight posterior pairs, those
adjacent to excurrent siphon (tenth) longest,
but shorter than longest anterior tentacles. A
single median tentacle (Fig. 6, mpt) between
first and second posterior pairs (Fig. 6, nos.
7, 8). Structure of individual tentacles as in D.

octotentaculata. Excurrent opening (Fig. 6,
exc) between ninth and tenth tentacle pairs.

Preserved animals not fully retracted into
gaping shell; mantle folds contracted to nar-
row rim along shell edge, exposing most of
shell surface; tentacles, cowl, and posterior
pouch contracted but still usually extending
beyond shell edges; foot usually completely
withdrawn into pallial cavity.

Shell (Figs. 26-30, 32): Shell generally 5-7
mm in length, nearly evenly oval, longer pos-
teriorly, equivalve, compressed, glossy, trans-
parent to translucent white. Small indentation
(Figs. 6, ind; 27, arrow; 28) at mid-ventral

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 13
30 TA

Figs. 26-32. Shell of Divariscintilla cordiformis (SEM). 26. Right valve, external view, 6.4 mm length, para-
type, USNM 859445. 27. Right valve, internal view, 6.4 mm length, paratype, USNM 859445. Arrow =
ventral indentation. 28. Close-up of ventral indentation, interior view. Scale bar = 100 um. 29. Hinge,
anterior to left. Scale bar = 200 рт. 30.Prodissoconch, 360 um length, paratype, HBOM 064:01862. Single
arrow = prodissoconch |-!! boundary. Double arrow = prodissoconch Il-dissoconch boundary. 31. Newly
released larval shell, 146 um length. 32.Microstructure, with internal surface at top. Scale bar = 5 um. (per,
periostracum).

14 MIKKELSEN 8 BIELER

margin, slanting anteriorly toward umbo, evi-
dent only on distal 1 mm or so of shell growth
(as evidenced by growth lines). Exterior
sculpture smooth except for fine concentric
growth lines, heavier at anteroventral margin,
anterior to indentation. Beaded radial ribs re-
stricted to distal 1 mm or so of shell edge,
most prevalent interiorly at posteroventral
margin (Fig. 27) and exteriorly at antero- and
posterodorsal margins (Fig. 26), forming dis-
tinct, fine crenulation at shell edge, absent
only at ventral margin immediately anterior to
indentation. Size large relative to mantle,
comprising approximately 65% of extended
mantle length. Valves held open at 20-30° an-
gle while crawling, incapable of complete clo-
sure. Adductor muscle scars subequal, faint.
Periostracum as in Divariscintilla octotentac-
ulata; also evident covering ventral indenta-
tion (Fig. 28, per).

Hinge line (Fig. 29) short, as in Divariscin-
tilla octotentaculata, with two small rounded
cardinal teeth.

Prodissoconch (Fig. 30) brownish-yellow,
approximately 360 um in length. Prodisso-
conch | corresponding in size to shell of newly
released larva, approximately 45% of length
of prodissoconch Il; sculpture not discernible
(surface abraded in adult shell). Prodisso-
conch Il sculptured with coarse concentric
growth lines. Demarcation between prodisso-
conch | and II stages (Fig. 30, single arrow),
and between proaissoconch ll and disso-
conch (Fig. 30, double arrow) as in Divaris-
cintilla octotentaculata.

Organs of the Pallial Cavity: Foot and shell
muscles (adductors, pedal retractors, pedal
protractors, including relative positions) as in
Divariscintilla octotentaculata. Muscles leav-
ing very faint attachment scars on shell (Fig.
27).

Visceral mass, labial palps and ctenidia as
in Divariscintilla octotentaculata. Palps with
approximately seven lamellae each side.
Outer demibranch approximately 35% smaller
than inner. Cilial currents on palps and ctenidia
unknown.

Single “flower-like organ” (see Mikkelsen &
Bieler, 1989) on anterior surface of visceral
mass just ventral to labial palps.

Digestive System: Similar to that of Divaris-
cintilla octotentaculata (relative positions of
gastric shield, style sac, digestive diverticula,
midgut, etc.), based on histological sections.
Limited number of specimens did not permit

confirmation of structure through gross dis-
section.

Suprabranchial Chamber. As in Divariscintilla
octotentaculata.

Nervous System: Arrangement of ganglia,
statocysts, and major nerves as described for
Divariscintilla yoyo (see Mikkelsen & Bieler,
1989: fig. 31). Additional tentacular nerves
arising (independently) from pallial nerve.

Reproductive System: Simultaneous her-
maphrodite. Overall gross morphology as in
Divariscintilla octotentaculata.

One specimen brooding larvae collected in
August 1987. Larvae released one day after
collection. Newly-released larval shells 136-
148 um in length (X = 143 pm, п = 40; Fig.
31). Adult preserved and subsequently sec-
tioned; apparently intact larvae found in su-
prabranchial chamber and throughout diges-
tive system, including intestine and rectum
(therefore not being digested).

Circulatory and Excretory Systems: As in Di-
variscintilla octotentaculata.

Distribution and Abundance

Known from the type locality at Peanut Is-
land, Palm Beach County, and from sand flats
at Ft. Pierce Inlet, St. Lucie County, Florida,
27°28.3'N, 80%17.9'W. Rare; only 6 speci-
mens known.

Etymology

An adjective, cordiformis, -e, from the Latin
cordis (heart) and the Latin forma (shape),
referring to the ventrally indented shell out-
line.

Remarks

As in Divariscintilla maoria (see Judd,
1971), the ventral indentation in the shell of
this species does not seem to be functionally
important. In both species, it is present in both
valves, anteriorly inclined, not developed in
juveniles, and not reflected by soft anatomy.

ECOLOGY AND BEHAVIOR

As in the two previously described com-
mensal galeommatids (Mikkelsen & Bieler,
1989), no specimens of the newly described
species were ever found physically attached

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 15

to a mantis shrimp, either in the field or in
museum specimens (HBOM). They are as-
sumed to be free-living within the vertical por-
tions of the U-shaped burrow, although spec-
imens were never visible at the opening prior
to pumping. Again as with the previous spe-
cies (Mikkelsen 8 Bieler, 1989), these clams
were never found free-living outside of the
burrows or associated with any other burrow-
ing invertebrate in the area (e.g. other mantis
shrimps, callianassid shrimps, polychaetes,
sipunculans), nor were any empty shells lo-
cated in dry collections (AMNH, DMNH,
FMNH, HBOM, USNM), probably because of
their fragile nature.

Divariscintilla octotentaculata was the most
frequently encountered commensal mollusk in
the Lysiosquilla burrows; of the 35 burrows
containing mollusks, 31 contained D. octoten-
taculata, 20 D. yoyo, 19 D. troglodytes, 8 D.
luteocrinita, 6 Cyclostremiscus beauii, 7 Cir-
culus texanus, and only 2 D. cordiformis. Di-
variscintilla octotentaculata was usually col-
lected with other commensals, occurring alone
in only 6 of the 31 samples. Divariscintilla lu-
teocrinita was always collected with other
commensals. Divariscintilla cordiformis was
collected once with D. octotentaculata, and
once alone. Densities of D. octotentaculata
varied greatly, ranging from 1-74 per sample
(x = 8.3, п = 31). Divariscintilla luteocrinita
was usually present in numbers of only one or
two specimens per burrow; one sample con-
tained four specimens. Divariscintilla cordi-
formis was encountered only twice, once as
two specimens, once as four. Only a small
number of burrows sampled contained com-
mensals. And, as previously reported (Mik-
kelsen & Bieler, 1989), it must be emphasized
that in no case could an entire burrow be sam-
pled using the yabby pump, which only effec-
tively samples its own length (0.5-1.0 m) of the
vertical parts of the U-shaped burrow. Esti-
mates of occurrence and/or density of any
clams living in the deeper horizontal section of
the burrow was thus not possible.

Animals of all three species spent most of
their time in the laboratory attached to the
glass surface of laboratory bowls by up to four
byssus threads, and it is assumed that this is
also their habit on the smooth walls of the
Lysiosquilla burrow. When dislodged, they
actively crawl about using an even, gliding
motion produced by ciliary action on the cen-
tral surface of the foot. This was equally ef-
fective on the underside of the water surface
as on glass. They made no attempts to bur-

row when offered a substrate of loose sand in
the laboratory.

All three Divariscintilla species previously
described (D. maoria, Judd, 1971: fig. 4; D.
yoyo and D. troglodytes, Mikkelsen & Bieler,
1989: fig. 32) are known to “hang” from a
vertical substrate by the posterior foot-exten-
sion, which in these species is extremely long
and elastic; byssus threads secreted by the
anteriorly located byssus gland are laid down
within the ventral groove of the foot and
emerge from the terminus to attach to the
substrate. The threads are secured within the
groove by secretions of the posteriorly lo-
cated byssus adhesive gland. Byssus and
byssus adhesive glands of similar morpholo-
gies are present in each of the three new spe-
cies described here and are assumed to func-
tion in the same way. This has been
confirmed for D. octotentaculata and D. luteo-
crinita, in which secretion of byssus threads,
accompanied by distinct pulsing of the byssus
gland area, was observed as described for D.
yoyo and D. troglodytes (Mikkelsen & Bieler,
1989). Following this activity, the clam hangs
from the posterior foot-extension, with the
byssus threads emerging from the posterior
terminus of the foot in the vicinity of the bys-
sus adhesive gland. The posterior foot-exten-
sion of these species, as well as of D. cordi-
formis, is not as elongated and extensible as
in those previously described, therefore the
distinctive “hanging” posture, wherein the
clam “dangles” from an elongated foot, is not
as pronounced.?

A peculiar interaction between pairs of Di-
variscintilla species was observed in the lab-
oratory on three occasions. In two instances
involving D. octotentaculata, one animal of a
pair was noted reaching its foot into the man-
tle cavity of the second specimen, either from
in front of or behind its partner. On one of
these occasions, the two individuals per-
formed this activity simultaneously, “facing”
one another, with each one reaching around
the visceral mass of the other to contact the
posterior surface with the tip of its foot (Fig.
33). Two specimens of D. yoyo have been

®Careful notes were not recorded on the behavior of D.
cordiformis in the early phase of the study, and the inavail-
ability of additional living specimens prevented confirma-
tion of the presence or absence of “hanging” behavior.
Morphology suggests that this behavior does occur, how-
ever, the shorter posterior extension and the absence of
any mention in our preliminary written observations indi-
cates that the “hanging” posture was inconspicuous, as in
D. octotentaculata and D. luteocrinita.

16 MIKKELSEN 8 BIELER

Fig. 33. Mating (?) behavior between two specimens of Divariscintilla octotentaculata, as observed in

laboratory.

noted performing this same type of activity. In
all three instances, the interaction was initi-
ated by the smaller of the two specimens of
the pair, and the activity was sustained for 3-7
minutes. Although there is no direct evidence,
it seems likely that this observed interaction is
part of some sort of reproductive activity, per-
haps stimulation of sperm transfer (external
gonadal openings are located on the poster-
odorsal surface of the visceral mass). Each of
the two specimens involved in the simulta-
neous behavior described above had larvae
in their suprabranchial chambers within three
weeks of the noted activity. (If this behavior is
in fact copulatory in function, then this con-
firms simultaneous hermaphroditism in these
species.)

Observed animals of these Divariscintilla
species do not respond to changes in light
intensity (e.g. photographic strobes) and did
not seek darkness when offered “artificial bur-
rows” (in the form of black plastic tubes) in the
laboratory.

DISCUSSION

Generic Placement

The genus Divariscintilla was redescribed
in a previous part of this study on Lysiosquilla-
associated mollusks (Mikkelsen & Bieler,
1989). Major emphasis was placed on two
features, the presence of flower-like organs
and a bipartite foot with both byssus and bys-
sus-adhesive glands, shared between these
species and the type species, D. maoria Pow-
ell, 1932 (the latter described by Judd, 1971).

We regarded the presence of an apparently
functionless (Judd, 1971) shell notch in D.
maoria, as well as the difference in degree of
shell coverage by the mantle, as species-
level rather than generic characters. At the
time, we also considered placement of D.
yoyo and D. troglodytes in the monotypic
genus Phlyctaenachlamys Popham, 1939,
based on P. lysiosquillina Popham, 1939.
However, there are a number of anatomical
characters that distinguish the Divariscintilla
species (flower organs, hindgut typhlosole,
interlamellar ctenidial junctions, relative size
of adductor muscles, hinge teeth; compare
data of Popham, 1939, and Mikkelsen &
Bieler, 1989).

Divariscintilla luteocrinita and D. cordi-
formis fit well within Divariscintilla, as rede-
scribed. In fact, these new species actually
bridge several morphological gaps between
D. maoria and D. yoyo/D. troglodytes, that is
shell notch, posterior shell prolongation, less
shell internalization, and more numerous pos-
terior tentacles. Divariscintilla octotentaculata
does not possess the flower-like organs; how-
ever, in all other taxonomic characters (hinge,
mantle, ctenidia, etc.), it does conform to the
redefined genus. The hinge teeth of this spe-
cies, comprised (as in other Divariscintilla
species) of only one small cardinal tooth in
each valve, was an important element in de-
ciding generic placement. Hinge structure is
presently considered taxonomically important
at the generic level in Galeommatoidea (P. H.
Scott, in litt., October 1990). Most members of
the superfamily possess more than one car-
dinal tooth in at least one valve; some show
distinct lateral teeth as well. The genus Scin-

NEW SPECIES OF COMMENSAL GALEOMMATIDAE

17

TABLE 1. Distinguishing characteristics of all described species of Divariscintilla: an expansion of table 1

from Mikkelsen & Bieler (1989).

D. maoria
(from Judd, D. D. D. D.
1971) D. yoyo troglodytes octotentaculata luteocrinita cordiformis
Shell:

General shape oval elongate- oval roundly roundly oval

pointed triangular triangular

Ventral indentation present absent absent absent absent present

Prolongation posterior anterior anterior posterior posterior posterior

Internal sculpture unribbed unribbed radially radially unribbed, with radially ribbed

ribbed ribbed internal (marginally)
(marginally) (marginally) thickening
Length relative to
extended
mantle length 68% 40% 50% 70% 70% 65%
Prodissoconch
length (шт) “small” 360 380 360 390 360
Mean newly-released
larval shell
length (шт) unknown 132 126 119 unknown 143
Mantle:

Color, thickness (not whitish, yellowish, whitish, thin yellowish, whitish, thin

given) thick thin thin

Extent covering margins entire, entire, anterior and entire, entire,

shell only umbonal anterior slit posterior thirds midline midline
foramen overlap overlap

Papillae very sparse, numerous, numerous, numerous, numerous,

small very small small, small, small, small,
evenly- longer at evenly- evenly-
distributed ventral distributed distributed
edge
Anterior tentacles 2 pairs 1 pair 2 pairs 2 pairs 3 pairs 6 pairs
Posterior tentacles 1 single 1 single 1 single, 1 single, 2 singles, 1 single,
1 pair 2-3 pairs 5 pairs, 8 pairs

+ 3-6 pairs
accessory,

+ 1 pair

club-shaped

Defensive
appendages 6-8 absent absent absent absent absent
present
Pedal protractor

muscle

insertion relative to

anterior adductor

muscle (not dorsal dorsal ventral ventral ventral

given)
Flower-like organs:
number 1 3-7 1 0 1 1
(usually 5)
Labial palps:

Lamellae per palp approx. 9 10-14 14-20 6-8 approx. 8 approx. 7
Ctenidia: smooth pleated pleated smooth smooth smooth?
Geographical range: New eastern eastern eastern eastern eastern

Zealand Florida Florida Florida Florida Florida

tillona Finlay, 1927, is the only other genus
known to us in which members have a single
Cardinal tooth in each valve. However, mem-

bers of the latter genus are all attached ecto-
commensals on echinoderms, and in addi-
tion, are distinguished by a highly specialized,

18 MIKKELSEN 8 BIELER

laterally compressed and furrowed foot. Both
of these considerations exclude the new
species described here. In view of the taxo-
nomic confusion present in this group, we
prefer to use an existing genus until a generic
revision of this superfamily based on both
shell and anatomical characters can be ac-
complished.

Coney (1990), in a review of “ventrally
notched galeommatid genera,” assigned Di-
variscintilla yoyo and D. troglodytes to the ge-
nus Phlyctaenachlamys, and reinstated
Divariscintilla as a monotypic genus. His rea-
soning revolved around (1) the shell notch, (2)
hinge teeth and ligament morphology, (3)
shell ultrastructure, (4) mantle tentacles, (5)
degree of shell coverage by the mantle, and
(6) ctenidial morphology. We disagree with
his taxonomic decisions and address these
points as follows:

(1) Shell notch: The newly described spe-
cies Divariscintilla cordiformis has an appar-
ently functionless ventral shell notch very sim-
ilar to that of the type species D. maoria, and
agrees well anatomically with the “un-
notched” species assigned to Divariscintilla.
Importantly, the notch does not influence
mantle or other soft-part morphology.

(2) Hinge and ligament: Coney (1990: 131,
135) offered conflicting statements regarding
hinge teeth. His table of generic characters
and description of Divariscintilla maoria
stated that there was one cardinal tooth in the
right valve, whereas his generic description
indicated two. Although his scanning photo-
micrograph of the right hinge (Coney, 1990:
fig. 11) seemed on first examination to reveal
two cardinal teeth, comparison with our pho-
tomicrographs as well as Coney’s subse-
quent text indicated that the posteriormost
“knob” was actually the resilium, not a sec-
ond cardinal tooth. Coney (1990) upheld the
Original interpretation by Powell (1932) that
the left valve was edentulous, although his
scanning photomicrograph (Coney, 1990: fig.
12) featured a small structure that may corre-
spond to Powell’s (1932: 67) “minute and
shapeless vestige” of a left cardinal. Addition-
ally, Coney (1990) noted a ridge in the left
valve that he called a posterior lateral tooth.
All Divariscintilla species (by our definition)
possess strengthened or thickened hinge
lines to varying degrees, but without develop-
ment into distinct and/or interlocking lateral
teeth/lamellae. As we lack relevant ontoge-
netic data on shell development, we do not
wish to infer homology in this instance and

prefer not to call these structures “lateral
teeth.”

According to Popham (1939: 66, text-fig. 6),
the hinge of Phlyctaenachlamys lysiosquillina
has two discrete cardinal teeth in the right
valve, one cardinal in the left, and an inter-
locking set of posterior laterals. This is dis-
tinctly different from the situation in Divaris-
cintilla yoyo, D. troglodytes, and D. maoria.

Ligament morphology in all Divariscintilla
species (by our definition) consists of an ex-
ternal amphidetic ligament (inappropriately
called “periostracal webbing” at one point in
our earlier paper; Mikkelsen & Bieler, 1989:
178) supported by a nymph, and an internal
opisthodetic resilium. The resilium is also ap-
parently present in Phlyctaenachlamys lysios-
quillina, but the external ligament was not
mentioned by Popham (1939).

In total, these findings do not agree with
Coney's (1990: 142) statement that “the
hinge teeth and ligament are remarkably sim-
ilar between the three species [/ysiosquillina,
yoyo, troglodytes], but are quite different than
those of Divariscintilla maoria. .. .”

(3) Shell ultrastructure: This feature was
also utilized by Coney (1990) in his discus-
sion of distinguishing characteristics of Di-
variscintilla and Phlyctaenachlamys species,
although he also (1990: 142) admitted that
the “shell ultrastructure of Phlyctaenach-
lamys lysiosquillina is unknown”. Studies of
the three new species described here (Figs.
13, 19-20, 32), as well as a reevaluation of
this character in D. yoyo and D. troglodytes
(see Mikkelsen & Bieler, 1989: figs. 11, 15),
indicate that both the inner and outer non-
cross-lamellar layers may be prismatic rather
than “homogenous” as previously labelled by
us. Variability in thickness and/or presence of
the various layers among individuals and
among different locations on a single valve
(e.g. extra internal layer in D. luteocrinita, see
above) has also been noted. A more exacting
study using more appropriate methodology
(e.g. sections rather than fractions, see Taylor
et al., 1973) is necessary before differences
at this level should be employed in taxonomic
decisions.

(4) Mantle tentacles: Coney (1990: 142)
maintained that “number and placement of
mantle tentacles and defensive appendages
is strongly similar between P/hlyctaenach-
lamys] lysiosquillina and those of P. yoyo and
P. troglodytes,” noting that none of these spe-
cies possess the numerous posterior defen-
sive tentacles seen in Divariscintilla maoria.

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 19

The three new species described here in Di-
variscintilla all possess a number of posterior
tentacles, albeit none “defensive.” The two
primary anterior tentacles, also mentioned
specifically by Coney (1990: 142) to combine
P. lysiosquillina, D. yoyo, and D. troglodytes,
are in fact also present in Divariscintilla mao-
ría.

(5) Shell coverage: Although there is a dis-
tinct difference in the degree of shell cover-
age by the mantle between the type species
and Divariscintilla yoyo and D. troglodytes,
the three newly described species show inter-
mediate conditions. All Divariscintilla species
(by our definition) show at least some degree
of shell exposure, thus differing from the con-
dition in Phlyctaenachlamys Iysiosquillina
(“the shell is completely embedded”; Pop-
ham, 1939: 65). A similar range of variability
was described for the genus Ephippodonta
Tate, 1889, by Arakawa (1960: 57), and can
also be found in Entovalva Vóltzkow, 1890,
sensu lato, wherein the genus Devonia was
distinguished by Winckworth (1930: 14) for a
species with incomplete shell coverage.

(6) Ctenidial morphology: Coney (1990:
142) noted that the ctenidia in Divariscintilla
maoria are smooth, whereas those of Phlyc-
taenachlamys lysiosquillina, D. yoyo and D.
troglodytes have been described as pleated.
However, unlike pleated (= plicate) gills in
other bivalve groups (e.g. Pecten; see Rice,
1897), pleating in these species is not based
on structural differences in the filaments
(pers. obs.; Popham, 1939: 71). Some degree
of “pleating” caused by contraction in pre-
served specimens has been noted during the
present study. An additional ctenidial charac-
ter separates P. lysiosquillina and the five Flo-
ridian species of Divariscintilla, in that the lat-
ter have interlamellar junctions (unknown for
New Zealand D. maoria).

Comparative characteristics for all six
known species of Divariscintilla are presented
in Table 1. The three new species described
here more closely resemble the type species,
D. maoria (see Judd, 1971: fig. 1), in general
morphology than do the other species previ-
ously described during this study (D. yoyo
and D. troglodytes, see Mikkelsen & Bieler,
1989: figs. 1, 2). Like D. maoria, the three new
species have posteriorly prolonged, relatively
large shells and smooth ctenidia (differences
in percent reduction of the outer demibranch
as cited in the text may not be reliable, as they
were taken from preserved specimens and
were affected by contraction). One species,

D. octotentaculata, has a shell that is similarly
incompletely covered by the mantle. Another
species, D. cordiformis, shows a similar ven-
tral indentation that is likewise apparently
functionless. These similarities effectively re-
move most of the dissimilarity between the
New Zealand type species and the included
eastern Florida species that existed at the
completion of the previous paper (Mikkelsen
8 Bieler, 1989). Divariscintilla maoria is the
only species in the genus for which detach-
able defensive papillae have been described.

Distribution

The peculiar pattern of geographic distri-
bution of Divariscintilla species, with now
five members in the western Atlantic and one
in New Zealand, is most likely a result of in-
sufficient sampling of burrow fauna. No eco-
logical niche separation between the five
sympatric species was recognized, leaving in-
teresting questions for future research.

Comparison with Other Genera

Divariscintilla was formerly treated as a
subgenus of Vasconiella Dall, 1899, by Cha-
van (1969), apparently on the basis of
notched shells. However, members of V. jef-
freysiana (P. Fischer, 1873), type and sole
species of the genus, possess a deep inden-
tation only in the comparatively smaller right
valve, and importantly, there are modifica-
tions in the right mantle and ctenidial tissues
corresponding to the notch (Cornet, 1982).
This species is also the only other galeomma-
toidean with a published account of a “flower-
like organ” (Table 2). Unfortunately, although
recognizably illustrated by Cornet (1982: fig.
5), no details on structure or possible function
were provided for the briefly mentioned
“rounded tubercle just under the labial palps”
(Cornet, 1982: 39). Vasconiella jeffreysiana is
probably also commensally associated with a
mantis shrimp, Lysiosquilla eusebia (Risso,
1816) (Table 2; Cornet, 1982). Differences of
Vasconiella from Divariscintilla (i.e. two cardi-
nal teeth in the left valve, lack of a posterior
foot-extension, ?lack of ctenidial interlamellar
junctions) prevent synonymy of the two gen-
era as currently defined, but clearly, their re-
lationship should be investigated further.

Mention has been made several times
above to the relationships of galeommatoide-
ans with certain phyla of host invertebrates
(e.g. echinoderms versus stomatopods). Evi-

20 MIKKELSEN 8 BIELER

TABLE 2. Galeommatoidean species, hosts, and occurrences of flower-like organs and hanging foot

structure (x = presence; — = absence).
Flower- Hanging
like foot
Host organs structure References(s)
Divariscintilla maoria Powell, Heterosquilla tricarinata x Xx Judd, 1971
1932 (Claus)
D. yoyo Mikkelsen & Bieler, Lysiosquilla scabricauda x x Mikkelsen & Bieler,
1989 (Lamarck) 1989
D. troglodytes Mikkelsen & Lysiosquilla scabricauda X X Mikkelsen & Bieler,
Bieler, 1989 (Lamarck) 1989
D. octotentaculata n. sp. Lysiosquilla scabricauda = x This study
(Lamarck)
D. luteocrinita n. sp. Lysiosquilla scabricauda x x This study
(Lamarck)
D. cordiformis n. sp. Lysiosquilla scabricauda x x This study
(Lamarck)
Parabornia squillina Lysiosquilla scabricauda x x Boss, 1965a; this study
Boss, 1965 (Lamarck)
Vasconiella jeffreysiana ?L. eusebia (Risso) x = Cornet, 1982
(P. Fischer, 1873)
Phlyctaenachlamys L. maculata (Fabricius) = x Popham, 1939
Iysiosquillina Popham,
1939
Ceratobornia longipes Callianassa major Say or = x Jeffreys, 1863; Dall,

(Stimpson, 1855)

C. cema Narchi, 1966

Upogebia affinis (Say)

Callianassa major Say

1899; Norman, 1891;
Jenner & McCrary, 1968
Narchi, 1966

dence suggests that host specificity applies
mainly to those cases where modifications for
locomotion and attachment have occurred.
Most “generalist” species (sensu B. Morton &
Scott, 1989; e.g. Mysella bidentata (Montagu,
1803), associated with a wide variety of inver-
tebrates [Boss, 1965b]) possess a simple foot
structure consisting of a strong crawling por-
tion (similar to the anterior foot of Divariscin-
tilla spp.) ending in a bluntly rounded “heel”
from which byssus threads emanate (B. Mor-
ton 4 Scott, 1989: fig. 1 [Lasaeal], figs. 9-10
[Pseudopythina]). This is probably the plesio-
morphic condition in the superfamily, com-
pared to the more derived states seen in
some of the host specialists. For example, the
flattened foot of Scintillona spp. (J. E. Morton,
1957) may be a modification for laterally ap-
plied locomotion among the vertical spines
and papillae of echinoderms to which they at-
tach (Dall et al., 1938: 145; Yamamoto &
Habe, 1974; O Foighil & Gibson, 1984: 75).
Also, the sucker-like anterior foot of Entovalva
(= Devonia) perrieri (Malard, 1903) (see An-
thony, 1916) allows attachment to the
smooth, outer body walls and/or inner cloacal

walls (Bruun, 1938) of burrowing holothuri-
ans.

An elongate posterior foot-extension for at-
tachment to a smooth vertical substrate is
found in three specialist genera: Divariscin-
tilla, Phlyctaenachlamys Popham, 1939, and
Ceratobornia Dall, 1899 (Table 2). All species
involved are associated with crustaceans that
produce smooth-walled burrows in sand
(Table 2). Most are believed to live attached
to the crustacean burrow walls (Ceratobornia.
cema may also “temporarily” attach to its host
(Narchi, 1966); the biology of C. longipes is
unknown). Although insufficiently studied in
Phlyctaenachlamys and Ceratobornia, all
species apparently possess the same
“hanging apparatus” comprised of an anterior
byssus gland, ventral groove, and posterior
byssus adhesive gland (see Popham, 1939:
fig. 2, P. lysiosquillina, hanging behavior not
specifically described; Narchi, 1966: figs. 1, 5,
C. cema; Dall, 1899: pl. 88, figs. 10-11, 13, C.
longipes). They also all possess similar gen-
eral morphologies of the pallial, ctenidial, di-
gestive, circulatory, excretory, and nervous
systems (C. longipes incompletely known).

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 21

TABLE 3. Distinguishing characteristics for the three galeommatoidean genera possessing the “hanging
foot.” See text for included species and sources of data. [L, left; R, right]

Divariscintilla

Shell internalization incomplete
Hinge:
Cardinal teeth ike le
Lateral teeth absent
Retraction into shell yes (1 sp.)
no (5 spp.)
Adductor muscles subequal

Flower-like organs present (5 spp.)

absent (1 sp.)

Interlamellar ctenidial present
junctions:

Hindgut typhlosole present

Hypobranchial gland present?

Supportive chondroid absent

edge in foot

Phlyctaenachlamys Ceratobornia
complete incomplete
AR PR e le
1 posterior, 1 posterior,
reduced reduced
no yes
posterior reduced subequal
absent absent
absent absent
absent absent
absent absent
absent present

The anatomical differences among the three
genera that presently prevent their synonymy
are listed in Table 3. Until the importance of
each of these characters can be reassessed,
it is uncertain whether possession of a
“hanging foot” reflects convergence or phylo-
genetic relationship, and the three genera are
best treated separately.

Another possible difference between Di-
variscintilla on one hand and Phlyctaenach-
lamys and Ceratobornia on the other lies in
the mode of reproduction. Members of Di-
variscintilla for which such data are available
(D. yoyo, D. troglodytes, D. octotentaculata,
D. cordiformis) are known to be simultaneous
hermophrodites. From literature data, it
seems that Phlyctaenachlamys lysiosquillina
and Ceratobornia cema are forms with sepa-
rate sexes (Popham, 1939, p. 80: “specimen
sectioned was a male”; and Narchi, 1966, p.
521: “sectioned specimen was a female”).
However, while simultaneous hermaphrodit-
ism can be documented from individuals with-
out observations over extended periods of
time, “males” or “females” could belong to
forms with consecutive hermaphroditism.
Protandrous and protogynous hermaphrodit-
ism have both been reported for the super-
family (summarized by Fox, 1979: tab. 11, as
Leptonacea).

Rhamphidonta Bernard, 1975, represented
by the single species A. retifera (Dall, 1899),
also possesses a bipartite foot but one which
is different from that discussed above, both
morphologically and functionally. According
to Bernard (1975), the foot of R. retifera is

anteriorly elongated, with the main enlarged
crawling portion located posteriorly. Members
of this species are not known to hang; they
burrow into sand to avoid illumination, and are
apparently free-living. The hinge of A. retifera
(see Bernard, 1975: fig. 1) is distinctly “mon-
tacutid.” Close relationship with the three
genera discussed above is unlikely.

The mantis shrimp Lysiosquilla scabri-
cauda also serves as host to Parabornia
squillina Boss, 1965, a galeommatoidean that
attaches to the inner surface of the abdominal
sclera of the shrimp (Table 2). Parabornia
squillina has been collected in the region of
this study (Sebastian Inlet, Peanut Island),
but has not been collected in burrows contain-
ing Divariscintilla species. Interestingly, the
animal of P. squillina also possesses a single
“flower-like organ” (pers. obs.), although this
was not mentioned in the original species de-
scription based on preserved material. (Boss,
1965a: fig. 3 shows an unexplained five-part
zig-zag outline for the visceral mass below
the gills and labial palps.) Its foot (Boss,
1965a: fig. 3) appears morphologically similar
to the “hanging type” described above; a
deep ventral groove ends with an opaque
white area at the end of a very short posterior
extension (pers. obs.). Boss (1965a: 4) inter-
preted the white area at the end of the foot as
the byssus gland (as did Judd [1971] for D.
maoria, and Narchi [1966] for Ceratobornia
cema). In view of the byssus adhesive gland
described here and previously (Mikkelsen &
Bieler, 1989), this organ (which was depicted
with internal lamellae by Boss, 1965a: fig. 3)

22 MIKKELSEN 8 BIELER

warrants reinvestigation. If this structure is the
primary byssus gland, what is the function of
the ventral groove? (Histological study of the
structure of the entire foot, and observations
of living animals producing byssus threads
would resolve this question.) Boss (1965a)
placed Parabornia in the Erycinidae (Erycini-
nae), in part, because its hinge has two car-
dinal teeth in each valve and a central resil-
ium. If hinge structure is eventually confirmed
as a reliable taxonomic character based on
phylogeny of the group, the “hanging foot”
and “flower-like organs” would need to be ex-
plained as results of evolutionary conver-
gence and/or symplesiomorphies.

Mating Behavior in Bivalves

The observed “mating” behavior (Fig. 33)
between individuals of Divariscintilla octoten-
taculata as well as of D. yoyo is of special
interest. Mating behavior has not been con-
firmed for bivalves (Mackie, 1984: 362).
Mackie (1984: 363) summarized the reported
cases of associations between females and
dwarf males in bivalves and concluded that
those interactions would qualify as mating be-
havior if it can be shown that the associations
are confined to a breeding period.

True copulatory behavior is absent in bi-
valves because they lack copulatory organs.
However, several authors (Clapp, 1951: 7;
Townsley et al., 1966: 49) have observed
male teredinids (Bankia spp.) inserting their
exhalent siphons into the inhalent siphons of
females during sperm release. Purchon
(1977: 295), and subsequently Mackie (1984:
362), interpreted this as the use of an intromit-
tant organ, the development of which would
be of “great survival value to species ... living
in isolated pieces of drift wood (Purchon,
1977: 295). Similarly, quasi-copulatory be-
havior in Divariscintilla would be advanta-
geous for these small bivalves with limited ga-
mete production, living in restricted, relatively
isolated populations within the burrows of
their hosts. Selective mating in burrows that
contain several other congeneric species
would guarantee successful fertilization and
make conservative use of male gametes. In
Our previous discussion of flower-like organs,
now known from five species of Divariscintilla,
we (Mikkelsen & Bieler, 1989: 191) specu-
lated that these organs might emit a phero-
mone for attracting reproductive partners
(another possibility, if they are indeed phero-

mone-emitting organs, would be the attraction
of larvae to settlement sites).

ACKNOWLEDGMENTS

The following are gratefully acknowledged:
Dr. Raymond B. Manning (USNM), for bring-
ing this interesting fauna to our attention;
William D. (“Woody”) Lee (SMSLP), for in-
valuable field help; Paul H. Scott (SBMNH),
for comments on generic placement; Patricia
A. Linley and Pamela Blades-Eckelbarger
(both НВО!) for SEM assistance; Tom
Smoyer (HBOl) for photographic work; Dr.
Kerry B. Clark (Florida Institute of Technol-
ogy, Melbourne) and John E. Miller (formerly
HBO!) for the use of and help with video re-
cording equipment; Paul S. Mikkelsen (for-
merly HBOI), for behavioral and developmen-
tal notes and the photograph in Fig. 1; Chris
Bauman (Vero Beach, Florida) for the illustra-
tion in Fig. 33; Paul H. Scott (SBMNH), Dr.
Kenneth J. Boss (MCZ), and two anonymous
reviewers for valuable comments on the
manuscript.

This research was supported in part by the
Smithsonian Marine Station at Link Port; the
cooperation of Dr. Mary E. Rice and her staff
is gratefully acknowledged. Funding for this
project was derived in part from a National
Capital Shell Club scholarship to P. M., and a
NATO Postdoctoral Research Fellowship at
SMSLP to R. B. This is Harbor Branch
Oceanographic Institution Contribution no.
862 and Smithsonian Marine Station Contri-
bution no. 273.

LITERATURE CITED

ANTHONY, R., 1916, Contribution a l'étude de
l’Entovalva (Synapticola) perrieri Malard, mol-
lusque acéphale commensal des synaptes. Ar-
chives de Zoologie Expérimentale et Générale,
55: 375-391, pls. 6-7.

ARAKAWA, K. Y., 1960, Ecological observations
on an aberrant lamellibranch, Ephippodonta mu-
rakamii Kuroda. Venus, Japanese Journal of Ma-
lacology, 21(1): 50-60, pls. 7-8.

BERNARD, F. R., 1975, Rhamphidonta gen. n.
from the northeastern Pacific (Bivalvia, Leptona-
cea). Journal de Conchyliologie, 112(3-4): 105-
HIS:

BIELER, R. 8 P. M. MIKKELSEN, 1988, Anatomy
and reproductive biology of two western Atlantic
species of Vitrinellidae, with a case of protan-

NEW SPECIES OF COMMENSAL GALEOMMATIDAE 23

drous hermaphroditism in the Rissoacea. The
Nautilus, 102(1): 1-29.

BOSS, K. J., 1965a, A new mollusk (Bivalvia, Ery-
cinidae) commensal on the stomatopod crusta-
cean Lysiosquilla. American Museum Novitates,
no. 2215: 11 pp.

BOSS, K.J., 1965b, Symbiotic erycinacean bi-
valves. Malacologia, 3(2): 183-195.

BRUUN, A. F., 1938, A new entocommensalistic
bivalve, Entovalva major n.sp., from the Red
Sea. Videnskabelige Meddelelser fra Dansk
naturhistorisk Forening i Kobenhavn, 102: 163-
167.

CHAVAN, A., 1969, Superfamily Leptonacea Gray,
1847. Pp 518-537, in: R. C. MOORE, ed.,
Treatise on Invertebrate Paleontology, Part N.
Mollusca 6. Bivalvia 2. Geological Society of
America (Boulder, Colorado) & University of Kan-
sas (Lawrence), ii + N491-N952 pp.

CLAPP, W. F., 1951, Observations on living Tere-
dinidae. Fourth Progress Report on Marine Borer
Activity in Test Boards Operated during 1950.
Report no. 7550, W. F. Clapp Laboratories, Inc.,
Duxbury, Massachusetts, 9 pp.

CONEY, C. C., 1990, Bellascintilla parmaleeana
new genus and species from the tropical eastern
Pacific, with a review of the other, ventrally
notched galeommatid genera (Bivalvia: Galeom-
matacea). The Nautilus, 104(4): 130-144.

CORNET, M., 1982, Anatomical description of Vas-
coniella jeffreysiana (P. Fischer, 1873) (Mollusca,
Bivalvia, Leptonacea). Journal of Molluscan
Studies, 48(1): 36-43.

DALL, W. H., 1899, Synopsis of the Recent and
Tertiary Leptonacea of North America and the
West Indies. Proceedings of the United States
National Museum, 21(1177): 873-897, pls. 87-88.

DALL, W. H., P. BARTSCH 8 H. A. REHDER,
1938, A manual of the Recent and fossil marine
pelecypod mollusks of the Hawaiian Islands. Ber-
nice P. Bishop Museum Bulletin, 153: iv + 233
pp., 58 pls.

ECKELBARGER, K. J., R. BIELER 4 P. M. MIK-
KELSEN, 1990, Ultrastructure of sperm develop-
ment and mature sperm morphology in three
species of commensal bivalves (Mollusca: Ga-
leommatoidea). Journal of Morphology, 205: 63-
75:

FOX, T. H., 1979, Reproductive adaptations and life
histories of the commensal leptonacean bivalves.
Ph.D. dissertation, University of North Carolina,
Chapel Hill, ix + 207 pp.

HUMASON, G. L., 1962, Animal Tissue Tech-
niques. W. H. Freeman & Co., San Francisco &
London, xv + 468 pp.

JEFFREYS, J. G., 1863, British Conchology, or an
account of the Mollusca which now inhabit the
British Isles and the surrounding seas. Vol. Il.
Marine Shells, comprising the Brachiopoda, and
Conchifera from the family of Anomiidae to that of
Mactridae. London, John Van Voorst, xiv + 465
pp., 8 pls.

JENNER, C. E. 8 A. B. MCCRARY, 1968, A “gas-

tropod” bivalve, commensal on Squilla empusa
(Abstract). Annual Report for 1968, American
Malacological Union: pp. 20-21.

JUDD, W., 1971, The structure and habits of Di-
variscintilla maoria Powell (Bivalvia: Galeomma-
tidae). Proceedings of the Malacological Society
of London, 39: 343-354.

MACKIE, G. L., 1984, Bivalves. Pp. 351-418,
in: A. S. TOMPA, et al., The Mollusca, Vol. 7, Re-
production. Academic Press, Orlando, xix + 486

pp.

MIKKELSEN, P. M. 8 R. BIELER, 1989, Biology
and comparative anatomy of Divariscintilla yoyo
and D. troglodytes, two new species of Galeom-
matidae (Bivalvia) from stomatopod burrows in
eastern Florida. Malacologia, 31(1): 1-21.

MORTON, B., 1973a, The biology and functional
morphology of Galeomma (Paralepida) takii (Bi-
valvia: Leptonacea). Journal of Zoology, 169(2):
133-150.

MORTON, B., 1973b, Some factors affecting the
location of Arthritica crassiformis (Bivalvia: Lep-
tonacea) commensal upon Anchomasa similis
(Bivalvia: Pholadidae). Journal of Zoology, 170:
463-473.

MORTON, B., 1975, Dymantic display in Ga-
leomma polita Deshayes (Bivalvia: Leptonacea).
Journal of Conchology, 28: 365-369.

MORTON, B., 1976, Secondary brooding of tem-
porary dwarf males in Ephippodonta (Ephipp-
odontina) oedipus sp. nov. (Bivalvia: Leptona-
cea). Journal of Conchology, 29: 31-39.

MORTON, B. & P. H. SCOTT, 1989, The Hong
Kong Galeommatacea (Mollusca: Bivalvia) and
their hosts, with descriptions of new species.
Asian Marine Biology, 6: 129-160.

MORTON, J. E., 1957, The habits of Scintillona ze-
landica (Odhner) 1924 (Lamellibranchia: Ga-
leommatidae). Proceedings of the Malacological
Society of London, 32(5): 185-188.

NARCHI, W., 1966, The functional morphology of
Ceratobornia cema, new species of the Erycina-
cea (Mollusca, Eulamellibranchiata). Anais da
Academia Brasileira de Ciéncias, 38(3-4): 513-
524.

NORMAN, A. M., 1891, Lepton squamosum (Mon-
tagu), a commensal. Annals and Magazine of

_ Natural History, (6) 7(39): 276-278.

O FOIGHIL, D. 8 A. GIBSON, 1984, The morphol-
ogy, reproduction and ecology of the commensal
bivalve Scintillona bellerophon spec. nov. (Ga-
leommatacea). The Veliger, 27(1): 72-80.

POPHAM, M. L., 1939, On Phlyctaenachlamys
lysiosquillina gen. and sp. nov., a lamellibranch
commensal in the burrows of Lysiosquilla macu-
lata. British Museum (Natural History), Great Bar-
rier Reef Expedition 1928-29, Scientific Reports,
6(2): 549-587.

PURCHON, R. D., 1977, The Biology of the Mol-
lusca, 2nd ed. Pergamon Press, Oxford, xxv +
560 pp.

RICE, E. L., 1897, Die systematische Verwert-
barkeit der Kiemen bei den Lamellibranchiaten.

24

MIKKELSEN 4 BIELER

Jenaische Zeitschrift für Naturwissenschaft, 31
(N.F. 24): 29-89, pls. 3-4.

TAYLOR, J. D., W. J. KENNEDY & A. HALL, 1973,
The shell structure and mineralogy of the Bi-
valvia. Il. Lucinacea—Clavagellacea, conclu-
sions. Bulletin of the British Museum (Natural
History), 22(9): 253-294, pls. 1-15.

TOWNSLEY, P. M., R. A. RICHY 8 P. C. TRUS-
SELL, 1966, The laboratory rearing of the ship-
worm, Bankia setacea (Tryon). Proceedings of
the National Shellfisheries Association, 56: 49-52.

VACCA, L. L., 1985, Laboratory manual of his-
tochemistry. Raven Press, New York, xvii + 578

WINCKWORTH, R., 1930, Notes on nomenclature.

Proceedings of the Malacological Society of Lon-
don, 19(1): 14-16.

YAMAMOTO, T. 8 T. HABE, 1974, Scintillona stig-

matica (Pilsbry) new to Japan. Venus, Japanese
Journal of Malacology, 33(3): 116.

Revised Ms accepted 1 August 1991

MALACOLOGIA, 1992, 34(1-2): 25-32

EMBRYONIC DEVELOPMENT OF BIOMPHALARIA GLABRATA
(SAY, 1818) (MOLLUSCA, GASTROPODA, PLANORBIDAE):
A PRACTICAL GUIDE TO THE MAIN STAGES

Toshie Kawano (Camey),' Kayo Okazaki? 8 Lillane Ré‘

ABSTRACT

The morphology of the gastropod mollusk Biomphalaria glabrata (Say, 1818) from the stage
proceeding the first egg cleavage to the hippo stage is presented. The morphogenetic changes
along its development were followed by in vivo observation and by cell lineage studies, with
description of the origin and function of the main structures that characterize the embryonic and

larval phases.

Key Words: Biomphalaria glabrata, Planorbidae, cleavage, embryonic, development.

INTRODUCTION

The studies of cell lineages in mollusks are
of great interest for research in experimental
embryology. The first ones date back to the
end of the last century, when different species
were investigated by many authors, such as
Neritina fluviatilis by Blochmann (1882), Um-
brella mediterranea by Heymons (1893),
Ischnochiton sp. by Heath (1899), Planorbis
trivolvis by Holmes (1900), Physa fontinalis by
Wierzejski (1905), Littorina obtusata by Dels-
man (1912, 1914), Limnaea stagnalis by
Raven (1946, 1958) and Verdonk (1965) and
Dentalium sp. by Dongen (1976). Camey &
Verdonk (1970) carried out studies on cell lin-
eage of the gastropod Biomphalaria glabrata
(Say, 1818); they analyzed its embryonic
stages, with emphasis on the cephalic region,
from the first egg cleavages to the veliger
stage.

This guide was prepared in order to de-
scribe the various stages of the embryonic
and larval development of B. glabrata and
was also especially planned for those initiat-
ing research activities or attending or giving
lectures on embryology. It covers the stages
between the first egg cleavages, which are
characterized by the spiral (helicoidal) type,
and the trochophore and veliger stages, all of
them occurring within the egg capsule.

MATERIAL AND METHODS

Egg masses of Biomphalaria glabrata (Say,
1818) were picked up in Belo Horizonte, State

of Minas Gerais, Brazil, and maintained for
several years in the laboratory, at 25°C, ina
climatic chamber.

The developmental stages were followed
and photographed in vivo by using a Zeiss
photomicroscope. From the first to the fourth
cleavages, the stages were photographed at
intervals of 1 to 30 minutes. From the fourth
cleavage on, the time intervals were of 5
hours. From the early trochophore stage on,
the larvae were immobilized with ethyl ether
for photography. These time intervals were
decided upon based on data obtained in pre-
vious observations of the embryonic develop-
ment of this species.

Stage 1 (Fig. 1B) was considered as the
starting point for the determination of the em-
bryo age due to the variation of the time in-
tervals elapsing between egg laying and first
egg cleavage, as pointed out by Raven (1946).
Due to the asynchrony of egg division in the
same egg-mass, zero time was considered to
be the time when 50% of the eggs were in
stage 1.

For the cell lineage study, the embryos were
prepared according to the technique de-
scribed by Verdonk (1965), which basically
consists of the following steps: the embryos
were decapsulated, fixed with 0.75% AgNO,
for a few seconds until the cell contours be-
came sharp, dehydrated in an alcohol series,
cleared with xylene, and mounted with Per-
mount between a slide and a coverslip. Em-
bryos and larvae were drawn using a Zeiss
camera lucida, and the nomenclature adopted
was that of Conklin (1897).

"Servico de Zoonoses Parasitárias e Parasitologia, Instituto Butantan Cx.Postal 065, CEP 01000, Sao Paulo, SP, BRAZIL.
“Divisáo de Radiobiologia, Departamento de Aplicacóes em Ciéncias Biológicas, Instituto de Pesquisas Energéticas e
Nucleares, Comissáo Nacional de Energia Nuclear/SP, Sáo Paulo, SP, BRAZIL

26 KAWANO, OKAZAKI & RÉ

j4p.b.

1A 1

15 16 17

| а.р. en
+. is uE
B 2 3

FIG. 1. Early development of Biomphalaria glabrata. (1A) Undivided egg with two polar bodies. (1B) Begin-
ning of first cleavage, with appearance of the cleavage furrow in the animal pole (0 hr). (2) Appearance of
the cleavage furrow, after 3 min. (3) After 4 min. (4) Two blastomeres attached to each other by a small
cytoplasmic area, after 10 min. (5), (6) and (7) the blastomeres present a gradually larger contact surface:
after 21, 26 and 38 min., respectively, (8) Emergence of the cleavage cavity after 45 min. (9) Cleavage cavity
appearing more clearly, after 75 min. (10) Second cleavage, after 80 min. (11) Four blastomere stage, after
91 min. (12) Beginning of the appearence of the cleavage cavity, at 98 min. (13) After 103 min. (14) After
123 min. (15) Almost rounded surface of the blastomeres, at 134 min. (16) Third cleavage with formation of
the first quartette of micromeres, at 160 min. (17) At 165 min. (18) At 173 min. (19) At 181 min. (20) Fourth
cleavage, at 230 min. p.b. = polar body, a.p. = animal pole, v.p. = vegetative pole, c.c. = cleavage cavity,

m = micromere, M = macromere.

RESULTS

Table 1 shows the main stages of the em-
bryonic development of B. glabrata, from the
beginning of the first cleavage to the hippo
stage.

The morphological features of B. glabrata
during the main stages of its embryonic and
larval development are illustrated in Figures
1-3. The embryonic and larval cells are sche-
matically presented in Figure 4.

Figure 1 presents a sequence of morpho-
genetic alterations covering the period of time
between the zygote (or fertilized egg) and the

fourth cleavage stage. Stage 1A shows the
zygote before the first cleavage; two polar
bodies (p.b.) can be seen at the upper apical
region, exactly in the animal pole. The diam-
eter of a viable egg is approximately 100 ¡um
long. Stages from 1 to 19 are equivalent to
those described for Limnaea stagnalis by
Raven (1946). The first cleavage furrow ap-
pears during stage 1B in the animal pole. This
phase corresponds to the beginning of the
first cleavage, which is total and passes
through the animal and vegetative poles.
Stages 2 and 3 occur approximately 3 or 4
minutes after stage 1B, respectively, with

EMBRYONIC DEVELOPMENT OF BIOMPHALARIA 27

TABLE 1. Main stages of the embryonic development of B. glabrata at 25°C.

Time interval

Number of stage

Embryonic stage between observations in figures
Beginning of the 1st cleavage 0 1B
2nd cleavage 80 min. 10
3rd cleavage 160 min. 16
4th cleavage 230 min. 20
blastula 15 hrs. 21
gastrula 26 hrs. 23
early trochophore 43 hrs. 25
late trochophore 66 hrs. 27
early veliger 96 hrs. 28
late veliger 120 hrs. 29
hippo stage 144 hrs. 30

cleavage furrow becoming gradually more
predominant; the cleavage of the animal pole
region is more rapid than that of the vegeta-
tive one. At stage 4 the cleavage furrow al-
most completely divides the zygote into two
blastomeres, both of them presenting a well-
rounded surface and linked to each other by
only a small cytoplasmic bridge. This phase
occurs approximately 10 minutes after stage
1B (Fig. 1, stage 4; Fig. 4, stage 31). After
stage 5, at approximately 21 minutes from the
beginning of zero time, the two blastomeres
approach each other, increasing their contact
surface, that is, they flatten against each
other, with the formation of a separating blas-
tomeric membrane (Fig. 1, stage 5). During
stages 6 and 7 (Fig. 1), the boundaries of the
blastomeres can hardly be defined; these
stages occur approximately 26 and 38 min-
utes after stage 1B, respectively.

The cleavage cavity, the function of which is
osmotic regulation, can be seen between the
two blastomeres at stage 8, approximately 45
minutes after stage 1B (Fig. 1). This cavity
increases in size, and both blastomeres have
an egg shape (Fig. 1, stages 8 and 9).

The beginning of the second cleavage (Fig.
1, stage 10) occurs 80 minutes after the first
cleavage. Because the blastomeres do not di-
vide synchronously, the cleavage furrow of
one of the blastomeres appears before the
other, so that this stage presents an asym-
metrical shape (Fig. 1, stage 10). This cleav-
age is meridional and total.

During stage 11 (Fig. 1), after approxi-
mately 91 minutes, the blastomeres show a
rounded shape but are not on the same plane
when observed obliquely; that is, blastomeres
A and C (Fig. 4, stage 32) are linked to each
other by the furrow in the animal pole, and

blastomeres B and D by the furrow in the veg-
etative pole.

At stage 12 (Fig. 1), after approximately 98
minutes, the cleavage furrow linking the alter-
nate blastomeres in the animal and vegeta-
tive poles of the egg can be seen more clearly
(Fig. 1, stage 12; Fig. 4, stage 32). At stages
13 (after 103 min) and 14 (after 123 min) (Fig.
1), the blastomeres are already more closely
joined, and the cleavage cavity between them
begins to appear. Material from the cleavage
cavity then starts coming out, with contraction
of the whole egg causing a visible reduction in
diameter (Fig. 1, stage 15).

The third cleavage is laeotropic and occurs
in the subequatorial plane of the egg, with the
formation of the first micromere quartette (1a
to 1d), approximately 160 minutes after the be-
ginning of the first cleavage (Fig. 1, stage 16;
Fig. 4, stage 33). A gradual increase of the
cleavage cavity is observed during stages 17
(after 165 min), 18 (after 173 min) and 19
(Fig. 1).

Stage 20 (Fig. 1) illustrates the egg at the
fourth cleavage stage, which is dexiotropic.
The fourth cleavage is subequatorial and oc-
curs approximately 230 minutes after stage
1B (Fig. 4, stage 34), originating the second
micromere quartette (2a to 2d).

The embryo reaches the blastula stage ap-
proximately 10 to 23 hours after the 1st cleav-
age (Figs. 21, 22).

Stages 35 and 36 (Fig. 4) show the embryo
with about 64 and 130 cells, respectively, and
presenting a set of cells forming a cross-like
figure (1a** to 1d**, 1a'? to 1d*? and 2a"' to
2d'' cell lineages) in the animal pole, which
gives origin to almost the whole head region.

Gastrulation occurs about 24 to 39 hours
after the first cleavage (Fig. 2, stages 23, 24A,

28 KAWANO, OKAZAKI 8 RÉ

21

24A 24B

FIG. 2. Different stages of development of Biomphalaria glabrata. (21) Blastula, 15 hrs. after 1st cleavage.
(22) Blastula, 21 hrs. and 30 min. after 1st cleavage (23) Gastrula, 26 hrs. after 1st cleavage. (24A) Late
gastrula as seen from the vegetative pole after 34 hrs. (24B) Lateral view of the 34 hrs. gastrula stage. (25)
Early trochophore, after 43 hrs. a.p. = animal pole, v.p. = vegetative pole.

24B; Fig. 4, stages 37 and 38). Flattening of
the vegetative pole region towards the animal
pole then occurs, followed by the invagination
of this region, with the formation of a spherical
opening, the blastopore, which then becomes
gradually smaller.

The early trochophore stage appears ap-
proximately 40 to 65 hours after the first cleav-
age, and is characterized by the first larval
movements (Fig. 2, stage 25; Fig. 3, stage 26)
by means of the ciliated cells of the prototroch,
which divides the larva in the pretrochal re-
gion, characterized by the presence of an api-
cal plate, head vesicle and cephalic plates,
and the posttrochal region, characterized by
the presence of a shell gland, stomodeum and
foot (Fig. 4, stage 39). The prototroch shows
a double row of cells, an upper one formed by
2d!11, 26112, 1a°', 1a22, 161 and 1622 cells,
and a lower ciliated one consisting of 2b***
and 2b'*' cells and of cells descending from
2b'** and 2b*' cells (Fig. 4, stage 39).

The late trochophore stage occurs approx-
imately 65 to 80 hours after the first cleavage,
with the larva having a slightly elongated, kid-
ney-like shape. In this phase, cells responsi-
ble for the formation of the head and foot can
be seen in its anterior region, and in the dor-

sal region one can see the shell gland (Fig. 3,
stage 27). Stage 40 (Fig. 4) shows a greater
development of the prototroch, leading to
more active larval movements. The pretrochal
region, located above the prototroch, is
formed by a set of cells that gives rise to such
larval structures as the apical plate (1а''' to
1d, 1212", 1b""2 and 1b = 0) copie
plates (cells located laterally to the apical
plate), and the head vesicle (1c?*, 1d?*, 1c°?,
id: taco 1 id iq igo
2a**, 2c**, 2d"'), which in turn will form the
head. In the posttrochal region, below the pro-
totroch, are the stomodeum (M), which will
give origin to the mouth, and cells that will
form the foot and the shell gland.

The time interval between 80 and 100
hours after the first cleavage corresponds to
the early veliger stage. The shell and foot are
then more developed (Fig. 3, stage 28). The
prototroch has evolved into the velum (Fig. 4,
stage 41). In this stage, the shell gland shifts
towards the right side.

The veliger stage occurs about 120 hours
after the first egg cleavage. The eyes become
more visible, and the elevation of the tentacle
regions can be observed, as well as the
mouth, foot and shell (Fig. 3, stage 29). Stage

EMBRYONIC DEVELOPMENT OF BIOMPHALARIA 29

FIG. 3. Different stages of development of Biomphalaria glabrata. (26) Early trochophore, 55 hrs. after 1st
cleavage. (27) Trochophore, after 66 hrs. (28) Early veliger, after 96 hrs. (29) Veliger, after 120 hrs. (30)
Hippo stage, 144 hrs. after first cleavage. E = eyes, F = foot, H = head, M = mouth, S = shell, Pr =

prototroch.

42 (Fig. 4) shows that the apical plates and
head do not undergo any changes, while the
cephalic plates go on dividing to form the ten-
tacles and eyes.

The hippo stage is a phase of larval devel-
opment that occurs within the egg capsule ap-
proximately 144 hours after the first cleavage.
At this stage, the eyes and tentacles in the
pretrochal region are already well developed,
and in the posttrochal region the foot has
grown in size and is much more differentiated.
The shell starts coiling and covers almost the
whole body (Fig. 3, stage 30). At 25°C, the
young snail hatches from the egg capsule be-
tween the sixth and ninth day after the first
cleavage.

DISCUSSION

Due to their semi-transparence and easy
maintenance in the laboratory, the eggs of B.
glabrata represent a suitable material for
studies on embryology, allowing a good in
vivo observation. At egg-laying, its egg-
masses already contain fertilized eggs and,
after approximately 30 minutes (at 25°C), the
emission of the first polar body occurs, as a
result of the first meiotic division. Approxi-
mately 60 minutes later, the second polar
body emerges. Both polar bodies remain in
the animal pole during the first cleavages.
Within 60 minutes after the emission of the
second polar body, the fusion of the male and

30 KAWANO, OKAZAKI & RÉ

FIG. 4. Outline of cell boundaries on developing eggs of Biomphalaria glabrata. (31) 1st cleavage with 2
blastomeres. (32) 2nd cleavage with 4 blastomeres, 80 min. after 1st cleavage. (33) 3rd cleavage with 1st
quartette of micromeres (1a to 1d) and with 8 blastomeres, after 160 min. (34) 4th cleavage with 2nd
quartette of micromeres (2a to 2d) and 12 blastomeres, after 230 min. (35) Blastula stage with 64 blas-
tomeres, showing the cross figure at the animal pole after 15 hrs. (36). Blastula with 130 cells (animal pole)
after 21 hrs. and 30 min. (37). Gastrula, after 26 hrs. (38) Gastrula after 34 hrs. (39) Early trochophore after
34 hrs. (40) Trochophora after 66 hrs. (41) Early veliger after 96 hrs. (42) Veliger stage after 120 hrs. M =
mouth, A.P. = apical plate, H.V. = head vesicle, C.P. = cephalic plate, V = velum, T = Tentacle, E
eyes, F = foot, Pr. = prototroch.

31

EMBRYONIC DEVELOPMENT OF BIOMPHALARIA

NIT

Dem

ps:

в
E

32 KAWANO, OKAZAKI & RÉ

female pronuclei occurs, the egg now being
ready for the first cleavage (Camey, 1968).

Except for cephalopods, the cleavage pat-
tern of mollusks is of the spiral type; that is,
blastomere cleavage is oblique in relation to
its axis and is of the determinative type, be-
cause the different egg regions are going to
form the future organs (Raven, 1958).

The developmental stages described here
are similar to those of Limnaea stagnalis, with
some differences. Cleavage is laeotropic or
reverse in B. glabrata (Camey & Verdonk,
1970) and dexiotropic in L. stagnalis (Ver-
donk, 1965). The differentiation of the type of
cleavage occurs according to the orientation
of the division spindle. When this is oblique in
relation to the axis of the egg in clockwise
direction, cleavage is dexiotropic, and when
the spindle is oblique but in a counterclock-
wise direction, cleavage is laeotropic. The
first indication of the orientation of the cleav-
age spindle in B. glabrata occurs during the
third cleavage (Camey & Verdonk, 1970).

In the trochophore stage, the shell gland is
shifted to the right side in B. glabrata (Camey,
1968) and to the left side in L. stagnalis (Ver-
donk, 1965).

When the helicoidal shell of an adult of L.
stagnalis is placed with the apex turned up, it
can be seen that the opening is towards the
right side. In B. glabrata, with a planispiral
shell, it is difficult to determine the direction of
the shell opening, except for its shift in a di-
rection opposite to that of L. stagnalis during
larval development. It may be concluded,
therefore, that the cleavage pattern and shell
opening of B. glabrata are laeotropic.

ACKNOWLEDGEMENTS

Thanks are due to Dr. Kaoru Hiroki for the
critical reading of this paper and to Miss. Vera
Helena Monezi for typing this manuscript.

LITERATURE CITED

BLOCHMANN, F., 1882, Ueber die Entwicklung der
Neritina fluviatilis. Mull. Zeitschrift fur Wissen-
shaftliche Zoologie. Leipzig. 36: 125-174.

CAMEY, T., 1968, Estágios iniciais do desenvolvi-
mento embrionario da Biomphalaria glabrata
(Say, 1818). Tese de Doutoramento. Belo Hori-
zonte, MG, Brasil.

CAMEY, T. & N. H. VERDONK, 1970, The early
development of the snail Biomphalaria glabrata
(Say) and the origin of the head organs. Nether-
lands Journal of Zoology, Leiden, 20(1): 93-121.

CONKLIN, E. G., 1897, The embryology of Crepi-
dula. Journal of Morphology, 13: 1-226.

DELSMAN, H. C., 1912, Ontwikkelingsgeschiede-
nis van Littorina obtusata. Thesis, Univ. of Am-
sterdam, Amsterdam, The Netherlands.

DELSMAN, H. C., 1914, Cf. Entwicklungsge-
schichte von Littorina obtusata. Tijdschrift Neth-
erlandsche Dierkundige Vereeniging. Leiden,
13(2): 170-340.

DONGEN, C. A. M. van, 1976, The development of
Dentalium with special reference to the signifi-
cance of the polar lobe. Utrecht, Rijksuniversiteit,
92 pp. PhD. Thesis.

HEATH, H., 1899, The development of /schnochi-
ton. Zoologische Jahrbucher, Anatomie, 12: 567—
656.

HEYMONS, R., 1893, Zur Entwicklungsgeschichte
von Umbrella mediterranea Lam. Zeitschrift für
Wissenschaftliche Zoologie, 36: 245-298.

HOLMES, S. J., 1900, The early development of
Planorbis. Journal of Morphology. 16: 369-458.

RAVEN, C. P., 1946, The development of the egg of
Lymnaea stagnalis L. from the first cleavage till
the trochophore stage, with special reference to
its “chemical embriology”. Archives Neerlan-
daises de Zoologie, Leiden, 7(3-4): 353-434.

RAVEN, C. P., 1958, Morphogenesis: the analysis
of molluscan development. London, Pergamon
Press. 311 pp. (International series of mono-
graphs on pure and applied biology. Division Zo-
ology, v. 2)

VERDONK, N. H., 1965, Morphogenesis of the
head region in Lymnaea stagnalis L. Utrecht, 133
pp. PhD. Thesis.

WIERZEJSKI, A., 1905, Embryologie von Physa
fontinalis L. Zeitschrift für Wissenchaftliche Zool-
ogie, 83: 502-706.

Revised Ms. accepted 31 August 1991

MALACOLOGIA, 1992, 34(1-2): 33-61

ALLOZYME FREQUENCIES IN ALBINARIA (GASTROPODA: PULMONATA:
CLAUSILIIDAE) FROM THE IONIAN ISLANDS OF KEPHALLINIA AND ITHAKA

Th. C. M. Kemperman & G. H. Degenaars'

Systematic Zoology Section, Population Biology Department, Rijksuniversiteit Leiden,

c/o National Museum of Natural History, P.O. Box 9517, 2300 RA Leiden, The Netherlands

ABSTRACT

In 42 samples comprising 11 of the 12 subspecies of the four species of Albinaria occurring
on the islands of Kephallinia and adjoining Ithaka, Greece, allozyme variation was studied at 28
gene loci by starch gel electrophoresis. To serve as outgroups, single samples of two Cretan
Albinaria species and both an Albinaria and an Isabellaria species from the Peloponnese were
included. The various taxa varied at 0.0% to 35.7% of the loci, with an average of 12.2%. In 71%
of the populations, a deficiency of heterozygosity was found. Taxa differ in the number of
populations that must be analysed to reach a point where further addition of populations does
not increase the genetic variation. The results show relatively low genetic distances between the
taxa from Kephallinia; between the subspecies of the two non-endemic species hardly any
allozyme differentiation is found. In the two endemic species, however, the subspecies are less
similar to each other and show conspicuous genetic resemblances with different non-endemic
species. The taxon from Ithaka differs markedly from all Kephallinian taxa and thus geographical
isolation seems to be an important factor with respect to genetic differentiation. There is evi-
dence for introgressive hybridization. In at least one case, we found indications for hybrizymes.
A Hennigian analysis did not result in a resolved phylogeny. The present classification of species
and subspecies, based on morphological and biogeographical data, is maintained only with
some doubt.

Key words: systematics, allozymes, genetic variation, hybridization, electrophoresis, Albina-

ria, Pulmonata.

INTRODUCTION
General Notes

Within the context of a study on the sys-
tematics and evolutionary history of the pul-
monate genus Albinaria Vest, 1867 (Gitten-
berger, in press), special attention has been
given to the Albinaria species of the neighbor-
ing lonian islands Kephallinia and Ithaka. The
aims of this subproject are (1) to unravel the
pattern of taxonomic relationships among the
various taxa and (2) to get some insight into
the population genetics and the evolutionary
history of these pulmonates. This goal is ap-
proached by using shell morphology (Kem-
perman & Gittenberger, 1988), radula mor-
phology (Kemperman et al., in prep.), genital
morphology, distributional patterns, and allo-
zyme variation (Degenaars, in press). In ad-
dition to the latter preliminary report, the

present paper deals with the biochemical data
in more detail.

The value of electrophoretic techniques for
systematic study is widely recognized (Avise,
1974; Buth, 1984; Richardson et al., 1986;
Ferguson, 1988; Nevo, 1988; Murphy et al.,
1990). Allozyme variation among samples of
different populations provides a data-set that
may be considered virtually independent of
the morphological data (Avise, 1974; Ember-
ton, 1988). Albinaria has been the subject of
electrophoretic studies before. Ayoutanti et al.
(1988) reported on an allozyme analysis of 23
Albinaria populations from 12 Aegean is-
lands, including Crete. Using 27 loci, these
authors tried to unravel the genetic affinities
among a group of Albinaria species from
many islands with generally only one or two
Albinaria taxa each. We limited our study to
the rich Albinaria fauna of only two, neighbor-
ing islands. The studies show some overlap in

“Present address: Hugo de Vries Lab., University of Amsterdam, Kruislaan 318, 1098 SM Amsterdam, The Netherlands.

34 KEMPERMAN 8 DEGENAARS

that we used, for example, Cretan material as
outgroup and partly analyzed the same en-
zyme systems.

Geography of Kephallinia and Ithaka

The largest island, Kephallinia (Fig. 1),
measures about 50 km from north to south
and somewhat more than 20 km across the
middle of the main part of the island (that is,
without the Paliki Peninsula), which is domi-
nated by a mountain ridge, reaching 1,628 m
altitude. All over the island are areas with
limestone rocks and/or rock faces inhabited
by Albinaria. Ithaka is only 23 km long and up
to 6 km wide. It consists of two large penin-
sulas joined by a narrow, high isthmus.

The present 130 m isobath can probably be
looked upon as the Pleistocene shore line of
approximately 18,000 B.P., when sea level
reached its lowest point (Andel & Shackleton,
1982: 447-448) during the Wúrm. Judging
from the isobaths on Admiralty Chart no. 203
(1988) and data given by Bird (1984: 81, fig.
41), we may assume that between approxi-
mately 29,000 and 12,000 years ago there
was a minimally 1 km-broad land connection
between Kephallinia and Ithaka, running from
Cape Dichalia, northeast of Sami on Kephal-
linia, and the south-southwest coast of Ithaka.
At present, this is a stretch of approximately 4
km sea water. North of this area, there have
not been relatively recent land bridges, be-
cause there the sea between Kephallinia and
Ithaka is over 145 m deep, with a depth of 182
m near Fiskardo, northeastern Kephallinia.
Thus, Ithaka and Kephallinia have been sep-
arate islands for at least 12,000 years.

Because the sea between Lixouri and Ar-
gostoli is at present only 25 m deep, we have
to accept that during glacial sea-level lower-
ings the entire Gulf of Argostoli between the
Paliki Peninsula and the main island was
land.

Distribution of Albinaria on Kephallinia
and Ithaka

In Figure 1, the ranges of the Albinaria taxa
from Kephallinia and Ithaka are indicated,
based on our data from approximately 800
localities (Kemperman, in prep. A). Different
shadings indicate different species, whereas
broken thick lines separate the ranges of sub-
species. Based on classical conchological
characters, four species are distinguished,
each with at least two subspecies.

On Kephallinia, two species are endemic,
both with limited ranges, namely A. jonica and
A. adrianae. On Ithaka no endemic Albinaria
species have been found.

Albinaria contaminata ranges on the lonian
islands from Lefkas to Zakynthos and on the
mainland from the southern Epirus to the
province of Phocis. It is the only Albinaria
known from Ithaka. On Kephallinia it is the
most widely distributed species, occurring al-
most everywhere, except for places along the
west coast and the Paliki Peninsula. The sec-
ond widespread species is A. senilis, occur-
ring on all the lonic islands, except for Za-
kynthos, and from the Epirus southward to the
mainland opposite Lefkas.

MATERIALS AND METHODS
Species Sampled

In total, 46 populations are examined elec-
trophoretically, 42 of which were from Kephal-
linia and Ithaka, representing the following
taxa (between square brackets, the number of
investigated populations: individuals is given):
Albinaria contaminata contaminata (Ross-
massler, 1835) [7: 128], A. c. incommoda
(Boettger, 1878) [4: 36], A. c. odysseus (Boett-
ger, 1878) [4: 53], A. c. liebetruti (Charpentier,
1852) [1: 18], A. senilis senilis (Rossmassler,
1836) [12: 140], A. s. flavescens (Boettger,
1878) [2: 27], A. s. kolpomyrtensis Kemper-
man & Gittenberger, 1990 [1: 15], A. adrianae
adrianae Gittenberger, 1979 [4: 52], A. a. du-
bia Gittenberger, 1979 [1: 19], A. jonica assi-
cola Kemperman & Gittenberger, 1990 [2: 36],
and A. j. jonica (Pfeiffer, 1866) [2: 16].

All specimens from Kephallinia and Ithaka
were collected by the first author in March
1988. See Appendix for detailed information
on the localities indicated in Figure 2.

As outgroups we used populations of A.
teres nordsiecki Zilch, 1977 [1: 8] and A. re-
beli Wagner, 1924 [1: 13] from Crete, and A/-
binaria spec. [1: 20] and /sabellaria edmundi
Gittenberger, 1987 [1: 20] from the Pelo-
ponnese (Parnon) (see Appendix). The out-
group material was collected by E. Gitten-
berger in April and June 1988.

In total 613 individuals are examined, an
average of 13 specimens per population.

Collecting and Transport

Populations were sampled at a location lim-
ited to 1 to 20 m°. The snails could be col-

ALLOZYME FREQUENCIES IN ALBINARIA

Albinaria species and subspecies

E contaminata MM senilis SS] jonica ZZZA adrianae
С =с. contaminata $ =s. senilis JA = j. assicola A =a. adrianae
I =c.incommoda F =s. flavescens JJ =}. jonica D =a. dubia
O =c. odysseus К =s. kolpomyrtensis
Г, =с. liebetruti
5 10
М =с. muraria о
km

FIG. 1. Map of Kephallinia and Ithaka, with distributions of Albinaria taxa.

35

36 KEMPERMAN & DEGENAARS

. contaminata contaminata
. c. incommoda

. с. odysseus

. с. liebetruti

. senilis senilis

. s. flavescens

. s. Kolpomyrtensis
. jonica jonica

. j. assicola

. adrianae adrianae
. a. dubia

binaria hybrids

O
O
e
|] -
о
о
A
A
%
0
x

Assos-Penins.742

Fiskardo

Stavros

Ormos Poleos

Cape Dichalia

FIG. 2. Map of Kephallinia and Ithaka, with collection sites of material used in allozyme studies.

lected easily, without any special device.
Forced into aestivation, they are not very de-
manding during transport. Plastic sandwich-
bags with some dry paper wipes proved to be
adequate for storage of several weeks.

Electrophoresis

Complete animals, including shells, were
ground in two drops of distilled water in ice-

cooled porcelain cups. Wicks (Whatman no. 3
filter paper; 2 x 5 mm) were used to apply the
homogenate on the gels. After 15 initial run-
ning minutes at approximately 16 V/cm, the
wicks were removed and the electric field was
raised to approximately 19 V/cm. Running
times depended on the buffer system that
was used (Table 1). The gel/buffer system
was ice-cooled. The complete set of enzyme
systems of a specimen was tested, usually in

ALLOZYME FREQUENCIES IN ALBINARIA

37

TABLE 1. Enzymatic systems and loci scored for electrophoretic analysis, with buffer systems employed

and stain references.

Enzymatic system (E.C.) Loci
Adenylate kinase (2.7.4.3) Ak
Aldehyde oxidase (1.2.3.1) Ao
Alkaline phosphatase (3.1.3.1.) Aph
Aspartate aminotransferase (2.6.1.1) Aat-1,2
a-Esterase (3.1.1.1) Est-1,2,3
Glucose-6-phosphate dehydr. (1.1.1.49) G6pd
Glucophosphate isomerase (5.3.1.9) Gpi
a-Glycerophosphate dehydrog. (1.1.1.8) a-Gpdh-1,2
Hexokinase (2.7.1.1) Hk
L-Iditol dehydrogenase (1.1.1.14) Idh
Isocitrate dehydrogenase (1.1.1.42) Isdh-1,2
Lactate dehydrogenase (1.1.1.27) Ldh
Leucine aminopeptidase (3.4.11.1) Lap
Malate dehydrogenase (1.1.1.37) Mah-1,2
Mannose phosphate isomerase (5.3.1.8) Mpi
NADH dehydrogenase (1.6.99.3) Nada-1,2
Phosphoglucomutase (5.4.2.2) Pgm
Phosphogluconate dehydrog. (1.1.1.43) Pgd
Superoxide dismutase (1.15.1.1) Sod-1,2
Xanthine dehydrogenase (1.2.1.37) Xdh

Migration Buffer* Stain reference
anodic TC 7 Shaw & Prasad (1970)
anodic Poulik Ayala et al. (1972)
anodic ANG 7 Shaw & Prasad (1970)
cathodic/anodic TEB9 Selander et al. (1971)
cathodic/anodic TC7 Shaw & Prasad (1970)
anodic TEB 9 Shaw & Prasad, mod.

Brewer (1970)

anodic Poulik Shaw & Prasad (1970)
anodic TEB 8 Shaw & Prasad (1970)
anodic TEB 8 Shaw & Prasad (1970)
anodic TEB 9 Shaw 8 Prasad (1970)
anodic MG 7 Shaw & Prasad (1970)
anodic TEB 8 Shaw & Prasad (1970)
anodic TEB 8 Shaw & Prasad (1970)
cathodic/anodic TC7 Shaw & Prasad (1970)
anodic TEB 9 Nichols et al. (1973)
anodic Poulik Menken (1980)

anodic Poulik Shaw & Prasad (1970)
anodic TEB 9 Shaw & Prasad (1970)
anodic TEB 8 Shaw & Prasad (1970)
anodic TEB 8 Shaw & Prasad (1970)

*) TC 7 = Tris—Citrate, pH 7, running time (rt) 2 h (Shaw & Prasad, 1970); Poulik = Poulik, gel pH 8.75-8.9 & electrode
pH 7.6-8.0, rt 3 h (Poulik, 1957); TEB 8 = Tris—EDTA—borate, pH 8, rt 3.5 h (Shaw 8 Prasad, 1970); TEB 9 =
Tris—EDTA—borate, pH 9, rt 4.5 h (Ayala et al., 1973). All at 350 V or 35 mA.

two days, for which the homogenate had to be
stored overnight at —80°C. For references
concerning the histochemical staining proce-
dures, see Table 1. All samples were scanned
for 28 loci from 20 enzyme systems. In Table
1 a list of the enzyme systems is given, with
the abbreviations used in the present article.
Because of the risk of a substantial decline in
detectability of the weaker enzyme systems,
we analyzed freshly prepared homogenates
only.

Data Analyses

Scoring of the loci resulted in a “single-in-
dividual” genotype dataset. Doubtful pairs of
bands were by convention scored as single
bands, and in case of a doubtful position of a
single band the alternative with the highest
frequency was accepted. Data analyses were
performed with the BIOSYS-1 computer pro-
gram, release 1.7 (Swofford & Selander,
1989). To assess the degree of genetic vari-
ability, we calculated per population and per
subspecies the mean observed heterozygos-
ity per locus (Ho), the mean heterozygosity
per locus based on Hardy-Weinberg expecta-
tions (He), the accompanying standard devi-
ations, as well as the percentage of polymor-

phic loci (Pp and Pt). Also allelic frequencies
were estimated per population and per sub-
species. In the latter case, populations with
morphologically determined hybrids have not
been taken into account. On the basis of the
allele frequencies, the genetic similarity coef-
ficients were computed for each pair of pop-
ulations or taxa, according to Nei’s (1972) ge-
netic distance (D) and Rogers’ (1972) genetic
distance (R). On the resulting genetic dis-
tance coefficient matrices, a UPGMA cluster
analyses (Sneath & Sokal, 1973) was per-
formed.

To achieve a phylogeny, we also carried
out a Hennigian analysis following Richard-
son et al. (1986: 337), Mooi (1989) and Em-
berton (1990). We used the computerpro-
gram HENNIG86 (Farris, 1988).

The ingroup is not polyphyletic, but consists
of a representative sample of subspecies of
the wide-spread species A. contaminata and
A. senilis and all the known subspecies of
the endemic species A. jonica and A. adrianae.
It is hypothesized that these species form a
monophyletic group and that both A. adrianae
and A. jonica have evolved on Kephallinia,
sharing a common ancestor with one of the
other species. Although our ingroup—all
the Albinaria species and subspecies from

38 KEMPERMAN 8 DEGENAARS

Kephallinia and Ithaka, except for A. contam-
inata muraria—does not consist of all the
known members of the clade, this does not
invalidate its use as a monophyletic group. In
principle one can never be sure whether all its
extinct and extant members are known. Many
valuable phylogenetic revisions have been
published in monographs dealing with incom-
plete ingroups due to geographical or practi-
cal restrictions. Our outgroup is composed of
not too closely related congeneric species
from relatively distant localities (Peloponnese
and Crete) and, in the case of /sabellaria ed-
mundi, a species of a genus of doubtful status
that is at least very closely related to Albinaria
(Gittenberger, 1987).

Classification on Shell Morphology
and Distribution

It should be emphasized that we primarily
classified the various taxa according to shell
morphology and distribution, following Nord-
sieck (1977) and Kemperman & Gittenberger
(1990). From the shell characters we mention
(1) the sculpture of the protoconch, (2) the
sculpture of the teleoconch (without cervix),
(3) the cervical sculpture, (4) the cervical
structures, (5) the microsculpture, (6) the
general shell shape, (7) the shell size, (8) the
apertural shape, (9) the shape of the apertural
lip, (10) the apertural attachment to the teleo-
conch, (11) the shape of the columellaris, (12)
the length of the parietalis, (13) the length of
the spiralis, (14) the presence of a paralellis,
(15) the shape of the palatalis, (16) the shape
of the lunella, and (17) the shape of the clau-
silium. These characters, which proved to
vary independently among the species and
subspecies, will be discussed in detail in
Kemperman (in prep. A). It turned out that A.
contaminata differs in eight independent char-
acters from A. senilis, in ten from A. jonica
and in eleven from A. adrianae. A. senilis dif-
fers in nine characters from A. jonica and in
four from A. adrianae, whereas A. jonica dif-
fers in seven characters from A. adrianae.
These morphological data, in combination
with the distributional patterns and the fact
that intermediate forms are an extreme minor-
ity found only in a few contact zones lead us
to consider these forms separate species.

RESULTS

Among all taxa, including the outgroups,
variation was found at 16 of the 28 loci (57%),

namely, Aat-1, Ak, Aph, Est-3, Gpi, Hk, Idh,
Lap, Ldh, Mdh-1, Mdh-2, Mpi, Nadd-1, Nadd-
2, Pgd, and Pgm. All the taxa, except for A. a.
dubia, of which only a single population could
be analysed, are polymorphic to some extent.
At up to 35.7% of the loci, with an average of
12.2% (+ 10.4%), variation has been demon-
strated (Table 2). The observed percentage
polymorphic loci per taxon (Pt) depends to
some extent on the number of scanned pop-
ulations. In a single population, only a part of
the specific allozyme variation is found.
Therefore, to estimate the amount of polymor-
phic loci in a certain taxon, a minimum num-
ber of populations must be studied. For A. c.
contaminata and A. s. senilis, we computed
the percentage polymorphic loci for one pop-
ulation, for two combined populations, and so
on, each time adding randomly specimens of
one more population. This procedure was re-
peated three times, starting and continuing
with different populations. The averaged val-
ues for the different numbers of populations
are indicated in Figure 3. From this figure it
becomes obvious that in A. s. senilis from four
populations on, the growth in variation is very
low. In A. c. contaminata, however, there is
still no important reduction in increase of vari-
ation after the addition of six populations.

The polymorphic loci per population (Pp)
varies between 0.0% and 17.9%, with an av-
erage of 7.5% (+ 4.9%). Two of sixteen pop-
ulations of A. contaminata and four of five
populations of A. adrianae proved to be in-
variable. The averaged values of Pp per
taxon vary between 0.0% for A. a. dubia and
10.7% for A. j. jonica.

In Figure 4, we plotted the Pp of the anal-
ysed populations of the various species
against the number of analysed individuals. It
turned out that there is only a limited relation
between the number of analysed individuals
and the Pp of that population (Table 2).

In all taxa that could be investigated, the
level of heterozygosity was low (Table 2).
With Ho = 0.030, A. s. senilis was the most
variable subspecies. No genetic variation was
found in the population of A. a. dubia, and
only very little was found in A. c. liebetruti (Ho
= 0.003), A. c. odysseus (Ho = 0.005) and
A. a. adrianae (Ho = 0.001). In all subspe-
cies, except for A. s. kolpomyrtensis, the
mean observed heterozygosity was below the
mean expected heterozygosity. The differ-
ences were often considerable, as in A. c.
liebetruti (Ho = 0.003 versus He = 0.017)
and A. c. odysseus (Ho = 0.005 versus He =

ALLOZYME FREQUENCIES IN ALBINARIA 39

TABLE 2. Percentages polymorphic loci per taxon and per population and the mean heterozygosity. For
abbreviations, see Results; “s” in an abbreviation means standard deviation. *: based on 702, 706-708,

710 and 712 only.

Taxon/loc. nr. P, pa Ps

contaminata 52m 9.2
713 1
715 Ute
716
721 10.
735

749

759

incommoda 14.3
723

727

772

Me
Liebetruti 3
odysseus ie
764

765

766

770
senilis
702

706

- oo, OW

4.4

8.0

NW WWONN NONONNF
ANO ano.

=> >
oo

SS 9.9

=
о
09
ae
WWW PPS

SO OI O Oo > WW

719 14.

flavescens 14.3
734 Be
751 10.
kolpomyrtensis 10.7 10.7
jonica 17.9 10.7
707 10.
708 10.
assicola 10.7

742 10.
744 Us
adrianae 356 0.9
725

NN NN

=

nN

Le)
wWooo
nooo

Albinaria sp.
Isabellaria

5
®
o
(4)
KW HN Y] Wo
ane 00

sP, Ho SH, Ho SH,
5.8 0.013 0.006 0.026 0.013
0.020 0.012 0.032 0.020
0.024 0.013 0.039 0.021
0.005 0.004 0.018 0.015
0.037 0.024 0.031 0.019
0.000 0.000 0.019 0.013
0.000 0.000 0.000 0.000
0.012 0.008 0.012 0.008
3.4 0.009 0.006 0.014 0.009
0.022 0.018 0.019 0.015
0.008 0.006 0.015 0.012
0.000 0.000 0.000 0.000
0.009 0.009 0.019 0.019
0.003 0.003 0.017 0.017
3.4 0.005 0.003 0.044 0.021
0.000 0.000 0.013 0.013
0.003 0.003 0.012 0.010
0.015 0.009 0.023 0.014
0.000 0.000 0.029 0.019
4.9 0.030 0.010 0.050 0.016
0.020 0.018 0.020 0.015
0.036 0.023 0.045 0.025
0.018 0.011 0.030 0.017
0.023 0.018 0.033 0.024
0.018 0.018 0.018 0.018
0.000 0.000 0.019 0.019
0.030 0.030 0.023 0.023
0.022 0.011 0.031 0.018
0.042 0.022 0.044 0.023
0.009 0.007 0.034 0.021
0.029 0.018 0.037 0.021
0.065 0.030 0.061 0.028
5.0 0.010 0.005 0.028 0.016
0.004 0.004 0.004 0.004
0.016 0.011 0.039 0.026
0.022 0.013 0.020 0.012
0.0 0.028 0.015 0.035 0.016
0.021 0.013 0.027 0.016
0.048 0.027 0.057 0.032
2.6 0.011 0.009 0.018 0.014
0.010 0.008 0.019 0.014
0.013 0.010 0.018 0.015
1.8 0.001 0.001 0.001 0.001
0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000
0.002 0.002 0.002 0.002
0.000 0.000 0.000 0.000
0.000 0.000 0.015 0.015
0.005 0.005 0.024 0.017
0.002 0.002 0.016 0.014
0.000 0.000 0.018 0.018

0.044). From the individual population sam-
ples in which Ho and He are not equal to zero
(Table 2), 13% have Ho > He, whereas 16%
have Ho = He. So, in 71% of the populations
there is a deficiency of heterozygosity.
Among the various taxa there is only little
allozymic differentiation (Table 3). All taxa

show the same allele at highest frequency for
all but nine loci, namely, Aat-1, Hk, Idh, Lap,
Ldh, Mdh-1, Mdh-2, Mpi and Pgm. Of these
loci, only Aat-1, Mdh-1 and Mdh-2 concern
Albinarias from Kephallinia or Ithaka. Although
Aat-1, Mdh-2 and Lap are highly polymorphic,
there is always a most common allele.

40 KEMPERMAN 8 DEGENAARS

a percentage polymorphic loci
+

0 1 2 3 а

5 6 7 8 9 10

number of populations

—— с. contaminata

trs senilis

FIG. 3. Allozyme variation in Albinaria subspecies at increasing number of populations. For explanation, see

Discussion.

Albinaria a. adrianae is monomorphic at the
Mdh-1 locus for an allele that is rare or absent
in the other taxa. Albinaria a. dubia is mono-
morphic at the Mdh-2 locus for an allele that is
not found in any other Kephallinian taxon, but
occurs in the Ithakian A. c. odysseus and in
the outgroup species, namely, Albinaria sp.
and /sabellaria edmundi from the Pelopon-
nese and A. teres nordsiecki and A. rebeli
from Crete. No other Kephallinian or Ithakian
taxa are monomorphic for rare alleles. The
outgroup species, however, have at the loci
Aat-1, Lap, Ldh, Mdh-1, Mdh-2, Mpi, Nadd-1,
Idh either unique alleles or alleles that they
share with other outgroup species.

Based on the estimated allelic frequencies,
coefficients of genetic distances (D & R) were
calculated for all pair-wise combinations of
populations as well as taxa (Table 4). The
results of the UPGMA cluster analyses, which
are based on the various matrices are given
in Figures 5 and 6, the population dendro-
grams, and Figures 7 and 8, the taxon den-
drograms. Intraspecific genetic distances in
the non-endemic species A. contaminata and
A. senilis on Kephallinia are very low,
whereas the differentiation among both the

subspecies of the endemic A. adrianae and A.
jonica is large enough to cluster them sepa-
rately with different groups of the non-
endemic species. The interspecific variation
of the non-endemic species is limited, but
large enough to enable a complete partition.
The genetic distance between the endemic
species is larger than that between the en-
demic and the non-endemic species.

There may be a correlation between ge-
netic distance and geographical isolation,
which is demonstrated by the genetic dis-
tance between the fully isolated subspecies of
both A. adrianae and A. jonica, as well as by
the allelic composition of the Ithakian A. c.
odysseus. The latter is genetically more dif-
ferentiated from the combined other Kephal-
linian taxa than these are from each other.
The distance between the Kephallinian-
Ithakian taxa and the Cretan-Peloponnesian
outgroup species also reflects the geograph-
ical isolation. However, among the outgroup
species, the differentiation between taxa from
Crete on the one hand and the Peloponnese
on the other hand does not strictly conform to
the geographical distances.

To test the goodness of fit of the trees we

ALLOZYME FREQUENCIES IN ALBINARIA 41

percentage polymorphic loci / population

0 5 10 15

20 25 30 35

number of individuals / population

—_——

contaminata

—+- senilis

—* jonica —&— adrianae

FIG. 4. Allozyme variation in Albinaria species within samples of different size. For explanation, see Dis-

cussion.

repeatedly calculated Neïs D and ran a
UPGMA cluster analysis while each time a
different subspecies was removed from the
data set. This procedure resulted in two types
of trees that differ only in the position of A. c.
odysseus and A. a. adrianae. When either A.
c. contaminata, À. c. incommoda or A. a. du-
bia was removed, a tree was obtained with A.
c. odysseus as the sister group of the com-
bined A. contaminata subspecies and A. j.
jonica. In case of the removal of A. c. contam-
inata and A. c. incommoda, A. a. adrianae
switched from the A. contaminata cluster to
the A. senilis cluster. However, when either A.
c. liebetruti, A. s. senilis, A. s. flavescens, A.
s. kolpomyrtensis, A. j. jonica, A. j. assicola or
A. a. dubia were removed the dendrogram
was identical to the one obtained when all
subspecies are included (Fig. 7). lt can be
concluded from this “jackknife” procedure
that the complete dendrogram is very stable
except for the position of A. c. odysseus and
A. a. adrianae.

To be used in a Hennigian analysis, Table 5
gives the characters of the various taxa, as
they are derived from the allele frequencies

(Table 3). There are 17 informative alleles.
The apomorphic character states are deter-
mined by outgroup comparison. Table 5
shows that a plesiomorphic state or states
have been proposed for 15 of the 16 polymor-
phic loci. Mdh-2 defines a large group to
which only A. a. dubia is not linked. Albinaria
a. dubia has no synapomorphies with any of
the other taxa. The goal of separating the taxa
into groups that are characterized by derived
character states turned out to be impractica-
ble. This means that a Hennigian analysis of
the entire group does not produce a resolved
phylogeny. The low number of synapomor-
phies distributed among the loci even troubles
partial analyses.

Some alleles are very locally distributed in
a certain species, sometimes exclusively
where its range comes close to that of an-
other species. This shared presence of alleles
may be considered then the result of intro-
gression. This applies to (1) the Nadd-1 c-
and d-allele in A. j. jonica at loc. 708, and in
the adjoining populations of A. s. senilis at
locs. 702 and 706; (2) Gpi alleles in A. s. se-
nilis at locs. 756 and 706, and A. c. contam-

42 KEMPERMAN 8 DEGENAARS

TABLE 3. Allele frequencies in populations for 16 polymorphic loci. N = number of individuals sampled
at the locus. For taxon abbreviations, see Figure 5.

sen sen sen jon jon
Locus 702 706 707 707 708

sen
708

sxj
709

Population

sen sen sen con con con sen cxs
709 710 712 713 715 716 719 721

Aph
(N) 8 6 12 5 2
a 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000
Est=3

Gpi
CN) 1 6 12 5 2
a 1.000 0.833 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000
с 0.000 0.083 0.000 0.000 0.000
а 0.000 0.000 0.000 0.000 0.000
е 0.000 0.083 0.000 0.000 0.000
f 0.000 0.000 0.000 0.000 0.000
g 0.000 0.000 0.000 0.000 0.000
Hk
(N) 6 3 1 3 1

Idh
CN) 5 9 14 6 2
a 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000

Lap
CN) 1 1 12 6 2
a 1.000 1.000 0.917 0.833 0.750
b 0.000 0.000 0.000 0.167 0.250
с 0.000 0.000 0.083 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000
e 0.000 0.000 0.000 0.000 0.000
f 0.000 0.000 0.000 0.000 0.000
g 0.000 0.000 0.000 0.000 0.000

Ldh
CN) 6 9 12 9 3
a 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000
а 0.000 0.000 0.000 0.000 0.000

inata at locs. 713, 715 and 716; (3) the c-
allele of the Ak locus in A. s. senilis at locs.
756 and 707 and in A. j. jonica at loc. 708; (4)
the b-allele of Aat-1 in A. c. liebetruti at loc.
746, in A. c. incommoda at loc. 773 and in all
Ithakian populations (Table 5). Therefore, a
Hennigian phylogenetic analysis has also
been performed on data in which the alleles
probably acquired by introgression were re-
moved. This, however, did not result in an
obvious improvement of the analyses.

2 10 16 20 24 17 12 5
1.000 1.000 1.000 0.975 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.025 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

1 1 6 9 24 5 11 5
1.000 1.000 1.000 1.000 1.000 1.000 0.909 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.091 0.000

1 8 10 17 18 14 12 5
1.000 1.000 1.000 0.676 0.667 0.750 0.917 1.000
0.000 0.000 0.000 0.029 0.028 0.179 0.000 0.000
0.000 0.000 0.000 0.059 0.056 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.056 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.083 0.000
0.000 0.000 0.000 0.059 0.056 0.000 0.000 0.000
0.000 0.000 0.000 0.176 0.139 0.071 0.000 0.000

1 10 16 9 7 13 1 1

1 10 10 21 16 13 12 5
0.000 0.500 1.000 1.000 0.906 0.962 0.958 1.000
1.000 0.500 0.000 0.000 0.000 0.038 0.042 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.094 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

2 10 6 24 25 10 12 5
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

(continued)

DISCUSSION
General

There is still much debate on the relevance
of various algorithms determining genetic dis-
tance coefficients and the combination of
these coefficients with clustering methods
(Nei et al., 1983; Buth, 1984; Richardson et
al., 1986; Emberton, 1988; Ferguson, 1988).
For a phenetic comparison of the allozyme

ALLOZYME FREQUENCIES IN ALBINARIA 43
TABLE 3. (Continued)
Mdh-1
(N) 6 9 14 8 3 К 4 1 9 6 19 29 1 17 7
а 0.000 0.000 0.000 1.000 1.000 0.000 1.000 0.000 0.000 0.167 1.000 0.931 1.000 0.000 0.857
b 0.750 1.000 1.000 0.000 0.000 1.000 0.000 1.000 1.000 0.583 0.000 0.017 0.000 1.000 0.143
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.017 0.000 0.000 0.000
е 0.250 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.083 0.000 0.000 0.000 0.000 0.000
+ 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.167 0.000 0.034 0.000 0.000 0.000
Mdh-2
(N) 6 9 14 8 3 if 4 1 9 6 20 29 2 17 7
a 1.000 0.889 1.000 0.063 0.000 0.786 0.250 1.000 1.000 1.000 1.000 0.897 1.000 0.676 1.000
b 0.000 0.111 0.000 0.000 0.000 0.214 0.000 0.000 0.000 0.000 0.000 0.103 0.000 0.324 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.000 0.000 0.000 0.938 1.000 0.000 0.750 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Mpi
(N) 1 6 12 6 2 4 3 2 10 1 23 15 14 12 5
a 1.000 1.000 0.958 1.000 1.000 1.000 1.000 0.750 1.000 1.000 1.000 1.000 1.000 1.000 0.700
b 0.000 0.000 0.042 0.000 0.000 0.000 0.000 0.250 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.300
а 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Nadd-1
(N) 14 9 14 8 3 7 4 1 9 16 18 26 22 17 7
a 0.929 0.611 1.000 1.000 0.667 1.000 1.000 1.000 1.000 1.000 0.917 1.000 1.000 1.000 0.929
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.036 0.167 0.000 0.000 0.167 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.036 0.111 0.000 0.000 0.167 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
е 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.083 0.000 0.000 0.000 0.000
+ 0.000 0.111 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
g 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.071
Nadd-2
(N) 14 9 14 8 3 7 4 1 9 15 15 23 22 17 if
a 1.000 1.000 1.000 0.813 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.188 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Pad
(N) 6 9 14 9 3 6 4 2 11 16 24 23 17 4 2
a 1.000 0.944 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.056 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Pgm
(N) 6 9 14 3 3 2 4 1 9 16 20 17 14 5 2
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.971 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.029 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Population
inc adr imc adr adr adr dub sxd sxd fla con kol ass ass lie
Locus 723 725 727 728 729 730 731 732 733 734 735 736 742 744 746
Aat-1
(N) 9 6 9 9 12 16 19 6 12 8 13 15 18 17 13
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.846 1.000 0.94 1.000 0.692
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.269
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.056 0.000 0.000
9 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
е 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.154 0.000 0.000 0.000 0.038
АК
(N) 9 6 9 3 3 6 8 4 11 8 8 6 1 1 16
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Aph
(N) 6 8 9 5 10 10 14 4 11 1 10 8 9 10 10
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(continued)

data, we have chosen the well known Nei’s
Genetic Distance (D). This index measures
the mean number of electrophoretically de-
tectable substitutions per locus, which accu-
mulated because the two populations di-
verged from a common ancestor (Buth,
1984). Nei’s D might also enable assumptions

concerning the period of time after the taxa
evolved from their most recent common an-
cestor, presuming that constant rates of evo-
lution occur within the lineages. No agree-
ment exists, however, on the translation of
genetic distances into time-spans (Rich-
ardson et al., 1986). After a comparison of

44 KEMPERMAN 8 DEGENAARS

TABLE 3. (Continued)

Est-3
(N) 9 8 9 1 10 10 6
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 .000 0.000 0.000 0.000 0.000 0.000
Gpi
(N) 6 6 9 3 13 16 8
a 1.000 1.000 1.000 1.000 1.000 0.969 1.000
b 0.000 0.000 0.000 0.000 0.000 0.031 0.000
c 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000
e 0.000 0.000 0.000 0.000 0.000 0.000 0.000
f 0.000 0.000 0.000 0.000 0.000 0.000 0.000
g 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Hk
(N) 6 6 9 3 12 16 2
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Idh
(N) 9 1 10 6 9 10 17
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Lap
(N) 6 1 8 1 9 10 2
a 0.750 1.000 0.813 1.000 1.000 1.000 1.000
b 0.250 0.000 0.000 0.000 0.000 0.000 0.000
c 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000
e 0.000 0.000 0.000 0.000 0.000 0.000 0.000
f 0.000 0.000 0.000 0.000 0.000 0.000 0.000
g 0.000 0.000 0.188 0.000 0.000 0.000 0.000
Ldh
(N) 9 6 9 9 12 6 19
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000
c 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Mdh-1
(N) 9 6 10 8 13 6 16
a 0.944 0.000 1.000 0.000 0.000 0.000 0.000
b 0.056 0.000 0.000 0.000 0.000 0.000 1.000
c 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 1.000 0.000 1.000 1.000 1.000 0.000
e 0.000 0.000 0.000 0.000 0.000 0.000 0.000
f 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Mdh-2
(N) 9 6 10 8 13 6 16
a 1.000 1.000 1.000 1.000 1.000 1.000 0.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 1.000
9 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Мрт
(N) 3 6 9 9 12 6 19
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Nadd- 1
(N) 9 14 10 8 13 16 16
a 1.000 1.000 0.950 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000
e 0.000 0.000 0.000 0.000 0.000 0.000 0.000
f 0.000 0.000 0.000 0.000 0.000 0.000 0.000
g 0.000 0.000 0.050 0.000 0.000 0.000 0.000

various combinations of genetic coefficients
and clustering methods, Nei et al. (1983) con-
cluded that Nei’s D in combination with an
UPGMA is the best choice for phylogenetic
analysis. However, a phylogenetic interpreta-
tion of the Neïs D dendrograms is seriously
affected by (a) the failure of Nei’s D to satisfy

4 11 1 10 8 16 18 18
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

4 11 1 10 8 10 10 10
1.000 0.909 1.000 1.000 0.938 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.091 0.000 0.000 0.063 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

1 1 8 13 15 18 8 12
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

8 14 10 13 15 9 10 16
1.000 1.000 0.950 1.000 0.867 1.000 1.000 1.000
0.000 0.000 0.050 0.000 0.133 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

6 1 1 10 9 10 10 10
0.917 1.000 1.000 1.000 0.889 1.000 1.000 1.000
0.083 0.000 0.000 0.000 0.111 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

6 12 1 13 9 17 8 18
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

6 13 10 13 14 9 8 8
0.000 0.000 0.000 1.000 0.000 0.000 0.000 1.000
0.917 1.000 1.000 0.000 1.000 1.000 1.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.083 0.000 0.000 0.000 0.000 0.000 0.000 0.000

6 13 10 13 14 9 8 8
0.750 0.846 1.000 0.846 1.000 0.222 0.250 1.000
0.250 0.154 0.000 0.154 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.778 0.750 0.000

6 12 1 10 9 9 10 10
0.333 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.667 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

6 13 10 13 6 18 18 18
1.000 0.500 1.000 1.000 1.000 0.972 0.944 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.028 0.000 0.000
0.000 0.500 0.000 0.000 0.000 0.000 0.056 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

(continued)

the triangle of equality, which may result
in biologically impossible negative branch-
lengths, and (b) the fact that the dendrograms
only show phenetic similarity or dissimilarity.
Rogers’ R (1972) meets the metrical require-
ments, which is therefore also applied in com-
bination with UPGMA cluster analyses.

TABLE 3. (Continued)

ALLOZYME FREQUENCIES IN ALBINARIA 45

Nadd-2
CN) 3 14 10
1.000 1.000 1.000
E 0.000

со
o
o
o
o
o
o
o
o

1

8

-000
0.

000

13
1.000
0.000

16
1.000
0.000

10
1.000
0.000
0.000

10
1.000
0.000
0.000

16
1.000
0.000

17
1.000
0.000
0.000

14
1.000
0.000
0.000

Gpi
CN) 12 10 10
a 1.000 1.000 1.000
b 0.000 0.000 0.000
c 0.000 0.000 0.000
d 0.000 0.000 0.000
e 0.000 0.000 0.000
f 0.000 0.000 0.000
g 0.000 0.000 0.000
Hk
(N) 4 17 12

Lap
CN) 11 10 9
a 1.000 0.950 0.833
b 0.000 0.000 0.167
с 0.000 0.000 0.000
d 0.000 0.000 0.000
e 0.000 0.000 0.000
f 0.000 0.050 0.000
g 0.000 0.000 0.000

O

oooo-_

oooo

ooOoooooo

0000000

The percentages of polymorphic loci per
population (Pp) (Table 2) are generally below
the values given by Nevo (1978: 126) for
other stylommatophoran gastropods. The av-
erage Pp of Kephallinian and Ithakian Albina-

6 13 10 13 14 18 18 6
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 000 0.000 0.000 0.000 0.000

2 14 2 13 9 17 18 18
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

2 13 2 13 8 18 18 12
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Population
ody ody inc inc ter reb Isa par
766 770 772 773
13 12 1 4 8 12 20 20
0.000 0.583 1.000 0.625 0.000 0.750 0.000 1.000
1.000 0.417 0.000 0.375 1.000 0.250 1.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
13 11 1 11 4 5 12 12
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
10 10 1 13 8 13 8 8
0.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
13 15 1 13 8 13 20 20
1.000 1.000 1.000 1.000 1.000 0.154 0.400 0.750
0.000 0.000 0.000 0.000 0.000 0.846 0.600 0.250
10 10 1 10 4 10 13 12
1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
10 10 1 1 4 10 1 1
1.000 1.000 1.000 1.000 0.000 1.000 1.000 1.000
0.000 0.000 0.000 0.000 0.750 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.250 0.000 0.000 0.000
13 15 4 11 4 10 12 12
1.000 1.000 1.000 1.000 1.000 0.000 1.000 0.000
0.000 0.000 0.000 0.000 0.000 1.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000
10 10 1 10 4 10 12 12
0.950 0.900 1.000 1.000 1.000 0.000 0.000 1.000
0.050 0.100 0.000 0.000 0.000 0.000 1.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 1.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
(continued)

ria taxa is 7.5% (+ 4.9%). Nevo listed mean
estimates of polymorphism which range from
6% to 100% with an average of 31.5% (+
26.6%) (excluding Rumina decollata). How-
ever, it is not clear from his data what is

46 KEMPERMAN 8 DEGENAARS

TABLE 3. (Continued)

Ldh
(N) 15 10 12 15 5 11 13 13 15 1 13 8 13 20 20
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.885 0.933 1.000 1.000 0.000 0.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 1.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.115 0.000 0.000 0.000 0.000 0.000 0.000 0.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.067 0.000 0.000 0.000 0.000 0.000 0.000
Mdh-1
(N) 16 7 13 15 6 5 13 13 5 4 13 8 3 20 18
a 1.000 0.000 0.000 0.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 0.000
b 0.000 1.000 1.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
c 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 1.000 1.000 1.000
d 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
e 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
if 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Mdh-2
(N) 16 7 13 15 6 5 13 13 >) 4 13 8 3 20 20
a 1.000 0.500 1.000 1.000 1.000 0.800 0.154 0.192 0.000 1.000 1.000 0.000 0.000 0.000 0.000
b 0.000 0.429 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.000 0.200 0.846 0.808 1.000 0.000 0.000 1.000 1.000 1.000 1.000
а 0.000 0.071 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Mpi
(N) 11 10 9 12 5 6 10 10 10 1 11 4 10 12 12
a 1.000 1.000 1.000 0.958 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.000 0.000 0.000 0.042 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Nadd-1
CN) 16 17 3 15 7 11 13 13 15 4 13 4 13 20 20
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000
с 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
е 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
+ 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
9 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Nadd-2
(N) 16 17 13 15 7 11 13 13 5 4 13 4 13 20 20
a 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Pad
CN) 15 10 12 15 5 11 13 13 15 4 13 8 13 20 20
a 1.000 1.000 1.000 0.500 0.900 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000
b 0.000 0.000 0.000 0.500 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
с 0.000 0.000 0.000 0.000 0.100 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Pam
CN) 16 10 13 3 4 6 10 10 10 4 13 8 13 20 20
1.000 1.000 0.769 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.025
b 0.000 0.000 0.192 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.975
c 0.000 0.000 0.038 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
meant by a population. According to our find- habitat preferences (Gittenberger, pers.

ings, the genetic variability depends on the
extent of the area which is considered to be
inhabited by a single population (Fig. 3). It is
uncertain, therefore, whether Nevo's P values
are comparable with our averaged Pp values
for each taxon.

Detailed results on mobility, dispersion and
homing in Albinaria are not yet available.
However, we have observed individuals mov-
ing on rockfaces for at least one meter during
a single rainy night. Baur (1988) found in bed-
rock-inhabiting Chondrina clienta (Wester-
lund), on the island of Oland, Sweden, an av-
erage distance of 291 cm was bridged during
six months, with a maximum of 814 cm. Ex-
cept for the more temporate climate to which
Chondrina clienta is confined, this species is
very well comparable with Albinaria species in

com.). In view of these data, and because our
samples are collected from an area limited to
1 to 20 m? of substratum, each sample is con-
sidered to belong to a panmictic population.
Genetic differences within a subspecies
proved to be present within distances less
than 200 m, for example in A. j. assicola at
locs. 744 and 742, in A. s. senilis and in A. j.
jonica at locs. 707 and 708, as well as in A. c.
contaminata at locs. 713, 715 and 716.
Nevo's (1978: 126) values for populations are
approximately of the same magnitude as
what we find within some whole subspecies
(Pt = of all populations together), namely,
32.1% for A. c. contaminata, 17.9% for A. c.
odysseus, but only 3.6% for A. c. liebetruti
and A. a. adrianae. In A. s. senilis, we looked
separately at the central west coast range

ALLOZYME FREQUENCIES IN ALBINARIA 47
TABLE 4. Matrix of genetic distance coefficients: above diagonal Roger's В, below diagonal Nei's D.
Subspecies 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

1 senilis ***** 0.028 0.026 0.058 0.053 0.062 0.115 0.087 0.046 0.058 0.056 0.187 0.220 0.165 0.235
2 flavescens 0.003 ***** 0.023 0.054 0.052 0.054 0.102 0.089 0.033 0.052 0.041 0.171 0.212 0.158 0.227
3 kolpomyrtensis 0.002 0.004 ***** 0.051 0.047 0.057 0.114 0.090 0.041 0.046 0.047 0.186 0.213 0.167 0.226
4 contaminata 0.037 0.039 0.038 ***** 0.015 0.021 0.077 0.058 0.071 0.046 0.079 0.181 0.221 0.169 0.228
5 incommoda 0.036 0.039 0.037 0.001 ***** 0.013 0.070 0.051 0.069 0.042 0.077 0.177 0.214 0.159 0.228
6 liebetruti 0.039 0.041 0.042 0.004 0.002 ***** 0.061 0.061 0.074 0.047 0.082 0.165 0.211 0.154 0.233
7 odysseus 0.085 0.083 0.093 0.051 0.049 0.037 ***** 0.088 0.108 0.107 0.088 0.128 0.211 0.116 0.239
8 jonica 0.071 0.069 0.075 0.035 0.036 0.040 0.062 ***** 0.059 0.087 0.088 0.191 0.224 0.167 0.234
9 assicola 0.022 0.017 0.023 0.058 0.059 0.062 0.091 0.041 ***** 0.066 0.035 0.172 0.215 0.162 0.219
10 adrianae 0.038 0.040 0.038 0.037 0.037 0.040 0.091 0.074 0.059 ***** 0.072 0.176 0.218 0.165 0.223
11 dubia 0.036 0.032 0.038 0.073 0.074 0.078 0.070 0.076 0.030 0.074 ***** 0.139 0.182 0.129 0.187
12 teres 0.184 0.179 0.194 0.189 0.186 0.170 0.114 0.194 0.183 0.190 0.147 ***** 0.161 0.125 0.255
13 rebeli 0.225 0.226 0.221 0.232 0.229 0.234 0.210 0.231 0.228 0.234 0.189 0.161 ***** 0.143 0.244
14 Isabellaria sp. 0.151 102158) 0.165 0.168 10-159 0.151 0.093 105159 0.164, 0.1717 0129 10.1227 0.142 ***** (05226
15 Albinaria sp. 0.242 0.241 0.242 0.243 0.245 0.250 0.244 0.245 0.236 0.244 0.199 0.284 0.264 0.249 *****

populations (locs. 702, 706-708, 710 and
712), which resulted in a Pt of 32.1%. Taking
all A. s. senilis material into account (thus in-
cluding 719, 752 and 756) leads to only a very
slight increase of Pt to 35%. The wide range
of the Pt values for the various taxa is to a
certain extent due to differences in the num-
ber of populations that could be investigated
(Table 2). A percentage of 3.6 in A. c. liebe-
truti is based on a single population. In fact,
this value cannot be compared with, for ex-
ample, 32.1% in A. s. senilis, because the
latter number is based on six populations. As
is obvious from Table 2, the range of variation
of species and subspecies is not reached in a
single population. This cannot be a conse-
quence of simply a larger number of speci-
mens that is studied when more populations
are investigated. Figure 4 shows that there is,
except for A. contaminata, only a small trend
towards an increasing P at populations with
higher numbers of analysed individuals. Ac-
cording to the results presented in Figure 3, a
minimal number of populations is needed to
reach a point where the addition of more pop-
ulations contributes relatively little to the ge-
netic variability. In case of A. s. senilis, this
number is five populations, whereas in case
of A. с. contaminata it is at least more than six
populations. This indicates that for A. c. con-
taminata the total variation has not yet been
established. It is concluded that there can be
important differences in genetic variability

throughout the range of Albinaria subspecies,
which recommends, at least in the case of
subspecies with large ranges, the use of min-
imally six populations. Because of the lack of
a sufficient amount of living material, we were
not able to investigate equal numbers of indi-
viduals per population and six or more popu-
lations for all taxa.

Nevo (1978) investigated the correlation
between several biological parameters and
the genetic variability observed in over 200
species of plants and animals. Some of his
parameters are briefly discussed here.

One might expect that the genetic variabil-
ity depends on the mode of reproduction. In
our material, there is a considerable defi-
ciency of heterozygosity in most populations
(Table 2). This pattern has also been found in
other hermaphroditic Mollusca (Selander 8
Kaufman, 1973; Nevo, 1978; Hillis et al.,
1987; Boato, 1988). A deficiency of heterozy-
gosity and a low proportion of polymorphic
loci are considered indicative of a breeding
system with incomplete panmixia. It has been
reported that a complete or near absence of
genetic variability within populations may oc-
cur in facultative self-fertilizing breeding sys-
tems, as is known, for example, for Rumina
decollata (see Selander & Kaufman, 1973),
Partula gibba (see Johnson et al., 1977) and
some Arion species (see McCracken & Se-
lander, 1980). The fact that most populations
of A. adrianae proved to be invariable (Table

48 KEMPERMAN 8 DEGENAARS

Distance

sen 702
sen 712
fla 734
kol 736
sen 707
sen 719
fla 751
sen 706
sen x dub 733
sen 708
sen 710
sen 756
sen x dub 732
ass 742
ass 744
sen 752
dub 731
sen 709
jon 707
sen x jon 709
jon 708
con 713
con 716
con 715
inc 723
inc 727
con 749
inc 772
con 759
con 735
lie 746
inc 773
con x sen 721
adr 725
adr 728
adr 729
adr 730
ody 764
ody 765
ody 770
ody 766

contaminata contaminata
c. incommonda

c. odysseus

c. liebetruti
senilis senilis
s. flavescens

s. kolpomyrtensis
jonica jonica
jonica assicola
adrianae adrianae
a. dubia

teres nordsiecki

=

o

1
AD AD E

. rebeli
Alb - Albinaria sp.
Isa - Isabellaria sp.

ter
| | Isa
reb

Alb

FIG. 5. Dendrogram of UPGMA cluster analysis on Neïs (1972) genetic distance coefficients. Cophenetic
correlation = 0.947.

ALLOZYME FREQUENCIES IN ALBINARIA 49

sen 702
sen 712
fla 734
kol 736
sen 710
sen 707A
sen 708
sen 719
fla 751
sen 706
sen x dub 733
dub 731
ass 742
ass 744
sen x dub 732
sen 756
sen752
sen 709
con 713
con 715
con 716
ING 727,
odk 749
inc 772
con/59
inc 723
con 735
lie 746
inc 773
con x sen 721
adr 725
adr 728
adr 729
adr 730

Distance

jon 707

sen x jon 709

jon 708
ody 764
ody 765
— ody 770
ody 766

ter
4 | Isa
reb

Alb

FIG. 6. Dendrogram of UPGMA cluster analysis on Rogers’ (1972) genetic distance coefficients. Cophenetic
correlation = 0.941. For legend, see Figure 5.

50 KEMPERMAN 3 DEGENAARS

Distance
40 33 327, 20
ee a fen ehe nl a ee en {= malen == fe = ==. pee
ehesten a НЕ ЕЕВС
40 33 27 20

5$. senilis

$. kolpomyrt.
$. flavescens
j. assicola

a. dubia

c. contaminata
c. incommoda
c. liebetruti
j. jonica

a. adrianae

c. odysseus
teres
Isabellaria sp.
rebeli

Albinaria sp.

FIG. 7. Dendrogram of UPGMA cluster analysis on Neïs (1972) genetic distance coefficients. Cophenetic

correlation = 0.960.

2) might be an indication for apomixis. How-
ever, there is no additional information that
could support this. The observations within,
for example, A. c. contaminata (namely, P =
17.9 in loc. 715 and P = 0.0 in loc. 749) show
that populations of a single Albinaria subspe-
cies may be strikingly different in genetic vari-
ability. Whether this is caused by multiple re-
productive systems or by one or more other
reasons, remains unknown. Because no in-
formation on the breeding system of Kephal-
linian and Ithakian Albinaria is present, it also

remains unclear whether the observed defi-
ciency of heterozygosity in the majority of the
sampled populations is caused by facultative
apomixis or by inbreeding. As has been
stated before, a sample is considered to be-
long to a panmictic population. Hence, it is
concluded that the Wahlund effect (Rich-
ardson et al., 1986) is probably not partly re-
sponsible for the paucity of the mean het-
erozygosity. It can also be suggested that the
convention of scoring doubtful pairs of bands
as single bands might have contributed to the

ALLOZYME FREQUENCIES IN ALBINARIA 51

Distance
40 ¿33 2 20
+----+----+----+----+----+----+----+--
Voss ae fe Pog POG pO pie
40 233 2 20

s. senilis

5. flavescens
5$. kolpomyrt.
Jj. assicola

a. dubia

c. contaminata
c. incommoda
c. liebetruti
a. adrianae

j. jonica

C. odysseus

teres

Isabellaria sp.
rebeli

Albinaria sp.

FIG. 8. Dendrogram of UPGMA cluster analysis on Rogers' (1972) genetic distance coefficients. Cophenetic

correlation = 0.963.

observed deficiency of heterozygotes. How-
ever, this cannot have influenced our values
substantially. Moreover, the effect is also re-
duced by the acceptance of the most frequent
alternative in case of a doubtful position of a
single homozygous band.

Most Albinaria habitats on Kephallinia and
Ithaka are seasonally subject to heat and
drought. Therefore, in summer at least, most
snails are in aestivation. Relatively long peri-
ods of severe drought and/or very high tem-
peratures might cause a high mortality, its ex-

tent depending on the local microclimate.
Especially in small populations, a high mor-
tality could result then in a bottleneck effect
and, therefore, a low genetic variability. Be-
cause we could not make reliable estimations
of population size, correlations between this
factor and allozyme variation cannot be inves-
tigated.

It is suggested (Nevo, 1978: 162) that a
relatively low genetic variability (low P value)
is characteristic for geographically restricted
habitat specialists, whereas a high variability

52 KEMPERMAN 8 DEGENAARS

TABLE 5. Character states for 16 polymorphic loci, showing the plesiomorphic states for the in-
group taxa. Synapomorphies are underlined, autapomorphies are in italics. For taxon abbreviations, see

Figure 5.

Taxon Locus as taxonomic character

Rat-1 AK Ah Est Gpi Hk Ich

Ingroup:

abcd fh

a a a
inc ab a a a a a a
a a a a
a a a

Outgroup:

ter b
reb ab
Isa b
Alb a

ооо ©
ооо ©
o
Ca
oo mo»
o
o © co ©

Plesiomorphic state:

ab a a a a a ab

is associated with more wide-spread habitat
generalists. This rule can be checked in our
material, consisting of two endemic habitat
specialists and two non-endemic, much more
wide-spread, less stenoecious species. In
populations of A. contaminata and A. senilis,
the mean P values vary between 3.6% and
10.7%, with an average of 8.0% for A. con-
taminata and 9.6% For A. senilis (Table 2). In
the endemic habitat specialist A. adrianae, we
found no polymorphic loci in one population of
A. a. dubia and three populations of A. a. adri-
anae; in a fourth population of the latter sub-
species, a P of 3.6% was found. Although
these differences are small, the data support
Nevo's rule. However, P values in the second
endemic species A. jonica contradict it. The
habitat specialist A. j. jonica, known only from
barren rock faces, has a mean P of 8.9%, and
the more catholic A. j. assicola has a mean P
of 10.7%, both moderately high values. Also,
the low P value of 3.6% for A. c. liebetruti,
which is considered a habitat generalist, does
not support Nevo’s suggestion. We do not
consider Nevo's rule falsified by our results,
however. It could equally well be argued that
there are only minor ecological differences in

Lap Ldh Mdh-1 Mdh-2 Mpi Nadd-1 Nadd-2 Pgd Pgm

abcd a abef ab abe acdf a ab abc

DO On»
oo 2
оао с
(ay С (ei С)
соо ©
соо ©
ооо о
о о @ =
w

abd a c ab a a a ab

the habitats of the various subspecies; they
are all represented in relatively warm, dry
limestone areas.

Apparently a certain genetic distance does
not reflect a divergence on either population,
subspecies, species or genus level. Gener-
ally, electrophoretic data alone are not suffi-
cient for final decisions on this matter (Gould
& Woodruff, 1978: 407; Gould & Woodruff,
1986: 464; Richardson et al., 1986: 308; Men-
ken & Ulenberg, 1987: 318). It may be impos-
sible to distinguish subspecies or even repro-
ductively isolated species on the basis of
biochemical characters (Avise, 1975; John-
son et al., 1977). Several authors list ranges
of Neïs D corresponding with various taxo-
nomic ranks (e.g., Davis et al., 1981; Menken
& Ulenberg, 1987; Thorpe, 1982). It becomes
obvious from these publications that the rank
indications for a certain group of taxa do not
necessarily hold for another group (Davis,
1984). The genetic distances found in Albina-
ría are very low, but, on the whole, the group-
ing of populations by UPGMA cluster analy-
ses of Nei’s D and Rogers’ Ris in congruence
with the results of the classical, morphological
approach.

ALLOZYME FREQUENCIES IN ALBINARIA 53

The amount of genetic divergence among
the Kephallinian and Ithakian Albinaria taxa
can be considered indicative of a recent radi-
ation of the group, assuming that a clear ge-
netic differentiation is secondary to the speci-
ation process. The occurrence of different
fixed alleles in several species, which are de-
fined on conchological and biogeographical
data, demonstrates that these taxa also have
individuality with regard to allozyme composi-
tion.

A Hennigian analysis of the allozyme data
did not result in a resolved phylogeny. The
results are contradictory and often inconsis-
tent with the conchological data (Kemperman,
in prep. A) and biogeographical patterns. It is
considered premature to discuss these differ-
ences in detail here.

There are several possible reasons for the
poor outcome of the Hennigian analyses.
Generally, the number of apomorphic charac-
ters states is too low to validate dichotomies
(Table 5). This number is even lowered when
alleles are removed from which the shared
presence is suspected to result from intro-
gression, rather than being based on apomor-
phy. Although, the resulting phylogeny should
become more realistic when the “pseudo”-
apomorphic characters based on introgres-
sion are removed, the results did not improve.
Another source of discrepancies might be the
use of characters based on very low allele
frequencies that are given the same weight as
characters based on high frequencies. Char-
acters based on low allele frequencies are
considered the least supportive, mostly be-
cause they are more error prone.

Albinaria spp. on Kephallinia and Ithaka

Albinaria contaminata: The subspecies A. c.
contaminata, A. c. incommoda, and A. c. lie-
betruti, cannot be recognized as separate en-
tities in the UPGMA cluster dendrograms of
Neïs D and Rogers’ В (Figs. 5, 6). Material
from locs. 727 (A. c. incommoda), 735, 749
and 759 (A. c. contaminata) and 772 (A. c.
incommoda), situated in various quarters of
the species range, is even identical according
to Neïs D dendrogram.

All phenetic analyses illustrate the relatively
high genetic distance between populations
from Ithaka on the one hand and those from
Kephallinia on the other hand. The cause of
this divergence is difficult to find out. Ithaka is
geologically older than Kephallinia (Hagn et

al., 1962: 185), but it has been unified with
Kephallinia several times in the past due to
sea-level oscillations. Nothing is known yet
about the early history of Albinaria in this re-
gion. The present situation on Kephallinia and
Ithaka is certainly affected by a series of sep-
arations and secondary contacts. It may be
questioned whether each geographical sepa-
ration resulted in genetic divergence and to
what extent the secondary contacts reestab-
lished a fully mixed gene-pool. The present
genetic divergence is at least the result of the
postglacial separation of minimally 12,000
years. lt cannot be excluded that it is the ef-
fect of an accumulation of smaller diver-
gences caused by the repetitive splitting and
fusion of ranges.

It remains a puzzling fact that, according to
both Nei’s and Rogers’ genetic distance co-
efficients, A. c. odysseus, from Ithaka, gets a
position in the UPGMA dendrograms (Figs.
5-8) opposite all Kephallinian Albinaria taxa,
thus including the not conspecific ones. This
implies that the amount of allozyme diver-
gence cannot be correlated with taxonomic
status, that is, with species boundaries. It
must be stressed here, however, that the po-
sition of A. c. odysseus is relatively unstable.
This is demonstrated by the jackknife proce-
dure, resulting in dendrograms in which A. c.
odysseus is grouped either among the other
A. contaminata subspecies or opposite the
combined Kephallinian taxa. Except for A. a.
adrianae in two cases, the subspecies remain
always in the same position.

An explanation for the large genetic dis-
tance between Ithakian and Kephallinian taxa
might be that, due to, for example, extensive
introgressive hybridization, a certain amount
of gene flow continues between all Kephallin-
ian taxa, opposing the genetic differentiation
among them. Although introgression among
Kephallinian taxa occurs, our biochemical
data are insufficient to demonstrate the level
of gene flow.

A population of A. contaminata sampled
near Fiskardo, northern Kephallinia (loc.
749), was identified as A. c. odysseus by
Rahle (1979: 215). Although the shells show
some resemblance to Ithakian A. c. odys-
seus, the allozyme data indicate no genetic
difference with other Kephallinian A. contam-
inata populations. The conchological similar-
ity might result from convergent evolution
only. However, because of the location of loc.
749, it can also be hypothesized that ancient
shipments between the harbour of Fiskardo

54 KEMPERMAN 8 DEGENAARS

and the opposite Mycenean harbour in the
Ormos Poleos on Ithaka caused human
transport of A. c. odysseus from Ithaka to
Kephallinia, followed by hybridization with the
local A. c. contaminata. If that is the case,
only some shell characters still witness the
introgression.

The range of A. c. liebetruti on Kephallinia
is very small, comprising probably less than
one km? along the coast east of Sami. This
area is situated close to the former land
bridge to Ithaka. Nevertheless, A. c. liebetruti
is not found on Ithaka. Maybe this is because
it could not penetrate an area occupied by
another conspecific taxon. lt is also possible
that this subspecies is only a relatively recent
human introduction on Kephallinia. If so, A. c.
liebetruti did not occur close to the Pleis-
tocene land bridge until recently. Most con-
chological characters of A. c. liebetruti from
Kephallinia are very similar to those of this
subspecies from Zakynthos, where it is widely
distributed (Ráhle, 1979; Kemperman, in
prep. A). This supports the second view.
Meanwhile, the results of the allozyme anal-
yses, especially with regard to the popula-
tions of locs. 746 and 773, and conchological
data suggest that there is introgressive hy-
bridization between A. c. liebetruti and A. c.
incommoda. Unfortunately, there are no allo-
zyme data for populations of the former sub-
species from Zakynthos.

There are no synapomorphies defining A.
contaminata as a monophyletic group (Table
5). Within A. contaminata, A. c. contaminata
shares a synapomorphic allele (d) of the
Aat-1 locus with A. c. odysseus and a syn-
apomorphic allele (e) of the same locus with
A. c. liebetruti; the shared presence of the
b-allele of the Aat-1 locus in A. c. incommoda,
A. c. liebetruti and A. c. odysseus is consid-
ered to be based on introgression, as dis-
cussed below. No other synapomorphies
were found within A. contaminata. With other
species, however, A. contaminata has sev-
eral apomorphies in common. The apomor-
phic b-allele of the Gpi locus is shared with A.
s. senilis and A. a. adrianae, whereas the
apomorphic alleles c and d of Gpi are exclu-
sively found in A. c. contaminata and A. s.
senilis. The presence of the Gpi alleles b, c
and d in A. c. contaminata and A. s. senilis is
further discussed below. The b-allele at Mah-
2 is the synapomorphic character state for A.
c. contaminata, A. s. senilis and A. s. fla-
vescens. Because these synapomorphies re-
sult in conflicting conclusions, they do not al-

low decisions on phylogenetic relations with
respect to A. contaminata.

Albinaria senilis: According to all distance co-
efficients, only minor genetic variation occurs
among the various populations of A. senilis.
There is no differentiation between the sub-
species. In both population dendrograms
(Figs. 5, 6), A. s. senilis from loc. 709 is the
sister group of the combined other A. senilis
samples. At locs. 708 and 709, A. s. senilis
lives truly sympatrically with A. j. jonica (locs.
707, 708). At those sites, some specimens
have been found that are morphologically in-
termediate between A. s. senilis and A. j. jon-
ica; allozyme analyses, however, group these
intermediate specimens with A. j. jonica (Fig.
5). The position in the UPGMA-dendrograms
of A. s. senilis of loc. 709, of which only two
specimens could be analysed, results from
the fixation of both the Mdh-1 b-allele and the
Lap b-allele. For Mdh-1 b, most other A. s.
senilis populations are also fixed; this allele is
absent in A. j. jonica. The Lap b-allele occurs
in some other populations of A. s. senilis, A. C.
contaminata and A. j. jonica. Although ac-
cording to the shell morphology this material
was convincingly identified as A. s. senilis, the
analysed enzyme systems indicate that the
specimens were hybrids between A. s. senilis
and A. j. jonica.

Within A. s. senilis two geographically iso-
lated populations, namely, loc. 752 on the Pa-
liki-Peninsula and loc. 756 from Kastro,
southeast of Argostoli, exhibit relatively high
genetic distances from the other A. senilis
populations. Both populations are not clearly
differentiated morphologically.

The results of the allozyme frequency anal-
yses did not support the taxonomic distinction
of A. s. kolpomyrtensis Kemperman 4 Gitten-
berger, 1990, which was based on concho-
logical and distributional data. In the dendro-
grams, this subspecies is closely linked to A.
s. flavescens (Figs. 5-8), which is the geo-
graphically adjoining subspecies.

There are no apomorphic character states
that point to A. senilis, $.1., as a monophyletic
group. This is not affected by limiting the anal-
yses to a subset of the outgroup species. The
apomorphic allele e of the Gpi locus is the
only derived character state defining a sub-
group within A. senilis, namely, the combined
A. s. senilis and A. s. kolpomyrtensis. All other
apomorphies are shared with one or more
other species, which results in conflicting phy-
logenetic trees. The apomorphic c-allele of

ALLOZYME FREQUENCIES IN ALBINARIA 55

Aat-1 defines A. s. senilis, A. s. flavescens
and A. j. assicola as a monophyletic group.
The c-allele of Ak groups A. s. senilis and A.
j. jonica together; however, the presence of
this allele in both taxa is considered to be the
result from introgression, as discussed below.
Both apomorphic alleles c and d of Nada-1
would point to the combined A. s. senilis, A. j.
jonica and A. j. assicola as a monophyletic
group, but the presence of these alleles in A.
s. senilis is regarded too as being due to in-
trogression. The cause of the distribution of
the “synapomorphic” alleles b, c and d of Gpi
in both A. s. senilis and A. c. contaminata is
further discussed below. The status of the b-
allele of Gpi with respect to A. a. adrianae and
A. s. senilis is unclear. The d-allele of Mdh-2
is shared by A. s. flavescens, A. j. jonica and
A. j. assicola, whereas the b-allele of Mdh-2
could be a synapomorphy for A. s. senilis, A.
s. flavescens and A. c. contaminata.

Albinaria jonica: The endemic A. jonica is di-
vided into two widely disjunct subspecies. Al-
binaria j. assicola occurs along the Kephallin-
ian northwest coast, including the Assos
Peninsula. Its range overlaps partly with that
of A. c. contaminata (Fig. 1), but the two spe-
cies remain separate in habitat. Albinaria j.
jonica is known from 3 to 4 km north of Ar-
gostoli, along the west coast of the main is-
land. It is found on the rocks along about 500
m of the coastal road, where it occurs in a
mixed population with A. s. senilis. Morpho-
logically intermediate specimens have occa-
sionally been found between A. j. assicola
and A. c. contaminata, as well as between A.
j. jonica and A. s. senilis.

With respect to both subspecies of A. jon-
ica, there is a striking contrast between the
results of morphological studies and those of
allozyme analyses. The taxa have been rec-
ognized as conspecific on the basis of con-
chological features (Kemperman & Gitten-
berger, 1990). Quite different from what is
found among subspecies of A. senilis and A.
contaminata, all UPGMA analyses demon-
Strate a relatively large genetic distance be-
tween A. j. assicola (locs. 742 and 744) and
A. j. jonica (locs. 707 and 708). Moreover, the
affinities between these and other taxa, as
indicated by the results of the UPGMA cluster
analyses, suggest that A. jonica is not a
monophyletic entity. Albinaria j. assicola is
grouped among A. senilis subgroups, al-
though it is partly sympatric not with this spe-
cies, but with A. contaminata. The position of

A. j. jonica is less constant. Nei’s D (Figs. 5, 7)
suggests that A. j. jonica and A. c. contami-
nata are sister groups; combined they are the
sister group of A. a. adrianae. After Rogers’ R
(Figs. 6, 8), A. j. jonica is the sister group of
the combined Kephallinian taxa.

Contrary to the results of the UPGMA anal-
yses is the outcome of a Hennigian analysis
defining A. jonica, s.l., as a monophyletic
group. At the Nadd-1 locus, c (present in locs.
708 and 742) and d (present in locs. 708 and
744) are synapomorphic alleles (Table 5); it is
assumed that the occurrence of both alleles in
A. s. senilis is the result of introgressive hy-
bridization. Another synapomorphy for A. jon-
ica would be the d-allele of Mdh-2, if one con-
siders its occurrence in A. s. flavescens (loc.
751 only) due to introgression. This could
then date back to times when the present
floor of the Gulf of Argostoli was dry land. The
c-allele of Ak in A. j. jonica, shared with A. s.
senilis (Table 5), will be discussed below. The
presence of the apomorphic c-allele of Aat-1
in A. j. assicola as well as in A. s. senilis and
A. s. flavescens, might result from convergent
evolution.

Albinaria adrianae: The endemic A. adrianae
is subdivided into two subspecies. This sub-
division, primarily based on shell characters,
is also strongly supported by the allozyme
data. The four analyzed samples of A. a. adri-
anae (locs. 725, 728, 729 and 730), proved to
be genetically very similar. According to
UPGMA population dendrograms based on
Neïs D and Rogers’ В (Figs. 5, 6), these pop-
ulations are even identical. Material from loc.
730, with a P value of 3.6%, differs only
slightly from the other three localities, which
were found to be monomorphic. Albinaria a.
adrianae has not a stable position within the
various dendrograms. According to Rogers'
R, A. a. adrianae and A. contaminata are sis-
ter groups. Corresponding to Neïs D, A. a.
adrianae is the sister group of the combined
A. contaminata and A. j. jonica. When Neïs D
is calculated while either A. c. contaminata or
A. c. incommoda are removed from the data-
set in a jackknife procedure, A. a. adrianae
switches from this position and becomes the
sister group of the combined A. senilis, A. j.
jonica and A. a. dubia. The second subspe-
cies, A. a. dubia, is clustered either as the
sister group of the combined A. senilis sub-
species and A. j. assicola (Nei's D) or as the
sister group of A. j. assicola alone (Rogers'
R). This results from the fixation of the Mdh-1

56 KEMPERMAN 8 DEGENAARS

allele b in A. a. dubia, as well as in A. j. assi-
cola, A. s. flavescens and A. s. kolpomyrten-
sis. However, no phylogenetic conclusions
can be drawn on this allele distribution as no
plesiomorphic character state can be deter-
mined for this locus (Table 5).

The allozyme analyses did not reveal one
or more synapomorphic character states
characterizing A. adrianae as a monophyletic
group (Table 5). Moreover, the absence of the
a-allele of MDH-2, which is a synapomorphy
for all subspecies but A. a. dubia, points to the
paraphyletic origin of A. a. adrianae and А. a.
dubia. With the b-allele of the Gpi locus, A. a.
adrianae has an apomorphic character state
in common with both A. s. senilis and A. c.
contaminata. Because the presence of this al-
lele in both A. s. senilis and A. c. contaminata
is considered to be the result of introgression
from A. c. contaminata to A. s. senilis, it is a
synapomorphic character state for A. c. con-
taminata and A. a. adrianae. Because in our
analyses A. a. dubia has no apomorphic char-
acter states at all, its phylogenetic position
remains puzzling.

Hybridization and Introgression: Among Albi-
naria species both hybridization and intro-
gression are known to occur (e.g. Fuchs 8
Käufel, 1936; Nordsieck, 1983; Gittenberger,
1991). For Kephallinian taxa, these phenom-
ena are reported by Ráhle (1979), Gitten-
berger (1979) and Nordsieck (1979), who
based their conclusions on morphological
characters. We found some biochemical data
pointing to introgression (Table 3).

(1) As mentioned above, the c- and d-al-
leles of the Nadd-1 locus are characteristic for
A. jonica, $.1.; in A. j. jonica they are present in
loc. 708. These alleles occur also in the ad-
joining A. s. senilis populations 702 and 706.
It is assumed, therefore, that their very local
presence in A. s. senilis results from intro-
gression.

(2) The c-allele of the Ak locus is only
known from A. s. senilis at locs. 756 and 707,
as well as from A. j. jonica at loc. 708. As
localities 707 and 708 are adjoining, intro-
gression of the Ak c-allele is hypothesized be-
tween A. s. senilis and A. j. jonica. Although
the distribution of this allele in A. s. senilis is
less local than it is in A. j. jonica, the direction
of the introgression is unclear.

(3) The b-allele of Aat-1 occurs in all Ith-
akian populations as well in A. c. liebetruti of
loc. 746 and A. c. incommoda of population

773. This is in roughly an area constituting
Ithaka and the central eastern part of Kephal-
linia. During the late Wúrm sea-level lowering,
Kephallinia and Ithaka were connected by a
land bridge located close to this area. There-
fore, introgression is a possible explanation.
Albinaria c. liebetruti might have received the
allele either during the late Wurm period di-
rectly from the same Ithakian source, or sec-
ondarily from local Kephallinian A. c. incom-
moda. Thus, whether or not A. с. liebetruti is a
recent invader in Kephallinia cannot be de-
cided on the basis of these data. For the pres-
ence of the Aat-1 b-allele in A. s. senilis of loc.
752, introgression cannot be hypothesized.
We have to accept either convergent evolu-
tion or indistinguishable, yet different electro-
morphs.

(4) The b-, c- and d-alleles of the Gpi locus
are found in the adjoining A. c. contaminata
populations at locs. 713 (alleles b and c), 715
(alleles b, c and d) and 716 (allele b). These
alleles are also found in the nearby A. s. se-
nilis populations at locs. 756 (alleles b, c, and
d) and 706 (allele b). Their local presence in
both species can be seen as the result of in-
trogression, but it can also be hypothesized
that the b-, c- and d-alleles of Gpi are exam-
ples of what Woodruff (1989: 282) called hy-
brizymes, “unexpected allelic electromorphs
associated with hybrid zones.” We found mor-
phologically intermediate specimens, indica-
tive of hybridization, in this area. Only the
presence of the Gpi-b allele in A. a. adrianae
of loc. 730 is not in agreement with the hy-
pothesis that hybrizymes are involved.

(5) Locally, both in A. c. contaminata pop-
ulation 715 and in A. s. senilis population 752,
the b-allele of the Pgm locus occurs. This can
hardly be the result of introgression, unless
ancestral populations once met during a gla-
cial sea-level lowering.

(6) At loc. 732, specimens morphologically
intermediate between A. s. senilis and A. a.
dubia are found, whereas at the nearby loc.
721, we collected specimens intermediate be-
tween A. s. senilis and A. c. contaminata.
Only at these two localities we found the c-
allele of the Mpi locus. Because loc. 732 is
situated in a hybrid zone, the Mpi c-allele
might be a hybrizyme (Woodruff, 1989). With
respect to the material of loc. 721, one might
postulate introgression between loc. 732 and
loc: 721:

(7) Some specimens of loc. 709 are con-
chologically intermediate between A. s. seni-
lis and A. j. jonica; the shells have the slen-

ALLOZYME FREQUENCIES IN ALBINARIA 57

derness of А. j. jonica, the apertural shape of
A. s. senilis, and an intermediate sculpture. In
contrast, the allozyme analyses point to a ge-
netic similarity with A. j. jonica (Figs. 5, 6).
This is, for example, indicated by the pres-
ence of the Mdh-2 d-allele, which might be
considered typical for A. j. jonica.

Apparently there are several localities on
Kephallinia and Ithaka where introgression
might explain the distribution of rare alleles
(1-4). As an alternative for introgression, the
presence of hybrizymes is suggested (4, 6).
Again, in all cases, one must remember the
possibility of convergent or parallel evolution
or the presence of electromorphs that are in-
distinguishable by our methods.

The Outgroup Species

All dendrograms (Figs. 5-8) show the
same topology for the species serving as out-
groups, namely, Albinaria sp. and /sabellaria
edmundi from the Peloponnesos, and A. teres
nordsiecki and A. rebeli from Crete. They are
linked outside the cluster formed by the spe-
cies from Kephallinia and Ithaka, with a ge-
netic distance varying between 0.18 and 0.41.
Albinaria t. nordsiecki is the sister group of
Isabellaria edmundi; the combination of A. t.
nordsiecki and |. edmundi is the sister group
of A. rebeli. This cluster is the sister group of
the combined Kephallinian and Ithakian A/bi-
naria taxa. The Albinaria sp. is grouped as the
sister group of all other taxa.

The grouping of /sabellaria edmundi within
a cluster of A/binaria species, with a genetic
distance far below the distance between a
Cretan and a Peloponnesian Albinaria spe-
cies, supports the view that /sabellaria sensu
auctt. might be of polyphyletic origin and
partly synonymous with Albinaria (Gitten-
berger, 1987: 79).

Electrophoretic Data, Morphoiogy
and Classification

In this paper, we do not discuss in detail the
results of morphological and morphometrical
analyses; these will be presented in a forth-
coming article. Nevertheless, some remarks
concerning electrophoretic data in relation to
morphology can be made already.

The genetic distance (D = 0.05 for Kephal-
linian and Ithakian taxa, and D < 0.284 when
including the outgroup-species) are very low

with regard to their status as separate species
(Menken & Ulenberg, 1987; Thorpe, 1982).
Apparently we are dealing with most closely
related “young” species, clearly character-
ized by their distribution and a series of inde-
pendent conchological characters, among
which some localized introgression still oc-
curs. This reminds one of what is described
for five Partula species by Johnson et al.
(1977: 122, 125), who found an D of 0.09.
Davis et al. (1981) mentioned 0.01 < D < 0.1
for the Elliptio complanata species group (Bi-
valvia: Unionidae). For Cerion species, the
following values are reported in the literature:
D = 0.015 (D < 0.056) (Gould & Woodruff,
1978: 407); D < 0.06 (Woodruff & Gould,
1980: 397); D + 0.05 (Gould & Woodruff,
1986: 464); and D < 0.01 (Gould & Woodruff,
1987: 343). Evidently, “taxonomic decisions
do not follow directly from the estimation of D”
(Gould 8 Woodruff, 1986: 466).

Preliminary results of allozyme analyses of
several Albinaria taxa from Crete are in line
with this statement. Although both the distri-
butional patterns and the conchological diver-
sification strongly remind one of what is re-
ported for the Kephallinian and Ithakian taxa
here, genetic distances between the Cretan
taxa were seven times larger (D = 0.35). This
is considered indicative of speciation events
that occurred further back in geological time
(Schilthuizen, pers. com., 1991).

For the Albinaria under study, the taxo-
nomic relationships, derived from morpholog-
ical and biogeographical data, are not in close
congruence with the patterns found in allo-
zyme variation. This becomes obvious sev-
eral times independently. Illustrative are the
widely separate positions of the subspecies of
both A. jonica and A. adrianae among differ-
ent clusters of A. contaminata and A. senilis,
according to the allozyme data. Despite the
fact that A. c. odysseus from Ithaka is only
slightly different morphologically from the
Kephallinian A. с. contaminata, the lthakian
subspecies has to be placed opposite the to-
tal group of four polytypic Albinaria species
represented in Kephallinia according to its al-
lozymes. In some cases, forms that are con-
sidered hybrid origin by morphological and
distributional data, cannot be recognized as
such electrophoretically.

Despite the puzzling facts that remain, the
electrophoretic analyses were an important
tool in the multidisciplinary approach towards
an understanding of the differentiation of
Kephallinian and Ithakian Albinaria.

58 KEMPERMAN 8 DEGENAARS

ACKNOWLEDGMENTS

We are indebted to Prof. Dr. E. Gitten-
berger, who initiated and supervised this
project and gave critical comments on the
various versions of this text. The second au-
thor wishes to express her gratitude to Dr. G.
M. Davis for teaching her the ins and outs of
allozyme analyses on molluscs at his labora-
tory at the Academy of Natural Sciences of
Philadelphia, USA; many thanks are due also
to C. Hesterman, who gave the daily technical
support there. At Amsterdam, we were guests
of Prof. Dr. S. B. J. Menken, who helped us
adapting and continuing our runs and also
commented on this paper. At Amsterdam,
daily technical support was given by W. van
Ginkel, Dr. J. W. Arntzen and Dr. W. M.
Scheepmaker, for which we are very grateful.
Dr. J. W. Arntzen also helped us with debug-
ging the data-set and introduced us to
BIOSYS. Dr. T. Backeljau of Brussels, Bel-
gium, is kindly acknowledged for his introduc-
tion to biochemical molluscan systematics.

LITERATURE CITED

ANDEL, T. H. VAN & J. C. SHACKLETON, 1982.
Late Paleolithic and Mesolithic coastlines of
Greece and the Aegean. Journal of Field Archae-
ology, 9: 445—454.

AVISE, J. C., 1974. Systematic value of electro-
phoretic data. Systematic Zoology, 23: 465-481.

AYALA, F. J., J. R. POWELL, M. L. TRACY, C. A.
MOURAO 8 $. PEREZ-SALAS, 1972. Enzyme
variability in the Drosophila willistoni group. IV.
Genic variation in natural populations of Droso-
phila willistoni. Genetics, 70: 113-139.

AYALA, F. J., D. HEDGECOCK, G. S. ZUMWAULT
& J. W. VALENTINE, 1973. Genetic variation of
Tridacna maxima, an ecological analog of some
unsuccessful evolutionary lineages. Evolution,
27: 117-191.

AYOUTANTI, A., C. KRIMBAS, S. TSAKAS & M.
MYLONAS, 1988. Genetic differentiation and
speciation in the Greek Archipelago: the genus
Albinaria. Biologia Gallo-Hellenica, 13: 155-160.

BAUR, B., 1988. Microgeographical variation in
shell size of the land snail Chondrina clienta. Bi-
ological Journal of the Linnean Society, 35: 247-
259.

BIRD, E. C. F., 1984. Coasts. An introduction to
coastal geomorphology. 3rd ed., Basil Blackwell,
New York, xv + 320 pp.

BOATO, A., 1988. Microevolution in Solatopupa
landsnails (Pulmonata: Chondrinidae): genetic
diversity and founder effects. Biological Journal
of the Linnean Society, 34: 327-348.

BREWER G. J., 1970. An introduction to isozyme

techniques. Academic Press, New York, xii +
186 pp.

BUTH, D. G., 1984. The application of electro-
phoretic data in systematic studies. Annual Re-
view of Ecology and Systematics, 15: 501-522.

DAVIS, G. M., 1984. Introduction to the second in-
ternational symposion on molluscan genetics.—
Malacologia, 25(2): 265-269.

DAVIS, G. M., W. H. HEARD, D. L. H. FULLER 4 C.
HESTERMAN, 1981. Molecular genetics and
speciation in Elliptio and its relationships to the
taxa of North American Unionidae (Bivalvia). Bi-
ological Journal of the Linnean Society, 15: 131—
150.

DEGENAARS, G. H. Allozyme variation analysis in
Albinaria (Gastropoda Pulmonata: Clausiliidae).
In: C. MEIER-BROOK, ed., Proceedings of the Tenth
International Malacological Congress, Tubingen,
1989. In press.

EMBERTON, K. C., 1988. The genitalic, allozymic,
and conchological evolution of the eastern north
american Triodopsinae (Gastropoda: Pulmonata:
Polygyridae). Malacologia, 21(1-2): 159-273.

EMBERTON, K. C., 1990. Acavid land snails of
Madagascar: subgeneric revision based on pub-
lished data (Gastropoda: Pulmonata: Stylom-
matophora). Proceedings of the Academy of Nat-
ural Sciences at Philadelphia, 142: 101-117.

FARRIS, J. S., 1988. Hennig86 reference, version
1.5. James S. Farris, Port Jefferson Station, New
York, 18 pp.

FERGUSON, A., 1988. Isozyme studies and their
interpretation. In: D. L. HAWKSWORTH, ed.,
Prospects in systematics. Systematics Associa-
tion, Spec. Vol. 36: 185-201.

FUCHS, A. 8 F. KAUFEL, 1936. Anatomische und
systematische Untersuchungen an Land- und
Süßwasserschnecken aus Griechenland und von
den Insel des Agáischen Meeres. Archiv fúr
Naturgeschichte (n.f.) 5(4): 541-662.

GITTENBERGER, E., 1979. Eine neue Albinaria
Art (Gastropoda, Clausiliidae) der Insel Kephal-
linia. Zoologische Mededelingen Leiden, 54(2):
83-85.

GITTENBERGER, E., 1987. Neue Taxa der soge-
nannten Gattung /sabellaria (Gastropoda Pulmo-
nata: Clausiliidae) vom Peloponnes. Basteria,
51: 79-84.

GITTENBERGER, E.. Diversification in Albinaria
(Gastropoda, Clausiliidae). п: С. MEIER-BROOXK,
ed., Proceedings of the Tenth International
Malacological Congress, Tübingen, 1989. In
press.

GITTENBERGER, E., 1991. What about non-
adaptive radiation? Biological Journal of the Lin-
nean Society, 43: 263-272.

GOULD, S. J. 8 D. S. WOODRUFF, 1978. Natural
history of Cerion VIII: Little Bahama Bank—a re-
vision based on genetics, morphometrics and
geographic distribution. Bulletin of the Museum
of Comparative Zoology, 148(8): 371-415.

GOULD, S. J. 8 D. S. WOODRUFF, 1986. System-
atics of Cerion on New Providence Island: A rad-

ALLOZYME FREQUENCIES IN ALBINARIA 59

ical revision. Bulletin of the American Museum of
Natural History, 182: 389—490.

GOULD, $. J. 4 D. $. WOODRUFF, 1987. System-
atics and levels of covariation in Cerion from the
Turks and Caicos Islands. Bulletin of the Museum
of Comparative Zoology, 151(6): 321-363.

HAGN, H., H. BERGMANN, B. BISCHOFF, K.
BRAUNE, G. DREMEL, G. HUG 8 W. OTT,
1968. Zur Neogen-Stratigraphie von Kephallinia
und Ithaka. Giornale di Geología, 2a, 35(2): 179—
188.

HILLIS, D. M., D. S. ROSENFELD 8 M. SANCHEZ,
1987. Allozymic variability and heterozygote de-
ficiency within and among morphologically poly-
morphic populations of Liguus fasciatus (Mol-
lusca: Pulmonata: Bulimulidae). American
Malacological Bulletin, 5(2): 153-157.

JOHNSON, M. S., B. CLARKE & J. J. MURRAY,
1977. Genetic variation and reproductive isola-
tion in Partula. Evolution, 31: 116-126.

KEMPERMAN, TH. C. M.. Systematics and evolu-
tionary history of the genus Albinaria (Gas-
tropoda: Pulmonata: Clausiliidae) on the lonian
islands of Kephallinia and Ithaka. Thesis, in prep-
aration A.

KEMPERMAN, TH. C. M.. Genitalia of Albinaria of
the lonian Islands Kephallinia and Ithaka. In
preparation B.

KEMPERMAN, TH. C. M. & E. GITTENBERGER,
1988. On morphology, function and taxonomic
importance of the shell ribs in Clausiliidae (Mol-
lusca: Gastropoda Pulmonata), with special ref-
erence to those in Albinaria. Basteria, 52: 77-
100.

KEMPERMAN, TH. C. M. 4 E. GITTENBERGER,
1990. Two new Albinaria subspecies (Gas-
tropoda: Pulmonata: Clausiliidae) from Kephal-
linia and Ithaka. Basteria, 54: 143-150.

KEMPERMAN, ТН. С. М. & С. D. E. POVEL 4 Р.Н.
Е. BOR. Differentiation in radulae of closely re-
lated pulmonates: a pilot study in Albinaria (Gas-
tropoda: Clausiliidae). In preparation.

MCCRACKEN, G. F. & R. K. SELANDER, 1980.
Self-fertilization and monogenic strains in natural
populations of terrestrial slugs. Proceedings of
the National Academy of Sciences of the USA,
77(1): 684—688.

MENKEN, S. B. J., 1980. Allozyme polymorphism
and the speciation process in small ermine moths
(Lepidoptera, Yponomeutidae). Thesis, Univer-
sity of Leiden.

MENKEN, S. B. J. 4 S. A. ULENBERG, 1987. Bio-
chemical characters in agricultural entomology.
Agricultural Zoology Reviews, 2: 305-360.

MOOI, R., 1989. The outgroup criterion revisited via
naked zones and alleles. Systematic Zoology,
38(3): 283-290.

MURPHY, R. W., J. W. SITES, ur., D. G. BUTH 4
С. Н. HAUFLER, 1990. Proteins |: Isozyme elec-
trophoresis. In: D. M. Hittis & С. MORITZ, eds.,
Molecular systematics. Sinauers Associates,
Inc., Sunderland, Massachusetts, pp. 45-126.

NEI, M., 1972. Genetic distance between popula-
tions. American Naturalist, 106: 283-292.

NEI, M., ЕР. TAJIMA 8 Y. TATENO, 1983. Accuracy
of estimated phylogenetic trees from molecular
data. Il. Gene frequency data. Journal of Molec-
ular Evolution, 19: 153-170.

NEVO, E., 1978. Genetic variation in natural popu-
lations: patterns and theory. Theoretical Popula-
tion Biology, 13: 121-177.

NEVO, E., 1988. Genetic diversity in nature, pat-
terns and theory. In: M. K. HECHT, B. WALLACE
8 G. T. PRANCE, eds., Evolutionary biology, vol.
18: 217-246. Plenum Press, New York.

NICHOLS, E. A., V. M. CHAPMAN 4 F. H. RUD-
DLE, 1973. Polymorphism and linkage for man-
nosephosphate isomerase in Mus musculus. Bio-
chemical Genetics, 8(1): 47-53.

NORDSIECK, H., 1977. Zur Anatomie und Syste-
matik der Clausilien, XVII. Taxonomische Revi-
sion des Genus Albinaria Vest. Archiv fúr Mol-
luskenkunde, 107(4/6): 285-307.

NORDSIECK, H., 1979. Eine neue Albinaria-art
von Kefallinia (Gastropoda: Clausiliidae). Archiv
für Molluskenkunde, 110(1/3): 63-66.

NORDSIECK, H., 1983. Neue Taxa rezenter euro-
páischer Clausilien, mit bemerkungen zur Bas-
tardierung bei Clausilien (Gastropoda: Clausilii-
dae). Archiv für Molluskenkunde, 114(4/6): 189—
Ze

POULIK, M. D., 1957. Starch gel electrophoresis in
a discontinuous system of buffers. Nature, 180:
1477-1479.

RÁHLE, W., 1979. Land- und Súfwassermollusken
von Kephallinia und Zakynthos (lonische Inseln).
Archiv fúr Molluskenkunde, 110(4/6): 199-224.

RICHARDSON, B. J., P. R. BAVERSTOCK & M.
ADAMS, 1986. Allozyme electrophoresis. A
handbook for animal systematics and population
studies. Academic Press, Sydney, xii + 410 pp.

ROGERS, J. S., 1972. Measures of genetic simi-
larity and genetic distance. Studies in Genetics,
7: 145-153.

SELANDER, R. K., M. H. SMITH, S. Y. YONG, W.
E. JOHNSON & F. B. GENTRY, 1971. Biochem-
ical polymorphism and systematics in the genus
Peromyscus. |. Variation in the old-field mouse
(Peromyscus polinotus). Studies in Genetics, 4:
49-90.

SELANDER, В. К. 8 D. W. KAUFMAN, 1973. Self-
fertilization and genetic population structure in a
colonizing land snail. Proceedings of the National
Academy of Sciences of the USA, 70(4): 1186-
1190.

SHAW, С. В. 8 В. PRASAD, 1970. Starch gel elec-
trophoresis, a compilation of recipes. Biochemi-
cal Genetics, 4: 297-320.

SNEATH, P. H. A. 8 R. R. SOKAL, 1973. Numerical
taxonomy. San Francisco, Freeman. xv + 573
Pp.

SWOFFORD, D. L. & R. B. SELANDER, 1989.
BIOSYS-1. A computer program for the analyses
of allelic variation in population genetics and bio-

60 KEMPERMAN 8 DEGENAARS

chemical systematics, release 1.7. Illinois Natural
History Survey, Champaign. v + 43 pp.

THORPE, J. P., 1982. The biological clock hypoth-
esis: Biochemical evolution, genetic differentia-
tion and systematics. Annual Reviews of Ecology
and Systematics, 13: 139-168.

WOODRUFF, D. S., 1989. Genetic anomalies as-
sociated with Cerion hybrid zones: the origin and
maintenance of new electromorphic variants
called hybrizymes. Biological Journal of the Lin-
nean Society, 36: 281-294.

WOODRUFF, D. S. & S. J. GOULD, 1980. Geo-
graphic differentiation and speciation in Cerion—
a preliminary discussion of patterns and pro-
cesses. Biological Journal of the Linnean
Society, 14: 389-416.

Revised Ms. received 18 September 1991

APPENDIX: LIST OF COLLECTION
LOCALITIES

Unless otherwise stated, the locality is on
Kephallinia island. Locality numbers corre-
spond with those on the map of Figure 2.

319 A. senilis senilis (Rossmássler)—Near junc-
tion to Moni Theotokou Sision (= 17.4 km
SE of Argostoli), on rock, 210 m alt.,
DH7017; 24-04-1987

702 A. senilis senilis (Rossmássler)—1 km E of
Argostoli, W of junction Angon—Sami/Poros,
on walls along road, 10 m alt., DH5626;
16-03-1988

706 А. senilis senilis (Rossmássler)—400 т М of
Ag. Konstantinos (= 1.5 km NE of Argostoli)
on rocks along road, 10 m alt., DH5627; 17-
03-1988

707 A. jonica jonica (L. Pfeiffer} —Akros Kokkinos
Vrachos, most western rocks along road,
(= 3.5 km N of Argostoli), 50 m alt.,
DH5429; 17-03-1988

707 A. senilis senilis (Rossmássler)—Akros Kok-
kinos Vrachos (= 3.5 km N of Argostoli),
most western rocks along road, 50 m alt.,
DH5429; 17-03-1988

708 А. senilis senilis (Rossmassler)—200 т N of
Akros Kokkinos Vrachos (= 3.7 km N of Ar-
gostoli); on rocks along road, 50 m alt., DH
5429; 17-03-1988

708 А. jonica jonica (L. Pfeiffer} —200 m N of Ak-
ros Kokkinos Vrachos (= 3.7 km N of Argos-
toli), on rocks along road, 50 m alt., DH5429;
17-03-1988

709 A. senilis senilis (Rossmässler)—500 m N of
Akrcs Kokkinos Vrachos (= 4 km N of Ar-
gostoli), along road on rocks, 55 m ait.
DH5429; 17-03-1988

710 A. senilis senilis (Rossmássler)—Spilia, $
exit village (= 1.7 km S of Argostoli), along

712

713

715

716

719

722

723

725

727

728

729

730

731

734

road on rocks covered with vegetation, 120
m alt., DH5524; 18-03-1988

A. senilis senilis (Rossmássler)—Lakythra,
50 m N of northern entrance (= 6.2 km SSE
of Argostoli), on rocks in field and on low
rock-face, W side of road, 230 m alt.,
DH5720; 18-03-1988

A. contaminata contaminata (Rossmáss-
ler)—Lakythra, 50 m N of northern entrance
(= 6.2 km SSE of Argostoli), on rocks in field
and on low rock-face, E side of road, 230 m
alt., DH5720; 18-03-1988

A. contaminata contaminata (Rossmáss-
ler)—0.5 km NW of Lakythra, in road-cleft on
hill-top (= 5.3 km SSE of Argostoli), W side
of road on rock-face, 240 m alt., DH5721;
18-03-1988

A. contaminata contaminata (Rossmáss-
ler)—0.5 km NW of Lakythra, in road-cleft on
hill-top (= 5.3 km SSE of Argostoli), E side
of road on rock-face, 240 m alt., DH5721;
18-03-1988

A. senilis senilis (Rossmássler)—Atsou-
pades, 200 m NW of junction for Arginia
(= 22.1 km ESE of Argostoli), on rocks
along road, 290 m alt., DH7416; 21-03-1988
A. adrianae dubia Gittenberger—W exit of
Markopoulon (= 24.1 km ESE of Argostoli),
on rocks along road, 285 m alt., DH7615;
21-03-1988

A. contaminata incommoda (Boettger)—300
m N of Ag. Georgios (= 24.3 km ESE of
Argostoli), on wet soft sand-‘stone’, 220 m
alt., DH7818; 21-03-1988

A. adrianae adrianae Gittenberger—Poros,
rock-faces N side of bridge in cleft (25.1 km
E of Argostoli), type loc., 10 m alt., DH8023;
21-03-1988

A. contaminata incommoda (Boettger)—180
m N of Poros-town bridge (= 25 km E of
Argostoli), on isolated rocks between
houses along coastal road, 1 m alt,
DH8023; 21-03-1988

A. adrianae adrianae Gittenberger—200 m
N of Poros-town bridge (= 25 km E of Ar-
gostoli), on isolated rocks between houses
along coastal road, 1 m alt., DH8023; 21-
03-1988

A. adrianae adrianae Gittenberger—Gorge
3.8 km SE of Poros (= 28.3 km ESE of Ar-
gostoli) on lower and higher rocks of gorge,
2-10 m alt., DH8220; 21-03-1988

A. adrianae adrianae Gittenberger—Poros,
N rocks along river E from bridge in cleft
(= 25.1 km E of Argostoli), 5-15 m alt.,
DH8023; 21-03-1988

A. adrianae dubia Gittenberger—2.5 km NW
of Ag.Georgios, type loc. (= 22.8 km ESE of
Argostoli) on high rock faces, 400 m alt.,
DH7618; 21-03-1988

A. senilis flavescens (Boettger)—1 km W of
Kourouklata (= 6.4 km N of Argostoli), on

735

736

742

744

746

749

751

752

756

759

ALLOZYME FREQUENCIES IN ALBINARIA 61

rocks along road, 140 m alt., DH5432; 22-
03-1988

A. contaminata contaminata (Rossmáss-
ler)—1 km NE of N exit of Angon (= 14 km
NNE of Argostoli); on rocks along road, 250
m alt., DH5641; 22-03-1988

А. $. kolpomyrtensis Kemperman 8 Gitten-
berger—2 km NE of N exit of Angon (= 15.3
km N of Argostoli), on rocks along road, 260
m alt., DH5641; 22-03-1988

A. j. assicola Kemperman & Gittenberger—
Assos Peninsula, rocks along lower part of
road to main entrance fortress (= 22.5 km
NNE of Argostoli), 2-5 m alt., DH5948; 22-
03-1988

A. j. assicola Kemperman & Gittenberger—
Assos Peninsula (= 22.5 km NNE of Argos-
toli), fortress and southern rocks, 100 m alt.,
DH5948; 22-03-1988

A. contaminata liebetruti (Boettger)—1 km
NE of Sami (= 17.2 km ENE of Argostoli),
on rocks W of Tomb-cave, 35 m alt.,
DH7035; 22-03-1988

A. contaminata (Rossmássler)—1 km S of
Fiskardo (= 30.8 km NNE of Argostoli), on
rocks along bay, 10 m alt., DH6356; 24-03-
1988

A. senilis flavescens (Boettger)—1.5 km S of
Livadi, = 1 кт N of Ag. Dimitrios (= 8.5 km
NW of Argostoli), on sand-stone wall along
road, 5 m alt., DH5033; 23-03-1988

A. senilis senilis (Rossmássler)—Lepada,
2.5 km S of Lixouri (= 4.5 km W of Argos-
toli), 100 m N of beach entrance on low
rocks along coast, 2 m alt., DH5125; 23-03-
1988

A. senilis senilis (Rossmassler)—Kastro Ag.
Georgios, inside fortress on stone walls
(= 7.0 km SE of Argostoli), 290 m alt.,
DH6021; 24-03-1988

A. contaminata contaminata (Rossmáss-
ler) —4 km NW of broadcasting plant on Me-
gas Oros (= Oros Aenos)(= 14.3 km E of
Argostoli), on rocks along road, 1260 m alt.
DH6723; 25-03-1988

764

765

766

770

772

773

A. contaminata odysseus (Boettger)—Ithaka
Isl., Perachori, in/on ruin Venetian Church in
lower part village (= 1.7 km S of Vathi), 170
m alt., DH7545; 26-03-1988

A. contaminata odysseus (Boettger)—Ithaka
Isl., 600 т $ of Agros, on rock face along
road-curve (= 4.5 km W of Vathi), 160 m
alt., DH7147; 26-03-1988

A. contaminata odysseus (Boettger)—Ithaka
Isl., W side of Poly-bay (= 11.2 km NW of
Vathi), on rocks and walls, 1-5 m alt.,
DH6855; 26-03-1988

A. contaminata odysseus (Boettger)—Ithaka
Isl., Kioni, most eastern port of town (= 9 km
N of Vathi), along coastal road on wall, 5 m
alt., DH7355; 26-03-1988

A. contaminata incommoda (Boettger)—2
km NW of Tzanata (= 21.8 km E of Argos-
toli), near bridge on sediment wall along old
road to Mon. Atrou, 130 m alt., DH7623; 27-
03-1988

A. contaminata incommoda (Boettger)—1.1
km E of Koulourata (= 16.9 km E of Argos-
toli), on high rock-faces along road, 420 m
alt., DH7128; 27-03-1988

Outgroup species, locality-numbers are not
on Figure 2.

1

A. teres nordsiecki Zilch—Greece, Crete,
Province of Lasithi, Kavousi ravine, 50 m alt;
28-04-1988

A. rebeli Wagner—Greece, Crete, Province
of Lasithi, NW of Orino, 620 m alt.; 6-04-
1988

A. nov. spec.—Greece, Peloponnesos,
Province of Arkadhia, E slope Parnon Moun-
tain at 8.5 km SW of Astros, shadowed rocks
in coniferious forest, 400 m alt., UTM FG43;
27-07-1988

Isabellaria edmundi Gittenberger—Greece,
Peloponnesos, Province of Arkadhia, high
rock-face along the forest road, 5 km (6 km
along the road) N of Kosmas, 975 m alt.,
UTM FG51; 27-07-1988

\
y
+

he

ni

MALACOLOGIA, 1992, 34(1-2): 63—73

EFFECT OF STARVATION AND HIBERNATION ON THE
FINE STRUCTURAL MORPHOLOGY OF DIGESTIVE GLAND CELLS
OF THE SNAIL HELIX LUCORUM

Vasilis K. Dimitriadis 4 Dimitris Hondros

School of Biology, Faculty of Sciences, Aristotle University of Thessaloniki,
Thessaloniki 540 06, Greece

ABSTRACT

Thirty seven days hibernation resulted in significant changes in the morphology of the diges-
tive gland epithelium of Helix lucorum compared to control animals: the number of the digestive
cells significantly decreased, but the number of the other cell types of digestive gland, calcium
and excretory cells significantly increased. The apical granules and the cisternae with dense
cores of the digestive cells significantly increased in number and size, while the calcium granules
of the calcium cells increased in number and showed more intense concentric rings compared
to those of the control animals. In the excretory cells, the large vacuoles significantly increased
in size, while in all the digestive gland cells the lipid inclusions decreased in number as hiber-
nation proceeded. In some cases, cytoplasmic material was extruded into the lumen, and in-

tercellular spaces appeared to be very dilated.
Forty days of starvation induced similar but less intense phenomena in the digestive gland of
Helix lucorum. The possible physiological function of the crop epithelial cells and the digestive

gland cells of Helix lucorum is discussed.

Key words: Helix lucorum, snail, digestive gland, fine structure, starvation, hibernation.

INTRODUCTION

Snails exposed to starvation or hibernation
respond by minimizing their energy require-
ments (Prosser, 1973). Carbohydrate metab-
olism is the principal energy source during hi-
bernation, whereas during starvation oxygen
uptake decreases sharply during the first day
and more slowly later (von Brand, 1931; von
Brand et al., 1948). However, little is known
about the biochemistry and the physiology, as
well as the fine structural morphology of
snails exposed to such conditions as hiberna-
tion and starvation. There is little information
about the fine structure of the gut and diges-
tive gland epithelium and other visceral tis-
sues of snails exposed to starvation and hi-
bernation (Sumner, 1965; Oxford & Fish,
1979; Janssen, 1985; Roldan, 1987). The
majority of the data available deals with the
fine structure and the physiology of the diges-
tive gland mainly under normal conditions
(see, for example, Sumner, 1965, 1966;
Owen, 1966; Walker, 1970; Bowen & Davies,
1971; Oxford & Fish, 1979; Pipe, 1986;
Roldan, 1987).

In the present study, the cells of the diges-
tive gland of Helix lucorum were studied with
the light and electron microscope under nor-

63

mal conditions. In addition, the morphology of
the digestive gland was studied in starving
and hibernating animals and was compared
with the morphology existing under normal
conditions.

MATERIALS AND METHODS

Adult snails of Helix lucorum (Gastropoda,
Pulmonata, Helicidae) were collected from
Edessa, northern Greece. The shell diameter
of the snails used in the present study varied
from 41 to 43 mm and the body weight from
about 19 to 21.5 g.

Snails, deprived of food, were kept in a cold
room at 4+1°C under a 9 h light (L): 15 h dark
(D) photoperiod for 8, 12, 22 and 37 days, in
order to be exposed to hibernating conditions
(Lazaridou-Dimitriadou & Saunders, 1986).
Control feeding snails were kept at 19+1°C
and 13L:11D photoperiod. Starved snails
were kept at 19+1°C and 13L:11D photope-
riod for 15, 25 and 40 days. Four animals from
each experimental group were dissected late
morning, and the digestive gland was fixed in
Karnovsky’s fixative (Karnovsky, 1965), post-
fixed in 2% osmium tetroxide, dehydrated and
embedded in Spurr’s resin. Sections were cut

64 DIMITRIADIS & HONDROS

FIG. 1. Drawing of the various cell types observed in the digestive gland epithelium of H. lucorum. CC,

calcium cell; DC, digestive cell; EC, excretory cell.

using a Reichert Om U3 ultramicrotome. The
sections were poststained with lead citrate
and uranyl acetate and examined under a Ja-
pan Electron Optics Laboratory Co. 100B
electron microscope operating at 80 KV. For
light microscopic observations, thick sections
were stained with 1% toluidine blue.
Morphometric evaluation was performed
according to Weibel (1979) and Steer (1981).
Samples of five cells were counted from every
digestive gland section of four different snails.
The volume densities of certain morphometric
parameters of crop and digestive gland or-
ganelles were determined from point counting
stereology, using a test square lattice with a
period d = 10 mm, equivalent to 1 um on the
specimen. A minimum of 480 points were
counted per cell. The mean absolute volume
of the various cells were estimated using the
mean absolute volumes of the nuclei (using,
depending of the cell type, the formula for a
spheroid of prolate spheroid) and the ratio of
nuclear/cytoplasmic volume. The mean total
volume per cell of the various cell compo-
nents were determined by their volume den-
sities and the absolute volumes of the cor-

responding cells. The percentages of the var-
ious epithelial cell types within the digestive
gland epithelium were determined from mea-
surements on light microscopic micrographs
at a final magnification of x 500. Cells were
identified on the basis of morphologic criteria
given in the Results section. Mean values and
standard deviations of the morphometric pa-
rameters were calculated and statistically
compared using Student's t-test, significant
level P < 0.05.

RESULTS

Morphology of the Digestive
Gland Epithelium

The digestive gland epithelium of Helix lu-
corum consisted of three cell types: digestive
cells, calcium and excretory cells (Fig. 1). In
cross sections of the gland tubules, the cells
were located around a central lumen, and the
whole structure was surrounded by connec-
tive tissue, muscular layers and a sys-

STARVATION IN HELIX LUCORUM 65

tem of haemocoelic spaces surrounded by
amoebocytes.

Digestive cells were the most frequent cell
type found in the digestive gland (Fig. 6).
They usually appeared in a columnar shape,
varied in size, and usually prossessed mi-
crovilli (Fig. 2). Digestive cells were charac-
terized by numerous granules and cisternae
of varying size and electron density. In the
apical portion of these cells, there was usually
a number of granules filled with a homoge-
neous material of a low electron density (Figs.
2, 3). In the middle region of the cells, smaller
cisternae were also observed containing two
or more intracisternal cores (Figs. 2, 3). At the
base of the digestive cells, a few lipid inclu-
sions were concentrated in clusters (Fig. 4).
The nuclei of the digestive cells were usually
located in the basal or in the middle regions of
the cell and were usually well defined.

Calcium cells were the second cell type
found in the digestive gland epithelium of H.
lucorum. Their number was smaller than that
of the digestive cells (Fig. 6). They were usu-
ally pyramidal in shape and did not usually
protrude into the lumen, but they were located
at the base of the epithelium. Their height was
less than that of the digestive cells (Fig. 6).
Calcium cells were usually filled with calcium
granules which were located mainly in the
basal region of the cells. Calcium granules
were spherical in shape (Fig. 5) and were of a
similar size throughout each cell. In high mag-
nification, the calcium granules showed inner
concentric rings. The nuclei of the calcium
cells were usually located in their basal re-
gion, where lipid inclusions were often ob-
served.

Excretory cells were the third cell type
found in the digestive gland epithelium of H.
lucorum (Figs. 12, 14). Their size was larger
than that of digestive cells, their shape varied
between ovoid and columnar, and their cyto-
plasm was characterized by one or more
large cisternae containing cores or amor-
phous mass of high electron density. These
cisternae occupied a large portion of the cell
volume (Fig. 12). The nucleus was usually po-
sitioned at the basal region of the excretory
cells and in their apical portion small vacuoles
of low electron density were observed.

Morphology of the Digestive Gland
Epithelium under Starving Conditions

Snails exposed to starving conditions
showed modified morphology of the digestive

cells compared to the control animals. After
40 days of starvation, a significant decrease
in the number of these cells were noticed (Fig.
6). A significant increase in the total volume
per cell of the apical granules located in the
digestive cells compared to the control ani-
mals was also noticed in the same period
(Figs. 7, 8). In contrast, the cisternae with the
electron dense cores did not change signifi-
cantly their total volume per cell after 40 days
of starvation compared to the control animals
(Fig. 7). Lipid inclusions decreased in number
and almost disappeared after 40 days of star-
vation compared to control animals, while
calcium granules increased their number and
consequently their total volume per cell (Figs.
6, 7). In high magnification, the calcium gran-
ules also showed more concentric rings in
starved snails (Fig. 11). However, the diame-
ter of the calcium granules was similar in both
starved and control animals.

Excretory cells significantly increased in
number and volume after 40 days of starva-
tion compared to the control animals (Figs. 6,
9). Moreover, the large cisternae of these
cells significantly increased in size (Figs. 7, 9)
so that they occupied over than half of the cell
volume. Generally, the Golgi complexes and
the rough endoplasmic reticulum of these
cells were quite distinguishable and well or-
ganised in this stage.

After 25 and 40 days of starvation, extru-
sions of cytoplasm of whole cells into the
gland lumen were apparent. Degeneration
phenomena were rarely observed, while no of
autophagic vacuoles were visible. The mem-
brane infoldings appeared more intense,
mainly in the basal region, and the intercellu-
lar spaces were more dilated than that of the
control animals.

Morphology of the Digestive Gland
Epithelium under Hibernating Conditions

After eight days of dormancy, the digestive
gland cells did not show significant morpho-
logical changes. The only change noticed
concerned the calcium cells, which showed
more electron-dense concentric rings com-
pared to those in control animals.

Thirty seven days exposure of H. lucorum
in hibernating conditions induced fine struc-
tural modification of digestive gland cells. The
digestive cells significantly increased in size,
while their number decreased by about 25%
(Fig. 6) compared to the control animals. The
total volume per digestive cell of the apical

DIMITRIADIS & HONDROS

66

A

a

>

= A €
$ 733
vo © Е
© осо
СВОЕ ©
5 538
¡CDS
2 о. =
ao || M
5 E
2 2330
Um=
E MEAR
о =o
o =%ou
О Е
o ое
—~ £20
> BED
O Sac
— MES
SB of?
FS tS Gh
Ф + o
D ЭР
So
- ©
9 GES.
6 28,1:
Е
esa 52 |
£s2E589
de, O Y co
AS .
u (#2)
zog
853037
ONIS
ECSUO JE
DS DID
DOC.230
= += J
2-4. oz
© £ © do :-
U>09928
Еее
Ee (ew oe
Eee
Se
асе =
ORE cease
IDO: Scenes
O6 à D 6 o 2
о 0. =
DO EDS.
оба
A
O0 OT 230
BOIEZG
NE
UE DONS
990506
шошош Е

. Bl, basal

ic rings

inner concentr

granules (CGs) with

FIG. 5. Calcium cells contain numerous calcium

infoldings; N, nucleus. Bar = 3 um.

STARVATION IN HELIX LUCORUM 67

150

0.8

0.6

0.4

Percentage of cells

0.2

0.0

10000

8000

6000

4000

Absolute volume of cells (1m3)

г. Control

Starved
Hibernated

Digestive cells Calcium cells Excretory cells

FIG. 6. Distribution (%) (A) and absolute volumes (B) of the various cell of control, starved (40 days) or
hibernated animals (37 days) observed in the digestive gland of H. lucorum. Significantly different values (P
< 0.05) among control and starved or control and hibernated animals are indicated by asterisks.

granules, as well as of the cisternae with the
electron dense cores, significantly increased
compared to the control animals (Fig. 7). Cal-
cium cells significantly increased in number
and slightly in size at that period (Fig. 6). As
hibernation proceeded, calcium granules in-
creased in number (Fig. 7) and appeared in
clusters. The concentric rings in the calcium
granules also had a higher electron density

and increased in number in relation to the
control animals (Fig. 11).

Another morphological effect of hibernation
was a significant increase in the absolute vol-
umes of the excretory cells compared to the
control animals (Fig. 6). After 37 days of hi-
bernation, the large cisternae of the latter
cells also greatly increased in size and often
occupied more than half of their cell volume

68 DIMITRIADIS 8 HONDROS

10000

8000

6000

4000

2000

Total volume per сей (um3)

Apical granules

digestive cells

Cisternae with Granules of

Я Control

ZA Starved

Hibernated

Large cisternae
of digestive cells dense cores of calcium cells of excretory

cells

FIG. 7. Total volume per cell of cell components of digestive, calcium and excretory cells of control, starved
(40 days) or hibernated cells (37 days) of digestive gland of H. lucorum. Significantly different values (P <
0.05) among control and starved or control and hibernated animals are indicated by asterisks.

(Figs. 7, 12, 13, 15). In all the cells studied,
the number of lipid inclusions observed at the
basal region greatly decreased 37 days after
the exposure of the animals to hibernation,
while the extracellular spaces formed by the
infoldings of the basal plasma membrane and
the interceilular spaces appeared very dilated
in relation to the cuntrol animals. Cytoplasmic
regions often extruded into the lumen. In ad-
dition, whole cells appeared to have degen-
erated and been released into the lumen (Fig.
10).

DISCUSSION

There are numerous studies on the mor-
phology and physiology of epithelial cells in
the digestive gland of the Pulmonata (Sum-
ner, 1965, 1966; Walker, 1970; Bowen 8 Dav-
ies, 1971; Oxford & Fish, 1979) that are con-
sistent with the fact that this, the largest
organ, participates in digestion, absorption
and storage of food material previously di-
gested in the crop and stomach lumen. How-
ever, there are conflicting reports about the
cell types that constitute the digestive gland
epithelium of Pulmonata. Thus, Krijgsman
(1925, 1928) considered digestive and excre-
tory cells as different cell types. David €
Gótze (1963) distinguished one category of
digestive calcium cells, whereas Billett & Mc-

Gee-Russell (1955) considered calcium and
excretory cells as distinct forms of digestive
cells. Other studies supported the existence
of three types of digestive gland cells in slugs:
digestive, calcium and thin cells (Abolins-Kro-
gis, 1961; Morton, 1979), considering excre-
tory cells as digestive cells.

After 40 days of starvation or 37 days of
hibernation, there was an increase in the vol-
ume of the digestive cells, as well as an in-
crease in the total volume per cell of the api-
cal granules in the digestive cells. This fact is
consistent with the results of Sumner (1965),
who reported that starvation resulted in vac-
uolisation of the digestive cells of Helix as-
persa, as well as in an increase in size of the
green granules compared to the control ani-
mals. Similar granules to the apical granules
of digestive cells of H. lucorum were reported
in the digestive glands of Arion hortensis, Lit-
torina littorea and Cepaea nemoralis (Bowen
8 Davies, 1971; Oxford & Fish, 1979; Pipe,
1986) and were characterized as hetero-
phagic, phagolysosomal or heterolysosomal
cisternae. On the contrary, Krijgsman (1925,
1928), Walker (1970) and Sumner (1965), in
Helix pomatia and Agriolimax reticulatus,
referred to similar granules as green vacuoles
due to their staining with light green. There
are also histochemical results in other snails
indicating that these cisternae display a pos-
itive PAS (Walker, 1969), acid phosphatase,

STARVATION IN HELIX LUCORUM 69

FIG. 8. 40 days' starvation. Apical portion of digestive cells possessing large number of apical granules
(DV,). DV,, cisternae with dense cores; Lu, lumen; Mv, microvilli; PM, plasma membrane. Bar = 3 um.
FIG. 9. 40 days' starvation. Excretory cells display many large cisternae (EV) with large electron dense
cores. Mv, microvilli; PM, plasma membrane; CC, calcium cell. Bar = 5 um.

FIG. 10. 22 days' hibernation. Part of the cytoplasm of a digestive gland cell shows degenerating phenom-
ena. Mi, mitochondria; G, granules. Bar = 3 um.

FIG. 11. 37 days’ hibernation. Calcium granules containing two or more inner concentric rings (CRs). Bar =
1 um.

70 DIMITRIADIS 8 HONDROS

FIGS. 12, 13. 37 days' hibernation. Part of large cisternae (EV) with large electron-dense cores occupying
great cytoplasmic areas of excretory cells. CGs, calcium granules; DV,, apical granules; DV,, cisternae with
dense cores; Li, lipid inclusions; Mv, microvilli. Fig. 12 Bar = 5 рт. Fig. 13 Bar = 6 pm.

FIGS. 14, 15. Light microscopic view of digestive gland epithelium of control (Fig. 14) and hibernating
animals (Fig. 15). Many excretory cells (EC) and calcium cells (CC) are observed in the epithelium of the

hibernating snails, compared to the control ones. DC, digestive cells; Lu, lumen. Fig. 14 Bar = 20 pm. Fig.
15 Bar = 20 pm.

STARVATION IN HELIX LUCORUM 71

esterase (Bowen, 1970; Bowen & Davies,
1971) and B-glucuronidase reactions (Billett
& McGee-Russell, 1955). This fact probably
indicates that elaboration of material with cer-
tain enzymes inside these cisternae occurs.
After 37 days of hibernation, the total vol-
ume per digestive cell of the cisternae with
dense cores, which probably correspond to
the green granules of Helix aspersa, was
found to have increased. Similar cisternae
have been reported in the equivalent cells of
H. aspersa, where they were characterized as
yellow granules and exhibited positive reac-
tions for calcium and lipofuscin, as well as
positive PAS activity (Sumner, 1965). Also,
Pipe (1986) observed similar cisternae in the
digestive cells of Littorina littorea and sug-
gested that they probably were cut off from
heterolysosomal vesicles and filled with
waste material. Sumner (1965) also sup-
ported the view that similar cisternae could
contain remnants of intracellular digestion
that must be extruded to the central lumen.
From the above-mentioned reports it could
be concluded that the apical granules and cis-
ternae with dense cores in digestive gland
cells are structures involved in endocytotic
processes. There are reports supporting the
existence of phagocytosis in the digestive
glands of most Gastropoda (Owen, 1966; Ox-
ford & Fish, 1979). Owen (1966) claimed that
the existence of phagocytosis in certain spe-
cies of Mollusca may be an adaptation to
functional needs. In the present study, the
fine structural characteristics of digestive cells
of H. lucorum support the view that these cells
are responsible for absorption and digestion
of food material, as well as production and
secretion of digestive enzymes. However,
phagocytosis of solid food material by the
cells examined was not observed. It is possi-
ble that nutrients are absorbed by small pi-
nocytotic vesicles forming the apical granules
and consequently are degraded by lysosomal
action, producing the cisternae with dense
cores. Thus, the increase of the total volume
per cell of the apical granules and of the cis-
ternae with dense cores could probably be
attributed to the accumulation of residual in-
digestible material inside the digestive cell,
reflecting decreased digestive and excretory
functions of these cells due to hibernation.
The absolute volume of the excretory cells,
and the volume of their large cisternae, in-
creased during hibernation. There is a con-
troversy about the origin and the function of
excretory cells. Thiele (1953) and Sumner

(1965) suggested that they are degenerating
calcium cells, as they contained only a small
quantity of DNA in their nuclei, little cytoplasm
and a small number of mitochondria. Fretter
(1952), Billett & McGee-Russell (1955) and
Walker (1970) proposed that they are differ-
entiated forms of digestive cells. Abolins-
Krogis (1961) found mucopolysaccharides,
proteins, lipids and small quantities of RNA in
similar vacuoles of Helix aspersa and sug-
gested that this material is used for shell re-
pair. In H. lucorum, the increase in size of the
large cisternae, is probably a result of accu-
mulation of excretory material inside the ex-
cretory cells, which probably reflects the min-
imizing of their excretory processes during
hibernation.

In the present study, the electron micro-
scopic examination of H. lucorum digestive
glands revealed the presence of three cell
types: digestive cells, calcium cells and ex-
cretory cells. The morphological features sug-
gest that excretory cells constitute a distinct
cell type. Morton (1975, 1979) supported the
view that excretory cells of Pulmonata are di-
gestive cells in different stages of a sequence
of cytological changes over the course of 24
hours. Also Walker (1970) supported the view
that excretory cells of Agriolimax reticulatus
digestive glands are degenerating cells full of
lipofuscin, which are formed from digestive
cells. If this hypothesis can be supported, the
excretory cells should be the final step in the
development of the digestive cells. In that
case, the excretory cells contain the digestive
remnants of ingested material, which are de-
graded by lysosomal action and are finally
discharged into the gland lumen. Arising from
this hypothesis are the following questions:
(1) Why is the population density of the ex-
cretory cells in H. lucorum digestive glands
apparently smaller than that of the digestive
cell? (2) Why are there no cell types interme-
diate between digestive and excretory cells in
the examined digestive glands?

Another characteristic feature that 37 days
hibernation induced was an increase in the
number of calcium cells. This is evidently due
to the participation of these cells in the forma-
tion of the calcareous epiphragm that covers
the peristome of the snail during hibernation.
The role of calcium in gastropods and other
molluscs is multiple: it is regarded as a com-
ponent being implicated in pH homeostasis
(Sminia et al., 1977; Akberalli et al., 1977;
Jokumsen & Fyhn, 1982), in reproduction
(Tompa & Wilbur, 1977; Fournié & Chetail,

72 DIMITRIADIS 8 HONDROS

1982), in regulation of freezing tolerance
(Murphy, 1977), as a component of mucus
(Barr, 1928), in metabolic waste production
(Kniprath, 1975), and in shell regeneration
and formation of the epiphragm during hiber-
nation (Abolins-Krogis, 1961, 1965; Wagge,
1951). In pulmonates, calcium appears as
carbonate (Wagge, 1951) or phosphate
(Krijgsman, 1928; Oxford & Fish, 1979), in a
crystalline form or as an amorphous mass
(Wagge, 1951). In the hepatopancreas of He-
Их aspersa (Howard et al., 1981), calcium
granules contain CaMgP,0,. These granules
are sites for accumulation of a wide variety of
cations, acting as a detoxification mechanism
that traps a number of dietary metals. How-
ever, the wide functions of these granules are
largely unknown (Taylor et al., 1988).

Autolysis and extrusions of cytoplasmic
material into the digestive gland lumen of H.
lucorum were increased significantly as star-
vation and hibernation proceeded. A similar
cell response is also referred to by Bowen &
Davies (1971) and Oxford & Fish (1979) in
Arion hortensis and Cepaea nemoralis, where
release of hydrolytic enzymes from the cyto-
plasm of digestive cells was related to an in-
crease in the rate of cell autolysis. In H. luco-
rum, the extrusion of cytoplasmic pieces into
the lumen was possibly related to the reduc-
tion of energy requirements of the epithelium
and the maintenance of low function activity
of the digestive gland cells.

By using light and electron microscopic
observations and stereological evaluation,
the present study provides fine structural in-
formation about how hibernation and starva-
tion affects the cells of the digestive gland of
H. lucorum. However, more information, es-
pecially from the use of cytochemical meth-
ods, is needed for a better understanding of
the physiology of the phenomenon of hiber-
nation in snails.

ACKNOWLEDGEMENTS

This work was supported financially by the
Greek Ministry of Agriculture. We are grateful
to Dr. M. Lazaridou-Dimitriadou, who pro-
vided the snails and kept them under normal,
starvation and hibernation conditions.

LITERATURE CITED

ABOLINS-KROGIS, A., 1961, The histochemistry
of the hepatopancreas of Helix pomatia (L.) in

relation to the regeneration of the shell. Arkiv für
Zoologie, 13:159-201.

ABOLINS-KROGIS, A., 1965, Electron microscope
observations on calcium cells in the hepatopan-
creas of the snail Helix pomatia (L.). Arkiv fúr
Zoologie, 18:85-92.

AKBERALLI, H. B., K. R. M. MARRIOTT 4 E. R.
TRUEMAN, 1977, Calcium utilization during
anaerobiosis induced by osmotic shock in a bi-
valve mollusc. Nature, 266:852-853.

BARR, R. A., 1928, Some notes on the mucus and
skin glands of Arion ater. Quarterly Journal of
Microscopical Science, 71:503-526.

BILLETT, Е. 4 $. M. MCGEE-RUSSELL, 1955, The
histochemical localization of B-glucuronidase in
the digestive gland of the Roman snail (Helix
pomatia). Quarterly Journal of Microscopical Sci-
ence, 96:35-48.

BOWEN, I. D., 1970, The fine structure localization
of acid phosphatase in the gut epithelium cells of
the slug Arion ater (L.). Protoplasma, 70:247-
270.

BOWEN, I. D. & P. DAVIES, 1971, The fine struc-
ture distribution of acid phosphatase in the diges-
tive gland of Arion hortensis (F.). Protoplasma,
73:78-81. Е

DAVID, Н. & J. GOTZE, 1963, Electronenmikrosko-
pische befunde an der Mitterdarmdruse von
Schmenken. Zeitschrift für Microscopisch-anato-
mische Forschung, 70:252-272.

FOURNIE, G. & M. CHETAIL, 1982, Evidence for a
mobilization of calcium reserves for reproduction
requirement in Deroceras reticulatum. Malacolo-
gia, 22:285-291.

FRETTER, V., 1952, Experiments with 92P and '3'|
on species of Arion and Agriolimax. Quarterly
Journal of Microscopical Science, 93:133-146.

HOWARD, B., P. C. M. MITCHELL, A. RITCHIE, K.
SIMKISS 8 M. G. TAYLOR, 1981, The composi-
tion of intracellular granules from the metal-accu-
mulating cells of the common garden snail (Helix
aspersa). Biochemical Journal, 194:307-311.

JANSSEN, H. H., 1985, Some histophysiological
findings on the midgut gland of the common gar-
den snail, Arion rufus (L.). Zoologischer An-
zeiger, (Jena) 215:33-51.

JOKUMSEN, A. & H. J. FYHN, 1982, The influence
of aerial exposure upon respiratory and osmotic
properties of hemolymph from two intertidal mus-
sels, Mytilus edulis and Modiolus modiolus. Jour-
nal of Experimental Marine, Biology and Ecology,
(B)61:189-203.

KARNOVSKY, M. J., 1965, A formaldehyde-glu-
taraldehyde fixative of high osmolarity for use in
electron microscopy. Journal of Cell Biology, 27:
137A-138A.

KNIPRATH, E., 1975, Das Wachstum des Mantels
von Lymnaea stagnalis (Gastropoda). Cytobiolo-
gie, 10:260-267.

KRIJGSMAN, B. J., 1925, Arbeitsrhythmus der Ver-
dauungsdrúsen bei Helix pomatia. 1. Die naturli-
chen bedingungen. Zeitschrift für Vergleichende
Physiologie, 2:264-296.

STARVATION IN HELIX LUCORUM 73

KRIJGSMAN, B. J., 1928, Arbeitsrhythmus der Ver-
dauungsdrusen bei Helix pomatia. 2. Sekretion,
Resorption und Phagocytose. Zeitschrift für Ver-
gleichende Physiologie, 8:187-280.

LAZARIDOU-DIMITRIADOU, M. & D. S. SAUN-
DERS, 1986, The influence of humidity, photo-
period, and temperature on the dormancy and
activity of Helix lucorum L. (Gastropoda, Pulmo-
nata). Journal of Molluscan Studies, 52:
180-189.

MORTON, B., 1975, The seasonal variation in the
feeding and digestive cycle of Amphibola crenata
(Martin 1784) (Gastropoda: Pulmonata). Forma
et Functio, 8:175-190.

MORTON, B., 1979, The diurnal rhythm and the
cycle of feeding and digestion in the slug Dero-
ceras carnuanae. Journal of Zoology, 187:135—
152.

MURPHY, D. J., 1977, A calcium dependent mech-
anism responsible for increasing freezing toler-
ance of the bivalve mollusc Modiolus demissus.
Journal of Experimental Biology, 69:13-21.

OWEN, G., 1966, Digestion. In: Physiology of Mol-
lusca, Vol. 2, К. М. WiLBUR & С. M. YONGE,
eds. Academic Press, New York, San Francisco.

OXFORD, С. S. 4 L. J. FISH, 1979, Ultrastructural
localization of esterase and acid phosphatase in
digestive gland cells of fed and starved Cepaea
nemoralis (L.). Protoplasma, 101:186-196.

PIPE, R. K., 1986, Light and electron microscopic
localization of ß-glucuronidase activity in the
stomach and digestive gland of the marine gas-
tropod Littorina littorea. Histochemical Journal,
18:175-186.

PROSSER, C. L., 1973, Comparative animal phys-
iology. W. B. Saunders Company, London.

ROLDAN, C., 1987, Ultrastructural modifications of
the epithelium in the anterior digestive tract in
starved specimens of Theba pisana. Iberus, 7:
153-164.

SMINIA, T., N. D. de WITH, J. L. BOS, M. E. Van
NIEUMEGEN, M. P. WITTER & J. WONDER-
GEM, 1977, Structure and function of the calcium
cells of the fresh water pulmonate snail Lymnaea
stagnalis. Netherland Journal of Zoology, 27:
195-208.

STEER, M. W., 1981, Understanding cell structure.
Cambridge University Press, Cambridge.

SUMNER, A. T., 1965, The cytology and cytochem-
istry of the digestive gland cells of Helix. Quar-
terly Journal of Microscopical Science, 106:173—
192.

SUMNER, A. T., 1966, The fine structure of the
digestive gland cells of Helix, Succinea and Tes-
tacella. Journal of the Royal Microscopical Soci-
ety, 85:185-192.

TAYLOR, M. G., K. SIMKISS, G. N. GREAVES & J.
HARRIES, 1988, Corrosion of intracellular gran-
ules and cell death. Proceedings of the Royal
Society of London, B234:463-476.

THIELE, G., 1953, Vergleichende unterschuchun-
gen uber den Feinbau und die Funktion der mit-
terdarmdrüse einheimischer Gastropoden.
Zeitschrift für Zellforschung und Microscopische
Anatomie, 38:87.

TOMPA, A. S. & K. M. WILBUR, 1977, Calcium
mobilization during reproduction in snail Helix
pomatia (L.). Nature, 270:53-54.

VON BRAND, T., 1931, Der Jahreszyklus im Stoff-
bestand der Weinbergschnecke (Helix pomatia).
Zeitschrift fúr Vergleichende Physiologie, 14:
200-264.

VON BRAND, T., M. O. NOLAN 4 E. R. MANN,
1948, Observations of the respiration of Austral-
orbis glabratus and some other aquatic snails.
Biological Bulletin, 95:199-213.

WAGGE, L. E., 1951, The activity of amoebocytes
and of alkaline phosphatases during the regen-
eration of the shell of the snail Helix aspersa.
Quarterly Journal of Microscopical Science, 92:
307-321.

WALKER, G., 1969, Studies on digestion of the
slug Agriolimax reticulatus (M.). Ph. D. Thesis,
University of Wales.

WALKER, G., 1970, The cytology, cytochemistry
and ultrastructure of the cell types found in the
digestive gland of the slug Agriolimax reticulatus
(M.). Protoplasma, 71:91-109.

WEIBEL, E. R., 1979, Stereological methods. Vol 1:
Practical methods for biological morphometry.
Academic Press, New York and London

Revised Ms. accepted 19 September 1991

ı 20»

ANNA +
ne

4 44 us! u

> o
N

PA 7
>
AN €
o

o о.

ae
En

o

ry IE de

ТЕ

HO Ei

. DES iio o
== = A
> ir
o = o
u
к

LE

MALACOLOGIA, 1992, 34(1-2): 75-86

SUPRASPECIFIC NAMES OF MOLLUSCS: A QUANTITATIVE REVIEW

Philippe Bouchet & Jean-Pierre Rocroi

Muséum National d'Histoire Naturelle, 55 Rue de Buffon, Paris 75005, France

ABSTRACT

The number of nomenclaturally available genus-group taxa described since 1758 for Recent
and fossil molluscs has been estimated by two methods. One result stands at 28,400 names;
another at 24,900 names, of which 12,700 are gastropods, 6,000 are cephalopods, 5,100 are
bivalves and 1,100 are in the smaller classes. The yearly increment appears to have remained
relatively stable since the late 19th century. It is presently at an average of 224 new genus-group
names per year, with Cephalopoda representing precisely one-third of the names. At least 20%
of the recently introduced taxa escape the Zoological Record, with the Soviet and paleontolog-
ical literatures especially underrepresented. In the last 30 years, journals and books published
in USSR, USA, and China contain altogether 50% of the new names. The total number of
nomenclaturally available genus-group names of Recent molluscs is on the order of 12,000. It
is estimated that the number of family-group names for molluscs is over 5,000.

Key words: genus-group names, family-group names, numbers, nomenclature, nomenclators,

trends, literature coverage.

INTRODUCTION

Although estimates of molluscan species
diversity (Nicol, 1969; Boss, 1971; for a criti-
cism, see Solem, 1978) exist, genus-group
names, as defined by Article 42 of the /nter-
national Code of Zoological Nomenclature,
have not received similar attention for many
years. In molluscs, as in other large groups of
Recent and fossil animals, taxonomic work is
hampered by the lack of adequate, up-to-
date, comprehensive manuals, and by the
vast number of journals and books containing
descriptions of new taxa. The fate of every
family- or genus-level taxonomic treatise is to
be incomplete or outdated the very year it is
published. In malacology, the currently avail-
able taxonomic academic treatises are 20-50
years old (Wenz, 1938-1944; Zilch, 1959—
1960; Moore, 1960-1971). In 1987, as a re-
sult of this frustration, we started to compile a
loose-leaf index to the new supraspecific
names proposed since 1960 for Recent and
fossil molluscs, exclusive of cephalopods. As
this work is now being extended to encom-
pass the older literature, we have found it in-
teresting to quantify the abundance of names
involved, to identify trends in current taxo-
nomic literature, and to evaluate the efficiency
of the supposedly most complete indexing
system, the Zoological Record. The present
paper documents these findings and dis-
cusses their implications.

75

Taxonomic research in malacology demon-
strates a strong division between cephalopod
and non-cephalopod literature. In particular,
ammonoid research is almost totally sepa-
rated from other fields of molluscan research:
malacological journals only exceptionally
contain papers on ammonites; ammonite and
other cephalopod specialists rarely interact
with other malacologists in national and inter-
national congresses. Our index, which re-
flects our own research interests, excludes
cephalopods and is itself an illustration of this
dichotomy. As a consequence, while some of
our results are concerned with Mollusca in
general, others deal only with Mollusca exclu-
sive of cephalopods.

METHODS

The number of genus-group taxa in Mol-
lusca is based on an evaluation of the number
of names contained in various catalogues,
nomenclators and checklists. Names enu-
merated in the sources listed in Table 1 were
counted. For the Nomenclator Zoologicus, we
counted the molluscan names in 50 pages
(chosen from five consecutive pages in ten
random samples) per volume. The total of
250 pages sampled represents about 5% of
4,816 pages included in the Nomenclator.
Only original spellings were considered.

76 BOUCHET 8 ROCROI

To index the supraspecific names pro-
posed after 1960, we first scanned the Mol-
lusca volumes of the Zoological Record tor
the relevant years. For Cephalopoda, only ge-
nus-group names were considered: we sim-
ply counted them without further checking.
For classes other than cephalopods, genus-
group and family-group names were consid-
ered separately. The original publications
were consulted, and all newly introduced
names were checked and recorded.

In a second phase, we used a combination
of approaches to find names (of non-Ceph-
alopods) omitted in the Zoological Record: (a)
We scanned biographical compilations of ma-
lacologists containing their lists of papers and
taxa (e.g. Powell by Cernohorsky, 1988;
Habe by Inaba 8 Oyama, 1977), and yearly or
cumulative indices for malacological journals
(e.g. Anonymous, 1979); (b) We searched
malacological and regional journals not cov-
ered by the Zoological Record that we knew
contained new malacological taxa (e.g. Noti-
ziario del CISMA, Roma; Schriften zur Mala-
kozoologie, Cismar); (c) We used library facil-
ities in Paris (Muséum National d'Histoire
Naturelle, Société Géologique de France,
Centre National de la Recherche Scientifique,
Université Pierre et Marie Curie and several
personal libraries), Leningrad (Zoological In-
stitute of the Academy of Sciences and All-
Union Geological Institute of the Ministry of
Geology), Londor (The Natural History Mu-
seum), Stockholm (Naturhistoriska Riksmu-
seet and University Library) and Frankfurt
(Senckenberg Library), including browsing
through collections of reprints in departmental
libraries; (d) We reviewed available papers for
secondary uses of supraspecific names pub-
lished elsewhere in unrecorded places; this
method proved particularly useful with the So-
viet literature; (e) The recent (since 1985)
Chinese paleontological literature is covered
by the quarterly Gushengwuxue Wenzhai
(English subtitle: Paleontological Abstracts),
and we discovered many names in this ab-
stracting journal; (f) Recently published cata-
logues of names (such as Vokes, 1980, 1990;
Vaught, 1989) were also scanned for omis-
sions, and several colleagues provided refer-
ences to obscure names.

Our data base is therefore far more com-
plete for non-cephalopods than for cephalo-
pods. Even for non-cephalopods, it is admit-
tedly still a little incomplete but provides the
best available source of supraspecific names
for the years 1960-1989.

RESULTS
Number of Genus-Group Taxa in Mollusca

We estimated the total number of genus-
group names by two independant methods.
One method relies primarily on a statistical
analysis of the names listed in Nomenclator
Zoologicus. The other is based on the various
catalogues and checklists available for se-
lected classes of the phylum.

Evaluation Based on
Nomenclator Zoologicus

(a) 1758-1965: We counted molluscan ge-
nus-group names in a set of 250 pages of the
Nomenclator Zoologicus, selected as de-
scribed under “Methods.” Excluding incorrect
subsequent spellings, there is an average
4.77 molluscan names per page, and the total
number of nomenclaturally available genus-
group names can be estimated at 23,000 for
the period 1758-1965. This number is a min-
imum because some names certainly have
escaped the Nomenclator Zoologicus. It is our
experience, however, that while omissions
appear to become more numerous with re-
cent volumes, as we document below, the
coverage of the Nomenclator Zoologicus is
fairly good for the years 1758-1935. We
therefore believe 23,000 names to be a reli-
able estimate.

(b) 1966-1989: Using the combination of
methods outlined above, our index of non-
cephalopod names for the period 1966-1989
contains 3,644 genus-group taxa. For the
same period, the Zoological Record lists
1,448 genus-group names of Cephalopoda. If
it is conservatively assumed that 20% of the
cephalopod names have escaped the Zoolog-
ical Record (see below), then the total num-
ber of cephalopod names introduced during
that period would be 1,810. The total number
of molluscan genus-group names introduced
in the years 1966-1989 is therefore approxi-
mately 5,400.

The total number of genus-group names in
Mollusca therefore amounts to circa 28,400
names.

Evaluation Based on Catalogues

A number of catalogues or counts are avail-
able for selected classes or subclasses of
Mollusca. We have counted the number of
genus-group names in comprehensive works,

SUPRASPECIFIC NAMES OF MOLLUSCS Ta

TABLE 1. Number of genus-group names of
molluscs, partitioned by class

Total Recent
Aplacophora 80 (a) 80 (a)
Monoplacophora 267 (b) 9
Polyplacophora 282 (c) 205 (a,c)
Bivalvia 5090 (d) 2043 (d)
Scaphopoda 63 (e) 57 (a)
Gastropoda 12721 9004 (a,g),
10451 (a)
Prosobranchia 7149 (f) 4112 (9),
5559 (a)
Opisthobranchia 982 (h) 817 (a)
Pulmonata 4590 (i) 4075 (a)
Hyolitha,
Rostroconchia, etc.. 362 (j) 0
Cephalopoda 6000 (k) 305 (I)
Total 24865 11703-13150
Sources:

(a) after Vaught (1989); (b) after Knight & Yochelson, and
Knight, Batten & Yochelson, in Moore (1960), plus incre-
ment; (c) after van Belle (1975-78), plus increment; (d)
after Vokes (1980), plus increment; (e) after Ludbrook, in
Moore (1960), plus increment; (f) after Schilder (1947) for
the period 1758-1932, an estimated increment of 45.5
names/yr. for the period 1933-1959 (see Table 4), and our
own counts for 1960-1989; (g) same as (f), but estimated
increment 24.3 names/yr; (h) after Zilch (1959—60), Rus-
sell (1971, 1986), and increment; (i) after Zilch (1959-60),
plus increment; (j) Hyolitha and Tentaculitida after Fisher,
in Moore (1962), plus increment; Rostroconchia after
Vokes (1980), plus increment; (k) Ammonoida and Nauti-
lida after Hewitt (1989), modified; (I) after Nesis (1987).

handbooks and catalogues, as listed in Table
1. Because none of these sources is com-
plete to 1989, we have added an increment,
based on our own index for the more recent
years. When these independent subcounts
are added, the result is 24,900 names (Figs.
153):

Number of Genus-Group Names of
Recent Mollusca

In this analysis, a name is classified as Re-
cent if it is based on a Recent type species.
Therefore, Recent and fossil constitute two
mutually exclusive categories: we are fully
aware that this is only an approximation of the
actual situation.

From the sources listed in Table 1, we con-
clude the number of genus-group names of
Recent Mollusca (Fig. 2; see also Fig. 4) to be
approximately 12,000, or 42% of all mollus-
can genus-group names. As a comparison,
Vaught (1989), who also listed incorrect sub-
sequent spellings, chresonyms (one name

may be tabulated several times), and some
genus-group names with fossil type species,
ennumerated approximately 15,000 names.

Discussion

The two estimates, 28,400 and 24,900, are
consistent with each other and differ only by
12%. We believe that the discrepancy be-
tween the two figures arises mainly from our
use of Schilder (1947) for an evaluation of the
number of names in the Prosobranchia.
Schilder's data were based on Wenz (1938—
44), which is fairly complete only for names
published before 1933. We do not know to
what extent Schilder corrected his counts with
the literature published in 1933-1947, and
this fact alone could explain a “loss” of sev-
eral hundred names.

Our results are intermediate between two
previously published estimates: (1) Schilder
(1949) estimated that there were 11,260 gen-
era established for “living and fossil shells.”
Because he did not explain how he obtained
his results, or what he meant with “shells,” it
is difficult to comment. We consider that this
very low figure is a gross underestimation of
the actual number of names involved, even
when one appreciates that Schilder made his
statement more than 40 years ago.

(2) Vokes (1967) stated he had a card file
containing 40,000 names of Mollusca ex-
tracted from volumes 1-5 of Nomenclator
Zoologicus, a much higher figure than our
own results indicates. We corresponded with
Dr. Vokes to discuss this difference, and we
have had access to the card-index that was
the basis for his 1967 and 1980 catalogues,
which he generously donated in 1991.

It appears that Vokes listed incorrect sub-
sequent spellings as well as original spellings:
on average, 22.5% of the names enumerated
in Vokes (1980) are incorrect subsequent
spellings, and as many as 37.9% of the 8,200
cards in his bivalve card-index are for nomen-
claturally unavailable names (incorrect sub-
sequent spellings, chresonyms, etc.). From
this evidence, we conclude that 40,000 was a
gross overestimate, and that it should not be
regarded as an accurate number of mollus-
can genus-group names.

Naming Activity in the Last 30 Years
In the period 1960-1989, 6,720 new mol-

luscan genus-group names have been intro-
duced. Numbers for classes other than ceph-

78 BOUCHET 8 ROCROI

Partition by class, 1758-1989

Bivalvia
33.97%

Cephalopoda

Aplacophora
Monoplacophora
Polyplacophora
Bivalvia
Scaphopoda
Gastropoda
Hyolitha, etc...
Cephalopoda

ENDOBESH

Gastropoda
44.71%

FIG. 1. Number of genus-group names of Mollusca introduced since 1758, partitioned by class.

Recent Mollusca, Partition by class

Cephalopoda

Bivalvia

Aplacophora
Monoplacophora
Polyplacophora
Bivalvia
Scaphopoda
Gastropoda
Hyolitha, etc...
Cephalopoda

Gastropoda

9000028 №

76.94%

FIG. 2. Number of genus-group names of Recent Mollusca introduced since 1758, partitioned by class.

alopods represent genus-group names ac- fossil, is presented in Table 2 and Figure 5.

tually found by us. For the Cephalopoda, we For simplicity, Caudofoveata and Solenogas-

use the count of names recorded by the Zoo- tra have been grouped under “Aplacophora,”

logical Record, corrected on the basis of an and the minor fossil classes are grouped un-

estimated 20% omission rate by the Zoologi- der a single entry. For the controversial con-

cal Record. tents of the Monoplacophora, we have fol-
The breakdown by class, and Recent vs. lowed Runnegar 4 Pojeta (1985).

SUPRASPECIFIC NAMES OF MOLLUSCS 79

Gastropod breakdown by subclass
1758-1989

Pulmonata
36.1%

Prosobranchia
56.2%

Opisthobranchia
7.1%

FIG. 3. Number of genus-group names of Gas-
tropoda introduced since 1758, partitioned by sub-
class.

Recent Gastropoda
breakdown by subclass

Pulmonata
39.0%

Prosobranchia
53.2%

Opisthobranchia
7.8%

FIG. 4. Number of genus-group names of Recent
Gastropoda introduced since 1758, partitioned by
subclass.

These results lead to several observations:

1. Cephalopod names can be estimated at
33% of the total on the basis of volume
7 of Nomenclator Zoologicus, the only
one that indicates class level. Counts for
the entire period 1960-1989 give
33.3%. It is remarkable that 98.2% of all
cephalopod genus-group names are
based on fossils.

2. Within the Gastropoda, which com-
prises 37.5% of the total, the number of
taxa can be further subdivided by sub-
class as presented in Table 3 and Fig-
ure 6.

Of all gastropod names, 62.6% are
based on Recent species, 37.4% on fos-
sil species. It is interesting to note that
61.7% of Archaeogastropoda names,
which date back to the early Paleozoic,
are based on fossils, whereas only

40.9% of Neogastropoda, which date
back to the Cretaceous, are based on
fossils. Considering that there are many
more taxonomists working on Recent
gastropods than on Paleozoic ones, this
would tend to indicate that name counts
do reflect some pattern of overall diver-
sity, rather than the activity of taxono-
mists.

3. Bivalves are, to a considerable extent,
dominated by paleontological research:
79.6% of all bivalve genus-group names
are based on fossil species.

Zoological Record Coverage

We have estimated the Zoological Record
coverage by two alternative methods.

1. Volume 7 (1956-1965) of the Nomen-
clator Zoologicus is largely based on the
Zoological Record (the same typograph-
ical errors that crept in the Zoological
Record also appear in the Nomenbclator,
e.g. Apollonia misspelled Appollonia).
We searched for 311 names (all the
names beginning with A, B or C) known
to us, published during this span of
years—56 names (18.0%) were found
missing in that volume.

2. For the period 1966-1989, the Zoolog-
ical Record listed some 2,848 genus-
group names of Mollusca exclusive of
Cephalopoda. (Names published in the
1950s but recorded by the ZR in the
1960s were not counted by us in this
total. Names published in 1989 and re-
corded by the ZA in the 1989-90 vol-
ume were counted.) Our own index con-
tains 3,644 names for the same period.
This indicates a coverage by the Zoo-
logical Record of 78.2%, or an omission
rate of 21.8%.

This is admittedly an optimistic esti-
mate, because our own index has also
certainly missed a number of names.
For instance, we know of another 85
names with incomplete references or no
references at all, which we have not yet
been able to trace to their primary
source. If these 85 names are consid-
ered when calculating the omission rate,
it then jumps to 23.7%.

Therefore, it seems fair to conclude that at
least 20% of new genus-group names have
been omitted in the last 35 years by a nomen-

80 BOUCHET 8 ROCROI

TABLE 2. Breakdown, by class, of new genus-group names introduced in 1960-1989

Total Increment
Paleozoic Mesozoic Cenozoic Fossil Recent Total (names/yr.)
Aplacophora = = = — 32 32 1.1
Monoplacophora 173 1 — 174 7 181 6.0
Polyplacophora 30 2 1 33 15 48 1.6
Bivalvia 301 525 275 1101 282 1383 46.1
Scaphopoda 7 1 1 9 24 33 1.1
Gastropoda 164 293 477 934 1566 2500 83.3
Hyolitha, Rostroconchia, 305 2 1 308 = 308 10.3
etc.
Cephalopoda 2197 38 2235 74.5
Grand Total 4756 1964 6720 224
3000
Mollusca genus-group names, 1960-1989
2000
O Recent
Cenozoic
1000 ES Mesozoic

had Paleozoic

<
<
o
iS
a
o
о
Ss
a
<

Bivalvia

Monoplacophora
Polyplacophora

Scaphopoda
Gastropoda
Hyolitha,etc...
Cephalopoda

FIG. 5. Number of genus-group names of Mollusca introduced in the period 1960-1989, partitioned by class.
All fossil Cephalopoda have been plotted under Mesozoic, which is admittedly an oversimplification.

clator considered to be the most complete in
terms of coverage of the taxonomic literature.

It is important to emphasize that omission
by the Zoological Record is not random. The
most imperfect coverage is of the Soviet liter-
ature, in particular the coverage of paleonto-
logical monographs. Altogether, probably as

much as one-third of names proposed in the
Soviet literature escapes the Zoological Rec-
ord. Other imperfectly covered literatures are
those from China, Japan, South America and
southern Europe. We confirm the generally
accepted belief that books are much more
poorly recorded than journals.

SUPRASPECIFIC NAMES OF MOLLUSCS 81

Gastropoda genus-group names

1960-1989

2000

O Recent
Cenozoic
E Mesozoic

1000 Ш Paleozoic

Pulmonata

Opisthobranchia

o
RS
о
=
$
>
a
o
я
o
ce
a

FIG. 6. Number of genus-group names of Gas-
tropoda introduced in the period 1960-1989, parti-
tioned by subclass.

Temporal Variation in Names Output

The year of publication of genus-group
names has been extracted from the samples
of the Nomenclator Zoologicus described
above, and grouped by periods of 30 years.
Table 4 shows the total number of genus-
group names proposed for each period of 30
years, and the yearly rate within each of these
eight periods.

An examination of the number of supraspe-
cific names contained per volume of Zoolog-
ical Record for the period 1960-1989 reveals
considerable variability in the yearly output
and indicates that no single randomly se-
lected year can be considered representative
for the period.

Table 4 and Figure 7 show a regular growth
of the yearly increment during the first 100
years, with a faster growth in the latter part of
the 19th century. The yearly output has sub-
sequently remained remarkably stable at
170-200 genus-group names per year since
1880, with a slight minimum in the period
1936-1965 (also noted by Solem, 1978).

Hewitt (1989) stated that the recent de-
cades are characterized by important “mono-
graphic bursts” in nautiloid and ammonoid
taxonomy. Such bursts have certainly oc-
curred elsewhere in selected groups and fau-

nas, such as the Aplacophora, for which 40%
of the names have been introduced in the last
30 years. However, contrary to a rather gen-
eral belief among many molluscan taxono-
mists, Table 4 and the graph suggest only a
moderate increase, not an overall explosion,
in the output of names over this 30-year pe-
riod.

Which Country Produces the
Most Molluscan Taxa?

To answer this question we have plotted for
1960-1989 the number of new genus-group
names (non Cephalopods only) by country of
origin of the author (Fig. 8), and separately by
country of publication. When multiple author-
ship is involved, each author is counted for
0.5 (two authors), 0.33 (three authors), etc.

In the period 1960-1989, authors from 53
countries are involved, with those from
USSR, USA and China accounting for a little
over half (50.3%) of the names. Below the 6th
rank (New Zealand, 4.1%), the percentage is
already below 5%, and the 1% mark is
reached at the 17th rank (Cuba).

National output should not be used as a
key to overail molluscan biological diversity of
the different parts of the world. While it is true
that, to some extent, the vast majority of the
Japanese naming activity focuses on the Jap-
anese (paelo)fauna, the naming activity of
many countries involves faunas from several
oceans or continents: a marine snail from the
Caribbean is just as likely to be named by an
American or a European author. In other
words, there is no immediate biological expla-
nation for the differences in numbers of mol-
lusc genus-group names described in each
country.

It is remarkable that New Zealand, Austra-
lia and Czechoslovakia, with national popula-
tions of respectively 3.4, 16.8 and 15.6 mil-
lion, rank among the 10 countries with the
highest output. New Zealand is then not only
the country with the highest number of sheep
per capita, but also the country with by far the
highest output of mollusc genus-group names
per capita!

Ranking by country of publication (Table 6)
is not significantly different from the ranking
by country of origin of authors. This means
that authors do not, as a rule, “expatriate”
their papers in journals of other countries.
However, Germany and, to a lesser extent,
Great Britain rank higher as publishers than

82 BOUCHET 8 ROCROI

TABLE 3. Breakdown, by subclass, of new gastropod genus-group names introduced in 1960-1989

Total

Paleozoic Mesozoic Cenozoic Fossil Recent Total

Prosobranchia 163 245 364 772 916 1688
Archaeogastropoda 148 61 26 235 146 381
Mesogastropoda 15 167 156 338 482 820
Neogastropoda — 17 182 199 288 487
Opisthobranchia = 24 21 45 168 213
Pulmonata 1 24 92 Ue 482 599
Total 164 293 477 934 1566 2500

TABLE 4. Estimates of number of genus-group names introduced during 30-year periods

1758—= 1788 1818— 1848— 1878— 1908— 1936— 1966-1989
period 1787 1817 1847 1877 1907 1935 1965 (1) (2)
time (yr) 30 30 30 30 30 28 30 24
names 171 874 2755 3002 5643 5947 5115 4333 5454
increment 5.7 29.1 91.8 100.1 188.1 212.4 170.5 180.5 227.3

(1) Names actually recorded by the Zoological Record (2) Names found by us (non-cephalopods) + cephalopod names
estimated on the basis of a 20% omission rate.

300

Mean yearly increment

200

Number of names

100

FIG. 7. Mean yearly increment of new genus-group names of Mollusca, calculated over 30-year periods.

as authors, whereas the opposite is true for available for molluscs has ever been pub-

New Zealand. lished. A fact well known to taxonomists is that
a new family may be erected without even
Family-Group Names mentioning that a new name is being intro-

duced. For that reason, we believe that our
No estimate of the number of family-group data base is slightly less complete for family-
names, as defined by Article 35 of the Code, group names than for genus-group names.

SUPRASPECIFIC NAMES OF MOLLUSCS 83

800

600

400

200

USSR
USA
China
Japan
Germany

Ranking by country of author

New Zealand
France
Australia
Great Britain
Czechoslovakia

FIG. 8. Ranking of number of genus-group names of Mollusca (exclusive of Cephalopoda) introduced in

1960-1989, partitioned by country of origin of author.

TABLE 5. Ranking of country of origin of author
as a function of total number of genus-group
names.

1 2 3 4

Country no. of names % Rank
USSR 970 21.6 1
USA 831 18.5 2
China 451 10.1 3
Japan 374 8.3 4
Germany 314 7.0 5
New Zealand 184 4.1 6
France 154 3.4 7
Australia 143 3.2 8
Great Britain 108 2.4 9
Czechoslovakia 96 2.1 10
Total 3625 80.7

The only partial checklist of family-group
names is that by Ponder 8 Warén (1988) for
the Caenogastropoda and Heterostropha. It
lists 833 names, of which 187 were intro-
duced in the period 1960-1987.

TABLE 6. Ranking of number of genus-group
names arranged by place of publication (1960—
1989)

Rank Country no. of names %
1 USSR 955 23
2 USA 901 20.1
3 China 446 9.9
4 Germany 415 9.3
5 Japan 390 8.7
6 Australia 154 3.4
7 France 146 3:3
8 Great Britain 143 3.2
9 New Zealand 133 3.0

10 Italy 99 2.2

1-10 3782 84.4

Our own index lists 1,102 new family-group
names proposed for molluscs exclusive of
cephalopods in the period 1960-1989. If we
assume that the ratio of names published be-
fore and after 1960 is the same for Caeno-

84 BOUCHET 8 ROCROI

gastropoda + Heterostropha as for the rest of
the molluscs, then the total number of fam-
ily-group names available for molluscs ex-
clusive of cephalopods can be estimated
at 4,900. No similar figure is available for
Cephalopoda, but Hewitt (1989) counted 409
nautiloid and ammonoid families.

In our view, this demonstrates the need for a
new nomenclatural tool that would be to fam-
ily-group names what the Index Animalium
(Sherborn, 1902; 1922-1932) and the No-
menclator Zoologicus are to species-group
and genus-group names respectively.

Epilogue

The sheer magnitude of the numbers dis-
cussed in this paper will certainly draw con-
trasting opinions among malacologists. Two
extreme views can be expected. In one view,
the newly named taxa are deemed to repre-
sent taxonomically valid units, and the current
naming activity simply demonstrates the
gross incompleteness of the knowledge on
the diversity of this phylum. In the opposite
view, most of the newly created names are
synonyms and the current naming activity is
the symptom of a system gone crazy.

Superfluity in molluscan nomenclature ap-
pears to be a recurrent concern among pro-
fessional taxonomists (e.g. Schilder, 1949;
Nicol, 1958; Boss, 1971, 1978). Schilder
(1949) estimated that 34% of the names avail-
able to classify prosobranchs were synonyms.
However, whereas 49% of the names estab-
lished in 1808-1857 were synonyms, the syn-
onymy ratio decreased to 37%, 34% and 21%
respectively for the next consecutive 25-year
periods (Schilder, 1949). This can be inter-
preted in two ways: either taxonomists have
been doing better work since the early 19th
century, or it takes a long period of time (in
excess of 75 years) before the value of a ge-
nus-group name can be properly assessed. It
is likely that both elements reflect the actual
situation, an opinion already expressed by
Schilder & Schilder (1947). Schilder (1947)
had also suggested that genus-group names
of Recent and fossil prosobranchs, described
and undescribed, would amount to about ap-
proximately 20,000, of which 5,000 would still
be extant. This prediction may not be as un-
realistic as it may first seem: in bivalves, fossils
outnumber Recent genus-group names in the
proportion of 2-3 to 1; and there are already
in the order of 5,000 Recent prosobranch ge-
nus-group names. If Schilder was correct, this

would mean that more than 10,000 fossil
prosobranch genera still await naming, a
daunting perspective!

Naming activity strongly reflects national
traditions. There used to be a time when ma-
lacologists from Australia and New Zealand
did not expect that their fauna might already
have been described by workers in other
parts of the world, and consequently engaged
in overnaming what they considered to be en-
tirely endemic faunas. The expression “an-
other creation” was even used for the Austra-
lian biota. It is clear that this scientific isolation
has now ended, and that the high level of mol-
luscan naming activity in New Zealand and
Australia is the result of a healthy descriptive
malacology there in a worldwide context.

Notwithstanding our unfamiliarity with Pa-
leozoic and Mesozoic faunas, we remain
greatly concerned by the introduction of new
taxa at the genus and family levels based on
poorly preserved fossils or molds. Certain
branches of malacology are also undoubtedly
suffering from “inbreeding,” a case in point
being the immense literature on the Neogene
Ponto-Caspian basins.

Despite these reservations, we do not
share the view that overnaming is the single
most important factor explaining yearly incre-
ments of 224 molluscan genus-group names,
and 40+ family-group names. In the last 30
years, whole new faunas have been discov-
ered, either Recent (e.g. oceanic hydrother-
mal vents) or fossil (e.g. lower Cambrian of
China); old faunas have been readressed us-
ing new techniques (e.g. scanning electron
microscopy, SCUBA diving) and new charac-
ters (e.g. anatomy of minute species). And,
perhaps most importantly, the phase of intel-
lectual stagnation that followed Thiele’s ep-
och-making Handbuch is giving way to stim-
ulating and provocative ideas. In this process,
superfluous names undoubtedly become in-
troduced, but we are convinced that these do
not minimize the considerable amount of gen-
uine discoveries being made every year.

The unexpectedly high omission rate of
Zoological Record should cause concern to
all taxonomists. Because this nomenclator is
the main bibliographical source of many (pa-
leo)zoologists, this factor alone poses an im-
portant threat to nomenclatural stability. We
believe that this justifies the establishment
of new criteria of availability of zoological
names, whereby a published name would
have to be registered by the International
Commission of Zoological Nomenclature be-

SUPRASPECIFIC NAMES OF MOLLUSCS 85

fore it is declared nomenclaturally available
(Bouchet, in press).

ACKNOWLEDGMENTS

For assistance with literature and/or calling
our attention to names not listed in the Zoo-
logical Record, we thank A. Bogan (Academy
of Natural Sciences, Philadelphia), S. Freneix
(Institut de Paléontologie, Paris), R. Janssen
(Senckenberg Museum, Frankfurt), A. Kabat
(National Museum of Natural History, Wash-
ington, D.C.), T. Kase (National Science Mu-
seum, Tokyo), |. Loch (Australian Museum,
Sydney), A. Lum (Natural History Museum,
London), Z.-G. Ma (Institute of Geology and
Paleontology, Academia Sinica, Nanjing), A.
Matsukuma (National Science Museum, To-
kyo), R. Petit (North Myrtle Beach, South
Carolina), Yu. Starobogatov and M. Dolgo-
lenko (Zoological Institute, Leningrad), H.
Vokes (Tulane University, New Orleans) and
A. Warén (Naturhistoriska Riksmuseet,
Stockholm). For assistance in English, we
thank A. Kabat. A. Foubert helped generating
computer-produced figures.

LITERATURE CITED

ANONYMOUS, 1979, Venus: Japanese Journal of
Malacology. Volumes 1-36 (1928-1977). The
Malacological Society of Japan, Tokyo.

BOSS, K. J., 1971, Critical estimate of the number
of Recent Mollusca. Occasional Papers on Mol-
lusks, 3(40):81-135.

BOSS, K. J., 1978, Taxonomic concepts and su-
perfluity in bivalve nomenclature. Philosophical
Transactions of the Royal Society of London,
(B)284:417-424.

BOUCHET, P. (in press), Proposal that a zoological
name becomes nomenclaturally available from
the date when the work containing its introduction
is received by the Secretariat of ICZN. Bulletin of
Zoological Nomenclature

CERNOHORSKY, W. O., 1988, Arthur William Ba-
den Powell (1901-1987). A brief biography and
bibliography with a list of molluscan taxa.
Records of the Auckland Institute and Museum,
25:1-38.

EDWARDS, M. A. & A. T. HOPWOOD, 1966, No-
menclator Zoologicus, vol. 6. The Zoological So-
ciety of London, London.

EDWARDS, М. А. & H. С. VEVERS, 1975, Nomen-
clator Zoologicus, vol. 7. The Zoological Society
of London, London.

GUSHENGWUXUE WENZHAI [= Paleontological

Abstracts], 1986-1991. Quarterly. Institute of
Geology and Paleontology, Nanjing.

HEWITT, R. A., 1989, Recent growth of nautiloid
and ammonite taxonomy. Paläontologische
Zeitschrift, 63:281-296.

INABA, T. & K. OYAMA, 1977, Catalogue of mol-
luscan taxa described by Tadashige Habe during
1939-1975, with illustrations of hitherto unfig-
ured species. Tokyo.

MOORE, R. C. (ed). Treatise on invertebrate pale-
ontology. Part |, Mollusca 1 (1960); Part N, Mol-
lusca 6 (vol. 1, 1969; vol. 2, 1969; vol. 3, 1971);
Part W, Miscellanea (1962). The Geological So-
ciety of America & The University of Kansas.

NEAVE, S. A., 1939-1940, Nomenclator Zoologi-
cus, vols 1-2 (1939), vols 3-4 (1940), vol. 5
(1950). The Zoological Society of London, Lon-
don.

NESIS, K. N., 1987, Cephalopods of the world. TFH
Publications, Neptune City.

NICOL, D., 1958, Trends and problems in pelecy-
pod classification (the genus and subgenus).
Journal of the Washington Academy of Sciences,
48(9):285-293.

NICOL, D., 1969, The number of living species of
molluscs. Systematic Zoology, 18:251-254.

NOMENCLATOR ZOOLOGICUS. See Neave; Ed-
wards & Hopwood; Edwards & Vevers.

PONDER, W. & A. WAREN, 1988, Classification of
the Caenogastropoda and Heterostropha. A list
of the family-group names and higher taxa. Ma-
lacological Review, Supplement 4:288-326.

RUNNEGAR, B. & J. POJETA, 1985, Origin and
diversification of the Mollusca. In: E. R. TRUE-
MAN & M. R. CLARKE, eds; The Mollusca, vol.
10, Evolution, 1-58. Academic Press, Orlando.

RUSSELL, H. D., 1971, Index Nudibranchia (1554 -
1965). Delaware Museum of Natural History,
Greenville.

RUSSELL, H. D., 1986, Index Nudibranchia. Sup-
plement I, 1966-1975. Department of Mollusks,
Harvard University. Special Occasional Publica-
tion 7.

SCHILDER, F. A., 1947, Die Zahl der Prosobran-
chier in Vergangenheit und Gegenwart. Archiv für
Molluskenkunde, 76:37—44.

SCHILDER, F. A., 1949, Statistical notes on mala-
cology. Proceedings of the Malacological Society
of London, 27(6):259-261.

SCHILDER, M. & F. A. SCHILDER, 1947, Zur
Entwicklung der Molluskenkunde. Archiv für Mol-
luskenkunde, 76:163-166.

SHERBORN, C. D., 1902, 1922-1932, Index Ani-
malium. Sectio | (1758-1800), 1902, Cambridge;
Sectio II (1801-1850), 1922-1932 (29 parts).
British Museum, London.

SOLEM, A., 1978, Classification of the land mol-
lusca. In: V. FRETTER & J. PEAKE, eds, Pulmo-
nates, vol. 2A. Systematics, evolution and ecol-
ogy, 49-97. Academic Press, London.

VAN BELLE, R. A., 1975-78, Sur la classification
des Polyplacophora. Informations de la Société
Belge de Malacologie, (4)5:121-134 (1975);

86 BOUCHET 4 ROCROI

(4)6:135-152 (1975); 5(2):15-42 (1977); 6(1):
3-28 (1978); 6(2):35-44 (1978).

VAUGHT, K. C., 1989, A classification of the living
Mollusca. American Malacologists, Melbourne,
Florida.

VOKES, H. E., 1967, Genera of the Bivalvia: a sys-
tematic and bibliographic catalogue. Bulletin of
American Paleontology, 51(232):1-394.

VOKES, H. E., 1980, Genera of the Bivalvia: a sys-
tematic and bibliographic catalogue (Revised
and updated). Paleontological Research Institu-
tion, Ithaca.

VOKES, H. E., 1990, Genera of the Bivalvia: a sys-
tematic and bibliographic catalogue—Addenda

and errata. Tulane Studies in Geology and Pale-
ontology, 23(4):97—120.

WENZ, W., 1938-1944, Handbuch der Paláozoo-
logie. Band 6, Gastropoda. Teil 1, Allgemeiner
Teil und Prosobranchia. Borntraeger, Berlin.

ZILCH, A., 1959-1960, Handbuch der Paláozoo-
logie. Band 6, Gastropoda Teil 2, Euthyneura.
Borntraeger, Berlin.

ZOOLOGICAL RECORD for the years 1960-1989,
1963-1990, Part 9, Mollusca. The Zoological So-
ciety of London (1963-1980); BioSciences Infor-
mation Service and the Zoological Society of
London (1981-1990).

Revised Ms. accepted 2 January 1992

MALACOLOGIA, 1992, 34(1-2): 87-97

SHELL PATTERN POLYMORPHISM IN A 13-YEAR STUDY OF THE LAND SNAIL
THEBA PISANA (MULLER) (PULMONATA: HELICIDAE)

Robert H. Cowie

Bishop Museum, Р.О. Box 190004, Honolulu, Hawaii 96817-0916, U.S.A.

ABSTRACT

Shell pattern gene frequencies in the helicid Theba pisana at Tenby, South Wales, did not
change over a 13-year period at a range of sites differing in frequency. This apparent stability
and the differences among sites may be maintained by selection, although the selective agent(s)
is(are) unknown. Alternatively, selection may be absent (or too weak to be detected), and the
differences among sites may be random due to the founder effect; in this case, stochastic
fluctuations in gene frequency would be small in these large populations and directional changes
would not be expected. The need for long-term ecological studies in order to address such

questions is emphasized.

Key words: Pulmonata, Theba pisana, genetics, polymorphism, selection, founder effect.

INTRODUCTION

Long-term field studies are essential in ev-
olutionary biology yet are rarely possible
(Cain, 1983; Cowie, 1989). Subtle but impor-
tant changes in the composition of communi-
ties, the relative distributions of species or the
morph frequencies in polymorphic popula-
tions, or indeed the absence of such changes,
can only be demonstrated by studies pursued
over many years.

In the extensively studied European land
snails Cepaea nemoralis (L.) and C. hortensis
(Muller), stability in morph frequencies over
periods of 5—50 years, as well as significant
changes over such periods, have been found
and attributed to natural selection (Cain,
1983; Murray 8 Clarke, 1978; Wall et al.,
1980). Other studies have failed to detect se-
lection (Cain & Cook, 1989; Cain et al., 1990).
Striking changes in the relative distributions of
the two species have also been reported
(Cowie & Jones, 1987). Longer-term studies
also have demonstrated significant changes
in morph frequencies, attributable to both nat-
ural selection and historical factors (Cain,
1983; Cameron & Dillon, 1984).

The rather strict criteria for the successful
study of gene frequencies over time (Murray
8 Clarke, 1978; and see also Endler, 1986,
pp. 73—75) are that (i) accurately dated and
localised data be available; (ii) the genetic
control of the polymorphism be understood
sufficiently to allow accurate scoring of geno-
types; (iii) there be genetic continuity over the
period of the study; and (iv) the selective sig-

87

nificance of the characters be understood.
The present study of morph frequencies in the
land snail Theba pisana (Müller) fulfills all
these criteria: the samples were all taken by
the author from the same group of sites on
specific dates and using the same collecting
techniques (Cowie, 1984a, b); the genetics of
the polymorphism have been elucidated by
breeding experiments (Cowie, 1984a); sam-
ples were taken from continuously extant
populations with overlapping generations ev-
ery year from 1977 to 1989 (except 1982);
and both visual selection by predators and
climatic selection appear to influence both
this polymorphism (Hazel & Johnson, 1990;
Heller, 1981; Heller & Gadot, 1984; Johnson,
1980, 1981) and other aspects of visually de-
tectable variation in T. pisana (Cowie, 1990).

Theba pisana is a coastal Mediterranean
land snail, the distribution of which extends
along the Atlantic coasts of north Africa and
western Europe as far north as the south-
western parts of England and Wales and the
east coast of Ireland. It has been introduced
to Australia, South Africa and California. A
five-year study of morph frequencies at six
sites at Tenby, South Wales (Cowie, 1984a),
failed to demonstrate any changes in morph
frequencies over this period despite a hint
that selection was acting against the un-
banded (i.e. pale) morphs. The aims of the
present study were to confirm or discount the
possible selection and to assess the stability
or otherwise of the morph frequencies over a
longer period. All shells are deposited in the
Mollusca Section of The Natural History Mu-

88 COWIE

seum (London) (Mollusca Section accession
number: 2355).

METHODS

Samples were collected from up to six sites
between 24 May and 31 August of each year
from 1977 to 1989 (except 1982) following the
precise protocols of Cowie (1984a, b). All
snails in a sample area were collected, so no
bias in morph frequency, due to differential
collection of certain morphs, was introduced.
A map of the site locations is given by Cowie
(1984a). Theba pisana is biennial at Tenby,
breeding only once at the end of its second
year and then dying. Summer samples taken
prior to the start of the late summer/early au-
tumn breeding season therefore have dis-
tinctly bimodal size frequency distributions,
allowing easy separation of two-year-old
adults from one-year-old juveniles by simple
graphical analysis (Cowie, 1984b). Morph
scores of the two age classes could, there-
fore, be analysed separately.

The data of Cowie (1984a) have been in-
corporated in the present analysis. The regu-
lar taking of more than one sample from some
of the six sites, sometimes on more than one
occasion during the 24 May-31 August pe-
riod, was discontinued from 1983 onwards as
being redundant: one sample only was taken
from each site, and on only one occasion (be-
tween 24 May and 31 August) in each year.
From 1986 onwards, a single sample only
was taken from each of three sites, represent-
ing the three broad sampling areas at Tenby
(see Cowie, 1984a).

Morphs were scored according to Cowie
(1984a). The discussion over whether 7.
pisana is pentataeniate or tetrataeniate
(Cowie, 1984a; Gittenberger 8 Ripken, 1987;
Heller, 1981) does not affect the scoring of
the polymorphism. The polymorphism in the
Tenby populations is controlled by three loci
with two alleles showing dominance at each.
Epistasy allows only four main phenotypes to
appear: plain unbanded, dotty unbanded
(with a row of dots along the mid-line of the
shell), dark five-banded and yellow five-
banded; these are illustrated by Cowie
(1984a). There is some blurring of the bound-
aries between the dark and yellow five-
banded morphs such that it has been neces-
sary to include an “intermediate” five-banded
category (Cowie, 1984a). The blurred distinc-
tions between the five-banded categories,

combined with the variable shell size at which
the dark banding lineolations appear, allow
only limited conclusions to be drawn regard-
ing this aspect of the variation, and indeed
juveniles were scored simply as banded with-
out distinction into dark, intermediate and
yellow. However, the distinction between
banded and unbanded shells, in both adults
and juveniles, is clear since, in these popula-
tions, the apex (in fact the protoconch) of
banded shells is always darker than that of
unbanded shells. This association of banding
pattern and apex colour is due to the very
close linkage of the banded/unbanded and
apex colour loci, combined with strong link-
age disequilibrium, due presumably to selec-
tion (Cain, 1984; Cowie, 1984a). Mistakes in
scoring are considered, then, to be insignifi-
cant.

RESULTS

Morph scores for all samples are given in
Table 1. The differences between sites in the
proportion of unbandeds, indicated by Cowie
(1984a), have been maintained. Sites 2, 3
and 4, which are less than 100 m apart and
more or less connected by suitable habitat
with T. pisana present, show similar frequen-
cies, most samples having 25—45% unband-
eds (average about 34%). Sites 1, 5 and 6,
the last two of which are close together but
separated from site 1 by about 0.5 km of cliffs
with no T. pisana, also have similar frequen-
cies, most samples having 10-25% unband-
eds (average about 17%).

Percentages of unbandeds through the
course of the study at each site, for adults and
juveniles separately, are shown in Figures 1—
6. Regression analyses (Feldman et al.,
1987) of these percentage data (arcsinh
square root transformed) on year reveal no
significant change (p>0.05 in all cases) in fre-
quencies of unbandeds in either adults or ju-
veniles at any site throughout the course of
the study, either when all samples or only
those of 30 or more individuals are included
(Table 2). Morph frequencies have not
changed significantly for 13 years. Two-way
log-likelihood G-statistic analyses (Sokal &
Rohlf, 1969) of the actual frequencies of un-
bandeds by year for each site separately
(Table 3) indicate some heterogeneity among
years (as detected by Cowie (1984a)) but no
obvious trend, and do not contradict the over-
all absence of change. (Because of the lack of

POLYMORPHISM IN THEBA PISANA 89

TABLE 1. Morph scores for all samples.

Site Wet. — ======9s8sSesccoss AU MES A ECC Totale SES Juveniles-------- Total Total
---unbanded--- --------- banded--------- adults ---unbanded--- banded juveniles
plain dotty dark intermed yellow plain dotty

1 1977 48 26 258 81 20 433 45 16 360 421 854

1978 27 9 150 36 4 226 42 22 294 358 584
1979 5 2 37 if 2 53 4 4 65 73 126
1980 Y 4 43 8 5 67 6 6 66 78 145
1981 3 3 31 10 3 50 17 4 72 93 143
1983 8 2 29 1 2 42 5 0 37 42 84
1984 5 3 24 1 1 34 if 2 55 64 98
1985 4 1 22 4 3 34 2 1 24 27 61
1986 3 1 22 1 2 29 8 4 91 103 132
1988 0 0 9 1 0 10 2 0 23 25 35
1989 3 1 12 3 4 23 2 0 13 15 38
2 1977 203 42 269 63 18 609* 245 43 372 660 1269
1978 133 51 204 67 21 476 90 22 164 276 752
1979 21 12 52 9 4 98 32 11 89 132 230
1981 81 39 159 59 20 358 47 16 88 151 509
1983 24 8 36 3 11 82 32 11 57 100 182
1984 14 7 35 4 4 64 36 20 129 185 249
1985 77 28 115 10 13 243 17 3 51 71 314
1986 18 5 35 5 4 67 43 6 91 140 207
1987 43 11 60 9 17 140 22 0 63 85 225
1988 39 15 57. 12 12 135 25 11 50 86 221
1989 28 10 54 9 6 107 18 if 38 63 170
3 1977 24 12 62 20 4 122 33 12 79 124 246
1978 38 6 70 16 6 136 37 18 113 168 304
1979 1 0 1 1 0 3 23 15 88 126 129
1980 9 5 23 if 2 46 20 10 65 95 141
1981 5 2 14 4 1 26 1 0 5 6 32
1983 19 5 26 8 10 68 5 5 35 45 113
1984 18 6 50 8 13 95 30 17 90 137 232
1985 32 14 81 14 10 151 14 0 30 44 195
4 1977 51 23 104 29 8 215 66 6 106 178 393
1978 55 35 143 33 if 273 106 65 297 468 741
1979 12 5 30 5 3 55 52 15 132 199 254
1980 34 17 44 13 2 110 42 15 110 167 277
1981 26 21 83 21 11 162 16 10 43 69 231
1983 17 4 39 8 10 78 17 12 77 106 184
1984 59 9 146 9 9 232 23 15 59 97 329
1985 65 15 115 9 10 214 8 0 32 40 254
5 1977 24 Y 119 38 15 203 13 2 53 68 271
1978 14 5 78 17 9 123 20 14 157 191 314
1979 1 0 10 2 2 15 3 2 6 11 26
1980 4 3 35 8 3 53 1 2 11 14 67
1981 4 1 16 1 3 25 1 1 2 4 29
1983 4 1 11 2 2 20 1 0 11 12 32
6 1977 23 10 139 39 10 221 16 4 104 124 345
1978 19 2 87 18 if 133 20 7 219 246 379
1979 0 0 16 12 1 29 3 2 29 34 63
1981 10 2 57 21 97 21 5 94 120 217
1983 11 1 103 14 13 142 7 1 57 65 207
1984 13 8 86 15 17 139 7 1 88 96 235
1985 22 5 82 5 9 123 10 3 130 143 266
1986 11 3 61 7 4 86 12 2 89 103 189
1987 9 4 49 7 12 81 28 2 111 141 222
1988 5 1 57 15 8 86 29 12 165 206 292
1989 10 5 5% 18 11 101 33 10 253 296 397
Total 1443 Si 3807 847 415 7043 1465 494 5232 7191 14234

In certain years, samples were not collected as follows: 1980—sites 2, 6; 1982—all sites; 1984—1989—site 5; 1986—
1989—sites 3, 4; 1987—sites 1, 3, 4, 5.

*There were 14 banded adults that were not scored as dark, intermediate or yellow. These have been included in analyses
of banded/unbanded but excluded from analyses within the banded class.

90 COWIE

407 ADS SITE 1 40> JUVS SITE
% %
30 30
20 20
10 10
0 0

1977 80 83 86 89 1977 80 83 86 89

FIG. 1. Percent unbandeds in adults (ADS) and juveniles (JUVS) at site 1 for 1977-1989, with 95%
confidence limits derived using the tables of Sokal & Rohlf (1973) supplemented by Goldstein (1964).

607 ADS ПЕ 607 JUVS SITESZ
DA %
50 50
40 40
30 30
20 20
0 0

1977 80 83 86 89 1977 80 83 86 89

FIG. 2. Percent unbandeds at site 2; details as in Fig. 1.

607 ADS SITE 3 607 JUVS SITE 3
% %
50 50
40 40
30 30
20 20
10 10
0 0

1977 80 83 86 89 1977 80 83 86 89

FIG. 3. Percent unbandeas at site 3; details as in Fig. 1. (The 1979 adult value is omitted as the sample was
too small to allow calculation of confidence limits.)

samples from some sites in some years, an ing against unbandeds, at least during the
overall three-way G-statistic analysis of fre- second year of life (Cowie, 1984a), is re-
quency by year by site was not performed). jected. Comparisons of unbanded frequen-

The suggestion that selection was operat- cies in adults with those in juveniles of the

e

POLYMORPHISM IN THEBA PISANA 91

604 ADS SITE 4 607 JUVS SITE 4
% %
50 50
40 40
30 30
20 20
10 10
0 0
1977 80 83 86 89 1977 80 83 86 89

FIG. 4. Percent unbandeds at site 4; details as in Fig. 1.

807 ADS SIMESS 807 JUVS SITE 5
% %
70 70
60 60
50 50
40 40
30 30
20 20
10 10
0 0
1977 80 83 86 89 1977 80 83 86 89

FIG. 5. Percent unbandeds at site 5; details as in Fig. 1. (The 1981 juvenile value is omitted as the sample
was too small to allow calculation of confidence limits.)

407 ADS SITE 6 407 JUVS SITE 6
% %
30 30
20 20
10 10
0
1977 80 83 86 89 1977 80 83 86 89

FIG. 6. Percent unbandeds at site 6; details as in Fig. 1.

92 COWIE

TABLE 2. Values of г? and р from the regression analyses of morph frequency (transformed) on year. If
any sample(s) or sub-sample(s) (e.g. unbanded adults when considering percent plain among adults)
contained fewer than 30 individuals, the regression was also computed omitting these samples and the

results are given in parentheses.

O Adults: ====3====>="=22 ass == gsi Ds ECS Juveniles ----------
% unbanded % plain % dark % yellow % unbanded % plain
2 2 2 2 2 2
r p г р г г p r p r p
1 0.160 0.222 0.109 0.351 0.012 0.747 0.006 0.828 0.286 0.090 0.351 0.055
(0.124) (0.392) (0.305) (0.256) (0.234) (0.332) (0.103) (0.439)
2 0.001 0.930 0.021 0.674 0.009 0.785 0.356 0.053 0.095 0.356 0.001 0.934

(0.015) (0.753)

3 0.086 0.481 0.035 0.657 0.011

4 0.033 0.667 0.430 0.077 0.174
(0.418) (0.165)

0.304

(0.133) (0.375)

0.807 0.264 0.193 0.044 0.620 0.025 0.711
(0.029) (0.749) (0.064) (0.838) (0.047) (0.680) (0.588) (0.075)

(0.117) (0.453) (0.267) (0.373)

o

175 0.503 0.411 0.087 0.006 0.859
(0.414) (0.241)

5 0.281 0.279 0.000 0.972 0.228 0.338 0.219 0.349 0.101 0.540 0.006 0.888
(0.961) (0.127) 0.7.08) (0.364) (0.893) (0.212) © = OSA

6 0.096 0.354 0.071 0.458 0.033 0.593 0.469 0.020 0.012 0.750 0.092 0.364
(0.017) (0.716) (0.015) (0.732) (0.403) (0.049) (0.472) (0.518)

“There were only two samples with 30 or more juveniles at site 5, so a regression based on just these two points has not

been calculated.

previous year (i.e. the population from which
those adults were derived) at each site (40
possible comparisons) do not show a signifi-
cant trend; the points in Figure 7 fall more or
less evently on either side of the line of equal
percent unbanded, and a Wilcoxon's signed
ranks test detected no significant difference
(p>0.1) in the frequencies of unbandeds be-
tween these adult and juvenile samples. (On
checking the data on which Table 3 of Cowie
(1984a) is based, the percent unbanded in
that table for juveniles in 1979 at site 1 was
found to be incorrect; the correct value is
11.0, not 32.6. Recalculation of the Wilcox-
on's tests of Cowie (1984a) but incorporating
this correction, in fact, indicates no significant
difference for the comparison of adults with
juveniles of the previous year—p>0.05; al-
though the less biologically meaningful com-
parison of adults and juveniles of the same

year remains significant, but less strongly
so—p<0.05.)

Within the unbandeds, there are no appar-
ent differences among sites in the proportion
of plain as opposed to dotty shells, most sam-
ples having 60-85% plain (average about
74%). As above, regression analysis (Table
2) detected no significant trends, while G-sta-
tistic analysis (Table 3) indicated some heter-
ogeneity among years but without suggesting
any trends. Again, morph frequencies show
no consistent change.

Within the bandeds, there are no apparent
differences among sites in the proportions of
dark, intermediate and yellow shells. Most
samples had 65-85% dark and 0-20% yel-
low (average about 75% and 9%, respec-
tively), with the remainder made up of inter-
mediates. Regression analysis revealed a
single significant change in morph frequen-

POLYMORPHISM IN THEBA PISANA 93

TABLE 3. G-statistic analysis of morph frequencies by year.

Site G dere p

adults juveniles adults juveniles adults juveniles

unbanded vs. banded

1 4.714 9.800 9% 10 n.s. ns.
2 92694 27.372 10 10 nes. <0.01
3 1215 4.426 7 7 n.s. n.s.
4 14.446 11.010 7) 7 <0.05 ASS
5 22890 7.430 5 5 nes: n.s.
6 17.646 23.762 9* 10 <0.05 <0.01
plain уз. dotty
i 3.502 7.340 I 6* 8 n.s.
2 1162858 118.416 10 9* n.s. <0°05
3 6.164 2.942 6* 5* n.s. n.s.
4 24.900 30.184 Y 6* <0/. 001, .<0/001
5 0.866 4.794 qx 3% n.s. n.s.
6 95730 9.030 9 10 ns. n.s.
dark vs. intermediate/yellow

il 17.248 = 9* = <0.05 >
2 22.400 = 10 = <0.05 =
3 5.690 - 6* > n.s. =
4 252072 - 1 - <0.001 =
5 2.180 = 5 = n.s. =
6 24.520 = 10 - <0.01 -

*Some years pooled to give adequate cell values.

94 COWIE

% UNBANDED — JUVENILES

10 20

30

60 70

40

50

“fo UNBANDED — ADULTS

FIG. 7. Scatter diagram of percent unbanded in adult samples against juvenile samples of the previous year
(see text for explanation). The diagonal indicates equal percentages. Open circles represent cases in which
the adult or juvenile sample, or both, contained fewer than 30 individuals.

cies—an increase in percent yellow at site
6—which is only just significant (p = 0.049)
when those samples of fewer than 30 individ-
uals are omitted from the analysis (Table 2).
This one significant result can be attributed to
chance, given the large number of regression
analyses performed, and does not indicate a
real change. G-statistic analysis (Table 3, with
frequencies of yellow and intermediate shells
combined) again indicates heterogeneity
among years (including at site 6, where the
regression analysis indicated a significant
trend) but this probably also reflects stochas-
tic effects associated with multiple testing,
combined with the difficulty, and consequent
inconsistency, of scoring morphs within the
banded class (see above). It is reasonable to
conclude, once again, that there has been no
consistent change in morph frequency over
the course of the study.

In the absence of any trends, selection co-
efficients have not been calculated (cf. Mur-
ray & Clarke, 1978; Wall et al., 1980).

DISCUSSION

Theba pisana was probably introduced ar-
tifically to the U.K. during the eighteenth cen-
tury (Cowie, 1984a; Turk, 1966, 1972). Cowie
(1984a) suggested that selection acting since
that time (over about 100 generations) may
have given rise to the characteristic appear-
ance of the morphs at Tenby and nearby lo-
calities in South Wales. While this may or may
not be true, the current differences among
sites in morph frequency, yet the lack of de-
tectable morph frequency change over the
13 years of the study, can be explained in a
number of ways (cf. Endler, 1986: 73—75), in-
volving both selection and stochastic pro-
cesses:

(1) Directional selection, following the initial
introduction, produced the differences among
sites; these differences are now being main-
tained by stabilising selection. While visual
selection by predators and climatic selection
are known to influence variation in shell pat-

POLYMORPHISM IN THEBA PISANA 95

tern and colour of soft parts in T. pisana
(Cowie, 1990; Hazel & Johnson, 1990; Heller,
1981; Heller & Gadot, 1984; Johnson, 1980,
1981), as in the better known helicids Cepaea
nemoralis and C. hortensis (Cain, 1983;
Cowie & Jones, 1985; Jones et al., 1977), it is
difficult to see what selective differences exist
among the three sampling areas (site 1, sites
2, 3 and 4, sites 5 and 6) because all occur
along a short length (about 0.75 km) of cliff
face in apparently fairly similar habitat; any
putative selective agent(s) is(are) unknown.
Nevertheless, subtle and arcane habitat dif-
ferences may be important, as has been sug-
gested as one of a number of explanations of
“area effects” in Cepaea (Cain & Currey,
1963a, b), and an influence of selection on
morph frequencies in T. pisana at Tenby can-
not be ruled out.

(2) The differences among sites arose from
founder effects; frequent extinction of local
populations, followed by recolonisation, at
this, the climatic edge of the species’ range
(Cowie, 1987), leads to morph frequencies
being the results of a succession of founding
events; directional selection operating simi-
larly at all sites has never been able to bring
the populations to equilibrium. None of the
present populations became extinct during
the course of the study, although numbers
were low at sites 1 and 5; selection may not
be sufficiently strong to have been detected
during the 13 years of the study. The sugges-
tion that apparently strong directional selec-
tion was operating over all sites against un-
bandeds, and perhaps against plain morphs
within the unbanded class (Cowie 1984a), is
rejected, with the explanation that the initial
five-year study was not long enough and the
apparent selection was a statistical artefact,
exagerated by a computational error (see
above). At this cold extreme of the range of T.
pisana, selection for darker colour (i.e. in-
creased banding) might be expected, and in-
deed, at all other localities in tnis part of South
Wales (Cowie, 1986), only five-banded (no
unbanded) snails are found (Cowie, 1984a,
1987; Fowles & Cowie, 1989). (The absence
of unbandeds at these other localities may
also be due to the founder effect if these
smaller populations are all derived ultimately
from the much larger and well-established
Tenby populations via small propagules con-
sisting only of banded snails—Cowie, 1984a;
Fowles & Cowie, 1989.)

(3) The differences among sites are due to
founder effects; selection is not operating;

and the differences are being maintained, or
at least changing only slowly, because large
population sizes prevent significant genetic
drift. Cowie (1984c) estimated effective
neighbourhood numbers as ranging up to
4130 individuals, large enough for drift to be
insignificant.

It is not possible to distinguish between
these various scenarios. Even though at least
some aspects of the biology of T. pisana, es-
pecially at Tenby, are fairly well understood,
there remain major gaps in our knowledge of
the selective pressures to which the snails are
subject, pressures that will differ from locality
to locality, so that studies of T. pisana in Israel
(Heller, 1981; Heller & Gadot, 1984), South
Africa (Hickson, 1972) and Australia (Baker &
Vogelzang, 1988; Hazel & Johnson, 1990;
Johnson, 1980, 1981) can only be of broad,
and not specific, relevance to our understand-
ing of factors controlling the polymorphism in
South Wales. Furthermore, Cain & Cook
(1989) and Cain et al. (1990), studying Ce-
paea nemoralis over periods of 18 years and
23 years, respectively, have indicated the
considerable difficulties involved in distin-
guishing among systematic changes in gene
frequencies due to selection, maintenance of
unchanging frequencies by selection, and
stochastic fluctuations (see also Endler,
1986, ch. 4); Cain & Cook (1989) suggest that
studies of 250 years or more may be neces-
sary. The unanswered questions posed by
the present study, reinforce the view (Endler,
1986; Wade & Kalisz, 1990) that only with
detailed ecological knowledge, necessarily in-
volving long-term studies, combined with ex-
perimentation, can the factors influencing
polymorphisms in natural populations be thor-
oughly understood.

ACKNOWLEDGEMENTS

This work was begun at Liverpool Univer-
sity and continued, in part, at University Col-
lege London. Scoring of post-1982 samples,
new analyses presented here and writing up
were done at the Bishop Museum. | thank
these institutions for the opportunities and fa-
cilities provided. | also thank Professor A. J.
Cain and Dr. M. S. Johnson for comments on
the manuscript, Dr. L. A. Mehrhoff for compu-
tational and statistical assistance and the var-
ious friends and colleagues, especially C. T.
French and A. M. Cassin, who assisted with
field work.

96 COWIE

LITERATURE CITED

BAKER, G. H. 8 B. K. VOGELZANG, 1988,
Life history, population dynamics and poly-
morphism of Theba pisana (Mollusca: Heli-
cidae) in Australia. Journal of Applied Ecol-
ogy, 25: 867-887.

CAIN, A. J., 1983, Ecology and ecogenet-
ics of terrestrial molluscan populations. In: W.
D. RUSSELL-HUNTER, ed., The Mollusca. Vol-
ume 6. Ecology. Academic Press, London.
pp. 597-647.

CAIN, A. J., 1984, Genetics of some
morphs in the land snail Theba pisana. Mal-
acologia, 25: 381-411.

CAIN, A. J., 4 L. M. COOK, 1989, Persis-
tence and extinction in some Cepaea popu-
lations. Biological Journal of the Linnean So-
ciety, 38: 183-190.

CAIN, A. J., L. M. COOK & J. D. CURREY,
1990, Population size and morph frequency in
a long-term study of Cepaea nemoralis. Pro-
ceedings of the Royal Society of London, Se-
ries B, Biological Sciences, 240: 231-250.

CAIN, A. J., 8 J. D. CURREY, 1963a, Area
effects in Cepaea. Philosophical Transactions
of the Royal Society of London, Series B, Bi-
ological Sciences, 246: 1-81.

CAIN, A. J., 4 J. D. CURREY, 1963b, The
causes of area effects. Heredity, 18: 467-471.

CAMERON, R. A. D. & P. J. DILLON, 1984,
Habitat stability, population histories and pat-
terns of variation in Cepaea. Malacologia, 25:
271-290.

COWIE, R. H., 1984a, Ecogenetics of
Theba pisana (Pulmonata: Helicidae) at the
northern edge of its range. Malacologia, 25:
361-380.

COWIE, R. H., 1984b, The life-cycle and
productivity of the land snail Theba pisana.
Journal of Animal Ecology, 53: 311-325.

COWIE, R. H., 1984c, Density, dispersal
and neighbourhood size in the land snail
Theba pisana. Heredity, 52: 391-401.

COWIE, R. H., 1986, Reduction in the dis-
tribution of the land snail Theba pisana in the
Tenby area, Dyfed. Nature in Wales, New Se-
ries, 4: 66-70.

COWIE, R. H., 1987, Rediscovery of Theba
pisana at Manorbier, South Wales. Journal of
Conchology, 32: 384-385.

COWIE, R. H., 1989, Of moths, snails and
population genetics. Bulletin of the British
Ecological Society, 20: 266-269.

COWIE, R. H., 1990, Climatic selection on
body colour in the land snail Theba pisana
(Pulmonata: Helicidae). Heredity, 65:123-
126.

COWIE, R. H. & J. S. JONES, 1985, Cli-
matic selection on body colour in Cepaea.
Heredity, 55: 261-267.

COWIE, R. H. & J. S. JONES, 1987, Eco-
logical interactions between Cepaea nemo-
ralis and Cepaea hortensis: competition, in-
vasion but no niche displacement. Functional
Ecology, 1: 91-97.

ENDLER, J. A., 1986, Natural selection in
the wild. Princeton University Press, Prince-
ton. xiii + 336 pp.

FELDMAN, D. S., R. HOFFMANN, J. GAG-
NON & J. SIMPSON, 1987, Statview II. Aba-
cus Concepts, Inc., Berkeley.

FOWLES, A. P. & R. H. COWIE, 1989,
Westward extension of the distribution of
Theba pisana in South Wales. Journal of
Conchology, 33: 186-187.

GITTENBERGER, E. & T. E. J. RIPKEN,
1987, The genus Theba (Mollusca: Gas-
tropoda: Helicidae), systematics and distribu-
tion. Zoologische Verhandelingen, 241: 1-59.

GOLDSTEIN, A., 1964, Biostatistics: an in-
troductory text. Macmillan, New York, ix +
272 pp.

HAZEL, W. N. & M. S. JOHNSON, 1990,
Microhabitat choice and polymorphism in the
land snail Theba pisana (Múller). Heredity,
65: 449—454.

HELLER, J., 1981, Visual versus climatic se-
lection of shell banding in the landsnail Theba
pisana in Israel. Journal of Zoology, 194: 85—
101.

HELLER, J. 4 M. GADOT, 1984, Shell poly-
morphism of Theba pisana—the effects of ro-
dent distribution. Malacologia, 25: 349-354.

HICKSON, T., 1972, A possible case of ge-
netic drift in colonies of the land snail Theba
pisana. Heredity 45: 7-14.

JOHNSON, M. S., 1980, Association of
shell banding and habitat in a colony of the
land snail Theba pisana. Heredity, 45: 7-14.

JOHNSON, М. S., 1981, Effects of migra-
tion and habitat choice on shell banding fre-
quencies in Theba pisana at a habitat bound-
ary. Heredity, 47: 121-133.

JONES, J. S., B. H. LEITH & P. RAWE
INGS, 1977, Polymorphism in Cepaea: a
problem with too many solutions? Annual Re-
view of Ecology and Systematics, 8: 109—
143.

MURRAY, J. J. 4 B. C. CLARKE, 1978,
Changes of gene frequency in Cepaea nem-
oralis over fifty years. Malacologia, 17: 317—
330.

SOKAL, R. В. & Е. J. ROHLF, 1969, Biom-
etry. Freeman, San Francisco, xxi + 776 pp.

POLYMORPHISM IN THEBA PISANA 97

SOKAL, В. В. & Е. J. ROHLF, 1973, Intro-
duction to biostatistics. Freeman, San Fran-
cisco, xiii + 368 pp.

TURK, S. M., 1966, The rediscovery of
Theba pisana (Müller) in Cornwall. Journal of
Conchology, 26: 19-25.

TURK, S. M., 1972, A study of the white
sand-hill snail, Theba pisana. Devon Trust for
Nature Conservation Journal, 4: 138-142.

WADE, M. J. & S. KALISZ, 1990, The
causes of natural selection. Evolution, 44:
1947-1955.

WALL, S., М. A. CARTER & В. С.
CLARKE, 1980, Temporal changes of gene
frequencies in Cepaea hortensis. Biological
Journal of the Linnean Society, 14: 303-317.

Revised Ms. accepted 16 December 1991

MALACOLOGIA, 1992, 34(1-2): 99-106

NULL ALLELES AND HETEROZYGOTE DEFICIENCIES AMONG
MUSSELS (MYTILUS EDULIS AND M. GALLOPROVINCIALIS)
OF TWO SYMPATRIC POPULATIONS

J. P. A. Gardner’

School of Biological Sciences, University College of Swansea, Swansea, SA2 8PP,
United Kingdom

ABSTRACT

The frequencies of null (enzymatically inactive) alleles estimated from two-banded pheno-
types at an esterase-D (dimeric enzyme) locus among two sympatric Mytilus edulis/Mytilus
galloprovincialis populations (Croyde and Whitsand) in southwestern England are 0.00823 and
0.00751 respectively. These values fall in the range of null allele frequencies predicted for British
mussel populations. At Croyde, there were significantly more M. galloprovincialis than M. edulis
null heterozygotes. This is thought to result from linkage of M. edulis null alleles to genes at a
selective disadvantage compared with M. galloprovincialis, which indirectly reduces null allele
frequency. Observed and predicted frequencies of null alleles are used to test the hypothesis
that heterozygote deficiencies (compared to Hardy-Weinberg expectations), which are often
observed among marine molluscan populations, might be caused by null alleles.

Key words: null alleles, heterozygote deficiency, Mytilus edulis, Mytilus galloprovincialis.

INTRODUCTION

Null (enzymatically inactive) alleles at loci
encoding enzymes have been reported in
populations of plants (Allendorf et al., 1982;
Lafiandra et al., 1987) and animals (Voelker
etal., 1980a, b; Langley et al., 1981; Skibinski
et al., 1983; Foltz, 1986; Katoh & Foltz, 1989).
Null alleles usually occur only rarely, often
<0.005 (Ayala et al., 1972; Voelker et al.,
1980a; Langley et al., 1981; Allendorf et al.,
1982), although higher frequencies have
been reported (Selander & Yang, 1969;
Coyne & Felton, 1977; Freeth & Gibson,
1985).

In southwestern England, the marine mus-
sels Mytilus edulis and M. galloprovincialis
occur sympatrically and hybridize (Skibinski,
1983; Gardner & Skibinski, 1988). The sys-
tematic status of M. galloprovincialis has
been reviewed by Seed (1978), Gosling
(1984) and Koehn (1991). One of the best
methods for distinguishing between the two
mussel types is electrophoresis of an es-
terase-D (Est-D) locus, which has a genetic
identity (/ value; Nei, 1972) of 0.039 in com-
parisons between M. edulis from south Wales
and M. galloprovincialis from Italy (Skibinski
et al., 1980). Est-D is a dimeric enzyme

(Harris & Hopkinson, 1976) that produces a
three-banded heterozygote phenotype. It is
a particularly good locus for observing null
alleles not only because it is the most effec-
tive locus for detecting differences between
the two mussel types, but also because two-
banded null allele phenotypes are easily
scored. However, estimates of this sort will
tend to underestimate overall null allele
frequency if the null allele produces a protein
that fails to dimerize with the normal protein,
that is, no heterodimer is produced, or if the
heterodimer is formed but is not enzymatically
expressed, that is, it is not visible on the gel.
Although the frequency of such events can
be determined by pedigree analysis and
biochemical methods, these techniques gen-
erally preclude investigations of the large
numbers (several thousands) of individuals
necessary for population studies.

A phenomenon often observed among ma-
rine molluscan populations is a significant
deficit of heterozygotes compared to Hardy-
Weinberg expectations (Milkman & Beaty,
1970; Berger, 1983; Skibinski et al., 1983;
Singh & Green, 1984; Zouros & Foltz, 1984;
Gaffney et al., 1990). It has been suggested
that null heterozygotes are one of the possi-
ble explanations for this deficit (Skibinski et

“Present address: Marine Sciences Research Laboratory, Memorial University of Newfoundland, St. John's, Newfoundland,

A1C 5S7, Canada

100 GARDNER

TABLE 1. Allele distributions in pure M. edulis and M. galloprovincialis populations, after Skibinski et al.

(1980) and Sanjuan et al. (1990).

Allele M edulis' M. edulis?
Est-D
76 0 0
82 0 0.005
90 0.036 0.024
93 0 0.001
100 0.946 0.941
103 0 0
107 0.004 0.004
110 0.015 0.020
118 0 0.003
Odh
80 0
100 0.020
15 0.930
129 0.050
140 0

M. galloprovincialis? M. galloprovincialis*

0.019 0
0.048 0.038
0.905 0.950
0 0
0.016 0.013
0.013 0

0 0

0 0

0 0
0.012

0.396

0.131

0.459

0.004

Mean allele frequency of M. edulis from Holland and Denmark (Sanjuan et al., 1990).

2Allele frequencies of M. edulis from Swansea, South Wales (Skibinski et al., 1980).

3Mean allele frequency of M. galloprovincialis from Spanish Mediterranean (Sanjuan et al., 1990).
4Allele frequencies of M. galloprovincialis from Venice, Italy (Skibinski et al., 1980).

al., 1983; Beaumont et al., 1985; Mallet et al.,
1985; Gaffney et al., 1990). Despite the fact
that few data exist concerning null allele fre-
quencies, it is usually assumed that null alle-
les do not occur at sufficiently high frequen-
cies to explain the observed heterozygote
deficits.

In this paper, genotype-dependent differ-
ences in the frequencies of two-banded null
heterozygotes are reported. Observed and
predicted frequencies of null alleles are ap-
plied to data from mussel populations to test if
null alleles can explain heterozygote deficien-
cies.

MATERIALS AND METHODS

During the period from October 1986 to
June 1989, 3,159 mussels from Croyde Bay,
north Devon, and 2,929 mussels from Whit-
sand Bay, south Cornwall, southwestern En-
gland, were collected from locations as previ-
ously reported (Skibinski, 1983; Gardner &
Skibinski, 1988). Mussels were electro-
phoresed, and stained for the esterase-D
(EST-D; EC 3.1.1.1) and octopine dehydro-
genase loci (ODH; EC 1.5.1.11). Some mus-
sels were also stained for the mannose phos-
phate isomerase locus (MPI; EC 5.3.1.8).
Tissue samples were prepared from hepato-
pancreas (liver) homogenized in an equal vol-

ume of distilled water. Following centrifuga-
tion at 0°C and 3000 rpm for 15 min, the
supernatant was used as the enzyme source.
EST-D, ODH and MPI were run on a Tris-citric
acid (pH 6.9) buffered gel (Grant & Cherry,
1985). The following staining methods were
employed; that of Ahmad et al. (1977) for
EST-D, Beaumont et al. (1980) for ODH, and
Grant & Cherry (1985) for MPI.

The migratory position (identity) of a null
allele cannot be determined because it can be
anodal or cathodal to the two-banded null
phenotype. Similarly, the identity of the ex-
pressed allele in the heterozygous state with
the null allele cannot be determined because
it is unknown which band of the phenotype is
the allele and which the heterodimer. It is
therefore not possible to determine individual
null allele frequencies, nor to determine the
frequencies of Est-D alleles in the heterozy-
gous state with the null alleles. For this rea-
son, the synthetic compound allele system
described by Skibinski (1983) was used to
classify null genotypes. At a given locus, the
compound E allele is obtained by pooling
those alleles at highest frequency in M. edu-
lis, and the compound G allele by pooling
those alleles at highest frequency in M. gal-
loprovincialis. For example, the Est-D* allele
occurs at frequencies of 0.01 and 0.96 at
Bude and Swansea, “pure” M. galloprovincia-
lis and M. edulis populations, respectively

NULL ALLELES AMONG HYBRID MUSSELS 101

(Edwards 8 Skibinski, 1987). The allele dis-
tributions among the two mussel types (Table
1) have been established by Skibinski et al.
(1980) and Sanjuan et al. (1990). Although
most alleles at a locus occur in both mussel
types, the frequency differences are quite
pronounced for the most common alleles.
Frequency differences of rare alleles are less
pronounced, but rare alleles contribute less
often to compound alleles. The compound
genotype of Est-D null heterozygotes was es-
timated by reference to the genotype at the
Odh locus, which is in tight linkage disequilib-
rium with the Est-D locus. There are signifi-
cant excesses of the E/E E/E (M. edulis-like)
and G/G G/G (M. galloprovincialis-like) dilo-
cus compound genotypes (Skibinski, 1983;
Gardner & Skibinski, 1988). Mussels with Est-
D null alleles were classified as M. edulis-like
if they possessed the Odh E/E compound
genotype, as intermediate or putative F1 hy-
brid if they exhibited the Odh E/G compound
genotype, and as M. galloprovincialis-like if
they possessed the Odh G/G compound geno-
type. For one null heterozygote, genotype
was assigned on the basis of the compound
genotype at the Mpi locus (which is also in
tight linkage disequilibrium with the Est-D lo-
cus; Gardner & Skibinski, 1988) because the
Odh genotype was not easily scored. Analy-
sis was Carried out to determine if null alleles
occur at significantly greater frequency
among M. edulis, M. galloprovincialis, or hy-
brid mussels.

Finally, null allele frequencies for three loci
were predicted for the Croyde and Whitsand
populations from the equation Fr = 2x/
(1 +x) (where x is null allele frequency and F,-
is a measure of deviation from Hardy-Wein-
berg expectations in the compound popula-
tion obtained by pooling all subdivisions) if
Hardy-Weinberg equilibrium (HWE) is as-
sumed and if null heterozygotes are scored
as homozygotes and null homozygotes are
inviable. F,, estimates have been obtained for
British M. edulis and M. galloprovincialis pop-
ulations, and for “pure” M. edulis populations
by Skibinski et al. (1983). The observed Est-D
two banded null phenotype frequencies and
the predicted null allele frequencies were both
applied to estimates of the D statistic (D =
(H,/H,)-1, where H, is the observed, and H,
the expected heterozygote frequencies),
which is commonly used to estimate devia-
tions from Hardy-Weinberg expectations (a
negative D value indicates a deficiency of het-
erozygotes compared with expectations). D

Statistics for the Croyde and Whitsand popula-
tions are given by Gardner & Skibinski (1988),
and have been recalculated to determine if
heterozygote deficiencies are attributable to
observed and predicted null allele frequen-
cies. New estimates of H, have been calcu-
lated by multiplying the original H, values of
Gardner & Skibinski (1988) by 1+ x, where x
is null allele frequency. Null allele frequencies
(x values) give a direct estimate of the under-
estimation of observed heterozygote fre-
quency and the overestimation of observed
homozygote frequency. New H, estimates
should be greater than the original H, esti-
mates because with the inclusion of null het-
erozygotes, heterozygosity will increase
when mussels incorrectly scored as homozy-
gotes are now correctly scored as heterozy-
gotes. However, the original estimates of H,
have been used because the identities of the
null alleles remain unknown, so new esti-
mates of H, have not been calculated. This
then gives a conservative test, in the sense
that (1) the calculated D values are slightly
greater than they should be, that is, the het-
erozygote deficiency at each locus is esti-
mated to be slightly smaller than it really is,
and (2) the corresponding X? and P (signifi-
cance) values are slightly less significant than
they should be. Because the method of esti-
mating null allele frequency assumes HWE,
the Fr values of Skibinski et al. (1983) for
“pure” M. edulis populations (which, on the
whole, are in HWE), and Ет values for M.
edulis and M. galloprovincialis populations
(which are in HWE less often) have both been
employed (Table 3).

RESULTS

At Croyde, two-banded null allele pheno-
types occurred at an average frequency of
0.00823. There were 1.93 times as many M.
galloprovincialis null heterozygotes as ex-
pected, based upon an hypothesis of equal
null allele frequency for all genotypes
(G=10.508, df=2, P=0.003). The E/E and
E/G compound genotypes showed null het-
erozygote frequencies that were 55.1% and
53.8% of their expected frequencies (Table
2). At Whitsand, null alleles were observed at
a mean frequency of 0.00751. There was no
significant association between null heterozy-
gote frequency and compound genotype (G
= 0.816, df = 2, P = 0.672). There were
non-significant excesses of E/G and G/G, and

102 GARDNER

TABLE 2. Frequencies of compound genotypes and null heterozygotes at an esterase-D locus for

mussels from Croyde and Whitsand.

Site

Compound genotype E/E
No. of null heterozygotes 6
No. of mussels without null alleles 1473
Observed % frequency of null heterozygotes
Expected number of null heterozygotes*

0.473

12.104

Croyde Whitsand
E/G G/G E/E E/G G/G
3 17 7 7 8

647 1013 1194 831 822

0.462 1.650 0.583 0.835 0.899

5.317 8.324 8.968 6.242 6.625

*assuming equal frequencies of null heterozygotes for all compound genotypes

a non-significant deficit of E/E null heterozy-
gotes compared with expectations (Table 2).
Inter-population comparisons reveal that
there were no significant differences between
the two populations in the total number of null
heterozygotes, or in the frequencies of M.
edulis null heterozygotes, hybrid (E/G) null
heterozygotes, and M. galloprovincialis null
heterozygotes.

D values calculated by Gardner & Skibinski
(1988) for the Ap, Est-D and Pgi loci at
Croyde and Whitsand are shown in Table 3.
In all cases, there were significant deficien-
cies of heterozygotes compared with expec-
tations. Recalculation of D for the Est-D locus
based upon the observed two-banded null
heterozygote frequencies does not reduce
the significance of the deficiencies at either
site. Changes in the associated X? values are
very small and do not have a significant effect
upon probability levels (Table 3). With the ex-
ception of the Ap locus at Whitsand, the same
is true when predicted null allele frequencies
(firstly for “pure” M. edulis populations, and
secondly for M. edulis and M. galloprovincialis
populations) estimated from the equation Fır
= 2x/(1+x) are used to re-calculate D val-
ues of the Ap, Est-D and Pgi loci.

DISCUSSION

Est-D null heterozygotes occur significantly
more often among M. galloprovincialis than
among M. edulis at Croyde. This is in line with
the findings of Skibinski et al. (1983), who
predicted an Est-D null allele mean frequency
of 0.018 in M. edulis populations and a mean
frequency of 0.071 in M. edulis and M. gallo-
provincialis populations. Null alleles most fre-
quently result from single point mutations
(Schwartz & Sofer, 1975; Voelker et al.,

1980b; Scallon et al., 1987), but can also be
due to small insertions that diminish transcrip-
tion (Burkhart et al., 1984; Gibson & Wilks,
1989; but see Jiang et al., 1988). Two sepa-
rate explanations can account for the lower
frequency of nulls in M. edulis than in M. gal-
loprovincialis. First, the M. edulis genome
might be less susceptible to the mutational
events that give rise to null alleles. Second,
tight linkage of M. edulis nulls to other loci in
the M. edulis genome, which experiences a
significantly higher age-dependent mortality
rate than M. galloprovincialis (Gardner 4 Ski-
binski, submitted), might indirectly reduce the
frequency of nulls in M. edulis. At Croyde,
comparison of the mean shell length of M.
edulis versus M. galloprovincialis with null al-
leles (mean + SD of 21.2 mm + 6.20, n = 6,
versus 39.7 mm + 4.02, n = 17) indicates
that M. edulis null alleles occur among
smaller, younger mussels. No M. edulis larger
than 30 mm length were found with null al-
leles, supporting the suggestion that in-
creased mortality experienced by M. edulis
compared to M. galloprovincialis acts indi-
rectly to reduce null allele frequency among
M. edulis. This second explanation is pre-
ferred because of supporting data, whereas
the first explanation requires the invoking of
differential mutation rates, for which there
is no direct evidence. Evidence of length-
dependent change in genotype frequencies
(Gardner & Skibinski, 1988) indicates that M.
edulis viability with regard to M. galloprovin-
cialis is greater at Whitsand than at Croyde,
providing an explanation for the non-signifi-
cant differences in genotype-dependent null
allele frequencies at Whitsand.

Few estimates of null allele frequencies ex-
ist for marine molluscan populations. Gaffney
et al. (1990) estimated mean null allele fre-
quencies in the coot clam, Mulinia lateralis, to

NULL ALLELES AMONG HYBRID MUSSELS

103

TABLE 3. Recalculation of D values to take into consideration (a) the observed two banded null
heterozygote frequencies at the esterase-D locus, and the predicted frequencies of null heterozygotes at
three loci for (b) “pure” M. edulis populations, and (c) for M. edulis and M. galloprovincialis populations.

CROYDE

Ap Est-D Pgi
‘Ho 0.379 0.196 0.457
‘He 0.462 0.505 0.651
'D 079? —0:611%% —0.298***
'n 422 474 501
2x — 0.00823 —
2Но — 0.198 —
“He — 0.505 —
=D — —0.608*** —
2n — 474 —
EX 0.0101 0.0304 0.0163
Ho 0.383 0.202 0.465
“He 0.462 0.505 0.651
=D TALE —0.600*** —0:287°°”
®n 422 474 501
x 0.0225 0.2903 0.0482
“Но 0.388 0.253 0.479
‘He 0.462 0.505 0.651
4D = 0.161" —0.499*** —0.264***
4n 422 474 501

WHITSAND
Ap Est-D Pgi
0.417 0.277 0.516
0.493 0.485 0.687
= 051.535 —0.430*** =0:250" =;
381 376 411
— 0.00751 —
= 0.284 —
= 0.485 —
— 0.4252 —
= 376 —
0.0101 0.0304 0.0163
0.422 0.285 0.524
0.493 0.485 0.687
—0.145* —0.413*** = 0.2375
381 376 411
0.0225 0.2903 0.0482
0.427 0.357 0.541
0.493 0.485 0.687
=0134NS —0.264*** 0.2137
381 376 411

x null allele frequency
Ho observed number of heterozygotes
He expected number of heterozygotes

D heterozygote deviation compared with Hardy-Weinberg expectations, calculated as D = (Ho/He)-1

NS not significant

*P<0.05

***P<0.001

n number of mussels

Data from Gardner & Skibinski (1988).

2Data from Gardner & Skibinski (1988) recalculated incorporating the observed frequencies of two banded null heterozygote

phenotypes.

SData from Gardner and Skibinski (1988) recalculated incorporating frequencies of predicted null alleles for “pure” M. edulis

populations (Skibinski et al., 1983).

“Data from Gardner & Skibinski (1988) recalculated incorporating frequencies of predicted null alleles for M. edulis and M.

galloprovincialis populations (Skibinski et al., 1983).

be 0.001 at the GPI locus and 0.0003 at the
AP2 locus. In common with other large stud-
ies, the null allele frequencies reported by
Gaffney et al. (1990) are for two-banded phe-
notypes of dimeric enzymes, and as in the
present study, must therefore be viewed as
minimum estimates. Because of the difficulty
of distinguishing one-banded null heterozy-
gotes from homozygotes, no null allele
frequency data are given for monomeric en-
zymes for the coot clam. No null heterozy-
gotes were observed at the other dimeric loci.
Skibinski et al. (1983) predicted that null allele
frequency in British populations of M. edulis
and M. galloprovincialis should be in the
range 0.003 to 0.102 for 11 allozyme loci. The

observed null allele frequencies reported here
(0.00823 at Croyde and 0.00751 at Whitsand)
are therefore consistent with predicted values
over several loci (Skibinski et al., 1983).
Mean null allele frequency estimated from
the equation F,; = 2x/(1 +x) for seven Est-D
alleles is 0.102 for M. edulis and M. gallopro-
vincialis, and for the pure M. edulis is 0.019
(Skibinski et al., 1983); that is, both values are
substantially greater than the observed fre-
quencies of two-banded null heterozygotes.
This might be accounted for by the undetect-
able presence of null alleles that do not pro-
duce a heterodimer, such individuals being
scored as homozygotes. This is particularly
important because it is often suggested that

104 GARDNER

null alleles might be responsible for observed
heterozygote deficiencies (Boyer, 1974; Ski-
binski et al., 1983; Beaumont et al., 1985;
Mallet et al., 1985; Foltz, 1986; Gaffney et al.,
1990). However, estimates from natural pop-
ulations suggest that null alleles are not com-
mon enough, often by orders of magnitude, to
account for homozygote excesses, and recal-
culation of D statistics (Table 3) seems to con-
firm this. D values increase (heterozygote de-
ficiencies decrease) in all cases as expected
when observed or predicted null allele fre-
quencies are used to re-estimate observed
and expected heterozygote frequencies.
However, changes in significance levels are
generally small and only effect probability val-
ues in one case. This indicates that null alle-
les generally do not occur at sufficient fre-
quency to significantly effect heterozygote
deficiencies. At Whitsand, the heterozygote
deficiency at the Ap locus reported by Gard-
ner & Skibinski (1988) is small but significant
(X? = 4.389, df = 1, P = 0.0362). Recalcu-
lation of D, adjusting for predicted null allele
frequency among “pure” M. edulis popula-
tions, reduces the X? value to 3.909 (df = 1,
P = 0.048), that is, to borderline significance.
Further recalculation adjusting for null allele
frequency among M. edulis and M. gallopro-
vincialis populations reduces the X? value to
3.361 (df = 1, P = 0.0668), that is, a non-
significant deficiency of heterozygotes. It
seems likely that only in such cases as this,
where heterozygote deficiencies are small but
(borderline) significant, can observed or pre-
dicted null allele frequencies contribute signif-
icantly to heterozygote deficiencies.

In the single case among marine bivalves
where very high frequencies of null alleles
have been observed (Foltz, 1986), it was sug-
gested (Gaffney et al., 1990) that the produc-
ton of aneuploid gametes (i.e., gametes with
one or more chromosomes, or chromosome
segments, absent) might better explain the
high frequency of apparent null alleles. There
is thus still no evidence from natural popula-
tions to indicate that heterozygote deficien-
cies result solely from high frequencies of null
alleles. Although null alleles appear likely to
contribute to heterozygote deficiencies, more
significant causes include selection, the
Wahlund effect, aneuploidy or imprinting, that
is, the differential expression of genetic ma-
terial, at the chromosomal or allelic level, de-
pending on whether the genetic material was
derived from the male or female parent
(Gaffney et al., 1990). In the case of the sym-

patric Mytilus populations of southwestern
England, selection and the Wahlund effect
are the most likely explanations for significant
heterozygote deficiencies (Skibinski, 1983;
Gardner & Skibinski, 1988).

ACKNOWLEDGEMENTS

Thanks to Eric Roderick for help in the field;
to Donna Gardner for data collation; to Andy
Beaumont, David Skibinski and three anony-
mous reviewers for helpful comments. The re-
search was supported in part by a NERC Stu-
dentship to the author.

LITERATURE CITED

AHMAD, M., D. O. F. SKIBINSKI & J. A. BEARD-
MORE, 1977, An estimate of the genetic varia-
tion in the common mussel Mytilus edulis. Bio-
chemical Genetics, 15: 833-846.

ALLENDORF, F. W., K. L. KNUDSEN & G. M.
BLAKE, 1982, Frequencies of null alleles at en-
zyme loci in natural populations of ponderosa
and red pine. Genetics 100: 497-504.

AYALA, F. J., J. R. POWELL, M. L. TRACEY, C.
MOURAO & S. PEREZ-SALAS, 1972, Enzyme
variability in the Drosophila willistoni group. IV.
Genic variation in natural populations of Droso-
phila willistoni. Genetics, 70: 113-139.

BEAUMONT, A. R., T. R. DAY & G. GADE, 1980,
Genetic variation at the octopine dehydrogenase
locus in Cerastoderma edule (L.) and six other
bivalve species. Marine Biology Letters, 1: 137—
148.

BEAUMONT, A. R., E. M. GOSLING, C. M. BEV-
ERIDGE, M. D. BUDD 4 G. M. BURNELL, 1985,
Studies on heterozygosity and size in the scallop,
Pecten maximus. Pp. 443-454, in: P. E. GiBBs,
ed., Proceedings of the 19th European Marine
Biology Symposium, Cambridge University
Press.

BERGER, E. M., 1983, Population genetics of ma-
rine gastropods and bivalves. Pp. 563-596, in:
W. D. RusseLL-HUNTER, ed., The Mollusca, Vol-
ume 6, Ecology, Academic Press, New York.

BOYER, J. F., 1974, Clinal and size-dependent
variation at the Lap locus in Mytilus, edulis. Bio-
logical Bulletin of the Woods Hole Oceano-
graphic Institute, Woods Hole, Mass., 147: 535—
549.

BURKHART, B. D., E. MONTGOMERY, C. H. LAN-
GLEY & В. A. VOELKER, 1984, Characterization
of allozyme null and low activity alleles from two
natural populations of Drosophila melanogaster.
Genetics, 107:295-306.

COYNE, J. A. 8 A. A. FELTON, 1977, Genic het-
erogeneity at two alcohol dehydrogenase loci in

NULL ALLELES AMONG HYBRID MUSSELS 105

Drosophila pseudoobscura and Drosophila per-
similis. Genetics, 87:285—304.

EDWARDS, C. A. & D. O. F. SKIBINSKI, 1987,
Genetic variation of mitochondrial DNA in mussel
(Mytilus edulis and M. galloprovincialis) popula-
tions from South West England and South Wales.
Marine Biology, 94: 547-556.

FOLTZ, D. W., 1986, Null alleles as a possible
cause of heterozygote deficiencies in the oyster
Crassostrea virginica and other bivalves. Evolu-
tion, 40: 869—870.

FREETH, A. L. & J. B. GIBSON, 1985, Alcohol de-
hydrogrenase and sn-glycerol-3-phosphate de-
hydrogenase null activity alleles in natural popu-
lations of Drosophila melanogaster. Heredity, 55:
369-374.

GAFFNEY, P. M., T. M. SCOTT, R. K. KOEHN 4 W.
J. DIEHL, 1990, Interrelationships of heterozy-
gosity, growth rate and heterozygote deficiencies
in the coot clam, Mulinia lateralis. Genetics, 124:
687-699.

GARDNER, J. P. A. & D. O. F. SKIBINSKI, 1988,
Historical and size-dependent genetic variation in
hybrid mussel populations. Heredity, 61: 93-105.

GARDNER, J. P. A. & D. O. F. SKIBINSKI, submit-
ted, Growth and mortality differences between
the mussels Mytilus edulis (L.), Mytilus gallopro-
vincialis (Lmk.) and their hybrids from two sym-
patric populations in S.W. England. Biological
Bulletin of the Woods Hole Oceanographic Insti-
tute, Woods Hole, Mass.

GIBSON, J. B. 8 A. V. WILKS, 1989, Molecular
structure of a naturally occurring alcohol dehy-
drogenase null activity allele in Drosophila mela-
nogaster. Biochemical Genetics, 27: 679-688.

GOSLING, E. M., 1984, The systematic status of
Mytilus galloprovincialis in western Europe: a re-
view. Malacologia, 25: 551-568.

GRANT, W. S. 8 M. I. CHERRY, 1985, Mytilus gal-
loprovincialis Lmk. in southern Africa. Journal of
Experimental Marine Biology and Ecology, 90:
179-191.

HARRIS, H. & D. A. HOPKINSON, 1976, Hand-
book of enzyme electrophoresis in human genet-
ics. North Holland Publishing Company, Amster-
dam.

JIANG, C., J. B. GIBSON, A. V. WILKS 4 A. L.
FREETH, 1988, Restriction endonuclease varia-
tion in the region of the alcohol dehydrogenase
gene: a comparison of null and normal alleles
from natural populations of Drosophila melano-
gaster. Heredity, 60: 101-108.

KATOH, M. 8 D. W. FOLTZ, 1989, Biochemical
evidence for the existence of a null allele at the
leucine aminopeptidase-2 (Lap-2) locus in the
oyster Crassostrea virginica (Gmelin). Genome,
32: 687-690.

KOEHN, R. K., 1991, The genetics and taxonomy
of the species in the genus Mytilus. Aquaculture,
94: 125-145.

LAFIANDRA, D., G. COLAPRICO, D. D. KASARDA
8 E. PORCEDDU, 1987, Null alleles for gliadin

blocks in bread and durum wheat cultivars. The-
oretical and Applied Genetics, 74: 610-616.

LANGLEY, C. H., R. A. VOELKER, A. J. LEIGH
BROWN, S. OHNISHI, B. DICKSON & E. MONT-
GOMERY, 1981, Null allele frequencies at allo-
zyme loci in natural populations of Drosophila
melanogaster. Genetics, 99: 151-156.

MALLET, A. L., E. ZOUROS, K. E. GARTNER-
KEPKAY, K. R. FREEMAN & L. M. DICKIE,
1985, Larval viability and heterozygote deficiency
in populations of marine bivalves: evidence from
pairs of matings of mussels. Marine Biology, 87:
165-172.

MILKMAN, R. & L. D. BEATY, 1970, Large-scale
electrophoretic studies of allelic variation in
Mytilus edulis. Biological Bulletin of the Woods
Hole Oceanographic Institute, Woods Hole,
Mass., 139: 430.

NEI, M., 1972, Genetic distance between popula-
tions. American Naturalist, 106: 283-292.

SANJUAN, A., H. QUESADA, C. ZAPATA & G. AL-
VAREZ, 1990. On the occurrence of Mytilus gal-
loprovincialis Lmk. on the N.W. coast of the Ibe-
rian Peninsula. Journal of Experimental Marine
Biology and Ecology, 143: 1-14.

SCALLON, B. J., C. D. DICKINSON & N. C.
NIELSEN, 1987, Characterization of a null-allele
for the Gy, glycinin gene from soybean. Molecu-
lar and General Genetics, 208: 107-113.

SCHWARTZ, M. 8 W. SOFER, 1976, Alcohol de-
hydrogenase-negative mutants in Drosophila:
defects at the structural locus? Genetics, 83:
125-136.

SEED, R., 1978, The systematics and evolution of
Mytilus galloprovincialis (Lmk.). Pp. 447—468, in:
B. BATTAGLIA & J. A. BEARDMORE, eds., Marine
organisms: genetics, ecology and evolution. Ple-
num Press, New York.

SELANDER, В. К. 8 S. Y. YANG, 1969, Protein
polymorphism and genic heterozygosity in a wild
population of the house mouse (Mus musculus).
Genetics, 63: 653-667.

SINGH, S. M. & R. H. GREEN, 1984, Excess of
allozyme heterozygosity in marine molluscs and
its possible biological significance. Malacologia,
25: 569-581.

SKIBINSKI, D. O. F., 1983, Natural selection in hy-
brid mussel populations. Pp. 283-298, in: G.S.
OXFORD 8 D. ROLLINSON eds., Protein polymor-
phism: adaptive and taxonomic significance.
Systematics Association Special Volume No. 24,
Academic Press, London & New York.

SKIBINSKI, D. O. F., J. A. BEARDMORE & T. F.
CROSS, 1983, Aspects of the population genet-
ics of Mytilus (Mytilidae; Mollusca) in the British
Isles. Biological Journal of the Linnean Society,
19: 137-183.

SKIBINSKI, D. O. F., T. F. CROSS & M. AHMAD,
1980, Electrophoretic investigation of systematic
relationships in the marine mussels Modiolus
modiolus L., Mytilus edulis L., and Mytilus gallo-
provincialis Lmk. (Mytilidae; Mollusca). Biological
Journal of the Linnean Society, 13: 65-73.

106 GARDNER

VOELKER, R. A., C. H. LANGLEY, A. J. LEIGH 1980b, Spontaneous allozyme mutations in
BROWN, S. OHNISHI, B. DICKSON, E. MONT- Drosophila melanogaster: rate of occurrence and
GOMERY & S. C. SMITH, 1980a, Enzyme null nature of the mutants. Genetics, 94: 961-968.
alleles in natural populations of Drosophila mel- ZOUROS, E. 4 D. W. FOLTZ, 1984, Possible ex-
anogaster: frequencies in a North Carolina pop- planations of heterozgote deficiencies in bivalve
ulation. Proceedings of the National Academy of molluscs. Malocologia, 25: 583-591.

Sciences, U.S.A., 77: 1091-1095.
VOELKER, В. A., H. E. SCHAFFER 4 T. MUKAI, Revised Ms. accepted 29 August 1991

MALACOLOGIA, 1992, 34(1-2): 107-128

A NEW PROBLEMATICAL HYGROMIIDAE FROM THE AEOLIAN ISLANDS
(ITALY) (PULMONATA: HELICOIDEA)'

Folco Giusti”, Giuseppe Manganelli? & Jorge V. Crisci?

ABSTRACT

Helicotricha n. gen. is proposed for a very small hygromiid from the Aeolian Islands, Italy: H.
carusoi n. sp. The new species has a shell with persistent postembrional hairs and is charac-
terized anatomically by: a right ommatophore retractor independent of the genitalia; a penial
nerve originating from the right cerebral ganglion; a vaginal complex with digitiform glands and
dart-sac complex consisting of two couples of stylophores disposed on opposite sides of the
vagina (each couple is formed by a larger dart-bearing outer stylophore and a smaller dartless
inner stylophore); and a penial complex with a very peculiar penial papilla. Direct anatomical
comparison of the new genus with other genera of the Hygromiidae suggested that it may be
very closely related to Microxeromagna. This hypothesis was subsequently found to be sup-
ported by the results of cladistic analysis. New evidence is thus provided to confirm that the
distinction between the Hygromiinae and the Trichiinae is artificial. A neotype is designated for
Helix aetnea Benoit, 1857, a junior synonym of Xerotricha conspurcata (Draparnaud, 1801), and
the presence of Helicopsis (s. str.) in western North Africa is confirmed.

Key words: Hygromiidae, Aeolian Islands, Italy, western Mediterranean, Helicella, Xerotricha,

Helicopsis, systematics.

INTRODUCTION

Recent insular equilibrium theory studies
based on analytical comparison of the mala-
cofaunas of the Tuscan and Aeolian archipel-
agos, ltaly (Piantelli et al., 1990), have moti-
vated new field research and study of the
material collected from the various islands. As
happened for the islands of the Tuscan Archi-
pelago (Giusti & Manganelli, 1989, 1990),
new data emerged for the Aeolians. A new
species of the Hygromiidae was identified
amongst material of Xerotricha conspurcata
(Draparnaud, 1801) collected in all the Aeo-
lian Islands. The Hygromiidae are a group of
helicoids of western Palaearctic distribution
recently separated from the Helicidae as a
distinct family and characterized by a bursa
copulatrix duct free from the diaphragm wall
and a variable number of stylophores.

The peculiar structure of the genitalia made
it difficult to establish relationships and ge-
neric status of the new species. Although the
2 + 2 structure of the dart-sac complex as-
sociated with the vagina, and the right om-
matophore retractor free of penis and vagina,
at first glance suggested a relationship with

“Notulae Malacologicae, LI.

the genus Helicopsis, details of the penial
complex suggested other possibilities. In par-
ticular, as in the case of Helicella (one large,
evident and one vestigial, externally invisible,
stylophore on opposite sides of the vagina)
and Xerolenta (one normal and one modified
stylophore on opposite sides of the vagina)
which appear to form monophyletic groups re-
spectively with Candidula (one large, evident
and one vestigial, externally invisible, stylo-
phore on one side of the vagina) and Xero-
munda (one normal and one modified stylo-
phore on one side of the vagina) respectively
(Hausdorf, 1988, 1990a; Giusti & Manganelli,
1989; Manganelli & Giusti, 1989), it seems
highly probable that the new genus (one
larger outer stylophore and one smaller dart-
less inner stylophore on opposite sides of the
vagina) is a member of the monophyletic
group to which also Microxeromagna (one
larger outer stylophore and one smaller dart-
less inner stylophore on one side of the va-
gina) belongs. Accordingly, in view of the fact
that Helicella, Xerolenta, Candidula, Xero-
munda are unanimously regarded as distinct
generic taxa and in view also of our consid-
erations on character weighting in establish-

“Dipartimento di Biologica Evolutiva, Universita degli Studi di Siena, Via Mattioli 4, 1-53100 Siena, Italy
SLaboratorio de Sistematica y Biologia Evolutiva, Museo de La Plata, Paseo del Bosque 1900, La Plata, Argentina

108 GIUSTI, MANGANELLI 8 CRISCI

ing systematic rank in the Hygromiidae (Man-
ganelli 8 Giusti, 1988), a new genus is intro-
duced for the new species.

MATERIALS AND METHODS

Empty shells and whole specimens were
collected in the litter or under stones and
wood on rocky slopes of the islands (locality
data follows species description). Living spec-
imens were left in water to drown for 24 h then
preserved in ethanol 75%. Relaxed material
was studied by optical microscopy (Wild
M5A). Bodies were isolated then dissected
using thin, pointed watchmaker's forceps. Im-
ages of isolated portions of body and genitalia
were drawn using a Wild camera lucida. Rad-
ulae were manually extracted from buccal
bulbs, then washed in pure 75% ethanol,
mounted on copper blocks with electron-con-
ductive glue, sputter-coated with gold and
photographed using a Philips 505 SEM.

All shell parameters—shell maximum di-
ameter, shell height, aperture maximum di-
ameter, aperture height—were measured in
variable numbers of shells from the different
islands using a Wild M5A microscope and a
millimetric lens. Whole shells were photo-
graphed under optical microscope. Whole
shells and shell surface details were photo-
graphed under optical and scanning electron
microscopes using the procedure described
for the radulae.

Detailed study of the genitalia followed,
particularly the distal portion (penis and va-
gina), the external and internal details of
which proved to be diagnostic characters in
similar previous studies (see literature cited in
the Discussion) (Table 2).

The entire set of character states was used
for classical evolutionary and cladistic analy-
sis to define the new genus.

Cladistic analysis was performed using the
method of phylogenetic systematics as origi-
nally developed by Hennig (1966), who main-
tained that only strictly monophyletic taxa
may be regarded as historical entities and
demonstrated that the only logical basis for
inferring monophyly is by showing synapo-
morphies. The distribution of the synapomor-
phies is determined by the parsimony crite-
rion (minimizing homoplasy). On the basis of
these synapomorphies, the taxa are ordered
into a specific pattern represented by a hier-
archical branching diagram.

The new genus Helicotricha belongs to the
family Hygromiidae, which is delimited by
the following unique combination of features:

bursa copulatrix free from diaphragm; vari-
able number of stylophores (2 + 2; 0 + 2;
1 + 1,0 + 1;0 + 0); stylophores, when
present and not extremely regressed or mod-
ified, forming dart-sac complex consisting of
one or two double units with a common base
and distinct distal sacs lying side by side in
the same plane; diverticulum of bursa copul-
atrix absent; digitiform gland tufts variable in
number inserted on proximal vagina not close
to where dart sacs, when present, open into
vagina. Helicidae are characterized by: bursa
copulatrix inserted into diaphragm; dart-sac
complex consisting of a single stylophore; di-
verticuium of bursa copulatrix duct present.
Helicidae and Hygromiidae share the follow-
ing characters: dart-sac complex, vaginal dig-
itiform glands.

Twenty (Table 1) genera are considered to
be the terminal taxa. The minute details of the
distal genitalia of some have recently been
reviewed (Schileyko, 1978a, 1978b; Giusti &
Manganelli, 1987, 1989; Hausdorf, 1988,
1990a, 1990b; Manganelli & Giusti, 1988,
1989).

In the case of the two sets of genera re-
viewed by Schileyko and Hausdorf, some
characters were uncertain or not discussed. A
question mark sometimes follows or substi-
tutes the character states in Table 3. A ques-
tion mark also substitutes the character states
when one character is known to be present
with more than two states in the same taxa
(character 5 in Xeromunda and Cernuella (s.
str.); character 16 in Xerotricha). Character
polarity was determined by outgroup compar-
ison methods (Watrous & Wheeler, 1981;
Maddison et al., 1984) using the family col-
lectively Helicidae as outgroup, because
there is no singie genus that is a clear sister
group. Three out fifteen characters (5, 10, 15)
had more than two character states. These
characters were treated as nonadditive. Two
characters (11, 14) were autoapomorphies,
with an additional autoapomorphy in a multi-
state nonadditive character (10).

The data was analyzed using a Wagner
parsimony algorithm from Farris’s phyloge-
netic program HENNIG86, applying the im-
plicit enumeration option for calculating trees
(version 1.5; Farris, 1989; see also Platnick,
1989) run on a IBM AT computer. When cla-
distic analysis yielded more than one tree, the
Nelsen consensus method was applied
(Nelsen, 1979). We also used the successive
weighting procedure (Farris, 1989), which cal-
culates weights from the best fits to the most
parsimonious trees, and applied them in the

A NEW HYGROMIIDAE FROM ITALY 109

TABLE 1. Acronyms, genus-group taxa, type species and bibliographical sources of anatomical data.
Some of the genera listed have a subgeneric division. For cladistic analysis, only species of

nominotypical subgenera have been considered.

Type-species
Glischrus (Helix) candidula,
Studer, 1820, = Helix
unifasciata Poiret, 1801

Helix eichwaldi Pfeiffer, 1846
Helix variabilis Draparnaud,
1801, = Cochlea virgata

Helix edentula Draparnaud,
Helix itala Linnaeus, 1758; cf.

Helix striata Múller, 1774

Helicotricha carusoi Giusti,
Manganelli 8 Crista, 1992
Helix cinctella Draparnaud,

Hygrohelicopsis darevskii
Helix holotricha O. Boettger,

Helix rubens von Martens, 1874
Helix stolismena Bourguignat,
in Servain, 1880, = Helix

Nanaja cumulata Schileyko,

Helix lubomirskii Slosarski, 1881
Helix (Pseudoxerophila)
bathytera Westerlund, in
Westerlund & Blanc, 1879
Helix hispida Linnaeus, 1758
Helix obvia Menke, 1828

Helix turbinata, sensu
Monterosato, 1892, non De
Cristofori & Jan, 1832) (cf.
Hausdorf, 1988; Manganelli &
Giusti, 1988; 1989; an
application to the 1.C.Z.N. is in
progress by Giusti & Manganelli
Helix explanata Múller, 1774

Acronyms Genus-group taxa
CAND Candidula Kobelt, 1871
CAUC Caucasigena Lindholm, 1927
CERN Cernuella Schluter, 1838
Da Costa, 1778
EDEN Edentiella Polinski, 1929
1805
HELL Helicella Férussac, 1821
Opinion 431
HELP Helicopsis Fitzinger, 1833
HELT Helicotricha Giusti €
Manganelli, 1992
HYGR Hygromia Risso, 1826
1801
HYGH Hygrohelicopsis Schileyko,
1978a Schileyko, 1978a
KOKO Kokotschashvilia Hudec &
Lezhawa, 1969 1884
LEUC Leucozonella Lindholm, 1927
MICR Microxeromagna Ortiz de
Zarate Lopez, 1950
armillata Lowe, 1852
NANA Nanaja Schileyko, 1978b
1978b
PLIC Plicuteria Schileyko, 1978a
PXER Pseudoxerophila Westerlund,
in Westerlund & Blanc, 1879
TRIC Trichia, Hartmann, 1840
XERL Xerolenta Monterosato, 1892
XERM Xeromunda Monterosato,
1892
XERS Xerosecta Monterosato, 1892
XERT Xerotricha Monterosato, 1892

Helix conspurcata Draparnaud,
1801

Sources
Hausdorf, 1988: (C. unifasciata,
C. gigaxii); personal
unpublished data on C. spadae,
C. intersecta, C. unifasciata
Schileyko, 1978a, 1978b
Hausdorf, 1988; Manganelli 8
Giusti, 1988

Schileyko, 1978a, 1978b

Hausdorf, 1988; Giusti &
Manganelli, 1989

Giusti & Manganelli, 1989;
Schileyko, 1978b;
Hausdorf, 1990b

this paper

Giusti & Manganelli, 1987
Schileyko, 1978a, 1978b
Schileyko, 1978a, 1978b
Schileyko, 1978a, 1978b
Hausdorf, 1988, 1990c;
Manganelli & Giusti, 1988
Schileyko, 1978b
Schileyko, 1978a, 1978b
Hausdorf, 1988

Schileyko, 1978a, 1978b
Hausdorf, 1988
Hausdorf, 1988, 1990a;
Manganelli & Giusti, 1989

Manganelli 4 Giusti, 1988
Hausdorf, 1988; Giusti &
Manganelli, 1989

weighting procedure until there were no
changes in successively produced trees.

SYSTEMATIC DESCRIPTION
Helicotricha n. gen
Diagnosis

Very small hygromiid with shell having per-
sistent postembronial hairs; anatomically
characterized by right ommatophore retractor
independent of genitalia; penial nerve from

right cerebral ganglion; vaginal complex with
digitiform glands and dart-sac complex con-
sisting of two pairs of stylophores, each cou-
ple comprising a larger dart-bearing outer sty-
lophore and a smaller dartless inner
stylophore; penial complex having a very pe-
culiar penial papilla.

Description

Shell: Small, hairy, opaque-brown in colour,
with white flecks. Spire consisting of approx-

110 GIUSTI, MANGANELLI 8 CRISCI

TABLE 2. List of characters

1—Penial nerve.

—From right cerebral ganglion = 0

—From right pedal ganglion = 1

Remarks: no data for Caucasigena, Edentiella, Hygrohelicopsis, Kokotschashvilia, Leucozonella, Nanaja,
Plicuteria. In Manganelli & Giusti (1989: 4) was wrongly reported for Xeromunda “from right pedal
ganglion.” Revision of original data indicates that penial nerve comes out of right cerebral ganglion.

2—Right ommatophore retractor.
—Between penis and vagina = 0
—Independent of penis and vagina = 1
Remarks: No data for Psudoxerophila.

3—Number of stylophores and/or their derivates forming the dart-sac complex.
—2+2=0
—0+2=1

4—Shape and position of stylophore groups in relation to vagina.

—Stylophores Trichia type: each stylophore group (each composed of an outer and an inner stylophore)
slender and entering vagina through a slender neck (Manganelli & Giusti, 1988: fig. 14 E) = 0
—Stylophores not Trichia type: each stylophore group (each composed of an outer and an inner
stylophore) wide and fused to inner walls of vagina for a long tract (Manganelli & Giusti, 1988: fig. 14 A)
= 1

Remarks: Helicopsis: based on the type species only (Giusti & Manganelli, 1989). Schileyko (1978b) and
Hausdorf (1990b) show drawings of the genitalia of H. striata and of some other species (H. likharevi, H.
retowskii) in which the situation is slightly different. A slightly different situation also occurs in the species
studied in the present paper and referred to as Helicopsis sp.

5—Shape and dimensions of stylophore groups.

—Each group formed by an inner and an outer stylophore of similar dimensions (Manganelli & Giusti,
1988: fig. 14 A) = O

—Each group consisting of large outer stylophore and small externally visible inner stylophore (Giusti &
Manganelli, 1989: fig. 9 A) = 1

—Each group consisting of large outer stylophore and externally visible residues of the inner stylophore
(Manganelli & Giusti, 1989: fig. 1 E) = 2

—Each group consisting of larger outer stylophore and very small, not externally visible, inner stylophore
(Schileyko, 1987a: fig. 43) = 3

—Each group consisting of large outer stylophore and extremely reduced not externally visible inner
stylophore (Giusti & Manganelli, 1989: fig. 9 C) = 4

Remarks: the scheme of the dart-sac complex in Helicella reproduced by Hausdorf (1988: fig. 8) is
incorrect: the inner stylophore, referred to as “Nebensack,” appears too large and externally visible.

6—Digitiform glands.

—All around vagina = 0

— Оп one side of vagina = 1

Remarks: situation not clear enough in drawings by Schileyko (1978a, 1978b) of the genitalia of
Edentiella, Nanaja and Plicuteria. The situation in Hygrohelicopsis and Leucozonella showed by the
same author (Schileyko, 1978a, 1978b) seems to indicate digitiform glands all around vagina. Species of
Cernuella (s. str.) show digitiform glands all around vagina (C. caruanae) or on one side of the vagina
(C. virgata).

7—Basal portion

—Stylophore groups opening directly into vagina without a wide basal dilated portion (Manganelli &
Giusti, 1988: fig. 14 E) = O

—Stylophore groups opening in a wide dilated basal portion (Manganelli 8 Giusti, 1989: fig. 1 E) = 1

8—Inner stylophores.

—With thin muscular walls and large internal cavity (Manganelli 8 Giusti, 1988: fig. 14 E) = O

— With thick muscular walls and small internal cavity (Manganelli 8 Giusti, 1988: fig. 14 A) = 1
Remarks: transverse and longitudinal sections of dart-sac complex in Edentiella, Kokotschashvilia,
Leucozonella, Nanaja, Plicuteria unknown. Situation as reported by Schileyko (1978a, 1978b) for
Caucasigena and Hygrohelicopsis not clear enough.

A NEW HYGROMIIDAE FROM ITALY 111

9—Opening of stylophores.

—Openings of inner and outer stylophore cavities into vagina close to each other (Manganelli 4 Giusti,
1988: fig. 14 E) = 0

—Openings of inner and outer stylophore cavities into vagina very far apart (Manganelli & Giusti, 1988:
fig. 14 A) = 1

10—Inner vaginal accessory structures.

—Opening of stylophores in a groove bordered by folds (Manganelli & Giusti, 1988: fig. 8 A) = O
—Opening of stylophores bordered by rows of papillae (Schileyko, 1987b: fig. 215) = 1

—Vagina with one tongue-like structure for each stylophore group (two tongue-like structures when 2
stylophore groups present) (Giusti & Manganelli, 1989: figs. 3, 9A) = 2

—Vagina with a groove-like structure for each stylophore group (unique tube-like structure when 2
stylophore groups present) (Giusti & Manganelli, 1989: figs. 7, 9 C) = 3

—Vagina with dart-gun through which dart is shoot (Manganelli & Giusti, 1989: figs. 4 E, 14 A) = 4

11—Joint of penis and vagina
—Penis joins vagina distally with respect to stylophores (Manganelli & Giusti, 1989: Fig. 5 F) = O
—Penis joins distal vagina level with stylophores (Manganelli 8 Giusti, 1989: fig. 8 B) = 1

12—Proximal penis.

—Proximal penis Helicopsis type: transverse sections reveal a duct in the lumen; this duct joins the
epiphallus lumen directly with the ejaculatory canal of the penial papilla (Giusti & Manganelli, 1989: fig. 8
F, H) =0.

—Proximal penis simple: transverse sections only show the penial walls (Manganelli & Giusti, 1988: fig.
11F) = 1.

Remarks: A structure, only apparently similar to those in Helicopsis, seems present in drawings by
Manganelli & Giusti (1990: figs. 2C, 3A, 4E for Xeromunda), Schileyko (1978b: Fig. 253 for “Helicella
candicans)”, Schileyko, in Damjanov & Likharev (1975: fig. 274, for “Helicella candicans” and fig. 278 for
“Helicella spiruloides”). This is due to the fact that the thin external layer of the penis has been detached
during dissection.

13—Glandular area on one side of terminal penis walls.

—absent = 0

—present (this paper: Fig. 3) = 1

Remarks: no data for Caucasigena, Edentiella, Hygrohelicopsis, Kokotschashvilia, Leucozonella, Nanaja,
Plicuteria, Pseudoxerophila; for Microxeromagna unpublished personal data.

14—Frenula.
—Penial papilla with no frenula joining it to the distal penis walls (Manganelli & Giusti, 1988: fig. 7 F-H)
=0

—Penial papilla joined by frenula to the distal penis walls (Manganelli 8 Giusti, 1988: fig. 6 C-D) = 1

15—Sections of penial papilla.

—Trichia type (Schileyko, 1978b: fig. 221) = 0

—Caucasigena type (Schileyko, 1978b: fig. 199) = 1

—Xerosecta type (Manganelli & Giusti, 1988: fig. 8 C) = 2

—Helicotricha type (this paper: Fig. 3C) = 3

—Microxeromagna type (Manganelli 8 Giusti, 1988: fig. 11 E) = 4

—Leucozonella type (Schileyko, 1978b: fig. 146) = 5

—Cernuella type (Manganelli 8 Giusti, 1988: fig. 7 G) = 6

Remarks: Species of Xerotricha have sections of penial papilla of Trichia type (X. apicina) and of
Cernuella type (X. conspurcata). Due to variability, Leucozonella is based only on the type-species.

imately 4 whorls separated by deep sutures,
last whorl angled at periphery. Umbilicus
open, wide approximately 1/5 of maximum
shell diameter. Aperture oblique, oval, lacking
internal rib. Peristome not thickened or re-
flexed. External surface of protoconch with
few faint growth lines and microsculpture con-

sisting of fine close longitudinal grooves. Ex-
ternal surface of teleoconch with superficially
reticulated periostracal layer and transverse
rows of very short hairs.

Genitalia: Vaginal complex with relatively long
distal vagina; dart-sac complex consisting of

112

GIUSTI, MANGANELLI & CRISCI

TABLE 3. Original data matrix used for cladistic analysis. All characters with more than two states are

treated as not additive.

Taxa

Outgroup
Candidula
Caucasigena
Cernuella
Edentiella
Helicella
Helicopsis
Helicotricha
Hygrohelicopsis
Hygromia
Kokotschashvilia
Leucozonella
Microxeromagna
Nanaja
Plicuteria
Pseudoxerophila
Trichia
Xerolenta
Xeromunda
Xerosecta
Xerotricha

O00000060VV%0VV%O0V%001V105-0|—
ni O O OO O O O O CIN
Spee. sSecesS ose oecla
KOASAS OO O OO O 36 o] AS
|=ONNDOO“DODOWOOKROOO A | м
EN O

two pairs of stylophores disposed on opposite
sides of vagina, each pair consisting of a
smaller inner and a larger outer stylophore;
outer stylophore containing slightly curved
darts of circular section near base and oval or
rhombic section thereafter; each inner stylo-
phore (called “accessory sacs” by Nordsieck,
1987, and Hausdorf, 1988; see Giusti & Man-
ganelli, 1989: 51, for a discussion of homol-
ogy and terminology of this structure) show-
ing wide, totally dartless internal cavity;
cavities of outer and inner stylophore of each
couple in communication and opening into va-
gina in a single opening bordered by two large
anteriorly fused pleats. Digitiform glands
forming two groups, each of two glands,
sometimes apically branched, arising from
opposite sides of distal vagina close to point
where bursa copulatrix duct arises. Bursa
copulatrix duct of medium length, with initial
portion slightly flared. Penial complex with fla-
gellum almost as long as epiphallus plus pe-
nis. Epiphallus (from end of vas deferens to
point of attachment of penial retractor) long,
almost twice penis length. Penis (from point of
attachment of penial retractor to genital
atrium) lacking distinct penial sheath and dis-
tally enlarged. Wide area yellow in colour
and covered with glandular tissue, on exter-
nal side of terminal portion of penis walls. Pe-

oo-$0-"0O000000000000000(N

Character

[to]
ER
o
EN
N
N
№
A
(a)
N
>
mb
on

Him +9929VO00201V%O00—-:%|0o0
O000000000001:00000100:%
DOOOOO-DPOOOLROOOWOROWO
O00000000000000000-:000
ооо.
120000%V%V2VV9OV%200V%0%00
oOOO00000000000000-000
9 D OO O © :9 © O B O1 © O © W © O) = ON — ON 9

nial papilla cylindrical, formed by a wide tube
with thin walls. Penial papilla lumen continu-
ous with that of proximal penis and epiphal-
lus. T-shaped (in transverse section) pilaster
running entire length of penial papilla and
joined to it by a peduncle, so that cross sec-
tion of papilla plus pilaster resembles card fig-
ure spades.

Right ommatophore retractor free of penis
and vagina.

Penial nerve apparently arising in the right
cerebral ganglion (according to Franc, 1968:
473, even if it comes from cerebral ganglion
the penial nerve originates in the pedal gan-
glion).

Origin of the Name

Helicotricha, gender feminine.
The small helicoid shell with hairy perios-
tracum suggested the name of the new genus.

Type species
Helicotricha carusoi n. sp.

Helicotricha carusoi n. sp.
[Figs. 1A-C, 2A-D, 3A—D, 4A-D, 5A-D, G,
6A-C]
Helicella (Xerotricha) conspurcata,—Giusti,

1973: 259-260 [partim, non Draparnaud,
1801].

A NEW HYGROMIIDAE FROM ITALY 113

FIG. 1. Shells of Helicotricha carusoi n. gen. n. sp. Holotype (A) and one paratype (B) from Alicudi Island:
Perciato, F. G. leg. 24.10.69. A shell from Salina Island: Pollara, R. Arcidiacono leg. 21.9.66 (C).

Helicopsis sp.,—Piantelli et al., 1990: Table 5
et passim.

Diagnosis

At present the only species of the genus
Helicotricha known. Specific coincides with
generic diagnosis.

Description

Shell (Figs. 1A-C, 2A-D): Shell small (Figs.
1A-C), hairy, low conical above, convex be-
low, opaque brown in colour, with white flecks.

White flecks widely distributed over shell sur-
face, concentrated above (near sutures and
near periphery of last whorl) to form irregularly
spaced spots of variable shape, below (from
periphery of last whorl to umbilicus) to form
white spiral bands of variable width and num-
ber (2-6). Spire depressed-conical, consisting
of 4-42 convex, regularly increasing whorls
separated by deep sutures; last whorl angled
at periphery, dilated, sometimes descending
slightly near aperture. Umbilicus open and
wide, approximately 1/5 of maximum shell di-
ameter. Aperture markedly oblique, oval, lack-

114 GIUSTI, MANGANELLI & CRISCI

FIG. 2. External shell surface of specimens of Helicotricha carusoi n. gen. п. sp. collected on Panarea Island,
D. Caruso & |. Marcellino leg. 27.5.67 (A-B) and Lipari Island: Monte Sant'Angelo, F.G. leg. 23.10.69 (C—D).
A: A view of first whorls. B: Detail of protoconch showing longitudinal grooves. C: Detail of last whorl with one
hair and reticular microsculpture of teleoconch. D: Detail of reticular microsculpture of teleoconch.

ing internal rib; peristome not thickened,
slightly reflexed at its columellar margin.
External surface of protoconch with few
faint growth lines near its end (Fig. 2A) and
microsculpture consisting of fine close spiral
grooves (Fig. 2B). External surface of teleo-
conch (Fig. 2A) with many growth lines, more
marked near sutures. Periostracal layer (Fig.

2C-D), superficially reticulated (Fig. 2D) to
form transverse rows of very short, often
hook-shaped hairs 0.1 mm in length (Fig. 2C).
Reticulation and hairs of caducous appear-
ance, absent in large portions of surface. Sur-
face of mineralized portion of shell (in areas
devoid of periostracal layer) seems to be
crossed by fine close spiral grooves.

A NEW HYGROMIIDAE FROM ITALY 115

FIG. 3. Genitalia of Helicotricha carusoi п. деп. п. sp. in specimens collected on Basiluzzo Islet, F. G. leg.
5.11.69. A-B: Two opposite views of the same genital apparatus (gonad excluded in A; gonad and part of
ovispermiduct excluded in B). C: Part of penial complex with distal penis opened to show penial papilla, a
section of proximal penis and two sections of penial papilla. D: Vagina opened to show its inner structure.
Explanations of the symbols used in Figures 3-5, 8: ag, albumen gland; bc, bursa copulatrix (gametolytic
gland); dbc, duct of bursa copulatrix; dg, digitiform glands; dsc, dart-sac complex; dp, distal penis; e,
epiphallus; f, flagellum; fc, fertilization chamber; fn, fenestration; fo, free oviduct; fr, frenula; g, penial papilla
(glans); ga, genital atrium; gp, genital pore; gt, glandular tissue; gtp, gland of the terminal penis; hd,
hermaphrodite duct; p, penis; po, prostatic portion of ovispermiduct; pp, proximal penis; pr, penial retractor
muscle; pv, proximal vagina; pw, penial wall; s, stripes; sr, seminal receptacle; t, talon; tsp, t-shaped pilaster
of the penial papilla; uo, uterine portion of ovispermiduct; v, vagina; vd, vas deferens.

116 GIUSTI, MANGANELLI & CRISCI

FIG. 4. Genitalia of Helicotricha carusoi n. gen. n. sp. in specimens collected on Filicudi Island: Stimpagnato,
F. G. leg. 28.10.69. A: Genital apparatus (gonad and part of ovispermiduct excluded). B—C: Part of two penial
complexes with distal penis opened to show penial papilla, section of epiphallus, proximal penis and penial
papilla (B), Section of proximal penis and two sections of penial papilla (C). D: Vagina opened to show its

inner structure.

Dimensions (N = 10): shell max. diam.:
4.5-5.4 mm; shell height: 2.5-3.4; aperture
max. diam.: 2.0-2.5 mm; aperture height:
1.7-2.1.

Dimensions of the holotype: shell max.

diam.: 5.3 mm; shell height: 2.8; aperture
max.: diam. 2.1 mm; aperture height: 2.1

Genitalia (Figs. 3A-D, 4A—D, 5A—D, G): Con-
voluted first hermaphrodite duct arising from

A NEW HYGROMIIDAE FROM ITALY 117

A-B 2mm
CG mm

FIG. 5. Outline and scheme of dart-sac complex (darts omitted) (A-B), talon (C-D) and dart (G) of Helicot-
richa carusoi n. gen. n. sp. in specimens collected on Filicudi Island: Stimpagnato, F. G. leg. 28.10.69, talon
(E) of Microxeromagna armillata from Corsica: Olmeto, F.G. & G.M. leg. 1.12.83 and talon (F) of Helicopsis
striata from Oland, parish Persnása (Jordhamn, Sweden).

118 GIUSTI, MANGANELLI 8 CRISCI

FIG. 6. Radula of Helicotricha carusoi n. gen. n. sp.
in a specimen collected on Basiluzzo Islet, F. G.
leg. 5.11.69. A: Central tooth and first lateral teeth.
B: 6th-12th lateral teeth. C: Extreme marginal
teeth.

multilobate gonad (not illustrated) and ending
in talon (i.e. fertilization chamber plus seminal
receptacle complex); talon lying on surface of
inner side of large albumen gland. Talon (Fig.
5C—D) with wide lateral fertilization chamber
embracing basal portion of seminal recepta-
cle complex. Receptacle complex very slen-
der, only slightly enlarged apically and con-
taining 2-3 very small distinct sperm tubes.
Ovispermiduct wide, plurilobate, consisting of
prostatic and uterine portions. Prostatic por-
tion continuing anteriorly into vas deferens
and proximal portion of penis complex. Penis
complex (Figs. 3A-B, 4A) consisting of penial

flagellum, epiphallus (i.e. from end of vas def-
erens to point of attachment of penial retrac-
tor muscle) and penis (i.e. from point of at-
tachment of penial retractor muscle to genital
atrium). Penial flagellum long (4.9—5.5 mm; N
= 3) and slender along entire length. Epiphal-
lus longer (3.0—4.0 mm; N = 3) than penis
(2.0-2.3 mm; N = 3) and narrower. Penis
lacking distinct penial sheath but with thin
bands of tissue arising where penial retractor
muscle ends and terminating near penis end.
Penis distally enlarged where it contains pe-
nial papilla (= glans). Wide area yellow in
colour and covered with glandular tissue (also
evident internally) on external side of terminal
portion of penis walls. Penis walls level with
base of penial papilla sometimes appear to
contain glandular tissue. Penial papilla (Figs.
3C, 4B-C) cylindrical, consisting of a wide
tube. Penial papilla lumen continuous with
proximal penis and epiphallus. T-shaped (in
transverse section) pilaster running alongside
penial papilla for entire length, joined to it by a
peduncle. Ejaculatory pore at apex of penial
papilla, slit-like, variable in width. T-shaped
pilaster basally connected to penis wall for
short distance at penial papilla base (Figs.
3C, 4C). Penial retractor muscle usually
short. Uterine portion of ovispermiduct con-
tinuing anteriorly into short (approximately 1
mm; N = 3), wide uterine canal (i.e. free ovi-
duct) ending in vagina level with point of entry
of duct (4-6 mm in length; N = 3) of bursa
copulatrix. Bursa copulatrix (i.e. gametolytic
gland) (Figs. 3A-B, 4A) sac-like and variable
in shape. Duct of bursa copulatrix 4-6 mm
long (N = 3), its initial portion slightly flared,
internally figured with series of fine pleats.
Dart-sac complex consisting of two pairs of
stylophores, each pair consisting of a smaller
inner dartless stylophore and a larger outer
dart-bearing stylophore. Opening of each pair
of stylophores clearly visible half way along
vagina on opposite sides (Figs. 3A-B, 4A,
5A-B). Two pairs of digitiform glands also
opening on opposite sides of proximal vagina
close to dart-sac complex. Digitiform glands
irregular in shape and sometimes apically bi-
furcated. Each digitiform gland and dart-sac
pair opens into the vagina within a furrow lat-
erally bordered by two large, distally fused
folds (Figs. 3D, 4D). Other smaller folds run
longitudinally on the internal vagina walls be-
tween two systems of larger folds. All folds
taper distally to end on internal genital atrium
walls. Dart of each outer stylophore very
small (1 mm in length; N = 2), basally circular

A NEW HYGROMIIDAE FROM ITALY 119

in transverse section, apically flattened-oval
or rhombic, tip arrowhead-shaped (Fig. 5G).

Other Anatomical Characters: Penial nerve
apparently originating in right cerebral gan-
glion (according to Franc, 1968: 473, even if it
comes from cerebral ganglion the penial
nerve originates in the pedal ganglion).

Right ommatophore retractor independent
of penis and vagina (i.e. not passing between
the penis and vagina).

Radula (Fig. 6A-C): Radula consisting of
many rows of teeth each according to formula
20-22 + C + 20-22. Central tooth with wide
basal plate and raised pointed upper vertices.
Body of tooth with very large mesocone and
two small ectocones, 1/3 of mesocone length.
First lateral teeth having wide basal plate but
with inner vertex missing; body with large
pointed mesocone and small pointed ectone
1/4 of mesocone length. At about 6th or 7th
lateral tooth, inner side of mesocone showing
slight protuberance developing into pointed
cusp in following lateral and marginal teeth.
Moving laterally, teeth maintaining same
shape but progressively smaller, with more
pointed cusps and reduced basal plates. Last
marginal teeth having mesocone apex with
2-3 cusps and ectotone split into 2-4 smaller
points.

Jaw: Jaw odontognathous, strongly ribbed
and devoid of central denticle.

Type Locality

Aeolian Islands, Alicudi Island: Perciato
[UTM references: 33SVC4465].

Typical Series

Holotype (shell) (Fig. 1A) and 7 paratypes
(2 shells + 3 dissected specimens + 2 spirit
specimens) from the type locality, F. G. leg.
24.10.69. Holotype and all the paratypes in
Giusti Collection, Dipartimento di Biologia Ev-
olutiva, Universita deg!i Studi di Siena, Via
Mattioli 4; 1-53100 Siena, Italy. Other material
examined (all from Aeolian Islands [UTM ref.:
33SVC, WC])

Alicudi Island [33SVC46]: Spano [[4467], F.
G. leg. 26.10.69 (2 spirit specimens + 2 dis-
sected specimens).

Basiluzzo Islet [33SWC07]: $. Bruno leg.

25.2.67 (7 spirit specimens), F. G. leg. 5.11.69
(1 spirit specimen + 1 dissected specimen); Е.
G. leg. 31.3.71 (3 spirit specimens).

Filicudi Island [33SVC66, 67]: Siccagni
[6070], F. G. leg. 29.10.69 (1 spirit specimen);
Stimpagnato [6168], F. G. leg. 28.10.69 (4
spirit specimens + 2 dissected specimens);
Between Canale and Monte Guardia [6268,
6368], F. G. leg. 28.10.69 (3 spirit speci-
mens); Zucco Grande [6270, 6370], F. G. leg.
30.10.69 (2 spirit specimens + 1 shell), F. G.
leg. 23.3.72 (2 shells).

Lipari Island [33SVC95, 96]: G. Marcuzzi
leg. 13.4.68 (1 shell + 1 spirit specimen);
Capistello [9556, 9656], F. G. leg. 27.4.70 (1
shell); Monte Sant’Angelo [9360, 9460], F. G.
leg. 23.10.69 (5 shells).

Lisca Bianca Islet [33SWC07]: Е. С. leg.
5.10.69 (7 spirit specimens)

Panarea Island [33SWC07]: D. Caruso 4 |.
Marcellino leg. 27.5.67 (5 shells), Punta del
Corvo [0576], F. G. leg. 30.3.71 (1 shell + 1
spirit specimen); Punta Milazzese [0575], F.
G. leg. 5.11.69 (1 spirit specimen + 1 dis-
sected specimen).

Salina Island [33SVC86, 87]: Capo Faro
[8870], F. G. leg. 25.4.70 (2 spirit specimens);
Lingua [8865], F. G. leg. 25.4.70 (2 spirit
specimens + 1 dissected specimen); Malfa
[8570], R. Arcidiacono leg. 17.9.66 (1 spirit
specimen); Monte dei Porri [8368, 8468], F.
G. leg. 26.4.70 (3 spirit specimens); Pollara
[8369], R. Arcidiacono leg. 21.9.66 (1 shell),
F. G. leg. 25.4.70 (1 spirit specimen); Rinella
[8566], F. G. leg. 26.4.70 (1 shell); Valle del
Santuario [8568], F. G. leg. 25.4.70 (1 spirit
specimen).

Stromboli Island [33SWC19, 20]: G. Mar-
cuzzi leg. 16.4.68 (28 spirit specimens).

Vulcano Island [33SVC95]: G. Marcuzzi
leg. 12.4.68 (2 shells); Porto [9652], F. G. leg.
27.4.70 (2 spirit specimens + 1 shell).

All the material in Giusti Collection, Dipar-
timento di Biologia Evolutiva, Universitá degli
Studi di Siena, Via Mattioli 4; 1-53100 Siena,
Italy.

Locality names and UTM references based
on the Official Map of Italy 1:25.000 Series M
891.

Origin of the Name

The new species is dedicated to Prof. Do-
menico Caruso, Director of the Department of
Animal Biology, University of Catania, Italy, in
token of highest esteem and companionship.

120 GIUSTI, MANGANELLI 8 CRISCI

DISCUSSION
Generic Status

As outlined in the introduction, the su-
praspecific status of the new species was dif-
ficult to establish. Some anatomical features
(2 + 2 stylophores and right ommatophore
retractor independent of penis and vagina)
suggested relationships with more than one
genera of Hygromiidae found in the western
Mediterranean and Europe: Xerotricha Mon-
terosato, 1892 (type species: Helix conspur-
cata Draparnaud, 1801), Helicella Férussac,
1821 (type species: Helix ¡tala Linnaeus,
1758) and Helicopsis Fitzinger, 1833 (type
species: Helix striata Múller, 1774) (see Git-
tenberger, 1969; Gittenberger 8 Manga,
1977; Schileyko, 1978a; Gittenberger 4
Raven, 1982; Giusti 8 Manganelli, 1989; Git-
tenberger et al., 1989; Hausdorf, 1988).

Following the scheme of classical evolu-
tionary systematics a first comparison was
made with Xerotricha, the type species of
which, X. conspurcata (Draparnaud, 1801),
has a small hairy shell that differs from Heli-
cotricha carusoi n. sp. only by virtue of its
longer periostracal hairs and its narrower um-
bilicus.

Xerotricha conspurcata, like both its conge-
neric species, X. apicina (Lamarck, 1822) and
X. nubivaga (Mabille, 1882) (Hausdorf, 1988;
Gittenberger et al., 1989, Giusti 8 Manganelli,
1989), shows a differently structured vaginal
complex characterized by a large dart-sac
complex consisting of two pairs of stylo-
phores, each constituted by a larger outer
dart-bearing stylophore and a very small inner
dartless stylophore (i.e. 2 + 2) (see Giusti 8
Manganelli, 1989: 51, for a discussion of ho-
mology and terminology of these structures).
The latter open into the vagina inside the slit
delimited by two large tongue-like structures
that are apically independent of each other
(Hausdorf, 1988: fig. 9; Giusti & Manganelli,
1989: figs. 3, 5, 9A-B). The penial papilla in
Xerotricha is also different, being simple and
without lateral pilaster (Giusti & Manganelli,
1989: figs. 2D, 4E).

Anatomically the new species is also simi-
lar to some small Helicella with shells that
have persistent postembryonic hairs (e.g. H.
corderoi Gittenberger & Manga, 1977—see
Gittenberger & Manga, 1977; Manga Gonza-
les, 1983; H. mangae Gittenberger & Raven,
1982—see Gittenberger 8 Raven, 1982; ac-
cording to Prieto, 1986, this nominal species

is a junior synonym of H. gonzalei Azpeitia,
1924). Nevertheless, Helicella has a dart-sac
complex which, although similar (consisting of
two opposite pairs of stylophores, each
formed by a very large apically pointed outer
stylophore and a very reduced, almost vesti-
gial, inner dartless stylophore, i.e. 2 + 2),
opens internally into a continuous pleated
tube-like structure contained in the vagina
(Hausdorf, 1988: fig. 8; Giusti & Manganelli,
1989: figs. 6A,E, 7, 9C). In Helicella, the pe-
nial papilla is also different, being simple and
without lateral pilaster (Hausdorf, 1988: fig. 8;
Giusti & Manganelli, 1989: fig. 6F).

Detailed comparison was then made with
Helicopsis*, the species of which have a very
similar genital scheme (Hesse, 1934; Gitten-
berger, 1969; Damjanov & Likharev, 1975;
Schileyko, 1978a; Grossu, 1983; Giusti 8
Manganelli, 1989; present paper: Figs. 5F,
8A-C). Helicopsis (s. str.) has so far been
identified with certainty only in central and ori-
ental Europe, and subgenus Xeroleuca Ko-
belt, 1877 (type species: Helix turcica Holten,
1802) has been reported in northwestern Af-
rica (Hesse, 1934; Zilch, 1960). There are two
records of Helicopsis (s. str.), one in Tunisia
(Ktari & Rezig, 1976) and one at Huelva,
Spain, (Gasull, 1972, 1985) that must be
checked. The species briefly studied by Ktari
& Rezig (1976) may correspond to a species
found by one of us in Morocco (Figs. 7A, 8A—
C). Helicopsis (s. str.) has a small, thick, ro-
bust, ribbed and hairless shell and a vagina
with a dart-sac complex constituted by two
opposite pairs of stylophores, each formed by
an outer and an inner stylophore (i.e. 2 + 2).
The outer stylophores are smaller and more
slender than those in Xerotricha and Helicella
and are more clearly distinguished from the
inner dartless stylophores which, in their turn,
are larger than those in Xerotricha and Heli-
cella. Moreover, the vagina in Helicopsis is
clearly different from that in Xerotricha and
Helicella, being internally devoid of tongue-
like or tube-like structures into which the sty-
lophores open (Giusti & Manganelli, 1989:
figs. 8C, 9D). In view of the above and be-
cause the new species and those of Helicop-

“While the paper was in press, Hausdorf (1990b) de-
scribed the genitalia of three species-group taxa supposed
to belong to genus Helicopsis: H. gittenbergeri п. sp., Н. $.
subcalcarata (Naegele, 1903) and H. subcalcarata neu-
berti n. subsp. Comparison with the latter taxa has been
omitted because no information about the penial papilla
structure, the only sure diagnostic character for Helicopsis,
was furnished.

A NEW HYGROMIIDAE FROM ITALY 121

FIG. 7. A: Shell of Helicopsis sp. collected at the foot of Mount Zerboum, Moulay Idris (Morocco). B: Neotype
of Helix aetnea Benoit, 1857, Nicolosi sull'Etna, C. Caroti leg. 1877 (Museum of Zoology, University of
Florence, Italy, MZUF no. 5049/1).

122

dp

GIUSTI, MANGANELLI & CRISCI

bw

FIG. 8. Genitalia of Helicopsis sp. in specimens collected at the foot of Mount Zerboum, Moulay Idris
(Morocco). A: Genital apparatus (gonad excluded). B: Vagina opened to show its inner structure. C: Part of
the penial complex with the penis opened to show the penial papilla.

sis (Schileyko, 1978a: fig. 237; Giusti 8 Man-
ganelli, 1989: fig. 8C) have stylophores
opening in a slit bordered by two large but
simple pleats running along the vagina walls,
it seems plausible to include the new species
in the genus Helicopsis. Nevertheless, the
structure of the penial complex in the new
species differs from that in Helicopsis. The

species of the latter genus, about which there
is detailed knowledge of genital structure,
have a penial papilla consisting of a central
tube wrapped in an external sheath that is
more or less laterally and basally fenestrated
and which basally expands to reach and to
fuse with the penial walls (Schileyko, 1978a:
figs. 235, 238—242, 244; Giusti 8 Manganelli,

A NEW HYGROMIIDAE FROM ITALY 123

1989: figs. 8D, F-H; present paper: Fig. 8C).
In so doing, the sheath distinguishes the cav-
ity of the distal penis (containing the penial
papilla) from that of the proximal penis, al-
though it does not impede communication
through its fenestrations. The cavity of the
proximal penis is traversed by a tube that has
sometimes a lateral pilaster connected by
frenula to the penis walls and which is basally
continuous with the epiphallus and apically
with the central tube of the penial papilla. This
peculiar penis structure, present not only in
the European species but also in the Magh-
rebian species studied herein, can be consid-
ered diagnostic for Helicopsis. Consequently,
the present new species cannot be included
in the genus Helicopsis.

One can argue that the penis structure in
the present new species may be derived from
that in Helicopsis through the reduction of the
penial papilla sheath (the lateral T-section pi-
laster may be a residue of the sheath) and the
loss the peculiar inner structure of the proxi-
mal penis. However this hypothesis seems
less probable than the following one.

Helicotricha carusoi n. sp. has a shell clearly
resembling in overall structure, microsculpture
and colour that of a well-known Mediterranean
species: Microxeromagna armillata (Lowe,
1852) (Giusti 8 Manganelli, 1989: pl. 7, figs.
A-E). The latter species, moreover, has a gen-
ital structure corresponding to that of H. caru-
soi n. sp. (similar talon, similar length and pro-
portions of the parts of the penial complex, a
yellow glandular area on distal penis walls; see
Ortiz de Zarate Y Lopez, 1950: fig. 22; Forcart,
1976: fig. 3; Clerx & Gittenberger, 1977: figs.
102-103; Falkner, 1981: fig. 2; Aparicio, 1982:
fig. 3; Manga Gonzales, 1983: fig. 12; Haus-
dorf, 1988: fig. 13; Manganelli & Giusti, 1988:
figs. 11A—F, 14H; present paper: Fig. 5E). It
only differs in the dart-sac complex, which has
0 + 2 stylophores (instead of 2 + 2), and in
some details of the penial papilla. Other mono-
phyletic groups have been hypothesized to be
formed by genera with 2 + 2 and 0 + 2 sty-
lophores (i.e. Helicella-Candidula and Xero-
lenta-Xeromunda; Hausdorf, 1988, 1990a,
Giusti & Manganelli, 1989; Manganelli 8
Giusti, 1989). Similarly, we hypothesize that
Helicotricha n. gen. forms a monophyletic
group with Microxeromagna. The former ge-
nus can have originated the latter by reduction
of the dart-sac complex; the origin of Helicot-
richa from Microxeromagna by duplication of
the dart-sac complex seems less probable ac-
cording to Schileyko's (1978a, 1984) recon-

struction of the phylogenetic relationships in
the Hygromiidae.

Many other genera of the European and
Russian Hygromiidae resemble Helicotricha
n. gen. in having the vaginal complex with
digitiform glands and dart-sac complex con-
stituted by two opposite pairs of stylophores,
each formed by a large, dart-bearing outer
stylophore and a smaller dartless inner stylo-
phore (i.e. 2 + 2): Hygrohelicopsis Schileyko,
1977 (type species: H. darevskii Schileyko,
1977), Leucozonella Lindholm, 1927 (type
species: Helix rubens von Martens, 1874),
Kokotschashvilia Hudec 8 Lezhawa, 1969
(type species: Helix holotricha Boettger,
1874), Caucasigena Lindholm, 1927 (type
species: Helix eichwaldi Pfeiffer, 1846), Pli-
cuteria Schileyko, 1977 (type species: Helix
lubomirskii Slosarski, 1881), Trichia Hart-
mann, 1840 (type species: Helix hispida Lin-
naeus, 1758), and Edentiella, Polinski, 1929
(type species: Helix edentula Draparnaud,
1801) (see Schileyko, 1978a, 1978b). Of
these, the genus closest to Helicotricha n. gen.
is Hygrohelicopsis because it shows the right
ommatophore retractor independent of penis
and vagina. Nevertheless, Hygrohelicopsis
can be easily distinguished by the structure of
distal genitalia showing inner stylophores ex-
tremely reduced, not visible from outside, dis-
tal vagina absent, and penial papilla bulbous
but simple without external sheath or lateral
pilaster. All the other genera are more distin-
guished from Helicotricha n. gen. because
they have the right ommatophore retractor
passing between penis and vagina, different
and larger shells, different penial papillae and
many other minor differences in the structure
of the genitalia (Schileyko, 1978a, 1978b).

Cladistic Analysis

The entire set of character states (Table 2)
utilized in the traditional approach was used
for cladistic analysis. The latter was at-
tempted even if the characters were few and
limited to genitalia structure. To avoid compli-
cating the analysis, only a limited number of
eastern European genera was considered:
those for which better description exist and
which are at least apparently more closely re-
lated to the new genus.

A total of 105 most parsimonious hypothe-
ses were generated by our data matrix. All
have 37 steps with a consistency index of
0.67 and rescaled retention index of 0.72 after
non-informative characters were excluded.

124 GIUSTI, MANGANELLI 8 CRISCI

OUT

CAUC
EDEN
HELP
KOKO
LEUC
PLIC
TRIC
HYGH

HYGR

CERN
CAND

HELL
PXER
XERM

XERL
NANA

XERT
HELT
MICR

XERS

FIG. 9. The Nelsen consensus tree of 105 cladograms.

The successive weighting procedure did not
discriminate among them, the main difference
being the position of the set of taxa HELP-
LEUC. The Nelsen consensus tree of 105 cla-
dograms (Fig. 9) showed that 11 monophyl-
etic groups appear in all of them, listed with
their synapomorphies as follows:

(1) All the taxa, except HELP, TRIC,
KOKO, PLIC, CAUC, EDEN, LEUC: 2
(1) (parallel with HELP and a reversion
in HYGM).

(2) All the taxa, except HELP, TRIC,
KOKO, PLIC, CAUC, EDEN, LEUC,
HYGH: 15 (6).

(3) HYGM, CERN: 3 (1), 9 (1), 10 (4) (the
first parallel with CAND, XERM, MICR,
XERS).

(4) CAND, HELL, PXER, XERL, XERM,
XERT, NANA, HELT, MICR, XERS: 8
(1) (parallel with HELP, TRIC).

(5) CAND, HELL: 1 (1), 5 (4), 10 (3) (the
first parallel with CERN).

(6) PXER, XERL, XERM: 7 (1).

(7) XERL, XERM: 5 (2).

(8) XERT, NANA, HELT, MICR, XERS: 13
(1) (with a reversion in XERS).
(9) XERT, NANA: 5 (1), 10 (2).
(10) HELT, MICR, XERS: 4 (0), 15 (2) (the
former is a reversion).
(11) MICR, XERS: 3 (1) (parallel with
HYGM, CERN, CAND, XERM).

Character 6 represented a synapomorphy
for the entire set of terminal taxa with a rever-
sion in MICR. Figure 10 is one of the 105
minimum-length trees. It is not a preferred
tree and is simply given to illustrate the evo-
lution of the characters.

As is evident, the main result of cladistic
study has been that of apparently supporting
the conclusion reached in the preceeding
paragraph, i.e. that Helicotricha is a sister
group of Microxeromagna-Xerosecta. It also
appeared to confirm that Helicotricha is not
closely related to Helicopsis as hypothesized
in the same paragraph.

Relationships with Xerosecta suggesied by
cladogram seem logical: Xerosecta is very

A NEW HYGROMIIDAE FROM ITALY 125

OUT
HELP
= TRIC
KOKO
PLIC
5
| = CAUC
= EDEN
+ LEUC
> HYGH
= о х НУСМ
22% CERN
=== CAND
fe =) CVS)
Ses HELL
PXER
= ХЕВМ
= XERL
я МАМА
=8 XERT
HELT
у 5 MICR
= х XERS

FIG. 10. One of the most parsimonious trees generated by the data matrix given by way of illustration of the
evolution of the characters (lines: synapomorphies; double lines: homoplasies; X reversions).

126 GIUSTI, MANGANELLI 8 CRISCI

close to Microxeromagna (Hausdorf, 1988,
1990c; Manganelli 8 Giusti, 1988).

The Suprageneric Systematics of
the Hygromiidae

The phylogenetic hypothesis concerning
Helicotricha n. gen. formulated above, in ad-
dition to those concerning Xerolenta-
Xeromunda and Helicella-Candidula, dis-
agrees with Schileyko’s proposal to
distinguish the Trichiinae (hygromiids with 2
+ 2 stylophores) from the Hygromiinae (hy-
gromiids with O + 2 stylophores).

The present data supports the idea that the
supposed members of one of these “subfam-
ilies” independently evolved from supposed
members of the other so that their derived
status is due to parallelism (Giusti 8 Manga-
nelli, 1987).

The consideration of a larger number of
characters, when eventually available (ge-
netic, cytological, etc.) could promote better
understanding of the phylogeny of Hygromi-
dae (and Helicoidea in general) and verify the
contention of Giusti et al. (1991) who, on the
basis of spermatozoa fine morphology, sug-
gested there are too many family group cat-
egories in the Helicoidea and that it is inop-
portune to produce new schemes and create
new taxa on the basis of old and insufficient
anatomical (or even shell!) characters.

Comparison with Old and Uncertain Taxa of
the Species Group

A search of the taxa described for Sicilian
fauna did not produce positive results. Apart
from Xerotricha conspurcata, X. apicina and
the two species recently revised by us and
recognized to belong to a distinct genus—
Schileykiella: S. reinae Pfeiffer, 1856, and S.
parlatoris, Bivona, 1839 (Manganelli et al.,
1989)—only one small hygromiid with a hairy
shell, Helix aetnea Benoit, 1857, is known in
Sicily.

Some shells were found in the Paulucci
Collection which, according to the label, were
collected in the type locality of H. aetnea,
studied by Benoit himself, and confirmed by
him to fully correspond to his H. aetnea (Mu-
seum of Zoology, University of Florence, Italy,
MZUF no. 5049). Our study of the minute
shell characters confirms Paulucci’s (1878: 6,
32) and Westerlund’s (1889: 302) identifica-
tion of H. aetnea as a juvenile of X. conspur-
cata. Because Benoit’s typical series is un-

traceable and there is no other possible
typical material in the principal malacological
collections, and because Benoit clearly con-
firmed that Paulucci’s material totally corre-
sponded to his species, we have selected a
neotype for this species (MZUF no. 5049/1)
illustrated in Figure 7B. Helix aetnea Benoit,
1857, consequently becomes a junior syn-
onym of X. conspurcata.

Geographic Distribution

All the Aeolian Islands. A few shells from
the island of Ustica (NW of Sicily) [UTM ref.:
33SUC38, 48] may belong to the new spe-
cies, but since they are sexually immature
and anatomical study is impossible, no con-
clusions can be drawn. Although its presence
in Sicily is highly probable, a search of the
literature and of our materials from Sicily, and
some Sicilian and Maltese Islands did not
bring to light any useful information.

Ecology

Helicotricha carusoi n. sp. has been found
to live under stones, dry leaves and pieces of
wood in many different places on single is-
lands, frequently together with Xerotricha
conspurcata. Like the latter species, the
present species is thus a xeroresistant ele-
ment, well adapted to Mediterranean habitats.

ACKNOWLEDGMENTS

We thank Mr. L. Gamberucci and Mrs. A.
Daviddi for technical assistance and Mrs. H.
Ampt for linguistic revision. This research was
supported by CNR, МР! 40% and МР! 60%
grants, and was carried out as part of a
project of the Escuela de Specialization en
Ambiente y Patologia Ambientale of the Uni-
versidad Nacional de La Plata, Argentina, and
Universita di Siena, Italy.

Thanks also to two anonymous revisers for
the detailed analysis of manuscript which im-
proved the quality of the paper.

LITERATURE CITED

APARICIO, M. T., 1982, Observations on the anat-
omy of some Helicidae from central Spain. Mal-
acologia, 22 (1-2): 621-626.

BENOIT, L., 1857, Illustrazione sistematica critica
iconografica de' testacei estramarini della Sicilia

A NEW HYGROMIIDAE FROM ITALY 127

Ulteriore e delle isole circostanti. i-xvi + 1-52
(+ 1-9 pls. ?), Napoli.

CLERX, J. P. M. 8 Е. GITTENBERGER, 1977, Eni-
ger über Cernuella (Pulmonata, Helicidae). Zool-
ogische Mededelingen (Leiden), 52: 27-56, pls.
1-2.

DAMJANOV, S. G. 4 I. LIKHAREV, 1975, Gas-
tropoda terrestria. Fauna Bulgarica, 4: 425 pp.
FALKNER, G., 1981, Lehmannia valentiana und
Microxeromagna vestita bei Grimaldi—neu fúr
die italienische Fauna. Mitteilungen der Zoologis-

chen Gesellschaft Braunau, 3: 388-391.

FARRIS, J. S., 1989, The retention index and the
rescaled consistency index. Cladistics, 5: 417-
419.

FORCART, L., 1976, Die Cochlicellinae und Heli-
cellinae von Palástina und Sinai. Archiv ftir Mol-
luskenkunde, 106 (4/6): 123-189.

FRANC, A., 1968, Sous-classe des pulmonés.—In
P. P. GRASSE, Traité de zoologie. Anatomie,
systématiques, biologiques, 5 mollusques gas-
téropodes et scaphopodes (3): 325-607.

GASULL, L., 1972, Descripcion de una nueva es-
pecie de Helicella de la provincia de Huelva, He-
licopsis (Helicopsis) altenai n. sp. (Gastrop. Pul-
mon.). Boletin de la Sociedad de Historia Natural
de Baleares, 17: 73-75.

GASULL, L., 1985, Fauna malacologica continental
de la provincia de Huelva. Miscellania Zoologica,
9: 127-143.

GITTENBERGER, E., 1969, Eine neue Art der Gat-
tung Helicopsis (Gastropoda, Helicidae, Helicel-
linae) aus Niederósterreich. Basteria, 33 (1-4):
63-68.

GITTENBERGER, E., M. R. ALONSO & M.
IBANEZ, 1989, Helix pavida Mousson, 1872, and
H. nubivaga Mabille, 1882 (Gastropoda, Heli-
cidae, Helicellinae) poorly known Helicellinae
from the Canary Islands. Basteria, 53 (4-6):
117-125.

GITTENBERGER, E. 8 M. Y. MANGA, 1977, Some
new species of the genus Helicella (Pulmonata:
Helicidae) from the province Leon. Zoologische
Mededelingen (Leiden), 51 (11): 177-189, pls.
1-2.

GITTENBERGER, E. 8 J. G. M. RAVEN, 1982, A
new Helicella (Helicidae, Helicellinae) from the
Cantabrian Mountains, Spain. Basteria, 46 (1-4):
79-83.

GIUSTI, F., 1973, Notulae malacologicae, XVIII. |
molluschi terrestri e salmastri delle Isole Eolie.
Lavori della Società Italiana di Biogeografia,
(n.s.), 3: 113-306, pls. 1-16.

GIUSTI, F. 8 G. MANGANELLI, 1987, Notulae ma-
lacologicae XXXVI. On some Hygromiidae (Gas-
tropoda: Helicoidea) living in Sardinia and in Cor-
sica. (Studies on the Sardinian and Corsican
malacofauna VI). Bollettino Malacologico, 23:
123-206.

GIUSTI, F. 8 G. MANGANELLI, 1989, Notulae ma-
lacologicae, XLV. A new Hygromiidae from the
Tyrrhenian islands of Capraia and Sardinia with
notes on the genera Xeromicra and Xerotricha

(Pulmonata: Helicoidea).
logico, 25 (1/4): 23-62.
GIUSTI, F. 8 G. MANGANELLI, 1990, Ciliellopsis
oglasae a new Hygromiidae from Montecristo Is-
land (Pulmonata: Hygromiidae). Journal of Con-

chology, 33: 269-277, pls. 27-29.

GROSSU, A. V., 1983, Gastropoda Romaniae 4.
Ordo Stylommatophora Suprafam.: Arionacea,
Zonitacea, Ariophantacea si Helicacea. Bucur-
esti, Editura Litera, 564 pp.

HAUSDORF, B., 1988, Zur Kenntnis der system-
atischen Beziehungen einiger Taxa der Helicelli-
nae lhering 1909 (Gastropoda: Hygromiidae). Ar-
chiv fúr Molluskenkunde, 119 (1/3): 9-37.

HAUSDORF, B., 1990a, Die Xeromunda-Arten des
griechischen Festlandes (Gastropoda: Hygromi-
idae). Archiv fúr Molluskenkunde, 119 (4/6): 407—
432.

HAUSDORF, B., 1990b, Zur Kenntnis einiger Arten
der Gattung Helicopsis Fitzinger aus Griechen-
land und der Türkei (Gastropoda: Hygromiidae).
Archiv für Molluskenkunde, 120 (1/3): 57-71.

HAUSDORF, B., 1990c, Uber die Verbreitung von
Microxeromagna armillata (Lowe, 1852) und Xe-
rotricha conspurcata (Draparnaud, 1801) in
Griechenland und der Türkei (Gastropoda, Pul-
monata: Hygromiidae). Malakologische Abhand-
lungen, 15: 55-62.

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

HESSE, P., 1934, Zur Anatomie und Systematik
palaearktischer Stylommatophoren. Zoologica
(Stuttgart), 33 (1), Heft 85: 1-59, pls. 1-9.

KTARI, M. H. & M. REZIG, 1976, La faune mala-
cologique de la Tunisie septentrionale. Bulletin
de la Societe des Sciences Naturales de la Tu-
пе, 11: 31-74.

MADDISON, W. P., M. J. DONOGHUE & R. MAD-
DISON, 1984, Outgroup analysis and parsimony.
Systematic Zoology, 33: 83-103.

MANGA GONZALES, M. Y., 1983. Los Helicidae
(Gastropoda: Pulmonata) de la provincia de
Leon. Leon, 394 pp.

MANGANELLI, G. & F. GIUSTI, 1988, Notulae ma-
lacologicae, XXXVIII. A new Hygromiidae from
Italian Apennines and notes on the genus Cer-
nuella and related taxa (Pulmonata: Helicoidea).
Bollettino Malacologico, 23: 327-379.

MANGANELLI, G. & F. GIUSTI, 1989, Notulae ma-
lacologicae, ХИН. Xeromunda Di Мапа Di Mon-
terosato in Italy (Pulmonata: Hygromiidae). Bol-
lettino Malacologico, 25 (1-4): 1-22.

MANGANELLI, G., I. SPARACIO & F. GIUSTI,
1989, New data on the systematics of two Sicilian
land snails, Helix parlatoris Bivona, 1839 and He-
lix reinae L. Pfeiffer, 1856 and description of
Schileykiella n.gen. (Pulmonata: Hygromiidae).
Journal of Conchology, 33: 141-156, pls. 10-16.

NELSEN, С. I., 1979, Cladistic analysis and syn-
thesis: principles and definitions, with a historical
note on Adanson’s “Familles des plantes”
(1763-1764). Systematic Zoology, 28: 1-21.

NORDSIECK, H., 1987, Revision des System der

Bollettino Malaco-

128 GIUSTI, MANGANELLI 8 CRISCI

Helicoidea (Gastropoda: Stylommatophora). Ar-
chiv für Molluskenkunde, 118: 9-50.

ORTIZ DE ZARATE Y LOPEZ, A., 1950, Observa-
ciones anatomicas y position sistematica de
varios helicidos espanoles. Boletin de la Real So-
ciedad Espanola de Historia Natural, 43: 21-85,
pls. 1-2.

PAULUCCI, M., 1878, Matériaux pour servir a
l'étude de la faune malacologique terrestre et flu-
viatile de l'Italie et des ses isles. iv + 54 pp.
Paris.

PIANTELLI, F., F. GIUSTI, F. BERNINI 8 G. MAN-
GANELLI, 1990, The mollusc and oribatid fauna
of the Aeolian and Tuscan Archipelagos and the
island equilibrium theory. In: Proceeding of the
international symposium “Biogeografical aspects
of insularity” (Roma 18-22 May 1987). Acca-
demia Nazionale dei Lincei, Atti dei Convegni
Lincei, 85: 117-154.

PLATNICK, N. I., 1989, An empirical comparison of
microcomputer parsimony programs. Il. Cladis-
tics, 5: 145-161.

PRIETO, C. P., 1986, Estudio sistematico y bio-
деодгайсо de los Helicidae (sensu Zilch, 1959—
60) (Gastropoda: Pulmonata: Stylommatophora)

del Pais Vasco y regiones adyacentes. Univer-
sidad del Pais Vasco, Facultad de Ciencias, doc-
toral thesis, 393 pp., 10 pls.

SCHILEYKO, A. A., 1978a, Molluscs. Land mol-
luscs of the superfamily Helicoidea. Fauna
SSSR, (n.s.) 17: 1-348 [in Russian].

SCHILEYKO, A. A., 1978b, On the systematics of
Trichia s. lat. (Pulmonata: Helicoidea: Hygromi-
idae). Malacologia, 17: 1-56.

SCHILEYKO, A. A., 1984, Molluscs. Terrestrial
molluscs of the suborder Pupillina of the fauna of
the USSR (Gastropoda, Pulmonata, Geophila).
Fauna SSSR, (n.s.) 130: 1-399. [in Russian].

WATROUS, L. & Q. WHEELER, 1981, The out-
group comparison method of character analysis.
Systematic Zoology, 30: 1-11.

WESTERLUND, C. A., 1889, Fauna der in der pa-
laarktischen Region lebenden Binnenconchylien.
Il. Genus Helix, 2, 473 + 30 pp. R. Friedlander &
Sohn., Berlin.

ZILCH, A., 1960, Euthyneura. In W. Wenz: Gas-
tropoda. Handbuch der Paläozoologie, 6 (2, 4th
part): 601-835 + i-xii.

Revised Ms. accepted 2 December 1991

MALACOLOGIA, 1992, 34(1-2): 129-141

STRUCTURE AND COMPOSITION OF THE SHELL OF THE
ARCHAEOGASTROPOD LIMPET LEPETODRILUS ELEVATUS ELEVATUS
(MCLEAN, 1988)

Stephen Hunt

Department of Biological Sciences, University of Lancaster, Bailrigg, Lancaster,
LA1 4YQ England

ABSTRACT

Shells from the hydrothermal vent-dwelling archaeogastropod limpet Lepetodrilus elevatus
elevatus are described in terms of their mineralogy, fine structure and composition. The shell is
mainly crossed-lamellar aragonite, with narrow inner and outer prismatic layers and no nacre.
The thick proteinaceous periostracum has an unusually high glycine content. Sulphur, detected
in the periostracum by X-ray microprobe analysis, cannot be attributed to either sulphur-con-
taining amino acids or free elemental sulphur. Periostracum and shell both contain phosphorus,
but metals other than calcium are present in traces only. Most shells examined show erosion in
the form of pits rich in sulphur and phosphorus. The significance of the unusual periostracal
composition in relation to the shell mineralogy, the propensity for shell erosion, and the chem-
istry of the vent water environment, is discussed.

Key words: archaeogastropod, Lepetodrilus, shell, calcification, aragonite, cross-lamellar, pe-

riostracum, polyglycine.

INTRODUCTION

Exploration of deep ocean hydrothermal
sites discovered in the 1980s revealed an un-
expected rich fauna with molluscs well repre-
sented. While large bivalves attracted much
attention, gastropods are most numerous,
mainly as small limpet forms, new and unique
to this special environment. Hickman (1983)
noted the presence in the Galapagos Rift and
East Pacific Rise (13°N and 21°N) sites of
three new archaeogastropod groups that
could not be assigned to then-recognised su-
perfamilies. These have been described by
McLean (1985, 1988) and by Fretter (1988).
Here | describe the architecture and compo-
sition of the shell of Lepetodrilus elevatus el-
evatus, a member of the archaeogastropod
superfamily Lepetodrilacea (McLean, 1988).

MATERIALS AND METHODS

Specimens of L. elevatus elevatus (see
McLean, 1988, for the systematic description
and list of the principal repositories of the type
material) were collected by Dr. F. Gaill of the
Centre du Biologie Cellulaire, CNRS, Ivry sur

_ Seine, France, at 2600 m depth during the

_ 1984 Biocyarise cruise using the submersible

Cyana at 12°48'N, 103°56’W.

129

Scanning electron microscopy was carried
out on shells that had been rinsed with dis-
tilled water, air dried, mounted with graphite
paste cement onto aluminium support stubs,
and sputter coated with gold. Microscopy was
performed with a Jeol JSM50A scanning elec-
tron microscope at 15kv. For microprobe
analysis, specimens were mounted on graph-
ite support stubs and were rotary coated with
carbon. Analyses were carried out at 15kv ex-
citation voltage by a Kevex energy dispersive
analysis system at 20ev per channel gain with
a preset live time of 100 sec.

For thin sectioning, periostracum was ob-
tained by shell dissolution with 0.1M hydro-
chloric acid, washed in distilled water, fixed
overnight in pH 7.0, 0.1M cacodylate buffer
containing 5% glutaraldehyde, and post-fixed
in cacodylate-2% osmium tetroxide for 30
min. Following dehydration through alcohols,
fixed material was embedded in Spurr's resin
and sectioned at 50 nm or less using a dia-
mond knife. Sections were stained with uranyl
acetate and Reynold's lead citrate for 30—60
min each. Sections were examined at 100kv
in a Jeol 100CX STEM microscope.

X-ray diffraction patterns were determined
using finely ground specimens of whole shell
held in glass capillaries mounted in a helium-
filled flat camera of working distance 3.5 cm
with Ni-filtered Cu/Ka radiation and pinhole

130

FIG. 1. Lepetodrilus elevatus elevatus. A. Lateral
view. B. Dorsal view. C. Ventral view.

collimation. Fibre diffraction patterns were
made using small strips of wet material
clamped in holders and slightly tensioned be-
fore drying. Standard patterns were made
with powdered calcite (Iceland spar) and ara-
gonite (scleractinian exoskeleton).

Amino acid analyses were carried out on
specimens hydrolysed in evacuated tubes
with 3M methane sulphonic acid containing
0.2% w/v tryptamine at 100°C for 48 hours.
Neutralized aliquots were analysed using an
LKB model 4101 amino acid analyser.

RESULTS
General Description of the Shells

Shells of L. elevatus elevatus are typically
about 2.25 mm high, with an oval aperture

FIG 2. Inner surface of the shell (i) showing the
folding over of the periostracum (p) at the shell
edge and the granular structure of the shell with
large and small scale irregularities (li and si) ar-
rowed. A. Low magnification. B. Same region at a
higher power.

about 4 mm by 3 mm (Fig. 1A-C). The exter-
nal surface has low concentric, closely
spaced growth ridges and a smoother apex.
An olive-brown periostracum folds over the lip
of the shell with a 0.2-0.3 mm inner overlap
(Figs. 1C, 2A). The inner surface of the shell
is smooth though higher magnifications re-
veal regularly distributed granules and shal-
low pitted irregularities (Fig. 2A, B).

Shell Mineralogy

Figure 3 shows the diffraction pattern of
powdered shell. The principal reflections are
aragonitic, with little or no calcitic contribution.

Ultrastructure of the Mineral Phase

Fractured edges of the shell show three
major shell layers—outer, middle and inner,
of which the middle is the major and com-
posed quite uniformly of the crossed-lamellar
structure type (Figs. 4-6). Between this layer
and the periostracum is a columnar layer of
irregular fibrous prismatic organization (Fig.
7). The innermost shell layer is columnar in
irregular simple prismatic form (Fig. 8). Here a
thin zone, no more than 2 um deep and
formed of either short prisms or granules, de-
marcates this layer from the crossed lamellar.

LEPETODRILUS SHELL STRUCTURE AND COMPOSITION 131

FIG 3. Powder diffraction patterns of A. Aragonite.
B, Lepetodrilus shell. C. Calcite.

There is no evidence of an innermost nacre-
ous layer being present.

Elemental Analysis

X-ray microprobe analyses of the main min-
eralized layers show calcium to be the only
detectable element present in significant
amounts (Fig. 9, Table 1), presumably as the
carbonate. The traces of sulphur and phos-
phorous would be unlikely to make an impor-
tant contribution to shell mineralogy even if
they were present as calcium phosphate and
sulphate. Although vent waters are transition
and heavy metal rich, none are detectable in
the shell mineral. Traces of magnesium, sili-
con and aluminium could indicate chance in-
corporation clay-forming suspensates. Roux
et al. (1985) have noted that concentrations of
trace elements in the aragonitic shell of the
hydrothermal vent bivalve Calyptogena did
not reflect vent-water trace element concen-
trations but were similar to those found in ara-
gonitic shell of other marine bivalves.

Periostracum

Figure 10 is a low-power transmission elec-
tron micrograph of sectioned periostracum.
No laminated, orthogonal or helicoidal fibrous
architecture of the type found in some marine
and terrestrial prosobranch gastropod perios-
traca (Bevelander & Nakahara, 1970; Hunt &
Oates, 1970, 1978, 1985) was found. The
faint bands in the plane of the periostracum
are probably depositional irregularities with-
out compositional or orientational signifi-
cance. The darker and more homogeneous
zone adjacent to the shell may represent a
phase of rapid deposition and tanning. The
inner surface has an irregular profile and
forms a thin, more electron-luscent layer than
the main body of periostracal protein into
which it merges via a narrow, more densely
staining zone. The outer surface is also irreg-
ular. What may be knobs or ridges adhere to
a thick layer of vacuolated granular material
that may be a deposit of microorganisms and
detritus. The remnants of bacteria are present
in the vacuoles.

Table 2 gives the amino acid analysis of the
periostracum, and Table 3 lists values for key
features of the composition enabling compar-
isons with other structural proteins. The peri-
ostracum has a high glycine content, over 70
residue percent. Glycine-rich proteins are
common among structural proteins, and
whereas many periostracal proteins are of
this type, it is unusual for those of marine gas-
tropods to be glycine-rich. Figures 11 and 12
show the composition of this periostracum
placed in relation to other gastropod periost-
raca. Its glycine content is higher than in any
other gastropod (62 species) recorded by us.
This, coupled with its hydrophobic index
(+ 47), ranks it with those of terrestrial proso-
branchs. The only periostraca with higher
glycine levels (over 80 residue percent) are
from bivalve molluscs at hydrothermal vent
sites. But, unlike the latter, the sulphur-con-
taining amino acid levels here are low or ab-
sent (Hunt, 1987). The high beta turn poten-
tial value, coupled with low values for helix
and beta sheet potential, point to a protein
secondary structure high in random folding.

Figure 13A shows the typical X-ray emis-
sion profile of the periostracum probed to
avoid surface deposits. Noteworthy is the
large sulphur peak. There are traces of iron
and copper and a little phosphorus (Table 4).
In the absence of calcium, the latter cannot be
accounted for as hydroxyapatite, though this

FIG. 4. Fractured edge of shell with the dorsal surface at the top and showing the periostracum (p) and the
three main calcified layers, outer (0), middle (m) and inner (i). Cross-lamellation (cl) is evident in the deeper

middle layer.

FIGS. 5, 6. Two views of the crossed-lamellar organization of the middle layer of the shell.
FIG. 7. The outer layer of the shell with the periostracum (p) overlying an irregular layer of aragonite

columns.

FIG. 8. The innermost layer of the shell composed of irregular columns sharply separated from the overlying
cross-lamellar material by a narrow zone of short prisms or granules.

is present in some _ bivalve periostraca
(Waller, 1983). The absence of sulphur-con-
taining amino acids demands another expla-
nation of the high sulphur content. Many mi-
croorganisms present at the hydrothermal
vents have an energy metabolism based on
oxidation of sulphur compounds. Deposition

of elemental sulphur as part of this process is
common and occurs in the gills of Calypto-
gena as part of a symbiotic relationship (Fel-
beck et al., 1985). Extraction, however, of the
periostracum with carbon disulphide only re-
moves traces of elemental sulphur (Fig.
13B,C, Table 4).

LEPETODRILUS SHELL STRUCTURE AND COMPOSITION 133

Ca

Al...
Mg 5 Sc
Ca

FIG. 9. Analysis of the shell by X-ray microprobe.
For this analysis, the beam was centred in the
crossed-lamellar layer, but closely similar patterns
were obtained for the outer and inner calcified lay-
ers. The analysis is energy dispersive, and the ab-
scissal axis is in terms of X-ray energy and is a
measure of increasing atomic number, whereas the
ordinate axis is a measure of received X-ray photon
counts at the detector.

X-ray diffraction patterns prepared from
flattened strips of periostracum show no ori-
ented fibre reflections. There is considerable
diffuse scatter and a weak ring corresponding
to a 0.387 nm spacing, which tends to confirm
the secondary structure as being a compact
and highly folded random coil. The helical
coiled polyglycines | and Il are absent.

Shell Erosion

All specimens examined had erosion pits,
most commonly near the shell apex. Many of
these pits penetrate the shell and are usually
partially filled or closed by brown or black ma-
terial. Figure 14 shows several such pits with
characteristic rounded profiles around single
or multiple coalescing sites of penetration.
The pitting is often associated with areas of
periostracal cracking that follow the pit profile
(Fig. 15).

Pits penetrate the shell in a series of ter-

raced steps. In several, the first layer re-
vealed beneath the periostracum is more
finely granular than the deeper terraces or the
coarser pit-bottom material (Figs. 16, 17).

Figure 18 and Table 5 show X-ray micro-
probe analyses of eroded and pitted areas. A
characteristic feature are the high levels of
sulphur and phosphorus in the pits and, as
might be expected, increasingly high calcium
counts from lower within them.

DISCUSSION

Leptodrilus elevatus elevatus is found in
huge numbers in close association with the
giant vestimentiferan tube worm Riftia at
depths of around 2600 m, close to the hydro-
thermal vents, where the sea water has high
concentrations of such reduced gases as hy-
drogen sulphide, together with dissolved sul-
phides of iron, manganese, zinc and copper
and low oxygen tension. Many of the animals
living there take advantage of chemo-
lithotrophic bacteria that base their metabo-
lism on oxidation of hydrogen sulphide and
other reduced forms of sulphur. Endosymbi-
otic relationships exist between vent bacteria
and bivalve molluscs (Felbeck et al., 1985), in
which hydrogen sulphide, concentrated from
the sea water by the mollusc, is transferred to
bacteria in the gills and used as the energy
source for synthesis of organic compounds
used by the animal. De Burgh 4 Singla (1984)
describe how a vent archaeogastropod limpet
also obtains nutrients by endocytosis of bac-
teria growing upon its gills.

The need to absorb oxygen at low concen-
trations from waters containing high metal
concentrations, as well as nutritional strate-
gies involving consumption of bacteria, which
may have absorbed metallic elements in their
cell coats (Juniper et al., 1986), might lead to
expectation that mineralization would yield
shells reflecting the sea water chemistry, but
this seems not to be so. Although traces of
magnesium, aluminium and silicon are nearly
always present in the shells, none of the

TABLE 1. Elemental composition of the main mineralized layer of the Lepetodrilus shell determined by
X-ray microprobe analysis. Values are as emitted X-ray counts for each integrated elemental peak as a
percentage of total counts integrated across the whole energy spectrum. For conditions, see text. These
values relate to Figure 9, on which the traces of magnesium and phosphorus barely show.

Element Na Mg Al Si B S Cl Ca
counts %total Vai 0.54 3.67 1.19 0.48 4.75 1.28 86.99

HUNT

FIG. 10. Transmission electron micrograph of a transverse section through the periostracum (p). The outer
surface with bacterial debris contamination (bc) is at the left and the inner shell contact surface (if) at the

right. The specimen has been decalcified.

TABLE 2. Amino acid composition of the periostracum of L. elevatus elevatus reported as residues of
each amino acid per thousand total amino acid residues.

Aspartic acid И Alanine
Threonine 6 Cysteine
Serine 8 Valine
Glutamic acid 11 Methionine
Proline 8 Isoleucine
Glycine 713 Leucine

10 Tyrosine 30
0 Phenylalanine 12
34 Histidine 14
2 Lysine 41
23 Arginine 22
49 Tryptophan—Not determined

heavier metallic elements appear to be in-
cluded during calcification. The only elements
achieving any prominence against the domi-
nant calcium component are phosphorus and
sulphur.

Traces of phosphorous in the shells might
be accounted for by phosphoprotein in the
matrix. Petit et al. (1980) have found calcium
“spherites” containing phosphorus in the vas-
cular channels and connective tissue imme-
diately adjacent to mantle edge cells of the
freshwater bivalve Amblema, implicating
these in shell formation. Marsh & Sass (1983)
have demonstrated calcium-binding phos-
phoprotein particles in the extrapalial fluid
and innermost shell lamella of estuarine and
marine bivalves.

Shell sulphur could be due either to traces

of calcium sulphate or perhaps sulphated
polysaccharides. Nautilus shell, for example,
contains polysaccharide sulphates in the
membrane overlying the nacreous crystals
(Crenshaw & Ristedt, 1976). Crassostrea has
sulphated proteins in the matrix (Wilbur &
Salleuddin, 1983). Organic ester sulphate
could also account for the high sulphur con-
tent of the periostracum.

Aragonite appears to be the predominant
form of crystaline calcium carbonate in the
shell. There are no precedents that would in-
dicate if very high hydrostatic pressures can
affect shell mineralization in molluscs, though
in vitro work shows that calcium carbonate
crystalization equilibria are pressure sensi-
tive—aragonite becomes the most stable form
at higher pressure and temperature (Mellor,

LEPETODRILUS SHELL STRUCTURE AND COMPOSITION 135

TABLE 3. Side-chain characteristics and conformational predisposing properties of the amino acid
composition of the Lepetodrilus and other molluscan periostracal proteins calculated from the data of
Table 2.

Av. other Av.
Side Chain Lepetodrilus marine gastropods? marine bivalves'°
Acidic' 28 237 60
Basic? 77 103 71
Polar? 149 501 264
Apolar* 136 300 189
Polar/Apolar lol 1.6 1.4
Hydrophobic index? +47 —226 +235
Helix potential? 215 513 270
Sheet potential” 229 428 352
Turn potential? 746 420 620

2Asp + Glu. His + Lys + Arg. ®Acidic + Basic + Thr + Ser + Tyr. “Pro + Ala + Val + lle + Leu + Phe. °Calculated
according to the method of Andersen (1979) by ranking each residue according to the characteristic energy change involved
in transfer of it between water and ethanol. Hydrophilic residues have negative values and hydrophobic positive. Summation
of the product of the i,, amino acid content and AG for the i,, transfer, for all amino acids, gives the hydrophobic index.
SCalculated from data in Chou 8 Fasman (1977) using the conformational frequency parameters P as weighting factors to
load the overall numbers of helix-promoting amino acids in the periostracal proteins. Thus, “helix potential” is given by the
number of residues (per 1000 total) of an amino acid multiplied by its Pa value and summed for the residues Glu, Met, Ala,
Leu, Lys, Phe, lle and Val (in decreasing order of helix-promoting ability). Typical values for a “random coil” protein (resilin),
a nearly 100% a-helical protein (tropomyosin), and a globular protein with low helix (2%) and high beta sheet (57%) content
(concanavalin A) are 350, 956 and 561, respectively (calculated from amino acid compositions in residues per 1000 total
residues) (Andersen & Weis-Fogh, 1964; Huddart & Hunt, 1975; Wang et al., 1971; Reeke et al., 1975). 7As for ©, but using
the beta sheet-promoting conformational parameters P, to weight and sum the contents of Val, lle, Tyr, Phe, Leu, Cys, Thr
and Met (Chou 8 Fasman, 1977). Values for resilin, tropomyosin and concanavalin A are 226, 344, and 532, respectively.
SAsp + Ser + Pro + Gly. These residues show high frequency in the four amino acids at a beta turn (Chou 4 Fasman,
1977). Values for resilin, tropomyosin and concanavalin A are 636, 174 and 388, respectively. *38 non-hydrothermal

species (Hunt, 1987). ‘°29 non-hydrothermal species (Hunt, 1987).

1960). Probably the chemistry of the organic
matrix is more important. Commenting upon
the entirely aragonitic form of the shell in the
hydrothermal vesicomyid C. magnifica and
the mixed aragonitic and calcitic form of a hy-
drothermal mytilid, Lutz (1982) observes that
in the eastern Pacific regions, the depths of
2000-3000 m, where sampling has taken
place, are well above reported calcite com-
pensation depths but substantially below
those for aragonite compensation.

Because the limpet is in a chemically hos-
tile niche, the aragonitic mineralogy might
seem to be a disadvantage being less stable
and slightly more soluble than the calcitic.
The unreliability of aragonite is seen where
the periostracum has been damaged and
rapid dissolution begins to penetrate the shell.

Clearly the periostracum offers much pro-
tection for the shell and probably because its
composition is atypical of marine gastropods.
Hunt & Oates (1985) note that there are prob-
ably two main families of periostracal proteins
in gastropods—those with a low glycine con-
tent and negative hydrophobic index, and

those with a medium to high glycine content
and positive hydrophobic index. The latter is
typical also of most marine and freshwater
lamellibranchs. The former have easily wet-
ted periostraca with high permeabilities to wa-
ter and ions; the latter have the opposite char-
acteristics. | have commented elsewhere
(Hunt, 1987) on the resistance of shells of the
latter type to attack by 2M hydrochloric acid
through an intact periostracum; Lepetodrillus
behaves in this way, showing only periostra-
cal blistering, carbon dioxide evolution, and
shell erosion where the periostracum is al-
ready damaged.

Marine prosobranch periostraca, for which
we have analyses available, do fall into both
groups, but the hydrophilic class are more
common. Assuming that the primary role of
periostraca is in shell deposition, then given
the type of massive calcitic shell formed
by many marine gastropods, an ion-imper-
meable barrier would not normally be an es-
sential protective feature where acid pH is a
less frequently encountered environmental
pressure. But here the environment is more

136 HUNT

500

400

300

APOLAR

200 N
TP
№
100
0
0 100 200

300

400 500 600

POLAR

FIG. 11. The relationship between apolar and polar amino acids, as defined in Table 3, for gastropod
periostraca. The different groups of gastropod are designated as follows:- TP, terrestrial prosobranchs; TPu,
terrestrial pulmonates; FWP, freshwater prosobranchs; FWPu, freshwater pulmonates; MP, marine proso-
branchs. Amino acid analyses are drawn partly from Meenakshi et al. (1969) and Degens et al. (1967) and
from analyses performed in this laboratory but not yet published. Numerals after the designations represent
numbers of species sampled, and the enclosed areas are the limits of apolar/polar value clusters —some
gastropod groups have periostraca of more than one type. The graphical line demarcates apolar periostraca
on the left from polar on the right. The star represents the position of the Lepetodrilus analysis.

extreme and shell mineralology unstable, de-
manding an impermeable and strong perios-
tracum.

Account may also have to be taken of the
lower oxygen tension at which the limpets
function. Operating metabolism partly anaer-
obically may mean that consistently elevated
levels of succinic acid could form in the extra-
pallial fluid within the shell (Crenshaw & Neff,
1969; Lutz & Rhoads, 1980), demanding a
constant internal regeneration of the shell to
compensate for acid dissolution. Better exte-
rior protection would then be advantageous.

Although the composition of the perios-
tracum is unusual, its hydrophobicity is still

inside the range for marine gastropods (Fig.
12). It is the glycine content that pulls it away
from the major groupings, and this factor too
contributes to the anomalous position on the
polarity distribution diagram. The almost
equal balance of polar and apolar character is
shared with some terrestrial and freshwater
prosobranchs, though not with most marine
species (Fig. 11). A high glycine content may
promote a densely packed, highly folded ran-
dom structure that density of which could be
as helpful in reducing ion-permeability as hy-
drophobicity. Lamellibranch periostraca from
hydrothermal species also have unusual
amino acid compositions with high glycine

LEPETODRILUS SHELL STRUCTURE AND COMPOSITION 137

600

200

Hydrophobic Index
o

'
№
о
o

-400

Glycine

FIG. 12. The relationship between hydrophobic in-
dex and glycine content for gastropod periostraca.
Hydrophobic index is defined in the footnote to Ta-
ble 3. Because glycine has only a hydrogen in the
side chain, it is assigned zero hydrophobicity (arbi-
trarily, as thermodynamically it will have a finite AG
value for transfer between water and a less polar
solvent) and is assumed to contribute neither net
polar or apolar value. Designations, numerals and
symbols have the same meanings as in Figure 11.

FIG. 13. Analysis of the periostracum by X-ray mi-

contents. That from Bathy modiolus thermo- croprobe. A. Periostracum untreated other than by
philus contains 85 residue percent though the dilute acid decalcification to remove the shell. В.
few remaining amino acids give it more hy- Periostracum extracted with carbon disulphide. C.
drophobic character (Hydrophobic index: The carbon disulphide extract dried onto the graph-
+ 156) (Hunt, 1987). ite stub.

TABLE 4. Elemental composition of the periostracum of Lepetodrilus determined by X-ray microprobe
analysis. A. Untreated material removed from the shell with dilute acid. B. Another specimen after
removal from the shell and extraction with carbon disulphide. C. The carbon disulphide extract
concentrated by evaporation at room temperature and dried onto the graphite-coated stub. These
compositions relate to Figure 13A—C. Traces of aluminium and silicon may originate from clay particles
on the surface of the material. Low levels of material recovered by CS, extraction do not make count
integration meaningful.

Element Al Si P 5 Fe Cu

Counts %total

A — 1.01 3.19 93.41 2.04 0.34
В 1.22 1.26 1.97 93.96 0.94 0.11

С {гасе {гасе = trace == trace

138 HUNT

FIG. 14. Scanning electron micrograph of the shell surface in the apical region showing erosion pits pen-
etrating the periostracum and the underlying calcified material.

FIG. 15. Detail of an erosion pit from the lower center of Figure 14. Note the cracking of the periostracum
at the right which seems to follow the profile of the pit.

FIG. 16. Detail of the pit shown in Figure 15 which illustrates the stepped progression of erosion.

FIG. 17. Detail from Figure 16 illustrating what appears to be residual strands and sheets of organic matrix
in and around the deepest pit.

LEPETODRILUS SHELL STRUCTURE AND COMPOSITION 139

A.

FIG. 18. Analysis of erosion pits by X-ray micro-
probe. A. Interior of a shell outside surface pit of the
type shown in Figure 17. B. Interior of a pit that has
penetrated the shell to the inside and probed from
within the shell.

Most of the limpet shells are eroded by
small pits showing that if the periostracum is
penetrated the vulnerable aragonitic shell will
be damaged. At 2000 m depth, wave action
can hardly be the cause of periostracal dam-
age. Most pits and periostracal cracks occur
near the shell apex, where the protein sheet is

oldest and curved more sharply to conform to
the shell's contour. Aging, chemical deterio-
ration, non-specific cross-linking could all
contribute to progressive stiffening and brittle-
ness causing cracking and porosity. Another
possibility is biodegradation. Baross & Dem-
ing (1985) note extensive colonization of the
limpets’ shell surfaces by microorganisms.
McLean (1988) remarks on heavy infestations
by an unidentified sedentary organism form-
ing globular irregularities on the shells of this
species and the related L. pustulosus. These
organisms may etch and damage the under-
lying periostracum. Herrera-Duvault & Roux
(1986) have studied corrosion in mussel
shells from the 13°N thermal vent site. After
predator damage or microboring, bacterial ac-
tivity destroys the organic matrix of the shell,
making the dissolution of the aragonite layers
easier.

High phosphorous and sulphur levels,
found here in the erosion pits, are difficult to
account for. The dark material seems to be
organic, as the microprobe detects no metals
other than calcium. Aragonite dissolution will
leave behind shell matrix proteins. Possibly
these are being secondarily tanned to form
products interacting with sulphur and phos-
phorous-containing inorganic species in the
water. Fibrous and sheet-like residues are
visible in the pit shown in Figure 17. A more
likely explanation is that pits provide a growth
site for microorganisms. Accumulated bacte-
rial material could be rich in sulphur and phos-
phorus as sulpho- and phospholipids and as
the phosphate groups of nucleic acids and
cell wall polymers.

ACKNOWLEDGEMENTS

| am indebted to Dr. Francoise Gaill for pro-
vision of specimens of Lepetodrilus, to Dr. K.
Oates for assistance with X-ray microprobe
analysis, and to Mr. Haydn Morris for carrying
out amino acid analyses.

TABLE 5. Elemental composition of eroded pits in the shell of Lepetodrilus determined by X-ray
microprobe analysis. A. Within a pit on the outside surface of the shell. B. Within a pitted area on the
inside surface where erosion has penetrated the shell from the outside. These analyses relate to Figures
18A and 18B.

Element Na Mg Al Si E S Cl Ca

Counts %total
A 0.36 3.58 — —
B 0.35 2.11 2.57 127

39.08
13.94

28.33 1.07
27.06 0.84

25:57
51.87

140 HUNT

LITERATURE CITED

ANDERSEN, $. O., 1979, Biochemistry of the in-
sect cuticle. Annual Review of Entomology, 24:
29-61.

ANDERSON, S. O. & T. WEIS-FOGH, 1964, Resi-
lin. A rubber-like protein in insect cuticle. Ad-
vances in Protein Chemistry, 2: 1-65.

BAROSS, J. A. & J. W. DEMING, 1985, The role of
bacteria in the ecology of black-smoker environ-
ments. Biological Society of Washington Bulletin,
6: 355-371.

BEVELANDER, G. & H. NAKAHARA, 1970, An
electron microscope study of the formation and
structure of the periostracum of a gastropod Lit-
torina littorea. Calcified Tissue Research, 1: 5-
12

CHOU, P. Y. & G. D. FASMAN, 1977, Secondary
structural prediction of proteins from their amino
acid sequence. Trends in Biochemical Science,
2: 128-131.

CRENSHAW, M. A. & J. M. NEFF, 1969, Decalci-
fication at the mantle-shell interface in molluscs.
American Zoologist, 9: 881-885.

CRENSHAW, M. A. & H. RISTEDT, 1976, His-
tochemical localization of reactive groups in sep-
tal nacre from Nautilus pompilius |. Pp. 355—
367, in N. WATABE & К. M. WiLBuR, eds., The
Mechanisms of mineralization in the inverte-
brates and plants, Univ. of South Carolina Press.

DE BURGH, М. E. & С. L. SINGLA, 1984. Bacterial
colonization and endocytosis on the gill of a new
limpet species from a hydrothermal vent. Marine
Biology, 84: 1-6.

DEGENS MENT AD AWSRENCERAS ЗАТ
PARKER, 1967, Paleobiochemistry of molluscan
shell proteins. Comparative Biochemistry and
Physiology, 20: 553-579.

FELBECK, H., M. A. POWELL, S. C. HAND &
G. N. SOMERO, 1985, Metabolic adaptations of
hydrothermal vent animals. Biological Society of
Washington Bulletin, 6: 261-272.

FRETTER, V., 1988, New archaeogastropod lim-
pets from hydrothermal vents; superfamily Lepe-
todrilacea Il. Anatomy. Philosophical Transac-
tions of the Royal Society (London), Series B,
319: 33-82.

HERRERA-DUVAULT, Y. & M. ROUX, 1986, Mo-
dalites et vitesse de dissolution de la coquille des
modioles du site hydrothermal de 13°N sur la dor-
sale du Pacifique oriental. Comptes Rendus de
l'Académie de Science de Paris, Série Ill, 302:
251-256.

HICKMAN, C. S., 1983, Radular patterns, system-
atics, diversity, and ecology of deep sea limpets.
Veliger, 26: 73-92.

HUDDART, H. & S. HUNT, 1975, Visceral muscle,
Blackie, Glasgow.

HUNT, S., 1987, Amino acid compositions of peri-
ostracal proteins from molluscs living in the vicin-
ity of deep sea hydrothermal vents: an unusual
methionine-rich structural protein. Comparative
Biochemistry and Physiology, 88B: 1013-1021.

HUNT, 5. 8 К. OATES, 1970, Fibrous protein ul-
trastructure of gastropod periostracum (Buc-
cinum undatum L.). Experientia, 26: 1196-1197.

HUNT, S. & K. OATES, 1978, Fine structure and
molecular organization of the periostracum in a
gastropod mollusc Buccinum undatum L. and its
relation to similar structural protein systems in
other invertebrates. Philosophical Transactions
of the Royal Society (London), Series B, 283:
414-459.

HUNT, S. 8 K. OATES, 1985, Helicoidal architec-
ture in the periostracum of a terrestrial proso-
branch Pterocyclus latilabrum Smith. Journal of
Molloscan Studies, 51: 336-344.

JUNIPER, 5. K., J. А. J. THOMPSON 4 S. E. CAL-
VERT, 1986, Accumulation of minerals and trace
elements in biogenic mucus at hydrothermal
vents Deep Sea Research, 33A: 339-347.

LUTZ, R.A., 1982, Shell microstructure and miner-
alogy of two species of bivalves from deep-sea
hydrothermal vents. American Malacological Bul-
letin, (1), Abstract, 101.

LUTZ, R. A. 8 D. C. RHOADS, 1980, Growth pat-
terns within the molluscan shell: an overview. Pp.
203-254, in D. С. RHOADS 4 В. A. LUTZ, eds.,
Skeletal growth of aquatic organisms, Plenum
Press.

MARSH, M. E. & R. L. SASS, 1983, Calcium-
binding phosphoprotein particles in the extrapal-
lial fluid and innermost shell lamella of clams.
Journal of Experimental Zoology, 226: 193-203.

MCLEAN, J. H., 1985, Preliminary report of the lim-
pets at hydrothermal vents. Biological Society of
Washington Bulletin, 6: 159-166.

MCLEAN, J. H., 1988, New archaeogatropod lim-
pets from hydrothermal vents; superfamily Lepe-
todrilacea |. Systematic descriptions. Philosoph-
ical Transactions of the Royal Society (London),
Series B, 319: 1-32.

MELLOR, J. W., 1960, A comprehensive treatise on
inorganic and theoretical chemistry, Volume 3:
824-827. Longmans, London.

MEENAKSHI, V. R., P. E. HARE, N. WATABE & K.
M. WILBUR, 1969, The chemical composition of
the periostracum of the molluscan shell. Compar-
ative Biochemistry and Physiology, 29: 611-620.

PETIT, H., W. L. DAVIS, В. G. JONES & Н. К.
HAGLER, 1980, Morphological studies on the
calcification process in the fresh-water mussel
Amblena. Tissue & Cell, 12: 13-28.

REEKE, G. N., J. W. BECKER 4 G. M. EDELMAN,
1975, The covalent and three-dimensional struc-
ture of concanavalin A. Journal of Biological
Chemistry, 250: 1525-1547.

ROUX, M., M. RIO & E. FATTON, 1985, Clam
growth and thermal spring activity recorded by
shells at 21°N. Biological Society of Washington
Bulletin, 6: 211-221.

WALLER, T. R., 1983, Dahllite in the periostracum
of Lithophaga nigra (Mollusca: Bivalvia) and its
taxonomic and functional implications. American
Malacological Bulletin, (1), Abstract, 101.

WANG, J. C., B. A. CUNNINGHAM & G. M. EDEL-

LEPETODRILUS SHELL STRUCTURE AND COMPOSITION 141

MAN, 1971, Unusual fragments in the subunit Shell formation. Pp. 235-287, in A. S. M.
structure of concanavalin A. Proceedings of the SALEUDDIN & К. M. WILBUR, eds., The Mol-
National Academy of Sciences of the USA, 68: lusca, 4(1). Academic Press.

1130-1134.

WILBUR, K. M. 8 A. S. M. SALEUDDIN, 1983, Revised Ms accepted 3 February 1992

ut

MALACOLOGIA, 1992, 34(1-2): 143-342

THE POMATIOPSIDAE OF HUNAN, CHINA
(GASTROPODA: RISSOACEA)

George M. Davis’, Cui-E Chen”, Chun Wu? ‚Tie-Fu Kuang*, Xin-Guo Xing?, Li Li®,
Wen-Jian Liu”, Yu-Lun Yan?

ABSTRACT

This is a monograph involving the systematics of the 17 species of pomatiopsid snails thus far
found in Hunan Province, the People's Republic of China. The Triculinae and Pseudobythinellini
(new tribe) of the Pomatiopsinae dominate the pomatiopsid fauna. Four new species are de-
scribed, three of Neotricula and one of Tricula. Detailed anatomical data are the basis for
describing each species. The anatomies of “Akiyoshia” and Lithoglyphopsis are presented for
the first time. Guoia is a new genus for some species previously considered to be Lithoglyphus
(a strictly European hydrobiid genus) or Lithoglyphopsis (restricted to China). The anatomical
data are the basis for phenetic and cladistic analyses to provide insight into those characters and
character-states that serve to distinguish taxa, and to assess relationships among genera of the
tribes Pachydrobiini (Halewisia, Neotricula, Pachydrobia, Jinhongia, Robertsiella, Guoia, Wu-
conchona, Gammatricula) and Triculini (Tricula, Delavaya, Fenouilia, Lithoglyphopsis, Lacunop-
sis). Hubendickia served as the outgroup genus of the remaining tribe of the Triculinae, the
Jullieniini. Biogeographic deployment of the genera of all three tribes of the Triculinae is mapped
on an area cladogram of relevant river systems. Genera of the Jullieniini dominate the lower
Mekong River. Tricula extends from Northern India down the Yangtze and upper Mekong rivers.
Tricula and Neotricula primarily flourish along the Yangtze River drainage. Four genera have
shells that are so similar that shell characters cannot be used to distinguish among the genera:
Tricula, Neotricula, Jinhongia and Gammatricula. Most of the species treated here are of medical
importance because they either transmit or are suspected of transmitting lung flukes of the
genus Paragonimus, or blood flukes of the genus Schistosoma.

Key words: Pomatiopsidae, biogeogrpahy, phenetics, cladistics, China, Hunan, Schistosoma,
Tricula, Neotricula, Paragoniumus, Yangtze River

INTRODUCTION

This work is one of a series of papers ded-
icated to establishing the detailed anatomy
and systematic relationships of genera of
southeast Asian and Chinese freshwater
prosobranch snails suspected of being mem-
bers of the Pomatiopsidae as defined in Davis
(1979, 1980). It is one more step in docu-
menting the extensive speciation that has oc-
curred within the tribes Triculini and Pachy-
drobiini of the Triculinae, which occur
primarily throughout southern China, in con-
trast to the extensive Jullieniini adaptive radi-
ation documented in the lower Mekong River

in Thailand, Laos, and Cambodia (Davis,
1979).

In this work, Guoia is established as a new
genus. The anatomies of “Akiyoshia” and
Lithoglyphopsis are presented for the first
time. Anatomical data are provided for clarifi-
cation of the genus Pseudobythinella. Several
nomenclatural problems are clarified or pre-
sented relative to necessary future work. Sev-
enteen nominal species are treated.

Phenetic and cladistic analyses are done to
provide insight into those characters and
character-states that serve to distinguish
taxa, and to assess relationships among gen-
era of the tribes Pachydrobiini and Triculini,

“The Academy of Natural Sciences of Philadelphia, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103-

1195 U.S.A.

?Institute of Parasitic Diseases, Zhejiang Academy of Medical Sciences, Hangzhou, Zhejiang Province, People's Republic

of China
Hunan Medical University, Changsha, Hunan Province

#Hengshan County Anti-Epidemic Station, Hengshan, Hunan, People's Republic of China
SAnhua County Anti-Epidemic Station, Zhuzhou, Hunan, People's Republic of China
SZhuzhou City Anti-Epidemic Station, Zhuzhou, Hunan, People's Republic of China

7Cili County Anti-Epidemic Station, Cili, Hunan, People's Republic of China

8Shimen County Anti-Epidemic Station, Shimen, Hunan, People's Republic of China

144 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

with one genus of the Jullieniini serving as the
outgroup. Biogeographic deployment of the
Triculinae is considered in conjunction with an
assessment of the evolution of river systems
in continental south Asia.

Most of the taxa treated here are of medical
importance because they either transmit or
are suspected of transmitting lung flukes of
the genus Paragonimus or blood flukes of the
genus Schistosoma.

MATERIAL AND METHODS

Localities are provided with the synonymy
section for each species. Sites 1 to 12 are
found on Figure 1. With reference to Figure 1,
be aware that maps published at different
times may provide different spellings for stan-
dard Chinese characters. Relevant here are:
Hsiang = Xiang; Tzu = Zi; Lin = Li. The
Yangtze River = the Chang Jiang (jiang =
river; shui = river). Dong Ting Lake (Lake =
Hu) = Tung t'ing Hu.

Abbreviations for institutions housing
voucher species are: ANSP = Academy of
Natural Sciences of Philadelphia (the A- num-
ber sequence indicates alcohol-preserved
specimens); IZAS = Institute of Zoology, Ac-
ademica Sinica; SMF = Senckenberg Mu-
seum, Frankfurt A.M. Germany; ZAMIP =
Zhejiang Academy of Medical Sciences, Insti-
tute of Parasitic Diseases, Hangzhou, China.
Voucher catalog numbers are provided in the
synonymy section for each species.

Methods are those of Davis 8 Carney
(1973), Davis et al. (1976), Davis (1979), and
Davis & Greer (1980). All dissections were
done with living material in Hunan, China us-
ing a Wild dissecting microscope and a Nikon
high intensity lamp. Scanning electron micro-
scopical analyses, photography of shells, rad-
ular preparations and computer analyses
were done in Philadelphia.

Descriptions of taxa employ only those
character and character-states useful to dis-
criminate among species of Pomatiopsidae.
Character-states common to all are not re-
peated here but are described elsewhere
(Davis, 1979, 1980).

Relative sizes of organs or shells are de-
fined by range in size (as in shells) or ratios to
facilitate comparison among taxa using terms
such as “small, “ “elongate, “ etc. This prac-
tice follows Davis et al. (1986) and all later
papers in this series of studies. The sizes and
ratio used are:

1. Shell length: large (= 5.0 mm), medium
(4.0—4.9 mm), small (2.1-3.9 mm), minute (=
2.0 mm) (based on mean size of the mature
size class with the greatest number of individ-
uals, realizing that mature snails may occur
with different numbers of whorls, e.g. 5.0, 5.5,
6.0 whorls).

2. Operculum attachment pad: width of at-
tachment pad = width of operculum: narrow
(= 0.35), wide (0.36-0.55), very wide
(= 0.56).

3. Osphradium length: length of osphra-
dium = length of gill (Davis et al. 1976, 1982,
1983): long (= 0.40), short (= 0.35).

4. Length of gill filament section Gf,: length
of Gf, + length of Gf, and Gf,: long (= 0.51),
medium (=normal) (0.31-0.50), short
(= 0.30).

5. Gill filament number: few (< 15), numer-
ous moderate (16-25), numerous (= 26).

6. Bursa copulatrix length: length of bursa
= length of the pallial oviduct: long (= 0.40),
short (= 0.38).

7. Albumen gland length: length of the al-
bumen gland (Ppo) = length of the entire pal-
lial oviduct: standard (= 0.45), short (< 0.42).

8. Male gonad length: gonad length
= length of digestive gland: short (< 0.50),
long (= 0.50).

9. Seminal receptacle duct length: very
short or absent (= 0.02 mm), long (> 0.02
mm).

10. Radula length: short (< 0.40 mm), me-
dium (0.41—0.59 mm), long (0.60 —0.79), very
long (= 0.80 mm).

11. Radula; mean rows of teeth: few (= 59),
moderate (60-69), many (70—84), very many
(= 85).

12. RPG ratio: length of the supraesoph-
ageal connective + sum of the length of the
right pleural ganglion, the supraesophageal
connective, and the supraesophageal gan-
glion; concentrated (= 0.29), moderately con-
centrated (0.30—0.49), elongate (0.50—0.67),
extremely elongate (= 0.68) (Davis et al.
1984, 1985).

13. Subesophageal connective length:
usual or standard triculine (0.02-0.09 mm);
none; long (= 0.10 mm) [based on average of
several measurements].

Multivariate Analysis

Shell data were analyzed for populations of
Guoia. The first data matrix involved 33 OTUs
(operational taxonomic units) (individual
shells) and 11 characters. The OTUs were:

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 145

am” 2: 113° =: À 0
HUBEI anate,
Rive,
Go (д
о ORS E
Ges anzhen (44) YW ee 6%
~ À \
о A AH
2 #4 Rive ; = ES
himen 4 e
(ЖП) >»
7 (21) OCili =
¿0 Shenglong BR) )
к (
ETS Xikou ) Y
iu O) 8 (40) AE) 4 J
Æ 0 Changde @ LOT
a
ping E
(NE +) €,
j ?
LAO Rive!
Rs) )
wan © Sa
(KM) <
AnhuaoQ 4,11
10 Jiangnan
JL
GM) Changsha eu
(к id)
HUNAN 3 per
= de
$ Ey y
2 =
LE Le
CH a” y
a Зе (
N x= 5 a
J 9 de x
A CES Baisha (A) 7 / g
j Sand Nany = O Hengshan (41 |) i <
- Der — Bjanme =
7 x oes m <=
27 AN ? Ox by A NET
Heng @ A po ?
2 è (At e) RS /
m
NM
DAS
= f
Я N
€ 2 anne.
Tr? j — $ Sanhe
\ AR 22 5 ЕЯ)
С j ER
== a & EY Г
я ; 4 ; > 2)
FANS y
2 A
x \
se G
$ Y, J PR
St A A Я = en
и г we Se
$ —
DER eh <
Ч A AZ % } à à
tes \ 4
@ Collection Sites wo < \ pe
; j
O County Seat ) TE ES
Gr r
Ç
© City ) GUANGDONG
| à
~— River ae
re)
Lake |
Y 24°
=)
€ a
11° 112° 113° 114

| FIG. 1. Map of Hunan, China, showing localities from which specimens described in this monograph were
collected.

146 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

Guoia viridulus, large class snails, males (1—
7), females (8-12); ANSP cataloged speci-
mens of Guoia viridulus collected over 90
years ago (18-21); small class snails, males
(22-26), females (27-32). Guoia fuchsianus
males (13, 14), females (15-17). ANSP cat-
aloged historic G. fuchsianus (33).

The characters were:

. Number of whorls

Length

. Width

. Length of body whorl

. Length of penultimate whorl
. Width of penultimate whorl
. Width of 3rd whorl

. Length of last three whorls
. Length of aperture

. Width of aperture

. Width of columellar shelf

The second matrix included individuals of
two populations from historic collections,
Guoia fuchsianus cataloged in the ANSP col-
lections: ANSP 98205 (34-38), ANSP 45961
(39-41). These OTUs were added to the
OTUs of matrix one to give a total OTU count
of 41. The character matrix was reduced be-
cause all OTUs 34-41 had eroded apices,
thus characters 1, 2, 7, and 8 could not be
measured for all individuals.

Computations were made using the 1974
version of NT-SYS (Rohlf et al., 1972). Char-
acters were standardized (standard deviation
and mean values). Similarity and distance co-
efficients were calculated and phenograms
produced using UPGMA. Cophenetic correla-
tions were calculated. Principal Component
Analysis (PCA) was done with components
extracted until eigenvalues became less than
1.0. A transposed matrix of the first three prin-
cipal components with their character loading
was postmultiplied by the standardized matrix
to yield a matrix of OTU projections in the
principal component space (Rohlf et al.
1972). The OTU locations in the three-dimen-
sional space were used as the initial configu-
ration for a nometric multidimensional scaling
(MDS) placement of Q-mode taxonomic dis-
tances between OTUs (Kruskal, 1964). Ordi-
nation was done followed by a Prim Network
(minimum Spanning Tree = MST). Subse-
quently a phenogram was produced based on
MDS ага the cophenetic correlation calcu-
lated.

A similar multivariate analysis was used to
assess phenetic relationships among 30

OTUs on the basis of shell data. The OTUs
involved species of Tricula, Neotricula or spe-
cies the shells of which resembled those of
species of Tricula. The analysis was done as
an aid for distinguishing among species on
the basis of shell characters.

Another similar multivariate analysis was
done to assess phenetic relationships among
the same species for which the shell analysis
was done. However, this analysis involved 48
anatomical character-states.

Cladistic Analysis

Scorings were not done in binary alone but
involved six multistate characters (Table 85).
Analysis was first done using the computer-
mediated program Hennig-86 (Farris, 1988).
Hubendickia served as the outgroup. Options
run were cc-. (character-states all unordered);
ie; tplot; tsave; nelsen. A final non-computer-
mediated cladogram was made after weight-
ing one character following establishing the
direction of evolution of its states.

Shell Characters

Shell character-states is illustrated in Figs.
2, 3. Aside from shell measurements there
are a number of diagnostic qualitative char-
acter-states of use to discriminate among
shells (Tables 1, 2).

Triculine shells have three predominant
shapes (Fig. 2A): globose, ovate-conic,
ovate-turreted. Many character-states involve
the aperture. The adapical end may be regu-
larly rounded or slightly constricted to form an
adapical apertural notch (Adn, Fig. 3D). The
constriction may be less pronounced (Fig.
3E). The adapical aperture and apertural
notch may be extended to form the adapical
apertural beak (Adb, Fig. 3A). The apertural
notch may have a noticeable internal groove
(Ngr, Fig. 3E) or lack a groove (Fig. 3A). The
adapical apertural notch may be bounded by
an adapical tooth or node on the inner lip (Adt,
Fig. 3A, D); or the adapical inner lip may lack
such a thickening (Fig. 3E, H).

The inner lip may (Ari, Fig. 3A) or may not
be fused (Gap, Figs. 3D, E) to the body whorl.
When separated from the body whorl the gap
(Gap) may be narrow (Fig. 3E) or wide (Fig.
3D). The inner lip may be straight, arched
(Fig. 3A), sigmoid or undulating (Fig. 3E), or
angled (Fig. 3D). The adapical part of the in-
ner lip may be thickened (Adil, Fig. 3D) or
thin; the abapical part of the inner lip the

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 147

TABLE 1. Twenty-six shell character and character-
states useful for describing shells of Triculinae as
illustrated in Figures 2 and 3.

Abil Abapical inner lip

Abs Abapical spout

Adb Adapical apertural beak
Adil Adapical part of inner lip
Adt Adapical tooth or node
Adn Adapical apertural notch
Agp Adapical apetural gap
Aol Adapical outer lip angle
Aols Adapical outer lip sinus
Ari Arched inner lip

Cr Crenulated suture

Ct Columellar tooth

Gap Gap between the body whorl and the
inner lip

L Length

Min Mid-lip notch
Nab Normal adapical aperture
Ngr Adapical notch groove

OI Outer lip
Sfo Scooped forward
Sin Sinuate outer lip

Stl Straight lip
Umc Umbilical chink

Var Varix

W Width

x distance from base of body whorl to
abapical end of aperture

y distance from edge of outer lip to edge of
body whorl

same. The inner lip may have a mid-lip notch
(Min, Fig. 3D).

The abapical end of the aperture may
project noticeably beyond the base of the
body whorl when the shell is viewed in aper-
tural view (Fig. 3E); it may not (Fig. 3F).
The abapical end of the aperture may be
spout-like (Figs. 3D, E); it may not (Fig. 3A,
F). The columella may have a tooth (Ct, Fig.
3F). It is necessary to break open the body
whorl of some shells from a population to de-
termine whether or not there is an internal
columellar node, tooth, or keel.

The adapical end of the aperture may be
pulled away from the body whorl leaving an
adapical gap (Agp, Fig. 3D, F) orthe adapical
aperture may be fused to the body whorl (Fig.
3A).

In side view the outer lip may be straight
(Stl, Fig. 3H) or sinuate (Sin, 3B). The outer
lip may be aligned with the axis of coiling (Fig.
3C, H) or scooped forward (Sfo, Fig. 3B). In
side view there may be an adapical depres-
sion or sinus in the outer lip (Aols, Fig. 3H)
caused by an adapical apertural notch.

Rotating the shell from apertural view to the
right (apex up) so that the inner lip is perpen-
dicular to the horizontal, the abapical part of
the inner lip may be deflected away from the
columella causing a lip deflection angle (Fig.
3G, 2B). Such an angle may not occur.

Gill Filament Shape

Considering the longest (thus mid-gill) gill
filaments, three shapes are found when the
filaments are examined so as to see the entire
leaflet (Fig. 4). The Gf, section may be flat,
have a modest dome, or be high domed.

Vital Staining

Davis (1967) discussed using aqueous
methylene blue and neutral red as vital stains.
The exact methods are as follows: small
quantities of powdered stain are dumped into
a quantity of tap water (approximately 500 ml)
and stirred until dissolved. Sufficient powder
is added so that when the solution is placed
before a light the solution is opaque. Wide-
mouth jars are used, one each for the two
stains. Wide mouth containers are specified
because the stain is used over and over
again. With the living animal pinned out on the
black parafin-wax layer in a 9-cm Petri dish,
the water covering the animal is poured off
and the stain poured on to cover the animal.
The stain is left on for 30 seconds to a minute,
then poured back into the storage bottle. The
animal is rinsed five to ten times to remove
excess stain and then covered with water to
continue the dissection.

Neutral red is used first to deliniate glandu-
lar tissue and the ducts of the female repro-
ductive system. The two regions of the pallial
oviduct are differentiated; the oviduct is
stained as well as the gonad and oocytes.

Methylene blue is used next; it seems to
harden and delineate the ducts of the female
reproductive system. The slender duct of the
seminal receptacle and the various other
components of the bursa copulatrix complex
of organs are made to stand out.

Bouins Solution

When all work with the living animal is fin-
ished, Bouins solution is most useful to elim-
inate problems caused by mucus, to delineate
mantle cavity structures, such as the gill fila-
ments, and to dissect the nervous system.
The nerves and ganglia stain bright yellow en-
abling one to see them better and to differen-

148

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 2. Shell character-states scored for 29 characters used for a multivariate analysis of phenetic
relationships among 21 species of Triculinae, the shells of which either are classified as Tricula or
Neotricula or resemble the shells of these two genera.

. Size:
. Shape:

. Aperture shape:

. Umbilicus:

. Whorl at suture:

. Teleoconch sculpture:

. Spiral sculpture at mid body-

whorl to shell base

. Protoconch whorls:
. Adapical aperture:
. Adapical aperture:

. Adapical aperture:
. Abapical aperture:
. Inner lip:

. Outer lip-side view:
. Outer lip-side view:

. Inner lip:
. Inner lip:
. Base of inner lip, side view:
. Inner lip:
. Inner lip:

. Adapical aperture:
. Columella within body whorl:

. Varix:

. Adapical aperture:

. Adapical outer lip angle:

. Base of body whorl:

. Base of shell at umbilicus:

. Base of shell at abapical lip:
. Shell attains 7.0 whorls:

small (0), medium (1), minute (2), long (3)

ovate-conic (0), ovate-turreted (1), turreted (2), cylindrical-conic
(3), globose-conic (4)

ovate (0), pyriform (1), distorted (2)

none (0), chink (1), clearly open (2)

smooth (0), crenulated (1)

none (0), spiral microsculpture at suture (1)

none (0), with (1)

smooth (0), wrinkled (1), malleated and/or pitted (2)

normal (0), with notch (1), with beak (2)

normal (0), with internal groove (1)

normal (0), with sinus (1)

normal (0), with spout (1)

straight (0), arched (1), sinuate (2), angled (3)

straight (0), sinuate (1)

parallel with axis of coiling (0), scooped forward (1), slanted back
(2)

no tooth (0), with tooth (1)

no notch (0), with notch (1)

straight (0), angled to form lip deflection angle (1)

thin (0), thick [= 0.10mm] (1), differentially thickened (2)

fused to body whorl (0), partly separated (1), totally separated by

narrow gap (2), widely separated (3)

fused to body whorl (0), slightly separated (1), widely separated
(2)

smooth (0), tooth or node (1), spiral keel (2), lamellae on aperture
side of columella (3)

none (0), slight (1), pronounced (2)

normal (0), with beak tubercle (1)

none (0), slight (1), extended (2)

normal (0), with keel spiraling down from umbilical area (1)

normal (0), with wide columellar shelf (1)

normal (0), with basal post (1)

no (0), yes (1)

tiate nerves from muscle fibers. Bouins solu-
tion strengthens the nerves such that they
can be handled better. Dissections of the
head involving the buccal mass, salivary
glands and associated nerves are facilitated.

Body Measurements

How the lengths of the gonad, digestive
gland, and body are measured is illustrated in
Davis & Carney (1973). Lengths of mantle
cavity organs are measured with the mantle
cut along the right side of the neck from the
mantle edge to the rear of the mantle cavity.
The mantle is then reflected to the left and
pinned out as shown in Figure 5. How lengths
of mantle cavity structures are measured is
illustrated in Figure 5. The ocular micrometer
is positioned and rotated along the dotted tra-
jectory. As shown in Figure 5, not all gill fila-
ments may be illustrated.

Abbreviations

a line perpendicular to mid-line “x” at
posterior edge of eye lobes

a’ Line perpendicular to mid-line “x” at
the anterior edge of penial base

Adb Adapical apertural beak

Ae Abapical embayment

Af Anterior foot;

Algo Anterior lobes of gonad

Als Adapical outer lip sinus

Apg Anterior pedal glands.

Apo Anterior pallial oviduct = capsule
gland

Apo-1 Anterior pallial oviduct = capsule
gland

Apo-2 Different tissue type anterior to cap-
sule gland.

Ast Anterior chamber of stomach

Au Auricle

Bc Basal crescent

Beak groove

LofA

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 149

Buccal mass
Base of penis

Bursa copulatrix

Cerebral commissure

Columellar muscle

Oviduct coil

Crescent ridge

Columellar shelf

Common sperm duct

Duct of bursa

Digestive gland

Duct of seminal receptacle
Eyebrow

Efferent branchial vein

Anterior edge of digestive gland
(Figs. 141, 142)

Dashed line indicates anterior limit
of digestive gland

Ejaculatory duct

Posterior end of mantle cavity
Dashed line indicates posterior limit
of prostate gland

Esophagus

Eye

Eyebrow

Fecal pellet

Thickened basal bar of gill filament
Slender terminal part of gill filament
Glandular lobe

Gonad

Section of gonad, the extent of which
is indicated by the dashed line, re-
moved to show seminal vesicle
Gonad

Patch of white granules

Granules or glands, white to lemon
yellow

Groove

Grey streak on Ast

Inner lip

Intestine

Intestine looping around anterior
end of style sac

Fecal pellet compressor section of
intestine

Anterior intestine

Kidney

Length

Length of aperture

Length of body whorl

Length of gill filament section Gf,
Length of gill filament section Gf,
Length of gill

Length (= width) of mantle collar
anterior to gill

Dashed line is trajectory used to
measure length of gill and length of
mantle cavity
Length of

shell with aperture

pressed down on a horizontal sur-
face

Length of osphradium

Left pleural ganglion

Length of penultimate whorl

Mantle collar

Neck

Adapical notch groove

Opening from stomach to digestive
gland

Opening of kidney into mantle cavity
Outer lip

Opening of spermathecal duct into
mantle cavity

Osphradiomantle nerve

Oocyte(s)

Opening into albumen gland (Ppo)
Opening of oviduct to albumen gland
(enlarged in Fig. 10)

Opening for sperm entry to pericar-
dium

Opening of oviduct into albumen
gland

Opening to vas deferens
Osphradium

Opening of spermathecal duct to
mantle cavity (Figs. 11, 21)

Point where oviduct fuses to, and
opens into pericardium
Osphradiomantle nerve

Oviducal seminal receptacle
Oviduct

Papilla

Pericardial bursa

Pellet compressor (= In.)
Pericardium

Penis

Penial filament

Posterior pallial oviduct = albumen
gland

Prostate

Pleurosupraesophogeal connective
Posterior chamber of stomach
Right cerebral ganglion

Yellow ridge

Right pleural ganglion

Radular sac

Spermathecal duct

Opening of oviduct into the pericar-
dium

Sperm duct

Pleurosupraesophageal connective
Supraesophageal ganglion

Salivary gland

Snout

Seminal receptacle

Style sac

Stylet

Pleurosubesophageal connective

150 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

Sug Subesophageal ganglion
Sv Seminal vesicle

Tn Tentacle

Twv Thin walled vestibule

Uc Umbilical chink

Vd Vas deferens

Vd, Posterior vas deferens
Vd. Anterior vas deferens
Ve Vas efferens

Vei Vein

Ven Ventricle

W Width (Fig. 110)

Wap Width of aperature
War White granular inclusions
Wg White granules

WofA Width of shell with aperature pressed
down on a horizontal surface

W Wall of neck

x Mid-line of snout-neck with snout an-
terior (up).

x Marks the same point in Figs. 20, 21
A and B; 117A, 118

Yri Yellow ridge along anterior and pos-

terior ends of the Ast

SYSTEMATICS

Taxa Treated

POMATIOPSIDAE

Pomatiopsinae
Pomatiopsini
Oncomelania
Oncomelania hupensis Gredler, 1881
Pseudobythinellini, Davis & Chen new
tribe
Akioyshia
“Akiyoshia” chinensis Liu, Zhang &
Chen, 1982
Pseudobythinella
Pseudobythinella chinensis (Liu &
Zhang, 1979)
P. shimenensis Liu, Zhang & Chen,
1982
Triculinae
Pachydrobiini
Guoia Davis & Chen gen. nov.
G. fuchsianus (Moellendorff, 1885)
G. viridulus (Moellendorff, 1888)
Neotricula
N. cristella (Gredler, 1887)
N. dianmenensis Davis & Chen, sp.
nov.
N. duplicata Davis & Chen, sp. nov.

М. ШИ Chen & Davis, sp. nov.
N. minutoides (Gredler, 1885)

Triculini
Lithoglyphopsis
L. modesta (Gredler, 1886)

Tricula
T. gredleri Kang, 1986
T. maxidens Chen & Davis,
SP. nov.
T. odonta Liu, Zhang & Wang 1983a

Nomina nuda:
Tricula gredleri
Akiyoshia odonta

Nomina Nuda

Two names were introduced into the litera-
ture for presumed Pomatiopsidae from Hunan
Province but without photographs, illustra-
tions, designation of types, or descriptions
that would permit one to identify the species
in question. Accordingly these are nomina
nuda. The nominal taxa involved are:

Tricula gredleri Feng et al., 1985 [Kang sp. nov.
Tricula gredleri Feng et al., 1986 [Kang sp. nov.
Akiyoshia (Saganoa) odonta Feng et al., 1985
[Kang sp. nov.]

Akiyoshia (Saganoa) odonta Feng et al., 1986
[Kang sp. nov.]

Higher Taxa Defined

The family Pomatiopsidae was differenti-
ated from the Hydrobiidae in Davis (1979);
two subfamilies were recognized, the Poma-
tiopsinae and Triculinae. Both subfamilies
have a spermathecal duct, a pomatiopsid
type central tooth (square to rectangular) with
pronounced basal cusps arising from the face
of the tooth (an exception includes the
Pseudobythinellini discussed below), cover
eggs (laid singly) with sand grains, and do not
brood young. The penis is without complex
glands or lobes. The differentiation of the po-
matiopsid subfamilies from other higher taxa
with spermathecal ducts was given in Davis et
al.(1985: 74-75). These higher taxa are the
Hydrobiidae: Littoridininae and Amnicolinae
(also see Hershler & Thompson, 1988).

The Pomatiopsinae are Gondwanian in dis-
tribution. Pomatiopsinae snails have an elon-

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 151

BE

GLOBOSE OVATE-CONIC OVATE-TURRETED

/

de
7
Lu
/ \
a \

FIG. 2. Shell characters used in describing species of Hunan Triculinae. A. Three major shell shapes. B.
Inner lip in side view showing a pronounced lip deflection angle.

152 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

р

Stl

Aols

Var

FIG. 3. Shell characters used in describing species of Hunan Triculinae. Labels are defined in Table 1.

gated spermathecal duct extending to the an-
terior end of the mantle cavity. In the tribe
Pomatiopsini, the eyes are in pronounced
bulges. These snails have a pedal crease, su-
prapedal fold and omniphoric groove. They
move by a step-wise mode. They have
evolved from freshwater to an amphibious
mode of existence, and in Japan Blanfordia
has become terrestrial (Davis, 1979, 1981).
The tribe Pseudobythinellini is placed in the
Pomatiopsinae because of the elongated

spermathecal duct. The central tooth is of the
Hydrobia type with a pair of cusps arising
from the lateral angles. The eyes are not in
pronounced eye lobes. There is no su-
prapedal fold, omniphoric groove or pedal
crease. Animals move by ciliary glide. The pe-
nis of species classified here as Pseudo-
bythinellini lacks a penial lobe with functioning
duct (in addition to the vas deferens) that wid-
ens to form a wide coiled mass or penial
gland in the cephalic haemocoel. The acces-

THE POMATIOPSIDAE OF HUNAN

FLAT

MODEST DOME HIGH DOMED

FIG. 4. Shapes of the Gf, segment of the largest gill
filaments.

FIG. 5. Mantle reflected to reveal mantle cavity or-
gans and how their lengths were measured. The
dashed line is the trajectory for measuring the
length of the gill. The length of the mantle cavity is
measured from Emc along the LGi plus LMa.

sory penial lobe and penial gland are charac-
teristic of the European genus Bythinella and
North American genus Amnicola of the Hy-
drobiidae: Amnicolinae (Davis et al., 1985;
Hershler & Thompson, (1988).

The Triculinae are Asian with a distribution
from northern India throughout South China
and southeast Asia. No Triculinae are found
in Japan (Shikoku to Okinawa). loganzen &
Starobogatov (1982) refer their new genus
Sibirobythinella to the Triculidae. However,
given the extensive convergences in structure
such as the spermathecal duct and central
tooth of the radula (reviewed in Davis et. al.,
1985) it is not at all certain that taxa from
Siberia or northeastern Russia relegated to

, CHINA (GASTROPODA: RISSOACEA) 153

the Triculidae by Russian workers are Tricu-
linae sensu Davis (1979, 1980). Sibiro-
bythinella is not, based on data provided, a
triculine snail. The shell and radula might in-
dicate placement in the Pseudobythinellini.
loganzen 8 Starobogatov (1982) place their
taxon in the family Triculidae apart from taxa
they erected to family status such as Littori-
dinidae, Pomatiopsidae and Stenothyridae,
all characterized by having a spermathecal
duct running independently of the pallial ovi-
duct from the mantle cavity to the bursa cop-
ulatrix. Unfortunately, they do not provide suf-
ficient data to differentiate their taxon from
European Hydrobiidae: Littoridininae, Amni-
colinae, or Pomatiopsinae: Pseudobythinel-
lini.

Triculinae snails have a short spermathecal
duct opening to the posterior end of the man-
tle cavity. The snails are aquatic, do not have
a suprapedal fold, pedal crease, or om-
niphoric groove. Only in a few derived genera
are the eyes in pronounced lobes. Snails
progress by ciliary glide.

Tribal, Generic and Species Descriptions

POMATIOPSINAE

Pomatiopsini
Type genus. Pomatiopsis Tryon 1862

Diagnosis. Ovate-conic to turreted shells.
Shell length greater than 2.5 mm; most
greater than 3.0 mm. Apical whorls are not
flattened. Spermathecal duct runs from the
bursa copulatrix to the anterior end of the
mantle cavity. Spermathecal duct not tightly
pressed to pallial oviduct. Sperm duct running
posteriorly from oviduct to the bursa or the
spermathecal duct at the bursa. Eyes are in
pronounced eye lobes. There is a suprapedal
fold, an omniphoric groove, and pedal crease.
Basal cusps of the central tooth of the radula
arise from the face of the tooth.

Oncomelania Gredler 1881

Type Species. Oncomelania hupensis Gred-
ler 1881: 120-121; pl. 6, fig. 5

Type locality. U—tschang—fu, March 1879.
[= Hubei Province; Wu—Tshan—fu; Yen,
1939].

154 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

Designation. By monotypy

Types. Bozen; lectotype and two paralecto-

types
Bozen No. 89. No. types at SMF (Zilch,
1974: 197)

Oncomelania hupensis Gredler, 1881

No detailed study of this well-known spe-
cies is presented here. This species is in-
cluded for completeness sake in monograph-
ing the Pomatiopsidae of Hunan, China. This
species is the vector for Schistosoma japoni-
cum and abounds in the marshes around
Dong Ting Lake and along the Yangtzee
River (Fig. 1) (Lou et al., 1982; Liu et al.,
1981). Detailed anatomical data for Oncomel-
ania have been published elsewhere (Davis
1967, 1968a, 1969a; Davis 8 Carney 1973).

Pseudobythinellini Davis & Chen
New Tribe

Type genus. Pseudobythinella Liu & Zhang
1979.

Diagnosis. Ovate shells less than 2.5 mm
long, with flattened apical whorls. The sperm
duct runs anteriorly from the oviduct to the
spermathecal duct; it is so tightly pressed to
the pallial oviduct that it is difficult to differen-
tiate it. Unlike the Pomatiopsini, the sperm
duct enters the spermathecal duct far anterior
to the bursa copulatrix; the duct of the bursa is
thus elongated (the continuation of the sper-
mathecal duct to the bursa from the point of
entry of the sperm duct). Eyes are not in pro-
nounced eye lobes. There is no suprapedal
fold, no omniphoric groove or pedal crease.
Animals are aquatic and move by ciliary glide.
Basal cusps of the central tooth of the radula
arise from the lateral angles of the tooth.

Synonymy. Erhaiini Davis et al. 1985: 69.

Discussion. The genera included in this tribe
are Akiyoshia and Pseudobythinella from
China. Liu & Zhang, 1979, described certain
species from China and classified them as
Bythinella and Pseudobythinella. They pro-
vided no anatomical data aside from penis
and radular illustrations. Pseudobythinella Liu
& Zhang, 1979, was described as different
from so-called Bythinella of China by (1) hav-
ing a tooth or node on the inner lip, and (2) the
central tooth having two pairs of basal cusps,
those on each side on the same level (not
Hydrobia-like).

The so-called Bythinella of China lack the
complex male reproductive system charac-
ters that serve to define the Hydrobiidae: Am-
nicolinae of Europe and North America in-
cluding genuine Bythinella (Europe) and
Amnicola (North America). Because Bythi-
nella does not occur in China, Davis & Kuo,’
(in Davis et al., 1985) described the genus
Erhaia to accommodate new species that Liu
and Zhang would have considered to be
Bythinella. Davis & Kuo (1985) raised the
tribe Erhaiini to accommodate Erhaia.

As a result of this study, Erhaia is placed in
the synonymy of Pseudobythinella. It is clear
that at least one species described as
Bythinella from China (B. chinensis Liu &
Zhang, 1979) has a tooth on the columella; it
is observed when the aperture is tilted. Upon
breaking open the shell it is seen that this
“tooth” is the terminus of a thick, glassy, spi-
ral columellar shelf or ledge. In summary,
there appears to be no basis for placing those
taxa with an overall similar anatomical ground
plan in separate genera on the basis of pres-
ence or absence of a tooth on the columella,
or where the tooth is prominently seen in the
aperture contrasted with the presence of a
columellar thickening seen only upon break-
ing the shell. The type species of Erhaia, E.
daliensis Davis & Kuo (in Davis et al., 1985)
has no columellar “tooth” in evidence in the
aperture even when the aperture is rotated to
examine as deeply as possible into the shell.
However, upon breaking open the body
whorl, a pronounced thickened columellar
ridge is seen. This glassy ridge is also found
in Erhaia kunmingensis Davis & Kuo 1985,
but it is not very pronounced and forms no
node.

The same situation involves the basal
cusps of the central tooth. The two basal
cusps on a side being on the same level or
one above the other is irrelevant to generic
definition. Most species of Pseudobythinella
(including the Chinese taxa described as
Bythinella) have only one pair of basal cusps
on the lateral angles.

The Pseudobythinellini are not classified as
Hydrobiidae but are in the Pomatiopsidae for
the following reasons: (1) There is evidence
that the position of the basal cusps of the cen-

Due to a change in China affecting the manner of spelling
Chinese words in English, Dr. Y. H. Kuo in the literature
until 1985 changed the spelling of his name to Y. H. Guo
(see Davis et al., 1985, 1986).

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 155

tral tooth derived from the pomatiopsine con-
dition, ¡.e. from the face of the tooth. In
Pseudobythinella kunmingensis from Yun-
nan, China most central teeth of most snails
had a single pair of basal cusps. However,
some central teeth had two pairs (Davis et al.,
1985: fig. 16F), the innermost arising from the
face of the tooth and lower than the outermost
pair arising from the laternal angle. This has
not, to our knowledge, been seen in any taxon
of Hydrobiidae that has a spermathecal duct.
(2) The position of the connections of the
spermathecal duct, duct of the bursa, and ovi-
duct are unlike any seen in the Hydrobiidae.
These Chinese species lack the complex pe-
nial lobes, three to four differentiated regions
of the pallial oviduct, and tendency to brood-
ing seen in the Hydrobiidae: Littoridininae. (3)
There is the biogeographic overlap of these
snails with at least one genus of the Pomati-
opsini, more if Akiyoshia, originally described
from Japan, is also a member of the Pseudo-
bythinella clade.

Akiyoshia Kuroda 8 Habe 1954

Type Species. Akiyoshia uenoi Kuroda &
Habe, 1954: 71-73, figs. 1-4

Type locality. Akiyoshi limestone cave, Ya-
maguchi Prefecture, Japan

Designation. By monotypy

Saganoa Kuroda, Habe 8 Tamu, 1957

Type species. Akiyoshia (Saganoa) kishiiana
Kuroda, Habe 8 Tamu in Kuroda € Habe,
1957: 186-187, figs. 5-8.

Type locality. Saga (= Arashiyama), western
foothills of Kyoto City, Japan

Designation. By original designation

Discussion. Species from China were de-
scribed as Akiyoshia (Saganoa) by Liu et al.
(1982). They were thus described because
the shells of those species are minute, tur-
reted, with roundish apertures, i.e. generally
corresponded to the conchological descrip-
tions given by Kuroda & Habe (1957). We
doubt that the Chinese species are Akiyoshia,
5.5., or Akiyoshia (Saganoa) due to the dis-
tance and differences in geological events
creating environments now inhabited by the
Chinese and Japanese snails. We are re-
minded that species once described as Tric-
ula from India, China, and the Ryukyu Archi-
pelago are now classified in four genera on

the basis of detailed anatomical data. We
note that Saganoa species are blind cave- or
well-dwelling snails; those collected from
China are stream-dwelling and not blind.
However, we will not describe a new genus
until the relevant Japanese snails are studied
and the data dictate such a course.

Akiyoshia (Saganoa) chinensis
Liu et al., 1982

Holotype. IZAS, HN 798002; Liu, Zhang &
Chen, 1982: 367, figs. 2-5 (in Liu et at. 1982).

Type Locality. Guzhang, Hunan Province,
Sept. 1979

Assigned Species. China only: Akiyoshia (Sa-
ganoa) chinensis Liu, Zhang & Chen, 1982;
A. (S.) yunnanensis Liu, Wang 8 Zhang,
1982. N=2 (in Liu et at. 1982).

Habitat

Refer to Tricula gredleri, D87-3. This spe-
cies was sympatric with T. gredleri and T.
maxidens. Snails were collected from a vil-
lage close to the type locality, approximately 5
km; the type locality is across the valley and
on the opposite hillside from D87-3.

Depository

Specimens are catalogued into the collec-
tions of ZAMIP, M0010; ANSP 373142,
A12658.

Description

Shell. Shells are minute, smooth, turreted
with deep sutures and slightly convex whorls
(Figs. GA—E, 7A—E, 8). Shells of mature ani-
mals have 4.5-5.5 whorls; the majority are
5.0-5.5 whorls. Shell lengths of mature snails
of 5.0 or more whorls range from 1.60-1.86
mm (Table 3). There is a slight umbilical
chink. The aperture is sub-circular; there is a
trace of a varix on some shells. In side view
the outer lip is slightly sinuate (Fig. 7B). With
the outer lip down (90° to the horizontal), the
outer lip is either straight or angled. There are
no apertural teeth or adapical notches.
Under SEM the following are observed: (1)
Spiral microsculpture can be seen at the base
of the shell (Fig. 7C), and on some whorls
(Fig. 7D). (2) The protoconch is smooth (Fig.
7D, E). (3) Breaking open the body whorl re-
veals no internal columellar teeth or spiral
ledges (Fig. 8A—E). The columella of the pen-

156

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 3. Shell measurements (mm) of topotypical Akioyshia chinensis. Mean + standard deviation
(range). N = number measured.

Large class
М =3 N=5
No. Whorls 5.05.25 5.5

Length (L) 1.77+0.02 1.84+0.01
(1.76—1.80) (1.82—1.86)

Width (W) 0.73+0.01 0.70+0.02
(0.72—0.74) (0.68—0.74)

L last three 1.53+0.02 1.51+0.03
whorls (1.52—1.56) (1.48—1.56)

L body whorl 0.91+0.03 0.88+0.03
(0.88 —0.94) (0.84—0.92)

L penultimate 0.36 0.37+0.01
whorl No. Var. (0.36 —0.38)
W penultimate 0.59+0.01 0.58+0.01
whorl (0.58—0.60) (0.58—0.60)
W 3rd whorl 0.48+0.02 0.49+0.04
(0.46—0.50) (0.44—0.54)

L aperture 0.61 +0.02 0.59+0.02
(0.60 —-0.64) (0.56—0.60)

W aperture 0.49+0.01 0.46+0.04
(0.48—0.50) (0.40—0.48)

x 0.20 0.15+0.04
No Var. (0.10—0.22)

y 0.09+0.03 0.10+0.04
(0.06-0.12) (0.04—0.14)

Small class
N =1 N=5 NA
4.5 5.0 5.5
1.64 1.66+0.04 1.68
(1.60—1.68)
0.70 0.66+0.02 0.66
(0.64—0.68)
1.44 1.44+0.02 1.40
(1.40- 1.46)
0.86 0.85+0.02 0.80
(0.84—0.88)
0.34 0.34+0.02 0.32
(0.32—0.36)
0.58 0.56+0.01 0.54
(0.54—0.56)
0.48 0.46 0.46
No Var.
0.60 0.55+0.02 0.54
(0.52—0.58)
0.46 0.31+0.01 0.44
(0.44—0.48)
0.16 0.17+0.08 0.18
(0.16—0.24)
0.12 0.07+0.04 0.12
(0.06—0.08)

ultimate whorl has minute calcareous nodes
(Fig. 8F). The inner edge of the inner lip has
raised calcareous pitted nodes (Figs. 8G, H);
the inner surface of the aperture is regularly
pitted (Fig. 8G, |).

External features. The head is white; there
are no white granules about the eye. There is
no Oncomelania-type “eyebrow.” The oper-
culum is corneous and paucispiral (Fig. 7F.).
The inner-lip edge is straight, i.e. the opercu-
lum shape is sub rectangular, not ovate. The
inner muscle attachment callus is compara-
tively small and weakly developed, about
27% the diameter of the operculum.

Mantle cavity. Measurements and counts of
mantle cavity organs are given in Table 4; see
Figure 9. The gill is central in the mantle cav-
ity and symmetrical about the osphradium.
The osphradium is short in females, long in
males relative to the length of gill. Being long
in this case is an artifact of the much reduced
gill. If the gill was the usual length, i.e. filling
most the length of the mantle cavity, the os-
phradium would (and should) be classified as
short.

There are only 11-14 gill filaments. Gf, is
long in females, medium (or normal) in males.
The total lengths of the longest gill filaments

are 0.22 + 0.03 mm. There is no patch of
white granules anterior to the osphradium
near to or partially on the mantle collar.

Female reproductive system. The body of an
uncoiled female without head or kidney tissue
is shown in Figure 10A. Measurements of rel-
evant organs are given in Table 4. Important
features are: (1) The gonad (Go) is a simple
sac located posterior to the stomach. (2) The
bursa copulatrix (Bu) is clearly seen posterior
to the pallial oviduct (Ppo). It is small. (3)
There are three histologically different sec-
tions to the pallial oviduct; the albumen gland
(Ppo), capsule gland (Apo-1), and a dense
white anterior region (Apo-2). These areas
are clearly discernable in gross dissection of
living animals. (4) The bursa complex of or-
gans is shown in Figure 11 in the same rela-
tive position as in Figure 10. The usual poma-
tiopsid seminal receptacle is lost; the function
of the seminal receptacle is relocated within a
U-shaped bend or twist of the oviduct (Osr,
Figs. 10, 11). (5) There is a pronounced, elon-
gated sperm duct (Sdu) connecting the ovi-
duct (Ov) to the spermathecal duct. The ovi-
duct enters the albumen gland (Ppo) just
posterior to the posterior end of the mantle
cavity (Emc). (6) The spermathecal duct (Sd)

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 157

FIG. 6. Shells of Akiyoshia chinensis (A-E); Pseudobythinella shimenensis (F-I), and P. chinensis (J-N).
The length of shell A is 1.77 mm; others are printed to same scale.

158 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 7. SEM pictures of shell (A-E) and operculum (F) of Akiyoshia chinensis. C. Enlargement of base of
shell showing rough growth lines crossed by fine spiral microsculpture. D, E. The apical whorls are smooth;
teleoconch roughened sculpture starts at 1.5 whorls. Spiral microsculpture is seen at 1.5 to 1.75 whorls. F.
Inner surface of operculum to left showing modestly developed attachment pad for muscles. Outer surface

is to right showing paucispiral coil.

runs to the anterior end of the mantle cavity to
open independently of the pallial oviduct.

Male reproductive system. An uncoiled male
is shown in Figure 12 without head and with
kidney tissue removed. Measurements of rel-

evant organs are given in Table 4. Important
features are: (1) The anterior lobes of the go-
nad (Go) are ventral to the posterior chamber
of the stomach. (2) There are numerous lobes
of the gonad that drain into a vas efferens
(Ve). (3) The vas deferens leaves the vas ef-
ferens at mid-gonad to slightly posterior to

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)
d ares я C ys

B = 5 Er

159

FIG. 8. SEM analysis of columella inside body whorl of Akiyoshia chinensis and sculptural attributes of
columella and inner lip. See text for details.

mid-gonad to begin coiling as the seminal
vesicle (Sv) dorsal to the gonad. (4) The pros-
tate (Pr) overlies the posterior end of the man-
tle cavity. (5) The prostate consists of dis-
cernable lobes. (6) The anterior vas deferens
(Vd,) leaves the prostate close to mid-pros-

tate. (7) The penis (Fig. 13 C) is simple with a
long penial filament (Pf) and small papilla
(Pa). A penial filament is defined as a terminal
length of penis that is distinctly narrowed
compared with the rest of the penis. This nar-
rowing is not simply due to the gradual taper-

160 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 9. Mantle cavity structures of Akiyoshia chin-
ensis

ing of the penis. No ejaculatory duct was
seen. An ejaculatory duct is a distinctly swol-
len muscle wrapped section of the vas defe-

rens found in the base of the penis or neck.
(8) The base of the penis (Bp) is to the right at
the snout-neck mid-line (x, Fig. 13A, B) and
oriented 65°-90° to the mid-line.

Digestive System. The digestive gland (Di)
covers the posterior chamber of the stomach
(Fig. 12). The stomach is shown in Figure 14
in the same relative position as in Figure 12.
There is no caecal appendix; the style sac
(including the intestinal loop) is 48% the
length of the stomach.

Radular statistics are given in Tables 5 and
6. Refer to Figures 15, 16. The cusp formula
most frequently encountered is

4(5)-1-(5)4; 4-1-4; 19-24; 16-21.
1(2)-1(2)

There are numerous cusps on the marginals
but there are significantly more on the inner

FIG. 10. Uncoiled female of Akiyoshia chinensis with head and kidney tissue removed (A) and gonad
showing oocytes (B).

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 161

TABLE 4. Lengths (mm) or counts of non-neural organs and structures of Akiyoshia chinensis. N =
number of snails used. Mean + standard deviation (range).

Body
Gonad
Digestive gland
Posterior pallial oviduct
(= albumen gland)
Anterior pallial oviduct
(= capsule gland)
Total pallial oviduct
= OV
Bursa copulatrix
BU
Duct of BU
BU - OV
Seminal receptacle
Duct of seminal receptacle
Mantle cavity
Gill (G)
Osphradium (OS)
95 = G
No. of filaments
Gf,
Gf,
Total Gf = TGF
ci Gr

Prostate

Seminal vesicle

Penis

Females (N = 4) Males (N = 2)
3.27=0:20 (№ = 3) 3.06
(3.04—3.40) (3.02, 3.10)
0.40+0.11 (N = 3) 1.36
(0.28 —0.50) (1.30, 1.42)
1.59+0.19 (N = 3) 0.80 (N = 1)
(1.40—1.78)
0.57+0.06 (М = 3) —
(0.50 —0.60)
0.52+0.07 (N = 3) —
(0.46 —0.60)
1.09+0.10 —
(0.96—1.20)
0.24+0.03 —
(0.20 —0.28)
0.43+0.08 —
(0.32—0.50)
0.22+0.02 —
(0.20—0.25)
0.73+0.05 0.70
(0.70 —0.80) (0.60, 0.80)
0.52+0.04 0.44
(0.50 —0.58) (0.40, 0.48)
0.17+0.04 0.21
(0.12—0.20) 0.20, 0.22
0.32+0.07 0.48
(0.25—0.39) (0.42, 0.55)

12+1.4 11.5
(11-14) (AZ)
0.11+0.03 (N = 5) M+F 0.12
(0.08—0.12) (0.10, 0.14)
0.10=0.02 (М = 5) М+Е 0.11
(0.08—0.14) (0.10, 0.12)
0.22+0.03 0.23
(0.18—0.26) (0.20, 0.26)
0.54+0.08 0.48
(0.46 —-0.67) (0.46, 0.50)
— 0.73
(0.70, 0.76)
— 0.76 (М = 1)

marginals (Х of 21.3 vs. 18.1 for the ощег or two basal cusps, they arise from the lateral
marginals). The central tooth is featured in angle of the central tooth. The bases of the

Figure 15C, E, H. The central raised “tongue” lateral angles are flared as in the above men-
running anterior—posterior on the face of the tioned species of Pseudobythinella.
tooth is flanked by deep eye-socket-like de- The shape of the lateral teeth is empha-

pressions (contrast lack of such deep holes sized in Figure 15G, H. Marginals are fea-

| on each side of the “tongue” in Pseudo- tured in Figure 16. The outermost cusp of the
bythinella kunmingensis and P. daliensis outer marginal is extremely elongated, a
’ (Davis et al., 1985: figs. 9, 16). Whether one unique feature (Fig. 16F—H).

162 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

jem
Ov 0.5mm

FIG. 11. Bursa copulatrix complex of organs of
Akiyoshia chinensis positioned as in Figure 10.

Nervous system. No data.
Remarks

This species differs from those classified as
Pseudobythinella by possessing a slender

Di

high-turreted shell without a trace of a tooth or
node on the columella as examined in the ap-
erture or within the body whorl. The standard
seminal receptacle is lost; its function is
moved into the oviduct. The bursa is a com-
paritively small sphere clearly visible posterior
to the pallial oviduct in contrast to the elon-
gated tubular bursa mostly, if not entirely, bur-
ied dorsal to the albumen gland in Pseudo-
bythinella. The central tooth of the radula has
deep-set socket-like depressions on either
side of the “tongue” on the face of the tooth;
these deep holes are lacking in Pseudo-
bythinella.

Pseudobythinella Liu & Zhang, 1979
Synonymy. Erhaia Davis & Kuo, 1985.

Type species. Pseudobythinella jianouensis
Liu 8 Zhang, 1979: 135-136, figs. 1-3.

Ast

0

.5mm

FIG. 12. Uncoiled male of Akiyoshia chinensis without head and kidney tissue removed. Some lobes of
gonad are removed to reveal seminal vesicle (Sv). Dashed line shows contour of gonad.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 163

A x „Ast ‚Pc

eS nn

FIG. 14. Stomach of Akiyoshia chinensis positioned
as in Figures 10 and 12.

0.33mm

Types. IZAS, FJ767701, holotype, plus para-
types.

Assigned Species. China only: P. jianouensis
Liu & Zhang, 1979; Bythinella chinensis Liu &
Zhang, 1979; B. hubeiensis Liu, Zhang &
Wang, 1983b; B. gongjianguoi Kang, 1983b;
B. wufengensis Kang, 1983a; B. watanensis
Kang, 1983a; P. liui Kang, 1983b; P. shime-

| nensis Liu, Zhang & Chen, 1982; В. Ш Kang,
1985; Erhaia daliensis Davis & Kuo, 1985; E.

| 0.25 mm
| kunmingensis Davis 8 Kuo, 1985. N = 11.
| FIG. 13. Position of base of penis of Akiyoshia chi- 9

| nensis on neck (А, В) relative to snout-neck mid-
line (x); penis (C).

Diagnosis. Bythinella—like shell, minute, with
smooth, flattened apical whorls. Apical whorls
may have faint spiral microsculpture. Col-

Type locality. Jian'ou, Fujian Province, Peo-
ple's Republic of China.

Designation. By monotypy with designated
type species.

umella with spiral glassy thickening that may
form a considerable keel or spiral ledge; the
ledge may terminate in the aperture appear-
ing as a “tooth” on the inner lip. Anatomical
criteria are based on four species for which

| TABLE 5. Radular statistics for Akiyoshia chinensis. Mean + standard deviation (range). N = number
| used. In mm except for width of central tooth in pm.

Females (N = 7) Males (N = 4)
Shell length 1.71+0.09 —
| (1.60— 1.78)
| Radular length 0.39+0.02 0.39+0.02
(0.36—0.42) (0.37—0.40)
Radular width 0.040.003 0.040.005
(0.038—0.048) (0.036—0.046)
Total rows of teeth 90.4+4.3 84.5+5.1
(84—96) (81—92)
No. rows о teeth 33+1.6 31-3123
forming (31-35) (30—33)
Central tooth width 10.5+0.5 (N = 21) 9.8+0.5 (N = 19)
(9.5—10.7) (8.9—10.2)

H, females. C, E, H. features central teeth; G, H.

( CCC SSS

ns, e a

4 (4 f it ( = |

< erat AA ere |

SS

2.
<
>
où
2
=
=
g
=
x
©)
=
<
=)
Se
=)
=
A]
Lu
ac
O
a)
>
<
о

FIG. 15. Radula of Akiyoshia chinensis. A-E, males; F—

Lateral teeth. D. features inner marginals.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 165

j > M * a aS Lau
A} ey | ee

{if

FIG. 16. Radula of Akiyoshia chinensis. A, B, F = males; CE, G, H = females. A. features inner marginals.
B, D. Basal morphology of attachment zone of marginals. F-H. Outer marginals. See text for details.

166 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 6. Cusp formulae for the radular teeth of Akiyoshia chinensis with the percent of radulae in which

a given formula was found at least once.

Central Teeth Lateral Teeth

4-1-4 57% 4-1-4 100%
1-1

5-1-5 57%
E

5-1-4 29%
1-1

5-1-5 29%
2-2
4-1-4 14%
2-2
5-1-5 14%
1-2

Inner Marginal Outer Marginal

teeth teeth

14 — 14%
15 — 14%
16 — 29%
17 — 71%
18 14% 43%
19 43% 71%
20 71% 43%
21 86% 29%
22 71% =
23 86% —
24 57% =
25 29%

27 EE 18.1+1.7

N = 70 N = 70

*Mean + standard deviation of cusp number for all teeth counted.

TABLE 7. Shell measurements (mm) of Pseudo-
bythinella chinensis. Mean + standard deviation
(range). Number measured = 5. All shells of
mature animals had 4.5 whorls.

Length (L) 1.77+0.06 (1.68—1.84)
Width (W) 0.85+0.03 (0.83—0.90)
L last three whorls 1.64+0.06 (1.56—1.68)
L body whorl 1.07+0.06 (1.00-1.16)

L penultimate whorl 0.36+0.01 (0.34—0.36)

W penultimate whorl 0.69+0.07 (0.56—0.74)
L 3rd whorl 0.53+0.01 (0.52—0.54)
L aperture 0.70+0.03 (0.66—0.72)
W aperture 0.54+0.04 (0.52—0.60)
x 0.16+0.03 (0.12—0.20)

y 0.04+0.01 (0.02—0.06)

we have data; the two species of this study
and P. daliensis and P. kunmingensis de-
scribed in Davis et al. (1985). The female go-
nad is a simple tube or tube with low lobes.
The oviduct has a wide 360° loop dorsal to
the bursa copulatrix. There is the standard
generalized seminal receptacle. Hydrobiid—
type central tooth. The lateral tooth lacks a
pronounced intermediate cusp. A lateral tooth

has a pronounced curved process projecting
posterior from the face of the tooth. The pro-
cess is slight, or if large it is straight in Akiy-
oshia chinensis. Stomach without caecal ap-
pendix. There is a tendency for hypertrophy of
the radular sac.

Pseudobythinella chinensis (Liu and Zhang)

Holotype. IZAS, HN766602, Liu & Zhang,
1979: fig. 4.

Type locality. | Lengshuijiang, Xinhua,
Hunan Province; July 1976. (approxi-
mately 27°4'N, 111°25’E)

Synonymy. Bythinella chinensis Liu & Zhang,
1979: 134, 136.
Erhaia chinensis, Davis et al., 1985
Pseudobythinella chinensis, this paper

Habitat

Snails were collected from Shen Keng Vil-
lage, Sanhe Town, Lingxian County, Zhu
Zhou Prefecture; 26°17'18" N, 113°31'50” E;
Figure 1, site 6. The field collection number

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 167

200Un

FIG. 17. SEM picture of shells (A-E) and opercula (F, G) of Pseudobythinella chinensis. Note that in A, no
“tooth” is seen on columella. Tilting shell to look inside body whorl reveals a swelling or “tooth” on columella.
C, D reveal no spiral microsculpture. See text for details. The apical whorls are smooth (E). Inner surface of
operculum (F); outer surface (С).

168 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 18. SEM of shells of Pseudobythinella chinensis broken to show raised spiral ridge and “tooth” on
columella inside body whorl. See text for discussion of sculpture on columella and inner lip.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 169

TABLE 8. Lengths (mm) or counts of non-neural organs and structures of Pseudobythinella chinensis. N
= number of snails used. Mean + standard deviation (range).

Body
Gonad
Digestive gland

Posterior pallial oviduct
(= albumen gland)
Anterior pallial oviduct
(= capsule gland)
Total pallial oviduct
= OV
Bursa copulatrix
BU
Duct of BU
BU - OV

Seminal receptacle
Duct of seminal receptacle
Mantle cavity

Gill (G)
Osphradium (OS)
©5-—@

No. of filaments
Gf,

Gf,

Total Gf = TGF
Gi Gr
Prostate

Seminal vesicle

Females (N = 2) Males (N = 1)
3.68 3.28
(3.42,3.94) —
0.32 0.90
(0.30,0.34) —
1.72 1.80
(1.70,1.74) —
1.20 —
(No. Var.)
0.38 Tr

(0.36,0.40)

0.32 —

(0.30,0.33)

0.04(N=1) —

0.12(N=1) —
= 0.52
És 0.24
= 12
= 0.10
ыы 0.10
E 0.20
ee 0.50
>32 0.64
EN 0.40
und, 0.52

Penis

assigned was D85-81. Snails came from a
small stream that flowed between the peaks
of big mountains; the elevation was 500 m.
The stream was heavily shaded by vegetation
lining the banks. The stream was 20-30 cm
wide, 10-20 cm deep with a bottom paved
with small rocks, leaves and mud.

Depository

Specimens are housed at ZAMIP, M0011;
ANSP, 373139, A12655.
Description

Shells. Shells are minute, smooth, and
conic-turreted (Figs. 6J-N; 17A-E; 18A-H).

They are 4.5 whorls ranging in length from
1.68 to 1.84 mm (Table 7); they are umbilicate
with a sub-ovate to sub-circular aperture. The
sutures are moderately deep; whorls with
slightly convex whorls. No tooth is visible in
the aperture in 60% of the shells while a thick-
ening on the columella is seen in the aperture
of 40% of the shells. Tipping the aperture up,
one can see a considerable tooth-like swell-
ing on the columella, the terminus of the pro-
nounced internal columellar keel. In side view
the outer lip is straight to slightly sinuate; in
the contrasting view the inner lip is straight
(outer lip down, 90° to the horizontal).

SEM analyses show the apical whorls to be
smooth (Fig. 17E). With the aperture tilted,

170

FIG. 19. The buccal mass of Pseudobythinella chi-
nensis with pronounced radular sac (A) and re-
flected mantle showing mantle cavity organs (B).

the columeller “tooth” is seen (Fig. 17C, D);
note the numerous calcareous pitted pustules
on the columella and “tooth”. With the shell
broken open, it is seen that the “tooth” seen
in the aperture is the terminus of a pro-
nounced columellar keel starting at mid-body
whorl and spiralling down to the aperture (Fig.
18A-E). An enlargement of the area of highest
concentration of pitted calcareous pustules
on the “tooth” (Fig. 18G) is given in Fig. 18H.
The entire surface of the keel is roughened
with calcareous pustules or micro-ridges.

External features. The head-neck is white;
there are no white granules about the eyes.
The operculum is corneous and paucispiral
(Fig. 17F, G). The shape is sub-rectangular.
The columeller edge is somewhat concave,
especially where the operculum passes over
the columeller “tooth” (Fig. 17C). The callus
on the inner surface is weakly developed, but
wide, 69% the width of the operculum (Fig.
17F).

Mantle cavity. The opened and reflected
mantle cavity is shown in Figure 19B. Mea-

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

surements and counts of relevant organs are
given in Table 8. The gill is centrally located in
the mantle cavity and the osphradium (Os) is
centered against the gill. The osphradium by
definition is long but this is an artifact of the
much reduced size of the gill. The usual po-
matiopsid gill nearly fills the length of the
mantle cavity. On this criterion the osphra-
dium is here considered short (circa 0.32
ratio). The Gf, is medium (= normal) length.
The length of the longest gill filament is ap-
proximately 0.20 mm.

Female reproductive system. The body of an
uncoiled female snail is shown in Figure 20
without head and with kidney tissue removed.
Measurements of relevant organs are given in
Table 8. Important features are: (1) The go-
nad (Go) is posterior to the stomach (Pst). (2)
The gonad consists of a simple sac with four
to five small undivided lobes protruding from
the posterior end. (3) The bursa (Bu) is either
completely covered by the albumen gland
(Ppo) or only protrudes slightly posterior to
the albumen gland (as in Fig. 20). (4) The
oviduct (Ov) makes a pronounced bend
pressing against the posterior edge of the
bursa (Bu). (5) The spermathecal duct (Sd)
runs to the anterior end of the mantle cavity to
open independently of the pallial oviduct. (6)
The sperm duct (Sdu) is unique in that it
branches off the spermathecal duct near the
anterior end of the spermathecal duct, far for-
ward of the posterior end of the mantle cavity
(Fig. 21). (7) The seminal receptacle (Sr) is
unique in that it branches from the sperm duct
(Sdu), not the oviduct, and in that it is entirely
anterior to the posterior end of the mantle
cavity (Fig. 21).

Male reproductive system. An uncoiled male
is shown in Figure 22 without head or kidney
tissue. Part of the gonad (Go) has been cut
away to reveal the seminal vesicle (Sv) coiled
dorsal to it. Measurements of relevant organs
are given in Table 8. Important features are:
(1) The gonad is posterior to the stomach. (2)
There is no vas efferens in the usual sense.
Thick lobes arise from a wide collecting gutter
that functions as a vas efferens. (3) The sem-
inal vesicle (Sv) arises from mid-gonad and
makes a small knot dorsal to the gonad. (4)
The posterior vas deferens (Vd,) runs as a
wide tube, swollen with sperm, from the sem-
inal vesicle to become a slender duct only at
the style sac. (5) The prostate overlies the

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 171

Ast

Sts

Di

pi er
0.52mm

FIG. 20. Uncoiled female Pseudobythinella chinensis with head and kidney tissue removed.

Osd

FIG. 21. Details and variation of bursa copulatrix
complex of organs of Pseudobythinella chinensis.
Figure 21A is in same orientation as in Figure 20. In
B, oviduct cut and reflected in direction of arrow to
show opening of oviduct (Oov) that attaches to al-
bumen gland

172 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Pst

Emc

|
0.5 тт

FIG. 22. Uncoiled male of Pseudobythinella chinensis without head or kidney tissue. Some of lobes of
anterior gonad are cut away (- - -) to show knot of seminal vesicle (Sv) dorsal to gonad.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 173

x

Bp

Vd

> я
0.2mm

FIG. 23. A. The orientation of base of penis (Bp) of
Pseudobythinella chinensis to snout-neck mid-line
(x) and lobes of eyes (a). B. Penis.

posterior end of the mantle cavity. (6) The
anterior vas deferens (Vd,) separates from
the prostate at mid-prostate. (7) The penis
(Fig. 23B) has an enormous muscular ejacu-
latory duct (Ej) in the base that continues to
coil for a short distance in the neck. (8) The
penis has an enormously elongated papilla, a
unique feature. (9) The base of the penis (Bp,
Fig. 23A) arises to the right of the snout-neck
mid-line x and at 90” to it.

Digestive system. The digestive gland (Di)
covers the posterior chamber of the stomach
(Pst, Figs. 20, 22). The radular sac (Rs) loops
up over the buccal mass (Bm, Fig. 19 A).
Radular statistics are given in Tables 9, 10.
The radula is minute, with an extraordinary
number of rows of teeth (97 per 0.48 mm)

TABLE 9. Radular statistics for Pseudobythinella
chinensis. Mean + standard deviation (range). N
= number used. In mm except for width of
central tooth in um.

Sex Unknown (N = 4)

Shell length 1.79+0.001
Radular length 0.48+0.01
Radular width 0.06+0.004

Total rows of teeth 97.3+6.7 (N = 3)
No. rows of teeth forming 10.3+4.0

Central tooth width 12.0+0.5 (N = 9)

(Fig. 24). The most commonly encountered
cusp formula is (4)5-1-5(4); 4(5)-1-(5, 6)4;
1-1

24-29; 17-22:

The inner marginal teeth have significantly
more cusps(mean of 26) than do the outer
marginals (mean of 20). The two outer cusps
of the outer marginals are specialized to form
a pincer-like process (Fig. 24G, H). The
“tongue” on the face of the central tooth is
broad and is not flanked by deep-set holes
(Fig. 24A, B).

Nervous system. No data.

Remarks

This species differs from the other three for
which we have anatomical data by the follow-
ing: (1) The bursa may somewhat protrude
posterior to the albumen gland. (2) the oviduct
makes a pronounced bend near the bursa;
the ascending and descending arms of the
bend are pressed together (contrast the open
360” loops or complex oviduct loops of the
other species). (3) The sperm duct is elon-
gated and extends anterior to the posterior
end of the mantle cavity to enter the sper-
mathecal duct, a unique character-state. (4)
The seminal receptacle branches off the
sperm duct, not the oviduct, a unique charac-
ter-state. (5) There is no vas efferens. (6) The
penis has an extremely elongated papilla, a
unique character-state.

Pseudobythinellu. shimenensis Liu,
Zhang & Chen

Holotype. IZAS, HN797904; Liu et al., 1982:
254, 256, fig. 1.

APTOITE

FIG. 24. Radula of Pseudobythinella chinensis. B. Central and inner marginal teeth emphasized. C, D.

Details of morphology of lateral teeth. E, F. Inner marginals emphasized. G, H. Outer marginals. See text for
details.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 175

TABLE 10. Cusp formulae for the radular teeth of Pseudobythinella chinensis with the percent of the four
radulae in which a given formula was found at least once.

Central Teeth Lateral Teeth

5-1-5 75% 4-1-4 100%
1-1

5-1-5 75%
5-1-4 50%
1-1 6-1-5 50%
4-1-4 25% 4-1-5 50%
1-1

5-1-4 50%

6-1-4 25%

Inner Marginal Outer Marginal

Teeth Teeth
ir. — 75%
18 — 75%
19 — 75%
20 — 75%
21 — 50%
22 — 50%
23 — 25%
24 25% 25%
25 100% =
26 100% =
27 75% —
28 25% —
29 25% =
Х 2260 20.0+2.1
N = 40 N = 33

*Mean + standard deviation of cusp number for all teeth counted.

Type locality. Shimen, Hunan Province, Oc-
tober 1979.

Synonymy. Pseudobythinella shimenensis
Liu, Zhang 4 Chen, 1982: 254-256.
Pseudobythinella shimenensis Davis et al.,
1985: 68.

Habitat

Snails were collected from Qingguandu Vil-
lage, Nanzhen Town, Shimen County,
Changde Prefecture; 29°56’36” N, 110°41'42”
E; Figure 1, site 3; topotypes. Snails were col-
lected from under stones and leaves at the
edge of a small pool of a stream. The water
was Clean, clear, cool.

Depository

Specimens are housed at ZAMIP, M0012,
ANSP 373137, A12653.

TABLE 11. Shell measurements (mm) of Pseudo-
bythinella shimensis. Mean + standard deviation
(range). Five shells measured, all 4.0 whorls.

Length (L) 1.98+0.04 (1.94—2.04)
Width (W) 15172002 (15141420)
L last three whorls 1.95+0.04 (1.92—2.02)
L body whorl 1.44+0.03 (1.41 —1.48)

L penultimate whorl
W penultimate whorl

0.38+0.01 (0.38—0.40)
0.82+0.02 (0.80 —0.84)

L 3rd whorl 0.48+0.02 (0.45—0.50)
L aperture 0.99+0.02 (0.96—1.00)
W aperture 0.76+0.03 (0.73—0.80)

x 0.24+0.12 (0.10—0.38)
y 0.07+0.04 (0.04—0.12)

Description

Shells. Shells are minute, smooth, ovate with
flattened apex (Figs. 6F-I, 25, 26). Shell mea-
surements are given in Table 11; shell lengths
range from 1.94 to 2.04 mm for shells of 4.0

176 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Я 200 UN ————

FIG. 25. SEM photographs of shells of Pseudobythinella shimenensis. A. Note slight swelling or tooth on
columella. C. Apical whorls showing spiral microsculpture starting at 0.5 whorls. D. Aperture rotated to
expose fully tooth on columella. E, F. Columella highly magnified to show sculpture pattern on inner lip,

columella, and tooth.

whorls. The aperture is broadly ovate to sub-
circular. The sutures are deep and the whorls
shouldered, convex. The inner lip is straight to
saddle-shaped with a narrow umbilicus above
the center of the saddle of 33%; no umbilicus,
66%. The columella has a tooth plainly in view

in the aperture. The face of the body whorl is
flattened in most specimens. The outer lip is
straight to slightly arched (Fig. 25B). In side
view, the adapical outer lip is fused to the
body whorl in 40%; separated by 0.05 mm +
0.03 mm (N = 5) from the body whorl in 60%.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 177:

FIG. 26. SEM photographs of shells of Pseudobythinella shimenensis broken open to reveal raised spiral
ridge on columella of body whorl that terminates as tooth. C, D. Large scale-like sculptural plates on
columella at aperture contrast with thin line-like raised ridges on spiral ridge within body whorl (G).

178 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

m
0.33 mm

FIG. 27. Head (A) and penis (B) of Pseudo-
bythinella shimenensis. A. Relationship of base of
penis to mid-line of snout-neck (x) and to posterior
end of eye lobes (a).

The inner lip is straight (outer lip down 90° to
horizontal).

SEM examination shows that the tip of the
apical whorl is smooth with fine spiral sculp-
ture seen starting in less than half a whorl
(Fig. 25C). Tilting the aperture somewhat, the
columellar “tooth” is clearly seen (Fig. 25D—
F). It is covered with raised calcareous lap-
pets (Fig. 25E), contrasted with the pustules
seen in P. chinensis. Upon breaking open
several shells, it is seen that the “tooth” is the
terminus of a columellar keel that starts mid-
body whorl (Fig. 26A, B, E-G). The calcare-
ous lappets begin on the columellar keel and
extend to the aperture (Fig. 26D, G); above
the keel there are minute raised calcareous
ridges (Fig. 26G).

External features. The head is white with a
small cluster of white granules about the me-
dial edge of the eye, i.e. an “eyebrow” (Ebr,
Fig. 27). There is a scattering of white gran-
ules posterior to the eyebrow at the side of the
neck. The operculum is corneous, paucispiral
and kidney-bean shaped with the columel-
lar—side concavity corresponding to the tooth
on the columella (Fig. 28). On the inner sur-
face of the operculum the attachment pad for
the muscle has a pronounced ridge (Fig. 28A,
C). The pad is relatively narrow, some 40%
the width of the operculum.

Mantle cavity. The reflected mantle showing
mantle cavity organs is given in Figure 29;
measurements and counts are given in Table
12. There are relatively few gill filaments (12—
13) yet these take up 78% the length of the
mantle cavity. The osphradium is mid-gill and
is long. Gf, is long; the longest gill filaments
are 0.36 mm long. There is no cluster of white
granules just anterior to the osphradium close
to the neck (Ne)—mantle collar (Ma) junction.

Female reproductive system. The body of an
uncoiled female with head and kidney tissue
removed is shown in Figure 30. Measure-
ments and counts of organs are given in Ta-
ble 12. Important features, to note are: (1)
The gonad is located posterior to the stom-
ach. It is a single small sac that has three or
four small protruding lobes created by pres-
sure of individual oocytes at the posterior end
of the gonad. (2) The albumen gland (Ppo)
covers most of the bursa copulatrix (Bu) and
extends beyond the bursa to cover most of
the style sac. The albumen gland curves
around the bursa leaving the posterior end of
the bursa exposed. Only in a few cases is the
bursa completely covered by the albumen
gland. (3) The spermathecal duct (Sd) is
tightly pressed to the pallial oviduct and
opens into the mantle cavity close to the an-
terior end of the capsule gland (Apo). (4) The
bursa copulatrix complex of organs is shown
in Figure 31 oriented in the same relative po-
sitions as in Figure 30. The bursa is short. (5)
The oviduct enters (Oov) the albumen gland
posterior to the posterior end of the mantle
cavity (Emc). The sperm duct is short and
branches off the spermathecal duct posterior
to the posterior end of the mantle cavity. (6)
The duct of the seminal receptacle (Dsr)
branches off the oviduct posterior to the
sperm duct (Fig. 31). The seminal receptacle

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 179

FIG. 28. Opercula of Pseudobythinella shimenensis. A, C. Inner surface. B, D. Outer surface. Note prom-
inent ridge on inner surface running along inner edge of attachment pad.

180 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 12. Lengths (mm) or counts of non-neural organs and structures of Pseudobythinella
shimenensis. N = number of snails used. Mean + standard deviation (range).

Females (N = 5) Males (N = 1)

Body 3.62+0.34 3.94
(3.26—4.04)

Gonad 0.47+0.09 2.00
(0.34—0.60)

Digestive gland 1.78+0.15 2.10
(1.68—2.04)

Posterior pallial oviduct (= albumen gland) — —

Anterior pallial oviduct (= capsule gland) — —

Total pallial oviduct 1.52+0.09 —

=OV (1.40—1.60) М = 4
Bursa copulatrix 0.49+0.11 —
ВУ (0.36—0.60) М = 4
Duct of ВУ 0.26 (N = 2)
No. var.

BU + OV 0.33+0.08 —
(0.23—0.40) М = 4

Seminal receptacle 0.11+0.02 —
(0.08—0.13) М = 4

Duct of seminal receptacle 0.16+0.05 =
(0.10—0.22) N = 4

Buccal Mass 0.46 (N = 1) —

Mantle cavity 0.98+0.11 .90
(0.84—1.10) М = 4

Gill (G) 0.76+0.07 .70
(0.70—0.84) N = 4

Osphradium (OS) 0.31+0.08 .28
(0.20 —0.40)

OSEIG 0.40+0.08 .40
(0.29—0.48) М = 4

No. of filaments 12.3+0.5 .28

(12-13)

Gf, 0.20 (N = 2) —

Gf, 0.16 (N = 2) —

Total Gf = TGF 0.36 (N = 2) =

Gia GE 0.55 (N = 2) —

Prostate — 1.00

Seminal vesicle — .80

Penis — .90

TABLE 13. Radular statistics for Pseudobythinella shimenensis. Mean + standard deviation (range). N
= number used. In mm except for width of central tooth in рт.

Shell length

Radular length

Radular width

Total rows of teeth

No. rows of teeth forming
Central tooth width

Sex Unknown
2.1+0.12 (1.9-2.2)N = 9

0.79+0.46 (0.71 —0.84) М = 5
0.071 +0.005 (0.064—0.076) N = 6

113+1.7 (111-115)

9+3 (6-14)

14.1+0.82 (13.2-15.3) N = 13

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 181

Sr

EF,
0.5mm

FIG. 29. Mantle cavity of Pseudobythinella shime-
nensis. Mantle cut and reflected to show mantle-
cavity structures.

0.5mm

FIG. 31. Details and variation of bursa copulatrix
complex of organs of Pseudobythinella shimenen-
sis. Figure 31A is in same orientation as in Figure
30. B. Bursa cut away to show seminal receptacle
(Sr) and oviduct coil (Coi).

| HA EZ
| Di 0.5mm
|

FIG. 30. Uncoiled female Psuedobythinella shimenensis with head and kidney tissue removed.

182

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

D A

0.5mm

Ma

FIG. 32. Uncoiled male of Pseudobythinella shimenensis without head or kidney tissue. Some lobes of
gonad removed (iine with arrows) to reveal seminal vesicle (Sv).

and the duct of the seminal receptacle are
well developed. (7) Posterior to the duct of the
seminal receptacle the oviduct makes an ir-
regular loop or double coil that is dorsal to the
bursa. In Figure 31A, the coil has been pulled
out a little from beneath (dorsal to) the bursa
to show the pattern of coiling. The coil is
shown in actual position in Figure 30.

Male reproductive system. The body of an un-
coiled male is shown in Figure 32 with head
and kidney tissue removed. Measurements
and counts of organs are given in Table 12.
Important features to note are: (1) The gonad
covers the stomach. (2) A portion of the go-
nadal lobes has been cut away to show the
seminal vesicle (Sv) that coils regularly dorsal
to the gonad. (3) There is a well-defined vas
efferens (Ve) with the posterior vas deferens
arising from the vas efferens at mid—to

slightly posterior to mid-gonad. (4) There are |
a number of bundles of testicular lobes arising |
from the vas efferens. (5) The prostate over-

lies the posterior end of the mantle cavity —

(Emc). (6) The anterior vas deferens (Vd,) -
leaves the prostate (Pr) close to the posterior
end of the mantle cavity. (7) The penis is sim-
ple and without discernable papilla (Fig. 27B). :
(8) The base of the penis (Bp) arises from the ©
neck to the right of the snout-neck mid-line (x)
and at 90° to it (Fig. 27A). There is no dis-
cernable ejaculatory duct.

Digestive system. The digestive gland covers |
the posterior chamber of the stomach of fe- |
males (Di, Fig. 30), and the entire stomach in -
males (Fig. 32). The radular sac (Rs, Fig. 33) :
is highly elongated, coiling dorsal to the buc- —
cal mass (Bm).

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 183

0.33mm

FIG. 33. Dorsal aspect of buccal mass of Pseudo-
bythinella shimenensis with dorsal part of nerve
ring. Note radular sac (Rs) coiling dorsally over
buccal mass (Bm).

Radular statistics are given in Tables 13
and 14. The most commonly encountered
cusp formula is

4(5)-1-(5)4; 2(3)-1- 3(4); 17-20, 16-18.
1-1

There is a very large number of rows of teeth
for the comparatively short length of radula
(113 per 0.79 mm). The radula is illustrated
in Figures 34, 35. Fig. 34B, D, E feature the
central tooth. The face of the tooth is moder-
ately raised as a tongue with a slight concav-
ity on either side extending beneath the basal
cusps. The lateral angles have the flared
ends typical of Pseudobythinella.

The lateral teeth are featured in Figures

34C, D, F and 35A, B, E. Two points are
important. (1) The basal process of the lateral
tooth is prominent and curved towards mid-
radula as also seen in P. chinensis but in con-
trast to the weak or slighly developed straight
basal process seen in Akiyoshia chinensis.
(2) As in the above mentioned taxa, there is a
gradation in size of the cusps on the lateral
tooth. The “1” of the 3-1-3 is not considerably
larger than the flanking cusps. Inner marginal
teeth (Figs. 34B, C, F; 35C, D, E, F) and outer
marginals (Figs. 34C, E; 35C, G) have numer-
ous small cusps. No one cusp is extra long
(i.e. with derived specialization).

Nervous system. A segment of the nervous
system is shown in Figure 33. The cerebral
commissure (Cc) is elongated. The dorsal
nerve ring is moderately concentrated (RPG
of 0.38, Table 15). Otherwise the dorsal as-
pect of the nerve ring is typical for the Tricu-
linae.

Remarks

This species differs from the others for
which we have anatomical data as follows: (1)
The glands about the eye form an “eyebrow”.
Pseudobythinella chinensis and P. daliensis
lack any glands about the eyes; there is a
scatter of glands in P. kunmingensis. (2) A
tooth is clearly visible in the aperture of the
shell. (3) The bursa is an elongated tube; it is
an elongated ovoid sac in the other species.
(4) The penis lacks an ejaculatory duct.

Pseudobythinella daliensis differs from the
other three species by having a small, not
minute shell. It does not have an elongated
radular sac as do the others.

Pseudobythinella kunmingensis _ differs
from the other three species by having a scat-
ter of glands about the eyes. The spermath-
ecal duct shares a common opening with the
capsule gland (Apo) (not shown in Fig. 31
where the spermathecal duct is cut short).

TRICULINAE

Pachydrobiini Davis & Kang, 1990

Type genus. Pachydrobia Cross & Fischer,
1876

Diagnosis. Genera of Triculinae in which the
spermathecal duct bypasses the pericardium
and the oviduct does not make a closed 360°
twist. With the exception of Wuconchona,

184 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

ed

AA des

h N) )
Ya |W hs E я *
y A E /

“à

7.

#

FIG. 34. Radula of Pseudobythinella shimenensis featuring central and lateral teeth (B—D, F) and outer

marginal teeth (E).

there is a sperm duct. The seminal receptacle
arises from the bursa or the duct of the bursa
in the plesiomorphic state; the seminal recep-
tacle is lost in the derived state and its func-
tion taken over by new structures.

Genera assigned. Guoia, Halewisia, Neotric-
ula, Pachydrobia, Robertsiella, Jinhongia,
Gammatricula, Wuconchona (N = 8).

The Lithoglyphopsis Problem

Considering all Asian genera of freshwater
rissoacean snails, it has been especially im-
portant to locate and study the type species of
Lithoglyphopsis for two reasons. (1) Lithogly-
phopsis was the taxon that caused early Eu-
ropean workers to include within the family
Hydrobiidae those Asian taxa now known to

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 185

FIG. 35. Radula of Pseudobythinella shimenensis featuring entire lateral tooth (A, B, E), inner marginals

(C-F), outer marginals (G).

be Pomatiopsidae: Triculinae. (2) The snail
transmitting Schistosoma mekongi Voge et
al., 1978, was originally described as Litho-
glyphopsis aperta Temcharoen, 1971.
Considering the first, the shells of certain
| Chinese species from Hunan Province so
much resembled shells of European Lithogly-
| phus that they were described as species of

Lithoglyphus (Gredler, 1881, 1886; Moellen-
dorff, 1888). Subsequently Thiele (1928)
noted that the morphology of the central tooth
of L. modestus differed slightly from that of
European Lithoglyphus and he therefore es-
tablished the genus Lithoglyphopsis, with L.
modestus Gredler, 1886, as its type species
by original designation. However, overall shell

186 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 14. Cusp formulae for the radular teeth of Pseudobythinella shinenensis with the percent of the
radulae in which a given formula was found at least once.

Central Teeth Lateral Teeth

4-1-4 80% 2-1-3 60%
1-1
5-1-5 40% 2-1-4 60%
1-1
5-1-4 20% 3-1-3 60%
1-1
4-1-4 20% 3-1-4 40%

Inner Marginal Outer Marginal

Teeth Teeth
14 = 20%
15 = 40%
16 == 60%
17 80% 100%
18 100% 80%
19 100% 40%
20 60% 20%
21 40% =
X* = 18.7+1.2 17.2+1.3
N = 50 N = 49

*Mean + standard deviation of cusp number for all teeth counted.

TABLE 15. Lengths of neural structures of Pseu-
dobythinella shimenensis. The mean is given and
with data when N = 2. N = number used.

Cerebral ganglion 0.20 (N = 2)
(No. Var.)
Cerebral commissure 0.09 (N = 2)
(0.08, 0.10)
Pleural ganglion
Right (1)* 0.10 (N = 1)
Left —
Pleuro-supraesophageal 0.12 (N = 1)
connective (2)*
Pleuro-subesophageal —
connective
Supraesophageal ganglion (3)* 0.10 (N = 1)
Subesophageal ganglion —
Osphradio-mantle nerve 0.04 (N = 1)
RPG ratio* = 2 - 1+2+3 0.38 (N = 1)

and raduiar characters persuaded Thiele
(1928) to include Lithoglyphopsis of the Hy-
drobiidae along with 12 other Asian Triculinae
genera within the Tribe Lithoglypheae along
with European Lithoglyphus. As late as 1974,
Brandt included the Asian genera in question
in the family Hydrobiidae, subfamilies Triculi-
nae, Cohilopinae, Rehderiellinae, and Litho-
glyphinae.

As for the second, Davis et al. (1976) stud-
ied this species and stated that it was most
closely related to Tricula, especially Tricula

burchi Davis, 1968b; they stated that Tricula
might be a suitable genus for “L.” aperta. In
1980, Davis referred L. aperta to Tricula.
Davis et al. (1976) pointed out that on the
basis of both shell and radula, Tricula aperta
could not be considered a species of Litho-
glyphopsis. However, the questions have re-
mained: What is Lithoglyphopsis? To which
genera is Lithoglyphopsis most closely re-
lated? Can one establish once and for all that
Lithoglyphopsis is a member of the Triculinae,
not a member of the Hydrobiidae: Lithoglyph-
inae? What is the potential for species of
Lithoglyphopsis to transmit a species of
Schistosoma?

As a result of our studies in Hunan Prov-
ince, it was clear that species historically con-
sidered to be Lithoglyphus in Hunan belong to
two genera: Lithoglyphopsis and a new genus
described here as Guoia.

Guoia Davis & Chen, genus nov.

Type Species. Lithoglyphus viridulus Moel-
lendorff, 1888

Etymology. Named for Dr. Guo Yuan Hua,
Institute of Parasitic Diseases, China National
Center for Preventive Medicine, Shanghai, for
his tireless efforts to discover and understand
species of snails involved in disease trans-
mission.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

Diagnosis. Shells small (<3.6 mm long), and
globose-conic. The spermathecal duct opens
into the rear of the mantle cavity. A long, slen-
der sperm duct connects the spermathecal
duct (at a position close to the posterior end of
the mantle cavity) to the oviduct close to
where the oviduct opens into the albumen
gland. Duct of the bursa is massive, running
directly anterior from the large bursa to form
the short spermathecal duct opening at the
rear of the mantle cavity. There is no seminal
receptacle; sperm are stored in a swelling of
the oviduct just posterior to the juncture of the
sperm duct and oviduct, or they are stored in
an outpocketing of the sperm duct at the junc-
ture to the oviduct. There is a thin corneous
stylet at the tip of the penis. The penis has a
glandular lobe. The ejaculatory duct is mas-
sive, extending posteriorly along the neck
from the base of the penis. Radula Tricula-
like.

Relationships. A member of the Neotricula
clade by virtue of (1) the oviduct travels from
gonad to albumen gland straight, without twist
or coil (contrast the Tricula clade), and (2) the
spermathecal duct does not enter the pericar-
dium. The closest generic similarity is with
Robertsiella of Malaysia in that (1) the penis
of the males has a similar stylet, and (2) the
spermathecal duct-duct of bursa-bursa con-
nections and relative postitions are the same.
There are differences. In Robertsiella, the
seminal receptacle is encapsulated in the
muscular wrapped duct of the bursa (Guoia
lacks a seminal receptacle). A section of the
oviduct of Guoia serves as a seminal recep-
tacle. The shell of Robertsiella is ovate-conic,
not globose-conic. The sperm duct of Guoia is
elongated, twisting over the bursa; it is very
short in Robertsiella. The duct of the bursa is
very short in Robertsiella, elongated in Guoia.

Assigned Species.
Lithoglyphus fuchsianus Moellendorff,
1885

Lithoglyphus viridulus Moellendorff, 1888

Thiele (1928: 365) noted that of Asian spe-
cies described as Lithoglyphus, L. fuchsianus
and L. viridulus, had radulae that corre-
sponded to the European Lithoglyphus,
whereas the radulae of L. modestus and L.
tonkinianus Bavay 8 Dautzenberg (a Viet-
namese species) had an entirely different
type of central tooth. He created the genus
Lithoglyphopsis to include the latter two spe-

187

cies and named “L. modesta” as the type
species. Yen (1929) placed all Chinese taxa
described as Lithoglyphus in the genus Litho-
glyphopsis including L. liliputanus Gredler,
1881 from “Kwangtung” (= Guangdong).
The shell of this last species is a miniature
version of L. modestus. Until the anatomy is
known its generic placement is incertae sedis.

Guoia viridulus (Moellendorff, 1888).

Lithoglyphus viridulus Moellendorff, 1888:
141, pl. 4, fig. 6, 6a-b

Lithoglyphus viridulus (Moellendorff, 1888).
Thiele 1928

Lithoglyphopsis viridulus, Yen, 1939

Types: $.1.; Lectotype, 4129, fig. in Yen,
1939: pl. 4, fig. 9

Paralectotypes, 4130, figured here; Figure 36
ALC.

Type Locality: Hunan

Habitat

Anhua County, Anhua Town, Zijiang River;
28°23'46" N, 111°12’41” E., Figure 1, Site 10.
Collected by Chen and Davis, 16 March 1987;
field collection number D87-1. Catalog num-
bers are: large class, ANSP 373147, A12663;
small class, ANSP 373148, A12664. D85-78,
small class, ANSP 373149, A 12665; ZAMIP
M0055. Snails were collected 1.6 km up-
stream from the town boat landing, along the
shores of a small island in the middle of the
river. Water flows through a stone breakwater
at the upstream end of the island. Between
the breakwater and the cobbles of the island
was a protected area with water some 30 cm
deep and with emergent vegetation. On the
undersides of rocks at the breakwater on the
protected side were numerous Guoia viridu-
lus and Lithoglyphopsis modesta. Associated
snail fauna included Stenothyra hunanensis
(Davis et al. 1988), Gyraulus sp., Radix sp.,
Semisulcospira sp., and a viviparid.

Introduction

There were two size classes of fully mature
snails living intermixed at the site (Fig. 37).
Throughout the description these will be re-
ferred to as the large class and small-class
snails. There are slight shape differences be-
tween the classes, yet we have felt it prudent
not to consider them belonging to different

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 36. Lectotype and paralectotypes of species of Guoia. A-C. Paralectotypes of Guoia viridulus (SMF
4130); A = 3.50 mm long. Guoia fuchsianus, D-F. D. Lectotype, L = 3.24 mm. E, F. Paralectotypes.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

FIG. 37. Four aspects of each of six shells of Guoia viridulus. AC. Large class; A = 3.6 mm long. D-F.
Small class; D = 2.8 mm long. A, B, D, E = females; C, F = males.

190 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Adb
Cs Se ( N
y N
Als —
Cr |
Ol
Bc
A Ae B C
10mm A res
Cs
Adb
Uc Ngr N Ngr Als
Cr Ol
Ol
Bc
D Ae E F
BE

1.0mm

FIG. 38. Illustrated details of the apertural region of shells of Guoia viridulus characteristic of the species.

species as we could find no anatomical differ-
ences between them, and as they live in mi-
cro-sympatry. The greater amount of anatom-
ical data was from the large-class snails
based on living specimens. Subsequently,
anatomical data were gathered in Philadel-
phia from alcohol-preserved small class
snails.

As there is considerable similarity among
shells of the two size classes of snails that we
refer to as G. viridulus, as well as G. fuch-
sianus, we did a multivariate analysis of shell
measurements of these taxa as well as older
museum specimens of these species in order
to attempt to assess differences among them.
The following anatomical description is based
on large class snails. Small class snails are
contrasted with large class individuals in the
remarks section for this species.

Description

Shells. Figures 36-39. Large-class (Figs.
37A-C, 39A-C). Shells small, globose-conic,
4.0—4.5 whorls, smooth but with rough growth
lines. Aperture sub-circular with complete
peristome, with pronounced apertural beak
(Fig. 38, Adb) and a pronounced beak groove
(Ngr). There is a wide columellar shelf (Cs), a
small basal crescent (Bc) with crescent ridge
(Cr). Some shells have an umbilical chink
(Uc), others do not. The abapical lip has an
embayment (Ae). The adapical outer lip is
flared out making somewhat of an angle with
the remaining outer lip. In side view, the lip is
sinuate forming a slight sigmoid embayment
(Als).

Females are larger than males (Table 16).
Size ranges (lengths) are: males, 2.8 to 3.6

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 191

FIG. 39. SEM photos of shells of Guoia viridulus (A-F) and G. fuchsianus (G-I). AC. Large form; D-F.
Small form. B, C, E, F, H, | = enlargements of apical whorls.

mm; females, 2.8-3.8 mm. Our samples com-
pare well with two lots of G. viridulus from the
collections (Table 16, column 3). They com-
pare well with paralectotypes (Figs. 36A-C
compared with Fig. 37A-C).

Small-class (Figs. 37D-F). Shells as in the
large class with the following differences.
Length is substantially smaller (Table 16) with

| males 2.0-2.1 mm long, females 2.2-2.8 mm

long. The small-class shells appear some-

_ what more conic than large class shells (com-

pare Fig. 37E vs. 37A). However, the width to
length ratio is not significantly different be-
tween classes; the length of the body whorl to
total height is slightly greater in the large size
class (0.84 + 0.03, 0.84 + 0.05 for large-
class; 0.80 + 0.02, 0.80 + 0.03 for small
class). The basal crescent and columellar
shelf are deeply depressed in small class
shells (Fig. 37D). The outer lip in side view

| has a deep sigmoid embayment (Fig. 38F).

SEM pictures of the apical whorls are
shown in Figure 39B, C, E, F. The large class
individuals inevitably had eroded apices; the
small class shell apices were smooth. Some
spiral microsculpture was noted at the shoul-
der of the body whorl of small class shells.

Multivariate analysis. The phenogram based
on distance coefficients (cophenetic correla-
tion 0.77) is given in Figure 40. Note the fol-
lowing: (1) Historic G. viridulus (nos. 18-21)
form a discrete unit within the wider variance
of 9 large-class Anhua G. viridulus collected by
us. (2) One Anhua G. viridulus (no. 5) clusters
with Baisha G. fuchsianus collected by us. (3)
The one historic G. fuchsianus that had com-
plete whorls so that it could be included in this
analysis clusters in the middle of the G. fuch-
sianus variance. (4) Two small-class Anhua
G. viridulus (nos. 27, 28) and two snails clas-
sified by us as large-class Anhua G. viridulus

192 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

1.775 1.525 1.275 1.025 0.775 0.525 0.275 0.025
RE Ton TRE
oA

ов
o2/\
“ná
10 À
18 V
19 V
21 М
20 V
оз A
os À

од А Guoia viridulus - Anhua
er V -historic material

ZX -large class males

be $ A -large class females

170
33 Y
1 O

15 ® Guoia fuchsianus
14 (O) & -historic material

A © - males
& - females
12 À

27 ES
28 №
20]
23 U
25 O
24 L]
26 O

[_] - small class males

a -small class females

29 Е
зо №
з2 №
31 [|

И И ЕН
1.775 1.525 1. 275 1.025 0.775 0.525 0.275 0.025

FIG. 40. UPGMA phenogram of distances for shell measurements of two species of Guoia. See text for
details.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

193

Guoia viridulus - Anhua
V - historic material
A - large class males
A - large class females

O small class males

| | small class females

Guoia fuchsianus

© - historic material

3) 5%

T als T T T T 5 iF
0.633 Г 3359/06
0.396 + у’ 4
1811 11 21 |
19
0.158 - 4
25 2
-0.079 4
10
A 8
3
-0.316 + 4
4 9
DER me = E —" 11 1 1 1 1
-1522 -1.154 -0.786 -0.418 -0.050 0.318 0.686 1.054 1.422

I

| FIG. 41. Ordination diagram following three dimensional scaling. The first two principal components are
given. Taxa are connected by a MST. See text for details.

(nos. 7, 12) form a sub cluster with G. fuch-
| sianus. (5) Most (82%) of the small-class G.
| viridulus form a discrete cluster at the bottom.
In the PCA analysis, there were three sig-
nificant components (Table 17); the first with
76.6% of the variance, the second with 13.49,
the third with 5.69 (total of 95.8%). Character
loadings showed the first component to be
one of size (Table 18); characters loading
heavily on the second component-axis are
number of whorls and length of the penulti-
mate whorl. Ordination on the first two PCA
axes following MDS is given in Figure 41. The
MST connects OTUs. Size increases from left
to right. Along the second axis, relative to
length, shells with more whorls, longer penul-
timate whorls, and wider 3rd whorls are to the

bottom. Proportionally shorter penultimate
whorls and narrower 3rd whorls are to the top.
The second axis is one of changes in trans-
lation, whorl expansion, and whorl number.
We deduce the following from the ordination
diagram and table of character loadings. (1)
We consider the small-class and large-class
snails to belong to the same species, G. viri-
dulus. With only two exceptions (nos. 5, 20)
the variance along the second axis ranges
from —0.554 to +0.182 for both size classes
showing the same considerable variance in
whorl translation and expansion. The only dif-
ferences between the size classes are clear-
cut differences in size, and the qualitative
differences of deeper lip sinuation and
depressed lip sinuation and depressed cres-

194

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 16. Shell measurements for populations of Guoia viridulis.

No. Whorls
Length (L)

Width (W)
L body whorl

L penultimate
whorl

W penultimate
whorl

W 3rd whorl

L last 3
whorls
L aperture

W aperture
W columellar

shelf
WE

Large Class (D87-1)

Males (N = 7)

4.0—4.5

3.21+0.28
(2.76-3.56)
2.57+0.16
(2.32-2.76)
2.69+0.28
(2.12—2.92)
0.33+0.07
(0.20—0.38)
1.16+0.13
(0.96— 1.28)
0.56+0.08
(0.48—0.68)
3.15+0.26
(2.72—3.48)
1.95+0.19
(1.56-2.12)
1.56+0.17
(1.24-1.76)
0.33+0.04
(0.28—0.40)
0.80+0.03
(0.77—0.84)

Females (М = 5)

4.0—4.5

3.36+0.38
(2.84—3.84)
2.61+0.29
(2.12—2.88)
2.82+0.31
(2.32-3.12)
0.40+0.05
(0.36—0.48)
1.22+0.13
(1.04—1.36)
0.56+0.08
(0.48—0.68)
3.28+0.34
(2.80—3.68)
2.00+0.19
(1.68—2.20)
1.62+0.22
(1.24—1.76)
0.31+0.10
(0.14—0.40)
0.78+0.03
(0.75-0.81)

ANSP 45501
& 98206 (N = 4)

TABLE 17. Percent of variance accounted for by

each principal component (PC).

PC

© D —

%

76.6
13.5

5.7

76.6
90.1
95.8

TABLE 18. Character loading on each PC axis.

. No. Whorls
. Length
. Width

000 —J O O1 BR À ND —

a bh

. Length of body whorl
. Length of penultimate whorl
. Width of penultimate whorl
. Width of 3rd whorl

. Length of last three whorls
. Length of aperture

. Width of aperture

. Width of columellar shelf

Accumulated %

4.5

3.44+0.16

(3.28 —3.60)
2.89+0.19

(2.72-3.12)
2.99+0.11

(2.88—3.12)
0.29=0.02
(0.28—0.32)
1.11+0.04
(1.08—1.16)
0.53+0.04
(0.48—0.56)
3.38+0.17
(3.20—3.52)
2.29+0.12
(2.16—2.44)
1.83+0.12
(1.68—1.96)
0.33=0.05
(0.28—0.40)
0.84+0.02
(0.82—0.87)

cent-columellar callus

Small Class (D87-1)

Males (N = 5)

4.0-4.25

2.12+0.06
(2.02-2.16)
1.66+0.08
(1.54—1.76)
1.70+0.06
(1.62—1.76)
0.24+0.02
(0.22—0.28)
0.80+0.04
(0.76—0.84)
0.44+0.04
(0.40—0.50)
2.05+0.06
(1.96-2.12)
1.26+0.05
(1.20 —1.32)
0.94+0.05
(0.88— 1.00)
0.13+0.02
(0.12—0.16)
0.78+0.03
(0.76—0.82)

Females (N = 6)

4.5

2.48+0.24

(2.24—2.80)
1.86+0.14

(1.68—2.08)
1.99+0.22

(1.80—2.32)
0.28=0.03

(0.24—0.32)
0.90=0.10
(0.78—1.04)
0.51 +0.06
(0.40 —-0.56)
2.41 +0.23
(2.20 —2.72)
1.44+0.11

(1.32—1.60)
1.10+0.09
(0.96 — 1.20)
0.11+0.03
(0.09—0.16)
0.75+0.03
(0.71—0.79)

in the small-class

snails. Note that two shells (nos. 7, 12) where
originally selected as large-class snails based
on visual impression of shape and apertural
characters, yet are clearly small-class snails
when measured. The length variance for
small-class shells was 2.02-2.84 mm, for

Components
р.

—0.724
0.012
0.210

AAA RÁ“

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 195

TABLE 19. Lengths (mm) or counts of non-neural organs and structures of large-class Guola viridulus.
Mean + standard deviation (range). N = number of snails used.

Females Males (N = 1)
Body 5.69+1.29 6.0
(4.36—7.24) М = 4
Digestive gland 2.23+0.49 2.8
(1.80-2.9)N = 4
Gonad 0.89+0.46 1.4
(0.46—1.44)
Total pallial oviduct 2.09+0.21 —
РО (1.80—2.30) М = 4
Bursa copulatrix 0.61+0.11 —
= Bu (0.50—0.76) N = 4
Bu - PO 0.29+0.03 т
(0.27—0.33) М = 4
Duct of bursa 0.12+0.06 —
(0.06—0.18) М = 3
Buccal mass 0.67+0.03 —
(0.64—0.70) М = 3
Mantle cavity 1.51+0.18 2.30
(1.30—1.66) М = 5
Osphradium 0.53+0.09 1.00
= Os (0.46—0.68) N = 5
Gill 13550517 2.00
= (1.14-1.50)N = 5
Os +G 0.40+0.05 0.50
(0.32—0.45) М = 5 (0.26 —-0.36)
No. gill filaments 24.4+2.1 25
(22>27)INi="5
Gf, 0.25+0.09 male + female
(0.12—0.34) N = 8
Gf, 0.37+0.07 male + female
(0.30—0.42) М = 8
Total Gf 0.57+0.17 male + female
= ТСР (0.42—0.74) М = 8
Gin = Wels 0.39+0.08 male + female
(0.29—0.49)
Prostate — 1.80
Seminal vesicle — 1.20
Penis — 3.33 (N = 2)
(3.06 —3.60)

large-class, 2.92-3.84 mm; thus for the spe-
cies there is a 90% increase in size from
shortest to longest mature specimen.

2) We consider the shells at the top of the
ordination clustered around mid-axis 1 as be-
longing to G. fuchsianus. They differ consid-
erably from G. viridulus in that all have 4.0
whorls, relatively narrower 3rd whorls and
shorter penultimate whorls. Only 1 individual
(no. 5) of G. viridulus (7%) grouped with

shells of G. fuchsianus.

External features. The head is shown in Figure

42. The snout is transparent but flecked with
| Spots of melanin pigment on the dorsal snout

and bars of pigment along the tentacles.

There is a small dense eyebrow (Eyb) about
each eye comprised of yellow glands. The
operculum is shown in Figures 43, 44C—F. It
is ovate, paucispiral, with a modest internal
attachment pad.

Mantle cavity. The reflected mantle is shown
in Figure 454. The mantle cavity structures are
typical for taxa of the Neotricula clade. The
opening of the spermathecal duct (Osd) is next
to the pericardium (Pe). The osphradium (Os)
is long and situated mid-gill. The terminal gill
filaments (Gf) are normal length (Tables 19,
20). The length of the longest filaments is 0.57
+ 0.17 mm long. The filaments are not pleated
and lack a pronounced crest (Fig. 45B).

196

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 20. Lengths (mm) or counts of non-neural organs and structures of alcohol-preserved small-class

Guoia viridulus. N = number of snails used.

Body
Digestive gland

Gonad
Posterior pallial

oviduct ( = albumen gland)

Anterior pallial

oviduct (= capsule gland)

Total pallial oviduct
Bursa copulatrix
Mantle cavity

Osphradium
= Os

No. gill filaments
Gf.
Gf,
Total Gf
= TGF
Gio UGE

Prostate

Penis

FIG. 42. Head and penis of Guoia viridulus.

6.3

Females Males (N = 1)

(5.60, 6.98) N = 2
2.75 =

(2.4,

3.1)N = 2

1.10 (N = 1) =

=
—
Es
&
z
[

1.3

2

(1.2, 1.34) N = 2

2.28+0.33 =
(1.9-2.5) М = 3

0.43+0.16 =
(0.25—0.56) N = 3

1.81 1.50
(1.62, 2.00) N = 2

0.67 0.62
(0.58, 0.76) N = 2

1.61 1.24
(1.42, 1.80) N = 2

0.41 0.41
(0.41, 0.42) N = 2

22
(21,

0.23+0.07

23) N = 2
male & female

(0.16—0.30) N = 3

0.30+0.08

male & female

(0.24—0.36) N = 3

0.53+0.13

male & female

(0.40 —0.66) N = 3

0.43+0.03 male & female
(0.40 —0.46) N = 3

Te 1.04

== 2.24

Female reproductive system. The body of an
uncoiled female without head and with kidney
tissues removed is shown in Figure 46A. Mea-
surements of relevant organs are given in Ta-
bles 19, 20. Important features to note are: (1)
The body is not squat, but regularly tubular
(contrast Lithoglyphopsis modesta). (2) The
posterior pallial oviduct (Ppo) does not bend
over the style sac (contrast L. modesta). (3)
The gonad (Go) is posterior to the stomach.
(4) Sperm enter the system at the rear of the
mantle cavity (Emc) through the opening of
the spermathecal duct. The spermathecal
duct does not enter the pericardium (contrast
taxa of the Tricula clade). (5) The pericardium
does not swell out into the mantle cavity. The
bursa copulatrix complex of organs is shown
in Figure 47. (6) The terminal end of the mus-
cular spermathecal duct (Sd) is a thin walled
vestibule (Twv). The spermathecal duct is
short or long; it is that section of duct between

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 197

SB SAauM

FIG. 43. Opercula of Guoia viridulus (A-D) and Guoia fuchsianus (E, F). A, B from large-class specimens;
| С, D from small-class specimens. A, С, E. Outer surfaces; В, D, Е. Inner surfaces.

198

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 21. Radular statistics for Guoia viridulus snails. Mean + standard deviation (range). In mm

except for width of central tooth in рт.

Radular length
Radular width

Total rows of teeth
Rows of teeth forming

Central tooth width

Radular length
Radular width

Total rows of teeth
Rows of teeth forming

Central tooth width

Large Class
Females (N = 4) Males (N = 3)
0.96+0.11 0.94+0.02
(0.84—1.08) (0.92—0.96)
0.12+0.01 0.12+0.01
(0.104—0.132) (0.112—0.128)
70+4 72225
(66—75) (69—74)
10+1.8 9.7+0.06
(8-12) (9-10)
27+3.5 26+2
(24-32) (24—28)
Small Class
0.67+0.05 0.61+0.05
(0.64—0.72) (0.60—0.62)
0.09+0.01 0.08+0.01
(0.08—0.10) (0.07—0.09)
69+3.6 69.7+6.7
(66-73) (64—77)
7.5+0.6 8.7+0.6
(7-8) (8-9)
18+1.6 173312
(16-20) (16-18)

the end of the mantle cavity (Emc) and the
point where the sperm duct (Sdu) arises from
the spermathecal duct. (7) The spermathecal
duct runs directly posterior from the mantle
cavity to become the duct of the bursa (Dbu)
directly posterior to which the thin-walled
bursa copulatrix (Bu) swells as a large oval
sac. (8) The sperm duct (Sdu) is a long con-
voluted duct transporting sperm from the
spermathecal duct to the oviduct close to
where the latter (Opo) enters the albumen
gland (Ppo). (9) The usual seminal receptacle
has been lost. When dissecting living speci-
mens one often observes a bright pink sheen
or glittering within a section of the oviduct
(Ov) just posterior to the juncture of the sperm
duct (Sdu) and oviduct (Osr, Fig. 47). This sec-
tion of oviduct assumes the function of the
seminal receptacle and is called the oviducal
seminal receptacle (Osr). The duct may be
considerably swollen with sperm (Fig. 47D,
E). (10) Posterior to the sperm storage area
(Osr) the oviduct is convoluted (Fig. 47B). (11)
The bursa is posterior to the posterior pallial
oviduct (Bu, Fig. 46). (12) The bursa is short.

Male reproductive system. The body of an un-
coiled male is shown in Figure 48 without

head but with kidney tissue (Ki) left in place.
Measurements of relevant organs are given in
Tables 19, 20. Important features are: (1) The
gonad consists of relatively large lobes drain-
ing into a vas efferens (Ve, Fig. 49). (2) The
seminal vesicle arises from the vas efferens
(Ve) approximately one third of the way pos-
terior from the anterior end of the gonad (Fig.
49). The juncture of the vas efferens and the
beginning seminal vesicle may be swollen
(Fig. 48). The coils of the seminal vesicle may
be ventral to or dorsal to the lobes of the go-
nad. (3) The gonad is posterior to the stom-
ach. (4) The prostate (Pr) overlaps the poste-
rior end of the mantle cavity (Emc). (5) The
penis has a glandular lobe (Fig. 42 Glo) on
the concave edge near the base. (6) The pe-
nis has a stylet (Sty, Fig. 42).

The stylet requires some comment. It is
very much like the stylet observed in Robert-
siella (Davis 8 Greer, 1980). It is corneous
and very fragile. In the living-moving male, it
is very obvious at 50X, projecting from the tip
of the penis (Fig. 42). However, in relaxed
and fixed specimens it is not in evidence.
Rarely at 50X one observed a pin-prick of
light reflecting from the tip of the stylet mostly
retracted into the penis. To observe the stylet
in preserved specimens, the penis is cut from

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 199

и —— 500UN

300UN

FIG. 44. Opercula of Guoia viridulus (C-F) and G. fuchsianus (A, B). C, D from large-class snails; E, F from
| small-class snails. A, С, E. Outer surfaces; В, D, F. Inner surfaces.

| the animal, mounted in water on a slide with
| cover slip, and examined at 400 X (Fig. 50).
The stylet is quickly dissolved when the penis
is placed in Clorox (0.5% Na hypochlorite).
Thus the stylets of Robertsiella and Guoia are
not robust as those found in Stenothyra
(Davis et al. 1986, 1988).

(7) The ejaculatory duct is massive and
highly muscularized (Ej, Figs. 42, 51). It ex-
tends out of the base of the penis and from
the posterior penis along the entire length of
_ the dorsal neck. The circular ejaculatory duct
is slightly left (Fig. 52), central, or to the right
of the snout-neck mid-line (Fig. 51).

Digestive system. The digestive gland covers
the posterior chamber of the stomach. The
paired salivary glands are the standard tricu-
line type. The radular sac does not coil up
over the buccal mass.

The radulae of large-class snails are shown
in Figures 53E-H and 54A—H. Teeth counts
and statistics are given in Tables 21, 22. The
most frequently encountered formula is:

3-1-3; 3(4)-1[2]-3(4); 10-13; 8-10.

The inner pair of basal cusps of the central
tooth are comparatively enormous. The cen-

200 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 22. Cusp formulae for the radular teeth of Guoia viridulus with the percent of the radulae in which
a given formula was found at least once.

Central Teeth Lateral Teeth Inner Marginal Teeth Outer Marginal Teeth
Large Class (N = 5 radulae)

3-1-3 60% 3-1[2]-4 60% 8 8 40%
3-3

3-1-3 20% 4-1[2]-3 60% 9 20% 9 60%
4-3

3-1-4 20% 3-1[2]-3 40% 10 60% 10 40%
4-4

4-13 20% 3-1[2]-5 20% 11 80% 11 20%
3-3

4-1-4 20% 4-1[2]-4 10% 12 60% 12 0
3-3

515 20% 13 40% 13 20%
3-3

5-1-4 20% 14 20% 14 20%
3-3 В

51-5 20% X= 15-16 10.4+2.3

33 San

Small Class (N = 5 radulae)

4-1-4 40% 4-2-3 100% 11 20% 11 0
3-3
3-2-4 80% 12 60% 12 40%
4-1-3 40%
4-3 4-2-4 60% 13 80% 13 80%
4-1-5 40%
3-3 5-2-3 60% 14 100% 14 100%
3-1-5 40% 5-2-4 60% 15 80% 15 60%
4-4
3-1-3 20% 3-2-3 40% 16 60% 16 0
4-3
4-2-4 40%
3-1-4 20% X=*7.40+0.9 6.3+0.6
4-4 4-2-5 20% N = 30 N = 30
3-1-3 20%
4-4
4-1-4 20%
4-4
4-1-3 20%
3-4
4-1-4 20%
-3

“Mean + standard deviation of cusp number for all teeth counted.

tral cusp at the anterior edge of the central 3(4)-1-(4)3; 3(4)-1[2]-3(4); 12-16; 12-15.
tooth is frequently grooved or split. 3-3

The radulae of small-class snails are Comparing both size classes, as one would
shown in Figures 54E-H and 55. Teeth counts expect, the small-class has a smaller radula
and statistics are given in Tables 21, 22. The except for total rows of teeth, which are not
most frequently encountered formula is: significantly different between classes. The

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 201

A

Emc

Osd

Ne

Dai

Gf,
ce Lm
ia Ebv
Sm Ez N

FIG. 45. Cut and reflected mantle to show mantle cavity organs of Guoia viridulus (A). Not all gill filaments

are shown. B. Single gill filament.

morphologies of the teeth of both size classes
are the same.

Nervous system. Measurements are given in
Table 23. The RPG ratio is 0.229 + 0.11, that
is, the pleuro-supraesophageal connective is
concentrated. The pleuro-subesophageal
‚ connective is lost in the fusion of ganglia. The
osphradio-mantle nerve is elongated (>0.12
mm) thus compensating for the low RPG ra-
| tio.

| Remarks

In comparing small- and large-class snails,
no qualitative anatomical differences were
found. The sizes of the bodies and organs
were not significantly different. However, the
small class animals were relaxed with sodium
nembutol prior to fixing in 8% formalin and
grading up to 70% ethyl alcohol. Presumably,
the extended fixed bodies would measure
longer than animals removed living from the
_ Shells and pinned out in a contracted state.
Because the animals of the two size classes
do not differ in details of anatomy, and be-

cause they live in microsympatry, they are
considered the same species.

While the shells are significantly different in
size, the range of variance along the second
principal component is the same for both size
classes indicating that the substantive differ-
ence is only one of size. What could account
for this? Has this phenomenon been seen be-
fore? A similar situation was reported for Tric-
ula xianfengensis Davis & Guo, 1986, in which
large and small class mature snails with the
same anatomy and shell shape were found in
a narrow ditch alongside a kitchen garden in
Xiaguan City, Yunnan Province. In both of
these cases, we suspect that the size classes
resulted from two different cohorts when egg
laying occurred at different seasons. It is pos-
sible that eggs laid late in the season with
growth extending into the cold weather months
resulted in a small size class due to stunting
caused by decreased rate of growth in the fall
and winter. Presumably egg laying and growth
in the late spring and summer would result in
optimal growth. A comparative molecular ge-
netic analysis of these size classes would be
most instructive and desirable.

202 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Ast

FIG. 46. Uncoiled female of Guoia viridulus with head and kidney tissue removed (A). B. Variation in gonad
morphology.

TABLE 23. Lengths of neural structures from four individuals of large class Guoia viridulus. Mean +
standard deviation (range).

Cerebral ganglion 0.31 +0.03 (0.26 —0.32)

Cerebral commissure 0.07 +0.02 (0.06—0.10)

Pleural ganglion
Right (1) 0.15+0.04 gun
Left 0.11+0.01 (0.10—0.12)

Pleuro-supraesophageal 0.08+0.04 (0.04—0.14)
connective (2)

Pleuro-subesophageal 0 0
connective

Supraesophageal 0.12+0.02 (0.10—0.14)
ganglion (3)

Subesophageal 0.12+0.02 (0.10—0.14)
ganglion

Osphradio-mantle nerve 0.15+0.03 (0.12—0.18)

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 203
Bu

FIG. 47. Bursa copulatrix complex of organs of Guoia viridulus. Organs in A are in same orientation as in
_ Figure 46. Figures В, С shows bursa flipped over to show dorsal aspect to reveal origin of elongated and
_ twisting sperm duct (Sdu). Figure D most closely approximates organ position in Figure 46 but with tissue
cleared away to show relationships of ducts relative to end of mantle cavity (Emc) and albumen gland (Ppo).
E. Blown up section of oviduct within which sperm are stored (Osr).

204 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 48. Uncoiled male of Guoia viridulus with head removed.

Guoia fuchsianus (Moellendorff, 1885)

Lithoglyphus fuchsianus Moellendorff, 1885:

169, 170

Lithoglyphus, fuchsianus, Moellendorff, 1888:

140. pl. 4, fig. 5, 5a—b.
Lithoglyphus fuchsianus, Thiele, 1928
Lithoglyphopsis fuchsianus, Yen, 1939

Type locality. Moellendorff, 1885: Hsiang-tan
provinciae sinensis Hunan.

Moellendorff, 1888: Hsiangtan and
Héngshan-hsien; Hunan.

Yen, 1929: Heng-dshou-fu, Hunan

Types. Lectoype: Sl, 4127. Figured by Yen
(1939: pl. 4, fig. 10); paralectotypes: SMF,
4128. Figured here, Figure 36E, F.

Habitat

Xiangjiang River at Baisha, Hengshan
County, 25°58’22”М, 112°45’55’E. Figure 1,

FIG. 49. Male gonad with anterior lobes cut away to
show origin of vas deferens from vas efferens (Ve),
and seminal vesicle (Sv) that lies dorsal to gonad.
Dashed line indicates extent of gonad cut away.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

x

FIG. 50. Tip of penis of preserved Guoia viridulus
as observed under compound microscope at 400X.
Sty, Stylet; Vd, Vas deferens.

FIG. 51. Relationship of ejaculatory duct to mid-line
(x) of snout-neck of Guoia viridulus (from small-
class specimen).

205

FIG. 52. Ejaculatory duct of Guoia viridulus slightly
to left of snout-neck mid-line.

locality 9. Collected by Davis, Chen, 8 Wu,
October 1985, field number D85-75; by Chen
8 Wu, 1986. Snails sympatric with Lithogly-
phopsis modesta on stones at the bottom of
the river in 2.0—2.5 m depth. Catalog numbers
are: D86-B, ANSP 373145; A12661. D85-75,
ANSP 373146; A12662.

Description

Shell. Shells (Figs. 36D-F, 39G-I, 56C—F,
57,58) are small (Table 24), and generally as
described for large class L. viridulus with the
following differences. The outer lip of G. fuch-
sianus is only slightly sinuate. There is a more
pronounced umbilical depression. The inner
lip is separated slightly from the body whorl by
a narrow depressed groove, the basal cres-
cent (Fig. 57, Bc) paved with shell growth in-
crements from body whorl to the inner lip
(Figs. 57, 58). The penultimate whorl of the
species is proportionally shorter than that of
G. viridulus and the third whorl less wide.
Overall, however, the shells have such in-
trapopulation variance that the species are
difficult to differentiate on the basis of shells
alone. Some 10% of the shells from one pop-
ulation are virtually indistinguishable from
shells of the other. An example of variance in
shape for G. fuchsianus is shown in Figure
58A—D, in which camera lucida drawings of
two shells from collection D86-B are illus-
trated with two shells from historic ANSP
98205. Because most of the historic G. fuch-

206 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

D

FIG. 53. Radulae of large size-class Guoia viridulus (E-H) and G. fuchsianus (A-D). A, E. = portions of
radulae. B-C, F-G. Central, lateral and inner marginal teeth; D, H. Outer marginal teeth.

207

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

‘Я :seinpei jo sjuauBas ‘| ‘3 ‘y ‘(1

1)

ee} ¡eurbrew 1810 ‘7 ‘H ‘а ‘4ез} ешблеш лэии! pue ¡esaje] “¡enuao ‘M PD ‘4 ‘9
SNUB/sy9n, BIOND Pue :SNINPHIA BIONH (H-3) ssejo-ezis |ешз pue (3-y) ssejo-azis эбле| jo эвпрен “pg ‘54

; N : qu N В " 4 !
{ AS а \ | |
\ Y As \ м j ‘à у 4 = Г Y 2
> Bu
er ” Wt <
4 И :
» Y

208 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

LA

NT FRENY
ь\ А АА 7
N +f a uf y XV | N |

=.

| z N \

A
LS

ji LV i
Baum.

FIG. 55. Radula of small size-class Guoia viridulus. A. Segment of radula; B, C. Central, lateral and inner

marginal teeth; D. Outer marginal teeth.

sianus had eroded shells so that characters 1,
7, and 8 could not be measured, the multivari-
ate analyses involving all populations of G.
viridulus and G. fuchsianus without these
three characters did not provide differentia-
tion of species or populations. Individuals
were simply spread out along a size gradient.
The historic G. fuchsianus specimens are
larger than field-collected specimens. Still,
basing size on the length of body whorl and
width, this species is somewhat smaller than
G. viridulus (compare Tables 24 and 16).

Anatomy. In 1985, most of the snails were
sexually immature so that complete anatom-
ical data could not be gathered. Organ mea-
surements and counts that could be made are
presented in Table 25. Sufficient observations
were made to demonstrate that Guoia fuch-
sianus is a distinct species, not to be con-
fused with G. viridulus. Only anatomical fea-
tures unique to this species will be presented
here.

(1) The duct ofthe bursa (Dbu, Fig. 59B) is
a considerably swollen duct that narrows be-
fore opening into the capacious bursa (Bu).

(2) The duct of the bursa is comparatively
long: 0.35 mm for G. fuchsianus, 0.12 for G.
viridulus. (3) The sperm storage area (Sr) is a
swelling that bulges out of the sperm duct
(Sdu) where the sperm duct opens into the
oviduct (Ov, Fig. 59). (4) The buccal mass is
longer (0.64—0.70 mm for G. viridulus, 0.70—
0.72 for G. fuchsianus). (5) The central ante-
rior cusp of the central tooth of the radula and
flanking cusps are considerably more poste-
riorly projecting than those of G. viridulus
(compare Fig. 54C, G, K). The anterior cusp
support and cusps of G. fuchsianus form a
prominent triangular-shaped projection poste-
riorly. This, coupled with the elongated central
cusp, causes the tip of the central cusp to
overlap the tooth immediately posterior. A ra-
tio will demonstrate the difference between
taxa. The length from anterior central edge to
posterior tip of the central cusp divided by
width of the anterior central cusp yields 0.75-
0.88 for G. fuchsianus and 0.57-0.71 for G.
viridulus. (See Tables 26, 27 for radular sta-
tistics.) (6) Measurements of neural struc-
tures are given in Table 28. The RPG ratio is
0.42, that is, the pleuro-supraesophageal

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 209

connective is moderately concentrated (con-
trast the concentrated condition in G. viridu-
lus, RPG ratio = 0.23).

The penis of two individuals had a well-de-
veloped stylet but no glandular lobe or pro-
nounced ejaculatory duct. The prostate was
fully developed and penis length was 1.5-1.6
mm long. Because the penis of G. viridulus
with a pronounced glandular lobe exceeded
3.0 mm in length, it is possible that the ones
we observed for G. fuchsianus were imma-
ture, underdeveloped, or, conversely, atro-
phied. We are not convinced that the penis of
G. fuchsianus lacks a glandular lobe or ejac-
ulatory duct.

Remarks

Conchological characters that separate this
species from G. fuchsianus are few and dis-
cussed above. Refer also to the multivariate
analysis presented under G. viridulus.

Wheareas it is clearly desirable to have
more anatomical data, the available data sub-
stantiate the placement of this species in the
genus Guoia. The six anatomical differences

| provided above convince us that we are deal-

ing with a distinct species. The penial data are

not of sufficient quality to permit us to make a

definitive comment except that a stylet is def-
initely present.

Neotricula Davis, 1986

Type Species. Lithoglyphopsis aperta Tem-
charoen, 1971: 103-104, pl. 7, fig. 14.

Tricula aperta (Temcharoen), Davis 1979,
1980.

Type locality. Mekong River at Ban Na on
Khong Island, Laos

Assigned Species. N. aperta (type for genus
assigned in Davis et al., 1986a); N. burchi
(Davis, 1968) [Thailand]; N. cristella (Gredler,
1887); N. dianmenensis Davis & Chen, sp.
nov.; N. duplicata Davis 8 Chen, sp. nov.; N.
lili Chen 8 Davis, sp. nov.; N. minutoides
(Gredler, 1885); N = 7.

Diagnosis. Shells small to medium sized,
ovate conic, smooth. Central tooth of radula
with several anterior cusps (contrast single tri-
angular blade as in Delavaya). The oviduct
runs from the gonad to the pallial oviduct with-
out making a loop or twist. The spermathecal
duct does not enter the pericardium but opens
into the posterior mantle cavity; it is a narrow
duct throughout. The duct of the bursa enters
a U-shaped bend to run into a discrete sem-

TABLE 24. Shell measurements (mm) for populations of Guoia fuschianus. Mean + standard deviation

(range).
Baisha ANSP: 98205, 45961
Males (N = 2) Females (N = 3) Whole Shell (N = 1) Eroded Shell (N = 8)
1. No. Whorls 4.0 4.0 4.0 —
2. Length (L) 2.26 2.79+0.12 2.96 —
(2.48, 2.76) (2.72-2.79)
3. Width (W) 2.20 2.36+0.07 2.32 2.44+0.12
(2.08, 2.32) (2.32—2.44) (2.28—2.60)
4. L body whorl 2.26 2.45+0.02 2.60 2.70+0.12
(2.16, 2.36) (2.44—2.48) (2.56—2.88)
5. L penultimate 0.24 0.23+0.05 0.24 0.30+0.04
whorl (0.20, 0.28) (0.20—0.28) (0.24—0.36)
6. W penultimate 0.90 0.92+0.07 0.86 1.01+0.08
whorl (0.80, 1.00) (0.88—1.00) (0.92—1.12)
7. W 3rd whorl 0.46 0.47+0.02 0.40 —
(0.40, 0.52) (0.44—0.48)
8. L last three whorls 2.60 2.77+0.09 2.92 —
(2.48, 2.72) (2.72—2.88)
9. L aperture 1.68 1.85+0.08 2.00 1.93+0.08
(1.56, 1.80) (1.76—1.92) (1.80—2.04)
10. W aperture 1.38 1.45+0.02 152 1.54+0.08
(1.28, 1.48) (1.44—1.48) (1.40—1.60)
11. W columellar shelf 0.18 0.25+0.02 0.28 0.28+0.04
(0.16, 0.20) (0.24—0.28) (0.20—0.32)
We 0.84 0.85+0.01 0.78 —
| No var. (0.84—0.85)

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 56. Shells of small-class Guoia viridulus (A, B) for comparison with shells of G. fuchsianus (C-F). Shell
A is 2.36 mm long; others printed at same scale. All are from females except E, which is from a male.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 211

TABLE 25. Lengths (mm) or counts of non-neural organs of female Guoia fuschianus.
tle 3 x
Body 4.9
Digestive gland 2.1
Bursa copulatrix 0.68 0.54 0.61
Duct of bursa 0.34 0.36 0.35
Buccal mass 0.70 0.72 0.71
Mantle cavity 2.0 1.70 2.20 1.97+0.25
Gill 1.8 1.50 2.00 1.77+0.25
Osphradium 0.9 0.7 0.84 0.81+0.10
Osphradium = gill 0.50 0.47 0.48
No. gill filaments 25 25 24.0+1.7
Gf, 0.16 0.22 0.19
Gf, 0.48 0.46 0.47
Total Gf 0.64 0.68 0.66
Habitat

FIG. 57. SEM photograph of a shell of Guoia fuch-
sianus.

inal receptacle in most species. The duct of
the seminal receptacle may shorten or be-
come lost. The spermathecal duct joins the
duct of the bursa (contrast joining the bursa
as in Halewisia). A slender duct, the sperm
duct, connects the U-shaped duct to the ovi-
duct.

Neotricula cristella (Gredler, 1887)

Paralectotypes. SMF 4242; plate 4, fig. 3, in
Yen, 1939

Type locality. Kiangshi Province, “in Quell-
wasser.”

Synonymy. Hydrobia cristella Gredler, 1887
Tricula cristella, Yen, 1939
Neotricula cristella, this paper

Material for this paper was collected from
Mojingtai, Hengshan Mountain, Nanyue
Town, Hengshan County, Hengyang Prefec-
ture; 27°15’N, 112°39’13’E; Figure 1 site 1.
Snails came from a small stream 0.4 km down
the mountain road from the Mojingtai Hotel
towards the Banshan Ting Temple. Collec-
tions number = D85-73; the collection was
made by Dr. Chen Cui—E, 28 Sept. 1985.

Depository

Specimens are deposited in ZAMIP,
M0001; in ANSP 368774 and A12146.

Description

Shell. Shells are small, narrowly ovate-conic,
of 5.0 to 5.5 whorls (Figs. 60H-L, 61A-D).
Lengths range from 2.40 to 2.84 mm (Table
29). The aperture is ovate; there is no umbi-
licus. The whorls at the suture are smooth
(not crenulated). SEM analyses reveals a
faint trace of spiral microsculpture on the
adapical surface of the whorls at the suture.
The inner lip is arched and widely separated
from the body whorl, the distance increasing
from abapical to adapical. Thus, the adaptical
end of the aperture is widely separated from
the body whorl. The adapical end of the ap-
erture has a wide notch; there is no internal
notch groove; there is no sinus. There is no
abapical spout. In side view, the outer lip is
slightly sinuate; it is not scooped forward. The
inner lip lacks nodes, teeth or notches. The
inner lip is thin throughout. In side view, the
inner lip is straight on some shells, angled to
form a deflection angle in others. Within the

212 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

FIG. 58. Illustration of four shells of Guoia fuchsianus aided by camera lucida. Shells A, D from ANSP 98205

(historic collections); shells B, C from D86-B.

body whorl, the columella is smooth. There is
a slight varix on some shells, no varix on oth-
ers. In apertural view, the lip projects beyond
the base of the shell 0.34 + 0.06 mm. SEM
analysis shows the protoconch to be slightly
wrinkled (Fig. 61C, D).

External features. The head is not pigmented.
There are no granules about the eyes (Fig.
62). There is a patch of white granules where
the mantle margin meets the neck. The oper-
culum is corneous, paucispiral and appears
to have two layers, a larger outer layer and a

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 213

TABLE 26. Radular measurements (mm) and counts for Guoia fuchsianus. Mean + standard deviation
(range). In mm except for width of central tooth in am.

Radular length

Radular width

Total rows of teeth

No. rows of teeth forming

Central tooth width

Female (N = 5) Male (N = 2)
0.90+0.05 0.80
(0.84 —0.98) (no variation)
0.11+0.003 0.11
(0.11—0.12) (0.10—0.11)
72.6+4.3 68.5
(66—78) (68—69)
23.8+1.1 22.5
(23-25) (22-23)
23.2+2.3 22
(22-26) (no variation)

TABLE 27. Cusp formulae for the radular teeth of Guoia fuschianus with the percent of radulae in which
a given formula was found at least once. N = 3 radulae.

Central Teeth Lateral Teeth

4-1-4 66% 3-1[2]-4 100%
2-2
4-1[2]-3 66%
3-1-3 33%
-3 3-1[2]-3 66%
4-1[2]-4 33%
2-1[2]-4 33%

Inner Marginal Teeth

Outer Marginal Teeth

10 33% 10 33%

11 66% 11 66%

12 100% 12 100%
13 33%

114-07 11.3+0.8

N = 26 N = 30

*Mean + standard deviation of cusp number for all teeth counted.

TABLE 28. Lengths (mm) of neural structures of
Guoia fuchsianus. Mean (data). N = 2

Cerebral ganglion
Cerebral commissure
Pleural ganglia
Right (1) 0.11 (0.10, 0.12)
Left 0.10 (no variation)
Supraesophageal 0.16 (0.12, 0.20)
connective (2)
Subesophageal connective 0

Supraesophageal 0.11 (0.10, 0.12)
ganglion (3)

Subesophageal ganglion
Osphradio-mantle nerve
RPG ratio (2 + 1+2+3)

0.29 (0.28, 0.30)
0.12 (no variation)

0.10 (no variation)
0.12 (0.10, 0.14)
0.42 (0.38, 0.46)

narrower inner layer (Fig. 61E, F). The inter-
nal attachment pad is prominent but only 40
to 50% the width of the operculum.

Mantle cavity. Mantle cavity organs are
shown in Figure 63A. Organ measurements
and counts are given in Table 30. The osphra-
dium is slightly anterior to mid-gill; it is oval
and small. There are 13 to 15 gill filaments, of

which only seven or eight are fully developed,
that is, with both Gf,, and Gf, elements prom-
inent. The anterior five filaments are widely
separated and without the Gf, part. Gf, is
long. The longest gill filaments are 0.38 +
0.09 mm long. The largest gill filaments in lat-
eral view are modestly domed (Fig. 63B). The
pericardium (Pe), opening of the kidney into
the mantle cavity (Oki) and the opening of the
spermathecal duct (Sd) into the mantle cavity
are shown in relationship to each other (Fig.
63A).

Female reproductive system. An uncoiled fe-
male without head and with kidney tissue re-
moved is shown in Figure 64. Measurements
of organs are given in Table 30. Important
features are: (1) The gonad (Go) is posterior
to the stomach, is small and consists of few
lobes. (2) The bursa copulatrix (Bu) is round
and situated directly posterior to the albumen
gland (Ppo). (3) The albumen gland is of nor-
mal size. (4) The bursa copulatrix complex of
organs is shown in Figure 65. Organs in Fig-
ure 65A have the same orientation as in Fig-
ure 64. The bursa is short. (5) The duct of the
seminal receptacle (Dsr) is a U-shaped con-

214

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

Ov

0.5mm

FIG. 59. Aspects of female reproductive system of Guoia fuchsianus. A. Female reproductive system
oriented in same position as in Figure 46A. Anterior pallial oviduct not shown. B. Bursa of A turned 180° over,
as indicated by the arrow, to show the relationships of ducts attaching to the bursa (Bu) and oviduct (Ov).

tinuation of the duct of the bursa (Fig. 65B, C).
К turns to the right side and tucks dorsal to the
bursa. (6) The duct of the seminal receptacle
(Dsr) is long. The seminal receptacle (Sr) is a
comparatively minute bulb. (7) The spermath-
ecal duct (Sd) is short and opens into the pos-
terior end of the mantle cavity (Emc). (8) The
spermathecal duct opens into the bottom of

the U-shaped bend formed by the duct of the
bursa at a point opposite the opening of the
duct of the seminal receptacle (Fig. 65A, B).

Male reproductive system. An uncoiled male
snail without head and with kidney tissue re-
moved is shown in Figure 66. Measurements
of organs are given in Table 30. Important

THE POMATIOPSIDAE OF

FIG. 60. Shells: Neotricula lili, A, B. Para
3.12 mm; others printed to same scale.

HUNAN, CHINA (GASTROPODA: RISSOACEA) 215

types. N. minutoides, C-G. N. cristella, H-L. Length of shell A is

216 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

DA

FIG. 61. SEM photographs of shells (A-D) and opercula (E, F) of Neotricula cristella. C, D. Details of apical
whorls. In E, F, inside surface of opercula shown on right side.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

TABLE 29. Shell measurements (mm) for male and female Neotricula cristella. Mean + standard

deviation (range). N = number measured.

Females Males
Whorls 5.0-5.25 (N = 4) 5:5 (М = 2) 5.0 (N = 3)
Length (L) 2.64+0.10 2.78 2.37+0.02
(2.52—2.76) (2.72-2.84) (2.36—2.40)
Width (W) 1.33+0.09 1.38 1.21+0.08
(1.24—1.44) (1.36—1.40) (151221528)
L body whorl 1.85+0.11 1.86 1.61+0.02
(1.76—2.00) (1.84—1.88) (1.60—1.64)
L penultimate whorl 0.40+0.01 0.45 0.36+0.03
(0.38—0.40) (0.40—0.50) (0.34—0.40)
W penultimate whorl 0.90+0.02 0.94 0.83+0.03
(0.88—0.92) (0.88—2.00) (0.80—0.86)
L last three whorls 2.46+0.11 2.56 2.20+0
(2.36 —2.60) (2.52—2.60) —
L aperture 1.21+0.08 1.14 1.07+0.06
(1.16-1.32) (1.12-1.16) (1.00—1.12)
W aperture 0.81+0.05 0.80 0.69+0.02
(0.76—0.88) — (0.68—0.72)
Tip apical whorl (W) 0.14+0.02 — =
(0.12—0.16) N = 6* — —
Diameter 1st whorl 0.28+0.02 — (0.26—0.30) N = 6*
x 0.37+0.02 0.34+0.06
(0.36—0.40) М = 3* (0.28—0.40)

————-—-——_Ц_—_-ЦЦцЦ-оцЦ-Ь——о—]—]о6—]———5---------Ц-цСд д

* all whorl classes

“distance from base of body whorl to abapical tip of aperture

features are: (1) The gonad (Go) is posterior
to the stomach. (2) The prostate (Pr) overlaps
the posterior end of the mantle cavity (Emc).
(3) The seminal vesicle (Sv) arises from the
vas efferens (Ve) about mid-gonad. (4) The
seminal vesicle is coiled lateral to the gonad,
like a spring or coils in a knot dorsal to the
gonad posterior to the stomach. It does not
continue onto the stomach. (5) The anterior
vas deferens (Vd,) leaves the prostate at the
posterior end of the mantle cavity. (6) The
penis is simple but with a very elongated pe-
nial filament. (Pf, Fig. 67). (7) The vas defer-
ens is highly coiled passing through the cen-
ter of the penis. There is no ejaculatory duct in
the base of the penis or in the neck. (8) The
orientation of the base of the penis (Bp) to the
snout-neck midline (x) and the posterior edge
of the eye bulges is shown in Figure 62. The
long axis of the penial base varies from 70° to
90° from “x” as shown.

Digestive system. The digestive gland covers
the posterior chamber of the stomach of fe-
males but is posterior to the stomach of
males. Radular statistics are given in Tables

31 and 32. There are 99 + 3.4 rows of teeth
along a radula 0.54 mm long. The most fre-
quently encountered formula is

3(2)-1-(2)3; 3-1[2]-3(4); 14-17; 12-15.
3(2)-(2)3

SEM photographs of radulae and teeth are
given in Figures 68, 69. Central teeth are fea-
tured in Figure 68C—F; they are typical of the
generalized triculine type. The morphology of
the entire lateral tooth is shown in Figure 68A,
B. The one unusual feature seen in the radula
of this species involves the dominant cusp of
the lateral tooth, the “1” of the 3-1-3; it is
deeply divided, almost forming two cusps.
(Fig. 68A—F, 69A—C).

Nervous system. Measurements are given in
Table 33. The RPG ratio of 0.36 shows that
the dorsal nerve ring is moderately concen-
trated.

Remarks

Conchologically, this species differs from
all other Neotricula by having a shell with (1)

218 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

А
1.0mm

FIG. 62. Head of a male snail of Neotricula cristella
showing the orientation of long axis of the penial
base (Bp) to snout-neck mid-line x.

the inner lip widely separated from the body
whorl, and (2) the adapical end of the aper-
ture widely separated from the body whorl.
The multivariate analysis involving all spe-
cies of Triculinae with shells closely resem-
bling those of Tricula or Neotricula indicates
different affinities depending on the analysis
used. The phenogram based on distance co-

efficients (Fig. 153) shows linkage with Gam-
matricula chinensis. The differences are: (1)
The sutures of N. cristella are smooth; those
of G. chinensis are crenulated. (2) The adapi-
cal aperture of the former has a notch; of the
latter, no notch. (3) The inner lip of the former
is thin, of the latter, thick. (4) The inner lip of
the former is widely separated from the body
whorl, of the latter, narrowly separated. (5)
The adapical end of the aperture of the former
is widely separated from the body whorl, of
the latter, slightly separated. (6) There is no
abapical outer lip deflection angle in the
former, a slight one in the latter.

The minimum spanning tree (Fig. 154)
based on distance coefficients yields a differ-
ent closest relationship, one with Tricula hud-
iequanensis. However, (1) T. hudiequanensis
has a medium length shell (N. cristella has a
small shell). (2) The adapical aperture of the
former does not have a notch. (3) The outer
lip of the former, in side view, is scooped for-
ward; it is straight in the latter. (4) The inner lip
of the former is thick; it is thin in the latter. (5)
The inner lip and adapical aperture are not as
greatly separated from the body whorl in the
former as in the latter.

The most distinguishing anatomical charac-
ter (considering only species of Neotricula) is
the long penial filament. This species shares
with N. /ilii the character-state of having very
many rows of teeth on the radula; with N. min-
ima, the state that the digestive gland covers
the posterior chamber of the stomach. The
U-shaped continuation of the duct of the
bursa into the duct of the seminal receptacle
is shared with N. burchi of northwest Thailand
and N. aperta of the lower Mekong River.

Neotricula dianmenensis Davis & Chen,
sp. nov.

Holotype: ZAMIP-M0034, Figure 70A.
Paratypes: ANSP 373143, A12659; ZAMIP
MOO04. Figure 70B—D; Figure 71A, B.

Type Locality: Jiepai Village, Dianmen Town,
Hengshan County, Hengyang Prefecture.
27°15'16’N, 112"33'31”E. Figure 1, Sie
Field collection D85-80.

Collection Date: October 1985.
Etymology: Named for the town of Dianmen.

Habitat

300 m above sea level. Snails were col-
lected from a small stream 20-25 cm wide and

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 219

FIG. 63. A. Mantle cavity structures of Neotricula cristella showing the relationship of the gill to pericardium
(Pe) and openings of the kidney (Oki) and spermathical duct (Sd). B. Single gill filament

Emc

Apo

Ma

FIG. 64. Uncoiled female Neotricula cristella with head and kidney tissue removed.

220 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 30. Lengths (mm) or counts of non-neural organs and structures of Neotricula cristella. Mean +
standard deviation, (range).

Females (N = 5) Males (N = 3)
Body 4.60+0.17 4.17+0.32
(4.34—4.76) (3.80—4.36)
Gonad 0.62+0.08 1.03+0.21
(0.54—0.70) (0.80— 1.20)
Digestive gland 2.06+0.17 1.93+0.31
(1.90 —2.30) (1.60 —2.20)
Posterior pallial oviduct 0.86+0.06 —
(= albumen gland) (0.76—0.90)
Anterior pallial oviduct 0.96+0.09 —
(= capsule gland) (0.90—1.10)
Total pallial oviduct 1.82+0.13 —
= OV (1.66—2.00)
Bursa copulatrix 0.38+0.05 =
= BU (0.34—0.46)
Duct of BU 0.19+0.05 —
(0.14—0.26)
BU = OV 0.21+0.04 —
(0.17—0.20)
Seminal receptacle 0.10+0.02 —
(0.08-0.12)
Duct of seminal 0.19+0.04 —
receptacle (0.16—0.24)
Mantle cavity 1.11+0.28 1.00+0.07
(0.76—1.44) (0.96—1.08)
Gill (G) 0.92+0.28 0.85+0.05
(0.60—1.28) (0.80—0.90)
Osphradium (OS) 0.23+0.08 0.24+0.02
(0.16—0.36) (0.22—0.26)
OS - @ 0.26+0.08 0.29+0.03
(0.18—0.37) (0.26—0.33)
No. of filaments 13.6+0.9 13.3+1.2
(13-15) (12-14)
Gf, 0.23+0.06 —
(0.16—0.30)
Gf, 0.16+0.03 =
(0.12—0.20)
Total Gf = TGF 0.38+0.09 —
(0.26—0.50)
СЪ = TIGE 0.61+0.04 —
(0.56—0.65)
Prostate — 0.81+0.02
(0.80—0.90)
Seminal vesicle — 0.38+0.07
(0.40—0.44)
Penis — 1.53+0.18
(1.40—1.74)
Buccal mass 0.50+0.06 —

(0.44—0.56)

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 221

0.5mm

FIG. 65. Details and variation of bursa copulatrix complex of organs of Neotricula cristella. Figure A is in
same orientation as in Figure 64. B. Bursa removed to show entire seminal receptacle (Sr) and duct (Dsr).
C. Bursa rotated to show interconnection of Dbu, Dsr and the spermathical duct (Sd).

FIG. 66. Uncoiled male of Neotricula cristella without head or kidney tissue.

222 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 31. Radular statistics for Neotricula TABLE 33. Lengths (mm) of neural structures of

cristella. Mean + standard deviation (range). N = Neotricula cristella. Mean + standard deviation
number used. In mm except for width of central (range). N = 3.
tooth in am. SS == —
Cerebral ganglion 0.23+0.06 (0.16—0.28)
Males and females (N = 4) Cerbral commissure 0.05+0.01 (0.04—0.06)
Shell length 2.58+0.30 (2.16—2.88) Pleural ganglion
Radular length 0.54+0.05 (0.48—0.59) Right (1)* 0.08+0.02 (0.06—0.10)
Radular width 0.06+0.002 (0.056—0.060) Left 0.10+0.04 (0.06—0.14)
Total rows of teeth 99 +3.4 (95-103) Pleuro-supraesophageal
No. rows of teeth 236.9 (16—32) connective (2)* 0.11+0.01 (0.10—0.12)
forming Pleuro-subesophageal
Central tooth width 12.9+0.8 (12.3-14) connective 0.09+0.06 (0.02—0.14)
aa ee Ten oe” «a SUpraesoohageal 0.11+0.01 (0.10—0.12)

ganglion (3)*
Subesophageal ganglion 0.11+0.01 (0.10—0.12)

Osphradio-mantle nerve 0.05+0.03 (0.02—0.08)

RPG ratio* = 0.36+0.01 (0.36—0.38)
py E)

10-15 cm deep. The flow was slow; the water
was clean and cool. The stream flows down
from Nanyue Mountain. The bottom of the
stream was paved with small rocks, sand, and
leaves. At stream side was short, scrubby
vegetation. There were some 50 snails per
stone.

Description

Shell. The shells are small, ovate-conic, of 5.0
to 5.5 whorls (Figs. 70A-D, 71A, B). Because
the apices of most mature snails are eroded,
FIG. 67. Penis of Neotricula cristella. it is not possible to be precise about lengths;

TABLE 32. Cusp formulae for the radular teeth of Neotricula cristella with the percent of radulae in which
a given formula was found at least once.

Central Teeth Lateral Teeth Inner Marginal Teeth Outer Marginal Teeth
3-13 93% 3-1[2]-3 79% 11 — 7%
= 3-1[2]-4 21% ER ds
2-1-2 21% 13 14% 71%
ee CONES) ees 14 43% 79%
= 9% 4-1[2]-4 7% 15 71% 50%
16 71% =
17 50% =
18 21% —
X — 1527 13.6+1.2
N = 138 N = 127

“Mean + standard deviation of cusp number for all teeth counted.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

WE 7A MES

de

FIG. 68. Radula of Neotricula cristella. See text for details.

223

224 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

В 44
АА et à / eu

FIG. 69. Radula of Neotricula cristella. See text for details.

225

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

FIG. 70. Shells of: Neotricula dianmenensis A-D. A. Holotype. Neotricula duplicata E-I from D85 collections.
F. Holotype. Neotricula lilii J-L. J. Holotype. A = 3.32 mm; other shells printed at same scale. Shells not

designated as holotypes are paratypes.

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

FIG. 71. SEM photographs of shells (A, В) and opercula (С, D) of Neotricula dianmenensis. D. Inner surface.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 227

TABLE 34. Shell measurements (mm) for Neotricula dianmenensis. Mean + standard deviation (range).
All except two shells were eroded, therefore lengths are less than for entire specimens. e = eroded.

*, probably a male.

No. Whorls

Length (L)

Width (W)

L body whorl

L penultimate
whorl

W penultimate
whorl

W 3rd whorl

L last three whorls

L aperture

W aperture

X

y

Shells used for dissections
Males (1)

Females (3)
3e

2.90+0.20
(2.76-3.12)
1.50+0.08
(1.44—1.60)
2.02+0.12
(1.92—2.16)
0.54+0.10
(0.48—0.62)
1.10+0.10
(1.04—1.20)

2.90+0.20
(2.76-3.12)
1.47+0.08
(1.40—1.56)
0.91+0.02
(0.88—0.92)
0.42+0.05
(0.40—0.48)

5.5
2.72

1.36
1.76
0.40
0.96

Holotype

4e
3.32

1.64
2.20
0.52
1.16
0.80
3.12
1.56
0.96
0.50

0.16

Types

Paratypes (5)
2.0—4.0e

2.89+0.25
(2.64—3.20)
1.50+0.06
(1.44—1.56)
1.96+0.08
(1.84—2.04)
0.52+0.05
(0.46—0.56)
1.08+0.06
(1.00—1.16)
0.77+0.05
(0.72—0.80) N = 3
2.85+0.19
(2.64—2.96) М = 3
1.43+0.08
(1.321452)
0.91+0.03
(0.88—0.94)
0.43+0.08
(0.32—0.52)
0.09+0.05
(0.04—0.16)

*Small
Paratype (1)

5.0
2.76

1.28
1.68
0.44
0.96
0.68
2.40
1.26
0.80
0.36

0.08

FIG. 72. Uncoiled female Neotricula dianmenensis with head and kidney tissue removed.

228 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 35. Dimensions (mm) or counts of non-neural organs and structures of Neotricula dianmenensis.
N = number of snails used. L = length. Mean + standard deviation (range).

o 5 EEE

Females (N = 3) Males (N = 1)
Body L. 4.9+0.53 3.9
(4.4—5.4)
Digestive gland L 1.99+0.36 1:9
(1.76—2.4)
Gonad L. 0.75+0.08 15
(0.70—0.84)
Pallial oviduct L. 1.58+0.16 =
РО (0.44—0.56)
Вигза copulatrix L. 0.50+0.06 —
= BU (0.44—0.56)
Duct of bursa L. 0.18 —
(0.16, 0.20) N = 2
BU + PO 0.32+0.07 —
(0.26—0.40)
Seminal receptacle L. 0.15+0.01 —
(0.14—0.16)
Duct of seminal 0 to 0.12 —
receptacle L. (see text)
Buccal mass L. 0.55+0.04
(0.52—0.60)
Mantle cavity L. 1.41+0.20 1.26
(1.20—1.60)
Osphradium L. 0.35+0.05 0.34
= Os (0.30—0.40)
Gill L. 1.23+0.21 1.06
= @ (1.00—1.40)
OC 0.29+0.03 —
(0.26—0.32) N = 4
No. filaments 18.3+1.5 ut
(17—20)
Gf, L 0.25 (males & females)
(0.20, 0.30) N = 2
GmE 0.21 М = (males & females)
Total Gf L. 0.46N = 2
NICE no variation
GR - СЕ 0.54 (males & females)
(0.44, 0.65) N = 2
Prostate L. — 1.0
Seminal vesicle L. — 0.70

Penis L.

naaa

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 229

Omc

FIG. 73. Details and variation of bursa copulatrix complex of organs of Neotricula dianmenensis. A, same
orientation as in Figure 72. B—D. Bursa complex flipped over to show where seminal receptacle enters the

base of bursa, and variation.

we can give no data on apical whorl charac-
ters. Lengths probably range from 2.76 to
3.44 mm (Table 34). There is possibly sexual
dimorphism, with the males smaller. The ap-
erture shape is distorted due to the pro-
nounced aperture beak. There is a narrow but
pronounced umbilicus. The whorls at the su-
ture are smooth (not crenulated). SEM anal-
ysis reveals a trace of spiral microsculpture
below the sutures (also seen at 50X). There is
an angulation of the inner lip just abapical to
the pronounced beak tubercle. The inner lip is
fused to the body whorl (most shells). The
inner lip is slightly separated from the body
whorl opposite the beak tubercle (some
shells). The adapical end of the aperture is
fused to the body whorl; there is a pro-
nounced apertural beak some 0.26 mm long

with a wide opening into the interior (0.12—
0.16 mm wide gap); there is no internal notch
groove. There is an apertural sinus. In side
view, the outer lip is straight; it is slightly
scooped forward. Facing the inner lip in side
view, there is a strong lip deflection angle.
There is no varix. In apertural view, the abapi-
cal lip extends beyond the base of the shell
0.43 + 0.08 mm.

The whorls are slightly convex and the su-
tures sharply defined but shallow.

External features. Details of the head are not
available. The operculum (Fig. 71C,D) is dis-
tinctive for its long, thin shape. The width to
length ratio is 0.44 + 0.03. As seen in Figure
71D, there may be an easily detached exter-
nal layer. The internal attachment pad is pro-

230 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Pst

Di

Ast

A KA
0.5 mm

FIG. 74. Uncoiled male of Neotricula dianmenensis without head or kidney tissue. Some lobes of gonad cut
away to show seminal vesicle that coils dorsal to gonad. Dashed line shows anterior limit of gonadal lobes.

nounced (Fig. 71D). The shape of the oper-
culum reflects the inner lip angulation.

Mantle cavity. Measurements and statistics
are given in Table 35. Mantle cavity structures
and relationships are normal for those of
Neotricula. The osphradium is mid-gill; it is
short. Gill filaments are normally developed;
Gf, is normal length. Gf, does not have a
pronounced crest.

Female reproductive system. The body of an
uncoiled female without head and with kidney
tissue removed is shown in Figure 72. Mea-
surements of organs are given in Table 35.
Important features are: (1) The gonad is pos-
terior to the stomach; it consists of few lobes.
(2) The bursa (Bu) is clearly seen posterior to
the posterior pallial oviduct (Ppo). (3) Sperm
enter the system at the posterior end of the
mantle cavity (Emc) by passing into the sper-
mathecal duct (Sd) that bypasses the pericar-
dium. The spermathecal duct runs a short dis-
tance to open into the bursa (Bu, Fig. 73A)
dorsal to the duct of the bursa (Dbu). (4) The

sperm duct (Sdu) runs from the bursa (Bu) to
the oviduct (Ov); there is no duct of the bursa.
(5) The seminal receptacle (Sr) arises from
the spermathecal duct where the latter joins
the bursa (Fig. 73B—D). (6) The duct of the
seminal receptacle varies in length from 0 (Sr
fused to the spermathecal duct, Fig. 73C) to
moderately long (Fig. 73D). (7) The oviduct
runs from gonad to pallial oviduct without
making a loop or twist. (8) The oviduct opens
into the pallial oviduct close to the posterior
end of the pallial oviduct. (9) The bursa cop-
ulatrix is round and short. (10) The length of
the albumen gland is standard (based on one
individual for which this could be discerned,
ratio of 0.47).

Male reproductive system. The body of an un-
coiled male is shown in Figure 74 without
head and with kidney tissue removed. The
anterior lobes of the gonad were removed to
reveal the seminal vesicle. Measurements
are given in Table 35. Important features are:
(1) The gonad consists of numbers of finely
divided lobes that drain into a vas efferens

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 231

x

FIG. 75. The penis (B). (A), orientation of the base
of the penis (Bp) of Neotricula dianmenensis to the
snout-neck mid-line (x).

TABLE 36. Radula statistics for Neotricula dian-
menensis. Mean + standard deviation (range). In
mm except the central tooth width in um. N = 4.
Shell lengths not available.

Radular length 0.45+0.06 (0.36 —0.48)

Radular width 0.08+0.002 (0.76—0.080)

Total rows of 58.8+4.2 (56—65)
teeth

No. rows of teeth 1.90+3.6 (14-22)
forming

Central tooth 14.1+0.6 (13.4- 14.9)
width N = 11

(Ve, Fig. 74). (2) The posterior vas deferens
arises from the vas efferens posterior to mid-
gonad and immediately coils as the seminal
vesicle (Sv) that is dorsal to the lobes of the
gonad. (3) The seminal vesicle coils posterior
to the stomach. (4) The gonad is entirely pos-
terior to the stomach. (5) The prostate is mas-
sive; it overlaps the posterior end of the man-
tle cavity. The posterior prostate overlaps the
entire style sac (Sts). (6) The penis is simple,
without lobes; it has a papilla (Fig. 75B). (7)
No ejaculatory duct was found. (8) The shaft
of the penis arises from the neck to the right of
the snout-neck mid-line (x, Fig. 75A) at an
angle of 22-23".

Digestive system. The digestive gland is pos-
terior to the stomach (only slight overlap on
the posterior chamber (Figs. 72, 74). The rad-

TABLE 37. Cusp formulae for the radular teeth of Neotricula dianmenensis with the percent of radulae (N
= 4) in which a given formula was found at least once.

Central Teeth Lateral Teeth

2-1-2 50% 3-1-4 75%
Fe 4-1-3 50%
3-1-3 25% 4-1-2 25%
2-3
nie E 2-1-3 25%
2-2 2-1[2]-2 25%
2-1-3 25% 31.2 25%
al 3-1-3 25%
3-1-4 25%

Outer Marginal Teeth

Inner Marginal Teeth (N = 3)
Y 25% 7 0
8 25% 8 0
9 25% 9 33%
10 25% 10 33%
11 100% 11 66%
12 100% 12 100%
13 75% 13 66%
14 25% 14 0
y E Er 11612
N = 38 30

“Mean + standard deviation of cusp number for all teeth counted.

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 76. Radula of Neotricula dianmenensis. A, E. Segments of the radula. B-D, F-H. Central, lateral and
inner marginal (C, D, H) teeth. See text for details.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 233

FIG. 77. Radula of Neotricula dianmenensis. Outer marginals are featured in C, E, F. Dominant cusp of
lateral tooth (i.e. the “1” of the 3-1-2) is either massive (D) or bifurcated (B).

TABLE 38. Lengths (mm) of neural structures of Neotricula dianmenensis. Mean + standard deviation
(range). N = 4.

Cerebral ganglion 0.23+0.01
Cerebral commissure 0.07+0.01
Pleural ganglia
Right (1) 0.13+0.01
Left 0.12— (no variation)
Pleuro-supraesophageal 0.13+0.04 (0.10—0.18)
connective (2)
Pleuro-subesophageal 0.08 N = 2 (0.06—0.10)
connective
Supraesophageal ganglion (3) 0.11+0.01 (0.10—0.12)
Subesophageal ganglion 0.12—
Osphradio-mantle nerve 0.04+0.01 (0.02 —0.04)

RPG ratio = 2 + 1+2+3 0.35+0.01 (0.34—0.36)

234 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 78. Paratypes of Neotricula duplicata from D87-2. Shell in Figure A is 3.52 mm long; other shells printed
at same scale.

ula sac does not coil dorsal to the nerve ring.
The radula is shown in Figures 76, 77; radular
statistics are given in Tables 36 and 37. The
radula is typical Triculini. The enlarged cusp
of the lateral tooth (the “1” of 3-1-2, etc.) is
frequently bifid or massive (Figs. 77A, B, D).
The typical radular formula is:

2(3)-1-(3)2, 3(4)-1[2]-2 to 4, 10-13, 10-13.
2-2

Nervous system. Standard Triculinae. Mea-
surements are given in Table 38. The RPG
ratio shows the supraesophageal connective
to be moderately concentrated.

Remarks

Conchologically, N. dianmenensis is most
similar to N. duplicata and N. lili (Figs. 153,
154). However, it differs from the other two by
having a distorted aperture shape (Table 2,
char. 3), an angled inner lip (char. 13), anda
pronounced beak tubercle (char. 24). Addi-

tionally, it differs from N. duplicata by having a
clearly open umbilicus (char. 4), an adapical
apertural sinus (char. 11), a sinuate outer lip
(char. 14), and a thin inner lip (char. 19).

Anatomically the greatest similarity is to N.
duplicata (Fig. 157), but it differs from that
species in seven characters (15% of the 46
comparable characters, Table 80). Neotric-
ula. dianmenensis has few rows of teeth
whereas N. duplicata has many (char. 42).
The gonad of the former is posterior to the
stomach; it overlaps the stomach in the latter.
The spermathecal duct is comparatively long
in the former, short in the latter (char. 19).
There is no duct of the bursa in the former; it
is long in the latter (char. 16). The osphradium
is long in the former, short in the latter (char.
9).

Neotricula duplicata Davis & Chen
Sp. nov.

Holotype. ZAMIP-M0035, Figure 70A.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 235

ATITÉE

AR

FIG. 79. SEM photographs of shells of Neotricula duplicata from D85-79. B, D show the abapical lip

deflection. C, E. Enlargements of apical whorls.

Paratypes. ANSP 373152, A12668; 373151,
A12667; ZAMIP M0003; Figure 70E, G-I; Fig-
ure 78B—E.

Type Locality. Huang-sha-ping Village, Jiang-
nan Town, Anhua County, Yiyang Prefecture.
28°21'24"E, 110%15'11'N. Figure 1, site 4.
Collection numbers D85-79, 5 October 1985,
Davis, Hoagland 8 Chen; D87-2, 18 March
1987, Davis 4 Chen.

Etymology. Named for the duplication of an-
atomical research effort on specimens col-

lected from this population in both 1985 and
1987, not realizing in 1987 that the population
had been dissected and analyzed in 1985.

Habitat

Some 400 m from Zijiang River, a small
stream flowed from a hill through rice fields.
The stream was about 10 cm deep, 14 cm
wide with a mud substratum. Snails were col-
lected in the upper narrow shaded part of the
stream from under stones. Associated mol-
luscan fauna: Gyraulus sp.

236 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

FIG. 80. SEM photographs of shells of Neotricula duplicata from D87-2. C-E. Enlargements of apical

whorls.

Description

Shell. Shells are illustrated in Figures 70E—
|, 78A-E, 79, 80. They are small, ovate-conic
(Tables 39-41). Because most snails are
eroded, it is difficult to analyze size classes
based on whorl numbers. It is clear that there
are two size classes of mature snails (Table
39); females are larger than males (Table 41).
Additionally, there appear to be two size
classes within each sex (Table 41). The size
range based on uneroded shells is 2.72—3.88

mm. The aperture is pyriform. Adapically
there is a strong apertural beak; there is no
beak tubercle. There may or may not be an
internal notch groove.

The whorls at the suture are smooth (not
crenulated). There is an umbilical chink. SEM
analysis reveals faint spiral microsculpture at
the shoulders of the whorls after the second
whorl (Fig. 80D, E).

The inner lip is straight to arched, narrowly
separated from the body whorl, and uniformly
thick. The adapical end of the aperture is only

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 237

TABLE 39. Measurements in mm of D85-79 shells of Neotricula duplicata. Mean + standard deviation
= number measured. Sex unknown.

(range). e = eroded apical whorls; (

No. Whorls
Length (L)

Width (W)

L last three whorls
L body whorl

L penultimate whorl
W penultimate whorl
W 3rd whorl

L aperture

W aperture

X

y

Holotype
large class

4e
3.44

1572

3.12

large class
Зе—4е (3)

3.03+0.06
(3.00—3.10)
1.57+0.10
(1.50—1.68)
2.91+0.16
(2.80—3.10)
2.12+0.14
(2.04—2.28)
0.45+0.02
(0.44—0.52)
0.99+0.07
(0.92— 1.06)
0.66+0.05
(0.60—0.70)
1.51+0.12
(1.44—1.64)
0.95+0.08
(0.88—1.04)
0.48+0.07
(0.40—0.52)
0.15+0.06
(0.08—0.16)

Paratypes

small class
3e—5e (5)

2.87+0.10e
(2.72—3.00)
1.46+0.04
(1.44—1.48)
2.67+0.06
(2.56-2.72)
1.94+0.02
(1.92—2.96)
0.47+0.02
(0.44—0.48)
0.97+0.02
(0.96—1.00)
0.65+0.02
(0.64—0.68)
1.36+0.04
(1.32—1.40)
0.86+0.04
(0.80—0.88)
0.44+0.06
(0.36-0.52)
0.15+0.04
(0.10—0.20)

TABLE 40 Measurements in mm of paratype shells of Neotricula duplicata from D87-2. Mean + standard
deviation (range). e = eroded apical whorls; ( ) = number measured. Sex unknown.

No. Whorls
Length (L) of eroded shell

Length of uneroded shells
Width (W)

L last three whorls

L body whorl

L penultimate whorl

W penultimate whorl

W 3rd whorl

L aperture

W aperture

X

Y,

5e
3.44

1.60
2.92
2.08
0.56
1.06
0.72
1.52
1.00
0.52

0.16

5.5 (3)

2.89+0.20
(2.72-3.12)
1.37+0.09
(1.32 1.48)
2.59+0.15
(2.48—2.76)
1.87+0.08
(1.80—1.96)
0.45+0.03
(0.42 —0.48)
0.91 +0.05
(0.88 —-0.96)
0.62+0.05
(0.58—0.68)
1.33+0.09
(1.28—1.44)
0.85=0.06
(0.84—0.92)
0.37=0.05
(0.32—0.40)
0.10+0.02
(0.08—0.12)

6.0 (2)

3.53
(3.46,3.60)
1.61
(1.60,1.62)
3.10
(3.00,3.20)
2.18
(2.08,2.28)

(0.52,0.56)

No variation
152
(1.44,1.60)
0.96

No variation
0.42
(0.40,0.44)
0.14
(0.12,0.16)

238 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 41. Measurements (mm) of shells of Neotricula duplicata from both collections where animals were
used for dissections. Mean + standard deviation (range). All except two shells were eroded, therefore lengths
are less than for entire specimens. e = eroded. *, probably a male. () = number measured.

Females
No. Whorls 2e—5e (3) 5.5 (1)
Length (L) of eroded 3.31+0.45e
shells (2.80—3.68)
Length of 3.60
uneroded shells
Width (W) 1.77+0.08 1.76
(1.68—1.84)
L last three whorls 3.24 (N = 2) 3.20
No var.
L body whorl 2.36+0.08 2.36
(2.28—2.44)
L penultimate 0.69+0.34 0.52
whorl (0.48 —1.08)
W penultimate 1.12+0.04 1.12
whorl (1.08—1.16)
W Зга whorl 0.64 (N = 2) 0.78
(0.56, 0.72)
L aperture 1.60+0.06 1.60
(1.56—1.68)
W aperture 1.07+0.08 1.08
(0.88—0.92)

Е
0.

.5 тт

FIG. 81. Нега of a male Neotricula duplicata from
D85-72 showing the relationship of the base of the
penis (Bp) to the mid-line of the snout-neck (x).

Males
6.0 (1) 4e—5e (2) 5.0(1)* 5.5 (2)
3.22
(3.12, 3.32)
3.88 2.88 3.12
No var
1.76 175% 1.48 1.56
(1.56, 1.58) No var
3.36 2.88 2.62 2.76
(2.76, 3.00) (2.72, 2.80)
2.36 2.10 1.96 1.96
(2.08, 2.12) (1.92, 2.00)
0.62 0.49 0.40 0.46
(0.42, 0.56) (0.44, 0.48)
1.16 1.05 0.96 1.01
(1.02, 1.08) (1.00, 1.02)
0.82 2.88 2.62 2.76
(2.76, 3.00) (2.72, 2.80)
1.64 1.46 1.32 1.38
(1.42, 1.48) (1.36, 1.40)
1.06 0.92 0.88 0.94
No var. (0.92, 0.96)

slightly separated from the body whorl. There
is no apertural sinus. In side view, the outer lip
is straight; the outer lip is slightly scooped for-
ward. In side view, the inner lip has a strong
deflection angle, some 140°. There is no
varix. The abapical lip just beyond the base of
the shell 0.48 + 0.07 mm; the abapical lip has
a spout.

SEM examinations of the apical whorls
(Figs. 79C—E; 80C-E) show that the tip of the
apical whorl is smooth, but that just beyond
the tip the shell is minutely wrinkled. Coarse
growth lines begin just before 1.75 whorls.

External features. The head is dark grey to
black. There may or may not be white gran-
ules close to the eyes extending posteriorly
back along the neck (Wg, Fig. 81). When
granules occur, they do not form a dense lu-
nate mass or “eyebrow.”

Opercula from both collections are shown
in Figures 82, 83. They are elongate-ovoid
and are composed of discernable flaky lay-
ers. The muscular attachment pad (Fig. 82B,
D) is pronounced and wide (66% width of the
operculum).

Mantle cavity. Mantle cavity organs are
shown in Figure 84; not all gill filaments are

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 239

FIG. 82. Opercula of Neotricula duplicata from D85-79. A, C. Outer surface; B, D. Inner surface. Note the
external loose, flaky layer (A, C, D).

DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

PE en

FIG. 83. Opercula of Neotricula duplicata from D87-2.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 241

TABLE 42. Lengths (mm) or counts of non-neural organs of Neotricula duplicata. Mean + standard
deviation (range). For D87-2 snails, N = 5 for females, N = 4 for males; for D85-79 snails, N = 5 for
females, N = 3 for males unless stated otherwise.

Females
D87-2 D85-79 D87-2 D5-79
Body 4.99+0.45 4.90+0.51 4.43+0.14 4.31+0.02
(4.46—5.70) (4.20—5.46) (4.30—4.56) (4.30—4.34)
Digestive gland 1.95+0.46 2.03+0.20 2.02+0.14 1.92+0.14
(1.30—2.50) (1.74—2.30) (1.86—2.20) (1.76—2.00)
Gonad 0.61+0.13 0.75+0.05 1.50+0.38 11-25] =
(0.48—0.74) М = 4 (0.70—0.80) (1.20—2.10) (1.16, 1.30)
Total pallial oviduct 1.70+0.20 2.04+0.34 — —
(= ТРО) (1.50—1.90) М = 3 (1.78—2.50) М = 4
Bursa copulatrix 0.47+0.12 0.48+0.05 — —
(= Bu) (0.30—0.60) N = 4 (0.44—0.56)
Duct of bursa 0.17+0.06 0.10 — —
(0.10—0.20) М = 3 No var. N = 2
Bu - TPO 0.32+0.12 0.23+0.04 —- —
(0.20—0.43) М = 3 (0.18—0.30)
Seminal 0.21+0.07 0.25+0.05 — —
receptacle (0.16—0.30) (0.18—0.26)
Duct of seminal 0.06 0.06 — —
receptacle No var. N = 2 N =1
Mantle cavity 1.51+0.18 1.48+0.07 1.33+0.04 1.35+0.19
(1.26-1.70) (1.40—1.56) М = 4 (1.30—1.38) (1.14—1.52)
Ctenidium 1.27+0.19 1.28+0.09 1.07 +0.09 1.15+0.19
(= CT) (1.00 — 1.50) (1.20—1.36) М = 4 (1.00— 1.20) (0.94— 1.30)
Osphradium 0.37+0.13 0.37+0.03 0.35+0.06 0.36+0.13
(= OS) (0.20 —0.48) (0.34—0.40) N = 3 (0.28 —0.36) (0.24—0.50)
No. gill 19.3+1.7 18.0+1.4 170152 23.6+7.1
filaments (17-21) (16-19) М = 4 (16-18) (16-30)
OS = © 0.30=0.11 0.29+0.04 0.33+0.07 0.32+0.10
(0.17—0.44) (0.23—0.28) N = 3 (0.23—0.40) (0.20—0.38)
Gf, 0.20+0.09 0.28+0.05 — —
(0.12—0.34) N = 8 (0.24—0.36) М = 4
Gf, 0.24+0.07 0.23+0.03 — —
(0.20 —0.30) М = 8 (0.20—0.26) М = 4
Total Gf 0.46+0.08 0.50+0.05 — —
(0.32—0.56) М = 8 (0.44—0.50) М = 4
Gio Gia Gt 0.43+0.15 0.54+0.07 — —
(0.25—0.63) (0.48—0.64) М = 4
Prostate — — 0.85+0.13 0.87+0.06
(0.70—1.00) (0.80—0.90)
Seminal vesicle — — 0.53+0.22 0.45+0.09
(0.20—0.70) (0.40—0.56)
Penis — — 1.16+0.27 151 == 0.22
(0.90—1.54) (1.36-1.76)

illustrated. Measurements and counts are
given in Table 42. The osphradium is approx-
imately mid-gill; it is short (ratio of 0.29). The
shape of the osphradium is lunate, not oval.
The number of gills ranges from 17 to 30. Gill

filament section Gf, is normal; the length of
the longer gill filaments is 0.46 to 0.50 mm
long. The shape of the gill leaflet in side view
is moderate to high domed. There is no cir-
cular patch of white or yellowish granules just

242 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Ga

Gf:

Os

Emc vA

Ma

FIG. 84. Mantle cavity organs of Neotricula duplicata from D89-79. Dashed line is the trajectory for mea-
suring length of gill and mantle cavity. Only three central gill filaments are illustrated.

In Bu

Ast

: А
Di 0.5mm

FIG. 85. Uncoiled female Neotricula duplicata with head and kidney tissue removed; from D85-79.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 243

Pst

Emc

Apo

AH
1.0 mm

FIG. 86. Uncoiled female Neotricula duplicata with head and kidney tissue removed from D87-2.

anterior to the osphradium next to the neck-
mantle collar (Ma) juncture.

Female reproductive system. The bodies of
uncoiled females without head or kidney tis-
sue are shown in Figures 85, 86. Organ mea-
surements are given in Table 42. Important
features are: (1) The gonad is posterior to the
stomach; it is relatively short and of few lobes.
(2) The albumen gland (Ppo) is of normal
length. (3) The bursa copulatrix (Bu) is clearly
visible posterior to the albumen gland; the
bursa is small and round to oval. (4) The ovi-
duct runs from the gonad to the albumen
gland without coiling. (5) The bursa copulatrix
complex of organs is shown in Figures 87, 88.
Figure 87A is positioned exactly as in Figures
85, 86. The spermathecal duct (Sd) opens
into the posterior mantle cavity; it is short and
swollen in most specimens. (6) The seminal
receptacle (Sr) arises from the dorsal surface
at the juncture of the duct of the bursa and the
spermathecal duct. (7) The duct of the semi-
nal receptacle (Dsr) slowly increases in diam-
eter to form a club-shaped (not spherical)
storage sac that lies dorsal to the bursa cop-

ulatrix. In some specimens, it is clear that the
seminal receptacle is a U-shaped continua-
tion of the duct of the bursa and that the sper-
mathecal duct attaches at the bottom of the
“U” (Fig. 88C). (8) The sperm duct (Sdu)
arises from the duct of the bursa on the ven-
tral side just posterior to the duct of the sem-
inal receptacle.

Male reproductive system. The posterior sec-
tion of an uncoiled male with kidney tissue
removed is shown in Figure 89. It is the same
in specimens from both D85 and D87 collec-
tions. The outline of the gonad (Go) is shown
(dashed line) but most of the lobes of the go-
nad are removed to show the seminal vesicle
(Sv) coiled dorsal to the gonad. Measure-
ments of organs are given in Table 42. Impor-
tant features are: (1) The gonad overlaps the
posterior part of the posterior chamber of the
stomach (Pst). (2) The prostate (Pr) overlaps
the posterior end of the mantle cavity (Emc)
and covers the style sac. (3) The seminal ves-
icle, while coiling dorsal to the gonad, does
not extend over the stomach. (4) The anterior
vas deferens leaves the prostate (Pr) close to

245 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

Bu

Emc

С О Ov

FIG. 87. Details and variation of bursa copulatrix complex of organs of Neotricula duplicata from D85-79.
Figure A is in same orientation as in Figures 85 and 86. B. Swelling of spermathecal duct (Sd) seen in most
individuals. C. Bursa complex seen in B flipped over 180° so that Emc is to left; this view shows juncture of
duct of the bursa (Dbu), seminal receptacle (Sr) and spermathecal duct (Sd). D. As in A, but with oviduct (Ov)
pulled to left to show the juncture of Dsr and Sd with duct of bursa.

Sd

TABLE 43. Radular statistics for Neotricula duplicata. Mean + standard deviation (range). In mm except
for the width of the central tooth in um. N = number counted.

D85-77 D87-2
Shell length (not eroded) 3.20+0.26 3.43+0.21
(2.84-3.60) N = 8 (4.44—5.20)
Radula length 0.55+0.04 0.57+0.04
(0.66—0.60) N = 11 (0.52—0.62) N = 10
Radula width 0.08+0.004 0.08+0.006
(0.072—0.088) N = 11 (0.078—0.092) N = 10
Total rows of teeth 70.4+4.1 60.1+8.0
(61-74)N = 8 (46—69) М = 10
No. rows of teeth forming 9.0+2.1 10.3+3.3
(7-13) М = 8 (6-15) М = 10
Central tooth width 15.5+0.8 17.1218
(14.6-16.8)N = 9 (15.8—19.0) М = 7
the posterior end of the mantle cavity (Emc). pilla is withdrawn and cannot be seen (living

(5) The penis is simple, with a small evertible tissue, fresh microscope mount). In others,
papilla (Fig. 90B). In some specimens, the pa- only the tip of the papilla can be seen being

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 245

u,
0.5mm

Emc

Sdu

FIG. 88. Details and variation of bursa copulatrix complex of organs of Neotricula duplicata from D87-2.
Figure A is in same orientation as in Figures 85 and 86. B-C. Variation in configuration of seminal receptacle
(Sr). B. Bursa complex rotated in the direction of arrow to show interconnections of Dbu, Csd, Sr and Sd. C.
Dorsal side of bursa complex to show that duct of bursa (Dbu) makes a U-shaped bend into duct of seminal
receptacle (Dsr). Swollen short spermathecal duct (Sd) enters bottom of U-shaped bend. Note: Common
sperm duct (Csd) is that portion of duct extending from duct of bursa between Sdu and Sd. Duct of bursa
is defined in all papers as that duct extending from bursa copulatrix (Bu) anteriorly to point where another

sperm duct connects to it, e.g. the Sdu.

pushed out and then withdrawn. (6) The base
of the penis (Bp, Figs. 81, 90A) is oriented on
the neck in diverse ways. The angle of the
long axis of attachment varias from parallel to
the mid-line of the snout-neck (x) through 90°.
The base may overlap the mid-line or be to
the right of mid-line. (7) No ejaculatory duct is
seen in the base of the penis or in the neck.

Digestive system. The digestive gland is pos-
terior to the stomach in both sexes. However,
in some males of the D87-2 collection both
digestive gland and gonad covered the pos-
terior chamber of the stomach.

Radular statistics are given in Tables 43
and 44. The most commonly encountered for-
mula in the D85-79 collection was:

2-1-2 ; 3-1[2]-3(2); 12-15; 11-14;
2(3)-(3)2
in the D87 2 collection it was:

3(2)-1-(2)3; 3(4)-1-3; 12-14; 11-13.
3(2)-(2)3

SEM photographs of radulae from snails of
both collections are given in Figures 91-94
No significant differences occur between the

246 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Ast

_OAA A
0.5mm

FIG. 89. Uncoiled male of Neotricula duplicata from D85-79. Anterior part of mantle cavity omitted. Posterior
lobes of the gonad (Go) removed to show coiled seminal vesicle (Sv) dorsal to gonad. Dashed line indicates

extent of gonad.

snails of the two collections; the tooth mor-
phologies are the same and are typical of the
generalized triculine type. Of note is the split
or fork in the dominant central cusp of the
lateral tooth (i.e. the “1” of the 3-1-3) (Fig.
91C, 92C, D, 93G, 94C).

The stomach has a noteworthy character.
The anterior chamber of the stomach (Ast,
Fig. 95) has at the anterior and posterior
edges prominent yellow ridges (Yri).

Nervous system. Measurements of neural
structures are given in Table 45. The RPG
ratio indicates a moderately concentrated
dorsal aspect of the nerve ring.

Remarks

These detailed anatomical studies of a sin-
gle population collected two years apart pro-
vide a unique opportunity to assess (1) how
well methods provide an adequate assess-
ment of qualitative and quantitative character-
stokes. The 1987 study is a control for the
1985 study, particularly as it was not remem-
bered in 1987 that the 1985 assessment had

been made. (2) Variation due to sampling (in
part) and actual quantitative changes that
could have occurred in two years. This is the
first time that we have obtained an indication
of the amount of variation that might be ex-
pected in such a study. This is of considerable
help in assessing how much difference may
actually exist when comparing two popula-
tions, each studied only once, e.g. between
N. cristella and N. dianmenensis of this
monograph.

Overall, there was excellent agreement in
qualitative anatomy when comparing the two
year classes. There was significant variation
in cusp counts between year classes, partic-
ularly involving the marginal teeth. More vari-
ation occurred in central tooth cusp numbers
in the 1987 year class.

Conchologically, N. duplicata is most simi-
lar to N. dianmenensis, N. lili, and Tricula
gredleri (Figs. 153, 154). Differences from N.
dianmenensis are given on page 234. Neotric-
ula duplicata differs from T. gredleri in three
characters (10%). The former has spiral mi-
crosculpture lacking in the latter (char. 6). The
former has an abapical spout lacking in the

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 247

TABLE 44. Cusp formulae for the radular teeth of Neotricula duplicata with the percent of radulae in
which a given formula was found at least once.

Central Teeth Inner Marginal Outer Marginal
(N = 10) Lateral Teeth Teeth Teeth
D85-79
2-1-2 70% 3-1[2]-3 100% 9 — 9 10%
3-3
2-1-2 30% 3-1[2]-2 50% 10 10% 10 20%
2-2
2-1-2 30% 4-1[2]-3 30% 11 20% 11 60%
3-2
2-1-2 20% 2-1[2]-3 20% 12 50% 12 70%
2-3
3-1-3 10% 13 100% 13 60%
2-2
3-1-2 10% 14 90% 14 40%
2-2
15 50% 15 —
16 30% 16 —
17. 20% We —
X M6 15 12.0+1.1
N = 108 120
D87-2
3-1-3 60% 3-1[2]-3 100% 11 — di 40%
3-3
2-1-2 50% 4-1[2]-3 40% 12 70% 12 90%
3-3
2-1-2 40% 3-1[2]-4 30% 13 80% 13 30%
2-3
3-13 30% 2-1[2]-3 10% 14 60% 14 20%
2-2
2-1-2 20% 3-1[2]-2 10% 15 20% 15 —
3-2
2-1-2 10% Х = 12.9=0.7 12103
2-2 N = 100 N = 100
3-1-3 10%
2-3
3-1-2 10%
3-3
2-1-2 10%
3-2
2-13 10%
3-3

*Mean + standard deviation of cusp number for all teeth counted.

248 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

FIG. 90. A. The position of base of the penis of Neo-
tricula duplicata relative to the snout-neck mid-line
(x). B. Penis.

latter (char. 12). The latter has an adapical
beak tubercle lacking in the former (char. 24).
Anatomical differences are considerable,
those used to define the genera among oth-
ers (Fig. 157).

Conchologically, N. duplicata differs from N.
lili by having a wrinkled protoconch, not a
smooth one (char. 8); having an adapical
beak instead of a notch (char. 9); having a
straight to arched inner lip, not a sinuate one
(char. 13); having a thick inner lip, not a thin
one (char. 19); by lacking a slight varix found
in the latter (char. 23); and by lacking an
adapical outer lip angle seen in the latter
(char. 25). Anatomically, these species differ
in nine characters (19%) (Figs. 157, 158). The
differences are: The former has an elongated

oval operculum with two or more layers (chars.
2, 3); the latter has an ovate operculum with
one layer. The opercular attachment pad is
very wide in the former, wide in the latter
(char. 4). Gill filament section Gf, is of me-
dium length in the former, long in the latter
(char. 8). The bursa shape is ovoid in the
former, round in the latter (char. 15). The
spermathecal duct is short in the former, long
in the latter (char. 19). The spermathecal duct
runs directly anterior from the duct of the
bursa to the mantle cavity in the former; it
slants away at an angle in the latter (char. 20).
The vas deferens bends away from the pros-
tate at the posterior end of the mantle cavity in
the former, at mid-prostate in the latter (char.
31). Neotricula lili has very many rows of
teeth on the radula; N. duplicata has many
(char. 42).

Neotricula lilii Chen & Davis, sp. nov.
Holotype. ZAMIP-M0002, Figure 70J.

Paratypes. ANSP 373138, A12654, Figure 70
K, L Figure 96A-D.

Type Locality. Chang Wang Village, Chuan-
xing Town, Lingxian County, Zhuzhou Prefec-
ture. 26°17'18"N, 113°41’3’E. Figure 1, Site
2.

Etymology. Named for Dr. Li Li, Zhuzhou Pre-
fectural Epidemic station, who collected this
species.

Habitat

The assigned field collection number was
D85-74. Snails came from a mountain stream
550 m above sea level. The stream was 10 to
15 cm deep, with small rocks, leaves and
mud. There was low vegetation beside the
stream.

Description

Shell. The shells are small, ovate-conic, with
5.5 to 6.0 whorls (Figs. 60A-B, 70J-L, 96A—
B). Measurements are given in Tables 46, 47.
Limited data suggest that females are larger
than males (Table 47) with the lengths of the
last three whorls of females ranging from
2.98-3.28; those for males, 2.72-2.80. The
aperture is pyriform. Adapically there is a wide
apertural notch; there is no beak tubercle.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

=

4343

FIG. 91. Radula of Neotricula duplicata from D85-79. Centrals, laterals and inner marginals featured in B, C.

D. Outer marginals.

TABLE 45. Lengths (mm) of neural structures of Neotricula duplicata. Mean + standard deviation

+

y, р
р
‘

(range). * = neural elements measured to calculate the RPG ratio. N = number measured.

Cerebral ganglion
Cerebral commissure

Pleural ganglion
Right (1)*

Left

Pleuro-supraesophageal
connective (2)*
Pleuro-subesophageal
connective
Supraesophageal
ganglion (3)*
Subesophageal
ganglion
Osphradio-mantle nerve

RPG ratio (2 + 1+2+3)*

D85-79 (N = 4)

0.26+0.03
(0.22—0.30)
0.05+0.02
(0.02—0.06)

0.14+0.03
(0.10—0.16)
0.12+0.03
(0.08—0.14)
0.14+0.04
(0.10—0.18)
0.06+0.05
(0.00—0.10)
0.13+0.02
(0.10—0.14)
0.11+0.02
(0.10—0.12)
0.08+0.01
(0.06—0.08)
0.34+0.08
(0.26-0.45)

D87-2 (N = 5)

0.24+0.03
(0.24—0.26)
0.04+0.01

(0.03—0.06)

0.13+0.02
(0.10—0.16)
0.12+0.01
(0.10—0.12)
0.11+0.02
(0.10-0.14)
0.06+0.02

(0.03-0.08) N = 4

0.12+0.02
(0.10-0.14)
0.11+0.01

(0.10-0.12)N = 4

0.08+0.04
(0.00—0.12)
0.31+0.03
(0.28—0.35)

250 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

4

4

de

n° 240) D
= yo
Pa {<

№
4
~

ar

р; #
г, E

i“
=> =

FIG. 92. Radula of Neotricula duplicata from D85-79. Centrals, laterals and inner marginals featured in B-D.
E, F. Outer marginals.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 251

q

55

33"

|

7
0 A [
{

-

>

| D = um
= Я |

FIG. 93. Radula of Neotricula duplicata from D87-2. Centrals, laterals and inner marginals featured in В, C,
E, G. D, H = outer marginals. A-D = males; E-H = females.

252 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

FIG. 94. Radula of female Neotricula duplicata from
D87-2.

Pst_

FIG. 95. Ventral aspect of the stomach of Neotricula
duplicata in same orientation as in Figures 85, 86.

There is no internal notch groove. There is a
pronounced adapical outer lip angle.

The whorls at the sutures are smooth (not
crenulated). There is an umbilicus. SEM anal-
ysis reveals faint spiral microsculpture at the
shoulders of the whorls after 1.75 whorls (Fig.
96A, C).

The inner lip is slightly sinuate and clearly
separated from the body whorl in most spec-
imens; it is uniformly thickened. In some
(< 20%), the adapical inner lip is fused to the
body whorl. The adapical end of the aperture
is only slightly separated from the body wall.
An apertural sinus is apparent examining the
outer lip in side view. In side view, the outer
lip, abapical to the sinus, is straight. The outer
lip is slightly scooped forward. In side view,
the inner lip has a strong deflection angle.
There is a slight varix. There is an abapical
spout. The abapical lip projects beyond the
base of the shell 0.50 + 0.05. SEM examina-
tion of the shell reveals that the apical whorl is
smooth. (Fig. 96D).

External features. The head is grey. There
are no granules about the eyes and no
“eyebrow.” The operculum is corneous and
paucispiral. The inner lip edge is straight abapi-
cally and slightly sinuate towards the aper-
tural beak end (adapically). The muscle at-
tachment callus is pronounced and narrow,
only 40% of the operculum width. A porous,
regularly pitted surface of the callus was
found in most specimens (75%) (Fig. 96E-H).

Mantle cavity. Measurements and counts of
mantle cavity structures are given in Table 48.
The ovoid osphradium is mid-gill; it is short.
There are 20-24 gill filaments. Gf, is long.
The length of the longest filament is 0.51 mm.
The larger gill filaments are moderately
domed. There is no spherical patch of white
granules anterior to the osphradium where
the mantle collar joins the neck.

Female reproductive system. The body of an
uncoiled female without head or kidney tissue
is shown in Figure 97. Organ measurements
are given in Table 48. Important features are:
(1) The gonad (Go) is posterior to the stom-
ach; it is short and consists of few groups or
bundles of lobes. (2) The bursa copulatrix
(Bu) is spherical and prominent posterior to
the albumen gland (Ppo). (3) The bursa is
short. (4) The albumen gland is standard (=
normal) length. (5) The bursa copulatrix com-
plex of organs is shown in Figure 98A, B in
the same orientation as in Figure 97. The nar-

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 253

Saaum

FIG. 96. SEM photographs of shells and operculum of Neotricula lili. C, D. Enlargements of apical whorls.
E-H. Inner surfaces of four opercula; insets provide enlargements of attachment pads to reveal micro-
stucture.

254 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 46. Shell measurements (mm) for the types of Neotricula lili. Mean + standard deviation (range).
e = eroded apical whorls. N = number measured.

Paratypes

Holotype Larger (N = 1) Smaller (N = 4)

No. Whorls 5:5 4e 4e to 5.5
Length (L) 3.20 3.20 2.98+0.18
(2.76—3.20)

Width (W) 1.52 1.68 1.51+0.10
(1.38—1.62)

L last three 2.76 3.00 2.70+0.11
whorls (2.54—2.80)

L body whorl 1.96 2.20 1.90+0.09
(1.76—1.96)

L penultimate 0.48 0.48 0.49+0.02
whorl (0.48—0.52)
W penultimate 0.98 1.12 1.03+0.05
whorl (0.96— 1.08)
W 3rd whorl 0.72 0.80 0.72+0.04
(0.68—0.76)

L aperture 1.40 1.60 1.39+0.10
(1.24—1.48)

W aperture 0.92 1.04 0.93+0.06
(0.86 — 1.00)

x 0.52 0.56 0.50+0.05
(0.44—0.56)

y 0.12 0.20 0.15+0.04
(0.12—0.20)

TABLE 47. Shell measurements (mm) for Neotricula lili used for anatomical work. Mean + standard

deviation (range). e = eroded apical whorls. N

number measured.

Males
Larger Size Class (N Small Size Class (N (N = 3)
No. Whorls 4e-6.0 5.75 Зе—4е
Length (L) 3.43+0.19 3.12 2.89+0.12
(3.20—3.56) (2.76—3.00)
Width (W) 1.63+0.08 1.48 1.47+0.06
(1.56—1.72) (1.40—1.52)
L last three 3.11+0.15 2.72 2.76+0.04
whorls (2.98- 3.28) (2.72—2.80)
L body whorl 2.21+0.12 1.92 1.95+0.06
(2.08- 2.28) (1.88—2.00)
L penultimate 0.55+0.04 0.48 0.49+0.04
whorl (0.52—0.60) (0.44—0.52)
W penultimate 1.13+0.06 1.00 1.03+0.04
whorl (1.08—1.20) (1.00—1.08)
W 3rd whorl — = —
L aperture 1.59+0.10 1.32 1.35+0.08
(1.48- 1.68) (1.28—1.44)
W aperture 0.99+0.02 0.96 0.89+0.02
(0.96 — 1.00) (0.88 —0.92)
x 0.52+0.06 0.52 0.45+0.02
(0.44—0.56) (0.44—0.48)

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 255

Ast

Pst

cx

1.0 mm

FIG. 97. Uncoiled female Neotricula lilii with head and kidney tissue removed.

256 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 98. Details and variation of bursa copulatrix
complex of organs of Neotricula lili. Figure A is in
same orientation as in Figure 97. B. Bursa flipped
over to show place where seminal receptacle (Sr)
enters common sperm duct (Csd). C. Variation in
size and shape of seminal receptacle.

row spermathecal duct dilates to form the
common sperm duct (Csd) from which opens
the sperm duct (Sdu) and the seminal recep-
tacle (Sr) (Fig. 98 B). (6) The sperm duct is
very short and to the left of the common
sperm duct. (7) The seminal receptacle arises
some 45” from the opening of the sperm duct;
it arises from the right dorso-lateral side of the
sperm duct. (8) The seminal receptacle varies
in shape from elliptical to sub-spherical (Fig.
98C); it has a very short duct (Dsr). (9) The
duct of the bursa (from the bursa to openings
of Dsr and Sdu) is extremely short. (10) The
distance from the posterior end of the mantle
cavity (Emc) to the bursa is short, only 0.16
mm to 0.32 mm. The spermathecal duct—
common sperm duct run straight posteriorly
from the end of the mantle cavity to the bursa.

Male reproductive system. A section of an un-
coiled male without kidney tissue is shown in
Figure 99. The anterior gonadal lobes are re-
moved to show the seminal vesicle (Sv). Or-
gan measurements are given in Table 48. Im-
portant features are: (1) The gonad (Go)
slightly overlaps the posterior chamber of the
stomach. (2) The seminal vesicle (Sv) arises
from the vas efferens (Ve) at mid-gonad to
slightly posterior to mid-gonad. (3) The pros-
tate (Pr) overlaps the posterior end of the
mantle cavity and covers half the style sac
(Sts). (4) The anterior vas deferens (Vd,) di-
verges from the prostate slightly anterior to
the posterior end of the the mantle cavity. (5)
The penis is simple with a papilla (Pa) that
everts (Fig. 100B). There is a massive con-
centration of white granules along the anterior
convex edge. (6) The penis arises to the right
of the snout-neck mid-line (Fig. 100A). The
long axis of the penial base (Bp) is swollen at
its base increasing the diameter from 0.14
mm to 0.22 mm. (7) No ejaculatory duct is
seen in the penial base or neck.

Digestive system. The digestive gland is pos-
terior to the stomachs of both males and fe-
males. Radular statistics are given in Tables
49 and 50. The radula is illustrated in Figures
101, 102. The central tooth is the generalized
triculine type (Fig. 101B, E). The dominant
cusp of the lateral tooth (the “1” of the 3-1-3)
is almost always bifurcated (Fig. 101). The
most commonly encountered cusp formula is

3(2)-1-(2)3; 3-[2]-3(4); 13-15; 12-14.
22

Nervous system. It is standard triculine. Mea-
surements are given in Table 51. The RPG
ratio of 0.44 indicates that the dorsal nerve
ring is moderately concentrated.

Remarks

Conchologically, N. lili is most similar to N.
duplicata and N. dianmenensis (Figs. 153—
155). The comparison of these three species
is given in the remarks section for the two
previously treated species. Anatomically, N.
liliiis most similar to N. cristella, a species that
has a very different shell (Figs. 154, 156-158).
There are five differences (Tables 80, 81)
(10%): N. МИ has an operculum with a single
layer whereas the operculum of N. cristella's
has two or more layers (char. 3). The former

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 257

TABLE 48. Lengths (mm) or counts of non-neural organs and structures of Neotricula lili. N = number
of snails used. Mean + standard deviation (range).

Body
Gonad
Digestive gland
Posterior pallial oviduct
(= albumen gland)
Anterior pallial oviduct
(= capsule gland)
Total pallial oviduct
= OV
Bursa copulatrix
= BU
Duct of BU
BU + OV
Seminal receptacle
Duct of seminal receptacle
Mantle cavity
Gill (G)
Osphradium (OS)
95-6
No. of filaments
Gf,
Gf,
Total Gf = TGF
СР. = ТСЕ
Prostate

Seminal vesicle

Penis

Females (N = 4) Males (N = 3)
4.96+0.82 4.27+0.08
(3.92—5.86) (4.64—4.80)
0.80+0.12 1.42+0.07
(0.70 —0.96) (1.36—1.50)
2.03+0.23 2.13+0.19
(1.80—2.30) (2.00—2.34)
1.00+0.08 —
(0.96-1.10)N = 3
1.14+0.16 “=
(0.96—1.26) М = 3
2.15+0.22 —
(1.92-2.36) М = 3
0.52=0.04 —
(0.50—0.56) М = 3
0.10 (М = 1) —
0.24+0.02 —
(0.23—0.26)
0.19+0.02 —
(0.16—0.20)
0.04+0.02 —
(0.02—0.06)
1.54+0.20 1.37+0.07
(1.20—1.56) (1.30—1.44)
1.19=0.15 1.22+0.07
(1.0—1.36) (1.16—1.30)
0.32+0.05 0.38+0.02
(0.28—0.38) (0.36—0.40)
0.27+0.07 0.31+0.02
(0.22—0.38) (0.28—0.33)
21.8+1.7 21.4+0.6
(20—24) (21-22)
0.32+0.03
(0.28—0.36) М = 6 Males 8 Females
0.19+0.03
(0.16—0.20) М = 6 Males & Females
0.51+0.05
(0.46—0.58) N = 6 Males & Females
0.63+0.04
(0.58—0.69) М = 6 Males & Females
— 1.02+0.13
(0.90— 1.16)
— 1.00+0.20
(0.80 — 1.20)
— 1.15+0.01
(1.50—1.52)

TABLE 49. Radular statistics for Neotricula lili.

has a moderate number of gill filaments whilst

Mean + standard deviation (range). N 4. In the latter has few (char. 6). The male gonad
Е wel OP Centre) tooth ayn: overlaps the stomach of the former; it is pos-
Shell length 3.00+0.17 (2.80 —3.16) terior to the stomach in the latter (char. 27).
Radular length 0.58+0.01 (0.57 —-0.60) The anterior vas deferens leaves the prostate
Radular width 0.08 (0.072— 0.080) at mid-prostate in the former; from the pros-
Total rows of teeth 87 (84, 90) (N = 2) tate at the posterior end of the mantle cavity in
a teeth 19 (16, 22) (N = 2) the latter (char. 31). The penial tip of the-
Central tooth width 17.2(16.2-17.9) N = 3 ea ee Sn ella

258 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 99. Uncoiled male of Neotricula lilii without head or kidney tissue. Anterior part of mantle cavity omitted
as is posterior end of digestive gland.

TABLE 50. Cusp formulae for the radular teeth of Neotricula lilii with the percent of the three radulae in
which a given formula was found at least once.

Inner Marginal Outer Marginal
Central Teeth Lateral Teeth Teeth Teeth
3-1-3 66% 3-1[2]-3 100% 12 — 33%
2-2
2-1-2 33% 3-1-3 66% 13 100% 66%
2-2
3-1-3 33% 4-1[2]-3 33% 14 100% 100%
2-3
3-1-4 33% 15 66% 33%
16 33% —
Х * 14.2+0.9 13.8+1.0
N = 30 N = 30

*Mean + standard deviation of cusp number for all teeth counted.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 259

x

FIG. 100. A. Relationship of base of the penis of
Neotricula lilii to mid-line of snout-neck (x) and to
posterior end of eye lobes (a). B. Penis.

Neotricula minutoides (Gredler, 1885)

Paralectotypes. SMF 4155; pl. 4, fig. 1, in
Yen, 1939.

Type locality. “Aus Quellen bei Hensan”;
Heng-shan-hsien, Hunan (Yen, 1939). Near
site 1, fig. 1 (Hengshan)

Synonymy. Bithynia minutoides Gredler,
1885

Hydrobia minutoides (Gredler, 1887)

Tricula minutoides, Yen, 1939

Neotricula minutoides, this paper

Habitat

Specimens collected for this paper came
from Tong Meng Village, Xikou town, Cili
County, Changde Prefecture; 29°13'58"N,
110°44'12’E; Figure 1 site 8. The field collec-
tion number was D85-83 on 7 October 1985.
Specimens were collected from Tong meng
village, Xikou Tow, Cili County, Changde Pre-
fecture; 29%13'58"N, 110°44'12"E; Figure 1
site 8. The assigned collection number was
D85-83. The habitat was a small perrenial
stream 15-25 cm wide and 5-10 cm deep at
an altitude of 550 m above sea level. The
stream was shaded; the flow was slow. The
bottom of the stream was paved with small
stones in mud. The water was clean and cool.
The sides of the stream had weeds and short
shrubery.

TABLE 51. Lengths of neural structures of Neotricula lili. Mean + standard deviation (range). N = 4.

*

Cerebral ganglion
Cerebral commissure

Pleural ganglion
Right (1)*

Left
Pleuro-supraesophageal connective (2)*
Pleuro-subesophageal connective
Supraesophageal ganglion (3)*
Subesophageal ganglion
Osphradio-mantle nerve
RPG ratio* = 2 = 1+2+3

= neural elements measured to calculate the RPG ratio.

0.25+0.01 (0.24—0.26)
0.09+0.05 (0.04—0.16)

0.11+0.01 (0.10—0.12)
0.10 (N = 3) No var.
0.17+0.03 (0.12—0.18)

0.08+0.03 (0.04-0.10) N = 3
0.11+0.03 (0.10—0.14)
0.08+0.04 (0.10-0.12) N = 3

0.02 (0, 0.04) N = 2
0.44+0.08 (0.33—0.50)

260 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Se D
FE 4

я ` s J
ogg
= #
¿ A
РЕ >
# ; LA y
"> hr rn
vd EM
р /

en Dm Sy x

FIG. 101. Radula of Neotricula lili. A. Segment of radula. В, E. Central teeth. С, D, Е. Lateral and inner
marginal teeth.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 261

2вим

FIG. 102. Radula of Neotricula lili. A. Right lateral and inner marginal teeth. B-F. Outer marginal teeth; В,
D, F = right; C, E = left outer marginal teeth.

262 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 52. Shell measurements (mm) of Neo-
tricula minutoides. Mean =+ standard deviation
(range). N = 5; all shells 5.5 whorls.

Length (L) 3.41+0.20 (3.14—3.60)
Width (W) 1.68+0.09 (1.60—1.80)
L last three whorls 3.07+0.18 (2.84—3.28)
L body whorl 2.28+0.15 (2.12—2.48)

L penultimate whorl
W penultimate whorl

0.50+0.04 (0.44—0.56)
1.09+0.05 (1.04—1.16)

W 3rd whorl 0.72+0.03 (0.68—0.76)
L aperture 1.53+0.13 (1.40—1.70)
W aperture 1.02+0.07 (0.96-1.12)
x 0.44+0.05 (0.40—0.52)
y 0.18+0.05 (0.12—0.24)
Depository

Specimens are deposited in ZAMIP,

№0006; ANSP, 373136, A12652.

Description

Shells. The shells are small, ovate-conic with
5.5 whorls (Figs. 60C-G, 103A-E). Lengths
range from 3.14-3.60 mm (Table 52). The
aperture is pyriform. Adapically there is no
notch or beak, nor is there an internal notch
groove. There is no adapical outer lip angle.

The whorls at the suture are smooth (not
crenulated). There is a pronounced umbilicus.
SEM analysis reveals faint traces of spiral mi-
crosculpture on the adapical part of the
whorls of the teleoconch.

The inner lip is thick and arched. It is clearly
separated from the body whorl by a narrow
gap. The adapical apertural lip is pulled away
from the body whorl but the interspace is filled
with shell layers (as is the inner side of the
inner lip). There is no apertural sinus. In side
view, the outer lip is straight and slightly
scooped forward. In side view, the inner lip is
Straight. There are no apertural teeth or
notches, no varix, no spout. The abapical lip
projects beyond the base of the shell 0.44 +
0.05 mm.

As seen with SEM, the protoconch is mi-
nutely wrinkled (Fig. 103C—E). Growth lines
begin on the teleconch at 2.0 whorls.

External Features. The head is dark grey,
with few or no white granules about the eyes.
The operculum is corneous and paucispiral
(Fig. 103F). It appears to grow in layers. The
internal attachment pad is prominent, some
49% the width of the operculum.

Mantle cavity. Mantle cavity structures are
typically triculine. Organ measurements are
given in Table 53. The osphradium is located
mid-gill; it is small. Only one gill filament was

measured; it was 0.58 mm long. For this fila-
ment Gf, was long.

Female reproductive system. An uncoiled fe-
male without head and with kidney tissue re-
moved is shown in Figure 104. Measure-
ments of organs are given in Table 53.
Important features are: (1) The gonad is pos-
terior to the stomach, is small, and consists of
one or two bundles of lobes. (2) The bursa
copulatrix (Bu) is round, small and situated
directly posterior to the albumen gland (Ppo).
It is not occluded by the posterior part of the
albumen gland. (3) The albumen gland is
short. (4) The bursa copulatrix complex of or-
gans is shown in Figure 105. The orientation
of organs in Figure 105A is the same as that
in Figure 104. The bursa (Bu) is short. (5) The
seminal receptacle (Sr) is spherical; it may or
may not have a duct. In either case, it opens
into the base of the duct of the bursa where
the latter runs into the spermathecal duct (Sd)
and also receives the sperm duct (Sdu). This
opening for the seminal receptacle is on the
dorso-lateral side of the bursa complex (Fig.
105C). (6) When there is a discernable duct
of the seminal receptacle, it is extremely
short, about 0.02 mm. (7) The spermathecal
duct is short and swollen; it is nearly a
Straight, as well as a direct continuation of the
duct of the bursa; it opens into the posterior
end of the mantle cavity (Emc).

Male reproductive system. The posterior half
of an uncoiled male with kidney tissue re-
moved is shown in Figure 106. Measure-
ments of organs are given in Table 53. Impor-
tant features are: (1) The gonad (Go) is
posterior to the stomach. It consists of several
bundles of lobes (removed in Figure 106 to
reveal the seminal vesicle) arising from the
vas efferens. (2) The prostate (Pr) overlaps
the posterior end of the mantle cavity (Emc.)
(3) The seminal vesicle (Sv) arises from mid-
vas efferens. (4) The seminal vesicle forms a
small knot of ducts dorsal to the gonad; it
does not continue onto the stomach. (5) The
anterior vas deferens (Vd,) leaves the pros-
tate near the posterior end of the mantle cav-
ity (Emc). (6) The penis (Figure 107B) is sim-
ple, slender and differs from that of any other
Hunan species by being highly extensible. (7)
The penis has no papilla nor was an ejacula-
tory duct found at the base of the penis or in
the neck. (8) The orientation of the base of the
penis on the neck is shown in Figure 107A. It
slightly overlaps the snout-neck mid-line; it is

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 263

E

ми

FIG. 103. SEM photographs of shells and opercula of Neotricula minutoides. Note in В that the ощег lip is
slanted. C-E. Details of protoconch and in C, beginning teleconch. F. Opercula with outer surface (left) and

inner surface (right).

orientated at an angle of 40” to the snout-neck
mid-line (x).

Digestive system. The digestive gland covers
the posterior chamber of the stomach of both
sexes. Radular statistics are given in Tables

54 and 55. There are 56.4 + 4.8 rows of teeth
along a radula of 0.51 mm length. The most
frequently encountered formula is

(2)3-1-3(2); 2 to 4-[2]-2(3); 12-14; 11-14.
2-2

264 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 53. Lengths (mm) or counts of non-neural organs and structures of Neotricula minutoides. N =
number of snails used. Mean + standard deviation (range).

Females (N = 5) Males (N = 1)
Body 5.02+0.13 4.40
(4.90—5.16)
Gonad 0.83+0.15 1.28
(0.70—1.00)
Digestive gland 2.10+0.28 1.90
(1.80—2.36)
Posterior pallial oviduct 0.80 —
(= albumen gland) (0.60, 1.00) N = 2
Anterior pallial oviduct 1.28 —
(= capsule gland) (1.16, 1.40) N = 2
Total pallial oviduct 2.02+0.34 =
= OV (1.76—2.40)
Bursa copulatrix 0.55+0.14 u
= BU (0.40—0.66)
Duct of BU — —
BU = OV 0.28+0.10 —
(0.17—0.35)
Seminal receptacle 0.10 —

(no var.) М = 2
Duct of Seminal receptacle = —

Mantle cavity 1.56 1.40
(1.36, 1.76) N = 2

Gill (G) 1.40 1.24
(1.20, 1.60) N = 2

Osphradium (OS) 0.41 0.44
(0.36, 0.46) N = 2

OS = G 0.29 0.35
(0.29, 0.30) N = 2

No. of filaments 17 18
(13, 21) N =2

Gf, 0.36 М = 1 —-

Gf, 0.22 М = 1 —

Total Gf = TGF 0.58 N = 1 —

Gi IGE 0.62 М = 1 —

Buccal mass 0.57 N = 1 -—

Prostate — 1.10

Seminal vesicle — 0.48

Penis — 2.60

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 265

Di

1

.Omm

FIG. 104. Uncoiled female Neotricula minutoides with head and kidney tissue removed.

266 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

ie 1.0mm

FIG. 105. Details and variation of the bursa copu-
latrix complex of organs of Neotricula minutoides.
Figure A is in the same orientation as in Figure 104.
B. Section of oviduct cut away at “x” to show con-
tinuation of duct of bursa (Dbu) into spermathecal
duct (Sd) with latter opening (Osd) into posterior
end of mantle cavity. C. Bursa complex flipped over
to show relationships of seminal receptacle to
sperm duct (Sdu), spermathecal duct (Sd) and duct
of bursa.

The radula is shown in Figure 108. A pro-
nounced bifurcation of the blade of the pri-
mary cusp of the lateral tooth may be found
(Fig. 108C, D, E). Otherwise tooth morpholo-
gies are standard triculine.

TABLE 54. Radular statistics for Neotricula
minutoides. Mean + standard deviation (range).
N = number used. In mm except for width of
central tooth in рт.

Females (N = 6)
Shell length

3.67+0.30 (3.40—4.00)
Radular length 0.51+0.06 (0.44—0.57)
Total rows of teeth 56.4+4.8 (50—60) N = 7
No. rows of teeth 10.9+6.3 (5-21) М = 8
forming
Central tooth
width

15.1+1.1 (13.4-16.8) № = 13

Nervous system. Measurements are given in
Table 56. The RPG ratio is 0.45; the dorsal
nerve ring is thus moderately concentrated.

Remarks

Conchologically, this species most closely
resembles Tricula bambooensis (Figs. 153—
155). Neotricula minutoides has a proportion-
ally much wider shell (compare illustrations
here with Davis et al., 1986: fig. 20A-E). Ad-
ditionally, the outer lip of the former, in side
view, is scooped forward; it is parallel with
the axis of coiling in the latter (char. 15). The
adapical aperture of the former is fused to
the body whorl; it is slightly separated from
the body whorl in the latter (char. 21). The
aperture of the former is pyriform; it is oval in
the latter (char. 3).

The most prominent anatomical differences
are those that separate the two genera.

TRICULINI Davis, 1979
Type genus. Tricula Benson, 1843

Diagnosis. Those genera of Triculinae in
which the spermathecal duct enters the peri-
cardium. The oviduct makes a 360° closed
twist. In the plesiomorphic state, the seminal
receptacle arises from the oviduct; in the de-
rived state, the seminal receptacle is lost and
its function taken over by derived structures
attaching to the oviduct-spermathecal duct
juncture. There is no sperm duct.

Genera assigned. Delavaya, Fenouilia, La-
cunopsis, Lithoglyphopsis, Tricula. (N= 5).

Lithoglyphopsis Thiele, 1928

Type species. Lithoglyphus modestus Gred-
ler, 1886 [1887]

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 267

Ast In

éáAE AH)
1.0mm

FIG. 106. Uncoiled male of Neotricula minutoides without head or kidney tissue. Anterior part of mantle
cavity omitted. Outline of gonad (Go) represented by a dashed line; lobes of gonad removed to show knot
of tubes of seminal vesicle (Sv).

TABLE 55. Cusp formulae for the radular teeth of Neotricula minutoides with the percent of radulae in
which a given formula was found at least once. N = 7 radulae. [ ] indicates one cusp support with a
split blade.

Inner Marginal Outer Marginal

Central Teeth Lateral Teeth Teeth Teeth
3-1-3 71% 3-[2]-2 57% 11 29% 11 100%
2-2
2-1-2 57% 2-[2]-3 29% 12 71% 12 71%
2-2
3-1-3 29% 4-1-3 29% 13 86% 13 71%
3-2
4-[2]-3 29% 14 71% 14 43%
3-1-3 14% у
2-3 3-1-2 14% X = 12.5+1.9 AT +1.6
2-1-3 14% N = 64 N = 64
ЕО 3-1-3 14%
4-[2]-2 14%

“Mean + standard deviation of cusp number for all teeth counted. N = number of teeth counted on 7 radulae.

268 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

x

Bp N

FIG. 107. A. Relationship of base of the penis of
Neotricula minutoides to mid-line of the snout-neck
(x) and to posterior end of the eye lobes (a). B.
Penis.

Type locality. Hen-Kiou-fee bis Pe-shang
(Gredler, 1886), = Heng-dshou-fu bis Pe-
shang, Hunan (Yen, 1939).

Designation. By Thiele, 1928

TABLE 56. Lengths (mm) of neural structures of
Neotricula minutoides. Mean + standard deviation
(range). N = number used = 2. * = neural
elements measured to calculate the RPG ratio.

Cerebral ganglion
Cerebral commissure

Pleural ganglion
Right (1)* 0.13 (0.12, 0.14)

Left 0.11 (0.10, 0.12)
Pleuro-supraesophageal
connective (2)*
Pleuro-subesophageal
connective
Supraesophageal
ganglion (3)*
Subesophageal ganglion
Osphradio-mantle nerve

RPG ratio* = 2+ 1+2+3

0.28 (0.26, 0.30)
0.06 (0.5, 0.6)

0.20 (no var.)

0.07 (0.04, 0.10)
0.11 (0.10, 0.12)

0.11 (0.10, 0.12)
0.05 (0.04, 0.06)
0.45 (no var.)

Lithoglyphopsis modesta (Gredler)

Types. Paralectotypes SMF 4214a. Figure
109A = specimen figured by Yen (1939). Fig-
ure 109B-D = additional type specimens.

Type locality. See above

Synonymy. Lithoglyphus modestus Gredler,
1886

Lithoglyphopsis modesta, Thiele, 1928
Lithoglyphopsis modesta, Wenz, 1939: 580,
Figure 1581

Lithoglyphopsis, modestus Yen, 1939: 43, pl.
4, Figure 7.

Habitat

The type locality is today known as Baisha,
Hengshan County from the Xiangjiang River;
25°58'22"N, 112°45'55’E, Figure 1, locality 9.
Snails were collected by diving to obtain
stones from the river bottom at a depth of
about 2.0-2.5 m. Collected by Chen and Wu
in 1986, collection number 86-B, ANSP
373144, A12660, ZAMIP M00054.

The locality from which snails were used for
most of the dissections was Anhua town, An-
hua County, Zijiang River; 28°23'46"N,
111°,12'41”Е, Figure 1, locality 10. The col-
lection number was D87-1, by Davis, Chen
and Xing on 16 March 1987, ANSP 373150,
A12666. Snails were collected 1.6 km up-
stream from the town boat landing. Collec-
tions came from the shores of a small island
in the middle of the river. Water flows through
a stone breakwater at the upstream end of the

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 269

14 У

\ Wi № fi 4

<

FIG. 108. Radula of Neotricula minutoides. A. Section of radula. В-Е. Central, lateral and inner marginal
teeth. F. Outer marginal teeth.

270 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 109. Shells of Lithoglyphopsis modesta. A-D, paralectotypes, SMF; A. specimen figured by Yen (1939).
E, F. Specimens from Anhua, D87-1. A = 4.56 mm long; other shells printed at same scale.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 271

TABLE 57. Shell measurements (mm) of Anhua and Baisha populations of Lithoglyphopsis modesta.

Mean + standard deviation (range). N

numbered measured.

No. Whorls

Length (L)

Width (W)

L body whorl

L penultimate
whorl

W penultimate
whorl

W 3rd whorl

L aperture

W aperture

L crescent

W crescent

W columellar

plate

Е “А”

W “Аг

3.0-3.5

5.18+0.04
(4.58—5.57)
4.87+0.33
(4.74—4.99)
4.91+0.40
(4.37-5.41)
0.15+0.04
(0.08—0.17)
1.12+0.15
(0.99—1.33)
0.44+0.04

(0.42-0.50) N = 3

4.43+0.19
(4.16—4.66)
3.62+0.23
(3.24-3.83)
2.31+0.19
(2.08—2.50)
0.42+0.06
(0.33—0.50)
0.73+0.06
(0.67—0.83)
5.53+0.25
(5.16-5.74)
4.20+0.27
(3.91-4.58)

Anhua
Female (N = 5)

Male (N = 6)
3.0-3.5

5.10+0.60
(4.16—5.66) М = 5
4.68+0.42
(3.91—5.16)
4.76+0.54
(3.91—5.24)
0.21—

No. var.
1.06+0.10
(0.99—1.25) М = 5
0.42+0.06
(0.33—0.50) М = 4
4.20+0.31
(3.66—4.58)
3.52+0.31
(3.00—3.83)
2.25+0.35
(1.83—2.58)
0.35+0.06
(0.25—0.42)
0.77+0.17
(0.50 —1.00)
5.26+0.44
(4.58—5.66)
4.08+0.33
(3.58—4.49)

Female (N = 5)

3.0—4.0

4.53+0.27
(4.16—4.91)
4.49+0.23
(4.24—4.83)
4.27+0.22
(4.07—4.65)
0.20+0.05
(0.17—0.25)
1.00+0.09
(0.91—1.04)
0.44+0.04
(0.42—0.50)
4.05+0.17
(3.90—4.07)
3.29+0.15
(3.15—3.49)
2.04+0.19
(1.83—2.32)
0.38=0.09
(0.25—0.50)
0.68+0.07
(0.58—0.75)
4.95+0.22
(4.73—5.23)
3.64+0.22
(3.49—3.94)

Male (N = 5)

3.5-4.0

4.46+0.31
(4.15—4.90)

4.58+0.32
(4.23—4.98)

4.19+0.32
(3.90—4.65)
0.252—

No. var.
1.04+0.11
(0.91—1.16)
0.46+0.50
(0.42—0.50)
4.03+0.20
(3.90—4.32)
3.32+0.30)
(3.07—3.74)
2.20+0.35
(1.83—2.57)
0.58+0.17
(0.49—0.83)
0.72+0.08
(0.66 —0.83)
5.08+0.30
(4.81 —5.48)
3.74+0.26
(3.49—4.07)

FIG. 110. Illustration to show three shell orienta-

W of A —

tions for measurements.

Description

island. Snails were collected from the bottom
surface of stones along the protected inner
edge of the breakwater where there was an
expansive shallow water area some 30 cm
deep and with emergent vegetation. Associ-
ated fauna included: Stenothyra hunanensis,
Semisulcospira sp, a species of Viviparidae, a
planorbid, and Radix sp.

Shells. Shells measurements are given in Ta-
ble 57 for the two populations. Shells are il-
lustrated in Figures 109—113. How measure-
ments are made for these globose shells is
given in Figure 110. Shells are medium to
long in length. They are nearly round in out-
line with the apex to base alignment as in

Figure 110C; they are globose and smooth.

The shell is dominated by the body-whorl.

272 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 111. Shells of Lithoglyphopsis modesta from Anhua (D87-1) А-С; “L.” liliputinus, D; and Guoia viri-
dulus, paralectotypes, E, F. Shell A is 4.72 mm long; other shells printed at same scale.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 273

FIG. 112. Shells of Lithoglyphopsis modesta from Baisha (86-B), A-D. A, B = males; C, D = females. “L.”
liliputinus, E, F. A = 4.72 mm; other shells printed to same scale.

274 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

ЗИМЕ

FIG. 113. SEM photographs of shells of Lithoglyphopsis modesta. А-С. Enlargements of apical whorls.
Spiral microsculpture is seen on second and third whorls. E-H. Enlargement of apertural areas showing
basal crescent (Bc), apertural beak (Ab), and beak or adapical apertural groove (Agr).

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 275

TABLE 58. Lengths (mm) or counts of non-neural organs and structures of Lithoglyphopsis modesta.
Mean + standard deviation (range). N = number of snails used. “See text for discussion of pallial
oviduct length.

Females Males (N = 2)
Body 7.52+1.52 7.85
(4.96 —9.40) М = 6 (7.50, 8.20)
Digestive gland 3.01 +0.60 3.65
(2.16-4.0) М = 6 (3.10, 4.20)
Gonad 1.96+0.21 3.35
(1.10-2.20)N = 6 (2.70, 4.00)
Posterior pallial 1.43+0.06 —
oviduct (1.40— 1.50) М = 6
(= albumen gland)
Anterior pallial 2.89 —
oviduct NEA
(= capsule gland)
Total pallial oviduct 4.23+0.33 —
= PO (3.70—4.60)
Bursa copulatrix 0.87 —
(0.80, 0.94) М = 2
Duct of Бигза 0.31+0.09 —
(0.22-0.40) М = 3
BUEZR® 0.22 —
(0.18—0.25) N = 2
Seminal receptacle 0.19 —
(0.18, 0.20) N = 2
Duct of seminal 0.33+0.04 —
receptacle (0.30—0.38) N = 3
Buccal mass 1.60+0.13 —
(1.50—1.78) М = 4
Mantle cavity 4.57 +0.26 4.08
(4.28 —-4.86) М = 5 (3.96, 4.20)
Osphradium 1.94+0.22 1-15
= Os (1.76—2.30) (1.0, 1.30)
Gill 4.20+0.26 3.70
= (4.28—4.86) (3.60, 3.80)
OST (© 0.47+0.05 0.31
(0.39—0.48) (0.26, 0.36)
No. filaments 39+2 41
(38—41) М = 4 no variation
Gf, 0.43+0.19 male + female
(0.20—0.80) М = 16
Gf, 1.04+0.14 male + female
(0.70-1.30)N = 16
Total Gf 1.48+0.21 male + female
= TGF (1.06—1.90) N = 15
GEZZIGE 0.29+0.12 male + female
(0.13—0.41) М = 16
Prostate — 2.52
(1.84, 3.20)
Seminal vesicle — 1.60
no variation
Penis = 3.67
(3.60, 3.74)

o 5

276 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 59. Radula statistics for Lithoglyphopsis modesta. Mean
except for the width of the central tooth in jm. N = number measured.

Greatest shell dimension
Radular length

Radular width

Total rows of teeth

No. rows of teeth forming

Central tooth width

Greatest shell dimension
Radular length

Radular width

Total rows of teeth

No. rows of teeth forming

Central tooth width

Females (N = 10)
Anhua Population (D87—1)

4.92+0.47
(4.00—5.36)
2.52+0.20
(2.26—2.90)
0.23+0.02
(0.20—0.27)
69+5
(62—77)
31=4
(25—36)
554
(48—58)

Baisha Population (1986)
4.79+0.25
(4.60—5.16)
2.60+0.36

standard deviation (range). In mm

Males (N = 7)

4.91+0.26
(4.44—5.20)
2.44+0.16
(2.20 —-2.66)
0.23+0.01
(0.22—0.24)
66+3
(61—70)
25+5
(19—32)
55-1
(54—56)

4.77 +0.32
(4.28—5.12)
2.78=0.13
(2.10—3.00)
0.21+0.01
(0.20—0.22)
74+3
(71—80)
37=2
(35—39)
49=2
(48—52)

TABLE 60. Cusp formulae or the radular teeth of Lithoglyphopsis modesta with the percent of radulae т
which a given formula was found at least once. N = number used and/or counted.

Central Teeth Lateral Teeth Inner Marginal Teeth Outer Marginal Teeth

Anhua Population (N = 10)

1 60% 0-1-3 80% 6 30% 6 20%

2-2

Ae 50% 2-1-0 70% U 90% 7 90%

3-3

Aue 40% 3-1-0 60% 8 90% 8 70%

3-2

“AS 10% 0-1-2 40% 9 30% 9 20%

2-3

4-1-0 10% 10 0 10 10%
X* = 7.4+0.8 7.5+0.9
N = 100
Baisha Population (N = 3)
1. 67% 0-1-3 100% 5 — 5 33%
2-2
2-1-0 67% 6 33% 6 100%
AE 567%
2-3 0-1-2 67% Y 67% 7 100%
1 33%

32 3-1-0 33% 8 67% 8 33%
X* = 7.40+0.9 6.3+0.6
N = 30

*Mean + standard deviation of cusp number for all teeth counted.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 277

TABLE 61. Dimensions of neural structures from five individuals of Lithoglyphopsis modesta. Mean +
standard deviation (range). In mm; L = length. * = neural elements measured to calculate the RPG

ratio.

Cerebral ganglion L

Pleural ganglion L
Right (1)*

Left
Cerebral commissure L

Pleuro-supraesophageal
connective L (2)*
Pleuro-subesophageal
connective L
Supraesophageal
ganglion L (3)*
Subesophageal
ganglion L
Osphradio-mantle nerve L
Pedal commissure L

RPG ratio* (2 + 1+2+3)

RER

1.0 mm

FIG. 114. Head of Lithoglyphopsis modesta from a
snail collected from D87-1.

The apex is a protruding nipple, a scant 7 to
8% of the overall shell length. In side view,
with axis of coiling vertical, the outer lip is
straight and slanted back (to the right) 35° to

0.55+0.07 (0.46 —0.64)
0.26+0.04 (0.20 —0.30)
0.18+0.02 (0.16—0.20)
0.30+0.07 (0.20—0.38)
0.19+0.03 (0.16—0.24)
0.02 (n variation)
0.29+0.03 (0.24—0.32)
0.20+0.03 (0.18—0.24)
0.27+0.08 (0.20—0.30)
0.23+0.11 (0.10—0.30)
0.26+0.03 (0.23—0.30)

the axis of coiling (Fig. 109, column 3). The
shell length (L) is always with the axis of coil-
ing vertical. Greatest overall length (L of A) is
with the shell resting on the aperture (Fig.
110C). The aperture is measured by tilting the
axis of coiling so that the plane of the aperture
is horizontal (Lap, Fig. 110B). There is a wide
columellar shelf 0.32 to 0.36 mm wide (Cs,
Fig. 110B); the edge at the aperture is straight
to slightly arched. There is a pronounced
crescent-shaped ridge to the left of the col-
umellar shelf with a somewhat depressed
concavity, the basal crescent (Bc), between
the ridge and the columella. The peristome is
complete. There is no umbilicus. The adapical
end of the aperture is produced into a beak-
like projection (Ab, Fig. 113G). The aperture
shape is round to broadly pyriform. The apex
of some individuals may have a reddish color;
the cleaned shell is horn yellow.

The SEM pictures of varied aspects of the
shell are shown in Figure 113. The apical
whorls are shown in Figure 113A-D. They are
invariably eroded. Spiral microsculpture is ev-
ident (Fig. 113A, B, D). Figure 113E-H shows
variation in the size of the basal crescent (Bc).
The apertural beak (Ab) is featured in Figure
113F-G showing the groove (Agr) running in-
side it.

External features. The head is shown in Fig-
ure 114; it is broad and squat; there are pro-
nounced bulging eye lobes at the base of
each tentacle. In these character-states the
head is similar to those of Fenouilia and La-

278 DAVIS, CHEN, WU, KUANG, XING, Ll, LIU 8 YAN

FIG. 115. Opercula of Lithoglyphopsis modesta. A, B from Baisha snails; C, D from Anhua snails. A, C.
External surfaces; B, D. Internal surfaces.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 279

Af Apg

AAA RR |

1.0mm

FIG. 116. Masses of anterior pedal glands (Apg) in
anterior foot seen through dorsal surface.

cunopsis. The granules shown on the head
vary in color from bright lemon yellow to light
yellow to white. There is a pronounced om-
niphoric groove; no suprapedal fold. The
operculum is shown in Figure 115; it is regu-
larly ovate, corneous, paucispiral without dis-
cernable internal pad; without ridges. The at-
tachment area is barely distinguishable.

Opening the anterior foot dorsally and re-
moving the buccal mass and supporting mus-
cle bands, one sees a mass of tubular ante-
rior pedal glands (Figure 116). These come
up as a bunch over the elongated pedal com-
missure.

Mantle cavity. The reflected mantle is shown
in Figure 117A. The mantle cavity is typical for
those of species of Triculini in which the sper-
mathecal duct (Sd) enters the pericardium
and the pericardium (Pe) swells out into the
mantle cavity. Organ measurements and sta-
tistics are given in Table 58.

The osphradium (Os) is long; the posterior
end may be considerably narrowed (Figs.
117, 119C). Variation in osphradial shape is
shown in Figure 119C. The terminal gill fila-
ment Gf, is short, that is the ratio is 0.29 +
0.12 (Table 58). The length of the longest fil-
aments is 1.48 + 0.21 mm. In lateral view the
filament is pleated and Gf, has a pronounced
dome (Fig. 117B).

The pericardium bulges out into the mantle
cavity and the opening into the pericardium
for sperm entry is clearly observable (Ope,
Fig. 117A). The mantle cavity organs and ar-
rangement is typical of the taxa of the Tricula
clade of Triculini.

Female reproductive system. The body of an
uncoiled female without head and with kidney
tissue removed is shown in Figure 118. Mea-
surements of the relevant organ systems are
given in Table 58. Important features to note
are: (1) The body is squat and fat as would be
expected from shell shape. (2) The dorsal
surface is densely pigmented with melanin.
(3) The posterior pallial oviduct (Ppo) makes
a pronounced bend over the style sac. (4) The
gonad covers the stomach. (5) The sperm en-
ter the system at the rear of the mantle cavity
through the pericardium (Ope, Fig. 117A).
The pericardium swells out into the mantle
cavity. The spermathecal duct (Sd) is a short
tube running from the pericardium to the swol-
len section of the oviduct (Fig. 119A). (6) The
oviduct makes a tight twist or coil dorsal to the
bursa copulatrix. (7) The seminal receptacle
(Dsr) arises from the oviduct posterior to the
duct of the bursa (Dbu, Fig. 119A). (8) The
bursa complex is dorsal to the posterior pallial
oviduct (= albumen gland). (9) The oviduct
opens into the pallial oviduct close to the pos-
terior end of the mantle cavity, i.e. not into the
posterior end of the pallial oviduct. (10) The
bursa is short. (11) The albumen gland length
is short if the albumen gland is measured
from the posterior edge on the stomach
thereby not including the bend over the style
sac (ratio of 0.35 + 0.02). However, measur-
ing along the bend the actual length (func-
tional length) averages 1.83 mm and this di-
vided by that pallial oviduct length averages
0.45 + 0.03, i.e. the functional length is stan-
dard.

Character-states 5—7 are the same as those
found in Tricula, Fenouilia, and Delavaya.

Male reproductive system. The body of an un-
coiled male is shown in Figure 120 without
head and with kidney tissue removed. Mea-
surements of relevant organs are given in Ta-
ble 58. Important features are: (1) The gonad
consists of moderately large lobes draining
into a vas efferens (Ve). (2) The posterior vas
deferens arises from the vas efferens at mid-
gonad or slightly posterior to mid-gonad and

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

280

зиэшец 116 aybuis y 'g ‘umous ae (Чо 9 95) зиэшец 1116 ¡e jon 'зиебло jo xajduoo (ng) xuyejndos esinq pue
(ag) unipreoued “(9u3) Аулеэ ejueu jo pue Jo11a]sod 0} diysuoyeja4 1194} pue seinjonys AyAe9 ajyuew Buimoys (y) эзиеш pajoa¡ja1 pue IND “ZL 1 ‘94

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 281

FIG. 118. Uncoiled female Lithoglyphopsis modesta with head and kidney tissue removed.

282 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 119. Bursa copulatrix complex of organs; A, B.
Variation in shape of the osphradium, C; refer also
to Os, Figure 117A. In Figure A, organs are in the
same orientation as in Figures 117 and 118. How-
ever, here the bursa copulatrix is cut away to show
the oviduct coil (Coi) and the seminal receptacle
(Sr) where it attaches to oviduct. B. Variation in the
shape of seminal receptacle.

forms a loosely coiled seminal vesicle (Sv)
visible to the left side of the gonad. (3) The
gonad covers the stomach (is ventro-lateral to
it). (4) The prostate is comparatively massive,
covering the style sac and part of the anterior
chamber of the stomach. (5) The penis is sim-
ple, without lobes (Fig. 121B). The anterior
end is extremely slender, that is, a long penial
filament (Pf). There is no papilla. (6) There is
a slender muscular ejaculatory duct (Ej) be-
neath the base of the penis in the neck. (7)
The shaft of the penis arises to the right of the
snout-neck mid-line (x, Fig. 121A at an angle
of. 307 = 02:

Digestive system. The digestive gland covers
the posterior chamber and most of the ante-
rior chamber of the stomach (Figs. 118, 120)
in males and females. The buccal mass and
salivary glands are shown in Figure 122A.
The paired salivary glands (Sg) are massive

and run posteriorly over the nerve ring. The
radular sac (Rs) is extremely elongate and
coils up dorsally along the buccal mass (Bm)
between the buccal mass and the right cere-
bral ganglion (Rcg, Fig. 122B).

The radula is shown in Figure 123. Teeth
counts and statistics are given in Tables 59
and 60. Diagnostic features are (1) the large,
singular triangular anterior cusp of the central
tooth, (2) the massive basal cusps on the face
of the central tooth, and (3) the massive dom-
inant cusp on the lateral tooth.

The stomach (Fig. 124) has a relatively
slender posterior chamber (Pst) that is
slanted from ventral to dorsal along its whole
length and is buried dorsal to the digestive
gland. The digestive gland covers part of the
anterior chamber (Ast) as does the prostate.
There is no caecal appendix.

Nervous system. There are several notewor-
thy features of the nervous system (Fig.
122B). (1) The RPG ratio has a mean of 0.26
(Table 61); the pleuro-supraesopheageal
connective is short. (2) The cerebral commis-
sure is elongate (exceeds 0.12 mm). (3) The
cerebral ganglion have melanin splotches
(Fig. 122B). (4) The pedal commissure is
elongate (exceeds 0.12 mm). (5) The osphra-
dio-mantle nerve is elongate (exceeds 0.12
mm).

Remarks

Lithoglyphopsis modesta could only be
confused with various species of Lacunopsis
on the basis of shell. Anatomically they differ
as the latter has several accessory seminal
receptacles; the generalized seminal recepta-
cle is lost.

Liu et al. (1980) described Lithoglyphopsis
ovatus, L. grandis, Lacunopsis yunnanensis
and L. auris as new species from Yunnan,
China. On the basis of the shell illustrations
given, and in the absence of anatomical data,
it is not possible to ascertain generic status or
discuss relationships. Lacunopsis yunnanen-
sis has a shell with a pronounced keel. That
together with the central tooth morphology
given indicates a possibility that a species of
Fenouilia is involved.

The shells of Lythoglyphopsis liliputanus
(Gredler, 1881) look like a miniature L. mod-
esta (Figs. 111D, 112E, F). Nothing is known

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 283

FIG. 120. Uncoiled male Lithoglyphopsis modesta without head and with kidney tissue removed. Two
bundles of gonadal lobes were cut away.

284 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

—o

1.0mm

FIG. 121. A. Relationship of base of penis (Bp) of
Lithoglyphopsis modesta to snout-neck mid-line (x).
B. Penis

of this species except for the shells. The type
locality is Lien-dshou-ho in Kwangtung [=
Guangdong Province], immediately south of
Hunan (Yen, 1939). The lectotype and para-
lectotypes are in the Bozen Museum, No. 109
(Zilch, 1974).

It is probable that there are additional spe-
cies of Lythoglyphopsis. The closest relation-
ship of L. modesta to any other Triculinae thus
far studied is with Fenouilia kreitneri (see
Davis et al. 1983). Both species have a fat
body, thus fitting a shell shape that is globose
or rather trochoid. The shells are quite differ-
ent. Lythoglyphopsis modesta has a globose-
conic shell where as Fenouilia has a trochoid
to trochoid-ovate shape. The shell of the
former is smooth, that of the latter has keels.
The characters of the head are the same. The
female reproductive system is virtually the
same except for two substantive differences.
One is the very short spermathecal duct of the
former bridging the oviduct to the pericardium
where the oviduct turns towards the bursa
complex. The spermathecal duct is elongated

1.0mm

a
FIG. 122. A. Dorsal aspect of the buccal mass of
Lithoglyphopsis modesta. B. Dorsal aspect of nerve
ring.

in the latter. The other is the bending around
of the posterior pallial oviduct thus covering
the bursa complex of organs (in the former).
In the latter, the pallial oviduct does not bend;
the bursa is posterior to the pallial oviduct.
The male reproductive system is the same in
both taxa with one substantive difference; the
penis of the former has a long penial filament
lacking in the latter. The radular sac is ex-
tremely elongated and coils up dorsally along
the buccal mass; it is short, that is, general-
ized in length in the latter (as also in Lacunop-
sis). The cerebral commissure of the former,
is over double that of the latter. The gill fila-
ments are different. In the former they are of
moderate length with a short section of Gf,
clearly seen. In the later, the gill filament Gf,
is elongated; Gf, is not visible with the mantle
reflected.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 285

FIG. 123. Radula of Lithoglyphopsis modesta. A, E segments of radula. C, D, G, H. Central and lateral teeth.
B, E, F. Lateral, inner and outer marginal teeth.

286 DAVIS, CHEN, WU, KUANG, XING, Ll, LIU 8 YAN

nn Es

1.0 mm

FIG. 124. Ventral aspect of the stomach of Litho-
glyphopsis modesta.

However, given this series of differences
between the two taxa, without more data for
more species, it is not possible to define ge-
neric limits clearly.

Tricula Benson, 1843

Type Species. Tricula montana Benson,
1843: 465

Type locality. Bhimtal, N. India
Designation. By monotypy

Assigned species. Based on anatomical
study; type species and Chinese species
only. T. montana Benson, 1843; T. bam-
booensis Davis & Zheng, 1986 (in Davis et
al., 1986b); 7. boilingi Davis, 1968; T. grego-
riana Annandale, 1924; T. hudiequanensis
Davis & Guo, 1986 (in Davis et al., 1986b); T.
ludongbini Davis 8 Guo, 1986; T. xianfengen-
sis Davis & Guo, 1986; T. xiaolongmenensis
Davis & Guo, 1986. T. gredleri Kang, 1986; T.
maxidens Chen & Davis, sp. nov; T. odonta
Liu, Zhang 8 Wang, 1983a. N = 11.

Diagnosis. Shells small to medium, ovate-
conic. Central tooth as in Neotricula. The ovi-
duct makes a tight 360° twist dorsal to the
bursa. The spermathecal duct enters the peri-
cardium; the spermathecal duct starts as a
wide duct when diverging from the oviduct
and narrows to a slender duct at the pericar-
dium. The duct of the seminal receptacle
arises from the oviduct, from the base of the
duct of the bursa, or from the duct of the
bursa. The duct of the bursa runs undimin-
ished in diameter into the oviduct close to the
opening of the oviduct into the albumen
gland. Some species of Tricula have a peri-

FIG. 125. Shells of Tricula gredleri. Figure A is 3.2
mm long; other shells are printed at same scale.

cardial bursa (Davis et al., 1986b: 515, fig.
40). The type species has a pericardial bursa
(Davis et al., 1986a: 434, fig. 6A).

Tricula gredleri Kang, 1986

Syntypes. Hubei Medical College, Depart-
ment of Parasitology, Wuhan City, Hubei
Province, People's Republic of China. Fig-
ured in Kang (1986: pl.1, fig. 2). Holotype not
segregated in Kang's collection. SMF
305653-305654

Type locality. Maluxi, “(28°56’М, 109°92’E)”,
Orientalis Commune, Guzhang County,
Hunan Province. Collected 18 Oct. 1983.

Etymology. Named for Vincent Gredler, the
German malacologist who first described Chi-
nese Tricula from Hunan Province.

Habitat

Specimens studied for this paper were col-
lected from a small stream at Yantuo Village,
Xiaoguanping Town, Guzhang County,
Xiangxi Prefecture; 28°42'7"N, 110%0'37"E
(Fig. 1, site 12), 20 May 1987, Li Chi-Jian.
The habitat was 700 m at the edge of the
stream shaded by short shrubs. The name of
the stream is You Shui He (he = small
stream). The field number is D87-3.

Depository
Specimens are catalogued into the collec-

tions of ZAMIP, M0008; ANSP 373141,
A12657.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 287

304un —

FIG. 126. SEM photographs of shells of Tricula gredleri. A, B. Whole shells of mature individuals; C—D.
Magnified view of the aperture showing small but deep umbilicus and pronounced apertural notch. E, F.
Eroded protoconchs; teleoconch begins at 1.75 whorls.

288 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

3984

FIG. 127. Opercula of Tricula gredleri. Outer surface at left (A); inner surface to the right (B).

Description

Shell. Shells are small, ovate-conic, and
smooth (Figs. 125, 126). Virtually all shells
are eroded at the apices so that a true under-
standing of length is not possible. Lengths do
exceed 3.30 mm (Table 62). The aperture is
pyriform. Adapically there is a pronounced
apertural beak with beak tubercle. There is no
internal notch groove. There is no adapical
outer lip angle.

The whorls at the suture are smooth (not
crenulated). There is an umbilical chink. The
inner lip is straight abapical to the beak tuber-
cle; it is separated from the body whorl.
Adapically shell material may be layered be-
tween the lip and the body whorl to the point
of fusion. The inner lip is uniformly thin. An
apertural sinus is seen when the outer lip is
seen in side view. Abapical to the sinus the lip
is straight to slightly scooped forward. In side
view, the inner lip is straight; there is no spout,
no varix.

The abapical lip projects beyond the base of
the shell 0.38 + 0.02 mm.

External features. The head is translucent;
there is no deposition of melanin pigment.
There are no white or yellow granules about

the eyes. The operculum (Fig. 127) is cor-
neous and paucispiral. It is unique in being
especially thickened abapically and sitting
cap-like on the foot. The attachment pad is
prominent, some 56% the width of the oper-
culum.

Mantle cavity. Mantle cavity organs are
shown in Figure 128. Organ measurements
and counts are given in Table 63. The osphra-
dium is positioned slightly anterior to mid-gill;
it is elliptical and short. There are 13 to 19 gill
filaments with both Gf, and Gf, elements
prominent. The longest gill filaments are 0.48
+ 0.09 mm long. Gill filament section Gf, is
long. In side view, the largest gill filaments are
moderately domed.

The portal for sperm entry into the pericar-
dium (Ope) is shown. The pericardium (Pe) is
considerably swollen and there is an enor-
mous pericardial bursa (Pbu) pushing the lin-
ing of the mantle cavity far forward into the
mantle cavity.

Female reproductive system. An uncoiled fe-
male without head and with kidney tissue re-
moved is shown in Figure 129. Measure-
ments of organs are given in Table 63.
Important features are: (1) The gonad is pos-
terior to the stomach. It is small and consists

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 289

Pbu

FIG. 128. Mantle cavity of Tricula gredleri showing the enormous pericardial bursa (Pbu) bulging into mantle
cavity.

TABLE 62. Shell measurements (mm) of Tricula gredleri syntypes and of shells of snails used for
dissections. All shells that had eroded apices = e. Mean + standard deviation (range). N = number
measured. M = male, F = female.

Syntypes (N = 2) For Anatomy (N = 5)
1: 2:

No. Whorls 4e 5e Зе—4е

Length (L) 3.08e 3.12e 3.16 (4e,M); 3.32 (4e,F)
Width (W) 1.40 1.36 1.52+0.11 (1.36—1.64)
L last three whorls 2.80 2.76 3.01+0.11 (2.84—3.16)
L body whorl 2.04 2.08 2.20+0.11 (2.12-2.36)
L penultimate whorl 0.44 0.40 0.54+0.02 (0.52—0.56)
W penultimate whorl 0.92 0.96 1.06+0.05 (1.00—1.12)
W 3rd whorl 0.64 0.60 0.71+0.04 (0.68—0.76)
L aperture 1.48 1.48 1.50+0.08 (1.40—1.56)
W aperture 0.80 0.88 0.99+0.05 (0.92—1.04)

x 0.38+0.02 (0.36—0.40)
y 0.18+0.06 (0.10—0.24)

290 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

_ SSS

1.0mm

FIG. 129. Uncoiled female Tricula gredleri with head and kidney tissue removed.

1.0mm

FIG. 130. Variation in gonad structure of female
Tricula gredleri.

of one to four bundles of lobes Figures 129,
130. (2) The bursa copulatrix (Bu) is round to
sub-triangular, minute, and clearly visible
posterior to the albumen gland (Ppo). (3) The
albumen gland is of normal size. (4) The
bursa copulatrix complex of organs is shown
in Figures 131, 132. In Figure 131 the orien-
tation of the bursa complex is the same as in
Figure 129. The 360° twist of the oviduct (Ov)
and the duct of the bursa arising from the ovi-
duct are seen in Figure 131; these two char-
acter-states are seen in all genera of the tribe
Triculini. (5) The species is unique in having
such an enlarged pericardial bursa (Pbu). (6)
In Figure 132A and B, it is shown that the duct
of the seminal receptacle arises from the ovi-
duct slightly posterior to the opening of the
duct of the bursa either slightly offset dorso-
laterally, or directly in line with the opening of
the duct of the bursa (Fig. 132 C). (7) The
seminal receptacle (Sr) is a small sphere. (8)
There is no spermathecal duct, or, at best, a
minute one; the oviduct appears, in gross dis-
section, to be fused to the pericardium (Sdo,

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

291

TABLE 63. Lengths (mm) or counts of non-neural organs and structures of Tricula gredleri. N = number
of snails used. Mean + standard deviation (range).

Body
Gonad
Digestive gland
Posterior pallial oviduct
(= albumen gland)
Anterior pallial oviduct
(= capsule gland)
Total pallial oviduct
= OV
Bursa copulatrix
= BU
Duct of BU
BU + OV
Seminal receptacle
Duct of seminal receptacle
Mantle cavity
Gill (G)
Osphradium (OS)
Os + G
No. of filaments
Buccal mass
Gf,
Total Gf = TGF
GE Gr
Prostate

Seminal vesicle

Penis

SS

*males and females

Females (N = 5)

4.96+0.45
(4.30—5.44)
0.85+0.11
(0.70—1.00)
2.17+0.32
(1.70—2.50)
1.05+0.19

(0.80—1.20) N = 4

1.13+0.15

(1.00—1.30) М = 4

2.14+0.17

(2.00—2.40)
0.31+0.01

(0.31—0.32)
0.17+0.01

(0.16-0.18) N = 3

0.14+0.01
(0.13—0.16)
0.17+0.01

(0.16—0.18) М = 4

0.08 +0.02
(0.06—0.10) N
1.42+0.13
(1.30—1.60)
1.07+0.17
(1.30—1.60)
0.35+0.08
(0.30—0.46) N
0.35+0.09
(0.25—0.46) N
16.8+2.3
(13-19)

0.58+0.09
(0.50—0.68)
0.26+0.06*
(0.20—0.36) N
0.22+0.08*
(0.18—0.26) N
0.48+0.09*
(0.38—0.62) N
0.54+0.03*
(0.50—0.58) N

Il

4

Males (N = 2)

1.97
(1.92, 2.02)

1.55
(1.50, 1.60)
1.11
(1.10, 1.12)
0.31
(0.24, 0.38)

28

(0.22, 0.34)
16.0
(No var.)

0.93
(0.86, 1.00)
0.70
(0.60, 0.80)
2.60
(2.00, 3.20)

292 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Bu

nn

0.33mm

FIG. 131. Details and variation of bursa copulatrix complex of organs relative to style sac (Sts), pericardium
(Pe) and posterior end of mantle cavity (Emc).

TABLE 64. Radular statistics for Tricula gredleri. Mean + standard deviation (range). N = number used.

In mm except for width of central tooth in рт.

Females (N = 3) Males (N = 4)
Shell length 3.41+0.17 3.30
(3.28 —3.60) (3.10, 3.40) N = 2
Radular length 0.49+0.10 0.51 +0.03
(0.39—0.60) (0.47—0.54)
Radular width 0.07+0.002 0.07+0.01
(0.072—0.076) (0.060—0.072)
Total rows of teeth 56+4.4 63.4+4.2
(53-61) (61—70)
No. rows of teeth forming 16.0+2.0 16.3+1.5
(14-18) (15-18)
Central tooth width ESO) 15413414,

(12.8—15.0) N = 8

(13.1-17.1)N = 17

Fig. 132B). (9) Sperm enter the female sys-
tem from the mantle cavity into the pericar-
dium hence to the oviduct.

Male reproductive system. An uncoiled male
without head and with kidney tissue removed
is shown in Figure 133. Two bundles of go-
nadal lobes are cut away to show the seminal

vesicle (Sv). Measurements are given in Ta-
ble 63. Important features are: (1) The gonad
fills the digestive gland (Di) and covers the
posterior and anterior chambers of the stom-
ach. (2) The prostate (Pr) overlaps the poste-
rior end of the mantle cavity (Emc). (3) The
seminal vesicle (Sv) arises from the vas effe-
rens (Ve) about mid-gonad. (4) The seminal

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 293

B

Bu

Dsr

FIG. 132. Details and variation of bursa copulatrix complex of organs of Tricula gredleri. A. Compared with
Figure 131, bursa copulatrix cut off and duct system rotated 90° to show position of seminal receptacle (Sr)
and its duct (Dsr). Complex in A is further rotated in B to show where duct of seminal receptacle (Dsr) opens
into oviduct (Ov). C. Section of oviduct oriented as in Figure 131 showing a variation in position of opening
of duct of seminal receptacle into oviduct, the usual Tricula position.

TABLE 65. Cusp formulae for the radular teeth of Tricula gredleri with the percent of the seven radulae
in which a given formula was found at least once.

Central Teeth Lateral Teeth Inner Marginal Teeth Outer Marginal Teeth
3-1-3 71% 2-[2]-3 86% 8 29% 8 14%
2-2
3-1-3 14% 3-[2]-2 86% 9 57% 9 71%
3-3
3-1-2 14% 2-[2]-4 43% 1075 71% 10 57%
-2
2-1-2 14% 4-[2]-2 14% 11 57% 11 71%
2-2
4-13 14%
3-2 Ñ
X * = 9.69+2.01 9.0+3.1
N = 70 N = 70

“Mean + standard deviation of cusp number for all teeth counted.
[] = blade bifurcates thus resembling two cusps.

294 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

Di

Go

AE
0.5mm

FIG. 133. Uncoiled male of Tricula gredleri without head or kidney tissue. Two sections of gonadal lobes
removed to reveal seminal vesicle coiled dorsal to gonad.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 295

FIG. 134. A. Relationship of base of penis of Tricula
gredleri to mid-line of snout-neck (x) and to poste-
rior end of eye lobes (a). B. Penis.

vesicle coils dorsal to the gonad from mid-
gonad anterior to cover the posterior chamber
of the stomach. (5) The anterior vas deferens
(Vd,) leaves the prostate near the posterior
end of the mantle cavity (Emc). (6) The penis
is unique among triculine taxa (Figure 134B).
The anterior end is blunt with a protudable
papilla (Pa) emerging from the edge of the
square penial tip at the concave side of the
penis. (7) The ejaculatory duct is massive and
extends as a wide muscular duct from the
base of the penis along the dorsal aspect of

TAO ey

FIG. 135. Stomach of Tricula gredleri oriented ex-
actly as in Figures 129, 131, and 133.

TABLE 66. Lengths of neural structures of Tricula
gredleri. Mean + standard deviation (range). N =
number used = 5. * - neural elements measured
to obtain the RPG ratio.

Cerebral ganglion
Cerebral commissare

Pleural ganglion
Right (1)* 0.12+0.02 (0.10—0.14)

Left 0.12+0.02 (0.10—0.14)
Pleuro-supraesophageal
connective (2)*
Pleuro-subesophageal
connective
Supraesophageal
ganglion (3)*
Subesophageal ganglion 0.12+0.02 (0.10—0.14)
Osphradio-mantle nerve 0.07+0.03 (0.04—0.12)

RPG ratio* = 0.37+0.04 (0.33—0.41)
Иез

0.24+0.02 (0.22-0.26)
0.04+0.02 (0.02—0.06)

0.14+0.02 (0.12-0.16)

0.14+0.03 (0.10-0.16)
0.12+0.02 (0.10—0.14)

the neck to the end of the neck. (8) The ori-
entation of the base of the penis to the snout-
neck mid-line is shown in Figure 134A. It is to
the right of the mid-line (“x”), and at an angle
of 45°-50° to it.

Digestive system. The digestive gland is pos-
terior to the stomach of both sexes. The stom-
ach is shown in Figure 135. Two features of
note are: (1) there is a groove (Gr) or crease
between the anterior chamber of the stomach
(Ast) and the section of stomach receiving the
esophagus (Es) and duct of the digestive
gland (Odi). (2) The anterior chamber (Ast) is
yellow with a centrally positioned grey area.
Radular statistics are given in Tables 64
and 65. There appears to be sexual dimor-
phism in the number of rows of teeth on the

296 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 136. Radula of Tricula gredleri. A. Section of radular ribbon. B-E. Central, lateral and inner marginal
teeth. F. Outer marginal teeth. Note pronounced bifurcation of the blade of the major cusp of lateral tooth

(especially in Figs. C—E).

radular ribbon; males have more (61 to 70 on
a radula 3.30 mm long). However, the sample
size is small. The most frequently encoun-
tered formula is

3-1-3 _; 2(3)-[2]-2(3); 9-11; 9-11.
(3)2-2(3)
SEM photographs of radulae are shown in

Figure 136. The dominant cusp of the lateral
tooth (Fig. 136B-E) has a blade that bifur-

cates. Otherwise the tooth morphologies are
standard triculine.

Nervous system. Measurements are given in
Table 66. The RPG ratio of 0.37 indicates that
the dorsal nerve ring is moderately concen-
trated. The pleuro-subesophagal connective
of this species is unusual for its length, equal
to that of the pleuro-supraesophageal con-
nective.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 297

Remarks

Conchologically, Tricula gredleri is most
similar to N. duplicata (Figs. 153-155). Con-
chological comparison with N. duplicata is
given in the remarks section for that species.
Anatomically, 7. gredleri has the generic-level
character-states that serve to differentiate it
from N. duplicata.

Tricula gredleri has seven unique anatom-
ical character-states (Tables 80, 81). The
operculum is cap-like, not a flat sheet (char.
2). The oviduct is fused to the pericardium;
there is no discernable length of spermathe-
cal duct (chars. 19,20). The penial opening of
this species arises form the concave edge of
a blunt tip (char. 34). The anterior chamber of
the stomach is lemon yellow (char. 40). The
pleuro-subesophageal connective is long
(char. 48).

Tricula gredleri belongs to the species
group within Tricula that has a pericardial
bursa T. montana, T. gredleri, T. gregoriana
and T. maxidens. Its closest relationship an-
atomically is with T. odonta.

Tricula maxidens Chen & Davis, sp. nov.
Holotype. ZAMIP, M0009, Figure 137A.

Paratypes. ANSP 373140, A12656, Figure
137B-E.

Type locality. Yantuo Village, Xiaoguanping
Town, Guzhang County, Xiangxi Prefecture;
28°42'7'N, 110%0'37"E. Figure 1, site 12.

Etymology. contraction from the Latin maxi-
mum (the greatest) and dens (tooth).

Habitat

These snails were collected in sympatry
with Tricula gredleri and Akiyoshia chinensis
from a small mountain stream at an altitude of
700 m. The field number assigned was D87-
3.

Description

Shell. Shells are small, smooth, and cylindri-
cal-conic (Figs. 137A-E; 138A-D, F). They
mature at 5.5 whorls with the largest size
class ranging in length from 2.08-2.24 mm
(Table 67). The sutures are deep and the
whorls are slightly convex. Cleaned shells are
glassy; there is no umbilicus. The aperture is
semi-circular with the inner lip a flat-straight
edge. The inner lip is fused to the body whorl.

The outer lip in side view, is straight, without
notch or sinuation. With the outer lip down
and 90” to the horizontal, the inner lip is seen
in side view with a slight angulation. The most
prominent feature is the relatively gigantic col-
umellar tooth seen in the aperture (Fig. 138B)
that continues inside the shell wrapping the
columella adapically. SEM analyses of the
apical whorls indicates that they are minutely
malleated to smooth (Fig. 138D, F).

External Features. The head is devoid of mel-
anin pigment. There are no white or yellow
granules or glands around the eyes. The
operculum (Fig. 138E) is corneous, paucispi-
ral and kidney-bean shaped. The embayment
on the columellar side corresponds to the
large tooth on the columella. The attachment
pad is narrow (33% of operculum width) and
moderately prominent.

Mantle cavity. Mantle cavity organs are
shown in Figure 139A. Organ measurements
and counts are given in Table 68. The osphra-
dium (Os) is mid-gill, and long. There are 10
to 15 well developed gill filaments. The long-
est gill filament are 0.27 + 0.01 mm long. The
Gf, part of the gill is long. Sperm enter the
pericardium at the rear of the mantle cavity
(Ope). The pericardium (Pe) is moderately
swollen and bulges out into the mantle cavity.
A discrete but small and delicate pericardial
bursa (Fig. 139D) was seen in two individuals,
it measured 0.30 x 0.14 mm.

Female reproductive system. An uncoiled fe-
male without head and with kidney tissue re-
moved is shown in Figure 140. Measure-
ments of organs are given in Table 68.
Important features are: (1) The gonad (Go) is
posterior to the stomach; it is small and with
bundles of lobes. (2) The bursa copulatrix is
round and small; half the bursa is dorsal to the
posterior end of the albumen gland (Ppo). (3)
The albumen gland is of standard length. (4)
The bursa copulatrix complex of organs is
shown in Figure 139. The orientation in Figure
139A is the same as that in Figure 140. The
bursa copulatrix (Bu) is short. (5) The seminal
receptacle (Sr) is spherical; it has a short duct
(Dsr) opening into the right ventro-lateral
edge of the oviduct (Ov) just posterior to the
opening of the duct of the bursa (Dbu) into the
oviduct. (6) The spermathecal duct (Sd) is
moderately long; it runs to open into the peri-
cardium (Pe). (7) The pericardium is swollen
with sperm and distends into the posterior
mantle cavity. In some specimens, a minute

298 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

FIG. 137. Shells of Tricula maxidens, A-E; Tricula odonta, F-K. A. Holotype; BE. Paratypes. Length of A
= 1.94 mm; other shells printed to same scale.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 299

FIG. 138. SEM photographs of shells and opercula of Tricula maxidens. A. Shells of mature individuals. Note
straight outer lip (right photo) and prominent columellar tooth. B. Enlargement of aperture showing columel-
lar tooth. C, D, F. Protoconch. Although encrusted, protoconch is seen to be vaguely malleated. E. Opercula:
right operculum shows outer surface with paucispiral growth line; left operculum shows inner surface with
narrow attachment pad.

300 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 67. Shell measurements (mm) of Tricula maxidens. Mean + standard deviation (range). b =

broken apical whorls. N = number measured.

Holotype
No. Whorls 5:5
Length (L) 1.94
Width (W) 0.86
L last three 1.82
whorls
L body whorl 1.24
L penultimate 0.40
whorl
W penultimate 0.68
whorl
W 3rd whorl 0.48
L aperture 0.76
W aperture 0.58
x 0.30
У 0.08

pericardial bursa was observed (Fig. 1390,
Pbu).

Male reproductive system. An uncoiled
male with head and kidney tissue removed is
shown in Figure 141. Lengths of non-neural
organs are given in Table 68. Important fea-
tures are: (1) The gonad (Go) covers the pos-
terior chamber of the stomach and most of
anterior chamber (Ast). (2) The prostate (Pr)
is relatively massive and overlaps the poste-
rior end of the mantle cavity (Emc); it covers
half the style sac. (3) The vas deferens arises
from mid-gonad to slightly anterior to mid-
gonad (Fig. 142). (4) The seminal vesicle
forms a short coil of tubes dorsal to the gonad
and posterior to the stomach. (5) The anterior
vas deferens (Vd,) leaves the prostate
slightly posterior to the anterior end of the
prostate (Fig. 141). (6) The penis is simple,
without papilla; it is slender (Fig. 143B). (7)
There is a pronounced ejaculatory duct (Ej) in
the base of the penis. (8) The orientation of
the long axis of the penial base (Bp) to the
snout-neck mid-line (x) is shown in Fig. 143A.
The penial base is to the right of the mid-line
and at an angle of 40°.

Digestive system. The digestive gland is pos-
terior to the stomach in both sexes. Radular

Paratypes
Large class (N = Small Class (N =
5.5 5.5

2.17+0.06 1.97+0.06
(2.08—2.24) (1.92—2.04)
0.92+0.03 0.82+0.03
(0.90—0.96) (0.80—0.86)
1.95+0.03 1.73+0.08
(1.92—2.00) (1.68—1.82)
1.34+0.04 1.16+0.10
(1.28—1.36) (1.08—1.28)
0.40+0.03 0.35+0.02
(0.36—0.44) (0.32—0.36)
0.71+0.01 0.62+0.02
(0.70—0.72) (0.60—0.64)
0.50+0.01 0.46+0.02
(0.48—0.52) (0.44—0.48)
0.86+0.02 0.73+0.06
(0.84—0.90) (0.68—0.80)
0.62+0.02 0.55+0.05
(0.60—0.64) (0.52—0.60)
0.34+0.02 0.32

(0.32—0.36) no variation
0.08+0.01 0.07+0.03
(0.06-0.10) (0.04—0.10)

statistics are given in Tables 69 and 70. There
are 59.8 + 6.4 rows of teeth along a radula of
0.29 + 0.04 mm (= equivalent of 207 teeth
per mm). The most frequently encountered
formula is

nS 3(4)-1-4; 12-14; 13-14.

The radula is shown in Figure 144; radula
morphology is of the generalized Tricula type.
The central anterior cusp of the central tooth
is long and dagger-like. The dominant cusp of
the lateral tooth is slender and dagger-like.

The stomach (Fig. 145) is the usual type
but with a diagnostic grey streak (Gs) running
across the ventral surface of the anterior
chamber (Ast).

Nervous system. Measurements not taken.
Remarks

Conchologically, among species of Triculi-
nae, T. maxidens is unique and clearly sepa-
rated from all other species (Figs. 153-155).
Unique character-states (Tables 2, 76) are
the cylindrical-conic shape (char. 2) and the
massive tooth nearly filling the aperture (char.
16). The columella within the body whorl has

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 301

FIG. 139. Details and variation of bursa copulatrix of organs of Tricula maxidens. Figs. A and B in the same
orientation as in Figure 140. C. Enlarged seminal receptacle. D. Pericardium with minute pericardial bursa

(Pbu).

a raised spiral keel. The shell seems more
similar to Pseudobythinella than to Tricula.
Anatomically, this species is most closely
allied phenetically to 7. xiaolongmenensis
(Figs. 156-158). It differs from that species in
11 character-states (26%); see Tables 80, 81.
Some of these differences are: In T. maxi-
dens the bursa is covered by the albumen
gland; in 7. xiaolongmenensis the bursa is
posterior to the albumen gland (char. 14). In
the former, the duct of the seminal receptacle
arises from the inside edge of the oviduct; it
arises from the outside edge in the latter
(char. 24). The former has a slight pericardial
bursa; the latter has no pericardial bursa
(char. 25). The radula is short in the former, of

medium length in the latter (char. 41). The
central anterior cusp of the central tooth is
long and dagger-like in the former, the stan-
dard type in the latter (char. 44).

Tricula odonta Liu, Zhang & Wang, 1983a

Holotype. SX 788104 (= IZAS; Beijing, Peo-
ple’s Republic of China). Figure 1.

Type locality Shangnan, Shaanxi Province.
24 October 1978

Distribution. Shangnan, Shangxian, Zhen-
ping, Tongguan, Ninggiang and Yuanqu of
Shanxi Province; Xingshan and Wufeng of
Hubei Province; Henan Province

302 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 68. Lengths (mm) or counts of non-neural organs and structures of Tricula maxidens. N =
number of snails used. Mean + standard deviation (range).

Body
Gonad
Digestive gland
Posterior pallial oviduct
(= albumen gland)
Anterior pallial oviduct
(= capsule gland)
Total pallial oviduct
= OV
Bursa copulatrix
= BU
Duct of BU
BU + OV
Seminal receptacle
Duct of seminal receptacle
Mantle cavity
Gill (G)

Osphradium (OS)

OS + G

No. filaments
Gf,

Gf,

Total Gf = TGF
GEIGE
Prostate

Seminal vesicle

Penis

Females (N = 5)

3.30+0.33

(2.90 —3.68)

0.47 +0.09

(0.36 —0.60)
1.32+0.19

(1.00— 1.48)
0.58+0.11
(0.54—0.70) N = 3
0.67+0.15
0.50-0.80)N = 3
1.23+0.23
(1.00—.150)
0.24+0.05
(0.22—0.30)
0.09+0.01
(0.08—0.10) N = 4
0.20+0.06
(0.16—0.30)
0.07+0.01
(0.06—0.08) М = 3
0.02

N=1

0.81+0.08
(0.74-0.90) N = 3
0.66+0.13
(0.54-0.80) N = 3
0.28

(0.24, 0.32) N = 2
0.39
(0.38, 0.40) N = 2
12.7+2.5
(10-15)
0.17+0.01
(0.16—0.18)
0.11+0.01
(0.10—0.12)
0.27+0.01
0.61 +0.04
(0.57—0.64)

Males (N = 2)

3.24
(3.22, 3.26)
1.31
(0.90, 1.72)
1.41
(1.40, 1.42)

0.75
(0.70, 0.80)
0.65

(0.52, 0.76)

a ии

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 303

FIG. 140. Uncoiled female Tricula maxidens with head and kidney tissue removed.

304 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Ast

|

0.5mm

FIG. 141. Uncoiled male of Tricula maxidens without head or kidney tissue.

FIG. 142. A. Figure to show seminal vesicle (Sv) of
Tricula maxidens in relationship to stomach. Most
of gonad and vas efferens cut away.

TABLE 69. Radular statistics for Tricula max-
idens. Mean + standard deviation (range). N =
number used. In mm except for width of central
tooth in шт.

Females (N = 4)

Shell length 2.19+0.08 (2.08—2.24)
Radular length 0.29+0.04 (0.240—0.204)
Radular width 0.03+0.01 (0.028—0.040)
Total rows of teeth 59.8+6.4 (55-69)
No. rows of teeth 1133159312)
forming
Central tooth 9.6+0.3 (9.0— 10.1)
width (N = 18)

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 305

x

FIG. 143. Tricula maxidens: A. Relationship of base
of penis to mid-line of snout-neck (x) and to poste-
rior end of eye lobes (a). B. Penis.

Habitat

Material for this study was collected on 7
October 1985 from Shenglong Village, San-
huokou Town, Cili County, Changde Prefec-
ture; 2938'35"N, 110°39'12”E; Figure 1, site
7. The field collection number assigned was
D85-82. Specimens came from a small
stream flowing to the Lishui River. The collec-
tor was Dr. Liu Wen Jian of Cili County Public
Health Station.

Depository

Specimens for this study were deposited in
ZAMIP, M0005; ANSP, 368773, A12145.

Description

Shell. Shell measurements are given in Table
71; See Figures 137F-K, 146. Mature males
and females are primarily 7.0 whorls; rela-
tively few are 6.0 or 6.5 whorls (Table 71).
Only one specimen of 7.5 whorls was ever
seen (within the size range of 7.0 whorls
snails of Table 71). Shells are medium sized
with a length range of 3.76 to 4.36 mm for
shells of 7.0 whorls (the dominant size class).
Shells are ovate-turretted. The whorls are
slightly convex; the sutures are deep. The
peristome is complete; the inner lip is widely
separated from the body whorl along its entire
length (0.08 + 0.02 mm, range 0.06-0.10;

TABLE 70. Cusp formulae for the radular teeth of Tricula maxidens with the percent of the seven radulae

in which a given formula was found at least once.

Central Teeth Lateral Teeth

3-1-3 100% 3-1-4 75%
2-2

4-1-4 75%
4-1-4 25%
22 2-1-3 25%
41-3 25% 5-1-4 25%
2-2
3-1-4 25%
2-2
2-1-3 25%
2=
3-1-3. 25%
2-3

Inner Marginal Outer Marginal

Teeth Teeth

12 75% 12 25%

13 100% 13 75%

14 100% 14 100%

15 25% 15 25%

16 25% 16 25%
X* = 13.4+0.9 13.9+0.9
N = 4 N = 40

* Mean + standard deviation of cusp number for all teeth counted.

306 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

FIG. 144. Radula of Tricula maxidens. A. Part of radular ribbon. B, C. Central, lateral and inner marginal
teeth. D, E. Outer marginal teeth.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

LA

In

Ast

Pst Gs

Odi

=
0.5mm

FIG. 145. Stomach of Tricula maxidens.

N=5). The aperture is pyriform. The adapical
end of the inner lip has a tooth-like swelling;
this, plus the thickening of the outer lip oppo-
site the swelling, enclose an adapical notch.
Mid-columella there is a marked indentation.
Although the inner lip is separated from the
body whorl, there is no umbilicus in some in-
dividuals; there is one in some. Facing the
aperture of the shell the adapical outer lip
may be sinuate (16%); in side view, the outer
lip is scooped forward (83%) or straight
(17%). When the outer lip is scooped forward
there is an adapical sinuation or notch. There

307

is no basal post (see Davis et al., 1986). The
cleaned shell is glistening, glassy.

The nuclear whorls (Fig. 146) have a rough,
wrinkled surface. There are 1.5 nuclear
whorls. The roughened surface may not be
discerned at magnifications less than 200 or
300X. The aperture has six diagnostic fea-
tures: (1) There is an adapical notch or beak
bounded by (2) a thickening or node on the
inner lip; (3) The inner lip has an angulation
located some 60% of the length of the inner
lip abapically from the adapical end of the ap-
erture. The inner lip is thickened noticeably
from the fulcrum of the angle to the apertural
notch. (4) There is an indentation at the ful-
crum of the angle. Abapical to the fulcrum of
the angle the inner lip edge is thin. (5) In side
view, the adapical end of the outer lip is
slightly sinuate, the remaining lip is straight.
(6) With the outer lip down, 90° to the hori-
zontal, the inner lip has a slight sinuation that
corresponds to a slight trough seen on the
abapical end of the inner lip in apertural view.

External features. The head of this species is
dark gray. No cluster of white granules or
glands about the eyes were observed.

The operculum is corneous. The typical
paucispiral growth line is not seen; the nu-
cleus is prominent (Fig. 147E-G). The oper-
culum has an odd irregular shape reflecting

TABLE 71. Shell measurements (mm) of mixed males and females of Tricula odonta. Mean + standard

deviation (range). N = number measured.

5

No. specimens (N) N =7

No. Whorls 7.0

Length (L) 4.09+0.22
(3.76—4.36)

Width (W) 1.52 +0.10
(1.48—1.60) N

L body whorl 2.09+0.09
(1.96—2.24)

L penultimate 0.69+0.06

whorl (1.16-1.26) N
W penultimate 1.20+0.05
whorl (1.16-1.26)N = 5

L last three whorls 3.25+0.17
(3.04—3.48)

L apertue 1.34+0.05
(1.28-1.40)N = 5

W aperture 0.89+0.02
(0.88—0.92) М = 5

x 0.37+0.08
(0.28—0.48) N = 4

y 0.19+0.07

(0.12—0.28) М = 4

N=2 N =1
6.5 6.0
3.40 3.00

(3.32, 3.48) =
1.40 1.28
1.86 =

(1.84, 1.88) ds
0.58 0.56

(1.02, 1.08) =
1.06 0.98

(1.02, 1.08) =
2.78 2.48

(1.71, 2.84) =
1.20 1.04
0.85 0.76

(0.84, 0.86) ae

308 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

180UM -

НАТ - 18QUN

FIG. 146. SEM photos of shell apical whorls of Tricula odonta. Teleoconch begins at about 1.50 to 2.0 whorls
(A, C, G). The protoconch is heavily wrinkled (B, D, F, H).

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

190 ———_—______

©

Ssaaum

FIG. 147. Radula (A-D) and opercula (E-G) of Tricula odonta. A. Left lateral, inner and outer marginal teeth.
B-D. Marginal teeth. E, G. Inner surfaces. Note irregular shape of opercula and thickened ridge along
columellar edge of operculum in G. F. Outer surface. All three opercula show the double layers (see text).

310 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN
Coi

FIG. 148. Details and variation of bursa copulatrix complex of organs and mantle cavity structures of Tricula
odonta. A in same orientation as in Figure 149. Not all gill filaments shown. B. Single gill filament.

FIG. 149. Uncoiled female Tricula odonta with head and kidney tissue removed.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

311

TABLE 72. Lengths (mm) or counts of non-neural organs and structures of Tricula odonta. Mean

standard deviation (range). N = number measured.

Females (N = 3) Males (N = 2)
Body 5.83+0.36 4.74
(5.60 —6.24) (4.7, 4.78)
Gonad 0.74+0.15 1.55
(0.60 —0.90) (1.50, 1.60)
Digestive gland 2.45+0.19 2.04
(2.24—2.60) (2.00, 2.08)
Mantle cavity 1.83+0.15 1.53
(1.80—2.00) (1.50, 1.60)
Osphradium 0.34+0.05 0.33
(0.30 —0.40) (0.30, 0.36)
Ctenidium = Gill 1.63+0.15 1.38
(1.50—1.80) (1.36, 1.40)
OS CG 0.21 +0.04 0.24
(0.17—0.25) (0.22, 0.26)
Gill filament No. 29.7+1.5 25.5
(28-31) (24, 27)
Buccal mass 0.57+0.06 —
(males + females; N = 3) (0.50—0.62)
Posterior pallial oviduct 0.92 —
= albumen gland) (0.64, 1.20) N = 2
Anterior pallial oviduct 1.12 —
(= capsule gland) (1.0, 1.24) (N = 2)
Total pallial oviduct 2.04
= OV (1.64, 2.44) М = 2 —
Bursa copulatrix 0.28+0.03
= BU (0.24—0.30) —
Duct of BU 0.21 +0.04
(0.16—0.24) —
BU = OV 0.14
Seminal receptacle 0.11+0.03 —
(0.08—0.14)
Gf, 0.26+0.02* --
(0.24—0.28)
Gf, 0.23+0.02* —
(0.20—0.24)
Total Gf = TGF 0.49+0.01* —
(0.48—0.50)
Gf, + TGF 0.53+0.04*
(0.50—0.58)
Prostate — 0.75
(0.70, 0.80)
Seminal vesicle — 0.66
N = 1
Penis — 1.62
(1.40, 1.84)

*males and females

312 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

Sv In

Sts

qG>EOÁ—— ene
0.75mm

FIG. 150. Uncoiled male of Tricula odonta without head or kidney tissue.

TABLE 73. Radular statistics for males and
females of Tricula odonta. Mean + standard
deviation (range). N = 5. In mm except for width
of central tooth in um.

Shell length: No shell measurements;
radulae came from
heads of snails used
in dissections.

Radular length 0.54+0.04 (0.52—0.62)

Radular width 0.08+0.004 (0.08—0.09)

Total rows of teeth 61.3+4.4 (56-69)

No. rows of teeth 7+1.7 (5-10)
forming

Central tooth width 19.8+2.9 (18.1 —23.6)

the pyriform shape of the aperture and the
PESTE mid-inner lip indentation. It is particularly nar-

row at the abapical end, not broadly or regu-

FIG. 151. A, B. Relationship of base of penis to larly rounded as in those species with an oval
mid-line of snout-neck (x) and to posterior end of | Operculum. The operculum consists of two
eye lobes (a) of Tricula odonta; C. Penis. layers that are readily separated if the oper-

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 313

35338

b
р.

> j ho +
| à <
LA

+25
3:
ae

y) Фи

FIG. 152. Radula of Тисша odonta. A. Рай of a radular ribbon. В. Central tooth. С. Left lateral tooth. D, F.
Left lateral and marginal teeth. Note pronounced bifurcation of the major cusp of lateral tooth. E, G, H.
Marginal teeth.

314 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 74. Cusp formulae for the radular teeth of Tricula odonta with the percent of the radulae in which
a given formula was found at least once. N = number of radulae used.

Central Teeth Lateral Teeth

N=6 N=6
3-1-3 50% 2-1[2]-3 83%
3-3

313 33%
3-1-3 33%
2-2 2-1-2 17%
2-1-2 33%
33 1-1PE2 17%
о 179%
4-4
31-3 17%
4-4

Inner Marginal Outer Marginal

Teeth (N = 3) Teeth (N = 4)
y= 70 2598
8 — 8 50%
9 33% 9 25%
10 66% 10 50%
11 33% 11 25%
12 66% 12 1050
13 66% 13 25%
14 33% 14 25%
15 25%
X* = 11.5+1.6 9.2+1.7
N=3 = 40

*Mean + standard deviation of cusp number for all teeth counted.

TABLE 75. Lengths of neural structures of four
individuals of Tricula odonta. Mean + standard
deviation (range). * = neural elements measured
to attain the RPG ratio.

Cerebral ganglion 0.24+0.03
(0.20—0.26)
Cerebral commissure 0.03+0.02
(0.02—0.06)
Pleural ganglion
Right (1)* 0.12+0.02
(0.10—0.14)
Left 0.12+0.02
(0.10-0.14)
Pleuro-supraesophageal
connective (2)* 0.20+0.07
(0.12—0.24)
Pleuro-subesophageal
connective 0.05+0.04
(0.02—0.10)
Supraesophageal ganglion (3)* 0.12+0.01
0.10-0.12)
Subesophageal ganglion 0.09+0.03
(0.06—0.12)
Osphradio-mantle nerve 0.05+0.02
(0.02 —0.06)
RPG ratio* = 2 + 1+2+3 0.45+0.10
(0.32—0.55)

culum is left too long in dilute Clorox used for
digesting and clearing away tissue fused to
the operculum There is an inner thickened
ridge on the columellar side. The attachment
callus is 54.5 + 3.5% of the width of the oper-

culum; it is prominent and much thickened in
some specimens.

Mantle cavity. Measurements and counts of
mantle cavity structures are given in Table 72.
See Figure 148A. The osphradium is short
(mean ratio 0.22 + 0.04; N = 5). The position
of the osphradium is approximately mid-gill
(Fig. 148A). The slender aspect of the gill fil-
ament (Gf,) is long (mean ratio 0.53 + 0.04;
N = 3). The length of the longest gill filament
is 0.49 + 0.01 mm long (N = 3). The shape of
the filaments is shown in Figure 148B. It is
high domed.

Female reproductive system. The body of an
uncoiled female with head and kidney tissue
removed is illustrated (Fig. 149). Measure-
ments of the relevant organs are given in Ta-
ble 72. Diagnostic features are: (1) The gonad
is posterior to the stomach. (2) The bursa
copulatrix (Bu) is posterior to the pallial ovid-
uct (Ppo) at mid-ventrolateral surface as
shown. (3) The bursa copulatrix complex of
organs is the same as seen in T. bollingi
(Davis, 1968) in that the oviduct makes a tight
loop (twist of 360°) dorsal to the bursa (Fig.
148A). (4) The seminal receptacle is spheri-
cal, posterior to the duct of the bursa (Dbu)
and fused to the surface of the oviduct on the
mid-ventral surface, not from the inner curva-
ture as seen in other species. (5) The sper-
mathecal duct enters the pericardium; the

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 315

pericardium is swollen with sperm. (6) Sperm
enter the pericardium at the posterior end of
the mantle cavity (Ope, Fig. 148A) (7) The
bursa copulatrix is minute (mean ratio, 0.15 +
0.04; N = 3). (8) The length of the posterior
pallial oviduct is standard.

Male reproductive system. The body of an un-
coiled male with head and kidney tissue re-
moved is shown (Fig. 150). Lengths of relevant
organs are given in Table 72. Important fea-
tures are: (1) The gonad either overlaps the
posterior chamber of the stomach or is pos-
terior to the stomach. (2) The prostate over-
laps the posterior end of the mantle cavity and
covers the anterior half of the style sac. (3) The
vas deferens arises from mid-vas efferens to
slightly anterior to mid-gonad. (4) The seminal
vesicle coils on the stomach; in some individ-
uals, the posterior vas deferens was not swol-
len to form the seminal vesicle (presumable
immature, just pre-sperm production individ-
ual). (5) The anterior vas deferens (Vd,)
leaves the prostate slightly anterior to mid-
prostate. (6) The penis is simple, without pa-
pilla (Fig. 151C). The opening of the vas def-
erens is comparatively large and pronounced.
(7) The base of the penis is 0.18 + 0.03 mm
posterior to the eye-lobes (Fig. 151A, B). Vari-
ation in the position of the base of the penis to
the snout-neck mid-line (x) is shown in Figure
151 A, B. A part of the penial base is to the left
of the mid-line. The long axis of the penial base
varies from 65° to 90° to the snout-neck mid-
line. (8) There is a large ejaculatory duct (Ej)
in the base of the penis (Fig. 151C).

Digestive System. The digestive gland is pos-
terior to the stomach. Radular statistics are
given in Tables 73 and 74. Radular morphol-
ogy is the generalized Tricula type (Figs.
147A-D, 152). There are 61 rows of teeth per
radula length of 0.54 mm. The major cusp of
the lateral teeth is massive and split thus giv-
ing the appearance of being two cusps. The
most commonly encountered formula is

3(2)-1 (23; 2(3)-[2]-3; 9-14; 7-14.
3(2)-(2)3

Nervous System. Lengths of neural structures
are given in Table 75. The RPG ratio of 0.45
shows the dorsal nerve ring to be moderately
concentrated.

Remarks

Conchologically, aside from Neotricula ap-
erta, T. odonta has the most distinctive shell

(Fig. 153) in that it attains 7.0 whorls (Tables
2, 76) (char. 29). It has a beak tubercle (char.
24) as do two other species: Tricula gredleri
and N. dianmenensis. The adapical aperture
is widely separated from the body whorl as is
seen in N. cristella (char. 21). The inner lip is
widely separated from the body whorl as it is
in N. cristella (char. 20). A unique state, the
inner lip is differentially thickened (char. 19).
Another unique state, the inner lip has a notch
(char. 17). Among species of Tricula, only this
species of those thus far studied, has an
abapical apertural spout (char. 12), an angled
inner lip (char. 13), and an abapical lip deflec-
tion angle. (char. 18).

Anatomically, Tricula odonta stands apart
from other species of Tricula (Tables 80, 81),
but is most similar to T. xianfengensis (Figs.
156-158). Tricula odonta differs from the latter
in 17 characters (37%). Most notable are the
difference in operculum shape that reflects
differences in shell apertural shape (char. 2),
the differences in Gf, length (char. 8), female
gonad length (char. 12) and seminal recepta-
cle duct length (char. 23). Tricula odonta has
a large ejaculatory duct; the latter does not
have an ejaculatory duct (chars. 36, 37). The
radula of the former is of medium length, of
the latter, long (char. 41). The dorsal nerve
ring is moderately concentrated in the former,
elongated in the latter (char. 42).

MULTIVARIATE RELATIONSHIPS
Shells

The 29 shell characters scored are given in
Table 2, the actual scores in Table 76. All
species of Tricula and Neotricula, as well as
other species having shells resembling those
of Tricula or Neotricula and for which anatom-
ical data are available, were included. The
sources for data for species not treated in this
paper are listed in Appendix |. The pheno-
gram based on UPGMA treatment of distance
coefficients is given in Figure 153. Distance
coefficients were used because the cophe-
netic correlation (phenogram to origninal ma-
trix) was r = 0.928 for distances; r = 0.725
using similarities.

It is clear that one cannot sort species to
genus on the basis of shell characters. Addi-
tionally, of the 21 species, four (19%) could
not be clearly distinguished on the basis of
the shell characters used; Tricula bollingi phe-
notypes 1 and 3 clustered with T. ludongbini,

316

DAVIS, CHEN, WU, KUANG, XING, Ll, LIU 8 YAN

TABLE 76. Scoring of 44 OTUs for 29 shell characters listed in Table 2. Shells from the same population
of some species varied for some characters. Such a species was divided into as many as three OTUs,
e.g. xin-1, xin-2, xin-3. All species for which there were adequate anatomical data were used.

Abbreviations are

: ape, N. aperta; bur, N. burchi; cri, N. cristella; dia, N. dianmenenensis; dup, N.

duplicata; lil, N. Ши; min, N. minutoides; bam, T. bambooensis; boll, T. bollingi; grd, T. gredleri; grg, T.
gregoriana; hud, T. hudiequanensis; lud, T. ludongbini; max, T. maxidens; mon, T. montana; odn, T.
odonta; xin, T. xianfengensis; xia, T. xiaolongmensis; chi, G. chinensis; jin, J. jinghongensis; niz, W.

niuzhuangensis.

Neotricula

1 2 3 4 5 6 Y 8 SOIT

Tricula (N = 11)

13 14 15 16. 117% 18. ey OZ 22

ape bur-1 bur-2 сп-1 cri-2 dia-1 dia-2 дир-1 dup-2 lil-1 Ш-2 тт-1 тт-2 bam-1 bam-2 bol-1 bol-2 bol-3 grd-1 grd-2 grg-1 grg-2

UY 0 0 0 0 0 0 0 0 00
2 4 1 1 0 0 0 0 0 0 OO
34 0 0 0 0 2 2 1 1 Me _1
4 0 0 1 0 0 2 2 1 1 CON
50 1 1 0 0 0 0 0 0 O)
6 0 1 1 1 1 1 1 1 1 al
TO, 0 0 0 0 0 0 0 0 оо
8 2 0 0 1 1 0 0 1 1 оо
922 0 0 1 1 2 2 2 2 WEA
10 1 0 0 0 0 0 0 1 0 оо
O 0 0 0 0 1 1 0 0 uv Y
ve © 1 0 0 0 1 1 1 1 Mesa
Ч 1 1 1 1 3 3 0 1 22
14 0 1 1 1 1 1 1 0 0 о о
152 1 1 0 0 0 1 1 1 We a
16 0 0 0 0 0 0 0 0 0 OO
17 0 0 0 оо 0 0 0 0 OO
18 0 0 0 1 0 1 1 1 1 |
1920 0 0 0 0 0 0 1 1 оо
20 0 0 0 3 3 1 0 2 2 23051
21 0 0 0 2 2 0 1 1 1 о 1
22 0 0 0 OO 0 0 0 0 оо
23 0 0 0 1 1 0 0 0 0 UA
24 0 0 0 0 0 1 1 0 0 оо
25 0 0 0 0 0 0 0 0 0 ит
26 1 0 0 0 0 0 0 0 0 оо
21551 0 0 0 0 0 0 0 0 оо
0 0 0 0 0 0 0 0 0 0 о

0 0 0 0 0 0 0 0 0 (YY

whereas phenotype 2 clustered with 7. xia-
olongmenensis phenotype 1. Tricula xia-
olongmenensis phenotype 2 did not have
close affinity to any other species in the upper
half of the phenogram. Tricula montana phe-
notype 3 clustered with T. gregoriana, not
with phenotypes 1 and 2 of Tricula montana.
Tricula xianfengensis phenotype 2 did not
have close affinity with any other species in
the upper half of the phenogram. While most
shells can be classified readily on the basis of
the characters used (exact measurements
and whorl numbers would allow further dis-
crimination), four of the 44 OTUs (only 9%) do
not cleanly segregate with their other conspe-
cific OTUs.

It is useful to compare the phenogram with

0

SOTO 0, 01 0101010 07 =7 979859 =O) = 1011S ON OSLO 95587 N=

0 0 0 0 0 0 0 0 1 1

90 ©00000000N-000-01+0000-0O 2 ON = O
oOOOO000 00 app “.0O00NO.0O000 0 0 0 00060
Oooooo°o=s+0Ot NFO 0 00 , oO 200 09 9 00 DS SO DO =
(=) foes (>) <> (=) (>) >) >) (> 9892 O eh MO TOMA OOOO ANOS NO (eo te} >)
O0 000 --4 0900 09 99 4-2 “+. 0 00000002 + ©
90 Ce) foley Toy (ole) Co) (eS) 01010101090 оо ооо о ооо
ooooo-jo0enwrveo0ocoo0o0o0oco=/"oqone-"o0co00co=+ +0
ooooo=}?o0oe-nwvroocjcoe-o0oco0o-Onereoooeo-+_o9o
ооо ооо ооо = = © OO o0 0c OO == ©
© O0 0000000000000--12000 000 ON = ©

a plot of the Prim Network (Minimum Span-
ning Tree [MST]) (Fig. 154) based on the dis-
tance coefficients (drawn to scale of actual
pathway distances). The two are useful in as-
sessing those species most closely resem-
bling a species of immediate concern. For ex-
ample, what species might be conchologically
confused with Tricula gredleri? An examina-
tion of Figures 153 and 154 indicates that N.
duplicata is the most closely allied pheneti-
cally. The several differences are then readily
found in Tables 2 and 76. With MST, only two
OTUs of different species do not link with
other conspecific OTUs (i.e. T. montana phe-
notype 3; T. bollingi phenotype 2). Thus, only
5% of the OTUs do not segregate cleanly.
Table 77 provides a ranking of species by a

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 317
TABLE 76. (Continued)
Gamma- Wucon-
Tricula (N = 11) Tricula tricula Jinhongia chona
23 PLS PE) 20 2 28 28 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44

hud-1 hud-2 lud-1 lud-2 max-1 max-2 топ-1 mon-2 mon-3 odn-1 odn-2 xin-1 xin-2 xin-3 xio-1 xio-2 chi-1 chi-2 jin-1 jin-2 niz-1 niz-2

1 1 0 0 0 0 0 0

D JO O1 BR WD =

5
oooo oo 00.1.0002. 000 0+000+00
O0OOOOOOO Pp 10001: 24200 0 0N COCO COON O =
(= {oy (=) fey (=) (=) COTO "TONO f=) (=) fe) fe) ©) (>) (>) ee) SS (=) (>) f=) |) (>) (>)
(=) (>) (=) (>) Ke) (>) te ey К=) (>) (>) (>) (>) Se) Ем (=) ©! fy TO [®) f=) Ce) re) 1) (>) (>)
(=>) (=>)! (Ss) fe) (©) fey foo sy es (=) fe) (=>) OM A SEO ZSZSH SEO (=) (=) t=) f=) (=) (>) (4)
(=) (=>) {>} (=) te Cy ter Е SE Si) (>) = TOMO f=) 91987 fe) Ike) fe) Ce) {> (4%)
(oy f=) (=) (oy) (oJ) Koy fo) Toye te) (=). (oe) f=) fe) 8797 8797 ee) 95982 985 9,09: SO
(=) (=) (=) (=) def (>) ei t=) “ACTO (>) (=) (> (> Cs TO CS (>) (>) Te (=) (=) (=) fe ON
(==) (= (>) (= (= (>) ote) ley 9797 9877 987 =) (=) (= (= Кен) (>) ZZ (>)

0000 -0O0N0N- 0000 р-он +000N 2+ = =

1 1 2 1 0 0 0 и о 0

кооооозоомомы O 2004 .Q + зоо - =
ye) fe) COTO (2) =) = (=7 (=) fe) yey) fe) f=) fe) fee) DS) ey Ss ey)
Ono OO = TOMO TONO 859297 ооо осо осо = Ono Oro) =
SO TO OS SCS Or ON NO SO (SO) © OO ооо Oo O NIO OO" = Onn
ooooooco-rH+00 000 0+00000N0000=> —
oooooocejvwA HS aoao0cdcaoo0ocoocooo0oe--nooco >» = —
oooo-oeetoen4t Ni 00004 но, ++00
oooo}-of0oee2e nN2rsFO O02 200000 0++2+AN O00
S-O0O0000 0 0 000000099 * + ыы 2QO OOO » » O0 O O
Oo=2=-0000000 1.0020 ua1.O000%» +--+ © =
OS "O TOTO осо но O" O TOTO ON O O O = oC SO = 1010 = O
© © © © © © © N° © © © = © © = © N° OOO = = © = © = = ©

TABLE 77. Ranking species on a consistency index for shell characters. The index =

1.0—[number

of character state differences between phenotypes in a population + 29 characters]. Numbers in ()

= number of phenotypes.

N. aperta (1) 1.0

N. cristella (2) 0.97
N. minima (2) 0.97
T. gredleri (2) 0.97
T. maxidens (2) 0.97
T. odonta (2) 0.97
N. burchi (2) 0.93
N. duplicata (2) 0.93
N. ШИ (2) 0.93
T. gregoriana (2) 0.93

T. ludongbini (2) 0.93
N. dianmenensis (2) 0.90
T. bambooensis (2) 0.90
T. hudiequanensis (2) 0.90
G. chinensis (2) 0.90
W. niuzhuangensis (2) 0.90
T. xiaolongmenensis (2) 0.86
T. xianfengensis (3) 0.83
T. bollingi (3) 0.79
T. montana (3) 0.79
J. jinhongensis (2) 0.76

consistency index based on the character-
state differences between OTUs of the same
species (1.0 = no differences). This table
helps clarify the problems of why T. montana
and 7. bollingi do not cleanly link with conspe-
cific OTUs (i.e. three classes of phenotype
varying over six character-states).

Principal Component Analysis (PCA) was
done. There are ten components that account
for 84.6% of the vaiance (Table 78). The first
three components account for only 45% of the
variance. Factor loading of characters for the
first ten PCs are given in Appendix Il. Ordina-
tion diagrams are not provided because: (1)

318 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 78. Principal component analysis of shell data: First ten principal components are listed with the

percentage of variance loading on each component.

Eigenvalue

6.16870
3.56713
3.27486
2.59528
2.45005
1.76301
1.38014
1.30153
1.10735
0.93280

Components

© © © -J O O1 BR OC MN —

—

most taxa cluster too close together at the
center of the plots; (2) the MST results (Fig.
155) are much less satisfactory than those
given in Figure 154; there are 14 mismatches
of phenotypes (i.e. 32%). Three dimensional
scaling produced even less acceptable re-
sults.

PCA analysis is useful in demonstrating the
utility of shell characters used for discriminat-
ing among species. The first component
(Table 79) is not one of size but involves ten
apertural characters that are most useful in
distinguishing among species. As examples:
(1) only four species (N. dianmenensis, N. lili,
T. gredleri, T. odonta) have a pronounced
adapical apertural sinus; (2) only three spe-
cies have an adapical apertural beak tubercle
(T. dianmenensis, T. gredleri, T. odonta); (3)
only T. odonta has a mid-lip inner lip notch
and a shell of 7.0 whorls.

The second component is defined on the
basis of seven components that include
sculpture, size, varix formation, and shell
base characters. The third component in-
volves three (shape, columellar and apertural
groove) characters, while the fourth is defined
by four characters, with teeth and internal
keels as well as sculpture and outer lip orien-
tation.

The Neotricula duplicata cluster of five taxa
(N. duplicata, N. lili, N. odonta, N. dianmen-
ensis, N. gredleri), best seen in Figures 153
and 154, have four to seven of the following
seven character states: an outer lip sinus,
adapertural beak or notch, abapical spout, a
beak tubercle, a thickened inner lip, the inner
lip completely separated from the body whorl,
and an angled or sinuate inner lip.

All species of Neotricula thus far examined,
except the alpha race of N. aperta, have spiral
microsculpture on the shoulders of the teleo-
conch whorls; not so for species of Tricula.

Percent Cumulative
21.27 21.27
12.30 33.57
11.29 44.86

8.45 53.81
8.45 62.26
6.08 68.34
4.76 73.10
4.49 77.59
3.82 81.41
3.22 84.62

Neotricula aperta is unique for its globose-
conic shape, wide columellar shelf, and keel
on the base of the body whorl bordering the
wide columellar shelf. Only T. maxidens has a
cylindrical-conic shell and a large columellar
tooth clearly visible in the aperture. Only J.
jinghongensis has spiral microsculpture at the
base of the body whorl and a basal post. Only
T. xiaolongmenensis has a domed tooth on
the columella inside the body whorl. Only W.
niuzhuangensis has a spiral keel on the col-
umella inside the body whorl that does not
extend basally so as to be seen as a tooth in
the aperture (in contrast to T. maxidens). Only
three species have crenulated whorls at the
suture: N. burchi, G. chinensis, J. jinghongen-
sis.

Anatomy

Thirty OTUs were scored for 48 character
states (Tables 80, 81). AUPGMA phenogram
based on distance coefficient is given (Fig.
156). Distances were used as the cophenetic
correlation because r = 0.960 is superior to
that using similarity coefficients (r = 0.865).
In this phenetic treatment, species do not
group neatly into well-defined generic clus-
ters. However, by eliminating characters with
missing data and using the Prim Network
(MST) (Fig. 157), well-defined generic group-
ings resulted.

In the PCA analysis, ten components were
extracted before eigenvalues dropped below
1.0; these components accounted for 89.8%
of the variance (Table 82). Factor loadings of
characters on each of the ten components are
given in Appendix III. Ordination on the first
two PC axes is given (Fig. 158) with OTUs
connected by the MST based on the original
distance matrix (with characters involving
missing data removed) (Fig. 157). The spe-

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 319

TABLE 79. Shell characters highly correlated on each of nine principal components. * = also loads =
0.440 on another axis.

COMPONENTS
Characters Loading |. APERTURAL/WHORLS
11 —0.849 sinus—adaptical aperture
24 —0.733 beak tubercle—adaptical aperture
13 0.715 inner lip shape
18 — 0.696 lip deflection angle
17 —0.665 notch, inner lip
29 —0.665 shell has 7.0 whorls
12 —0.653 abapical spout
20 —0.636 inner lip separated from body whorl
19 — 0.632 inner lip thickness
9 — 0.606 adapical notch, beak
3 — 0.582 apertural shape
|. SIZE, SCULTPURE, VARIX, SHELL BASE
6 +0.812 spiral microsculpture
5 + 0.653 whorl at suture crenulated
28 + 0.596 basal post
122 +0.497 abapical spout
1 —0.461 size
8 —0.461 protoconch sculpture
23 +0.458 varix
Ш. SHAPE, KEEL, COLUMELLAR SHELF: N. APERTA AXIS
26 +0.810 keel at external base of shell
2 + 0.751 columellar shelf
10 +0.728 inner adapical apertural groove
2 + 0.645 shape
IV. STRUCTURE ON COLUMBELLA INSIDE OR OUTSIDE SHELL;
OUTER LIP
16 = 0.7418 inner lip tooth
7 —0.611 spiral sculpture at shell base
22 — 0.585 internal columellar keel
15 + 0.443 outer lip scooped forward
V. APERTURE, WHORL CRENULATION
3* +0.574 apertural shape
57 — 0.499 crenulated suture
19* — 0.459 inner lip thickened
9* + 0.450 adaptical apertural notch, beak
VI. VARIX, LIP SEPARATION
235 —0.553 varix
21 —0.496 adaptical aperture separated from body whorl
20* — 0.444 inner lip separated from body whorl
VII. LIP ANGLE, SIZE
25 — 0.696 lip angle
Ye 0.570 aize
VIII. LIP SINUATION
14 + 0.734 lip sinuation
IX. PROTOCONCH SCULPTURE
8* —0.494 protoconch sculpture

cies fall cleanly into generic clusters on the
ordination diagram. Through use of Figures
157 and 158, it is clear which species pairs
are most similar anatomically; e.g., Tricula
montana is most similar to T. gregoriana.
The ordination diagram and an assessment
of anatomical characters with high loadings
on each of the ten principal components are

useful for assessing OTU relationships. The
first and second components clearly separate
genera (Table 83). Characters 17, 18, 24, 21
clearly serve to separate generic clusters.
Species with the 360° open oviduct circle
complex of organs with the spermathecal duct
entering the pericardium are to the left upper
and lower quadrants. Species that have lost

320 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

APE N. 2.25
BUR-1 N. 0.48
BUR-2 1.07
MIN- 1 0.38
MIN-2 0.69
BAM-1 0.53

MAX-1 1.03
MAX-2 1.70
JIN-1 1.19

Е =F + + + — ==
2.400 2.000 1.600 1.200 0.800 0.400 0.000

FIG. 153. Phenogram based on UPGMA treatment of distance coefficients involving shell characters from
phenotypes (1-3) of 21 species (Tables 76, 77). G, Gammatricula; J, Jinhongia; N, Neotricula; T, Tricula; W,
Wuconchona. APE, N. aperta; BUR, N. burchi; BAM, T. bambooensis; BOL, T. bollingi; CRI, N. cristella; CHI,
С. chinensis; DIA, N. dianmenensis; DUP, N. duplicata; GRD, T. gredleri; GRG, T. дгедопапа; HUD, T.
hudiequanensis; JIN, J. jinhongensis; LIL, N. ИИ; LUD, T. ludongbini; MAX, T. maxidens; MIN, N. minutoides;
MON, 7. montana; NIZ, W. niuzhuangensis; ODN, T. odonta; XIO, T. xiaolongmenensis; XIN, T. xianfen-
gensis.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 321

[ E BUR1 JINA JIN.2
J

BOL.2]T
T
Bota! BOL.3 fron wap)
CDE
(gun © on) "
я
1.0

T T
MONA MON.2
ODN.1
GRD.2 - Ti
ÓN GRDA ОР

Ne 2
A

FIG. 154. Minimum spanning tree (MST) based on distance coefficients involving shell characters treated in
Tables 2 and 76. The tree is en to scale using inter-taxon distances. See caption for Figure 153 for

defining abbreviations.

the seminal recepatacle and have the sper-
mathecal duct opening into the mantle cavity
are in the upper right quadrant.

While major groupings along the 1st com-
ponent are established by the four highly cor-
related characters given above, species
placements are dictated by the 14 other char-
acters with loadings = 0.678. Tricula gredleri
is far to the left because of character 37; it
alone, of all the OTUs, has a massive ejacu-
latory duct. Wuconchona niuzhuangensis is
far to the right because it shares with Gam-
matricula chinensis character states for char-
acters 21 and 35; the function of the seminal
receptacle is moved to the inside of the ovi-

duct, and the concave edge of the penis has a
white muscular zone. Wuconchona is dis-
placed to the right of Gammatricula due to
characters 17, 28, and 29. Wuconchona is
unique in having a discrete U-shaped bend of
the oviduct, a very short male gonad, and no
vas efferens.

Distribution along the second principal
component is due to the interaction of 12
characters. Tricula gredleri, Neotricula di-
anmenensis, and N. minutoides are at the
bottom of the ordination diagram because of
characters 2, 19, 20, 34, and 40. Unique to 7.
gredleri are operculum shape, fusion of the
oviduct to the pericardium, the lemon color of

322 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

ES

GRD.2

JIN.2

is Fiz MED ZN
/ ATI BUR.1 BUR.2
1 BOL.2

FIG. 155. MST based on shell distance coefficients following PCA analysis. Abbreviations are defined in

caption for Figure 153. See text for details.

the anterior chamber of the stomach, and the
position of the penial opening. Neotricula di-
anmenensis and N. minutoides have few
rows of teeth per radula. Tricula minutoids
has a highly extensible penis. Jinhongia, Wu-
conchona and Gammatricula are at the top
because of the loss of the seminal receptacle
(and duct).

Comparing Figures 156-158, it is seen
which species are most similar anatomically
and the affect of intrapopulation variation on
distance measures. Linkages at a value <=
0.40 (distance coefficients) clearly involve
population variation. Linkages = 0.67 involve
discrete species.

There are species clusters of phenetically
closely related species. The biggest cluster,
one that has the least distances among spe-
cies, involves the Tricula bollingi complex: T.

bollingi, T. bambooensis, T. hudiequanensis
and 7. ludongbini. These species occur in
Yunnan, China. Within this complex, 7. hud-
iequanensis is unique in that it has a scatter of
a few white glands about the eyes (char. 1).
The attachment pad of the operculum is very
wide (char. 4). The shell is most similar to that
of T. bambooensis (Figs. 153-155). However,
there is a 12.5% difference between T. hud-
iequanensis and T. bambooensis (chars. 1, 4,
6, 8, 14, 37). Within the T. bollingi complex, T.
bollingi is unique in that the female gonad is
posterior to the stomach (char. 11), the radula
is of medium length, not short (char. 41), and
the radula has a moderate number of rows of
teeth, not few (char. 42). The shell of 7. bollingi
is most similar to that of 7. ludongbini.

A closely allied species pair involves 7.
montana from India and 7. gregoriana from

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 323

TABLE 80. Anatomical characters and character-states (n = 48) showing one or more differences
among 20 species belonging to the genera Tricula or Neotricula, or where the shells indicate affinity to
these two genera.

—

OI ah WwW №

ZE

22.

. Glands about the eyes
. Operculum shape:

. Operculum:
. Operculum attachment pad:
. Operculum; thick internal ridge:

. Gill filament number:

. Gill filament shape:

. Gf, length:

. Osphradium length:

. Circular grandular mass anterior to

osphradium:

. Gonad:
. Gonad:
. Bursa:
. Bursa:

. Bursa:
. Bursa; duct:
. Oviduct at bursa:

. Spermathecal duct:
. Spermathecal duct:
. Spermathecal duct:

Seminal receptacle:

Seminal receptacle duct:

. Seminal receptacle duct:
. Seminal receptacle arises from:

. Pericardial bursa:
. Albumen bland:

. Gonad:

. Gonad:

. Vas efferens:

. Seminal vesicle:

. Vas deferens—2:
. Penial tip:

. Penis:
. Penial opening:

External Features (N = 5, 10%)
no glands (0), scatter of a few glands (1), dense
concentration of glands (2)
ovate (0), mid-columellar indentation (1), elongate-oval (2),
irregular shape (3), cap-like (4)
single layer (0), two or more layers (1)
wide (0), narrow (1), very wide (2)
none (0), has (1)

Mantle Cavity (N = 5, 10%)
few (0), moderate (1), many (2)
flat (0), moderate dome (1), high domed (2)
long (0), medium (1), short (2)
long (0), short (1)
none (0), has (1)

Female Reproductive System (N = 16, 33%)
behind stomach (0), overlaps stomach (1)
long (0), short (1), very short (2)
short (0), long (1), minute (2)
posterior to pallial oviduct [Ppo] (0), covered wholly or
considerably by Ppo (1), ventral to Ppo (2), postero-lateral
to Ppo (3)
round (0), triangular to subtriangular (1), ovoid (2)
= 0.04 mm long (0), none (1)
makes 360° tight twist (0), runs to Ppo without twist or bend
(1), no twist but has discrete bend (2)
enters pricardium (0), enters mantle cavity (1)
long (0), short (1), none, fused to pericardium (2)
slants at an angle from duct of bursa or oviduct to mantle
cavity, or at an angle from anterior-posterior axis (0), runs
directly anterior from duct of bursa to mantle cavity (1),
none, oviduct fused to pericardium (2)
arises from oviduct or duct of bursa (0), lost, function
removed to inside swelling between duct of bursa and
spermathecal duct (1), lost, function removed to inside
oviduct (2)
U-shaped continuation of duct of bursa (0), arises from
oviduct (1), from juncture of duct of bursa and oviduct (2),
from duct or bursa (3), lost (4)
= 0.02 mm long (0), very short to ductless (1), seminal
receptacle and duct lost (2)
inside edg of oviduct in coil (0), outside edge of oviduct coil
(1), from duct of bursa (2), seminal receptacle lost (3)
none (0), slight (1), large (2), extremely enlarged (3)
normal (0), short (1)

Male Reproductive System (N = 11, 23%)
posterior to stomach (0), overlaps stomach (1)
long (0), short (1)
has (0), does not have (1)
coils posterior to stomach (0), coils also on stomach (1),
coils only on stomach (2)
leaves prostate at Emc (0), leaves prostate at mid-prostate
(1)
no papilla (0), has papilla (1), long penial filament (2)
normal (0), highly extensible (1)
center of penial tip (0), from concave edge of blunt penial
tip (1)

(continued)

324 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 80. (Continued).

35. Penis:

36. Ejaculatory duct:

External Features (N = 5, 10%)

no white muscular zone on concave edge (0), has such
zone (1)
none (0), in base of penis (1), extends from penial base into

the neck (2), only in neck (3)

37. Ejaculatory duct:

38. Digestive bland:

39. Anterior chamber of stomach
[Ast]:

40. Ast lemon yellow color:

41. Radula:

42. Radula:

43. Radula central tooth:

none (0), slight (1), large/prominent (2), massive (3)

Digestive System (N = 9, 19%)

posterior to stomach (0), covers Pst (1)
no streak (0), with grey-melanin streak (1)

no (0), yes (1)

short (0), medium length (1), long (2), very long (3)
no. rows teeth: few (0), moderate no. (1), many (2)
has paired swellings at posterior face of tooth between

innermost basal cusps (1), lacks (0)

44. Radula central tooth:
45. Lateral tooth major cusp:
46. Lateral tooth major cusp:
47. RPG ratio:
(2)
48. Length pleuro-subesophageal
connective:

Yunnan, China. They are similar anatomically
(Fig. 157) and conchologically (Figs. 153,
154). They differ in seven anatomical charac-
ters (15%).

PHYLOGENETIC RELATIONSHIPS

Phylogenies involving 19 genera of the
three tribes of Triculinae were recently pub-
lished (Davis et al. 1990; Davis 4 Kang, 1990;
Davis, 1992). To these are now added the two
additional genera (Guoia, Lithoglyphopsis),
additions made possible because of this
study. In this analysis, only one genus of the
Jullieniini is used (outgroup), and the 13 gen-
era grouped in the Triculini and Pachydrobiini.
The characters and character-states used are
listed in Table 84. There are 17 characters
listed under synapomorphies, an additional
14 as autapomorphies. Scores are tabulated
in Table 85. The result of the Hennig-86 anal-
ysis was 20 equally parsimonious trees with a
length of 37, a consistency index of 86 and an
ri (retention index; Farris, 1989) index of 91.
Four of the trees and the Nelsen consensus
tree are given (Figs. 159-163). All show the
separation of the three tribes with the same
genera assigned to each of the tribes. The
only difference among the 20 trees for genera

central cusp: generalized type (0), long and dagger-like (1)
=)

normal (0), displaced towards ощег edge (1)

moderately concentrated (0), elongated (1), concentrated

usual (0), none (1), long (2)

of the Triculini was the tricotomy of Fenouilia,
Lacunopsis, Lithoglyphopsis or, alternatively,
the divergence of Fenouilia from Lacunopsis
and Lithoglyphopsis. The separation ((Fen-
ouilia) (Lacunopsis, Lithoglyphopsis)) is war-
ranted as Fenouilia has a trochoid shell, and
the other two genera have globose shells.
The greatest shifting about involves genera
of the Pachydrobiini. The consensus tree
reduces the shifting to produce a polycotomy
for Neotricula, Halewisia, Pachydrobia, and
Jinhongia. In all trees, Gammatricula and Wu-
conchona diverge from the same node sepa-
rated from the node that supports the sub-
cluster of ((Robertsiella) (Guoia-a, Guoia-b)).
A single tree resolves with the ordering of
character-states involving (1) the position of
the seminal receptacle and its duct, (2) the
loss of the seminal receptacle and the direc-
tion of evolution of the location of the struc-
tures functionally replacing the seminal re-
ceptacle, and (3) an analysis of homoplasies
(Fig. 164). The generalized position of the
seminal receptacle is a branch off the oviduct
posterior to the duct of the bursa and anterior
to a coiling of the oviduct between the gono-
pericardial duct and the bursa copulatrix (out-
groups Hydrobia, Hydrobiidae; Pomatiopsis,
Pomatiopsinae: Pomatiopsini). This position
is seen in the genera of the tribe Triculini. In
the Triculini and Pachydrobiini, the position of

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 325

CRI N 069
LIL М 091
DUP-1 0.31
DUP-2 N 1.20
BUR N o76
XIO T 0.97
GRG T 1.03
BAM-1 0.33
BAM-2 0.66
BOL T o.82
HUD-1 0.33
HUD-2 0.78
LUD-1 0.38
LUD-2 T 0.19
LUD -3 0.94
XIN T 113
ODN-1 0.31
ODN-2 1.28
JIN-1 0.29
JIN-2 0.99
CHI-1 0.24
CHI-2 1.42
MON T 1.44
MAX T 153
DIA N 147
MIN-1 0.20
MIN-2 1,69
GRD -1 0.26
GRD-2 2.33
NIZ WwW.

к
2.400 2.000 1.600 1.200 0.800 0.400 0.000

FIG. 156. Phenogram based an UPGMA treatment of distances based on а! 48 anatomical characters from
1 to 2 phenotypes of 20 species. See Tables 80, 81. Abbreviations are defined in the caption for Figure 153.

326 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

TABLE 81. Scoring of 30 OTUs for 48 character-states listed in Table 80: bur-1, bur-2, etc. indicates that
shells of Neotricula burchi exhibit more than one character-state for one or more characters. bur, burchi;
cri, cristella; dia, dianmenesis; dup, duplicata; lil, ШГ, min, minutoides; bam, bambooensis; bol, bollingi;
grd, gredleri; grg, gregoriana; hud, hudiequanensis; lud, ludongbini; max, maxidens; mon, montana; odo,
odonta; xian, xianfengensis; xio, xiaolongmenensis; chi, chinensis; jin, jinhongensis; niz, niuzhuangensis.
NC = missing data.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
N.cri N.dia N.dup1 N.dup2 М.И N.mini N.min2 N.bur T.mon T.bami Tbam2 Tboll Tgredi Tgred2 Tgreg

т © NG 0 1 0 0 1 2 1 0 0 0 0 0 2
2 70 2 2 2 0 0 0 0 0 0 0 0 4 4 0
$ 1 1 1 1 0 1 1 0 0 0 0 0 0 0 0
4 0 0 2 2 0 0 0 NC 0 0 0 0 0 0 0
50 0 0 0 0 0 0 0 0 0 2 0 0 0 0
6 0 1 1 1 1 1 1 1 1 2 2 1 1 1 2
TA NC 1 2 1 NC NC NC 1 2 1 2 1 1 NC
8 0 1 1 1 0 0 0 NC 0 1 1 1 0 0 0
9 0 1 1 1 1 1 1 1 1 0 1 1 1 1
10220 0 0 0 0 0 0 0 1 0 1 0 0 0 0
11270 0 0 0 0 0 0 0 0 0 1 0 0 0 0
de y 1 1 1 1 1 1 1 1 1 2 0 1 1 1
1300 0 0 0 0 0 0 0 0 2 2 2 2 2 0
14 0 0 0 0 0 0 0 3 0 2 2 2 2 2 0
150 0 2 2 0 0 0 2 2 2 0 2 0 1 2
16 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0
WA 1 1 1 1 1 1 1 0 0 0 0 0 0 0
181 1 1 1 1 1 1 1 0 0 0 0 0 0 0
19 0 0 1 1 0 1 1 0 0 0 0 0 2 2 0
20 0 1 1 1 0 1 1 0 0 0 0 0 2 2 0
2.1220 0 0 0 0 0 0 0 0 0 1 0 0 0 0
220 0 0 0 0 0 0 0 3 1 0 1 1 1 3
2320 0 0 0 0 1 1 0 0 0 0 0 0 0 0
2A 2 2 2 2 2 2 2 2 0 0 0 0 0 2
25) 10 0 0 0 0 0 0 0 2 0 0 0 3 3 2
26 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0
27 0 0 1 1 1 0 0 0 1 1 1 1 1 1 1
28 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
29 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
300 0 0 0 0 0 0 0 1 1 1 1 1 1 1
310 0 0 0 1 0 0 1 1 1 1 1 0 0 it
32 2 1 1 1 1 0 1 0 1 0 0 1 1 1 1
33) 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0
34 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0
35 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
36 0 0 0 0 0 0 0 1 2 1 1 1 2 р 3
I/O 0 0 0 0 0 1 1 2 1 1 1 3 3 2
38 af 0 0 0 0 1 1 0 0 0 0 1 0 0 0
39 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0
40 0 0 0 0 0 0 1 0 0 0 0 0 1 1 0
413 1 1 1 1 1 0 1 2 2 2 1 1 1 1
42 3 0 2 2 3 0 0 1 2 2 2 1 1 2 2
43 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0
4 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0
AA 1 1 1 1 1 0 0 1 0 0 0 1 1 0
46 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
47 0 0 0 0 0 0 0 NC 0 0 0 0 0 0 0
48 0 0 0 0 0 0 0 NC NC 0 0 0 2 2 0

327

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA)

TABLE 81. (Continued)

25 26) 27 28 29 30
Jjin2 Wniz Gchii

24

23
Todo20 Тжап Txiao Jjin1

17 18 19 20 21 22
Thud2 Tludi Tlud2 Tlud3 Tmax Todol

16
Thud1

Gchi2

NC

10

11

NC

12
13
14
15
16
uz
18
19
20
21

22
23

24

25

26
27

NC

28

29
30
31

32

0

36

37

38
39
40
41

46
47

NC
NC

made in a single step. To obtain the uncoiled
oviduct seen in Neotricula (contrast Triculini),
first the duct of the seminal receptacle, then

the spermathecal duct would have had to mi-
of the bursa. With this accomplished, the

large scale rearangements of ducts were
grate along the oviduct to move onto the duct

ceptacle shifts as follows: along the oviduct to
the base of the duct of the bursa (one species
of Tricula); — to stem off the duct of the bursa
(one species of Tricula, nearly all Neotricula,
Halewisia); — to stem off the sperm duct
(Pachydrobia). The presumption is that no

the opening into the duct of the seminal re-

328 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

NEOTRICULA
MIN

DIA DUP1 LIL CRI

TRICULA

FIG. 157. MST based on distance coefficients in-
volving all anatomical characters treated in Tables
80 and 81. The tree is drawn to scale using inter-
taxon distances. See caption for Figure 153 for de-
fining abbreviations. Generic groupings have been
circled.

0.97 |

TRICULA FE

T T T 1
-0.6 0.0 0.6 1.2 1.8

FIG. 158. Ordination diagram following РСА and
MST seen in Figure 157. All characters scored in
Table 81 as NC were removed in this analysis to
permit the PCA treatment. The ordination is on the
first two PCA axes. Generic groupings have been
circled. See text for details. See caption for Figure
153 for defining abbreviations.

TABLE 82. Principal Component Analysis of anatomical data (matrix of 41 x 30). The first ten principal
components are listed with the percentage of variance loading on each component.

Component Eigenvalue

8.87575
6.86082
5.58656
3.54706
2.70292
2.60035
2.12427
1.83148
1.61804
1.05972

© © © -J O O1 BR OC ND —

—

Percent Cumulative
21.65 21.65
16.73 38.38
13.63 52.01

8.65 60.66
6.59 67.25
6.34 73.59
5.18 78.77
4.47 83.24
3.95 87.19
2.58 89.77

proximal section of the duct of the bursa (as
seen in Tricula) would become the sperm
duct, linking the new bursa complex to the
oviduct to facilitate fertilization. In no case is a
species of Pachydrobiini seen with the semi-

nal receptacle arising from the oviduct! The
most probable sequence is the move of the
seminal receptacle from the oviduct to the
duct of the bursa (situation seen in Tricula
species b). Then the spermathecal duct mi-

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 329

TABLE 83. Characters with high loadings, that define each of ten components. **High loading on more

than one axis (above).

Characters Loadings
UT + 0.934
18* + 0.838
24* + 0.734
27 —0.678
23 + 0.658
21 + 0.649
37 — 0.644
36 — 0.623
6 — 0.616
28 +0.610
29 + 0.610
46 +0.610
35 + 0.599
13 — 0.580
38 + 0.579
30 —0.546
41 —0.460
26 + 0.519
20 —0.871
31 + 0.847
45 —0.784
19 —0.723
15 + 0.696
22 + 0.665
2 —0.601
34 —0.575
40 —0.575
42 + 0.522
39 — 0.449
33 —0.471
46** + 0.632
28** + 0.632
29** + 0.632
2 +0.596
40** + 0.589
34** + 0.589
10 + 0.573
25 + 0.570
19% + 0.514
30** + 0.485
5 + 0.469
45% + 0.447
3 — 0.446
32 + 0.446
44 — 0.631
Gas + 0.545
30** + 0.529
14 —0.502
26% + 0.470
3 + 0.646
12 —0.484

COMPONENT

|. GENERIC ORGANIZERS*, N = 18

oviduct 360” circle

spermathecal duct enters pericardium

origin of seminal receptacle

male gonad overlaps stomach

seminal receptacle duct length

seminal receptacle lost

ejaculatory duct prominence

position of ejaculatory duct

gill filament number

male gonad length

presence of vas efferens

position of lateral tooth major cusp

white muscular zone on penis

bursa size

digestive gland covers Pst

poisition of seminal vesicle

radula length

albumen gland length

|. SPERMATHECAL DUCT, RADULA, PENIS, STOMACH,
N = 12

orientation spermathecal duct

where vas efferens leaves prostate

major cusp of lateral tooth

length spermathecal duct

bursa shape

configuration duct of seminal receptacle

operculum shape

position of penial opening

color of Ast

number of rows of teeth

melanin streak on Ast

penis highly extensible

Ш. MALE REPRODUCTIVE, RADULA, OPERCULUM,
STOMACH, N = 14

poisition major cusp lateral tooth

male gonad length

has/has not: vas efferens

operculum shape

Ast color

position penial opening

gland mass anterior to osphradium

pericardial bursa

length spermathecal duct

position serminal receptacle

ridge on operculum

lateral tooth major cusp

opercular layers

penial tip

IV. RADULA, OPERCULUM, REPRODUCTIVE, N = 5

central cusp, central tooth

opercular ridge

position, seminal vesicle

bursa position

albumen gland length

V. OPERCULUM, FEMALE REPRODUCTIVE, MANTLE
CAVITY, N = 4

opercular layers

female gonad length

(continued)

330 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

TABLE 83. (Continued).

Characters Loadings COMPONENT
232 +0.424 length duct seminal receptacle
10* —0.421 gland mass anterior to osphradium
VI. STOMACH, PENIS, FEMALE REPRODUCTIVE, N = 6
39°” + 0.583 melanin streak on Ast
33” +0.530 highly extensible penis
26*™* + 0.482 albumen bland length
SON +0.475 digestive bland covers Pst
325 —0.455 penis tip
16 —0.435 duct of bursa
VII. CENTRAL TOOTH, OSPHRADIUM, OPERCULUM,
N=4
9 +0.617 osphradium length
43 + 0.465 swelling on face of central tooth
5° — 0.443 opercular ridge
44 —0.417 central cusp of central tooth
VII. CENTRAL TOOTH, N = 1
43** —0.610 swellings on face of central tooth
IX. BURSA, TEETH, N = 2
16 —0.611 duct of bursa
42** + 0.523 rows of teeth: number
X. BURSA, М = 1
1 + 0.538 position of bursa

grates along the oviduct to the duct of the
bursa, with subsequent shift of the sperma-
thecal duct from opening into the pericardium
to opening into the posterior end of the mantle
cavity next to the pericardium, and the uncoil-
ing of the oviduct (as seen in Neotricula spe-
cies of type a, and b). Subsequently, either
the duct of the seminal receptacle migrates
onto the sperm duct, or the point where the
spermathecal duct joins the duct of the bursa
moves distally towards the bursa leaving an-
terior to it that proximal piece of duct that once
was the duct of the bursa, with the seminal
receptacle attached to it (as seen in Pachy-
drobia). In this scenario, both Halewisia and
Pachydrobia evolved from progenators ге-
sembling Neotricula species of type a (e.g. N.
aperta, N. cristella). In Halewisia the sper-
mathecal duct migrated to open into the bursa
copulatrix. The shell is a derived type (Davis,
1979: 111). Pachydrobia has evolved a large
first-order adaptive radiation (Davis, 1992) of
greater than 14 species in which the shell has
modified from the generalized Tricula type
variously: larger, thick shells with thick lips;
tendency to have pronounced macrosculp-
ture (nodes, spines); tendency to shell as-
symetry.

There is parallel evolution in the Triculini
and Pachydrobiini involving loss of the semi-
nal receptacle and duct with new structures

arising in different locations to accomodate
the functions of the seminal receptacle. The
solution to this loss is quite different in the two
tribes, hence, no difficulty arises regarding
correct tribal classification or assessment of
intergeneric relationships. The tendency to-
wards loss is minor in the Triculini, involving
only one genus, Lacunopsis; it is major devel-
opment in the Pachydrobiini involving the ma-
jority of genera (five of eight). In Lacunopsis,
two or more accessary seminal receptacles
arise to the exterior of the juncture of the ovi-
duct and spermathecal duct. In the Pachydro-
biini, solutions to the loss of the seminal re-
ceptacle are all within the relevant ducts (duct
of the bursa, sperm duct, oviduct). The most
parsimonious direction of evolution is as fol-
lows: the duct of the seminal receptacle is
lost, the seminal receptacle becomes fused to
the duct of the bursa close to or at the bursa
(as in N. dianmenensis); — The seminal re-
ceptacle is replaced by an outpocketing within
the encapsulated duct of the bursa (as in
Robertsiella); — the sperm storage area
shifts anteriorly to a swelling at the juncture of
the duct of the bursa, spermathecal duct and
sperm duct (as in Jinhongia); — the next shift
is into the sperm duct (Guoia-a), and finally
the oviduct (Guoia-b) — and farther posterior
within the oviduct (Wuconchona and Gamma-
tricula).

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 331

TABLE 84. Characters and character-states used for the cladistic analysis.

№ —

BH 0

SYNAPOMORPHIES

. Spermathecal duct opens to pericardium: (a) (0); (b) to posterior mantle cavity (1).
. Oviduct makes: (a) a closed, tight 360° twist posterior to duct of the bursa (0); (b) an open 360°

circle (1); (c) does not have the 360° twist or circle (2).

. Sperm duct: (a) absent (0); (b) present and short (1); (c) present and elongated (2).
. Spermathecal duct: (a) has primitive position between pericardium and oviduct as seen in

Pomatiopsis (Pomatiopsinae) (0); (b) joins oviduct before oviduct twist (1); (c) joins oviduct 180° or
more into oviduct circle (2); (d) joins duct of bursa, or common sperm duct (3); (e) joins bursa (4).

. Seminal receptacle: (a) arises from primitive condition from oviduct posterior to the duct of the bursa

(as in Pomatiopsis) (0); (b) from left side of the oviduct and pressed to the contour of the oviduct
circle complex (1); (c) joins duct of bursa or common sperm duct (2); (d) joins the sperm duct (3); is
lost; function replaced by: (e) an outpocketing of the encapsulated duct of the bursa (4); (f) an
internal duct swelling at the juncture of bursa and spermathecal duct (5); (g) an outpocketing of
sperm duct or swelling of sperm duct at juncture of sperm duct and oviduct (6); (h) within the oviduct
(7); (i) two or more seminal receptacles arise from the oviduct (outside the duct) at the juncture of
the oviduct with the spermathecal duct (8).

. Duct of serminal receptacle(s): (a) short (0); (b) long (1); (c) lost due to fusion of the seminal

receptacle to the duct of the bursa (2); (d) lost as the usual seminal receptacle is lost (3).

il):

. Gonad 1 or 2, or 3 finger-like tubes: (a) (0); (b) (1).

. Vas deferens runs to style sac, turns posteriorly to form seminal vesicle: (a) (0); (b)
. Central tooth: (a) anterior cusps 5 or more (0); (b) only one large triangular blade (1
. Foot shape, power: (a) slender, weak (0); (b) wide, powerful (1)

. Head shape: (a) standard (0); (b) wide, squat head; eyes in pronounced lobes (1)

. Shell shape: (a) ovate-conic (0); (b) globose (1); (c) trochoid (2)

. Shell size (a) small, thin (0); (b) increased length, thickness (1)

. Penis with chitenous stylet: (a) (0); (b) (1).

. Spermathecal duct with vaginal section: (a) (0); (b) (1).

. Bursa copulatrix elongated relative to length of albumen gland: (a) (0); (b) (1).

. Penis has a strongly developed muscular white zone at concave edge: (a) (0); (b) (1).

(
)

AUTAPOMORPHIES
Fenouilia

. Gill filament section Gf, extremely elongated (1)

Lithoglyphopsis

. Radular sac extremely elongated (1).
. Posterior pallial oviduct bends 180° around (1).
. Extremely long cerebral commissure (1).

Jinhongia
5f (above). Function of the seminal receptacle within an internal swelling at the juncture of the bursa
and spermathecal duct (1).

Robertsiella
5e (above). Function of the seminal receptacle within a cavitation of the duct of the bursa (1).

. Common sperm duct, duct of the bursa encapsulated (1).

Guoia

. The penis has a glandular lobe (1).
. The sperm duct is not only much elongated, it coils (1).

Wuconchona

. The columella of the shell within the body whorl has a raised spiral ridge (1).
. The spermathecal duct is short (1).
. The male gonad is a wide sac, few lobes, floor of sac opens to a vas deferens, ¡.e. no vas efferens

(1).

. The dominant cusp of the lateral tooth is displaced towards the outside edge of the tooth (1).
. The oviduct makes a U-shaped bend (1).

(continued)

332 DAVIS, CHEN, WU, KUANG, XING, Ll, LIU 8 YAN

TABLE 84. (Continued).

Gammatricula
30. The posterior half or third of the prostate is smooth (1).

Pachydrobia

31. The shell lip is especially thickened (1).

5d (above). The seminal receptacle arises from the sperm duct.

Halewisia
4e (above). The spermathecal duct joins the bursa copulatrix.

TABLE 85. Scoring 18 OTUs for synapomorphies listed in Table 84. Hub, Hubendickia (outgroup of
Jullieniini); Tr-a, Tricula species a, Tr-b, species b; Delavaya; Fen, Fenouilia; Lac, Lacunopsis; Lit,
Lithoglyphopsis; Ne-a, Neotricula species a, Ne-b, species b, Ne-c, species c; Hal, Halewisia; Pac,
Pachydrobia; Gu-a, Guoia species a, Gu-b, species b; Rob, Robertsiella; Jin, Jinhongia; Wuc,

Wuconchona; Gam, Gammatricula.

1 2 3. 4 5 6 7 8 9

UO 1 12 ie! 145" 15 16 1 1

Hub Tr-a Tr-b Del Fen Lac Lit Ne-a Ne-b Ne-c Hal Pac Gu-a Gu-b Rob Jin Wuc Gam

1.0 0 OF 10" 20207 207221 1
2 0 о OO e 2
3 0 0 OF HO FON O OL 1
4 2 1 1 1 1 la 3 3
5 1 0 2. 0. 0.8 02 2
Ou 0 OOO OF 0 RO 1
AE) 0 07 OO O 0) о 0
8 1 0 о ооо 0
9720 0 ON 1 ent 0 0
10 0 0 OO I fe a 0 0
110 0 Qu SOL Ud 0 0
12000 0 A ee LOS 0 0
13 0 0 OF tal 1 yo u 0 0
14 0 0 Oy 3057 01 1010-5050 0
15 0 0 О. 00. 0..0 -0 0
16 0 0 ms, OOO ON 0
17220 0 0..0. 707 O On 0 0

There is clear-cut parallelism in the intra-
ductal shifting of sperm storage. The Robert-
siella-Guoia clade has highly derived charac-
ter-states: the chitenous sheath on the penial
papilla, the vaginal section of the spermathe-
cal duct, the thick, straight and relatively short
continuous duct running from the end of the
mantle cavity into the bursa, and the loss of
the seminal receptacle. The clade is relatively
old within the 12 + 4 million years in which the
triculine macro adaptive radiation (Davis,
1992) has been evolving. The presence of
Robertsiella in Malaysia and Guoia in Hunan,
China, indicate both the age of the clade in
and its origin in northern Burma or Yunnan,
China. Three character-states involving the
sperm storage location are found within this
one clade: (1) within an outpocketing of the
encapsulated duct of the bursa (Robertsiella),

1

oOOoooooooooyDpD@—npn

oOo-O0O0O0O0O0O0O0O0O0—-DPEDD—
OO0OO-10O00O0OO OO — & CO — ND
oO--0000000090PDPD—
O0o0- "00000000 JO D D —
O0o0-=-00000000R20-=n —
OO0OOOOOOOO O OO O1 GO py —
1000000000 VJOON —
1000000000 очно — M —

(2) within the sperm duct next to the oviduct
(Guoia-a), (3) within the proximal oviduct
(Guoia-b). The parallel situation is seen in the
genera Wuconchona and Gammatricula that
are found only (thus far) in southeastern
China. In these genera, the simplified gonad,
penial-characters (but lacking the chitenous
penial stylet), and sperm storage in a more
distal section of oviduct indicate a clade of
derived taxa clearly divergent from the Rob-
ertsiella-Guoia clade.

Other parallelisms are simply partitioned
and understood. In each second order radia-
tion (Davis, 1980, 1992) there are trends for
increasing size (seen in shell length), increas-
ing complexity in sculpture (from smooth —
ribs — reticulate sculpture, spiral cords,
spines or nodes), and changes in shape
(ovate-conic to turreted, globose, assymetry

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 333

HUBENDICKIA

DEL AVAYA

FENOUILIA
LACUNOPSIS
LITHOGLYPHOPSIS
NEOTRICULA-b

NEOTRICULA -c
NEOTRICULA-a

GAMMATRICULA
WUCONCHONA
ROBERTSIELLA
GUOIA-a

DEL AVAYA

FENOUILIA
LACUNOPSIS
LITHOGLYPHOPSIS
NEOTRICULA - b

NEOTRICULA- c

NEOTRICULA-a

HALEWISIA

PACHYDROBIA
JINGHONGIA

WUCONCHONA
GAMMATRICULA

HUBENDICKIA

TRICULA-b

TRICULA-a

DELAVAYA

FENOUILIA
LACUNOPSIS

LITHOGLYPHOPSIS

NEOTRICULA-b
NEOTRICULA-c
NEOTRICULA-a
HALEWISIA
PACHYDROBIA
JINGHONGIA
GAMMATRICULA
WUCONCHONA
ROBERTSIELLA
GUOIA-a

GUOIA-b
HUBENDICKIA
TRICULA-b
TRICULA- a
DELAVAYA
FENOUILIA
LACUNOPSIS

LITHOGLYPHOPSIS

NEOTRICULA-a
HALEWISIA
GAMMATRICULA
WUCONCHONA
JINGHONGIA
ROBERTSIELLA
GUOIA- a
GUOIA-b

FIGS. 159-162 Four representatives of the 20 computer-generated cladograms following the Hennig-86
treatment of data given in Tables 85 and 86. See text for details.

(see Davis, 1979). The primitive small ovate-
conic shell seen in Tricula, Neotricula, Jinhon-
gia and Gammatricula is the basic platform
from which splendid first order adaptive radi-
ations arose, most with distinctive patterns of
shell development (e.g. Pachydrobia, Jullie-
nia, Hubendickia, Lacunopsis. However,
while one can discern the shells of Lacunop-
sis from those of Hubendickia or Pachydro-
bia, there is convergence in the globose shell
types such that the shell character-states of
Guoia, Lithoglyphopsis and Lacunopsis
merge. Fortunately, there is extreme anatom-
ical divergence between Guoia, Lacunopsis
and Lithoglyphopsis; accordingly these gen-
era are readily classified in one or the other of
the relevant second order radiations.

The length of the duct of the seminal recep-
tacle increases with increasing anatomical
complexity in the Jullieniini and Triculini. The
trend is particularly pronounced in the Jullie-

niini (Davis, 1979; 1991). There is a parallel
increase in the length of the sperm duct in-
volving some derived genera of the Triculini
and Pachydrobiini; it is particularly evident in
Guoia.

OLD PROBLEMS RESOLVED

Two unsolved questions of a decade ago
can now be answered. (1) What is Lithogly-
phopsis and its relationship to Lithoglyphop-
sis aperta Temcharoen, 1971? (2) Given the
impact of the Himalayan orogeny as a driving
force of the spectacular triculine radiation, too
few species of Tricula or species morpholog-
ically close to Tricula had been found a de-
cade ago (Davis, 1980). There were numer-
ous species of Pachydrobia, Lacunopsis,
Hubendickia, etc. known a decade ago, but
only three species of Tricula, sensu Davis,

334 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN

HUBENDICKIA
NEOTRICULA- a
NEOTRICULA-b
NEOTRICULA-c
HALEWISIA
PACHYDROBIA

JINGHONGIA
WUCONCHONA
GAMMATRICULA
ROBERTSIELLA
GUOIA-a

DELAVAYA

FENOUILIA
LACUNOPSIS
LITHOGLYPHOPSIS

FIG. 163. The nelsen consensus tree resolving the
20 trees generated following the Hennig-88 treat-
ment of data given in Tables 85 and 86.

1979, had been found in Southeast Asia. A
few species from Burma had been described
as Tricula on the basis of shells (reviewed in
Davis, 1980). It was predicted that more
would be found (Davis, 1980). How extensive
was the Tricula radiation?

Concerning the first question, it is evident
that Lithoglyphopsis is most closely related to
Fenouilia and Delavaya anatomically and to
Lacunopsis conchologically. Lithoglyphopsis
is a derived genus in the tribe Triculini. Litho-
glyphopsis aperta, the snail host for Schisto-
sosma mekongi, is the type species for the
genus Neotricula (Davis et al., 1986a). Neo-
tricula is an anatomically generalized genus
in the tribe Pachydrobiini. Thus, while Litho-
glyphopsis and Neotricula are members of
the same subfamily, they are highly divergent
genera within the Triculinae.

The second issue is now resolved. With re-
search initiated in Yunnan, China, in 1983
(Davis et al., 1986b) and with this and other
ongoing research, it is clear that Tricula,
sensu Davis, 1979, comprises an extremely
large radiation primarily located in southern
China. The concept of Tricula sensu Davis,
1979, has changed. The anatomical data sup-
port four genera with a Tricula-type shell: Tric-
ula, Neotricula, Gammatricula, and Jinhongia.
Anatomical data have been published for
seven species of Neotricula, eleven species
of Tricula and one each for Jinhongia and
Gammatricula. These 20 species are one
third of the estimated 60 species (conserva-
tive estimate) with Tricula-like shells that

we think occur throughout southern China,
Burma and northern Vietnam (Davis, 1992).

BIOGEOGRAPHY

Biogeography at a glance is provided in
Figure 165. An area cladogram for river evo-
lution is given with the number of species of
each genus for which we have anatomical
data listed for each region of river. Overall
Triculinae biogeography has been reviewed
recently (Davis, 1992). Only a few points
need to be emphasized here. (1) Thus far,
Neotricula has its greatest deployment in the
mid to lower Yangtze River, whereas Tricula
has more species in the upper Yangtze River.
(2) The distribution of Guoia, Robertsiella,
Tricula and Neotricula show that the major
innovative anatomical developments in the
Triculini and Pachydrobiini had occurred be-
fore the Yangtze-Mekong river drainages
were completely separated.

(3) The tribes Triculini and the derivative
Pachydrobiini dominate the Yangtze River
drainage; no Jullieniini have been found, thus
far, in the mid or lower Yangtze River drain-
ages. Conversely, the Jullieniini dominate the
Mekong River drainage with their greatest ra-
diation in the lower Mekong River. (4) Only
Pacydrobia of the Pachydrobiini and La-
cunopsis of the Triculini have undergone ex-
plosive adaptive radiations (considering all
genera of the Triculini and Pachydrobiini); this
has occurred in the lower Mekong River. (5)
The most derived anatomical character-
states are found in the lower river systems.

(6) There are considerable ecological dif-
ferences among tribes. With the exception of
Kunmingia living in the shallow basin of a
limestone spring, the other genera of the Jul-
lieniini live in a large river environment. Con-
sidering the Triculini and Pachydrobiini, only
Lacunopsis, Lithoglyphopsis, Guoia, Halewi-
sia and Pachydrobia live in large rivers.
Fenouilia lives on the bottom of a large lake,
as did Parapyrgula (presumed extinct). Tric-
ula, Neotricula, Jinhongia and Gammatricula
live in upland shaded small streamlets of
pure, cold water. Only Neotricula aperta has
adjusted to living in a large river environment.

There are numerous species of Neotricula
and Tricula in Hubei Province and provinces
to the east, species that are in the process of
being studied. We have seen at least one ad-
ditional species of Gammatricula. The proven
distribution of Tricula is from northern India to

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 335

с.
(=
=
i HUBENDICKIA | =
2b 4c 5b 7b 8b 6b =
=
TRICULA-a
—
| TRICULA-b =
5c G
E
4b 2
DELAVAYA
| FENOUILIA
9b 13b O
OO LITHOGLYPHOPSIS
10b11b Pure
PAS) LACUNOPSIS
6d 5i
[| O HALEWISIA
(i Ä 30 31-4e
ECS NEOTRICULA-b
i NEOTRICULA-a
5c T
>
2 NEOTRICULA- c о
НЮ = PACHYDROBIA <
16 2c 3b,c 4d |=)
5d 136 66 31 2
о
19 SYNAPOMORPHIES JINHONGIA =
Ника sf OO ROBERTSIELLA =
5 22 =
15 HOMOPLASIES > GUOIA-a
>] 146156 CH) i
24 23 =
4 LOSSES er 3c GUOIA-b
OOOO WUCONCHONA
38 TREE LENGTH | | 3a 24 25 26 27 28, 29
5h 16b17b @ GAMMATRICULA
AUTAPOMORPHIES © 30

FIG. 164. The cladogram resuiting from weighting one character in particular, the seminal receptacle, and
the direction of evolution of character states with the primitive condition being 5a (= score 0) (see Table 85)
(all Triculini except Lacunopsis); the shift of the duct of the seminal receptacle to the duct of the bursa (5c
= score 2); the loss of the seminal receptacle, its function transferred into the duct of the bursa (5e* and
5e on the cladogram); the move within the duct system to the juncture of duct of the bursa, sperm duct and
spermathecal duct (5f) — to the sperm duct (5g) — to the oviduct (5h). Parallelisms are accounted for; see
text for details.

the East China Sea. As discussed in Davis highly probable that species of Neotricula will
(1992), there are numerous diverse Triculinae be found among them in the upper Yangtze
in Yunnan that have not yet been studied. It is River drainage. While this study has done

DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

LITHOGLYPHOPSIS-1
NEOTRICULA-5
TRICULA-3
WUCONCHONA- 1

GAMMATRICULA-1

LOWER YANGTZE R.

GUOIA-2
DELAVAYA-1
KUNMINGIA-1*
LACUNOPSIS-?
NEOTRICULA-?
TRICULA-3
UPPER YANGTZE R.
FENOUILIA-1
PARAPROSOSTHENIA*-E?
PARAPYRGULA*-E?
TRICULA-3
TRICULA-1 LAKE
ERHAI

HALEWISLA
NEOTRICULA-1 HUBENDICKIA*
MEKONG R. TRICULA-1 HYDRORISSOIA*
HUBENDICKIA-1 JULLIENIA*
JINHONGIA-1 THAI DRAINAGE KARELAINIA*
LACUNOPSIS>1 LACUNOPSIS
TRICULA-1 NEOPROSOSTHENIA*

UPPER MEKONG R.

YUNNAN

NEOTRICULA-1
PACHYDROBIA
PACHYDROBIELLA*
SADUNIELLA*

LOWER MEKONG R.

ROBERTSIELLA-2

MALAYSIA

FIG. 165. An area cladogram for rivers showing the location of triculine genera. Those genera with the
asterisk belong to the tribe Jullieniini. The question mark indicates the possibility that the genus might be
found there. E? indicates probable extinction. With the exception of Pachydrobia and Lacunopsis in the lower
Mekong River, each with more than ten species, the number of species of each genus of the Triculini and
Pachydrobiini studied anatomically is given at each location along the drainage systems. Tricula-1 is Tricula

montana of northern India. See text for further details.

much to reduce the probability that new gen-
era of Triculinae will be found (we would ex-
pect one or two more, especially in Yunnan),
there is considerably more to be learned con-
cerning species of Tricula and Neotricula and
their patterns of distribution.

ACKNOWLEDGEMENTS

Use of the facilities of Hunan Medical Col-
lege and the Zhejiang Academy of Medical
Sciences is gratefully acknowledged. The
drawings were made by Davis with render-
ings by Elizabeth Carrossa and Susan Tram-
mell. Scanning electron microscope work was
done by Dr. Chen Cui-E and Caryl Hester-
man. Photographs of the shells were taken
and printed by Dr. Chen and Caryl Hester-
man. Graphics were done by Susan Tram-

mell. Dr. John Hendrickson is thanked for his
help in executing the NTSYS programs. The
work was supported by the National Natural
Science Foundation of China award No.
3860607 to Dr. Chen, and М.1.Н. award A1
11373 to Dr. Davis.

LITERATURE CITED

ANNANDALE, N., 1924, The molluscan hosts of
the human blood fluke in China and Japan, and
species liable to be confused with them. In
E. C. FAUST, and H. E. MELENEY, Studies on
schistosomiasis japonica. American Journal of
Hygiene Monographic Series, No.3: 269-294, 1

|

pl.

BENSON, W. H., 1843, Description of Campto-
ceras, a new genus of the Lymnaeidae, allied to
Ancylus, and of Tricula, a new type of form allied
to Melania. Calcutta Journal of Natural History,
3(12): 465-468.

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 337

BRANDT, R. M., 1974, The non-marine aquatic
Mollusca of Thailand. Archiv für Molluskenkunde,
105: 1-423.

CROSSE, H. & P. FISCHER, 1876, Mollusques flu-
viatiles, recueillis au Cambodge, par la Mission
scientifique francaise de 1873. Journal de Con-
chyliologie, 24: 313-342, pls. 10-11.

DAVIS, G. M., 1967, The systematic relationship of
Pomatiopsis lapidaria and Oncomelania hupen-
sis formosana (Prosobranchia: Hydrobiidae).
Malacologia, 6(1-2): 1-143.

DAVIS, G. M., 1968a, A systematic study of On-
comelania hupensis chiui. Malacologia, 7(1): 17-
70.

DAVIS, G. M., 1968b, New Tricula from Thailand.
Archiv für Molluskenkunde, 98: 291-317.

DAVIS, G. M., 1969a, Reproductive, neural and
other anatomical aspects of Oncomelania min-
ima (Prosobranchia: Hydrobiidae). Venus, Japa-
nese Journal of Malacology, 28(1): 1-36.

DAVIS, G. M., 1969b, Oncomelania and the trans-
mission of Schistosoma japonicum: a brief re-
view. In HARINASUTA, C., ed., Proceedings of
the Fourth Asian seminar on parasitology and
tropical medicine, schistosomiasis and other
snail-transmitted helminthiasis, 24-27 February,
Manila. pp. 93-103.

DAVIS, С. M., 1979, The origin and evolution of the
gastropod family Pomatiopsidae, with emphasis
on the Mekong River Triculinae. Monograph of
the Academy of Natural Sciences of Philadelphia
No. 20: 1-120.

DAVIS, G. M., 1980, Snail hosts of Asian Schisto-
mosoma infecting man: origin and coevolution. In
J. I. BRUCE, et al., eds., The Mekong schisto-
some, Malacological Review, Supplement 2:
195-238.

DAVIS, G. M., 1981, Different modes of evolution
and adaptive radiation in the Pomatiopsidae
(Prosobranchia: Mesogastropoda). Malacologia,
21(1/2): 209-262.

DAVIS, G. M., 1992, Evolution of prosobranch
snails transmitting Asian Schistosoma; coevolu-
tion with Schistosoma: a review. Progress in Clin-
ical Parasitology, in press.

DAVIS, G. M. & P. CARNEY, 1973, Description of
Oncomelania hupenis lindoensis, first intermedi-
ate host of Schistosoma japonicum in Sulawesi
(Celebes). Proceedings of the Academy of Nat-
ural Sciences of Philadelphia, 124(1): 1-34.

DAVIS G. M. 8 G. GREER, 1980, A new genus and
two new species of Triculinae (Gastropoda:
Prosobranchia) and the transmission of a Malay-
sian mammalian Schistosoma sp. Proceedings
of the Academy of Natural Sciences of Philadel-
phia, 132: 245-276.

DAVIS, G. M. 8 Z. B. KANG, 1990, The genus Wu-
conchona of China (Gastropoda: Pomatiopsidae:
Triculinae): anatomy, systematics, cladistics, and
transmission of Schistosoma. Proceedings of the
Academy of Natural Sciences of Philadelphia,
142: 119-142.

DAVIS, G. M., C. E. CHEN, X. G. XING 8 C. WU,

1988, The Stenothyridae of China, No. 2:
Stenothyra hunanensis. Proceedings of the
Academy of Natural Sciences of Philadelphia,
140(2): 247-266.

DAVIS, G. M., V. KITIKOON 8 P. TEMCHAROEN,
1976, Monograph on “Lithoglyphopsis” aperta,
the snail host of Mekong River schistosomiasis.
Malacologia, 15(2): 241-287.

DAVIS, С. M. Y. H. KUO, К. E. HOAGLAND, Р. L.
CHEN, H. M. YANG, & D. J. CHEN, 1984, Kun-
mingia, a new genus of Triculinae from China:
phenetic and cladistic relationships. Proceedings
of the Academiy of Natural Sciences of Philadel-
phia, 136: 165-193.

DAVIS, G. M., Y. H. GUO, K. E. HOAGLAND, P. L.
CHEN, L. C. ZHENG, H. M. YANG, D. J. CHEN
8 Y. F. ZHOU, 1986b, Anatomy and systematics
of triculine (Prosobranchia: Pomatiopsidae: Tri-
culinae) freshwater snails from Yunnan, China,
with descriptions of new species. Proceedings of
the Academy of Natural Sciences of Philadel-
phia, 138(2): 466-575.

DAVIS, G. M., Y. H. KUO, K. E. HOAGLAND, P. L.
CHEN, H. M. YANG & D. J. CHEN, 1983, Ad-
vances in the Systematics of the Triculinae (Gas-
tropoda: Prosobranchia): the genus Fenouilia of
Yunnan, China. Proceedings of the Academy of
Natural Sciences of Philadelphia 135: 177-199.

DAVIS, G. M., Y. H. KUO, K. E. HOAGLAND, P. L.
CHEN, Н. M. YANG 4 D. J. CHEN, 1985, Erhaia,
a new genus and new species of Pomatiopsidae
from China (Gastropoda: Rissoacea). Proceed-
ings of the Academy of Natural Sciences of Phil-
adelphia, 137: 48-78.

DAVIS, G. M., Y. Y. LIU 8 Y. G. CHEN, 1990, New
Genus of Triculinae (Prosobranchia: Pomatiop-
sidae) from China: Phylogenetic Relationships.
Proceedings of the Academy of Natural Sciences
of Philadelphia, 142: 143-165.

DAVIS, G. M., N. V. S. RAO 4 K. E. HOAGLAND,
1986a, In search of Tricula (Gastropoda: Proso-
branchia): Tricula defined, and a new genus de-
scribed. Proceedings of the Academy of Natural
Sciences of Philadelphia, 138(2): 426-442.

FARRIS, J. S., 1988, Hennig 86, Version 1.5. Stony
Brook, N.Y. 18 pp.

FARRIS, J. S., 1989, The retention index and the
rescaled consistency index. Cladistics 5: 412-
419.

FENG, X. C., F. H. ZHOU & S. J. ZHANG, 1985,
New discovery of first intermediate hosts of Par-
agonimus skrjabini—Tricula gredleri Kang sp.
nov. and Akiyoshia (Saganoa) odonta Kang sp.
nov. Hunan Medicine, 2(4): 65.

FENG, X. C., F. H. ZHOU 4 S. J. ZHANG, 1986,
Discovery of new snails, first intermediate hosts,
transmitting Paragonimus in Guzhang County.
Chinese Public Health, 5(5): 38-39.

GREDLER, P. V., 1881, Zur Concylien-Fauna von
China. Ill. Stúck. Jahrbücher der Deutschen
Malakozoologischen Gesellschaft, 8: 110-132.

GREDLER, P. V., 1885, Zur Concylien-Fauna von
China. VIII. S.1-19. Selbstverlag, Bozen.

338 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

GREDLER, P. V., 1886, Zur Concylien-Fauna von
China. IX. Malakozoologische Blatter (n.s.) 9: 1—
20. Cassel.

GREDLER, P. V., 1887, Zur Conchylien-Fauna von
China. XII. Stuck. Nachrictsblatt der Deutschen
Malakozoologische Gesellschaft, No. 19 (11
u.12): 168-178.

GUO, Y. H. & J. R. GU, 1985, Studies on the inter-
mediate host of Schistosoma and Paragonimus:
1. Tricula jinhongensis, a new species of Tricula
from Yunnan Province. (Gastropoda: Hydrobi-
idae). Acta Zootaxonomica Sinica, 10(3): 250-
252.

HERSHLER, В. & Е. G. THOMPSON, 1988, Notes
on morphology of Amnicola limosa (Say, 1817)
(Gastropoda: Hydrobiidae) with comments on
status of the subfamily Amnicolinae. Malacolog-
ical Review, 21: 81-92.

IOGANZEN, B. G. 4 Y. A. STAROBOGATOV,
1982, A finding of a freshwater mollusk of the
family Triculidae (Gastropoda: Prosobranchia) in
Siberia. Zoologichesky Zhurnal, 61(8): 1141-
1147.

KANG, Z. B., 1983a (Sept.), A new genus and three
new speices of the family Hydrobiidae (Gas-
tropoda: Prosobranchia) from Hubei Province,
China. Oceanologia et Limnologia Sinica, 14(5):
499-505.

KANG, Z. B., 1983b (Nov.), Two new molluscan
hosts for Paragonimus skrjabini. Oceanologia et
Limnologia Sinica, 14(6): 536-541.

KANG, Z. B., 1985, Descriptions of a new species
of Bythinella from China. Acta Hydrobiologica
Sinica, 9(1): 84-88.

KANG, Z. B., 1986, Descriptions of eight new
minute freshwater snails and a new and rare spe-
cies of land snail from China (Prosobranchia: Po-
matiopsidae, Hydrobiidae; Hydrocenidae). Archiv
für Molluskenkunde, 117: 73-91.

KRUSKAL, J. B., 1964, Multidimensional scaling by
optimizing goodness of fit to a nonmetric hypoth-
esis. Psychometrika, 29: 1-27.

KURODA, T. 8 T. HABE, 1954, New aquatic gas-
tropods from Japan. Venus, 18(2): 71-78.

KURODA, T. 8 T. HABE, 1957, Trolobiontic aquatic
snails from Japan. Venus, 19: 183-196.

LIU, Y. Y., T. K. LOU, Y. X. WANG & W. Z. ZHANG,
1981, Subspecific differentiation of oncomelaniid
snails. Acta Zootaxonomica Sinica, 6(3): 253-
266

LIU, Y. Y., Y. X. WANG 8 W. Z. ZHANG, 1980, On
new species and new records of freshwater
snails of the family Hydrobiidae from Yunnan,
China. Acta Zootaxonomica Sinica, 5(4): 358—
368.

LIU, Y. Y. 8 W. Z. ZHANG, 1979, On new genus
and species of freshwater snails harbouring cer-
cariae of lung flukes from China. Acta Zootaxo-
nomica Sinica, 4(2): 132-136.

LIU, Y. Y., W. Z. ZHANG & C. E. CHEN, 1982,
Pseudobythinella shimenensis, sp. nov., a new
aquatic Paragonimus cercariae carrying snail

from Hunan Province. Acta Zootaxonomica Sin-
ica, 7(3): 254—256.

LIU, Y. Y., W. 2. ZHANG, Y. Х. WANG, С.Е. CHEN
& S. Z. CHEN, 1982, Discovery of Akiyoshia
Kuroda et Habe (Hydrobiidae: Mollusca) from
China with descriptions of two new species. Acta
Zootaxonomica Sinica, 7(4): 364-367.

LIU, Y. Y., W. Z. ZHANG, & Y. X. WANG, 1983a,
Studies on Tricula (Prosobranchia: Hydrobiidae)
from China. Acta Zootaxonomica Sinica, 8(2):
135-139.

LIU, Y. Y., W. Z. ZHANG & Y. X. WANG, 1983b,
Three new species of Hydrobiidae (Gastropoda:
Prosobranchia) from China. Acta Zootaxonomica
Sinica, 8(4): 366-369.

LOU, Т. K., Y. Y. LIU, W. Z. ZHANG 8 У. X. WANG,
1982, A discussion on the classification of On-
comelania (Mollusca). Sinozoologica, 2: 97-117.

MOELLENDORFF, O.F. VON, 1885, Diagnoses
specierum novarum sinensium. Nachrichtsblatt
der deutschen Malakozoologischen Gesell-
schaft, 11(u.12): 161-170.

MOELLENDORFF, O.F. VON, 1888, Materialien
zur Fauna von China. Malakozoologische Blatter
(n.f.) 10: 132-163, pl. 4.

ROHLF, F. J., J. KISHPAUGH 4 D. KIRK, 1972,
NT-SYS: Numerical taxonomy system of multi-
variate statistical programs. Technical Report,
State University of New York, Stony Brook, NY.

TEMCHAROEN, P., 1971, New aquatic mollusks
from Laos. Archiv. für Molluskenkunde, 102: 91—
109.

TRYON, G. W., 1862, Notes on American fresh wa-
ter snails, with descriptions of two new species.
Proceedings of the Academy of Natural Sciences
of Philadelphia, 14: 451—452.

THIELE, J., 1928, Revision des systems der Hydro-
biiden und Melaniiden. Abteilung fur Systematik,
Okologie und Geographie der Tiere, 55: 351—
402.

VOGE, M., D. BRUCKNER & J. I. BRUCE, 1978,
Schistosoma mekongi sp. n. from man and ani-
mals, compared with four geographic strains of
Schistosoma japonicum. Journal of Parasitology,
64(4): 577-584.

WENZ, W., 1939, Gastropoda, Part 3, Prosobran-
chia. Pp. 555-581 in O. H. SCHINDEWOLF, ed.,
Handbuch der Paláozoologie, Berlin (Borntrae-
ger).

YEN, T. C., 1939, Die chinesischen Land-und
Süsswasser—Gastropoden des Naturmuseums
Senckenberg. Abhandlungen der Senckenber-
gischen Naturforschenden Gesellschaft, 444: 1—
234.

ZILCH, A., 1974, Vinzenz Gredler und die Er-
forchung der Weichtiere Chinas durch Franzi-
kaner aus Tirol. Archiv fúr Molluskenkunde, 104:
171-228.

Revised Ms. accepted 16 December 1991

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 339

APPENDIX 1. Sources of data for species of Triculinae used in the multivariate analyses
(excluding species treated in this paper).

Neotricula aperta (Temcharoen, 1971): in Davis et al., 1976 [Type species of genus; designated
in Davis et al., 1986a].

Neotricula burchi (Davis, 1968)

Tricula montana Benson, 1843: in Davis et al., 1986a [Type species of genus]

T. bambooensis Davis 4 Zheng, 1986: in Davis et al. 1986b

T. bollingi Davis, 1968

T. gregoriana Annandale, 1924: in Davis et al., 1986b

T. hudiequanensis Davis 8 Guo, 1986: in Davis et al., 1986b

T. ludongdini Davis & Guo, 1986: in Davis et al., 1986b

T. xianfengensis Davis & Guo, 1986: in Davis et al., 1986b

T. xiaolongmenensis Davis & Guo, 1986: in Davis et al., 1986b

Jinhongia jinhongensis (Guo 8 Gu, 1985): in Davis et al., 1986b [Type species of genus]
Gammatricula chinensis Davis, Liu 8 Chen, 1990: in Davis et al., 1990 [Type species of genus]
Wuconchona niuzhuangensis Kang, 1983: in Davis & Kang, 1990a [Type species of genus]

340 DAVIS, CHEN, WU, KUANG, XING, LI, LIU 8 YAN

APPENDIX 2. Factor loading of characters for the ten principal components that collectively account for
85% of the variation in this study of shell characters.

PRINCIPLE COMPONENTS

|. ll. Il. IV. V.
ile —0.0410545 —0.4614740 0.1951133 0.1102495 —0.1964896
2 0.3371738 —0.2702949 0.6452890 —0.2870722 —0.2273258
3. —0.5820084 0.0153342 0.2508085 0.0428066 0.5737585
4. —0.4883084 —0.0937884 —0.3041671 0.3050418 —0.0443432
5: 0.1598786 0.6532829 —0.1058710 —0.0463305 —0.4989291
6. —0.2746159 0.8122425 0.0071554 —0.0500111 0.1822269
Uh 0.2803561 0.2844659 0.2436223 —0.6108634 —0.2804339
8. 0.1706873 —0.4612060 0.0912007 0.4000807 —0.1060740
g: —0.6064097 0.0550309 0.4268830 0.0825438 0.4504842
10. —0.0511060 —0.0086817 0.7289513 0.3538858 0.0138379
uve —0.8494929 0.0619407 0.1342359 —0.1052565 0.2478488
12. —0.6526540 0.4972618 0.2145493 —0.2189934 —0.0605389
13. —0.7146944 0.2136463 0.0176418 —0.0641302 0.0094302
14. 0.1056108 0.2477635 —0.3376958 0.0983843 —0.0247723
15. —0.1938874 0.2239324 0.2971711 0.4426535 —0.3136139
16. 0.3325563 —0.1555008 0.1835824 —0.7154313 0.0046761
72 —0.6648433 —0.3846053 0.0730909 —0.3156229 —0.4205736
18. —0.6955758 0.3660246 0.1990413 —0.2286217 0.2015936
19. —0.6319250 —0.3369728 —0.1161682 0.0574202 —0.4591451
20. —0.6360441 —0.1127735 —0.2589680 0.1512705 —0.3323932
21. —0.4095072 —0.3166545 —0.2127306 —0.1982544 —0.2723410
22. 0.3875425 —0.1664845 0.1605595 —0.5852705 0.3093103
23. —0.0710536 0.4584002 —0.2259200 0.1122301 —0.2893657
24. —0.7326382 01977977 0.0873077 —0.2158382 0.1143209
25. —0.0983867 0.3013054 —0.2336317 0.1312115 —0.1079600
26. 0.0980229 —0.0845837 0.8104326 0.3794045 —0.1045079
27. 0.1178917 0.2651106 0.7510733 0.1976523 —0.3326711
28. 0.1214556 0.5956626 0.1896647 —0.1906758 —0.4167448
29. —0.6648433 —0.3846053 0.0730909 —0.3156229 —0.4205736
Vi. VII. МИ. IX. X.
1 0.3384070 —0.5698774 0.0460096 —0.1038296 —0.2015172
2 —0.1341500 —0.1682082 0.1296563 0.2262901 0.1179613
3 0.1338242 0.0179137 0.0716831 0.1584757 0.0337647
4. 0.1847772 —0.0621649 —0.3912900 0.3915480 —0.0423710
5: 0.1626572 —0.0789925 0.2185103 0.1225756 —0.0940029
6. —0.1021466 0.0043270 0.0444586 —0.0901887 0.2147945
7 —0.0002180 0.1959912 —0.1031375 0.0608091 —0.2179681
8 0.0615272 0.1587772 —0.2981737 —0.4940681 —0.0164338
9. —0.2213138 —0.0213113 0.1639645 —0.1460116 —0.2374873
10. —0.2431751 0.0326452 0.1561172 0.0089133 —0.0482282
ile —0.1235609 0.0485087 —0.0126241 0.0552378 —0.3203997
Wes 0.1602039 0.0045610 —0.0563521 —0.0245823 —0.0996069
13. 0.2771966 —0.1739794 —0.0983610 0.0695070 0.4416130
14. 0.2559430 —0.1571185 0.7337301 0.0273879 0.1133017
15. —0.0909669 0.2889332 —0.2158527 0.4116663 0.2163310
16. —0.3295022 0.0224967 —0.0264935 0.2984503 0.0080422
17 0.1965575 —0.1292952 —0.0028319 —0.0987587 0.1400116
18. —0.0673200 —0.1059933 —0.1987022 —0.2522537 0.2059356
19. —0.1305618 0.1619921 —0.0220588 0.1714101 —0.1204920
20. —0.4444652 0.1209360 0.1189897 —0.0587002 —0.0058012
21. —0.4959528 0.1709711 0.2976215 —0.1487671 0.1337223
22. —0.1349609 —0.2227013 —0.2268552 —0.0828126 0.1662128
23. —0.5534215 —0.2637156 —0.0907283 —0.2368980 0.0500118
24. 0.1807102 0.1188745 0.2030360 0.0449544 —0.2547120
25. —0.3337661 —0.6962163 —0.1929729 0.1613878 —0.2710739
26. —0.0871736 —0.1570683 0.1745502 0.0772054 0.1175377
27. 0.0513585 —0.0302295 —0.0577648 —0.1412964 —0.0056331
28. 0.3292032 0.2276822 —0.1024080 —0.2274468 —0.2240125

29. 0.1965575 —0.1292952 —0.0028319 —0.0987587 0.1400116

THE POMATIOPSIDAE OF HUNAN, CHINA (GASTROPODA: RISSOACEA) 341

APPENDIX 3. Factor loading of characters for the ten principal components that collectively account for
90% of the variation in this study of anatomical characters. Only 41 characters are used as no missing
data were allowed in the PCA analysis.

—0.0971430
0.2857560
0.2334109

—0.6159816

—0.1030642
0.3783801
0.2808698

—0.5800852

—0.2141655

—0.0625617
0.1441796
0.9340978
0.8379920
0.0948649

—0.0477885
0.6492946
0.2164818
0.6580347
0.7339792

—0.3732890
0.5188465

—0.6784778
0.6098325
0.6098325

—0.5456913

—0.0988049
0.4414210
0.3388348

—0.3332398
0.5994937

—0.6232941

—0.6437625
0.5793707
0.2811599

—0.3332398

—0.4597675

—0.1782957

—0.0840292
0.1616456
0.0864995
0.6098325

Vi.

—0.1394164
0.0093270
0.0449117
0.1163917
0.1417184

—0.1248724

—0.1854555
0.2550922
0.2836786

—0.0788765

—0.4354456

—0.1534960

—0.1997098

PRINCIPLE COMPONENTS

—0.6013535
—0.0616901
0.0746406
0.2368112
0.2758602
0.2220442
—0.3578114
0.2057583
—0.0802545
0.6955844
—0.2877685
—0.0764409
—0.1581357
—0.7225091
—0.8711195
0.4411572
0.6650421
0.2659786
0.1832732
—0.4181728
—0.1851264
0.1793504
0.1973294
0.1973294
0.1716460
0.8467538
0.0242099
—0.4707434
—0.5745382
0.3691855
0.2722790
0.1028952
—0.1093442
—0.4485988
—0.5745382
0.4187379
0.5215056
0.1112269
0.1293033
—0.7839667
0.1973294

VII.

—0.3474175
—0.0592813
—0.4425233
0.0395406
0.6167990
0.2536923
—0.0913738
—0.1352908
—0.1617582
0.2099368
—0.3648156
0.0609385
0.1097419

0.5959546
—0.4464561
0.4689891
—0.1616069
0.2265553
0.5730512
0.3977414
0.3564111
0.1442952
0.1884190
—0.1940611
—0.0112246
—0.2415173
0.5135236
0.2605822
0.2990977
0.3564631
0.1374653
—0.1162168
0.5697897
0.0220450
0.0918947
0.6315968
0.6315968
0.4846009
—0.0251339
0.4463780
—0.3449156
0.5893458
0.3543632
0.2957349
0.3961242
—0.1040323
—0.3590495
0.5893458
0.0153485
0.2159982
0.1647247
—0.1539454
0.0446516
0.6315968

МИ.

—0.0792670
—0.2851638
—0.2218537
—0.1044403
0.2681562
—0.3529955
0.0851 108
0.1456703
0.3841972
0.2237864
—0.2176404
0.1527720
0.1298337

IV.

0.0668697
0.1288039
0.5445427
0.3824686
0.1676757
0.1711390
0.2035502
0.3719478
—0.5024069
—0.0751765
0.0395566
—0.1052855
—0.2205638
—0.0776931
—0.1511693
—0.2707346
—0.3670552
—0.1748385
—0.3264529
—0.3883381
0.4697214
—0.1980585
0.2793996
0.2793996
0.5285202
—0.2078032
—0.3216799
0.2567195
—0.3239226
—0.0498222
—0.2475745
—0.3618655
—0.1968725
0.0503694
—0.3239226
0.3484048
—0.1292289
—0.0415819
—0.6310220
0.1787427
0.2793996

IX.

0.1056858
0.1587237
0.3159829
—0.0334291
0.4167362
—0.0161008
0.1708288
—0.1650157
—0.0299036
0.0080864
—0.6109018
0.0182737
0.0507008

V.

0.2552416
0.6460538
0.1455881
0.3790016
0.3476608
—0.4213342
—0.4838175
0.3222399
—0.1107264
0.0585363
0.0007327
0.1197626
0.2052664
0.1336098
0.1926126
0.3700605
0.2019860
0.4240264
0.1854340
—0.1400759
—0.1053183
—0.2374376
—0.1918658
—0.1918658
0.0935103
—0.2906712
—0.0195030
—0.0462704
0.1852541
0.3475240
0.0005819
0.0745969
—0.3085475
—0.3279842
0.1852541
—0.0427419
0.0143310
—0.3936273
—0.2087496
0.0927740
—0.1918658

X.

0.0854577
—0.0517160
0.1471501
0.2700133
0.1386042
0.1129820
0.1926282
0.1249281
0.5384613
0.3954614
0.1593562
0.1012241
0.1254707

(continued)

342 DAVIS, CHEN, WU, KUANG, XING, LI, LIU & YAN
APPENDIX 3. (Continued)
PRINCIPLE COMPONENTS

VI. VII. МИ. IX. X.
19. 0.1677881 0.2533934 0.1785646 0.0104447 —0.0195058
20. 0.0126496 0.1760192 0.0752291 —0.1513470 0.0383810
21. 0.1104381 —0.0205072 —0.0538776 —0.1350167 —0.0573354
22: 0.2026985 0.0578036 —0.3427233 —0.0726498 —0.0024815
23. 0.3357801 —0.0910891 —0.2196621 0.1255442 0.1376799
24. —0.1652149 0.2716005 —0.1927697 —0.0061319 0.0910426
25. 0.0315530 0.1971248 —0.2439671 —0.0636229 —0.0579249
26. 0.4822483 0.2524238 0.0512107 —0.0957911 —0.0476433
27. —0.0037041 0.0253443 0.1257655 0.2416408 —0.0378403
28. 0.0709616 —0.1122485 0.1196700 —0.0827769 —0.0295994
29. 0.0709616 —0.1122485 0.1196700 —0.0827769 —0.0295994
30. 0.1328322 —0.1671411 —0.2060764 0.1016164 0.0785346
31. 0.1569631 0.0579261 0.1100781 —0.2140089 —0.0485999
32. —0.4554683 0.1161356 0.0894265 0.0458263 —0.2058747
33. 0.5295037 0.3717229 —0.1185204 —0.0160033 0.0198906
34. 0.1115931 0.0793780 0.1069966 —0.0967904 —0.0698220
35. 0.0258687 0.0397086 —0.0296042 —0.2666091 —0.2045914
36. 0.1449843 0.1298960 —0.2864864 —0.1526312 —0.0600707
37. 0.2668363 —0.1405495 —0.3199402 0.0507502 0.0056072
38. 0.4748601 —0.1419334 0.0072793 0.2376274 —0.0643782
39. 0.5829850 0.0145123 —0.1912881 0.0057201 —0.1940589
40. 0.1115931 0.0793780 0.1069966 —0.0967904 —0.0698220
41. —0.0477399 0.3526877 0.0829260 —0.2145180 —0.0505431
42. —0.3470802 —0.0603708 0.1326352 0.5233748 —0.3710022
43. —0.2444863 0.4647836 —0.6101984 0.0604030 0.1866010
44. 0.3225032 —0.4169769 —0.1464836 0.2691902 0.0783850
45. —0.3467677 0.1343245 —0.2636901 0.3034776 0.0343297

46. 0.0709616 —0.1122485 0.1196700 —0.0827769 —0.0295994

MALACOLOGIA, 1992, 34(1-2): 343-354

CROP EPITHELIUM OF NORMAL FED, STARVED AND HIBERNATED SNAILS

HELIX LUCORUM: A FINE STRUCTURAL-CYTOCHEMICAL STUDY

V. K. Dimitriadis, D. Hondros & A. Pirpasopoulou

Department of Genetics, Development and Molecular Biology, School of Biology,
Aristotle University of Thessaloniki, Thessaloniki, 54006, Greece

ABSTRACT

The crop epithelium of the snail Helix lucorum consists of four cell types: ciliated and unciliated
columnar cells, mucous cells, and basal cells. The morphologic features of the epithelium show
that the digestive activities in the crop lumen are probably mediated by digestive enzymes
regurgitated from the digestive gland and the stomach, rather than secreted by the crop cells
themselves. On the other hand, the crop epithelium is probably responsible for the absorption of
nutrients from the crop lumen. The mucous cells secrete a periodate-reactive, non-sulfated and
non-carboxylated material, whereas the dense bodies observed in the columnar cells show a
positive reaction for acid phosphatases, as well as for periodate-reactive, sulfated and carbox-
ylated glycoconjugates and possibly are of lysosomal origin. Thirty seven days of hibernation
increased the number of mucous cells and the number of the dense bodies in the columnar cells,
while the amount of glycogen particles and lipid inclusion drastically decreased in non-mucous
cells compared to controls. Hibernation did not alter the carbohydrate content of mucous cells.
In addition, hibernation caused increased appearance of extrusion of cytoplasmic regions into
the crop lumen, as well as lysis of certain columnar cells compared to controls. Forty days of

starvation induced similar but less intense phenomena in the crop epithelium.
Key words: Helix lucorum, snail, crop, fine structure, hibernation, starvation, cytochemistry.

INTRODUCTION

The crop of Pulmonata consists of a thin
tube connected to the oesophagus and fol-
lowed by the main stomach. There are few
data about the structure and function of the
cells that constitute the pulmonate crop,
which is regarded as a region where food is
stored for certain periods (Runham, 1975).
Furthermore, there is little available informa-
tion on the morphology and function of these
cells at the ultrastructural level (Roldan, 1987;
Roldan 8 Garcia-Corrales, 1988). As far as
the digestive enzymes of the digestive tube of
molluscs are concerned, there are contradic-
tory views about their origin, for example
whether they are of exogenous or endoge-
nous origin or both (Fretter, 1952; Jeuniaux,
1954; Owen, 1966; Charrier, 1990).

In addition, little is known about the bio-
chemistry, physiology and fine structural cell
morphology of snails exposed to such condi-
tions as hibernation and starvation, and that
literature mainly refers to the digestive gland
epithelium (Sumner, 1965; Oxford & Fish,
1979; Janssen, 1985; Dimitriadis & Hondros,
in press). To our knowledge, the only related
information at the ultrastructural level comes

343

from the snail Theba pisana, in which starva-
tion resulted in the disappearance of lipid
droplets and glycogen particles in the colum-
nar crop cells (Roldan, 1987).

The present study was designed: (1) to ex-
amine the fine structural features of the crop
epithelium of the snail Helix lucorum; (2) to
cytochemically characterize the crop cells;
and (3) to study the effect of starvation and
hibernation on crop cell morphology. One fur-
ther question was also addressed: is the car-
bohydrate content of the mucous cells altered
in starved and hibernated snails compared to
controls?

MATERIALS AND METHODS

Adult Helix lucorum (Gastropoda, Pulmo-
nata, Helicidae) were collected from Edessa,
northern Greece. The largest shell diameter
of the snails used in the present study varied
from 41 to 43 mm, and the body weight was
between 19 and 21.5 g.

The effect of hibernation on crop cells was
examined in snails kept in a cold room at
4+1°C under a photoperiod of 9 h light : 15 h
dark (Lazaridou-Dimitriadou & Saunders,

344 DIMITRIADIS ET AL.

1986), for 37 days. Starved snails were kept
at 19+1°C and a photoperiod of 13L : 11D for
40 days. Control fed snails were kept at 19 +
1°C and a photoperiod 13L : 11D.

The tissues were fixed in Karnovsky's fixa-
tive (Karnovsky, 1965), postfixed in 2% os-
mium tetroxide, dehydrated, and embedded
in Spurr’s resin. Sections were cut using a
Reichert OmU3 ultramicrotome, post-stained
with uranyl acetate and lead citrate, and ex-
amined under a JEOL 100B electron micro-
scope operating at 80 KV. For light micro-
scopic observations, thick sections were
stained with 1% toluidine blue.

For carbohydrate cytochemistry, finely
minced pieces of crop were incubated over-
night in low iron diamine (LID) (Takagi et al.,
1982) or in high iron diamine (HID) (Spicer et
al., 1978; Sannes et al., 1979), treated with
osmium tetroxide and embedded in Spurr's
resin. Thin sections of these specimens were
stained with the thiocarbohydrazide-silver
proteinate (TCH-SP) sequence. Control tis-
sues were exposed т 1 M MgCl, in place of
LID or HID. Specimens without osmium
tetroxide treatment were used for the postem-
bedding periodate-thiocarbohydrazide-silver
proteinate (PA-TCH-SP) method (Thiéry,
1967). Control sections were stained without
the periodate treatment. Thin sections of
buffer- and osmium tetroxide-treated tissues
post-stained with uranyl acetate and lead cit-
rate were used to characterize normal crop
epithelial ultrastructure.

To demonstrate acid phosphatase activity,
a modified method similar to that proposed by
Barka & Anderson (1962) was applied. Tis-
sues sections approximately 1 mm in thick-
ness were fixed in glutaraldehyde and then
incubated in a medium containing 0.2 M tris/
maleate (pH 5) as a buffer an 0.1 M B-glyc-
erophosphate as substrate at 37° for 15-30
min. Control sections were incubated in the
absence of substrate.

Morphometric evaluation was performed
according to the methods described by Wei-
bel (1979) and Steer (1981). Samples of five
cells were analysed from each crop section of
four different snails. The volume density of
the dense bodies and of the lipid inclusions in
the supranuclear cytoplasm of the columnar
cells was determined from point counting ste-
reology, using a test square lattice with a pe-
riod а = 10 mm, equivalent to 1 рт on the
specimen. A minimum of 480 points were
counted per cell. The distribution of columnar
and mucous cells within the crop epithelium

were determined by measuring nucleated
cells of both types on light microscope micro-
graphs at a final magnification of x 500. Mean
values and standard deviations of the mor-
phometric parameters were calculated and
Statistically compared using Student's t-test,
significant level P<0.05.

RESULTS
Morphology of the Crop Epithelium

The light and electron microscope exami-
nations of the crop epithelium of Helix luco-
rum revealed that it is surrounded by connec-
tive tissue, muscular layers and a system of
haemocoelic spaces surrounded by amoebo-
cytes. The crop epithelium consists of four
cell types: ciliated and unciliated columnar
cells, mucous cells, and basal cells (Fig. 1).

The ciliated and unciliated cells are colum-
nar epithelial cells displaying similar morphol-
ogy to each other. Ciliated cells are the most
frequent cell type observed in the crop epithe-
lium. As far as the electron microscope tech-
niques permit, it is found that the ciliated and
unciliated cells are gathered in groups or ir-
regularly dispersed in the crop epithelium.

The nucleus of both ciliated and unciliated
cell types is usually in the middle portion of
the cells (Figs. 1, 2). Their cytoplasm contains
small quantities of rough endoplasmic reticu-
lum and a small number of Golgi complexes
usually located in the supranuclear cyto-
plasm. Numerous mitochondria are predomi-
nantly in the apex of the cells (Fig. 2). The
apical plasma membrane of the ciliated cells
forms well-developed microvilli, and long cilia,
displaying the “9 + 2” system of microtubules,
are regularly dispersed among the microvilli
(Figs. 2, 6). A well-developed microvillar bor-
der, similar to that observed in the ciliated
cells is also present in the unciliated cells
(Figs. 1, 2). Both ciliated and unciliated cells
possess moderate quantities of glycogen par-
ticles, lipid inclusions, and electron-dense
bodies in cytoplasmic areas adjacent to the
nuclei.

In certain cases, columnar cells showing
similar morphologic characteristics but pre-
senting a cytoplasm with a lower electron
density than the adjacent ciliated and uncili-
ated cells are apparent in crop epithelium.

A characteristic feature of columnar cells is
the presence of many electron-dense bodies
in cytoplasmic areas adjacent to their nuclei

CROP EPITHELIUM OF HELIX LUCORUM 345

SO ETE E =
MC> MC5

FIG. 1. Drawing of the various cell types observed in the crop epithelium of Helix lucorum. BC, basal cell;
CC, ciliated cell; CT, connective tissue; MC,_;, mucous cell in various stages of their development; MS,

muscles; UC, unciliated cell.

(Figs. 2, 17). The dense bodies react posi-
tively for periodate-reactive, sulfated and
carboxylated glycoconjugates when the
PA.TCH.SP, HID.TCH.SP and LID.TCH.SP
cytochemical sequences are applied, respec-
tively (Figs. 3, 4). Columnar cells display large
numbers of glycogen particles that react
strongly to periodate-reactive polysaccha-
rides by the PA.TCH.SP technique (Fig. 3);
they also contain lipid inclusions, usually lo-
cated in the middle and basal portion of the
cells, as well as a few infoldings of the basal
plasma membrane.

The third cell type, mucous cells, show a
cytoplasm that is largely composed of mu-
cous secretory granules (Figs. 1, 7). Mito-
chondria and rough endoplasmic reticulum
are dispersed throughout the remaining cyto-
plasm, and numerous Golgi complexes (Fig.
9) are frequently observed with their trans
face in opposition to a mucous granule. The
nucleus of these cells is usually situated in
their middle portions. In many cases, the
rough endoplasmic reticulum appears swol-
len and contains crystalline-like material (Fig.
8). Mucous granules are usually spherical in
shape, are 1-2 ¡um in diameter, and contain a
fibrillar matrix without cores (Figs. 7, 8).

Mucous cells are irregularly located across
the epithelium. Mucous cells of different sizes
and epithelial orientations from each other are
observed in transverse sections of the crop
epithelium (Fig. 1). The development of mu-
cous cells seems to occur in following stages:
(1) Spheroid or ovoidal mucous cell appear in

the base of the epithelium via differentiation of
basal cells. (2) The mucous cells increase in
size and extend their dorsal side towards the
apex of the epithelium. (3) The mucous cells
further increase in size; their dorsal side is in
close attachment to the luminal surface of the
epithelium. (4) The mucous cells are trans-
epithelially oriented. Their luminal surface is
limited and their apical region is narrow. (5)
The mucous cells are finally secreted into the
crop lumen.

At the same time, mucous cells showing
all the above-mentioned morphologic stages
are present in transverse sections of the crop
epithelium. Mucous cells displaying a non-
transepithelial orientation are also presented
in the apex of crop epithelium.

Mucous cells react positively to the
PA.TCH.SP cytochemical method indicating a
content rich in periodate-reactive polysaccha-
rides (Figs. 9, 10). Most of the reaction prod-
ucts are observed in a weblike structure within
the matrix of the secretory granules. A strong
PA.TCH.SP positive reaction is also found in
the Golgi complexes and glycogen particles of
the mucous cells (Fig. 9). The crystalline-like
material inside the rough endoplasmic reticu-
lum lack PA.TCH.SP reaction product (Fig.
10). Often, mucous cells located near the apex
of the epithelium display a large cisterna in
their apical portion (Fig. 11); this cisterna dis-
plays a negative reaction for carbohydrate and
acid phosphatase (Fig. 11).

Mucous granules react negatively to the
HID.TCH.SP and LID.TCH.SP sequences

DIMITRIADIS ET AL.

346

CROP EPITHELIUM OF HELIX LUCORUM 347

(Fig. 6), indicating absence of sulfated and
carboxylated glycoconjugates.

Finally, the fourth cell type, basal cells, are
elongated to oval, and their ventral surfaces
are attached to the basement membrane.
Few mitochondria and little rough endoplas-
mic reticulum are found in these cells.

Control sections of all the cytochemical
techniques used showed negative reactivity.

Morphology of the Crop Epithelium of
Starved and Hibernated Snails

Snails starved for 40 days or hibernated for
37 days presented large regions of crop epi-
thelium where ciliated and unciliated cells
were morphologically similar to those of con-
trols. In the latter cells, however, a significant
increase in the volume density of dense bod-
ies in the hibernated snails (Fig. 13), as well
as a significant decrease in the volume den-
sity of lipid inclusions in both starved and hi-
bernated snails was noted. The presence
also of glycogen particles was less apparent
in the experimental snails than in controls.

The application of the PA.TCH.SP,
HID.TCH.SP and LID.TCH.SP techniques in
the starved and hibernated snails indicated a
carbohydrate content of mucous cells similar
to that identified in controls, that is, the pres-
ence of periodate-reactive, non-sulfated and
non-carboxylated polysaccharides was ap-
parent in both experimental and control
snails.

Mucous cells in starved and hibernated
snails significantly increased in numbers (Fig.
12). In the experimental snails, the number of
the luminal surface-oriented mucous cells
was increased, while the mucous cells lo-
cated at the base of the epithelium was de-
creased compared to controls. In many

cases, mucous cells were smaller in size
compared to controls and were positioned in
the apex of the epithelium, displaying a non-
transepithelial orientation. The latter was
more apparent in the hibernated snails.

In the starved and hibernated snails, there
were also other crop epithelial regions where
increased lytic phenomena compared to con-
trols were observed. The latter phenomena
included lysis of certain cells (Fig. 15), as well
as disorganization of apical cell regions (Fig.
16) and extrusion of cytoplasmic regions into
the crop lumen (Fig. 17). The lytic phenom-
ena were more intense in the hibernated than
in the starved snails.

DISCUSSION

The available data concerning the role of
the crop epithelium in Pulmonata is inconsis-
tent. According to Owen (1966), the crop is
implicated in food storage and enzyme secre-
tion. In addition, van Weel (1961) suggested
the existence of extracellular digestion in the
crop lumen by enzyme activity of variable or-
igin.

It is not clear, however, whether the en-
zymes are secreted by the crop wall cells
(Owen, 1966), transferred from the digestive
gland (Hirsch, 1917; Fretter, 1952), or pro-
duced by bacteria in the crop lumen (Jeu-
niaux, 1954). In the present study, the colum-
nar cells of the snail Helix lucorum did not
show morphological features that support
their extensive secretory function. Such char-
acteristics would be, for example, large
amounts of rough endoplasmic reticulum, as
well as large number of Golgi complexes and
secretory granules produced by the Golgi
complexes. Therefore, the results of the

FIG. 2. Transverse view of crop epithelium. In the apical portion of the ciliated (CC) and unciliated cells (UC)
there are numerous mitochondria (Mi), while in their middle portion there are abundant lipid inclusions (Li).
Note the similarity in the morphology of ciliated and unciliated cell. The asterisk shows a columnar cell
presented less electron dense cytoplasm compared to the adjacent cells. Ci, cilia; Dbs, dense bodies; Lu,
crop lumen; MC, mucous cell; N, nucleus. Bar = 3 um.

FIG. 3. Detection of periodate-reactive polysaccharides (PA.TCH.SP technique). A positive PA.TCH.SP
reaction is shown in the dense bodies (Db) and glycogen particles (arrow) in a columnar cell. Un-counter-
stained section. Bar = 0.4 um.

FIG. 4. Detection of sulfated polysaccharides (HID.TCH.SP technique). A positive HID.TCH.SP reaction is
shown in a dense body (Db) in a columnar cell. Un-counterstained section. Bar = 0.3 am.

FIG. 5. Detection of acid phosphatase. A positive reaction for acid phosphatase indicates a dense body (Db)
in a columnar cell. Un-counterstained section. Bar = 0.35 шт.

FIG. 6. Detection of carboxylated polysaccharides (LID.TCH.SP technique). A negative LID.TCH.SP reac-
tion is noticed in the mucous granules (MG) of a mucous cell. Note the adequate ability of the technique to
stain the mitochondria (Mi) and the cilia (Ci) of a ciliated cell. Un-counterstained section. Bar = 1.5 um.

DIMITRIADIS ET AL.

348

CROP EPITHELIUM OF HELIX LUCORUM 349

Percentage of cells

Columnar cells

= Control

Starved
Hibernated

Mucous cells

FIG. 12. Percentage (%) of the columnar (ciliated and unciliated cells) and mucous cells in the crop
epithelium of H. lucorum. Significantly different values (P<0.05) between control and starved or control and

hibernated snails are indicated by an asterisk.

present study are consistent with the notion
(see, for example, Oxford, 1977) that diges-
tive activities in the crop of snails are probably
mediated by digestive enzymes regurgitated
from the digestive gland or the stomach,
rather than by enzymes secreted by the crop
cells.

The presence of a large number of glyco-
gen particles in the columnar cells indicates
that the cytoplasm of these cells is a site of
carbohydrate storage. Because other studies
show cellulase and chitinase activity in the
crop lumen of Helix (Jeuniaux, 1955; Stras-
dine & Whittaker, 1963; Koopmans, 1967;
Flari & Charrier, in press), it is possible that
the columnar cells of crop epithelium of H.
lucorum absorb oligosaccharides from crop
lumen, part of which are stored as glycogen
particles. In addition, the presence of a very

well-developed microvillar border and numer-
ous mitochondria in the apex of crop colum-
nar cells of H. lucorum, as well as the pres-
ence of moderate quantities of lipid inclusion
in their cytoplasm, possibly indicates an ex-
tensive absorptive function coexisting with
their function as cells that propel mucus. In
the snail Theba pisana, an absorptive func-
tion is also suggested for the crop epithelial
cells (Roldan & Garcia-Corrales, 1988).

The dense bodies observed in the digestive
cells of H. lucorum crop have been also found
in the crop or intestinal columnar and mucous
cells in different gastropods (pulmonates and
prosobranchs) (Carriker & Blistad, 1946;
Bowen, 1970; Boquist et al., 1971; Lufty 8
Demian, 1976; Angulo et al., 1986). In the
present study, the positive acid phosphatase
reaction established the lysosomal origin of

FIG. 7. Mucous granules (MG) containing a fibrillar matrix without cores inside a mucous cell. Bar = 2 um.
FIG. 8. Crystalline-like material (asterisk) appears inside the swelled rough endoplasmic reticulum of a
mucous cell. MG, mucous granule. Bar = 0.3 um.

FIG. 9. An intense PA.TCH.SP positive reaction show the saccules of the Golgi complexes (Go), the mucous
granules (MG) and the glycogen particles (arrow) in a mucous cell. Un-counterstained section. Bar = 0.6
um.

FIG. 10. The granules of a mucous cell show a reticulated PA.TCH.SP positive reaction. Note the negative
reaction on the crystaline-like inclusions inside the rough endoplasmic reticulum (arrow). Un-counterstained
section. Bar = 0, 35 um.

FIG. 11. A mucous cell located near the apex of the epithelium displays PA.TCH.SP positive mucous
granules and a large cisterna showing a negative reaction (asterisk). Mv, microvilli; N, nucleus. Bar = 3 um.

350 DIMITRIADIS ET AL.

0.14

0.12

0.10

0.08

0.06

0.04

Volume density

0.02

0.00

Dense bodies

M Control
Starved

Hibernated

Lipid inclusions

FIG. 13. Volume density of dense bodies and lipid inclusions in the supranuclear cytoplasm of columnar cells
(ciliated and unciliated cells). Significant different values (P<0.05) between control and starved or control

and hibernated snails are indicated by an asterisk.

these structures. A positive acid phosphatase
reaction was also observed in apical vacuoles
and granules of the crop and intestinal colum-
nar cells, and in all the intestinal gland cells of
the slug Arion ater (Bowen, 1970; Angulo et
al., 1986). The lysosomal origin of the dense
bodies of crop digestive cells of H. lucorum is
additionally supported by the positive
PA.TCH.SP, HID.TCH.SP and LID.TCH.SP
reactions, indicating the presence of period-
ate-reactive, sulfated and carboxylated glyco-
conjugates, respectively. The association of
the latter reactions with lysosomal activities is
also referred to in studies on mammalian leu-
kocytes (Bruyn et al., 1975, Spicer et al.
1978), as well as in studies on H. lucorum

digestive gland cells (Dimitriadis & Liosi, in
press).

There is little information about carbohy-
drate histochemistry and cytochemistry of the
molluscan digestive epithelium. In the slug
Arion ater (Angulo et al., 1986), intestinal
gland cells showed either a positive reaction
for acid and sulphated mucosubstances or a
positive one for neutral polysaccharides. In
the molluscs Semerula maculata (Varute &
Patil, 1971) and Venus mercenaria (Zacks,
1965), sulfated acid and neutral polysaccha-
rides, respectively, were found in their intes-
tinal mucocytes. The mucous cells of H. luco-
rum crop epithelium, as the present study
indicates, were positive to the PA.TCH.SP re-

FIG. 14. 25 days of starvation. Numerous dense bodies (Dbs) are located in the apical portion of certain

columnar cells. Bar = 1.9 pm.

FIG 15. 40 days of starvation. A crop epithelial cell in a stage of its lysis. N, nucleus Bar = 1 um.

FIG. 16. 37 days of hibernation. Cisternae with membranic materials are located in the apex of a ciliated
cells. Bar = 1 um.

FIG. 17. 37 days of hibernation. A columnar cell (asterisk) is extruded into the crop lumen. This phenomenon
is more apparent in the experimental snails than in controls. Note the numerous dense bodies (arrows) in
the supranuclear cytoplasm of two columnar cells. Mi, mitochondria; Mv, microvilli; N, nucleus. Bar = 1.5
pm.

351

CROP EPITHELIUM OF HELIX LUCORUM

352 DIMITRIADIS ET AL.

action and negative to both HID.TCH.SP and
LID.TCH.SP sequences, indicating periodate-
reactive glycoconjugates, which lack sulfated
and carboxylated esters. However, it is pos-
sible that the polysaccharides of mucous cells
in H. lucorum crop epithelium contain hyalu-
ronic and sialic acid, which are stained by al-
cian blue (pH = 2.5), as was demonstrated in
mucocytes in the intestinal gland cells of the
slug Arion ater (Angulo et al., 1986).

Morphology of the Crop Epithelium of
Starved and Hibernated Snails

The existing data on morphological
changes in the fine structure of the crop epi-
thelial cells of Pulmonata under starvation or
hibernation is not sufficient to give a clear in-
sight into the physiology of the cells under
these conditions. There are reports indicating
biochemical and physiological differences in
enzyme secretion, changes in the concentra-
tion of the components of extracellular body
fiuids (Meenakshi, 1956), changes in the os-
motic pressure and respiratory rate (Ghiretti,
1966), and changes in major metabolic prod-
ucts (Florkin, 1966) due to hibernation or star-
vation.

The light and electron microscope observa-
tions of the crop cells of H. lucorum after 40
days of starvation and 37 days of hibernation
showed a significant increase in number of
the dense bodies located near the nuclei of
the columnar cells compared to controls. This
increase probably reflects increased lysoso-
mal activity and/or increased intracellular di-
gestion of nutrients by these cells during star-
vation and hibernation.

The increased lytic phenomena and extru-
sions of cytoplasmic regions into the crop lu-
men of the snails H. lucorum during starvation
and hibernation compared to controls are
probably related to the reduction of the en-
ergy requirements and the maintenance of
low function activity by the crop epithelial cells
during the experimental periods. In contrast to
our observations, autophagic vacuoles were
formed in tissues of other organisms under
starvation (Bowen, 1968; Bauer et al., 1977)
that were related to cell autophagy. The non-
appearance of autophagic vacuoles in the
crop cells of H. lucorum during starvation and
hibernation probably indicates the snail's ad-
aptation to these conditions.

Another effect that starvation and hiberna-
tion induced in the crop epithelium of H. luco-
rum was the increase in the number of mu-

cous cells in the starved and hibernated
snails compared to controls. It is uncertain
whether the increase in the number of mu-
cous cells in the starved and hibernated
snails is a result of increased mucous cell de-
velopment or of the accumulation of mucous
cells in the crop epithelium, reflecting de-
creased mucous cell secretion due to starva-
tion and hibernation.

In conclusion, by using light and electron
microscopic observations, in combination
with cytochemical characterization and mor-
phometric evaluation, the present study pro-
vides information about the effect of starva-
tion and hibernation on the cells of the crop
epithelium of the snail H. lucorum. The results
of the present study should be considered as
preliminary. Additional information, especially
by the use of cytochemical techniques, is
needed in order to clarify the physiology of
snails during such conditions as hibernation
and starvation.

ACKNOWLEDGMENTS

This work was financially supported by the
Greek Ministry of Agriculture. We are grateful
to Dr. M. Lazaridou-Dimitriadou for the provi-
sion of the snails and their keep under nor-
mal, starvation and hibernation conditions.

LITERATURE CITED

ANGULO, E., J. MOYA & I. de la VEGA, 1986,
Histochemical observation on the intestinal epi-
thelium of the slug Arion ater. Cuadernos Inves-
tigacion Biologica (Bilbao) 9: 59-65.

BARKA, T. & P. J. ANDERSON, 1962, Histochem-
ical methods for acid phosphatase using hexazo-
nium parasolinin as coupler. Journal of His-
tochemistry and Cytochemistry, 10: 141-153.

BAUER, R., W. RUDIN & H. HECKER, 1977, Ultra-
structure changes in midgut cells of female
Aedes aegypti L (Insecta, Diptera) after starva-
tion or sugar diet. Cell and Tissue Research, 177:
215-219.

BOQUIST, L., S. FALKEMER & B. K. MEHROTRA,
1971, Ultrastructural search for homologues of
pancreatic beta-cells in the intestinal mucosa of
the mollusc Buccinum undatum. General Com-
parative Endocrinology 17: 236-239.

BOWEN, 1. D., 1968, Electron microscopic studies
on autophagy in the gut epithelial cells of the lo-
gust Schistocerca gregaria. Histochemical Jour-
nal, 1: 141-151.

BOWEN, I. D., 1970, The fine structure localization
of acid phosphatase in the gut epithelium cells of

CROP EPITHELIUM OF HELIX LUCORUM 353

the slug Arion ater (L.). Protoplasma 70: 247-
270.

BRUYN, P. P. H. de, S. MICHELSON & R. B.
BECKER, 1975, Endocytosis, transfer tubules
and lysosomal activity in myeloid sinusoidal en-
dothelium. Journal of Ultrastructure Research,
53: 133-151.

CARRIKER, М.В. 8 N. M. BILSTAD, 1946, Histol-
ogy of the alimentary system of the snail Lym-
naea stagnalis Transactions. American Micro-
scopical Society 65: 250-275.

CHARRIER, M., 1990, Evolution, during digestion,
of the bacterial flora in the alimentary system of
Helix aspersa (Castropoda: Pulmonata): a scan-
ning electron microscope study. Journal of Mol-
luscan Studies, 56: 425-433.

DIMITRIADIS, V. K. & D. HONDROS, Effect of star-
vation and hibernation on the fine structural mor-
phology of digestive gland cells of the snail Helix
lucorum. Malacologia (in press).

DIMITRIADIS, V. K. & M. LIOSI, Ultrastructural lo-
calization of periodate-reactive complex carbohy-
drates and acid-alkaline phosphatases in diges-
tive gland cells of fed and hibernated Helix
lucorum (Mollusca: Helicidae). Journal of Mollus-
can Studies (in press).

FLARI, V. & M. CHARRIER, Contribution to the
study of carbohydrases in the digestive tract of
the edible snail Helix lucorum L. (Gastropoda:
Palmonata: Stylommatophora) in relation to its
age and its physiological state. Journal of Mol-
luscan Studies (in press).

FLORKIN, M. 1966, Nitrogen metabolism. Pp.
309-351, in: K. M. wiLBUR & С. M. YONGE, eds,
Physiology of Mollusca. Vol. 2. Academic Press,
New York, San Francisco.

FRETTER, V., 1952, Experiments with 92P and *?”/
on species of Arion Hellix and Agriolimax. Quar-
terly Journal of Microscopical Science 93: 133-
146.

GHIRETTI, F., 1966, Respiration. Pp. 175-208, in:
К. M. wILBUR 8 С. М. YONGE, eds., Physiology of
Mollusca. Vol. 2. Academic Press, New York,
San Francisco.

HIRSCH, G. C., 1917, Ernahrungsbiologie fleis-
chfressender Gastropoden. |. Makroskopischer
Bau, Nahrung, Nahrungsaufnahme, Verbauung,
Sekretion. Zoologishe Jahrbúcher, Abteilung
Allgemeine Zoologie und Physiologie 35: 357-
504.

JANSSEN, H. H., 1985, Some histophysiological
findings on the mid-gut gland of the common gar-
den snail, Arion rufus (L.) Zoologischez Anzeiger
(Jena) 215: 33-51.

JEUNIAUX, C., 1954, Sur la chitinase et la flore
bacterienne intestinale des mollusques gastero-
podes. Bulletin Academie Royale de Belgique,
Classe des Sciences 28: F7.

KARNOVSKY, M. J., 1965, A formaldehyde-glu-
taraldehyde fixative of high osmolarity for use in
electron microscopy. Journal of Cell Biology 27:
137A-138A.

KOOPMANS, J. J. C., 1967, The nature and origin

of cellulases in Helix pomatia. Acta Physiologica
et Pharmacologica Neerlandica 14: 349-350.

LAZARIDOU-DIMITRIADOU, M. & D. C. SAUN-
DERS, 1986, The influence of humidity, photo-
period and temperature on the dormancy and
activity of Helix lucorum L. (Gastropoda, Pul-
monata). Journal of Molluscan Studies, 52: 180—
189.

LUFTY, R. G. 8 E. S. DEMIAN, 1976, The histology
of the alimentary system of Marisa cornuarietis
(Mesogastropoda : Ampullariidae). Malacologia
5: 375-422.

MEENAKSHI, V. R., 1956, Physiology and hiberna-
tion of the apple snail Pila virens (L.). Current
Science 10: 321-322.

OWEN, G., 1966, Digestion. Pp. 53-96, in: K. M.
WILBUR & C. M. YONGE, eds., Physiology of Mol-
lusca. Vol. 2. Academic Press, New York, San
Francisco.

OXFORD, G. S., 1977, Multiple sources of esterase
enzymes in crop juice of Cepaea (Mollusca, He-
licidae). Journal of Comparative Physiology 122:
375-383.

OXFORD, G. S. & L. J. FISH, 1979, Ultrastructural
localization of esterase and acid phosphatase in
digestive gland cells of fed and starved Cepaea
nemoralis (L.). Protoplasma 101: 186-196.

ROLDAN, C., 1987, Ultrastructural modification of
the epithelium in the anterior digestive tract in
starved specimens of Theba pisana (Mollusca,
Gastropada, Pulmonata) /berus, 7: 153-164.

ROLDAN, C. & B. GARCIA-CORRALES, 1988,
Anatomy and Histology of the alimentary tract of
the snail Theba pisana (Gastropoda: Pulmo-
nata). Malacologia, 28: 119-130.

RUNHAM, N. W., 1975, Alimentary canal. Pp. 53—
105, in: V. FRETTER & J. PEAKE, eds., Pulmonates.
Vol. 1. Functional anatomy and Physiology. Lon-
don, Academic Press.

SANNES, P. L., S. SPICER 8 T. KATSUYAMA,
1979, Ultrastructural localization of sulfated and
complex carbohydrates with a modified iron di-
amine procedure. Journal of Histochemistry and
Cytochemistry 27: 1108-1111.

SPICER, S. S., J. H. HARDIN & M. E. SETSER,
1978. Ultrastructural visualization of sulfated
complex carbohydrates in blood and epithelial
cells with the high iron diamine procedure. His-
tochemical Journal, 10: 435-452.

STEER, M. W., 1981, Understanding cell structure.
Cambridge University Press, Cambridge.

STRASDINE, С. A. & D. В. WHITTAKER, 1963, On
the origine of the cellulase and chitinase of Helix
pomatia. Canadian Journal of Biochemistry and
Physiology 41: 1621-1626.

SUMNER, A. T., 1965, The cytology and cytochem-
istry of the digestive gland cells of Helix. Quar-
terly Journal of Microscopical Science 106: 173—
192.

TAKAGI, M., T. В. PARMLEY, $. $. SPICER, Е. В.
DENYS & M. E. SETSER, 1982, Ultrastructural
localization of acidic glycoconjugates with the low

354 DIMITRIADIS ET AL.

iron diamine method. Journal of Histochemistry
and Cytochemistry, 30: 471-476.

THIÉRY, J. P., 1967, Mise en evidence des
polysaccharides sour coupes fines en micro-
scopie électronique. Journal of Microscopie 6:
987-1018.

VARUTE, A. T. 8 V. A. PATIL, 1971, Histochemical
analysis of molluscan stomach and intestinal al-
kaline phosphatase: a sialoglycoprotein. His-
tochemie, 25: 77-90.

WEIBEL, E. R., 1979, Stereological methods. Vol.

1: Practical methods for biological morphometry.
Academic Press, New York and London.

WEEL, P.B. van, 1961, The comparative physiology
of digestion in molluscs. American Zoologist 1:
245-272.

ZACKS, $. |., 1955, The cytochemistry of the
amoebocytes and intestinal epithelium of Venus
mercenaria (Lamellibranchiata), with remarks on
a pigment resembling ceroid. Quarterly Journal
of Microscopical Science, 96: 57-71.

Revised Ms. accepted 9 March 1992

MALACOLOGIA, 1992, 34(1-2): 355-369

GENETIC SIMILARITIES AMONG CERTAIN FRESHWATER MUSSEL
POPULATIONS OF THE LAMPSILIS GENUS IN NORTH CAROLINA

Alan E. Stiven' and John Alderman?

ABSTRACT

The objective of this research was to explore ecological and genetic similarities among several
southeastern freshwater mussel populations, focusing especially on two rare and potentially
endangered “subspecies” of Lampsilis radiata residing in river systems in the Piedmont and
coastal plain of North Carolina. Conchological analyses and the examination of genetic identity
patterns derived from 11 electrophoretic loci in six different populations or species formed the
basis of our interpretations.

Two currently recognized subspecies of Lampsilis radiata (radiata and conspicua) are now
quite rare in North Carolina waters, and one, L. r. conspicua, may be endemic to the state.
However, we find that in spite of the difference in unadjustable sample mean shell length, they
do not differ in general shell morphometry. For example, covariance analyses indicated homo-
geneity of shell length-height regression coefficients, and no difference in adjusted mean
lengths. A similar conchological comparison of these two “subspecies” with the smaller sized
recently described Lampsilis fullerkati of Lake Waccamaw also failed to find any significant
conchological differences among these populations.

Nei’s unbiased genetic distance and Rogers modified genetic distance derived from electro-
phoretic analyses consistently separated on established Ellipito, Leptodea and Lampsilis spe-
cies. However, these genetic distance measures did not distinguish between the two rare Lamp-
silis radiata “subspecies” nor between these two groups and Lampsilis fullerkati: Nei's unbiased
identity ranged from 0.945 to 0.978 among these three populations. Phenograms and a Wagner
tree derived from Rogers distance supported evolutionary similarities among these three pop-
ulations.

We also found evidence of site effects (conchological differences in different habitats) in
Lampsilis cariosa and Leptodea ochracea, the two species examined for such a phenomenon.
Genetic identity values among these intraspecific but allopatric populations ranged from 0.922—
0.982. We suggest that until contrasting genetic and distinctive biological data are forthcoming,
the two “subspecies” of L. radiata and possibly L. fullerkati simply be considered as a multiple
population complex of Lampsilis radiata.

Key words: freshwater mussel, Lampsilis radiata, ecology, genetics, North Carolina, genetic

distance, phenogram, Wagner tree.

INTRODUCTION

Freshwater bivalves constitute a large and
very diverse assemblage of species in North
Carolina and other southeastern states (R.I.
Johnson, 1970; Burch, 1975; Davis, 1984).
Many of these mussels and clams are highly
vulnerable to pollutants, and changes in their
abundance, survival and reproductive status
can be a warning signal of the declining
health of an estuary or river system (Goldberg
et al., 1978). Because many freshwater mus-
sels in North Carolina are already in a state of
decline (Clarke, 1983), it is imperative that
work focus on the assessment of potentially
rare and endangered species of freshwater
mussels, especially in the genera Lampsilis,

Lasmigona, Elliptio, and Alasmidonta (Clarke,
1983; Keferl & Shelley, 1988; Alderman,
1988). The development and use of the “rare
and endangered” classification, however, has
been hampered by insufficient ecological, ge-
netic and systematic information, as well as
confusion even over the species concept in
this group. Many early mussel investigators
relied on shell morphology for distinguishing
species, but this approach is often suspect,
especially in light of the extensive concholog-
ical divergence in this group (Davis, 1984).
The freshwater mussel genus Lampsilis is
probably one of the most successful and spe-
cialized groups of the Unionidae (Davis,
1984), with more than 20 living species in
North America. At least five of these species

"Department of Biology, University of North Carolina, Chapel Hill, North Carolina 27599-3280, U.S.A.
2Nongame Program, North Carolina Wildlife Resources Commission, Raleigh, North Carolina 27604-1188, U.S.A.

356 STIVEN 8 ALDERMAN

are found in North Carolina waters and some
may be endemic to the state. Currently, the
American Malacological Union recognizes
two subspecies of Lampsilis radiata (Gmelin,
1791), L. г. radiata (eastern lampmussel) and
L. r. conspicua (l. Lea, 1872) (Carolina fat-
mucket) (Turgeon et al., 1988). The eastern
lampmussel is found largely in coastal plain
waters and occupies a broad geographic
range from the Pee Dee drainage north to the
St. Lawrence River. Collections of this “sub-
species” in North Carolina waters during the
past four years include sites in the Chowan
River, Swift Creek, Fishing Creek and the
lower main channel of the Tar River (Clarke,
1983; Alderman, 1988). Shelley (1987) lists
its occurrence in the Tar, Neuse, and Cape
Fear rivers. The second “subspecies,” the
very rare Carolina fatmucket, is apparently
found only in North Carolina and may be en-
demic to this state. It has been collected in the
tributaries of the Pee Dee River drainage (e.g.
the Little and Uwharrie rivers), and recently in
the Flat River of the upper Neuse River drain-
age basin (Alderman, 1988, 1989). In prior
collections of these two “subspecies,” L. r.
radiata often exhibited smaller shells (Al-
derman, 1989). Lampsilis r. radiata is also
found in waters containing anadromous fish,
and its larval glochidia may be parasitic on
such fish. If these two populations are suffi-
ciently different to warrant “species” status,
then the rarer L. r. conspicua is a prime can-
didate for federal listing as “endangered.”
This study utilized biochemical and popula-
tion genetic methods (Rogers, 1972; Davis et
al., 1981; Davis, 1984) to probe several ques-
tions about the genetic relatedness among
certain North Carolina mussel populations.
Our prime question dealt with ecological and
genetic differences in the two Lampsilis radi-
ata “subspecies.” In addition, we explored
three other questions within this genus. The
first focused on the recent discovery of a
small population of a Lampsilis-like species in
the Deep River near its confluence with the
Cape Fear River. This population is com-
prised of only large old specimens and is
thought to be declining. Clarification of its ge-
netic structure and possible systematic status
is required (Alderman, 1990). The second
question related to the genetic and ecological
status of Lampsilis cariosa (Say, 1817).
Mostly large L. cariosa are now found in sites
formerly containing both young and adults
(e.g. Tar and upper Neuse drainages) (Al-
derman, personal observation), and its level

of genetic heterozygosity is unknown. Finally,
Lampsilis fullerkati R. |. Johnson, 1984, a
Lake Waccamaw species, was first discussed
by Kat (1983) and then described as a new
species by R. I. Johnson (1984). Kat (1983)
noted that the genetic structure of this species
was similar to Lampsilis radiata from New
Brunswick and Nova Scotia. To help clarify
the systematic status of Lampsilis fullerkati,
we compared its genetic structure with North
Carolina Lampsilis radiata specimens as well
as other Lampsilis species.

METHODS
Study Sites and Collections

The description of all collection sites, as
well as the species/populations and numbers
of mussels collected, along with their ANSP
catalog number (except the two rare Lampsi-
lis radiata subspecies as noted below), are
given in Figure 1 and Appendix 1.

Collections of mussels were made from
shallow water by hand and from deeper areas
by snorkeling and SCUBA. Mussel species of
prime interest to this study were generally too
sparse to use quadrat sampling to estimate
relative abundance. Specimens of the two
rare “subspecies,” L. r. radiata and L. r. con-
spicua, were returned to their original sites
following shell measurements and the re-
moval of a small tissue sample from the man-
tle for use in genetic analysis. Individuals of
all other mussel species collected were sac-
rificed, tissue samples removed, shell mea-
surements obtained, and shell and remaining
soft body tissue preserved. Shell measure-
ments included maximum length, height, and
width. Leptodea ochracea (Say, 1817), for-
merly placed in Lampsilis, was identified as
such by the lack of a distinct mantle flap (this
flap is located ventral to the incurrent aperture
in Lampsilis).

Genetic Methods

All animals were processed within 10 days
of collection. Shells were pried open and tis-
sue samples of mantle and liver (hepatopan-
creas) were removed, combined and pro-
cessed, except in the case of the rare
“subspecies” as noted above. Tissues were
chopped on cold glass with a razor blade,
then homogenized (sonified) in an equal vol-
ume of grinding buffer (0.1 M Tris, 0.001 M

LAMPSILIS IN NORTH CAROLINA 357

T > { < ) ad AL, Г +
Mapids Lake \ | À

> Boe} \

IN

À | \

а |
ur

\
À NAAA
Reservoir \, a
\ /

ang > = y Wi AR

Townsend Lake

E A | 7 Еа 8 Везегуои` as >
> SEE FEIN aten \ | > : y SES = ) N ae \
\ ep Pop an jews Lake 7 ~ ( E = x SAT Sy “7 | TN Ñ
=> Yadkin: ‘Pee Dee \ \ | > ( ХУ LT
X i [ | Lexigpon Nor пуле servoir N Reverb er eos y À N Ve Lat | ) \
Wgokour Shos Is La ke y! и a > \ \ y УЧ E (
ET high Rock Lake — TOS NM Tar-Pamlico) (3 y au. = DY
+ 0% Túc kertown 8$ Servo!r AN f 0 à ENS 4 Sa Tari 2
A Ge ; \ \ № SL 4< rramuskeet yy
na UN AOS N SES BETA MESA LAS j
N S Ne Ne ST = SA NE e ) Se x | y ASS e TNT
a —~ \ \ 1 — y [< ) y
she SN AEN | | «IS Neuse RNCS = ya
\ { ? Lake m. I\¢ 4 A \ Am „u y A
\Y IA | 3 E LA >
app N MEANS Es oh и
A / у ‘ у y ne Ie < J ral 1 (és / \ \ ) | N NU / A я
VHA Vag ки) 7a J \ Ш | AN Sy a\\ \
NE 1 ff me); Is Lake \ \\&abb Fe) | > AS 7
IN _/ р 1 Cake \ о Саре Fear | nce Bke wide dy APT /
TAS MANN { TIA New РН
= ( 2 BUN WMA Suggs Mi ола \\ I SINE AIO
INN \\ ZARN S| Black Lak WAS, \
ZEN NT E Lake) | ) =
UN Lumber Ne OX Wy, ® E
Ne a SI
(/ D N WN AS JE
> О: N N | 14
Lake Wafcama iy `_ я
у ) Es }
OIT MAO 7
)

FIG. 1. Map of eastern North Carolina showing major river systems and the location of the collection sites
(A-G) for this study. See Appendix | for details of each site.

EDTA, and 5 x 10 ° M NADP, pH 7.0) for
about 20-50 sec. The sample was then cen-
trifuged for 30 min at 12,000 g. Sample ex-
tracts were stored at —70°C. Protein extracts
were then analyzed by horizontal starch gel
electrophoresis (Electrostarch, Lot 89) follow-
ing procedures outlined by Shaw & Prasdad
(1970), Selander et al. (1971), Davis et al.
(1981), Kat (1983), Davis (1984), and Stiven
(1989). The 11 enzyme loci used (Table 1)
correspond to most of the loci utilized by oth-
ers in molluscan genetic research (e.g. Davis
et al., 1981; Kat, 1983). The numbering of loci
and lettering of alleles were from the most
anodal region back to the origin on the gel.
Sample tissues of Elliptio complanata (Light-
foot, 1786) and Leptodea ochracea were run
with the other species and subspecies to pro-
vide allelic consistency. Aminopeptidase 1
and 2 (Ala-1 and Ala-2) are associated with
the substrate leucyl-alanine. Leucine ami-
nopeptidase (Lap) or a-aminoacyl-peptide hy-
drolase (cytosol) was resolved using L-leucyl-
B-naphthylamide and fast black K salt. The
position of this locus differed from the Ala loci
on the same gel.

The genetic data were analyzed using
Swofford & Selander’s (1981) BIOSYS-1
computer program. P, the proportion of loci

that are polymorphic under the 0.95 criterion,
and H, the direct-count heterozygosity, were
measures of genetic variability. Levels of het-
erozygote deficiency were also estimated by
D = (Но-НЕ)/НЕ, where Но and HE are the
observed and expected frequencies of het-
erozygotes respectively (Koehn et al., 1973).
Departure of genetic frequencies from Hardy-
Weinberg expectations were assessed by
chi-square analysis adjusted for small sam-
ple sizes (Levene, 1949). All statistical anal-
yses were carried out using SYSTAT (Wilkin-
son, 1990).

Genetic distance/similarity among species-
populations were measured by Nei’s nonmet-
ric unbiased genetic distance and genetic
identity (Nei, 1972, 1978) and Rogers metric
modified distance (Rogers, 1972; Wright,
1978). Each has its specific assumptions and
limitations, but all are related to Е т (Wright's
measure of genetic divergence-Wright, 1978;
Hartl & Clark, 1989). Nei’s unbiased mea-
sures are less sensitive to small sample sizes
than are Nei’s biased measures (Nei, 1978).
There is no upper bound on the magnitude of
Nei’s genetic distance measure. We note also
that when distance measures are relatively
large between pairs of species, and heterozy-
gosity is low, the construction of phenograms

358 STIVEN 8 ALDERMAN

TABLE 1. Isoenzymes and buffers utilized in the genetic analysis of freshwater mussels from North

Carolina waters.

Locus/Isoenzyme (No. Loci)

Phosphoglucose isomerase (1)
Phosphoglucomutase (2)

Peptidase (2)

Leucine aminopeptidase (1)

Malate dehydrogenase (1)
Phosphogluconate dehydrogenase (1)
Mannose 6-phosphatase isomerase (1)
a-glycerophosphate dehydrogenase (1)
Superoxide dismutase (1)

'Selander et al. (1971)

Abbrev. E.C. No Buffer
Gpi 5.3.1.9 LiOH'
Pgm 2.7.51 LiOH
Ala 3.4.1.2 LiOH
Lap 3.4.11.1 LiOH
Mah IES CM-6?
6-Pgd 112143 CM-6
Mpi 5.3.1.8 CM-6
a-Gpdh AS ТЕВ-93
Sod SA TEB-9

28.4 g citric acid per liter, to pH 6 with N-(3aminopropyl) morpholine. Gel buffer 1 part electrode: 19 parts water.

3Davis et al. (1981)

can be carried out fairly reliably with only a
few representative individuals for a species
(Nei, 1978). Phenograms were constructed
for Neïs unbiased genetic distance (Nei,
1978) and for modified Rogers genetic dis-
tance (Rogers, 1972; Wright, 1978) using the
unweighted pair group method of construction
(Sneath & Sokal, 1973). A Wagner tree phe-
nogram (Farris, 1972; Felsenstein, 1983) was
also constructed using Rogers modified met-
ric distance coefficient (Wright, 1978), with
the tree rooted at the midpoint of the greatest
patristic distance. Roger's metric measure
satisfies the Wagner requirement of triangle
inequality. This method does not assume
constant evolutionary rates (Felsenstein,
1983).

RESULTS

Ecological and Shell Parameters

Only nine individuals of the rare L. r. radiata
were obtained during many searches (by
hand during wading, and SCUBA in deeper
waters), some lasting many hours. Most were
found in fine to medium sand and organic silt
substrates. In the lower Chowan River site,
Leptodea ochracea was about three times as
abundant as L. r. radiata. In contrast, a 40-
min search on 5-17-90 by two persons in the
South Flat River yielded 40 Elliptio compla-
nata and seven individuals of other species.
Lampsilis r. conspicua, also rare, was ob-
tained from the Flat River in Durham County.
This subspecies was found primarily in
coarse sand especially along sand bars.

Figure 2 illustrates the shell length vs.

height regressions for L. r. conspicua and L. r.
radiata. Although these “subspecies” differed
significantly in mean shell length (F,3 =
29.5, P<0.0001) (Table 2), covariance anal-
ysis indicated that the length-height regres-
sion coefficients did not differ (F, 53, = 0.61, P
= 0.44), suggesting similar relative concholo-
gies. Shell length, adjusted for differences in
shell height, was 105.7 mm and 104.1 mm for
L. r. conspicua and L. r. radiata respectively,
and this difference is not significant (F, 53 =
0.63, P = 0.43). Shells of L. fullerkati from
Lake Waccamaw were considerably smaller
than the two subspecies (Table 2). However,
covariance analysis of length-height regres-
sions among L. г. conspicua, L. г. radiata, and
L. fullerkati indicated no significant differ-
ences (F> s, = 0.32, P = 0.49), and adjusted
mean lengths also did not differ among these
three Lampsilis populations (F263 = 2.72, P
= 0.073). In contrast, shell length-height re-
gression comparisons among all combina-
tions of Elliptio complanata, Lampsilis cari-
osa, and Leptodea ochracea were significant.

Leptodea ochracea and Lampsilis cariosa
showed evidence of the “site effects” that
arise from conchological differences in popu-
lations of the same species drawn from differ-
ent sites (Kat, 1982; Hinch et al., 1986), or
conversely, conchological similarities among
different species found in the same site (Horn
8 Porter, 1981). In L. cariosa, mean shell
length was larger (F,:, = 46, P<0.001; and
Tukey test) at the Deep River site (116.4 mm)
than at the Cape Fear (60.6 mm) or Tar River
(61.1 mm) sites. Similarly, Leptodea ochracea
was much larger in the Chowan River (92.5
mm) than in Lake Waccamaw (37.6 mm) (F, 72
= 847, P<0.0001; and Tukey test).

LAMPSILIS IN NORTH CAROLINA 359

150

130

110

LENGTH (mm)
©
©

ы
о

rc: ©
50 [.г.г. ©

30
10 20 30 40 50 60 70 80 90 100

НЕСНТ (тт)

FIG. 2. Plot of shell length against shell height (mm) for samples о the freshwater mussels, Lampsilis radiata
conspicua (open circles) and L. r. radiata (closed circles) from North Carolina waters. Refer to Appendix | for
collecting sites.

TABLE 2. Summary of shell measurements for the freshwater mussels Lampsilis radiata radiata,
Lampsilis radiata conspicua, and Lampsilis fullerkati from collections taken in North Carolina waters,
1990. Values are means + SE.

Shell Parameters Е. г. conspicua L. г. radiata L. fullerkati
Sample Size 23 9 35

Length 113.0 + 2.88 83.1 + 4.87 47.1 + 0.89
Height 64.6 + 1.65 46.4 + 2.57 23.5 + 0.44
Width 40.2 + 1.67 30.0 + 2.27 5:5 ЕЕТО:25
Genetic Analyses species/subspecies are given in Table 3. Av-

Diversity and Heterozygosity: Alleleic fre- erage direct-count heterozygosity and P-
quencies and measures of genetic diversity values were high in the more common and
derived from the 11 loci for all populations/ wide-spread species, Elliptio complanata and

360 STIVEN & ALDERMAN

TABLE 3. Alleleic frequencies, sample sizes, and locus-specific direct-count heterozygosity, along with
percentage of loci polymorphic (P at the 0.95 criterion), and mean direct-count heterozygosity (H + 1
S.E.) for the nine populations/species of freshwater mussels from North Carolina waters.

Population’
Locus 1 2 3 4 5 6 7 8 9
PGI
A 0.000 0.125 0.000 0.294 0.000 0.000 0.086 0.000 0.000
B 0.000 0.833 0.929 0.706 0.000 0.000 0.000 0.000 0.000
C 0.000 0.000 0.000 0.000 0.000 0.000 0.729 0.935 1.000
D 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
E 0.067 0.042 0.071 0.000 0.316 0.718 0.186 0.065 0.000
F 0.767 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
G 0.078 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
H 0.089 0.000 0.000 0.000 0.684 0.282 0.000 0.000 0.000
(N) 45 12 7 17 19 55 35 23 9
H (dc) 0.133 0.333 0.143 0.235 0.526 0.309 0.314 0.130 0
PGM-1
A 0.011 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
B 0.744 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
C 0.044 0.167 0.214 0.147 0.000 0.000 0.000 0.000 0.000
D 0.189 0.625 0.571 0.559 0.974 1.000 0.814 1.000 1.000
E 0.011 0.167 0.000 0.294 0.026 0.000 0.171 0.000 0.000
F 0.000 0.042 0.214 0.000 0.000 0.000 0.014 0.000 0.000
(N) 45 12 7 1% 19 55 35 23 9
H (dc) 0.356 0.417 0.286 0.471 0.053 0 0.314 0 0
PGM-2
A 0.000 0.042 0.000 0.029 0.000 0.000 0.000 0.043 0.000
B 0.044 0.000 0.000 0.000 1.000 0.964 0.000 0.022 0.000
C 0.944 0.792 0.857 0.706 0.000 0.027 1.000 0.935 1.000
D 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
E 0.011 0.167 0.071 0.265 0.000 0.000 0.000 0.000 0.000
F 0.000 0.000 0.071 0.000 0.000 0.009 0.000 0.000 0.000
(N) 45 12 7 17 19 55 35 23 9
H (dc) 0.067 0.250 0.286 0.471 0 0.073 0 0.130 0
ALA-1
A 0.000 0.000 0.000 0.000 0.053 0.055 0.000 0.000 0.000
B 0.011 0.792 0.714 0.676 0.947 0.936 0.229 0.457 0.722
C 0.067 0.208 0.214 0.324 0.000 0.000 0.029 0.043 0.000
D 0.744 0.000 0.071 0.000 0.000 0.009 0.714 0.500 0.278
E 0.122 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
F 0.056 0.000 0.000 0.000 0.000 0.000 0.029 0.000 0.000
(N) 45 12 7 17 19 55 35 23 9
H (dc) 0.244 0.417 0.429 0.529 0 0.018 0.286 0.391 0.333
ALA-2
A 0.000 0.000 0.000 0.059 0.000 0.036 0.000 0.348 0.000
B 0.274 0.625 0.643 0.647 1.000 0.964 1.000 0.652 1.000
C 0.726 0.375 0.357 0.294 0.000 0.000 0.000 0.000 0.000
(N) 42 12 7 17 19 55 35 23 9
Н (dc) 0.244 0.583 0.714 0.294 0 0 0 0 0
[АР
А 0.022 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
В 0.678 0.000 0.000 0.000 0.079 0.000 0.000 0.000 0.000
С 0.011 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
D 0.133 0.000 0.000 0.250 0.000 0.000 0.000 0.000 0.000
E 0.011 0.000 0.071 0.313 0.000 0.000 0.000 0.000 0.000
F 0.000 0.375 0.571 0.250 0.000 0.000 0.000 0.109 0.000
G 0.011 0.292 0.214 0.125 0.000 0.000 1.000 0.761 1.000
H 0.067 0.083 0.071 0.063 0.000 0.000 0.000 0.022 0.000
| 0.000 0.000 0.071 0.000 0.000 0.000 0.000 0.000 0.000
J 0.067 0.250 0.000 0.000 0.921 1.000 0.000 0.109 0.000
(N) 45 12 7 16 19 54 1 23 9
H (dc) 0.333 0.667 0.857 0.375 0.158 0 0 0.087 0

LAMPSILIS IN NORTH CAROLINA 361

TABLE 3. (Continued)

Locus 1 2 3 4
MDH

A 0.000 0.000 0.000 0.000
B 0.012 0.167 0.000 0.000
C 0.628 0.833 1.000 1.000
D 0.360 0.000 0.000 0.000
(N) 43 12 7 112

H (dc) 0.023 0 0 0
6-PGD

A 0.000 0.042 0.000 0.059
B 0.000 0.000 0.143 0.000
C 0.000 0.917 0.786 0.941
D 0.000 0.000 0.000 0.000
E 0.326 0.042 0.071 0.000
F 0.674 0.000 0.000 0.000
(N) 43 12 7 17

H (dc) 0.372 0.167 0.143 0.118
MPI

A 0.023 0.000 0.000 0.000
B 0.682 0.000 0.000 0.000
C 0.295 0.792 0.000 0.000
D 0.000 0.000 0.000 0.324
E 0.000 0.208 1.000 0.676
(N) 44 12 7 Uz

H (dc) 0.068 0.250 0 0.647
a-PGD

A 0.000 1.000 1.000 0.765
B 1.000 0.000 0.000 0.235
(N) 45 12 7 17

H (dc) 0 0 0 0.353
SOD

A 0.000 0.000 0.000 0.000
B 1.000 1.000 1.000 1.000
(N) 45 12 7 1174

H (dc) 0 0 0 0

B 81.8 81.8 63.6 81.8

H + 0.169 0.280 0.260 0.318
SIE 0.044 0.070 0.090 0.064

Population'

5 6 Y 8 9
0.158 0.009 0.000 0.000 0.000
0.842 0.991 0.214 0.000 0.167
0.000 0.000 0.000 0.000 0.000
0.000 0.000 0.786 1.000 0.833

19 55 35 23 9
0.316 0.018 0.429 0 0.333
0.000 0.000 0.000 0.000 0.000
0.000 0.009 0.000 0.000 0.000
0.000 0.000 0.686 0.413 0.500
1.000 0.973 0.000 0.000 0.000
0.000 0.018 0.314 0.587 0.500
0.000 0.000 0.000 0.000 0.000

19 55 35 23 9
0 0.018 0.514 0.391 0.556
0.000 0.000 0.000 0.000 0.000
0.000 0.018 0.000 0.000 0.056
0.974 0.900 0.029 0.348 0.389
0.000 0.000 0.000 0.022 0.000
0.026 0.082 0.971 0.630 0.556

19 55 35 23 9
0.053 0.127 0 0.304 0.556
1.000 1.000 1.000 0.870 1.000
0.000 0.000 0.000 0.130 0.000

19 55 35 23 9
0 0 0 0.087 0
1.000 1.000 0.000 0.000 0.000
0.000 0.000 1.000 1.000 1.000

19 55 35 23 9
0 0 0 0 0

36.4 27.3 45.5 72.7 36.4
0.100 0.051 0.169 0.138 0.162
0.052 0.028 0.061 0.046 0.071

11 -Elliptio complanata, S. Flat River, Durham Co.
2-Lampsilis cariosa, Cape Fear River, Cumberland Co.
3-Lampsilis cariosa, Tar River, Nash Co.

4-Lampsilis cariosa, Deep River, Moore Co.
5-Leptodea ochracea, Chowan River, Gates Co.
6-Leptodea ochracea, Lake Waccamaw, Columbus Co.
7-Lampsilis fullerkati, Lake Waccamaw, Columbus Co.
8-Lampsilis radiata conspicua, Flat River, Durham Co.
9-Lampsilis radiata radiata, Chowan River, Gates Co.

Lampsilis cariosa. Lower heterozygosity esti-
mates are associated with the two rare Lamp-
silis radiata “subspecies,” Lampsilis fullerkati
and Leptodea ochracea, which exhibited a
very low heterozygosity level.

Elliptio complanata from the South Flat
River showed the largest number of polymor-
phic loci deviating from Hardy-Weinberg ex-
pectation (six of nine) (0.95 criterion for poly-

morphic loci), whereas Leptodea ochracea
from Lake Waccamaw had three of four.
Lampsilis r. conspicua had three of eight, and
L. r. radiata none. All significant departures
came from deficiencies in heterozygotes. For
example, in Е. complanata, “О” ranged from
0.36 to —0.95 among loci, Leptodea ochra-
cea from —0.66 to —1.0, and in L. г. conspicua
from 0:21:t0 150:

362

STIVEN & ALDERMAN

TABLE 4. Matrix of Nei (1978) unbiased genetic identity (below diagonal) and unbiased genetic distance
(above diagonal) for the nine populations/species of freshwater mussels from North Carolina waters.

Population/Species 1 2

1. Elliptio complanata 0.853
2. Lampsilis cariosa-Cape Fear (а о
3. Lampsilis сапоза-Таг В. 0.406 0.921
4. Lampsilis cariosa-Deep В. 0.438 0.914
5. Leptodea ochracea-Chowan В. 0.118 0.484
6. Leptodea ochracea-L. Wacc. 0.113 0.483
7. Lampsilis fullerkati 0.410 0.657
8. L. radiata conspicua 0.445 0.661
9. L. radiata radiata 0.396 0.708

3

0.902
0.082
0.959
0.329
0.338
0.704
0.655
0.680

4

0.825
0.090

5

2.138
0.726
ПА,

6

2.179
0.729

7%

0.892
0.420
0.350
0.404
1.049
1.004
0.945
0.952

9

0.927
0.345
0.386
0.435
0.823
0.819
0.049
0.022

E22 2 2 2

Genetic Relatedness Among and Within the
Populations: Genetic relatedness among the
nine populations was explored through pair-
wise associations using Neïs (1978) unbi-
ased genetic distance and identity measures
(non metric) (Table 4) and Rogers modified
distance metric measure (Rogers, 1972;
Wright, 1978) (Table 5). Some of the greatest
distances were found between species of the
two genera, Elliptio and Lampsilis, confirming
prior taxonomic designations based largely
upon morphological work. Notably, some of
the smallest distance values occurred be-
tween L. r. conspicua and L. r. radiata, the two
rare subspecies. In addition, these rare sub-
species showed corresponding small dis-
tance values with another species, L. fuller-
kati, a mussel first collected from Lake
Waccamaw. Kat (1983) provided morpholog-
ically and electrophoretic descriptions of this
population, and В. |. Johnson (1984) judged it
to be sufficiently distinct from L. r. radiata to
warrant separate species status. Rogers
modified distance (Table 5), which produces
less compaction at the extremes, yielded cor-
responding results, with L. fullerkati being al-
most as genetically close to L. r. radiata as L.
r. conspicua is to L. r. radiata.

Two species, Lampsilis cariosa and Lep-
todea ochracea, were sampled from different
river systems (Fig. 1, Appendix 1) thus provid-
ing measures of intraspecific genetic similar-
ity among sites. Nei’s (1978) unbiased ge-
netic identity among the three poulations of
Lampsilis cariosa from the Cape Fear, Tar,
and Deep rivers ranged from 0.914—0.959.
Heterozygosity values for these populations
(Table 3) were also comparable (0.26—0.32).
The populations of Leptodea ochracea from
the Chowan River and Lake Waccamaw (Fig.
1, Appendix 1) were genetically identical
(Nei’s identity was 0.982; Rogers metric dis-

tance was 0.134). However, the Lake Wacca-
maw population exhibited a low heterozygos-
ity value (0.051, almost half that of the
population from the Chowan River), and
these within-species genetic distance com-
parisons essentially correspond to the levels
found for the L. r. conspicua-L. r. radiata-L.
fullerkati association (Tables 4, 5).

Evolutionary relationships among the nine
populations were explored by performing a hi-
erarchical cluster analysis using the un-
weighed pair-group method with arithmetic
averaging (Sneath & Sokal, 1973). The result-
ing phenograms, one based upon Nei’s unbi-
ased genetic distance (Nei, 1978) and the
other based upon unmodified Rogers genetic
distance (Rogers, 1972; Wright, 1978) are
given in Figures 3 and 4 respectively. Both
confirm the close genetic similarity between
the two rare “subspecies” L. r. conspicua and
L. r. radiata, as well as their very close simi-
larity to the Lake Waccamaw L. fullerkati.
Lampsilis cariosa lies next in genetic related-
ness to this group of three “species/subspe-
cies,” with Leptodea ochracea and Elliptio
complanata exhibiting significant genetic dis-
tance not only from each other but also from
the Lampsilis group. The phenogram based
upon Nei’s unbiased distance tended to com-
pact values at the lower end of the scale (Fig.
3), whereas the phenogram derived from
modified Rogers distance, a metric scale, pro-
vided a less compacted view of the genetic
relationships among the Lampsilis radiata
subspecies and L. fullerkati (Fig. 4). Even
viewed in this manner, the relationships are
still exceedingly close.

A possible evolutionary or phylogenetic
pathway among the nine populations is given
by the Wagner tree (Fig. 5) (Farris, 1970,
1972). It is based upon Rogers metric modi-
fied distance measure. Construction assump-

LAMPSILIS IN NORTH CAROLINA 363

TABLE 5. Matrix of Rogers modified distance for the nine populations/species of freshwater mussels
from North Carolina waters.

Population/Species 1 2 3 4 5 6 7 8 9

1. Elliptio complanata RN

2. Lampsilis cariosa-Cape Fear 0/6270

3. Lampsilis cariosa-Tar R. 0.652 0.269 *****

4. Lampsilis cariosa-Deep R. 0.609 0.262 0.211 *****

5. Leptodea ochracea-Chowan R. 0.835 0.647 0.747 0.729 *****

6. Leptodea ochracea-L. Wacc. 0.840 0.650 0.745 0.731 0.134 *****

7. Lampsilis fullerkati 0.665 0.516 0.490 0.500 0.749 0.742 *****

8. L. radiata conspicua 0.629 0.499 0.514 0.518 0.716 0.714 0.219 *****

9. L. radiata radiata 0.680 0.484 0.515 0.522 0.702 0.703 0.213 0.162 *****
DISTANCE

1.20 1.00 .80 .60 .40 .20 .00

+ ---- 4-4 4-44 de e e dd

Elliptio complanata-S. Flat R.
Lampsilis cariosa-Cape Fear R.
Lampsilis cariosa-Tar R.

Lampsilis cariosa-Deep R.
Lampsilis fullerkati-L. Waccamaw
Lampsilis radiata conspicua-Flat R.
Lampsilis radiata radiata-Chowan
Leptodea ochracea-Chowan R.

/ aptodea ochracea-L. Waccamaw

+ ---- 4 4-4 E dd

1.20 1.00 .80 .60 .40

.20 .00

FIG. 3. Phenogram of nine North Carolina freshwater mussel populations based upon Nei (1978) unbiased
genetic distance and using the unweighted pair group method of construction. Farris (1972) “f”-value =

4.672.

tions are discussed in Felsenstein (1983) and
Swofford & Olsen (1990). The method of root-
ing is based upon midpoint location (Farris
1972, 1981). This method usually leads to
trees having their roots on the branch leading
to outgroup species (Felsenstein, 1983). This
tree placed the two rare Lampsilis subspecies
and L. fullerkati at approximately the same
distance from the root, and into the same
common evolutionary assemblage. Lampsilis
cariosa lies next in genetic relatedness to the
three aforementioned species-subspecies.
Leptodea ochracea and Elliptio complanata,

as expected, exhibited significant genetic dis-
tances not only from each other but also from
the Lampsilis group. Populations within both
Lampsilis cariosa and Leptodea ochracea
have relative distances from the root that dif-
fer little from that among the L. radiata sub-
species-L. fullerkati complex.

With respect to specific loci, both L. radiata
“subspecies” also share the same major
(most common) allele for all loci, and do so in
approximately the same frequency (Table 3).
In six of the loci, slight differences among the
two subspecies do exist, ranging from certain

364 STIVEN 8 ALDERMAN
DISTANCE

.80 .67 53 .40 27 13 .00

+ ---- 4 4-4 4444444
Elliptio complanata-S. Flat R.
Lampsilis cariosa-Cape Fear R.
Lampsilis cariosa-Tar R.
Lampsilis cariosa-Deep R.
Lampsilis fullerkati-L. Waccamaw
Lampsilis radiata conspicua-Flat R.
Lampsilis radiata radiata-Chowan
Leptodea ochracea-Chowan R.

Leptodea ochracea-L. Waccamaw

+ ---- 4-4 44 Hd dd dd nd
.80 .67 .53 .40 27 .13 .00

FIG. 4. Phenogram of nine North Carolina freshwater mussel populations based upon modified Rogers
genetic distance (Rogers, 1972; Wright, 1978) and using the unweighted pair group method of construction.
Farris (1972) “f”-value = 0.780.

DISTANCE FROM ROOT

.00 .07 .14 .21 .28 .35 .42
4-4 4444444 dd

Elliptio complanata-S. Flat R.

Lampsilis cariosa-Cape Fear R.

Lampsilis cariosa-Tar R.

Lampsilis cariosa-Deep R.

Lampsilis fullerkati-L. Waccamaw

Lampsilis radiata conspicua-Flat R.

Lampsilis radiata radiata-Chowan R.

Leptodea ochracea-Chowan В.
Leptodea ochracea-L. Waccamaw

SE ie ee a ee ee CESSE EE Zu
.00 .07 .14 .21 .28 .35 .42

FIG. 5. Wagner tree depicting phylogenetic relationships of nine North Carolina freshwater mussel species
and populations. Coefficient used was modified Rogers distance (Rogers, 1972; Wright, 1978). Tree is
rooted at midpoint of longest path. Farris (1972) “f”-value is 0.893.

LAMPSILIS IN NORTH CAROLINA 365

loci being polymorphic in one but not the other
(Pgi, Pgm-1, Pgm-2, a-Pgdh) to one locus
showing an allele in one subspecies that is
not present in the other (Mpi).

DISCUSSION AND CONCLUSIONS

Genetic Relationships Among Lampsilis,
Leptodea and Elliptio spp.

Freshwater mussels are one of the most
diverse and abundant groups of macroinver-
tebrates in the United States, but also one of
the most notorious for lack of systematic
agreement among investigators (Davis &
Fuller, 1981). Because conchological differ-
ences among bivalves are so environmentally
dependent, the use of shell morphology in
separating groups, especially above the spe-
cies level, can often be misleading (Davis 8
Fuller, 1981). Molecular genetic methodolo-
gies have grown in importance in helping sort
out differences among the Unionacea (Davis et
al., 1981; Davis & Fuller, 1981; Kat, 1983; Kat
& Davis, 1984). However, opinions differ over
the association between genetic distance,
level of systematic divergence and specia-
tion, as well as the broader question of the
applicability of molecular genetic tools (G.B.
Johnson, 1977) and their use in phylogenetic
interpretations (Felsenstein, 1988; Swofford
& Olsen, 1990). Avise (1976) suggested in
general terms that when using Nei’s (1972,
1978) genetic identity (I), values of 0.9-1.0
would be expected among populations within
a species, 0.8 to 0.89 among subspecies
within a species, and a value < 0.8 among
different species. Thorpe (1982), in a litera-
ture review, found that 97% of Nei I-values
used for erecting new species were below
0.85, and 98% of the I-values associated with
within species comparisons exceeded 0.85.
In freshwater mussels, however, systematic
interpretations based upon genetic distances
vary, often as a function of allopatry-sympa-
try. Kat (1983) argued that a value of 0.878
found among Lampsilis spp. and L. radiata is
characteristic of distinct species among radi-
ating clades. This is consistent with Davis et
al. (1981), who reported values > 0.9 among
apparent sympatric species of Elliptio in Flor-
ida and Lake Waccamaw in North Carolina.
The two “sympatric” Elliptio species in Lake
Waccamaw had different shell characters
states, differed in allelic frequencies at three
loci, and preferred different habitats. Habitat

differences were not apparent in the two Flor-
ida sympatric Elliptio species. On the other
hand, Kat & Davis (1984) interpretated a
value of 0.954 between two allopatric popu-
lations of Elliptio complanata in adjacent
drainages in Nova Scotia as consistent with
the Avise (1976) scheme. Thus, Davis et al.
(1981) argue that in established conchologi-
cally distinct sympatric populations, genetic
identity values of >0.9 could indicate different
species, yet recognize that in allopatric situa-
tions, genetic identity values for different spe-
cies of the same genus would more likely be
less than 0.9.

Our intraspecific work with Lampsilis cari-
osa and Leptodea ochracea populations pro-
duced unbiased genetic identities (Nei, 1978)
among separated sites of 0.922-0.959 and
0.982 for these species, respectively. These
values are consistent with those suggested
by Avise (1976), Davis et al. (1981), and
Thorpe (1982) to encompass the expected
variation and range within allopatric popula-
tions of a single species. Both phenograms
(Figs. 3, 4) illustrate the identical genetic na-
ture of these two species groups. Thus, the
large-shelled L. cariosa type sampled from
the Deep River is probably L. cariosa. It is
genetically indistinct from the other two pop-
ulations and also shows no signs of pro-
longed bottleneck effects (e.g. diminished
heterozygosity).

Among paired associations within the four
species of Lampsilis, we found identity values
ranging from 0.357 to 0.863, consistent with
Avise’s (1976) expectations for variation
among different species, as well as values
found by Kat (1983) for three distinct species
of Lampsilis. For example, we found identity
values of 0.382 and 0.439 between Chowan
River Leptodea ochracea and L. r. conspicua
and L. r. radiata respectively, whereas Kat's
(1983) value for a comparable association
was 0.31. Genetic identity measures between
Leptodea ochracea and Lampsilis fullerkati
ranged from 0.35 to 0.366, and Kat’s (1983)
corresponding value was about 0.49. We also
found values from 0.33 to 0.48 between Lep-
todea ochracea and L. cariosa, clearly imply-
ing distinct species status for each according
to guidelines in Avise (1976). Identity values
among Elliptio, Leptodea and Lampsilis gen-
era in our study ranged from 0.115 to 0.445
(Table 4), consistent with Avise (1976) and
also Davis et al. (1981). Again the two phe-
nograms (Figures 3, 4) clearly depict these
patterns.

366 STIVEN 8 ALDERMAN

Heterozygosity Levels

In examining relative heterozygosity levels
among the various populations, we found that
the two rare Lampsilis “subspecies” had H-
values that were about 40% lower than that of
populations of the more abundant and com-
mon Lampsilis cariosa. They were about the
same as that of the genetically identical
Lampsilis fullerkati from Lake Waccamaw and
the common Elliptio complanata. Thus, we
cannot associate these low heterozygosity
levels with a prolonged bottleneck effect (Nei
et al., 1975). Leptodea ochracea populations
exhibited even lower heterozygosity levels as
well as low polymorphism, a finding consis-
tent with Kat’s (1983), especially for Lake
Waccamaw. We note that reliability of genetic
identity comparisons among populations by
genetic distance measures when sample
sizes are small is greatly improved when het-
erozygosity levels are low (Nei, 1978), as in
the case of the rare L. radiata “subspecies.”

One possible explanation for the heterozy-
gote deficiencies may be that a population in
any one site is simply a mix of age classes
and genotypes derived from a variety of pop-
ulations some distance away that possess dif-
ferent gene frequencies. Dispersal of mussel
larvae may be substantial (river currents,
glochidia on fish hosts), and the Wahlund ef-
fect may be a partial explanation for heterozy-
gote deficiencies (Dillon, 1988).

The Eastern Lampmussel and the Carolina
Fatmucket: Distinct Subspecies?

The two rare L. radiata populations (subspe-
cies) are currently recognized by the Ameri-
can Malacological Union as L. r. conspicua
(Carolina fatmucket) and L. r. radiata (eastern
lampmussel) (Turgeon et al., 1988). Turgeon
et al. (1988) and Alderman (personal obser-
vation) have reported that sampled L. r. con-
spicua shells were larger and heavier than L.
r. radiata, and that differences existed in their
current range and preferred habitat. Lamp-
silis r. radiata resides in the lower parts of
drainage basins of the eastern North Ameri-
can coastal plain, discontinuously from South
Carolina to the St. Lawrence, whereas the
very rare L. r. conspicua seems endemic to
North Carolina and is currently restricted to
the Pee Dee River drainage basin and the
upper Neuse River basin. Previous work on
these populations differentiated substrate af-
finities (Alderman, 1989), with the Carolina

fatmucket found more often in course sands,
gravel and cobble substrates, whereas the
eastern lampmussel occurred more often in
medium to coarse sands. Our own collecting
of these two groups was consistent with these
reported substrate preferences. Sample shell
size data from this study, even though based
on relatively small numbers, confirmed prior
evidence of a larger L. r. conspicua, that is,
about 1.4 times greater than L. r. radiata
(Table 2). There is also strong evidence for
“site effects” in the two species examined
(Lampsilis cariosa and Leptodea ochracea)
for this phenomenon, and this process cannot
be eliminated as a basis of the shell differ-
ences. In addition, covariance analysis of
shell regressions clearly indicated strong con-
cohological similarities between these two
“subspecies.”

Nei’s (1978) unbiased genetic identity be-
tween these two “subspecies” is 0.978 (and
an unbiased genetic distance of 0.022) (Table
4). Rogers metric modified distance (Rogers,
1972; Wright, 1978) was 0.162, hardly distin-
guishable from the value of 0.134 between
the two separated populations of Leptodea
ochracea, and even smaller than 0.211 to
0.229 among the allopatric populations of
Lampsilis cariosa. Clearly, Nei’s (1978) iden-
tity between the two “subspecies” is consis-
tent with Avise’s (1976) categorization of pop-
ulations within a species (0.9—1.0). It does not
approach the subspecies category (0.80—
0.89), let alone values differentiating species
(<0.8). It is also in the range of values we
found among the allopatric populations of
Lampsilis cariosa and Leptodea ochracea. In
addition, we note that the distinct Lake Wac-
camaw species, Lampsilis fullerkati, has Nei
(1978) unbiased identity values of 0.945 and
0.952 in association with L. r. conspicua and
North Carolina L. r. radiata respectively.
Those are above the 0.88 recorded for the
same species comparisons by Kat (1983), but
Kat's L. radiata specimens came from Cana-
da’s maritime provinces. Recall also that
Davis et al. (1981) argued that values >0.9
could be found among sympatric freshwater
mussel species that recently underwent spe-
ciation. However, these two so-called sub-
species of L. radiata are currently not sympat-
ric, and, in fact, are found for the most part in
different river basins with the exception of the
Neuse River (В. 1. Johnson, 1970). Lampsilis
r. radiata seems confined to the coastal plain
tidewater areas of the river systems and does
not appear to overlap with L. r. conspicua.

LAMPSILIS IN NORTH CAROLINA 367

One may also speculate the most pre-existing
L. radiata populations above the tidewater ar-
eas have simply been extirpated.

Thus, the genetic identity and distance
measures (Tables 4, 5), and the graphical il-
lustrations given by the phenograms and
Wagner tree (Figs. 3—5) point out the almost
identical genetic similarity between the two
“subspecies” of L. radiata, as well as the neg-
ligible genetic distance separating these
“subspecies” from the recently designated
species, L. fullerkati. Conchological data for
these three populations are also similar (co-
variance analysis) and suggest either the be-
ginnings of evolutionary divergence or simply
evidence of wide phenotypic plasticity coming
from different shell growth rates and patterns
in different sites. The genetic similarity of
these two L. radiata populations differ little
from that found among poulations of the same
species living in different sites (e.g. Leptodea
ochracea and Lampsilis cariosa). Clearly, re-
search on reciprocal transplant growth of
young individuals (preferably of both “sub-
species” and L. fullerkati) would be useful in
Clarifying this possible shell phenotypic plas-
ticity. Answers to questions about reproductive
isolation and potential hybridization among
these populations also would shed further
light on this genetically similar complex. Until
evidence to the contrary appears, we suggest
that L. r. conspicua and L. r. radiata and per-
haps L. fullerkati be considered simply as al-
lopatric “populations” of Lampsilis radiata.

We are aware that much of the work on the
application of molecular genetic techniques to
systematic issues has been in one direction,
that of defining and erecting new species and
subspecies. For example, Ayala (1975) insti-
tuted several subspecies of Drosophila willis-
toni based on allozyme techniques and Nei’s
(1972) genetic distance. Highton et al. (1989),
also utilizing allozymes, established 16 spe-
cies of Plethodon salamanders from the sin-
gle Plethodon glutinosis complex based on a
minimum Nei distance of 0.15 or an Nei ge-
netic identity (1) of 0.85. Such taxonomic split-
ting is quite common in the systematic litera-
ture, including that of freshwater mussels.
Our suggestion, therefore, of regrouping the
two and possibly three previously recognized
allopatric subspecies/species into one spe-
cies complex, based upon very high levels of
genetic identity as well as similar concholo-
gies, is probably an uncommon event, yet ap-
propriate until more distinctive biological spe-
cies properties become evident.

ACKNOWLEDGMENTS

Special thanks go to Christopher McGrath
for his help with the field work. We are also
appreciative of the helpful comments of two
reviewers as well as the editor, George Davis.
The research was partly supported by a grant
from the Nongame Program, North Carolina
Wildlife Resources Commission.

LITERATURE CITED

ALDERMAN, J. M., 1988, Tar River spiny mussel
annual performance report. Annual Performance
Report for the N. C. Wildlife Resources Commis-
sion-Wildlife Management, October 1987-June
1988, 41: 192-218.

ALDERMAN, J. M., 1989, Lampsilis radiata (Gmelin,
1791). Unpublished report to the Nongame Pro-
gram, North Carolina Wildlife Resources Com-
mission, Raleigh, North Carolina. 6 pp.

ALDERMAN, J. M., 1990, Lampsilis cariosa (Say
1817). Pp. 97-100, in Report on the Conserva-
tion Status of North Carolina's Freshwater and
Terrestrial Mollusca Fauna, W. F. ADAMS, ed.,
North Carolina Wildlife Resources Commission.
Mimeo, 245 pp.

AVISE, J. C., 1976, Genetic differentiation during
speciation. Pp. 106-122, Molecular evolution,
F.J. AYALA, ed. Sinauer. Sunderland, Massachu-
setts.

AYALA, F. F., 1975, Genetic differentiation during
the speciation process. Evolutionary Biology 15:
151-185.

BURCH, J. B., 1975, Freshwater unionacean clams
(Mollusca: Pelecypoda) of North America, re-
vised ed. Michigan: Malacological Publications.
204 pp.

CLARKE, A. H., 1983, Status survey of the Tar
River spiny mussel. Final Project Report for the
U. S. Fish and Wildlife Service. Ashville, North
Carolina, mimeo.

DAVIS, G. M., 1984, Genetic relationships among
some North American Unionidae (Bivalvia): sib-
ling species, convergence, and cladistic relation-
ships. Malacologia, 25: 629-648.

DAVIS, G. M. £ S. L. FULLER, 1981, Genetic re-
lationships among recent Unionacea (Bivalvia) of
North America. Malacologia, 20: 217-253.

DAVIS, G. M., W. H. HEARD, S. L. FULLER 4 C.
HESTERMAN, 1981, Molecular genetics and
speciation in Elliptio and its relationships to other
taxa of North American Unionidae (Bivalvia). Bio-
logical Journal of the Linnean Society, 15: 131-
150.

DILLON, R. T., 1988, The influence of minor human
disturbance on biochemical variation in a popu-
lation of freshwater snails. Biological Conserva-
tion, 43: 137-144.

FARRIS, J. S., 1970, Methods for computing Wag-
ner trees. Systematic Zoology, 19: 83-92.

368 STIVEN 8 ALDERMAN

FARRIS, J. S., 1972, Estimating phylogenetic trees
from distance matrices. American Naturalist,
106: 645-668.

FARRIS, J. S., 1981, Distance data in phylogenetic
analysis. In: V. A. Funk & D. R. Brooks, eds.,
Advances in cladistics: Proc. First Meeting of the
Willie Hennig Soc., N.Y. Botanical Garden,
Bronx, New York.

FELSENSTEIN, J., 1983, Parsimony in systematics:
biological and statistical issues. Annual Review
of Ecology and Systematics, 14: 313-333.

FELSENSTEIN, J., 1988, Phylogenies from molec-
ular sequences: inferences and reliability. Annals
of Research in Genetics, 22: 521-565.

GOLDBERG, E. D. plus nine other authors, 1978,
The mussel watch. Environmental Conservation,
5: 101-105.

HARTL, D. L. & A. G. CLARK, 1989, Principles of
population genetics, 2nd ed. Sinauer. Sunder-
land, Massachusetts. 682 pp.

HIGHTON, R., G. A. MAHA & L. R. MAXSON,
1989, Biochemical evolution in the slimy sala-
manders of the Plethodon glutinosis complex in
the eastern United States. Illinois Biological
Monographs, 57: 1-153.

HINCH, $. G., В. С. BAILEY 8 В. H. GREEN, 1986,
Growth of Lampsilis radiata (Bivalvia: Unionidae)
in sand and mud: a reciprocal transplant experi-
ment. Canadian Journal of Fisheries Science,
43: 548-552.

HORN, К. J. 8 H. J. PORTER, 1981, Correlation of
shell shape in Elliptio waccamawensis, Leptodea
ochracea, and Lampsilis sp. (Bivalvia, Union-
idae) with environmental factors in Lake Wacca-
maw, Columbus County, North Carolina. Bulletin
of the American Malacological Union, for 1981:
1-4.

JOHNSON, G. B., 1977, Assessing electrophoretic
similarity: the problem of hidden heterogeneity.
Annual Review of Ecology and Systematics, 8:
309-328.

JOHNSON, В. I., 1970, The systematics and zoo-
geography of the Unionidae (Mollusca: Bivalvia)
of the Southern Atlantic Slope region. Bulletin of
the Museum Comparative Zoology, 140: 263—
449.

JOHNSON, В. 1., 1984, A new mussel, Lampsilis
(Lampsilis) fullerkati (Bivalvia: Unionidae) from
Lake Waccamaw, Columbus County, North
Carolina, with a list of the other unionid species of
the Waccamaw River system. Occasional Pa-
pers on Mollusks, Museum Comparative Zool-
ogy, Harvard University, 4: 305-320.

KAT, P. W., 1982, Effects of population density and
substratum type on growth and migration of El-
liptio complanatum (Bivalvia: Unionidae). Mala-
cological Review, 15: 119-127.

KAT, P. W., 1983, Morphological divergence, ge-
netics, and speciation among Lampsilis (Bil-
valvia: Unionidae). Journal of Molluscan Studies,
49: 133-145.

KAT, P. W. 8 G. M. DAVIS, 1984, Molecular genet-
ics of peripheral populations of Nova Scotia

Unionidae (Mollusca: Bivalvia). Biological Jour-
nal Linnean Society, 22: 157-185.

KEFERL, E. P. 8 R. M. SHELLEY, 1988, The final
report on a status survey of the Carolina heel-
splitter, Lasmigona decorata and the Carolina
elktoe, Alasmidonta robusta. Prepared for the
U.S. Fish and Wildlife Service and the N.C.
State Museum of Natural History.

KOEHN, R. K., F. J. TURANO & J. B. MITTON,
1973, Population genetics of marine pelecypods.
Il. Genetic differences in microhabitats of Modio-
lus demissus. Evolution, 27: 100-105.

LEVENE, H., 1949, On a matching problem arising
in genetics. Annals of Mathematical Statistics,
20: 91-94.

NEI, M., 1972, Genetic distance between popula-
tions. American Naturalist, 106: 283-292.

NEI, M., 1978, Estimation of average heterozygos-
ity and genetic distance from a small number of
individuals. Genetics, 89: 583-590.

NEI, M., T. MARUYAMA 8 В. CHAKROBORTY,
1975, The bottleneck effect and genetic variabil-
ity in populations. Evolution, 29: 1-10.

ROGERS, J. S., 1972, Measures of genetic simi-
larity and genetic distance. Studies in Genetics,
University of Texas Publications, 7213: 145-153.

SELANDER, R. K., M. H. SMITH, Y. YANG, W. E.
JOHNSON 4 J. B. GENTRY, 1971, Biochemical
polymorphism and systematics in the genus Per-
omyscus. |. Variation in the field mouse (Peromy-
scus polionotus). Studies in Genetics VI, Univer-
sity of Texas Publications, 7103: 49-90.

SHAW, C. R. 4 R. PRASAD, 1970, Starch gel elec-
trophoresis of enzymes-A compilation of recipes.
Biochemical Genetics, 4: 297-320.

SHELLEY, R. M., 1987, Unionid mollusks from the
upper Cape Fear River basin, North Carolina,
with a comparison of the fauna of the Neuse, Tar,
and Cape Fear drainages (Bivalvia: Unionacea).
Brimleyana, 13: 67-89.

SNEATH, P. H. 8 R. R. SOKAL, 1973, Numerical
taxonomy: the principles and practice of numer-
ical classification. W. H. Freeman, San Fran-
cisco. 573 pp.

STIVEN, A. E., 1989, Population biology of two
southern Appalachian land snails (Mesomphix
spp.): variation among six sites with differing dis-
turbance histories. Oecologica, 79: 372-382.

SWOFFORD, D. L. & G. J. OLSEN, 1990, Phylog-
eny reconstruction. Pp: 411-501, D. М. HırLıs &
C. Moritz, eds., Molecular systematics. Sin-
auer, Sunderland, Massachusetts.

SWOFFORD, D. L. & R. B. SELANDER, 1981,
Bioysis-1: a Fortran program for the comprehen-
sive analysis of electrophoretic data in population
genetics and systematics. Journal of Heredity,
72: 281-283.

THORPE, J. P., 1982, The molecular clock hypoth-
esis: biochemical evolution, genetic differentia-
tion and systematics. Annual Review of Ecology
and Systematics, 13: 139—168.

TURGEON, D. D., plus nine other authors, 1988,
Common and scientific names of invertebrates

from the United States and Canada: Mollusca.
American Fisheries Society Special Publication

16: 277 pp.

LAMPSILIS IN NORTH CAROLINA 369

ulations. Vol. 4. Variability within and among nat-
ural populations. University of Chicago Press,

Chicago, Illinois.

WILKINSON, L., 1990, SYSTAT: The system for

statistics. SYSTAT, Inc., Evanston, Illinois.

WRIGHT, S., 1978, Evolution and genetics of pop-

Revised MS accepted 24 March 1992

APPENDIX |. Species and ANSP catalog numbers, number electrophoresed, water system, site code
(Fig. 1), and description of collecting sites for populations of North Carolina freshwater mussels used in
this study. The rate L. r. radiata and L. r. conspicua were returned alive to collecting site.

Species ANSP
Lampsilis radiata
radiata’
Lampsilis radiata
conspicua'
Elliptio 392038
complanata 392039
Leptodea 392040
ochracea? 392041
Lampsilis cariosa 392042
392043
392044

Lempsilis fullerkati 392044

‘Returned alive to collecting site, no ANSP number.

No.

9

23

44

19

55

11

7
17

35

County
Gates

Durham
Person
Gates
Columbus

Cumberland

Nash
Moore

Columbus

2Formerly called Lampsilis ochracea (see Turgeon et al., 1988).

Water System (Code) Location

Chowan River (C). Below HiWay 13/
158 bridge, Hertford-Gates county line
Flat River (B). HiWay 501, east on SR
1471 to bridge, 200 m below bridge

S. Flat River (A). HiWay 501, west on
SR 1123, south on SR 1125 to bridge,
200 m below bridge

Chowan River (C) (see above)

Lake Waccamaw (D). Collections just
off shore State Park, 2 m depth

Cape Fear River (F). HiWay 401, site at
confluence of Carvers Ck. and Cape
Fear R.

Tar River (E). Main channel from Spring
Hope to HiWay 581 bridge

Deep River (G). Via HiWay 22 west on
SR 1456 bridge, upstream 100 m.

Lake Waccamaw (C) (see above)

LH mms A 21
de sr teen eel,

rite 4% má IB s и. : y u Cra Ml

a, dé rá К
e rs u > 7 u
то eus A 4, pe Tal I.
= Y u =. of ea м Ба % AY Mi |
7 ve Pp = > eg > we es at NG

a
м

5 pro
A A A A ЧТ у RL
e AAA Enz e > a oo
> 4) ee ¡O Y AAA AMET Li |
z ME de HN yu o
o Al AE ft de | a
ee a
L
A u A
4 y
o ve
en
q
i o.

a a
= =
| . € D RE 27:
è
j в CS о _
'D ay
u er æ o = > Y DEZ
> i 2 =
u u u e y р > р 5 > E Ar
о Gase y Can A > jot AA
PSA 0 MNAE. sy pa
o >

MALACOLOGIA, 1992, 34(1-2): 371-376

INDEX

Taxa in bold are new; page numbers in
bold indicate pages on which new taxa are
indicate
systematic descriptions of previously named

described; underscored pages

taxa; pages in italics indicate figured taxa.

adrianae, Albinaria 37, 38, 40, 41, 47, 52,

adrianae, Albinaria adrianae 34-36, 39, 41,

46, 47-52, 56, 60
aetnea, Helix 121, 126
Agriolimax reticulatus 68, 71
iyoshia 143, 154, 155
uenoi 155
Akiyoshia (Saganoa) chinensis 148, 155-
156, 158-162, 166, 297; 157-160, -162-
165
kishiiana 155
odonta 150, 152
yunnanensis 155
Alasmidonta 355
Albinaria 33-61
adrianae 37, 38, 40, 41, 47, 52, 55-57
adrianae adrianae 34-36, 39, 41, 46,

47-52, 56, 60
adrianae dubia 34-36, 38-41, 47-52,
56, 60

contaminata 34-38, 40, 41, 47, 52-55,
57

contaminata contaminata 34-36, 38, 39,

41, 46-51, 53, 54, 56, 57, 60, 61

contaminata incommoda 34-36, 39, 41,
42, 47-51, 53, 54, 60, 61

contaminata liebetruti 34-36, 38, 39,
41, 42, 46-54, 56, 61

contaminata muraria 38

contaminata odysseus 34-36, 38, 39,
41, 46, 47-51, 53, 54, 57, 61

jonica 37, 38, 40, 41, 47, 52, 55, 57

jonica assicola 34-36, 39, 41, 46, 47-
51, 54-56, 61

jonica jonica 34-36, 38, 39, 41, 42, 46,
48-52, 55, 60

rebeli 34, 39, 40, 47-51, 61

senilis 34-38, 40, 41, 52, 54-55, 57

senilis flavescens 34-36, 39, 41, 47-51,

55, 56, 60, 61
senilis kolpomyrtensis 34-36, 38, 39,
41, 47-51, 54, 56, 61
senilis senilis 34-36, 38, 39, 41, 42,
46-51, 54-57, 60, 61
teres 39, 47, 50, 51
teres nordsiecki 34, 40, 48, 49, 57, 61
Aligena elevata 10
Amblema 134
Ammonoida 77
Amnicola 153, 154
Amnicolinae 153
aperta, Lithoglypopsis 185, 186, 333, 334
aperta, Neotricula 209, 218, 315-317, 320,
322, 325-328, 334

aperta, Tricula 209
apicina, Xerotricha 111, 120, 126
Aplacophora 77, 78, 80
тре 82
Апоп 4
ater 352
hortensis 68, 72, 87, 95
armillata, Helix 109
armillata, Microxeromagna 117, 123
Arthritica crassiformis 11
aspersa, Helix 68, 71, 72
assicola, Albinaria jonica 34-36, 39, 41, 46,
47-51, 54-56, 61
ater, Arion 352
bambooensis, Tricula 266, 286, 316-317,
320, 322, 325-328, 339-342
Bathymodiolus thermophilus 137
bathytera, Helix a 109
beauii, Cyclostremiscus 1, 15
Biomph aria glabrata 25- 32; 26, 28-31
Bithynia minutoides 259
Bivalvia 77, 78, 80
Blanfordia 153
bollingi, Tricula 286, 314, 315-317, 320,
322, 325-328, 339-342
burchi, Neotricula 209, 218, 316-318, 320,
322, 325-328, 339-342
burchi, Tricula 186
Bythinella 154
chinensis 154, 163, 166
gong en 163
ubeiensis 163
Ш 163
watanensis 163
wufenensis 163
Caenogastropoda 83, 84
Calyptogena 131, 132
magnifica 135
candicans, Helicella 111
Candidula 107, 109, 112, 126
intersecta 109
spadae 109
unifasciata 109
candidula, Glischrus (Helix) 109
cariosa, Lampsilis 356, 358, 360-367, 369
caruanae, Cernuella 110
carusoi, Helicotricha 109, 112-119, 123,
126; 113-118
Caucasigena 109-112, 123
cema, Ceratobornia 20, 21
Cepaea 95
nemoralis 68, 72, 87, 95
Cephalopoda 77, 78, 80
Ceratobornia cema 20, 21
longipes 20
Cerion 57
Cernuella 108, 109, 111, 112
caruanae 110
virgata 110
Certatobornia 20, 21

372 INDEX

chinensis, Akiyoshia (Saganoa) 148, 155-
156, 158-162, 166, 297; 157-160, 162-
165

chinensis, Bythinella 154, 163, 166

chinensis, Erhaia 166

chinensis, Gammatricula 218, 316-318,
320-322, 325-328, 339-342

chinensis, Pseudobythinella 148, 166, 169-
170, 173, 175; 157, 167, 168, 170-174

Chondrina clienta 46

cinctella, Helix 109

Circulus texanus 1,15

clienta, Chondrina 46

Cochlea virgata 109

Cohilopinae 186

complanata, Elliptio 57, 357-367, 369

conspicua, Lampsilis radiata 356, 358-367,
369

conspurcata, Helicella (Xerotricha) 112

conspurcata, Helix 109, 120

conspurcata, Xerotricha 107, 111, 120, 126

contaminata, Albinaria 34-38, 40, 41, 47,
52-55, 57

contaminata, Albinaria contaminata 34-36,
38, 39, 41, 46-51, 53, 54, 56, 57, 60,
61

corderoi, Helicella 120

cordiformis, Divariscintilla 11-12, 14, 15-
215 3, 4 13

crassiformis, Arthritica 11

Crassostrea 134

cristella, Hydrobia 211

cristella, Neotricula 150, 209, 211-214,
217-218, 220, 222, 246, 256, 315-317,
320, 322, 325-328, 330; 215, 216, 219,
221, 222-224

cristella, Tricula 211

cumulata, Nanaja 109

Cyclostremiscus beauii 1, 15

daliensis, Erhaia 154, 163

daliensis, Pseudobythinella 161, 166, 183

darevskii, en icopsis 109, 123

decollata, Rumina 45, 47

Delavaya 266, 279, 332-336

Dentalium 25

Devonia 19, 20

dianmenensis, Neotricula 150, 209, 218,
222, 227-231, 233-234, 246, 256, 315-
318, 320-322, 325-328, 330; 225-227,
229-233

Divariscintilla 2, 19-22
cordiformis 11-12, 14, 15-21; 3, 4, 13
luteocrinita 6, 9-11, 15-18, 20; 3, 8,

LE
maoria 1,10, 14-21
octotentacula 2, 4-6, 9, 12, 14-17, 19-
LES 4. 7. 1216

troglodytes 1, 15-21
yoyo 1, 5, 6, 15-22

dubia, Albinaria adrianae 34-36, 38-41, 47-
52, 56, 60

duplicata, Neotricula 150, 209, 234-238,
241, 243-249, 256, 297, 316-318, 320,
322, 325-328; 225, 234-236, 238-240,
242-246, 248-252

Edentiella 109-112, 123

edentula, Helix 109, 123
edmundi, Isabellaria 34, 38, 57, 61
edulis, Mytilus 99-106
eichwaldi, Helix 109, 123
elevata, Aligena 10
elevatus, Leptodrilus elevatus 129-141;
130, 131, 134, 138
Elliptio 355, 362, 365
complanata 57, 357-367, 369
Entovalva 19
perrieri 20
Eon 19
rhaia 162
chinensis 166
daliensis 154, 163
kunmingensis 154, 163
Erhaiini 154
Erycinidae 22
Erycininae 22
explanata, Helix 109
Fenouilia 266, 277, 279, 282, 284, 324,
331, 332-336
kreitneri 284
flavescens, Albinaria senilis 34-36, 39, 41,
47-51, 55, 56, 60, 61
fluviatilis, Nertina 25
fontinalis, Physa 25
fuchsianus, Guoia 146, 150, 190, 192, 193,
195, 204-205, 208-209, 211, 213; 188,
191, 197, 199, 206, 207, 210-212, 214
fuchsianus, Lithoglyphopsis 204
fuchsianus, Lithoglyphus 187, 204
fullerkati, Lampsilis 356, 358-367, 369
Galeomma polita 10
takii 10
Galeommatidae 1-24
galloprovincialis, Mytilus 99-106
Gammatricula 184, 320-322, 325-328, 330-
336
chinensis 218, 316-318, 320-322, 325-
328, 339-342
Gastropoda 77, 78, 80
gibba, Partula 47
gittenbergeri, Helicopsis 120
glabrata, Biomphalaria 25-32; 26, 28-31
Glischrus (Helix) candidula 109
gongjianguoi, Bythinella 163
gonzalei, Helicella 120
grandis, Lithoglyphopsis 282
gredleri, Neotricula 318
gredleri, Tricula 150, 152, 155, 246, 286,
288-293, 295-297, 315-318, 320-322,
325-328; 286-290, 292-296
gregoriana, Tricula 286, 297, 316-317, 319,
320, 322, 325-328, 339-342
Guoia 142, 150, 186-187, 199, 324, 330-
336
fuchsianus 146, 150, 190, 192, 193,
195, 204-205, 208-209, 211, 213;
188, 191, 197, 199, 206, 207, 210-
212, 214
viridulus 146, 150, 187, 190-196, 198-
201, 203, 208, 209; 188-191, 196,
197, 199, 201-208, 210, 272
Gyraulus 187, 235

Halewisia 184, 211, 324, 327, 330, 332-

336
Helicella 107, 109, 110, 112, 120, 126
candicans 111
corderoi 120
gonzalei 120
mangae 120
spiruloides 111
Helicella (Xerotricha) conspurcata 112
Helicopsis 107, 109-112, 120, 122; 121,
122
ittenbergeri 120
ikharevi 110
retowskii 110
striata 110, 117
subcalcarata neuberti 120
subcalcarata subcalcarata 120
Helicotricha 109-112; 123, 124, 126
а 109, 112-119, 123, 126; 113-
1
Нейх 349
аетеа 121, 126
armillata 109
азрегза 68, 71, 72
cinctella 109
conspurcata 109, 120
edentula 109, 123
eichwaldi 109, 123
explanata 109
hispida 109, 123
holotricha 109, 123
itala 109, 120
lubomirskii 109, 123
lucorum 63-73, 343-354; 63, 66, 69,
70, 345, 346, 348, 351
obvia 109
pomatia 68
rubens 109, 123
stolismena 109
striata 109, 120
turbinata 109
turcica 120
unifasciata 109
variabilis 109
Helix (Pseudoxerophila) bathytera 109
Heterostropha 83, 84
hispida, Helix 109, 123
holotricha, Helix 109, 123
hortensis, Arion 68, 72, 87, 95
hubeiensis, Bythinella 163
Hubendickia 146, 332-336

hudiequanensis, Tricula 218, 286, 316-317,

320, 322, 325-328, 339-342
hunanensis, 'Stenothyra 187, 271
hupensis, Oncomelania 148, 153, 154
Hydrobia 153, 324

cristella 211

minutoides 259
yo ka 153, 186, 324

ygrohelicopsis 109-112, 123

darevskii 109, 123
Hygromia 109, 112
Hygromiidae 107-128
Hygromiinae 126
Hyolitha 77, 78, 80

INDEX 373

incommoda, Albinaria contaminata 34-36,
39, 41, 42, 47-51, 53, 54, 60, 61

intersecta, Candidula 109

Isabellaria 39, 47-51, 57
edmundi 34, 38, 57, 61

Ischnochiton 25

itala, Helix 109, 120

jeffreysiana, Vasconiella 19, 20

jianouensis, Pseudobythinella 162, 163

jinhongensis, Jinhongia 316-318, 320, 322,
325-328, 339-34

Jinhongia 184, 320, 322, 324-328, 330-336

jinhongensis 316-318, 320, 322, 325-
328, 339-342
jonica, Albinaria 37, 38, 40, 41, 47, 52, 55,
57

jonica, Albinaria jonica 34-36, 38, 39, 41,
42, 46, 48-52, 55, 60
Jullieniini 324, 332
kishiiana, Akiyoshia (Saganoa) 155
Kokotschashvilia 109-112, 123
kolpomyrtensis, Albinaria senilis 34-36, 38,
39, 41, 47-51, 54, 56, 61
kreitneri, Fenouilia 284
kunmingensis, Erhaia 154, 163
kunmingensis, Pseudobythinella 155, 161,
166, 183
Kunmingia 334
Lacunopsis 266, 277, 282, 284, 324, 330,
332-336
yunnanensis 282
Lampsilis 355-369
cariosa 356, 358, 360-367, 369
fullerkati 356, 358-367, 369
radiata 356, 365-367
radiata conspicua 356, 358-367, 369
radiata radiata 356, 358-367, 369
Lasaea 20
Lasmigona 355
Leptodea 365
ochracea 356-358, 360-367, 369
Leptodrilus elevatus elevatus 129-141; 130,
131, 134, 138
pustulosus 139
Leucozonella 109-112, 123
liebetruti, Albinaria contaminata 34-36, 38,
39, 41, 42, 46-54, 56, 61
li, Bythinella 163
likharevi, Helicopsis 110
ШИ, Neotricula 150, 209, 218, 234, 246,
248, 252, 254, 256-259, 316-318, 320,
322, 325-328; 215, 225, 253, 255, 256,
258-261
liliputanus, Lithoglyphus AO 282; 272, 273
Limnaea stagnalis 25, 26, 3
Lithoglypheae 186
Lithog Yphinae 186
Lithoglyphopsis 143, 184-187, 266, 324,
331, 332-336
aperta 185, 186, 333, 334
fuchsianus 204
grandis 282
modesta 150, 185, 187, 196, 205, 266,
268, 271, 275- TT 279: 282, 284,
286; 270, 272-274, 277-286

374 INDEX

ovatus 282
viridulus 187, 205
Lithoglyphus 186
fuchsianus 187, 204
liliputanus 187, 282; 272, 273
modestus 268
tonkinianus 187
viridulus 186, 187
littorea, Littorina 68, 71
Littorina littorea 68, 71
obtusata 25
liui, Pseudobythinella 163
longipes, Ceratobornia 20
lubomirskii, Helix 109, 123
lucorum, Helix 63-73, 343-354; 63, 66, 69,
70, 345, 346, 348, 351
ludongbini, Tricula 286, 315-317, 320, 322,
325-328, 339-342
luteocrinita, Divariscintilla 6, 9-11, 15-18,
202378 11
Iysiosquillina, Phlyctaenachlamys 16, 18-21
maculata, Semerula 352
magnifica, Calyptogena 135
mangae, Helicella 120
maoria, Divariscintilla 1, 10, 14-21
maxidens, Tricula 150, 155, 286, 297,
300-302, 304-305, 316-318, 320, 322,
325-328; 298, 299, 301, 303-307
mediterranea, Umbrella 25
mercenaria, Venus 352
Mesogastropoda 82
Microxeromagna 107, 109, 111, 112, 124
armillata 117, 123
minima, Neotricula 218
minutoides, Bithynia 259
minutoides, Hydrobia 259
minutoides, Neotricula 150, 209, 259, 262-
264, 266-268, 316-317, 320-322, 325-
328; 215, 263, 265-269
minutoides, Tricula 259
modesta, Lithoglyphopsis 150, 185, 187,
196, 205, 266, 268, 271, 275-277, 279
282, 284, 286; 270, 272-274, 277-286
modestus, Lithoglyphus 268
Mollusca 75-86
Monoplacophora 77, 78, 80
montana, Tricula 286, 297, 316-317, 319,
320, 322, 325-328, 339-342
muraria, Albinaria contaminata 38
Mytilus edulis 99-106
galloprovincialis 99-106
Nanaja 109-112
cumulata 109
Nautilida 77
Nautilus 134
nemoralis, Cepaea 68, 72, 87, 95
Neogastropoda 82
Neotricula 146, 148, 184, 187, 209, 211,
217, 218, 230, 286, 315, 318, 320, 322-
328, 330, 332-336
aperta 209, 218, 315-317, 320, 322,
325-328, 334
burchi 209, 218, 316-318, 320, 322,
325-328, 339-342
cristella 150, 209, 211-214, 217-218,
220, 222, 246, 256, 315-317, 320,

322, 325-328, 330; 215, 216, 219,
221, 222-224
dianmenensis 150, 209, 218, 222,
227-231, 233-234, 246, 256, 315-
318, 320-322, 325-328, 330; 225-
227, 229-233
duplicata 150, 209, 234-238, 241, 243-
249, 256, 297, 316-318, 320, 322,
325-328; 225, 234-236, 238-240,
242-246, 248-252
redleri 318
ilii 150, 209, 218, 234, 246, 248, 252,
254, 256-259, 316-318, 320, 322,
325-328; 215, 225, 253, 255, 256,
258-261
minima 218
minutoides 150, 209, 259, 262-264
266-268, 316-317, 320-322, 325-328;
215, 263, 265-269
odonta 318
Nertina fluviatilis 25
neuberti, Helicopsis subcalcarata 120
niuzhuangensis, Wuconchona 316-318,
320-322, 325-328, 339-342
nordsiecki, Albinaria teres 34, 40, 48, 49,
57, 61
nubivaga, Xerotricha 120
obtusata, Littorina 25
obvia, Helix 109
ochracea, Leptodea 356-358, 360-367, 369
octotentacula, Divariscintilla 2, 46, 9, 12,
14-17, 19-22; 3, 4, 7, 12, 16
odonta, Akiyoshia (Sagonoa) 150, 152
odonta, Neotricula 318
odonta, Tricula 150, 286, 297, 301, 305,
307, 311-312, 314-315, 316-318, 320,
322, 325-328; 298, 308-310, 312, 313
odysseus, Albinaria contaminata 34-36, 38,
39, 41, 46, 47-51, 53, 54, 57, 61
Oncomelania 153-154
hupensis 148, 153, 154
Opisthobranchia 77, 79, 81, 82
ovatus, Lithoglyphopsis 282
Pachydrobia 183, 184, 324, 327, 330, 332-
336

Pachydrobiini 150, 183, 324, 328, 330, 334
Parabornia 22

squillina 20, 21
Parapyrgula 334
Partula 57

gibba 47
Pecten 19
perlatoris, Schileykiella 126
perrieri, Entovalva 20
Phlyctaenachlamys 20, 21

lysiosquillina 16, 18-21
Physa fontinalis 25
pisana, Theba 87-97, 343, 350
Planorbidae 271
Planorbis trivolvis 25
Plicuteria 109-112, 123
polita, Galeomma 10
Polyplacophora 77, 78, 80
pomatia, Helix 68
Pomatiopsidae 143-342
Pomatiopsinae 148, 153, 324

INDEX 375

Pomatiopsini 153, 324
Pomatiopsis 324
Prosobranchia 77, 79, 81, 82
Pseudobythinella 143, 154, 155, 161, 162-
163, 166, 301
chinensis 148, 166, 169-170, 173, 175;
157, 167, 168, 170-174
daliensis 161, 166, 183
jianouensis 162, 163
kunmingensis 155, 161, 166, 183
liui 16
shimenensis 150, 163, 173, 175, 178,
180, 182-183, 186; 157, 176-179,
181-185
Pseudobythinellini 148, 153, 154-155
Pseudopythina 20
Pseudoxerophila 109-112
Pulmonata 77, 79, 81, 82
pustulosus, Leptodrilus 139
radiata, Lampsilis 356, 365-367
men Lampsilis radiata 356, 358-367,
9
Radix 187, 271
rebeli, Albinaria 34, 39, 40, 47-51, 61
Rehderiellinae 186
reinae, Schileykiella 126
reticulatus, Agriolimax 68, 71
retifera, Rhamphidonta 21
retowskii, Helicopsis 110
Rhamphidonta retifera 21
Robertsiella 184, 187, 199, 324, 330-336
Rostroconchia 77, 78, 80
rubens, Helix 109, 123
Rumina decollata 45, 47
Saganoa 155
Scaphopoda 77, 78, 80
Schileykiella 126
perlatoris 126
reinae 126
Scintilla 1, 6
Scintillona 17, 20
Semerula maculata 352
Semisulcospira 187, 271
ou Albinaria 34-38, 40, 41, 52, 54-55,

senilis, Albinaria senilis 34-36, 38, 39, 41,
42, 46-51, 54-57, 60, 61

shimenensis, Pseudobythinella 150, 163,
173, 175, 178, 180, 182-183, 186; 157,
176-179, 181-185

Sibirobythinella 153

spadae, Candidula 109

spiruloides, Helicella 111

squillina, Parabornia 20, 21

stagnalis, Limnaea 25, 26, 32

Stenothyra 199
hunanensis 187, 271

stolismena, Helix 109

Striata, Helicopsis 110, 117

Striata, Helix 109, 120

subcalcarata, Helicopsis subcalcarata 120

takii, Galeomma 10

Tentaculitida 77

teres, Albinaria 39, 47, 50, 51

texanus, Circulus 1, 15

Theba pisana 87-97, 343, 350

thermophilus, Bathymodiolus 137
tonkinianus, Lithoglyphus 187
Trichia 109-112, 123
Trichiinae 126
Tricula 146, 148, 186, 196, 218, 266, 279,
286, 301, 315, 320, 322-328, 332-336
aperta 209
bambooensis 266, 286, 316-317, 320,
322, 325-328, 339-342
bollingi 286, 314, 315-317, 320, 322,
325-328, 339-342
burchi 186
cristella 211
gredieri 150, 152, 155, 246, 286, 288-
293, 295-297, 315-318, 320-322,
325-328; 286-290, 292-296
gregoriana 286, 297, 316-317, 319,
320, 322, 325-328, 339-342
hudiequanensis 218, 286, 316-317,
320, 322, 325-328, 339-342
ludongbini 286, 315-317, 320, 322,
325-328, 339-342
maxidens 150, 155, 286, 297, 300-
302, 304-305, 316-318, 320, 322,
325-328; 298, 299, 301, 303-307
minutoides 259
montana 286, 297, 316-317, 319, 320,
322, 325-328, 339-342
odonta 150, 286, 297, 301, 305, 307
311-312, 314-315, 316-318, 320,
322, 325-328; 298, 308-310, 312,
313
xiangfengensis 201, 286, 315-317, 320,
22, 325-328, 339-342
xiaolongmenensis 286, 301, 316-318,
320, 322, 325-328, 339-342
Triculidae 153
Triculinae 149, 150, 153, 183, 186, 218,
266, 324, 334-336, 339-342
Triculini 150, 266, 279, 324, 327, 330, 333
trivolvis, Planorbis 25
ie Divariscintilla 1, 15-21
turbinata, Helix 109
turcica, Helix 120
uenoi, Akiyoshia 155
Umbrella mediterranea 25
unifasciata, Candidula 109
unifasciata, Helix 109
Unionidae 355
variabilis, Helix 109
Vasconiella 19
jeffreysiana 19, 20
Venus mercenaria 352
virgata, Cernuella 110
virgata, Cochlea 109
viridulus, Guoia 146, 150, 187, 190-196,
198-201, 203, 208, 209; 188-191, 196,
197, 199, 201-208, 210, 272
viridulus, Lithoglyphopsis 187, 205
viridulus, Lithoglyphus 186, 187
Viviparidae 187, 271
watanensis, Bythinella 163
Wuconchona 183, 184, 320-322, 325-328,
330-336
niuzhuangensis 316-318, 320-322, 325-
328, 339-342

376 INDEX

wufenensis, Bythinella 163

Xerolenta 107, 109, 112, 126

Xeroleuca 120

Xeromunda 107-112, 126

Xerosecta 109, 111, 112, 124

Xerotricha 108, 109, 111, 112, 120
apicina 111, 120, 126
conspurcata 107, 111, 120, 126
nubivaga 120

xiangfengensis, Tricula 201, 286, 315-317,

20, 322, 325-328, 339-342

xiaolongmenensis, Tricula 286, 301, 316-
318, 320, 322, 325-328, 339-342

yoyo, Divariscintilla 1, 5, 6, 15-22

yunnanensis, Akiyoshia (Saganoa) 155

yunnanensis, Lacunopsis 282

MALACOLOGIA, VOL. 34

CONTENTS

PHILIPPE BOUCHET & JEAN-PIERRE ROCROI
Supraspecific Names of Molluscs: A Quantitative Review .................

ROBERT H. COWIE
Shell Pattern Polymorphism in a 13-Year Study of the Land Snail Theba
pisana (Müller) (Pulmonata: Helicidae) ....................................

GEORGE M. DAVIS, CUI-E CHEN, CHUN WU, TIE-FU KUANG, XIN-GUO XING,
LI LI, WEN-JIAN LIU 8 YU-LUN YAN
The Pomatiopsidae of Hunan, China (Gastropoda: Rissoacea) ............

V. K. DIMITRIADIS, D. HONDROS & A. PIRPASOPOULOU
Crop Epithelium of Normal Fed, Starved and Hibernated Snails Helix luco-
rum: A Fine Structural-Cytochemical Study ...............................

VASILIS K. DIMITRIADIS & DIMITRIS HONDROS
Effect of Starvation and Hibernation on the Fine Structural Morphology of
Digestive Gland Cells of the Snail Helix lucorum ..........................

J. P. A. GARDNER
Null Alleles and Heterozygote Deficiencies Among Mussels (Mytilus edulis
and M. galloprovincialis) of Two Sympatric Populations ...................

FOLCO GIUSTI, GIUSEPPE MANGANELLI 8 JORGE V. CRISCI
A New Problematical Hygromiidae from the Aeolian Islands (Italy) (Pulmo-
ME 2. 7.2002 a ee oon ES Ue a

STEPHEN HUNT
Structure and Composition of the Shell of the Archaeogastropod Limpet
Lepetodrilus elevatus elevatus (Mclean, 1988) ............................

TOSHIE KAWANO (CAMEY), KAYO OKAZAKI & LILLANE RÉ
Embryonic Development of Biomphalaria glabrata (Say, 1818) (Mollusca,
Gastropoda, Pianorbidae): A Practical Guide to the Main Stages ..........

TH. C. M. KEMPERMAN & G. H. DEGENAARS
Allozyme Frequencies in Albinaria (Gastropoda: Pulmonata: Clausiliidae)
from the lonian Islands of Kephallinia and Ithaka .........................

PAULA M. MIKKELSEN & RUDIGER BIELER
Biology and Comparative Anatomy of Three New Species of Commensal
Galeommatidae, with a Possible Case of Mating Behavior in Bivalves ....

ALAN E. STIVEN AND JOHN ALDERMAN
Genetic Similarities Among Certain Freshwater Mussel Populations of the
Lampsiis Genus in North Carolina ccoo ec unies

75

87

143

343

63

99

107

129

25

33

| РАТИ % u
.
UVA

o

| °
o as | on LAC y UE LL мг a TI =
paa ne ао ПА PCR
= АВР be de |

ES PP IA
ee (aap Ay we LEN sr al
— gf = я van > р

2 1.00% й АД
LE

>

я

var y ri (22) AA р A у Vall
roy e é 2 à y m
u te: o a Y aA SID it 2 few
| ( В ¡ue 4 м} ur Buy 4
Go?
y № MAI SAY ne
EC mis
Eole Gers u a
=> a
DIN
a A e O
x
mw DCR в sa el
u
es
=
= Dz A MA
E&
40 44%
©

*

WHY NOT SUBSCRIBE TO MALACOLOGIA?

ORDER FORM

Your name and address

Send U.S. $26.00 for a personal subscription (one volume) or U.S. $45.00 for an
institutional subscription. Make checks payable to “MALACOLOGIA.”

Address: Malacologia, Academy of Natural Sciences
1900 Benjamin Franklin Parkway
Philadelphia, PA 19103-1195, U.S.A.

MALACOLOGIA, 1992, 34(1-2):

34(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

1992

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 рт.

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 20.0cm
(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
2 or ‘3. Letters and numbers should be at
least 3mm high after reduction. Several
drawings or photographs may be grouped
together to fit a page. Photographs are to be
high contrast. High contrast is especially im-
portant for histological photographs.

13. All illustrations are to be numbered
sequentially as figures (not grouped as plates
or as lettered subseries), and are to be ar-
ranged as closely as possible to the order in
which they are first cited in the text. Each
figure must be cited in the text.

14. Scale lines are required for all nondi-
agrammatic figures, and should be conve-
nient lengths (e.g., “200 шт,” 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 if
grant or institutional support is available.

SUBSCRIPTION COSTS

25. Effective August 1992, personal sub-
scriptions are U.S. $26.00 and institutional
subscriptions are U.S. $45.00. Back issues
and single volumes: $35.00 for non-
institutional purchaser; $45.00 for institutional
purchaser. There is a one dollar handling
charge for all purchases of single volumes.
Address inquiries to the Subscription Office.

VOL. 34, NO. 1-2 MALACOLOGIA

CONTENTS

PAULA M. MIKKELSEN 8 RÚDIGER BIELER
Biology and Comparative Anatomy of Three New Species of Commensal
Galeommatidae, with a Possible Case of Mating Behavior in Bivalves ....
TOSHIE KAWANO (CAMEY), KAYO OKAZAKI & LILLANE RÉ
Embryonic Development of Biomphalaria glabrata (Say, 1818) (Mollusca,
Gastropoda, Planorbidae): A Practical Guide to the Main Stages ..........
ТН. С. М. KEMPERMAN & G. H. DEGENAARS
Allozyme Frequencies in Albinaria (Gastropoda: Pulmonata: Clusia
from the lonian Islands of Kephallinia and Ithaka ........................
VASILIS K. DIMITRIADIS & DIMITRIS HONDROS
Effect of Starvation and Hibernation on the Fine Structural Morphology of
Digestive Gland Cells of the Snail Helix lucorum .............. th Pea ah
PHILIPPE BOUCHET & JEAN-PIERRE ROCROI
Supraspecific Names of Molluscs: A Quantitative Review .................
ROBERT H. COWIE Be;
Shell Pattern Polymorphism in a 13-Year Study of the Land Snail Thebes sl u
pisana (Müller) (Pulmonata: Helicidae) ....................................
J. P. A. GARDNER | ME.
Null Alleles and Heterozygote Deficiencies Among Mussels (Mytilus edulis. р ды
and M. galloprovincialis) of Two Sympatric Populations ................... 9
FOLCO GIUSTI, GIUSEPPE MANGANELLI 8 JORGE V. CRISCI IM
A New Problematical Hygromiidae from the Aeolian Islands (Italy) (Pulmo- =
nata: Helicoidea) .......:...... a A E IR оо Eu
STEPHEN HUNT
Structure and Composition of the Shell of the Archaeogastropod pale
Lepetodrilus elevatus elevatus (Mclean, 1988) ............................
GEORGE M. DAVIS, CUI-E CHEN, CHUN WU, TIE-FU KUANG, XIN-GUO XING,
LI LI, WEN-JIAN LIU & YU-LUN YAN
The Pomatiopsidae of Hunan, China (Gastropoda: Rissoacea) ........... Pb.
V. K. DIMITRIADIS, D. HONDROS 4 A. PIRPASOPOULOU E
Crop Epithelium of Normal Fed, Starved and Hibernated Snails Helix luco-
rum: A Fine Structural-Cytochemical Study .............................. x
ALAN E. STIVEN AND JOHN ALDERMAN rt
Genetic Similarities Among Certain Freshwater Mussel Populations of the ==
Lampsilis Genus in North Carolina ........................................ O 5

О

072

HE

Are Y A
AA LT a Ha

DE

En
Kurses

ее ee True ны < 4 4 . > x А г si , N = : har ot Ve
PAP LITER he A ; a и eon x 4 e > Ss geo:
Ber

HEN De mB nen
Ay epee teu ey

A A A AAA HE A UP т LENE AI av AE q ¢ : т 4 } a
а OT UA ET TE PTS CITE vr X 5 ‘ - 3 + re
TB мент ae NU I ENT MA UR An a ETA .
CAC A OT Fu SS an у
, : : ое Ва UE инок
Parties 4 . h : iento HONTE
rare CURE : Me SAR Ab о NOR SOE ARA

МИНЫ y . AA NANO
APO ED MLE Майк x AN RDS
PORTE nu
3 PE
a o A A Pb p " р ° » PERS Eu E
A NENA GG $ ave
A wort AAA LHe
Takin AP AA Tai oe AGE APT A
Am nern чу HAT HPA A ит
MIE AAA Hf MEME a
LAIA DA OIGA AGA eg A

MORNE MERE HSH a
PLE MUN MIA? отно р f 4 y ea
A AE м BR ANSE AA о WENN рен Y 4 a o reee
2 ee ge ho CNA ат ель a CRE >
прет as QU UT WANE a ди A bi и Чу CET
MERE GAMERA YEW) DIR EE RAB UL OTN OM WBA LA VAT PAS YBN BUY river Y ama al Mero ee etry nip
DOTE Te тия VA LI NE 4 Va UT але VA VRP UN À Rae | f Peng ton Y E ARA PRA PS NN e
MAGA EA UA IA AA und YP OM GE LAE Фо р ARE ORS CEE м v vw и ЧА М Fe о
PRA AAA и A LANGE 7 ИИ AMIE ATE AI VAN OA RE apt РЦ , AO
IRAN

ME NY
и BAM ky ay
lé € к WRU NOMA Ea HAE NDA PS LENS
DCE Y y e ыы
4 p * A ме 4 UA
» AS

Roe
E paren