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

Full text of "Malacologia"

ЧЧГ4 



HARVARD UNIVERSITY 

Library of the 

Museum of 

Comparative Zoology 



VOL 35 1993 



MALACOLOGIA 



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







Publication dates 


Vol. 


28, 


No. 1-2 


19 January 1988 


Vol. 


29, 


No. 1 


28 June 1988 


Vol. 


29, 


No. 2 


16 Dec. 1988 


Vol. 


30, 


No. 1-2 


1 Aug. 1989 


Vol. 


31, 


No. 1 


29 Dec. 1989 


Vol. 


31, 


No. 2 


28 May 1990 


Vol. 


32, 


No. 2 


7 June 1991 


Vol. 


33, 


No. 1-2 


6 Sep. 1991 


Vol. 


34, 


No. 1-2 


9 Sep. 1992 


Vol. 


35, 


No. 1 


14 July 1993 



t»"^ MCZ 

LIBRARY 
VOL 35, NO. 1 1993 

JUL 2 1 1993 

HARVARD 
UNIVERSITY 



MALACOLOGIA 



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



MALACOLOGIA 

Editor-in-Chief: 
GEORGE M. DAVIS 

Editorial and Subscription Offices: 

Department of Malacology 

The Academy of Natural Sciences of Philadelphia 

1900 Benjamin Franklin Parkway 

Philadelphia, Pennsylvania 19103-1195, U.S.A. 



EUGENE COAN 

California Academy of Sciences 

San Francisco, CA 



Co-Editors: 



Assistant Managing Editor: 

CARYL HESTERMAN 

Associate Editors: 



CAROL JONES 
Denver, CO 



JOHN B. BURCH 
University of Michigan 
Ann Arbor 



ANNE GISMANN 
Maadi 

Egypt 



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



KENNETH J. BOSS 

Museum of Comparative Zoology 

Cambridge, Massachusetts 

JOHN BURCH, President 

MELBOURNE R. CARRIKER 
University of Delaware, Lewes 

GEORGE M. DAVIS 
Secretary and Treasurer 

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



JAMES NYBAKKEN 

Moss Landing Manne Laboratory 

California 

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

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

SHI-KUEI WU 

University of Colorado Museum, Boulder 



Participating Members 

EDMUND GITTENBERGER JACKIE L VAN GOETHEM 

Secretary, UNITAS MALACOLOGICA Treasurer, UNITAS MALACOLOGICA 

Rijksmuseum van Natuurlijke Koninklijk Belgisch Instituut 

Historie voor Natuurwetenschappen 

Leiden, Netherlands Brüssel, Belgium 



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

ELMER G. BERRY, 
Germantown, Maryland 



Emeritus Members 

ROBERT ROBERTSON 

The Academy of Natural Sciences 

Philadelphia, Pennsylvania 



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



Copyright © 1 993 by the Institute of Malacology 



1993 
EDITORIAL BOARD 



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 Stieffield 
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 



E. GITTENBERGER 
Rijksmuseum van Natuurlijke Historie 
Leiden, Netherlands 

F. GIUSTI 
Università di Siena, Italy 

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

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

A. V. GROSSU 
Universitatea 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 
Göteborg, Sweden 

S. HUNT 
Lancashire 
United Kingdom 

R. JANSSEN 

Forschungsinstitut Senckenberg, 
Frankfurt am Main, Germany 

R. N. KILBURN 
Natal Museum 
Pietermaritzburg, South Africa 

M. A. KLAPPENBACH 

Museo Nacional de Historia Natural 

Montevideo, Uruguay 



J. KNUDSEN 

Zoologisk Institut & Museum 

Kebenhavn, Denmark 

A. J. KOHN 

University of Washington 

Seattle. U.S.A. 

A. LUCAS 

Faculté des Sciences 

Brest, France 

C. MEIER-BROOK 
Tropenmedizinisches Institut 
Tübingen, Germany 

H. К. MIENIS 

Hebrew University of Jerusalem 

Israel 

J. E. MORTON 
The University 
Auckland, New Zealand 

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

R. NATARAJAN 
Marine Biological Station 
Porto Novo, India 

J. 0KLAND 
University of Oslo 
Norway 

T. OKUTANI 
University of Fisheries 
Tokyo, Japan 

W. L. PARAENSE 

Instituto Oswalde Cruz, Rio de Janeiro 

Brazil 

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

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 

Ol Z. Y. 

Academia Sinica 

Qingdao, People's Republic of China 



D. G. REÍD 

The Natural History Museum 

London, United Kingdom 

N. W. RUNHAM 

University College of North Wales 

Bangor, United Kingdom 

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

A. STAÑCZYKOWSKA 
Siedlce, Poland 

F. STARMÜHLNER 

Zoologisches Institut der Universität 

Wien, Austria 

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

W. STREIFE 
Université de Caen 
France 

J. STUARDO 
Universidad de Chile 
Valparaiso 

S. TILLIER 

Muséum National d'Histoire Naturelle 

Paris, France 

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

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, 1993, 35(1): 1-7 



ADULT AND JUVENILE FLASHES IN THE 
TERRESTRIAL SNAIL DYAKIA STRIATA 

Jonathan Copeland^ & Maryellen Maneri Daston^ 

ABSTRACT 

Photomultiplier recordings were used to categorize the flash types produced by caged adults 
and juveniles of the terrestrial bioluminescent snail Dyakia striata. Simple and modulated flashes 
were produced by both adult and juvenile snails. Flash duration and interflash interval were 
measured in both adults and juveniles. Adult flashes were less bright than juvenile flashes, and 
adult flashes were usually simple (non-modulated) flashes. Interflash intervals were usually 
longer for adult snails than juveniles. These findings are interpreted in terms of the neural control 
of this unusual effector organ. 

Key words: bioluminescence, Dyakia, behavior. 



INTRODUCTION 

Dyakia striata (Ahophantidae), found in 
Singapore and Malaysia (Parmentier & 
Barnes, 1975) is the only terrestrial snail 
known to be luminescent. It produces light 
from a luminescent organ, called the organ of 
Haneda (reviewed in Haneda, 1981), located 
within the head-foot. Discrete flashes of light, 
sometimes single-peaked and sometimes 
multiple-peaked, are produced (Haneda, 
1981 ; Parmentier & Barnes, 1975). Occasion- 
ally, glows occur (Haneda, 1981). 

Luminescence was once thought to occur 
only in juvenile snails and then disappear 
(Haneda, 1981; Martoja & Bassot, 1970; Par- 
mentier & Barnes, 1975). However, more re- 
cent studies have shown that it can some- 
times persist to adulthood (Copeland & 
Maneri, 1984; Counsilman et al., 1987; Cope- 
land & Daston, 1989). 

Because previous workers had studied ju- 
venile luminescence only (Haneda, 1981; 
Parmentier & Barnes, 1975), here, the 
flashes of adult and juvenile snails are com- 
pared. Differences in bioluminescence be- 
tween young and adults have been found in 
other bioluminescent systems, and these dif- 
ferences have often been instructive in terms 
of neural and biochemical control (Herring, 
1978). 

MATERIALS AND METHODS 

Snail flashes were recorded using a tripod- 
mounted photomultiplier tube (RCA 6655-A) 



that modulated the carrier frequency of a volt- 
age controlled oscillator (A. R. Vetter, Inc.). In 
this way, the snail flashes, which were rela- 
tively slow, were sensed by the photomulti- 
plier and this signal then modulated the high 
frequency oscillator. The high frequency os- 
cillator signal increased and decreased in 
parallel with changes in the light intensity. 
This high frequency signal was stored on a 
portable A.C. tape recorder (SONY 3600). 
Later, the tape recorded signals were played 
back through a demodulator unit and then into 
a chart recorder (Grass Model 79B). The sec- 
ond tape recorder channel was used to record 
voice commentary simultaneously from the 
observer. 

Flashes were recorded from snails placed 
either in a 10 gallon glass aquarium (adults) 
or a 50 mm diameter beaker (juveniles). 
Flashes from adult snails were recorded us- 
ing a tripod-mounted photomultiplier which 
could be repositioned by the observer who 
simultaneously noted the occurrence and 
type (simple, modulated) of the flash. Adult 
snails moved considerably less than juvenile 
snails (Copeland & Daston, 1989). Flashes 
from juvenile snails were recorded with no ob- 
server present. These snails were placed in a 
beaker that faced the photocell. Because the 
juvenile snails moved a good deal, aluminum 
foil was wrapped around most of the beaker 
to ensure that flashes would be reflected to- 
ward the photomultiplier tube regardless of 
the orientation of the snail. 

A snail would usually retract into the shell 
completely when picked up and transferred to 



^Department of Biology, Georgia Southern University, Statesboro, Georgia 30450-8042, U.S.A. 

^Department of Cell Biology and Anatomy, University of Cincinnati School of Medicine, Cincinnati, Ohio 45267, U.S.A. 



COPELAND & DASTON 



the aquarium or beaker. Therefore, the first 
ten minutes of data from each one-hour re- 
cording session were ignored to allow time for 
the snail to recover from this disturbance. 
Most measurements with the photomultiplier 
were made in complete darkness. However, 
several observations of the movements of the 
snail's body while it flashed were made which 
used dim red illumination to silhouette the 
snail's body. Copeland (1988) showed that 
there was no response to red light when neu- 
ral recordings were made from the optic nerve 
of D. striata. 

All measurements were made from the 
chart recorder traces. Flash duration (from 
baseline to baseline) was measured, as was 
interflash interval (interval from the beginning 
of one flash to the beginning of the subse- 
quent flash). Also, the number of peaks in 
each flash were counted. A peak was consid- 
ered to have occurred when the flash de- 
creased rapidly in amplitude (but not com- 
pletely) to baseline. 

Adult snails were collected in Singapore 
and tested at 27-29°C. Juvenile snails were 
raised from eggs hatched in the lab. They 
were kept in 5 cm x 30 cm plastic cages with 
sterilized potting soil on the bottom. Cages 
were misted daily. Juvenile snails were fed 
meat and vegetable Gerber's baby food (Ma- 
son & Copeland, 1988) which was changed 
every other day. A 12:12 light:dark cycle and 
28°C were maintained. Juvenile recordings 
were made at 28°C. 



RESULTS 

Flash Types and Patterns 

Adult Flash Types: The type of luminescence 
spontaneously produced by adult D. striata 
ranges from a discrete bright flash (Fig. 1A, 
first three flashes) to a very weak low intensity 
glow-like flash (Fig. 1A, 4th flash). Time from 
baseline until flash peak was variable but less 
than one second. 

The flashes of seven adult snails were 
viewed. They flashed continuously (no inter- 
flash interval greater then 60 sec) for 19-45 
minutes within the total one hour recording 
period (first 10 minutes ignored). These 
flashes, when viewed directly or monitored in- 
directly via the photomultiplier, were catego- 
rized as simple flashes (with a single peak), 
which were symmetrical (Fig. 1 B, symmetrical 
rise and fall of flash) or asymmetrical (Fig. 




JA_ 



10 sec 

FIG. 1. Flashes recorded from freely moving adult 
snails with a tripod-mounted photomultiplier tube. 
The records read from left to right, with time in the 
X-axis and flash intensity in the y-axis. Simple and 
modulated flashes are shown. A-C, simple flashes; 
D-G, modulated flashes; A (first three) and C, 
asymmetrical flashes (quenched slowly); A (fourth 
flash) appeared as dim weak glow (not a flash). 



1С), and modulated flashes (with more than 
one peak). In modulated flashes, an intensity 
modulation produced a pulsation of light (Fig. 
1 D-G). Sometimes, the pulsation could be re- 
solved into two discrete flashes (Fig. IG). 
Flashes with three or four peaks occurred, but 
these were rare (< 1%) in adult snails. 

Both simple and modulated flashes in adult 
D. striata last from 0.5 to 6 seconds (Fig. 2A), 
although there was a tendency for simple 
flashes to be shorter than modulated flashes. 
This difference in flash duration was signifi- 
cant in snails 2 and 3 but not snail 1 in Figure 
2A (t-test, p < 0.05). 

All adult snails showed both simple and 
modulated flashes, although the ratio of sim- 
ple:modulated flashes varied from about 1:1 
to 2:1 in the seven snails viewed. 

Usually, several flashes of one kind would 
be followed by several flashes of the other 
kind, but the two types of flashes (simple or 
modulated) could be interspersed. No obvi- 
ous correlation was seen between snail be- 
havior and flash type. 

The interflash interval for the animals illus- 



ADULT AND JUVENILE FLASHES IN DYAKIA 



SIMPLE FLASHES 



(0 
Ш 

z 
< 

-J 
u. 

u. 
о 

ш 

ta 
S 
э 
z 



в 



(О 

ш 

X 
(О 



U. 

II. 
о 

ее 
ш 
еа 

S 



30 



20 



10 



Id 



I 



30 



20 



10 



J 



M 



20 



10 



п 



¿Uk 



2.0 4.0 6.0 1.0 2.0 4.0 S.O t.O 



2.0 4.0 6.0 8.0 



20 
10 



Jnn 



20 
10 



MODULATED FLASHES 

2 3 

20 

\ "IS 



2.0 4.0 60 8.0 



2.0 4.0 6.0 8.0 

FLASH DURATION 
(SECONDS) 



2.0 4.0 6.0 8.0 



IS 
10 

S 



IS 

10 

s 




66 MINUTES 



[W^nn4 nn ^n 



s 10 16 20 25 30 35 40 45 60 55 60 65 70 76 80 

2 n 



63 MINUTES 



ЧпП-^ 



s 10 16 20 26 30 35 40 45 80 66 60 66 70 78 80 



Па... 



31 MINUTES 



JOJL. 



S 10 16 20 25 30 35 40 48 60 55 60 66 70 76 60 

INTERFLASH INTERVAL 
(SECONDS) 

FIG. 2. Flash duration for sirnpie (top row) and modulated (bottom row) flashes produced by three different 
adult snails (animals #1-3) over sample periods of 56 (left column, top), 53 (middle column, top), and 31 
(right column, top) minutes respectively. B. Interflash intervals from the same snails during the same one 
hour test sessions as in A. All data measured from photomultiplier records. 



COPELAND & DASTON 



trated in Figure 2A is shown in Figure 2B. The 
interflash interval for these individuals varied 
between 2-80 seconds, with a mean inter- 
flash interval of 18.0 ± 1.2 S.D. sec. 

The adult flash was yellow-green in color 
and weak in intensity. Indeed, some dark ad- 
aptation was necessary before an observer 
could easily see the flash. (When compared 
by eye to the flash of the firefly Pteroptyx mal- 
lacae, the flash of D. striata was considerably 
weaker in intensity.) However, during the 
most intense flashes, the entire anterior part 
of the snail was illuminated. 

Because any movement of the head-foot, 
which contains the luminescent organ of D. 
striata, could create the illusion of multiple 
peaks when the luminescent organ was 
viewed by a stationary photomultiplier, on 
several occasions a flashing adult D. striata 
was viewed using weak red backlighting to 
produce a silhouette. When modulated 
flashes occurred, the head-foot was continu- 
ously extended against the substrate. Thus, 
the modulated flashes could not have oc- 
curred because a continuously glowing lumi- 
nescent organ was moved in and out of the 
shell like a shutter, something that has been 
found with other luminescent organs in other 
animals (Herring, 1978). 

Juvenile Flash Types: Juvenile flash types in 
D. striata were similar to adult flash types: 
simple and modulated flashes occurred, as 
did glows. As in the adult, the color of the 
flash was yellow-green, but the flash was 
considerably brighter to the eye. Little dark 
adaptation was necessary to view juvenile 
snail flashes, and many of the flashes ap- 
peared to the eye to contain pulsations. In 
fact, the juvenile flash could be so bright and 
had such a range of intensities when com- 
pared to the adult flash that it was difficult to 
obtain complete records from all the juvenile 
snails tested (N = 10) because many of the 
flashes from some snails saturated the pho- 
tomultiplier tube, thus preventing multiple 
flashes from being recorded. 

The results from two juvenile snails whose 
flashes were within the range of the photo- 
multiplier for the entire test period are shown 
in Figure 3. They flashed continuously (no in- 
terflash interval greater than 60 seconds) for 
18 to 30 minutes. Simple flashes lasted 0.5- 
2.5 seconds and modulated flashes lasted 
0.5-5.5 seconds (Fig. 3). The difference be- 
tween simple and modulated flashes was sig- 
nificant (t-test, p < 0.02). The ratio of simple: 



modulated flashes was less than 1 :2 for one 
snail and 1 :7 for the other snail. Many modu- 
lated flashes had three peaks or more (12- 
41%). 

The interflash interval from the two juvenile 
snails shown in Figure 38 varied between 2 
and 50 seconds. Mean interflash interval (N 
= 2) was 9.8 ± 0.5 S.D. sec. 



DISCUSSION 

Adult D. striata produce weak intensity 
flashes that are usually simple flashes. The 
average interflash interval is about 18 sec- 
onds (Fig. 2). Adult simple flashes are usually 
shorter in duration than adult modulated 
flashes (Fig. 2). Juvenile flashes are much 
brighter to the eye and many appear to twin- 
kle with multiple peaks. Most juvenile flashes 
are modulated flashes and have an average 
interflash interval of about 10 seconds (Fig. 
3). Juvenile simple flashes are also shorter in 
duration than juvenile modulated flashes. 

These findings extend the observations of 
Haneda (1981) and Parmentier & Barnes 
(1975), who noted the presence of simple 
flashes and flashes with multiple peaks (mod- 
ulated flashes) in juvenile D. striata but did 
not quantify these flashes and did not com- 
pare the flashes of juveniles and adults. 

Because it is now known that adult flashes 
occur in D. striata and that adult and juvenile 
flashes differ, it might be instructive to look at 
flash similarities and differences from the per- 
spective of neural and biochemical control of 
flashing. 

Virtually nothing is known about the neural 
control of bioluminescence in D. striata. No 
reflex-evoked luminescence (flashes, glows, 
scintillations) occur in response to tactile stim- 
ulation (Parmentier & Barnes, 1975) as it 
does in many bioluminescent organisms 
(Herring, 1978), but flashing can occur as fast 
as 0.5 Hz (Parmentier & Barnes, 1975). How- 
ever, photic stimuli, either from a flashing 
conspecific snail or an electric torch, can 
change the flash rate of a flashing snail (Cope- 
land & Daston, 1989). Additionally, ultrastruc- 
tural evidence exists for the presence of 
nerve endings in the luminescence organ 
(Maneri, 1985). These facts, plus the rapid 
rise time of the flash, suggest that flashing in 
D. striata is under nervous control. 

Even less is known about biochemical con- 
trol of bioluminescence in D. striata. Haneda 
(1963), using dried and crushed bodies of 



ADULT AND JUVENILE FLASHES IN DYAKIA 

SIMPLE FLASHES 



20 30 40 50 




10 20 30 40 50 

MODULATED FLASHES 



П 




20 30 40 50 



п_У 



FLASH DURATION 
(SECONDS) 



4n^ 



23 MINUTES 



25- 



^JU 



30 MINUTES 



Д^^^^'^Л^ 



18 MINUTES 



INTERFLASH INTERVAL 
(SECONDS) 

FIG. 3. Flash duration for simple (top row) and modulated (bottom row) flashes produced by two different 
juvenile snails (animals #4-5) over sample periods of 23 (left column, top), 30 (middle column, top), and 18 
(right column, top) minutes respectively. Data in A2 and ЛЗ are from the same snail. B. Interflash interval 
from the same snails during the same one hour test sessions as in A. All data measured from photomultiplier 
records. 



COPELAND & DASTON 



snails, could not find evidence of a lucifehn- 
luciferase reaction with hot or cold water ex- 
tracts. He did, however, find microscopic ev- 
idence for granules in the cells of the 
luminescent organ which emitted a golden 
autofluorescence when viewed with a fluores- 
cence microscope. Isobe et al. (1988) ex- 
tracted a green fluorescent substance from D. 
striata (presumed to be the luminescent sub- 
stance) that is probably different from the lu- 
minescent substance in fireflies. 

Previous work in other bioluminescent sys- 
tems, such as fireflies, have used the obser- 
vations of flashes and their kinetics to sug- 
gest physiological and biochemical control 
mechanisms. For example, natural lumines- 
cence, such as continuous glow, intermittent 
glow, pulsation, and flash in fireflies (Buck, 
1948), and experimentally induced lumines- 
cence in fireflies, such as pseudoflash, hy- 
poxic glow, and scintillation (Buck, 1948; 
Harvey, 1951; Carlson, 1968) have all been 
used to support both the oxygen-control hy- 
pothesis of flash (Buck, 1948) and the ner- 
vous-system-control hypothesis (McElroy, 
1947, 1951; Carlson, 1961). 

The initiation of a flash in fireflies involves 
more than the chemical addition of the lumi- 
nescent reactants. In vitro, it takes 60 msec 
for light production to occur if oxygen is added 
to a mixture of enzyme and substrate that has 
already formed an enzyme-substrate com- 
plex (DeLuca & McElroy, 1974). The same 
reaction takes several hundred milliseconds 
to develop if just enzyme and substrate are 
added in the presence of oxygen (DeLuca & 
McElroy, 1974). In adult fireflies, where a tra- 
cheal end organ is in the pathway between 
nervous system and photocyte (Smith, 1963), 
light production usually takes less than 100 
msec to occur from the time the action poten- 
tials leave the 6th and 7th abdominal ganglia 
(Case & Buck, 1963). In larval fireflies, where 
the nervous system ends directly on the pho- 
tocytes, light production can take up to a sec- 
ond to occur from the time the action poten- 
tials leave the 8th abdominal ganglion. In 
firefly larvae, the light production is a slow 
glow, not a rapid flash (Carlson, 1968). 

The number of peaks and the intensity of 
the flash in juveniles suggest that a difference 
may exist in adult and juvenile luminescent 
organ peripheral neural control and biochem- 
istry, a possibility reinforced by the ultrastruc- 
tural findings of Maneri (1985), where differ- 
ences between adults and juveniles in the 
size and density of photocyte granules were 



seen. Perhaps the larger, more electron- 
dense photocyte secretory droplets of juve- 
nile snails contain more concentrated lu- 
ciferin, or perhaps the photocytes are 
activated more often or more vigorously by 
the nervous system in juveniles. 

In addition to peripheral changes, central 
changes may also occur. For example, the 
decrease in interflash interval in juveniles is 
paralleled by an increased locomotion in the 
juveniles (Copeland & Daston, 1989). Addi- 
tionally, because simple flashes are usually of 
shorter duration than modulated flashes, the 
latter might be modulated because they are 
showing facilitation or summation. Summa- 
tion, at least in skeletal and some smooth 
muscle, is due to both central nervous system 
activation at a rapid rate and peripheral effec- 
tor inability to respond 1 :1 to each central ner- 
vous system stimulus (Eckert et al., 1990). 

Whether these differences reflect matura- 
tion or some other process, such as senes- 
cence (Martoja & Bassot, 1970), is not clear. 
Additionally, the actual locus of the changes, 
be they central, peripheral, or both, is also not 
known. 



AKNOWLEDGMENTS 

This work was supported in part by a grant 
from the National Geographic Society. We 
thank Dr. A. D. Carlson for a critical reading of 
an earlier version of the manuscript and also 
thank an anonymous reviewer for many help- 
ful comments and saint-like patience, both of 
which vastly improved the manuscript. 



LITERATURE CITED 

BUCK, J. в., 1948, The anatomy and physiology of 
the light organ in fireflies. Annals of the New York 
Academy of Science, 49: 397-482. 

CARLSON, A. D., 1 961 , Effects of neural activity on 
the firefly pseudoflash. Biological Bulletin, Marine 
Biological Lab, Woods Hole, 121: 265-276. 

CARLSON, A. D., 1968, Neural control of firefly 
bioluminescence. Advances in Insect Physiol- 
ogy, 6: 51-96. 

CASE, J. F. & J. B. BUCK, 1 963, Control of flashing 
in fireflies. II. Role of the central nervous system. 
Biological Bulletin, 125: 234-250. 

COPELAND, J-, 1988, Optic nerve response to 
photic stimulation in Dyakia (Quantula) striata. 
Comparative Biochemistry and Physiology. A89: 
391-400. 

COPELAND, J. & M. M. DASTON, 1989, Biolumi- 



ADULT AND JUVENILE FLASHES IN DYAKIA 



nescence in the terrestrial snail Dyakia (Quan- 
tula) striata. Malacologia, 30: 317-324. 

COPELAND, J. & M. MANERI, 1984, Biolumines- 
cence and communication in the terrestrial snail 
Dyakia (Quantuia) striata. Society for Neuro- 
science Abstracts, 10: 396. 

COUNSILMAN, J. J., D. LOH, S. Y. CHAN, W. H. 
TAN, J. COPELAND & M. MANERI, 1987, Fac- 
tors affecting the rate of flashing and loss of lu- 
minescence in Asian land snail, Dyakia striata, 
Veliger, 29: 394-399. 

DeLUCA, M. & W. D. McELROY, 1974, Kinetics of 
the firefly luciferase catalysed reactions. Bio- 
chervistry, 13: 921-925. 

ECKERT, R., D. RANDALL & G. AUGUSTINE, 
1988, Animal phiysiology. W. Freeman, New 
York. 

HANEDA, Y., 1963, Further studies on a luminous 
land snail, Quantuia striata, in Malaya. Yokusuka 
City Museum Science Report, 8: 1-7. 

HANEDA, Y., 1981, Luminous activity of the land 
snail Quantuia striata. Pp. 257-265, in м. a. de- 
LUCA & w. D. MCELROY, eds., Bioluminescence and 
chemiluminescence. Academic Press, New York. 

HARVEY, E. N., 1951, Bioluminescence. Academic 
Press, New York. 

HERRING, P. J., 1978, Bioluminescence in action. 
Academic Press, New York. 

ISOBE, M., D. UYAKUL, T GOTO & J. J. COUN- 
SILMAN, 1988, Dyakia bioluminescence-1. Bio- 
luminescence and fluorescence spectra of the 
land snail, D. striata. Japanese Journal of Cell 
Biology, 25: 791-795. 



Mcelroy, W. D., 1947, The energy source for bio- 
luminescence in an isolated system. Proceed- 
ings of the National Academy of Science, U.S.A., 
33: 342-345. 

Mcelroy, W. D., 1951, Properties of the reaction 
using adenosine triphosphate for biolumines- 
cence. Journal of Biological Chemistry, 191: 
547-557. 

MANERI, M., 1985, Bioluminescence and sexual 
maturity in the terrestrial snail, Dyakia strata. 
Masters Thesis, University of Wisconsin-Mil- 
waukee. 

MARTOJA, M. & J. M. BASSOT, 1970, Étude his- 
tologique de complexe glandulaire pedieux de 
Dyakia strata, Goodwin et Austin, gastéropode 
pulmoné données sur l'organe lumineux. Vie et 
l[/lillieu, Serie A: Biologie Marine, XXI, Fase. 2-A: 
395-452. 

MASON, J. & J. COPELAND, 1988, The incidence 
and variety of Lehmannia valentiana conjoined 
twins: related breeding experiments (Gastro- 
poda, Pulmonata). Malacologia, 28 (1-2): 17-27. 

PARMENTIER, J. & A. BARNES, 1975, Observa- 
tions on the luminescence produced by the Ma- 
layan gastropod Dyakia striata. Malayan Nature 
Journal 28: 173-180. 

SMITH, D. S., 1963, The organization and innerva- 
tion of the luminescent organ in a firefly, Photuris 
pennsylnvanica (Coleóptera). Journal of Cell Bi- 
ology : 6: 323-359. 



Revised Ms. accepted 20 April 1 992 



MALACOLOGIA, 1993, 35(1): 9-19 

THE LUMINESCENT ORGAN AND SEXUAL MATURITY IN DYAKIA STRIATA 

Maryellen Maneri Daston^ & Jonathan Copeland^ 

ABSTRACT 

Dyakia striata, a snail found in Singapore and Malaysia, is the only terrestrial mollusc known 
to be luminescent. It produces flashes of light by means of a discrete luminescent organ in the 
head-foot. Previous studies of D. striata emphasized juvenile snail luminescence and its loss 
with sexual maturity. We, however, subsequently discovered that luminescence persisted in 
large snails that were probably adults. Here, the gross and ultrastructural anatomy of the re- 
productive system and the luminescent organ were compared between three snail categories: 
small snails with a luminescent organ, large snails with a normal luminescent organ, and large 
snails incapable of luminescence. We found that loss of luminescence did not coincide with 
sexual maturity. Mature gametes were found in the ovotestis of large snails capable of light 
production. Thus, some large D. striata were adults, possessed a structurally normal lumines- 
cent organ, and could flash. Because there is no good external marker for sexual maturity in D. 
striata, this leaves open the possibility that the flash is involved in reproductive behavior. 

A comparison of the D. striata light organ with the light organs of two other mollusks suggests 
that the luminescence in D striatia is intraglandular and not intracellular. 

Key words: Dyal<ia, luminescence, behaviour. 



INTRODUCTION 

Dyakia striata (Ariophantidae), found in 
Singapore and Malaysia (Parmentier & 
Barnes, 1975), is the only terrestrial gastro- 
pod known to be luminescent. It produces 
flashes of light similar to those of a firefly by 
means of a discrete luminescent organ 
(Haneda, 1981; Copeland & Daston, 1989). 

The luminescent organ of D. striata, called 
the organ of Haneda (Martoja & Bassot, 
1970), is a complex, histologically discrete 
lantern in which light production is thought to 
be intracellular (Haneda, 1963, 1981; Bassot 
& Martoja, 1968; Martoja & Bassot, 1970). 
The organ of Haneda, located within the 
pedal gland complex in the anterior head-foot 
(Parmentier & Barnes, 1975: fig. 1) is modi- 
fied glandular tissue. It lies between the inter- 
mediate gland and the basal gland and con- 
sists of an epithelial integument, connective 
tissue, and photocytes (Martoja & Bassot, 
1970). 

That luminescence in D. striata occurs only 
in juvenile snails was first noted by Haneda 
and confirmed by others (reviewed by 
Haneda, 1981). At the onset of sexual matu- 
rity, the entire luminescent organ was thought 
to be reabsorbed by phagocytes and replaced 
by an absorption cyst (Bassot & Martoja, 



1968; Martoja & Bassot, 1970). The disap- 
pearance of the luminescent organ was sup- 
posed to coincide with the first maturation di- 
vision of the gametes (Martoja & Bassot, 
1970). However, our field collections pro- 
duced large-sized, apparently non-juvenile 
snails that were luminescent (Copeland & 
Maneri, 1984; Copeland & Daston, 1989). 

The purpose of this study is to determine if 
large luminescent D. striata were sexually 
mature and to investigate differences be- 
tween luminescent and non-luminescent 
large snails. Thus, we looked at the gross re- 
productive anatomy and the ultrastructure of 
the ovotestis and the ultrastructure of the or- 
gan of Haneda in small and large D. striata, 
and related this to light production. The gross 
reproductive anatomy has not been described 
for D. striata, nor has the ultrastructure of the 
luminescent organ or any part of the gonad. 



MATERIALS AND METHODS 

Snails were collected in public parks in Sin- 
gapore over a six-week period. The gross 
anatomy dissections were done in the field 
using freshly collected snails. Living snails 
were fixed and then prepared for electron mi- 
croscopy. Dyakia striata is difficult to maintain 



^Department of Anatomy and Cell Biology, University of Cincinnati School of Medicine, Cincinnati, Ohio 45267, U.S.A. 
^Department of Biology, Georgia Southern University, Statesboro, GA 30450-8042, U.S.A. 



10 



DASTON & COPELAND 



in laboratory culture. It Is thin shelled and, 
thus, difficult to ship from Singapore to the 
United States, so the sample size in all cate- 
gories is small. 

The luminescent organ was viewed in the 
intact snail using a non-invasive ultraviolet 
light technique (Copeland & Maneri, 1984; 
Copeland & Daston, 1989). This allowed 
large snails to be classified as with or without 
a "visible" luminescent organ. 

Because Copeland & Maneri (1984) and 
Counsilman et al. (1987) observed that all 
snails capable of light production show fluo- 
rescence when stimulated with an ultraviolet 
light, and because the luminescence of snails 
in captivity was often very infrequent (Maneri, 
1985; Counsilman et al., 1987), we assumed 
that snails with a "visible" luminescent organ 
(bright yellow-green dot near the mouth on 
the ventral surface of the head-foot in re- 
sponse to stimulation with ultraviolet light) 
could flash and that all snails with a "non- 
visible" luminescent organ (no fluoresence in 
response to ultraviolet light stimulation but a 
luminescent organ was subsequently found 
by dissection) could no longer flash. Some of 
the large snails and all of the small snails 
were directly observed to produce flashes. 

Two large snails (23.0 mm and 22.0 mm 
shell diameter) with "visible" luminescent or- 
gans, two large snails (23.0 mm shell diame- 
ter) with "non-visible" luminescent organs, 
and two small snails (4.5 mm and 5.0 mm 
shell diameter) were selected for ultrastruc- 
tural studies. 

The snails were anesthetized (ten min in a 
freezer) and then dissected in a chilled mol- 
luscan saline (Copeland & Gelperin, 1983). 
The ovotestis and organ of Haneda of large 
snails were removed and immediately placed 
in fixative. The ovotestis of the small snails 
could not be isolated due to its undeveloped 
and fragile state and, thus, no small snail 
ovotestes were included. To ensure uniform 
fixative penetration, the mature ovotestis was 
first cut into small pieces. The organ of 
Haneda was small enough (about 1 mm x 
0.5 mm) to be fixed whole. The tissues were 
fixed in 2% glutaraldehyde in 0.1 M caco- 
dylate buffer, then post fixed in osmium te- 
troxide in the same buffer (Eaken & Bran- 
denburger, 1975). The tissues were then 
dehydrated in an ethanol series and embed- 
ded in Spurr's low viscosity embedding me- 
dium. Thin sections were cut using a glass 
knife on a Porter-Blum MT-II Ultramicrotome 
and then placed on a 300-gauge copper grid. 



The specimens were viewed using a Hitachi 
HU-11B-2 electron microscope. 

The gross anatomy of the reproductive sys- 
tem was examined in freshly caught animals. 
Eleven small snails, the most abundant D. 
striata found, were dissected. Nine large 
snails with a "visible" luminescent organ were 
dissected, as were three large snails with a 
"non-visible" luminescent organ. These latter 
were the most difficult to find in a collection. 
The reproductive organs were isolated in mol- 
luscan saline and sketched while viewed 
through a 30 X dissecting microscope. 



RESULTS 



Gross Anatomy 



The small snails (shell diameter 13-1 6 mm; 
N = 4) had small, poorly developed repro- 
ductive systems when compared to the large 
snails (shell diameter = 20 mm, N = 9). Typ- 
ical small snail and large snail reproductive 
systems are shown in Figure 1 A and in Figure 
IB, C, respectively. The small snail reproduc- 
tive system was relatively small and undevel- 
oped compared to that of the large snails. 

A comparison between a large snail with a 
"visible" luminescent organ and a large snail 
with a "non-visible" luminescent organ is 
shown in Figure 1 B, С The snail with a "vis- 
ible" luminescent organ had an expanded 
dart gland (lobes were separated and ex- 
panded), a swollen dart gland duct, and a dart 
in the dart sac (Fig. 1 B). These features were 
also seen in four other large snails that had a 
luminescent organ. The snail with a "non-vis- 
ible" luminescent organ had a more compact 
dart gland (the lobes were tightly folded to- 
gether), a narrower dart gland duct, and no 
dart in the dart sac (Fig. 1С). These features 
were also found in two additional snails with a 
"non-visible" luminescent organ. The sper- 
moviduct of the snail with the "visible" lumi- 
nescent organ was swollen in comparison to 
the snail with no luminescent organ. Both an- 
imals had a reddish spermatheca. 

Microscopic Anatomy 

Ovotestis: The ovotestis of all of the large 
snails (N = 2 with "visible" luminescent or- 
gan and N = 2 with "non-visible" luminescent 
organ) contained mature spermatozoa. Ma- 
ture sperm were identified by the appearance 
of the axoneme of the f lagellum in cross sec- 



LUMINESCENT ORGAN AND SEXUAL MATURITY 



11 




12 



DASTON & COPELAND 



tion (Tompa, 1984). A group of spermatozoa 
surrounding a Sertoli cell is shown in Figure 
2A and a cross section of a flagellum at 
higher magnification in Figure 2B. The Sertoli 
cells are the largest of the four general cell 
types found in the acinus (sperm, oocytes, 
follicle cells, and Sertoli cells) (Tompa, 1984). 
Normally, stylommatophoran oocytes range 
from 50-200 |хт (Tompa, 1984). No cells of 
that size were found in the ovotestis. 

Luminescent Organ: Organ ofHaneda: All lu- 
minescent organs (N = 2 large-sized snails 
with "visible," N = 2 large-sized snails with 
"non-visible," and N = 2 small-sized snails 
with "visible," luminescent organs) showed 
an integument of dorsal ciliated epithelium, a 
ventral simple squamous epithelium, and 
large granular photocytes surrounded by con- 
nective fibers (Figs. 3, 4). 

Photocytes were recognized by the large 
secretory droplets that comprised much of the 
cytoplasm (Bassot & Martoja, 1968; Martoja & 
Bassot, 1 970). The size and appearance of the 
droplets varied among the different snail 
groups. The average droplet size for the large 
snails with a "visible" luminescent organ was 
0.14 (xm ± 0.02 S.D. (N = 15) (Fig. 3C) and 
2.4 ^JLM ± 0.56 S.D. (N = 15) for large snails 
with a "non-visible" luminescent organ (Fig. 
3D). For small snails, the average droplet size 
was5.8M-m±2.15S.D. (N = 15) (Fig. 4C, D). 

The substance in the droplets of the large 
snails with "visible" luminescent organs was 
homogeneous and was only slightly electron- 
dense (Fig. 3B, C), whereas the material in 
the droplets of the large snails with "non-vis- 
ible" luminescent organs contained a granu- 
lar substance (Fig. 3D). The substance in the 
droplets of the small-sized snails was homo- 
geneous and electron dense (Fig. 4B, C). 

Structures that have the ultrastructural 
characteristics of axon terminals (Taue, 1977; 
Heuser & Reese, 1974) were found between 
and directly beneath the integumentary epi- 
thelium in one large snail with a "visible" lu- 
minescent organ (Fig. 5A, B). Connective fi- 
bers (Fig. 5B) were also found that show the 
characteristic striated feature of collagen in 
longitudinal section at high magnification 
(Porter & Bonneville, 1968). 

When dissected, the organ of Haneda was 
shaped like a flattened discus. It was yel- 
lowish in appearance, and consisted of an 
epithelial integument which surrounded pho- 
tocytes. A reconstruction of the entire lumi- 
nescent organ is shown in Figure 6. 



DISCUSSION 
Sexual Maturity and the Luminescent Organ 

The reproductive systems of large D. striata 
(both with and without a "visible" luminescent 
organ) were well developed (Fig. 1), sug- 
gesting that reproductive maturity is not oblig- 
atorily linked to the loss of the organ of 
Haneda. In Figure 1, the large snail with a 
"visible" luminescent organ had a dart in its 
dart sac, suggesting a propensity for mating 
(Tompa, 1984). Using the red spermatheca 
as a criterion for prior mating (Tompa, 1984), 
both large snails shown in Figure 3 had al- 
ready mated at least once. The small-sized 
individuals, an example of which is shown in 
Figure 1A, possessed luminescent organs, 
undeveloped genitalia, and undeveloped dart 
glands and, thus, were probably sexually im- 
mature juvenile). 

Using ТЕМ, sperm was found in large-sized 
snails that had "visible" luminescent organs 
(and in those with "non-visible" luminescent 
organs as well) (Fig. 2). Taken together, we 
conclude that luminescence occurs in sexually 
mature individuals. This contradicts earlier 
studies, which described luminescence in D. 
striata as juvenile luminescence and indicated 
that luminescence was lost at sexual maturity 
(Haneda, 1981; Martoja & Bassot, 1970). 

It is possible that in the previous studies too 
few large-sized snails were found for adult lu- 
minescence to have been seen (i.e., sampling 
bias). For example, we searched for snails for 
1-2 hours every other day for two weeks at 
one collection site at the Institute of Educa- 
tion, National University of Singapore. At this 
site, 59 small snails were found. The ratio of 
those with a "visible" luminescent organ to 
those with a "non-visible" luminescent organ 
was 3.5:1. At the same site, 21 large snails 
were found, and the ratio of "visible":"non- 
visible" luminescent organ in these snails was 
0.4:1 (Copeland & Maneri, 1984). The num- 
ber of snails collected at this site was about 
average, as was the size distribution. Had we 
only collected a small number of snails of both 
size, the probability of finding a large snail 
with a "visible" luminescent organ might have 
been low. Additionally, adult snails flash less 
often than juvenile snails (Copeland & Das- 
ton, 1989), so adult luminescence might be 
easily overlooked. Also, we used an ultraviolet 
light to determine that snails possessed a lu- 
minescent organ. 



LUMINESCENT ORGAN AND SEXUAL MATURITY 13 













FIG. 2. Ovotestis of an adult snail. A. Sertoli cell (sc) with a group of sperm tails (arrow) (6300 x ). B. High 
magnification view of sperm tails in cross-section showing the axoneme (arrow) (54,300 x ). 



Cellular Structure and Function of the Organ 
of Haneda 

The organ of Haneda is discus-shaped and 
yellow. It consists of a dorsal ciliated epithe- 



lium, a ventral simple squamous epithelium, 
and large granular photocytes surrounded by 
connective fibers (Figs. 3-5). This confirms 
the morphology described by Bassot & Mar- 



14 



DASTON & COPELAND 




FIG. 3. Luminescent organ of adult snails. A. Ciliated epithelial cells; c, cilia (35,700). B. Photocyte with 
numerous mitochondria (m) (49,500 x); С Photocyte with secretory droplets (sd) (42,000 x); D. Photocyte 
with secretory droplets (sd) (7,500 x ). B, C, snail with "visible" luminescent organ; D, snail with "non-visible" 
luminescent organ. 



LUMINESCENT ORGAN AND SEXUAL MATURITY 



15 





FIG. 4. Luminescent organ of a juvenile snail with a "visible" luminescent organ. A. Ciliated epithelia cells; 
c, cilia (1 7,900 X). B. Border between ciliated epithelium (ep) and photocytes (sd, secretory droplets 
(4,000 X). C, D, material within the photocytes (C = 1 5,000 x; D = 1 3,000 x). 



toja (1968) and Martoja & Bassot (1970) us- 
ing light microscopy. 

Little is known about the mechanisms of 
light production in D. striata. The lumines- 
cence is thought to be intracellular, but this 



belief is inferential: a substance stored in the 
secretory droplets of the luminescent organ is 
believed to contain the luminescent substrate 
and enzyme, and the reaction is suspected to 
take place inside the photocytes (Bassot & 



DASTON & COPELAND 














^^^Г'^ 



FIG. 5. Evidence for neural innervation of the luminescent organ. Axon terminals (arrows) from the lumi- 
nescent organ of an adult snail. A. Ciliated epithelial cells (53,000 x). B. Beneath the ciliated epithelium, 
collagen fibers are seen (72,000 x). Abbreviations: c, cilia, со, collagen fibers. 



LUMINESCENT ORGAN AND SEXUAL MATURITY 
Pre-buccal Canal 



17 



Pre 




Foot Muscle 

FIG. 6. Reconstruction of a luminescent organ (organ of Haneda) in cross section. CE, ciliated epitheliurn; 
SE, simple squamous epithelium; N, nucleus; SD, secretory droplets; CF, collagen fibers. Scale; width of 
organ of Haneda = 1 mm. 



Martoja, 1968; Marloja & Basset, 1970; 
Haneda, 1963, 1981). What is known is that 
the luminescent substance in D. striata tests 
negatively to a luciferin-luciferase reaction 
(Haneda, 1 963) and, from spectrophotometric 
evidence that used extracted luminescent or- 
gans, that the luminescent substance of D. 
striata is different from firefly luciferin (Isobe 
eta!., 1988). 

The organ of Haneda is part of the pedal 
gland complex of D. striata. This pedal com- 
plex is larger In D. striata than it is in other 



stylommatophorans, in which only the dorsal 
gland and the pedal gland have been found 
(Martoja & Bassot, 1 970). Glands of the pedal 
complex usually secrete mucus extracellu- 
larly for use in locomotion (Barr, 1926; Mar- 
toja & Bassot, 1970; Kater, 1977). 

The structure of the organ of Haneda is 
similar to the structure of the luminescent or- 
gan in the two other known luminescent non- 
cephalopod mollusks (Nichol, 1960; Bowden, 
1950). In these other mollusks, the lumines- 
cence is associated with the secretion of mu- 



18 



DASTON & COPELAND 



cus from glands. In Pholas dactylus, a marine 
bivalve, the luminescent organ consists of a 
ciliated columnar epithelium that lies over the 
glandular cells which expel their secretions 
through the surface epithelium. The glandular 
cells are of three types: mucus secreting cells 
and two types of photocytes. Here, the lumi- 
nescence is under the control of the nervous 
system and is thought to be extracellular 
(Nichol, 1960). Latia neritoides, a freshwater 
limpet, has photocytes that are histologically 
similar to P. dactylus and D. striata. However, 
instead of being confined to a discrete organ, 
the photocytes are scattered over the body of 
the limpet in small clusters that lie beneath 
the surface cuboidal epithelium within the 
loose subepithelial tissue. Mucocytes, melan- 
ophores, and muscle fibers are found inter- 
mingled among the photocyte clusters. Lumi- 
nescence in L. neritoides is extracellular and 
does not involve the nervous system (Bow- 
den, 1950). 

The histological similarity between D. stri- 
ata, P. dactylus, and L. neritoides could indi- 
cate similar function: extracellular secretion of 
a luminescent mucous. Thus, although lumi- 
nescence in D. striata might be intracellular 
(Haneda, 1963, 1981; Martoja & Bassot, 
1970), it could also be extracellular and even 
intraglandular. It is possible that the lumines- 
cent substance is secreted from the photo- 
cytes and remains localized within the organ of 
Haneda. 

The difference in the appearance of the 
secretory droplets in the photocytes in the 
three types of snails examined (Figs. 3, 4) 
could be correlated with differences in the in- 
tensity of luminescent activity (Copeland & 
Daston, 1992, this issue). For example, Cope- 
land & Daston show that small snails have 
brighter flashes than large snails when the 
flashes are viewed either by eye or with a 
photomultiplier. Small snails have the largest 
secretory droplets (Fig. 4). The secretory 
droplets in small snails possess a substance 
that was homogenous but not electron-dense. 
Large snails with "non-visible" luminescent 
organs have intermediate-sized secretory 
droplets, but these are granular and non- 
homogenous (Fig. 3D). The granular appear- 
ance could represent a degenerative form of 
the luminescent substance. 

There was no indication of the phagocyto- 
sis of the photocytes described earlier (Bas- 
sot & Martoja, 1968; Martoja & Bassot, 1970). 
Some of the large snails with a "visible" lumi- 
nescent organ had photocytes with a highly 



convoluted plasma membrane (Fig. 3B), but 
unlike the findings of Martoja & Bassot 
(1970), no phagocytes were found in the in- 
dentations (Fig. 3B). 

One of the adult snails with a "visible" lu- 
minescent organ exhibited variability in the 
appearance of the photocytes: in some 
cases, the cytoplasm was crowded with mito- 
chondria and the plasma membrane was con- 
voluted, whereas in other cases the photo- 
cytes had secretory droplets in the cytoplasm 
and even a membrane. Some of the possible 
explanations for this phenomenon are: (1) 
there are two types of photocytes; (2) the two 
forms represent cells in different phases of a 
production-secretion cycle; or (3) they repre- 
sent a concentration of different organelles in 
different regions of a single cell. 

Thus, mature gametes, photocytes, plus 
the presence of secretory droplets and nu- 
merous mitochondria (Figs. 2, 3), suggest 
that luminescence can perist into adulthood in 
D. striata. 



Luminescence, Gonadal Maturity, 
and Behavior 



Stylommatophorans usually exhibit simulta- 
neous hermaphroditism or protandry (Tompa, 
1 984). In terms of gonadal maturation, oocytes 
usually start to differentiate first, but the sperm 
develop faster and, thus, are first to reach ma- 
turity (Runham & Hunter, 1970). 

In D. striata, we found that the large snails 
have large, well-developed gonads and ma- 
ture sperm (Figs. 1, 2), and are, therefore, 
adults. Small snails have undeveloped go- 
nads (Fig. 1), and are, thus, juveniles. Some- 
where along the continuum of snail sizes, 
sexual maturity is reached, but an external 
marker for sexual maturity is not yet known. 

Because luminescence in D. striata is not a 
juvenile-only luminescence (Haneda, 1981; 
Martoja & Bassot, 1970), as was previously 
thought, it is possible that it might play a role 
in mating behavior in D. striata. The presence 
of two types of adult snails, some with a "vis- 
ible" luminescent organ and some with a 
"non-visible" luminescent organ, and the 
commonplace nature of simultaneous her- 
maphroditism or protandry in stylommatopho- 
rans, is a stimulus for further research on 
analysis of communication by biolumines- 
cence in D. striata. As yet, the behavioral sig- 
nificance of the flash of D. striata remains 
enigmatic. 



LUMINESCENT ORGAN AND SEXUAL MATURITY 



19 



ACKNOWLEDGMENTS 

We thank the National Geographic Society 
for support for the field collection in Singa- 
pore, and Dr. A. D. Carlson for a critical read- 
ing of the manuscript. We also thank an anon- 
ymous reviewer of the manuscript for helpful 
comments. This reviewer obviously spent 
considerable time and energy to positively 
communicate ways in which the manuscript 
could be improved. We have learned much 
and are grateful. 



LITERATURE CITED 

BARR, R. A., 1926, Some observations on the 
pedal gland of Milax. Quarterly Journal of Micro- 
scopical Science, 70: 641-671 . 

BASSOT, J. M. & M. MARTOJA, 1968, Presence 
d'un organe lumineux transitoire chez le gas- 
teropode pulmone, Hemiplecta weinkauffiana 
(Crosse et Fischer). Comptes Rendus des Se- 
ances de l'Académie des Sciences, Paris, 266: 
105-1047. 

BOWDEN, B. J., 1950, Some observations on a 
luminescent freshwater limpet from New Zea- 
land. Biological Bulletin, 99 (3): 373-380. 

COPELAND, J. & M. M. DASTON, 1989, Biolumi- 
nescence and communication in the terrestrial 
snail Dyakia (Quantula) striata. Malacologia, 30: 
317-324. 

COPELAND, J. & A. GELPERIN, 1983, Feeding 
and a serotonergic interneuron activate an iden- 
tified autoactive salivary neuron in Umax maxi- 
mus. Comparative Biochemistry and Physiology, 
76: 21-30. 

COPELAND, J. & M. MANERI, 1984, Biolumines- 
cence and communication in the terrestrial snail 
Dyakia (Quantula) striata. Society for Neuro- 
science Abstracts, 10: 396. 

COUNSILMAN, J. J., D. LOH, S. Y. CHAN, W. H. 
TAN, J. COPELAND & M. MANERI, 1987, Fac- 
tors affecting the rate of flashing and loss of lu- 
minescence in an Asian land snail, Dyakia stri- 
ata. Veliger, 29: 394-399. 

EAKIN, R. M. & J. L. BRANDENBURGER, 1975, 
Retinal differences between light-tolerant and 
light-avoiding slugs (Mollusca: Pulmonata). Jour- 
nal of Ultrastructural Research, 53: 382-394. 

HANEDA, Y., 1963, Further Studies on a luminous 
land snail, Quantula striata, in Malaya. Science 
Report of the Yokusuka City Museum, 8:1-9. 



HANEDA, Y., 1973, Further studies on a luminous 
land snail, Quantula striata, in Malaya. Science 
Report of the Yokusuka City Museum, 8: 1-9. 

HANEDA, Y., 1981, Luminous activity of the land 
snail Quantula striata. Pp. 257-265, in м. deluca 
& w. MCELROY, eds.. Bioluminescence and 
chemiluminescence. Academic Press, New York. 

HEUSER, J. E. & T. S. REESE, 1974, Morphology 
of synaptic vesicle discharge and reformation at 
the frog neuromuscular junction. Pp. 59-77, in 
M. v. L. BENNET, ed., Synaptic transmission and 
neuronal interaction. Raven Press, New York. 

ISOBE, M., D. UYAKUL, T. GOTO & J. J. COUN- 
SILMAN, 1988, Dyakia bioluminescence — 1. 
Bioluminescence and fluorescence spectra of 
the land snail, D. striata. Japanese Journal of 
Cell Biology 25: 791-795. 

KATER, S., 1977, Calcium electroresponsiveness 
and its relationship to secretion in molluscan exo- 
crine gland cells. Pp. 195-214, in w. м. cowan & 
J. A. FERRENDELLi, eds.. Approaches to the cell bi- 
ology of neurons. Society for Neuroscience, Be- 
thesda, Maryland. 

MANERI, M., 1985, Bioluminescence and sexual 
maturity in the terrestrial snail Dyakia striata. MS 
thesis. University of Wisconsin-Milwaukee. 

MARTOJA, M. & BASSOT, J. M., 1970. Étude his- 
tologique du complexe glandulaire pedieux de 
Dyakia striata, Godwin et Austin, gasteropode 
pulmone données sur l'organe lumineux. Vie et 
Mil lieu, Serie A: Biologie Marine, XXI, Fase. 2-A: 
395-452. 

NICHOL, J. A. C, 1960, Histology of the light or- 
gans of Pholas dactylus Lamellibranchia). Jour- 
nal of the Marine Biology Association, United 
Kingdom, 39: 29-32. 

PARMENTIER, J. & A. BARNES, 1975, Obsen/a- 
tions produced by the Malayan gastropod Dyakia 
striata. Malayan Nature Journal, 28: 173-180. 

PORTER, K. & M. A. BONNEVILLE, 1968, Fine 
structure of cells and tissues. Lea & Febiger, 
Philadelphia. 

RUNHAM, N. & P. HUNTER, 1970. Terrestrial 
slugs. Hutchinson, London. 

TAUC, L., 1977, Transmitter release at cholinergic 
synapses. Pp. 64-78, in g. cottrell & p. usher- 
wood, eds.. Synapses. Academic Press, New 
York. 

TOMPA, A. S., 1984, Land snails (Stylommato- 

phora). Pp. 47-140, in A. TOMPA, N. VERDONK & J. 

VAN DER BiGGELAAR, eds.. The Mollusca, Volume 7. 
Academic Press, New York. 



Revised Ms. accepted 20 April 1 992 



MALACOLOGIA, 1993, 35(1): 21-41 

A POPULATION STUDY OF THE BIVALVE IDAS ARGENTEUS JEFFREYS, 1876, 

(BIVALVIA: MYTILIDAE) RECOVERED FROM A SUBMERGED WOOD BLOCK IN 

THE DEEP NORTH ATLANTIC OCEAN 

Harlan K. Dean 

Department of Invertebrates, Museum of Comparative Zoology, Harvard University, 
Cambridge, Massachusetts, 02138 U.S.A. 

ABSTRACT 

A large population of the wood-associated, deep-sea bivalve Idas argénteas was recovered 
from a wood block submerged for 1 1 years at 3,600 m depth at Deep Ocean Station 2 (DOS 2) 
in the western Atlantic south of New England. Acetate peels of the inner shell layer revealed a 
series of annual growth lines which were utilized to establish a relationship between shell length 
and age. Individuals recovered from wood panels also deployed at DOS 2 but submerged for 
much shorter periods were also examined using the acetate peel technique, and the number of 
growth lines generally coincided with the length of time spent on the bottom. Evidence for 
seasonality in the deep sea is reviewed, and the annual variation in the settlement of organic 
material from overlying photosynthetic layers is invoked as an important environmental cue to 
deterministic growth of the filter feeder /. argenteus. Analysis of a crystal size gradient in the 
region between successive growth lines in the inner shell layer lends support to gradual envi- 
ronmental change at DOS 2 and also to the Lutz-Rhoads (1980) model of annual shell depo- 
sition. Age-size frequency analysis revealed numerical dominance by the third and fourth year 
classes, perhaps due to what Roughgarden et al. (1985) characterized as "limit cycles." The /. 
argenteus living on the wood block functioned as protandric hermaphrodites, spending their first 
six years as males and the remainder of their existence as females. Increase in shell length of 
/. argenteus fits both the Gompertz and Power function growth models. The analysis of size- 
specific growth rates indicates that /. argenteus lacks the high growth rate displayed during the 
first year but shows a slower decrease in size-specific growth rates with age compared to 
shallow-water and freshwater bivalves. Specimens from the wood panels were larger than 
equal-aged individuals from the wood block, most likely due to a higher food quality and quantity 
on the wood panels. Idas argenteus is capable of colonizing patches of organic material in the 
deep sea probably a consequence of high reproductive potential and a planktotrophic larval 
stage. Whereas shallow-water opportunists are capable of a rapid increase in population size 
following settlement of a new site, /. argenteus can only increase population size upon reaching 
sexual maturity the year following settlement. 

Key words: deep-sea ecology, bivalve, opportunist, growth line analysis, protandry, population 
structure, size frequency analysis, growth rate, shell microstructure, seasonality, larval settle- 
ment. 



INTRODUCTION 

Many known deep-sea bivalves (vy^lth the 
exception of those living at the hydrothermal 
vents and sulfide/methane seeps) are small, 
with low metabolic and growth rates, and ap- 
parently require a long time to reach maturity 
(Turekian et al., 1975; Grassle, 1978; Smith & 
Minga, 1983; Grassle, 1986). Recolonization 
studies of sediment trays in the deep sea in- 
dicate low recruitment rates as well as low 
rates of population increase (Grassle, 1977; 
Levin & Smith, 1984). There is increasing ev- 
idence, however, that what have been de- 
scribed by Pearson & Rosenberg (1978) as 



"enrichment opportunists" occur in the deep 
sea and survive by specifically finding and ex- 
ploiting organically enriched sites (Grassle & 
Morse-Porteous, 1987; Smith & Hessler, 
1987; Desbruyères & Laubier, 1988). 

Turner (1973) was the first to describe 
deep-sea opportunistic species associated 
with organic material. Turner (1973, 1977, 
1981) found that wood placed on the deep- 
sea floor was rapidly colonized by pholad bi- 
valves belonging to the subfamily Xyloph- 
againae, a group of obligate deep-sea wood 
borers. Large numbers of these opportunistic 
borers rapidly colonize submerged wood, and 
probably reach sexual maturity rapidly — esti- 



21 



22 



DEAN 



mated by Turner (1973) to take as little as 
three months — and render the nutrients 
in cellulose accessible to other deep-sea 
species. Desbruyères et al. (1980, 1985) re- 
ported rapid colonization of organic aggre- 
gates and floes by the polychaete Ophryotro- 
cha sp., whereas Grassle & Morse-Porteous 
(1987) found Ophryotrocha sp. and Capitella 
spp. most abundant in those sediment trays 
containing decaying Sargassum. More re- 
cently, Desbruyères & Laubier (1988), work- 
ing in the deep Atlantic, reported a new genus 
and species of scale worm recovered from 
organically enriched substrates. The settle- 
ment of organic material in the deep sea 
appears to be a type of disturbance that pro- 
vides an important source of spatial hetero- 
geneity in what was previously viewed as a 
uniform homogeneous environment (Grassle 
& Morse-Porteous, 1987). 

In June 1986, a wood block was recovered 
from DOS 2 (38°18.4'N, 69°35.6'W) 350 km 
south of Cape Cod by the research vessel 
DSRV/ALVIN as part of Turner's ongoing 
study of deep-sea wood-boring pholads. This 
block was riddled with mostly abandoned 
pholad burrows within which lived a large 
number (7,872) of the wood-associated deep- 
sea bivalve Idas argenteus (family Mytilidae). 
Recovery of this material provided a unique 
opportunity to study several aspects of the life 
history and population biology of a bivalve in- 
habiting an organically enriched environment 
in the deep sea. 



MATERIALS AND METHODS 

Living specimens of /. argenteus (Figs. 1 , 2) 
were taken from a wood block (I.D. number 
N-1 7) approximately 30 cm on a side that had 
been placed at DOS 2 as part of a 12-block 
"wood island" in July 1975 and retrieved on 
28 June 1986. Each block was enclosed in a 
plastic mesh bag to hold together the crum- 
bling wood during recovery. Block N-1 7 was 
removed from the wood island using ALVIN's 
mechanical arm, placed in a vinyl-lined milk 
crate, and brought to the surface in ALVIN's 
collecting basket. 

Aboard ship, many specimens of /. argen- 
teus were immediately removed from the 
wood block and placed in 5% buffered forma- 
lin. The block was then broken into small 
pieces and also fixed in 5% buffered formalin. 
After fixation, all samples were washed and 
transferred to 95% ethanol. In the laboratory, 



both the wood block and the panels were dis- 
sected using a Stanley knife, and all speci- 
mens of /. argenteus visible through a lOx 
lens were removed from the wood chips. 

Specimens of /. argenteus were also recov- 
ered from nylon mesh-covered wood panels 
(57.6 X 14.5 X 2.3 cm) that had been ex- 
posed for periods of 1 1-47 months (Table 2) 
near the wood island. Once extracted from 
the sediment, the panels were placed in re- 
trieval boxes equipped with a locking top to 
prevent loss of material during their return to 
the surface (Turner, 1977). These wood pan- 
els were fixed while on the bottom with glut- 
araldehyde, which was released upon closure 
of the retrieval box lid, or on board ship with 
either 5% buffered formalin or 2% glutaralde- 
hyde. 

Length measurements of the shells repre- 
sent the maximum distance between the an- 
terior and posterior margin of the valves taken 
parallel to the ventral margin. All length mea- 
surements were made using a Wild M-8 dis- 
secting microscope equipped with an ocular 
micrometer (at 50 x each unit of measure 
was equal to 19.4 jxm). Direct length mea- 
surements were made of all wood block spec- 
imens > 2.72 mm in length and < 0.97 mm. 
Individuals between 0.97-2.72 mm in length 
(N = 6,196) were randomly subsampled and 
the size frequency distribution of this subsam- 
ple (N = 770) was adjusted to the total sam- 
ple size of 7,872 in order to construct the size 
frequency distribution of the entire wood block 
population. The U.S. National Marine Fisher- 
ies Normal Distribution Separator Program 
(NORMSEP) was used to divide the size fre- 
quency distribution into age classes based on 
the results of growth line analysis. 

Growth line studies were made of the inner 
shell layer of 102 specimens recovered from 
the wood block. Valves were removed from 
fixed individuals, air dried, and embedded in 
EPO-TEC 301 , a transparent epoxy. The em- 
bedded valves were filed down along the axis 
of maximum growth and the exposed surface 
polished with 240, 800 and 3200 grit polishing 
compounds. The polished cross sections 
were etched using 2% HCl (by volume) for 5 
to 8 minutes. Once dry, the etched surface 
was flooded with acetone and a sheet of ac- 
etate placed over the surface. Following 
evaporation of the acetone, the acetate sheet 
was peeled off, mounted in EPO-TEC 301, 
and growth lines in the inner shell layer ex- 
amined using light microscopy. 

Thirty nine individuals from the panels were 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



23 




FIGS. 1 , 2. Scanning electron micrographs of specimens of Idas argenteus recovered from the wood block. 
1 . Exterior of left valve showing dense periostracal hairs. 2. Inner surface of right valve. Scale bar = 1 .0 mm. 



analyzed in order to confirm the annual nature 
of the growth lines. Valves fronn the larger in- 
dividuals found on six wood panels were pol- 
ished and acetate peels made using the pro- 
cedures described above. 
Some polished and etched valve surfaces 



were also examined with the scanning elec- 
tron microscopy (SEM). The embedded 
valves were mounted on aluminum stubs with 
double-sided tape, coated with a 700 Â layer 
of gold-palladium, and viewed using an AMR- 
1000 electron microscope. The analysis of 



24 



DEAN 




FIG. 3. Camera lucida drawing of an acetate peel taken from the polished surface of the valve of Idas 
argenteus. Five growth lines within the inner shell layer are indicated. 



calcium carbonate crystal size was conducted 
using an enlargement of the SEM micrograph 
shown in Figure 5. Thirty-one equally spaced 
transects were drawn perpendicular to these 
two growth lines, and the length of the 
transect across each individual crystal was 
recorded. The relationship between these es- 
timates of crystal size and the distance of 
each crystal from the older of the two growth 
lines was analyzed using linear regression. 

During the removal of valves for growth line 
analysis the reproductive state of each spec- 
imen was noted using a dissecting micro- 
scope. Occasionally, gonadal smears were 
examined under a compound microscope to 
confirm the identification of their sexual state. 

Analysis of shell growth rates was carried 
out using the statistical package FISHPARM 
(Saila et al., 1988). Specific growth rate (the 
rate of growth divided by size, G) was esti- 
mated using the equation: 

G = (S2-Si)/Si, 

where Si = shell length at the beginning of 
time interval T, and Sg = shell length at the 
end of time interval T (Kaufmann, 1981). 



RESULTS 

The shell of /. argenteus is composed of 
three separate crystalline layers (Figs. 3, 4). 
The outer layer consists of irregular simple 
prisms (sensu Carter, 1980) approximately 12 
p.m long and 1.7 |лт in diameter, oriented 
roughly parallel to the shell surface (Fig. 4). 
This outer layer forms a series of closely 
spaced concentric lines on the exterior sur- 



face of the valve, but distinctive growth layers 
associated with these external lines were not 
apparent. 

The middle shell layer is composed of 
sheets of nacreous tablets varying from 0.4 to 
0.8 M.m in thickness. This layer is relatively 
thin in the umbonal region of the valve; it ex- 
pands, however, to make up much of the 
thickness of the shell at the valve edge (Fig. 
3). No growth lines were apparent in this 
sheet nacreous layer. 

The inner layer of shell has a fine, complex 
crossed lamellar microstructure (Figs. 4, 5). 
This layer is divided into a series of bands by 
fine lines running parallel to the shell growth 
axis (Fig. 3). The bands of shell material be- 
tween each pair of lines extend along the axis 
of growth, with each successive growth band 
extending somewhat further from the um- 
bonal region than its antecedent (Fig. 3). SEM 
examination revealed little that was remark- 
able about the crystalline microstructure of 
the inner shell layer in the vicinity of these 
lines (Fig. 5). These fine lines (hereafter re- 
ferred to as growth lines), present in the inner 
shell layer of /. argenteus, were used to de- 
termine the ages of these clams. 

Growth lines in the inner shell layer were 
counted in valves of known length to establish 
a relationship between size and age (Table 
1). The smallest specimens examined dis- 
played a single growth line, whereas the larg- 
est individual in the wood block population 
(7.15 mm in length) possessed nine growth 
lines in its inner shell layer. The number of 
fine lines in the inner shell layer of /. argen- 
teus increases in concert with increase in 
valve length. Although there is some size 



POPULATION STUDY OF A DEEP-SEA BIVALVE 

штат 



25 



I M L 



щтщщщшщщ 





FIGS. 4, 5. Scanning electron micrograph of a cross section of the shell of Idas argenteus from the region 
indicated by the arrow in Fig. 3. Arrows indicate five growth lines in the inner shell layer. 5. Scanning electron 
micrograph of the fine complex crossed lamellar inner shell layer of Idas argenteus. Arrows indicate two 
growth lines, ol = outer shell layer; ml = middle shell layer; il = inner shell layer; imi = innermost growth 
layer. Scale bars = 10 |xm. 



26 



DEAN 



TABLE 1. Results of the growth line analysis from acetate peels of sectioned valves of specimens 
recovered from wood block N-17. The size range of individuals encountered, as well as the number of 
specimens analyzed (N), is given for each age/growth line class. 



Number of growth lines 



Shell 


length (mm) 


Minimum 




Maximum 


0.91 




1.26 


1.35 




1.70 


1.44 




1.90 


1.91 




2.62 


2.40 




3.75 


3.04 




4.29 


4.08 




5.88 


5.44 




6.85 


7.15 




7.15 



Number of specimens 



6 
10 

9 
12 
15 
15 
24 
10 

1 



overlap, age classes based upon growth line 
number form distinct shell length size classes. 

Figure 6 includes the reconstructed size 
frequency distribution (solid line) of the popu- 
lation of /. argénteas taken from wood block 
N-17. Also included in this figure are the nine 
component normal distributions (dotted lines) 
generated by the normal distribution separa- 
tor program NORMSEP. This program fits 
normal curves to the size frequency data 
based upon the size range of each age class 
derived from growth line analysis (Table 1). 
The number of individuals in each age class 
(the area under each of the nine normal 
curves) and the mean size of each year class 
are also included in Figure 6. 

Growth line counts were also made of 
larger specimens recovered from wood pan- 
els submerged for periods of 1 1 to 47 months. 
This allowed the scrutiny of growth line pro- 
duction over much shorter periods of time 
than the eleven years of wood block submer- 
gence and was used to corroborate the inter- 
pretation of these fine lines as annual growth 
markers. Results indicate that the number of 
growth lines in /. argenteus is indeed congru- 
ent with a yearly deposition of shell layers in 
the inner shell (Table 2). Only specimens 
taken from a panel submerged for 35 months 
and a panel submerged for 47 months pos- 
sessed a number of growth lines other than 
would be predicted based upon the number of 
years submerged. In these two cases, there 
were fewer growth lines than expected, per- 
haps a consequence of a delay in the time of 
initial colonization by /. argenteus or of an in- 
creased death rate due to higher prédation by 
epifaunal organisms on the less protected 
wood panels (Williams & Turner, 1986). 

To determine the reproductive strategy of /. 



argenteus, 101 specimens of known age 
(based on the results of acetate peel analysis) 
were dissected and the reproductive state of 
the gonads recorded (Table 3). All members 
of the first year class examined were found to 
be sexually immature. Sexually ripe males 
were present in the second to seventh year 
classes whereas ripe females occurred in the 
sixth to eighth year classes. Specimens with 
unripe gonads were present over the entire 
size range of the clams analyzed. Four her- 
maphroditic individuals were encountered 
possessing both ripe ovaries and testes. In 
these four instances, the ovaries were well 
developed while the testes were quite small 
but still contained spermatozoa (confirmed 
with gonadal smear analysis). 



DISCUSSION 

Shell Fine Structure 

The shell fine structure of /. argenteus is 
similar to that reported in other members of 
the family Mytilidae (Taylor et al., 1969) and 
agrees with an earlier description (Carter et 
al., 1990) of a single specimen of /. argenteus 
(Yale Peabody Museum 9596) collected from 
2,900 m depth "off Marthas Vineyard." Carter 
et al. (1 990) reported that the simple prismatic 
outer shell layer of this species was calcific 
whereas the nacreous middle shell layer and 
inner fine complex crossed lamalla of the in- 
ner shell layer was composed of aragonitic 
crystals. The presence of a calcitic outer shell 
layer has been noted in several subfamilies of 
the Mytilidae, especially in mytilid species 
from cold water habitats. (Taylor et al., 1969; 
Carter, 1 980: fig. 5). These authors report that 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



27 



2001 




N 



1 


119 


0.93 


2 


1384 


1.32 


3 


2126 


1.75 


4 


1877 


2.23 


5 


1538 


2.85 


6 


608 


3.59 


7 


199 


4.60 


8 


18 


5.82 


9 


3 


6.87 



1.0 2.0 3.0 4.0 5.0 
Length (mm) 



8 

6.0 



9 



7.0 



FIG. 6. Size frequency analysis of the wood block population of Idas argénteas. Solid line is the size 
frequency derived from direct measurement of shell length (all specimens > 2.72 mm and < 0.97 mm) or 
derived from direct measurement of a random subsample (specimens > 0.97 mm and < 2.72 mm). Dashed 
lines are the age classes derived from the normal distribution separator program NORMSEP based upon the 
size ranges of each growth line class. N = the number of individuals in each age/growth line class based 
on the normal curves (dotted lines) dehved from NORMSEP. X = mean valve length of each age/growth line 
class. 



tropical or warm-v\/ater mytilids generally pos- 
sess shells composed entirely of aragonite. 
Idas argénteas, living in the cold waters of 
3,600 m depth, has a prismatic, outer calcitic 
layer similar to that in other mytilids from 
colder regions. 

The greater width of the innermost band of 
fine complex crossed lamella in the aragonltic 
inner shell layer (Fig. 4, imi) tends to support 
the general description of annual growth line 
deposition by Lutz & Rhoads (1980). This 
model postulates that an extended period of 
shell deposition is followed by a period of dis- 
solution of a portion of this newly laid down 



shell material. The Lutz-Rhoads hypothesis 
suggests that during extended shell closure a 
buildup of organic acids due to anaerobic 
conditions leads to a reduction in pH of the 
extrapallial and mantle fluids to such levels 
that calcium carbonate crystals are dissolved. 
A concentration of less soluble organic matrix 
would occur in the 'egion between two depo- 
sitional periods resulting in what would then 
be recognized as a growth line. 

The innermost growth band of /. argénteas, 
which is wider relative to those laid down pre- 
viously, may be the current year's deposit of 
calcium carbonate crystals produced during a 



28 



DEAN 



TABLE 2. Results of the growth line analysis from acetate peel of sectioned valves of specimens 
recovered from the panels. The valve length and the number of growth lines in the inner shell layer is 
given for the largest individuals on the wood panel successfully analyzed. The number of months of 
panel submergence and the number of individuals (N) of Idas argenteus recovered from each panel are 
also included. *, See text. 



Length (mm) 


Number of lines 


Length (mm) 


Number of 1 


nes 


Panel N-37 


11 months 


N 


= 6 


Panel N-76 


35 months N 


= 1577 


1.05* 


1 






2.16 
2.23 


3 
3 




Panel N-39 


23 months 


N 


= 221 


2.23 
2.25 


3 
3 




1.35 


1 






2.27 


3 




1.47 


1 






2.43 


3 




1.51 


1 






2.78 


3 




2.00 


2 












Panel N-30 


24 months 


N 


= 129 


Panel N-78 
2.18 


35 months N 
3 


= 363 


1.47 


1 






2.33 


2 




1.82 


2 






2.47 
2.61 
2.66 
2.86 
2.90 


3 
3 
2 
3 
3 




Panel N-93 


25 months 


N 


= 71 








1.29 


1 












1.45* 


2 












1.51* 


2 






Panel N-55 


47 months N 


= 2068 


Panel N-82 


35 months 


N 


= 79 


2.47 
2.48 


3 
3 




1.59* 


2 






2.57 


3 




1.82 


2 






2.58 


2 




1.84 


2 






2.67 
2.72 


3 
3 




Panel N-83 


35 months 


N 


= 424 


2.76 


3 




1.90 


2 






2.98 


3 




1.90 


2 






3.10 


3 




2.18 


3 






3.68 


3 





TABLE 3. Results of the gonadal analysis of specimens prepared for growth line analysis. The number of 
individuals examined (N) and their reproductive state are presented for each age/growth line class. 



Number of 
Lines 



Number of 
Specimens 



Male 



Female 



Hermaphrodite 



Unripe 



5 


— 


— 


6 


2 


— 


6 


2 


— 


12 


10 


— 


14 


8 


2 


14 


7 


1 


23 


1 


16 


9 


— 


6 


1 


— 


— 



period of growth prior to collection of the 
block. This band of crystals \NOu\a have been 
partially eroded during a subsequent non- 
growth period to a width similar to the previ- 



ously formed growth bands seen in the inner 
shell layer. This scenario is strongly sup- 
ported by examination of the crystal size gra- 
dient in this innermost growth band (dis- 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



29 



cussed below relative to deterministic growth 
in the deep sea), which indicates the occur- 
rence of a period of shell crystal deposition 
extending beyond that seen in previously laid 
down growth bands. An expected concentra- 
tion of organic material at each growth line 
was not evident upon SEM examination of the 
shell of /. argénteas (Fig. 5), and this aspect 
of the Lutz-Rhoads hypothesis of shell growth 
is not supported by these results. 

Growth Lines 

Growth lines, such as those seen within the 
inner shell layer of /. argénteas, have been 
interpreted as being produced annually in 
many shallow-water bivalves (Rhoads & 
Panella, 1970; Lutz & Rhoads, 1980; Fritz & 
Lutz, 1986). This has been documented in 
mark-and-recovery experiments with Merce- 
naria mercenaria (Linné) (Panella & Мас- 
Clintock, 1968), Spisula solidissima Dillwyn 
(Jones et al., 1 978), Anadara granosa (Linné) 
(Richardson, 1987), Муа arenaria Linné 
(MacDonald & Thomas, 1980), Mytilus edulis 
(Linné) (Lutz, 1976), and Corbicula fluminea 
(Müller) (Fritz & Lutz, 1986). Further support 
for yearly deposition of growth lines has been 
given by Jones et al. (1983), who analyzed 
annual cycles in oxygen isotopic variations in 
the shell growth increments of Spisula solidis- 
sima. 

Whereas internal growth lines within the in- 
ner shell layers have been reported from 
deep-water bivalves, it has not been demon- 
strated that these growth lines represent 
yearly depositional events. Work with Yoldia 
thraciaeformis from a submarine canyon off 
the southeastern Grand Banks of Newfound- 
land at 895-1 ,500 m by Hutchings & Haed- 
rich (1984) and Gilkinson et al. (1986) noted 
the presence of distinctive growth lines, but 
they could only assume that they were laid 
down annually. The data presented here for /. 
argenteus provide the first strong evidence for 
annual growth patterns in deep-sea bivalves. 

Wood block N-17 provided a large number 
(7,872) of specimens of /. argenteus, thus al- 
lowing growth line analysis over a wide range 
of shell lengths (Table 1 ). The results of these 
analyses present a very clear picture of a di- 
rect relationship between the number of 
growth lines and shell length, as well as an 
estimate of the size range of individuals in 
each age class based upon growth line num- 
ber. The largest individual in the population 
exhibited nine growth lines, indicating that it 



was collected while in its ninth year, two years 
less than the period of submergence of the 
wood block. Idas argenteus may not have col- 
onized the wood block until some time after 
the deep-sea wood boring pholads had colo- 
nized and begun the conversion of the wood 
block to more accessible forms of organic ma- 
terial (Turner, 1977, 1981). Additionally, large 
numbers of /. argenteus would not be avail- 
able for settlement until a pioneering colony of 
adults had become established on the iso- 
lated wood island. Finally, given the low num- 
ber of individuals in the older year classes, 
any individuals that could have colonized the 
wood block immediately after submergence 
would probably have had little chance of sur- 
vival to their tenth or eleventh year due to high 
annual mortality rates. The absence of a tenth 
and eleventh age class is thus not surprising, 
and a population age structure of nine year 
classes strongly supports the interpretation of 
the growth lines as representative of some 
annual cycle in shell growth. 

More telling evidence of the annual nature 
of the growth lines in /. argenteus are the re- 
sults of the analyses of the largest individuals 
recovered from wood panels submerged 
close to the wood block but for much shorter 
periods of time. One would expect rapid col- 
onization of these panels by both the pholads 
and /. argenteus soon after emplacement due 
to the large numbers of larvae emanating 
from the previously established wood island, 
and there should be close agreement be- 
tween the number of growth lines in the shells 
of larger specimens of /. argenteus and the 
number of years submerged. The maximum 
number of growth lines observed in speci- 
mens from seven of the nine panels exam- 
ined did indeed parallel the number of years 
the panel was on the bottom (Table 2). The 
larger individuals from panel N-82, which was 
submerged for 35 months, possessed only 
two growth lines, whereas those from panel 
N-55, which was submerged for 47 months, 
exhibited a maximum of only three growth 
lines. These two exceptions may perhaps be 
the result of susceptibility of /. argenteus to 
prédation by epifaunal organisms on the 
wood panels (Williams & Turner, 1986) either 
prior to the exposure of the pholad tunnels 
upon breakdown of the panel surface or per- 
haps following the eventual crumbling and 
disintegration of the panel. Most important is 
that there is generally a one-to-one relation- 
ship between the number of growth lines in 
the inner shell layer and the number of years 



30 



DEAN 



of submergence of the wood, thus providing 
powerful supporting evidence for annual 
growth periods in /. argénteos. 

Deterministic Shell Growth in the Deep Sea 

Seasonal variation as well as annual 
spawning cycles have been implicated in 
shell layer deposition by bivalve mollusks. For 
many shallow-water temperate species, 
growth lines appear to reflect periods of little 
or no shell growth during the winter when 
temperatures are at a minimum (Panella & 
MacClintock, 1968; Williamson & Kendall, 
1981 ; Jones et al., 1983; Fritz & Lutz, 1986). 
Richardson (1987) suggested that growth 
lines in the shells of the subtropical Anadara 
granosa may reflect exposure to low salinity 
waters during the annual intermonsoon pe- 
riod. Both Turekian et al. (1982) and Trut- 
schler & Samtleben (1988) noted that the 
growth lines in Árctica islándica Linné and As- 
tarte elliptica (Brown) were produced coinci- 
dent with seasonal minima in their food sup- 
ply and may simply be a reflection of slow 
growth due to nutritional deficiency. Cessa- 
tion of shell growth during spawning periods 
when available energy is channelled toward 
the production of sperm and eggs may also 
result in growth lines (Pannella & Mac- 
Clintock, 1968; Thompson et al., 1980; Gal- 
lucci & Gallucci, 1982). 

In the deep-sea environment, both temper- 
ature and salinity change very little (Sanders 
et al., 1965; Mantyla & Reid, 1983; Grassle & 
Morse-Porteous, 1987) and cannot be in- 
voked to explain annual shell growth events. 
In the only previous studies of growth lines in 
a deep-sea bivalve, Hutchings & Haedrich 
(1984) and Gilkinson et al. (1986) assumed 
that Yoldia thraciaeformis formed these lines 
either in response to seasonal fluctuations in 
food supply or as a "marker" of the reproduc- 
tive cycle (Gilkinson et al., 1986). These two 
factors may also provide an explanation for 
seasonal shell growth by /. argénteas. 

The specimens of /. argenteus observed in 
this study were apparently filtering suspended 
material from the water column. Many speci- 
mens, especially those taken from the wood 
panels, were observed with ingested material 
within the stomach and in the posterior por- 
tion of the intestine. SEM study revealed that 
the ciliation patterns of the gill filaments with 
long latero-frontal cilia, are typical of those 
seen in other filter feeding bivalves (Fiala- 
Métivioni et al., 1986). There were also sub- 



stantial amounts of what are presumed to be 
food particles on the frontal cilia of the gill 
surface and in the ventral food groove similar 
to that seen in other bivalves known to be 
actively engaged in filter feeding (Foster- 
Smith, 1975). Based on these observations, it 
is believed that /. argenteus is filtering sus- 
pended material either drifting down from the 
overlying waters or derived from the actions 
of the wood-boring pholads and other organ- 
isms associated with the wood island. 

Recently, specimens of Idas washingtonius 
(Bernard, 1978) with symbiotic bacteria in 
their gill filaments were reported from the 
deep Pacific Ocean attached to the bones of 
a whale carcass (Smith et al., 1989). These 
authors suggested that /. washingtonius may 
be augmenting its nutrient intake by sulfate 
reduction in a manner similar to that de- 
scribed by Felbeck & Somero (1982) and 
Grassle (1986) for several deep-sea vent 
species. The relative importance of such a 
chemoautotrophic food source to the total en- 
ergy budget of these deep-sea bivalves and 
to that of shallow-water bivalves known to 
possess the enzymes necessary for sulfate 
reduction is unknown (Somero et al., 1983). If 
such a symbiotic relationship does exist for /. 
argenteus, it could perhaps explain the large 
number of individuals (7,872) on a single 
wood block. Regardless of any possible con- 
tribution by sulfate reduction to the energy 
budget of /. argenteus, any appreciable en- 
ergy intake gained through suspension feed- 
ing could impart a seasonal component to its 
overall energy budget. 

There is growing evidence for appreciable 
seasonal variability in the deep-sea environ- 
ment (see Tyler, 1988, for a review). Perhaps 
most cogent to this discussion is evidence of 
a rapid transport of organic matter from the 
surface waters resulting in annual pulses in 
food supply to the deep-sea benthos. Turner 
(1973) and Wolff (1979) first called attention 
to a seasonal influx of plant remains to the 
deep sea, and sediment trap studies have in- 
dicated that particulate material settling on 
the bottom at depth is coupled to the seasonal 
plankton blooms in the overlying surface wa- 
ters (Honjo, 1980; Deuser et al., 1981; Ittek- 
kot et al., 1984; Matsueda et al., 1986). Pho- 
tographic records and direct sampling have 
recorded the settlement of large amounts of 
phytodetritus on the bottom shortly after phy- 
toplankton blooms in the surface waters (Bil- 
lett et al., 1983; Lampitt, 1985; Riemann, 
1989). Several studies have documented 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



31 



what is usually a rapid response by deep-sea 
benthic communities to these pulses of food 
material (Turner, 1973, 1977, 1981; Gooday, 
1988; Gooday & Lambshead, 1989; Graf, 
1989; Lambshead & Gooday, 1990; Gooday 
&Turley, 1990). 

The seasonal phytoplankton bloom in the 
northwestern Atlantic occurs from November 
to April (Menzel & Ryther, 1961), whereas 
sediment trap studies conducted southeast of 
Bermuda indicate that the highest influx of or- 
ganic material reached 3,200 meters from 
January to May or June (Deuser et al., 1 981 ). 
Idas argenteus is most likely exposed to 
greatest food supplies from January to June 
as a result of the rapid settlement of in- 
creased amounts of organic material derived 
from photosynthetic activities occurring in the 
surface waters. 

The availability of an enriched food supply 
in the deep sea may also extend beyond the 
time of high productivity in the surface waters 
due to both the fall phytoplankton bloom and 
the intermittent resuspension of previously 
settled particulate matter similar to that docu- 
mented at the HEBBLE site by Lampitt (1985) 
and recorded at DOS 2 by Rowe & Gardner 
(1979). Bottom currents are capable of creat- 
ing a nepheloid (cloudy water containing sus- 
pended solids) layer close to the bottom with 
a higher suspended load than the overlying 
waters (Jumars & Gallagher, 1982). Temporal 
variation in these deep-sea currents has been 
well documented (Dickson et al., 1982; 
Grassle & Morse-Porteous, 1987, for the 
DOS 2 sample site; Csanady et al., 1988), as 
have abyssal storms associated with the Gulf 
Stream Current (Hollister & McCave, 1984). 
These deep-sea currents are of magnitudes 
capable of resuspending particulate matter, 
allowing deep-sea suspension feeders an ex- 
tended period of increased food availability 
perhaps greater than that indicated by sedi- 
ment trap studies conducted well above the 
bottom. Such resuspended material, which 
would enhch the near-bottom nepheloid layer, 
as well as the particulate material settling 
from the overlying surface waters, could re- 
sult in a seasonal variation in food supply to 
such deep-sea benthic organisms as /. argen- 
teus. 

The presence of annual growth lines in /. 
argenteus could also be the result of seasonal 
spawning events. The presence of small 
numbers of first-year clams on the wood block 
indicates that some spawning and settlement 
must have occurred previous to the collection 



date of June 28th. Settlement must occur at 
least through September because there was 
a large number of sexually mature individuals 
on the wood block and a large number of very 
small, presumably recently settled clams on 
the panels recovered between late July to 
early September. Inspection of the 39 larger 
specimens taken from the panels disclosed 
that only one individual (recovered in late 
July) possessed a ripened gonad; the other 
39 specimens were unripe. These observa- 
tions indicate that spawning of /. argenteus 
may perhaps be completed by late July, at 
least in the wood panel populations. If shell 
growth in /. argenteus does cease during an 
annual spawning season or at least during a 
season of maximum spawning (Rokop, 
1 974), then the growth lines visible in the shell 
could be a reflection of a spawning period 
rather than a cycle of food availability. 

The pattern of crystal deposition at a 
growth line has been found to differ between 
a growth line associated with spawning and 
one attributed to seasonal change in the en- 
vironment (Kennish, 1980). Lutz (1976) and 
Lutz & Rhoads (1978, 1980) have character- 
ized the microstructure of spawning breaks 
in Geukensia demissa (Dillwyn) and Mytilus 
edulis as consisting of a series of normal 
width nacreous crystal tablets that are inter- 
rupted abruptly by a growth line break. This 
growth break is succeeded by deposition of a 
series of thin crystals laid down during a pe- 
riod of reproductive stress followed by a re- 
turn to normal width crystals once spawning 
is completed. Annual growth lines associated 
with variation in an environmental factor, such 
as water temperature, are associated with 
gradual, rather than abrupt, change in crystal 
deposition (Wada, 1961 ; Kennish, 1980). Lutz 

6 Rhoads (1980), for example, described reg- 
ular hexagonal nacreous tablets in the inner 
shell layer of G. demissa that gradually be- 
came smaller and less regular as water tem- 
perature declined. 

The shell microstructure of /. argenteus is 
similar to that noted in response to long-term 
seasonal changes by shallow-water bivalves 
(Lutz & Rhoads, 1980; Kennish, 1980). Figure 

7 shows the running average (N = 3) of crys- 
tal size measured as crystal overlap along 31 
transects drawn perpendicular to the two 
growth lines shown in Figure 5. The crystals 
gradually increase in size along these 
transects in the direction of growth away from 
a growth line (upward in Figs. 4 and 7). Ad- 
ditionally, a linear regression of crystal size 



32 



DEAN 



20 1 



Growth Line 




Growth Line 



S 






OX) 

с 
о 



с 

ее 

Q/3 



3.0 3.5 4.0 4.5 5.0 5.5 6.0 
Crystal size (1 unit = 0.96 дт) 

FIG. 7. Running average (N = 3) of the length of crystal overlap in the fine complex crossed lamellar inner 
shell layer of Idas argenteus along transects drawn across the two growth lines in Fig. 5. 



for the region between these two growth lines 
with distance from the older (lower) growth 
line was found to be highly statistically signif- 
icant (p < 0.0000). Based on these observa- 
tions, it seems that following the establish- 
ment of an annual growth line small crystals 
are deposited, with crystal size becoming in- 
creasingly larger as shell growth progresses. 
Based on the analysis of crystal size in the 
most recently deposited growth band (the 
innermost band adjacent to the mantle), the 
peaks in crystal size approaching each 
growth line in Figure 5 are thought to repre- 
sent true maxima. Bands of shell material 
are much narrower between successive 
growth lines, presumably due to dissolution of 
a portion of these older bands following their 
seasonal deposition, as postulated in the 
Lutz-Rhoads (1980) hypothesis. The newly 
deposited layer of crystals in the innermost 
layer has not yet been subjected to the ero- 
sion thought to occur at the mantle-shell in- 
terface during extended periods of shell clo- 



sure between growth periods. The crystals in 
this band have a similar size distribution to 
those found between the growth lines; how- 
ever, the right tail of the curve, indicating de- 
creasing crystal size following a seasonal 
maximum, is more extensive. As mentioned 
above, variation in crystal size deposition by 
shallow-water bivalves has been correlated 
with environmental conditions, with crystal 
size being reduced in times of stress and re- 
duced growth (Wada, 1961; Kennish, 1980; 
Lutz & Rhoads, 1 980). If crystal size gradients 
in the shell layers of /. argenteus reflect sea- 
sonal trends in relative environmental condi- 
tions and coincident growth, then it is appar- 
ent that some sort of seasonal optimum had 
occurred prior to retrieval of the wood block. 
The shell microstructure in the inner shell 
layer of /. argenteus indicates shell deposition 
in response to a seasonal gradual change in 
the environment. As previously noted, the 
most apparent environmental variable capa- 
ble of imposing this type of effect upon shell 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



33 



growth at DOS 2 is food availability. The grad- 
ual increase in crystal size deposition follow- 
ing production of a growth line may reflect 
increased food supply due to submergence of 
organic material produced in the photic zone 
during the spring phytoplankton bloom. The 
reduction in crystal size following a seasonal 
maximum (seen best in the innermost growth 
band) may reflect a decreased food availabil- 
ity later in the growth period. 

Because food is a factor in the regulation of 
gametogenesis (Giese & Pearse, 1974), it is 
probable that there is a coupling of food avail- 
ability with the spawning period as well as 
with the production of shell growth lines in the 
deep sea. The peak in crystal size between 
successive growth lines noted in the inner 
shell layers could reflect a shift from the chan- 
neling of available energy to the production of 
the metabolically expensive organic matrix 
(Palmer, 1983) necessary for shell growth to 
the production of gametes. To attribute the 
production of growth lines in the shell of /. 
argenteus entirely to deviations in food supply 
would be to neglect the metabolic stress of 
reproduction. Variation in food supply and the 
channeling of available energy to reproduc- 
tive processes is most likely an interactive re- 
lationship, and presumably both would affect 
the shell growth pattern of /. argenteus. 

Population Size Frequency 

As may be seen in Figure 6, the wood block 
population is numerically dominated by the 
third and fourth year classes. This size fre- 
quency distribution is believed to be a true 
representation of the wood block population 
rather than a sampling artifact. Although 
some individuals may have been washed off 
the block during retrieval, it is doubtful that 
such loss would occur preferentially to the 
smallest individuals in the population, that is 
that 1.3 mm specimens would be preferen- 
tially dislodged from the wood block relative to 
1 .75 mm specimens. The very low number of 
newly settled, first-year individuals suggests 
that retrieval of the wood block occurred prior 
to the period of greatest larval settlement. 
Many of the individuals in the block had hpe 
gonads and were about ready to spawn at the 
time of retrieval in late June. The abundance 
of very small, newly settled young on panels 
recovered in late August and September sug- 
gests that the major settlement of larvae oc- 
curs some time in late summer and that the 



dearth of first-year individuals is not the result 
of sampling. 

Numerical dominance by older age/size 
classes is not unusual for populations of ma- 
rine organisms (Gaines & Roughgarden, 
1985; Hughes, 1985, 1990; Roughgarden et 
al., 1985; Breen et al., 1991) and has been 
reported for several deep-sea invertebrate 
populations (Allen & Sanders, 1973; Rex et 
al., 1979; Tyler & Pain, 1982). This type of 
age-size frequency distribution was also re- 
ported for the deep-sea bivalves Nuculana 
pernula and Yoldia thraciaeformis by Hutch- 
ings & Haedrich (1984). These authors sug- 
gested that intense prédation by boring gas- 
tropods and benthic fish selects for fast 
growing individuals that quickly reach a size 
refuge from predators. This explanation, how- 
ever, does not address the predominance of 
older age classes (five to ten years based on 
external or internal shell growth lines) in their 
collections. 

Roughgarden et al. (1985) and Gaines & 
Roughgarden (1985) have postulated that 
populations limited by habitat space and hav- 
ing high, density-independent larval settle- 
ment rates would exhibit what they termed 
"limit cycles." In this model, a wave of numer- 
ically dominant year classes moves through 
the population with time, appropriating much 
of the available habitat. In the case of /. argen- 
teus, the third and fourth year classes may 
inhabit many of the life-sustainable sites on 
the wood block, thus preventing the success- 
ful recruitment of younger age classes. As 
these dominant age classes move through 
the population and become less numerous 
due to density-dependent mortality, a larger 
amount of suitable habitat becomes available 
for successful larval settlement, leading to the 
eventual establishment of another generation 
of numerically dominant age classes. 

Reproductive Strategy 

Analysis of gonadal development (Table 3) 
indicates that the /. argenteus in the wood- 
block population at DOS 2 are protandric her- 
maphrodites. In the four year classes follow- 
ing the first year of sexual immaturity, those 
individuals observed with ripe gonads were 
exclusively males. Females occurred in the 
fifth and sixth year classes, but the majority of 
sexually ripe individuals in these age classes 
were males. With a single exception, all indi- 
viduals in the seventh year class and older 
were females. It appears that members of the 



34 



DEAN 



wood block /. argenteus population spend 
their first five or six years as males and sub- 
sequent years as female. The environment 
has been shown to be a major determinant of 
the sexual strategy of an opportunist such as 
/. argenteus (Charnov & Bull, 1977), and 
protandry would not necessarily be the opti- 
mum strategy in all environments. In a newly 
colonized habitat where there are no preex- 
isting females, it would be expected that 
some of the first sexually mature individuals 
of /. argenteus would be female. 

According to the size-advantage model of 
Ghiselin (1969), protandric mollusks gener- 
ally have a very patchy distribution with only 
limited adult mobility. These generalizations 
seem true of /. argenteus, which is character- 
ized as living associated with sunken wood 
(Dell, 1987; Waren, 1991) and is nonmotile as 
an adult. Males living in such small, isolated 
communities are thought to have limited op- 
portunity for successful mating because the 
restrictive factor is the number of eggs pro- 
duced by the females of the population 
(Wright, 1988). Under such conditions, there 
would be little gained by producing large 
amounts of sperm, and there would be no re- 
productive advantage to being a large male. 
There is usually a direct relationship between 
female fecundity and female size in the Mol- 
lusca (Hoagland, 1978). Idas argenteus may 
be viewed as maximizing its reproductive suc- 
cess by being male when small and switching 
its sex later in life when its larger size would 
maximize its output of eggs. 

Growth Rates 

Estimates of annual growth in /. argenteus 
were derived from the mean valve lengths of 
the nine age classes shown in Figure 6. The 
change in length from one year class to the 
next was divided by the size at the beginning 
of the growth period, resulting in a size-spe- 
cific growth rate that could be compared with 
similarly derived growth rates from other bi- 
valves much different in size. The assumption 
is made that variations in growth rate due to 
year-to-year environmental variability are 
negligible and that each individual follows the 
same schedule of growth during its lifetime. 
As has been noted (McNew & Summerfelt, 
1978; Kaufmann, 1981), the use of the mean 
length for each year class tends to dampen 
any yearly variations in growth, making this a 
resilient method for the analysis of the growth 
strategy of a species. 



The resultant annual size-specific growth 
rates for /. argenteus were found to change 
little over the eight growth intervals, exhibiting 
only a slight downward trend with increasing 
age (Fig. 8, solid line). This growth pattern 
exhibited a statistically highly significant fit with 
the Gompertz (R^ = .998) and Power curve 
(R^ = .995) growth models, whereas the Ex- 
ponential (R^ = .914), Logistic (R^ = .926) 
and Von Bertelanffy growth models (R^ = 
.800) fit less effectively. Both the Gompertz 
and Power growth models include a reduction 
in growth rate with age, but the former as- 
sumes asymptotic growth to a size maximum 
and the latter is an indeterminant growth 
model. Due to the low number of individuals 
and greater standard deviations of the older 
age classes commonly encountered in size 
frequency distributions (MacDonald & Pitcher, 
1979; Gage, 1985), it is not possible to deter- 
mine whether the growth of /. argenteus is 
determinate or indeterminate from these data. 

Also included in Figure 8 are the size-spe- 
cific growth rates derived from previously 
published age-length data for two freshwater 
species {Lampsilis radiata and Anadonta 
grandis from f\/lcCuaig & Green, 1983) and 
three shallow-water marine species {Cerasto- 
derma edule and Modiolus modiolus from 
Seed & Brown, 1978, and Spisula solidissima 
from Jones et al., 1978). The size-specific 
growth pattern of /. argenteus differs greatly 
from these bivalves, which all exhibit elevated 
growth rates in their first year followed by a 
precipitous drop in growth by the second 
year. By the third or fourth year, the size-spe- 
cific growth rates of all five of these freshwa- 
ter and shallow-water species are lower than 
those of /. argenteus. Only M. modiolus (the 
only other member of the family Mytilidae in 
Figure 8) approached the rate of growth ex- 
hibited by /. argenteus in the older age 
classes. The deep-water bivalve /. argenteus 
lacks the rapid growth exhibited early in life by 
the shallow-water marine and freshwater spe- 
cies but experiences a slower reduction in 
growth with increasing age. 

It is difficult to make comparisons of the 
growth rates of /. argenteus with other deep- 
sea bivalves not associated with the vents 
and seeps as so few such studies have been 
conducted. Early growth estimates were car- 
ried out on Tindaria callistiformis collected 
from 3,800 m depth in the North Atlantic by 
Turekian et al. (1975). These authors em- 
ployed radio-chemical dating techniques to 
establish a life span of approximately 100 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



35 






о 
и 

о 



ел 

О) 



31 



2- 



1- 







.-о---- 




Idas argenteus 
Lampsilis radiata 
Anodonta grandis 
Cerastoderma edule 
Modiolis modiolis 
Spisula solidissima 



- - . ; J*- -.::-- ^m :: •• 



-" г 



T ' 1 ' г 



^ 



;;;;;;;з.. 



012345678 
Annual Growth Period 

FIG. 8. Size specific growth rates of Idas argenteus (solid line) and five species of marine sfiallow-water and 
freshwater bivalves (dotted lines). Lampsilis radiata Gmelin and Anodonta grandis Say from data in McCuaig 
& Green (1983); Cerastoderma edule (Linné) and Modiolus modiolus (Linné) from data in Seed & Brown 
(1978); Spisula solidissima Dillwyn from data in Jones et al. (1978). 



years and a resultant very slow growth rate of 
0.084 mnn/year. Unfortunately, the variance in 
their data (s.d. = 38 years) plus the use of 
external rather than internal growth lines as 
annual markers (see Lutz & Rhoads, 1980) 
makes their estimates of longevity and growth 
rate highly questionable. 

Hutchings & Haedrich (1984) included age 
determinations based on internal growth lines 
for Yoldia thraciaeformis collected 895-1 ,500 
m deep in the northwestern Atlantic Ocean, 
making it possible to derive size specific 
growth rates from their data. The size-specific 
grov\/th rate of four- to eight-year-old speci- 
mens of Y. thraciaeformis ranged from 0.07 to 
0.18. These growth rates are comparable to 
those of the similarly aged fresh and shallow- 
water species included in Figure 8 but are 
lower than those for specimens of /. argen- 
teus of comparable age from the wood block 
population. 



Rhoads et al. (1 982) carried out in situ mea- 
surements of growth for specimens of the 
large mussel, Bathymodiolus thermophilus 
Kenk & Wilson, 1 985, from the Galapagos Rift, 
and size-specific growth rates were generated 
using estimated values from their figure 4. 
Comparisons were made between individuals 
collected from a densely populated area and 
from a less densely populated region periph- 
eral to the mussel beds. For two specimens 
from the dense mussel bed, estimated to be 
five years old based on growth lines, the size- 
specific growth rates were 0.27 and 0.29, 
whereas a specimen estimated to be eight 
years old had a specific growth rate of 0.14. 
Eight- to fourteen-year-old specimens of B. 
thermophilus taken from the less densely pop- 
ulated peripheral region had size specific 
growth rates ranging from 0.04-0.15. Lutz et 
al. (1985, 1988) have indicated that this cor- 
relation between growth rates and proximity to 



36 



DEAN 



the hydrothermal vents are most likely the con- 
sequence of an elevated food supply. 

The size-specific growth rates for the mus- 
sel bed specimens of the Galapagos Rift are 
comparable with, while those specimens from 
the periphery of the mussel bed are lower 
than, those of /. argénteas taken from the 
wood block at DOS 2. Apparently, these high 
size-specific growth rates for /. argénteas are 
the consequence of the organic enrichment of 
the region surrounding the wood island due to 
the actions of the wood-boring pholads 
(Turner, 1973, 1977, 1981). 

The analysis of specimens from the panels 
also presents evidence that food availability 
may be a major determinant of growth for /. 
argénteas. Included in Table 2 are the lengths 
of specimens with ages determined by growth 
line analysis, and it is apparent that these 
clams are larger than their age cohorts grow- 
ing on the block. Those specimens with shell 
lengths that do not exceed the range of the 
normal curve (and thus fall within the size 
range) for their age class in the wood block 
population have been marked with an asterisk 
in Table 2. Growth of /. argénteas is appar- 
ently more rapid in specimens inhabiting the 
panels than in specimens living on the block. 

The major difference between the wood 
panels and the wood block was that the wood 
panels contained large numbers of pholads 
that were providing copious supplies of fecal 
material to the organisms on and around the 
panels (Turner, 1981). The posterior intes- 
tines of the majority of specimens examined 
from the wood panels were filled with yellow- 
ish fecal material, in contrast to the speci- 
mens from the wood block, which usually 
had little or no visible material in their guts. 
Additionally, after eleven years of submer- 
gence and processing by benthic organisms, 
the organic material derived from the wood 
block was probably of much lower quality 
than that of the younger (one to four years) 
wood panels. Alongi (1992), in his study of 
deep-sea benthic communities in the west- 
ern South Pacific, found that much of the 
wood and plant material encountered was 
well aged, with C:N ratios exceeding 300:1 
(as compared to 18:1 for fresh algal mate- 
rial), indicating low nutritional value. Food 
therefore seemed to be more abundant on 
the panels and may have been of higher qual- 
ity, resulting in higher growth rates and indi- 
cating that food availability is a limiting factor 
to the growth of /. argénteas in the deep 
sea. 



Opportunists in the Deep Sea 

Two life history traits that give opportunistic 
species an ability to colonize under-exploited 
areas of suitable habitat are a high dispersive 
capability and a facility to rapidly increase pop- 
ulation size (Turner, 1973, Grassle & Grassle, 
1974). These traits allow long distance move- 
ment by pioneering individuals and the ability 
to maximize the exploitation of that resource. 
The results of the present study indicate that 
/. argénteas possesses both of these at- 
tributes. 

The small prodissoconch I (length = 110 
ixm) of /. argénteas indicates an egg size as- 
sociated with bivalves possessing plank- 
totrophic larvae, and the well-developed pro- 
dissoconch II (approximately 500 ixm) is an 
indication of an appreciable free-swimming 
phase (Turners Lutz, 1984). Individual repro- 
ductive output is apparently quite large, with 
an estimated 3,000 eggs in varying stages of 
development observed within the ovaries of a 
single female 5.26 mm in length. By broad- 
casting large numbers of free-swimming lar- 
vae into the water column with the capability 
of remaining suspended for an extended pe- 
riod of time, /. argénteas has the dispersal 
capabilities necessary for successful coloni- 
zation of an ephemeral deep-sea habitat. 

Based on what has been learned from the 
wood block and panel studies, /. argénteas 
increases its population size by means of lar- 
val settlement. The abundance of small indi- 
viduals found on several of the wood panels 
(1,500-2,200 specimens <1.2 mm in length 
on two panels colllected in late July) indicated 
dense settlement by larvae undoubtedly orig- 
inating from the previously established wood 
island population. Grassle & Morse-Porteous 
(1 987) also reported large numbers of juvenile 
specimens of /. argénteas in the organically 
enriched sediments surrounding the wood is- 
land at both DOS 1 and DOS 2. Whereas the 
larvae of /. argénteas have the capacity to 
colonize distant isolated patches, it may often 
be more advantageous to settle close to the 
home site when unexploited substratum re- 
mains available. It is known that the planktonic 
larvae of shallow-water invertebrates often 
display great variability in the length of the 
competent phase, which may be greatly af- 
fected by the presence of an appropriate set- 
tlement site (Scheltema, 1986; Knowlton & 
Keller, 1986). The high reproductive capacity 
of /. argénteas ensures dense settlement of 
the wood island area by those larvae remain- 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



37 



ing close to the homesite, perhaps due to 
chemosensory cues similar to those described 
for shallow-water species (Burke, 1986). 

Results of this study indicate that while /. 
argénteas has a high reproductive potential 
and is capable of rapid population increase by 
dense larval settlement of an established site, 
it lacks the capacity seen in shallow-water op- 
portunists immediately following the coloniza- 
tion of a new site. The generation time of a 
shallow-water opportunist, such as Capitella 
sp., for example, is approximately 30 to 40 
days (Grassle & Grassle, 1974), whereas at 
DOS 2 /. argénteas is not capable of repro- 
duction until the year following settlement. 
The few pioneering larvae that successfully 
colonize an isolated patch of organic matter 
would experience a delay prior to the full ex- 
ploitation of the available resource. Popula- 
tion size could not increase until the pioneer- 
ing individuals were sexually mature and able 
to produce large numbers of larvae. 

Colonization rates of organically enriched 
sediment trays in the deep sea are quite low 
when compared to similar studies in shal- 
lower waters (Levin & Smith, 1984; Desbru- 
yères, 1985; Grassle & Morse-Porteous, 
1987). For many species, the pattern of col- 
onization on sediment trays deployed by 
Grassle & Morse-Porteous (1987) at DOS 2 
was a small initial settlement followed by in- 
creasing densities with time. For four of the 
more common species colonizing these sed- 
iment trays, Grassle & Morse-Porteous 
(1987) indicated maximum times to maturity 
much greater than those of similar opportun- 
ists from shallower waters. The bivalve Nu- 
cula cancellata collected from these trays 
was, for example, estimated to have a maxi- 
mum maturation time of two years. The de- 
pendence upon colonization by planktonic lar- 
vae and the preliminary delay in population 
increase due to slow maturation time was 
used by Grassle & Morse-Porteous (1987) to 
explain the slow colonization rates reported 
for the deep-sea benthos. The sexual matu- 
rity of the deep-sea organic enrichment op- 
portunist /. argénteas, which occurs a year 
after initial settlement, lends further support to 
the view that deep-sea opportunists differ 
from those in shallow water in the rate of their 
response to patches of organic enrichment. 

ACKNOWLEDGMENTS 

This study would not have been completed 
without the assistance of Ruth Turner (Har- 



vard University) who graciously allowed me 
free access to her laboratory and to the ma- 
terials collected from her deep-sea wood is- 
land studies. Richard Lutz (Rutgers Univer- 
sity) reviewed an earlier draft and provided 
support and direction in the correction of a 
misinterpretation of my original growth line 
analysis. Early direction was provided by 
Judy Grassle, Fred Grassle, Roger Green 
and especially Felicita D'Escrivan and Peter 
Schweitzer of Pat Lohmann's lab (WHOI). 
Nicholas Butterfield (Harvard University) 
provided advise and allowed access to the 
necessary grinding and polishing equipment. 
Michael Fogarty (NMF-Woods Hole) contrib- 
uted the NORMSEP and FISHPARM pro- 
grams, and Frank Almeida (NMF-Woods 
Hole) made programming changes in NORM- 
SEP to accommodate my data. At the Mu- 
seum of Comparative Zoology (Harvard Uni- 
versity), Robin Pinto did the SEM work while 
AI Coleman printed Figures 1 and 2. Robert 
Buteau provided his computer expertise and 
helpful advice throughout this project. Ken 
Boss, Robert Bullock, George Davis and an 
anonymous reviewer offered constructive crit- 
icisms of earlier drafts of this manuscript. The 
recovery of the wood block, SEM and photo- 
graphic work for Figures 1 and 2 were sup- 
ported by the Office of Naval Research 
through Dr. Ruth Turner under Contract no. 
N00014-84-C-0258 with Harvard University. 



LITERATURE CITED 

ALONGI, D. M., 1992, Bathymétrie patterns of 
deep-sea benthic communities from bathyal to 
abyssal depths in the western South Pacific (So- 
lomon and Coral Seas). Deep-Sea Research, 39: 
549-565. 

ALLEN, J. A. & H. L. SANDERS, 1973, Studies on 
the deep-sea Protobranchia (Bivalvia): the fami- 
lies Siliculidae and Lametilidae. Bulletin of the 
Museum of Comparative Zoology 1 45: 263-31 0. 

BILLETT, D. S. M., R. S. LAMPITT, A. L RICE & R. 
F. С MANTOURA, 1983, Seasonal sedimenta- 
tion of phytoplankton to the deep-sea benthos. 
Nature, 3022: 520-522. 

BREEN, P. A., С GABRIEL & T. TYSON, 1991, 
Preliminary estimates of age, mortality, growth, 
and reproduction in the hiatellid clam Panopea 
zelandica in New Zealand. New Zealand Journal 
of tsarine and Freshwater Research, 25: 231- 
237. 

BURKE, R. D., 1986, Pheromones and the gregar- 
ious settlement of marine invertebrate larvae. 
Bulletin of Marine Science, 39: 323-331 . 

CARTER, J. G., 1980, Environmental and biologi- 



38 



DEAN 



cal controls of bivalve shell mineralogy and mi- 
crostructure. Pp. 69-113, 627-643, in: D. С 
Rhoads, & R. a. Lutz, eds.. Skeletal growth of 
aquatic organisms, Plenum Press, New York. 

CARTER, J. G., R. A. LUTZ & M. J. S. TEVESZ, 
1990, Shell microstructural data for the Bivalvia. 
part VI. Orders Modiomorphoida and Mytiloida. 
Pp. 391-41 1 , in: J. G. Carter, ed., Skeletal bio- 
mineralization: patterns, processes, and evolu- 
tionary trends. Vol. 1, Van Nostrand Reinhold, 
New York. 

CHARNOV, E. L & J. BULL, 1977, When is sex 
environmentally determined: Nature, 266: 828- 
830. 

CSANADY, G. T., J. H. CHURCHILL & B. BUT- 
MAN, 1988, Near-bottom currents over the Con- 
tinental Slope in the Mid-Atlantic Bight. Continen- 
tal Shelf Research, 8: 653-671 . 

DEBRUYÈRES, D., J. Y. BERVAS & A. KHRI- 
POUNOFF, 1980, Un cas de colonization rapide 
d'un sediment profond. Oceanologica Acta, 3: 
285-291 . 

DEBRUYÈRES, D., F. GAILL, L. LAUBIER & Y. 
FOUGUET, 1985, Polychaetous annelids from 
hydrothermal vent ecosystems: an ecological 
overview, in: M. L. Jones, ed., The hydrothermal 
vents of the eastern Pacific; an overview. Bulletin 
of the Biological Society of Washington, 6: 103- 
116. 

DEBRUYÈRES, D. & L LAUBIER, 1988, Exploita- 
tion dune source de matière organique concen- 
trée dans l'océan profond: intervention d'une an- 
nèlide polychète nouvelle. Comptes Rendus de 
l'Académie des Sciences, Paris, 30(111): 329- 
335. 

DELL, R. K., 1987, Mollusca of the family Mytilidae 
(Bivalvia) associated with organic remains from 
deep waters off New Zealand, with revisions of 
the genera Adipicola Dautzenberg, 1 927 and Ida- 
sola Iredale, 1915. National l\/luseum of New 
Zealand Records, 3(3): 17-36. 

DEUSER, W. G., E. H. ROSS & R. F. ANDERSON, 
1981, Seasonality in the supply of sediment to 
the deep Sargasso Sea and implications for the 
rapid transfer of matter to the deep oceans. 
Deep-Sea Research, 28: 495-505. 

DICKSON, R. R., W. J. GOULD, P. A. GURBUTT & 
P. D. KILLWORTH, 1982, A seasonal signal in 
ocean currents to abyssal depths. Nature, 295: 
193-198. 

FELBECK, H. & G. N. SOMERO, 1982, Primary 
production in deep-sea hydrothermal vent organ- 
isms: roles of sulfide-oxidizing bacteria. Trends in 
Biochemical Science, 7: 201-204. 

FIALA-MÈTIVIONI, A., С METIVIER, A. HERRY & 
M. LE PENNEC, 1 986, Ultrastructure of the gill of 
the hydrothermal-vent mytilid Bathymodiolus sp. 
Marine Biology, 92: 65-72. 

FOSTER-SMITH, R. L., 1975, The role of mucus in 
the mechanism of feeding in three filter-feeding 
bivalves. Proceedings of the f^alacological Soci- 
ety of London, 41 : 571-588. 

FRITZ, L. W. & R. A. LUTZ, 1986, Environmental 



perturbations reflected in internal shell growth 
patterns of Corbicula fluminea (Molluscs: Bi- 
valvia). Veliger, 28: 401-417. 

GAGE, J. D., 1985, The analysis of population dy- 
namics in deep-sea benthos. Pp. 201-212, in: 
p. E. GIBBS, ed.. Proceedings of the 19th European 
t\Aarine Biological Symposium, Cambridge Uni- 
versity Press, Cambridge. 

GAINES, S. & J. ROUGHGARDEN, 1985, Larval 
settlement rate: a leading determinant of struc- 
ture in an ecological community of the marine 
intertidal zone. Proceedings of the National 
Academy of Science, 82: 3707-371 1 . 

GALLUCCI, V. F. & B. B. GALLUCCI, 1982, Repro- 
duction and ecology of the hermaphroditic cockle 
Clinocardium nuttallii (Bivalvia: Cardiidae) in Gar- 
rison Bay. Marine Ecology Progress Series, 7: 
137-145. 

GHISELIN, M. T., 1969, The evolution of hermaph- 
roditism among animals. Quarterly Review of Bi- 
ology, 44: 189-208. 

GIESE, A. С & J. S. PEARSE, 1974, Introduction: 
general principles. Pp. 1-49, in: A. С Giese & J. 
S. Pearse, eds., Reproduction of Marine Inverte- 
brates. Volume 1 Academic Press, New York and 
London. 

GILKINSON, K. D., J. A. HUTCHINGS, P. E. 
OSHEL & R. L HAEDRICH, 1986, Shell micro- 
structure and observations on internal banding 
patterns in the bivalves Yoldia thraciaeformis 
Storer, 1838, and Nuculana pernula Müller, 1779 
(Nuculanidae), from a deep-sea environment. 
Veliger, 29: 70-77. 

GOODAY, A. J., 1988, A response by benthic For- 
aminifera to the deposition of phytodetritus in the 
deep sea. Nature, 332: 70-73. 

GOODAY, A. J. & P. J. D. LAMBSHEAD, 1989, The 
impact of seasonally deposited phytodetritus on 
benthic foraminiferal populations in the bathyal 
northeast Atlantic: the species response. Marine 
Ecology Progress Series, 58: 53-67. 

GOODAY, A. J. & С M. TURLEY, 1990, Re- 
sponses by benthic organisms to inputs of or- 
ganic material to the ocean floor: a review. Philo- 
sophical Transactions of the Royal Society of 
London, A, 331: 119-138. 

GRAF, G., 1989, Benthic-pelagic coupling in a 
deep-sea benthic community. Nature, 341 : 437- 
439. 

GRASSLE, J. F., 1977, Slow recolonization of 
deep-sea sediment. Nature, 265: 618-619. 

GRASSLE, J. F., 1978, Diversity and population 
dynamics of benthic organisms. Oceanus, 21: 
42-49. 

GRASSLE, J. F., 1986, The ecology of deep-sea 
hydrothermal vent communities. Advances in 
Marine Biology, 23: 302-362. 

GRASSLE, J. F. & J. P. GRASSLE, 1974, Oppor- 
tunistic life histories and genetic systems in ma- 
rine benthic polychaetes. Journal of Marine Re- 
search, 32: 253-284. 

GRASSLE, J. F., & L. S. MORSE-PORTEOUS, 
1987, Macrofaunal colonization of disturbed 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



39 



deep-sea environments and the structure of 
deep-sea benthic communities. Deep-Sea Re- 
search, 34: 1911-1950. 

HOAGLAND, K. E. 1978, Protandry and the evolu- 
tion of environmentally-mediated sex change: a 
study of the Mollusca. Malacologia, 1 7: 365-391 . 

HOLLISTER, С D. & I. N. McCAVE, 1984, Sedi- 
mentation under deep-sea storms. Nature, 309: 
220-225. 

HONJO, S., 1980, Material fluxes and modes of 
sedimentation in the mesopelagic and bathype- 
lagic zones. Journal of Marine Research, 38: 53- 
97. 

HONJO, S., 1982, Seasonality and interaction of 
biogenic and lithogenic particulate flux at the 
Panama Basin. Science, 218: 883-884. 

HUGHES, T. P., 1985, Population dynamics and life 
histories of early successional corals. Proceed- 
ings of the Fifth International Coral Reef Con- 
gress, Tahiti, 2: 101-106. Antenne Museum- 
Ephe, Moorea, French Polynesia. 

HUGHES, T. P., 1990, Recruitment limitation, mor- 
tality, and population regulation in open systems: 
a case study. Ecology, 71: 12-20. 

HUTCHINGS, J. A. & R. L HAEDRICH, 1984, 
Growth and population structure in two species of 
bivalves (Nuculanidae) from the deep sea. tsa- 
rine Ecology Progress Series, 17: 135-142. 

ITTEKKOT, v., W. G. DEUSER & E. T. DEGENS, 
1984, Seasonality in the fluxes of sugars, amino 
acids and amino sugars to the deep ocean: Sar- 
gasso Sea. Deep-Sea Research, 31: 1057- 
1069. 

JONES, D. S., I. THOMPSON & W. AMBROSE, 
1978, Age and growth rate determinations for the 
Atlantic surf clam Spisula solidissima (Bivalvia: 
Mactracea), based on internal growth lines in 
shell cross-sections. Marine Biology, 47: 63-70. 

JONES, D. S., D. F. WILLIAMS & M. A. ARTHUR, 
1 983, Growth history and ecology of the Atlantic 
surf clam, Spisula solidissima (Dillwyn) as re- 
vealed by stable isotopes and annual shell incre- 
ments. Journal of Experimental Marine Biology 
and Ecology, 73: 225-242. 

JUMARS, P. A. & E. D. GALLAGHER, 1982, Deep- 
sea community structure; Three plays on the 
benthic proscenium, Pp. 217-225, in W. G. 
Ernst & J. G. Morin, eds.. The environment of 
the deep sea. Prentice Hall, Englewood Cliffs, 
NJ. 

KAUFMANN, К. W., 1981 , Fitting and using growth 
curves. Oecologia (Berlin), 49: 293-299. 

KENNISH, M. J., 1980, Shell microgrowth analysis: 
Mercenaria mercenaria as a type example for re- 
search in population dynamics. Pp. 255-294, in 
D. С Rhoads & R. A. Lutz, eds., Skeletal growth 
of aquatic organisms. 

KENK, V. С & В. WILSON, 1985, A new mussel 
(Bivalvia, Mytilidae) from hydrothermal vents in 
the Galapagos Rift zone. Malacologia, 26: 253- 
271. 

KNOWLTON, N. & B. D. KELLER, 1986, Larvae 
which fall far short of their potential: highly local- 



ized recruitment in an alpheid shrimp with ex- 
tended larval development. Bulletin of Marine 
Science, 39: 213-223. 

LAMBSHEAD, P. J. D. & A. J. GOODAY, 1990, The 
impact of seasonally deposited phytodetritus on 
epifaunal and shallow infaunal benthic foraminif- 
eral populations in the bathyal northeast Atlantic: 
the assemblage response. Deep-Sea Research, 
37: 1263-1283. 

LAMPITT, R. S., 1985, Evidence for the seasonal 
deposition of detritus to the deep-sea floor and its 
subsequent resuspension. Deep-Sea Research, 
32: 885-897. 

LEVIN, L. A. & С R. SMITH, 1984, Response of 
background fauna to disturbance and enrichment 
in the deep sea; a sediment tray experiment. 
Deep-Sea Research, 31: 1277-1285. 

LUTZ, R. A., 1976, Annual growth patterns in the 
inner shell layer of Mytilus edulis (L.). Journal of 
the Marine Biological Association of the United 
Kingdom, 56: 723-731. 

LUTZ, R. A. & D. С RHOADS, 1978, Shell struc- 
ture of the Atlantic ribbed mussel, Geukensia de- 
missa (Dillwyn): a réévaluation. Bulletin of the 
American Malacological Union, for 1978: 13-17. 

LUTZ, R. A. & D. С RHOADS, 1980, Growth pat- 
terns within the molluscan shell: an overview. Pp. 
203-254, in: D. С Rhoads & R. A. Lutz, eds.. 
Skeletal growth of aquatic organisms. Plenum 
Press, New York. 

LUTZ, R. A., L. W. FRITZ & R. M. CERRATO, 1 988, 
A comparison of bivalve (Calyptogena magnifica) 
growth at two deep-sea hydrothermal vents in the 
eastern Pacific. Deep-Sea Research, 35: 1793- 
1810. 

LUTZ, R. A., L. W. FRITZ & D. С RHOADS, 1985, 
Molluscan growth at deep-sea hydrothermal 
vents. In: M. L. Jones, ed.. The hydrothermal 
vents of the eastern Pacific; an overview. Bulletin 
of the Biological Society of Washington, 6: 199- 
210. 

MacDONALD, B. A. & L. H. THOMAS, 1980, Age 
determination of the soft-shell clam Mya arenaria 
using shell internal growth lines. Marine Biology, 
58: 105-109. 

MacDONALD, P. D. M. & T. J. PITCHER, 1979, 
Age-groups from size-frequency data: a versatile 
and efficient method of analyzing distribution 
mixtures. Journal of the Fisheries Research 
Board of Canada, 36: 987-1001 . 

MANTYLA, A. W. & J. L. REID, 1983, Abyssal char- 
acteristics of the world ocean waters. Deep-Sea 
Research, 30: 805-833. 

MATSUEDA, H., N. HANDA, I. INOUE & H. TA- 
KANO, 1986, Ecological significance of salp fecal 
pellets collected by sediment traps in the eastern 
North Pacific. Marine Biology, 91: 421-431. 

McCUAIG, J. M. & R. H. GREEN, 1983, Unionid 
growth curves derived from annual rings: a base- 
line model for Long Point Bay, Lake Erie. Cana- 
dian Journal of Fisheries and Aquatic Science, 
40: 436-442. 

McNEW, R. W. & R. С SUMMERFELT, 1978, Eval- 



40 



DEAN 



uation of a maximum-likelihood estimator for 
analysis of length-frequency distributions. Trans- 
actions of the American Fisheries Society, 107: 
730-736. 

MENZEL, D. W. & J. H. RYTHER, 1961, The an- 
nual cycle of primary production in the Sargasso 
Sea off Bermuda. Deep-Sea Research, 7: 351- 
367. 

PALMER, A. R., 1983, Relative costs of producing 
skeletal organic matrix versus calcification: evi- 
dence from marine gastropods. Marine Biology, 
75: 287-292. 

PANNELLA, G. & С. MacCLINTOCK, 1968, Biolog- 
ical and environmental rhythms reflected in mol- 
luscan shell growth. Journal of Paleontology, 42: 
64-80. 

PEARSON, T. H. & R. ROSENBERG, 1978, Mac- 
robenthic succession in relation to organic en- 
richment and pollution of the marine environ- 
ment. Oceanography and Marine Biology Annual 
Review, 16: 229-311. 

REX, M. A., С A. VAN UMMERSEN & R. D. 
TURNER, 1979, Reproductive pattern in the 
abyssal snail Benthonella tenella (Jeffreys). Pp. 
173-188, in: S. E. Stancyk, ed.. Reproductive 
ecology of marine invertebrates. Belle W. Baruch 
Library in Marine Science No. 9, University of 
South Carolina Press, Columbia. 

RHOADS, D. L, R. A. LUTZ, R. M. CERRATO & 
E. C. REVELAS, 1982, Growth and prédation ac- 
tivity at deep-sea hydrothermal vents along the 
Galapagos Rift. Journal of Marine Research, 40: 
503-516. 

RHOADS, D. С & G. PANNELLA, 1970, The use of 
molluscan shell growth patterns in ecology and 
paleoecology. Lethaia, 3: 143-161. 

RICHARDSON, С A., 1987, Microgrowth patterns 
in the shell of the Malaysian cockle Anadara gra- 
nosa (L.) and their use in age determinations. 
Journal of Experimental Marine Biology and 
Ecology, 1 1 1 : 77-98. 

RIEMANN, P., 1989, Gelatinous phytoplankton de- 
tritus aggregates on the Atlantic deep-sea bed. 
Structure and mode of formation. Mahne Biology, 
100:533-539. 

ROKOP, F. J., 1974, Reproduction patterns in the 
deep-sea benthos. Science, 186: 743-745. 

ROUGHGARDEN J., Y. IWASA & С BAXTER, 
1985, Demographic theory for an open marine 
population with space-limited recruitment. Ecol- 
ogy, 66: 54-67. 

ROWE, G. T. & W. D. GARDNER, 1979, Sedi- 
mentation rates in the slope water of the north- 
west Atlantic Ocean measured directly with sed- 
iment traps. Journal of Mahne Research, 37: 
581-600. 

SAILA, S. В., С. W. RECKSIEK & M. H. PRAGER, 
1988, Basic fishery science program. A compen- 
dium of microcomputer programs and manual of 
operation. Developments in aquaculture and fish- 
ery science, 18. Amsterdam, Elsevier, 230 pp. 

SANDERS, H. L, R. R. HESSLER & G. R. HAMP- 
SON, 1965, An introduction to the study of deep- 



sea benthic faunal assemblages along the Gay 
Head-Bermuda transect. Deep-Sea Research, 
12:845-867. 

SCHELTEMA, R. S., 1986, On dispersal and plank- 
tonic larvae of benthic invertebrates: an eclectic 
overview and summary of problems. Bulletin of 
Marine Science, 39: 290-322. 

SEED, R. & R. A. BROWN, 1978, Growth as a 
strategy for survival in two marine bivalves 
Cerastoderma edule and Modiolus modiolus. 
Journal of Animal Ecology, 47: 283-292. 

SMITH, С R. & R. R. HESSLER, 1987, Coloniza- 
tion and succession in deep-sea ecosystems. 
Trends in Ecology and Evolution, 2: 359-363. 

SMITH, K. L., Jr. & K. R. HINGA, 1983, Sediment 
community respiration in the deep sea. Pp. 331- 
370, in G. T. RowE, ed.. Deep sea biology, the 
sea, Vol. 8, John Wiley and Sons, New York. 

SMITH, С R., H. KUKERT, R. A. WHEATCROFT, 
P. A. JUMARS & J. W. DEMING, 1989, Vent 
fauna on whale remains. Nature, 341 : 27-28. 

SOMERO, G. N., J. F. SIEBENALLER & P. W. 
HOCHACHKA, 1983, Biochemical and physio- 
logical adaptations of deep-sea animals. Pp. 
261-330, in G. T. RowE, ed., Deep sea biology, 
the sea, Vol. 8, Wiley, New York. 

TAYLOR, J. D., W. J. KENNEDY & A. HALL, 1969, 
The shell structure and mineralogy of the Bi- 
valvia: introduction; Nuculacea-Trigonacea. Bul- 
letin of the British Museum (Natural History) Zo- 
ology Supplement, 3: 1-125. 

THOMPSON, I., D. S. JONES & D. DREIBELBIS, 
1980, Annual internal growth banding and life 
history of the ocean quahog Árctica islándica 
(Mollusca: Bivalvia). Marine Biology, 57: 25-34. 

TRUTSCHLER, K. & С SAMTLEBEN, 1988, Shell 
growth of Astarte elliptica (Bivalvia) from Kiel Bay 
(Western Baltic Sea). Mahne Ecology Progress 
Series, 42: 155-162. 

TUREKIAN, K. K., J. K. COCHRAN, D. P. 
KHARKAR, R. M. CERRATO, J. R. VAISNYS, H. 
L SANDERS, J. F. GRASSLE & J. A. ALLEN, 
1975, Slow growth rate of a deep-sea clam de- 
termined by 228-Ra chronology. Proceedings of 
the National Academy of Sciences, USA, 72: 
2829-2832. 

TUREKIAN, K. K., J. K. COCHRAN, Y. NOZAKI, I. 
THOMPSON & D. S. JONES, 1982, Determina- 
tion of shell deposition rates of Árctica islándica 
from the New York Bight using natural 228-Ra 
and 228-Th and bomb-produced 14-C. Limnol- 
ogy and Oceanography, 27: 737-741 . 

TURNER, R. D., 1973, Wood-boring bivalves, op- 
portunistic species in the deep sea. Science, 
180: 1377-1379. 

TURNER, R. D., 1977, Wood, mollusks, and deep- 
sea food chains. Bulletin of the American Mala- 
cological Union for 1976: 13-19. 

TURNER, R. D., 1981, "Wood islands" and the 
"thermal vents " as centers of diverse communi- 
ties in the deep sea. The Soviet Journal of Ma- 
rine Biology, 7(1): 1-9. 

TURNER, R. D. & R. A. LUTZ, 1984, Growth and 



POPULATION STUDY OF A DEEP-SEA BIVALVE 



41 



distribution of mollusks at deep-sea vents and 
seeps. Oceanus, 27: 54-62. 

TYLER, P. A., 1988, Seasonality in the deep sea. 
Oceanography and Marine Biology Annual Re- 
view, 26: 227-258. 

TYLER, P. A. & S. L. PAIN, 1982, The reproductive 
biology of Plutonaster bifrons, Dytaster insignis, 
and Psilaster andromeda (Asteroidea: Astropec- 
tinidae) from the Rockall Trough. Journal of the 
Marine Biological Association of the United King- 
dom, 62: 869-887. 

WADA, K., 1961, Crystal growth of molluscan 
shells. Bulletin of the National Pearl Research 
Laboratory, 7: 703-728. 

WAREN, A., 1991, New and little known mollusca 
from Iceland and adjacent areas. II. Sarsia, 76: 
53-124. 



WILLIAMS, A. B. & R. D. TURNER, 1986, Squat 
lobsters (Galatheidae: Munidiopsis) associated 
with mesh-enclosed wood panels submerged in 
the deep sea. Journal of Crustacean Biology, 6: 
617-624. 

WILLIAMSON, P. & M. A. KENDALL, 1981, Popu- 
lation age structure and growth of the trochid 
Monodonta lineata determined from shell rings. 
Journal of the Marine Biological Association of 
the United Kingdom, 61 : 101 1-1026. 

WOLFF, T, 1976, Utilization of seagrass in the 
deep sea. Aquatic Botany, 2: 161-174. 

WRIGHT, W. G., 1988, Sex change in the Mollusca. 
Trends in Ecology and Evolution, 3: 137-140. 



Revised Ms. accepted 1 8 August 1 992 



MALACOLOGIA, 1993,35(1): 43-61 

EVOLUTIONARY RELATIONSHIPS AND EXTREME GENITAL VARIATION IN A 
CLOSELY RELATED GROUP OF PARTULA 

Michael 8. Johnson\ James Murray^ & Bryan Clarke^ 

ABSTRACT 

The land snails Partula otaheitana, P. jackieburchi, and P. affinis, endemic to Tahiti, are 
genetically very similar species with complex morphological relationships. There is great varia- 
tion among the species in the morphology of the reproductive system, P. jackieburchi having 
originally been placed in the genus Samoana because of its genital characters. Individuals with 
characteristics intermediate between the species have been found in several populations. Mul- 
tivariate analysis of morphological variation among 108 individuals from 14 sites shows that 
different combinations of the species may be distinct in sympatry, but that the distinctions break 
down at some sites. The morphology of genitalia is correlated with the morphology of shells in 
comparisons between species, and in comparisons between various intermediate forms, but not 
in comparisons within species. This pattern suggests that the correlation is due to intergradation 
between species, rather than to geographic variation within the separate species. Laboratory 
hybhds between P. otalieitana and P. jackieburchi have genitalia with charactehstics similar to 
those of many intermediate individuals found in the wild. Quantitative comparisons with the 
related genus Samoana show that the differences in genital anatomy between species in the P. 
otaheitana group are as great as, or greater than, the overall differences between genera. Our 
results show that even large differences in genital anatomy do not necessarily bhng about 
reproductive isolation, and they demonstrate the complexity of relationships within the endemic 
radiation on Tahiti. 



INTRODUCTION 

Land snails of the genus Partula have ra- 
diated on many high islands of the Pacific, 
and show their greatest diversity in the Soci- 
ety Islands (Cowie, 1992). The radiation on 
Moorea has been studied in the most detail, 
and has revealed complex patterns of varia- 
tion in reproductive relationships, morphol- 
ogy, and molecules (e.g. Crampton, 1932; 
Murray & Clarke, 1 980; Johnson et al., 1 986a; 
Murray et al., 1991). The species on Tahiti 
apparently represent a more recent radiation 
derived from a Moorean ancestor (Johnson et 
al., 1986b). Although the Tahitian species 
have not been as thoroughly studied, they too 
display a challenging array of diversity. The 
most confusing variation is in the Partula ota- 
heitana group. 

This group, which is endemic to Tahiti, is 
now considered to include the three species 
P. otaheitana (Bruguière, 1789), P. jackiebur- 
chi {Konäo, 1980), and P. affinis Pease, 1868 
(Kondo & Burch, 1979, 1983; Kondo, 1980; 
Johnson et al., 1986c). On the basis of their 
shell morphology, Crampton (1916) appor- 



tioned the variation represented by these taxa 
among eight subspecies of P. otaheitana, and 
this assignment was adopted in a recent anal- 
ysis of geographical variation (Emberton, 
1982). However, P. o. affinis, the most distinc- 
tive of the "subspecies," is widely sympatric 
with P. o. rubescens Reeve, 1850, "its very 
antithesis in most respects" (Crampton, 1916: 
185). Whereas P. o. rubescens is large, al- 
most entirely sinistral, and generally yellow or 
red, P. o. affinis is generally small, usually 
dextral, and typically brown (Crampton, 1916, 
color plates). The two sympatric forms also 
have distinct genital anatomies (Kondo & 
Burch, 1979; Kondo, 1980), supporting the 
view that they are separate species. 

Although the morphology of the reproduc- 
tive system can often be useful in clarifying 
relationships (e.g. Reid, 1986), this appears 
not to be so for the P. otaheitana group, in 
spite of the differences between P. affinis and 
P. otaheitana. It was on the basis of genital 
morphology that P. jackieburchi was sepa- 
rated from P. o. rubescens. Although the 
shells of the two taxa are virtually indistin- 
guishable, the anatomical differences are so 



^ Department of Zoology, University of Western Australia, Nedlands, Western Australia 6009, Australia. 

^Department of Biology, University of Virginia, Charlottesville, Virginia 22901, U.S.A. 

^Departniient of Genetics, School of Biological Sciences, Queens Medical Centre, Nottingham NG7 2UH, United Kingdom. 



43 



44 



JOHNSON, MURRAY & CLARKE 



striking that P. jackieburchi was initially de- 
scribed as a member of the genus Samoana 
(Kondo, 1980). Later studies of allozymes, 
however, showed that P. jackieburchi is very 
similar to other species of Partula, and very 
different from Samoana (Johnson et a!., 
1986c). Indeed, P. otaheitana, P. jackieburctii, 
and P. affinis cannot be distinguished by their 
allozymes or their mitochondrial DNA (Murray 
etal., 1991). 

The genital characteristics of P. jackiebur- 
chi show a strong convergence toward the 
genus Samoana, resulting in great anatomi- 
cal diversity within this closely related group 
of Partula species. While attempting to dis- 
cover the relationships of P. jackieburchi and 
P. otaheitana, we have found another level of 
complexity. At several sites there are snails 
that do not fit the anatomical descriptions of 
any species. This finding was perhaps antic- 
ipated by Kondo's (1968) summary of unpub- 
lished observations: "A curious instance of a 
species having 3 distinct forms of genitalia 
occurs in Tahiti. Five of the 8 varieties (or 
subspecies) of P. otaheitana dissected show 
that two of them vary in anatomy according to 
valleys." 

We have tried to find out whether the pecu- 
liar anatomical types represent geographic 
variation within, or genetic exchange be- 
tween, taxa. Few studies of reproductive 
anatomy in gastropods quantify the variation 
within or between taxa. In the highly variable 
P. otaheitana group, however, such quantifi- 
cation is essential. In this paper we report the 
results of multivariate analyses of genital mor- 
phology and shell characters in samples of P. 
otaheitana, P. jackieburchi, P. affinis, and var- 
ious types of intermediates, and compare 
them with data from laboratory hybrids be- 
tween P. otaheitana and P. jackieburchi. We 
also compare the Partula species with two 
species of Samoana. 



uncertain placement (Table 1). The sampling 
localities are concentrated in the eastern half 
of Tahiti Nui, the region where P. otaheitana 
and P. affinis are sympatric. All the securely 
identified P. otaheitana are P. o. rubescens, 
except those from Sample 801 (P. o. crassa 
"Pease" Garrett, 1884) and Sample 778 (P. o. 
amabilis Pfeiffer, 1846). All are sinistral, ex- 
cept two dextrals in Sample 778. All the P. 
affinis are dextral, except three sinistrals in 
Sample 791. As well as the samples of 
Partula, three individual Samoana diaphana 
Crampton & Cooke, 1953, from Moorea (one 
from Uufau; two from Faatoai) and seven S. 
attenuata (Pease, 1 864) (five from Hotutea on 
Moorea; two from Tiarei on Tahiti) were in- 
cluded to allow comparison between the two 
genera. 

Hybrids were obtained from laboratory mat- 
ings between P. otaheitana from Papehue 
(Sample 801) and P. jackieburchi from Ma- 
haena (Sample 780). Experimental matings 
within and between the species were set up to 
test the relative fertility of the interspecific ' 
matings, and the viability and fertility of the | 
hybrids. The parents of the matings were 
wild-caught juveniles reared to maturity in iso- 
lation. Laboratory conditions were as de- 
scribed in earlier studies (Murray & Clarke, 
1966). Unfortunately, neither the experimen- 
tal matings nor the controls were very suc- 
cessful. Not enough young were produced to 
allow comparisons of fertilities. Nevertheless, 
mature offspring were produced by two inter- 
specific matings. From one mating, both par- 
ents and three mature offspring were dis- 
sected. The parents of the second mating 
died, and were in too poor a condition for 
measurement of the anatomical traits, but two 
mature offspring of that mating were dis- 
sected. 

Measurements 



MATERIALS AND METHODS 

Samples 

We examined 108 adult Partula from 14 
sites. Their locations are shown in Figure 1 , 
and a summary of the samples is given in 
Table 1 . The snails were initially identified us- 
ing the anatomical drawings by Kondo & 
Burch (1 979) and Kondo (1 980). The samples 
contain 24 obvious P. otaheitana, 22 P. jack- 
ieburchi, 17 P. affinis, and 45 specimens of 



Seventeen anatomical characters were 
measured in each snail (Fig. 2): length of vas 
deferens (LVD), coded as (stretched taut 
between penis and oviduct), 1 ("normal"), or 
2 (heavily convoluted); length of penis 
(LPEN), including epiphalus, from its tip to the 
junction with the vagina; angle of retractor 
(ARET), measured on the side of entry of the 
vas deferens, between a line along the out- 
side of the retractor and a line tangent to the 
penis at the point of attachment (to nearest 
15°); angle of insertion of vas deferens (AVD) 
(to nearest 15°); distance from vas deferens 



ANATOMICAL VARIATION IN PARTULA 



45 




TAHITI NUI 



FIG. 1 . Map of Tahiti, showing sampling sites for the Partula otaheitana group. Sample codes as in Table 1 . 



TABLE 1 . Samples dissected for quantitative study of genital morphology in the Partula otaheitana group 
on Tahiti. Sample codes as in Fig. 1. 



Sample 


Valley 


P. otaheitana 


P. jackieburchi 


P. affinis 


unplaced 


801 


Papehue 


4 






3 


778 


Hamuta 


6 








794 


Papenoo 


1 




1 


10 


779 


Faarumai 


7 




4 




776 


Tiarei 


2 




1 




784 


Tiarei 


3 








786 


Tiarei 




1 






742 


Tiarei 


1 


9 






780 


north fHahaena 




9 




1 


791 


south Mahaena 






7 




792 


south Mahaena 






4 


25 


793 


south Mahaena 








3 


774 


Faone 




3 






813 


Faone 








3 


TOTAL 




24 


22 


17 


45 



46 



JOHNSON, MURRAY & CLARKE 




FIG. 2. Diagram showing the traits measured in the analysis of genital morphology. The traits LSG and LALB 
are not shown. See text for explanation. 



to retractor (VDRET), measured on the prox- 
imal side of each; length of spermatheca, 
from its tip to junction with vagina (LSP); dis- 
tance from the genital pore to junction of the 
spermatheca with the vagina (LFSP); length 
of vagina from its junction with the spermath- 
eca to the beginning of the oviduct (LVAG); 
width of penis at the vas deferens (WPVD); 
width of penis at one quarter of its length from 
the genital pore (WPEN1); width of penis at 
three quarters of its length (WPEN3); dis- 
tance from entry of vas deferens to the junc- 
tion of the penis with the vagina (HVD); width 
of the spermatheca at its midpoint (WSP2); 
width of the spermatheca at one quarter of its 
length (WSP1); width of the spermatheca at 
three quarters of its length (WSP3); length of 
shell gland (LSG); length of albumen gland 
(LALB). 

Although our interest in this group of spe- 
cies was initiated by Kondo's (1980) descrip- 
tion of P. jackieburchi, we soon became 
aware that the overal variation of genital mor- 
phology transcends the specific problems 
raised by that work. It is this overall variation, 
and not the specific taxonomic questions, that 



is the focus of this study. We did not select the 
anatomical traits specifically with the P. ota- 
heitana group in mind, so they do not repli- 
cate the set of traits used by Kondo (1968, 
1980). Except for one addition (LFSP), they 
are the traits used previously to represent 
variation in Partula on Moorea (Murray & 
Clarke, 1980). Therefore, our selection of 
characters should not introduce any bias 
stemming from our perception of variation in 
the P. otaheitana group. Nevertheless, the set 
of traits is sufficiently comprehensive that it 
should reflect the major variations described 
by Kondo. 

The shells of all but eight of the dissected 
Partula were also measured, producing 13 
variables (for detailed descriptions, see Mur- 
ray & Clarke 1980): length of shell (SHLEN); 
width of shell (SHWID); length of aperture 
(APLEN); width of aperture (APWID); length 
of spire (SPILEN); width of spire (SPIWID); 
width of upper suture (SUTWID); width of lip 
(LIPWID); thickness of lip (LIPTHIC); height 
of shell (SHHT); height of spire (SPHT); angle 
between columella and long axis of aperture 
(APANG); number of whorls (WHORL). 



ANATOMICAL VARIATION IN PARTULA 



47 



Measurements were made with vernier cal- 
ipers to the nearest 0.1 mm. Anatomical mea- 
surements of the genitalia were made on 
camera lucida images, projected on a ground 
glass screen at a magnification of 5. All mea- 
surements were made by one person to en- 
sure the consistency of any individual bias. 
The anatomical data are given in the Appen- 
dix. 

Analyses 

In morphometric studies, variation in size 
can overwhelm other components of varia- 
tion. In order to minimize redundancy among 
the characters, it is important to correct for the 
underlying effect of size, and there are sev- 
eral possible approaches to this problem. Ra- 
tios are sometimes used, but they have se- 
vere statistical problems, and can produce 
misleading results (Atchley & Anderson, 
1978). A more reliable approach is to use re- 
gression analysis, and adjust the variables to 
a standard size. Here, the relevant regression 
is that within species, rather than that in the 
total sample. A variable independent of size 
within species but correlated with size among 
species should not be "corrected" for size, 
because we are interested in species differ- 
ences. We have used the length of the shell 
(SHLEN) as a measure of size. Within each 
species, each anatomical and shell variable 
was tested in a regression against SHLEN. If 
the average of the three intraspecific r^ values 
was greater than 0.5, the variable was trans- 
formed. The transformed value was: 

y' = y + m(Average SHLEN — SHLEN) 

where y is the original measurement, and m is 
the weighted average of the slopes of the 
within-species regressions (weighted by r^). 
Seven of the thirteen shell characters were 
transformed: SHWID; APLEN; APWID; 
SPILEN; SPIWID; SUTWID; SHHT. None of 
the genital traits required correction, as they 
were not significantly correlated with SHLEN 
within species. Three transformations were 
made to reduce redundancy among the ana- 
tomical characters themselves. HVD was 
scaled by its intraspecific regression on 
LPEN, in the manner described above. Be- 
cause HVD is a part of LPEN, the transforma- 
tion is an obvious one. Since WSP1 , WSP2, 
and WSP3 are the widths of the spermatheca 
at different positions, a clearer indication of 
the relative widths is provided by expressing 



WSP1 and WSP3 as their differences from 
WSP2. 

Because of damage, some anatomical 
measurements were missing in nine speci- 
mens of Partula (three with one missing 
value, two with two, and four with three). 
Missing values exclude an individual from 
many types of multivariate analysis. To avoid 
losing information, missing values were re- 
placed by estimates derived from a multiple 
regression. Each variable with a missing 
value was used as the dependent variable 
with all of the other characters as indepen- 
dent variables in a multiple regression, calcu- 
lated from all the specimens without missing 
values. Each missing value was then re- 
placed by a calculated one based on the data 
available for the individual concerned. To test 
the usefulness of this approach, we tested the 
regression equations on the individuals for 
which we had complete data. For all the rel- 
evant characters, the correlation between ac- 
tual values and the values predicted by the 
regressions was greater than 0.8, indicating 
that the estimates were reasonably accurate. 

The modified data were analysed by two 
kinds of multivariate techniques. We used a 
principal components analysis of the genital 
characters to give a summary of the variation 
that was independent of our initial classifica- 
tion of the specimens. We used varimax rota- 
tion to produce axes that were the most easily 
interpretable in terms of the original variables. 
After the principal components had confirmed 
that the differences between species could be 
quantified, we used discriminant functions to 
maximize the separation between the groups. 
The functions then gave scores for the indi- 
viduals initially classified as "unplaced." The 
data on shell variation were subjected to a 
separate discriminant analysis. The analyses 
were carried out using the SPSS-X routines at 
the University of Virginia. 



RESULTS 

Differences Between the Species 

The principal components confirmed our vi- 
sual impressions about the range of variation 
in genital anatomy. The first two axes (repre- 
senting 37.2% and 10.6% of the original vari- 
ation) show a clear separation of P. otaheit- 
ana from P. jackieburchi and P. affinis, and a 
weaker separation of P. jackieburchi from P. 
affinis (Fig. 3). Factor 1 separates P. otafieit- 



48 



JOHNSON, MURRAY & CLARKE 

-I- 




FIG. 3. Results of the principal components analysis of variation in genital morphology. Polygons enclose 
conspecifics of readily identifiable individuals. Circles = P. otaheitana; open triangles = P. jackieburchi; 
filled triangles = P. affinis; X = unknown. 



ana from P. affinis. Partula jackieburchi 
broadly overlaps the others, but with interme- 
diate average scores. High scores on this axis 
reflect the large, chunky shape of the P. ota- 
heitana penis, with strong positive loadings 
for LPEN and WSP2, and reasonably strong 
ones for some other traits (Table 2). Factor 2 
separates P. jackieburchi from the others. 
The strong negative loading of HVD and the 
positive loadings of VDRET, WPVD, and 
WPEN3 give P. jackieburchi negative scores, 
which reflect the distal insertion of the vas 
deferens into the relatively narrow penis. Pop- 



ulations within a species overlap each other 
on both axes, indicating that geographic vari- 
ation is small compared to the differences be- 
tween the species. Two more factors have 
eigenvalues greater than one, but they do not 
improve the separation of P. jackieburchi from 
P. affinis. The "unplaced" snails are variously 
intermediate, but spread over a wide range 
(Fig. 3). 

The principal components illustrate two im- 
portant points that underly later analyses. 
First, both the differences between species 
and the peculiarities of the "unplaced" snails 



ANATOMICAL VARIATION IN PARTULA 



49 



TABLE 2. Varimax factor loadings of traits in the 
principal components analysis of genital mor- 
phology in the Partula otaheitana group. Only 
traits with loadings greater than 0.5 on either of 
the first two principal components are included. 



Variable 


PCI 


PC2 


LPEN 


0.735 


0.334 


VDRET 


0.506 


0.729 


LSP 


0.694 


0.280 


LFSP 


0.646 


0.146 


WPVD 


0.373 


0.718 


WPEN3 


0.312 


0.745 


HVD 


0.018 


-0.843 


WSP2 


0.830 


0.195 


LSG 


0.830 


0.066 



are shown clearly. Because the analysis does 
not use our a priori groupings, it confirms that 
the difficulty of identifying specimens was 
genuine. Second, the measured characters 
do a reasonably good job of quantifying the 
visual classification. Thus we can be confi- 
dent, in moving to the discriminant analysis, 
that we are not making artificial groups. The 
principal components show that the specific 
groups are objectively recognizable, and the 
discriminant functions can be used to express 
their differences most effectively. 

Discriminant analysis of P. otaheitana, P. 
jackieburchi, and P. affinis gives a picture sim- 
ilar to that given by the principal components, 
but, as expected, a clearer separation of the 
species (Fig. 4). The first discriminant func- 
tion separates P. otaheitana from the others. 
This function is positively correlated with 
WPVD, WPEN3, and VDRET and negatively 
correlated with HVD, so that high scores rep- 
resent the club-like shape of the penis in P. 
otaheitana, and its proximal insertion of the 
vas deferens. The second discriminant func- 
tion separates P. jacl<ieburchi and P. affinis, 
mainly by the smaller size of P. affinis (Table 
3). 

The discriminant functions based on geni- 
talia correctly group all the members of the 
three species identified in the initial classifica- 
tion. Those based on shell characters do not 
do so well. The shells of 24 P. otaheitana, 17 
P. jaciiieburchi, and 17 P. affinis were ana- 
lyzed, and the discriminant analysis incor- 
rectly classified 12% of the specimens from 
each species. Nearly all the separation be- 
tween the species was by the first function, on 
which P. jackieburchi is intermediate between 
P. otaheitana and P. affinis, which do not 



overlap. The variable most strongly correlated 
with this function is shell length (Table 4). 

Connections Between the Species 

The possiblity of genetic exchange be- 
tween anatomically different species is dem- 
onstrated by the hybrids between P. otaheit- 
ana and P. jackieburchi from the laboratory 
crosses. In the discriminant analysis of the 
genital morphology, the parents of mating 
MJ430 lie with their respective conspecifics, 
whereas the offspring are almost exactly in- 
termediate (Fig. 4). Drawings of the genitalia 
of these hybrids, their parents, and a repre- 
sentative P. affinis are shown in Figure 5. The 
parents of the second mating (MJ431) could 
not be dissected, but the two mature offspring 
of that mating have scores on the first discrim- 
inant function that lie between those of the 
parental species. One of the offspring is close 
to the group from MJ430, but the other has a 
lower score for the second discriminant func- 
tion, placing it between P. otaheitana and P. 
affinis. Although all of them lie between the 
parental species, the hybrids span a wide 
range of discriminant scores. 

In the analysis of genital morphology, the 
"unplaced" snails also show a wide range of 
intermediate values, overlapping the specific 
groups, and bridging the gaps between them 
(Figs. 2, 3). We were able to measure the 
shells of 44 "unplaced" snails and assign 
them scores from the discriminant functions 
based on the identified groups. The relation- 
ship between the variation in genital morphol- 
ogy and the shells can be seen by comparing 
the individual scores on the first discriminant 
functions for each set of traits (Fig. 6). Taken 
together, these two functions completely dis- 
tinguish P. otaheitana, and nearly separate P. 
jackieburchi and P. affinis. The scores on the 
two functions are significantly correlated both 
for the combined sample of identifiable indi- 
viduals (r = -0.74, P<0.001) and for the 
"unplaced" snails (r = -0.57, P<0.001). 
Nevertheless, it is clear from Figure 6 that 
many of the unknowns have shells like P. ota- 
heitana but intermediate genitalia. Further- 
more, the association of the two sets of traits 
is between groups, most clearly between P. 
otaheitana and P. affinis. They are not corre- 
lated within any of the three species (Fig. 6). 

Using these analyses, we can look in detail 
at each of the samples with "unplaced" 
snails. Discriminant scores for the genital 
morphology of these snails blur the distinc- 



50 



JOHNSON, MURRAY & CLARKE 




DFl 

FIG. 4. Discriminant scores from the analysis of genital morphology. Symbols as in Fig. 3. Additional 
symbols: P = parents for mating MJ430; H = F^ from MJ430; h = F, from MJ431 . 



TABLE 3. Pooled within-groups correlations be- 
tween the traits and the discriminant functions in the 
analysis of differences in genital morphology be- 
tween P. otaheitana, P. jackieburchi, and P. affinis. 
Only traits with a correlation of at least 0.4 with one 
of the two functions are included. 



Variable 



DF1 



DF2 



WPVD 


0.743 


0.042 


WPEN3 


0.581 


-0.049 


VDRET 


0.534 


0.129 


HVD 


-0.404 


0.190 


WPEN1 


0.080 


0.489 


LPEN 


0.341 


0.477 


LSP 


0.219 


0.438 



tions between the three species, but each 
sample has its own characteristics (Fig. 7). 

Sample 801 from Papehue on the western 
side of Tahiti is the source of the P. otaheitana 
parents in the experimental matings. The 
sample has seven snails, all of which are sin- 
istral. Four of them are clearly P. otaheitana. 
One of the "unplaced" snails also falls within 
P. otaheitana, but the other two are interme- 
diate between P. otaheitana and the other two 
species (Fig. 7). 



TABLE 4. Pooled within-groups correlations 
between the traits and the discriminant functions 
in the analysis of differences in shells between P. 
otaheitana, P. jackieburchi, and P. affinis. Only 
traits with a correlation of at least 0.4 with one of 
the two functions are included. 



Variable 



DFl 



DF2 



SHLEN 
SPWID 
APWID 



0.857 
-0.039 
-0.251 



-0.451 

-0.485 

0.420 



Sample 794 is from the lower section of the 
large central valley of Papenoo. It includes 
typical P. otaheitana, but it also spans the 
range of intermediates, suggesting connec- 
tions between P. otaheitana and either P. af- 
finis or P. jackieburchi, or both (Fig. 7). With 
one exception, the individuals with intermedi- 
ate genitalia have shells that resemble P. ota- 
heitana. 

The sample from north Mahaena (780) is 
not problematical. The one doubtful individual 
is clearly P. jackieburchi, making a total of ten 
P. jackieburchi. Sample 793 from south Ma- 



p. otaheitana 




ANATOMICAL VARIATION IN PARTULA 
P. jackieburchi 





1 mm 





Hybrid 




FIG. 5. Reproductive anatomies of the parents (P. otaheitana and P. jackieburchi) and the F, hybrids of 
laboratory mating MJ430, drawn from camera lucida images. A typical P. affinis from sample 779 is included 
for comparison. 



haena, however, does have peculiar individ- 
uals. This sample contains three large sinis- 
tral snails with pink shells, taken from a high 
ridge. One lies within P. otaheitana, but the 
other two are anatomically intermediate (Fig. 
7). 

Sample 792, also from south Mahaena, is a 
more complicated mixture. With the exception 
of four variously intermediate individuals, the 
discriminant analysis of the genitalia made 
this group overlap, but offset from, unambig- 
uous P. affinis (Fig. 7). There is a range of 
shell types connecting P. affinis with the other 
species. The group Is polymorphic for the di- 
rection of coiling. Seven Individuals are dex- 
tral, including the four snails that were clearly 
P. affinis on visual inspection of their genitalia. 
These four also have shells that are typical of 
P. affinis, so they pose no problem. Tfie mul- 
tivariate analyses showed that the other three 
dextrals are also P. affinis, although the shell 
of one of them is not clearly so. 

Among the sinistrals, variation connects all 
three species. Several have genitalia similar 
to P. affinis, but most of these are displaced 
from the clear P. affinis group containing the 
dextral individuals (Fig. 7). Others have shells 
like P. otahieitana but genitalia of intermediate 
character. Within the group of sinistrals, 



scores on the first discriminant functions for 
genitalia and shells are significantly corre- 
lated (r = -0.474, P = 0.026). To examine 
this variation more closely, a separate princi- 
pal components analysis was made using the 
genitalia of Sample 792 alone (Fig. 8; Table 
5). The first axis, representing 22.5% of the 
variation, separates two of the sinistrals from 
all the others. With high loadings from LPEN, 
LFSP, WSP2, and LSG, this component is 
similar to the first component in the analysis 
of all specimens (Table 2). The high scores of 
the two distinct Individuals reflect their larger 
size and greater similarity to P. otafieitana. 
They have large, yellow shells with a pink 
apex, typical of P. o. rubescens or P. jacl<ie- 
burcfii. The second principal component 
(16.7% of the variation) confirms the differ- 
ence between the dextrals and the sinistrals. 
The dextrals, which include typical P. affinis, 
all have relatively high scores. The sinistrals, 
in contrast, span the range of scores, but are 
concentrated at the lower end (Fig. 8). A low 
score on the second component indicates a 
penis that is relatively thick in the middle re- 
gion and thin at the distal end, and a relatively 
long spermatheca (Table 5), suggesting 
some similarities to P. jackieburcfii. The snails 
with low scores tend to have shells with some 



52 



JOHNSON, MURRAY & CLARKE 




-2024 
DFl for GENITALIA 

FIG. 6. Relationship between the discriminant scores based on analyses of shells and genitalia. Symbols as 
in Figs. 3 & 4. 



yellow or pink, similar to P. o. rubescens or P. 
jackieburchi. From this analysis, it is clear that 
this Is a heterogeneous sample, which cannot 
be explained simply as aberrant P. affinis. 

The final sample with individuals that were 
difficult to identify is number 81 3, In the south- 
eastern valley Faone. This sample includes 
seven snails, only three of which could be dis- 
sected. Two shells are brown dextrals, typical 
of P. affinis. The dissected dextral also has 
genitalia typical of P. affinis (Fig. 7). The other 
five snails are large sinistrals, with the ap- 
pearance of either P. o. sinistrorsa "Pease" 
Garrett, 1884, or P. a. producta Pease, 1864, 
which are sympatric and conchologically in- 
distinguishable in southwestern portion of Ta- 
hiti Nui (Kondo & Burch, 1983). Four of these 
have the cestata banding morph, whereas the 
fifth is apex, both morphs being common in P. 



o. sinistrorsa (see Crampton, 1916). One of 
the dissected sinistrals has genitalia interme- 
diate between P. affinis and P. otatieitana, 
whereas the other is within the range of typi- 
cal P. affinis (Fig. 7). 

Taken together, these samples suggest 
connections between P. affinis and P. otatieit- 
ana, and possibly P. jackieburctii. Although 
each sample has its unique features all the 
samples with anatomically intermediate snails 
contain individuals that lie unambiguously 
within one of the three species. Thus, we 
have not found any purely intermediate pop- 
ulations. 

Comparisons Between 
Partula and Samoana 

In order to see how the differences be- 
tween the species of Partula compare with 



ANATOMICAL VARIATION IN PARTULA 



53 




794 




-6 -4 -2 




6 8 -6-4-2 

SCORE ON DFl 

Fig. 7. Discriminant scores from the analysis of the genital morphology for samples with "unplaced" snails. 
Sample codes as in Fig. 1 and Table 1. Polygons indicate areas occupied by typical P. otaheitana, P. 
jackieburchi, and P. affinis as in Fig. 4. Open circles = sinistral unplaced; filled circles = dextral unplaced; 
+ = individuals originally in the known groups. 



the differences between the genera, a dis- 
criminant analysis of the genitalia was made, 
using the four groups P. otaheitana, P. jack- 
ieburchi, P. affinis, and the combined samples 
of Samoana attenuata and S. diaphana. The 
overall separation of these groups is good, 
and all the snails were correctly placed in their 
prescribed groups. The separation on the first 
two axes is essentially the same as in the 
earlier analysis of Partula alone: P. otaheitana 
is separated from the others on the first, 
whereas P. jackieburchi and P. affinis are sep- 



arated on the second (Fig. 9). The two spe- 
cies of Samoana are intermediate but over- 
lapping with P. jackieburchi and P. affinis. 
Thus, the major separation is between the 
species of Partula, not between the genera. 
This is not surprising for P. jackieburchi, which 
was at one time placed within Samoana, but it 
was not expected for P. affinis. On the third 
discriminant axis there is partial separation of 
Samoana from P. jackieburchi and P. affinis 
(Fig. 9). The trait contributing the most to that 
separation is the relative width of the proximal 



54 



JOHNSON, MURRAY & CLARKE 



2- 



ÇVI 

и 

û- 04 




осо 



-1- 



-2 



-2 



-1 



PCI 



Fig. 8. Principal components scores for the analysis of genital morphology within Sample 792. Polygon 
encloses dextral individuals. Open circles = sinistrals; filled circles = dextrals. 



TABLE 5. Varimax factor loadings of traits in the 
principal components analysis of genital morphol- 
ogy in Sample 792. Only traits with loadings greater 
than 0.5 on either of the first two principal compo- 
nents are included. 



Variable 


PC1 


PC2 


LPEN 


0.634 


0.049 


LSP 


0.356 


-0.644 


LFSP 


0.752 


-0.200 


WPVD 


-0.101 


0.772 


WPEN1 


0.080 


0.884 


WPEN3 


-0.154 


0.845 


WSP2 


0.783 


-0.140 


LSG 


0.877 


-0.074 



section of the penis (WPEN1 ). The low scores 
of S. attenuata and S. diaphana reflect the 
stout penis with thickened middle region. 



DISCUSSION 

The complexity of variation revealed in this 
study is important both for understanding the 
radiation of Partula on Tahiti and for tackling 
general problems of snail systematics. Our in- 
terest began with Kendo's (1 980) discovery of 
a dramatically different anatomical form within 
P. o. rubescens, and his description of that 
form as Samoana jackieburchi. Comparisons 
of allozymes showed this placement to be in- 
correct, as this taxon clearly lies within 
Partula, and is genetically very similar to P. 
otaheitana and P. affinis (Johnson et al., 
1986c). Later work on mitochondrial DNA has 
confirmed the close association of these three 
species (Murray et al., 1991). 

The present study shows clearly that the 
overall differences in genital morphology are 
between the species, and not between the 



ANATOMICAL VARIATION IN PARTULA 



55 



b 


A 


■ 


д 




— 1 1 1_ 




4 


Д Д 




д д 








2 


^ ;^ Д 




Д дд ^ 




О 


О 


1 

n 


X Д ^ ^ 

X : Д 




^Х А^ Д 
л X ^ 




о о 

Og)0 


О О 



-1 
-2- 


■Ж'А i "'^ 

^A ! 


Д Х^ А 1 
^ А 




^ о 

о 


-3 


A i A 


■ 


А 








-4. 


là 


■ 


А 








-S 
















4-2 2 -6 -4 -2 С 


2 


4 


6 8 




DF3 






DF1 







Fig. 9. Discriminant scores for the analysis of genital morphology in P. otaheitana (circles), P. jackieburchi 
(open triangles), P. affinis (filled triangles), and S. attenuata and S. diaphana (X). Scores for P. otaheitana 
on the third discriminant function span a wide range, and are omitted for clarity. 



genera. There are two conclusions to be 
drawn from the comparison of Partula and 
Samoana. First, If there are consistent differ- 
ences separating the genera, we have not 
measured them. However, because the anal- 
yses within Partula discriminate the main 
groups already recognized, our chosen set of 
characters has provided a reasonable de- 
scription of the variation. The multivariate 
analyses show that the definition of the 
groups does not depend on some special 
weighting of certain "important" characters. 
The second conclusion is that, regardless of 
whether there are other anatomical differ- 
ences between the genera, there is conver- 
gence of anatomical characteristics between 
P. jackieburchi (and P. affinis) and Samoana. 
Convergence, rather than retention of ances- 
tral characteristics, is indicated by the fact 
that the Tahitian species of Partula are appar- 
ently derived from l\/loorean ancestors 
(Johnson et al., 1986b), but none of the 
Moorean species share the anatomical char- 
acteristics with Samoana (Murray & Clarke, 
1968, 1980). 

Even more interesting than this conver- 
gence is the demonstration, by the experi- 
mental matings, that snails with "generically 
different" genital morphologies can inter- 
breed, producing viable hybrids. It is signifi- 
cant in this respect that the laboratory hybrids 
between P. jackieburchi and P. otaheitana 



have intermediate morphologies. They show 
no sign of aberrant genitalia that might sug- 
gest developmental problems (cf. Murray & 
Clarke, 1980). As discussed below, the field 
results also suggest that these species can 
exchange genes, despite their anatomical dif- 
ferences. A similar situation occurs on 
Moorea, where Partula aurantia Crampton, 
1932, has a large, club-like penis, which dis- 
tinguishes it from all other species on the is- 
land, but does not prevent its hybridization 
with P. suturalis Pfeiffer, 1855 (Murray & 
Clarke, 1968). It is clear that, in Partula at 
least, differences in genital morphology have 
little impact on reproductive isolation, and do 
not necessarily have special value as taxo- 
nomic characters. In this light, we must view 
with caution the proposed taxonomic revision 
of the Tahitian Partulidae based solely on re- 
productive anatomy (Kondo & Burch, 1983). 
The complexity of the P. otaheitana group 
has long been recognized on the basis of the 
variation in their shells (Crampton, 1916). 
Rather than simplifying the complexity, our re- 
sults increase it. It is important, however, to 
exclude possible artefacts before attempting 
to interpret the multivariate patterns of varia- 
tion in genital morphology. Measurement er- 
rors, state of preservation, and reproductive 
state can have marked effects on analyses of 
genital morphology (e.g. Emberton, 1985, 
1989). Some of the variation of discriminant 



56 



JOHNSON, MURRAY & CLARKE 



scores within the clearly defined groups or 
among siblings from the laboratory crosses 
might be due to such errors. However, the 
ability of our multivariate analyses to recog- 
nize the groups described by Kondo (1968, 
1980; Kondo & Burch, 1983) indicates that 
the major variations are real. Furthermore, 
the intermediacy of the laboratory hybrids 
provides strong evidence that we are looking 
at heritable differences between groups. 
Thus, we can be confident that any spurious 
variation in our measurements is small 
enough to justify examination of the geo- 
graphical and taxonomic patterns of the vari- 
ation in the P. otaheitana group. 

Based on our analyses, it is clear that some 
combinations of species are distinct in sym- 
patry, without any sign of interbreeding. 
Partula affinis can coexist with either P. ota- 
heitana or P. jackieburchi. The situation be- 
tween P. otaheitana and P. jacl<ieburchi is not 
as clear. Tiarei is the only valley in which both 
have been found, and they are found together 
only in Sample 742. Even that case is mar- 
ginal, however. The genitalia of 34 individuals 
from that site were examined (ten of which 
were measured for this study). Only one was 
P. otaheitana, and 33 were P. jackieburchi. 
About 1 .5 km lower down the valley, near site 
776, a sample of 1 7 individuals was examined 
(but not measured), and all were P. otaheit- 
ana. Attempts to collect along a transect be- 
tween the sites were not very productive, be- 
cause the snails were scarce, but the few 
snails obtained were P. otaheitana. In our 
samples outside Tiarei, distinct P. otaheitana 
were found only to the north and west, and 
distinct P. jackiebruchi only to the south 
(Table 1). Thus, it appears that P. otaheitana 
and P. jackieburchi are, at least locally, para- 
patric replacements. However, there is some 
uncertain evidence for the occurrence of P. 
otaheitana to the south in Mahaena (see 
below), and much more sampling would be 
needed to describe the geographical distribu- 
tions of the two species. 

In contrast to the coexistence, or abrupt 
transition, between species is the existence of 
variously intermediate individuals at several 
sites. It is difficult to know how much of this 
intermediacy is due to geographic variation 
within species and how much to exchange of 
genes between species. The possibility of 
gene exchange is shown by the laboratory 
hybrids between P. jackieburchi and P. ota- 
heitana, and by the fact that in the discrimi- 
nant analysis the hybrids lie amongst the "un- 



placed" snails from the field samples (Fig. 4). 
Gene exchange is also suggested by the cor- 
relation between genital anatomy and shell 
shape among the "unplaced" snails and be- 
tween species, but not within species (Fig. 6). 
However, the strength of the evidence for hy- 
bridization differs from sample to sample. 

One difficulty is that hybrids are not easy to 
identify. Although they are intermediate in 
their anatomy,'even the sibling hybrids show 
a wide range of discriminant scores (Fig. 4). It 
is therefore difficult to separate hybrids of P. 
otaheitana and P. jackieburchi Uom hybrids of 
P. otaheitana and P. affinis. In Sample 794 
from Papenoo, for example, the snails vary 
from obvious P. affinis, with small, brown, 
dextral shells, to P. otaheitana, with large, 
pink or yellow, sinistral shells. All the individ- 
uals with intermediate genital morphologies, 
however, have shells like P. o. rubescens, 
with no sign of introgression from P. affinis. 
Since typical P. otaheitana occur on either 
side of this valley, it seems unlikely that the 
intermediates represent an unusual geo- 
graphic variant of P. otaheitana. It is not clear, 
however, whether P. otaheitana is exchang- 
ing genes with P. affinis (without any apparent 
effect on the shells) or with P. jackieburchi 
(which has not been reported from Papenoo). 

Similar problems apply to other samples. In 
Sample 801 from Papehue, for example, 
there are typical P. otaheitana and apparent 
hybrids, but the shells are all typical of P. ota- 
heitana. Furthermore, neither P. affinis nor P. 
jackieburchi is known from the western series 
of valleys. Similarly, Sample 793 from Ma- 
haena includes P. otaheitana and possible 
hybrids with P. jackieburchi, but the presence 
of P. jackieburchi has not been established. 
Although exchange of genes between spe- 
cies seems to be the most likely explanation 
for these samples, we cannot exclude the 
possibility of local differentiation. 

The most convincing evidence for hybrid- 
ization is in Sample 792, also from Mahaena. 
In this chirally polymorphic population, the 
dextrals are typical P. affinis, but the sinistrals 
show a spread between P. affinis and P. ota- 
heitana for both genital and shell morphology. 
Taken together, samples 792 and 793 sug- 
gest that a thorough search would reveal typ- 
ical P. otaheitana in Mahaena. 

Another connection between P. affinis and 
P. otaheitana is suggested by Sample 813 
from Faone, the southernmost valley in this 
study. Whereas the dextral individual is 
clearly P. affinis, with a small, brown shell, the 



ANATOMICAL VARIATION IN PARTULA 



57 



sinistrals have shells typical of P. o. sinis- 
trorsa (Crampton, 1916, plate 30), and geni- 
talia either like P. affinis or intermediate be- 
tween P. affinis and P. otaheitana. Crampton 
(1916) did not find P. o. sinistrorsa in Faone, 
but reported large numbers from the valleys 
that connect to its southern ridge. Kondo & 
Burch (1983) also found large sinistrals with 
genitalia like P. affinis in Faone. They consid- 
ered these to be the subspecies P. a. pro- 
ducta, which they say is conchologically indis- 
tinguishable from P. o. sinistrorsa. If their 
interpretation is correct, their subspecies P. a. 
affinis and P. a. producta are sympatric. In 
either case, the sinistral individual with inter- 
mediate genitalia indicates a connection be- 
tween P. affinis and P. otahieitana at the 
southern end of Tahiti Nui. 

These results pose more questions than 
they answer. Regardless of how we explain 
the existence of intermediate specimens, the 
variation in genital morphology fills the gaps 
between the currently recognized species. Al- 
though these species retain their distinctness 
in some areas, the connections demonstrate 
the complexity of the group. Faced with this 
variation, it is clear that only comprehensive 
study, based on intensive geographic sam- 
pling, dissection of large samples, and quan- 
titative analysis will resolve the relationships 
within the P. otahieitana group. These species 
are now almost certainly extinct in the wild 
(Murray et al., 1988), so that further work 
must rely on preserved specimens. 



ACKNOWLEDGMENTS 

We thank Jane Prince for the painstaking 
measurements. Financial support was pro- 
vided by the Australian Research Grants 
Scheme and the U.S. National Science Foun- 
dation (BRS 83-15097). 



LITERATURE CITED 

ATCHLEY, W. R. & D. ANDERSON, 1978, Ratios 

and the statistical analysis of biological data. 

Systematic Zoology, 25: 71-78. 
COWIE, R. H., 1992, Evolution and extinction of 

Partulidae, endemic Pacific island land snails. 

Ptiiiosoptiicai Transactions of the Royal Society, 

Ser. В., 335: 167-191. 
CRAMPTON, H. е., 1916, Studies on the variation, 

distribution, and evolution of the genus Paríala. 



The species inhabiting Tahiti. Carnegie Institu- 
tion of Washiington Publications, 228: 1-31 1 . 

CRAMPTON, H. E., 1932, Studies on the variation, 
distribution, and evolution of the genus Paríala. 
The species inhabiting Moorea. Carnegie Institu- 
tion of Washington Publications, 310: 1-335. 

EMBERTON, K. C, 1982, Environment and shell 
shape in the Tahitian land snail Paríala otaheit- 
ana. f^alacologia, 23: 23-35. 

EMBERTON, K. C, 1985, Seasonal changes in the 
reproductive gross anatomy of the land snail Tri- 
odopsis tridentata tridentata (Pulmonata: Polygy- 
ridae). Malacologia, 26: 225-239. 

EMBERTON, K. C, 1989, Retraction/extension 
and measurement error in a land snail: effects on 
systematic characters. Malacologia, 31: 157- 
173. 

JOHNSON, M. S., J. MURRAY & B. CLARKE, 
1986a, Allozymic similarities among species of 
Paríala on Moorea. Heredity, 56: 319-327. 

JOHNSON, M. S., J. MURRAY & B. С CLARKE, 
1986b, An electrophoretic analysis of phylogeny 
and evolutionary rates in the genus Paríala from 
the Society Islands. Proceedings of the Royal 
Society of London, Ser. В, 227: 161-177. 

JOHNSON, M. S., J. MURRAY & В. CLARKE, 
1986c, High genetic similarities and low het- 
erozygosities in land snails of the genus Samo- 
ana from the Society Islands, l^alacologica, 27: 
97-106. 

KONDO, Y., 1968, Partulidae: preview of anatom- 
ical revision. 7776 Nautilus, 81 : 73-77. 

KONDO, Y., 1980, Samoana ¡ackieburchi, new 
species (Gastropoda: Pulmonata: Partulidae). 
Malacological Review, 13: 25-32. 

KONDO, Y. & J. B. BURCH, 1979, Extrusive genital 
anatomies and their internal postures in Paríula 
affinis of Tahiti. fHalacological Review, 16: 101- 
106. 

KONDO, Y. & J. B. BURCH, 1983, Two amend- 
ments to Crampton's monograph on Tahitian 
Partulidae. t^alacological Review, 12: 79-84. 

MURRAY, J. & B. CLARKE, 1966, The inheritance 
of polymorphic shell characters in Paríula (Gas- 
tropoda). Genetics, 54: 1261-1277. 

MURRAY, J. & B. CLARKE, 1968, Partial reproduc- 
tive isolation in the genus Paríula (Gastropoda) 
on Moorea. Evolution, 22: 684-698. 

MURRAY, J. & B. CLARKE, 1980, The genus 
Paríula on Moorea: speciation in progress. Pro- 
ceedings of the Royal Society of London, Ser. В, 
211:83-117. 

MURRAY, J., E. MURRAY, M. S. JOHNSON & B. 
CLARKE, 1988, The extinction of Paríula on 
Moorea. Pacific Science, 42: 150-153. 

MURRAY, J., O. С STINE & M. S. JOHNSON, 
1991, The evolution of mitochondrial DNA in 
Paríula. Heredity, 66: 93-104. 

REID, D. G., 1986, The littorinid molluscs of man- 
grove forests in the Indo-Pacific region. British 
Museum (Natural History), London, 238 pp. 

Revised Ms. accepted 25 June 1 992 



58 



JOHNSON, MURRAY & CLARKE 



о 


ш 




Ü. 


X 


^ 


ш 








fl) 


о 


0) 


> 


W 


Q. 


ci 


^ 


3 




о 


СО 


U) 


< 


СП 


> 


с 


_| 


СП 








ф 


Q. 


Í- 


О) 


(П 


11 


о 


_1 


о: 


Q. 


ш 


СП 


^ 


_1 






с 


H 




ш 
ее 


о 


Q 


о 


> 


^ 




Q. 


п 


О 


■> 


b 


< 


га 


H 


с 


ш 


ф 


гг 


О) 


< 






о 




<л 


-¿ 




ш 


>< 


Ü. 


го 


_| 


с 




го 


Q 


к. 


> 


^ 


_| 


го 




m 


Q. 


Û 


U) 


X 





Q 


го 


^ 


(П 


Ш 




0. 


ш 


Ü. 




< 


со 



coco^o•^ooг^c^cN1-■^ooг^o•^0)oюooa)ooc^JЮ^-••-^.oo■•-■^(rз^n<D•r-1-coo5Ln 

OI^Or^^C\l'-COOC\JC<5^0)OiCO-^CJ)->-C3505CO'^0)CD^r^O)0)I^COCOCO<OCON.N.aD 
CMt-CVÍ-t-CVIi-t-t-CMCMt-CM т--^ ■.-,- ^^i-^,-^T-t-i--t-T-,--i-C\jT-^,- 

oo)cчJ(rîcмc^J■■-ю<o■*cocDoocDln^-oo^-•^(^loo•^o<û■^o>oc^Joo^-ocDooo)cocз5oo 

05Cn<MCnC0Oi-OC0OC0000000C0CMC0-*r^OT-C\JCDOr^<0->-00r^C0C\jmO'í-inT-C\J 
-.-•^CVJi-CMCJCVJCsJ'.-CMCNJ^T-^CVJCVJCVJ-.-T-CMCJCVJ-.-CJ-r-i-CVJf-i-T-i-T-i-T-CJCVJCVJ 



CsJOlinCVJh-'^'íinOOOiCOO 
C0C\JCOCVJCOC\JC0C0CVJC\JC0C\JCMC4JC\JCMC0 



•■-•^0>0'>-r^O'^C4jr^T-C\lCOt^CD(r)COOO-'-h-C\JOOOOOI^ 

■*01-ютгю1Г)а>тГ1Л<01Г>ооосо^а)Г^сп<ос1)Ююсо->- 

CMC4JC\JCMCO->-i-T-C\JCVJC\JC4J(r)CVJC\JCVJC\J-i-T--.-i-i-C\JC\JCO 



'-•^ОС0О00'*Г^ОЮ0>Г^1-С0т-С000ООСи(0С0О00'3-^С0001ПС0ЮС0О-^Г^<0Ю 

05^-■^-•т-^.oc^Jaзcoo5ooco(r5cзo)^^■^-o)^co^^co^-юcoo(D■*■^-c^JCЧJ1-^^•*•^ю■^ 

i-i-C\JC\J'-C\JCOC\JC\i^COCMi-'^i-->-C\J^-^i-f-->-C\JC\JC\JC\Ji-C4JCM<MC\JC\JC\J-.-C\JC\JCO 

О1-Г~-С0С0т-СМО-г-->-1ПГ^С0ОЮО'^1П1П1^<01ПГ^С0С0<0С01^С0т-ОС0С0О(0С005 
OOOOr^Or^OO>CDi-Tj-C\J^OO)<J)-^C\JOOr^O>CTlOC\JOC\J'-i-CMC\J<£)CDinr^O'^'^-^ 

00С01ПС0->-С000т--^сМт--.-Г>.Г^Гч.-.-С0С0Г^'!1-Г^<ОГ^'^''-О'4-С00)а)С\1СиС00)Ю00С0 
CDr^TtC0r^0000T-C\IC0C\JC\JC0O00<O0)i-0>00h-mc0OT--.-Lnir)OC\JC0T-C\Jl0C\J-^C0 
C\JC4ICVJC\JC\JC\JC\JCOC\IC\JCOC\JC4JCOC4JC\JC\J^'i-C\JC\JCVJCr)COC0C0C\JC\JCO-r-'i-^'i-i-COC0CO 

•■-■•-Ю<0Си1^ОС0СГ)О00Си00<М<МСЧ1С0СМ00т)-С1)т}-00С0Г^0и1-'^00СЧ1Ю1-СМ1Г)'>-1Л00 

o)■*юaзooюc^JaïoooO'-oor^<Dююк■•-■^oo<oю^*r^^^oooocDr^■■-■^-'-c\J•^■*coco 



оооооооюоююоююооотюююююотооюооооошююю 

Г^Г^ОООООО^-СОСОЮС35<01ПСООио01Г>СОО)СОСОСОСО(ОЮС01Г>ООСОООООаЭСОООСОСОСГ)СО 



•^OOOOinCDOOOOi-i-r^OOC\JO)OOOCOh-r^Cr)(DCOh-COCOi-inOO'^-'-r^Oa)OOCX305in 
(r)<DC0CMC0C\|in->-T-Tt00<£)C4ir^000)OinOt^-.-T-ininr^CDir)00h-0>'-00'^ir>a)C0m 
■^CN-^CO'^CO-^TtcOCO'^COCOCOCvICVJ-^T-CVlOJCOCOCOCOCOCOCOCVJCOT-CVJ-^CVJ'i-CO-*'^ 



ro ro ro ro ro ro ro 



^ ^ ^ ^ ^ 



^ ъи ^ \u ^ vu vu vu vu vu vu vu ^ ^ ^ ^ 
DO ЗО ЗООООООО 3 3 D 3 



J£ JÈ J¿ ^ 



го го го го го го го 



^^ ,— ,— vu vu vu vu vu vu vu Ч— Ч— **— Vf «t— vu vu vu 

^ ^ ^ ^^ i^rf ^^ ^^ ^^ ^-t ^-1 ^ ^ ^ 4^ ^ ^^ ^^ ^^ 

зззэооооооогагогогогоооо 



■i-i-T-T-T-T-0000000OCO0O-*'^-*'^TÍ-Tt'^T}-Tl-Tl-(J)a)CT)O)CT)O5O5O>a>O)O)CDCDCD'3- 

oooooot^r-h-r^:^t^CT)0>05050)a)05a)CT>ci)r^r^^-f^^-r^h-r^tv.|v.r^r^is.h.oo 



ANATOMICAL VARIATION IN PARTULA 59 

СМ0}С0С0'^00СМС0С\11-(0ЮС0->-Ют'^^1-^-'^1Г)ч-1Г)<£)Г^'^00'^Г^О1-С01-'О 
i-T)-T-OCDCOOOCOa5050T-(DC\J-*0)<»CV10)I^OC\JO)0-*05COOOCO'^ÇDOOt^O)Q4 
CNCM-^(MT-C\J-.--i-^-i-C>JCMi-C\JCvJ->-,-CM-.-i-^^t-C41CVJC\JC4jT-i-C\Ji-T-i-T-3 



.с 



cr)or^■^r^oю^-c^J<oocooo^*юoocмíDooooo>lnlr)cчJcoooooc^JЮ^-o>oco^< 

00OC\JOC0Kr--CDC0Tr00C0O0)r^^tN-O>^Tfir)'*inC\JC0Tj-|v.00CMf^OC0'^'>-Ci 

■,-C\J-.-^-^,-^-i--.-T--f--i-C\jT-'-i-T-i--t-T--t--i-C\Ji-T-l-i-^i-C\|-.-T--^T- 



C\JCJ00C\Jh-'*O-*C3>ir)<û^-CDC0h-r-C0Or^0>C\jT-C\JtnCD^'^t^-'-<DT-mC0->- 
C0-*<00)C0m0)ir)00C0'^tOO'-00i-C0CDC\JO00^t^0000C0a)0)C\JC0h-O05 
C0COCVJi--'-C4jT-C\J-i-C\JC4JC\JC\JC0CMC\JCVICMC\JCOC0T-C\Ji--t--.-i-T-,-CVjT-T-C\J-^ 



00'^(ОЮ<01-0'^Г^-^ООС010Г^Г-СОГ^->-001Г>001-СОООЮ->--^(^05ЮСЧ]Ю'>- 
1-СГ)ОС005СООСОЮЮСОООО)СГ)05СМ'^-^С001П'^0<ОСОСЧ|СО->-<00'-СОО>т- 
COC\JCVJC\J^TfTtCJ^^-.-i- C\J-.-i-C\JC\JCVJCOCVJC\lCMi-C\JC\J-.-T--.-C\J-i-CNi-i- 

СО(ОЮСО->-С01--*СОСОСОО)ООСиО)ОС\]1-Г^ОО-'-'^Ют-СО<Г)'^Г~-т-ОСО->-050 

r^oo)r^oocoinT-cvjoor^r~-oooocDoœc\j<Dcoooc\Joocooooo<oi^c\jir)i-com 



CT)O'-CD0)h-0)<ÛC0T-T-O>ÇDC000<£)in00C000C0i-mCT)'*mh-C00iinC\JOCMi- 
т-0)1-1-0005'^ЮС01-СМ00О001-->-ЮС0С00>001-1ПС0С0а)00ОС00>Г^Ю(0т- 
COCVJ^i- 1- CM-i-C\Jt-t-COCMC\JCMC\JC\JCOCMC\J(MC\J^i-^i-C\|t--i-,-C\J^i- 



■^С\](^ОЮ00ОЮ'^00Г^051ПС0С0ЮС0С0ОС0С0С00)О'*Г>-ОГ~-Ю'^Г^С0'*Г^ 
т1-ЮО->-0)1Г>^<00'*ЮЮСОООСМОС\]Юа>^-СОЮО'^СО-^0^--*1ЛО'^СО-* 
TtTtC>JC\l^C\JC\IC4JC\JC4ICVJC\JC0COC4ICO(NC\JC\JCOcr)C\IC0C\JC\JC4JC\JC\JC\JC\JC\JC\JC\JC\J 



(O^OOüOÜOOOÜüWüOÜOÜOÜCJ 

о о .2..2..î5,.52,.5..î5..î2,.55,.w .55,0 .5..52,.î5..2..î5,.î2..î2..î5. 5 зэззэззЗззэ 



^^tCDCVJC\JC\JCVJC\IC>JCMCViC\ICVJOOOOOOOOCVJCMC\JC4ÍCVJC\ICVJC\JCVJCMC4IC\JC\J 
OOOOOOTj--^TtTj-Tj-i!tTj-Tj-'^Tl-OOOOOOOOOOOOaD000505050)0505050)0)CT)0)0)0) 



60 



JOHNSON, MURRAY & CLARKE 



со 



■.-^-C0050T-i-<ÛQOC\JOOC\JCOCDO)CDOC\J<û-^<£)COO)OiO)nO 
(û<nOïCOCOC\lt^CD->-COt^m-^inmO)ir>00'*0)'*CDCOOO'a-QO-^ 

i--.-i-i--.-i--i-^CMi-T-i-T-T--.--t--^i-^CVJi-C>Ji--<-l--.-T-C4l 



^CDCDC^JOO(DCDCDЮ■*CDOOCOOOCOOЮO>^-OOOt^<£)•^C^JC\JCO 
COC\l'!t'^CO-^COCMOC\ICMT-C\J^OCOh-^-h-CDCDCDC\Ji-CVJi-C\JCO 
■t--i--.--.-t--^-i-T--i--i-T--.-'^^t--i-C4IOJ-.-i-i-T--.-i-T-t-T-i- 



^-^<£)'^ОЮ1П1-СОГ^.ЮООООО'*СОСОГ^СОСО'>-1^1П'^^С\ЛСОСО 
i-0000)00005(ÛCOOT-<J)0>CDCOO)->-COTj-r^OOKO'*C\JCMTl-CO 
C\JC\J^-.-i-i-i-i-CVJC\JCM'-i-i--'-T-COCOCO'^i-C\JCVJ'^i--i-'.--.- 



OOh-OCMi-Oir^COOOO>inOOCMCOOOCO«)CDO-*'^t^T-C4JCOC\JCOt~~ 
CMC\JCOCOC\JC\J-.-T-C\Ji-i-i-CNJi-i-C\JCMCMC\Ji- •^COCVJCOCTCNJ 



тоо)1-ч-|^оюсот-о-^юо5со'а-1Г)ог>-соооососоЮ'^1-о 

Г^-*0>С\]-*-.-1Г>00Г^О1Г)С0а5Г^-^00С\]С0О-*С000ЮО00-^СО1П 
1-СМ C\Ji--.-f-i--.-CMi-T-T-T-C\Ji-C0CVJC0-^C\J'>-i-T--f-C\Ji-C4J 



■^оос0т-соооооо>сг)а)0оог^со1пю'*г^осо<01-с\]сос\;т'^ 
coo'^a)a)'*h-ino>cr)-^i-c\jif>ooooi^'>-co'*cr)mcvji^oor^h-c4j 

C\JC\Ii- ■^■r-i-CMT-i-CMC4JC\Ji-i--^CVIC0C\J-^C\jT--.-^-t--!-i-C\J 



CD00C00>T-<OO0000«r>Cr)aD(DCDh-CDT-C0a>O^-CMC0t^O'ia><D 
C\Jir)COCVjr~-COin00ir)ir)COCOCMCDC0CO<D00CD COCOi-lDCOCVJ 



юооююоюоооююошюооооооооюооюо 

С0050)<ГЗСОСОСОО)СЗ)С\10<005СОСОСМСОСЛСОООООО>ООС01ПЮСОЮ 



oooooooooooooooooooooooooooin 
ojascoooooooooajcooooooooocooocooooooooooooooooooooococo 



C\JC0r^'^inCNJ'Î^OC0t^OLnh-05r^'^C\ja)<û''-0>OOr^'^CT>C\J 
■^С01-СиЮ'^-*-*0)Ю^»-(М0иС0-^С0Г^-'-ОГ^00Г^С0Г^ОС0Г^С0 
CMCVJCMC\JC\l'>-CVJC4JCMC\JC\JC\JCMC\ICMC\JC0'*'^''-'-C\JC\J-^CVJi-i-'^ 






333333333 



333333. ™,.™,.™,03(0CÖ(0C0(C 



Of-COO'^inoOOi-r-CO'^Oi-CVJCOh-T-CVJCO'^-'-CNjT-CNJCO'^inr^ 
CMCJCVJCOCOCOCOCO-^-^-^lOininintn •^■^ 

<NC\JCVJC\JCMC4JC\JC\JC\JCMC\JC\JC4JC\JC\IC\JCTC0CO^'^'Î--^'^'-^'^-^ 



ANATOMICAL VARIATION IN PARTULA 



61 



о CO CM CO in o) r^ 

CM 1- CsJ C\J 00 CO t^ 
■.- Cvl ■r- CM ■— Cv) -^ 



r- in 00 -^ 00 CO r^ 
O CD о Ю -^ in If 
1- •.- CM -^ 1- CM CM 



CM о 1- 1- ^ 00 CO 
CM 1^ ■>- 1^ 00 о 1^ 
1- -I- CM 1- Ч- CM T- 



o) f- о Ti- CO о r~- 

O CO OO CO CO о Tt 
CM CM CM f- -^ CM -^ 



CD en CD LO T- CO T- 
CM en O) CO CO CO 00 
CM T- T- I- ,- CO CVJ 



CO t 1- CM to t^ CM 

in 00 1- 00 1- CO 1- 
1- Ч- CO 1- CM CM CM 



4- -Si J¿ ^ ^ -sc m 

га ^ '^ '^ ^ <= s 

"^ D 3 3 3 3 о 

00 (J) -^ CD h- CM CO 



1— о CO CO CO ■^ •^ 

CT) CO 1- T- 1— (П en 
r^ r^ 00 00 oo f^ r~- 



CM 00 O) CD ^ -^ CD 
<n en о Ю CD Ю CO 
1- 1- CM -^ 1- CM CM 



1- CM CM CO CM r^ in 

CD CD о CM о (П en 
■.- 1- CM CM CM T- ■>- 



1- CM CO CO en •^ -^ 

CO ■* CM о CT) in r^ 
1- CM CVJ CO T- t- CM 



о CM о ■* CM Ю CO 

r^ en 1- CO CM CM CO 

■>- CM CO CM CM CM CO 



CO •>- en CM CO tT CO 

CO CM t^ CO r^ ^ CM 

^ CO CM CO CM CM CO 



WJ vu ^ 

с 
га 
Eo-.-,-CMcoi-cM 

о 
га 
к о о о •.- -^ 

raœo^^^^^ 
— ir^oo^S^^^ 



MALACOLOGIA, 1993, 35(1): 63-77 

GENITAL MORPHOLOGY OF CARACOLUNA LENTICULA (MICHAUD, 1831), 

WITH A NEW PROPOSAL OF CLASSIFICATION OF HELICODONTOID GENERA 

(PULMONATA: HYGROMIOIDEA) 

Carlos E. Prieto, Ana I. Puente, Kepa Altonaga & Benjamin J. Gomez 

Department of Animal Biology and Genetics, Faculty of Sciences, University of tfie Basque 
Country, P. O. Box 644, 48080-BILBAO, SPAIN 

ABSTRACT 

The genital system of Caracollina lenticula (Michaud, 1831) has been studied in many Iberian 
populations, revealing a high morphological diversity affecting mainly the stimulatory apparatus. 
The general pattern (mucous gland plus "appendix" plus dart sac) appears sometimes modified 
due to the absence of the "appendix" or the mucous gland, or even both of them simultaneously; 
whenever the "appendix" is absent, the dart sac is also lacking. Observations carried out in 
serial sections show that the mucous gland is attached to the "appendix" and that the so called 
"appendix" is an organ where secretion elaborated by the mucous gland is accumulated, thus 
corresponding to the accessory sac in the sense of Nordsieck (1987). 

Caracollina lenticula was placed in the Helicodontinae by Hesse (191 8). In this paper, a critical 
review of the classifications of the Helicodontinae (Nordsieck, 1987, Schileyko, 1991) is made. 
We agree with Nordsieck in considering the Helicodontinae to be a polyphyletic assemblage of 
genera and thus an artificial group, but there are two main points of discordance: Ciliella is 
related to Hygromiinae (Hygromiidae) on the basis of its anatomy and shell microsculpture, 
which implies a nomenclatorial change for the Nordsieck's "Ciliellinae," once Ciliella is excluded. 
Moreover, all genera of this group, including Caracollina and Oestophora (which were errone- 
ously considered devoid of accessory sac), have a dart sac with accessory sac and mucous 
gland (except secondary losses) and, therefore, a subdivision based on the stimulatory appa- 
ratus alone is unjustified. Consequently, Schileyko's classification of this group in four subfam- 
ilies is also rejected. 

We propose the division of the "Helicodontinae" into two unrelated families, Helicodontidae 
and Trissexodontidae. The inclusion of Helicodontidae in the superfamily Hygromioidae is un- 
clear, because it has a penial caecum and lacks a penial papilla, whereas Trissexodontidae is 
considered a primitive taxon of Hygromioidea, and the general pattern of its stimulatory appa- 
ratus next to the plesiomorphic condition of Hygromioidea. 

Key words: Helicodontidae, Thssexodontidae, Caracollina, anatomy, morphology, classifica- 
tion. 



INTRODUCTION 

Caracollina Beck, 1837, is a typical Medi- 
terranean genus; its unique species, C. len- 
ticula (Michaud, 1831), is circummediterra- 
nean (Forcart, 1965), also being present in 
the Canary Islands, Azores, Madeira and 
Cape Verde islands (Backhuys, 1975). 

Caracollina lenticula is an almost unmistak- 
able species; its shell has been fully de- 
scribed by many authors (see below). Its gen- 
ital morphology is also characteristic, but it 
shows several morphs. On the other hand, 
many published interpretations of its genital 
system, mainly concerning the "appendix" lo- 
cated on the dart sac, are discrepant. 

In spite of these disagreements, no studies 
on variability and taxonomy of С lenticula 



have been published, and its systematic po- 
sition has remained in the Helicodontinae 
from Hesse (1918) until Nordsieck (1987), 
who proposed the new tribe Caracollinini, 
placing it together with the Ciliellini and the 
new tribes Trissexodontini and Oestophohni 
in the subfamily Ciliellinae. Nordsieck (1987) 
divided Hesse's Helicodontinae into two sub- 
familes: Ciliellinae and Helicodontinae. More 
recently, Schileyko (1991) reunited these two 
subfamilies into the Helicodontidae, and he 
raised Caracollinini to subfamilial rank, the 
Caracollinae. 

Routine dissections carried out to identify 
material collected to study the geographic dis- 
tribution of С lenticula on the Iberian Penin- 
sula (Puente et al., 1990) have provided new 
information about its genital morphology and 



63 



64 



PRIETO ET AL. 




FIGS. 1 , 2. Shell microsculpture of Caracollina lenticula. (1) Protoconch; scale, 100 \x.m. (2) Protoconch and 
first whorl of the teloconch; scale, 50 \хт. 



have allowed us to reevaluate the nature of 
the "appendix" or "upper stylophore" and to 
suggest a new classification of the Helicodon- 
tinae sensu Hesse, 1918. 



MATERIAL AND METHODS 

The studied material of C. lenticula has been 
listed in Puente et al. (1990). Additional ma- 
terial from three localities in Jaén province 
has been studied: Vilches-Guadalén: 3 km 
(VH5427), Martos (VG1575), and Jimena 
(VG5688). Specimens were drowned before 
being preserved in 70% ethanol. Fresh dis- 
sected genital systems of some specimens 
from Jérica (Valencia, YK0620) were fixed 
in Bouin's fluid (Culling, 1974), dehydrated 
with alcohol and embedded in parafin wax; the 
genital organs between the free oviduct and 
atrium were serially sectioned at 8 |xm and 
stained with Masson's Haemalum in combi- 
nation with picroindigocarmine (Martoja & 
Martoja-Pierson, 1970) for histological obser- 
vations. 



DESCRIPTION 
Caracollina lenticula (Michaud, 1831) 

Shell 

Bibliographical Data: Michaud (1831: 43; pi 
15, figs. 15-17); Moquin-Tandon (1855, t. II 
109; Atlas: pi. 10, figs. 15, 16); Haas (1929 
241, fig. 74); Germain (1930: 236; pi. 3, figs 



69-71 ; pi. 1 2, figs. 355, 356); Nobre (1941 : 85; 
pi. 1 5, fig. 9; pi. 1 6, figs. 4-6); Zilch (1 960: 693, 
fig. 2418); Gasull (1965: 59); Backhuys (1975: 
223; p. 27, figs. 79-80); Gasull (1975: 103; p. 
3, fig. 31bis); Mateo (1978: 13; fot. 14); Ker- 
ney, in Kerney et al. (1983: 304 + flg.). 

Comments: The examined material agrees 
conchologically with most of the descriptions 
listed above and, therefore, a new shell de- 
scription is omitted here. (An error must have 
occurred in Michaud's original description, 
because he states "sept tours de spire," but 
only 4.5 whorls can be counted in his figure.) 
The shell microsculpture, which has remained 
unknown until now, is described. 

Shell Microsculpture (Figs. 1,2): The proto- 
conch has one whori and is characteristically 
sculptured by small, regulariy interrupted spi- 
ral crests; from the beginning of the telo- 
conch, these crests change gradually to form 
a delicate reticulated microsculpture, which is 
superposed on the typical longitudinal ribs. 



Radula 

Bibliographical Data: Hesse (1931 
Giusti (1970: 102; pi. 14, figs. 1-3). 



49); 



Genital System 

Bibliographical Data: Moquin-Tandon (1855, 
t. II: 109; Atlas: pi. 10, fig. 14); Schuberth 
(1892: 9; pi. 1 , fig. 9); Hesse (1918: 104); Ger- 
main (1930: 235; fig. 182); Hesse (1931: 49; 
pi. 7, fig. 61a-d); Odhner (1931: 84; fig. 36); 



MORPHOLOGY OF CARACOLLINA 



65 



Ortiz de Zarate & Ortiz de Zarate (1961 : fig. 
3); Giusti (1970: fig. 20); Nordsieck (1987: 30; 
fig. 22); Schileyl<o (1991: 208; fig. 8-XVIII). 

Description (Figs. 3-7, 12): Right ommato- 
pfiore retractor muscle between penis and va- 
gina. Atrium, two to four times longer than 
wide, with an enlarged proximal part and, 
usually, an outside visible fold; on the oppo- 
site side, around the penial orifice, there is 
internal ring-shaped fold showing some volu- 
minous sub-epithelial goblet-gland cells with 
narrow necks that open on the epithelial sur- 
face (Fig. 12). The penis is cylindrical, with an 
enlarged distal part, twisted above the atrium, 
and covered by a penial sheath. In the prox- 
imal end of the penis, there is a very small, 
slender and elongate penial papilla, which is 
perforated by a central duct. The penial re- 
tractor muscle is attached to the diaphragm. 
The epiphallus is cylindrical, one to three 
times the penis length, usually double, and 
elbow-shaped at its middle. There is no fla- 
gellum, and the epiphallus/vas deferens tran- 
sition is evident. The vas deferens is enlarged 
at its origin and decreases gradually distally. 
The vagina is thicker than the penis and has 
an evident muscular protuberance in its distal 
third, which constitutes a low, broad dart sac 
containing a small dart. The dart is very small, 
hook-shaped, with a furrow on its convex side 
(Fig. 6). The external surface of the dart sac 
has an U-shaped muscular crest with the U 
branches directed towards the oviduct; from 
the и vertex arises an "appendix," very slen- 
der at its insertion on the dart sac but greatly 
enlarged distally, cylindrical, muscular and 
bent. In the proximal third of the vagina, there 
is a single mucous gland, generally bifurcated 
at the middle; the mucous gland duct is at- 
tached to the vagina wall until it communi- 
cates with the "appendix" duct. The bursa 
copulatrix is very small, oval or rounded in 
shape, with a slender duct one to two times 
the penis length. The free oviduct, which is as 
long as the atrium, is progressively enlarged 
from the insertion of the bursa copulatrix duct 
to the separation of the broaded vas defer- 
ens. Running along the free oviduct and the 
proximal part of the vagina, there is a muscu- 
lar band originating from the spermoviduct 
that ends attached to the vagina wall. 

Other Morphoiogies (Figs. 8-11): Besides 
the morphology of the genital system de- 
scribed above, which is the most frequent and 
the only one that exists in most of the popu- 



lations examined, some modifications in the 
stimulatory apparatus have been observed. 

(1) Very reduced mucous gland (Fig. 10): 
The mucous gland appears as a small rudi- 
ment; the other parts appear unaltered. It has 
been observed from Plasenzuela (Cáceres 
province, QD5462). 

(2) Absence of mucous gland (Fig. 8): This 
has been observed in three of four specimens 
collected from Porcuna-Bujalance (Jaén 
province, VG9492); in two specimens from 
the same locality, the other parts of the stim- 
ulatory apparatus appear unaltered, but in the 
third, the "appendix" is reduced to a small 
swelling. 

(3) Absence of "appendix" (Fig. 11): Five 
out of ten specimens examined from Vilches- 
Guadalén (Jaén province, VH5427) show very 
variable forms of mucous gland — bifurcate, bi- 
furcate but with reduced branches, simple — 
but both the "appendix" and the dart sac are 
absent. In these specimens, the vagina is 
much shorter than in those specimens from 
the same locality with complete stimulatory 
apparatus (four out of ten examined speci- 
mens). 

(4) Absence of both mucous gland and "ap- 
pendix" (Fig. 9): The simultaneous absence 
of both structures is accompanied by a short- 
ening of the vagina, which causes alterations 
in the proportions of the genital system: the 
penis/atrium + vagina length ratio is 1/1, in 
contrast to 1/1 .5-2.5 in typical specimens. As 
in the previous case, the absence of "ap- 
pendix" is related to the lack of dart sac. This 
morphology has been observed in one out of 
ten examined specimens from Vilches-Guad- 
alén, one of the four specimens collected 
from Porcuna-Bujalance, and in all the 14 
adult and subadult specimens from La 
Guardia de Jaén (Jaén province, VG3977). 

Histological Observations (Fig. 13): The 
proximal portion of the vagina has a thick 
muscular and connective wall, with muscular 
fibres oriented in any direction; the low-co- 
lumnar epithelium is folded, becoming cuboi- 
dal towards the distal portion, where the vag- 
inal wall enlarges laterally due to the 
presence of a thick dart sac (Fig. 13a). 

The mucous gland wall consists of a single 
high-columnar epithelium, the cells of which 
have many small mucous secretory vesicles 
concentrated in the apical region; these ves- 
icles seem to be detaching from the epithelial 
cells towards the mucous gland lumen, which 
is full of mucus. A very thin wall of mainly 



66 



PRIETO ET AL. 




FIGS. 3-7. Genital system of Caracollina lenticule. (3) Dalias (Almería, WF1174). (4) Tavira (Algarve, 
PB201 1 ). (5) El Villar (Huelva, PB9974). (6) Dart from a specimen of Jerez de la Frontera (Cádiz, QA51 63). 
(7) Scheme of the stimulatory organ. Abbreviations: as, accessory sac; b, bursa copulathx; bd, bursa 
copulatrix duct; d, dart; ds, dart sac; ep, epiphallus; mg, mucous gland; p, penis; pr, penial retractor muscle; 
V, vagina; vd, vas deferens; vm, vaginal muscle. Scale, 1 mm. 



MORPHOLOGY OF CARACOLLINA 



67 




FIGS. 8-1 1 . Defective genital systems of Caracollina lenticule. (8) Porcuna-Bujalance (Jaén, VG9492), 
without mucous gland. (9) La Guardia de Jaén (Jaén, VG3977), without mucous gland or accessory sac. (1 0) 
Plasenzuela (Cáceres, QD5462), with rudimentary mucous gland. (11) Vilches-Guadalén: 3 km (Jaén, 
VH5427), without accessory sac. Scale, 1 mm. 



connective tissue surrounds the epithelium 
(Fig. 13a). 

The wall of the "appendix" is thick and 
mainly muscular, with dense muscular fibres 
mostly circularly oriented; the epithelium is 



cuboidal, lacking secretory cells (Fig. 13a). 
Nevertheless, the lumen of this organ is full of 
secreted material with the same mucous ap- 
pearance as the mucous gland secretions. 
The base of the mucous gland is a narrow 



68 



PRIETO ET AL. 




FIG. 12. Two histological sections of the genital atrium and penial distal region of a Caracollina lenticula 
specimen from Jérica (Valencia, YK0620) (left, upper section). Abbreviations: af, annular fold; ga, genital 
atrium; gc, goblet-gland cells; ip, inner penis; pp, penial papilla; pr, penial retractor muscle; ps, penial sheath; 
pw, penial wall. Scale, 100 p-m. 



duct through which the secretory products, 
elaborated in the upper region, are dis- 
charged; the epithelial cells have lost their 
glandular nature becoming cuboidal (Fig. 
13b). This secretory duct fuses with the vag- 
inal wall over the lateral thickening and runs 
within the vaginal wall as a duct totally inde- 
pendent of the vaginal lumen, which is sur- 
rounded by connective and muscular walls 
(Fig. 13c). More distally, the "appendix" itself, 
after being bound by muscular bands, fuses 
with the vagina and, after a short distance in 
which three lumina run together, the mucous 
gland duct flows into the lumen of the "ap- 
pendix" duct (Fig. 1 3d-f ); close to the junction 
of both ducts (approximately, 25 ixm out- 
wards), the upper end of the dart sac cavity 
begins to appear. The lumina of dart sac and 
"appendix" duct are covered by dense mus- 
cular fibres, mostly circularly oriented, and 
both are embedded In the enlarged vaginal 
wall (Fig. 13g). The "appendix" duct evagi- 
nates into the dart sac cavity, until the former 
becomes a very narrow duct that opens into 
the hollow side of the dart (Fig. 13h-i); the 
opening of the "appendix" duct is controlled 
by a thickening of the connective tissue of its 
walls, which operates as a terminal valve. 



DISCUSSION 

Morphological Diversity of the Genital 
System of С lenticula 

As it has been stated above, the genital 
system of С lenticula shows distinct morphol- 
ogies affecting mainly the stimulatory appara- 
tus. The most frequent morphology is the 
presence of a complete stimulatory appara- 
tus, that is dart sac plus "appendix" plus 
forked or simple mucous gland. The different 
descriptions of the stimulatory apparatus 
mentioned in the literature and in the material 
studied are listed in Table 1 . 

The only descriptions in the literature not 
observed among our specimens is that de- 
picted by Moquin-Tandon (1855, t. II: 109): 
"Point de poche a dart. Une seule vésicule 
muqueuse simple, vermiforme, flexueuse, a 
peine renflée au sommet (...). Vagin assez 
développé, se dilatant brusquement en un 
corps irrégulièrement obové, un peu au des- 
sous de la vésicule vermiforme," and that by 
Germain (1930: 235): "1 seule glande multl- 
fide simple, vermiforme, flexueuse (...); pas 
du sac du dard." Although Moquin-Tandon 
stated that there is no dart sac, he mentioned 



MORPHOLOGY OF CARACOLUNA 

.3 b 



69 




FIG. 13. Microscopical sections of the vaginal structures of Caracollina lenticule of a specimen from Jérica 
(Valencia), (a) Mucous gland, accessory sac and vagina sections, (b) Conversion of the mucous gland into 
a mucous gland duct, (c) Fusion of the mucous duct with the vagina wall, (d) Binding of the accessory sac 
to the vagina wall by muscular bands, (e) Fusion of the accessory sac to the vagina wall, (f ) Flowing of the 
mucous duct into the accessory sac duct, (g-i) Accessory sac duct running into the hollow side dart. Symbols: 
1 , lumen of the vagina; 2, mucous gland and mucous gland duct; 3, accessory sac and accessory sac duct; 
3', accessory sac duct below its fusion with the mucous gland duct; 4, dart sac lumen with the dart. Scale, 
100 M-m. 



a well-developed vagina with a strong dilata- 
tion, which can only correspond to the dart 
sac. This suggests that the "appendix" could 
had been accidentally lost during the dissec- 
tion (due to the narrowness and extreme fra- 



gility of the lower part of the "appendix") be- 
cause, according to our observations, the lack 
of the "appendix" is always related to the ab- 
sence of the dart sac and reduction of the 
vagina length. 



70 



PRIETO ET AL. 



TABLE 1 . Bibliographical descriptions of the genital system of C. lenticula. 



Appendix 



Mucous gland 



References and searched localities 



PRESENT 



ABSENT 



BIFURCATE 



SIMPLE 

BIFURCATE 
SIMPLE 



Schubert (1892): Tanger, Barcelona 

Hesse (1931): Oran, Mallorca, Tenerife (v. major) 

Odhner (1931): Canary Islands 

Giusti (1970): Pianosa Island 

Hesse (1931): Palermo, Tenerife, Gran Canaria 

О. Zarate & О. Zarate (1961): La Rábida (Huelva) 

Soos (1933)( + ): Maltese Islands 

Moquin-Tandon (1855): S-France 

Germain (1930)(*): S-France 



( + ) taken from Ortiz de Zarate & Ortiz de Zarate (1961) 
(*) who states "quelquefois bifide" also. 



We have also noticed other variations not 
described before, such as a extremely re- 
duced mucous gland, the lack of mucous 
gland, and the simultaneous absence of both 
mucous gland and "appendix." 

Defective morphologies of the stimulatory 
apparatus have been observed in specimens 
from three localities, all of them in Jaén prov- 
ince, although specimens from intermediate 
and neighbouring localities have complete 
stimulatory apparatus. These observations 
suggest a tendency towards the reduction of 
the stimulatory apparatus in this area; it is 
even completely absent in all the 14 adult and 
subadult specimens sampled from La 
Guardia de Jaén. We consider that the dis- 
tinct described morphologies are within the 
scope of the polymorphism of С lenticula. 
Nevertheless, we cannot exclude the possi- 
bility that the specimens without stimulatory 
apparatus could constitute a local subspecies 
and, thus, the intermediate morphologies 
would correspond to intermediate forms. In- 
tensive sampling from the Jaén area should 
be made to solve this question. 

Interpretation of the "Appendix" 

Authors dealing with the genital system of 
С lenticula have given different names to the 
"appendix" on the dart sac, as a result of dif- 
ferent interpretations of this organ. Schuberth 
(1892) regarded it as a somewhat extended 
dart sac, whereas Odhner (1931) mentioned 
a long muscular appendix, and Hesse (1931) 
an appendicula. Giusti (1970), in a drawing of 
the genital system, pointed out a vaginal di- 
verticulum, and Schileyko (1973) considered 
it as a second mucous gland. Recently, Nord- 
sieck (1987) indicated that С lenticula has no 



accessory sac near the dart sac, although 
there is a dart sac appendix. Finally, Schi- 
leyko (1991) emphasized that Caracollina 
"posseses a pair of stylophores," the upper 
stylophore (= "appendix") being modified 
into a hydrostatic pump. 

Our observations suggest that the muscu- 
lar "appendix" is an organ where the secre- 
tion elaborated by the mucous gland before 
copulation is stored. The opening of the ter- 
minal valve of the "appendix" duct allows the 
mucous secretion to flow into the hollow dart 
face. During mating, this secretion would be 
injected into the haemocoel of the partner 
through the dart injuries, accompanied by the 
simultaneous contraction of the muscular wall 
of the "appendix," in order to stimulate the 
copulation or to reduce the courtship duration, 
as it has been stated in other stylommato- 
phores (Tompa, 1984; Adamo & Chase, 
1990; Gómez, 1991). On the other hand, the 
secretions of the goblet-gland cells located in 
the penial opening seem to aid sperm trans- 
fer. 

Thus, the muscular "appendix" of С lentic- 
ula corresponds to the accessory sac in Nord- 
sieck's terminology. This conclusion is in con- 
trast to Schileyko's idea, regarding the 
"appendix" in Caracollina as a modified upper 
stylophore. In the remaining Hygromioidea, 
the homologization of the upper stylophores 
(never with darts) with true dart sacs, pro- 
posed by Schileyko (1991), is very doubtful. 
In this sense, the structure and function here 
shown for Caracollina and Hygromia (Prieto & 
Puente, in press-2) lead us to support Nord- 
sieck's (1987) hypothesis, which considers 
the upper sacs as accessory sacs, directly 
and primarily originated for the accumulation 
of mucous gland secretions. 



MORPHOLOGY OF CARACOLLINA 



71 



Critical Review of the Classifications of 
the Helicodontoids 

The first anatomical diagnosis for Helico- 
dontinae, as a subfamily of Helicidae, was pro- 
vided by Hesse (1918), and included genera 
with a dart sac (Oestophora Hesse, 1907; 
Drepanostoma Porro, 1836; and Mastigophal- 
lus Hesse, 1918), as well as genera lacking a 
dart sac {Helicodonta Férussac, 1819; Cana- 
hella Hesse, 1918; Caracollina; Soosia Hesse, 
1918; and Trissexodon Pilsbry, 1895), plus 
some incertae sedis {Helix buvignieri Mi- 
chaud, H. hispánica Gude, and H. turriplana 
Morelet, among others). Some statements 
about these genera have been later corrected: 
Hesse (1931, 1934) considered that Caracol- 
lina is monotypical and possesses a dart sac 
with dart, which was figured by Odhner (1 931 ), 
and that Drepanostoma and Lindholmiola 
Hesse, 1931, do not have a dart sac. 

Later, Gittenberger (1968) showed that Tris- 
sexodon has a dart sac with dart and a mus- 
cular ligament between the stimulatory appa- 
ratus (dart and accessory sacs, and 
sometimes the base of the mucous gland) and 
the spermoviduct, and he suggested a relation 
between the mucous gland and accessory 
sac. He proposed to divide Helicodontinae into 
two groups that might be unrelated subfami- 
lies, although these were neither named nor 
formalized. The first group would include 
Oestophora, Mastigophallus, Oestophorella 
Pfeffer, 1929, Trissexodon, and perhaps Cil- 
iella Mousson, 1872, whereas Helicodonta, 
Drepanostoma, Lindholmiola, Átenla Gitten- 
berger, 1968, Soosia, and perhaps Caracol- 
lina would constitute the second. 

Schlleyko (1978: 57) considered Helico- 
dontidae as a family within Helicoidea, and 
recognized its heterogeneity, subdividing it 
into four groups headed by Trissexodon, 
Lindholmiola, Helicodonta, and Oestophora, 
respectively. In contrast, Nordsieck (1987) 
recognized two unrelated lines within "Helico- 
dontinae" (= Helicodontidae sensu Schl- 
leyko), Ciliellinae and Helicodontinae, both 
belonging to Hygromiidae. This reorganiza- 
tion agrees in outline with the groups sug- 
gested by Gittenberger, except in including 
Caracollina in the Ciliellinae (approximately 
corresponding to Gittenberger's first group) 
and Soosia into Eloninae (Xanthonychidae). 
The Ciliellinae was divided into four tribes: 
Trissexodontini (with dart sac and accessory 
sac, and a small dart), Oestophorini (without 
accessory sac, with dart sac and darts of 



different sizes, or lacking dart sac), Caracol- 
linini (with dart sac, without accessory sac, 
but with an appendix, and a very small dart) 
and Ciliellini (without stimulatory apparatus at 
all). The Helicodontinae was divided into two 
tribes: Helicodontini (dart sac transformed 
into an appendix, without dart, and the penial 
retractor muscle arising from the columellar 
muscle) and Lindholmiolini (without appendix, 
the penial retractor muscle arising from the 
diaphragm). According to Nordsieck (1987), 
the unique characteristics that relate both 
subfamilies are the depressed shell and the 
tendency towards the reduction of the stimu- 
latory apparatus, both conditioned by the 
endogeous way of life. We agree with Nord- 
sieck's classification in recognizing two unre- 
lated groups, which will be substantiated fur- 
ther as two families within Hygromioidea, and 
in the generic composition of each group, with 
an exception for Ciliella. 

Three features permit us consider the Cil- 
iella does not belong to the helicodontoid 
groups: 

(1) The genital system, with a broad penis, 
wrinkled tongue-shaped penial papilla and 
short, enlarged flagellum, with a short vagina 
without stimulatory apparatus and with a wide 
bursa copulatrix duct (Manganelli et al., 
1989), is not related to any genus of these 
groups. 

(2) The shell surface is covered by numer- 
ous radially arranged, nail-like scales and 
rows of minute longitudinal crests (Manganelli 
et al., 1989), which is very similar to the shell 
surface of two Hygromiidae genera: Cryp- 
tosaccus Prieto & Puente (Prieto & Puente, in 
press-1 ) and Mengoana Ortiz de Zarate, 1 949 
(Outeiro, 1988). This characteristic is not 
present in any helicodontoid genus. 

(3) The habitat and way of life of Ciliella are 
clearly distinct from those of the helicodon- 
toids; it lives on vegetation near streams in 
montane habitats (Germain, 1930; Kerney et 
al., 1983; personal observations) as do other 
species of Hygromiidae, e.g., Hygromia, Men- 
goana or Euomphalia. 

Therefore, we consider the Ciliella belongs 
to Hygromiidae and is close to Hygromiinae. 
This possible new systematic placement of 
Ciliella would require nomenclatorial changes 
in the classifications of both Nordsieck and 
Schlleyko: the "Ciliellinae" of Nordsieck 
(1987), minus Ciliella, should be named Tris- 
sexodontinae, and the "Ciliellidae" of Schl- 
leyko (1991), minus Ciliella, should be named 
Halolimnohelicidae. 



72 



PRIETO ET AL. 



Nevertheless, we disagree with Nordsieck's 
diagnosis for Oestophorini and Caracollininl. 
The former has a stimulatory apparatus con- 
sisting of a dart sac with a little dart, and a large 
accessory sac (Manga, 1983; unpublished 
data), contrary to the large dart sac with a long 
dart inside it figured by Nordsieck (1987: fig. 
21 ) based on an erroneous drawing of Oesto- 
phora barbula (Rossmässler, 1838) by Schil- 
eyko (1971); Caracollininl, as indicated by 
Schileyko (1991) and shown above, is char- 
acterized by having a long accessory sac in- 
stead of an appendix. Therefore, the diagnosis 
for both Oestophorini and Caracollininl agree 
with the one for Thssexodontini and, thus, Nor- 
dsieck's tribal division is not longer valid. 

Recently, Schileyko (1991) included Ciliel- 
linae and Helicodontinae sensu Nordsieck 
(excluding Ciliella and Canariella) plus Soosia 
within Helicodontidae, a family of Hygromio- 
idea. The reconstruction of the evolutionary 
pathways of Helicodontidae and its division 
into subfamilies and tribes made by Schileyko 
are unsatisfactory in many aspects: 

(1 ) The attachment point of the penial re- 
tractor muscle is unclear in the hypothetical 
hygromioid ancestral form: it appears attached 
to the diaphragm in Schileyko's figs. 2-III and 
5-III, and to the columellar muscle in his figs. 
8-1 and 9-1. Moreover, the penial retractor mus- 
cle reverses once more to appear attached to 
the diaphragm in his figs. 8-11 (scheme of ev- 
olution of the Ciliellidae) and 9-11 (scheme of 
the Hygromiidae); within the Helicodontidae, 
Schileyko suggests a very unparsimonious 
way to explain the presence of a penial-col- 
umellar muscle in Helicodontinae, with parallel 
reversions to a penial-diaphragmatic muscle 
in all the remaining subfamilies. 

(2) In Schileyko's fig. 8, both Caracollina 
and Trissexodon derive from Mastigophallus, 
but in his classification, Caracollina is sepa- 
rated as a subfamily from Trissexodontinae 
(with Mastigophallus and Trissexodon). 
Doubtful as well is the derivation of Gittenber- 
geria Schileyko, 1991, and Helicodontinae 
from an "intermediate link" common to both, 
suggesting a close phylogenetic relationship 
for them, when Schileyko (1991: 206) sup- 
poses that "the roots of the origin of Gitten- 
bergeria should be looked for among the 
forms close to Trissexodon. " 

(3) The most important criticism is that 
some genital schemes utilized by Schileyko 
are erroneous. The case of Oestophora has 
been mentioned before; another example is 
his representation of the genital system of Git- 



tenbergeria turriplana (Schileyko, 1971). We 
have observed in this species a single bir- 
ramous mucous gland flowing into the vagina 
and, by a narrower duct, also into the long 
accessory sac, which is in turn flowing into 
the vaginal side of the dart sac. Within the 
dart sac, an annulated papilla, located below 
the insertion point of the sac accessory has 
been observed; no dart has been found. The 
dart and sacs accessory are apically con- 
nected with the spermoviduct by means of a 
conspicuous muscular ligament (unpublished 
data). 

A Proposed New Classification 

As a result of these comments, we believe 
that previous classifications are unsatisfac- 
tory in both nomenclatorial and diagnostic as- 
pects, and we propose a new one for the he- 
licodontoid genera. 



HELICODONTIDAE Kobelt, 1904 

Diagnosis: Shell planorboid (although some 
genera have a depressed shell) with very 
open umbilicus and a smooth surface always 
with hairs. Genital system with a sac (absent 
in Lindholmiola, Soosia and Atenia) without 
dart; one undivided mucous gland beside the 
sac; penis covered by a sheath, with a small 
caecum between the slender proximal and 
the widened distal parts of the penis (Gitten- 
berger, 1968, ior Atenia; Prieto, 1986: fig. 7B, 
Gittenberger et al., 1970: fig. 183, and Nord- 
sieck, 1 989, for Helicodonta; Schileyko, 1 971 : 
fig. 2-IV, for Lindholmiola); there is neither pe- 
nial papilla nor flagellum. Penial retractor 
muscle attached to the columellar muscle, but 
to the diaphragm in Lindholmiola; the attach- 
ment point is unknown for Atenia (Gitten- 
berger, 1968). 

Geographic distribution: Central and south- 
ern Europe, with one genus extending to the 
Iberian Mediterranean region {Atenia), where 
it is endemic. 

Composition: Helicodonta Férussac, 1819; 
Drepanostoma Porro, 1836; Falkneha Nord- 
sieck, 1989; Lindholmiola Hesse, 1931; Ate- 
nia Gittenberger, 1968; and perhaps Soosia 
Hesse, 1918. 

Comments: The following features appear to 
be synapomorphic: planorboid shell; absence 
of dart sac; undivided mucous gland; penis 



MORPHOLOGY OF CARACOLUNA 



73 



with a small caecum and lacking both 
flagellum and penial papilla. The lack of 
these structures is convergent with other 
groups: the dart sac is absent in some 
Hygromiidae (Euomphaliinae, Metafruticicoli- 
nae, and some Trichiinae and Hygromiinae, 
and Ciliella) and in one genus of Trissex- 
odontidae (see below); either the flagellum or 
the penial papilla are absent in some genera 
of Trissexodontidae, and neither of the two is 
present in Oestophora (Schileyko, 1971). 
The most striking feature is the presence of 
a small caecum, which is unknown in the 
remainder Hygromioidea, and could be the 
main synapomorphic character for this family. 
It is not clear whether the penial-columellar 
retractor muscle is synapomorphic for Helico- 
dontidae (modified secondarily to a penial- 
diaphragmatic muscle in Lindholmiolinae) or 
for Helicodontinae only (and unchanged in 
Lindholmiolinae). It is also unclear whether 
the dartless sac is homologous to the dart 
sac, as suggested by Nordsieck (1987), or to 
the accessory sac, although Schileyko (1991) 
considers it to be a small branch of the 
mucous gland. In any case, the relationships 
of Helicodontidae with Hygromioidea are not 
well supported, and both taxa could be 
unrelated. 

The systematic position of Soosia is doubt- 
ful; Nordsieck (1986, 1987) considers it to 
belong to the Eloninae (Xanthonychidae, 
Helicoidea), whereas it is related to Heli- 
codontinae by Schileyko (1991). The defec- 
tive genital system of Soosia, which lacks ac- 
cessory sac, mucous glands and flagellum, 
makes its systematic placement difficult, but 
the morphology of its genital system, penial- 
diaphragmatic retractor muscle, shell mor- 
phology and geographic distribution (Grossu, 
1983) suggest a probable relationship to Lind- 
holmiola. 

Helicodontidae can be divided into two sub- 
families, as already proposed by Schileyko 
(1978): 

HELICODONTINAE Kobelt, 1904 

Diagnosis: Planorboid shell. Genital system 
with accessory sac, tubular mucous gland; 
penial-columellar retractor muscle; inner pe- 
nis (only known for Helicodonta) with spinu- 
lose semicircular folds and a long, strong, lon- 
gitudinally divided distal pleat (Schileyko, 
1971, 1978, 1991). Chromosome number n 
= 27? (only known for Helicodonta; Rainer, 
1967). 



Composition and Comments: Helicodonta, 
Drepanostoma and Falkneria. Atenia seems 
to be related to these genera because of its 
planorboid shell, tubular mucous gland and 
geographic distribution, but the absence of 
accessory sac, a condition of Lindholmioli- 
nae, together with the unknown insertion of 
the penial retractor muscle, make its system- 
atic placement difficult. The synapomorphic 
features of this group appear to be the plan- 
orboid shell and the penial-columellar retrac- 
tor muscle, although this last character is con- 
sidered plesiomorphic for Hygromioidea by 
Schileyko (1991), as it has been previously 
discussed. 

LINDHOLMIOLINAE Schileyko, 1978 

Diagnosis: Lenticular shell. Genital system 
with a corrugate mucous gland (absent in 
Soosia), without accessory sac; penial-dia- 
phragmatic retraction muscle; inner penis with 
small flaccid folds. 

Composition and Comments: Lindholmiola 
and perhaps Soosia (see above). The syn- 
apomorphic features of this group are the ab- 
sence of accessory sac (convergent with Ate- 
nia) and the corrugation of the mucous gland. 

TRISSEXODONTIDAE Nordsieck, 1987 

Diagnosis. Shell regularly ribbed and flat- 
tened, never with hairs. Genital system with 
an accessory sac, usually long and large, 
flowing into the dart sac (except in Gasulliella 
Gittenberger, 1980, in which the stimulatory 
apparatus is completely absent), with their 
upper ends connected to the spermoviduct by 
a muscular ligament (except in Caracollina, in 
which it is attached to the vagina wall; it has 
not been described for Mastigophallus, but its 
presence is probable); dart short and curved 
(canaliculate in Caracollina); one or two bifur- 
cate mucous glands flowing into the base of 
the accessory sac (in Oestophora they are 
connected to the vagina); penis covered by a 
penial sheath, with a penial papilla deeply sit- 
uated (but absent in Oestophora; Schileyko, 
1971) and a moderate-sized to long flagellum 
(reduced in Oestophorella and absent in Car- 
acollina, Oestophora and Gittenbergeria; 
Schileyko, 1991). Penial retractor muscle at- 
tached to the diaphragm. Chromosome num- 
ber n = 30? (only known for Oestophora; 
Ramos & Aparicio, 1985). 

Geographic Distribution: Iberian Peninsula, 
northwest Africa and ?Macaronese Islands. 



74 



PRIETO ET AL. 



Composition: Trissexodon Pilsbry, 1895; 
Caracollina Beck, 1837; Oestophora Hesse, 
1907; Mastigophallus Hesse, 1918; Oesto- 
phorella Pfeffer, 1929; Gasullia Ortiz de 
Zarate & Ortiz de Zarate, 1961; Suboesto- 
phora Ortiz de Zarate & Ortiz de Zarate, 1 961 ; 
Gasu///e//a Gittenberger, 1980; Gittenbergeria 
Schileyko, 1991; and perhaps Spirorbula 
Lowe, 1852, endemic from Madeira Islands 
and with a stimulatory apparatus that reminds 
one of that of Caracollina (see Schileyko, 
1991). 

Comments: As it has been commented pre- 
viously, Ciliella is not related to this group 
and, therefore, the name Ciliellinae, sensu 
Nordsieck, is not available. On the other 
hand, Canariella Hesse, 1918, according to 
Nordsieck (1987), is related to Oestophora, 
but is included in Ciliellidae by Schileyko 
(1991) (= Halolimnohelicidae, if Ciliella is re- 
moved from this family). 

In contrast to the Helicodontidae, the syn- 
apomorphic features of Trissexodontidae can- 
not be readily established because the general 
structure of the genital system that we can 
deduce for this group (one bifurcate mucous 
gland flowing into the usually great accessory 
sac which, in turn, flows into the dart sac, and 
penis with penial papilla and flagellum) could 
be the plesiomorphic condition of Hygromio- 
idea. On this assumption, the double stimula- 
tory apparatus present in Hygromiidae (at 
least, in some subfamilies), as well as in Vi- 
cariihelicinae and Halolimnohelicinae (in- 
cluded by Schileyko, 1991, in Ciliellidae, see 
above), is a derivative condition from a prim- 
itive single stimulatory apparatus, represented 
in Trissexodontidae and Helicodontidae, and 
(secondarily?) in Hygromiinae. This supposi- 
tion is contrary to the plesiomorphic condition 
proposed for Hygromioidea by Nordsieck 
(1987) and Schileyko (1991), who consider 
that the single stimulatory apparatus is a con- 
vergent derivative condition. 

In the resolution of this dilemma, i.e., single 
vs. double stimulatory apparatus as the ple- 
siomorphic condition for Hygromioidea, other 
data can be used, e.g., the insertion of the 
mucous glands and the chromosome num- 
ber. 

(1) Schileyko (1991) considered the primi- 
tive position of the mucous glands of Hygro- 
mioidea to be around the vagina above the 
upper sacs. Most Hygromiidae have this ar- 
rangement, but there is, at least, one case 
with another disposition: Ponentina Hesse, 



1921, with double stimulatory apparatus, 
shows one bifurcate mucous gland attached 
to each of the accessory sacs, and these, in 
turn, are attached to the vaginal side of the 
dart sacs, which bear darts (Manga, 1983; 
Prieto, 1986). In "Ciliellidae" sensu Schi- 
leyko, the two subfamilies with sacs have, ac- 
cording to Schileyko (1 991 ), bifurcate mucous 
glands attached to the base of the respective 
dartless sacs, which are very small, but these 
flow into the sacs in, at least, Vicariihelix ki- 
vuensis Verdcourt and Halolimnohelix seri- 
cata Pilsbry (Verdcourt, 1981). In Helicodon- 
tidae, there is one mucous gland near the 
base of the small dartless sac (if present). In 
Trissexodontidae, the bifurcate mucous gland 
flows into the accessory sac; in Suboesto- 
phora, in which the mucous gland appears to 
be completely divided into two forked glands 
again, these flow independently into the base 
of the large accessory sac (unpublished ob- 
servations). 

The presence of a single or bifurcate mu- 
cous gland flowing into the accessory sac in 
some representatives of all Hygromioidea 
families suggests that this configuration is 
plesiomorphic respect to the insertion of the 
mucous glands into the vagina, which hap- 
pens mostly in Hygromiidae. On the other 
hand, only Trissexodontidae and Hygromi- 
idae have sacs with darts, and in both families 
there are some cases where the accessory 
sacs are attached to the dart sacs: this occurs 
in all Trissexodontidae genera with stimula- 
tory apparatus and clearly in the hygromiid 
Ponentina; in the other hygromiids, whenever 
accessory and dart sacs are present, they are 
always closely attached and, in some cases, 
accessory sacs flowing into dart sacs can be 
seen (Schileyko, 1978). Again, an accessory 
sac flowing into the dart sac can be deduced 
as a plesiomorphic condition, rather than as a 
separate implantation of both on the vagina, 
which has been used as an argument to pro- 
pose the existence of upper and lower stylo- 
phores. 

(2) The chromosome number is unknown 
for many stylommatophores, but some num- 
bers are clearly indicative: within the Heli- 
coidea, the Ariantinae and Euparyphinae (He- 
licidae) have n = 29-30, whereas the 
Helicinae has n = 22, 25-27, 30, and the 
Elonidae has n > 29 (M. T. Aparicio, personal 
communication); within the Xanthonychoidea, 
the Bradybaenidae has n = 28-29 and the 
Monadeniinae (Xanthonychidae) has n = 29. 
The most common number appears to be n = 



MORPHOLOGY OF CARACOLLINA 



75 



29, a fact that agrees with the chromosome 
number of the related Camaenoidea and Me- 
sodontiodea, in which n = 29 is the most 
common number (Patterson & Burch, 1978). 
Therefore, Nordsieck (1987) suggests that 
this number is plesiomorphic for Helicoidea 
and related superfamilies. Nevertheless, the 
chromosome number of Hygromiidae is 
lower, with n = 23-26 (Thchiinae and Eu- 
omphaliinae) and n = 21, 23-26 (Hygromii- 
nae) (Patterson & Burch, 1978; Aparicio, 
1983; Ramos & Aparicio, 1985), but surpris- 
ingly higher in Oestophora, n = 30 (Ramos & 
Aparicio, 1985). This suggests that the chro- 
mosome number of Hygromiidae is apomor- 
phic in relation to that of Trissexodontidae. 
The chromosome number of Helicodonta, n 
= 27 (Rainer, 1967), is also unusual within 
Hygromioidea, but no conclusion about it is 
possible. 

Therefore, the two discussed features of 
Trissexodontidae, mucous gland flowing into 
the accessory sac and high chromosome 
number, suggest that this family is a primitive 
group. Because all Trissexodontidae genera 
have a single stimulatory apparatus (except in 
Gasulliella, in which it is completely reduced; 
Gittenberger, 1980), we conclude that this 
condition is plesiomorphic for Hygromioidea. 

There is another typical character of Tris- 
sexodontidae: the muscular ligament between 
the upper ends of both dart and accessory 
sacs and the spermoviduct. Nevertheless, this 
character seems to be plesiomorphic as well, 
because in addition to its presence in all Tris- 
sexodontidae genera (it can also be seen in 
Gasulliella — where dart and accessory sacs 
are absent — as a thin muscular line along the 
vagina wall; unpublished observations), it is 
also visible as a thin connective bridle in some 
Hygromiinae (Hygromiidae with single stimu- 
latory apparatus) as, for example, Cryptosac- 
cus (Prieto & Puente, in press-1), and in some 
Helicidae (Helicoidea) as, for example, Iberus 
Montfort, 1810 (Garcia San Nicolás, 1957, de- 
scribed as a "duct" between the dart sac and 
the spermoviduct). 

The function suggested by us for this mus- 
cular ligament is to maintain the stimulatory 
apparatus joined to the vagina to avoid a float- 
ing location in the haemocoel; the stimulatory 
apparatus would be primitively connected to 
the vaginal tract by the dart sac alone, be- 
cause the accessory sac with the mucous 
gland flowing into it was attached to the dart 
sac. This structure would be related to an elon- 
gate asymmetric stimulatory apparatus. 



In consequence, we cannot recognize any 
synapomorphic character in the genital sys- 
tem of Trissexodontidae; the only one syn- 
apomorphy that we suggest for this group is 
the regularly ribbed shell associated with the 
lack of hairs, which does not occur in any 
other group of Hygromioidea. 

At present, a subfamiliar division of Trissex- 
odontidae seems inappropriate to us, be- 
cause its genital structure is rather conserva- 
tive in spite of some modifications of the 
general pattern, for example, loss of flagellum 
{Caracol Una, Gittenbergeria, Oestophora), 
loss of penial papilla (Oestophora), loss of the 
stimulatory apparatus {Gasulliella), or pres- 
ence of two bifurcate mucous glands {Sub- 
oestophora, Gasullia, Oestophorella, Masti- 
gophallus). These modifications could have 
happened several times during the evolution 
of this group. Therefore, analysis of possible 
evolutionary pathways into Trissexodontidae 
requires further research: a solid taxonomic 
revision based on accurate dissections and 
investigation of characters (e.g., chromosome 
number, enzymatic analysis, shell micro- 
sculpture, distribution patterns) overlooked 
previously. 



ACKNOWLEDGMENTS 

This research was supported by a predoc- 
toral research grant conceded by the Depart- 
ment of Education, Universities and Research 
of the Basque Government to A. I. Puente, 
and by the "Fauna Ibérica M" project (PB89- 
0081 ) of the Spanish General Directorate for 
Scientific and Technical Research (DGICYT). 



LITERATURE CITED 

ADAMO, A. S. & R. CHASE, 1990, The "love dart" 
of the snail Helix asperse injects a pheromone 
that decreases courtship duration. Tt)e Journal of 
Experimental Zoology, 255: 80-87. 

APARICIO, M. T., 1983, Estudio morfológico y ci- 
totaxonómico de algunos Helicidos de la fauna 
española, en especial de la región central. 
Colecc. Tesis Doctorales, 29. Universidad Com- 
plutense de Madrid. 299 pp. 

BACKHUYS, W., 1975, Zoogeography and taxon- 
omy of the land and fresh-water molluscs of the 
Azores. Backhuys & Meesters, Amsterdam, 350 
pp., 97 map., 32 pis. 

CULLING, С F. A., 1974, Handbook of histological 
and histochemical techniques. Buttenworths, 
London, 712 pp. 



76 



PRIETO ET AL. 



FORCART, L., 1965, Rezente Land- und Süss- 
wassermollusken der süditalienischen Land- 
schaften Apulien, Basilicata und Calabrien. Уег- 
handelingen Naturforschunge Gesellschaft in 
Basel, 78(1): 59-184. 

GARCÍA SAN NICOLÁS, E., 1957, Estudios sobre 
la biología, la anatomía y la sistemática del 
género Iberus Montfort, 1810. Boletín de la Real 
Sociedad Española de Historia Natural (Biolo- 
gía), 55(2/3): 199-390 + 29 lám. 

GASULL, L., 1965, Algunos moluscos terrestres y 
de agua dulce de Baleares. Boletín de la So- 
ciedad de Historia Natural de Baleares, 11(1- 
2-3-4): 7-161. 

GUSULL, L., 1975, Fauna malacológica terrestre 
del sudeste ibérico. Boletín de la Sociedad de 
Historia Natural de Baleares, 20: 5-148, 4 pl. 

GERMAIN, L, 1930, Mollusques terrestres et fluvi- 
átiles. In: Faune de France. Lechevalier, Paris, 
477 pp., 13 pis. 

GITTENBERGER, E., 1968, Zur Systematik der in 
die Gattung Trissexodon Pilsbry (Helicidae, He- 
licodontinae) gerechneten Arten. Zoologische 
Mededelingen, 43(13): 165-172. 

GITTENBERGER, E., 1980, Three notes on Iberian 
terrestrial gastropods. Zoologische Mededelin- 
gen, 55(17): 201-213. 

GITTENBERGER, E., W. BACKHUYS & T. E. RIP- 
KEN, 1970, De Landslakken van Nederland. 
Koninklijke Nederlandse Natuurhistorische Ve- 
reniging, Amsterdam, 177 pp. 

GIUSTI, F., 1970, Notulae malacologicae. XII. 
L'Isola de Pianosa e lo scoglio La Scola (Arci- 
pelago Toscano). Annali del Museo Cívico di Sto- 
ria Naturale di Genova, 78: 59-148, 15 pis. 

GÓMEZ, В. J., 1991, Morphological and histologi- 
cal study of the genital ducts of Cryptazeca mo- 
nodonta (Pulmonale, Orthurethra), with special 
emphasis on the auxiliary copulatory organ. 
Zoomorphology, 111: 95-102. 

GROSSU, A. v., 1983, Gastropoda Romaniae. 
Ordo Stylommatophora 4. Suprafam.: Arionacea, 
Zonitacea, Ariophantacea si Helicacea. Litera, 
Bucuresti, 564 pp. 

HAAS, F., 1929, Fauna malacológica terrestre у de 
agua dulce de Cataluña. Trabajos del Museo de 
Ciencias Naturales de Barcelona, 13: 1-491. 

HESSE, P., 1918, Die subfamilie Helicodontinae. 
Nachrichtsblatt der Deutsche Malakozoologische 
Gessellschatt, 50: 99-110. 

HESSE, P., 1931, Zur Anatomie und Systematik 
palearktischer Stylommatophoren. Zoológica, 
31(81): 1-118, 16 pis. 

HESSE, P., 1934, Zur Anatomie und Systematik 
palearktischer Stylommatophoren. Zoológica, 
34(85): 1-57, 9 pis. 

KERNEY, M. P., R. A. D. CAMERON & J. H. JUNG- 
BLUTH, 1983, Die Landschnecken Nord- und 
Mitteleuropas, P. Parey, Hamburg und Berlin, 
384 pp., 24 pis. 

MANGA, M. Y., 1983, Los Helicidae (Gastropoda, 
Pulmonata) de la provincia de León. Diputación 



Provincial de León, Institución "Fray Bernardino 
de Sahagún," León, 394 pp. 

MANGANELLI, G., I. SPARACIO & F. GUISTI, 
1 989, 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. 

MARTOJA, R. & M. MARTOJA-PIERSON, 1970, 
Técnicas de histología animal. Toray-Masson, 
Barcelona, 350 pp. 

MATEO, В., 1978, Estudio comparado de los mo- 
luscos terrestres de Menorca. B. Mateo, Mahón, 
56 pp. 

MICHAUD, A. L. G., 1831, Complément de l'his- 
toire naturelle des mollusques terrestres et fluvi- 
átiles de la France, de J.P.R. Draparnaud. Lipp- 
mann, Verdun, 116 pp. 

MOQUIN-TANDON, A., 1855, Histoire naturelle 
des mollusques terrestres et fluviátiles de 
France. T. Il + Atlas. J.B. Baillière, Paris, 648 pp. 
+ 92 pp., 54 pis. 

NOBRE, A., 1941, Fauna malacológica de Portu- 
gal. II. Moluscos terrestres e fluviais. Memorias e 
Estudos do Museu Zoológico da Universidade de 
Coimbra, 124: 1-277, 30 pis. 

NORDSIECK, H., 1986, Das System der tertiären 
Helicoidae Mittel- und Westeuropas (Gas- 
tropoda: Stylommatophora). Heidia, 4(1): 109- 
120, pis. 15-17. 

NORDSIECK, H., 1987, Revision des Systems der 
Helicoidea (Gastropoda: Stylommatophora). Ar- 
chiv für Molluskenkunde, 118(1/3): 9-50. 

NORDSIECK, H., 1989, Falkneria n. gen., eine 
neue Gattung der Helicodontinae (Gastropoda, 
Stylommatophora: Hygromiidae). Heidia, 1(5/6): 
165-168, pl. 25. 

ODHNER, N. H., 1931, Beiträge zur Malakozoolo- 
gie der Kanarischen Inseln, Lamellibranchien, 
Caphalopoden, Gastropoden. Arkiv for Zoologi, 
23A(1 4): 52-115. 

ORTIZ DE ZARATE, A. & A. ORTIZ DE ZARATE, 
1961, Moluscos terrestres recogidos en la pro- 
vincia de Huelva. Boletín de la Real Sociedad 
Española de Historia Natural (Biología), 59: 1 69- 
190. 

OUTEIRO, A. M., 1988, Gasterópodos de O Courel 
(Lugo). Tesis Doctoral. Universidad de Santiago, 
Santiago de Compostela, 627 pp., 1 lám. 

PATTERSON, С. M. & J. В. BURCH, 1978, Chro- 
mosomes of pulmonale molluscs. Pp. 171-217 
in: V. Fretter & J. Peake, eds., Pulmonates, Vol. 
2A. Systematics, evolution and ecology. Aca- 
demic Press, London, 540 pp. 

PRIETO, С. е., 1986, Estudio sistemático y bio- 
geográfico de los Helicidae sensu Zilch, 1959- 
60 (Gastropoda: Pulmonata: Stylommatophora) 
del País Vasco y regiones adyacentes. Tesis 
Doctoral. Universidad del País Vasco, 393 pp, 10 
lám. 

PRIETO, С E. & A. I. PUENTE, in press-1, Un 
nuevo Hygromiinae (Pulmonata: Helicoidea: Hy- 
gromiidae) del norte de la Península Ibérica, 



MORPHOLOGY OF CARACOLUNA 



77 



Cryptosaccus astuhensis n. gen., n. sp. Archiv 
für Molluskenkunde, 123. 

PRIETO, С. E. & А. I. PUENTE, in press-2. El 
género Hygromia Risso, 1826 en la Peninsula 
Ibérica, con descriptción de Hygromia gofasi sp. 
nov., у consideraciones sobre la interpretación 
functional del aparato estimulador de Hygromi- 
idae. Bulletin du Muséum National d'Histoire Na- 
turelle, Paris. 

PUENTE, A. I., С E. PRIETO & К. ALTONAGA, 
1 990, Nuevos datos sobre la distribución de Car- 
acollina lenticula (Michaud 1831) (Gastropoda: 
Pulmonata: Helicoidea) en la Península Ibérica. 
Cuadernos de Investigación Biológica (Bilbao), 
16: 101-113. 

RAINER, M., 1967, Chromosomenuntersuchungen 
an Gastropoden (Stylommatophora). Malacolo- 
gia, 5(3): 341-373. 

RAMOS, M. A. & M. T. APARICIO, 1985, A cyto- 
taxonomic study of some Spanish and Portu- 
guese Helicidae (Pulmonata: Geophila). Malaco- 
logical Review, 18: 73-82. 

SCHILEYKO, A. A., 1971, The taxonomic status of 
the Helicodontinae (Pulmonata, Helicidae). 
Naucn. Kokl, Vyss. Skoly. Biol. Nauki., 12: 7-16 
(in Russian). 

SCHILEYKO, A. A., 1973, Comparative character- 
istics of Palearctic families of terrestrial Molluscs 



from the superfamily Helicoidea. Zoologicheskii 
Zhurnal, 52(4): 492-506 [in Russian]. 

SCHILEYKO, A. A., 1978, Nazemnye molljuski 
nadsemejstva Helicoidea. Fauna SSSR, Moll- 
juski, 3(6). Zoologicheskii Institut, Akademija 
Nauk SSSR, Novaja Serija, 1 1 7: 384 pp. Lenin- 
grad. 

SCHILEYKO, A. A., 1991, Taxonomic status, phy- 
logenetic relations and system of the Helicoidea 
sensu lato. Archiv für Molluskenkunde, 120(4/6): 
187-236. 

SCHUBERTH, O., 1892, Beiträge zur vergle- 
ichenden Anatomie des Genitalapparates von 
Helix mit besonderer Berücksichtigung der Sys- 
tematik. Archiv für Naturgeschichte, 58(1): 1-65, 
4 pis. 

TOMPA, A., 1984, Land snails (Stylommatophora). 
Pp. 47-140. in: K. M. Wilbur, ed., The Mollusca, 
vol. VII: Reproduction. Academic Press, London. 

VERDCOURT, В., 1981, Contributions to the 
knowledge of the Helicidae-Bradybaeninae of 
Zaïre (Mollusca, Gasteropoda). Revue de Zoolo- 
gie Africaine, 95(3): 525-556, pis. 3-5. 

ZILCH, A., 1959-60, Euthyneura. In: W. Wenz, 
Handbuch der Palaözoologie, 6(2). Gebrüder 
Borntraeger, Berlin, 835 pp. 

Revised Ms. accepted 30 July 1992. 



MALACOLOGIA, 1993, 35(1): 79-87 

MELANISM IN THE LAND SNAIL HEUCELLA CANDICANS (GASTROPODA, 
HELICIDAE) AND ITS POSSIBLE ADAPTIVE SIGNIFICANCE 

Alois Honèk 

Department of Entomology, Research Institute of Plant Production, Ruzyné 507, 
16106 Praha 6, Czecfioslovakia 

ABSTRACT 

Shell banding polymorphism in 184 local populations of Helicella candicans (Pfeiffer) from 
western Czechoslovakia was investigated. The shells are white with up to nine dark brown 
bands, which may fuse. There was large within- and among-population variation in shell band- 
ing. An "index of melanisation," indicating proportion of shell surface covered with extended or 
fused bands, revealed geographic patterning of dark phenotypes. The frequency of dark forms 
was higher in some areas, due perhaps to decrease of incident sunshine by fog, clouds or 
industrial air pollution. High and dense vegetation cover were also associated with melanism. In 
the laboratory, temperature of irradiated dark shells increased more rapidly than that of light 
shells, and the thermal equilibrium of the former was higher. The differences were greatest on 
a white background and with low ambient temperature. In areas of reduced sunshine, dark 
individuals may be at an advantage, especially during the autumn breeding period. When ex- 
posed to sunshine during summer dormancy, light forms may also be able to maintain lower 
body temperature than dark forms. 



INTRODUCTION 

Helicella candicans (Pfeiffer) is a small he- 
licid gastropod (shell diam. 9-20 nnm). In Bo- 
hemia, western Czechoslovakia, it inhabits 
dry steppes on calcium-rich soils, particularly 
on the southern slopes of hills along the Ohfe 
(Eger) and Labe (Elbe) rivers, in the Central 
Bohemian Karst, and in a few other sparsely 
distributed localities (Lozek, 1956). Oviposi- 
tion was observed in late summer and early 
autumn. During dry periods in June to Sep- 
tember, the animals aestivate attached to dry 
herbaceous vegetation. 

The very diffuse nature of the variation is 
perhaps why the shell banding polymorphism 
of H. candicans has been little studied. Geo- 
graphic variation in the proportions of different 
phenotypes is considerable. I have developed 
a system that enables the degree of melan- 
ism of the shell to be classified. I explored 
variation in melanism at a number of localities 
in Bohemia and attempted to establish the re- 
lationship between this variation and local mi- 
croclimate. 



MATERIALS AND METHODS 

In 1987-1989, H. candicans was collected 
at 184 sites in central and western Bohemia. 
At each site, all shells were sampled from an 



area, the size of which varied according to 
snail abundance. This prevented collecting 
bias favouring certain morphs due to differ- 
ences in relative crypsis to the collector. The 
minimum distance between the sites was 
150 m. At each site, 50-150 living or well- 
preserved dead individuals were collected, 
and the density and height of vegetation 
cover was evaluated, specifically to estimate 
how it may shade the surface in late summer 
and early autumn, during the H. candicans 
breeding season. The vegetation was ranked 
into seven crude subjective categories that 
proved usable for quantification of plant cover 
effects on H. candicans melanism. 

The dorso-ventral ly compressed shell of H. 
candicans is white, with one to nine dark 
brown to black bands (Fig. 1). The single dor- 
sal band is variable in width and may extend 
over the whole dorsal surface when the edges 
of the band become diffuse. There are zero to 
six lateral bands, the width of which vary less 
than that of the dorsal band. Adjacent bands 
may fuse to form a belt consisting of up to six 
original bands. There are zero to two narrow 
ventral bands. Individuals with diffuse dark 
coloration of the dorsum and with a lateral belt 
consisting of four or five fused bands were 
termed "dark" forms. Individuals having a thin 
dorsal band only were termed "light" forms. 

Shell coloration was classified according to 
the degree of melanisation, i.e. the proportion 



79 



80 



HONEK 




FIG. 1. Variation in shell banding pattern in H. can- 
dicans. 1-2, light and dark shells viewed dorsally. 
3-7, shells with different nunnbers of lateral bands. 
8-1 2, shells with 2-5 lateral bands fused into belts. 
Specimens 2 and 12 are examples of "dark" indi- 
viduals. 



of the shell surface colored dark, calculating 
an "index of nrielanisation." This index was 
calculated as follows. The dorsal band width 
was scored as: < 0.15 mm, 0.15-0.39 mm, 
0.40-0.69 mm, 0.70-1.00 mm, or > 1.00 
mm, these classes being given scores of 0.5, 
1, 2, 3, 4, respectively. Lateral bands were 
split into three width classes: < 0.15 mm, 
0.15-0.30 mm, and > 0.30 mm, with scores 
of 0.5, 1 , and 2, respectively. Ventral bands, if 
present, were scored as 0.5 or 1 . Every fusion 
of two adjacent bands was given a score of 2. 
The number of fused bands could be deter- 
mined in most shells because one whorl back 
from the shell aperture the color of fusions is 
usually lighter than the color of bands. The 
index of melanisation for an individual shell 



was the sum of scores for all bands and all 
fusions. Individual indices varied between 0.5 
(shells with traces of a dorsal band only) to 25 
(dark individuals). The average index of mel- 
anisation for a population was the arithmetic 
mean of the individual indices for all shells in 
the sample from that population. 

The temperature increase inside shells un- 
der incident radiation was measured using 
dead shells of 13-14 mm diameter (mea- 
sured 1/4 whorl back from the shell aperture). 
A dark and a light shell were filled with petro- 
leum jelly, thermocouples were inserted into 
the shell cavities, and the shells were placed 
simultaneously on a wooden block painted 
black or white, irradiated with a 60 W or a 200 
W lamp from a distance of 25 cm. At the start 
of each experiment, the temperature in the 
shells was allowed to approach ambient. After 
switching on the light, the temperature in the 
shells was read (with 0.1 °C accuracy) every 
30 sec for 1 minutes. The experiments were 
made at low (average within shell tempera- 
ture at the start 12.1°C) and high (average 
starting temperature 25.9°C) ambient temper- 
atures. All measurements were repeated with 
two pairs of shells, twice with each pair. 

Our explanation of the variation in banding 
(see Discussion) points to an influence of me- 
teorological factors that decrease the amount 
of solar radiation reaching the earth's surface. 
No map indicating local variation of these fac- 
tors with sufficient precision is available. 
Some relevant data (Fig. 2) were compiled 
from Vesely (1953) (number of overcast days 
per year, a map based on data from 270 me- 
teorological stations in Czechoslovakia from 
1926-1950) and SIádek (1977) (per cent 
days with fog per year, tabular data for nine 
meteorological stations within the study area 
from 1971-1975). The distribution of frequent 
autumn local fogs is based on the author's 
experience over several years and on consul- 
tation with local inhabitants. 



RESULTS 

There was a large inter-population variation 
in average shell melanisation. However, the 
distribution of dark populations (with average 
index of melanisation > 11.0) was not com- 
pletely random (Fig. 3). Many dark populations 
were found along the northwest section of 
Labe River, and several dark populations were 
also found further east along this river. Dark 
populations were found also near the cement 



MELANISM IN THE LAND SNAIL 



81 



Ш m 150 140 130 


j^^^jí¿fec<ry / /^ 


'\*^ _LJ^!^-\; >ч y^^ 


^_^jv---y"^ V,^ V / 


\ ^J--"^^ I ^> 






-f^ Ч .—vi /тш^^. 




^^:jí-MiJ^"% 


" ^||Sk,9^^_^x:j|p^ / /^ ) \\ ""^"""--^ Л 


"^ÑA/^T^" 


/ /^"1<г"~^^'^ 7 








"Чй.— .-•» • 


л г "w ^ ~~ '' V 




^^ &7 


Ш'"~-,^ ^ .206"^- --^,¿^/ 




%-N 


Щ ^----^ ч. Î f'' 






\\ '^^Ъ\ J ^'^.Si 




é- %*se^ ^ 


V^^^ ^-^uiiHiî>sîj2^;Ô ^^^^ ''''^' 




'ttiSASfcrgJStóíííiii-' 


13.2' ( \Jj\ \v 




"^~^130 


\|^.6 J30 ^120 






^/ 1 



FIG. 2. Selected climatic data for the region of western Czechoslovakia shown in Fig. 3 (see right left upper 
insert in Fig. 3 for position of the region). The map indicates: (1 ) The iso-lines of the number of overcast days 
per year (an overcast day means 80-100% average cloud cover calculated from observations at 07.00, 
14.00 and 21.00 h). (2) Per cent days with fog per year (italics) at nine meteorological stations (from left: 
Zatec, Doksany, Praha-Ruzyné, Praha-Karlov, Tièice, Brandys nad Labem, Lysá. Insert: Beroun, Kladno). 
(3) The areas of frequent occurrence of fogs (shaded). 



factory in Kráiúv Dvùr in the Bohemian Karst 
(Fig. 3, asterisk on left insert). The populations 
with intermediate indices of melanisation were 
scattered over the whole area. Light popula- 
tions (lOM < 9.0) prevailed in the hilly area of 
the Bohemian Karst (Fig. 3, insert). Despite 
this geographic pattern of distribution, there 
was a large local variation in lOM, and popu- 
lations at sites closer than 0.5 km sometimes 
had quite different indices of melanisation. 

Populations from habitats with dense and 
tall vegetation tended to be darker than pop- 
ulations of short grass steppes. I found a 
weak but significant relationship between in- 
dex of melanisation and plant density (Fig. 4) 
or vegetation height (r^ = 2.7%, p < 0.05). 
Frequency of populations with high proportion 
of dark (lOM = 25) individuals also increased 
with vegetation density (r^ = 0.4%). These 
populations were more frequent at sites with 
tall vegetation than at short grass steppes 
(Fig. 4). However, the relationship between 
plant density or height and percent of dark 
shells was not significant. Low statistical sig- 
nificance was the consequence of many zero 
values for proportions of dark individuals in 
populations under each type of vegetation. 



Dark and light shells differed in their rates 
of heating when exposed to radiation under 
experimental conditions. The rate of temper- 
ature increase and differences between dark 
and light shells depended on ambient temper- 
ature, intensity of radiation and color of the 
background (Fig. 5). The differences in within- 
shell temperature increased during the first 
six minutes of irradiation, when the tempera- 
ture of dark shells increased faster than tem- 
perature of light ones. The highest differences 
were attained at low ambient temperature, 
with high intensity of radiation, on a white 
background. The maximum differences after 
the thermal equilibria were attained (approxi- 
mately 10 minutes from the start of the irradi- 
ation) were about 2.5°C (Table 1). The ther- 
mal equilibria at low ambient temperature 
were highest on a black background, where 
the temperature excess over ambient was 
about 10°C. 



DISCUSSION 

Many factors including selection (by pred- 
ators or climatic factors) and historical events 



82 



HONEK 







MELANISM IN THE LAND SNAIL 



83 



1 - 



60 w 



21- 



1- 



OOOO 



OqOOOOOOOOqqO 



UJ 
U- ' 



1 - 



T 1 1 г 



200 W 



26 ^C 



oooo 



OqO 



ooo 



oOo°°°° 



2- 



>»8S' 



ФФ 



888888*8<»» 



nOOO 
О • 



о 



oo 



1 1 Г- 

oooo 



12 °C 



0|-(»Cill# 0Э 

T 1 1 — ( 1 1 1 1 1 1 1 г 

12 3 4 5 6 7 8 12 3 4 5 6 7 8 



FIG. 4. Vegetation cover and shell melanisation. Top: Density of plant cover DEN and index of melanisation 
lOM, regression у = 0.41 3x + 7.84, t = 2.767, p < 0.01, coefficient of determination r^ = 4.0%, p<0.05. 
Bottom: Average height of the plant stand and proportion of dark individuals MEL in populations, regression 
у = 0.01 5 X + 0.807, t = 1.764, coefficient of determination r^ = 1.7%, n.s. Symbols: 1-4 cases, and 
> 5 cases with similar proportion of dark individuals. Total number of investigated sites is 184. 



(founder effect), and an extensive random 
variation (genetic drift) influence the compo- 
sition of populations of polymorphic snail spe- 
cies. In addition, microhabitat choice of differ- 
ent morphs may also vary composition of 
populations. This plurality of evolutionary 
forces and behavioral effects also makes dif- 



ficult the causal explanation of population 
structure in species with shell banding poly- 
morphism (cf. Jones, 1973; Jones et al., 
1977; Cain, 1983; Hazel & Johnson, 1990). 

Helicella candicans is a typical example of 
species with variation that cannot be ex- 
plained by a simple mechanism. There is a 



FIG. 3. Geographic variability of the index of melanisation (lOM) in the valleys of Ohfe and Labe rivers, and 
in the area of Central Bohemian Karst (left lower insert). The position of the areas shown on the territory of 
western Czechoslovakia is indicated in the right upper insert. Asterisks indicate major sources of industrial 
aerial pollution. Each circle represents one collecton site. Open: lOM <8.9, with central spot: 9.0 <IOM 
<10.9, solid: lOM >1 1 .0. Localities included: 1 . Pi'ivlaky, 2-3. Stroupeé, 4. Zatec, 5. Leneáice, 6. Milá, 7-9. 
Rana, 10-11. Chraberce, 12. Chozov, 13-15. Dobromèhce, 16. Zidovice, 17. Koèetice, 18-21. Kfesín, 22. 
Dubany, 23-25. Libochovice, 26-27. Klapy, 28. Radovesice, 29. Zabovi'esky nad Ohi^i, 30. Bfezany nad 
Ohfl, 31-34. Doksany, 35-37. Libochovany, 38-39. Veiké Zernoseky, 40. Zalhostice, 41-44. Litomèi'ice, 
45. Velky Újezd, 46. Ki'eèice, 47. Encovany, 48. Polepy, 49-51. Vrutice, 52. Hoèt'ka, 53. Brzánky, 54. 
Kochovice, 55-59. Stèti, 60-61 , Radouñ, 62. Cakovice, 63. Straöi, 64-66. Poöepice, 67. Jeèovice, 68-69. 
Libèchov, 70. Vehiovice, 71 . Mëlnicka Vrutice, 72. I^aly Újezd, 73. Vavi=ine6, 74-75. Kelské Vinice, 76. 
Tuhañ, 77-80. Tuhañské Vétruáice, 81-83. Cervená Píska, 84-86. Pfívory, 87-88. Nedomice, 89-91. 
Di'ísy, 92. Byèice, 93. Ceàelice, 94. Konètopy, 95-97. Sudovo HIavno, 98-100. Kosteiní HIavno, 101. Krpy, 
102. Skorkov, 103. Tufice, 104. Pi'erov nad Laberp, 105-109. Semice, 110. Roudnice, 111. Ctmëves, 112. 
Kostomlaty pod Ripem, 113-115. Libkovice pod Ripem, 116-117. Nové Ouholice, 118. Mlöechvosty, 119. 
Úzice, 120. Veliká Ves, 121-122. Praha. 123. Slavíky, 124-128. Suchomasty, 129-132. Vinai'ice, 133-137. 
Váeradice, 138. Liten, 139-140. Korno, 141-145. Meñany, 146-151. Tobolka, 152-155. Koledník, 156. 
Jarov, 157-159. Tetín, 160-167. Beroun, 168-174. Srbsko, 175. Karlètejn, 176-177. Hlásná Trebáñ, 178. 
Mofinka, 179. Moi'ina, 180. Bubovice, 181. Lodénice, 182. Vrbice, 183. VIkov pod Oèkobrhem, 184. 
Hradöany. The localities are designated with names of the nearest village. 



84 



HONEK 



15- 



10 






•••• 



•••. 



•• 



••• 
• ••• 



• ••• 



• • 



••• 



•••• 



•••• 

■ - 
■ 



••• 



•••• 
• ••• 






•• 



15- 



10 



3 4 

DEN 






•• 






■ 



■ 

-j- 



10 



20 



30 



40 cm 50 



PLANT HEIGHT 



FIG. 5. The differences in warming up of the light and dark shells of H. candicans, under 60 W (left) and 200 
W (right) 1аглр, at 26°C (above) and 12°C (below) ambient temperatures. The circles indicate differences in 
within-shell temperature read every 30 s from the start of the experiment. Open circles, white ground surface, 
solid circles, black ground surface. Each circle represents the mean of three measurements; standard errors 
for all means were between 0.20°C and 0.29°C. 



MELANISM IN THE LAND SNAIL 



85 



TABLE 1. Average temperature (°C) excess (± SE) over ambient after 10 minutes of irradiation of dark 
(D) and light (L) shells, at two ambient temperatures. Starting temperature is an average of temperatures 
established within the shells left to cool to ambient temperature, at the start of the irradiation. 











Light source 












200 W 








60 W 




Starting temperature 


D 




L 


D 




L 


White background surface 
















12.1°C 


7.9 




5.3 




2.1 




1.5 




±0.5 




±0.4 




±0.1 




±0.2 


25.9°C 


9.8 




8.6 




4.1 




3.9 




±0.6 




±0.2 




±0.2 




±0.4 


Black background surface 
















12.1°C 


11.0 




9.6 




3.5 




2.8 




±0.9 




±1.1 




±0.7 




±0.3 



large within- and among-population variation 
in shell banding, and a weak association be- 
tween environment factors and melanism. 
The genetic basis of polymorphism in H. can- 
dicans Is unknown, but a genetic component 
in shell banding polymorphism may be in- 
ferred from analogy with other helicids (e.g. 
Wolda, 1969), and here I assume that a ge- 
netic control of shell banding polymorphism 
does exist. The large individual variation at all 
localities studied indicates an important inde- 
terministic component affecting the variation 
of shell banding forms (cf. Cameron et al., 
1980; Cameron & Dillon, 1984; Ratel et al., 
1989). Although a large proportion of variation 
may be random, a part of variation may have 
adaptive significance. 

The only significant factor of shell melani- 
sation that could be demonstrated from this 
study is climatic selection. I suppose that the 
reduced incident solar radiation may favour 
dark populations. This is indicated by in- 
creased frequency of dark populations in ar- 
eas with frequent fogs and increased cloudi- 
ness. This particularly applies to area around 
the northwest section of Labe River (Fig. 3). 
This river crosses the Ceské Stfedohofí 
Mountains through a narrow valley. In this re- 
gion, there are several chemical factories and 
electric plants using lignite (Fig. 3, asterisks) 
that are sources of air pollution. These factors 
favour the origin of local fogs, which often ap- 
pear in the autumn, decreasing solar radiation 
reaching the earth's surface. The greater 
cloudiness in this area also decreases solar 
radiation reaching the earth's surface (Fig. 2). 
Several dark populations were found further 
east along the Labe River where local fogs 



are also frequent. Local fogs and aerial pol- 
lution may affect the occurrence of dark pop- 
ulations near the cement factory in Králúv 
Dvúr (Fig. 3, insert), whereas the light popu- 
lations prevailed in the rest of the hilly area of 
Bohemian Karst with relatively clean air, low 
cloudiness and low fog frequency (Fig. 2, 
insert). Plant cover may also reduce the in- 
tensity of incident solar radiation, and several 
examples of increased melanisation under 
dense and tall plant stands were found. 

The shell banding polymorphism in H. can- 
dicans may have adaptive significance re- 
lated to different thermoregulation properties 
of dark and light morphs (cf. Tilling, 1983; Et- 
ter, 1988). High index of melanisation and in- 
cidence of dark shells were associated with 
environments where sunshine was reduced. 
Variation in other snail species provides par- 
allel examples of association between shell 
color and microclimate (cf. Heller & Volokita, 
1 981 a; LIvshits, 1 981 ; Nevo et al., 1 981 ; Em- 
berton, 1982; Nevo et al., 1982; Heller & Ga- 
dot, 1984; Ramos, 1984, 1985; Sacchi, 1984; 
Vicario et al., 1988; Hazel & Johnson, 1990). 
I suggest that dark shell coloration may help 
to maintain increased body temperature on 
cool and overcast days. Such conditions are 
frequent in the autumn, the breeding season 
of H. candicans, particulariy at localities near 
rivers and sources of air pollution, which both 
contribute to frequent fog. Then, a quicker in- 
crease of body temperature during the short 
spells of sunshine may confer some advan- 
tage on darks (cf. Heller & Volokita, 1981b). 

On the other hand, being dark may also 
have negative consequences. The snails are 
particulariy sensitive to overheating and des- 



86 



HONEK 



iccation when active, and there is a selection 
for pale body color in warm areas (Cowie & 
Jones, 1985; Cowie, 1990). Light individuals 
may maintain lower thermal equilibria than 
dark individuals, the coloration which may 
then become a disadvantage. I have no data 
on mortality, but I suppose that at the steppe 
localities, e.g. on the southern slopes of hills 
in the Bohemian Karst, heat stress from solar 
radiation may affect survival. 

Although the advantage that arises from 
different thermoregulation properties of dark 
and light morphs probably contributes to dif- 
ferentiation of phenotype frequencies among 
the populations, climatic selection explains 
only a very small fraction of among-popula- 
tion variation in shell melanisation. Further 
study may reveal other selection forces, and I 
suppose that a great proportion of variation is 
random. 



ACKNOWLEDGMENTS 

I thank Prof. A. J. Cain of the University of 
Liverpool, and two anonymous reviewers for 
critical reading and valuable comments on the 
MS, and Martin Vákáf of Technical University 
of Prague for assistance in measuring within- 
shell temperatures. 



LITERATURE CITED 

CAIN, A. J., 1983, Ecology and ecogenetics of ter- 
restrial molluscan populations. Pp. 597-647, in: 
W. D. Russell-Hunter, ed., The Mollusca. Vol 6. 
Ecology, Academic Press. London, New York, 
San Francisco. 

CAMERON, R. A. D., M. A. CARTER & M.A. 
PALLES-CLARK, 1980, Cepaea on Salisbury 
Plain: patterns of variation, landscape history 
and habitat stability. BiologicalJournal of the Lin- 
nean Society, 14: 335-358. 

CAMERON, R. A. D. & P. J. DILLON, 1984, Habitat 
stability, population histories and patterns of vari- 
ation in Cepaea. Malacologia, 25: 271-290. 

COWIE, R. H., 1990, Climatic selection on body 
colour in the land snail Ttieba pisana (Pulmo- 
nata: Helicidae), Heredity, 65: 123-126. 

COWIE, R. H. & J. S. JONES, 1985, Climatic se- 
lection on body colour in Cepaea. Heredity, 55: 
261-267. 

EMBERTON, K. C, 1982, Environment and shell 
shape in the Tahitian land snail Paríala otaheit- 
ana. Malacologia, 23: 23-35. 

ETTER, R. J., 1988, Physiological stress and color 
polymorphism in the intertidal snail Nucella lapil- 
lus. Evolution, 42: 660-680. 



HAZEL, W. N. & M. S. JOHNSON, 1990, Microhab- 
itat choice and polymorphism in the land snail 
Ttieba pisana (Müller). Heredity, 65: 449-454. 

HELLER, J. & M. GADOT, 1984, Shell polymor- 
phism of Tfieba pisana — the effects of rodent dis- 
tribution. Malacologia, 25: 349-354. 

HELLER, J. & M. VOLOKITA, 1981a, Shell-banding 
polymorphism of the land snail Xeropicta vestalis 
along the coastal plain of Israel. Biological Jour- 
nal of thie Linnean Society, 16: 279-284. 

HELLER, J. & M. VOLOKITA, 1981b, Gene regu- 
lation of shell banding in a land snail from Israel. 
Biological Journal of the Linnean Society, 16: 
261-277. 

JONES, J. S., 1973, Ecological genetics and natu- 
ral selection in molluscs. Science, 182: 546-552. 

JONES, J. S., B. H. LEITH & P. RAWLINGS, 1977, 
Polymorphism in Cepaea: a problem with too 
many solutions? Annual Review of Ecology and 
Systematics, 8: 109-143. 

LIVSHITS, G. M., 1981, Survival, behaviour and 
spatial distribution of shell morphs in a population 
of the snail Brephulopsis bidens (Pulmonata). 
Oecologia, 51 : 220-226. 

LOZEK, v., 1956, Klid ôeskoslovenskych mëkkyàù 
[Key to Czechoslovak Mollusca]. Vydavatelstvo 
Slovenskej Akademie Vied. Bratislava. 

NEVO, е., с. bar-el & a. BEILES, 1981, Genetic 
structure and climatic correlates of desert land- 
snails. Oecologia, 48: 199-208. 

NEVO, е., с. bar-el, a. BEILES & Y. YOM-TOV, 
1 982, Adaptive microgeographic differentiation of 
allozyme polymorphism in landsnails. Genética, 
59: 61-67. 

RAMOS, M. A., 1984, Polymorphism of Cepaea ne- 
moralis (Gastropoda, Helicidae) in the Spanish 
occidental Pyrenees. Malacologia, 25: 325-341 . 

RAMOS, M. A., 1985, Shell polymoфhism in a 
southern peripheral population of Cepaea nem- 
oralis (L.) (Pulmonata: Helicidae) in Spain. Bio- 
logical Journal of the Linnean Society, 25: 1 97- 
208. 

RATEL, M. O., J. GÉNERMONT & M. LAMOTTE, 
1989, Relation entre polymorphisme et milieu 
chez les Cepaea nemoralis (Moll. Pulmones) de 
la région parisienne. Bulletin de la Société 
Zoologique de France, 113: 145-154. 

SACCHI, С F., 1984, Population ecology of Ce- 
paea nemoralis and С vindobonensis along the 
north Adriatic coasts of Italy. Malacologia, 25: 
315-323. 

SLÁDEK, I., 1977, Studium geografického 
rozlození potenciálu znediéténi ovzduéí na území 
CSR. [Geographie distribution of factors affecting 
air pollution in the Czech Socialist Republic]. Un- 
published report. Hydrometeorologicky ústav 
Praha, 84 pp. 

TILLING, S. M., 1983, An experimental investi- 
gation of the behaviour and mortality of artificial 
and natural morphs of Cepaea nemoralis (L.). Bi- 
ological Journal of the Linnean Society, 19: 35- 
50. 



MELANISM IN THE LAND SNAIL 87 

VICARIO, Д., L. I. MAZON, A. AGUIRRE, A. ES- oslovakia]. Ústfední správa geodezie a kar- 

TOMBA & С LOSTAO, 1988, Variation in popu- tografie. Praha. 

lations of Cepaea nemoralis (L.) in North Spain. WOLDA, H., 1969, Genetics of polymorphism in the 

Biological Journal of the Linnean Society, 35: landsnail Cepaea nemoralis. Genética 40- 475- 

217-227. 502. 
VESELY, A., ed., 1953, Atlas podnebi Ceskoslo- 

venské republiky. [Climatological atlas of Czech- Revised Ms. accepted 26 June 1992 



MALACOLOGIA, 1993, 35(1): 89-98 

DAILY MOVEMENT PATTERNS AND DISPERSAL IN THE 
LAND SNAIL AR I ANTA ARBUSTORUM 

Anette Baur & Bruno Baur 
Institute of Zoology, University of Basel, Rheinsprung 9, CH-4051 Basel, Switzerland 

ABSTRACT 

The relationship between daily movements of individuals and their dispersal over longer periods 
was studied in two natural populations of the land snail Arianta arbustorum in Switzerland. In a 
forest clearing, daily movements of individually marked snails ranged from to 4.44 m (median 
0.58 m); the frequency distribution of the distances traveled fitted a function with exponential 
decay. The snails showed no preference in direction of movement. Further, the directions chosen 
by an individual on consecutive days were independent from each other. These findings agree 
with the assumptions of a random movement model. In a 1-m wide belt of tall grass and forbs 
along a ditch, daily movements of A. arbustorum were exponentially distributed and ranged from 
to 1.57 m (median 0.40 m). The snails' movements were confined to favourable vegetation; 
individuals that reached the edge of the belt did not enter the drier surroundings (a mown 
meadow); instead they continued to move in a new direction within the belt. 

Using characteristics of the movement pattern of the A. arbustorum population in the forest 
clearing, we simulated snail dispersal in habitats of different shape over longer periods. The 
simulations showed that snails dispersed significantly longer distances in a two-dimensional 
habitat than in linear habitats of 1 and 8 m width. A comparison with literature data on helicid 
snails dispersing in two-dimensional habitats (meadows, pastures) and linear habitats (roadside 
verges, river embankments, hedges) supports this result. 

Key words: Arianta arbustorum, Gastropoda, dispersal, gene flow, movement pattern, habitat. 



INTRODUCTION 

The distances moved by organisms be- 
tween locations where they are born and 
where they mate and reproduce are important 
determinants of population structure. From a 
population genetics perspective, vagility can 
strongly influence effective population size 
and the rate of gene flow, especially when 
populations are spatially structured by discon- 
tinuities of suitable habitats or resources. Re- 
stricted gene flow, in turn, can lead to genetic 
differentiation of local populations as a result 
of locally differing selection pressures or ge- 
netic drift. 

Dispersal in non-flying animals is often con- 
fined to suitable habitat. Type and heteroge- 
neity of habitat, local population density and 
such individual characteristics as body size, 
age, nutritional condition and homing ten- 
dency have been assumed to influence dis- 
persal in terrestrial gastropods (e.g. Cain & 
Currey, 1968; Greenwood, 1974; Pollard, 
1975; Oosterhoff, 1977; Dan, 1978; Cook, 
1979, 1980; Lind, 1988, 1989; Baker & 
Hawke, 1990). The purpose of this study is 
twofold. First, we quantify the relationship be- 
tween daily movement patterns of individuals 



of the land snail Arianta arbustorum (L.) and 
the distances dispersed during periods of dif- 
ferent lengths. Second, we examine the effect 
of habitat form (either two-dimensional or lin- 
ear) on distances dispersed. 

Dispersal is defined here as the distance 
travelled by a snail in its daily activity during 
periods longer than one day (Endler, 1977). 
Daily movement, or distance covered per day, 
is defined as the straight line between the po- 
sitions of an individual on two successive 
days. We assume that the snails live in rela- 
tively homogeneous habitats, and conse- 
quently in the present context ignore directed 
seasonal migrations between hibernation, 
aestivation and oviposition sites as described 
for Helix pomatia (Edelstam & Palmer, 1950; 
Pollard, 1975; Tischler, 1973; Lind, 1989), 
Theba pisana (Johnson & Black, 1979; 
Johnson, 1981; Lebel, 1991) and Cernuella 
virgata (Baker, 1988a, b). 

MATERIALS AND METHODS 

The Species 

Arianta arbustorum is a simultaneously her- 
maphroditic helicid gastropod that is common 



89 



90 



BAUR & BAUR 



in moist habitats in northwestern and central 
Europe (Kerney & Cameron, 1979). Shell 
growth is restricted to spring and summer and 
is completed after one or several hibernations 
with the formation of a shell lip at the edge of 
the shell aperture, with adult snails measuring 
16-20 mm in shell diameter (Baur & Raboud, 
1988; Baur, 1990). The mean adult life span 
of A. arbustorum is 3-4 years, but a maxi- 
mum longevity of 14 years after reaching sex- 
ual maturity has been recorded (Baur & 
Raboud, 1988). 

Locomotory activity occurs only under par- 
ticular physical conditions, temperature, pho- 
topehod and air humidity being the important 
determinants (Cameron, 1970a, b). During 
periods of drought and heat, A. arbustorum 
aestivates either buried in the soil or attached 
to leaves and stems of plants (Frömming, 
1954; В. Baur, 1984, 1986). During winter the 
animals hibernate in the soil (Frömming, 
1954;Terhivuo, 1978). 

Recording of Movement Patterns 

Daily movements of A. arbustorum were re- 
corded in a grass-covered clearing, 20 x 30 m 
in size, in a coniferous forest 10 km south of 
Basel, Switzeriand (47°28'N, 7°34'E; altitude 
360 m a.s.l.). A grid of 25 units, each 4 m^ in 
area, was set up in the central part of the 
clearing by marking the corners of each unit 
with a stake. Sixteen subadult (individuals 
with a shell diameter > 8 mm but without a 
reflected lip at the shell aperture) and 51 adult 
(individuals with a reflected lip) A. arbustorum 
were collected within the clearing and individ- 
ually marked on their shells with numbers 
written in permanent felt pen on a spot of cor- 
rection fluid (Tipp-Ex). The shell diameter of 
each snail was measured to the nearest 0.1 
mm with vernier callipers. Marking and mea- 
suring were carried out in the field, and the 
snails were released immediately at their orig- 
inal positions. On 1 1 consecutive days in April 
and five days in May 1990 the grid and the 
adjacent area within 5-8 m were carefully 
searched for marked A. arbustorum. The po- 
sition of each marked snail was recorded by 
measuring the distances to the nearest two 
stakes of the grid; based on these data, co- 
ordinates were calculated. Field work was al- 
ways done in the late afternoon; therefore the 
snails' positions usually represent their day- 
time resting sites. 

Using the coordinates of the position of 
each snail, we calculated: (1) the distance be- 



tween the positions on two successive days 
(to the nearest cm), (2) the angle of each daily 
displacement relative to the grid system ( = 
orientation of movement), and (3) the angle 
(measured in a counter-clockwise direction) 
between two successive daily displacements. 

To test the accuracy of the method, the 
daily positions of 32 snails were marked with 
numbered flags. The distances between suc- 
cessive positions were measured directly and 
compared with those calculated from coordi- 
nates using correlation analysis. The direct 
measurement of displacements was simple, 
but did not allow any estimate of angles be- 
tween successive movements. The calcu- 
lated distances covered were highly corre- 
lated with those measured directly (r = 0.998, i 
d.f. = 60, p < 0.001), indicating a high accu- 
racy of the coordinate method. 

To estimate dispersal over a longer period, 
the clearing was carefully searched for A. ar- 
bustorum 30 days after initiation of the study. 
Later observations (after two and three 
months) indicated that some snails had 
reached the clearing's edge, which consisted 
of stands of blackberry (Rubus corylifolius). 
However, no snails were found in blackberry 
stands and in the adjacent pine forest, indi- 
cating that this type of habitat was repellent to 
the snails and thus influenced their move- 
ments. 

Daily minimum and maximum air tempera- 
tures were obtained from a minimum-maxi- 
mum thermometer placed 10 cm above 
ground in the clearing. Data on precipitation 
and duration of sunshine were recorded at 
Aesch and Schönenbuch, situated 3 and 8 km 
away from the clearing. During the study, the 
weather was favourable for snail activity: daily 
minimum temperatures ranged from 2.5 to 
14.0°C and maximum temperatures from 10.5 
to 21 .0°C. Precipitation was distributed fairiy 
evenly over the period and occurred on 10 of 
the 1 6 days. 

Daily movements of A. arbustorum were 
also monitored in a 1-m-wide and 50-m-long 
belt of forbs and grass in a subalpine pasture 
at Potersalp, 1290 m a.s.l., in the eastern 
Swiss Alps (47°17'N, 9°20'E). Snail densities 
of up to 6.8 adults per m^ were recorded (B. 
Baur, 1986). The height of the vegetation in 
the belt was 30-50 cm. A partly overgrown 
ditch (5-20 cm wide) ran down the middle of 
the belt. The meadows adjacent to both sides 
of the belt were cut to a height of 7-10 cm. 
For detailed description of the habitat and lo- 
cal climate, see B. Baur (1986, site A). 



DAILY MOVEMENTS AND DISPERSAL IN A LAND SNAIL 



91 



In September 1 981 , 60 A. arbustorum were 
individually marked with numbers in India ink 
on 1 mm X 2 mm pieces of paper glued onto 
their shells. Shell size was measured as 
above. Marking was carried out in the field, 
and the snails were immediately released at 
their original positions. A grid of 1 m^ units (1 1 
squares in a line) was set up to enable re- 
cording of the positions of marked snails. 
Daily displacements of snails were recorded 
as above during five successive days. 

Air temperature and relative humidity were 
recorded by a thermohygrograph 10 cm 
above ground in the belt of tall vegetation. 
During this study, minimum air temperature 
ranged from 0.8 to 2.4°C, and maximum air 
temperature from 3.2 to 10.5°C. Humidity in 
the vegetation belt averaged 86.5% (range 
79.4-94.8%). 

In the vegetation belt, a second experiment 
was conducted to examine dispersal of A. ar- 
bustorum over a longer period using the same 
grid. On 16 August 1981, 92 A. arbustorum 
were marked with dots of car-lacquer; individ- 
uals from each grid unit were marked with a 
different coloured lacquer. Snails were 
marked in the field and released as described 
above. After ten months, the grid and the ad- 
jacent area within 10-20 m were carefully 
searched. The positions of marked individuals 
were recorded. Dispersal of snails was deter- 
mined by calculating the distances between 
the grid units where the snails were marked 
and recovered (distance between neighbour- 
ing units = 1 m). 

Simulation Model 

A model of random movement was used in 
computer simulations to examine dispersal of 
A. arbustorum over longer periods. Random 
movement can be assumed if (1 ) traveling an- 
imals do not prefer any direction, (2) the di- 
rection of movement does not depend on the 
direction of preceding movements, and (3) 
the distance moved by each animal is an ex- 
ponential random variate (Pielou, 1969). The 
pattern of distances covered per day by A. 
arbustorum in the clearing indicates that 
these assumptions were fulfilled as long as 
the snails did not reach its edge (see Re- 
sults). 

To simulate dispersal in a two-dimensional 
habitat, we assumed a uniform distribution of 
angles of orientation (no preference for any 
direction). For each snail, x (= number of 
days) random variâtes generated from the ex- 



ponential distribution of daily distances cov- 
ered (Fig. la) were assigned to a random di- 
rection (derived with an accuracy of Г from a 
uniform distribution in the interval from 0° to 
360°). Daily movements were summed by 
vector addition of Cartesian coordinates re- 
sulting in a final distance moved from the or- 
igin. The entire simulation procedure was re- 
peated for 1,000 "snails," each of them 
"moving" X days from a common starting 
point (X = 10, 20, 30, ... 110, 120 days). We 
assume that 120 days correspond approxi- 
mately to one year of activity in A. arbustorum 
living in lowland populations in Central Eu- 
rope (cf. В. Baur & Raboud, 1988). 

To simulate dispersal in linear habitats of 1 
and 8 m width, for each "snail" random vari- 
ates were generated from the exponential dis- 
tribution of daily distances moved in the clear- 
ing (Fig. la), and a random direction from a 
uniform distribution was assigned to each 
variate. If a "snail" reached one of the edges 
of the linear habitat, a new random direction 
among the angles possible within the favour- 
able habitat was generated, and the "snail" 
moved from its position at the edge the re- 
maining part of the daily distance in this new 
direction. Daily net movements were summed 
as described above. 

Dispersal in linear habitats (river embank- 
ments, roadside verges) is often measured in 
one dimension (i.e. distances dispersed along 
the X-axis only are considered) (Goodhart, 
1 962; B. Baur, 1 984, 1 986; A. Baur & B. Baur, 
1990). To compare simulated dispersal in 
two-dimensional and linear habitats with liter- 
ature data, we also calculated the distances 
dispersed along one axis in our simulations 
for both habitat forms. 



RESULTS 

Movement Patterns in Natural Populations 

In the clearing, the recovery rate of marked 
A. arbustorum averaged 47.5% (range 20.0 - 
71.4%) after 24 h. A total of 119 daily dis- 
tances moved by 50 A. arbustorum were re- 
corded. The distances covered within a day 
ranged from to 4.44 m (median value: 0.58 
m), and their frequency distribution fitted a 
function with exponential decay (Fig. la). A 
proportion of the snails (28.6%, Fig. la) re- 
mained inactive or moved very short dis- 
tances (< 25 cm), even in 24-h intervals with 
favourable weather conditions (rainy nights). 



92 



BAUR & BAUR 



N = 119 




30- 
20- 
10- 


:« 


У--'/ 






В 

N = 45 




1 1 



О 0.5 1.0 1.5 2.0 2.5 3.0 3.5 О 0.5 1.0 1.5 

Dispersal (m) 

FIG. 1 . Frequency distribution of distances moved per day by A. arbustorum in (a) a forest clearing and (b) 
a belt of grass and forbs (1 m wide). Exponential functions were fitted to the distributions: (a) y = 21 .510 
g-oiox ^ ^ QjQ t ^ g74_ (jf = 12, p < 0.001; (b) y = 55.060 e" °^^^ r^ = 0.88, t = 6.00, d.f. = 5. 
p < 0.01 ; X = distance in cm and y = frequency (%). 



The mean distance covered per day (all snails 
considered) was positively correlated with 
daily mininnum temperature (r = 0.59, n = 
16, p = 0.016), and negatively correlated with 
the number of sunshine hours (r = -0.76, n 
= 16, p < 0.001). Thus, snails moved larger 
distances during relatively warm nights, 
whereas sunny days restricted their move- 
ments. The distances moved per day were 
not influenced by the age-class of the snails 
(0.88 m in subadults vs. 0.92 m in adults; 
Mann-Whitney U-test, n = 119, p > 0.4). We 
cannot exclude that data about the most- and 
the least-mobile snails are underrepresented, 
because snails moving long distances are 
less likely to be recovered than those moving 
less far and individuals remaining inactive for 
several days are often buried In the soil. How- 
ever, these sources of bias may balance to 
some extent. 

Representative movement tracks of A. ar- 
bustorum recorded in the clearing are illus- 
trated in Figure 2a. Overall, the snails showed 
no preference in direction of movement (Ray- 
leigh test, n = 119, p > 0.1). Furthermore, 
the direction chosen by a traveling snail was 
independent of that of the preceding day 
(Rayleigh test, n = 45, p > 0.2). Finally, the 
snails moved equal distances in all directions 
(Kruskal-Wallis test, d.f. = 5, p > 0.6, analy- 
sis based on sectors of 60°). Six A. arbusto- 



rum were recovered 30 days after marking. 
The distances dispersed averaged 3.43 m 
(range 0.77-6.28 m). 

In the vegetation belt, the recovery rate of 
marked A. arbustorum averaged 42.0% 
(range 33.3-50.0%) after 24 h. A total of 45 
daily distances covered by 25 A. arbustorum 
were recorded. The distances covered were 
exponentially distributed and ranged from to 
1 .57 m (median = 0.40 m) (Fig. 1 b). As in the 
clearing, a proportion of the snails (28.9%, 
Fig. 1b) were inactive or moved distances 
<25 cm even in 24-h intervals with favour- 
able weather conditions. Subadult and adult 
A. arbustorum did not differ in the distances 
covered (0.26 m vs. 0.48 m, Mann-Whitney 
U-test, n = 45, p > 0.05). 

Representative movement tracks of A. ar- 
bustorum living in the vegetation belt along 
the ditch are illustrated in Figure 2b. The 
snails showed no preferred direction of move- 
ment (Rayleigh test, n = 45, p > 0.8). Like- 
wise, the direction chosen by a moving snail 
was independent of that of the preceding day 
(Rayleigh test, n = 18, p > 0.8). Repeated 
observations during the day revealed that the 
snails did not enter the drier surroundings (a 
mown meadow); individuals that reached the 
edge of the vegetation belt continued their 
movements in a new direction within the fa- 
vourable habitat. The repeated returning at 



DAILY MOVEMENTS AND DISPERSAL IN A LAND SNAIL 



93 



A 





FIG. 2. Representative movement tracks of individuals of A. arbustorum in (a) a clearing and (b) a vegetation 
belt (1 m wide). Dots indicate the snails' positions on consecutive days and arrows the directions of 
movement. Dashed line indicates movement in two days. 



the edges may result in shorter distances dis- 
persed in a linear than in a two-dimensional 
habitat. This suggests that the pattern of dis- 
persal of A. arbustorum is influenced by the 
form of the habitat. 

In the second experiment performed in the 
vegetation belt, 13 out of the 92 marked A. 
arbustorum were recovered after ten months. 
The distances dispersed along the ditch av- 
eraged 6.2 m (range: 0-15 m). 

Simulated Dispersal 

Simulated mean dispersal for 1 ,000 snails 
in a two-dimensional habitat increased from 
4.0 m in 10 days to 14.5 m in 120 days (dis- 
persal in two dimensions considered: Fig. 3a), 
the maximum distances dispersed being 15.1 
m and 39.6 m, respectively. 

The form of the habitat had a significant 
effect on snail dispersal: in linear habitats the 
animals dispersed shorter distances per time 
unit than in a two-dimensional habitat (Fig. 
3a). Furthermore, the width of the linear hab- 
itat influenced snail dispersal (Fig. 3a). When 
dispersal along one axis was considered, the 
distances dispersed per time unit decreased, 
but the difference between habitat forms re- 
mained (Fig. 3b). 

Literature data suggest that helicid snails 
disperse larger distances in two-dimensional 
habitats than in linear habitats, supporting the 
results of our simulation (Table 1). For exam- 
ple, mean dispersal of Cepaea nemoralis was 



found to be 10 m in one year in a grassland in 
England and 4.7 m along a slope of a river 
bank (a linear habitat). Dispersal of A. arbus- 
torum averaged 4.9 m in three months in a 
clearing in central Sweden. Corresponding 
figures for roadside verges of 2 and 2.5 m 
width with similar vegetation were 2.2 m and 
2.9 m, respectively. 



DISCUSSION 

This study indicates that long-term dis- 
persal of land snails can be estimated on the 
basis of daily movements. Our simulation 
model incorporates several assumptions: (1) 
the distribution of daily distances moved does 
not change during the activity season, (2) the 
length of the activity season is fixed (in our 
case 1 20 days), (3) the structure of the habitat 
is homogeneous, and (4) the snails show no 
homing behaviour. 

Our simulations may accurately estimate 
snail dispersal, presupposing that the as- 
sumptions are fulfilled. In the field, the daily 
activity of snails and the distance moved in a 
day are mainly determined by abiotic factors 
(e.g. humidity, changes in temperature, light), 
time of the year, and endogenous rhythms 
(Dainton, 1954; Bailey, 1975; Rollo, 1982; 
Dainton & Wright, 1985; Ford & Cook, 1987; 
Munden & Bailey, 1989). The length of the 
activity period (time from arousal in spring un- 
til hibernation in late autumn) of snails in nat- 



94 



BAUR & BAUR 



TD 

ел 

и 

D 

О. 
ел 



а) 



151 



10- 



5- 




Two-dimensional 
habitat 

Linear habitats: 

(8 m) 
(Im) 



О 20 40 60 80 100 120 



ел 
(ú 
О 

С 

cd 
ч— > 
ел 



Ь) 



10 п 



5- 








Two-dimensonal 
habitat 

Linear habitats: 

(8 m) 
(Im) 



— ' — I — ' 1 ' — I — ' 1 ■ — I ' — I 

20 40 60 80 100 120 



Time (days) 



FIG. 3. Simulated dispersal of snails in habitats of different form for periods of 10, 20 1 10, 120 days. 

Each point represents the mean dispersal for 1,000 snails. For details of the simulation model, see Material 
and methods. Dispersal is calculated in two dimensions (a) and in one dimension (b). 



DAILY MOVEMENTS AND DISPERSAL IN A LAND SNAIL 



95 



ce 

E 
E 

3 

Ю 






CO 



°^J2 0) 
m rö 3 

lo ^ 



™ га ^ 
£í? Ф g 

ФСС 

Û-£E 



J3 -g E 
«5- 



raâ 
•^ ra 

«I 

Xa 
(It 



,- •.- in 
in in 05 

C35 С3> ^ 

s s ^ 

о о ^ 

E E¿ 
га га ü 
_i _i со 



ÜOÜ 



ю 


о 


CD 


Г-- 


C\J^ 


£i 


2;^ 


со 










1^ 


■1- 


00 


т- 


ai 


CXJ 


со 


c\i 






CVJ 


со 




СП 


со 




1Л 


£ 


^ 


со 


га 
ф 


с 
о 

Е 


с 
о 

Е 


га 
ф 


см 


m 


со 


C\J 



■Q S" 
■о îi: 

ф ::г^ — ^ 

.t; т 1— со 

" fc^œ 
^°'-;s Oí 
см C-«î '- 

>« 5 со r^ 
2 S a.r 

!= Ф с D 

5 >- Ф ra 

■^ '" Dû OÛ 



: Ol 



о 

О) 



3 

га 

об 

га 
CD 



3 ^ 



ÜO 



, . 


о 


о 


о 


о 


см 


г^ 


со 


о 


о 


in 


со 


со 


<т> 


00 


со 


00 


in 


¿ 


т— 












Cvj^ 


£S 




















со 


о 


со 


со 


со 


00 


■* 


■* 


со 




со 


■^ 


■г- 


■* 






см 





in со 1- т- со 

"о со CJ) Tt 

см см со ■* со 



со ora 

Ф с <1> 

>s с >% 

■Î- см 1- 



о о о 
Е Е Е 



III I II I I 



со 



о -о ^ 

"О ф -" 

■о ^ я га Ç 
(U га 
с " 

с (О 



с: "С о 

га Ф 

C3)JZ 



сз) 5 i5 '^ 
СП о « Е 

Ос^ф«Е^|^с 
гага^га5о2о2 



ф - Ф 



ф 



ф 



ф ф га 

. . о Ü с 

ся с с t 

га 2 5 <1> 
;gu.u.O 

га 

га 75 
Во 

о о 



t со е с 

^ съ с с 

г ее О) (Ь 

о а с с 
г 5) . . 
нООО 



С 

га 

СЗ) 

с 
ш 



с 
« 

с 05 
га^ 
ф 5 

i:» со 

.3 



со 
со 3 

с ^ 

О С 
•С со 



с 
ф 
тз 
ф 

со 



см 
со 


со 

О) 


<л 


■*" 






•ч_^ 


■* 


с 
га 


00 
О) 


тз 




о 


3 


о 


га 


о 


CD 



га 
CD 


га 

CD 


об 


об 


3 

га 
со 


3 

га 
CD 



о 



о 



■* 


со 




о 

см 


см 


f— 


со 


1 




см 


1 


о 



inOinin COCDOCO-T- 

CD'í-r^'^ r^-r^Tflh'^ 

CD in ^т^ CO CO CO CM ■>- 

COOOCOOO CJ)0>CDCJ)CO 

COOOOCM СЛСОСО'^'"- 



r~- T- к CO 



CO ■* о CM in 



OCON-O COCMCOOCVI 

CMCO'tr^ -r-CMCMCMCD 



s ^ « « 

9? s Ф Ф 



£ ¿ ^ £ -g 

ceceo 
о о о о с 
ЕЕ Е е| 
1- со 1- со 1- 



:¿ 5 

с о 
га -о 

J3 л 

^Е 
о-о 

(U Ф 

о со 

со CJ) 



ф 

ф 0) ф' ф ♦; 

■о Е'таото)^ j; 

^ ф 2 Ф s a.¡¿ 

> -Р > -Р ~ 
•- "- Ф "О 



га 



■" ф Ф 0) Ф 

ф -g -С -g -С •= га 

£ со ^ со ^ -= СП 

о "О СЯТЭ СЗ) ™ с 

со га 3 га 3 -5 о 
о о о о 



s s 



<J о^ 



!û "со 

!i 

га <ü 
ф с 
с 
ZiO 



96 



BAUR & BAUR 



ural populations is relatively well known (e.g. 
Dan, 1978; B. Baur & Raboud, 1988). How- 
ever, at present no data are available about 
the number of days the snails actually show 
locomotory activity in natural populations (but 
see Bailey, 1989a, for activity under experi- 
mental conditions). This represents a major 
problem for any simulation of dispersal. 

Dispersal in land snails has been shown to 
be affected by type and height of vegetation 
(Cain & Currey, 1968; Cowie, 1980, 1984; 
Baker & Hawke, 1990), local population den- 
sity (Greenwood, 1974), snail size (Szlavecz, 
1986; A. Baur & B. Baur, 1988), homing ten- 
dency (Cook, 1979, 1980; Bailey, 1989b), and 
time of the year (Cameron & Williamson, 
1977; B. Baur, 1984, 1986). Possible effects 
of habitat structure, type of vegetation and lo- 
cal population density on daily distances 
moved and thus on dispersal were beyond 
the scope of this study. Furthermore, we con- 
sidered exclusively fully grown and almost 
fully grown individuals of A. arbustorum which 
did not differ in movement behaviour. The dis- 
tribution of daily distances moved may 
change in the course of the activity season. 
Helicid snails have been observed to move 
farther during the reproductive season than in 
autumn shortly before hibernation (Cameron 
& Williamson, 1977; B. Baur, 1984, 1986; A. 
Baur & B. Baur 1990). Detailed data on sea- 
sonal variation of daily distances moved are 
so far lacking. 

The marking procedure, type of release 
(crowded at a central point or individually at 
original positions) and searching procedure 
significantly influence snail dispersal over 
shorter periods (Oosterhoff, 1977; Cowie, 
1980). We tried to minimise the latter effects 
by marking the snails in the field and releas- 
ing them immediately at the positions where 
they were found. However, monitoring of snail 
movements in natural habitats needs re- 
peated recoveries of individually marked 
snails. Intense and repeated searching pro- 
cedures damage the vegetation and change 
the microclimate, which in turn may alter the 
snails' behaviour (Cameron & Williamson, 
1977). Consequently, to record daily move- 
ments, the search intensity should be moder- 
ate, and reduced recovery of marked snails 
must be accepted. Recovery of marked indi- 
viduals is further reduced by the snails' rest- 
ing behaviour. During the activity season, A. 
arbustorum frequently rests for periods of up 
to several days buried in the soil. A proportion 
of snails remain inactive in the soil even under 



conditions favourable for activity (Peake, 
1978). For example. Helix aspersa was active 
in a test arena during 67% of nights with fa- 
vourable conditions (Bailey, 1989a). 

In the vegetation belt, we observed during 
the day that individuals reaching the edge of 
the belt generally did not enter the suboptimal 
surroundings, but rather continued their 
movement in a new direction within the fa- 
vourable habitat. The adjoining mown 
meadow may constitute an unsuitable habitat 
to A. arbustorum for several reasons. The 
short vegetation of the meadow retains less 
humidity and hence, curtails the snails' activ- 
ity. Daily fluctuations in temperature may be 
more extreme and insolation more intense in 
short grass than in the tall vegetation of the 
belt. Furthermore, the short vegetation makes 
snails more vulnerable to visually hunting 
predators (the song thrush, Turdus philome- 
los, is an important predator of A. arbustorum 
in that area; B. Baur, 1984). Finally, due to 
repeated cutting, different species of grass 
dominated the meadow (grass is not a major 
constituent of the diet of A. arbustorum; Fröm- 
ming, 1954; Speiser & Rowell-Rahier, 1991). 

Literature data revealed that snails dis- 
persed shorter distances in linear habitats 
than in unlimited two-dimensional habitats 
supporting the results of our simulation study. 
The fact that dispersal is reduced in linear 
habitats may be of importance for estimates 
of effective population size and rate of gene 
flow. 



ACKNOWLEDGEMENTS 

We thank S. E. R. Bailey, G. H. Baker, T. 
Ebenhard, J. Shykoff, S. Ulfstrand and an 
anonymous reviewer for comments on the 
manuscript and A. Ulfstrand for drawing the 
figures. Financial support was received from 
the Swiss National Science Foundation (grant 
31-26258.89). 



LITERATURE CITED 

BAILEY, S. E. R., 1975, The seasonal and daily 
patterns of locomotor activity in the snail Helix 
aspersa Müller, and their relation to environmen- 
tal variables. Proceedings of the Malacological 
Society London, 41 : 415-428. 

BAILEY, S. E. R., 1989a, Daily cycles of feeding 
and locomotion in Helix aspersa. Haliotis, 19: 23- 
31. 



DAILY MOVEMENTS AND DISPERSAL IN A LAND SNAIL 



97 



BAILEY, S. E. R., 1989b, Foraging behaviour of 
terrestrial gastropods: integrating field and labo- 
ratory studies. Journal of Molluscan Studies, 55: 
263-272. 

BAKER, G. H., 1988a, Dispersal of Theba pisana 
(Mollusca: Helicidae). Journal of Applied Ecol- 
ogy, 25: 889-900. 

BAKER, G. H., 1988b, The dispersal of Cernuella 
virgata (f^ollusca: Helicidae). Australian Journal 
of Zoology, 36: 513-520. 

BAKER, G. H. & B. G. HAWKE, 1990, Life history 
and population dynamics of Tfieba pisana (Mol- 
lusca: Helicidae) in a cereal-pasture rotation. 
Journal of Applied Ecology, 27: 16-29. 

BAUR, A. & B. BAUR, 1988, Individual movement 
patterns of the minute land snail Punctum pyg- 
maeum (Draparnaud) (Pulmonata: Endodon- 
tidae). Veliger, 30: 372-376. 

BAUR, A. & B. BAUR, 1990, Are roads barriers to 
dispersal in the land snail Arianta arbustorum? 
Canadian Journal of Zoology, 68: 613-617. 

BAUR, В., 1984, Dispersion, Bestandesdichte und 
Diffusion bei Arianta arbustorum (L.) (/Mollusca, 
Pulmonata). Ph.D. Thesis, University of Zürich. 

BAUR, В., 1986, Patterns of dispersion, density 
and dispersal in alpine populations of the land 
snail Arianta arbustorum (L.) (Helicidae). Holarc- 
tic Ecology 9: 117-125. 

BAUR, В., 1990, Seasonal changes in clutch size, 
egg size and mode of oviposition in Arianta ar- 
bustorum (L.) (Gastropoda) from alpine popula- 
tions. Zoologischer Anzeiger, 225: 253-264. 

BAUR, В. & С. RABOUD, 1988, Life history of the 
land snail Arianta arbustorum along an altitudinal 
gradient. Journal of Animal Ecology, 57: 71-87. 

BENGTSON, S.-A., A. NILSSON, A. NORD- 
STROM & S. RUNDGREN, 1976, Polymorphism 
in relation to habitat in the snail Cepaea hortensis 
in Iceland. Journal of Zoology, London, 178: 
173-188. 

CAIN, A. J. & J. D. CURREY, 1968, Studies on 
Cepaea. III. Ecogenetics of a population of Ce- 
paea nemoralis (L.) subject to strong area ef- 
fects. Philosophical Transactions of the Royal 
Society London Series B, 253: 447-482. 

CAMERON, R. A. D., 1970a, The survival, weight- 
loss and behaviour of three species of land snail 
in conditions of low humidity. Journal of Zoology, 
London, 160: 143-157. 

CAMERON, R. A. D., 1970b, The effect of temper- 
ature on the activity of three species of helicid 
snail (Mollusca: Gastropoda). Journal of Zool- 
ogy, London, 162: 303-315. 

CAMERON, R. A. D. & P. WILLIAMSON, 1977, Es- 
timating migration and the effects of disturbance 
in mark-recapture studies on the snail Cepaea 
nemoralis (L.). Journal of Animal Ecology, 46: 
173-179. 

COOK, A., 1979, Homing in gastropods. Malacolo- 
gia, 18:315-318. 

COOK, A., 1980, Field studies of homing in the 
pulmonale slug Umax pseudoflavus (Evans). 
Journal of Molluscan Studies, 46: 100-105. 



COWIE, R. H., 1980, Observations on the dispersal 
of two species of British land snail. Journal of 
Conchology, 30: 201-208. 

COWIE, R. H., 1984, Density, dispersal and neigh- 
bourhood size in the land snail Theba pisana. 
Heredity, 52: 391-401. 

DAINTON, B. H., 1954, The activity of slugs. I. The 
induction of activity by changing temperatures. 
Journal of Experimental Biology, 31 : 1 65-1 87. 

DAINTON, B. H. & J. WRIGHT, 1985, Falling tem- 
perature stimulates activity in the slug Arion ater. 
Journal of Experimental Biology, 118: 439-443. 

DAN, N., 1978, Studies on the growth and ecology 
of Helix aspersa t\/lüller. Ph.D. Thesis, University 
of Manchester. 

EDELSTAM, С & С. PALMER, 1950, Homing be- 
haviour in gastropods. Oikos, 2: 259-270. 

ENDLER, J. A., 1977, Geographic variation, speci- 
ation and dines. Princeton University Press, 
Princeton. 

FORD, D. J. G. & A. COOK, 1987, The effects of 
temperature and light on the circadian activity of 
the pulmonale slug Umax pseudoflavus Evans. 
Animal Behaviour, 35: 1754-1765. 

FRÖMMING, E., 1954, Biologie der mitteleuropäis- 
chen Landgastropoden. Duncker & Humblot, 
Berlin. 

GOODHART, С В., 1962, Variation in a colony of 
the snail Cepaea nemoralis. Journal of Animal 
Ecology 31 : 207-237. 

GREENWOOD, J. J. D., 1974, Effective population 
numbers in the snail Cepaea nemoralis. Evolu- 
tion, 28: 513-526. 

JOHNSON, M. S., 1981, Effects of migration and 
habitat choice on shell banding frequencies in 
Theba pisana at a habitat boundary. Heredity, 
47: 121-133. 

JOHNSON, M. S. & R. BLACK, 1979, The distribu- 
tion of Theba pisana on Rottnest Island. Western 
Australian Naturalist 14: 140-144. 

KERNEY, M. P. & R. A. D. CAMERON, 1979, A 
field guide to the land snails of Britain and north- 
west Europe. Collins, London. 

LAMOTTE, M., 1951, Recherches sur la structure 
génétique des populations naturelles de Cepaea 
nemoralis (L.). Bulletin Biologique de la France et 
de la Belgique, Supplement 35: 1-239. 

LEBEL, T., 1991, The distribution of the Mediterra- 
nean snail, Theba pisana (Mollusca: Helicidae), 
on Rottnest Island, Western Australia. Western 
Australian Naturalist, 18: 217-222. 

LIND, H., 1988, The behaviour of Helix pomatia L. 
(Gastropoda, Pulmonata) in a natural habitat. Vi- 
denskabelige Meddelelser fra Dansk Naturhisto- 
risk Forening, 147: 67-92. 

LIND, H., 1989, Homing to hibernating sites in Helix 
pomatia involving detailed long-term memory. 
Ethology 81:221-234. 

MUNDEN, S. К. & S. E. R. BAILEY, 1989, The 
effects of environmental factors on slug behav- 
iour. In I. HENDERSON, ed., Slugs and snails in 
world agriculture. Monograph 41: 349-354. Brit- 
ish Crop Protection Council, Thornton Heath. 



98 



BAUR & BAUR 



MURRAY, J. J., 1962, Factors affecting gene fre- 
quency in some populations of Cepaea. Ph.D. 
Thesis, University of Oxford. 

OOSTERHOFF, L. M., 1977, Variation in growth 
rate as an ecological factor in the landsnail Ce- 
paea nemoralis (L.). Netfierlands Journal of Zo- 
ology, 27: 1-132. 

PEAKE, J., 1978, Distribution and ecology of the 
Stylommatophora. In v. fretter & j. peake, eds., 
Pulmonates. Vol. 2A: Systematics, evolution and 
ecology. Academic Press, London. 

PIELOU, E. C, 1969, Mathematical ecology. John 
Wiley & Sons, New York. 

POLLARD, E., 1975, Aspects of the ecology of He- 
lix pomatia L. Journal of Animal Ecology, 44: 
305-329. 

ROLLO, С. D., 1982, The regulation of activity in 
populations of the terrestrial slug Umax maximus 
(Gastropoda: Limacidae). Researcfi in Popula- 
tion Ecology, Kyoto, 24: 1-32. 

SCHNETTER, M., 1951, Veränderungen der ge- 



netischen Konstitution in natürlichen Popula- 
tionen der polymorphen Bänderschnecken. Zoo- 
logischer Anzeiger, Supplement ^ 5: 192-206. 

SPEISER, B. & M. ROWELL-RAHIER, 1991, Ef- 
fects of food availability, nutritional value, and al- 
kaloids on food choice in the generalist herbivore 
Arianta arbustorum (Gastropoda: Helicidae). Oi- 
kos, 62:306-318. 

SZLAVECZ, K., 1986, Food selection and nocturnal 
behavior of the land snail l\Aonadenia hillebrandi 
mariposa A. G. Smith (Pulmonata: Helmintho- 
glyptidae). Veliger, 29: 183-190. 

TERHIVUO, J., 1978, Growth, reproduction, and hi- 
bernation of Arianta arbustorum (L.) (Gas- 
tropoda, Helicidae) in southern Finland. Annales 
Zoologici Fennici, 15: 8-16. 

TISCHLER, W., 1973, Zur Biologie und Oekologie 
der Weinbergschnecke {Helix pomatia). Faunis- 
tisch-ökologische f^itteilungen, 4: 283-298. 

Revised Ms. accepted 28 October 1 992 



MALACOLOGIA, 1993, 35(1): 99-117 

GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 
MÜLLER STUDIED UNDER LABORATORY CONDITIONS 

Luc Madec & Jacques Daguzan 

Laboratoire de Zoologie et d'Ecophysiologie (LA. INRA) Université de Rennes I, 
Campus de Beaulieu Av. du Général Ledere, 35042 Rennes CEDEX, France 

ABSTRACT 

The reproductive characteristics of the land snail Helix aspersa were investigated under 
artificial conditions in ten populations exposed to contrasting selective pressures in their natural 
environments. Two of them were studied for two different years. 

Significant geographic variation was detected not only in fecundity (clutch number, clutch size 
significantly related to shell size) but also in the timing of mating and egglaying. Thus, seasonal 
adjustments (breeding season and duration), related to the geographic location of populations, 
seemed to be partially preserved under uniform laboratory conditions. 

In order to assess the extent of genetic or environmental determination of variation in these 
characters, three successive generations of snails from four ecologically distinct regions were 
reared under the same artificial conditions. This experiment revealed that a large proportion of 
the initially observed variation in natural populations from Lorient and Toulouse, France, and in 
snails from St. Denis, La Réunion, was environmentally induced. Animals born and reared in the 
laboratory exhibit similar traits: they mate two or three times, lay a mean of 1 .3 clutches corre- 
sponding to between 120 and 130 eggs per snail. On the other hand, snails from Algeria retain 
their natural characteristics (larger shell size, larger clutches with larger eggs) under artificial 
conditions. 

In the context of intraspecific life-history variation of Helix aspersa, observed combinations of 
traits might illustrate two tactics: (i) Snails from Algeria have a large size {H. a. maxima), which 
allows them to have a higher egg production in comparison with "norms" of the species (i.e. all 
other known populations), but not with respect to their shell volume (smaller than possible clutch 
volume). This production could compensate for a high mortality, which would affect all age 
categories in the field, (ii) Life-history patterns of populations from more or less recently colo- 
nized habitats, always dependant on human activities, would be considered as the second tactic 
of the species: stable populations of smaller adults with a smaller egg production and consid- 
erable plasticity in life-history traits. 

Key words: Helix aspersa, reproduction, geographic variation, phenotypic plasticity. 



INTRODUCTION 

The helicid land snail Helix aspersa Müller, 
native to the western Mediterranean area, is 
now very abundant in human-modified habi- 
tats of northwestern Europe. This wide distri- 
bution leads to geographic variation in annual 
activity rhythms. Thus, the breeding season is 
restricted to spring and summer in northern 
localities, to autumn or even winter in the 
Mediterranean area (Chevallier, 1983). Peri- 
ods of activity are followed in northern lati- 
tudes by hibernation, which has a diapause 
value (Bailey, 1983; Lorvelec & Daguzan, 
1990), in southern ones by estivation, which, 
in some cases, is only a warm torpor. Sacchi 
(1971) suggested that reproduction Is poten- 
tially continuous and might occur during all 
sufficiently wet and warm periods of the year. 
Thus, the annual activity rhythm and life cycle 



of this species present a high degree of flex- 
ibility, of which an important part can be ob- 
served in the same population. Previous stud- 
ies have also documented variation in the 
seasonality of reproductive activity (Potts, 
1975; Crook, 1980; Madec & Daguzan, 1991) 
and geographic variation in egg production 
per snail (Guemene & Daguzan, 1982). 

In other pulmonate landsnails, several life- 
history traits (growth rate, age at maturity, 
adult size, and life span) often covary with 
reproductive characters (Peake, 1978; 
Calow, 1983; Cowie, 1984). Some combina- 
tions clearly adapt the populations to local cli- 
matic conditions (Baur & Raboud, 1988). 
However, such covariation need not be ad- 
aptative, and it is therefore necessary to de- 
termine the genetic component of the varia- 
tion. Quantitative genetic methods should 
permit this determination (heritabllities and 



99 



100 



MADEC & DAGUZAN 



genetic correlations), but their use often pre- 
sents many technical difficulties. Another ap- 
proach consists of transplant experiments to 
artificial conditions to observe if natural con- 
trasts remain constant through several gener- 
ations of laboratory culture or if the progenies 
converge to a common form (Clarke et al., 
1978; Brown, 1985). 

The first approach has yielded estimates of 
heritability for shell characters, including a 
significant genetic component of shell size 
variation among populations (Clarke et al., 
1978; Goodfriend, 1986). The inheritance of 
variation in Helix aspersa shell size, which is 
very extensive in natural conditions and 
strongly correlated with fecundity, has been 
studied using both the first (Crook, 1980; 
Panella, 1982) and second approaches (Ma- 
dec, 1989a). In this way, laboratory colonies 
of four natural populations characterized by 
large differences in adult shell size showed 
the strong influence of the environment (cli- 
mate, population density) in determining 
small size (dwarfs from the island of La Ré- 
union) and a primary role played by the ge- 
netic component in the determination of the 
giant size of individuals from Algeria {Helix 
aspersa maxima Taylor). However, the great 
phenotypic plasticity shown by the other 
snails {Helix aspersa aspersa Müller) could 
be itself under genetic control. 

The present study reports on: (i) natural vari- 
ability in reproductive traits of Helix aspersa 
examined in samples from ten localities cov- 
ering its whole ecological range. (Because the 
experiments took place under uniform labora- 
tory conditions, this comparative study was 
designed primahly to obtain information on 
variation in reproductive potential of the spe- 
cies, but can also be used to discuss the dis- 
turbances in activity rhythms of transplanted 
snails from contrasting habitats.) (ii) examina- 
tion of the persistence of variation under the 
same conditions, following the continuous 
rearing of three generations of snails from four 
source populations with different life histories. 



MATERIAL AND METHODS 

Relevant reproductive behaviour of Helix 
aspersa has been described by Tompa 
(1984) and Adamo & Chase (1988). 



Origin and Maintainance of Animals 

Random samples were collected from col- 
onies covering the whole range of the spe- 
cies. Snails were taken as adults from their 
natural environments from April to May 1983 
or/and 1985, just after the natural hibernation 
for samples from France and Balearios, and 
after the winter activity for snails from Algeria; 
the annual activity rhythm of snails from La 
Réunion is not known, but animals were ac- 
tive or just attached with strong mucus to var- 
ious hard surfaces when they were collected. 
French populations sampled included (Fig. 1): 
Lorient (northwest). Surgères (central-west), 
Toulouse (southwest), Belmont (east), Lyon 
(central-east), Avignon (southeast), and Bas- 
tía (Corsica). Colonies from Lorient, Belmont 
and St. Denis de La Réunion were sampled 
twice. A comparative study of colonies from 
Lorient and other Breton populations had al- 
ready shown that the only significant variation 
between samples concerned the start of the 
breeding season (Madec & Daguzan, 1991); 
in the present study, we used only the sample 
from Lorient to represent this region and re- 
ferred, if necessary, to the others. In addition, 
we also studied a sample from a population 
recently introduced by man from Brittany (Ma- 
dec, 1991) to St. Denis de La Réunion, a vol- 
canic island of the Mascarene Archipelago 
(Indian Ocean), a sample from Palma de 
Mallorca (Balearles), and another from Alger 
(Algeria). Snails from this last sample belong 
to a different subspecies, namely Helix as- 
persa maxima, initially described by Taylor in 
1883, more recently studied by Chevallier 
(1983). Climatic data for each locality are il- 
lustrated in Figure 1 . 

From the natural populations, two from 
France (Lorient, Toulouse) and those from St. 
Denis and Algeria were selected to represent 
the most important variations of reproduction 
in this species. However, the breeding of the 
Algerian stock could not be maintained and 
consequently, only the results from the sam- 
ple of snails collected in the natural popula- 
tion and a sample of the F6 generation of an 
experimental population obtained from collab- 
orating researchers^ are presented here. For 
the others, four generations were identified as 
follows: 

— AS generation: snails collected as adults 
in their natural environment; 



M. с Bonnet, Institut National de la Recherche Agronomique, Domaine du Magneraud, Surgères. 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 101 

T IMP 




FIG. 1. Location of the ten sampled sites (except St. Denis de la Réunion), sampling dates, and diagrams 
of relation during one year between rainfall (P: mean monthly rainfall, mm) and temperature (T: mean 
monthly temperature, °C). 



— JS generation: snails, collected as juve- 
niles in their natural environment, which be- 
came mature in artificial conditions; 

— Fl generation: offspring of random 
crosses between individuals of the AS gener- 
ation; 



— F2 generation: offspring of random 
crosses between individuals of the Fl gener- 
ation. 

In the laboratory, snails of the AS generation 
remained into an artificial hibernation (5±1°C; 
60 ±5% R.H.; 0L:24D light cycle) for one week, 



102 



MADEC & DAGUZAN 



except for samples from La Réunion and Al- 
geria, which were kept directly in the breeding 
conditions. For the others, revival was trig- 
gered in a room at 12°C, 80% R.H. and a 
12L:12D light cycle, in which snails were fed 
again. For the reproduction experiment, snails 
were reared in controlled temperature and rel- 
ative humidity rooms maintained at 20±1°C, 
80±5% R.H. and a 16:8 light:dark cycle. They 
were housed in polythene containers (50 x 30 
X 10 cm; 29 X 18x7 cm) with a biomass 
density per cage of approximatively 1 8 kg/m^ 
(1 3-1 5 individuals in small boxes and 35 in the 
others). These values were selected as opti- 
mal for breeding activity of snails living in west- 
ern France, e.g. Surgères or Lorient (Da- 
guzan, 1 981 ; Le Guhennec & Daguzan, 1 983). 
For snails from Algeria, which are larger, the 
density was 30 kg/m^ (8-10 individuals in 
small boxes). At least two replicate cages were 
used per population to take possible "cage 
effects" into account. Furthermore, the loca- 
tion of boxes in the rearing room was changed 
each day, and adjoining boxes always con- 
tained snails from different populations. 

All individuals were fed with the same com- 
posite food supplied ad libitum and renewed 
at least twice a week. Water was available in 
a watering place, and the synthetic foam cov- 
ering the cage bottom was kept moist and 
washed every day. Laying jars containing a 
moist and light soil (sterilized compost) were 
placed in the cages, two in small cages and 
four in large ones. A jar was replaced by an- 
other as soon as a snail laid in it. Afterwards, 
jars with clutches were transferred to an incu- 
bator (T = 20±rC; R.H. = 100%; 12L:12D). 

For the JS, Fl, and F2 samples, growth 
and reproduction occurred under the same 
conditions of temperature (20°C), photope- 
riod (16L:8D), and humidity (80% R.H.), and 
with the same diet. However, during growth, 
snails were sorted, and densities modified ac- 
cording to snail size to avoid the effects of 
crowding (Madec, 1989a). After the growth 
period, which finished approximatively three 
months after birth in F1-F2 generations, 
snails were induced into artificial hibernation 
for three months (5°C; 60% R.H.; 0L:24D light 
cycle). Revival was triggered in a room at 
12°C, 80% R.H. and a 12L:12D light cycle, in 
which snails were fed again. 

Methods 

Adult Measures and Monitoring: Adult shell 
height and maximum breadth were measured 
to the nearest 0.1 mm using a vernier calliper; 



each animal was numbered with an adhesive 
stamp. Mating and egg-laying in Helix as- 
persa have durations of about eight hours and 
18 hours respectively, so two daily observa- 
tions (08:00 hr; 18:00 hr) permitted monitoring 
of all layings and 97% of the matings (per- 
centage based on dart presence in a cage 
without mating observation). Dates when in- 
dividuals resumed activity and dates of death 
were also recorded. The length of the repro- 
ductive season was different for each popu- 
lation because it was based on the end of 
layings, which generally coincided with the 
start of a higher mortality. 

Egg Collection and Measures: Each clutch 
was identified by its parentage and its position 
(1st, 2nd, 3rd clutch of the same snail), date 
of laying, its size (number of eggs), and 
hatching date. Of each clutch from AS, JS, 
Fl, and F6 populations, 30 eggs chosen at 
random were weighed (±0.01 g) and their di- 
ameter (diameters when ovoid) measured 
with a dial calliper (±0.01 mm). After that, all 
the eggs were replaced in a soil cavity, and 
the laying-jar was covered by a plexiglass 
plate before being placed in the incubator. 
Newly hatched juveniles emerging from the 
soil were counted, removed and the durations 
of incubation and hatching noted. From each 
hatching, 30 individuals chosen at random 
were weighed. 

Statistical Methodology 

Data analysis was performed using the 
STAT-ITCF (1988) programs. Where possi- 
ble, contingency tables were studied with the 
help of x^ tests; samples with quantitative 
data were compared with analysis of variance 
followed by S.N.K. multiple comparisons 
tests, if the F was significant. The t-test was 
previously used to compare the means of the 
different cages of the same sample. When 
differences were not significant (P > 0.05), 
we used one set of data per sample. When 
non-normality or heterogeneity in variances 
were detected or could not be tested, non- 
parametric statistics were adopted (see 
Results). 

RESULTS 

Variation Between Samples in Reproductive 
Activity Under Artificial Conditions 

Timing Fluctuations: Significant variations 
between AS snail samples were observed not 
only in the dates of resumption or termination 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 103 



of mating and laying activities, but also in the 
rhythm of these activities during the breeding 
period (Fig. 2). Thus, mean numbers of days 
between revival and the mating and laying ac- 
tivities measured for the ten first reproducing 
individuals for each sample were significantly 
different (Kruskal-Wallis tests; P < 0.001). 
According to the non-parametric test of mul- 
tiple comparisons with a level of significance 
P = 0.01 (Scherrer, 1984), snails from north- 
ern France formed two homogeneous groups 
(Lorient/Surgères; Lyon/Belmont), in which 
snails started to reproduce after one week, 
significantly earlier than snails from Toulouse 
and Avignon, which started to mate more than 
two weeks after revival, from Bastia and 
Palma (on average eight weeks during which 
many snails had reformed an epiphragm), 
and from Algeria, which were not sexually ac- 
tive before October (24 weeks), as they were 
in their natural environment. The level of sex- 
ual activity of snails from St. Denis was al- 
ways low, but this sample was relatively close 
to the group Lorient/Surgères. Groups consti- 
tuted according to first ovipositions gave sim- 
ilar indications, but northern ones were disso- 
ciated and the sample from St. Denis was 
close to Belmont (Lyon < Belmont < St. Den- 
is < Lorient < Surgères, with P < 0.05). In 
addition, there was no significant variation be- 
tween samples of the same populations sam- 
pled for two years (Belmont 1983/1985; St. 
Denis 1985/1986), either for distributions of 
matings numbers per week, or for oviposi- 
tions (Kolmogorov-Smirnov tests; P > 0.05). 
Thus, snails seemed to reproduce gradually 
later from northern to southern populations. 

The phase of reproductive activity in- 
crease up to a peak (first mode of distribu- 
tions of mating numbers per week and, to a 
lesser degree, oviposition numbers) con- 
firmed the distinctions between AS samples. 
Snails from Belmont and Lyon reached a high 
level of reproductive activity in only one week 
and then remained at it for several weeks 
(Fig. 2). On the other hand, we observed a 
slow progression to a single peak for both 
mating and laying activities in the Bastia and 
Palma samples; peaks were followed by a 
fast decrease of reproductive activity, which 
stopped completely three weeks after these 
maxima. In between, other distributions were 
not very different, but the sample from Lorient 
was close to those from Belmont and Lyon, 
and the sample from Avignon was close to 
those from Bastia and Palma. 

The most contrasting curves of seasonal 



activity are shown in Figure 3. In addition to 
the differences between eastern and south- 
ern populations (accentuated by high de- 
grees of skewness of the distributions), we 
noted that effective lengths of the breeding 
period in these two contrasting samples (12- 
1 3 weeks) were shorter than in others (1 4-1 6 
weeks). 

Over three generations in the laboratory 
(JS, Fl , F2, only F6 for Algeria because of the 
small size of the JS sample), the timing of 
both mating (mainly due to a shift in the Tou- 
louse population) and oviposition converged 
among all four populations (Fig. 4). These 
snails tended to produce clutches earlier than 
their conspecifics from the field (Kolmogorov- 
Smirnov tests; Toulouse and Alger, P 
<0.001; Lorient, 0.07>P>0.01; La Réunion 
AS-F2, P = 0.05, N.S. for the other compar- 
isons). Frequency distributions of matings 
and layings per week in the Fl , F2, and (F6) 
generations were not significantly different in 
the four populations (x^ tests: matings, P = 
0.08; layings, P = 0.65). 

Variation in Number of Matings and Clutclies: 
AS populations differed significantly in terms 
of mean rates of mating and egglaying (x^ 
tests; P < 0.001 ); total numbers of matings or 
clutches per sample varied approximately be- 
tween ten (Alger) to 100 (Belmont) (Table 1). 
However, the numbers of matings and 
clutches produced per individual were also 
variable in the same population (Fig. 5). Dis- 
tributions of snails according to their total 
number of matings were significantly different 
between AS populations (x^ test; P < 0.001); 
these variations in level of reproductive activ- 
ity led us to distinguish three significantly dif- 
ferent groups (Simultaneous Test Procedure 
with a significance level P = 0.05): a first 
group of samples with a high level of individ- 
ual activity (Belmont, Lyon, Toulouse, Avi- 
gnon; distributions with a mode of three mat- 
ings per snail), a second with moderate 
activity (Lorient, Surgères; 20% of the snails 
did not mate), and a third group (Bastia, 
Palma, St. Denis) with a low level of activity 
(samples with at least 55% of snails with at 
most one mating). The comparison of distri- 
butions of snails according to their total num- 
ber of clutches led to the distinction of only 
two groups with significantly different levels of 
egglaying activity. Thus, there was sharp con- 
trast between AS samples from mainland 
France and insular ones. 
When distributions of AS and JS individuals 



104 



MADEC & DAGUZAN 



20- 



Lorient 



u 




ll 



до- 
зе - 

% 20- 



Д 



1 2 3 4 S 6 7 8 9 10 И 12 13 14 IS 16 17 18 19 



40 п 



10 - 



Belmont 




1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 



40- 
30- 

20 - 
10 



О 



Toulouse 



il 




Surgères 



шш 



I 2 3 4 S 6 7 8 9 10 И 12 13 14 15 16 17 18 19 



40-1 



20- 



Lyon 



UIjjl 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 



Avignon 



.1 .1 ,1 , 



y 



к 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 



40-1 



30- 



Bastia 



■ ■ лушш 



30 
% 20 

10 H 







Palma 



11Ш 




1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 



40- 




St.-Denis 


% 20- 


■ 


10- 
0- 


id 


l|.í,|],ii,íl,i|j,n 



% 20- 



Alger 



I mu 



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 

Time (weeks) 



12345 12232425262728 

Time (weeks) 



FIG. 2. Weekly variations of mating (solid) and oviposition (crosshatched) numbers, according to the origin 
of snails (expressed as % of the total number of individuals per sample). 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 105 
60t 
40+ 

^ Palma 



% 




Lorient 



Belmont 



1 23456789 10 11 
Time (weeks) 



FIG. 3. Evolution in fortnights of matings (solid line) and clutches (stippled line), expressed as % of their 
respective total numbers within four natural populations of Helix aspersa. 



according to their total number of matings 
were compared, we observed that among all 
the populations (Lorient, Toulouse, St. Den- 
is), only the AS sample from St. Denis was 
unique because 40% of snails had no mating 
activity. In addition, the mode for populations 
raised in the laboratory from field-collected ju- 
veniles was a single mating (Fig. 6). Distribu- 
tions of individuals according to their clutch 
production were also significantly different. 
For the same origin, the total number of 
clutches laid by snails from JS generation 
was always higher than the one of AS gener- 
ation. Moreover, snails from Toulouse (AS, 
JS) were characterized by the highest individ- 
ual clutch production (Table 2). The multiple 
comparisons between all AS and JS popula- 
tions led to three homogeneous groups (Si- 
multaneous Test Procedure with P < 0.01; 
[1]: population JS Toulouse; [2]: populations 
AS-JS Lorient/AS Toulouse/JS St. Denis; [3]: 
populations AS St. Denis/Algeria). 

With the exception of one sample (St. Den- 
is, F2), all Fl, F2, (and F6) populations pre- 



sented a mating rate of 100%, snails often 
mating twice during the period of reproduc- 
tion. Distributions of individuals according to 
their total number of matings were not signif- 
icantly different (x^ test; P = 0.32). Even if the 
total number of matings in the F2 population 
of La Réunion was very low, only one snail did 
not mate during the breeding season. Distri- 
butions of Fl and F2 individuals according to 
their total ovipositions were remarkably ho- 
mogeneous (x^ test; P = 0.68). Snails from 
the Algerian-F6 sample gave a very different 
result: 50% of them produced at least four 
clutches and, on average, twice as many as 
the others (Fig. 7). 

Finally, with the exclusion of these F6 snails, 
all animals born and reared in laboratory con- 
ditions behaved in the same way: the total 
number of matings by sample was low, but 
their distribution among individuals was equal; 
snails had a similar oviposition activity, which 
was expressed, after nearly 1 2 weeks of repro- 
duction, by about 60 clutches for 45 individu- 
als, corresponding to 1 .3 clutches per snail. 



106 



MADEC & DAGUZAN 




^55 



^S^ 



^ V <■ «. V 



s 






fi 



CS — . 





^5В 



<N ^ 



^ 



1^^ 



^^ 



5В 




^^В 




"So 



f- с 



(D 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 107 



о 
о 

JO 

E 
о 



QJ 








Ü 




(0 


^—C^ 






ra 


о 


^ 




Ü 


Ф 


ФТЗ 


> 


CO 


о 


-o 


3 


с 


E 





'^ с 
I— ü 



(0 




с 


со 


О) 


00 


Q 


<У> 










СО 




ся 




0) 


in 
00 


Q 


сз) 










СО 




(0 

Е 


оо 


л 


ст> 


CL 


^ 


СО 


со 


(Л 


00 


со 


сз) 


Ш 


^ 






С 


in 


о 


00 


>чСТ) 1 


_| 


у— 










с 




о 


ю 


Е 


g 


ф 


^ 


CQ 








С 




о 


СГ) 


Е 


00 


ф 




со 




с 








о 


со 


с 


оо 


О) 


о 


> 


т— 


< 




ф 




ел 




3 


in 


о 


оо 


3 

о 


ел 


ь 




со 




Ф 






со 


-Ф 


со 


и><у> 1 


3 


^ 


ел 








с 


со 


ф 


00 




CD 


О 




_1 


~-^ 












со 




ф 




>. 




СП 




с 


с 


Q. 


]д) 


Е 




со 


О 


Ä 



^¿ 

E ö) 

со ^ 

со со 



+1 с;] 



со со 

+1 +1 



о ^ 



см 


00 О) 


сЬ 


in со 


+1 


+1 +1 


со 


со 1- 


Ö 


CÓ ai 
со со 


,- 


у- СО 


CD 


0С> Tt 


+ 1 


+ 1 +1 


00 


t^ CO 


Ö 


5Р ^ 
1^ оо 


т- 


OvJ О 


Ö 


ocj -^ 


+1 


+1 +1 


г^ 


о со 


Ö 


со 00 


о 


со 

со^ 
1- in 


+1 


+1 +1 


1- 


со 00 


'" 


со со 
О) 00 


,- 


Ю 1- 


о 


О) tri 


+1 


+1 +1 


ю 


00 ^ 


■*" 


cvj in 
c\j 00 


сЬ 


CVJ 
1- in 


+1 


+1 +1 


'S- 


CO ^ 



CO 

y- in 

+ 1 +1 

CM r^ 

in CM 



о 




о 




о 


1- in 


+1 


IV. 

in 


+1 


CO 

00 


+1 


+1 +1 


CO 


a> 


1^ 


t^ 


CO 


in 1- 


CO 




cvj 




■r-^ 


iri CM 



y- Ю 

+ 1 +1 



+1 
in 

8 




Ö 

00 






+1 
см 




с> 
со 






+1 

СП 
Ö 




+1 
со 
см 
00 


+1 
со 






■^ 


ф 


со 






^^ 


ф 


со 




Ф 


ф 






о 


■^^ 


п 


О) 




"о 


¿ 


п 


ф 




J3 


о. 






Ф 


со 


Е 


с 




ф 




Е 


SZ 

о 

О 

о 




Е 


со 
о .£ 


ф 

N 


e" 

E, 


с 

со 

ф 


СП 

с 
го 
Е 


3 

с 
с 

со 

ф 


Е 
о 


тз 

с 

ф 
о. 


« 

с 
го 
ф 


со 

CJ) 

с 
го 


3 

с 

с 
го 
Ф 


"О 

с 
ф 

Q. 


3 
с 
с 
го 
Ф 


SI 
О 
3 



108 



MADEC & DAGUZAN 



LORIENT 




SURGERES 




12 3 4 5 6 



BELMONT 




LYON 




12 3 4 5 6 



12 3 4 5 6 



TOULOUSE 




AVIGNON 




12 3 4 5 6 



bastía 




PALMA 




60 - 
50- 
40 - 
% 30- 
20 - 
10 - 



ST.-DENIS 



80 -1 



ALGER 




12 3 4 5 6 

Numbers ofmalings and clutches per aiail 




12 3 4 5 

Numbers of malinus and dulchcs m snail 



FIG. 5. Distributions of the snails (in %) according to their total number of matings (solid) and clutches 
(crosshatched) in the ten natural populations studied. 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 109 



% 60n 
50 
40- 

js 30- 

20 
10 



0' 



LORIENT 



Ë 



1н,1пИ 



60- 
50- 

40 
30 
20 
ЮН 
О 



TOULOUSE 



£ 




12 3 4 5 6 7 



12 3 4 5 6 7 



60 - 
50 - 
40 - 
30 - 
20 
10 
О 



ST. DENIS 



t 



w\ 



12 3 4 5 6 7 




12 3 4 5 6 7 



12 3 4 5 6 7 



1 



6 7 



60 - 
50- 
40 - 
F2 30 - 
20 - 
10 - 




Ú 



12 3 4 5 6 7 



60 - 
50- 
40 
30 
20 
10 




IL 



60 - 










50- 




40 - 




— 




■ 




/ 


^ 


30- 




/ 


/ 


■ 




/ 
/ 


^ 


20 - 




/ 


;; 


10- 




7\ 

/ 


/ 


; 


, 




/ 


/ 


', 


0- 


■ 


/ 


/ 


Ш 



I 



12 3 4 5 6 7 
Number of matings and clutches per snail 



FIG. 6. Distributions of the snails (in %) according to their total number of matings (solid) and clutches 
(crosshatched) in different generations of the three populations considered. 



110 



MADEC & DAGUZAN 



о 


о 


> 


ф 






ü 


тз 


3 




■о 


(5 


отз 




с 


Q. 


ш 


Ш 




ce 


to 




+1 


C\j 






X 


ш 


^— ^ 


_l 


и 


< 


с 



W 



со 



со 




т- 




т- 


C\j 


■,- 






о 


00 


d 


h- 


о 




■^ 


■^ 


^ 


+1 


rv! 


+1 


CD 


+1 


+1 


+1 


|< 


•г-^ 


о 


О) 


CD 


СО 


•* 


■* 


CVJ 


00 


■^ 


cvj 




^ 




^ 


о 


C\j 






со 










со 


о> 







-Ч; ^ т- 



■^ со 



in 

^ со 

+1 +1 

(О C4J 



т- (О 

СП со 

+1 +1 

CD со 



о ч- со 

+1 +1 +1 

со 00 -^ 



О) -ч- 

00 со 

+1 +1 

г^ о 



+1 +1 

ел 00 

■г^ со 

о СП 

см 

см 

^ г^ 
■>- со 

+1 +1 

со 00 

г^ cvi 
см О) 



•^ со 

+1 +1 

Tt CD 



о ■■-со 



■^ со 
о со 



CD о 

со о 
со •^ 



см ю 

т-' Ю 

СП -^ 




GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 1 1 1 




12 3 4 5 6 7 

Numbers of matings and clutches per individual 



80-1 



60- 



% 



40- 



20 - 



F6 




12 3 4 5 6 7 

Number of matings and clutches per individual 



FIG. 7. Distributions of the snails (in %) according to their total number of matings (solid) and clutches 
(crosshatched) in two generations of Algerian snails. 



Changes in Number of Eggs and Young 

In AS populations, the number of eggs of 
the first clutch (N1) was significantly higher 
than the next ones for snails that laid at least 
two clutches (t-test; P < 0.001), and linearly 
related to shell size, with a highly significant 
correlation coefficient, except in the Algerian 
sample in which only seven clutches have 
been considered (Table 3). In addition, the 
nine regression lines compared were signifi- 
cantly different (ANCOVA; P < 0.001). Thus, 
for a given shell size, snails from a population 
with on average larger individuals were in- 
clined to lay larger clutches. 

Differences between samples (without the 
Algerian one) in mean first clutch size and 
mean shell size were also highly significant 
(ANOVA; P < 0.001 ), and there was, as might 
be expected, the same differences between 
samples for the two characters after multiple 
comparisons tests (Table 4): snails from 
southern populations seemed to be larger 
and to lay larger clutches, whereas insular 
snails size was reduced, as was their mean 
clutch size, especially for the sample from St. 
Denis. 

In addition, samples with the lower mean 
clutch size were also those with the lower 
mean number of clutches laid per snail. As AS 
populations did not differ in hatching success 
for "healthy" clutches (Table 1), mean num- 
bers of young produced per snail presented 
the same differences or homogeneities be- 
tween them as those observed for mean num- 
bers of eggs. However, some clutches were 
infected by various parasites, mainly nema- 



todes, to different degrees according to their 
origin (from a maximum of 28.9% in Belmont 
1985 to a minimum of 5.1% in St. Denis 1985 
with, respectively, hatching success of 27.7% 
and 53.9%; no apparent infection in samples 
from St. Denis 1986 and Algeria). 

For the populations studied through four 
generations, individual shell breadth and first 
clutch size were introduced in a two-way 
(generation, origin) ANOVA with replication. 
Each factor and their interaction have highly 
significant effects (P < 0.001 ) and therefore, 
the population classification according to N^ 
led to the following conclusions (Table 5): 

— Significant differences were observed 
only between AS populations. The homoge- 
neity of all the other populations for this char- 
acter was the result of a decrease of the 
mean value in JS, Fl , and F2 samples from 
Toulouse with respect to the AS generation 
and, in contrast, an increase of the mean 
clutch size in successive experimental gener- 
ations from La Réunion. Differences between 
snails from Lorient were not significant, what- 
ever the generation was. 

— The F1 samples from Lorient and Tou- 
louse were characterized by small clutches, 
which could be associated with a relatively 
low number of clutches produced per snail. 
Thus, snails born and reared in the laboratory 
laid clutches with a number of eggs indepen- 
dent of parental origin and between 90 and 
100. 

The mean numbers of eggs deposited per 
AS-JS snail during the season (total fecun- 
dity) showed differences between populations 
in accordance with the preceding compari- 



112 



MADEC & DAGUZAN 



TABLE 3. Relationship between first clutch size N1 (dependent variable) and shell breadth 
10) in Helix aspersa from ten natural populations. P: level of significance of r. 



Origin 


N 


Slope 


Intercept 


r 


P 


Lorient 


40 


0.88 


-164.6 


0.60 


*" 


Surgères 


51 


0.74 


-115.5 


0.44 


** 


Toulouse 


55 


0.84 


-131.2 


0.51 


*** 


Avignon 


56 


0.82 


-141.5 


0.46 


** 


Lyon 


27 


1.30 


-291.2 


0.68 


*** 


Belmont 


60 


0.90 


-179.9 


0.66 


*** 


Bastia 


41 


0.82 


-149.3 


0.60 


*** 


Palma 


44 


0.65 


-95.3 


0.63 


... 


St. Denis 


29 


0.98 


-182.4 


0.74 


*** 


Alger 


7 


0.15 


+ 121.5 


0.10 


NS 



TABLE 4. Classification of natural populations according to shell breadth and first clutch size N1 (S.N.K. 
test; P < 0.05) 



Shell breadth classification 






Clutch size 


classification 






Terms used 


Means 




SNK test 


Terms used 


Means 




SNK test 




Toulouse 


33.5 


A 






Toulouse 


150.2 


A 








Avignon 


33.3 


A 






Avignon 


132.8 


A 








Surgères 


32.4 


A 






Surgères 


124.3 


A 








Belmont 


30.8 




В 




Lorient 


104.9 




В 






Lorient 


30.5 




В 




Belmont 


97.3 




В 


С 




Lyon 


29.4 




В 


С 


Palma 


95.1 




В 


С 


D 


Palma 


29.3 




В 


С 


Lyon 


91.1 




В 


С 


D 


Bastia 


28.4 






С 


Bastia 


83.8 






С 


D 


La Réunion 


26.3 








D La Réunion 


74.1 








D 



TABLE 5. Classification of AS, JS, Fl and F2 samples according to shell breadth and first clutch size 
(S.N.K. test; P < 0.05) 



Shell breadth classification 




Clutch 


size classification 




Terms used 


Means 




SNK test 


Terms used 


Means 




SNK test 


AS-Toulouse 


33.3 


A 






AS-Toulouse 


145.9 


A 






F2-Toulouse 


32.6 


A 


В 




F2-La Réunion 


102.8 




В 




F2-Lorient 


32.0 


A 


В 


С 


Fl -La Réunion 


101.2 




В 




F2-La Réunion 


31.9 


A 


В 


С 


F2-Lorient 


100.5 




В 




JS-Toulouse 


31.5 




В 


С 


JS-Toulouse 


100.3 




В 




Fl -Toulouse 


31.3 




В 


С 


F2-Toulouse 


100.1 




В 




Fl -Lorient 


31.1 




В 


С 


AS-Lorient 


100.1 




В 




Fl -La Réunion 


30.8 




В 


С 


JS-Lorient 


99.7 




В 




JS-Lorient 


30.6 




В 


С 


Fl -Toulouse 


91.2 




В 


С 


AS-Lohent 


30.4 






С 


Fl -Lorient 


90.7 




В 


С 


JS-La Réunion 


27.5 






D 


JS-La Réunion 


80.8 






С D 


AS-La Réunion 


26.4 






D 


AS-La Réunion 


74.1 






D 



sons of clutch size. However, we noticed that 
all JS snails have laid more eggs than the 
corresponding AS populations (Tables 1 , 2). 
In spite of the results relative to F1 genera- 
tions from Lorient and Toulouse, it did appear 
that eggs numbers produced per snail born 



and reared under artificial conditions con- 
verged among the three populations. 

For snails from Algeria, there was no sig- 
nificant relationship between clutch size (N1) 
and shell breadth, but the mean numbers of 
eggs of clutches of both AS, JS and F6 snails. 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 113 

TABLE 6. Reproductive characters, shell size and rлortality of Algerian Helix aspersa frorn three 
generations under artificial conditions (x ± standard error). 



Generation 


AS 


JS 


F6 


Sample size 


35 


10 


20 


Shell breadth (mm) 


44.2 ± 0.3 


42.5 ± 0.1 


44.2 ± 0.2 


Mean rate of matings (%) 


40.1 


100.0 


100.0 


Mean number of matings per individual 


0.6 ± 0.1 


3.2 ± 0.4 


3.4 ± 0.1 


Mean rate of layings (%) 


20.0 


80.0 


95.0 


Mean number of clutches per individual 


0.3 ± 0.1 


3.0 ± 0.6 


3.1 ±0.1 


Mean number of eggs per individual 


42.1 ± 15.3 


443.8 ± 104.9 


608.3 ± 59.1 


Clutch size 


163.7 ±28.5 


160.6 ± 17.6 


186.6 ± 12.0 


Mean weight of eggs (mg) 


39.2 ± 4.3 


41 .6 ±3.8 


41.0 ±3.2 


Hatching success (%) 


94.8 


78.7 


82.4 


Adult mortality (%) 


31.4 


20.0 


20.0 



which seemed to have preserved character- 
istics of snails from the field (shell and clutch 
sizes), were higher than all the others (Table 
6). The only important difference between 
generations (total fecundity) was a conse- 
quence of the number of clutches produced 
per snail and could be attributable to physio- 
logical disturbance of AS snails, as the JS 
results suggested. Because the populations 
did not differ significantly in hatching success, 
the mean number of young produced per Al- 
gerian snail that had laid eggs was by far the 
highest. 

Mortality During the Breeding Season 

There was no significant difference be- 
tween AS populations in total number of dead 
snails during the same breeding period (Table 
1). However, in 1985, the majority of snails 
survived, except in the sample from Algeria; 
on average, only 7.5% of the snails collected 
in 1985 died, versus 25% in 1983 (x^ test; P 
< 0.001). 

The numbers of snails dead in Fl and es- 
pecially F2 generations were comparable 
from one population to the other (Table 2). 
Differences among AS or JS generations 
could be attributable to acclimatization prob- 
lems, especially for the JS sample from St. 
Denis, which had been subjected to an artifi- 
cial hibernation. 



DISCUSSION 

In the present study, snails were reared un- 
der uniform artificial conditions, whatever 
their origin. For AS and JS samples, variation 
in reproductive characters may consequently 



be genetically determined or induced by en- 
vironmental factors prior to the snails' cap- 
ture. This prior conditioning could include 
many factors, such as time of year, duration 
of activity suspension, or the reserves carried 
over winter which are able to contribute to 
modification of fecundity (Brown et al., 1985; 
Baur & Raboud, 1988). Furthermore, varia- 
tion in egg production of Helix aspersa cannot 
be dissociated from shell size, itself depen- 
dent on several proximate factors that act on 
growth rate and age at maturity. One may 
also suspect interactions between genotypes 
and laboratory conditions and differences in 
acclimatization ability, which lead to a change 
of reproductive activity for snails adapted to 
other proximal conditions, in comparison with 
their real potential expressed in the field. For 
example, we can assume that reproductive 
characteristics of snails from La Réunion and 
Algeria, for which spring and summer are not 
(or not necessarily) the breeding season, are 
affected not only by the starting date of the 
experiment, but also by the 16L:8D cycle se- 
lected in the laboratory as an optimal combi- 
nation for reproduction of snails from western 
France (Daguzan, 1981 ; Le Guhennec & Da- 
guzan, 1983). Therefore, total egg production 
of snails from Breton samples duhng the rear- 
ing period is not different from the annual egg 
production of snails of the same populations 
living in the field. However, the length of their 
breeding period and the timing of mating and 
oviposition may be notably shorter, according 
to a variation in proximate factors (Madec & 
Daguzan, 1991). In natural environments, the 
time of year of breeding takes gradually place 
from spring (Brittany) to winter (Algeria), with 
possibly two breeding seasons (spring and 
autumn) or, in contrast, a short and single pe- 
riod in the late spring for mountain popula- 



114 



MADEC & DAGUZAN 



tions (Belmont). Even if the present work 
gives no precise evaluations, it seems that 
seasonal adjustments are partially retained 
under laboratory conditions and may lead, 
when local conditions are very different (late 
autumn or winter reproduction), to important 
disturbances (snails from Algeria). Under cli- 
matic conditions of La Réunion, it is possible 
that reproduction of Helix aspersa occurs 
throughout the year (Fig. 1), and then eggs 
deposited by a snail during this experiment 
would represent just a little part of its annual 
egg production in natural conditions. 

The continuous rearing of three genera- 
tions of snails from four populations with con- 
trasting reproductive characteristics (Lorient, 
Toulouse, St. Denis, Algeria) demonstrates 
that the major proportion of phenotypic varia- 
tion observed in H. a. aspersa (all populations 
except the Algerian one) is environmentally 
induced. Thus, differences between AS sam- 
ples, for the most part, disappear when snails 
are reared for two generations in the same 
environment, whatever the initial degree of 
variation and the characters concerned. The 
phenomenon is already perceptible among in- 
dividuals that in the beginning of their lives 
had very different ecological constraints (JS 
generation). Helix a. aspersa seems to be 
charactehzed by the ability to respond to en- 
vironmental changes with a large range of 
phenotypes, which suggests an important 
plasticity. However, this experiment does not 
allow us to explain the specific differences ob- 
served in AS populations or to give precise 
estimates of the respective effects of environ- 
mental and genetic components. In addition, 
other factors could interfere before the initia- 
tion of reproduction in the laboratory. Thus, 
we have to consider the age of snails when 
reproduction occurs (six-seven months for 
JS, F1 , and F2 individuals; unknown for AS 
snails from La Réunion; at least two years for 
the others). In this regard. Le Calve (1988) 
emphasizes that an older snail has a ten- 
dency to mate more often but seldom to lay. 
Their clutch size is higher and correlated with 
smaller eggs. Young adults (JS) produce 
clutches at a rate higher than that of adults 
from the corresponding AS generation which 
are, on average, older. On the other hand, 
when shell-size effects are removed, clutch 
size of young adults seems to be smaller. 
These results are different from those of 
Wolda (1963) for Cepaea nemoralis but, in 
each case, it seems that a balance finds its 
expression in an egg production per snail for 



one breeding season not very different from 
one age class to the other. 

Snails from Algeria (/-/. a. maxima) seem to 
have developed a specific combination of re- 
productive traits. Egg weight (or size; r^^g = 
0.94), clutch size, and number of eggs pro- 
duced per snail in one season indicate a 
higher reproductive investment for an Alge- 
rian snail and, at the species level, lead to 
surprising relationships as, for example, the 
positive one between egg size and egg num- 
bers. However, we should have weighted 
these values by the size of animals, and in 
addition, results of this experiment should be 
considered with caution because of the small 
size of the samples. Furthermore, we are not 
able to know if the extent of reproductive in- 
vestment affects the survivorship of snails, 
only one breeding season being studied in 
laboratory conditions. Nevertheless, variation 
in these large snails may have a specific ge- 
netic basis and thus, is not a part of the plas- 
ticity that characterizes H. a. aspersa. 

In order to discuss these combinations of 
traits and to compare them with other Hell- 
cidae, we have to integrate the variation in 
reproductive characters in the species' life 
history and in the context of its natural envi- 
ronment. Unfortunately, relevant field data on 
other life-history traits, their genetic compo- 
nents, and local ecological constraints are un- 
available or are imprecise. Nevertheless, the 
two opposite trends, illustrated in the extremes 
by populations from St. Denis (recently intro- 
duced) and Alger (natural distribution area), 
can be useful for the understanding of the life- 
history variability of Helix aspersa. Additional 
data (Chevallier, 1983; Madec, 1988, 1989b) 
are used to specify the identity of the two forms 
in Table 7. 

Differences between these two patterns 
are obviously related to their respective hab- 
itats. Our purpose is then to compare two 
contrasting habitats and possible life-history 
solutions adopted by the species, with the 
help of predictions of theoretical life-history 
models. In this respect, the general demo- 
graphic classification of habitats (Begon et al., 
1987) allows consistent hypotheses about in- 
terpretation of observed patterns by looking 
at the mortality factors affecting infra-popula- 
tions of juveniles and adults. 

At St. Denis de La Réunion, ameliorating 
effects of altitude (900 m., decrease of tem- 
perature) and proximity of the ocean (in- 
crease of humidity) lead to a climatic regime 
favourable to a long growing period (annual 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 1 1 5 
TABLE 7. Summary of life-history traits observed in Helix aspersa from La Reunion and Algeria. 



Population from Algeria 



Population from St. Denis 



•Thicker shells 
•Adult size larger 

• Later maturity 

• Longer length of life 

• More offspring, smaller/parent size 



•Thinner shells 
•Adult size smaller 

• Earlier maturity 
•Shorter length of life 

• Fewer offspring, larger/parent size 



cycle) and an extended breeding season, 
which allows, if necessary (calcium not easily 
available at this basaltic site), egglaying of 
several small clutches per snail. In addition, 
large size of eggs in comparison to shell size 
of adults (Madec, 1988) seems to be obtained 
at the expense of the number per clutch, and, 
if not only a phylogenetic constraint, this 
would have an explanation in a high popula- 
tion density, because favourable climatic con- 
ditions avoid a high mortality of eggs and 
young; juveniles may be advantaged by large 
size because of strong intraspeclfic competi- 
tion. The small size of adults could also be 
related to high population density, which acts 
on growth rate via the mucus secreted for lo- 
comotion, as demonstrated for Helix aspersa 
(Dan & Bailey, 1982; Lucarz & Gomot, 1985) 
and other Helicidae (Oosterhoff, 1977; Cam- 
eron & Carter, 1979). Moreover, snails from 
La Réunion are characterized by their thinner 
shells, perhaps related to the calcium defi- 
ciency and the high rainfall (Goodfriend, 
1986). This low resource (calcium) allocation 
for growth and maintenance, which probably 
does not affect snail survival, would lead to a 
higher (and earlier) egg production. In other 
habitats colonized by H. a. aspersa (western 
Europe, USA), populations exhibit notably dif- 
ferent features (larger adult size, larger 
clutches); this variability could be partially ex- 
plained by high egg and juvenile mortality by 
desiccation, frost and prédation (Potts, 1975; 
Daguzan, 1982), which is also a characteristic 
of numerous other Helicidae in Europe 
(Wolda & Kreulen 1973; Pollard, 1975; Cowie, 
1984). Thus, lower population density (growth 
rate increase) and longer length of growth 
lead to an increase of adult size, conse- 
quently larger clutches, which counterbalance 
higher juvenile mortality (Peake, 1978). On a 
smaller scale. Potts (1972) noticed that two 
neighbouring colonies of Helix aspersa in Cal- 
ifornia (one living on waste ground, another in 
a garden) produce such different demo- 
graphic traits as, in this experiment, popula- 
tions from La Réunion and Surgères, only by 



reason of daily watering. Finally, this first 
trend seems to be the result of a considerable 
flexibility in life-history traits, which allows H. 
a. aspersa to successfully colonize a large 
range of unstable habitats. 

By contrast, snails from Algeria (/-/. a. max- 
ima) have larger shells, which are twice as 
thick as those from La Réunion, obtained af- 
ter a growth period of, at least, three years, 
including long suspensions of activity during 
summer. This greater shell volume allows the 
production of larger clutches with significantly 
larger eggs (Madec, 1 988). The present study 
gives no pertinent information on egg produc- 
tion per breeding season for Algerian AS 
snails because the experiment began when 
they were preparing to aestivate in the field. 
However, data on JS and F6 generations, 
which confirm larger clutch and egg sizes, in- 
dicate that sexually active snails lay on aver- 
age three clutches during the breeding period 
under laboratory conditions, that is to say a 
mean number of eggs per snail between 450 
and 600. Moreover, because these character- 
istics are genetically determined, an allomet- 
ric relationship seems to exist, which leads in 
/-/. a. maxima to a decrease of the proportion 
of shell volume allocated to clutch volume in 
comparison to H. a. aspersa "norms," despite 
their higher mean egg and clutch sizes. With 
reference to the theory, an efficient protection 
against abiotic mortality (and perhaps such 
other factors as predators) represented by a 
larger shell in adults as in juveniles is related 
to other features: delayed maturity, smaller 
reproductive allocation, and investment in a 
large size (protection) leading to an increase 
of residual reproductive value. In this respect, 
H. a. maxima differs from other Mediterra- 
nean Helix, which seem to fit this model bet- 
ter, because of a small clutch size with larger 
eggs {Helix lucorum: Staikou & Lazaridou- 
Dimitriadou, 1988; Helix texta: Heller & Ittiel, 
1990). In addition, our hypothesis remains 
speculative because not only is nothing 
known about residual reproduction but also a 
proportion of the observed variation has no 



116 



MADEC & DAGUZAN 



genetic basis. Thus, the life-cycle length vari- 
ability is essentially environmentally induced, 
because snails from all populations, including 
Algerian ones, reach maturity from three to 
six months after birth under laboratory condi- 
tions (Madec, 1989b). This observation raises 
the problem of the precise localization of nat- 
ural populations of this form, and the neces- 
sity of studying several of them in order to 
define the degree of variation of its life cycle in 
particular ecological conditions. Similarly, 
Heller & Ittiel (1990) show that in unstable 
populations of Helix texta, a low population 
density, caused by a massive prédation of 
adults, allows a very rapid growth of young. 
An other density-dependent mechanism, also 
related to prédation and climate (semi-arid 
environment), pressures on two slopes of a 
wadi, leads to an important variation of fecun- 
dity in nearby populations of Trochoidea 
seeizen/ (Yom-Tov, 1972). 

Finally, a valid comparison with the predic- 
tions of life-history models requires a field 
study on tactics used by Helix asperse to re- 
spond to various selection pressures, i.e. to 
test: (i) the hypothesis based on an adapta- 
tive plasticity in life-cycle traits in H. a. as- 
persa, which lives in "favourable" but often 
human perturbed environments and which 
could explain its widespread geographic and 
ecological distribution; (ii) the hypothesis of a 
specific combination adopted by H. a. max- 
ima as a response to harsh conditions of its 
reduced distribution area. 



ACKNOWLEDGEMENTS 

Thanks are given to Dr. L. M. Cook and two 
anonymous reviewers for helpful comments 
and linguistic revision. 



LITERATURE CITED 

ADAMO, S. A. & R. CHASE, 1988, Courtship and 
copulation in the terrestrial snail Helix aspersa. 
Canadian Journal of Zoology, 66: 1446-1453. 

BAILEY, S. E. R., 1983, The photoperiodic control 
of hibernation and reproduction in the landsnaii 
Helix aspersa Müller. Journal of Molluscan Stud- 
ies, Supp. 12A: 2-5. 

BAUR, B. & С RABOUD, 1988, Life-history of the 
landsnaii Arianta arbustorum along an altitudinal 
gradient. Journal of Animal Ecology, 57: 71-87. 

BEGON, M., J. L. HARPER & С R. TOWNSEND, 
1987, Life-history variation. Pp. 501-538 in: м. 
BEGUN, J. L. HARPER & 0. R. TOWNSEND, eds., Ecol- 



ogy: Individuals, Populations and Communities. 
Blackwell Scientific Publications, Oxford. 

BROWN, K. M., 1985, Intraspecific life-history vari- 
ation in a pond snail: the roles of population di- 
vergence and phenotypic plasticity. Evolution, 
39: 387-395. 

BROWN, K. M., D. R. DEVRIES & B. K. LEATH- 
ERS, 1985, Causes of life history variation in the 
freshwater snail Lymnaea elodes. Malacologia, 
26: 191-200. 

CALOW, P., 1983, Life cycles patterns and evolu- 
tion. Pp. 649-677 in: w. D. russell-hunter, ed., 
The Mollusca, Vol. 6: Ecology. Academic Press, 
London. 

CAMERON, R. A. D. & M. A. CARTER, 1979, Intra 
and inter-specific effects of population density on 
growth and activity in some helicid land snails 
(Gastropoda: Pulmonata). Journal of Animal 
Ecology 48: 173-179. 

CHEVALLIER, H., 1983, Les escargots comesti- 
bles commercialisés en Europe occidentale. Pp. 
21-35 in: j. DAGUZAN, ed.. L'escargot et IHélicicul- 
ture. Informations techniques des services vété- 
rinaires, Paris. 

CLARKE, В., W. ARTHUR, D. T. HORSLEY & D. T. 
PARKIN, 1978, Genetic variation and natural se- 
lection in pulmonate molluscs. Pp. 219-271 in: v. 
FRETTER & J. PEAKE, ods., Pulmonates, Vol. 2A. Ac- 
ademic Press, London. 

COWIE, R. H., 1984, The life-cycle and productivity 
of the land snail Theba pisana (Mollusca: Heli- 
cidae). Journal of Animal Ecology, 53: 31 1-325. 

CROOK, S. J., 1980, Studies on the ecological ge- 
netics of Helix aspersa. Ph. D. dissertation, Uni- 
versity of Dundee, Dundee. 

DAGUZAN, J., 1981, Contribution à l'élevage de 
l'escargot petit-gris Helix aspersa Müller. I-Re- 
production et eclosión des jeunes en bâtiment et 
en conditions thermohygrométriques contrôlées. 
Annales de Zootechnie, 30: 249-272. 

DAGUZAN, J., 1982, Contribution à l'élevage de 
l'escargot petit-gris Helix aspersa Müller. Il-Evo- 
lution de la population juvénile de l'éclosion à 
l'âge de 12 semaines en bâtiment et en condi- 
tions d'élevage contrôlées. Annales de Zootech- 
nie, 31: 87-110. 

DAN, N. & S. E. R. BAILEY, 1982, Growth, mortality 
and feeding rates of the snail Helix aspersa at 
different population densities in the laboratory, 
and the depression of activity of helicid snails by 
other individuals or their mucus. Journal of /Mol- 
luscan Studies, 48: 257-265. 

GOODFRIEND, G. A., 1986, Variation in land snail 
shell form and size and its causes: a review. Sys- 
tematic Zoology, 35: 204-223. 

GUEMENE, D. & J. DAGUZAN, 1982, Variations 
des capacités reproductrices de l'escargot petit- 
gris Helix aspersa Müller selon son origine géo- 
graphique: accouplement et ponte. Annales de 
Zootechnie, 31 : 369-390. 

HELLER, J. & H. ITTIEL, 1990, Natural history and 
population dynamics of the land snail Helix texta 



GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 1 1 7 



in Israël (Pulmonata, Helicidae). Journal of Mol- 
luscan Studies, 56: 189-204. 

LE CALVE, D., 1988, Influence de l'âge sur les 
comportements d'accouplement et de ponte 
chez l'escargot petit-gris Helix aspersa Müller. Mé- 
moire, I.S.P.A., Rennes. 

LE GUHENNEC, M. F. & J. DAGUZAN, 1983, Rôle 
de la lumière sur la reproduction de l'escargot 
petit-gris. Helix aspersa Müller. Comptes Rendus 
de l'Académie des Sciences, Paris, 297(111): 141- 
144. 

LORVELEC, O. & J. DAGUZAN, 1990. Etude, en 
conditions climatiques naturelles, de la variation 
saisonnière de l'activité lomotrice chez l'escargot 
Helix aspersa Müller. Colloque de l'INRA, Régu- 
lation des cycles saisonniers chez les inverté- 
brés. Dourdan, 52: 61-64. 

LUCARZ, A. & L. GOMOT, 1985, Influence de la 
densité de population sur la croissance diamé- 
trale et pondérale de l'escargot Helix aspersa 
Müller dans différentes conditions d'élevage. 
Journal of f^olluscan Studies, 51 : 1 05-1 1 5. 

MADEC, L., 1988, Origine et importance des diffé- 
rences affectant la forme et la taille des oeufs 
chez l'escargot petit-gris Helix aspersa Müller. 
Haliotis, 19: 143-152. 

MADEC, L., 1989a, Variations géographiques de la 
taille et de la forme des coquilles d'Hélix aspersa 
Müller. Evolution de ces caractères au labora- 
toire. Bulletin de la Société Zoologique de 
France, 114:85-100. 

MADEC, L., 1989b, Étude de la différenciation de 
quelques populations géographiquement sépa- 
rées de l'espèce Helix aspersa t\ñüller: aspects 
morptiologiques, écophiysiologiques et bioctii- 
miques. Ph. D. dissertation. University of 
Rennes, Rennes. 

MADEC, L., 1991, Genetic divergence in natural 
populations of the land snail Helix aspersa Müller. 
Journal of Molluscan Studies, 57: 483-487. 

MADEC, L. & J. DAGUZAN, 1991, Variabilité de la 
reproduction examinée au laboratoire entre pop- 
ulations naturelles d'Hélix aspersa Müller de la 
région Bretagne. Reproduction Nutrition and De- 
velopment, 31: 551-559. 

OOSTERHOFF, L. M., 1977, Variation in growth 
rate as an ecological factor in the landsnail Ce- 
paea nemoralis (L). Motherland Journal of Zool- 
ogy, 27: 1-132. 

PAN ELLA, F., 1 982, Effect of one cycle of divergent 
selection for shell length in Helix aspersa Müller. 
Annales de Génétique et de Sélection Animale, 
14(3): 421-426. 



PEAKE, J., 1978, Distribution and ecology of the 
Stylommatophora. Pp. 429-526 in: v. fretter & j. 
PEAKE, eds., Pulmonates, Vol. 2A. Academic 
Press, London. 

POLLARD, е., 1975, Aspects of the ecology of He- 
lix pomatia L. Journal of Animal Ecology, 44: 
305-329. 

POTTS, D. C, 1972, Population ecology of Helix 
aspersa, and the nature of selection in favourable 
and unfavourable environments. Ph. D. disserta- 
tion. University of California, Santa Barbara. 

POTTS, D. C, 1975, Persistence and extinction of 
local populations of the garden snail Helix as- 
persa in unfavourable environments. Oecologia, 
21:313-334. 

SACCHI, С F., 1971, Ecologie comparée des gas- 
téropodes pulmones des dunes Méditerra- 
néennes et Atlantiques. Natura, 62(3): 277-358. 

SCHERRER, В., 1984, Biostatistique, g morin, éd. 
Chicoutimi, 850 pp. 

STAIKOU, A., M. LAZARIDOU-DIMITRIADOU & N. 
FARMAKIS, 1988, Aspects of the life cycle, pop- 
ulation dynamics, growth and secondary produc- 
tion of the edible snail Helix lucorum L. in Greece. 
Journal of Molluscan Studies, 54: 139-155. 

STAT-ITCF, 1988, Manuel d'utilisation, version 4. 
Institut technique des céréales et des fourrages, 
ed., Paris. 268 pp. 

TAYLOR, J. W., 1914, Monograph of the land and 
freshwater Mollusca of the British Isles. Taylor 
Brothers, Leeds. (See pp. 236-273). 

TOMPA, A. S., 1984, Landsnails (Stylommato- 
phora). Pp. 47-131, In: w. D. RussELL-HUNTER, ed.. 
The Mollusca, Vol. 7: Reproduction. Academic 
Press, London. 

WOLDA, H., 1963, Natural populations of the poly- 
morphic landsnail Cepaea nemoralis (L.). Fac- 
tors affecting their size and their genetic consti- 
tution. Archives Néerlandaises de Zoologie, 
15(4): 381-471. 

WOLDA, H. & D. A. KREULEN, 1973, Ecology of 
some experimental populations of the landsnail 
Cepaea nemoralis L. II. Production and survival 
of eggs and juveniles. Netherland Journal of Zo- 
ology 23: 168-188. 

YOM-TOV, v., 1972, Field experiments on the ef- 
fects of population density and slope direction on 
the reproduction of the desert snail Trochoidea 
(Xerocrassa) seetzeni. Journal of Animal Ecol- 
ogy, 41: 17-22. 



Revised Ms accepted 1 7 November 1 992 



MALACOLOGIA, 1993, 35(1): 119-134 

ANATOMY AND FUNCTIONAL MORPHOLOGY OF THE FEEDING 

STRUCTURES OF THE ECTOPARASITIC GASTROPOD 

BOONEA IMPRESSA (PYRAMIDELLIDAE) 

John B. Wise 
Department of Biology, George Washington University, Washington, D.C. 20050, U.S.A. 

ABSTRACT 

The ectoparasitic snail Boonea impressa (Say, 1822) feeds on a variety of invertebrates. In 
tfie laboratory, Boonea impressa parasitized both Crassostrea virginica (Gmelin, 1791) and 
Geul<ensia demissa (Dillwyn, 1817), positioning itself on the edge of the host's shell, thus 
providing access to the host's mantle tissue exposed when the bivalve is open. Feeding struc- 
tures of Boonea Impressa include: (1) an acrembolic or completely invaginable proboscis, (2) a 
buccal sac comprised of sucker, mouth, stylet with separate buccal opening, and stylet bulb, (3) 
a muscular buccal pump, (4) a pair of salivary glands, and (5) a coiled esophagus. These enable 
the snail to feed once the extended proboscis locates the host's soft tissue, which is penetrated 
by the stylet. Subsequently, the muscular action of the buccal pump removes host hemolymph. 
Retraction of the everted proboscis and the muscles involved in this process are examined and 
discussed. Scanning electron microscopy and transmission electron microscopy revealed de- 
tails of the feeding structures (e.g., tufts of cilia apically located on the papillae of the proboscis) 
previously unknown for this genus. When B. impressas feeding structures were compared to 
those of selected European pyramidellids described in the literature, morphological and ultra- 
structural differences became apparent. These differences further support the retention of this 
species in Boonea. 

Key words: Boonea impressa, Pyramidellidae, ectoparasite, feeding structures, histology, 
functional morphology. 



INTRODUCTION 

Boonea impressa (Say, 1822), commonly 
cited as {Odostomia impressa), is an ectopar- 
asite within the large gastropod family Pyra- 
midellidae, which feeds on the body fluids of 
invertebrates (Hopkins, 1956; Wells, 1959; 
Allen, 1958; Robertson & Orr, 1961; Schel- 
tema, 1965; Cheng, 1967; Abbott, 1974; Rob- 
ertson, 1978; Robertson & Mau-Lastovicka, 
1979). It commonly inhabits the littoral and 
sublittoral zones of the western Atlantic from 
New Jersey, USA, to Quintana Roo, Mexico 
(Robertson, 1978). 

Recent studies have examined aspects of 
this ectoparasite's population dynamics, be- 
havior, and its effects on Crassostrea virgin- 
ica (Gmelin, 1791) (White et al., 1984, 1985; 
Ward & Langdon, 1986; Powell et al., 1987a, 
1987b; White et al., 1988a, 1988b). Boonea 
impressa can be deleterious to oysters by re- 
ducing growth, net productivity, and survival 
rates, while also effectively altering valve 
movement and lowering filtration rates (White 
et al., 1984; Ward & Langdon, 1986). In ad- 
dition, White et al. (1 987) have suggested that 
B. impressa may be a vector for the oyster 
pathogen Perkinsus marinus. 



To date, no detailed anatomical studies 
have been conducted on species within the 
genus Boonea (formerly included in Odosto- 
mia Fleming, 1817; Robertson, 1978). Al- 
though White et al. (1985) cursorily examined 
a portion of B. impressas alimentary system 
in a comparison of Texas and North Carolina 
specimens and European pyramidellids, an 
understanding of the structural and functional 
morphology of Boonea impressa is lacking. 
The objectives of this investigation were: (1) 
to describe the morphology and function of 
feeding structures and (2) to compare these 
structures with those of selected European 
pyramidellids described in the literature. 



MATERIALS AND METHODS 

Boonea impressa was collected from the 
Folly River and Inlet Creek oyster reefs near 
Charleston, South Carolina, from 1984 to 
1986. Each collection yielded approximately 
200 snails, which were maintained in an 
aquarium of filtered sea water. 

Snails (3-6 mm shell length) were re- 
moved from their shells with a vise or pliers. 
Snails were dissected under a dissecting mi- 



119 



120 



WISE 



croscope equipped with an ocular microme- 
ter. Photographs were taken with a camera 
mounted on a Nikon Labophot microscope or 
a Zeiss Tessavar. 

Snails were decalcified using a commercial 
agent (Decal) to prepare serial sections of the 
entire snail. In order to section the proboscis 
in its extended condition, snails were relaxed 
in a sea water and Sevin-acetone solution 
(Carriker & Blake, 1959) prior to decalcifica- 
tion. Tissue was fixed in 10% seawater for- 
malin, effectively dehydrated in alcohol, 
cleared in xylene, and embedded in paraffin. 
Sections were cut at 2-5 |xm and stained with 
hematoxylin (Ehrlich acid alum or Gills) and 
with eosin-Y. Photographs were taken with a 
photomicrographic system (model PM-10AK) 
mounted on an Olympus BH2-D0 micro- 
scope. 

Snails for histochemical studies were de- 
calcified prior to fixation in B-4 (consisting of 
0.1% glutaraldehyde, 6% HgCi2. and 1% so- 
dium acetate) for 5 h. Tissue was treated as 
described above. Once sections were cut (3- 
5 ixrn) they were deparaffinized, dezinkahzed 
with Lugol's iodine, hydrated, and placed in a 
solution of HID (high iron diamine) overnight 
(Sheenan & Hrapchak, 1980). They were 
then thoroughly rinsed with distilled water and 
counter-stained with alcian blue (Ph 2.5) for 
30 min. After rinsing, the tissue was dehy- 
drated, cleared in xylene, and mounted. 

Scanning electron microscopy was used to 
examine the gross and ultrastructural mor- 
phology of the alimentary structures. Speci- 
mens were relaxed in Sevin-acetone, re- 
moved from their shells and fixed in 2.5% 
glutaraldehyde, in a sodium cacodylate buffer 
and sea water solution. Following fixation, tis- 
sue was rinsed in cacodylate buffer, effec- 
tively dehydrated in ethanol, critical point 
dried, coated with gold-palladium, and exam- 
ined with a JEOL JSM-35C scanning micro- 
scope operating at 20 kev. 

For transmission electron microscopy, 
snails were treated with Sevin-acetone and 
seawater solution, decalcified, and rinsed 
thoroughly in sea water. Denuded snails were 
fixed for 24 h in a 2.5% glutaraldehyde-ca- 
codylate solution, washed in cacodylate 
buffer and post-fixed in osmium tetroxide 
(Shennan & Hrapchak, 1980). Following os- 
mication, snails were rinsed in distilled water, 
effectively dehydrated in a series of graded 
ethanol, and placed in propylene oxide. Spec- 
imens were transferred to a 1:1 solution of 
propylene oxide and 812 embedding resin 




1.0 mm 



ÏJ4 



FIG. 1 . Boonea impressa at the edge of valve of 
Crassostrea virginica, with proboscis (P) extended, 
feeding suctorially on the bivalve's mantle. 



and agitated overnight with an Adam's nuta- 
tor. Next, specimens were placed in a 2:1 so- 
lution of embedding resin and propylene ox- 
ide for 7 h. Once the snails had been placed 
in pure embedding resin, infiltration by the 
supporting medium was again facilitated by 
agitation for 24 h. The specimens were vac- 
uum infiltrated for 4 h and then placed in a 
mold and oriented. Thin sections were cut 
with a Sorvall M22 ultramicrotome, stained 
with UALC (uranyl acetate and lead citrate), 
and examined with a JEOL 100 Selectron mi- 
croscope. 



DESCRIPTIVE MORPHOLOGY 

The external anatomy of Boonea impressa 
is typical of the Pyramidellidae. This species 
has a well-developed, tentaculate head, a 
pair of eyes located beneath the epithelium 
medial to the tentacles, and a large opercu- 
lated foot tapered posteriorly (Fig. 1). The 
mentum located just ventral to the head ex- 
tends as a shelf over the propodium. A capa- 
cious mantle cavity narrows posteriorly, ex- 
tending to the most anterior position of the 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 



121 



visceral mass. The right anterior portion of the 
mantle edge forms a short canal or siphon. 
Other mantle cavity features characteristic of 
the family include opposing dorsal and ventral 
ciliated strips (responsible for the transport of 
water into and out of the mantle cavity), a 
palliai kidney, a simple apectinate osphra- 
dium, and a pigmented mantle organ (Fig. 
2A). 

The epidermis of the anterior region (tenta- 
cled head, foot, and mantle) is composed of 
one layer of cuboidal or columnar cells (Fig. 
ЗА) that are usually ciliated and have basal 
nuclei. The head-foot and mantle have large 
subepidermal gland cells that are basophilic. 
These cells contain granulated droplets 
(spheroids), which discharge between the 
epidermal cells; no ducts are present. Prelim- 
inary tests utilizing HID/AB (high iron diamine- 
alician blue) show that a majority of these 
cells stain purple-black, indicating the pres- 
ence of sulfated mucins. A few (inside the 
dorsum of the mentum) stain pale blue by ali- 
cian blue, indicating the presence of nonsul- 
fated acidic mucins. The pedal gland lies in a 
medial position just above and parallel to the 
ventral surface of the foot (Fig. ЗА). This 
gland is an invaginated thin layer of ciliated 
epithelial tissue that surrounds a lumen. The 
epithelia are encircled by an aggregate of 
gland cells, staining dark purple by hematox- 
ylin and eosin and also containing sulfated 
mucins. The opening of the pedal gland is 
located midline on the underside of the pos- 
terior portion of the foot. 

The pedal sinus complex traverses the 
length of the lower foot and is comprised of 
numerous sinuses surrounded by nucleated 
connective tissue (Fig. ЗА). The columellar 
muscle, located behind the foot and extend- 
ing posteriorly to the visceral mass, is com- 
posed of smooth muscle. Numerous muscle 
fibers radiate from the columellar muscle into 
the head-foot, including those interspersed 
throughout the gland cells and hemolymph si- 
nuses. 

The cephalic hemocoel is visible without 
dissection once the shell has been removed. 
The hemocoel is bordered by the columellar 
muscle ventrally and by the floor of the mantle 
cavity dorsally (Figs. 2B, ЗА). It terminates 
posteriorly at the visceral mass, and anteriorly 
it extends to just behind the head. The major- 
ity of the alimentary structures are located 
within the cephalic hemocoel. 

When retracted (Fig. 28), the proboscis, re- 
ferred to as the introvert, is completely in- 



verted, and largely within the cephalic hemo- 
coel. This inversion results in the looping of 
the introvert into three consecutive upright 
u's. The introvert extends posteriorly from its 
opening or aperture, passes through the 
nerve ring, and joins the buccal sac (com- 
prised of sucker, mouth, stylet with separate 
buccal opening and stylet bulb) located well 
within the cephalic hemocoel (Fig. 28). The 
temporary lumen created by this inversion is 
mainly bordered by the papillae of the probos- 
cis. Beneath the papillae and extending the 
length of the proboscis is a layer containing 
both circular and longitudinal muscles (Fig. 
38, C). A basal lamina extends between the 
papillae and this layer of muscle, which ap- 
pears mesh-like in light microscopy. Internal 
to this is a layer of connective tissue border- 
ing the lumen, which is present when the pro- 
boscis is protracted (Fig. 38; see Fig. 2C for 
the position of the proboscis and other feed- 
ing structures when the proboscis is extend- 
ing). It is through this connective tissue that 
secondary retractor muscles of varying length 
pass to insert at points along the proboscis 
(Fig. 3C). 

The everted proboscis appears rough and 
pustuloso, with the greatest concentration of 
papillae anterior to the tips of the tentacles 
(Figs. 20, 4A). The proximal portion of the 
proboscis within the boundaries of the tenta- 
cles, although tuberculate with scattered clus- 
ters of cilia, is non-papillate (Fig. 4A). The pa- 
pillae are flattened and compressed when 
first everted from the temporary lumen; how- 
ever, once in position on the external surface 
of the protracted proboscis, these papillae be- 
come tumescent (Fig. 48). Cilia extend from 
the center of each papilla as apical tufts. Each 
papilla is composed of several elongate cells 
containing organelles and darkly colored 
secretory granules, the number of which var- 
ies among papillae. Each papilla contains a 
central cell from which the cilia (possessing a 
9 + 2 microtubule arrangement) originate (Fig. 
4C, D). The papillae are bordered apically by 
fusiform microvilli covered by a glycocalyx. 

The introvert joins the buccal sac at two 
locations. Just outside the sucker, the papil- 
lae are replaced by simple cuboidal cells that 
attach directly to the sucker (Fig. 5A). These 
have numerous cilia, presumably of a tactile 
nature, that extend well into the temporary lu- 
men. Beneath the cells are the aforemen- 
tioned layers of muscle and connective tissue 
extending posteriorly to insert at the base of 
the sucker beside the primary retractor mus- 



122 



DCS 




1.5mm 




FIG 2 A Generalized representation of pallia! complex. Mantle skirt cut on left side and reflected to the 
right В Scherлatic of Boonea impressa in the non-feeding posture, with proboscis retracted. Mantle re- 
moved and cephalic hemocoel opened to expose alimentary structures in "natural Pps'^'O"- ,Г!^/^/^®Р^ 
of salivary glands. Salivary glands shown upright to reveal location to right of buccal pump II. C. Schematic 
of partially protracted proboscis, with buccal pump I uncoiling as it is pulled forward. Note new position of th^ 
buccal sac, now lying just anterior to head. A = anus; Bpl = buccal pump I; Bpll = buccal pump II; BS - 
buccal sac; DCS = dorsal ciliated strip; H = heart; К = kidney; MO = mouth; P = proboscis; E - 
esophagus- PMO = pigment mantle organ; SGL = salivary gland; VCS = ventral ciliated stnp. 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 



123 



T 
GLC 




-\Ш?"^см^^^" 




500 pm ^ ^- 



PSC 





м^*„>'-г^^^^.^ 



FIG. 3. A. Section through head-foot, mantle, and cephalic hemocoel. B. Transmission electron micropho- 
tograph of internal proboscis morphology. Note lamina between papillae, layer of circular and longitudinal 
muscle and thin layer of connective tissue beneath muscle layers. С Longitudinal section of inverted 
proboscis. BL = basal lamina; CH = cephalic hemocoel; CM = columella muscle; CRM = circular muscle; 
CT = connective tissue; F = foot; GLC = gland cell(s); LM = longitudinal muscle; MA = mantle; M = 
muscle; PA = papilla(e); PG = pedal gland; L = temporary lumen; PSC = pedal sinus complex; SRM = 
secondary retractor muscle(s); T = tentacle; VM = visceral mass. 



cle. The prinnary retractor nnuscle, the base of 
which is attached to the columellar muscle, 
extends into the cephalic hemocoel to insert 
on ether side of the sucker (Fig. 5A). 

The buccal sac has tv^^o major components: 
the stylet bulb and the buccal sucker (Fig. 5B, 



C). The stylet bulb, extending posteriorly, 
curves dorsally to lie beneath the most ante- 
rior portion of the buccal pump. Within the 
posterior portion of the stylet bulb is a cres- 
cent-shaped lumen, surrounded by the mus- 
cles of the stylet bulb (Fig. 5A). The stylet 



124 



WISE 




V Iß , 



^ 



4pm 



FIG. 4. A. Scanning electron microphotograph of partially extended proboscis. В. Inmú papillae on external 
surface of the proboscis, each with apical tuft of cilia. С Transmission electron microphotograph of individual 
papillae; each papilla comprised of several elongate cells, delineated by distinct cell membranes. D. Central 
cell from which papillary cilia originate. Cilia possess a 9 + 2 microtubule arrangement. С = cilia; CC = 
central cell; СЕМ = cell membrane; MV = microvilli; N = nucleus; P = proboscis; PA = papilla; T = 
tentacle. 



bulb's shape varies from round to oblong. The 
globe-shaped buccal sucker is comprised of a 
thick muscular wall comprised of numerous 
columnar cells arranged in a stack-like man- 
ner that surrounds the elevated inner labium 
(Fig. 5A). Within, the sucker the labium ap- 
pears smooth and corpulent. The center of 
the labium contains an aperture through 
which the stylet emerges. Dorsal to this open- 
ing is the true mouth, located at the junction 
between the inside sucker wall and the base 
of the labium (Fig. 5A, C). The oral tube ex- 
tends posteriorly from this opening, to join the 
buccal pump at the buccal pump-buccal sac 
junction. The oral tube is bordered ventrally 
by simple cuboidal cells and lined dorsally by 
a thin layer of flattened epithelium (Fig. 5A). 
The stylet, which lies within a cavity behind 



the sucker, is surrounded by a cuticular 
sheath. This cuticular sheath opens anteriorly 
to extend as a hood over the stylet's apex 
(Fig. 5B, D). The sheath, indented ventro-me- 
dially, has a prominent longitudinal dorsal 
ridge (Fig. 5D). The stylet is broad at its base 
and tapers distally, with the apex emerging 
through the opening in the sheath. Dorsally, 
the surface of the stylet, distal to its base, is 
notched by a series of parallel grooves that 
terminate prior to its apex. The medial inden- 
tation is bordered on either side by uneven, 
laterally grooved ridges (Fig. 5E). Retractor 
muscles within the base of the stylet insert at 
the buccal sac wall (Fig. 5A). The two salivary 
ducts, after entering the buccal sac from the 
buccal pump, unite to form a common duct, 
which enters the lower portion of the stylet 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 



125 




ГС ^ 








FIG. 5. A. Histological section through buccal sac. B. Scanning electron microphotograph of buccal sac; 
portion of buccal sac surrounding stylet and cuticular enclosure removed. Stylet bulb intact. С Globe-shaped 
sucker; within sucker is true mouth and stylet aperture. D. Scanning microphotograph of anterior part of 
cuticular sheath enclosing stylet (note prominent ridge). E. Cross-sectional view of stylet. Bpl = buccal 
pump I; С = cilia; CS = cuticular sheath; L = lumen; LA = labium; MO = mouth; ОТ = oral tube; PRM 
= primary retractor muscle; R = ridge; RM = retractor muscle; S = stylet; SA = stylet aperture; SB = 
stylet bulb; SCC = simple cuboidal cell; SD = salivary duct; SGL = salivary gland. 



126 



WISE 



and continues internally along its length (Fig. 
5A). 

The buccal pump can be divided into two 
distinct regions (Fig. 2B); the anterior portion 
of the buccal pump (termed buccal pump I) is 
an elongate cylindrical structure that pos- 
sesses an outer covering of very thin epithe- 
lium enclosing a layer of circular muscle (Fig. 
6A, B). Internal to this layer is a mathx of cells 
and muscle fibers that extends to the triangu- 
lar lumen. A large part of this organ is com- 
posed of tightly packed elongate muscle cells 
(Fig. 6C), which radiate outward from the lu- 
men to lie adjacent to the layer of circular 
muscle encircling this structure. Distinct 
bands of muscle fibers, anchored within a 
layer of connective tissue internal to the cu- 
ticular layer lining the lumen, pass between 
the muscle cells to insert just beneath the ex- 
ternal epithelium. Buccal pump I increases in 
diameter along the last quarter of its length 
prior to uniting with the remainder of the buc- 
cal pump. The large posterior portion of the 
buccal pump (termed buccal pump II) curves 
downward and then bends anteriorly, allowing 
accommodation within the confines of the 
cephalic hemocoel (Fig. 2B). This portion of 
the buccal pump (with the exception of its 
central lumen) is composed almost solely of 
muscle tissue (Fig. 6A). This segment of the 
buccal pump, elliptical in cross section, is cov- 
ered by a thin layer of furrowed epithelium, 
not unlike that covering buccal pump I (Fig. 
6D). Buccal pump II is similar to the buccal 
pump I in wall composition, but lacks buccal 
ducts and has a greater overall diameter and 
larger elliptical lumen. It is composed prima- 
rily of muscle fibers that radiate from the lu- 
men and extend to a layer of circular muscle 
located just underneath the peripheral layer 
of epithelium of the pump. The same kind of 
myofilament bands present in buccal pump I 
intermittently traverse the width of buccal 
pump II to anchor within a cuticularized layer 
lining the lumen (Fig. 6E). At the junction of 
buccal pump I and buccal II is a ring of mus- 
cle. 

The esophagus originates at a point below 
and just posterior to where the buccal pump is 
divided into two distinct sections (Fig. 2C). 
Elongate cilia are present at the junction of 
the buccal pump II and esophagus. This sec- 
tion of the esophagus coils repeatedly as it 
extends downward and then posteriorly to join 
the stomach, located within the visceral mass. 
The esophagus is very irregular and uneven 
along its length, surrounded by a thin layer of 



epithelium and muscle (Fig. 7A). The lining of 
the central lumen has numerous folds cov- 
ered with uniformly distributed cilia (Fig. 7B). 
Connecting the salivary glands to the ali- 
mentary canal are the salivary gland ducts 
(Fig. 6A, B). The ducts enter the ventral side 
of the buccal pump I, just anterior of the buc- 
cal pump Ixbuccal pump II juncture, and ex- 
tend the length of this section of the alimen- 
tary canal. The salivary ducts are comprised 
of a lumen encircled by multiple layers of cir- 
cular and longitudinal muscle. Epithelial tis- 
sue lining these ducts can occlude the lumen 
(Fig. 7C). The salivary glands lie together on 
the right side of buccal pump II within the 
cephalic hemocoel and are composed of vari- 
ably sized cells located along a central canali- 
culus, which extends to the vesicle-like struc- 
ture distally (Fig. 7A). The cells are tightly 
packed with a fine granular substance. The 
glands show differential staining along their 
lengths. This varies among individual snails, 
with no discernable pattern. The vesicle-like 
structure at the distal portion of the buccal 
gland is apparently a lumen lined with epithe- 
lium that extends the length of the gland to 
line the canaliculus. No cilia project from the 
epithelium lining the lumen of this distal por- 
tion, although the lining of the canaliculus is 
ciliated. Scanning electron microscopy con- 
firmed the presence of numerous secretory 
granules within the gland (Fig. 7D). With the 
exception of the striated outer surface, the cil- 
iated canaliculus, and the distal sac-like por- 
tion of this structure, this organ is composed 
solely of acinar secretory packets. 



DISCUSSION 

Anatomical studies of Boonea impressa 
shows that its external anatomy is very similar 
to the European pyramidellid species de- 
scribed by Fretter & Graham (1949), Maas 
(1965), and Ankel (1949) (Table 1 lists the 
taxa they examined). There are, however, 
both configurational and ultrastructural differ- 
ences, particularly concerning feeding struc- 
tures. These are discussed below, as is the 
generic assignment of Boonea impressa. 

Large gland cells that stain differentially by 
hematoxylin and eosin lie beneath the epithe- 
lial layer in B. impressa, and are scattered 
throughout the head-foot and mantle. These 
cells produce and release granulated spheres 
that transude the intercellular matrix, migrate 
between the epithelial cells, and eventually 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 



127 




i- -. 



FIG 6 Feeding structures. A. Cross sections of buccal pump I lying to one side of larger buccal риглр II. В. 
Scanning electron microphotograph of buccal pump I in cross section. Etched outer covering encloses 
internal layer of muscle fibers extending to the lumen. С Transmission electron microphotograph of buccal 
pump I in oblique section; numerous cells radiate from lumen to lie adjacent to layer of circular muscle 
encircling esophagus. D. Scanning electron microphotograph of buccal pump II covered by a thin layer of 
epithelium, comprised of myofibrils. E. Transmission electron microphotograph of buccal pump II in cross 
section Note circular muscle, longitudinal muscle, and muscle perpendicular to the organ's axis. Bpl - 
buccal pump I; Bpll = buccal pump 11; CR!^ = circular muscle; L = lumen; M = muscle; SD = salivary 
ducts. 



128 



WISE 




FIG. 7. A. Histological section of the esophagus and a single salivary gland. B. Scanning electron micro- 
photograph of interior of the esophagus. С Transmission electron microphotograph of single salivary duct 
in transverse section. Salivary duct enclosed by multiple layers of circular and longitudinal muscle. D. 
Scanning electron microphotograph of a cross-section of a salivary gland, composed of innumerable secre- 
tory granules with the exception of striated outer surface, ciliated lumen, and distal sac-like portion. С = cilia; 
CRM = circular muscle; EP = epithelium; LM = longitudinal muscle; E = esophagus; SG = secretory 
granules; SD = salivary duct; SGL = salivary gland. 



coat the ciliated exterior. No ducts lead from 
these gland cells to the external surface of the 
gastropod. This is contrary to observations of 
Fretter & Graham (1949), who found that in 
European pyramidellids they examined, the 
large gland cells of the head-foot had well- 
defined ducts, with non-mucoidal products. 
For B. impressa, a majority of these cells pos- 
sessed sulfated mucins (a major constituent 
of mucus), whereas a small number, located 
just Inside the dorsal surface of the mentum, 
contained nonsulfated acidic mucins. There- 
fore, these ductless cells function in the pro- 
duction of the mucus that coats the external 
surface of the mantle and head. 

The pedal gland of Boonea impressa con- 
tains sulfated mucins. Based on the arrange- 
ment of the gland cells, the presence of cili- 



ated epithelium, and its position within the 
foot, this structure is similar to the lateral 
streak or aggregate of cells, located on either 
side of the foot and dorsal to the sole, de- 
scribed by Fretter & Graham (1949) in Odos- 
tomia unidentata and other species they ex- 
amined (Table 1). On the basis of bundles of 
long cilia, associated with the lateral streak, 
these authors thought that it might function as 
a sensory organ. I did not observe the bun- 
dles of cilia in Boonea impressa, and my find- 
ings suggest that these same cells comprise 
the pedal gland in B. impressa (Fig. 2A). In B. 
impressa, the pedal gland is responsible for 
the formation of the suspensory thread with 
which this snail fastens itself to its surround- 
ings. An attachment thread has also been ob- 
served in other pyramidellids (Ponder, 1973; 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 



129 



Hoffman, 1979; J. E. Ward, 1985, pers. 
comm.). 



FEEDING STRUCTURES AND THEIR 
FUNCTIONAL MORPHOLOGY 

The feeding structures of Boonea impressa 
enable this gastropod to feed suctorially on a 
number of hosts. The proboscis is capable of 
extending to a length equal to or greater than 
the snail's shell, enabling it to reach its host's 
soft tissues. The stylet perforates the host's 
tissue, presumably once the muscular sucker 
is firmly attached to the host. The forward 
movement of the stylet is accomplished by 
the compression of the stylet bulb's crescent- 
shaped lumen. Retractor muscles ensure the 
return of the stylet to its original position 
within the stylet cavity (Figs. 5A). The dorsal 
surface of the stylet possesses a combination 
of grooves and ridges enabling the stylet to 
penetrate the host's tissue readily (Fig. 5E). 
The opening of the true mouth, through which 
host hemolymph and perhaps torn tissue 
fragments enter the alimentary canal, is con- 
nected to buccal pump I by the oral tube (Fig. 
5A, C). Contractions of only buccal pump II 
draw host hemolymph into the alimentary ca- 
nal. Located at the junction of buccal pump I 
and buccal pump II is a ring of muscle that 
closes this passageway when contracted, 
thereby forcing host hemolymph into the 
esophagus once the lumen of buccal pump II 
is compressed. Elongate cilia, present at the 
junction of the buccal pump II and esophagus, 
facilitate movement. Cilia within the esopha- 
gus (Fig. 7B), in conjunction with possible 
peristaltic movement, convey host hemo- 
lymph to the stomach. 

Movement of the proboscis involves a com- 
plex series of events. Protraction of the pro- 
boscis is presumably hydraulic, a conse- 
quence of the compression of the cephalic 
hemocoel and the redistribution of he- 
molymph. Retraction of the proboscis is ac- 
complished by the contraction of specific 
muscles. The most obvious of these, and pos- 
sibly the most important, is the primary retrac- 
tor muscle. Figure 8A shows the muscle's po- 
sition when the proboscis is retracted; 
however, once the proboscis is extended 
(Fig. SB), this muscle is brought forward as 
the mouth moves to its most anterior position 
at the tip of the completely protracted probos- 
cis. Contraction of the primary retractor mus- 
cle initiates the often rapid invagination of the 



proboscis. In concurrence with the contrac- 
tion of the primary retractor muscle, the sec- 
ondary retractor muscles contract sequen- 
tially, starting with those at the most anterior 
portion of the extended proboscis. The sec- 
ondary retractor muscle arrangement in the 
right anterior portion of the snail is shown 
(simplified) in Figure ВС. Only three of the 
approximately 24 secondary retractor mus- 
cles are illustrated. The axis or pivot point for 
the secondary retractor muscles is located in 
the head just behind the eye. From this point, 
two of the muscles extend anteriorly into the 
proboscis, and the third muscle extends pos- 
teriorly to attach to a portion of the proboscis 
that is still within the cephalic hemocoel. If the 
proboscis were fully protracted, the most pos- 
terior secondary retractor muscle would even- 
tually lie anterior to the other two secondary 
retractor muscles. If, however, the proboscis 
is retracted, the most anterior secondary re- 
tractor muscle would contract, resulting in the 
inversion of the most anterior portion of the 
proboscis. 



SYSTEMATIC CONCLUSIONS 

In the process of resolving some of this 
family's taxonomic problems, Robertson 
(1978) excluded three Western Atlantic Amer- 
ican pyramidellids from the genus Odostomia 
Fleming, 1813, where they were originally as- 
signed and proposed a new genus, Boonea, 
to accommodate them. His actions were 
based on differences (e.g., in protoconch 
shape, operculum configuration, excurrent si- 
phon, penial complex, pigmented mantle or- 
gan coloration, and in the location of the com- 
mon gonoduct opening) between these 
species and European species once consid- 
ered congeneric. As additional substantiation 
of Robertson's decision, this study compared 
the feeding structures of B. impressa to liter- 
ature accounts of the feeding structures of the 
European odostomians (including the type 
species of Odostomia, Odostomia plicata) de- 
scribed by Ankel (1949), Fretter & Graham 
(1949), and Maas (1965). 

The feeding structures of Boonea impressa 
follow the general anatomical scheme de- 
scribed for other odostomians, with some im- 
portant exceptions. Structurally, the proboscis 
of B. impressa is unlike those of the odosto- 
mian species described by Fretter & Graham 
(1949) and Maas (1965) (Table 1). The Euro- 
pean species examined by Fretter & Graham 



130 




1.0mm 





FIG. 8. Retractor гли8с1е8 of the proboscis. A. Scherлatic representation of primary retractor when proboscis 
completely inverted. Primary retractor muscle originating at columella muscle, extends into cephalic hemo- 
coel to pass through the sheath of the proboscis and insert on either side of buccal sucker's base. B. Primary 
retractor when proboscis partially protracted. Primary retractor muscle carried forward during extension, 
lying posterior to proboscial tip. С Schematic of secondary retractor muscle arrangement (right lateral view 
of head region). Only three of approximately 24 secondary retractor muscles illustrated. BS = buccal sac; 
CG = cerebral ganglion; CM = columella muscle; P = proboscis; PRM = primary retractor muscle; SRM 
= secondary retractor muscle. 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 
DGC 



131 




X600 



NGLC 




У SMGLC 




250pm 



FIG. 9. European odostomians. A. Longitudinal section of proboscis of Odostomia unidentata. Papillae 
consist of three to four cells. Extending from subepithelial cells located within the layer beneath papillae are 
ducts that pass through center of papillae to open apicaliy (redrawn from Fretter & Graham, 1949). B. 
Schematic of internal proboscial arrangement of Odostomia eulimoides. Papillae comprised of large-celled 
epithelium; note large gland cell duct extending between papillae to open externally (redrawn from Maas, 
1965). С. Schematic of feeding structures of O. eulimoides (redrawn from fHaas, 1965). BS = buccal sac; 
BPI = buccal pump I; BPII = buccal pump II; CRM = circular muscle; DGC = gland cell duct; ED = 
excretory duct; EPT = papilla (tran. sec); ES = esophagus; GLC = gland cell(s); LGLC = large gland cell; 
LM = longitudinal muscle; N = nucleus; NGLC = nucleus of gland cell; P = proboscis; PA = papilla; RM 
= retractor muscle; SD = salivary duct; SGL = salivary gland; SMGLC = small gland cell. 



132 



WISE 



TABLE 1 Morphological and ultrastructural differences between feeding structures of Boonea impressa 
and those of selected European odostomians (listed below). 



This Study 



*Maas; Ankel 



"Fretter & Graham 



(1) Proboscis: 

(a) papillae composed of 
numerous, elongate cells. 

(b) beneath papillae a layer of 
circular & longitudinal muscle 
enclosed by connective 
tissue. 

(c) no gland cells or ducts. 



(2) Buccal pump: 

divided into two regions, with 
Bpl twice the length of Bpll. 

(3) Salivary ducts: 

enter buccal sac & then 
stylet bulb, without exiting 
alimentary canal. 



(a) papillae composed of large- 
celled epithelium. 

(b) internal to papillae a layer of 
circular muscle, above a 
layer of longitudinal muscle. 

(c) beneath the layers of muscle 
an aggregate of large & 
small gland cells, the larger 
with ducts that terminate 
between the papillae 
externally. 

divisible into approximately 
equal length sections. 

exit alimentary canal just 
behind buccal sac and then 
enter stylet bulb. 



(a) papillae composed of only 3 
or 4 cells. 

(b) beneath papillae a of layer 
gland cells with ducts that 
extend to exterior via the 
center of the papillae. 

(c) internal to the glandular 
layer, a layer of longitudinal 
muscle. 



not divisible, uniform. 



exit alimentary canal 
posterior to buccal sac & 
then enter stylet bulb. 



'Examined in detail Odostomia eulimoides, O. plicata, and Liostomia clavula, with cursory attention given to Odostomia 
rissoides, Chirysallida spiralis, and C. obtusa. 

"Examined in detail Odostomia unidentata, O. plicata, and O. lukisii, with some attention given to O. scaiaris, (= O. 
rissoides), O. trífida, and Chrysallida spiralis. 



(1949) (e.g., Odostomia unidentata), pos- 
sessed papillae comprised of only three to 
four cells, containing large basal nuclei, side 
by side v^^ithin the neck of the papillae, and 
arranged so that they formed a narrow base, 
widened medially and then tapered to a blunt 
apex (Fig. 9A). Present within each papilla 
(along its longitudinal axis) was a duct that 
extended from a subepithelial gland cell lo- 
cated within the connective tissue of the wall 
of the proboscis. Fretter & Graham (1949) 
also determined that beneath the layer of 
gland cells, and underneath the epithelium of 
the buccal region, was an array of muscle fi- 
bers that comprised part of the mechanism for 
the retraction of the proboscis. Maas (1965) 
investigated several other odostomian spe- 
cies (e.g., Odostomia eulimoides; Table 1) 
and found the papillae of the proboscis to be 
comprised of large-celled epithelium (Fig. 
9B). Internal to the papillae is a layer of cir- 
cular muscle that lies above longitudinally ori- 
ented band of muscle. Beneath the muscle, a 
glandular layer contains a mixture of small 
and large (30jxm) gland cells. According to 
Maas (1965), the larger gland cells have 
ducts that pass through the layer of muscle, 
terminating between the papillae. 
Boonea impressa differs from the de- 



scribed European snails in several ways, the 
most noteworthy being in the histology of the 
proboscis (Table 1). Papilla are each com- 
posed of numerous elongate cells bordered 
internally by a layer of both circular and lon- 
gitudinal muscle (Figs. 38, C, 4B-D). This 
layer of muscle is enclosed by a thin layer of 
connective tissue. No gland cells or ducts are 
present within the papillae or the proboscis. 
Each papilla has a central cell from which cilia 
protrude as an apical tuft. Only a single spe- 
cies, Liostomia clavula, examined by Maas 
(1965) possessed papillary cilia. 

All the European odostomian species in- 
vestigated by Ankel (1949) and Maas (1965) 
have two well-developed buccal pumps that 
are delineated in part by a narrowing at their 
junction (Fig. 9C). Fretter & Graham (1949) 
examined some of the same species but did 
not consider the buccal pump as two separate 
entities: they treated the structure as a single 
pump and stated that it was histologically uni- 
form along its length (Table 1). Maas (1965) 
disagreed with Fretter & Graham's (1949) de- 
scription of the buccal pump, although Maas 
did not examine O. lukisii, one of the species 
Fretter & Graham (1949) used as an exam- 
ple. According to Mass (1965), O. pilicata (a 
species examined by Fretter & Graham), has 



FEEDING STRUCTURE MORPHOLOGY OF BOONEA 



133 



two buccal pumps (Bp I and Bp II respec- 
tively) that are histologically discrete (Fig. 
90). The first buccal pump (like the buccal 
pump I of B. impressa) has a trifid lumen (di- 
vided into three lobes with narrow sinuses), 
and the second buccal pump is flattened lat- 
erally, not dorso-ventrally as described by 
Fretter & Graham (1949). My investigations 
indicate that B. impressa possesses a buccal 
pump divided into anterior and posterior re- 
gions, similar to that for the species described 
by Maas (1965). However, the Bp! of Boonea 
impressa is very elongate and twice the 
length of the Bpll, whereas in all examined 
European odostomians, the buccal pump is 
divided into approximately equal sections 
(Fig. 90). The Bp! of Boonea impressa is a 
well-developed structure comprised chiefly of 
muscle cells that surround a triangularly 
shaped lumen (Fig. 6A, 60). This is contrary 
to White et al. (1985), who determined that 
this portion of the feeding apparatus of B. im- 
pressa was a poorly developed tube. 

The other major difference between B. im- 
pressa and the European pyramidellids is 
the way in which the salivary ducts traverse 
the alimentary canal and enter the buccal 
sac (Table 1). Both Fretter & Graham (1949) 
and Maas (1965) described the salivary 
ducts as entering the first buccal pump just 
anterior to the junction between the two buc- 
cal pumps (Fig. 90). Prior to entering the 
buccal sac, they exit the buccal pump (i.e., 
the alimentary tract) to then enter the stylet 
bulb. My study demonstrates that the salivary 
ducts of B. impressa pass into the ventral sur- 
face of the Bpl, traverse the length of this por- 
tion of the buccal pump, and eventually ex- 
tend into the buccal sac. However, at no point 
do the salivary ducts leave the buccal pump, 
and within the buccal sac they unite to form a 
single duct that enters the base of the hollow 
stylet and extends to the stylet's apex. These 
differences provide further evidence that Rob- 
ertson (1978) was correct in excluding B. im- 
pressa and other eastern American "odosto- 
mians" from the genus Odostomia. 



AOKNOWLEDGEMENTS 

This paper is part of a master's thesis com- 
pleted at the Oollege of Charleston and is 
contribution no. 105 from the Ghee Marine 
Biological Laboratory, Oollege of Charleston, 
Charleston, South Carolina. I am indebted to 
the members of my committee, Charles 



Biernbaum, Robert T. Dillon Jr., William A. 
Roumillat, and Samuel Spicer, for their time, 
energy, and expertise. I thank Karen Swan- 
son and Richard Houbrick for assistance with 
the illustrations and Bob and Jan Ashcraft for 
their help with the SEM and ТЕМ procedures. 
Richard Houbrick, Jerry Harasewych, and 
Robert Hershler critically reviewed the first 
draft of this manuscript and offered many 
helpful suggestions for its betterment. I am 
especially thankful to Marianne and my par- 
ents for all their support. Winston Ponder and 
one unidentified reviewer contributed com- 
ments that were most useful in improving this 
paper. 

LITERATURE CITED 

ABBOTT, R. T., 1974, American sea shells. D. Van 
Nostrand Co., Inc., New York. 541 pp. 

ALLEN, F. J., 1958, Feeding habits of two species 
of Odostomia. The Nautilus, 72: 11-15. 

ANKEL, W. E., 1949, Die Nahrungsaufnahme der 
Pyramidelliden. Vertiandlungen Deutsche Zoolo- 
gische Gesselschaft (Kiel). 1949: 478-484. 

CARRIKER, M. R. & J. W. BLAKE, 1959, A method 
for full relaxation of muricids. The Nautilus, 73(1 ): 
16-21. 

CHENG, T. C, 1967, Marine molluscs as hosts for 
symbiosis with a review of known parasites of 
commercially important species. Pp. 276-285 In 
F. S. Russell, ed., Advances in Marine Biology, 
Academic Press, London and New York. 

FRETTER, V. & A. GRAHAM, 1949, The structure 
and mode of life of the Pyramidellidae, parasitic 
opisthobranchs. Journal of the Marine Biological 
Association of the United Kingdom, 28: 493-532. 

HOFFMAN, D. L., 1979, An attachment structure in 
an epiparasitic gastropod. The Veliger, 22: 75- 
77. 

HOPKINS, S. H., 1956, Odostomia impressa, par- 
asitizing southern oysters. Science, 124: 628- 
629. 

MAAS, D., 1965, Anatomische und Histologische 
Untersuchungen am Mundapparat der Pyra- 
midelliden. Zeitschrift fuer Morphologie Oeko- 
logic der Tiere, 54: 566-642. 

PONDER, W., 1973, Pseudoskenella depressa 
Gen. Et sp. NOV., An ectoparasite on Galeolaria. 
Malacological Review, 6: 119-123. 

POWELL, E., M. WHITE, E. WILSON & S. RAY, 
1 987a, Small-scale spatial distribution of a pyra- 
midellid snail ectoparasite Boonea impressa, in 
relation to its host Crassostrea virginica on oyster 
reefs. Marine Ecology, 8: 107-130. 

POWELL, E., M. WHITE, E. WILSON & S. RAY, 
1 987b, Change in host preference with age in the 
ectoparasitic pyramidellid snail Boonea impressa 
(Say). Journal of Molluscan Studies, 53: 285- 
286. 

ROBERTSON, R., 1978, Spermatophores of six 



134 



WISE 



eastern North American pyramidellid gastropods 
and their systematic significance (with the genus 
Boonea). Biological Bulletin, 155: 360-382. 

ROBERTSON, R. & V. ORR, 1961 , Review of pyra- 
midellid hosts, with notes on an Odostomia par- 
asitic on a chiton. The Nautilus, 74: 85-91 . 

ROBERTSON, R. & T. MAU-LASTOVICKA, 1979, 
The ectoparasitism of Boonea and Fargoa (Gas- 
tropoda: Pyramidellidae). Biological Bulletin, 157: 
320-333. 

SCHELTEMA, A. H., 1965, Two gastropod hosts of 
the pyramidellid gastropod Odostomia bisutura- 
lis. The Nautilus, 79: 7-10. 

SHEENAN, D. С & В. В. HRAPCHAK, 1980, The- 
ory and practice of histotechnology. С V. Mosby 
Company. 481 pp. 

WARD, J. E., 1985, Univ. of Delaware, Lewes, Del- 
aware (data obtained by personal communica- 
tion). 

WARD, J. E. & С LANGDON, 1986, Effects of the 
ectoparasitic Boonea (= Odostomia) impressa 
(Say) (Gastropoda: Pyramidellidae) on the 
growth rate, filtration rate and valve movements 
of the host Crassostrea virginica (Gmelin). Jour- 
nal of Experimental Marine Biology and Ecology, 
99: 163-180. 

WELLS, H. W., 1959, Notes on Odostomia im- 
pressa (Say). The Nautilus, 72(4): 140-144. 

WHITE, M. E., E. N. POWELL & С L KITTING, 



1984, The ectoparasite gastropod Boonea 
(= Odostomia) impressa: population ecology 
and the influences of parasitism on oyster growth 
rates. Marine Ecology, 5(3): 283-299. 

WHITE, M. E., С L KITTING & E. N. POWELL, 

1985, Aspects of reproduction, larval develop- 
ment, and morphometries in the pyramdellid Boo- 
nea impressa (= Odostomia impressa) (Gas- 
tropoda; Opisthobranchia). The Veliger, 28(1): 
37-51 . 

WHITE, f^. E., E. N. POWELL, S. M. RAY & E. A. 
WILSON, 1987, Host-to-host transmission of 
Perkinsus marinus in oyster (Crassostrea virgin- 
ica) populations by the ectoparasitic snail Boo- 
nea impressa (Pyramidellidae). Journal of Shell- 
fish Research, 6(1): 1-5. 

WHITE, M. E., E. N. POWELL, S. M. RAY, E. A. 
WILSON & С E. ZASTROW, 1988a, Metabolic 
changes induced in oysters {Crassostrea virgin- 
ica) by the parasitism of Boonea impressa (Gas- 
tropoda: Pyramidellidae). Comparative Biochem- 
istry Physiology, 90A(2): 279-290. 

WHITE, M. E., E. N. POWELL & S. M. RAY, 1988b, 
Effect of parasitism by the pyramidellid gastropod 
Boonea impressa on the net productivity of oys- 
ters (Crassostrea virginica). Estuahne, Coastal, 
and Shelf Science, 25: 359-377. 

Revised MS accepted 22 November 1992 



MALACOLOGIA, 1993, 35(1): 135-140 

INFLUENCIA AMBIENTAL SOBRE EL CRECIMIENTO ALOMÉTRICO DE LA 

VALVA EN NACELLA (PATINIGERA) DEAURATA (GMELIN, 1791) 

DEL CANAL BEAGLE, ARGENTINA 

Elba Morriconi y Jorge Calvo 

Centro Austral de Investigaciones Científicas (CONICET), C. C. 92 (9410), Ushuaia, 

Tierra del Fuego, Argentina 

ABSTRACT 

Environmental influence on shell allometric growth in Nacelle (Patinigera) deaurata (Gmelin, 
1791) from the Beagle Channel, Argentina. 

The allometric relationships among a variety of shell characters were studied in P. deaurata, 
which inhabits the lower intertidal zone in Beagle Channel. Shell height and weight as well as 
inner volume were significantly higher in specimens living on coasts exposed to strong wave 
action. It is suggested that individuals inhabiting exposed surfaces are obliged to have a stronger 
grip, and consequently the mantle does not extend past the edge, resulting in shell height 
increase. The variations observed are related to the different exposures to wave action. Des- 
iccation is not an important factor in the habitat of this species. 

Key words: morphology, allometry, environmental influence, intertidal zone, limpets, Nacella, 
Prosobranchia. Palabras clave: morfología, alometn'a, influencia ambiental, intermareal, lapas, 
Nacella, prosobranquios. 



INTRODUCCIÓN 



MATERIAL Y MÉTODOS 



Los gasterópodos presentan valvas que 
varían su morfometría general y las propor- 
ciones entre los distintos parámetros estruc- 
turales de la valva en relación con las varia- 
ciones del ambiente, generando alometras 
en el crecimiento. Los factores ambientales 
que producirían cambios más marcados so- 
bre la morfometría valvar serían el oleaje o 
corrientes intensas y la exposición a la dese- 
cación (Balaparameswara Rao & Ganapati, 
1971; Vermeij, 1973, 1980; Branch, 1975; 
Bannister, 1975; Branch & Marsh, 1978; Low- 
ell, 1984; Simpson, 1985). Debido a que las 
lapas se encuentran en habitats muy varia- 
dos, desde el intertidal superior al inferior, en 
zonas expuestas y protegidas, resultan un 
adecuado material para analizar las influen- 
cias ambientales en la morfología valvar. 

Nacella (Patinigera) deaurata (Gmelin, 
1791) habita el intertidal inferior quedando 
expuesta a la desecación solamente en las 
mareas de sicigia. Por ello las variaciones 
morfológicas que presenta pueden correla- 
cionarse fundamentalmente con el grado de 
exposición al oleaje. El propósito de esta in- 
vestigación fue comparar los diferentes pa- 
rámetros estructurales en Nacella (P.) deau- 
rata, colectada en dos localidades con 
diferente grado de exposición. 



Los muéstreos se realizaron en dos loca- 
lidades (Fig. 1): (a) Punta Occidental (PC) 
(54°50'S., 68°20'W.: área expuesta a los 
vientos dominantes del SO, fuerte oleaje, de- 
clive suave, con abundancia de coralináceas 
incrustantes como Pseudolitophillum sp. y 
Synartrophitum sp. (Mendoza, 1988) y ejem- 
plares aislados de Macrocystis pirifera. (b) 
Bahía Lapataia (BL) (54°52'S., 68°35'W.): 
costa orientada hacia el norte, protegida de 
los vientos dominantes, con fuerte pendiente 
y denso cinturón de Macrocystis pirifera. Las 
lapas fueron extraídas por buceo autónomo. 
Se separaron las partes blandas de las val- 
vas, las que se secaron al aire durante varios 
días hasta que el peso no varió. 

Las características de las valvas que se 
consideraron fueron las siguientes (Flg. 2): 
Largo Total (LT), desde el extremo anterior al 
posterior, altura total (AT), desde el apex per- 
pendicularmente a la base, ancho (A), diá- 
metro máximo tomado perpendicularmente a 
LT, perímetro (P) y área basal (AB). Estas 
medidas fueron tomadas al milímetro inferior 
con un calibre vernier. Además se determina- 
ron el peso de la valva (PV) con una precisión 
de 0.01 gramo y el volumen interno (VI). Este 
fue obtenido llenando las valvas con arena 
fina tamizada a 600 mieras determinándose el 



135 



136 



MORRICONI & CALVO 




FIG. 1 : Ubicación de las localidades de muestreo. A: Punta Occidental (54°50'S, 68°20'W) y Bahía Lapataia 
(54°52'S, 68°35'W) área sombreada. В: Punta Occidental, zona expuesta (área sombreada). La flecha 
señala los vientos dominantes. C: Bahía Lapataia, zona protegida (área sombreada). La flecha señala los 
vientos dominantes. 



peso de la misma. Luego se pesó 1 cm^ de 
arena, calculándose el volumen correspon- 
diente a cada valva. La transformación peso 
de arena a volumen se realizó promediando el 
peso de diez réplicas de 1 cm^ de arena. 

Las valvas utilizadas fueron seleccionadas 
empleando números al azar de la colección 
total de valvas (PO: 662 ejemplares; BL: 628 
ejemplares). Las valvas dañadas o con epi- 
biontes fueron descartadas del muestreo. 



El rango de LT considerado comprendió 
valvas de 13 a 65 mm estableciéndose clases 
de 5 mm. En una primera selección se 
tomaron diez valvas para cada clase en am- 
bas localidades; posteriormente y a los efec- 
tos de disminuir la dispersión de las variables 
dependientes se aumentó a 20 por clase el 
número de valvas de las clases mayores de 
36 mm. Las variables (AT, PV y VI) fueron 
tomadas como dependientes del LT, caicu- 



CRECIMIENTO ALOMETRICO VALVAR EN N. (P.) DEAURATA 



137 



Pta. Occidental 



Lapataia 





AT 



•LT- 



fem 



FIG. 2: Vista lateral de las valvas de Nacella (Patinigera) deaurata provenientes de ambas localidades de 
muestreo. 



TABLA 1 : Regresión de AT, PV y VI sobre LT para zonas expuestas (PO) y protegidas (BL). 



Localidad 



Y = a + b*X 



(A)PO 


AT = -5.32 + (0.52 * LT) 


0.96 


< 0.001 


161 


(B)BL 


AT = -2.52 + (0.37 * LT) 


0.96 


< 0.001 


167 


Localidad 


IgY = a + (b*lgX) 


r 


P 


N 


(C) PO 


lg PV = -5.43 + (3.68* Ig LT) 


0.98 


< 0.001 


161 


(D)BL 


lg PV = -4.68 + (3.11 * Ig LT) 


0.98 


< 0.001 


167 


(E)PO 


lg VI = -5.04 + (3.58 * Ig LT) 


0.99 


< 0.001 


161 


(F)BL 


1g VI = -4.86 + (3.40 * Ig LT) 


0.98 


< 0.001 


167 



TABLA 2: LT/AT. 
ecuación (A) y Ьз 


Test de homogeneidad de las pendientes (Hq : b, = 
la pendiente de la ecuación (В) de la Tabla 1 . 


bg) siendo bi la pendiente de la 


Fuente de 
variación 


Suma de Grados de Cuadrado 
cuadrados libertad Medio 


F 


P 


Localidad 

LT 
Localidad * LT 

Error 


64.552 1 64.552 
12599.998 1 12599.998 
354.555 1 354.555 
923.549 324 2.85 


22.646 

4420.336 

124.385 


1Л 1Л 1Л 
o o o 
o Ö Ö 

o o o 
o o o 



lándose las ecuaciones de regresión corre- 
spondientes. Cuando fue necesario se realizó 
la transformación logarítmica de los datos a 
fin de ajusfarlos a la ecuación de la recta. 

RESULTADOS 

Se analizó la relación entre LT y los dife- 
rentes parámetros estructurales, calculán- 
dose las ecuaciones de regresión correspon- 



dientes por el método de los cuadrados 
mínimos. Las relaciones A/LT, P/LT y AB/LT 
no presentan diferencias significativas entre 
las pendientes de las rectas de regresión co- 
rrespondientes a cada localidad de muestreo. 

Relación LT— AT 

La relación LT — AT se ajusta a una recta 
en las dos zonas de muestreo consideradas 



138 



MORRICONI & CALVO 




FIG. 3: Rectas de regresión entre AT/LT, PV/LT y 

VI/LT para Punta Occidental ( ) y Bahía La- 

pataia ( ). 



(Tabla 1). La comparación entre las pen- 
dientes de las rectas de regresión de ambas 
localidades muestra diferencias significativas 



(Tabla 2). A igual LT las valvas de Punta Oc- 
cidental son más altas que las de Lapataia 
(Fig. 3). 

Relación LT— PV 

Esta relación se ajusta a una curva poten- 
cial tanto en Punta Occidental como en La- 
pataia por lo que se realizó la transformación 
logarítmica de la misma (Tabla 1). La com- 
paración entre las dos rectas resultantes 
muestra que las pendientes son diferentes 
(Tabla 3) siendo mayor el PV en Punta Occi- 
dental, para las LT consideradas (Fig. 3). 

Relación LT— VI 

Se ajusta de igual manera a una curva po- 
tencial en las dos localidades, por lo que se 
hizo la transformación logarítmica correspon- 
diente (Tabla 1), comparándose las dos rec- 
tas; éstas muestran pendientes significativa- 
mente diferentes (Tabla 4). Se observa que el 
VI es mayor para cada clase de LT en Punta 
Occidental (Fig. 3). 

DISCUSIÓN 

El análisis de las posibles influencias am- 
bientales sobre la morfología valvar se ha 
intentado en repetidas oportunidades, con re- 
sultados a veces contradictorios, especial- 
mente por la dificultad para analizar por se- 
parado la influencia de la turbulencia del 
agua y de la exposición a la desecación. 

La relación entre la resistencia ofrecida a 
las corrientes de agua y la forma de la valva 
de diferentes especies de lapas fue analizada 
expehmentalmente por Denny (1989). Este 
sostiene que la influencia de la forma de la 
valva en relación a la resistencia ofrecida a 
las corrientes no es tan crítica para la sobre- 
vida y por lo tanto es de un restringido valor 
adaptative. Orton (1932) sugiere que la 
acción de las olas sobre la altura de las val- 
vas de las lapas tendría un efecto insignifi- 
cante sobre la forma de las mismas en P. 
vulgata. Tampoco Balaparameswara Rao y 
Ganapati (1 971 ) hallan diferencia de altura en 
Cellana radiata que habita costas desprote- 
gidas con respecto a la población que vive en 
zonas protegidas. 

Por el contrario, Ebling et al. (1962) en Pa- 
tella aspersa encontraron lapas con valvas 
cuya altura aumentaba significativamente en 
las poblaciones que vivían permanentemente 
sumergidas y sometidas a fuertes corrientes. 



CRECIMIENTO ALOMETRICO VALVAR EN N. (P.) DEAURATA 139 

TABLA 3: LT/PV. Test de homogeneidad de las pendientes (Hq : b, = Ьг) siendo b, la pendiente de la 
ecuación (C) y Ьг la pendiente de la ecuación (D) de la Tabla 1 . 



Fuente de 
variación 


Suma de Grados de 
cuadrados libertad 


Cuadrado 
Medio 


F 


P 


Localidad 

IgLT 
Localidad *lgLT 

Error 


0.51 1 

108.086 1 

0.742 1 

2.948 324 


0.51 
108.086 
0.742 
0.009 


56.098 

11879.686 

81 .584 


<0.000 
<0.000 
<0.000 


TABLA 4: LT/VI. 
ecuación (E) y bj 


Test de homogeneidad de las pendientes (Hq : b, = 
, la pendiente de la ecuación (F) de la Tabla 1 . 


Ьг) siendo bi la pendiente de la 


Fuente de 
variación 


Suma de Grados de 
cuadrados libertad 


Cuadrado 
medio 


F 


P 


Localidad 

IgLT 
Localidad *1gLT 

Error 


0.027 1 

114.137 1 

0.072 1 

1 .424 324 


0.027 

114.137 

0.072 

0.004 


6.046 

25969.286 

16.357 


<0.014 
<0.000 
<0.000 



Walker (1972) en Patinigera polaris y Simp- 
son (1985) en Nacella macquarensis relacio- 
nan la intensidad alométhca del incremento 
de la altura de la valva respecto de la longitud 
con la mayor turbulencia del agua. En Ce- 
llana radiata provenientes de diferentes nive- 
les maréales, Balaparameswara Rao y Gana- 
pati (1971) concluyen que presentan mayor 
altura los individuos que están sujetos a 
mayor desecación. 

Vermeij (1973, 1978) halla que en varias 
especies de lapas la altura de la valva es 
mayor en las que habitan los niveles super- 
iores de la costa, sugiriendo que una valva 
más alta incrementaría la capacidad de 
reserva de agua y la resistencia a la deseca- 
ción. Coincidentemente, Bannister (1975) 
prueba experimentalmente que P. lusitanica, 
que habita en la zona superior del intertidal, 
resiste mejor la desecación que P. caerulea, 
que vive en la zona inferior del mismo; la 
mayor resistencia es vinculada al incremento 
de altura de la valva, que determina un mayor 
volumen interno. 

Las poblaciones de N. (P.) deaurata 
investigadas habitan el intertidal inferior y el 
subtidal somero, por lo que la desecación no 
influiría en la altura de las valvas como ocurre 
en otras especies. En esta especie, compa- 
rando lapas de igual LT provenientes de zo- 
nas expuestas (PO) y protegidas (BL) se 
comprueba una AT significativamente mayor 
para las primeras (Tabla 2, Fig. 3). 

Balaparameswara Rao y Ganapati (1971) 
comparan С radiata que vive en el intertidal 
superior e inferior y en zonas expuestas 



y protegidas. Estos autores encuentran 
que son más pesadas las valvas de las 
que habitan el intertldal superior, pero no ha- 
llan diferencias en zonas con distinta exposi- 
ción. 

En N. (P.) deaurata se produce un incre- 
mento del peso de la valva con el aumento de 
LT, expresándose esta relación en una curva 
potencial (Tabla 1). De la comparación entre 
poblaciones de zonas expuestas y protegidas 
se desprende una diferencia significativa, 
siendo las primeras más pesadas (Tabla 3, 
Fig. 3) 

Baxter (1983) no encuentra diferencias en 
la relación volumen-longitud en P. vulgata ha- 
bitando sitios con poca y mucha exposición al 
oleaje. 

Las valvas de N. (P) deaurata presentan, 
para una misma longitud, mayor volumen in- 
terno en las zonas expuestas (Punta Occi- 
dental) que en las protegidas (Bahía Lapa- 
taia) siendo las diferencias significativas 
(Tabla 4, Fig. 3). No se encontraron diferen- 
cias significativas entre el A, P y AB de la 
valva, en lapas de igual LT provenientes de 
ambas zonas de muestreo. Al no diferen- 
ciarse los parámetros mencionados se evi- 
dencia que el mayor volumen que presentan 
las lapas provenientes de Punta Occidental 
se debe a la mayor altura de las valvas. 

Kopp (1980) relaciona la mayor exposición 
a la desecación durante la baja marea en el 
mejillón Mytilus californianus con individuos 
que presentan valvas más anchas y pesadas. 
Una alometría similar, generando valvas más 
altas y pesadas en las lapas que están ex- 



140 



MORRICONI & CALVO 



puestas a cierto tipo de stress (desecación, 
exposición al oleaje) es encontrada por Orton 
(1932). Este autor argumenta que los estímu- 
los para mantener la valva fuertemente ad- 
herida al sustrato ocasionan la retracción del 
borde del manto. De esa manera disminuiría 
el crecimiento periférico y por lo tanto aumen- 
taría el crecimiento en altura de la valva. 
Kopp (1980) establece una relación análoga 
entre la forma de la valva y la extensión o 
retracción del borde del manto, apoyándose 
en pruebas experimentales. 

Se considera que un proceso similar daría 
lugar a un mayor engrosamiento de la valva 
que conduciría a un aumento de su peso. El 
incremento en altura sin cambio en la super- 
ficie o perímetro de la base aumentaría el vo- 
lumen interno. 



AGRADECIMIENTOS 

Los autores desean expresar su agradeci- 
miento a Gustavo Suarez y Regina Silva por 
su colaboración en la medición de las valvas, 
a Lucas Ramos por su ayuda en el procesa- 
miento de los datos, a Pedro Medina y Rafael 
Pastorino por su participación en la recolec- 
ción de las muestras, y a Miguel Barbagallo 
por la confección de los dibujos y gráficos. 
Esta investigación es parte del Proyecto de 
Investigación y Desarrollo (PID № 266): Bio- 
logía reproductiva de moluscos y equinoideos 
del Canal Beagle. Implicancias ecológicas y 
fisiológicas, financiado por el Consejo Nacio- 
nal de Investigaciones Científicas y Técnicas, 
Argentina. 



LITERATURA CITADA 

BALAPARAMESWARA RAO, B. y P. N. GANA- 
PATI, 1971, Ecological studies on a tropical lim- 
pet, Cellana radiata. Structural variations in the 
shell in relation to distribution. Marine Biology, 
10:236-243. 

BANNISTER, J. v., 1975, Shell parameters in rela- 
tion to zonation in Mediterranean limpets. Marine 
Biology, 31 : 63-67. 

BAXTER, J. M., 1983, Allomethc relationships of 
Patella vulgata L. Shell characters at three adja- 
cent sites at Sandwick Bay in Orkney. Journal of 
Natural History, 17: 743-755. 



BRANCH, G. M., 1975, Ecology of Patella species 
from the Cape Peninsula, South Africa. IV. De- 
siccation. Marine Biology, 32: 179-188. 

BRANCH, G. M. y A. С MARSH, 1978, Tenacity 
and shell shape in six Patella species: adaptive 
features. Journal of Experimental Marine Biology 
& Ecology 34: 111-130. 

DENNY, M., 1989, A limpet shell shape that re- 
duces drag: laboratory demonstration of a hydro- 
dynamic mechanism and an exploration of its ef- 
fectiveness in nature. Canadian Journal of 
Zoology 67:2098-2106. 

EBLING, F. J., J. A. SLOANE, J. A. KITCHING & H. 
M. DAVIES, 1962, The ecology of Lough Ine XII. 
The distribution and characteristics of Patella 
species. Journal of Animal Ecology, 31 :457-470. 

KOPP, J. C, 1980, Growth and the intertidal gra- 
dient in the sea mussel Mytilus californianus 
Conrad, 1837. The Veliger, 22: 51-56. 

LOWELL, R. В., 1984, Desiccation of intertidal lim- 
pets: effects of shell size, fit to substratum and 
shape. Journal of Experimental Marine Biology & 
Ecology 77:197-207. 

MENDOZA, M. L., 1988, Consideracines biológicas 
y biogeográficas de las Corallinaceae (Rho- 
dophyta) de las costas de la Isla Grande de Tie- 
rra del Fuego. Gayana Botánica, 45:163-171. 

ORTON, J. H., 1932, Studies on the relation be- 
tween organism and environment. Proceedings 
of Liverpool Biology Society, 46:1-16. 

SIMPSON, R. D., 1985, Relationship between allo- 
methc growth, with respect to shell height, and 
habitats for two patellid limpets, Nacella (Patini- 
géra) macquariensis Finlay, 1927, and Cellana 
tramoserica (Holten, 1802). The Veliger, 28:18- 
27. 

VERMEIJ, G. J., 1973, Morphological patterns in 
high-intertidal gastropods: adaptative strategies 
and their limitations. Marine Biology, 20:319- 
346. 

VERMEIJ, G. J., 1978, Biogeography and adapta- 
tions: patterns of marine life. Harvard University 
Press: Cambridge, Mass. 352 pp. 

VERMEIJ, G. J., 1980, Gastropod shell growth rate, 
allometry, and adult size: environmental implica- 
tions. Pp. 379-394 in D. с RHOADS & R. A. LUTZ, 

eds.. Skeletal growtti of aquatic organisms, Ple- 
num Press, New York. 
WALKER, A. J. M., 1972, Introduction to the ecol- 
ogy of the Antarctic limpet Patinigera polaris 
(Hombron and Jacquinot) at Signy Island, South 
Orckney Islands. Britisti Antarctic Survey Bulle- 
tin, 28:49-69. 



Revised Ms. accepted 1 7 December 1 992 



MALACOLOGIA, 1993, 35(1): 141-151 

A NEW DEEP-WATER HYDROTHERMAL SPECIES OF NUCULANA 
(BIVALVIA: PROTOBRANCHIA) FROM THE GUAYMAS BASIN 

J. A. Allen 

University Marine Biological Station, Millport, Isle of Cumbrae, Scotland, KA28 OEG\ United 
Kingdom, and Woods Hole Océanographie Institution, Massachusetts, 02543, U.S.A. 

ABSTRACT 

A new deep-water species of Nuculana is described that occurs in the southern trough of the 
Guaymas Basin and is associated with a hydrothermal vent system. The species, N. grasslei, is 
characterized by a large, ornamented prodissoconch, but in other respects it differs little in its 
gross morphology from other species of Nuculana. Such specializations that do occur relate to 
the hostile sulphurous environment in which it lives. Particularly important in this regard is the 
thickened periostracum and the large volume of pigmented blood. 

Keywords: Nuculana, Protobranchia, hydrothermal vents. 



INTRODUCTION 

This paper describes the gross morphology 
of a new species of Nuculana taken from the 
southern trough of the Guaymas Basin in the 
Gulf of California at a depth of 2000 m, adja- 
cent to a position where hydrothermal fluid at 
between 270-31 4°C percolates through a 
thick layer of pelagic sediment and through 
chimneys (Lonsdale et al., 1980; Simoneit & 
Lonsdale, 1982; Grassle et al., 1985; Berg & 
Van Dover, 1987). 

Juvenile and adult specimens were taken 
during a series of dives by DSRV Alvin in Jan- 
uary 1982 and August 1985 (listed in Jones, 
1985, and Berg & Van Dover, 1987). In the 
Guaymas Basin, there are black smokers, 
and the sediments from the study area smell 
strongly of hydrogen sulphide. On this sedi- 
ment, large patches of the filamentous bacte- 
rium Beggiatoa are present. The soft sedi- 
ment benthic communities comprise a few 
species in great numbers, but their composi- 
tion varies over short distances (Grassle et 
al., 1985). Samples of plankton containing lar- 
vae of the Nuculana were taken within the 5 m 
of water column above the sea bed (Berg & 
Van Dover, 1987). The methods employed to 
collect the specimens are reported by Grassle 
et al. (1985) and Berg & Van Dover (1987). 

I am very grateful to Dr. J. Frederick 
Grassle for allowing me to examine this ma- 
terial, to Dr. Cindy Lee Van Dover for permis- 
sion to copy from SEM photographs of larvae, 



and to the director and staff of the Woods 
Hole Océanographie Institution for their help 
over many years. 

DESCRIPTION 

Genus Nuculana Link 1 807 
Type species (OD): 

Arca rostrata Brugière, 1 789, 

ex Chemnitz MS, = Arca pernula 

Müller, 1779. 

Shell robust, moderately and posteriorly 
elongate; rostrum truncate, usually bicahnate, 
moderately compressed, strong concentric 
sculpture; umbo anterior; posterior ventral 
margin slightly sinuate; occasionally with ra- 
dial ribs; escutcheon present; hinge teeth 
chevron-shaped; ligament external with cen- 
tral internal part. 

Nuculana grasslei, new species 

Type locality: Guaymas Basin, south 
trough, 27°03'N, 111°23'W, 2003 m. 



Holotype: USNM 
No. 859482 

Paratypes: USNM 
No. 859481 



1 specimen 

specimens selected 
by J. A. A. from the type 
locality. 



Named in honour of Dr. J. F. Grassle, friend 
and colleague of many deep-sea voyages 
and participant in the Guaymas Expedition. 



^Address for correspondence. 



141 



142 
Material 



ALLEN 



Dive No. 



Depth (m) 



Specimens 
Examined 



(Number 
Collected) 



Position 



Equipment Date 



Alvin 1 1 68 2003 

Alvin1169 1998 

Alvin 1170 2019 

Alvin 1174 2011 

Alvin 1175 1997 

Alvin 1176 2022 

Alvin 1607 2012 

Alvin 1 608 2002 

Alvin 1614 2004 

Alvin 1628 2000 

(1-5 above bottom) 
Alvin 1629 2000 

(3-4 above bottom) 
ВС— Box Core 
TC— Tube Core 
SS — Scoop Sample 
PT — Plankton Tow 



25 
3 
8 



(50) 

(3) 
(16) 

(7) 

(1) 

(1) 
(152) 

(4) 

(1) 

(2) 
(5 postlarva) 



27°03'N, 111°23'W 



27°03'N, 
27°01'N, 
27°01'N, 
27°03'N, 
27°01'N, 
27°05'N, 
27°07'N, 
27°07'N, 
27°00'N, 



111°25'W 

111°25'W 

1ir24'W 

111°23'W 

111°25'W 

111°24.5'W 

111°24.4'W 

111°24.4'W 

111°24.5'W 



— (1 postlarva) 27°00'N, 111°25.5'W 

(225 cm^ area sampled) 

(35 cm^ area sampled) 

(63 mm mesh bag over metal frame) non-quantitative 

(0.4 m^, 183 jjL, mesh) non-quantitative 



SS 
TC 
ВС 
ВС 
ВС 
ВС 
ТС 
ТС 
ТС 
ВС 
РТ 

РТ 



10-1-82 

11-1-82 
12-1-82 
17-1-82 
18-1-82 
19-1-82 
29-7-85 
31-7-85 
6-8-85 
23-8-85 

23-8-85 



Samples reported in Grassle et al. (1985) and Berg & Van Dover (1987). 



Shell Description (Figs. 1-4) 

Shell elongate, stout, bluntly rostrate, equi- 
valve — although central portion of ventral mar- 
gin of right valve may slightly overlap left valve 
as a consequence of strong concentric orna- 
mentation; broad concentric ridges extend 
over central region of shell from faint postehor 
radial ridge to close to anterior margin, those 
close to umbonal region less conspicuous 
than those ventral to them; fine, closely 
spaced concentric striae extend anterior and 
posterior to ridges, with line of ridges marked 
by heavier striae; two faint radial ridges extend 
from umbo to posterior ventral margin; umbo 
anterior (position at approximately 38% total 
length), relatively large, beaks inturned; an- 
tero-dorsal margin smoothly curved near 
umbo, but in large specimens somewhat flat- 
tened anteriorly; postero-dorsal margin more 
or less straight or even slightly concave in 
large specimens, angulate at point opposite 
posterior limit of hinge plate; posterior margin 
broadly truncate and slightly gaping; ventral 
margin for most part an even, shallow curve, 
except posteriorly between limits of radial 
ridges, where it is sinuate (this corresponds to 
position of feeding aperture); escutcheon and 
lunule outlined by faint ridges; hinge plate 
moderately broad, continuous ventral to 



umbo; hinge teeth chevron-shaped, number 
increasing with increasing shell length, 17 an- 
terior and 25 posterior teeth in specimen 26.3 
mm total length, of these 6 or 7 on each side 
of umbo are more leaf-like than those more 
posterior, 1 1 anterior and 1 5 posterior in spec- 
imen 13.7 mm total length; ligament predom- 
inantly opisthodetic, small internal part at- 
tached to resilium, which occupies a dorsal 
position on hinge plate and separates anterior 
and posterior hinge tooth series; external part 
comprises small portion anterior to umbo and 
moderately elongate portion posterior to 
umbo, latter somewhat extended by fused 
periostracum; periostracum golden-yellow, 
much thickened and strongly held within perio- 
stracal groove. 

Prodissoconch large, 275-283 ^im total 
length, ornamented with 9-10 reticulated 
concentric ridges and 10-11 radial reticula- 
tions. 

Length of largest shell examined: 26.3 mm. 

Internal Morphology 

The gross morphology of the body organs 
is typically nuculanid in form (Fig. 5) and dif- 
fers little from descriptions of shallow-water 
species (Yonge, 1939). 



NEW DEEP-WATER HYDROTHERMAL SPECIES 



143 




FIG. 1 . Nuculana grasslei. Lateral view of the shell of the holotype from the left side and an internal view of 
the hinge region of the right valve of a specimen of similar size (bar = 1 mm). 



The mantle is relatively unspecialized. 
Three typical folds are present at the mantle 
margin. Antero-ventrally the middle sensory 
fold is somewhat enlarged to form a simple 
anterior sense organ. Posteriorly there is a 
shallow siphonal embayment enclosing com- 
bined inhalent and exhalent siphons. The in- 
halent siphon is unfused both dorsally and 
ventrally (Fig. 6). Nevertheless, the integrity 
of the siphonal lumena is maintained by the 
apposition of thickened central and ventral 
longitudinal ridges on the inner siphonal sur- 
face. The inhalent siphon is somewhat 
shorter than the exhalent. There is no sipho- 
nal tentacle present, as is the case in other 
species of Nuculana (e.g. Yonge, 1939); how- 
ever, a small lobe is present at the posterior 
limit of the left and right inner mantle folds 
where they meet the ventral margins of the 



mantle embayment. These are not homolo- 
gous to the protobranch tentacle and proba- 
bly represent the termination of the main re- 
jection tract of the mantle that is present on 
the inner surface of the inner muscular mantle 
fold. Their function presumably is to guide 
pseudofaeces to the inhalent siphon so they 
may be ejected on contraction of the shell 
valves. There is a simple feeding aperture im- 
mediately anterior to the siphonal embay- 
ment. Here the middle sensory and the inner 
muscular lobes of the mantle are widened 
and somewhat folded. The feeding aperture 
of N. grasslei is much simpler than that of 
many deep-sea nuculanid protobranchs 
(Allen & Hannah, 1989). Numerous fine radial 
muscles are present within the mantle to the 
inside of the marginal folds. The adductor 
muscles are relatively small and unequal in 



144 



ALLEN 





FIG. 2. Nuculana grasslei. Lateral views of shells from the right side to show variation in shape with 
increasing shell size. The figure includes a dorsal view of the hinge region of the next but largest shell 
illustrated and enlarged internal and external views of valves of a juvenile shell (bars = 1 mm). 



size. The posterior muscle is oval in cross 
section, v\/ith "quick" and "catch" portions of 
equal size. The anterior muscle is crescent- 
shaped, with a narrow elongate "catch" por- 
tion running the length of the anterior face. 

The gills are well developed and extend 
horizontally and parallel to the postero-dorsal 
shell margin from the mid-visceral region to 
the siphonal embayment. In the largest spec- 



imen examined, there are approximately 1 50 
broad gill plates on each demibranch. These 
are comparable to those described by Yonge 
(1939). The plates of each demibranch alter- 
nate in their attachment to the axis. Each axis 
extends posteriorly beyond the posterior plate 
as an extremely long, fine filament. Unlike the 
condition in other nuculanid protobranchs, 
these do not appear to be attached to the 



NEW DEEP-WATER HYDROTHERMAL SPECIES 

5^ 



145 




FIG. 3. Nuculana grasslei. Drawing from SEM photographs of the lateral external surface of the left valve and 
the internal surface of the right valve of a planktonic postlarva (with kind permission of Dr. С L. Van Dover) 
(bar = 0.1 mm). 




FIG. 4. Nuculana grasslei. Dorsal view of shell to 
show external detail of hinge region (bar = 1.0 
mm). 



respective left and right central ridges sepa- 
rating the inhalent from the exhalent siphon. 
Whether this is a consequence of preserva- 
tion and a tenuous attachment has been lost 



cannot be determined at present. They pre- 
sumably act as do axial extensions in other 
protobranchs, as guides to the transport of 
faecal rods from anus to exhalent siphon. It 
may be speculated that in this particular case 
they have become greatly extended to ensure 
disposal far distant from the feeding aperture. 

The palps are moderate in size, with rela- 
tively broad sorting ridges on their inner 
faces. As in the case of the gill plates, the 
number of ridges on each face varies with the 
size of the specimen — 39 in a specimen 26.3 
mm total length and 14 in a specimen 3.0 mm 
total length. The palp probóscides are broad 
and long, even in the contracted, preserved 
state. In life they must be capable of consid- 
erable extension beyond the shell. 

The foot and viscera are extensive. The 
muscular foot is broad. The sole is deeply di- 
vided and fringed with papillae. There is a 
small "byssal" gland in the heel of the foot at 
the point where it joins the sole. The pedal 
retractor muscles are not particularly well de- 
veloped. There Is a posterior pair inserted an- 
tero-dorsal to the posterior adductor muscle 
and two pairs of anterior retractor inserted pos- 
tero-dorsal to the anterior adductor muscle. 

The mouth lies somewhat posterior to the 
ventral edge of the anterior adductor muscle. 
The oesophagus is elongate and opens dor- 
sally on the anterior face of the stomach. The 
stomach and combined style sac are moder- 
ately large and lie vertically within the body. 
Because of the brittle nature of the preserved 
specimens and because the digestive diver- 
ticula adhere closely to the stomach wall, little 
detail of the stomach was observed. Never- 
theless, a well-developed dorsal hood and an 
extensive gastric shield are present. A small 
number of grooves comprising the posterior 
sorting area were identified. There is no doubt 



146 



ALLEN 



ppVGGI Kl HT GO 



ST 



HG 




FIG. 5. Nuculana grasslei. Semidiagrammatic drawing of the internal morphology of a specimen from the 
right side (bar = 1 .0 mm). AA, anterior adductor muscles; AS, anterior sense organ; BG, "byssal" gland; CG, 
cerebral ganglion; OP, "catch" portion of adductor muscle; DG, digestive diverticula; FA, feeding aperture; 
FT, foot; GA, extension of gill axis; Gl, gill; GO, gonad; HG, hindgut; HT, heart; Kl, kidney; PA, posterior 
adductor muscle; PG, pedal ganglion; PL, palp; PP, palp proboscis; PR, pedal retractor muscle; QP, "quick" 
portion of adductor muscle; SE, siphonal embayment; SI, combined siphon; ST, stomach; VG, visceral 
ganglion. 




FA 



FIG. 6. Nuculana grasslei. Enlarged detail of the 
siphon and postlarval margin of the left mantle (bar 
= 0.1 mm). DR, dividing ridge; ES, exhalent si- 
phon; FA, feeding aperture; IF, inner mantle fold; 
IS, inhalent siphon; MT, mantle tentacle; SN, si- 
phonal nerve; VM, ventral margin of inhalent si- 
phon. 



that the morphology of the stomach differs lit- 
tle from the typical deep-sea nuculanid stom- 
ach (Allen & Hannah, 1989). The hindgut 
takes a typical course. From the style sac, it 
passes posterior to the stomach to the dorsal 
margin of the viscera. It then describes a loop 
on the right side of the body (Fig. 7), reaching 
the internal face of the anterior adductor mus- 
cle before passing posteriorly along the mid 
dorsal margin of the body, through the peri- 
cardium and ventricle of the heart, over the 
posterior adductor muscle to the anus. There 
is a typhlosole along the length of the hindgut; 
the faecal rods are typically compact with a 
groove moulded by the typhlosole. The diges- 
tive diverticula are very extensive with fine tu- 
bules that permeate the entire visceral mass. 
The heart is exceptionally large. Paired lat- 
eral auricles are each supplied anteriorly via a 
major vessel from the gill axis. The blood vol- 
ume also appears to be large. In all speci- 
mens, the contraction of the body on preser- 



NEW DEEP-WATER HYDROTHERMAL SPECIES 



147 



Gl 



HG 

DH 
OE 




АД ' 

FIG. 7. Nuculana grasslei. Dorsal view of the inter- 
nal morphology of a specimen to show the course 
taken by the hind gut and the disposition of the right 
gill (bar = 1.0 mm). ДА, anterior adductor muscle; 
DH, dorsal hood; Gl, gill; HG, hind gut. 



vation has forced blood to various parts of the 
body, particularly the sinuses of the mantle 
margin and the gill and gill axis. These are 
swollen with congealed red-pigmented blood. 

The kidney consists of paired brown-pig- 
mented intercommunicating sacs, lying be- 
tween the heart and the posterior adductor 
muscle. It is particularly well developed. 

The nervous system follows the typical pro- 
tobranch plan. The paired cerebral ganglia 
are slender and not well developed. Similarly, 
the visceral ganglia, although somewhat 
larger than the cerebral, are also small in 
comparison with other deep-sea nuculanids. 
From each visceral ganglion, there is a major 
nerve to the gill axis, to the siphon, and to the 
mantle edge (Fig. 5). The pedal ganglia are 
large and lie at the interface of foot and vis- 
cera, anterior and close to the ventral limit of 
the hindgut. 

Paired gonads were seen In specimens 



>18 mm total length. The major portion of the 
gonad lies anterior to the heart and dorsal and 
posterior to the stomach. From there, it 
spreads thinly across the lateral surface of the 
digestive gland. The gonadial ducts traverse 
the lateral faces of the kidney to open in the 
supramantle cavity. No fully mature gonad 
was present in the specimens examined. 

Shell Growth 

Because of the wide difference in the size 
of the specimens examined, it was possible to 
obtain some information on the change in 
shape of the shell with increasing size. 

The prodissoconch is oval and large (275- 
283 |xm total length) equivalve and approxi- 
mately equilateral (Fig. 3). The prodissoconch 
of the post-larva illustrated by Berg & Van Do- 
ver (1987), and by kind permission redrawn 
here for comparison with the prodissoconchs 
present on the adult shells, has a reticulated 
ornamentation that is presently without paral- 
lel in the Protobranchia and almost so in bi- 
valves in general. 

Post-prodissoconch shell growth immedi- 
ately begins to take on adult proportions. The 
anterior growth is less than the posterior, and 
the disparity in the numbers of teeth on the 
hinge plates is immediately apparent, with 
two anterior and three posterior teeth present 
in the smallest post-larval shells (480 ixm total 
length) in the collection. The teeth are on a 
broad and continuous hinge plate (Figs. 1, 2). 
The outline of the shell gradually changes 
with growth, and by the time the shell is 10 
mm long the adult proportions are established 
(Figs. 2, 8). Thus, the percentage ratio of 
height over length to length over the first five 
millimeters of growth changes from 75% to 
65%. At the same time, the shell becomes 
more rostrate, with the post-umbonal length 
increasing in relation to total length, while the 
shell becomes more slender. This change in 
shape with size is typical of all deep-sea pro- 
tobranchs (Allen & Hannah, 1989). 

With increasing size (age), the umbonal re- 
gion of the shell becomes increasingly 
eroded. All specimens of more than 10 mm 
total length show erosion to some degree. In 
the case of the larger specimens (Fig. 9), an 
area equivalent to the outline of a 10-mm 
shell may be affected and to such an extent 
that all that remains is the thin innermost layer 
of shell. In this extreme condition, the umbo is 
completely lost, with the ligament and the re- 
mains of the hinge plate in which the hinge 



148 



ALLEN 



70 



о 50 
& 



30 




т 1 1 1 г- 

5 



10 15 

Length (mm) 



20 



25 



FIG. 8. Nuculana gasslei. Plot of the percentage ratios of height to length (open circles), width to length 
(closed circles) and post umbonal length to length (open squares) against length. 




FIG. 9. Nuculana grasslei. Lateral view of a large shell from the left side to show the extent of corrosion (bar 
= 1.0 mm). 



teeth are clearly visible, standing out as a 
crest to the shell (Fig. 9). In addition, the area 
over the insertion of the posterior adductor 
nnuscle also becomes eroded. 

Comparisons have been made with known 
species, with particular attention being paid to 



those from off the Pacific coast of America 
and from deep water. The combined shell 
characters of N. grasslei are unlike those of 
any other described species (Abbott, 1974; 
Bernard, 1983; Dall, 1890, 1896, 1897, 1908, 
1916; Dall & Bartsch, 1910; Hertlein & Strong, 



NEW DEEP-WATER HYDROTHERMAL SPECIES 



149 



1940; Moore, 1983; Oldroyd, 1935; Willett, 
1944). The main points of recognition of N. 
grasslei include the shell outline, in which the 
postero-dorsal margin is angulate and the 
postero-ventral margin is sinuous, the large 
and anteriorly placed umbo, the slightly flat- 
tened antero-dorsal margin, and the form and 
spacing of the concentric ribs. Furthermore, 
no other description includes reference to an 
ornamented prodissoconch, though this does 
not preclude unnoted occurrence in other 
species. It must be said that the prodisso- 
conch in N. grasslei is striking, and a similar 
presence in other species is unlikely to have 
been overlooked by earlier authorities. 

Although large by deep-sea protobranch 
standards (few species obtain a length of 
more than 5 mm), N. grasslei is not large in 
comparison with other species of Nuculana. 
For example, N. pernula (Müller, 1 779) from 
shallow Arctic seas is similar in size, as too is 
N. taphria Dall, 1 897, from the shallow water 
of California and Baja California. 

Discussion 

The investigation reported here is limited to 
the gross morphological description of a new 
deep-sea hydrothermal species. Detailed mi- 
croscopical examination was not made in the 
knowledge that Dr. Richard Gustafson of Rut- 
gers University was studying various organs 
in detail. 

For the most part, the functional morphol- 
ogy of N. grasslei differs little from that of 
other species of Nuculana from slope or shelf 
seas. There are no characters that differ so 
significantly to warrant separation at generic 
level. Nevertheless, there are a few unusual 
characters that relate to the habitat of the spe- 
cies and at least one that is unrelated to the 
habitat of the adult. The former include the 
thick periostracum and the large volume of 
pigmented blood; the latter refers to the orna- 
mented prodissoconch. 

The periostracum varies in thickness but 
measures up to 40 (xm a figure that is twice 
that of N. minuta (Müller, 1776) of a similar 
size (pers. obs.). It is probable that the thick- 
ened nature of the periostracum relates to the 
sulphurous nature of the habitat. Muds smell- 
ing of hydrogen sulphide must be acidic and 
thus corrosive to the shell. The thickened pe- 
riostracum clearly protects the shell up to a 
third of the life of the animal as measured by 
shell length, i.e. to the size when gonads are 
developing. Similarly, the large blood volume 



must also relate the the nature of the habitat. 
Hydrogen sulphide will affect oxygen levels of 
the overtying sea water as well as that within 
the sediment. A large oxygen carrying capac- 
ity of the blood would be expected on a priori 
grounds. It is known that protobranchs in par- 
ticular can survive anoxic conditions for long 
periods of time (Doeller et al., 1988; pers. 
obs.). Thus, all things being equal, it would be 
expected that protobranchs could survive the 
conditions pertaining at seeps and vents with 
little modification. In fact, there is circumstan- 
tial evidence that protobranchs can survive 
reducing conditions in marine muds better 
than most bivalves, possibly with the excep- 
tion of members of the Lucinacea. In recent 
laboratory experiments, three species of Nu- 
cula have survived anoxic conditions for more 
than three weeks (pers. obs.). 

Although common to all species of Nucu- 
lana, the lack of the siphonal tentacle is per- 
haps of interest, as too is the relatively poorly 
developed nervous system. Again, it may be 
speculated that this may be preadaptive in 
that N. grasslei lives in sediments in which 
there is ample food material in the form of 
bacterial mats at the surface. In such a situ- 
ation, specialized sensory assistance in food 
gathering is of minimal importance. 

The ornamented prodissoconch is striking. 
On first reflection, little evolutionary advan- 
tage would seem to accrue from this reticula- 
tion. As in all bivalves it is protective, not in 
terms of prédation, but in terms of the protec- 
tion it affords against the dissolution of the 
shell at a weak and vulnerable point. When 
the prodissoconch is eventually lost from the 
surface of the growing adult shell, it exposed 
a small area of calcium carbonate to the 
umbo, a part of the shell that is relatively thin. 
In the case of N. grasslei, the prodissoconch 
remains in place for a relatively long period, 
protecting the shell against corrosion until the 
animal is beginning to mature. As soon as it is 
lost, corrosion occurs at the place where it 
had been. What function the reticulate orna- 
mentation plays is much less certain. Reticu- 
late ornamentation is characteristic of some 
protobranchs (e.g. Nucula sulcata Bronn, 
1831) (Allen, 1954). Whereas in the adult or- 
namentation may assist in the maintenance of 
the position of the shell within the sediment 
(Stanley, 1970), it hardly seems likely in the 
case of the newly settled prodissoconch. 

Unlike better known vent bivalves, Calypto- 
gena magnifica Boss & Turner, 1980, and 
Bathymodiolus thermophilus Kenk & Wilson, 



150 



ALLEN 



1985, N. grasslei is not exceptionally large. 
This may be related to its deposit rather than 
its suspension feeding habits, its digestive 
physiology, and to the apparent lack in the gill 
of symbiotic chemoautotrophic bacteria of the 
type present in Calyptogena and Bathymodi- 
olus, although other types of bacteria are 
present (Gustafson, pers. comm.). These lat- 
ter may bear relationship to the large volume 
of pigmented blood observed in the speci- 
mens examined. The pigment is almost cer- 
tainly haemoglobin. This is known to be 
present in other vent bivalves and in some 
other nuculanid protobranchs (Wittenberg, 
1985). It would appear that this is part of an 
efficient oxygen carrying system in relatively 
low oxygen pressures (Wittenberg, 1985). 

The large size of the prodissoconch indi- 
cates a large heavily yolked egg, probably in 
the order of 200 + ixm. (No adults with mature 
ova were present in the samples.) It is not 
unusual for vent invertebrates to have leci- 
thotrophic larvae (Gage & Tyler, 1991). Al- 
though this does not appear to restrict the 
ability of vent species in general to colonize 
new vents as they occur, at present Nuculana 
grasslei is known only from the Guaymas Ba- 
sin in the Gulf of California. 



LITERATURE CITED 

ABBOTT, R. T., 1 974, American sea shells: the ma- 
rine Mollusca of the Atlantic and Pacific coasts of 
North America. 663 pp. Van Nostrand Reinhold 
Co, New York. 

ALLEN, J. A., 1954, A comparative study of the 
British species of Nucula and Nuculana. Journal 
of the /Marine Biological Association of the United 
Kingdom, 33: 457-472. 

ALLEN, J. A. & F. J. HANNAH, 1989, Studies on 
the deep-sea Protobranchia. The subfamily Le- 
dellinae (Nuculanidae). Bulletin of the British Mu- 
seum (Natural History), Zoology, 55: 123-171. 

BERNARD, F. R., 1983, Catalogue of the living Bi- 
valvia of the eastern Pacific Ocean: Bering Strait 
to Cape Horn. Canadian Special Publication of 
Fisheries and Aquatic Sciences, 61: 1-102. 

BERG, С J. and С L. VAN DOVER, 1987, Bentho- 
pelagic macrozooplankton communities at and 
near deep-sea hydrothermal vents in the eastern 
Pacific Ocean and the Gulf of California. Deep- 
Sea Research, 34: 379-401 . 

DALL, W. H., 1890, Scientific results of explorations 
by the U.S Fish Commission steamer "Albatross." 
VII. Preliminary report on the collection of Mol- 
lusca and Brachiopoda obtained in 1887-1888. 
Bulletin of the U. S. National Museum, 12: 219- 
362. 



DALL, W. H., 1896, Diagnoses of new mollusks 
from the west coast of America. Proceedings of 
the U. S. National Museum, 18: 7-20. 

DALL, W. H., 1897, Notice of some new or inter- 
esting species of shells from British Columbia 
and the adjacent region. Bulletin of the Natural 
History Society of British Columbia, 2: 1-18. 

DALL, W. H., 1908, Reports of the dredging oper- 
ations off the west coast of Central America to the 
Galapagos, to the west coast of Mexico, and in the 
Gulf of California, in charge of Alexander Agassiz, 
carried out by the U.S. Fish Commission steamer 
"Albatross" during 1891, Lieut. -Commander Z. L. 
Tanner U.S.N., commanding. XXXVII. Reports on 
the scientific results of the expedition to the east- 
ern tropical Pacific, in charge of Alexander Agas- 
siz, by the U.S. Fish Commission steamer "Alba- 
tross" from October, 1904, to March, 1985, Lieut. - 
Commander L. M. Garrett, U.S.N., commanding. 
XIV. The Mollusca and Brachiopoda. Harvard Uni- 
versity, Bulletin of the Museum of Comparative 
Zoology 43: 205-487. 

DALL, W. H., 1916, Diagnoses of new species of 
marine bivalve molluscs from the northwest coast 
of America in the United States National Mu- 
seum. Proceedings of the U. S. National Mu- 
seum, 52: 393-417. 

DALL, W. H. & P. BARTSCH, 1910, New species of 
shells collected by Mr. John Macoun at Barkely 
Sound, Vancouver Island, British Columbia. 
Memoirs of the Geological Survey Branch, Ca- 
nadian Department of Mines, 14-N: 5-22. 

DOELLER, J. E., D. W. KRAUS, J. M. COLACINO, 
& J. B. WITTENBERG, 1988, Gill hemoglobin 
may deliver sulphide to bacterial symbionts of 
Solemya velum (Bivalvia, Mollusca). Biological 
Bulletin, 175:388-396. 

GAGE, J. D. & P. A. TYLER, 1991, Deep-sea biol- 
ogy: a natural history of organisms at the deep- 
sea floor. Cambridge University Press, 504 pp. 

GRASSLE, J. F., L. S. BROWN-LEGER, L 
MORSE-PORTEOUS, R. PETRECCA, & I. 
WILLIAMS, 1985, Deep-sea fauna of sediments 
in the vicinity of hydrothermal vents. In M. L. 
JONES, ed.. The hydrothermal vents of the east- 
ern Pacific: an overview. Bulletin of the Biological 
Society of Washington, 6: 429-442. 

HERTLEIN, L G. & A. M. STRONG, 1940, Mol- 
lusks of the west coast of Mexico and Central 
America. Part I. Zoológica, New York Zoological 
Society, 25: 369-430. 

JONES, M. L., ed., 1985, The hydrothermal vents 
of the eastern Pacific: an overview. Bulletin of the 
Biological Society of Washington, 6: 1-566. 

LONSDALE, P. F., J. L BISCHOFF, V. M. BURNS, 
M. KASTNER & R. E. SWEENEY, 1980, A high- 
temperature hydrothermal deposit on the seabed 
at the Gulf of California spreading center. Earth 
and Planetary Science Letters, 49: 8-20. 

MOORE, E. J., 1983, Tertiary marine pelecypods of 
California and Baja California: Nuculidae through 
Malletiidae. U. S. Geological Survey Professional 
Paper, 1228-A: 1-108. 



NEW DEEP-WATER HYDROTHERMAL SPECIES 



151 



OLDROYD, I. S., 1935, Two new west American 
species of Nuculanidae. Nautilus, 49: 13-14. 

SIMONEIT, B. R. T & P. F. LONSDALE, 1982, Hy- 
drothermal petroleum in mineralized mounds at 
the seabed of Guaymas Basin. Nature, 295: 
198-202. 

STANLEY, S. M., 1970, Relation of shell form to life 
habits of the Bivalvia (Mollusca). Geological So- 
ciety of America Memoir, 125: 296 pp. 

WILLETT, G., 1944, New species of mollusks from 
Redondo, California. Bulletin of the Southern 
Calif ornian Academy of Sciences, 43: 71-73. 



WITTENBERG, J. В., 1985, Oxygen supply to in- 
tracellular bacterial symbionts. In м. к. jones, ed., 
The hydrothermal vents of the eastern Pacific: an 
overview. Bulletin of the Biological Society of 
Washington, 6: 301-310. 

YONGE, С M., 1939, The protobranchiate Mol- 
lusca: a functional interpretation of their structure 
and evolution. Transactions of the Royal Society 
of London, B, 230: 79-147. 



Revised Ms. accepted 29 April 1 992 



MALACOLOGIA, 1993, 35(1): 153-154 



LETTERS TO THE EDITOR 



REPLY TO "SUPRASPECIFIC NAMES OF MOLLUSCS: 
A QUANTITATIVE REVIEW" 

M. A. Edwards^ & M. J. Thorne^ 

ABSTRACT 

The article Supraspecific names of Molluscs; a quantitative review' by Phillipe Bouchet and 
Jean-Pierre Rocroi, contains some misapprehensions about the Zoological Record. This article 
seeks to correct them. 

Key words: Literature coverage, Mollusca, Taxonomic names. Zoological Record 



"Critics will certainly find it easy to discover defi- 
ciencies in the volume, but we may doubt whether 
they will realize the extent of the work involved in 
it." (Sharp, 1902) 

This comment, made by the then editor of the 
Zoological Record, is, apparently, as true to- 
day as it was nearly a century ago. 

The recent article by Bouchet & Rocroi 
(1992) discusses the numbers of supraspe- 
cific names in Mollusca, and takes the Zoo- 
logical Record to task for what they estimate 
to be an omission rate of 20% in respect of 
those names, particularly in the period 1960- 
1989. 

Those responsible for the Zoological 
Record are not averse to criticism, but the 
Mollusca must be considered in the context of 
the wide field of literature on all animal groups 
which the Record endeavours to search with 
the limited resources at its disposal. Although 
the annual growth in the number of new mol- 
luscan names may have remained reason- 
ably stable, the growth in the literature most 
certainly has not. 

Each annual volume of the Zoological 
Record cowers the recent literature relating to 
nearly 50 different animal groups. To locate 
relevant work, over 6,500 serials are 
searched, as available, together with some 
1 ,500 or more books and reports; from these, 
65-70,000 individual items are indexed each 
year. In addition, names described in works 
published in earlier years are constantly com- 



ing to light. These are included in that volume 
of the Record being indexed at the time of 
discovery, which makes an omission rate im- 
possible to define in the long term. 

Reference is made to the imperfect cover- 
age of some literature, in particular that from 
China, Japan and the former Soviet Union. 
While this is not disputed, it must be appreci- 
ated that access to this material is often diffi- 
cult, and the linguistic skills required to index 
it are expensive to obtain. Nevertheless, de- 
tails of additional publications are always wel- 
come. (Of those titles mentioned in the article, 
the two primary publications are covered in 
the Record, but the Chinese secondary pub- 
lication is not because abstracts are not nor- 
mally indexed.) 

Each section of the Zoological Record car- 
ries a request to authors to provide copies of 
recent publications for indexing purposes, 
and considerable efforts are made to obtain 
literature not previously covered. 

It is inevitable, however, that workers in a 
particular field in touch with colleagues will 
have more complete listings than the Record, 
and no doubt more opportunities to visit librar- 
ies abroad, to "browse" through reprint col- 
lections, and to check bibliographic compila- 
tions which may span many years. To do this 
on the scale required for all animal groups 
indexed in the Record would be beyond the 
resources available. 

Bouchet & Rocroi also say that the Record 



'The Zoological Society of London, Regent's Park, London NW1 4RY, England. 
^BIOSIS, U.K.. Garforth House, 54 Micklegate, York, North Yorkshire Y01 1LF, England. 

153 



154 



EDWARDS & THORNE 



is "supposedly the most complete indexing 
system," "a nomenclátor considered to be the 
most complete . . ." and go on to state that the 
"unexpectedly high omission rate . . . should 
cause concern to all taxonomists. Because 
this nomenclátor is the main bibliographical 
source of many (palaeo) zoologists . . .". They 
then suggest that names should be registered 
before they can be declared nomenclaturally 
available. 

The Record has never claimed to be com- 
plete, that would be impossible, but it is evi- 
dently still considered to be "the main biblio- 
graphical source" and no other more 
comprehensive work in the zoological field is 
known. As regards the registration of names, 
Zoological Record staff are working with the 
International Commission on Zoological No- 
menclature to establish such a register, 
though of course for Zoological Record pur- 
poses names would still have to be indexed 
whether or not they were registered. 

Compilation and production of the Zoolog- 
ical Record is an excessively expensive un- 
dertaking. Throughout its long history there 
have always been appeals for funds but little 
response from those who, while insisting on 
its continuation, are unwilling to provide suffi- 
cient financial support and rely on the publish- 



ers (The Zoological Society and now BIOSIS) 
to subsidize it. 

If the article by Bouchet & Rocroi helps to 
highlight the difficulties faced by the Zoologi- 
cal Record and thereby increases interest in 
and support for this unique publication, it will 
have served a useful purpose. Othenwise the 
biological community should seriously con- 
sider what the effects might be should the 
Record cease publication. 



LITERATURE CITED 

BOUCHET, PHILIPPE & JEAN-PIERRE ROCROI, 
1992, Supraspecific names for molluscs: a quan- 
titative review. Malacologie, 34:75-86. 



The editor-in-chief of Malacologie welcomes let- 
ters that comment on vital issues of general im- 
portance to the field of Malacology, or that com- 
ment on the content of the journal. Publication is 
dependent on discretion, space available and, in 
some cases, review. Address letters to: Letter to 
the Editor, Malacologia, care of the Department of 
Malacology, Academy of Natural Sciences, 19th 
and the Parkway, Philadelphia, PA 19103. 







Publication dates 


Vol. 


28, 


No. 1-2 


19 January 1988 


Vol. 


29, 


No. 1 


28 June 1988 


Vol. 


29, 


No. 2 


16 Dec. 1988 


Vol. 


30, 


No. 1-2 


1 Aug. 1989 


Vol. 


31, 


No. 1 


29 Dec. 1989 


Vol. 


31, 


No. 2 


28 May 1990 


Vol. 


32, 


No. 2 


7 June 1991 


Vol. 


33, 


No. 1-2 


6 Sep. 1991 


Vol. 


34, 


No. 1-2 


9 Sep. 1992 



AWARDS FOR STUDY AT 
The Academy of Natural Sciences of Philadelphia 

The Academy of Natural Sciences of Philadelphia, through its Jessup and 
McHenry funds, makes available each year a limited number of awards to support 
students pursuing natural history studies at the Academy. These awards are pri- 
marily intended to assist predoctoral and immediate postdoctoral students. Awards 
usually include a stipend to help defray living expenses, and support for travel to and 
from the Academy. Application deadlines are 1 March and 1 October each year. 
Further information may be obtained by writing to: Chairman, Jessup-McHenry 
Award Committee, Academy of Natural Sciences of Philadelphia, 1 900 Benjamin 
Franklin Parkway, Philadelphia, Pennsylvania 19103-1195, U.S.A. 



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 

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



MALACOLOGIA, 1993, 35(1): 



35(1) 



MALACOLOGIA 



1993 



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

4. Unless indicated otherwise below, con- 
tributors should follow the recommendations 
in the Council of Biology Editors (CBE) Style 
Manual (ed. 5, 1983) available for U.S. 
$24.00 from CBE, 9650 Rockville Pike, 
Bethesda, MD 20814, U.S.A. 

5. Be brief. 

6. Manuscripts must be typed on one side 
of good quality white paper, double-spaced 
throughout (including the references, tables 
and figure captions), and with ample margins. 
Tables and figure captions should be typed on 
separate pages and put at the end of the 
manuscript. Make the hierarchy of headings 
within the text simple and consistent. Avoid 
internal page references (which have to be 
added in page proof). 

7. Choose a running title (a shortened 
version of the main title) of fewer than 50 
letters and spaces. 



8. Provide a concise and informative Ab- 
stract summarizing not only contents but re- 
sults. A separate summary generally is super- 
fluous. 

9. Supply between five and eight key 
(topic) words to go at the end of the Abstract. 

10. Use the metric system throughout. Mi- 
cron should be abbreviated ц,т. 

11. Illustrations are printed either in one 
column or the full width of a page of the 
journal, so plan accordingly. The maximum 
size of a printed figure is 13.5 x 20.0 cm 
(preferably not as tall as this so that the cap- 
tion does not have to be on the opposite 
page). 

12. Drawings and lettering must be dark 
black on white, blue tracing, or biue-lined 
paper. Lines, stippling, letters and numbers 
should be thick enough to allow reduction by 
Уг or Уз. Letters and numbers should be at 
least 3 mm high after reduction. Several 
drawings or photographs may be grouped 
together to fit a page. Photographs are to be 
high contrast. High contrast is especially im- 
portant for histological photographs. 

13. All illustrations are to be numbered 
sequentially as figures (not grouped as plates 
or as lettered subseries), and are to be ar- 
ranged as closely as possible to the order in 
which they are first cited in the text. Each 
figure must be cited in the text. 

14. Scale lines are required for all nondi- 
agrammatic figures, and should be conve- 
nient lengths (e.g., "200 ixm," not "163 jxm"). 
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-institu- 
tional purchaser; $45.00 for institutional pur- 
chaser. There is a one dollar handling charge 
per volume for all purchases of single vol- 
umes. Address inquiries to the Subscription 
Office. 



VOL. 35, NO. 1 MALACOLOGIA 1993 

CONTENTS 

JONATHAN COPELAND & MARYELLEN MANERI DASTON 

Adult and Juvenile Flashes in the Terrestrial Snail Dyakia striata 1 

MARYELLEN MANERI DASTON & JONATHAN COPELAND 

The Luminescent Organ and Sexual Maturity in Dyakia striata 9 

HARLAN K. DEAN 

A Population Study of the Bivalve Idas argénteas Jeffreys, 1876, (Bivalvia: 
Mytilidae) Recovered from a Submerged Wood Block in the Deep North 
Atlantic Ocean 21 

MICHAEL S. JOHNSON, JAMES MURRAY & BRYAN CLARKE 

Evolutionary Relationships and Extreme Genital Variation in a Closely 
Related Group of Partula 43 

CARLOS E. PRIETO, ANA I. PUENTE, KEPA ALTONAGA & BENJAMIN J. GOMEZ 
Genital Morphology of Caracollina lenticula (Michaud, 1831), with a 
New Proposal of Classification of Helicodontoid Genera (Pulmonata: 
Hygromioidea) 63 

ALOIS HONÉK 

Melanism in the Land Snail Helicella candicans (Gastropoda, Helicidae) 

and its Possible Adaptive Significance 79 

ANETTE BAUR & BRUNO BAUR 

Daily Movement Patterns and Dispersal in the Land Snail 

Arianta arbustorum 89 

LUC MADEC & JACQUES DAGUZAN 

Geographic Variation in Reproductive Traits of Helix aspersa Müller Studied 
under Laboratory Conditions 99 

JOHN B. WISE 

Anatomy and Functional Morphology of the Feeding Structures of the Ecto- 
parasitic Gastropod Boonea impressa (Pyramidellidae) 119 

ELBA MORRICONI Y JORGE CALVO 

Influencia Ambiental Sobre el Crecimento Alométrico de la Valva en Nacella 
(Patinigera) deaurata (Gmelin, 1 791 ) del Canal Beagle, Argentina 135 

J. A. ALLEN 

A New Deep-Water Hydrothermal Species of Nuculana (Bivalvia: Protobran- 
chia) from the Guaymas Basin 141 

M. A. EDWARDS & M. J. THORNE 

Letters to the Editor 153 



MCZ 
LIBRARY 

i/OL 35, NO. 2 L (i IV93 ''^^^ 

HARVARD 
UNÍVER3ÍTY 



MALACOLOGIA 



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



MALACOLOGIA 

Editor-in-Chief: 
GEORGE M. DAVIS 

Editorial and Subscription Offices: 

Department of Malacology 

The Academy of Natural Sciences of Philadelphia 

1900 Benjamin Franklin Parkway 

Philadelphia, Pennsylvania 19103-1195, U.S.A. 



EUGENE COAN 

California Academy of Sciences 

San Francisco, CA 



Co-Editors: 



Assistant hAanaging Editor: 

CARYL HESTERMAN 

Associate Editors: 



CAROL JONES 
Denver, CO 



JOHN B. BURCH 
University of Michigan 
Ann Arbor 



ANNE GISMANN 

Maadi 

Egypt 



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



KENNETH J. BOSS 

Museum of Comparative Zoology 

Cambridge, Massachusetts 

JOHN BURCH, President 

MELBOURNE R. CARRIKER 
University of Delaware, Lewes 

GEORGE M. DAVIS 
Secretary and Treasurer 

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



JAMES NYBAKKEN 

Moss Landing Marine Laboratory 

California 

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

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

SHI-KUEI WU 

University of Colorado Museum, Boulder 



Participating Members 

EDMUND GITTENBERGER JACKIE L VAN GOETHEM 

Secretary, UNITAS MALACOLOGICA Treasurer, UNITAS MALACOLOGICA 

Rijksmuseum van Natuurlijke Koninklijk Belgisch Instituut 

Historie voor Natuurwetenschappen 

Leiden, Netherlands Brüssel, Belgium 



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

ELMER G. BERRY, 
Germantown, Maryland 



Emeritus Members 

ROBERT ROBERTSON 

The Academy of Natural Sciences 

Philadelphia, Pennsylvania 



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



Copyright © 1993 by the Institute of Malacology 



1993 
EDITORIAL BOARD 



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. 

С 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 



E. GITTENBERGER 
Rijksmuseum van Natuurlijke Historie 
Leiden, Netherlands 

F. GIUSTI 
Université di Siena, Italy 

A. N. GOLIKOV 
Zoological Institute 
St. Petersburg, Russia 

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

A. V. GROSSU 
Universitatea Bucuresti 
Romania 

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 
Göteborg, Sweden 

S. HUNT 
Lancashire 
United Kingdom 

R. JANSSEN 

Forschungsinstitut Senckenberg, 
Frankfurt am Main, Germany 

R. N. KILBURN 
Natal Museum 
Pietermaritzburg, South Africa 

M. A. KLAPPENBACH 

Museo Nacional de Historia Natural 

Montevideo, Uruguay 



J. KNUDSEN 

Zoologisk Institut & Museum 

Kobenhavn, Denmark 

A. J. KOHN 

University of Washington 

Seattle, U.S.A. 

A. LUCAS 

Faculté des Sciences 

Brest, France 

С MEIER-BROOK 
Tropenmedizinisches Institut 
Tübingen, Germany 

H. К. MIENIS 

Hebrew University of Jerusalem 

Israel 

J. E. MORTON 
The University 
Auckland, New Zealand 

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

R. NATARAJAN 
l^arine Biological Station 
Porto Novo, India 

J. 0KLAND 
University of Oslo 
Norway 

T. OKUTANI 
University of Fisheries 
Tokyo, Japan 

W. L. PARAENSE 

Instituto Oswalde Cruz, Rio de Janeiro 

Brazil 

J. J. PARODIZ 
Carnegie hAuseum 
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 



Ql Z. Y. 

Academia Sínica 

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. SEGERSTRÄLE 
Institute of Marine Research 
Helsinki, Finland 

A. STAÑCZYKOWSKA 
Siedlce, Poland 

F. STARMÜHLNER 

Zoologisches Institut der Universität 

Wien, Austria 

Y. I. STAROBOGATOV 
Zoological Institute 
St. Petersburg, Russia 

W. STREIFE 
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 
Kallaroo, Western Australia 

H. ZEISSLER 
Leipzig, Germany 

A. ZILCH 

Forschungsinstitut Senckenberg 

Frankfurt am Main, Germany 



MALACOLOGIA, 1993, 35(2): 155-259 



PHYLOGENETIC ANALYSIS OF THE RAPANINAE 
(NEOGASTROPODA: MURICIDAE) 

Silvard P. Kool 

Mollusk Department, Museum of Comparative Zoology, Harvard University, 
Cambridge, Massachusetts 02138, U.S.A. 

ABSTRACT 

The generic level revision and phylogenetic analysis of the gastropod subfamily Rapaninae 
Gray, 1853 (Prosobranchia: Neogastropoda: Muricidae), presented here is based primarily on 
gross anatomy (female and male reproductive systems, alimentary system, mantle cavity or- 
gans), radular, opercular, and protoconch morphology, and shell ultrastructure. Results reveal 
that Rapaninae includes most members previously allocated to the Thaidinae Jousseaume, 
1888. The type species of most recognized rapanine genera were studied for character selec- 
tion. Eighteen characters were determined for cladistic analyses, and results were compared 
with additional data derived from egg capsule morphology and biogeographic data. 

The cladistic analyses show (1) that the former Thaididae/nae of authors is polyphyletic and 
should be divided into two (monophyletic) groups; (2) that family status is not justified for either 
of these groups; (3) that Rapana Schumacher, 1817, is monophyletic with Thaidinae, resulting 
in synonymization of Thaidinae Jousseaume, 1888, with Rapaninae Gray, 1853; and (4) that 
several genera belonging to the Rapaninae merely deserve subgeneric status. 

The genera Nucella Röding, 1798, Forreria Jousseaume, 1880, Trochia Swainson, 1840, 
Acanthina Fischer von Waldheim, 1807, and Haustrum Perry, 1811, are placed in Ocenebrinae 
Cossmann, 1903 {sensu Kool, 1993); the genera Cymia Mörch, 1860, Rapana Schumacher, 
1817, Stramonita Schumacher, 1817, Concholepas Lamarck, 1801, Dicathais Iredale, 1936, 
Drupa Röding, 1798, Plicopurpura Cossmann, 1903, Pinaxia H. & A. Adams, 1853, Nassa 
Röding, 1798, Vexilla Swainson, 1840, Cronia H. & A. Adams, 1853, Morula Schumacher, 1817, 
Thais Röding, 1798, Purpura Bruguière, 1789, and Mancinella Link, 1807, are placed in Ra- 
paninae. The taxa Vasula (y/lörch, 1860, Tribulus Sowerby, 1839, and Neorapana Cooke, 1918, 
are allocated subgeneric status under Thais. 

"My Thais, thou hast seen these filthy snails crawling towards thee with 
their sticky sweat . . . Thais, Thais, Thais, . . . say if thou wilt go mad with 
them!" 

Anatole France, Thais 



INTRODUCTION 

Of all large littoral prosobranchs, none are 
more conspicuous and perplexing, in a taxo- 
nomic sense, than gastropods belonging to 
the Rapaninae ["Rapananina"] Gray, 1853, 
herein shown to include Thaidinae Jous- 
seaume, 1888 {sensu Kool, 1989 [= Thaid- 
Idae/nae of authors, in partem]). Rapaninae, 
sensu Kool (from this point onward referred to 
as Rapaninae), comprises many more genera 
than Rapaninae of authors. The Rapaninae is 
a group of predatory gastropods belonging to 
the family Muricidae Rafinesque, 1815, in the 
superfamily Muhcoidea {sensu Ponder, 1973; 
see below). Most rapanines live in the rocky 
intertidal zone where wave energy can be 
very high, but members of the genus Rapana 
Schumacher, 1817, are subtidal. Rapanines 



prey on a variety of invertebrates (mollusks, 
polychaetes, crustaceans, cnidarians, etc.; 
see Kool, 1987), although some are known to 
eat invertebrate and vertebrate carrion; some 
species are specialists (for example, coral 
feeders), others generalists. 

My initial assumption was that the Thaid- 
idae/nae of authors was a conglomerate of 
disparate taxa, and that para- and polyphyly 
would be rampant in this "waste-basket 
group." Although Rapaninae have been com- 
monly used for ecological (Spight, 1982; J. D. 
Taylor, 1984), environmental (Bryan et al., 
1986, 1987), genetic (Palmer, 1984, 1985), 
physiological (Carriker et al., 1978), and bio- 
chemical (Huang & Mir, 1972) research, little 
is known about the evolutionary relationships 
among the members of this group, and its sta- 
tus among other muhcid groups. 



155 



156 



KOOL 



Taxonomic History 

Traditionally, the superfamily Muricoidea 
Rafinesque {sensu Thiele [as Muricacea]) 
has been divided into several different fami- 
lies (Table 1). Ponder (1973) advocated inclu- 
sion of several other neogastropod families in 
Muricoidea, so that Muricoidea, sensu Thiele, 
is almost equivalent to Muhcidae, sensu Pon- 
der. Unless noted otherwise, Muhcidae will 
herein be equivalent to Muricoidea, sensu 
Thiele. 

Members of the Muhcidae have an often 
spiny shell, usually bearing a distinct, some- 
times long, anterior siphonal canal. An ana- 
tomical feature shared by most Muhcidae is 
the accessory bohng organ, located in the 
foot, and used for chemically dissolving shell 
matehal. Naticids have an accessory bonng 
organ as well, but this structure apparently 
has ahsen independently in these distinct 
groups. Most Muhcidae have a long radular 
nbbon with a row of tri- or pentacuspid rachid- 
ian (central) teeth, each of which is flanked by 
a lateral tooth. The th- and pentacuspid 
rachidian morphology occurs also in other 
Neogastropoda (for example, Buccinidae). 

The taxonomy and phylogeny of the Muri- 
cidae have been in a state of confusion for 
over two centunes. Taxonomic problems 
within the Muhcidae as a whole impede our 
understanding of all groups within this taxon. 
For example, due to the vague boundaries of 
many higher muhcid taxonomic groups, the 
limits of lower groups can not be set, and vice 
versa. Keen (1971a: 35) pointed out that "dis- 
tinctions between subfamilies within the Mu- 
hcidae are not always clear-cut, . . ." This 
taxonomic confusion results in a lack of un- 
derstanding of the phylogeny of all muhcid 
groups. 

Familial and subfamilial arrangements of 
Muricidae differ greatly among authors. A se- 
lection of arrangements and authors is listed 
in Table 1. For example, Cossmann (1903) 
recognized five subfamilies within the Muri- 
cidae: Ocenebhnae [authors and dates of 
taxa given in Table 1], Muhcinae, Trophoni- 
nae, Typhinae, and Rapaninae; he included 
the members of the Thaididae/nae of authors 
in the Purpuhdae as a separate family. Thiele 
(1929) included two families, Muhcidae and 
Magilidae, and did not list subfamilies. Wenz 
(1941) included the same two families, but 
subdivided the Muhcidae into the subfamilies 
Muhcinae, Rapaninae, Columbahinae, and 
Drupinae (Thaidinae of authors). Keen 



(1971a) recognized the families Muricidae, 
Columbahidae, Sarganidae, Coralliophilidae, 
Moreidae, and Thaididae; she subdivided 
the Thaididae into the subfamilies Thaidinae, 
Rapaninae, and Drupinae. Radwin & D'Attilio 
(1971) subdivided the Muhcidae into the 
families Muhcidae, Columbahidae, Ra- 
panidae, Coralliophilidae, and Thaididae. 
Ponder (1973) reduced the number of super- 
families in the Neogastropoda and included 
the Buccinidae, together with 16 other fami- 
lies in the Muricoidea, and followed Coss- 
mann's (1903) subdivision of the Muhcidae. 
Harasewych (1983) showed that the Colum- 
bahinae do not belong within the Muhcidae 
but instead in the Turbinellidae. Ponder & 
Waren (1988) include in Muhcidae the sub- 
families Muricinae, Thaidinae (with Rapani- 
nae in synonymy), Coralliophilinae, Sargani- 
nae, and Moreinae. 

Of the subgroups of the Muhcidae, the 
group formehy known as Thaidinae (or as 
Thaididae) Jousseaume, 1888 (ohginal spell- 
ing "Thaisidae"), is probably the most prob- 
lematic and in need of comprehensive revi- 
sion. Some authors have ranked this group as 
a subfamily, but many have given it family 
rank (Table 2). 

The family-subfamily controversy is a result 
of a poor understanding of genus-level rela- 
tionships within the Rapaninae and of rela- 
tionships between Rapaninae and the other 
muricid taxa. The genehc allotment for the 
many rapanine species is highly suspect, as 
genehc boundaries are usually ill-defined. 
Many muricid genera of uncertain status have 
been placed in Thaididae/nae of authors, re- 
sulting in a conglomerate of disparate taxa. 
Therefore, Thaididae/nae of authors, as well 
as other higher level muricid taxa, are proba- 
bly para- and/or polyphyletic. 

Taxonomic controversy in Rapaninae has 
existed from the time when rapanine genera 
were given their own group-name and rank- 
ing. Menke (1828) considered the group as a 
superfamily and used the name Purpuracea. 
Swainson (1835, 1840) referred to this group 
as Purpuhnae. Broderip (1839) ranked this 
group as a family (Purpuhdae). The family- 
level designation has been used most fre- 
quently since then. Other synonyms of Thai- 
didae/nae of authors (and thus in partem of 
Rapaninae, as defined herein) are Conchole- 
padidae Perrier, 1897, Purpuradae Leach, 
1852, Thaisidae Jousseaume, 1888, Thaidae 
Cooke, 1919, Drupinae Wenz, 1941, Thaisid- 
inae Kuroda & Habe, 1971, Thaidiidae Atap- 



PHYLOGENY OF RAPANINAE 



157 



attu, 1972, and Nucellinae Kozloff, 1987 (see 
also Ponder & Waren, 1988). 

The oldest rapanine generic name still in 
use is Purpura, introduced by Martini (1777). 
Due to the controversial history of Purpura 
(see treatment of this genus), Keen (1964) 
proposed that the names "Purpurinae," "Pur- 
puridae" and "Purpuracea" be placed on the 
Official Index of Rejected and Invalid Family- 
Group Names in Zoology and to place Thai- 
didae Suter, 1913 [originally as "Thaisidae"], 
on the Official List of Family-Group Names in 
Zoology. The Commission acted on this peti- 
tion (ICZN, Opinion 886, 1969) and placed 
Purpuracea Menke, 1828, and Purpurinae 
Swainson, 1840 [sic], on the Official Index of 
Rejected and Invalid Family-Group Names in 
Zoology. Furthermore, the Committee ruled 
that Purpuridae Broderip, 1839, and Thaid- 
idae Suter, 1913, be placed on the Official List 
of Family-Group Names in Zoology, and that 
Purpuridae not have priority over Thaididae. 
From this point on, the stem "Thaid-" has 
been used most frequently for rapanine gas- 
tropods (Table 2). As Cernohorsky (1980) 
pointed out, "Thaididae Jousseaume, 1888" 
(originally as "Thaisidae"), predates Thaid- 
idae Suter. Lehtinen (1985) petitioned to 
adopt the original spelling "Thaisidae" to 
avoid homonymy with the spider family Thai- 
didae Lehtinen, 1967 (based on the genus 
Thaida), but later withdrew his petition. 

Convergent Shell Morphology: Roots of 
Taxonomic Discord 

The main reason for the plethora of taxo- 
nomic arrangements for muricid groups is a 
poor understanding of muricid phylogeny. 
The characters on which all past taxonomic 
schemes were based are distilled primarily 
from external shell morphology. These fea- 
tures are readily visible but are misleading in 
that they may have resulted from convergent 
and/or parallel evolution. 

Many authors have pointed out that shell 
morphology within a species is effected by 
environmental influences. For example, envi- 
ronmental factors often dictate a particular 
shell shape and/or shell color. Examples of 
ecophenotypic variation are given in a num- 
ber of papers on muhcids (primarily the genus 
Nucella Röding) (Agersborg, 1929; Vermeij, 
1975, 1979, 1982; Palmer, 1979; Vermeij & 
Currey, 1980; Etter, 1987; Day, 1990) and on 
other gastropod groups as well (S. J. Gould, 
1971; Cain, 1981). If environmental influ- 



ences are strong enough to cause high selec- 
tion pressures at the population level, selec- 
tive forces may also have caused conver- 
gence in shell shape among species. Shell 
convergence among species may thus be 
high, and any taxonomic scenario for the Mu- 
ricidae (or other gastropod group) based ex- 
clusively or primarily on shell morphology is 
therefore highly suspect. 

Evidence for the phenomenon of environ- 
mentally induced shell shape is given for the 
species Nucella lapillus. Cooke (1895, 1919) 
pointed put that stunted, short-spired speci- 
mens of Nucella lapillus occurred in very ex- 
posed areas, whereas those living in sheltered 
areas had high-spired shells with a relatively 
small aperture. Crothers' (1973, 1974) studies 
on ecophenotypic variation of Nucella lapillus 
reported similar results to those of Cooke. 
Kitching et al. (1 966) were able to demonstrate 
experimentally that morphs of Nucella with 
wide apertures had greater adhesive power to 
cling to intertidal rocks than did the morphs 
with narrower apertures, thus providing an ad- 
aptationist explanation for variation in shell 
shape. Other characters derived from shell 
morphology correlating with environment are 
color patterns and sculpture (Agersborg, 
1929; Etter, 1987). 

Besides wave action, other environmental 
influences reportedly play a role in determin- 
ing aspects of shell morphology. Bala- 
parameswara Rao & Bhavarayana (1976) 
were able to correlate shell morphology sta- 
tistically in Drupa tuberculata with tempera- 
ture and desiccation at different intertidal lev- 
els. Moore (1936) suggested that the great 
intraspecific variation in shell shape in Nu- 
cella was due to differential feeding. Bändel 
(1984) showed that juveniles of Stramonita 
haemastoma floridana would "change" into 
typical Stramonita haemastoma in the labora- 
tory when food levels were kept artificially 
high. Hallam (1965) stated that a combination 
of such factors as food availability, salinity, 
oxygen concentration, temperature, turbidity 
and agitation, and population density, may in- 
duce stunting in mollusks and other inverte- 
brates. Wilbur & Owen (1964), in discussing 
allometric growth in mollusks, pointed out that 
growth rates for different bodily parts may not 
be equal; thus shell shape may depend on a 
snail's age. They also showed that this allom- 
etry may also partly be due to a combination 
of several environmental factors. 

Many authors have noted population differ- 
ences in shell shape in different muricidae 



158 



KOOL 



TABLE 1 . Important supraspecific taxonomic arrangements for muricids. 



Authors 



Taxonomic Names 



Fischer, 1887 
Cossmann, 1903 



Thiele, 1929 
Wenz, 1941 



Radwin& D'Attilio, 1971 



Keen, 1971a 



Ponder, 1973 



Golikov & Starobogatov, 1 975 



PECTINIBRANCHIATA 
MURICIDAE Rafinesque, 1815 
CORALLIOPHILIDAE Chenu, 1859 
RHACHIGLOSSA 
MURICIDAE Rafinesque, 1815 
MURICINAE Rafinesque, 1815 
OCENEBRINAE Cossmann, 1903 
TROPHONINAE Cossmann, 1903 

(incl. Forreria) 
TYPHINAE Cossmann, 1903 
RAPANINAE Gray, 1853 
PURPURIDAE Broderip, 1839 
(incl. thaidines s.l.) 
CORALLIOPHILIDAE Chenu, 1859 
MURICACEA Rafinesque, 1815 
MURICIDAE Rafinesque, 1815 
MAGILIDAE Thiele, 1925 
MURICACEA Rafinesque, 1815 
MURICIDAE Rafinesque, 1815 
RAPANINAE Gray, 1853 

(incl. Forreria) 
COLUMBARIINAE Tomlin, 1928 
MURICINAE Rafinesque, 1815 
DRUPINAE Wenz, 1941 
(incl. thaidines s.l.) 
MAGILIDAE Thiele, 1925 
(incl. Coralliophila) 
MURICACEA Rafinesque, 1815 
COLUMBARIIDAE Tomlin, 1928 
RAPANIDAE Gray, 1853 
CORALLIOPHILIDAE Chenu, 1859 
THAIDIDAE Jousseaume, 1888 
MURICIDAE Rafinesque, 1815 
(7 subfamilies) 
MURICACEA Rafinesque, 1815 
MURICIDAE Rafinesque, 1815 

(5 subfamilies) 
COLUMBARIIDAE Tomlin, 1928 
CORALLIOPHILIDAE Chenu, 1859 
MOREIDAE Stephenson, 1941 
SARGANIDAE Stephenson, 1923 
THAIDIDAE Jousseaume, 1888 
THAIDINAE Jousseaume, 1888 
DRUPINAE Wenz, 1941 
RAPANINAE Gray, 1853 
MURICACEA Rafinesque, 1815 
MURICIDAE Rafinesque, 1815 

(not specific about subfamilial divisions) 
BUCCINIDAE Rafinesque, 1815 
(and all other rachiglossate 
families usually attributed 
superfamilial status by other authors). 
MURICOIDEA Rafinesque, 1815 
MURICIDAE Rafinesque, 1815 
VASIDAE H. & A. Adams, 1853 
CORALLIOPHILIDAE Chenu, 1859 
THAIDIDAE Jousseaume, 1888 



{continued) 



PHYLOGENY OF RAPANINAE 



159 



TABLE 1 . {Continued) 



Ponder & Waren, 1 988 



MURICOIDEA Rafinesque, 1815 
MURICIDAE Rafinesque, 1815 
MURICINAE Rafinesque, 1815 

(incl. Trophoninae, Ocenebrinae, etc.) 
THAIDINAE Jousseaume, 1888 

(incl. Rapaninae) 
CORALLIOPHILINAE Chenu, 1859 
MOREINAE Stephenson, 1941 
7SARGANINAE Stephenson, 1923 



TABLE 2. Ranking of thaidine higher taxa since Thaididae, Jousseaume, 1888, by a selection of authors. 

Family Rank 
Thaididae: Medley, 1918; Iredale, 1937; Clench, 1947; Korobkov, 1955; Pchelintsev & Korobkov, 1960; 

Keen, 1964, 1971a, b; Strausz, 1966; Jung, 1969; Radwin & D'Attilio, 1971, 1972; Vokes, 1972; 

Golikov & Starobogatov, 1975; Petuch, 1982; Harasewych, 1984; Kensley, 1985; Kensley & Pether, 

1986. 
Thaisidae: Suter, 1909; Stewart, 1927; Iredale & McMichael, 1962; Powell, 1961; Miller, 1970. 
Thaidiidae: Atapattu, 1972. 
Thaidae: Cooke, 1919. 

Purpuridae: Cossmann, 1903; Lamy, 1928; Coomans, 1962; Settepassi, 1971; Abbott, 1974. 
Concholepadidae: Perrier, 1897. 

Subfamily Rank 
Thaidinae: Cernohorsky, 1969; Beu, 1970; Emerson & Cernohorsky, 1973; Rosewater, 1975; Rehder, 

1980; Emerson & D'Attilio, 1981; Fujioka, 1985a. 
Thaisidinae: Kuroda & Habe, 1971. 
Drupinae: Wenz, 1941; Hertlein, 1960. 
Purpurinae: Baker, 1895. 

No Separate Rank 
Muricidae: Thiele, 1929; Demond, 1957; Barnard, 1959; Arakawa, 1962, 1964, 1965; D. W. Taylor & 
Sohl, 1962; Habe, 1964; Wu, 1965a, 1968, 1973, 1985; Habe & Kosuge, 1966; Maes, 1966, 1967; 
Powell, 1979. 



but have not investigated causes for this phe- 
nomenon (Colton, 1916, 1922; Kincaid, 1957; 
Berry & Crothers, 1968, 1970; Cowell & 
Crothers, 1970; Hoxnnark, 1970, 1971; Lar- 
gen, 1971; Crothers, 1973; Spight, 1973). 

If environment causes high intraspecific 
variation in shell morphology among muricids 
(and gastropods generally), it is not surprising 
that convergence in shell shape is a fre- 
quently recognized phenomenon (Ponder, 
1973; Davis, 1979; Signer, 1982; Harasew- 
ych, 1984; Vermeij & Zipser, 1986). Similar 
shell shapes may have evolved in response 
to similar environmental pressures. Thus, 
convergence in shell shape is probably the 
major underlying cause of existing taxonomic 
controversies within the Thaldidae/nae of au- 
thors and other murlcid groups. 

Of course, shell morphology can be deceiv- 
ing in another way as well: major differences 
in external shell morphology may obscure a 
possibly close phylogenetic relationship, 



which may — as does convergence — result in 
paraphyletic and/or polyphyletic groups. 

Radular morphology is the second-most uti- 
lized criterion on which to base taxonomic 
groups within Thaididae/nae, although radular 
characters are almost always used in conjunc- 
tion with shell characters (Cooke, 1919; 
Thiele, 1929; Clench, 1947; Arakawa, 1962, 
1964; Wu, 1968, 1985; Radwin & D'Attilio, 
1971, 1972, 1976; Emerson & Cernohorsky, 
1973; Bändel, 1984; Harasewych, 1984; Fu- 
jioka, 1985a). Troschel (1866-1893) used 
radular characters as the sole basis for his 
classification. 

Although radular characters in Thaididae/ 
пае of authors and other molluscan groups 
have been applied cautiously, no studies cor- 
relating radular morphology and diet existed 
until recently (Kool, 1986, 1987) to indicate 
whether this caution is justified. Radular char- 
acters have often been regarded as, at most, 
moderately indicative of relationship, in par- 



160 



KOOL 



ticular, when radular characters do not show 
congruence with shell shape. In this case, 
adaptationist explanations usually have been 
invoked in which radular morphology is 
postulated to have evolved as a direct re- 
sponse to dietary habits (Arakawa, 1 964 [Ra- 
paninae, sensu Kool]; Wu, 1965a [Rapani- 
nae, sensu Kool]; Powell, 1964 [Turhdae]; 
see also Kool, 1987). Several authors (Ar- 
akawa, 1962; Radwin & D'Attilio, 1972; Wu, 
1973; Fujioka, 1985a) have mentioned intra- 
generic differences in rapanine radulae. How- 
ever, the generic determinations and bound- 
aries used by these authors were based on 
shell morphology, and may therefore have 
been invalid. A detailed investigation by Kool 

(1987) showed that radular morphology in 
Thaididae/nae of authors does not reflect 
diet, but is indicative of relationships as de- 
termined by anatomy [i.e. "soft" anatomy (not 
including radula)]. 

However, some degree of caution is nec- 
essary. Sexual dimorphism in radulae has 
been reported for several genera in Rapani- 
nae: Nassa (Maes, 1966), Drupella Thiele, 
1925 (Arakawa, 1957; Fujioka, 1982), Morula 
(Fujioka, 1984), and Cronia (Fujioka, 1984). 
Furthermore, Fujioka (1985a) and DiSalvo 

(1988) observed ontogenetic changes in the 
radulae of several rapanine species, and Fu- 
jioka (1985b) also found seasonal aberrant 
radular formation to occur in two species of 
rapanines. Anatomical [not including radula] 
data are probably the most reliable morpho- 
logical data in reflecting phylogenetic relation- 
ships. Molluscan anatomists, such as Ponder 
(1973), Ho.ubhck (1978), and Davis (1979), 
have demonstrated the importance of ana- 
tomical characters as opposed to characters 
derived from external shell morphology in es- 
tablishing phylogenetic relationships. It is now 
generally agreed that a reliable phylogenetic 
explanation for any molluscan group must be 
based on a robust set of anatomical data. 

In contrast to the vast amount of descrip- 
tive data on shell morphology, and the infor- 
mation available on radular morphology, very 
little is known about the anatomy of represen- 
tatives of the Rapaninae and other muricid 
groups. Most anatomical studies are either 
superficial or focus on specific aspects of 
anatomy, such as the alimentary system 
(Righi, 1964; Wu, 1965a; Rajalakshmi Bhanu 
et al., 1980, 1981a, b; Carriker, 1981; Shya- 
masundari et al., 1985), and the reproductive 
system (Houston, 1976; Gallardo & Garrido, 
1989; Shlakshmi, 1991). Haller (1888) pre- 



sented an exceptionally detailed anatomical 
study of Concholepas concholepas (Bru- 
guière, 1789), and anatomical information is 
also available on Nucella (Fretter, 1941; A. 
Graham, 1941, 1949; Fretter & Graham, 
1962; Harasewych, 1984; Houston, 1976) 
and Acanthina (Wu, 1985). Several anatomi- 
cal reports exist on a variety of other muricid 
taxa, e.g. Urosalpinx Stimpson, 1865 (Car- 
riker, 1943, 1955; Carriker et al., 1972), Tro- 
phon Montfort, 1810 (Harasewych, 1984; E. 
H. Smith, 1967), and Rapana (Chukhchin, 
1970). 

Recently, the topic of "imposex" (the occur- 
rence of male characters in female snails, in 
particular a penis) in especially Muricidae has 
received much attention (Feral, 1976; Hall & 
Feng, 1976; Bryan et al., 1986, 1987;Gibbs& 
Bryan, 1986; Gibbs et al., 1987; Bright & Ellis, 
1990). The occurrence of imposex is highly 
correlated with environmental pollution by the 
chemical tributyltin. 

Another non-conchological feature that 
may be of use in unraveling evolutionary re- 
lationships among rapanines is egg capsule 
morphology. Aspects of egg capsule mor- 
phology of muricids have been treated by a 
variety of authors (Lebour, 1936, 1945; Amio, 
1957; Ganaros, 1958; D'Asaro, 1966, 1970a, 
b, 1986; Gohar & Eisaway, 1967; Bändel, 
1976; Tirmizi & Zehra, 1983). The most com- 
prehensive work on muricid egg capsules to 
date is by D'Asaro (1991), who provided de- 
tailed descriptions for the egg capsule mor- 
phology of a wide variety of muricids. 

Hypothesis and Objectives 

The working hypothesis of this study is that 
a classification resulting from cladistic analy- 
ses of a data set of primarily anatomical char- 
acters will differ from all previous classifica- 
tions and will be far more reliable than those 
based primarily on shell shape. The new clas- 
sification will reveal which names and taxo- 
nomic levels should be applied to one or more 
monophyletic groups. 

This first comprehensive comparative ana- 
tomical study will establish a testable infer- 
ence of phylogeny and a classification not only 
for those taxa traditionally included in Thaid- 
idae/nae of authors, but also for other muricid 
groups. Furthermore, this study will provide a 
framework onto which other taxa can be added 
more easily, after limits of different taxa are set 
by identification of synapomorphies. 



PHYLOGENY OF RAPANINAE 



161 



MATERIALS AND METHODS 
Compilation of Morphological Data 

Eighteen type species (herein referred to as: 
Concholepas concholepas (Bruguière, 1789), 
Cronia amygdala (Kiener, 1835), Cymia tecta 
(Wood, 1828), Dicathais órbita (Gmelin, 
1 791 ), Drupa morum Röding, 1 798, Haustrum 
haustorium (ОглеПп, 1791), Mancinella 
alouina (Röding, 1798), Morula uva (Röding, 
1798), Nassa serta (Bruguière, 1789), Neora- 
pana muricata (Broderip, 1832), Nucella lapil- 
lus (Linnaeus, 1 758), Pinaxia versicolor {Gra'^, 
1839), Purpura pérsica (Linnaeus, 1758), 
Stramonita haemastoma (Linnaeus, 1767), 
Thais nodosa (Linnaeus, 1758), Tribulus 
planospira (Lamarck, 1822), Vasula melones 
(Duelos, 1832), and Vexilla vexilla (Gmelin, 
1791)], and one "non-type species," Plicopur- 
pura patula {Linnaeus, 1758), representing 19 
genera usually placed in Thaididae/nae of au- 
thors, were studied in detail (Appendix 1 ). Two 
additional type species, also usually placed in 
Thaididae/nae of authors, Acanthina mon- 
odon (Pallas, 1774) and Trochia cingulata 
(Linnaeus, 1771), were examined on a rela- 
tively low number of characters. Furthermore, 
one taxon belonging to Rapaninae of authors, 
Rapana rapiformis (Born, 1778), one taxon 
belonging to Muricinae, Muricanthus ful- 
vescens (Sowerby, 1841), and one taxon in- 
certae sedis, Porrería belcheri (Hinds, 1844), 
were examined in detail. A fossil taxon incer- 
tae sedis, Ecphora cf. quadricostata (Say, 
1824) was examined also. Twenty-four of the 
above-mentioned taxa (excluding Ecphora) 
were subjected to cladistic analyses per- 
formed with Hennig86 (Farris, copyright 
1988). 

The database used to address questions of 
muricid phylogeny consisted primarily of ana- 
tomical data, but also included data from pro- 
toconch, operculum, radula, and shell ultra- 
structure. Anatomical variation within and 
among species was determined by dissection 
of a variety of specimens. Most voucher spec- 
imens are deposited in the National Museum 
of Natural History, Smithsonian Institution, 
Washington, D.O., U.S.A.; others are at the 
Academy of Natural Sciences, Philadelphia, 
Pennsylvania, U.S. A, or at the Museum of 
Comparative Zoology, Harvard University. 

Field work was done at many geographical 
locations throughout the Pacific and western 
Atlantic oceans, and in numerous habitats 
(rocky intertidal, mangrove forest, etc.), allow- 



ing a variety of ecological and behavioral ob- 
servations (spawning, feeding, etc.). When 
possible, egg capsules of rapanine species 
were collected during spawning. 

Both living and preserved specimens were 
used in this study. Living animals were main- 
tained in tanks of running sea water and ob- 
served periodically before being sacrificed. 
Prior to dissection, animals were de-shelled 
using a vice and observed under a dissecting 
microscope. In some cases, a 7.5% isotonic 
solution of magnesium chloride was used to 
relax the animals. Snails were dissected while 
alive to observe color patterns, gross anat- 
omy, and variability within an individual in 
structures such as the penial flagellum. Dis- 
sected animals were fixed in 10% formalin 
and preserved in 70-75% ethyl alcohol for 
further study. Preserved museum material 
was frequently in poor condition due to incom- 
plete penetration of preservative, and pro- 
vided limited information. 

Some morphological data were obtained 
from histological sections and study of critical- 
point dried specimens using the Hitachi S-570 
and Cambridge Stereoscan (100 and 250 MK 
II) scanning electron microscopes at the U.S. 
National Museum of Natural History. Palliai 
gonoducts were embedded in paraffin and 
sectioned at 7, 10, or 15 micrometers, de- 
pending on the size of the animal and the 
degree of detail desired. They were normally 
stained using triple PAS stain, although other 
stains (Masson's and Cason's) were occa- 
sionally used. 

Morphological analyses resulted in a data 
matrix consisting of 18 characters and 64 
character states. These characters were de- 
rived from the protoconch, shell ultrastruc- 
ture, operculum, mantle cavity complex 
(ctenidium, osphradium), female and male re- 
productive and alimentary systems, and rad- 
ula, and were used in cladistic analyses. 

Because shell morphology is known to be 
under the influence of environmental selec- 
tion pressures, the only shell characters used 
in cladistic analyses are those taken from lar- 
val shells and shell ultrastructure (see below). 

Description of Characters 

A variety of philosophies advocate different 
ways of choosing and justifying characters for 
reconstructing phylogeny. For example, some 
authors argue that characters displaying par- 
allelism and convergence should not be used 
in phylogenetic analyses. However, parallel- 



162 



KOOL 



isms and convergences are only recognizable 
after analyzing the branching patterns of phy- 
logenetic trees. Once a convergence be- 
tween two synapomorphic states is recog- 
nized, the character in question should not be 
automatically discarded, because this results 
in loss of information and may in addition, 
lead to a reduction in resolution within or 
among branches of the tree. A case of ho- 
moplasy should be re-evaluated and re-di- 
vided into character states (perhaps with the 
tree topology based on other characters as a 
guide). Parallelisms and convergences, after 
all, provide valuable information about the 
manner in which different organisms adapt to 
possibly similar circumstances, and they indi- 
cate areas requiring more detailed study. Fur- 
thermore, those character states of a (par- 
tially homoplasious) character that are not 
homoplasious and occur only once in a 
branching sequence are additional synapo- 
morphies and add to the resolution of the cla- 
dogram. 

Convergence in external shell morphology 
is known to exist. Judging from the variety of 
taxonomic arrangements based on shell mor- 
phology and the results from the cladistic anal- 
yses presented herein, characters taken from 
the external morphology of the teleoconch 
have been very misleading in assessing rela- 
tionship (Kool, 1988b). For these reasons, I 
have not included characters from external 
shell morphology in the cladistic analyses 
presented here. However, with the obtained 
branching pattern as a frame work, "good" 
(i.e. reflecting relationship) characters from 
the external shell morphology can be identified 
and could be added in future analyses. 

Most of the characters used in the phylo- 
genetic analysis are anatomical characters 
(reproductive system, alimentary system [ex- 
cluding radula], mantle cavity, etc). The other 
characters were taken from shell ultrastruc- 
ture, protoconch, operculum, and radula. 

To avoid duplication of figures (often only 
differing in only minor details [e.g. length of 
accessory salivary glands]), general lay-outs 
of different morphological systems with their 
individual structures and organs are illus- 
trated in Figures 3 (whole animals, reproduc- 
tive systems, alimentary system, mantle cav- 
ity organs), 4 (female reproductive system), 5 
(male reproductive system), and 6 (rachidian 
tooth). 

I made no a priori assumptions about the 
validity of characters in reconstructing phylog- 
eny and used all characters analyzed. For ex- 



ample, a variety of authors has expressed 
suspicion about the phylogenetic significance 
of radular morphology in a variety of groups 
(Kool, 1987). Diet is often suspected to be the 
driving force behind the evolution of radular 
characters. Although this may be true for 
some groups, the matter has never been thor- 
oughly investigated. I have shown elsewhere 
(Kool, 1987) that there is very little correlation 
between radular morphology and dietary hab- 
its in rapanine gastropods, but that high cor- 
relation is present between relationship 
(based on anatomy) and radular morphology. 
The results of this study (Kool, 1987) show 
that inclusion of radular characters is indeed 
justified for reconstructing phylogeny and that 
characters, which were often assumed a pri- 
orHo be under the influence of environmental 
factors and thus non-reflective of relationship, 
need testing against an independent data set 
(reflecting phylogeny) prior to unqualified 
prejudice against that particular suite of char- 
acters. 

The list of characters follows the sequence 
in which these characters are described in 
each species. 

Protoconch: Most of the protoconchs (and, 
where possible, the embryonic shell) were de- 
scribed from scanning electron micrographs, 
but a few descriptions were based on pub- 
lished drawings. Whorls, seen in apical view, 
were counted from the end of protoconch II 
spiraling inward. In some cases, the exact 
number of whorls could not be given due to 
poor preservation of the protoconch. Most 
data were derived from SEM micrographs of a 
single specimen, but other data from light mi- 
croscopy were frequently added. 

Characters: 

1 . Number of whorls and sculpture 

(a) multispiral (more than two and a 
quarter whorls); sculptured (e.g. 
Figs. 10D, 19C) 

(b) paucispiral (fewer than two whorls); 
smooth (e.g. Figs. 15C, 28C) 

(c) multispiral; smooth (e.g. Fig. 9C) 

(d) paucispiral; sculptured (e.g. Fig. 
23D) 

2. Transition into teleoconch 

(a) outward-flaring lip (e.g. Fig. 10D, E) 

(b) smooth transition (e.g. Fig. 26B, C) 

Shell Morphology: Shell measurements 
(height and width) were taken from large adult 
specimens in the USNM collection and do not 



PHYLOGENY OF RAPANINAE 



163 



represent maximum sizes. Height was mea- 
sured from the apex (tip of earliest whorl) to 
the most distal point of the anterior siphonal 
canal, or apertural lip, whichever yielded the 
highest number; aperture height includes the 
apertural lip. Shell width is defined here as the 
distance between the apertural lip (or close to 
it to avoid inclusion of spines or knobs) and 
the other side of the body whorl (not including 
spines or knobs). Percentage measurements 
of the body whorl and aperture are relative to 
total shell height, and percentage is rounded 
off to a whole number and a multiple of five. A 
consistently present incision in the posterior- 
most portion of the apertural lip was consid- 
ered as a posterior siphonal canal. A large 
number of museum lots was examined for 
color descriptions. 

Shell ultrastructural data were obtained us- 
ing scanning electron microscopy. Shell frag- 
ments of at least two specimens (depending 
on ambiguity or difficulty of interpretation of 
data) provided data on the kinds and combi- 
nations of shell layers. Fragments were cut 
out from the central region of the apertural lip 
with a diamond saw at some distance (about 
one-half of a whorl away) from the apertural 
lip edge, and broken collabrally. The fracture 
surfaces were observed and the different lay- 
ers identified. In some cases, the fracture sur- 
face was polished; this process facilitates rec- 
ognition of the different layers. 

In the descriptions of the ultrastructure of 
the shells, the layers are listed in consecutive 
order beginning with the innermost layer (ad- 
jacent to the animal). All layers described for 
any of the taxa treated herein are present in, 
for example, Purpura; Figure 18F can be 
used for general reference. An approximate 
range for the thickness of each layer is given 
relative to all shell layers combined. 

Characters: 

3. Calcific outer layer 

(a) absent (e.g. Figs. 13F, 24D) 

(b) present, thick > 25% of total (e.g. 
Figs. 15G, 26F) 

(c) present, thin < 20% of total (e.g. 
Figs. 80, 25D, 18F, e) 

4. 45° innermost aragonitic layer 

(a) absent (e.g. Fig. 25D) 

(b) present (e.g. Figs. 14E, 110, H, 18F, 
a) 

Operculum: In the descriptions of the oper- 
cular morphology, terms such as "bracket- 
shaped" and "arch-shaped" are used to de- 



scribe the shape of growth lines on both the 
outside surface, referred to as "free surface" 
and the inside surface, referred to as "at- 
tached surface." In older specimens, the 
bracket-shaped growth lines often lose their 
horizontal portions, resulting in growth lines 
running straight from top to bottom. The terms 
"left side" and "right side" (on either surface) 
are used in reference to an operculum with its 
apex situated upward (the apex actually being 
the posteriormost end of the operculum). The 
vertical position of the nucleus varies among 
taxa; the description "in center right" denotes 
a nucleus located midway on an imaginary 
line running from the apex to the lower end of 
the operculum. The size of the operculum cor- 
responds closely to the size of the shell aper- 
ture (given in shell description), unless noted 
otherwise. No notation of color and color pat- 
terns was made; color often reflects the age 
and thickness of the operculum and varies 
among individuals of the same species. 

Character: 
5. Morphology of operculum (shape, posi- 
tion of nucleus) 

(a) operculum ovate; terminal nucleus in 
lower right (Fig. 1A) 

(b) operculum D-shaped, upper end 
rounded; lateral nucleus in lower 
right (Fig. ID) 

(c) operculum D-shaped, tapered at 
lower end, and with S-shaped left 
(adjacent to columella) edge; lateral 
nucleus in lower right (Fig. 1 F) 

(d) operculum inverted tear-shaped; lat- 
eral nucleus in lower right (Fig. IB) 

(e) operculum D-shaped; lateral nucleus 
in center right (Fig. 1С) 

(f) operculum ovate-elongate, tapered 
at lower end; lateral nucleus in upper 
right (Fig. IE) 

Foot and Mantle Cavity: The anatomical de- 
scriptions are given as follows. In a first para- 
graph, most of the external characteristics are 
listed (coloration and morphology of tentacles 
[e.g. Fig. 3B, t], head-foot region, kidney [e.g. 
Fig. 38, C, k], hypobranchial gland [e.g. Fig. 
3B, C, hg], nephhdial gland [anteriorly of the 
kidney; usually visible on left side of live ani- 
mals]), followed by data on accessory boring 
organ and (for females) ventral pedal gland 
(e.g. Fig. 4A, B, abo, pg). 

The second and third paragraphs treat the 
osphradial and ctenidial morphologies (e.g. 
Fig. 3D, OS, ct). The length of the osphradium 



164 



KOOL 




g Г 



nu 



\ 

nu 

в 











FIG. 1 . Morphologies of muricid opercula, showing free surface (facing to the outside) and attached surface 
(facing inside), respectively. A, Muricanthus fulvescens. B, Rapana rapiformis. C, Thais nodosa. D, Forreria 
belcheri. E, Vexilla vexillum. F, Cronia amygdala; gr, growth lines; nu, nucleus; h, rim of callus. 



is measured fronn the postehormost end (Fig. 
3D, pos) to the anteriormost tip (Fig. 3D, ant) 
along the central axis separating both pectins. 
Similarly, the length of the ctenidium (gill) is 
measured along the ctenidial efferent blood 
vessel (Fig. 3D, cv). Absolute measurements 
are not given; only relative size (osphradium 



vs. ctenidium). The term "symmetrical in 
shape" is used rather than "symmetrical" be- 
cause although there often is symmetry along 
the longitudinal (central) axis in the overall 
shape of both pectins, in none of the taxa 
examined was the number of osphradial 
lamellae equal between the left and the right 



PHYLOGENY OF RAPANINAE 



165 




FIG. 2. Rod structures located in hypobranchial gland of Morula nodulosa. A, surface of hypobranchial gland 
with rod structure in center (arrow), SEM (bar = 20 (xm). B, cross section through rod structure, SEM (bar 
= 2 |xm). 



pectin; the right pectin (directly adjacent to the 
ctenidium) consistently bears (about 25%) 
more lamellae than the left one. The general 
shape of the ctenidium (usually elongate half- 
moon-shaped [Fig. 3D, ct], or D-shaped) and 
osphradium (usually ovate-elongate) with left 
(Fig. 3D, los) and right pectins, is variable at 
least within some taxa, as is the morphology 
and number of individual lamellae of both or- 
gans. The edge of the ctenidial lamella adja- 
cent and parallel to the support rod is referred 
to as the ventral edge (Fig. 3D, Ir); the other 
free edge as the lateral edge (Fig. 3D, le). The 
size of the ctenidial lamellae is described as a 
relation between width and depth (the latter 
term was chosen over "height" because the 
lamellae in situ hang down). 

Characters: 

6. Rodlike structures in hypobranchial gland 

(a) absent 

(b) present (Fig. 2A, B) 

7. Ventral pedal gland and accessory bor- 
ing organ 

(a) sharing one duct (e.g. Fig. 4B) 

(b) having separate ducts (e.g. Fig. 4A) 

(c) accessory boring organ absent 

8. Osphradial length relative to ctenidial 
length 

(a) osphradial length less than one-half 
ctenidial length 

(b) osphradial length at least one-half 
ctenidial length 

Female Reproductive System: The repro- 
ductive organs of the female palliai gonoduct 
are listed and described in the same order in 



which the dissections were made (anterior to 
posterior), beginning with the vaginal opening 
and the vagina (Fig. 4C, v), followed by the 
bursa copulatrix (Fig. 4C, be), capsule gland 
with left and right lobes (Figs. 3E, eg, 4C, Ic, 
re), ventral channel (Fig. 4C, vc), ovi-sperm 
duct (connecting capsule gland with albumen 
gland; Fig. 4E-H, osd), ingesting gland (Fig. 
3E, ig), albumen gland (with or without pos- 
terior seminal receptacles; Figs. 3E, ag, 4E- 
H), and the gonad (Fig. 3E, ov). 

Characters: 
9. Bursa copulatrix 

(a) sacklike, separate from lumen of 
capsule gland (Fig. 4C, be) 

(b) continuous with capsule gland (Fig. 
4D, be) 

10. Posterior seminal receptacles around al- 
bumen gland 

(a) absent (Fig. 4F, G) 

(b) 1-3 with duet branching off ovi- 
sperm duct (Fig. 4E, psr) 

(c) many (usually at least 7 or 8) (Fig. 
4H, psr) 

1 1 . Morphology of albumen gland 

(a) diverticulum of oviduct (Fig. 4F) 

(b) arch-shaped, elongate (Fig. 4G) 

(c) staff-shaped (Fig. 4E) 

(d) omega-shaped, roundish (Fig. 4H) 

Male Reproductive System: Descriptions of 
the organs of the male reproductive system 
follow the same format as those of the female 
system. The penis (Figs. 3B, C, p, 5A-F, I) is 
described, followed by the penial vas defer- 
ens (Fig. 5A, B, D, pvd), cephalic vas defer- 



166 



KOOL 




PHYLOGENY OF RAPANINAE 



167 



ens, prostate (Figs. 3B, pr, 5G, H), prostate 
duct (Fig. 3B, pd), seminal vesicles (Fig. 3C, 
vs) and the testis (Fig. SB, С te). The term 
"large" as referred to penis size is to be taken 
relative to tentacle size; a penis which mea- 
sures more than twice the size of the tenta- 
cles is referred to as "large." Changes in pe- 
nial morphology within the same individual 
are a common phenomenon in most species. 
The penis can be extended or condensed, 
and its shape can thus be altered. In a relaxed 
state, however, the penial shape does not 
vary much among individuals of the same 
species. Penial variation in living specimens 
facilitated evaluation of penial shapes in pre- 
served specimens. 

Characters: 

12. Morphology of penis 

(a) elongate, gradually tapering (Fig. 5A) 

(b) straight to lightly curved, with 
pseudo-papilla (Fig. 5B) 

(c) strongly recurved, with large side 
lobe (Fig. 5E, I) 

(d) strongly recurved, club-shaped (Fig. 
5F) 

(e) strongly recurved, with flagellate 
pseudo-papilla (Fig. 5D) 

(f) slightly recurved, gradually thinning 
to flagellate morphology (Fig. 5C) 

13. Morphology of penial vas deferens 

(a) duct well developed, semi-closed by 
interlocking lateral ridges (Fig. 5A, 
pvd) 

(b) duct minute, open, adjacent to pos- 
terior edge of penis 

(c) duct minute, semi-closed by loosely 
overlapping ventral and dorsal sides 
of penis; adjacent to posterior edge 
of penis (Fig. 5B, pvd) 



(d) coiling duct within a larger duct (duct- 
within-a-duct system) (Fig. 5D, pvd) 
14. Morphology of vas deferens of prostate 
(palliai vas deferens) 

(a) open to mantle cavity in posterior 
portion (Fig. 5H, prv) 

(b) closed to mantle cavity (Fig. 5G, prv) 

Alimentary System: The alimentary system 
(exclusive of radula) is treated in two para- 
graphs; one for structures of the anterior por- 
tion of the alimentary system (Fig. 3F), such 
as the proboscis (pb), accessory salivary 
glands (ra, la), salivary glands (Isg), valve of 
Leiblein (vL), mid-esophageal glandular folds 
[on portion of mid-esophagus between nerve 
ring (nr) and duct to gland of Leiblein; meg], 
gland of Leiblein (gL), the other for the pos- 
terior structures, such as the stomach (e.g. 
Fig. 3G, H), rectal gland (Fig. 3C, E, rg), and 
anal opening. Size references for the acces- 
sory salivary glands are relative to shell 
height (see below). Size of the proboscis is 
given relative to the size of the gland of 
Leiblein ("large" translates into almost equal 
in size to gland of Leiblein). The portion of the 
mid-esophagus containing glandular folds is 
referred to as "long" when it stretches from 
the nerve ring to the duct to the gland of 
Leiblein. The posterior blind duct of the gland 
of Leiblein is either long (duct longer than 
one-half of length of gland), or short (duct 
shorter than one-fourth of length of gland); no 
intermediate values were found. 

The posterior portion of the stomach is 
herein considered that portion with is directly 
adjacent to the esophagus; a lateral exten- 
sion means an extension of the central mixing 
area of the stomach. The term "stomach 
typhlosole" (Fig. 3C, stt) refers to the foldlike 



FIG. 3. Anatomy of selected rapanines and their organs. A-C, E, whole animals removed from shell. A, 
Plicopurpura patula, male with mantle skirt cut longitudinally to expose head ( x 1 ). B, Morula uva, male, left 
side (xlO). C, Morula uva, male, right side (xlO). D, ctenidium and osphradium of Morula uva, with 
lamellae ( x 15). E, Morula uva, female, right side ( x 10). F, generalized representation of anterior portion of 
alimentary tract found in rapanines. G-H, morphologies of muricid stomach and intestine, inside views. G, 
Nucella lapillus. H, Muricanthus fulvescens; ag, albumen gland; ant, anterior end; eg, capsule gland; cm, 
columellar muscle; cme, cut mantle edge; ct, ctenidium; cv, ctenidial efferent vessel; dd, digestive diverticula; 
dg, digestive gland; dgL, posterior duct of gland of Leiblein; f, foot; g, gonad; gL, gland of Leiblein; h, heart; 
hg, hypobranchial gland; ig, ingesting gland; in, intestine; int, intestinal typhlosole; is, incurrent siphon; к 
kidney; la, left accessory salivary gland; le, lateral edge; los, left osphradial pectin; Ir, lamellar support rod 
(ventral edge); Isg, left lobe of salivary gland; m, mouth; ma, mantle; meg, mid-esophageal folds; nr, nerve 
ring; 0, operculum; od, oviduct; ov, ovary; p, penis; pb, proboscis; pd, prostate duct; pet, longitudinal folds 
of the posterior esophagus; pes, posterior esophagus; pos, posterior end; pr, prostate; psr, posterior seminal 
receptacles; r, rectum; ra, right accessory salivary gland; rg, rectal gland; s, sole; sf, folds on gastric wall of 
stomach; si, siphon; st, stomach; stt, stomach typhlosole; t, tentacle; ta, terminal ampulla; te, testes; vL, 
valve of Leiblein; vm, visceral mass; vs, vesícula seminalis. 



168 



KOOL 



abo 



abo 




FIG. 4. Morphologies of muricid female reproductive structures. A, B, sagittal cross sections through anterior 
foot of female, viewed from right. A, ventral pedal gland and accessory boring organ separate (e.g. Nucella 
lapillus). B, ventral pedal gland and accessory boring organ combined (e.g. Thais nodosa). C, schematic 
representation of anterior palliai gonoduct of female non-thaidine muricid (e.g. Nucella lapillus), viewed from 
left, with cross section. D, schematic representation of anterior palliai gonoduct of female thaidine (e.g. 
Plicopurpura patula), viewed from left, with cross section. E-H, albumen gland morphologies in Muricidae, 
viewed from right. E, e.g. Morula uva. F, e.g. Muricantus fulvescens. G, e.g. Nucella lapillus. H, e.g. 
Stramonita haemastoma; abo, accessory boring organ; ag, albumen gland; be, bursa copulatrix; Ic, left lobe 
of capsule gland; od, oviduct; osd, ovi-sperm duct; pg, ventral pedal gland; psr, posterior seminal recepta- 
cles; re, right lobe of capsule gland; tf, transverse furrow; v, vagina; vc, ventral channel; vf, ventral flange. 



PHYLOGENY OF RAPANINAE 



169 




FIG. 5. Morphologies of muricid male reproductive structures. A-F, I, penial rлorphoíogies in Muricidae A 
Muricanthus fulvescens, with cross section. B, Nucella lapillus, with cross section. C, Nassa serta D Thais 
nodosa, with cross section. E, Morula uva. F, Cymia tecta. I, Cronia amygdala. G-H, scherлatic represen- 
tation of prostate morphologies in Muricidae, with cross section. G, e.g. Thais nodosa. H, e.g. Nucella 
lapillus; po, penial opening; prv, prostate vas deferens; pvd, penial vas deferens; si, side lobe. 



170 



KOOL 



structure which usually borders the posterior 
mixing area and can be continuous with what 
Fretter & Graham (1962) refer to as "typhlo- 
sole 2," located in the intestine (e.g. Fig. 3G, 

int). 

Characters: 

1 5. Length of accessory salivary glands 

(a) right gland minute, nearly undetect- 
able; left one absent 

(b) both left and right glands very long 
(nearly one-half of shell height) 

(c) both glands short to medium (less 
than one-quarter of shell height; Fig. 
3F, la, ra) 

(d) both glands absent 

(e) right gland very long (nearly one-half 
of shell height); left gland absent 

16. Length of posterior blind duct of gland of 
Leiblein 

(a) duct at least one-half of length of 
gland (Fig. 3F, dgL) 

(b) duct shorter than one-half (usually 
less than one-fourth) of length of 
gland 

Radula: Radulae (2-6 per species) were dis- 
sected from living and preserved animals, 
cleaned in potassium hydroxide, and exam- 
ined using scanning electron microscopy. For 
the sake of consistency, only scanning elec- 
tron micrographs were used for analyzing 
radular structures. Four micrographs were 
taken of the central portion of each radular 
ribbon. The first two micrographs (one includ- 
ing lateral teeth, one excluding lateral teeth) 
were taken perpendicular to the radular rib- 
bon. The radula was then tilted laterally to an 
angle of 40° to obtain a lateral view of the 
morphology of the cusps and denticles on the 
rachidian tooth. Finally, the radula was tilted 
laterally to an angle of about 85° to examine 
the edge of the rachidian tooth and the an- 
gles, sizes and locations of its cusps and den- 
ticles, in an area from which the lateral teeth 
had been cut away with a surgical knife. 

The morphology of the radula is described 
starting with the rachidian tooth (Fig. 6B), fol- 
lowed by the lateral teeth. The cusps (three or 
five) on the rachidian are described beginning 
with the central cusp (Fig. 6B, cc), followed by 
the inner lateral denticle (ild), lateral cusp (Ic), 
the marginal area (ma), marginal denticles 
(d), and marginal cusp (mc). The marginal 
area is defined as the more or less horizontal 
area on the outside of the lateral cusp, ex- 
tending to — if present — the marginal cusp. 



Size of lateral cusps is given relative to size of 
central cusp ("nearly equal" translates into 
75% or more of central cusp length). The po- 
sition of the inner denticle(s) is against the 
base of the inner edge of the lateral cusp, 
unless noted otherwise. Size of inner lateral 
denticle is relative to lateral cusp. Size of lat- 
eral teeth is given relative to rachidian width. 
An approximate range of the length of the rad- 
ular ribbon is given, where available, relative 
to shell height. 

Characters: 

1 7. Orientation of marginal cusp of rachidian 
tooth 

(a) marginal cusp absent or in same 
plane as lateral cusp (and marginal 
denticles, if present) (e.g. Fig. 7F) 

(b) marginal cusp in different plane than 
lateral cusp (forming an approxi- 
mately 75° angle), on antero-posteri- 
orly widened base (e.g. Fig. 15E, F) 

18. Morphology of rachidian tooth 

(a) marginal area and cusps absent; in- 
ner lateral denticle small, free from 
and between lateral and central 
cusps; lateral cusps nearly equal in 
length to central cusp (Fig. 24E) 

(b) marginal area and cusps absent; in- 
ner lateral denticle larger than lateral 
cusp, free from and between lateral 
and central cusps; lateral cusps 
nearly equal in length to central cusp 
(Fig. 11D) 

(c) marginal area absent, marginal 
cusps small; one or more small inner 
lateral denticles; lateral cusps nearly 
equal in length to central cusp (Figs. 
15E, F, 26D, E) 

(d) marginal area absent, marginal 
cusps small; inner lateral denticle 
small; central cusp much longer than 
lateral cusps and reclining, forming 
angle with them (Fig. 8H) 

(e) marginal area wide, smooth, mar- 
ginal cusps absent; inner lateral den- 
ticle small, free from but adjacent to 
lateral cusp; central cusp much 
longer than lateral cusps (e.g. Fig. 
8D) 

(f ) marginal area and cusps absent; sev- 
eral faint inner lateral denticles; lat- 
eral cusps nearly equal in length to 
central cusp (Fig. 25C, E) 

(g) marginal area absent, marginal 
cusps small; one or more inner lat- 
eral denticles; lateral cusps nearly 



PHYLOGENY OF RAPANINAE 



171 



equal in length to central cusp (e.g. 
Fig. 7F) 
(h) marginal area wide, with multiple 
denticles and small marginal cusps; 
inner lateral denticle small; lateral 
cusps nearly equal in length to cen- 
tral cusp (e.g. Fig. 18D) 
(i) marginal area and cusps absent; in- 
ner lateral denticle absent; central 
cusp much longer than lateral cusps 
(Fig. 111) 
(j) short marginal area with small mar- 
ginal cusps; inner lateral denticle 
small or absent; lateral cusps nearly 
equal in length to central cusp which 
is wide at base (e.g. Fig. 22E) 
Note: both Neorapana and Tribulus have 
larger, wider central cusps relative to the lat- 
eral cusps. These lateral cusps (those of 
Neorapana without inner lateral denticle) are 
bent somewhat sideways, which, in the case 
of Neorapana, resulted in the loss of any mar- 
ginal area. If the Hennig86 program would al- 
low for scoring of more than ten character 
states, a separate character state would have 
been assigned to Neorapana and Tribulus. 
However, overall morphology of the rachidian 
tooth strongly suggests homology among the 
four genera scored for with "(j)." 

Taxa which could not be scored due to a 
limited number of character-state entries in 
Hennig86 are mentioned below. They are all 
synapomorphic and thus would not have in- 
fluenced the topology of the tree. 

Nassa — similar to "(i)," but female specimens 
with small free-standing inner lateral den- 
ticle (Fig. 13G). 

Plicopurpura — similar to "(i)," but with slit in 
central cusp (Fig. 17E). 

Vexilla — similar to "(i)," but with base of cen- 
tral cusp nearly as wide as rachidian (Fig. 
23C). 

Phylogenetic Analysis 

Data pertaining to the reproductive and al- 
imentary systems, mantle cavity, radula, 
operculum, protoconch, and shell ultrastruc- 
ture were subjected to cladistic analyses. No 
data were derived from external shell mor- 
phology. 

Three steps were necessary to commence 
the cladistic analysis: (1) identification of po- 
tentially homologous characters; (2) division 
of each individual character into character 
states; and (3) polarization of character 



states, for which the outgroup method was 
applied. Homology was regarded as two very 
similar structures with similar location and 
function. 

The outgroup method was used to deter- 
mine the ancestral state of each character. 
The outgroup criterion is based on the as- 
sumption that character states present in the 
sister group (outgroup) and the group studied 
(ingroup) is the plesiomorphic or "primitive" 
condition (Hennig, 1966). The outgroup 
method was thus used to determine the "zero 
state." Use of an outgroup further allows ap- 
plication of the parsimony criterion; it is as- 
sumed that the hypothesis based on the low- 
est number of character changes ("steps") is 
the best solution for the available data, be- 
cause it explains the data in the most eco- 
nomical way and is thus based on the small- 
est number of assumptions made about the 
evolutionary process (Farris, 1979, 1982; Lip- 
scomb 1984). 

The muricine Muricanthus fulvescens 
(Sowerby, 1841) (also known as Murex ful- 
vescens and Hexaplex fulvescens) appeared 
suitable to serve as outgroup in the cladistic 
analysis for several reasons: (1) the Murici- 
nae is a sister group of the Rapaninae; (2) 
many live-collected and well-preserved spec- 
imens were available to provide all data nec- 
essary for anatomical studies; (3) most of the 
structures and characters derived from rapa- 
nine anatomy are present also in l\/luricanthus 
Swainson, 1840, although their "states" may 
be very different. 

The character states of multi-state charac- 
ters were left unordered, because no realistic 
assumptions about character state evolution 
could be made a priori. For example, ontoge- 
netic criteria could not be applied because 
only adult specimens of the type species were 
available. 

Only a few continuous (or quantitative) 
characters (e.g. size, or numbers) were used 
due to the arbitrary nature of "cut-off points." 
Qualitative characters were more easily di- 
vided into character states. 

The Hennig86 cladistic computer package 
was used to derive a repeatable, testable, rel- 
atively objective, most parsimonious, and 
most informative hypothesis with the avail- 
able database. The results herein were very 
similar to previous results (Kool, 1989) ob- 
tained with a slightly different data set using 
other computer packages (PAUP [Swofford, 
copyright 1985]; and PHYSYS [Farris & Mick- 
evich, copyright 1985]). 



172 



KOOL 





В 



FIG. 6. А, egg capsule of Cymia teda, apical view. B, schematic representation of composite rachidian tooth 
of muricids (frontal view); cc, central cusp; d, denticles on marginal area; eh, exit hole; ild, inner lateral 
denticle; Ic, lateral cusp; ma, marginal area; mc, marginal cusp; st, stalk. 



One of the advantages of using cladistics is 
the predictive power of the obtained trees. To 
test the robustness and predictive power of 
the phylogeny proposed herein, a few taxa 
were examined on those characters which re- 
vealed themselves during early stages of the 
analysis as unique synapomorphies for cer- 
tain clades. This "spot checking" allowed for 
unambiguous placement of taxa for which 
only limited data were available. Based on the 
cladistic analyses, limits were set for each 
group after synapomorphies for each group 
were identified. 

Cladograms never yield a final solution for 
evolutionary relationships among taxa, and 
the phylogeny presented herein should be 
taken only as a testable hypothesis for the 
evolutionary history of the Rapaninae (as de- 
fined herein) and its position in the Muricidae. 



RESULTS 

The genera formerly included in Thaididae/ 
пае are treated in alphabetical order. A chro- 
nologically arranged synonymy of each genus 
is given, including author, date, page, and in- 
formation on the type species. The type spe- 



cies of the valid genus name is given, fol- 
lowed by the correct binomen and a 
synonymy. New combinations are omitted. A 
"Remarks" section provides for a short dis- 
cussion of the taxonomic history and place- 
ment by different authors (usually including 
Cossmann, 1903, Thiele, 1929, and Wenz, 
1941) of the genus and (type) species. 

Different aspects of morphology (proto- 
conch, teleoconch, anatomy, radula, egg cap- 
sules) of each species are described in detail, 
followed by (if available) data on the biology 
(ecology and geographic distribution) of each 
taxon. Not treated is the fossil history of each 
taxon, as most of this information, given by 
Thiele (1929) and Wenz (1941), is out of date 
and highly suspect (see "Congruence with 
Fossil Record"). 

A less detailed treatment is provided for 
Muricanthus fulvescens, used as outgroup. 
Porrería beleben, a taxon incertae sedis, and 
Rapana rapiformis. I should mention that it 
was not known initially that Rapana was 
monophyletic with most members of Thaidi- 
nae of authors. Only limited data were avail- 
able on the taxa Acanthina monodon and Tro- 
chia cingulata (both usually included in 
Thaididae/nae of authors), but the available 



PHYLOGENY OF RAPANINAE 



173 



data were used in the cladistic analysis, par- 
tially to test for character robustness. 

Although many of the descriptions of the 
anatomy of the type species are based on 
dissections of living animals, most observa- 
tions were based on preserved specimens. 
Illustrations of anatomy are schematic in or- 
der to standardize and elucidate the shared 
morphologies rather than to show individual 
idiosyncrasies due to intraspecific variation. 

Descriptions of taxa traditionally grouped in 
Thaididae/nae of authors 

Genus Concholepas Lamarck, 1801 
(Fig. 7A-F) 

Concholepas Lamarck, 1801 : 69. 

Concholepa Deshayes, 1830: 256 (error for 
Concholepas). 

Conchopatella Herrmannsen, 1847: 291 (in- 
troduced in synonymy). 

Type Species: Concholepas peruviana La- 
marck, 1801, by monotypy, = Concholepas 
concholepas (Bruguière, 1789); synonym: 
Buccinum concholepas Bruguière, 1 789. 

Remarks: Lamarck introduced the species C. 
peruviana as type of the genus Concholepas 
and may have considered it a different spe- 
cies from Buccinum concholepas Bruguière. 
More likely, he renamed it without regard for 
priority to avoid tautonomy (an unpopular no- 
menclatural procedure at the time). However, 
these two taxa are synonymous, and the ear- 
lier name, С concholepas, has priority. The 
genus has one living and several fossil repre- 
sentatives (Vokes, 1972; Kensley, 1985). 
Malier (1888) gave an extensive description of 
the anatomy of this species, emphasizing the 
nervous system. 

Shell: Protoconch (Fig. 7C, D) squat (wider 
than high), smooth, of 2.5-3 whorls, with 
slightly impressed suture, and with outward- 
flaring lip (DiSalvo, 1988) (eroded from fig- 
ured specimen) and sinusigeral notch. Teleo- 
conch (Fig. 7A, B) of 2-3 whorls and 
exhibiting high rate of whorl expansion. Adult 
shell up to about 125 mm in height, 95 mm in 
width. Suture slightly impressed, nearly 
canaliculate on final whorl. Body whorl and 
aperture reaching beyond apex. Body whorl 
robust, rounded "patelliform," sculptured with 
11-13 spiral, lamellose cords, with one spiral 
thread in interspaces. Lamellose sculpture 
most common in juveniles, often persisting in 



adults. Aperture oval, extending beyond shell 
spire. Apertural lip with crenate edge, corre- 
sponding to spiral cords. Anterior siphonal ca- 
nal short, wide and open; posterior siphonal 
canal absent. Columella flat or somewhat 
concave, continuous with apertural lip, and 
reaching from beyond apex to anterior sipho- 
nal canal. Siphonal fasciole similar to axial 
ribs but more elevated. One or two labial 
toothlike structures adjacent to siphonal fas- 
ciole on apertural lip. Shell uniformly dark red- 
dish brown; aperture white; columella white, 
occasionally with light brown areas. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (15-20%); aragonitic layer with 
crystal planes onented parallel to growing 
edge (15-20%); calcific layer (60-70%) (Fig. 
7E). 

Operculum: D-shaped (about one-third size 
of aperture), with lateral nucleus in center 
right (compare Fig. 1С). Free surface with 
bracket-shaped growth lines; attached sur- 
face usually with one bracket-shaped growth 
line and with callused, glazed rim (about 35- 
40% of opercular width) on left. 

Anatomy: (based on preserved animals 
only): Cephalic tentacles long and wide. Ten- 
tacles a uniform, medium brown. Head-foot 
and sole of foot mottled dark brown. Mantle 
edge smooth and following shell contour, with 
very long brown incurrent siphon. Pinkish and 
yellow hypobranchial gland positioned within 
thin, upright, lateral epithelial ridges. Kidney 
dull caramel brown. Pedal gland in females 
well developed, with accessory boring organ 
in proximal portion. 

Osphradial length less than one-fourth 
ctenidial length; osphradial width less than 
ctenidial width. Osphradium symmetrical in 
shape along lateral and longitudinal axes. Os- 
phradial lamellae attached along small por- 
tion of their base. 

Antehormost portion of ctenidium straight, 
extending farther anteriorly than osphradium. 
Anterior ctenidial lamellae distinctly wider 
than deep; posterior lamellae deeper than 
wide. Lateral and ventral edges of ctenidial 
lamellae concave, lateral edge occasionally 
straight. Distal tips of ctenidial support rods 
extending beyond lateral edge as papillate 
projections. 

Vaginal opening situated on tapering ante- 
rior end of palliai oviduct and located directly 
beneath anal opening. Bursa copulathx an 



174 



KOOL 






FIG. 7. Concholepas concholepas. A, shell (67 mm), apertural view. B, shell (67 mm), abapertural view. C, 
protoconch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, shell 
ultrastructure, SEM (bar = 50 ц.т). F, radula, SEM. 



PHYLOGENY OF RAPANINAE 



175 



open chamber in interior vagina and open to 
anterior portion of capsule gland. Posterior 
part of palliai oviduct with ventral sperm chan- 
nel consisting of two ventrally located flanges 
each facing one another and perpendicular to 
capsule gland lobes. Ventral channel in ante- 
rior portion of palliai oviduct very small. In- 
gesting gland located between capsule gland 
and albumen gland, continuing on left side of 
albumen gland, comprising many small, inter- 
connected chambers, and lined with dark yel- 
low epithelium. Seminal receptacles on dorsal 
periphery of albumen gland small, elongate- 
oval, white. Albumen gland small, omega- 
shaped. The external lay-out of the female 
reproductive system in this species and the 
species following hereafter is superficially 
similar to that shown in Figure 3E and in Kool 
{1988b, fig. 3C). 

Penis dorso-ventrally flattened, wide, with 
large folds along posterior border (in young 
individual examined), or angular (in older 
ones). Penial shaft curved, with long and thin 
flagellate tip. Vas deferens as thin duct- 
within-a-duct system (Fig. 5D, pvd) occupying 
about one-fifth of penial width. Prostate gland 
solid, white, adjacent to spongy, white, rectal 
wall. Duct of prostate closed off from mantle 
cavity but sometimes visible through epithe- 
lium. Seminal vesicles comprised of small, 
white or orange outpocketings. Testicular 
duct following periphery of gonad. 

Proboscis whitish, thinner than width of 
gland of Leiblein. Paired accessory salivary 
glands of equal length, long, worm-shaped, 
slightly less than one-half of shell height. Left 
accessory gland located under and separate 
from salivary gland but loosely connected to it 
by many strings of connective tissue. Right 
accessory gland ventral to proboscis and 
slightly ventral to salivary glands. Salivary 
glands cream brown, consisting of many 
small portions, larger in mass than accessory 
salivary glands, partially located between 
gland of Leiblein and proboscis, or partially 
between nerves emanating from nerve ring. 
Valve of Leiblein elongate, irregularly shaped, 
surrounded by salivary glands but not at- 
tached to them. Salivary ducts attached some 
distance from valve of Leiblein; valve sepa- 
rated from nerve ring. Portion of mid-esopha- 
gus with glandular folds long; folds well de- 
veloped. Major portion of posterior 
esophagus free and looped along side of 
gland of Leiblein, but small area of posterior 
esophagus closely attached to it. Gland of 
Leiblein coiled counterclockwise, forming two 



folds, brown grey, of hard consistency, with 
thick outer covering with "interwoven" strings 
of connective tissue. Blind posterior duct of 
gland of Leiblein more than one-half length of 
gland itself. The lay-out of the alimentary sys- 
tem in this and the following species is similar 
to that shown in Figure 3F. 

Stomach buried in digestive gland, with 
center projecting deep into visceral mass, and 
with lateral extension. Interior epithelium 
forms many (about 20) distinct folds, the larg- 
est central and perpendicular to typhlosole. 
Folds on right portion of stomach curve into 
central fold; folds of left portion perpendicular 
to stomach typhlosole. One diverticulum 
present. Stomach typhlosole well developed, 
continuing onto stomach wall. Intestinal 
typhlosole wide and shallow. Several minute 
folds on right side of intestinal typhlosole in 
intestinal groove. Anal opening distinct, wide, 
varying from thin- to thick-walled. Anal papilla 
poorly developed. Rectal gland well devel- 
oped, green, adjacent to entire length of pal- 
liai gonoduct. 

Radula: Central cusp on rachldian with wide, 
somewhat constricted base (Fig. 7F); lateral 
cusps pointing outward; inner lateral denticle 
located on base of lateral cusp and one-half 
its length; several knobby outer denticles on 
base of lateral cusp; marginal cusp very 
small. Lateral teeth long, thin, wide-based, 
nearly total rachidian width. 

Egg Capsules: Large, about 20 mm in height 
(Gallardo, 1973), elongate, slightly curving, 
with undulating surface, and resting on short, 
thin stalk, about 1 mm in length. Capsules 
arranged in clusters, close to one another, 
each containing up to 13,000 eggs (Gallardo, 
1979). Eggs up to 158-160 jxm in diameter 
(Gallardo, 1979). 

Ecology: Concholepas concholepas is one of 
the few rapanine gastropods of direct eco- 
nomic importance and of culinary value to 
man, who is this species' major predator on 
the west coast of South Amenca (Castilla & 
Duran, 1985). Thus, a substantial number of 
papers have been published on its ecology 
(Gallardo, 1973, 1979, 1980; Gallardo & Per- 
ron, 1982; Castilla & Cancino, 1976; Castilla 
& Duran, 1985). Egg capsules are usually 
found in the sublittoral zone; planktotrophic 
veliger larvae hatch from them probably 
spending up to several weeks in the plankton 



176 



KOOL 



before settlement (Gallardo, 1979). Adults 
live and spawn in the rocky intertidal zone, 
where they feed on barnacles and mussels 
(Gallardo, 1979; Kool, 1987). DuBois et al. 
(1 980) reported specimens living at a depth of 
40 m. DiSalvo (1988) describes the veliger 
stages. Beu (1970) suggested that fossil rel- 
atives of the Recent species lived in much 
deeper waters. 

Distribution: Eastern Pacific, from central 
Peru to southern Chile (Beu, 1970; Disalvo, 
1988). 

Genus Cronia H. & A. Adams, 1853 
(Fig. 8A-D) 

Cronia H. & A. Adams, 1853: 128 (as a sub- 
genus of Purpura). 

Type Species: Purpura amygdala Kiener, 
1835, by monotypy, = Cronia amygdala 
(Kiener, 1835); synonyms: 7Buccinum avel- 
lana Reeve, 1846; IPurpura aurantiaca Hom- 
bron & Jacquinot, 1852; ^Purpura pseu- 
damygdala Hedley, 1 902. 

Remarks: The taxon Cronia was introduced 
by H. & A. Adams (1853: 128) as a subgenus 
of Purpura "Aldrovandus" [correct author: 
Bruguière, 1789), with one species listed. 
Cossmann (1 903: 68) placed Cronia as a sec- 
tion under the subgenus Polytropalicus Rov- 
ereto, 1899, genus Purpura. Dall (1909: 50) 
allotted Cronia to Thais. Thiele (1929: 294) 
and Wenz (1941: 1113) placed Cronia as a 
subgenus under Drupa. Fujioka (1985a) and 
Cernohorsky (1982, 1983) used Cronia as a 
full genus. 

The species described below resembles 
Kiener's (1835) figures of Purpura amygdala 
but appears more similar to Medley's (1902) 
figures of Purpura pseudamygdala. Kiener's 
figures of Purpura amygdala bear more re- 
semblance to the figures of Medley's Purpura 
pseudamygdala than to Mombron & Jacqui- 
not's figures of Purpura aurantiaca, which is 
most likely conspecific with Buccinum avel- 
lana Reeve, 1846. I strongly suspect all four 
"species" to be geographical or ecopheno- 
typic variants of the same species. Cooke 
(1919: 107) explained that Medley restricted 
the amygdala form to the southeast coast of 
Australia, and introduced Cronia pseu- 
damygdala for the "species" from Queens- 
land. Closer examination of the types, ranges 
of variation, and the anatomy of these four 



"morphs" is necessary before definite state- 
ments on this matter can be made. 

Shell: Protoconch tall, conical, smooth, of 
about four adpressed whorls, and with out- 
ward-flaring lip and sinusigeral notch (Medley, 
1902: pi. 29, figs. 4-5). Teleoconch (Fig. 8A, 
B) of 6-7 adpressed, high-spired, fusiform 
whorls. Adult shell up to about 30 mm (includ- 
ing 3 mm siphonal canal) in height and 1 5 mm 
in width. Body whorl about 65-70% of shell 
height, rounded, heavily sculptured with five 
pronounced spiral cords, one of them directly 
below suture, and with 3-4 fine, delicately 
lamellose spiral lines at regular intervals from 
one another, between each pair of major spi- 
ral cords. Spiral cords bear 8-9 knobs at reg- 
ular intervals towards the base. Knobs 
aligned to form about nine thick axial ribs per 
whorl. Aperture elongate, about 60% of shell 
height. Apertural lip slightly thickened, with 
seven denticles. Anterior siphonal canal well 
developed, short, deep and semi-closed; pos- 
terior siphonal canal absent. Siphonal fasci- 
cle well developed, delicately lamellose, free 
from callus on lower columella. Columella 
with heavy callus deposition. Shell grey 
brown; knobs on axial ribs white or light 
brown; aperture light orange brown, espe- 
cially on columella and lip edge. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (25-30%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (70-75%) (Fig. 8C). 

Operculum: D-shaped, with S-shaped left 
edge, tapered at lower end, with lateral nu- 
cleus in lower right (compare Fig. IF). Free 
surface with staff-shaped growth lines; at- 
tached surface with about 5-7 arch- and 
bracket-shaped growth lines and with cal- 
lused, glazed rim (about 30-40% of opercu- 
lar width) on left. 

Anatomy (based on living and preserved ma- 
terial): Mead-foot and siphon brown with 
green, yellow and white specks, cephalic ten- 
tacles long. Mantle edge smooth, following 
aperture contour; incurrent siphon long. My- 
pobranchial gland large, perpendicular to 
mantle wall, with small, thin, black, rodlike 
structures embedded in it (compare Fig. 2A, 
B). Kidney green in males, brown in females. 
Nephridial gland green in females. Pedal 
gland as simple duct, combined with large ac- 
cessory boring organ (Fig. 4B). 
Osphradial length equal to or slightly more 



PHYLOGENY OF RAPANINAE 



177 




FIG. 8. A-D, Crania amygdala. A, shell (28 mm), apertural view. B, shell (28 mm), abapertural view. C, shell 
ultrastructure, SEM (bar = 0.10 mm). D, radula, SEM (bar = 30 ¡im). E-H, Cymia tecta. E, shell (55 mm), 
apertural view. F, shell (55 mm), abapertural view. G, shell ultrastructure, polished surface, SEM (bar = 0.30 
mm). H, radula, SEM (bar = 45 jim). 



178 



KOOL 



than one-half ctenidial length; osphradium 
and ctenldium about equal In width. Osphra- 
dium symmetrical in shape along lateral axis; 
right pectin wider than left. Osphradial lamel- 
lae attached along more than one-half of their 
base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior and posterior ctenidial lamellae 
wider than deep. Lateral and ventral edges of 
ctenidial lamellae usually sharply concave. 
Distal tips of well-developed ctenidial support 
rods not extending beyond lateral edge. 

Vaginal opening round, situated on distal 
end of short, attached tube and located below 
and posterior to anal opening. Bursa copula- 
trix a dorso-ventral slit, continuous with cap- 
sule gland and ventral channel (Fig. 4D). Ven- 
tral sperm channel formed by large rolled 
flange originating from ventral epithelium and 
lying below both capsule gland lobes. Duct 
from ovi-sperm duct enters mushroom- 
shaped, orange-brown (in living animals) in- 
gesting gland, which lies between capsule 
gland and albumen gland (compare Fig. 3E). 
Second duct branching off ovi-sperm duct 
more posteriorly, forming single, elongated, 
grey seminal receptacle lying above albumen 
gland (compare Fig. 3E, psr). Sperm appar- 
ent from iridescence in receptacle. Albumen 
gland omega-shaped, usually turned side- 
ways, lying on posterior portion. 

Penis with large side lobe (Fig. 51), basi- 
cally oval in cross section, with bulbous tip on 
long thin shaft. Triangular muscular side lobe 
(Fig. 51, si) pointing toward head and tenta- 
cles. Penial duct as duct-within-a-duct system 
(compare Fig. 5D, pvd) occupying about one- 
fourth of penial width. Testicular duct brown 
and seminal vesicles weakly developed. 
Prostate duct closed to mantle cavity. Pros- 
tate solid, light brown (in living animals), di- 
rectly adjacent to rectum, without layer of con- 
nective tissue separating both structures. 
Testis brown. 

Proboscis much wider than width of gland 
of Leiblein. Paired accessory salivary glands 
both equally short (2 mm), stubby, much less 
than half of shell height. Left accessory sali- 
vary gland embedded in intertwined salivary 
glands; right accessory salivary gland sepa- 
rated from salivary glands. Salivary glands in- 
tertwined, light orange, larger than accessory 
salivary glands and with granular appear- 
ance. Valve of Leiblein elongate, free from 
salivary glands. Salivary gland ducts attached 
to esophagus at base of valve of Leiblein, 



which lies adjacent to nerve ring. Glandular 
folds on mid-esophagus resulting in slight 
thickening of mid-esophagus. Duct between 
esophagus and gland of Leiblein poorly de- 
veloped. Posterior esophagus separated from 
gland of Leiblein along entire length. Gland of 
Leiblein coiled counterclockwise, forming two 
folds, flat, creamy brown, soft, appearing 
granular. Posterior blind duct about one-half 
of length of gland of Leiblein. 

Stomach very large, with large sorting area 
having weak lines arranged randomly. Large, 
posteriorly located, unciliated area and two 
digestive diverticula present. Intestinal typhlo- 
sole well developed, but stomach typhlosole 
variable in size. Anal opening inconspicuous; 
anal gland poorly developed, running dorsally 
along less than one-half of palliai gonoduct. 

Radula: Ribbon length about 20% of shell 
height (Fig. 8D). Rachidian with long, thin 
central cusp; lateral cusp with convex inner 
edge and smooth, concave outer edge; inner 
lateral denticle small, separate from lateral 
cusp; large, smooth, horizontal area between 
lateral cusp and edge of rachidian. Lateral 
teeth curved, smooth, slightly larger than half 
the rachidian width. 

Egg Capsules: Unknown. 

Ecology: Specimens of Cronia amygdala 
were collected on an intertidal offshore coral 
reef fringing a mangrove forest at Cockle Bay, 
Magnetic Island, Queensland, Australia. Abe 
(1983) reported Cronia margariticola (Brod- 
erip) to be a scavenger, preying upon a wide 
variety of food items, or feeding on eggs of 
Thais clavigera (Küster). 

Distribution: West, north, and east Australia 
(Eisenberg, 1981) and Pacific Ocean (Cerno- 
horsky, 1972). 

Genus Cymia Mörch, 1860 
(Fig. 8E-H) 

Cuma Humphrey, 1797 (rejected work). 

Cuma Swainson, 1840: 87 {non Milne-Ed- 
wards, 1 828) [type: Cuma sulcata Swain- 
son, 1840, by monotypy, ^ Cymia tecta 
(Wood, 1828)]. 

Cymia Mörch, 1860: 97 (replacement name 
for Cuma Swainson; as subgenus of Ra- 
pana). 

Cumopsis Rovereto, 1899: 105 (unnecessary 
replacement name for Cuma Swainson; 
as subgenus of Purpura). 

Cyma Rovereto, 1899: 105 (error for Cymia). 



PHYLOGENY OF RAPANINAE 



179 



Type Species: Cuma sulcata Swainson, 
1840, by monotypy, = Cymia tecta (Wood, 
1828); synonyms: Buccinum tectum Wood, 
1828; Purpura angulifera Duelos, 1832. 

Remarks: Swainson (1840: 87) placed Cuma 
in the subfamily Pyrulinae, family Turbinell- 
idae, and included only one species, Cuma 
sulcata. Mörch introduced Cymia as a re- 
placement name for Cuma Swainson, which 
was pre-occupied, and placed it under Ra- 
paría. Rovereto (1899: 105) synonymized 
Cuma Swainson with his replacement name, 
Cumopsis, allotted it to Purpura, and did not 
list any other species to be included in this 
subgenus. Korobkov (1955: 299) considered 
Cymia to be a subgenus of Thais. 

Shell: Protoconch unknown. (Protoconch of 
Cymia brightoniana Maury "a little more than 
one whorl" [Jung, 1969: 497]). Teleoconch 
(Fig. 8E, F) heavy, fusiform, oblong, of 7-8 
adpressed whorls, with high spire and shallow 
suture. Early whorls sculptured with spiral, in- 
cised lines. Adult shell up to about 70 mm in 
height, 50 mm in width. Body whorl about 65- 
70% of shell height, sculptured with 8-10 
large, spinose knobs on periphery of very pro- 
nounced, centrally located shoulder of each 
whorl. Suture adjacent to and following lower 
contours of these knobs. Twenty-five to 30 
deeply incised spiral grooves on body whorl, 
several crossing knobs. Aperture moderately 
large, about 70% of shell height. Apertural lip 
thin, reflecting pattern caused by incised 
lines. Anterior siphonal canal short, wide, 
open; posterior siphonal canal poorly devel- 
oped or absent. Heavy, central fold on col- 
umella. Siphonal fascicle curving, well devel- 
oped, only partially covered by moderate 
callus layer on fasciole. Shell white, yellow, 
grey-brown; aperture and columella white to 
very light orange. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (30-35%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (30-40%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (15-20%); calcific layer (15-20%) (Fig. 
8G). 

Operculum: D-shaped, with strongly concave 
left edge (to accommodate fold on shell fas- 
ciole), with lateral nucleus at center right 
(compare Fig. 1С). Free surface with bracket- 
shaped growth lines Indented in center; at- 
tached surface with about 4-6 arch- and 



bracket-shaped growth lines and with cal- 
lused, glazed rim (about 30-35% of opercular 
width) on left. 

Anatomy (based on preserved animals only): 
Cephalic tentacles short, stubby, with black 
blotches. Head-foot mottled black. Mantle 
edge crenate (following aperture lip contour). 
Incurrent siphon protruding farther than man- 
tle edge. Sole of foot with many, primarily lat- 
erally crossing, shallow grooves, resulting in 
pustulate pattern. Pedal gland large, sepa- 
rated from accessory boring organ, but adja- 
cent to it. Small lateral folds on wall of distal 
part of pedal gland; proximal part smooth. Ac- 
cessory boring organ large, compact, cham- 
ber-shaped, adjacent to pedal gland in fe- 
males. 

Osphradial length less than one-half ctenid- 
ial length; osphradium and ctenidium about 
equal in width. Osphradium symmetrical in 
shape along longitudinal axis; usually wider 
anteriorly. Osphradial lamellae attached 
along large portion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium, or osphradium extending slightly farther 
anteriorly. Anterior ctenidial lamellae wider 
than deep; posterior lamellae deeper than 
wide. Lateral and ventral edges of ctenidial 
lamellae variable in shape. Distal tips of 
ctenidial support rods extending beyond lat- 
eral edge as papillalike projections. 

Vaginal opening elongated, located directly 
below anal opening. Bursa copulatrix be- 
tween vaginal opening and capsule gland. 
Vertical flange large, folded, emanating from 
dorsal wall of bursa. Flange thin, straight, ver- 
tical, folded at tip prior to entering capsule 
gland. Bursa copulatrix continuous with ante- 
rior part of capsule gland. Flange minute, 
folded at 45° angle in most of capsule gland. 
Large second bursa between capsule gland 
and small albumen gland of the omega- or 
arch-shaped type. Ingesting gland with single 
chamber. 

Penis (Fig. 5F) large, thick, strongly re- 
curved, angular in cross section, with terminal 
papilla. Penial vas deferens tubular, about 
one-third of penis width. Cephalic vas defer- 
ens poorly developed. Prostate gland round 
in cross section, clearly separated from rectal 
wall, and with prostate duct closed off from 
mantle cavity. Posterior sperm storage area 
small but elongate, running horizontally on 
border line of gonad and digestive gland, dor- 
sal to prostate. 



180 



KOOL 



Proboscis muscular, thick, half as wide as 
gland of Leiblein. Paired accessory salivary 
glands very long, thin, of equal length, more 
than one-half of shell height. Right accessory 
salivary gland in dorsal right anterior corner of 
buccal cavity; left gland intertwined with sali- 
vary glands between proboscis and gland of 
Leiblein. Salivary gland mass dorsal, much 
smaller than accessory salivary glands. Valve 
of Leiblein elongate, free from salivary gland 
mass, adjacent to nerve ring. Salivary gland 
ducts attached to anterior portion of esopha- 
gus directly anterior to valve of Leiblein. Mid- 
esophageal folds indiscernible. Nerve ring 
adjacent to thin, long duct joining esophagus 
and gland of Leiblein. Posterior esophagus 
adjacent to lower left of gland of Leiblein. 
Gland of Leiblein spiral, forming two folds ori- 
ented antero-posteriorly, dark brown, of hard 
consistency. Posterior blind duct approxi- 
mately one-half of length of gland of Leiblein, 
running into dorsal branch of the afferent re- 
nal vein but not reaching kidney. 

Stomach U-shaped, but with large posterior 
widening. Sorting area with 10-15 folds ex- 
tending over only half its surface. Sorting area 
adjacent to intestinal typhlosole with minute 
folds and ridges parallel to it. Two digestive 
diverticula present. Intestinal typhlosole large. 
Rectum embedded in spongy tissue. Anal pa- 
pilla covering anal opening. Rectal gland long 
and thin; anal opening well developed. 

Radula: Ribbon length about 25% of shell 
height (Fig. 8H). Rachidian tooth with narrow 
central cusp; central cusp reclining, thus 
pointing in different direction than lateral 
cusp; inner lateral denticle nearly united with 
lateral cusp, which thus appears very wide; 
outer edge of lateral cusp straight, without 
denticulatlon; area between lateral cusp and 
edge of rachidian narrow, without denticles; 
wide marginal cusp pointing forward and par- 
allel to lateral extension on rachidian base. 
Lateral teeth smooth, about three-fourths of 
rachidian width. 

Egg Capsules: About 6 mm in height, ele- 
vated on wide stalk 1 mm long (Fig. 6A). Cap- 
sule vase-shaped, with oval, flat top; one side 
more elevated than other (normally continu- 
ing gradually in top layer of capsule); exit hole 
central, oval, located at slightly horizontal tip 
of capsule. All capsules appearing to be in- 
terconnected with basal membrane. Egg cap- 
sules examined (ANSP 355766) deposited on 
free side of operculum. 



Ecology: Specimens were found living on in- 
tertidal rocks on mud flats, but also on mud 
among mangrove roots. 

Distribution: Eastern Pacific, from Costa 
Rica to Ecuador (Keen, 1971b). 

Genus Dicathais Iredale, 1936 
(Fig. 9A-F) 

Dicathais Iredale, 1936: 325. 

Type Species: Buccinum órbita Gmelin, 
1791, by original designation, = Dicathais ór- 
bita (Gmelin, 1791); synonyms: Buccinum 
succinctum Martyn, 1784 (non-binominal); 
Purpura textilosa Lamarck, 1816; Purpura 
scalaris Menke, 1828 {non Schubert & Wag- 
ner, 1829); Purpura aegrota Reeve, 1846; Di- 
cathais vector Thornley, 1 952. 

Remarks: Iredale (1936: 325) removed suc- 
cincta from the genus Neothias Iredale, 1912 
(type: N. sm/Y/7/ Brazier, 1889, by original des- 
ignation; emended [unjustified] by Iredale to 
Neothais [1915: 473]), recognized órbita 
Gmelin as its valid name and designated Di- 
cathais órbita as type of Dicathais. Wenz 
(1941: 1124) synonymized Dicathais with 
Neothias. 

Controversy exists about the number of Di- 
cathais species. Cooke (1919: 97) observed 
differences between the radulae of "Thais 
succincta (= órbita)" and "T. textilosa." 
These and three other names {aegrota, sca- 
laris, and vector) are now considered to be 
geographical variants of one another (Phillips 
et al., 1973; Powell, 1979). The form here de- 
scribed is typical Dicathais órbita. 

Shell: Protoconch (Fig. 9C, D) low, smooth, 
of about four adpressed whorls, with outward- 
flaring lip and sinusigeral notch. Teleoconch 
(Fig. 9A, B) of 5-6 adpressed whorls. Adult 
shell up to about 85 mm in height, 60 mm in 
width. Spire less than one-third shell height. 
Suture impressed, canaliculate in final whorl. 
Penultimate and body whorls sculptured with 
eight, solid spiral cords and with many minute 
spiral, incised lines; body whorl about 85% of 
shell height. Aperture large, ovate, about 70- 
75% of shell height. Apertural lip thin, deeply 
scalloped due to spiral cords. Interior of aper- 
tural lip deeply grooved. Columella rounded 
or concave, with callus layer more pro- 
nounced toward posterior end. Anterior siph- 
onal canal a short but deep notch; posterior 
siphonal canal absent. Siphonal fasciole 
curved, about equally, or slightly more ele- . 



PHYLOGENY OF RAPANINAE 



181 






FIG. 9. Dicathais órbita. A, shell (58 mm), apertura! view. B, shell (58 mm), abapertural view. C, protoconch, 
side view, SEM (bar = 0.20 mm). D, protoconch, apical view, SEM (bar = 0.20 mm). E, shell ultrastructure, 
SEM (bar = 30 |xm). F, radula, SEM (bar = 40 |xm). 



182 



KOOL 



vated than spiral cords and adjacent to edge 
of lower, more heavily callused portion of col- 
umella. Shell white yellow to light brown (the 
latter especially in juveniles); aperture white 
yellow and columella white. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (25-50%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (20-25%); calcific layer (20-55%) 
(most pronounced at ribs) (Fig. 9E). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface usually with one bracket-shaped 
growth line and with callused, glazed rim 
(about 35-45% of opercular width) on left. 

Anatomy: (based on living and preserved an- 
imals): Cephalic tentacles long, uniform 
black. Head-foot mottled black. Mantle edge 
crenate, following contour line of spiral ribs. 
Incurrent siphon long, uniform dark brown to 
black. Accessory boring organ large, dorsal to 
pedal gland. 

Osphradial length about one-half ctenidial 
length; osphradial width between one-fourth 
and one-half ctenidial width. Osphradium 
symmetrical in shape along lateral and longi- 
tudinal axes. Osphradial lamellae attached 
along very small portion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior and posterior ctenidial lamellae 
usually wider than deep. Lateral and ventral 
edge of ctenidial lamellae concave. 

Vaginal opening a slit, situated on end of 
thick, tubular, partially detached, distal end of 
palliai gonoduct, and located directly below 
anal opening. Bursa copulatrix a channel, 
with flange, emanating from ventral lobe of 
capsule gland, forming oval, semi-closed ven- 
tral channel. Farther posteriorly ventral lobe 
of capsule gland absent and ventral channel 
located under right lobe of capsule gland. In- 
gesting gland on left of posterior part of cap- 
sule gland, with central and many smaller 
white-walled chambers; gland nearly as large 
as capsule gland, visible on exterior of body 
as large, dirty white granular mass. Row of 
pink, iridescent seminal receptacles on dorsal 
periphery of albumen gland. Albumen gland 
shape difficult to discern in adults; morphol- 
ogy in juveniles resembling both omega- 
shaped and arch-shaped types. Pseudo-pe- 
nis usually present, either as small appendix 



or equal in size and shape to penis of male 
specimens. 

Penis large, strongly recurved, with long 
flagelliform tip, occupying entire space be- 
tween tentacles and palliai complex, oval in 
cross section, with penial vas deferens as 
duct-within-a-duct system occupying nearly 
total width of penis. Cephalic vas deferens 
well developed, with internal, meandering tu- 
bular duct (similar to penial vas deferens). 
Prostate solid, dirty white, with accumulations 
of white granules. Prostate duct as closed 
tube adjacent to thin, cream-colored rectal 
wall. 

Proboscis very large, unpigmented, slightly 
less than, or equal in width to, gland of 
Leiblein. Paired accessory salivary glands 
long and thin, each adjacent to salivary 
glands; left accessory salivary gland some- 
times slightly longer than right one, and both 
about one-fourth of shell height. Salivary 
gland lobes inseparable; right portion under 
proboscis, extending to right anterior corner 
of buccal cavity. Valve of Leiblein elongate, 
irregularly shaped, separate from salivary 
gland mass. Salivary ducts attached to 
esophagus some distance from valve of 
Leiblein. Portion of mid-esophagus with glan- 
dular folds long, but poorly developed, except 
for short, widened section of mid-esophagus; 
widened section located adjacent to duct of 
gland of Leiblein. Duct between esophagus 
and gland of Leiblein thin. Posterior esopha- 
gus embedded in lower left side of gland of 
Leiblein. Gland of Leiblein spiral, forming two 
folds, of hard consistency, cream-colored, 
covered with thick, strawlike outer membrane. 
Posterior blind duct slightly less than length of 
gland of Leiblein. 

Stomach with large posterior projection. 
Ten to fifteen sizable folds on stomach wall. 
Two digestive diverticula present. Stomach 
typhlosole indistinct, poorly developed. Intes- 
tinal typhlosole thick, well developed. Long, 
wide rectal gland dark green. Rectal wall, at 
minute anal opening, pointing dorsally. 

Radula: Ribbon length about 40-45% of 
shell height (Fig. 9F). Central cusp on rachid- 
ian constricted at base; lateral cusps with 
large inner denticle attached midway; lateral 
cusps convex on inner edge, concave on 
outer edge; several faint, knobby, outer den- 
ticles on upper half of lateral cusp, and well- 
developed denticles at base; lateral cusp 
edge continuing down to well-developed mar- 
ginal cusp; rachidian base with lateral exten- 



PHYLOGENY OF RAPANINAE 



183 



sion. Lateral teeth nearly equal in length to 
rachidian width. 

Egg Capsules: About 9 mm in height, 6 mm 
wide, interconnected by basal membrane 
(Hedley, 1905). Dorsal surface of capsule 
elongate, rhomboidal, with elongate slit along 
longest axis. Hedley (1905) found egg cap- 
sules of "Purpura" succincta deposited on the 
ascidian Cynthia praeputialis Heller. Each 
capsule contains up to about 5,000 eggs 
(Phillips, 1969). 

Ecology: Dicathais órbita has been observed 
clinging tightly to rocks between large sea- 
squirts in the low intertidal zone of Botany 
Bay, Australia. It feeds on the barnacle Tes- 
seropora rosea (Kraus) and displays patterns 
of vertical migration between shelter areas 
(lower intertidal) and high concentrations of 
prey (high intertidal) (Fainweather, 1988). It 
has also been observed on rocks, partially 
buried in sand. The western Australian vari- 
ant Dicathais "aegrota" lives on limestone 
reef platforms where wave action is heavy 
(Phillips, 1969). It therefore seeks shelter in 
pockets and crevices, or partly buries itself (or 
gets buried) in the sand. Feeding usually oc- 
curs at high tide and at night (Phillips, 1969). 
Its varied prey consists mostly of mollusks 
(primarily Cronia "avellana") and malacostra- 
can crustaceans (Phillips, 1969). Large trem- 
atode parasites were present in several spec- 
imens I collected in Botany Bay (New South 
Wales, Australia), which had made these in- 
dividuals sterile. Phillips (1969) also found 
tremátodos in D. "aegrota." Some known 
predators of Dicathais are octopods, other Di- 
cathais individuals (at least under laboratory 
conditions), and perhaps crustaceans. Cronia 
"avellana" and Crustacea are known to feed 
on Dicathais egg capsules (Phillips, 1969). 

Distribution: Australia, Tasmania, Norfolk Is- 
land, Lord Howe Island, Kermadec Island, 
and New Zealand (Philips et al., 1 973; Powell, 
1979). 

Genus Drupa Röding, 1798 
(Fig. 10A-E) 

Drupa Röding, 1798: 55. 

Canrena Link, 1807: 126 [type: Murex neritoi- 
deus Linnaeus, 1 767 by subsequent des- 
ignation, Iredale, 1937: 256, = Drupa 
morum Röding, 1798, in partem]. 

Sistrum Montfort, 1810: 594 [type: Sistrum al- 
bum Montfort, 1810, by original designa- 



tion, = Murex ricinus Linnaeus, 1 758, = 
Drupa ricinus (Linnaeus, 1758)]. 

Ricinula Lamarck, 1816: 1, pi. 395 [type: 
Ricinula hórrida Lamarck, 1816, by sub- 
sequent designation. Children, 1823: 56 
(as Ricinula horida), = Drupa morum 
Röding, 1798]. 

Ricinulus Lamarck; Chenu, 1859: 174 (invalid 
emendation for Ricinula Lamarck). 

Ricimula A. A. Gould, 1855: 263 (error for 
Ricinula Lamarck). 

Ricinella Schumacher, 1817: 240 [type: Ri- 
cinella purpurata Schumacher, 1 81 7, by 
subsequent designation, Iredale, 1937: 
256, = Drupa rubusidaeus Röding, 
1798]. 

Pentadactylus Mörch, 1852: 87 [non 
Schultze, 1760, пес Gray, 1840] [type: 
Murex ricinus Linnaeus, 1 758, by subse- 
quent designation. Baker, 1895: 186, = 
Drupa ricinus (Linnaeus, 1758)]. 

Drupina Dal I, 1923: 303 [type: Ricinula digi- 
tata Lamarck, 1816, by original designa- 
tion, = Drupa grossularia Hading, 1798]. 

Type Species: Drupa morum Röding, 1798, 
by subsequent designation, Rovereto, 1899: 
105; synonyms: Nerita nodosa Linnaeus, 
1758 {in partem); Murex neritoideus Lin- 
naeus, 1767 {in partem); Ricinula globosa 
Martyn, 1784 (non-binominal); Ricinula hórr- 
ida Lamarck, 1816; Ricinella violácea Schu- 
macher, 1817; Ricinula horida Lamarck, Chil- 
dren, 1823 (error for hórrida). 

Remarks: Cossmann (1903: 68) considered 
Ricinula { = Drupa) a full genus. Thiele (1 929: 
295) subdivided the genus Drupa into the 
subgenera Drupa (sections Drupa, Morula, 
and Drupina), Cronia (sections Cronia, 
Morulina, Usilla, Muricodrupa), Phrygio- 
murex, Maculitriton, and Drupella. Wenz 
(1941: 1113) included the subgenera Drupa, 
Morulina, Usilla, Cronia, Muricodrupa, Phry- 
giomurex, Maculitriton, Morula, and Drupella 
in Drupa. Keen (1971b: 553) placed Drupa in 
the Drupinae. Emerson & Cernohorsky 
(1973) divided Drupa into the subgenera 
Drupa, Ricinella and Drupina on the basis of 
shell morphology. 

Shell: Protoconch similar to that of Drupa 
grossularia (Fig. 10D, E), tall, conical, consist- 
ing of at least 3.5 adpressed whorls [exact 
count could not be made from available spec- 
imen], with small subsutural plicae, intercon- 
nected by three thin spiral ridges, but other- 



184 



KOOL 




FIG. 10. A-C, Drupa morum. A, shell (35 mm), apertural view. B, shell (33 mm), abapertural view. C, radula, 
SEM (bar = 25 pirn). D-E, Drupa grossularia. D, protoconch, side view, SEM (bar = 0.10 mm). E, proto- 
conch, apical view, SEM (bar = 0.10 mm). 



wise smooth, and with outward-flaring lip; si- 
nusigeral notch covered by teleoconch. Te- 
leoconch (Fig. 10A, B) globose but flat on ap- 
ertural side, low-spired, of 3-4 adpressed 
whorls. Adult shell up to about 40 mm in 
height, 35 mm in width. Body whorl about 85- 
90% of shell height, dome-shaped, robust, 
thick, and sculptured with five rows of spiral 
bands of seven heavy, sometimes spinelike, 
axially arranged knobs. Largest knobs on 
second and third row, knobs on fifth row 
weakest. Thin, lamellose, spiral, microscopic 
riblets over entire whorl. Aperture about 95- 
100% of shell height; apertural opening nar- 
row, elongate. Interior of apertural lip heavily 
callused, with pair of wide teeth, each pair 
comprising 2-4 denticles; in addition, two 
weak, separate denticles near anterior sipho- 
nal canal; interior of aperture with weak den- 



ticles at previous growth intervals. Anterior si- 
phonal canal a short and open notch; 
posterior siphonal canal absent. Columella 
heavily callused, curving inward in center, 
and with three strong columellar teeth. Three 
to four well-developed knobs on siphonal fas- 
ciole. Shell white, knobs dark brown to black; 
aperture and columella purple. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45° angle to growing 
edge (0-15%; lacking in some specimens); 
aragonitic layer with crystal planes oriented 
perpendicular to growing edge (15-35%); 
aragonitic layer with crystal planes oriented 
parallel to growing edge (40-55%); aragonitic 
layer with crystal planes oriented perpendic- 
ular to growing edge (5-10%). Presence of 
calcific layer questionable. 



PHYLOGENY OF RAPANINAE 



185 



Operculum: D-shaped, tapered at lower end, 
with lateral nucleus in center right (compare 
Fig. 1С). Free surface with bracket-shaped 
growth lines; attached surface with about 4-7 
bracket-shaped growth lines and with cal- 
lused, glazed rim (about 35-40% of opercu- 
lar width) on left. 

Anatomy (based on living and preserved an- 
imals): Mantle edge, siphon and cephalic ten- 
tacles light green with white flecks; distal por- 
tion of tentacles dark brown with white tip. 
Side of foot white with many green dots; sole 
of foot light green with white specks. Minute 
accessory boring organ with long duct dorsal 
to long, thin pedal gland. 

Osphradial length slightly more than one- 
half ctenidial length; osphradium and ctenid- 
ium about equal in width. Osphradium sym- 
metrical in shape along lateral and 
longitudinal axes. Osphradial lamellae at- 
tached along small portion of their base. 

Anteriormost portion of ctenidium bending 
below osphradium. Anterior ctenidial lamellae 
wider than deep; posterior lamellae almost as 
wide as deep. Lateral edge of ctenidial lamel- 
lae concave; ventral edge straight. 

Vaginal opening small, elliptical, situated 
on dorsal side of rodlike, tubular, partially de- 
tached extension of palliai gonoduct and lo- 
cated directly below anal opening. Bursa cop- 
ulatrix consisting of main channel and 
connecting chamber on right side, the latter 
continuous with capsule gland. Ventral chan- 
nel initially located under ventral lobe, farther 
posterior under right lobe, and formed by 
large, complex flange with longitudinal ridges. 
Ventral flange emanating from ventral epithe- 
lium. Ingesting gland dark brown, consisting 
of several small chambers filled with floccu- 
lent brown material; located on left side and 
partially ventral to capsule gland, extending to 
left side of albumen gland. Seminal recepta- 
cles white, located on dorsal periphery of 
omega-shaped albumen gland. 

Penis large, strongly recurved, with small 
papilla-like tip. Penial vas deferens as duct- 
within-a-duct system occupying one-fourth of 
penial width. Cephalic vas deferens a well- 
developed duct-within-a-duct system. Pros- 
tate white, C-shaped in cross section (antero- 
posterior view), with large C-shaped lumen 
separating left and right lobes; folded over 
and under rectum, thus enveloping it. Seminal 
vesicles yellowish white. 

Proboscis long, unpigmented, narrower 
than gland of Leiblein. Esophagus attached to 



ventral surface of proboscis by numerous, 
thin muscle threads. Accessory salivary 
glands absent. Large paired salivary gland 
lobes separate; right gland under proboscis; 
left one dorsal, extending between left side of 
proboscis and gland of Leiblein. Valve of 
Leiblein short, separate from salivary glands. 
Caplike structure present on anterior portion 
of valve of Leiblein. Salivary ducts attached to 
esophagus a short distance from valve of 
Leiblein. Valve of Leiblein adjacent to nerve 
ring. Glandular folds on mid-esophagus indis- 
cernible. Esophagus directly attached to car- 
amel brown gland of Leiblein. Posterior 
esophagus embedded along left side of gland 
of Leiblein. Gland of Leiblein spiral, forming 
two folds (three "lobes"). Posterior blind duct 
shorter than gland itself, but larger than one- 
half of gland length. 

Stomach tubular, very elongate; distinct 
lines or small folds on posterior mixing area, 
and one diverticulum present. Stomach 
typhlosole and intestinal typhlosole well de- 
veloped. Anal opening conspicuous. Rectal 
gland appearing integrated with hypobran- 
chial gland and separated from rectum by ep- 
ithelial layer. 

Radula: Ribbon length about 30% of shell 
height (Fig. IOC). Central cusp of rachidian 
constricted at base; inner lateral denticle on 
base of lateral cusp attached almost along its 
entire side; outer edge of lateral cusp straight, 
lateral denticles absent; six to seven elongate 
marginal denticles on slightly sloping, narrow 
marginal edge, with one or two fused with 
base of lateral cusp; marginal cusp thicker 
and longer than marginal denticles. Lateral 
teeth curved, longer than one-half of rachid- 
ian width. 

Egg Capsules: Unknown. 

Ecology: Much information is available on 
the ecology of several species of Drupa. J. D. 
Taylor (1983) has extensively studied the 
ecology and in particular the feeding habits of 
Drupa species. Besides general information 
on feeding habits, species and sizes of prey 
from different geographic region were listed 
and discussed (J. D. Taylor, 1983). Drupa 
morum feeds mainly on eunicid polychaetes, 
such as Lysidice sp. (Bernstein, 1970), but 
occasionally also on Lepidonotus sp., Peri- 
nereis sp. and Eurythoe complánala (Pallas) 
(J. D. Taylor, 1984; Thomas & Kohn, 1985). 
Drupa ricinus feeds on Dendropoma gregaria 
(Thomas & Kohn, 1985). 



186 



KOOL 



J. D. Taylor (1971) reported finding Drupa 
morum on the outside of cobbles and boul- 
ders, and stated that Drupa species tend to 
live on vertical surfaces. I have found Drupa 
morum living on Intertidal limestone benches, 
where wave action can be very high. Thomas 
& Kohn (1985) reported three species of 
Drupa living on a windward, seaward plat- 
form. Drupa morum lives subtidally as well, 
with individuals reaching a large size in this 
habitat. Emerson & Cernohorsky (1973) re- 
ported Drupa morum living at a depth of 40 m. 
I have collected Drupa grossularia at 10 m 
depth on Niue Island (central South Pacific). 

Distribution: Indo-Pacific (between 35°N and 
35°S), from Red Sea to Easter Island, Pitcairn 
Island, and Clipperton Island (Emerson & 
Cernohorsky, 1973). 

Genus Haustrum Perry, 1811 
(Fig. 11A-D) 

Haustrum Perry, 1811, pi. 44. 

Lepsia Hutton, 1884: 222 [type: Buccinum 
haustrum Martyn, 1784 [non-binomial], 
by subsequent designation, D. H. Gra- 
ham, 1941: 155, = Haustrum hausto- 
rium (Gmelin, 1791)]. 

Type Species: Haustrum zealandicum Perry, 
1811, by subsequent designation, Iredale, 
1915:474, = Haustrum haustorium {GmeWn, 
1791); synonyms: Buccinum haustrum Mar- 
tyn, 1784 (non-binominal); Buccinum hausto- 
rium Gmelin, 1791. 

Remarks: Haustrum haustrum is a rejected 
name (ICZN, Opinion 479, 1957: 407), be- 
cause it was published in a non-binominal 
work. Thiele (1929: 296) and Wenz (1941: 
1117) both recognized Haustrum as a genus. 

Shell: Protoconch not seen, but reported as 
having ". . . about 2 smooth whorls, . . ." 
(Suter, 1913: 422). Teleoconch (Fig. 11 A, B) 
light, ovate, of 5-7 whorls, and with im- 
pressed suture, low spire, and high whorl ex- 
pansion rate. Adult shell about 65 mm in 
height, 45 mm in width. Body whorl dome- 
shaped, about 85% of shell height, smooth, 
with 40-50 incised fine, spiral lines. Aperture 
very large, about 80% of shell height; aper- 
tural lip thin, without denticles, but showing 
grooved pattern at edge of lip. Columella flat- 
tened to concave, with heavy callus layer and 
axial fold. Anterior siphonal canal moderately 
short; posterior siphonal canal absent. Siph- 
onal fasciole slightly curved, covered with cal- 



lus. Shell brown grey, grooves white; col- 
umella white, with brown smudge on upper 
region; aperture white, with thin brown rim on 
edge. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (25-30%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (45-50%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (5-7%); calcitic layer (15-20%) (Fig. 
11C). 

Operculum: D-shaped, upper end rounded, 
with lateral nucleus in lower right (compare 
Fig. ID). Free surface with staff-shaped 
growth lines; attached surface with about 1-3 
arch-shaped growth lines and with callused, 
glazed rim (about 30-35% of opercular width) 
on left. 

Anatomy (based on preserved animals only): 
Head-foot and tentacles unpigmented to faint 
yellowish. Kidney light cream brown. Diges- 
tive gland dark green. Cephalic tentacles 
short and stubby. Mantle edge follows con- 
tour of aperture. Incurrent siphon very short, 
not extending beyond mantle edge. Small ac- 
cessory boring organ dorsal to wide pedal 
gland with folds (Fig. 4B). 

Osphradial length less than one-half ctenid- 
ial length; osphradium and ctenidlum equal In 
width or osphradial width slightly less than 
ctenidial width. Osphradium symmetrical in 
shape along lateral and longitudinal axes. Os- 
phradial lamella attached along one-half of 
their base. 

Anteriormost portion of ctenidlum straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae wider than 
deep; posterior lamellae about as wide as 
deep. Lateral edge of ctenidial lamellae con- 
vex; ventral edges concave. Distal tips of 
ctenidial support rods extending beyond lat- 
eral edge as papillalike projections (more pro- 
nounced in posterior lamellae). 

Vaginal opening round, with diameter one- 
half that of capsule gland, situated on end of 
short tube, and located directly below anal 
opening. Bursa copulatrix running dorso-ven- 
trally, splitting into capsule gland on right, and 
blind sac on lower left. Ventral channel 
minute, present only for short distance be- 
neath ventral and left lobe, then present as 
few, thin ridges emanating from ventral epi- 
thelium; posteriorly, ventral channel formed 



PHYLOGENY OF RAPANINAE 



187 












FIG. 11. A-D, Haustrum haustorium. A, shell (48 mm), apertural view. В, shell (48 mm), abapertural view. 
C, shell ultrastructure, SEM (bar = 0.10 mm). D, radula, SEM (bar = 25 ц.т). E-l, Mancinella alouina. E, 
shell (44 mm), apertural view. F, shell (44 mm), abapertural view. G, shell ultrastructure, SEM (bar = 0.20 
mm). H, shell ultrastructure, polished surface, SEM (bar = 0.20 mm). I, radula, SEM (bar = 40 |xm). 



188 



KOOL 



by flange originating from ventral epithelium, 
with minute longitudinal ridges (inward projec- 
tions in cross section). Albumen gland arch- 
shaped, very elongate. Ovary olive green. 

Penis small, lightly curved, smooth, and 
dorso-ventrally flattened. Penial duct open 
(perhaps due to poor preservation), very nar- 
row, dorsal and along posterior margin of pe- 
nis. Cephalic vas deferens closed, visible ex- 
ternally as thin, clear white line directly below 
surface. Duct continuing posteriorly on inte- 
rior of mantle as open canal before entering 
prostate. Prostate small, solid, grey, opaque 
with dorso-ventral slit, adjacent to rectal wall. 
Seminal vesicles convoluted, poorly devel- 
oped, dirty white. 

Proboscis large, unpigmented, narrower 
than gland of Leiblein. Right accessory sali- 
vary gland long, thin, nearly one-half of shell 
height, located in right upper anterior corner 
of buccal mass, extending posteriorly and 
ventrally, adjacent to right side of salivary 
glands. Left accessory salivary gland absent. 
Yellow salivary gland mass consisting of elon- 
gate portions of glandular material with multi- 
tude of small threads. Well-developed left part 
of salivary mass about equal in size to right 
accessory salivary gland. Valve of Leiblein 
elongate, partially attached to salivary glands. 
Salivary ducts attached at varying distances 
from valve of Leiblein, which lies at least one 
length away from nerve ring. Portion of mid- 
esophagus with glandular folds long; folds 
poorly developed. Well-developed, long duct 
between esophagus and gland of Leiblein, 
nearly or about as thick as posterior esopha- 
gus. Posterior esophagus attached by minute 
threads of connective tissue to lower left por- 
tion of gland of Leiblein. Gland of Leiblein 
large, spiral, forming two folds, of hard con- 
sistency, light brown, with external strawlike 
membrane thickest in older specimens. Pos- 
terior duct very short (few mm), terminating 
with ampulla. 

Stomach U-shaped, with large posterior 
mixing area. About 20 distinct folds, oriented 
towards center, on stomach wall, with minute 
lines crossing over. Yellow layer overlays 
grey, opaque folds. Two digestive diverticula 
present. Intestinal typhlosole well developed, 
with small, small parallel folds in intestinal 
groove. Intestine with many small lateral folds 
of varying sizes. Rectum very large in diam- 
eter. Rectal gland undetectable from outside 
due to dark brown to black hypobranchial 
gland. Anal opening large, well defined, with 
upward-pointing anal papilla. 



Radula: Ribbon length approximately 20- 
25% of shell height (Fig. 11D). Short central 
cusp of rachidian wide at base; elongate, nee- 
dle-shaped, well-developed, cusplike inner 
denticles separate from lateral cusps, and 
nearly as long as central cusp; outer edge of 
short and wide lateral cusps straight, devoid 
of denticles, sloping towards rachidian base. 
Lateral teeth thin, smooth, slightly longer than 
one-half of rachidian width. 

Egg Capsules: Oval to circular, about 6 mm 
in height, with large, central, ovate exit hole. 
All capsules attached at common basal mem- 
brane (D. H. Graham, 1941). 

Ecology: This species lives in the intertidal 
on rocks (Powell, 1979). 

Distribution: New Zealand (Powell, 1979) 
and southern Australia (W. F. Ponder, per- 
sonal communication). 

Genus Mancinella Link, 1807 
(Fig. 11E-I) 

Mancinella Link, 1807: 115. 

Type Species: Mancinella aculeata Link, 
1807, by absolute tautonymy through its cited 
synonym. Murex mancinella Linnaeus, 1758 
(ICZN, Opinion 911, 1970: 20), = Mancinella 
alouina (Röding, 1798); synonyms: Man- 
cinella mancinella (Linnaeus, 1758), species 
dubium, rejected name (ICZN, Opinion 911, 
1970: 21); Volema alouina Röding, 1798; 
IVolema glacialis Röding, 1798; Purpura 
gemmulata Lamarck, 1816. 

Remarks: Cossmann (1903: 71) placed Man- 
cinella in the synonymy of Purpura Bruguière. 
Thiele (1929: 297), Clench (1947: 83), Keen 
(1971b: 549) and Abbott (1974: 1118) used 
Mancinella as a subgenus of Thais. Wenz 
(1 941 : 1 1 1 8) used Mancinella as a full genus. 
Cernohorsky (1969: 296-297) stated that 
Mancinella mancinella Linnaeus, 1758, is the 
type of the genus by tautonymy, although the 
Linnaean taxon is a composite species. Cer- 
nohorsky points out that it is clear that Lin- 
naeus only described one of the specimens 
(Mancinella mancinella of authors) in the 
"Murex mancinella" box in the Linnaean col- 
lection. However, Vokes (1970) noted that 
Linnaeus' description does not fit any of the 
specimens in the box. Vokes followed F. A. 
Smith (1913: 287) and considered Murex 
mancinella a nomen dubium. Keen (1964) pe- 
titioned the ICZN that Mancinella gemmulata 



PHYLOGENY OF RAPANINAE 



189 



(Lamarck, 1 81 6) ( = M. aculeata Link) be des- 
ignated as the type of Mancinella. The ICZN 
ruled (Opinion 91 1, 1970: 20) that Mancinella 
aculeata be the type species of the genus 
Mancinella. An available earlier name for 
Mancinella aculeata is Röding's Volema 
alouina. 

Shell: Protoconch unknown. Teleoconch 
(Fig. 1 1 E, F) strong, oval, squat, of about five 
adpressed whorls. Adult shell up to about 60 
mm in height, 40 mm in width. Globose body 
whorl about 95% of shell height and sculp- 
tured with five spiral rows of 9-10 occasion- 
ally spinelike, axially arranged knobs. Largest 
knobs on second and third row, knobs on fifth 
row weakest. About ten narrow minute ridges 
between rows. Aperture large, about 75% of 
shell height. Apertural lip with 10-12 spiral 
striae beginning about 1 cm from apertural 
edge. Siphonal canal moderately developed, 
deep, semi-closed. Columella flat to slightly 
concave, with angular curve in lower portion 
forming part of short, open anterior siphonal 
canal; posterior siphonal canal absent. Siph- 
onal fasciole with 5-6 knobs. Shell cream 
brown, knobs rusty brown, especially when 
worn; aperture and columella light to dark or- 
ange, with apertural striae dark orange. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45° angle to growing 
edge (15-20%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (25-30%); aragonitic layer with crystal 
planes oriented parallel to growing edge (30- 
40%); aragonitic layer with crystal planes ori- 
ented perpendicular to growing edge (7-9%); 
calcific layer (4-6%) (Fig. 11G, H). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface with about 4-7 bracket-shaped 
growth lines and with callused, glazed rim 
(about 35-45% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Head-foot and tentacles rusty, light to 
dark brown. Kidney olive green. Hypobran- 
chial gland bright light green. Digestive gland 
grey brown. Mantle edge smooth; incurrent 
siphon extending far from mantle edge. Ac- 
cessory boring organ dorsal to pedal gland 
(Fig. 4B). 

Osphradial length slightly more than one- 
half ctenidial length; osphradial width nearly 
equal to ctenidial width. Osphradium symmet- 
rical in shape along lateral axis; right pectin 



wider than left. Osphradial lamellae attached 
along very small portion of their base. 

Anteriormost portion of ctenidium straight, 
extending slightly farther anteriorty than os- 
phradium. Anterior and posterior ctenidial 
lamellae as deep as wide. Lateral edges of 
ctenidial lamellae faintly S-shaped; ventral 
edges concave. 

Vaginal opening central, slightly protruded 
on short tubular oviduct and located below 
and posterior to anal opening. Bursa copula- 
trix short, as part of vagina and anterior to 
capsule gland. Ventral channel formed by 
small flange originating from ventral epithe- 
lium. Ventral flange with few longitudinal 
ridges and located under ventral lobe. Ingest- 
ing gland a single chamber (not visible from 
outside). Albumen gland of the omega- or 
arch-shaped type, with many long, white sem- 
inal receptacles on dorsal periphery. Ovary 
yellow (in preserved specimens). 

Penis strongly recurved, with flagelliform 
tip, dorso-ventrally flattened. Penial vas def- 
erens as central, minute duct-within-a-duct 
system occupying about one-sixth of penial 
width. Cephalic vas deferens thin, running 
along mantle prior to entering prostate. Pros- 
tate small, yellow, with central duct, smaller in 
diameter than adjacent rectum. 

Proboscis large, unpigmented, nearly equal 
in width to gland of Leiblein. Paired accessory 
salivary glands very small, short, thin; left 
gland located in left anterior portion of buccal 
mass adjacent to salivary gland mass; right 
accessory salivary gland located in right an- 
terior portion of buccal mass, adjacent to pro- 
boscis. Salivary glands small, yellowish, lo- 
cated to left of proboscis, and anterior to 
gland of Leiblein. Salivary ducts attached to 
anterior portion of esophagus directly anterior 
of valve of Leiblein. Valve of Leiblein elon- 
gate, adjacent to nerve ring. Folds on mid- 
esophagus nearty indiscernible. Duct be- 
tween mid-esophagus and gland of Leiblein 
short and much thinner than posterior esoph- 
agus. Posterior esophagus adjacent to lower 
left portion of gland of Leiblein. Gland of 
Leiblein spiral, forming two folds, of hard con- 
sistency, yellowish, with thin external mem- 
brane. Posterior duct about one-half of length 
of gland of Leiblein and with terminal ampulla. 

Stomach nearty rectangular, with large pos- 
terior mixing area. About 12-15 folds on 
stomach wall, oriented towards center of 
stomach. Two digestive diverticula present. 
Stomach typhlosole only moderately devel- 
oped. Intestinal typhlosole thin. Intestinal wall 



190 



KOOL 



with many minute lateral lines and small folds. 
Intestinal groove with few thin longitudinal 
folds. Rectum with moderate diameter. Anal 
opening well defined, with anal papilla. 

Radula: Ribbon length about 25% of shell 
height (Fig. 1 1 1). Rachidian with thick, needle- 
shaped central cusp; short, wide lateral cusps 
smooth, with outside edge sloping to rachid- 
ian edge. Lateral teeth smooth, about three- 
fourths of rachidian width. 

Egg Capsules: Unknown. 

Ecology: Mancinella alouina lives from the in- 
tertidal to subtidal zones on sheltered rocks, 
whereas Mancinella echinulata occurs in 
crevices on exposed reefs (Kilburn & Rippey, 
1982). Remains of small crustaceans were 
present in the rectum of several animals ex- 
amined. 

Distribution: Red Sea and throughout Indo- 
Pacific (Cernohorsky, 1969). 

Genus Morula Schumacher, 1817 
(Fig. 12A-G) 

Morula Schumacher, 1817: 68, 227. 

Tenguella Arakawa, 1965: 123 [type: Purpura 
granulata Duelos, 1832, by original des- 
ignation, = Morula granulata (Duelos, 
1832)]. 

Type Species: Morula papulosa Schuma- 
cher, 1817 {non Philippi, 1849), by monotypy, 
= Morula uva (Röding, 1798); synonyms: 
Drupa uva Röding, 1798; Ricinula nodus La- 
marck, 1816; Ricinula áspera Lamarck, 1816; 
Ricinula morus Lamarck, 1822; Purpura 
sphaeridia Duelos, 1832; Ricinula alba 
Mörch, 1852; 7Sistrum striatum Pease, 1868; 
1 Morula nodilifera Habe & Kosuge, 1966. 

Remarks: Thiele (1929: 295) and Wenz 
(1941: 1114) considered Morula a section of 
the subgenus Drupa in the genus Drupa. 
Morula granulata was designated as type 
species of Tenguella Arakawa, 1965, based 
on radular characters (presence and number 
of marginal denticles). However, the number 
of marginal denticles is variable in both spe- 
cies and overlap occurs. Tenguella is herein 
considered synonymous with Morula. 

Shell: Protoconch (Fig. 12C, D) tall, conical, 
of at least 4.25 adpressed whorls [exact count 
could not be made from available specimen], 
sculptured with 3 spiral cords of small bead- 
like pustules directly below suture, but other- 



wise smooth, and with outward-flaring lip; si- 
nusigeral notch covered by teleoconch. 
Teleoconch (Fig. 12A, B) ovate, of 5-6 ad- 
pressed whorls, with moderately high spire. 
Adult shell up to about 27 mm in height, 17 
mm in width. Body whorl about 80% of shell 
height, sculptured with five spiral rows of 12 
short but well-developed knobs. One spiral, 
faintly lamellose ridge between rows with 
deep groove on each side. Elongate aperture 
about 68% of shell height. Apertural opening 
narrow, due to pair of heavy denticles pointing 
inward. Two smaller denticles located on 
lower end. Anterior siphonal canal very short, 
semi-closed; posterior siphonal canal absent. 
Columella concave; lower part with several 
faint denticles. Siphonal fascicle strongly 
curved, previous edges still visible, not knob- 
like. Shell white, knobs black; aperture and 
columella pink to violet purple. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (15-25%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (75-85%) (Fig. 12F). 

Operculum: D-shaped, with S-shaped left 
edge, tapered at lower end, with lateral nu- 
cleus in lower right (Fig. IF). Free surface 
with bracket-shaped growth lines; attached 
surface with about 4-6 bracket-shaped 
growth lines and with callused, dull rim (about 
30-35% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Head with long cephalic tentacles em- 
anating from common base. Lower part of 
head-foot mottled black and white to uniform 
black on lower portion; upper part with white 
and orange flecks. Tentacles uniform black at 
bases, white distally, or white with small black 
lateral band at eye levels. Mantle edge 
crenate, folded; underside of mantle with 
black and white patches. Incurrent siphon uni- 
form black, or with white flecks. Kidney cara- 
mel brown. Digestive gland dark brown. Sole 
white with central, opaque, white speckled 
band, oriented antero-posteriorly. Accessory 
boring organ large, with short duct opening 
close to anteriorly located pedal groove. Hy- 
pobranchial gland very large, divided into red 
brown, white, and green portions, and with 
black rods of unknown composition pointing 
towards mantle cavity. Ventral pedal gland 
combined with accessory boring organ. 

Osphradial length slightly greater than one- 
half ctenidial length (Fig. 3D); osphradial 



PHYLOGENY OF RAPANINAE 



191 




FIG. 12. Morula uva. A, shell (25 mm), apertural view. B, shell (25 mm), abapertural view. C, protoconch, 
side view, SEM (bar = 60 |xm). D, protoconch, apical view, SEM (bar = 60 |xm). E, penis, viewed 
postero-anteriorly, SEM (bar = 0.20 mm). F, shell ultrastructure, SEM (bar = 0.10 mm). G, radula, SEM (bar 
= 10 M,m). 



192 



KOOL 



width equal to or slightly greater than ctenidial 
width. Osphradium more tapered at posterior 
end; right pectin slightly wider than left. Os- 
phradial lamellae attached along most of their 
base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae deeper than 
wide; posterior lamellae as deep as wide. Lat- 
eral edges (Fig. 3D, le) of ctenidial lamellae 
concave; ventral edges straight. Distal tips of 
ctenidial support rods extending beyond lat- 
eral edge as papillalike projections. 

Vaginal opening a short slit (more rounded 
in juveniles) situated on distal end of tubular 
extension of palliai gonoduct and located be- 
neath anal opening. Bursa copulatrix as 
dorso-ventral slit open to vagina and contin- 
uous with capsule gland. Vagina continuing 
as ventral channel with large, circular ventral 
flange with many longitudinal and well-devel- 
oped ridges; flange positioned below left lobe 
of capsule gland anteriorly, smaller, flattened, 
and below both lobes posteriorly. Ventral 
channel branching away from capsule gland, 
forming large posterior bursa. Branch of 
bursa continuing as oviduct, larger portion as 
blind sac. Bursa connected to single-cham- 
bered ingesting gland with short duct. Ingest- 
ing gland larger than albumen gland and 
black when viewed from outside. Albumen 
gland staff-shaped, with anterior portion being 
much shorter and less developed. Few sem- 
inal receptacles (3-5) at dorsal side branch- 
ing from ovi-sperm duct prior to it connecting 
to albumen gland. Ovary white to yellow. [The 
female reproductive system of Morula granu- 
lata was described in detail by Srilakshmi 
(1991)]. 

Penis (Fig. 5E, 12E) very large, strongly re- 
curved, round in cross section, V-shaped, 
with flattened, large side lobe; distal end of 
penis varying in length and attached by small 
connection to proximal part of penis. Penial 
vas deferens as duct-within-a-duct system 
occupying about one-fifth of penial width. 
Cephalic vas deferens minute, describing "Z" 
pattern. Prostate solid, glandular, opaque, 
white opaque or dark brown, with closed duct; 
prostate much larger than rectum and not 
separated from it by layer of epithelium. Sem- 
inal vesicles well developed, white to dark or- 
ange brown. 

Proboscis large, equal in width to gland of 
Leiblein, occasionally folded and horseshoe- 
shaped, laying against left side of gland of 
Leiblein. Paired accessory salivary glands 



club-shaped, small, equal in length, much 
smaller than one-half of shell height; left ac- 
cessory salivary gland embedded in left sali- 
vary gland; right gland separate. Salivary 
glands very large, much larger than acces- 
sory salivary glands and almost as large as 
gland of Leiblein, located dorsally either as 
separate lobes or solid mass. Salivary ducts 
attached close to valve of Leiblein. Valve of 
Leiblein short, with caplike structure on ante- 
rior end, and lying adjacent to nerve ring, sep- 
arate from salivary glands. Glandular folds of 
mid-esophagus nearly indiscernible. Duct be- 
tween mid-esophagus and gland of Leiblein 
very thin. Posterior esophagus separate from 
gland of Leiblein. Gland of Leiblein spiral, 
forming two folds, of soft consistency, consist- 
ing of small cavities, dark brown, lacking 
strawlike membrane. 

Stomach as wide tube with few very large 
folds and many minute folds on stomach wall 
of posterior mixing area. Small unciliated area 
between posterior mixing area and intestine. 
Stomach and intestinal typhlosoles very well 
developed. One diverticulum present directly 
anterior to esophagus. Anal opening incon- 
spicuous but with very large papilla. Thin rec- 
tal gland along entire capsule gland. 

Radula: Ribbon length about 15% of shell 
height (Fig. 12G). Central cusp on rachidian 
tooth needle-shaped, with moderately wide 
base; lateral denticle separate from lateral 
cusp; outer and inner edge of lateral cusp 
straight, smooth; several stubby marginal 
denticles present on wide, horizontal edge of 
rachidian; wide, short marginal cusp. Lateral 
teeth strongly curved, smooth, with wide 
base; about one-half of rachidian width. 

Egg Capsules: Unknown. 

Ecology: Common on intertidal limestone 
benches, where it feeds almost exclusively on 
vermetid gastropods (Kay, 1971 ; Miller, 1970; 
J. D.Taylor, 1976, 1984). 

Distribution: Indo-Pacific, from Red Sea to 
Isla Guadalupe and Clipperton Island (Cerno- 
horsky, 1969; Keen, 1971b). 

Genus Nassa Röding, 1798 
(Fig. 13A-G) 

Nassa Röding, 1798: 132 {non Lamarck, 
1799, = /Vassar/us Duméril, 1806). 

lopas H. & A. Adams, 1853: 128 [type: Вис- 
cinum sertum Bruguière, 1789, by sub- 



PHYLOGENY OF RAPANINAE 



193 




FIG. 13. A-C, F-G, Nassa serta: A, shell (40 mm), apertural view. B, shell (44 mm), abapertural view. C, 
larval shell, side view, SEM (bar = 25 jim). F, shell ultrastructure, SEM (bar = 0.10 mm). G, radula, SEM, 
(bar = 25 |xm). D-E, Nassa "francolina" D, protoconch, side view, SEM (bar = 80 ц,т). E, protoconch, 
apical view, SEM (bar = 80 |xm). 



194 



KOOL 



sequent designation, Baker, 1895: 185, 
= Nassa serta (Bruguière, 1789)]. 

Jopus Schaufuss, 1869 (error for lopas). 

Jopas Baker, 1895: 185 (unjustified emenda- 
tion of lopas). 

Type Species: According to a number of au- 
thors (Winckworth, 1945; Iredale & Mc- 
Michael, 1962; Cernohorsky, 1969), Dall 
(1909) subsequently designated Nassa picta 
Röding, 1798, as the type species of Nassa. 
However, Dall (p. 47) does not list the name 
picta, but rather "Purpura sertum Lam" as 
type of Nassa, which was not one of the spe- 
cies included by Röding and is therefore un- 
available. I can find no valid subsequent des- 
ignation and here designate the type species 
as Nassa picta Röding, 1798, = Nassa serta 
(Bruguière, 1789); synonyms: Buccinum ser- 
tum Bruguière, 1789; Buccinum coronatum 
Gmelin, 1791; ?Stramonita hederacea Schu- 
macher, 1817; ?Buccinum francolinus Bru- 
guière, 1789; Buccinum situla Reeve, 1846. 

Remarks: Cossmann (1903: 68) considered 
Nassa a full genus (as ¡opas), and included, 
besides lopas s.s, Taurasia Bellardi, 1882. 
Thiele (1929: 296) used Jopas and included 
the subgenera Jopas ( = Nassa) and Vexilla. 
Wenz (1941 : 1 1 16) used Nassa and included 
the subgenera Nassa, Vexilla, and Taurasia. 
Controversy exists about whether the ge- 
nus Nassa contains one or two species. The 
nominal species serta and francolina can be 
separated on the basis of shell sculpture and 
geographic distribution (see "Distribution"). 
Individuals from the Pacific Ocean, tradition- 
ally grouped under N. serta, have shells with 
relatively coarse spiral ribs, whereas the 
shells of Indian Ocean specimens have very 
fine spiral lines and appear nearly smooth. I 
suspect, however, that future research will 
show that these taxa are conspecific, consid- 
ering the range of variation in sculptural pat- 
terns In many other rapanine species. 

Shell: Embryonic shell (Fig. 13C) with well- 
developed beak and pattern of spiral rows of 
microscopic volcanolike pustules. Protoconch 
(Fig. 13D, E; typical N. francolina) tall, coni- 
cal, of at least 4.25 adpressed whorls [exact 
count could not be made from available spec- 
imen], with subsutural plicae interconnected 
by three thin spiral ridges, but otherwise 
smooth, and with outward-flaring lip; sinusig- 
eral notch covered by teleoconch. Teleo- 



conch (Fig. 13A, B) elongate, slender, fusi- 
form, of 6-7 adpressed whorls. Adult shell up 
to about 70 mm in height, 35 mm in width. 
Body whorl rounded, about 85-90% of shell 
height. Body whorl sculptured with about 30 
small, spiral cords of minute pustules, nearly 
smooth in typical N. francolina. Aperture elon- 
gate, large, about 75% of shell height, curved 
angularly at base to form part of siphonal ca- 
nal. Apertural lip smooth interiorly, but 
crenate at edge, corresponding to external 
pattern of small ridges. Siphonal notch wide 
and open. Columella lightly callused and 
rounded. Posterior siphonal canal absent, but 
protrusion of columellar callus directly across 
from similar protrusion on inside of apertural 
lip forming canal in posteriormost end of ap- 
erture. Siphonal ridge with similar pattern as 
on body whorl, slightly curved, adjacent to 
columellar callus. Shell with varying color pat- 
terns comprising combinations of cream (usu- 
ally as median band running around body 
whorl), light and dark brown spiral bands 
which may consist of blotches; aperture white 
with some yellow tinges towards edge, and 
dark brown crenulations on edge, corre- 
sponding with dark brown spiral ridges; top of 
columella yellow white, caramel brown at 
base. 

Sfiell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (45-50%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (30-35%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (15-20%) (Fig. 13F). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface without distinct growth lines and with 
callused, glazed rim (about 45-55% of oper- 
cular width) on left. 

Anatomy (based on living and preserved an- 
imals): Cephalic tentacles long, uniform 
black, with distal halves of tips white. Head- 
foot uniform black, lightly spotted with white. 
Mantle edge simple and straight. Incurrent si- 
phon long, uniform black. Hypobranchial 
gland brown to yellow. Kidney brown. 
Nephridial gland S-shaped, wide, opaque. Di- 
gestive gland dark brown. Sole of foot yellow, 
with pattern of thin ridges. Accessory boring 
organ with long duct. Pedal gland large, lo- 
cated under accessory boring organ (Fig. 4B). 
Osphradial length equal to or greater than 
ctenidial length; osphradium and ctenidium 



PHYLOGENY OF RAPANINAE 



195 



about equal in width. Osphradium symmetri- 
cal in shape along lateral and longitudinal 
axes. Osphradial lamellae of right pectin at- 
tached along one-half of their base; those of 
left pectin attached along entire base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior and posterior ctenidial lamellae 
much deeper than wide. Lateral and ventral 
edges of ctenidial lamellae variable in shape. 
Distal tips of ctenidial support rods extend- 
ing beyond lateral edge as papillalike projec- 
tions. 

Vaginal opening slit-shaped, with two lon- 
gitudinal flanges in opening and located be- 
low and posterior to anal opening. Bursa cop- 
ulatrix as large storage area with fine 
horizontal lines, continuous with capsule 
gland. Small, circular flange originating from 
ventral epithelium, under small ventral lobe of 
anterior portion of capsule gland; flange 
minute, hooklike posteriorly, perpendicular to 
capsule gland lobes. Flange split at base in 
central portion of capsule gland. Ingesting 
gland as large thin-walled chamber contain- 
ing granular, caramel brown material. Semi- 
nal receptacles on dorsal periphery of omega- 
shaped albumen gland elongate to club- 
shaped, white, nearly reaching oviduct. Ovary 
orange. 

Penis long, thin, slightly recurved, flagelli- 
form, oval in cross section (Fig. 5C). Penial 
vas deferens as duct-within-a-duct system 
occupying one-fourth of penial width. Cepha- 
lic vas deferens thin, inconspicuous. Prostate 
small, white, with central duct, separated from 
very large rectum by epithelial layer. Seminal 
vesicles well developed, white. 

Proboscis very large, equal in width to 
gland of Leiblein, white. Paired accessory sal- 
ivary glands thin, equally long, about one- 
third of shell height. Left accessory gland ad- 
jacent to salivary gland mass; right gland in 
anterior right area of buccal cavity separate 
from salivary gland mass. Paired accessory 
salivary glands equal in size to salivary gland 
mass. Salivary glands inseparable, oriented 
dorso-ventrally. Valve of Leiblein elongate, 
not embedded in salivary glands. Salivary 
ducts attached to anterior portion of valve of 
Leiblein. Valve of Leiblein adjacent to nerve 
ring. Portion of mid-esophagus with glandular 
folds short, well developed. Duct between 
mid-esophagus and gland of Leiblein distinct, 
but thinner than esophagus. Posterior esoph- 
agus attached to lower left portion of gland of 
Leiblein. Gland of Leiblein spiral, forming one 



fold, light brown, with strawlike membrane. 
Posterior blind duct of gland of Leiblein longer 
than one-half of length of gland itself and 
opening into dorsal branch of renal afferent 
vein, extending beyond kidney opening. 

Stomach as wide tube with large posterior 
mixing area. Large number of folds on stom- 
ach wall of posterior mixing area; folds ori- 
ented towards stomach center; each one con- 
taining many lateral folds, directing small 
particles laterally. Stomach typhlosole well 
developed with two digestive diverticula at 
base; intestinal typhlosole narrow but distinct. 
Several small elongate folds in intestinal 
groove. Large bulbous papilla extending from 
dorsal rectal wall, lying over very small anal 
opening. Large thick orange gland over palliai 
gonoduct. Rectal gland dark green, thin, 
alone; entire capsule or prostate. 

htadula: Ribbon length about 25% of shell 
height (Fig. 13G). Rachidian with thin central 
cusp; inner lateral cusp denticle separate 
from lateral cusp in males; denticle may be 
absent, especially in narrower rachidian tooth 
of females (see Maes, 1966); lateral cusps 
smooth, less developed in female specimens 
relative to central cusp; outer edge of lateral 
cusps sloping nearly straight down to edge of 
rachidian. Lateral teeth very wide at base and 
as long as rachidian width. 

Egg Capsules: Cylindrical, 6-8 mm in 
height; base wide, 1-2 mm in length. Some 
appearing to consist of four sides, base con- 
stricted lengthwise along axes. All capsules 
attached to basal membrane. Exit hole on cir- 
cular apical plate, usually slightly off center. 

Ecology: Nassa serta lives under boulders 
and coral rubble on limestone benches and 
reef flats of the Pacific Ocean. Analysis of 
stomach contents revealed rachidian teeth of 
Nassa radula, suggesting cannibalism. Some 
specimens were found laying egg capsules 
under a large piece of coral rubble at low tide. 

Distribution: Indian Ocean, from Cocos-Keel- 
ing Islands (Maes, 1967: 132) throughout 
tropical Pacific Ocean (Abbott & Dance, 
1982) (typical Nassa serta); in remainder of 
Indian Ocean (Cernohorsky, 1969) usually re- 
ferred to as Nassa francolina. 

Genus Neorapana Cooke, 1918 
(Fig. 14A-F) 

Neorapana Cooke, 1918: 7 (as a subgenus of 
Acanthina Fischer von Waldheim, 1807). 



196 



KOOL 




FIG. 14. Neorapana muricata. A, shell (45 mm), apertural view. B, shell (45 mm), abapertural view. C. 
protoconch, side view, SEM, (bar = 0.20 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, shell 
uitrastructure, SEM (bar = 0.20 mm). F, radula, SEM (bar = 35 fim). 



PHYLOGENY OF RAPANINAE 



197 



Type Species: Purpura muricata Broderip, 
1832, by original designation, = Neorapana 
muricata [Broderip, 1832]; synonyms: Pur- 
pura truncata Duelos, 1832; Monoceros tu- 
berculatum Sowerby, 1835, ex Gray Ms. 

Remarks: Cooke based his separation of 
Neorapana from Acanthina s.s. on radular 
characters. The shell of N. muricata resem- 
bles that of species of Acanthina in having a 
labial tooth. This single character was the pri- 
mary criterion for inclusion of this species in 
the genus Acanthina by several authors. 
Thiele (1929: 297) allotted Neorapana section 
status under the subgenus Mancinella of the 
genus Thais. Wenz (1941: 1118) considered 
Neorapana a subgenus of Thais. Keen 
(1971b: 554) considered Neorapana a full ge- 
nus in the Rapaninae. 

Specimens of Neorapana muricata used in 
this study are representatives of typical 
Neorapana tuberculata (Sowerby, 1835); N. 
muricata has a greater distribution, ranging 
from Guaymas, Mexico, to Ecuador, whereas 
typical N. tuberculata ranges from Cabo San 
Lucas, Mexico, throughout the Gulf of Califor- 
nia to Mazatlán, Mexico (Keen, 1971b), thus 
partially overlapping in range with N. muri- 
cata. I regard the latter as merely a form or 
variant of the former; intergrading shell forms 
suggest conspecificity. Detailed anatomical 
and molecular studies, however, could show 
these forms to be different species. But until 
such a study has been performed, I will con- 
tinue considering these two names to be syn- 
onyms, with muricata having priority over tu- 
berculata. 

Shell: Protoconch (Fig. 14C, D) tall, conical, 
of at least 3.25 adpressed whorls [exact count 
could not be made from available specimen], 
with faint, small subsutural plicae and micro- 
scopic pustules (last whorl), and with out- 
ward-flaring lip; sinusigeral notch covered by 
teleoconch. Because the descriptions of N. 
muricata beyond the shell morphology are 
based on "tuberculate" specimens, a descrip- 
tion of the tuberculate shell morph follows. Te- 
leoconch (Fig. 13A, B) large, heavy, conical, 
of 5-6 adpressed whorls. Adult up to about 
60 mm (80 mm in typical N. muricata) in 
height, 45 mm (70 mm in typical N. muricata) 
in width. Body whorl about 85-90% of shell 
height, somewhat dome-shaped, sculptured 
with well-developed shoulder, and bearing 
four rows of spiral bands of 6-7 knobs. Su- 
ture lying adjacent to and following lower con- 
tours of second row of knobs on penultimate 



whorl. First row of knobs on angular shoulder, 
highly developed and with discontinuous 
ridge on knobs. Second, third and fourth rows 
consecutively less developed. Knobs of two 
uppermost rows lying directly under and 
above each other, as do third and fourth row, 
but knobs on latter pair not axially aligned with 
knobs on first two rows. Five to eight narrow, 
delicately lamellose spiral ridges between 
pairs of rows of knobs. Aperture large, about 
80-90% of shell height. Apertural lip with 
12-16 ridges on inside surface, most pro- 
nounced on last growth increment. Edge of lip 
crenate and thin. Anterior siphonal canal 
short, well developed in some specimens, but 
only a notch in others; posterior siphonal ca- 
nal poorly developed. Columella lightly to 
heavily callused, rounded to concave. Sipho- 
nal fascicle strongly curved, bending outward 
and free of callus margin. Shell cream to yel- 
low orange brown; columella white to yellow; 
interior apertural lip white to yellow orange. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45°-angle to grow- 
ing edge (15-20%); aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (25-30%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (30-40%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (5-8%); calcific layer (8-15%) (Fig. 
14E). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface with about 3-6 bracket-shaped 
growth lines and with callused, glazed rim 
(about 45-50% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Head-foot mottled black on white 
base. Mantle edge crenate, following aperture 
contour. Siphon long, black and white, ex- 
tending some distance beyond mantle edge. 
Hypobranchial gland with cottonlike appear- 
ance. Digesting gland caramel brown (one 
male examined) or dark olive green (one fe- 
male examined). Accessory boring organ rel- 
atively small, dorsal to narrow ventral pedal 
gland in females (Fig. 4B), with small trans- 
verse folds on transition zone. 

Osphradial length about one-half ctenidial 
length; osphradial width less than one-half 
ctenidial width. Osphradium symmetrical in 
shape along lateral and longitudinal axes. Os- 



198 



KOOL 



phradial lamellae attached along small por- 
tion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior and posterior ctenidial lamellae 
wider than deep. Lateral edge of ctenidial 
lamellae strongly concave; ventral edge mod- 
erately concave or S-shaped. Distal tips of 
ctenidial support rods extending beyond lat- 
eral edge as papillate projections. 

Vaginal opening slit-shaped, situated on 
distal end of short, attached, tubular exten- 
sion of palliai gonoduct, and located below 
and slightly posterior to anus. Bursa copula- 
trix small, with large inner ridges; bursa in 
open connection with vagina and located on 
right side of it, continuous with capsule gland. 
Large, complex ventral flange located under 
right lobe of capsule gland. Ingesting gland 
very large, dark brown, filled with dark brown 
granular chunks; single chambered, with 
small tubes connecting walls; extending from 
dorsal left posterior portion of capsule gland 
to left of albumen gland. Albumen gland 
omega-shaped, tilted strongly backwards. 
Seminal receptacles on dorsal periphery of 
albumen gland white. 

Penis strongly recurved, elongate, thick, 
muscular gradually tapering, and oval in cross 
section. Penial vas deferens as minute duct- 
within-a-duct system occupying one-eighth of 
penial width. Prostate white, with large longi- 
tudinal central opening closed, directly adja- 
cent to rectum. Seminal vesicles well devel- 
oped, orange or white. 

Proboscis black and white, much thinner 
than gland of Leiblein. Paired accessory sal- 
ivary glands thin, equally long, about one- 
third of shell height; left gland adjacent to sal- 
ivary gland, right one largely separate from 
salivary gland. Paired salivary glands as 
joined mass, each lobe consisting of many 
worm-shaped strands connected by small 
ducts. Valve of Leiblein elongate, separate 
from salivary gland mass, a considerable dis- 
tance from nerve ring. Salivary ducts attached 
to anterior portion of esophagus directly an- 
terior of valve of Leiblein. Glandular folds on 
mid-esophagus inconspicuous. Duct between 
gland of Leiblein and esophagus poorly de- 
veloped. Posterior esophagus attached to 
posterior lower left side of gland of Leibleln. 
Gland of Leiblein large, spiral, forming one 
fold with hole in center for passage of anterior 
aorta, of hard consistency, yellow to cream, 
and with thin strawlike membrane. Posterior 
blind duct of gland of Leiblein about one-half 



of length of gland of Leiblein and entering dor- 
sal branch of afferent renal vein. 

Stomach tubular, with large posterior mix- 
ing area, with 6-15 folds on stomach wall ori- 
ented towards center of stomach. Stomach 
typhlosole very large, sometimes continuing 
up left portion of stomach wall. Intestinal 
typhlosole thin, flat. Several small folds in in- 
testinal groove. Wide, thick fold demarcating 
entrance of intestine in older female speci- 
mens. Smooth area adjacent to thick fold. 
Two large digestive diverticula present. Rec- 
tum of moderate diameter, embedded in 
spongy connective tissue. Long papilla lying 
over distinct but small anal opening. Wide 
rectal gland adjacent to most of prostate and 
capsule gland. 

Radula: Rachidian with thick, wide central 
cusp, nearly one-third of rachidian width (Fig. 
14F); inner edge of lateral cusps convex, 
outer edge slightly concave; outer edge of lat- 
eral cusp sloping steeply towards marginal 
edge of rachidian, and with faint minute folds 
on lower base. Lateral teeth with wide bases 
and curving "hooked" tips; length of lateral 
teeth greater than rachidian width. 

Egg Capsules: Unknown. 

Ecology: Neorapana muricata lives on boul- 
ders in the intertidal zone but may occur in the 
sublittoral. I found many specimens partially 
buried in sand at the sand-rock interface; it is 
not clear whether this resulted from burrowing 
behavior or from sediment accumulation. 
Small crabs were present in the mantle of two 
specimens of Neorapana muricata. The diet 
of this species is not known. 

Distribution: Eastern Pacific, from eastern 
Baja California, Mexico, to Ecuador (Keen, 
1971b). 

Genus Nucella Röding, 1 798 
(Fig. 15A-G) 

Nucella Röding, 1798: 130. 

Polytropa Swainson, 1840: 80, 305 [type: 
Buccinum lapillus Linnaeus, 1758, by 
subsequent designation, Gray, 1847: 
138, = Nucella lapillus (Linnaeus, 
1758)]. 

Polytropalicus Rovereto, 1899: 105 (unnec- 
essary replacement name for Polytropa 
Swainson; section of Purpura) {nomen 
dubium). 



PHYLOGENY OF RAPANINAE 



199 




FIG. 15. Nucella lapillus. A, shell (32 mm), apertural view. B, shell (32 mm), abapertural view. C, protoconch, 
side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, shell ultrastructure, 
SEM (x55). F, radula, SEM (bar = 20 |xm). G, radula, side view, SEM (bar = 10 м,т). 



200 



KOOL 



Type Species: Buccinum filosum Gmelin, 
1791, by subsequent designation, Stewart, 
1927: 386 (footnote 260), = Nucella lapillus 
(Linnaeus, 1758); synonyms: Buccinum lapil- 
lus Linnaeus, 1758: 739; Nucella theobroma 
Röding, 1798; Purpura imbricata Lamarck, 
1822; Purpura bizonalis Lamarck, 1822; Pur- 
pura buccinoidea Blainville, 1829; Purpura 
céltica Locard, 1886; Coralliophila rolani Bog\ 
& Nofroni, 1984. 

Remarks: Cossmann (1903: 68) recognized 
Rovereto's subgenus Polytropalicus, not real- 
izing that it was an unnecessary replacement 
name for Polytropa. Thiele (1929: 298) in- 
cluded the sections Nucella, Acanthina, 
Acanthinucella Cooke, 1918, and Neothias 
(as Neothais; unjustified emendation) in the 
genus Nucella. Wenz (1941: 1123) raised 
these sections to subgeneric status under Nu- 
cella. Nucella species have often been placed 
in Thais and Purpura. For detailed information 
on the taxonomic history of the type species 
designation for Nucella, see Rehder (1962) 
and Kool & Boss (1992). 

Shell: Protoconch (Fig. 15C, D) short, coni- 
cal, of about 1.25 smooth whorls, and with 
impressed suture; transition with teleoconch 
smooth. Teleoconch (Fig. 15A, B) highly poly- 
morphic, but usually elongate, oval, of 6-7 
adpressed whoris. Adult shell up to about 55 
mm in height, 30 mm in width. Body whorl 
rounded, about 80% of shell height, smooth 
or sculptured with pattern of 15 spiral, occa- 
sionally lamellose ridges. Aperture oval, 
about 65% of shell height; apertural lip wide, 
inside smooth, occasionally with 3-4 denti- 
cles on edge of thickened lip. Anterior sipho- 
nal canal short, open or semi-closed; poste- 
rior siphonal canal absent. Columella with 
moderate amount of callus, flat to concave, 
with angular curve in lower portion to form 
part of siphonal canal. Siphonal fascicle 
poorly developed, adjacent to callus layer. 
Shell color variable: white, grey, yellow, 
brown, orange-red; often with banding pat- 
terns of these colors; aperture and columella 
white. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (15-25%) (not always present); ara- 
gonitic layer with crystal planes oriented par- 
allel to growing edge, occasionally colored 
reddish brown (15-35%); calcific layer (40- 
85%) (Fig. 15G). 



Operculum: D-shaped, upper end rounded, 
with lateral nucleus in lower right (compare 
Fig. ID). Free surface with staff-shaped 
growth lines; attached surface with about 3-5 
arch-shaped growth lines and with callused, 
glazed rim (about 35-40% of opercular width) 
on left. 

Anatomy (based on living and preserved an- 
imals): Head-foot light yellow to white, with 
elongate, thin cephalic tentacles and short an- 
terior siphon. Mantle edge smooth, straight. 
Sole of foot with ridges. Small nephridial gland 
arching over pericardium. Large accessory 
boring organ separated from adjacent, equally 
large pedal gland present in females (Fig. 4A). 

Osphradial length slightly more than one- 
third ctenidial length; osphradial width less 
than one-half ctenidial width. Osphradium 
symmetrical in shape along lateral axis; right 
pectin usually wider than left. Osphradial 
lamellae attached along one-half of their 
base. 

Anteriormost portion of ctenidium straight, 
extending slightly farther anteriorly than os- 
phradium. Anterior ctenidial lamellae wider 
than deep or as wide as deep; posterior 
lamellae as wide as deep. Lateral edge of 
ctenidial lamellae varying from strongly con- 
vex to straight; ventral edge straight. Distal 
tips of ctenidial support rods extending be- 
yond lateral edge as papillalike projections. 

Vaginal opening round with slightly swollen 
surrounding edges and located below and 
posterior to anus. Bursa copulatrix a large di- 
verticulum, connected to vagina by wide ven- 
tral passage. Ventral channel formed by two 
small interlocking flanges located under ven- 
tral lobe of capsule gland, one arising from left 
lobe, the other from ventral epithelium. Albu- 
men gland arch-shaped, elongate. Single- 
chambered ingesting gland extending be- 
tween capsule gland and albumen gland. 
Ovary yellow to light golden in living speci- 
mens. Pseudo-penis usually present in fe- 
males. 

Penis dorso-ventrally flattened, straight or 
lightly curved, and with abruptly tapering, 
papillalike end. Penial vas deferens as 
minute, simple duct, semi-closed by overlap- 
ping ventral and dorsal sides of penis. Ceph- 
alic vas deferens well developed. Prostate 
gland bilobed, white, with dorso-ventral slit 
partially open to mantle cavity. Vas deferens 
poorly developed, whitish, separated from 
rectum by epithelial layer. Testis light brown 
to golden in living specimens. 



PHYLOGENY OF RAPANINAE 



201 



Paired accessory salivary glands extremely 
long, usually longer than one-half of shell 
height; left gland intertwined with salivary 
gland mass, right one separate from salivary 
gland mass and located in right anterior cor- 
ner of buccal cavity. Salivary gland mass in 
center of dorsal buccal cavity between gland 
of Leiblein and short, pear-shaped valve of 
Leiblein. Salivary ducts attached to anterior 
portion of esophagus at some distance from 
valve of Leiblein. Glandular folds on mid- 
esophagus indiscernible. Duct between mid- 
esophagus and gland of Leiblein short, thick. 
Esophagus attached to left side of gland of 
Leiblein in horseshoe-shape. Gland of 
Leiblein spiral, of hard consistency, yellowish. 
Posterior blind duct very short, with terminal 
ampulla. 

Stomach tubular, with 8-12 large folds on 
stomach wall oriented toward center of stom- 
ach. Stomach typhlosole extending upwards 
on left portion of posterior mixing area. Intes- 
tinal typhlosole thick, wide. Two digestive di- 
verticula present. Large papilla lying over 
equally large anal opening. Rectal gland 
sometimes not apparent. 

Radula: About 30-35% of shell height (Fig. 
15E, F). Rachidian widening dramatically 
from cusp bases toward base of rachidian; 
central cusp of rachidian thin, somewhat con- 
stricted at base; inner lateral denticle low on 
base of lateral cusp, and occasionally bifur- 
cate; straight outer edge of lateral cusp with 
several short denticles at base; base of lateral 
cusp adjacent to base of large marginal cusp; 
marginal cusps in different plane than lateral 
cusps (about 75° angle) and parallel to elon- 
gate lateral extension at base of rachidian 
tooth, resulting in bifid rachidian edge. Lateral 
teeth shorter than rachidian width. 

Egg Capsules: Oval-elongate, vase-shaped, 
up to about 9 mm in height, 3 mm in width, 
each attached with short, thin base about 1 
mm long. Apex tapered with central exit hole. 
Capsules deposited some distance from 
other capsules but interconnected by base. 
Each capsule contains up to 600 embryos, 
94% of them being nurse eggs (Crothers, 
1985). 

Ecology: Probably more is known about 
Nucella ecology than that of any other muri- 
cid. Nucella lapillus and its western American 
congeners have been the topic of many com- 
prehensive studies (Kincaid, 1957; Crothers, 
1985) and Ph.D. dissertations (Emien, 1966; 



Spight, 1972; Etter, 1987). Nucella feeds on 
barnacles and mussels (Largen, 1967; Mur- 
doch, 1969; Connell, 1970; Crothers, 1973; 
Spight, 1982) in the rocky intertidal zone and 
is eaten by crabs and birds (Spight, 1976). 
Moore (1 938) reported winter and spring to be 
the main spawning period. 

Studies show that environmental factors 
(wave action, food availability, etc.) drastically 
influence shell morphology (Cooke, 1895; Ag- 
ersborg, 1929; Colton, 1922; Moore, 1936). 

Distribution: North Atlantic Ocean from 
southern Portugal to Novaya Zemblya 
[records from the western Mediterranean 
(Nordsieck, 1968, 1982), Azores, Morocco, 
Senegal, and Canary Islands (Adanson, 
1757) are highly suspect (Cooke, 1915) and 
need confirmation]; Great Britain; Ireland; Ice- 
land; Greenland; New Jersey, U.S.A., to 
northern Canada (Abbott, 1974) (For exten- 
sive list of geographical range and localities, 
see Cooke, 1915.) 



Genus Pinaxia H. & A. Adams, 1853 
(Fig. 16A-E) 

Pinaxia H. & A. Adams, 1853: 132. 
Conothais Kuroda, 1930: 1 [type: Conothais 
citrina Kuroda, 1930, by monotypy]. 

Type Species: Pinaxia coronata H. & A. Ad- 
ams, ex A. Adams MS, 1853, by monotypy, = 
Pinaxia versicolor (Gray, 1839); synonyms: 
Pyrula versicolor Gray, 1839; ?Conothais cit- 
rina Kuroda, 1930. 

Remarks: Cossmann (1903: 68) allocated 
section status to Pinaxia under lopas (lopas) 
[= Nassa], whereas Thiele (1929: 297) used 
Pinaxia as a section of Thais {Thais). Wenz 
(1941: 1121) allotted subgeneric status to 
Pinaxia under Thais. Fujioka (1985a: 242) 
considered Conothais congeneric with 
Pinaxia. I agree with Fujioka based on inter- 
grades between Conothais citrina and 
Pinaxia versicolor. 

Shell: Protoconch (Fig. 16C, D) tall, conical, 
of about four adpressed whorls, with small 
subsutural plicae and several microscopic 
pustules (last whorl), and with outward-flaring 
lip and sinusigeral notch. Teleoconch (Fig. 
16A, B) small, conical to bulbous, smooth, of 
4-6 adpressed whorls. Adult shell up to about 
25 mm in height, 15 mm in width, with thin, 



202 



KOOL 




FIG. 16. Pinaxia versicolor. A, shell (17 mm), apertura! view. B, shell (17 mm), abapertural view. C, proto- 
conch, apical view, SEM (bar = 0.10 mm). D, protoconch, side view, SEM (bar = 0.10 mm). E, radula, SEM 
(bar = 10 |xm). 



cream brown periostracurn. Body whorl about 
90% of shell height, smooth, usually with 
heavy shoulder with 6-7 inconspicuous wide 
swellings or knobs. Aperture about 80% of 
shell height, elongate, narrow. Upper part of 
thin apertural lip nearly straight, lower end 
curved. Apertural lip with elongate (4-6 mm) 
riblets starting about one mm from edge. An- 
terior siphonal canal a poorly developed 
notch; posterior siphonal canal absent. Col- 
umella nearly straight, margin rounded, with 
little callus. Siphonal fasclole forming thin, 
slightly elevated ridge adjacent to callus on 
lower columella. Shell yellow to orange with 
10-1 1 thin, continuous or discontinuous, spi- 



ral, dark brown bands (although banding pat- 
tern may be absent); apertural lip and col- 
umella yellow to orange brown. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (10-15%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (70-75%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (15-25%). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free side with 
bracket-shaped growth rings; attached side 
without or with 1-2 bracket-shaped growth 



PHYLOGENY OF RAPANINAE 



203 



lines and with callused, glazed rim (about 30- 
45% of opercular width) on left. 

Anatomy (based on poorly preserved ani- 
mals only): Head-foot predominantly brown, 
uniform black at periphery. Cephalic tentacles 
elongate, brown dorso-centrally, black on pe- 
riphery, and with white tips. Mantle edge sim- 
ple, smooth, following contour of aperture, 
and brown on inside. Siphon long, brown with 
white specks, extending substantial distance 
beyond mantle edge. Large accessory boring 
organ dorsal to ventral pedal gland in females 
(Fig. 4B). 

Osphradium and ctenidium about equal in 
length; both about equal in width. Osphra- 
dium symmetrical in shape along lateral and 
longitudinal axes. Osphradial lamella at- 
tached along small portion of their base. 

Anteriormost portion of ctenidium bending 
towards anterior portion of osphradium; both 
equidistant from mantle edge. Anterior ctenid- 
ial lamellae wider than deep; posterior lamel- 
lae as deep as wide. Lateral and ventral 
edges concave. 

Vaginal opening below and posterior to 
anal opening. Ventral channel located near 
left side of capsule gland, consisting of single, 
hooked flange which originates from ventral 
epithelium. Large ventral lobe in anterior por- 
tion of capsule gland. Ingesting gland be- 
tween capsule gland and albumen gland. Al- 
bumen gland omega-shaped, large, tilted 
backwards. Low number of white seminal re- 
ceptacles on dorsal side of albumen gland. 

Penis large, slightly recurved, dorso- 
ventrally flattened, elongate, with flagelliform 
tip. Penial vas deferens as central duct- 
within-a-duct system occupying about one- 
third of penis width. Cephalic vas deferens 
a well-developed duct-within-a-duct system, 
inconspicuous from outside. Prostate small, 
closed, solid, yellow, lacking prominent duct, 
adjacent to narrow, white-walled rectum. 
Seminal vesicles well developed, golden, or- 
ange or white. 

Proboscis thinner than gland of Leiblein, 
unpigmented. Paired accessory salivary 
glands stubby, club-shaped, short, of equal 
length, much less than one-half of shell 
height; left gland completely loose from sali- 
vary gland mass; right accessory salivary 
gland adpressed to salivary gland mass. Sal- 
ivary glands soft, cottonlike, located dorsally 
in buccal cavity, larger than accessory sali- 
vary glands. Valve of Leiblein elongate, adja- 
cent to salivary gland mass and nerve ring. 



and with cap structure on anterior end. Sali- 
vary ducts attached to anterior portion of 
esophagus at base of valve of Leiblein. Por- 
tion of mid-esophagus with glandular folds 
long; folds poorly developed. Duct between 
gland of Leiblein and esophagus as thick as 
or thicker than posterior esophagus. Esopha- 
gus free from gland of Leiblein. Gland of 
Leiblein spiral, forming one fold between two 
attached lobes, with central hole for passage 
of anterior aorta, of hard consistency, yellow, 
with strawlike outer membrane. Posterior 
blind duct of gland of Leiblein nearly equal in 
length to gland itself. 

Tubular stomach with about ten folds. Rec- 
tal gland not apparent. Small anal opening on 
tubular extension of rectum. Anal papilla ab- 
sent. 

Radula: Ribbon length about 20-25% of 
shell height (Fig. 16E). Central cusp on 
rachidian tooth thin, needle-shaped, straight 
or bent to either side (artifact?); small back- 
ward extension present at central cusp base 
close to rachidian base; inner lateral denticle 
on lower half of lateral cusp; outer edge of 
lateral cusp straight, with one outer denticle 
on base of lateral cusp, three more well-de- 
veloped denticles on wide, horizontal mar- 
ginal edge; lateral cusps nearly equal in 
length to central cusp; large marginal cusp 
more than one-half of lateral cusp length; lat- 
erally extending lobe on rachidian edge and 
rachidian base somewhat widened antero- 
posteriorly. Lateral teeth slender with wide 
bases, hooked at distal ends, and longer than 
one-half of rachidian width. 

Egg Capsules: Unknown. 

Ecology: Pinaxia versicolor lives on intertidal 
sandflats with rocks and algae. Rehder & 
Ladd (1973) reported this species from the 
subtidal zone. 

Distribution: Indo-Pacific, from Mauritius (Dri- 
vas & Jay, 1 987) to Japan (Abbott & Dance, 
1982). 

Genus Plicopurpura Cossmann, 1903 
(Fig. 17A-F) 

Plicopurpura Cossmann, 1903: 69 (as section 
of Purpura). 

Microtoma Swainson, 1840: 72 (non Laporte, 
1832) [type: Buccinum patulum Lin- 
naeus, 1785, by subsequent designation, 
Herrmannsen, 1847:42, = Plicopurpura 
patula (Linnaeus, 1758)]. 



204 



KOOL 




A '^ 




В 




FIG. 17. Pliœpurpura patula. A, shell (53 mm), apertura! view. B, shell (53 mm), abapertura! view. C, 
protoconch, side view, SEM (bar = 70 ц.т). D, protoconch, apical view, SEM (bar = 0.10 mm). E, radula, 
SEM (bar = 20 ji,m). F, shell ultrastructure, SEM (bar = 0.15 mm). 



PHYLOGENY OF RAPANINAE 



205 



Purpurella Dall, 1871: 110 {non Robineau- 
Desvoidy, 1853, nee Bellardi, 1883; as 
subgenus of Purpura) [type: Purpura col- 
umellaris Lamarck, 1816, by original des- 
ignation, = Plicopurpura columellahs 
(Lamarck, 1816)]. 

Microstoma Paetel, 1875: 126 (error for Mi- 
crotoma Swainson). 

Patellipurpura Dall, 1909: 50 [type: Buccinum 
patulum Linnaeus, 1758, by monotypy, 
= Plicopurpura patula {Unnaeus, 1758); 
as section of Thais]. 

Patellapurpura Abbott, 1974: 180 (error for 
Patellipurpura Dall). 

Type Species: Purpura columellaris Lama- 
rck, 1816, by original designation, = Pli- 
copurpura columellaris (Lamarck, 1816); syn- 
onyms: IBuccinum patulum Linnaeus, 1758; 
Haustrum dentex Perry, 1811 [nomen obli- 
tum; ICZN, Opinion 886, 1969: 129]; Purpura 
pansa A. A. Gould, 1853. 

Remarks: Cossmann (1903: 69) introduced 
Plicopurpura, because the earlier name, Pur- 
purella Dall, was preoccupied. Dall (1909: 50) 
erected Patellipurpura for the Caribbean spe- 
cies patula, which lacks a columellar fold as 
found In Plicopurpura and placed both Patel- 
lipurpura and Plicopurpura as sections under 
Thais. Thiele (1929: 296) followed Cossmann 
in recognizing Plicopurpura and Purpura s.s. 
as sections of the genus Purpura, and synon- 
ymized Patellipurpura with Purpura s.s. (see 
below). Wenz (1941: 1115) accorded full ge- 
neric status to Plicopurpura and included Pli- 
copurpura and Patellipurpura as subgenera. 
Keen (1971b: 552) indicated that Plicopur- 
pura is perhaps a nodose subgenus of Pur- 
pura. Kool (1 988b) showed that Plicopurpura is 
sufficiently different from Purpura to warrant 
separate generic status. 

Traditionally three species/subspecies 
were included in this genus: Plicopurpura col- 
umellaris, P. patula, and P. patula pansa. Pli- 
copurpura patula occurs in the Caribbean 
Province and has been separated from pop- 
ulations in the eastern Pacific since the clo- 
sure of the Isthmus of Panama; based on the 
fact that P. patula no longer Interbreeds with 
P. columellaris in nature, I consider these two 
taxa separate species on the basis of inter- 
rupted gene flow. Keen (1971b: 552) allotted 
full species status to the two eastern Pacific 
species: P. columellaris and P. pansa. How- 
ever, Wellington & Kurls (1983) provided ev- 
idence for conspecificity of these two nominal 
species. I suspect this species complex to 



consist of two species: one in the Caribbean, 
the other in the eastern Pacific (see "Re- 
marks" under treatment of Stramonita). Mo- 
lecular data may demonstrate the actual de- 
gree of divergence. 

Shell: Protoconch (Fig. 17C, D) moderately 
tall, conical, of about 2.25 adpressed whorls, 
with numerous faint subsutural plicae and mi- 
croscopic pustules (last whorl), with outward- 
flaring lip and sinusigeral notch. Teleoconch 
(Fig. 17A, B) large, oval, of 5-6 adpressed 
whorls, and with high whorl-expanslon rate. 
Adult shell up to about 85 mm In height, 55 
mm in width. Body whorl dome-shaped, about 
90% of shell height. Body whorl sculptured 
with 7-8 spiral rows of nodules (most pro- 
nounced and nearly spinelike on many juve- 
nile specimens) with four small striae be- 
tween rows. Aperture wide, oval, about 80% 
of shell height. Apertural lip smooth on inside, 
crenate on edge, corresponding to pattern of 
striae on outside. Anterior siphonal canal a 
poorly developed notch; posterior siphonal 
canal well developed in older specimens. Col- 
umella flattened, wide, with acute angle of 
135° in lower portion. Siphonal fascicle a 
slightly elevated uneven ridge. Shell grey 
white to light brown; apertural lip white, with 
darker areas indicating dark pattern on out- 
side surface; edge of lip caramel brown, with 
blotched dark brown crenulations; columella 
caramel brown (sometimes partially white) 
frequently with sizable dark brown upper pa- 
rietal blotch. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (30-35%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (10-15%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (60-70%) (Fig. 17F). Presence of cal- 
cific layer questionable; scored with "?" in cla- 
distic analysis. 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface with about 4-6 arch- and bracket- 
shaped growth lines and with callused, glazed 
rim (about 30-35% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals; Fig. ЗА): Head-foot nearly uniform 
black. Elongate cephalic tentacles black ex- 
cept for white distal tips. Grooved sole of foot 
yellowish. Mantle edge slightly crenate, fol- 
lowing aperture contours. Incurrent siphon 



206 



KOOL 



black, extending beyond mantle edge. Pedal 
gland combined with well-developed acces- 
sory boring organ (Fig. 4B). 

Osphradial length about one-half ctenidial 
length; osphradial width about one-fifth 
ctenidial width. Osphradium symmetrical in 
shape along lateral and longitudinal axes. Os- 
phradial lamellae attached along small por- 
tion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae much wider 
than deep; posterior lamellae about as deep 
as wide. Lateral and ventral edge of ctenidial 
lamellae varying from concave to convex. Dis- 
tal tips of ctenidial support rods extending be- 
yond lateral edge as papillalike projections. 

Vaginal opening situated on distal end of 
loose, tubular extension of palliai gonoduct, 
curled towards mantle or toward buccal mass, 
and located below and posterior to anal open- 
ing. Bursa copulatrix a dorso-ventral chamber 
connecting with vagina, continuous with cap- 
sule gland. Small ventral lobe in anterior por- 
tion of capsule gland, lying over ventral chan- 
nel, which is formed by small, heavily ciliated, 
circular flange with longitudinal folds and 
grooves. Capsule gland embedded in spongy 
connective tissue. Posteriorly, ventral sperm 
channel divided into two branches: one uncil- 
iated, leading into ingesting gland; the other 
ciliated, leading to albumen gland. Albumen 
gland omega-shaped. Ingesting gland single- 
or double-chambered, extending from poste- 
rior lower left part of capsule gland to left of 
anterior part of albumen gland. Seminal re- 
ceptacles located at dorsal periphery of ante- 
rior portion of albumen gland. Females occa- 
sionally with minute pseudo-penis. 

Penis large, strongly recurved, oval in cross 
section, tapering distally or with extended, 
flagelliform tip. Penial vas deferens as duct- 
within-a-duct system occupying about one- 
seventh of penial width. Cephalic vas defer- 
ens thin, inconspicuous, in straight line from 
penis to prostate. Prostate closed, directly ad- 
jacent to rectum, both embedded in opaque 
spongy connective tissue. Seminal vesicles 
well developed, brown. 

Proboscis moderately muscular, one-half of 
gland of Leiblein width, semi-transparent, with 
pink odontophores (visible in living speci- 
mens). Paired salivary glands usually equal in 
length (but right accessory salivary gland oc- 
casionally shorter); both glands elongate, 
thin, adjacent to salivary glands, about one- 
third of shell height. Salivary glands often 



joined, globular in appearance, larger than 
accessory salivary glands. Salivary ducts at- 
tached to anterior portion of esophagus at 
some distance from valve of Leiblein. Anterior 
portion of esophagus widened, forming elon- 
gate valve of Leiblein, adjacent to salivary 
glands. Portion of mid-esophagus with glan- 
dular folds short, swollen; folds poorly devel- 
oped. Duct between mid-esophagus and 
gland of Leiblein well-developed, about equal 
to posterior esophagus width. Posterior 
esophagus adjacent to gland of Leiblein, con- 
nected to it by connective tissue, or separate. 
Gland of Leiblein spiral, forming two lobes 
with dorso-ventral opening for anterior aorta, 
caramel brown, covered with thick, strawlike 
outer membrane. Posterior blind duct of gland 
of Leiblein narrow, elongate, longer than 
gland itself, and entering dorsal branch of af- 
ferent renal vein. 

Stomach tubular, with small posterior mix- 
ing area with about ten large folds on right 
two-thirds of interior stomach; left portion 
smooth. Two digestive diverticula present. 
Stomach typhlosole and intestinal typhlosole 
thin. Rectal gland long, thin, dark green, ad- 
jacent to entire length of capsule gland. Rec- 
tum large in diameter, embedded in spongy 
connective tissue without separation from 
capsule gland or rectum by epithelial layer. 
Anal opening small, well defined, with distinct 
anal papilla. 

Radula: Ribbon length about 45% of shell 
height (Fig. 17E). Central cusp of rachidian 
tooth elongate, needle-shaped, with slightly 
widened base and elongate median slit in 
central cusp extending from base of rachidian 
to slightly below tip; small inner lateral denti- 
cle separate from but directly adjacent to cen- 
tral and lateral cusps; lateral cusps smooth, 
with concave outer edge and convex inner 
edge; outer edge of lateral cusp sloping 
steeply down to rachidian base. Lateral teeth 
thin, strongly curved, equal in length to 
rachidian width. 

Egg Capsules: Flat and rounded, up to about 
4 mm in width; flat, round top of capsule with 
central, circular exit hole. Each capsule con- 
taining 50-100 eggs measuring about 0.24 
mm in diameter (Lewis, 1960). These data 
are very different from descriptions given by 
Kool (1989) of Plicopurpura columellahs. Be- 
cause the descriptions of Kool are based on 
specimens that were collected without the an- 
imal that laid them (ANSP 324406), they are 
probably based on eggs of a different spe- 



PHYLOGENY OF RAPANINAE 



207 



cies. The explanation that the egg capsule 
morphology of the two species is very differ- 
ent appears less likely. 

Ecology: Plicopurpura patula occurs from the 
splash zone and low intertidal to shallow sub- 
tidal, on hard substrates (often limestone plat- 
forms) in high-energy environments. It feeds 
on such mollusks as chitons (Clench, 1947; 
Lewis, 1960; Bändel, 1987; Kool, 1987) and 
nerites (Britton & Morton, 1989), and also on 
barnacles (Lewis, 1960; Kool, 1987). As de- 
scribed by Bändel (1987), Plicopurpura para- 
lyzes a chiton with a purple staining secretion, 
pulls it off the substrate, and, while holding it 
with its foot, eats it. Bändel noted that Pli- 
copurpura feeds in the splash zone because 
the paralyzing secretion would lose much of 
its effect by dilution when the animal is sub- 
merged. However, many rapanines are 
known to paralyze their prey, yet feed when 
submerged (Kool, personal observation). 
Breeding occurs in August and September 
(Lewis, 1960). 

Distribution: Western Atlantic, from central 
east Florida throughout West Indies to Brazil 
and Bermuda (Abbott, 1 974). Occurrence of a 
Plicopurpura-Wke shell on Mauritius (Drivas & 
Jay, 1987) needs further investigation. 

Genus Purpura Bruguière, 1789 
(Fig. 18A-G) 

Purpura Bruguière, 1789: 15 {non Röding, 
1798, пес Lamarck, 1799). 

Type Species: Buccinum persicum Lin- 
naeus, 1758, by subsequent designation, 
ICZN, Opinion 886, 1969: 128, = Purpura 
pérsica (Linnaeus, 1758); synonym: IPurpura 
inerma Reeve, 1846. 

Remarks: The generic name "Purpura" was 
first used by Martini (1777) and subsequently 
by Martyn (1784) and Meuschen (1787), all of 
which are non-binominal works. Bruguière 
formally introduced Purpura as a genus in 
1789, but did not mention any species. Three 
years later, Bruguière (1792) included the 
nominal species Purpura tubifer Bruguière, 
1792, which would make this the type species 
by subsequent monotypy. Unfortunately, this 
taxon is now regarded as a species of Typhis 
Montfort, 1810 (Muricidae: Typhinae). Later, 
Lamarck (1799, 1801) cited P. pérsica as the 
sole species in the genus, which did not result 
in P. pérsica being the type species by mono- 
typy, as Bradley & Palmer (1963: 252) incor- 



rectly stated it to be. To resolve this matter, 
Bradley & Palmer (1963) and Keen (1964) 
proposed, by petition to the International 
Committee of Zoological Nomenclature, that 
Purpura pérsica be designated type species 
of Purpura. Purpura pérsica officially became 
the type of Purpura after publication of ICZN, 
Opinion 886 (1969). Detailed nomenclatural 
history on this genus is given by Dall (1905), 
Winckworth (1945), Dodge (1956), Bradley 
and Palmer (1963), and Keen (1964). 

Cossmann listed Purpura pérsica as the 
sole example of the genus Purpura. Thiele 
(1929: 296) incorrectly cited Purpura patula 
as type of Purpura, and synonymized Patel- 
lipurpura with this genus. He recognized the 
sections Purpura and Plicopurpura (type spe- 
cies Purpura columellaris Lamarck, 1816). 
Wenz (1941: 1125), and later Pchelintsev & 
Korobkov (1960: 207), used Plicopurpura 
Cossmann for Purpura s.l., and Purpura Mar- 
tyn for the muhcine "Purpura" foliata. Keen 
(1971b: 552) synonymized the genera Pli- 
copurpura and Patellipurpura with Purpura. 
Kool (1 988b) argued for separation of Plicopur- 
pura and Purpura. 

Shell: Protoconch (Fig. 18C, E) tall, conical, 
of about three adpressed whorls [exact count 
could not be made from available specimen] 
with outward-flaring lip and sinusigeral notch. 
Sculptural pattern unknown (due to erosion). 
Teleoconch (Fig. 18A, B) with high whorl ex- 
pansion rate, large, heavy, oval, of about six 
adpressed whorls. Adult shell up to about 115 
mm in height, 90 mm in width. Body whorl 
dome-shaped, about 95% of shell height, 
sculptured with minute spiral grooves and 
7-1 5 slightly elevated spiral ridges, with one 
to several less elevated, thinner ridges in be- 
tween these; surface shiny, appearing 
smooth. Aperture very wide, oval, about 85% 
of shell height. Anterior siphonal canal short, 
wide, open; posterior siphonal canal deep, 
well developed. Apertural lip smooth, crenate 
towards edge, corresponding with outside 
groove pattern. Columella flat to concave, 
wide with moderate callus layer, with angular 
curve in lower portion of columella bordering 
wide, shallow anterior siphonal canal. Sipho- 
nal fasciole a slightly elevated ridge, adjacent 
to columellar callus. Shell grey brown; spiral 
ridges with color pattern of alternating dark 
brown and white; dark brown portions of up- 
per two ridges often elevated to form spiral 
cords of small beads; apertural lip bluish 
white, with about 30 spiral, dark brown lines 



208 



KOOL 




FIG. 18. Purpura pérsica. A, shell (61 mm), apertural view. B, shell (61 mm), abapertural view. C, proto- 
conch, side view, SEM (bar = 0.10 mm). D, radula, SEM (bar = 50 ц-т). E, protoconch, apical view, SEM 
(bar = 0.10 mm). F, shell ultrastructure, sawed surface, SEM (bar = 0.25 mm); a, aragonite (crystal planes 
oriented in 45° angle to growing edge); b, aragonite (crystal planes oriented perpendicular to growing edge); 
c, aragonite (crystal planes oriented parallel to growing edge); d, aragonite (crystal planes oriented perpen- 
dicular to growing edge); e, calcite. G, detail of fracture zone of layer b (Figure 18F), SEM ( x 700). 



PHYLOGENY OF RAPANINAE 



209 



continuing far into the aperture, with almost 
uniform, narrow (5-10 mm), black band along 
edge; columella orange on inside, with 
blotches of dark brown, cream and blue grey 
on upper parietal region. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45° angle to growing 
edge (Fig. 18F, a) (15-25%); aragonitic layer 
with crystal planes oriented perpendicular to 
growing edge (Fig. 18F, b, G) (20-25%); ara- 
gonitic layer with crystal planes oriented par- 
allel to growing edge (Fig. 18F, c) (35-55%); 
aragonitic layer with crystal planes oriented 
perpendicular to growing edge (Fig. 18F, d) 
(5-15%); calcific layer (5-10%) (Fig. 18F, e). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface with about 1-2 bracket-shaped 
growth lines and with callused, glazed rim 
(about 35-40% of opercular width) on left. 

Anatomy (based on preserved animals only): 
Head-foot region flecked with dark brown to 
black (often in vertical striae) on light yellow 
background. Elongate tentacles dark brown 
with light yellow tips. Mantle edge straight, 
smooth, unpigmented. Incurrent siphon 
brown black, extending some distance be- 
yond mantle edge. Anterior lobes of foot light 
brown. Kidney yellowish, not distinct. Acces- 
sory boring organ minute, dorsal to pedal 
gland and located in anteriormost portion of 
foot. 

Osphradial length about one-half ctenidial 
length; osphradial width between one-fourth 
and one-third ctenidial width. Osphradium 
symmetrical in shape along lateral and longi- 
tudinal axes, occasionally more tapered ante- 
riorly. Osphradial lamellae attached along 
small portion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae much wider 
than deep; posterior lamellae deeper than 
wide. Lateral edge of ctenidial lamellae vari- 
able; ventral edge concave. 

Vaginal opening on tubular extension of 
palliai gonoduct and located directly below 
anal opening. Small bursa copulathx a hori- 
zontal slit open to vagina and continuous with 
capsule gland. Minute ventral sperm channel 
formed by semi-circular flange originating 
from the ventral epithelium, located under 
ventral lobe. Ventral lobe initially small, be- 
coming larger posteriorly, finally disappear- 



ing. Posterior ventral channel with one minute 
flange below larger flange. Lower half of cap- 
sule gland opaque; upper portion yellow or- 
ange, flocculent. Ingesting gland with several 
to many sizable chambers surrounded by 
loose, white connective tissue, extending 
from left side of capsule gland to albumen 
gland. Albumen gland omega-shaped, tilted 
onto posterior half. Seminal receptacles on 
dorsal periphery of albumen gland. Ovary 
light brown. 

Penis large, strongly recurved, and flat- 
tened dorsoventrally at distal end, with large 
flagellar papilla curved along shaft. Penial 
duct as duct-within-a-duct system occupying 
one-third of penial width. Cephalic vas defer- 
ens meandering towards prostate. Prostate 
closed, large, similar to capsule gland in fe- 
males; embedded in spongy tissue, not dis- 
tinctly separated from rectum. Small, dark 
brown seminal vesicles. 

Proboscis very large, larger than gland of 
Leiblein, connected to dorsal wall of buccal 
cavity with small muscle bundles. Paired ac- 
cessory salivary glands elongate, thin, equal 
in length, less than one-half of shell height; 
right accessory salivary gland loose in right 
anterior buccal cavity; left gland partially ad- 
jacent to salivary gland. Very large salivary 
glands nearly equal in size to gland of Leiblein 
and partially located below proboscis. Sali- 
vary ducts attached to anterior portion of 
esophagus close to anterior part of valve of 
Leiblein. Salivary gland mass partially ventral 
to proboscis. Valve of Leiblein thin, elongate, 
adjacent to salivary glands. Portion of mid- 
esophagus with glandular folds long. Duct be- 
tween mid-esophagus and gland of Leiblein 
nearly equal in diameter to posterior esopha- 
gus. Posterior esophagus embedded in lower 
left portion of gland of Leiblein. Gland of 
Leiblein spiral, forming two folds, of hard con- 
sistency, thick, light caramel brown, with 
strawlike outer membrane. Blind posterior 
duct of gland of Leiblein much longer than 
gland itself. 

Stomach with large, deep posterior mixing 
area. Three-fourths of whole posterior mixing 
area occupied by 25 small folds; anterior one- 
fourth (adjacent to intestine) smooth, proba- 
bly non-ciliated. Two large digestive divertic- 
ula present. Stomach typhlosole thin. 
Intestinal typhlosole absent. Rectum thick- 
walled dorsally, with small internal longitudi- 
nal folds; rectum embedded in spongy tissue, 
separated from capsule gland by distinct layer 
of epithelium. Anal opening distinct, with up- 



210 



KOOL 



ward-pointing papilla at anal opening. Rectal 
gland moderately wide, extending along en- 
tire length of capsule or prostate gland; gland 
green in females, but usually pink with traces 
of green in males. 

Radula: Ribbon length about 30-35% of 
shell height (Fig. 18D). Rachidian wide, with 
needle-shaped central cusp; straight lateral 
cusps nearly equal in width to central cusp; 
with or without (can vary within same speci- 
men) single minute denticle on base of inner 
edge of lateral cusp; outer edge of lateral 
cusp with one denticle on base; 4-7 well-de- 
veloped, long, thin denticles on horizontal 
marginal area; very well-developed marginal 
cusp nearly equal in size to lateral cusps. Lat- 
eral teeth smooth, slightly curved, about 
three-fourths of rachidian width. 

Egg Capsules: Short, dirty yellow, up to 6 
mm in height, 5 mm in width, each with flat, 
widened base; bases usually confluent, cap- 
sules occasionally deposited on top of one 
another; flat, oval top of capsule with central, 
circular exit hole. Each capsule containing ap- 
proximately 160-200 eggs measuring about 
0.2 mm in diameter (Tirmizi & Zehra, 1983). 

Ecology: This species occurs in the rocky 
subtidal zone (Tirmizi & Zehra, 1983), often in 
high energy environments (B. Smith, personal 
communication), where it feeds, among other 
items, on limpets, as determined from doco- 
glossate rachidian teeth found in gut-content 
analysis. 

Distribution: Indo-Pacific, from Mauritius (Dri- 
vas & Jay, 1 987) to Marquesas Islands (Sal- 
vat & Rives, 1975). 

Genus Stramonita Schumacher, 1 81 7 
(Fig. 19A-F) 

Stramonita Schumacher, 1817: 68, 226. 

Type Species: Buccinum haemastoma Lin- 
naeus, 1767, by subsequent designation, 
Gray, 1847: 138, = Stramonita haemastoma 
(Linnaeus, 1767); synonyms: Thais grísea 
Röding, 1798; Thais metallica Röding, 1798; 
Thais nebulosa Röding, 1798; Thais stellata 
Röding, 1798; Purpura f leridana Conrad, 
1837; Purpura consul Reeve, 1846; Purpura 
forbesii Dunker, 1 853; Thais floridana haysae 
Clench, 1927; Thais (Stramonita) hidalgoi 
Coen, 1946; 7 Thais (Stramonita) langi 
Clench, 1948. 



Remarks: Most authors have considered 
Stramonita to be a subgenus of Thais Röding 
1798 (Cossmann, 1903: 68; Wenz, 1941 
1120; Woodring, 1959: 222; Keen, 1971b 
549). Thiele (1929: 297) placed Stramonita as 
a section of Thais s.S., genus Thais. Ko- 
robkov (1955: 299) considered Stramonita a 
subgenus of Thais. (Kool, 1987: 118) ac- 
corded Stramonita full generic status. Sub- 
specific status may be accorded to several of 
the taxa placed in synonymy with Stramonita 
haemastoma {"Thais" haemastoma haysae 
Clench, 1927; "Purpura" floridana Conrad, 
1837), but further anatomical, genetic (see 
Liu et al., 1991), and molecular studies are 
necessary prior to separation. Based on ex- 
periments in the laboratory, Bändel (1976: 
118) concluded that S. floridana is only an 
ecological form of S. haemastoma. 

The tropical eastern Pacific species Stra- 
monita biserialis (Blainville, 1832) deserves 
separate species status because it occurs on 
the west side of the Isthmus of Panama and 
has thus been genetically isolated from west- 
ern Atlantic populations for 2-3 million years 
(see "Remarks" under treatment of Plicopur- 
pura). 

Shell: Embryonic shell with pattern of spiral 
rows of microscopic, volcanolike, cone- 
shaped pustules. Protoconch (Fig. 19C, D) 
tall, conical of at least 3.5 adpressed whorls 
(exact count could not be made from avail- 
able specimen), with outward-flaring lip; si- 
nusigeral notch covered by teleoconch. First 
three whorls with faint shoulder with thin ridge 
sculptured with small plicae; last whorl with 
shoulder more pronounced and bearing nu- 
merous microscopic pustules; numerous 
small subsutural plicae on each whorl. Teleo- 
conch (Fig. 19A, B) highly variable, fusiform 
to more oval-shaped, of 7-8 whorls, with 
varying degree of prominence of suture. Adult 
shell up to about 90 mm in height, 55 mm in 
width. Body whorl about 75-85% of shell 
height, rounded or with distinct shoulder, 
sculptured with one or two spiral cords with 
faint knobs and with dense pattern of 30-40 
narrow but distinct ridges. Aperture moder- 
ately wide, about 60% of shell height. Aper- 
tura! lip with crenulations continuing into ap- 
erture as narrow, tall ridges. Anterior siphonal 
canal a short, wide notch; posterior siphonal 
canal present in many adult specimens, but 
poorly developed, flanked on left by small 
protrusion of columellar callus. Columella 
rounded, slightly curved, with little or no cal- 



PHYLOGENY OF RAPANINAE 



211 




FIG. 19. Stramonita haemastoma. A, shell (33 mm), apertura! view. B, shell (33 mm), abapertural view. C, 
protoconch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, radula, 
SEM (bar = 25 ц.т). F, Shell ultrastructure, fracture surface, SEM (bar = 0.15 mm). 



212 



KOOL 



lus. Siphonal fascicle directly adjacent to cal- 
lus, with spiral ridge as on rest of whorls. Shell 
flecked with dark brown, grey, and white, usu- 
ally forming semi-axial patterns; lower col- 
umella white to orange on callused region; 
upper columella with color pattern similar to 
that on outside of shell; apertural lip white to 
orange, with dark brown between distal ends 
of internal ridges and crenulations. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (10-20%) (lacking in some speci- 
mens); aragonitic layer with crystal planes ori- 
ented parallel to growing edge (30-40%); 
calcific layer (40-60%) (Fig. 19F). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface with about 3-5 bracket-shaped 
growth lines and with callused, glazed rim 
(about 30-35% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Head-foot mottled and blotched with 
grey black on white background. Cephalic 
tentacles uniform grey, with black tips. Large 
mantle covering total head-foot, crenate, with 
a few, caramel-brown antero-posterior elon- 
gate flecks on edge. Incurrent siphon very 
thick, short, mottled with grey black. Hypo- 
branchial gland pink. Accessory boring organ 
oval, 2 mm long, with duct (about 4 mm), lo- 
cated dorsal to pedal gland in females (Fig. 
4B). 

Osphradial length about one-third ctenidial 
length; osphradial width one-half ctenidial 
width. Osphradium symmetrical in shape 
along lateral and longitudinal axes, or slightly 
more tapered posteriorly. Osphradial lamella 
attached along small portion of their base. 

Anteriormost portion of ctenidium straight, 
extending farther anteriorly than osphradium. 
Anterior and posterior ctenidial lamellae wider 
than deep. Lateral edges of ctenidial lamellae 
varying from convex (anterior) to concave 
(posterior); ventral edges straight. 

Vaginal opening a simple hole situated on 
end of attached tubular extension of palliai 
gonoduct (in typical S. haemastoma morphs; 
in rounded morphs, vagina more elongate) 
and located below and slightly anterior to anal 
opening. Bursa copulathx extending along 
entire capsule gland and measuring one-half 
of gland height. Anterior part of bursa narrow, 
oriented dorso-ventrally, but circular posteri- 



orly, with intricately branching ridges. Well- 
developed ventral flange perpendicular to 
capsule gland lobes, originating from spongy, 
epithelial tissue on left side of capsule gland 
or from left lobe of capsule gland. Ingesting 
gland large, usually black, solid, with material 
similar to that found in rectal gland. Albumen 
gland arch-shaped, occasionally with anterior 
and posterior lobes disjunct to form arch, and 
with black or white seminal receptacles at pe- 
riphery. Small, pseudo-penis occasionally 
present in females. 

Penis in males thick, strongly recurved, 
blunt, dorso-ventrally flattened. Penial vas 
deferens as duct-within-a-duct system occu- 
pying about one-sixth of penial width. Ceph- 
alic vas deferens simple, running directly be- 
low epithelium. Prostate small, yellow, with 
wide central duct, adjacent to much larger 
rectum. 

Proboscis thin, long. Paired accessory sal- 
ivary glands elongate, of equal length, thin, 
one-third of shell height. Left accessory sali- 
vary gland adpressed to salivary gland mass, 
partially intertwined with it; right accessory 
salivary gland loose in anterior right buccal 
cavity, ventral to proboscis. Salivary gland 
mass equal in size to one accessory salivary 
gland, located in dorsal buccal cavity between 
gland of Leiblein and proboscis. Salivary 
ducts adjacent to esophagus directly anterior 
to valve of Leiblein. Portion of mid-esophagus 
with glandular folds long. f\/lid-esophagus di- 
rectly attached to gland of Leiblein. Gland of 
Leiblein of hard consistency, spiraled coun- 
terclockwise (forming two "folds" and three 
"lobes"), enveloped by thin strawlike mem- 
brane, varying in color from cream to light 
brown posteriorly to darker brown anteriorly. 
Posterior blind duct of gland of Leiblein long, 
about one-half of gland length, terminating in 
dorsal branch of afferent renal vein. Posterior 
esophagus loosely attached to left side of 
gland of Leiblein. 

Stomach large, with several large folds ori- 
ented toward intestine. Single large vertical 
fold with several thin ridges on both sides, 
perpendicular to and continuous with well-de- 
veloped stomach typhlosole. Two digestive 
diverticula present. Intestinal typhlosole well 
developed, continuing on stomach wall, de- 
marcating intestine from stomach. Several 
small ridges in intestinal canal. Ciliary move- 
ment on stomach wall directed toward intes- 
tine. Rectum very wide. Rectal gland green. 
Anal opening well developed, with pro- 
nounced anal papilla. 



PHYLOGENY OF RAPANINAE 



213 



Radula: Ribbon length about 25% of shell 
height (Fig. 19E). Rachidian with needle- 
shaped central cusp; lateral cusps with well- 
developed inner denticle high on cusp, occa- 
sionally with one or two additional denticle(s) 
below; outside edge of lateral cusp concave, 
with row of several well-developed denticles 
continuing up to large marginal cusp; rachid- 
ian base with lateral extension. Lateral teeth 
about equal in length to rachidian tooth. 

Egg Capsules: Vase-shaped, large, each 
with concave and convex sides, up to about 
13 mm in height, 2.5 mm in width. Apical plate 
usually flat or slightly concave, variable in 
contour, with round to oval, off-center exit 
hole. Two sutures extending from basal plate 
of each capsule to apical plate. Capsules ar- 
ranged in clusters, with concave sides adja- 
cent to convex sides and with confluent 
bases, each containing 150-800 embryos. 
Hatching occurs after about 15 days 
(D'Asaro, 1966). Boone (1984) reported a 
case of egg capsules attached to floating 
wood. 

Ecology: This species occurs in low- and 
high-energy intertidal environments. It also 
lives in mangrove habitats and on Phrag- 
matopoma reefs. It feeds on a variety of prey, 
such as mussels (Burkenroad, 1931), oysters 
(Bändel, 1976), barnacles (Cake, 1983), and 
polychaetes {Phragmatopoma sp.) (Kool, 
1987). A variety of ecological topics was 
treated by Gunter (1979). I found this species 
usually to be relatively inactive during low 
tide, but feeding when submerged at high 
tide. Females often congregate prior to 
spawning, which usually occurs from April to 
May. 

Distribution: Eastern Atlantic Ocean, from 
Mediterranean Sea to West Africa; western 
Atlantic Ocean, from North Carolina through- 
out the West Indies to Brazil (Abbott, 1974). 

Genus Thais Röding, 1798 
(Fig. 20A-F) 

Thais Röding, 1798: 54. 

IThalessa H. & A. Adams, 1853: 127 [type: 
Murex hippocastanum Linnaeus, 1758, 
by subsequent designation, F. С Baker, 
1895: 183 (Suppressed by ICZN, Opin- 
ion 911, 1970: 20), = Thais aculeata 
(Deshayes, 1844)]. 

7Menathais I rédale, 1937: 256 [type: Purpura 



pica Blainville, 1832, by original designa- 
tion, = Thais tuberosa {Roá\r\g, 1798)]. 

IThaisella Clench, 1947: 69 [type: Purpura 
trinitatensis Guppy, 1869, by original 
designation, = Thais trinitatensis 
(Guppy, 1869)]. 

IReishia Kuroda & Habe, 1971: 146 [type: 
Purpura bronni Dunker, 1861 , by original 
designation, = Thais bronni (Dunker, 
1861)]. 

Type Species: Murex fucus Gmelin 1791, by 
subsequent designation, Iredale, 1915: 472 
(ICZN, Opinion 886, 1969: 128), = Thais no- 
dosa (Linnaeus, 1758); synonyms: Nerita no- 
dosa Linnaeus, 1758 [in partem]; Murex neri- 
toideus Linnaeus, 1767 [in partem] [also cited 
as neritoides Linnaeus]; Thais lena Röding, 
1798; Thais meretricula Röding, 1798; Pur- 
pura ascensionis Ouoy & Gaimard, 1833. 

Remarks: Troschel (1866-1893: 130) placed 
Thais as a subgenus in the genus Stramonita. 
Cossmann (1903) did not list Thais. Thiele 
(1929: 297) included the following subgenera 
under the genus Thais: Mancinella, with sec- 
tions Mancinella, Neorapana and Tribulus; 
and Thais, with sections Thais, Stramonita, 
Cymia, Pinaxia, Trochia, and Agnewia. Wenz 
(1941: 1120) included the subgenera Stra- 
monita, Entacanthus, Cymia, Pinaxia, Tro- 
chia, and Agnewia under the genus Thais. 
Fujioka (1985a: 243) recognized both Reishia 
and Thaisella as subgenera of Thais. 

Iredale (1915: 472) provided a type species 
designation {"Thais neritoides = Murex fucus 
Gmel") in a synopsis of Dall's (1909) work. 
Stewart (1 927: 386) listed Thais fucus as type 
species of Thais but recognized Thais nodosa 
as a valid name by explaining that Murex neri- 
toideus was an unnecessary substitute for 
Nerita nodosa Linnaeus, both being based on 
the same figures. Stewart then synonymized 
the nominal species fucus, neritoideus, lena, 
and nodosa. In 1937 (p. 256) Iredale listed 
". . . Thais lena Bolten [sic] = Murex fucus 
Gmelin, . . ." as the type species, with this 
type species fixed as Murex fucus Gmelin, 
1 791 , by subsequent designation by Iredale 
(1915) (ICZN, Opinion 886, 1969: 128). Fur- 
thermore, the nominal species nodosa, the 
oldest available name, acquired official status 
in the same opinion. 

Thais nodosa meretricula from Ascension 
Island is herein considered synonymous with 
Thais nodosa nodosa. The number of black 
dots on the columella, often cited as a distinc- 
tive character for separating the two forms, is 



214 



KOOL 




FIG. 20. Thais nodosa. A, shell (45 mm), apertural view. B, shell (25 mm), abapertural view. C, protoconch, 
side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, shell ultrastructure, 
fracture surface, SEM (bar = 0.50 mm). F, radula, SEM (bar = 25 |xm). 



PHYLOGENY OF RAPANINAE 



215 



variable in both and shows overlap. Speci- 
mens from the African mainland are usually 
nodose, whereas most, but not all, specimens 
from Ascension Island are smooth. 



Shell: Protoconch (Fig. 20C, D) conical, of at 
least two adpressed whorls (exact count 
could not be made from available specimen), 
and with outward-flaring lip; sinusigeral notch 
covered by teleoconch. Sculptural pattern ob- 
scured by erosion, except for several micro- 
scopic pustules observed around lip region. 
Teleoconch (Fig. 20A, B) with high whorl ex- 
pansion rate, large, ovate to nearly round, of 
4-5 adpressed whorls. Adult shell up to about 
70 mm in height, 55 mm in width (form mer- 
etricula has the largest representatives). 
Body whorl dome-shaped, usually exceeding 
95% of shell height, occasionally with aper- 
ture reaching beyond apex. Thais nodosa 
form nodosa sculptured with five (sometimes 
four) spiral rows of 8-9 knobs (occasionally 
spinelike) and with about 35 narrow, low, spi- 
ral ridges, 4-6 of them between rows of 
knobs; knobs on second and third rows larg- 
est. Thais nodosa form merethcula with 
rounded body whorl sculptured with about 35 
narrow, low spiral ridges. Both forms with 
wide, oval aperture usually exceeding 95% of 
shell height. Apertural lip thick, with crenula- 
tions on edge corresponding to ridge pattern 
on outer surface; inside smooth and polished. 
Anterior siphonal canal as poorly developed 
notch; posterior siphonal canal poorly devel- 
oped in most specimens, well developed in 
others. Columella with wide, flat, heavily cal- 
lused parietal region and with moderately an- 
gular curve in lower region. Siphonal fasciole 
a well-developed ridge lying behind callus on 
lower parietal region. Shell dirty white to 
brown, columella white, with 1-4 large brown 
black spots (although overlap occurs, usually 
1-2 in Thais nodosa form nodosa; 3-4 in Г. 
nodosa form merethcula) arranged in vertical 
row; aperture and apertural edge white. 



Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45° angle to growing 
edge (30-50%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (5-15%); aragonitic layer with crystal 
planes oriented parallel to growing edge (20- 
25%); aragonitic layer with crystal planes ori- 
ented perpendicular to growing edge 
(5-10%); calcitic layer (5-10%) (Fig. 20E). 



Operculum: D-shaped, with lateral nucleus in 
center right (Fig. 1С). Free side with bracket- 
shaped growth lines; attached side with about 
4-6 bracket-shaped growth lines and with 
callused, glazed rim (about 30-35% of oper- 
cular width) on left. 

Anatomy (based on preserved animals only): 
Head-foot and long cephalic tentacles mottled 
with black. Mantle edge straight, simple, fol- 
lowing contour of aperture. Anterior siphon 
extending substantial distance beyond mantle 
edge. Sole of foot a pattern of pustules and 
ridges. Nephridial gland yellow. Kidney grey 
brown. Accessory boring organ dorsal to 
pedal gland in females (Fig. 4B). 

Osphradial length slightly more than one- 
half ctenidial length; osphradial width slightly 
less than ctenidial width. Osphradium sym- 
metrical in shape along lateral axis; right pec- 
tin distinctly wider than left one. Osphradial 
lamellae deeper than wide, attached along 
very small portion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae wider than 
deep; posterior lamellae deeper than wide. 
Lateral edge of ctenidial lamellae varying 
from concave (anterior) to straight or convex 
(posterior); ventral edge varying from slightly 
concave (anterior) to distinctly concave (pos- 
terior). 

Vaginal opening round, situated on poste- 
riorly curved tubular extension of palliai gon- 
oduct and located directly below anal open- 
ing. Ventral flange small, crescent-shaped, 
originating from ventral epithelium. Ventral 
channel under large ventral lobe. Ingesting 
gland on left and posterior sides of capsule 
gland. Several seminal receptacles on dorsal 
periphery of omega-shaped albumen gland. 

Penis strongly recurved, dorso-ventrally 
flattened, with short thick flagelliform tip (Fig. 
5D). Vas deferens as tube-within-a-tube sys- 
tem occupying about one-fifth of penial width. 
Prostate white yellow, embedded in spongy 
connective tissue, with closed duct, similar to 
capsule gland in females. Seminal vesicles 
pale yellow. 

Proboscis very large, about equal in width 
to gland of Leiblein. Paired accessory salivary 
glands thin, long, less than one-half of shell 
height; right gland usually few millimeters 
longer than left; left gland intertwined with sal- 
ivary gland mass, right gland free of salivary 
gland mass and located ventrally in anterior 
buccal cavity. Salivary gland mass in dorsal 



216 



KOOL 



buccal cavity. Valve of Leiblein small, elon- 
gate, adjacent to salivary gland mass. Sali- 
vary ducts attached to anterior portion of 
esophagus close to anterior part of valve of 
Leiblein. Duct between mid-esophagus and 
gland of Leiblein not pronounced. Posterior 
esophagus adjacent to lower left gland of 
Leiblein. Gland of Leiblein spiral, forming two 
folds, of hard consistency, dark brown with 
thin but distinct strawlike membrane. Poste- 
rior blind duct of gland of Leiblein more than 
one-half of gland length. 

Tubular stomach smooth or with many 
small folds oriented toward center. Stomach 
with two digestive diverticula, but without in- 
testinal typhlosoles (possibly not visible due 
to bad preservation). Rectal gland long, 
green. Anal opening small, indistinct, with 
anal papilla equal in size to opening. 

Radula: Ribbon length about 30% of shell 
height (Fig. 20F). Rachidian with wide central 
cusp; inner edge of lateral cusp straight to 
convex, with large denticle at base; outer 
edge of lateral cusp straight or concave, with 
1-2 small denticles on base; 1-2 more den- 
ticles on slightly sloping marginal edge; mar- 
ginal cusp large. Lateral teeth about equal in 
length to rachidian width. 

Egg Capsules: Unknown. 

Ecology: Thais nodosa lives in the rocky in- 
tertidal zone (Rios, 1970; Abbott & Dance, 
1982). 

Distribution: Eastern Atlantic, from western 
Africa (Bernard, 1984), to Ascension Island 
(Rosewater, 1975) and Cape Verde Islands 
(Nordsieck, 1968); western Atlantic, 
Fernando de Noronha Island, off Brazil (Rios, 
1970). 

Genus Tribulus Sowerby, 1 839 
(Fig. 21A-E) 

Tribulus (Klein) Sowerby, 1839: 107. 

Planithais (Bayle) Fischer, 1884: 645 [type: 
Purpura planospira Lamarck, 1822: 240, 
by monotypy, = Tribulus planospira (La- 
marck, 1822)]. 

Type Species: Purpura planospira Lamarck, 
1822, by monotypy, = Tribulus planospira 
(Lamarck, 1822); synonyms: Haustrum pic- 
tum Perry, 1811 [rejected name; ICZN, Opin- 
ion 886, 1969: 129]; Purpura lineata Lamarck, 
1816 [nomen oblitum, Old, 1964: 48]. 



Remarks: Sowerby (1839) formally intro- 
duced this name taken from an unpublished 
manuscript by Klein. H. & A. Adams (1853: 
126) used Tribulus as a subgenus of Purpura. 
Cossmann (1903: 68) listed Tribulus (as 
Planithais) as a section of Purpura s.S.; Thiele 
(1929: 297) gave it section rank under Man- 
cinella s.S.; Wenz (1941: 1118) included 
Tribulus as a subgenus of Mancinella, 
whereas Keen (1971b: 550) placed it under 
Thais. Old (1964: 47-48) pointed out that the 
nominal species pictum Perry, 1811 (see 
above), and lineata Lamarck, 1816, are nom- 
ina oblita. Therefore, Lamarck's taxon Pur- 
pura planospira, which he based on his own 
drawing of P. lineata, is the valid name and 
the type species of Tribulus by monotypy. 

Shell: Protoconch (Fig. 210, D) tall, conical, 
of 3.5-4 adpressed whorls and with outward- 
flaring lip; sinusigeral notch obscured by te- 
leoconch. Sculptural pattern obscured by ero- 
sion. Teleoconch (Fig. 21 A, B) large, oval to 
nearly round, of 3-4 adpressed whorls; dor- 
sal sides of last whorls forming flat plateau. 
Adult shell up to about 75 mm in height, 60 
mm in width. Body whorl and aperture reach- 
ing beyond apex. Body whorl dome-shaped, 
sculptured with 1-5 wide, low, spiral ridges 
between six lamellose, high ridges; first three 
adapical ridges most pronounced, top two 
most adjacent to each other. Apertural open- 
ing very wide, oval, usually reaching total 
shell height or extending beyond shell spire. 
Apertural lip thick, with elongate denticles on 
edge corresponding to ridge pattern on out- 
side surface; inside smooth and polished, 
with traces of denticle pattern from previous 
growth stages. Anterior siphonal canal a 
wide, completely open notch; posterior siph- 
onal canal absent. Columella concavely 
curved. Parietal region very wide, heavily cal- 
lused, with large, deep, central indentation 
which partially excavates parietal region; sev- 
eral elongate denticles on lower portion of pa- 
rietal region. Siphonal fasciole as ridge, re- 
sembling fifth and sixth body whorl ridges, 
lying behind callused lower portion of col- 
umella. Shell dirty white to uniform orange 
brown to dark brown; columella white, with 
orange brown blotches and black streak in 
white indentation of parietal region; denticles 
on columella and apertural lip orange brown, 
remainder of lip white. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45° angle to growing 
edge (10-15%) (lacking in many specimens); 



PHYLOGENY OF RAPANINAE 



217 




FIG. 21. Tribulus planospira. A, shell (50 mm), apertural view. B, shell (50 mm), abapertural view. C, 
protoconch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, radula, 
SEM (bar = 35 (xm). 



aragonitic layer with crystal planes oriented 
perpendicular to growing edge (25-30%); 
aragonitic layer with crystal planes oriented 
parallel to growing edge (25-30%); aragonitic 
layer with crystal planes oriented perpendic- 
ular to growing edge (5-10%); calcitic layer 
(25-30%). 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped grov\/th lines; attached 
surface with about 4-6 bracket-shaped 
growth lines and with callused, glazed rim 
(about 30-35% of opercular width) on left. 



Anatomy (based on poorly preserved male 
animals; no female specimens available): 
Head-foot red brown. Anterior siphon dark 
brown, extended some distance from mantle 
edge. Small accessory boring organ dorsal to 
small pedal gland (Fig. 4B). 

Osphradial length about one-half ctenidial 
length; osphradial width less than one-half os- 
phradial width. Osphradium symmetrical in 
shape along lateral and longitudinal axes. Os- 
phradial lamellae attached along very small 
portion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 



218 



KOOL 



dium. Anterior and posterior ctenidial lamellae 
wider than deep. Lateral edge of ctenidial 
lamellae varying from straight to concave; 
ventral edge straight. 

Penis strongly recurved, with long flagellum 
recurved along penial shaft. Penial vas defe- 
rens as centrally located duct-within-a-duct 
system occupying about one-fifth of penis 
width. Seminal vesicles well developed, 
golden brown. 

Proboscis unpigmented, narrower than 
gland of Leiblein. Accessory salivary glands 
thin, long. Salivary gland mass light brown, 
larger than accessory salivary glands. Gland 
of Leiblein spiral, caramel-brown, with straw- 
like external membrane. Mid-esophagus di- 
rectly attached to gland of Leiblein over small 
portion. Posterior esophagus adjacent to left 
lower gland of Leiblein. Anal opening well de- 
veloped, with anal papilla attached to wall. 

Radula: Ribbon length about 30% of shell 
height (Fig. 21 E). Rachidian with very wide 
central cusp, constricted at base; inner edge 
of lateral cusps straight to convex, with single 
denticle at base; outer edge of lateral cusps 
straight to concave, with several small denti- 
cles at base; base of outer edge of lateral 
cusp concavely sloping to large marginal den- 
ticle. Lateral teeth thin, smooth, longer than 
width of rachidian. 

Egg Capsules (identification uncertain; de- 
posited on valve of a pectinid, USNM 96840; 
egg capsule size corresponding with size of 
pedal gland): Small, laterally flattened, up to 
4.5 mm in height, each capsule rectangular in 
cross section, consisting of four distinct 
plates: front and back plate 2-2.5 mm in 
width, side plates 0.5-1 mm in width; front 
plate vase-shaped, side plates of equal dis- 
tance along total surface with central exit hole 
separating side plates. Capsule attached by 
all sides (stalk absent). Capsules deposited in 
row, with front plates adjacent to back plates. 

Ecology: Tribulus planospira lives on vertical 
hard substrates in the high-energy intertidal 
zone (J. H. McLean, personal communica- 
tion). 

Distribution: Eastern Pacific, from Cabo San 
Lucas, Mexico, to Ecuador (Keen, 1971b) and 
Galápagos Islands (Sabelli & Tommasini, 
1979). 

Genus Vasula Mörch, 1860 
(Fig. 22A-E) 

Vasula Mörch, 1860: 99 (as a subgenus of 
Purpura). 



Vascula Woodring, 1959: 223 (error for Va- 
sula Mörch) (as a subgenus of Thais). 

Type Species: Purpura melones Duelos, 
1832, by monotypy, = Vasula melones (Du- 
elos, 1832); synonym: Purpura crassa Blain- 
ville, 1832. 

Remarks: Cossmann, Thiele and Wenz did 
not use this name. Keen (1971b: 550) allotted 
Vasula subgeneric status under Thais, follow- 
ing Woodring (1959: 223). 

Shell: Protoconch of about 3.5 whorls, other- 
wise unknown. Teleoconch (Fig. 22A, B) 
solid, squat, elongate-ovate, of 6-7 ad- 
pressed whorls. Adult shell up to about 50 
mm in height, 35 mm in width. Body whorl 
about 90% of shell height, globose, but often 
with heavy shoulder and straight side, and 
sculptured with numerous (35-45) fine, 
nearly equidistant, spiral grooves; otherwise 
smooth. Apertural opening moderately wide, 
about 75-80% of shell height. Apertural lip 
rounded or J-shaped, depending on develop- 
ment of shoulder; inside smooth and pol- 
ished, crenate on edge. Anterior siphonal ca- 
nal a short, wide notch; posterior canal poorly 
developed. Columella rounded, nearly 
straight, with moderate callus layer. Siphonal 
fascicle forming slightly elevated ridge, 
slightly covered with callus on upper part. 
Shell dark brown with continuous or discon- 
tinuous spiral patterns of white blotches; col- 
umella pigmented with light brown, pink, 
white, yellow and/or orange; apertural lip whit- 
ish yellow, often with pinkish tint, and with 
narrow continuous or discontinuous black 
band along edge. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented in 45° angle to growing 
edge (10-15%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (25-30%); aragonitic layer with crystal 
planes oriented parallel to growing edge (55- 
60%) (Fig. 22C). Presence of calcific layer 
questionable. 

Operculum: D-shaped, with lateral nucleus in 
center right (compare Fig. 1С). Free surface 
with bracket-shaped growth lines; attached 
surface with callused, glazed rim (about 30- 
35% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Head-foot mottled black; tentacles 
black on proximal half of distal tips. Mantle 
edge smooth. Long anterior siphon extending 
far beyond mantle edge. Digestive gland car- 



PHYLOGENY OF RAPANINAE 



219 




FIG. 22. Vasula melones. A, shell (45 mm), apertural view. B, shell (45 mm), abapertural view. C, shell 
ultrastructure, polished fracture surface, SEM (bar = 0.20 mm). D, radula, SEM (bar = 35 ^.m). E, radula, 
rachidian row, SEM (bar = 20 |xm). 



amel-brown. Well-developed, elongate ac- 
cessory boring organ close to foot sole. 

Osphradial length slightly more than one- 
half ctenidial length; osphradial width slightly 
more than ctenidial width. Osphradium sym- 
metrical in shape along lateral and longitudi- 
nal axes. Osphradial lamellae attached along 
small portion of their base. 

Antehormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae wider than 
deep; posterior lamellae deeper than wide. 
Lateral and ventral ctenidial lamellae con- 
cave. 



Vaginal opening enlarged, protruding from 
short, tubular extension of palliai gonoduct, 
and located below and slightly posterior to 
anal opening. Bursa copulatrix as dorso-ven- 
tral slit connected to vagina, continuous with 
capsule gland. Large hook-shaped, ventral 
flange originating from ventral epithelium, lo- 
cated under ventral lobe of capsule gland, 
and minute posterioriy. Ingesting gland 
slightly dorsal to posterior portion of capsule 
gland, with many very small chambers filled 
with black granular material. Seminal recep- 
tacles on dorsal periphery of omega-shaped 
albumen gland. 



220 



KOOL 



Penis large, strongly recurved, with elon- 
gate flagelliform tip. Penial vas deferens as 
duct-within-a-duct system. Testis whitish. 

Proboscis unpigmented, about as wide as 
gland of Leiblein. Paired accessory salivary 
glands long, thin, about one-half of shell 
height; left gland adjacent to proboscis and 
left salivary gland, right gland in anterior part 
of buccal cavity adjacent to proboscis and 
right salivary gland. Salivary glands sepa- 
rated by withdrawn proboscis. Duct between 
mid-esophagus and gland of Leiblein very 
short. Posterior esophagus adjacent to lower 
left side of gland of Leiblein. Gland of Leiblein 
spiral, forming two folds, of soft consistency, 
light brown, without strawlike membrane. 

Stomach thin-walled, with 20-30 thin, 
nearly parallel folds and small folds, each ori- 
ented towards stomach center. Several mi- 
croscopic folds on small portion of posterior 
mixing area adjacent to intestine. Large stom- 
ach typhlosole as thin flange partially lying 
over small folds. Two digestive diverticula 
present. Intestine smooth-walled, with wide 
intestinal typhlosole and very thin folds in in- 
testinal groove. Thin-walled, wide rectum with 
small crystals and black granular material. 
Rectal gland dark green to black, adjacent to 
most of capsule gland in females. Small pa- 
pilla above small but distinct anal opening. 

Radula: Centra! cusp on rachidian con- 
stricted at base (Fig. 22D, E); lateral cusps 
straight; inner denticle small (occasionally bi- 
cuspid) and nearly free from lateral cusp; sev- 
eral small marginal denticles at base of lateral 
cusp, on narrow, somewhat sloping marginal 
area; marginal cusp pronounced, larger than 
marginal denticles; rachidian base with lateral 
extension. Lateral teeth smooth, nearly total 
rachidian width. 

Egg Capsules: Unknown. 

Ecology: During low tide, animals were found 
in shady areas on groups of rocks and boul- 
ders overgrown with barnacles and different 
species of oysters. 

Distribution: Eastern Pacific, from Mexico to 
Peru and Galápagos Islands (Keen, 1971b). 

Genus Vexilla Swainson, 1840 
(Fig. 23A-E) 

Vexilla Swainson, 1840: 300. 

Provexillum Hedley, 1918: 93 [type: Strombus 
vexillum Gmelin, 1791, by monotypy, = 
Vexilla vexillum (Gmelin, 1791)]. 



Type Species: Vexilla picta Swainson, 1840, 
by monotypy, = Vexilla vexillum (Gmelin, 
1791); synonyms: Strombus vexillum Gmelin, 
1791; Purpura taeniata Powys & Sowerby, 
1835. 

Remarks: Swainson (1840: 300) placed this 
genus in the subfamily Nassinae. Cossmann 
(1903: 68) considered Vexilla a valid genus; 
Thiele (1929: 296) placed it as a subgenus 
under Nassa {Jopas). Wenz (1 941 : 1117) fol- 
lowed Thiele's arrangement but used Nassa 
instead of Jopas. Most recent authors recog- 
nized this genus. 

Shell: Protoconch (Fig. 23D, E) very short, 
domelike, of about two adpressed whorls, 
sculptured with small subsutural plicae on last 
whorl, and with outward-flaring lip; sinusigeral 
notch obscured by teleoconch. Teleoconch 
(Fig. 23A, B) elongate-oval, of 3-4 ad- 
pressed whorls. Adult shell up to about 25 
mm in height, 15 mm in width. Body whorl 
rounded, elongate, smooth, up to about 95% 
of shell height. Apertural opening elongate, 
about 80% of shell height. Apertural lip 
slightly curved to J-shaped; inside of apertural 
lip smooth, polished, with crenulations on 
edge continuing inward as small ridges for 
short distance. Anterior siphonal canal a 
poorly developed notch. Postenor siphonal 
canal flanked on left by small protrusion of 
columellar callus. Columella rounded to flat, 
with little callus, curving inward at lower por- 
tion. Siphonal fascicle forming slightly ele- 
vated ridge. Shell usually colored with eight 
pairs of dark brown and cream, narrow, spiral 
bands; cream bands occasionally with red- 
dish narrow line in center. Columella and pa- 
rietal region white, sometimes with light or 
dark brown streak on lower end, occasionally 
continuing upward along inside of columella; 
interior apertural lip white, with faint, light 
brown lines (traces of color pattern on edges 
of previous growth stages); edge white with 
faint light brown blotches between crenula- 
tions and denticles corresponding to banding 
pattern on outside shell surface. 
Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (30-35%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (40-45%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (25-30%). 

Operculum: Ovate-elongate, tapered at 
lower end, with lateral nucleus in upper right 
(Fig. 1 E). Free surface without distinct growth 



PHYLOGENY OF RAPANINAE 



221 




FIG. 23. Vexilla vexillum. A, shell (14 mm), apertura! view. B, shell (14 mm), abapertural view. C, radula, 
SEM (bar = 20 Jim). D, protoconch, apical view, SEM (bar = 50 (лт). E, protoconch, side view, SEM (bar 
= 50 |xm). 



lines; attached surface also without distinct 
growth lines and with callused, glazed rim 
(about 45-50% of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Head-foot mottled dark brown on 
opaque grey. Cephalic tentacles long, mottled 
dark brown on grey, with many white dots, 
white at tips. Mantle edge simple, straight. 
Anterior siphon long, extending beyond man- 
tle edge. Nephridial gland thin, short, dorsal to 
heart. Females with small, shallow ventral 



pedal gland close to anterior part of foot. Bor- 
ing organ apparently absent. Sole of foot with 
small, shallow pustules. 

Osphradial length slightly more than one- 
half ctenidial length; osphradium and ctenid- 
ium about equal in width. Osphradium sym- 
metrical In shape along lateral and longitudinal 
axes. Osphradial lamellae triangular, attached 
along small portion of their base. 

Anteriormost portion of ctenidium straight, 
equidistant from mantle edge with osphra- 
dium. Anterior ctenidial lamellae wider than 



222 



KOOL 



deep; posterior lamellae deeper than wide, or 
as deep as wide. Lateral edge of ctenidial 
lamellae concave; ventral edge straight. 

Vaginal opening an elongated slit below 
and slightly posterior to anal opening. Semi- 
circular ventral flange (originating from epi- 
thelium) located below right lobe. Albumen 
gland omega-shaped, with white, silvery sem- 
inal receptacles on dorsal periphery of albu- 
men gland. 

Penis flagelliform, slightly recurved, oval in 
cross section, folded at gradually tapering tip. 
Penial duct as minute duct-within-a-duct sys- 
tem occupying one-eight of penial width. 
Cephalic vas deferens minute, inconspicu- 
ous. Palliai vas deferens appearing open to 
mantle cavity (in specimens from USNM 
718391) or closed (in specimens from Ha- 
waii). Prostate solid, with ventral duct, adja- 
cent to rectum. Seminal vesicles white. 

Proboscis short and wide, equal in width to 
gland of Leiblein. Accessory salivary glands 
absent. Two large, orange (white in USNM 
718391) distinctly separated salivary glands, 
one between proboscis and gland of Leiblein, 
other in right anterior part of buccal cavity; 
both glands in dorsal buccal cavity, multilob- 
ular. Valve of Leiblein short, with caplike 
structure on anterior end continuing smoothly 
into anterior portion of esophagus, some dis- 
tance from nerve ring and adjacent to left sal- 
ivary gland. Salivary ducts attached to ante- 
rior portion of esophagus at considerable 
distance from valve of Leiblein. Mid-esoph- 
ageal folds inconspicuous (possibly due to 
overall poorly developed, thin esophagus). 
Duct between mid-esophagus and gland of 
Leiblein short, thinner than esophagus itself. 
Posterior esophagus loose from gland of 
Leiblein, occasionally looped at anteriormost 
fold of gland of Leiblein. Gland of Leiblein spi- 
ral, forming two folds, of hard consistency, 
brown (yellowish white and soft in specimens 
from USNM 718391), lacking strawlike outer 
membrane. Posterior duct of gland of Leiblein 
shorter than gland itself, terminating in dorsal 
branch of afferent renal vein. 

Stomach as wide, U-shaped tube with sev- 
eral to many folds on stomach wall of posterior 
mixing area oriented toward center of stom- 
ach. Two digestive diverticula present. Stom- 
ach typhlosole lacking or poorly developed, 
located some distance from posterior mixing 
area edge, thus interrupting folds. Intestinal 
typhlosole distinct. Rectal gland thin, along en- 
tire capsule gland or prostate. Anal opening 
inconspicuous, with large anal papilla. 



Radula: Ribbon length about 25% of shell 
height (Fig. 23C). Rachidian tooth with ex- 
tremely wide central cusp extending along 
most of rachidian base; few small serrations 
at base of side of central cusp; lateral cusps 
smooth, one-third of central cusp length, slop- 
ing down toward edge of rachidian. Lateral 
teeth serrated along nearly entire length, 
much longer than rachidian width. 

Egg Capsules: Unknown. 

Ecology: This species occurs on high-energy 
rocky shores in the low intertidal zone on the 
sea urchins Colobocentrotus and Echi- 
nometra on which it feeds (Kay, 1979; Kool, 
1987: 120). 

Distribution: Indo-Pacific, from eastern Africa 
(Kilburn & Rippey, 1982) to Hawaii (Kay, 
1979). 

Descriptions of Taxa Traditionally 
Considered Belonging to Outgroups of 
Thaididae/nae of Authors 

To evaluate taxonomic positions of the taxa 
described above at the subfamilial and famil- 
ial levels, and to examine the boundaries of 
monophyletic groups, other muricid taxa, not 
believed to be in Thaididae/nae of authors, 
were studied and scored for the same char- 
acters. Choice of taxa depended on such cri- 
teria as availability and previous taxonomic 
placement. For example, Muricanthus ful- 
vescens represents the Muricinae, Rapana 
rapiformis the Rapaninae of authors, and Por- 
rería beleben is a taxon incertae sedis. 



Muricanthus fulvescens (Sowerby, 1841) 
(Fig. 24A-F) 

Shell: Protoconch (Fig. 24C, F) very tall, con- 
ical, of 4.5-4.75 adpressed whorls, with out- 
ward-flaring lip and sinusigeral notch. First 
two whorts smooth, later whorls with micro- 
scopic pustules. Protoconch I nearly as wide 
as first whorl of Protoconch II. Teleoconch 
(Fig. 24A, B) very large, wide, fusiform, mul- 
tispined, of about eight whorls, with im- 
pressed suture, and with long, well-developed 
siphonal canal. Adult shell up to about 185 
mm in height, 105 mm in width. Body whort 
about 85-90% of shell height, sculptured with 
7-9 varices, each with about ten spiny knobs 
open on anterior side. Knobs on varices inter- 



PHYLOGENY OF RAPANINAE 



223 




FIG. 24. Muricanthus fulvescens. A, shell (136 mm), apertural view. B, shell (136 mm), abapertural view. C, 
protoconch, side view, SEM (bar = 0.25 mm). D, shell ultrastructure, fracture surface, SEM (x35). E, 
radula, SEM (bar = 50 jxm). F, protoconch, apical view, SEM (bar = 0.10 mm). 



224 



KOOL 



connected by folds and ridges. Apertural 
opening round; aperture (including anterior si- 
phonal canal) about 70% of shell height. Ap- 
ertural lip semi-circular, thin, except when en- 
forced with knobs on varix; inside smooth and 
shiny; crenulations on edge elongated, con- 
tinuous with row of small denticles. Anterior 
siphonal canal long, wide, almost completely 
closed, straight, without callus, about 
40-45% of shell height; posterior siphonal 
canal absent. Columella rounded, parietal re- 
gion narrow, with moderate callus layer, oc- 
casionally partially detached at margin. Siph- 
onal fascicle well developed, with former 
distal ends of siphonal canal forming angle 
with one another. Shell whitish yellow with 
light and dark brown spiral, continuous or dis- 
continuous lines and blotches; columella and 
apertural lip white. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (30-40%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (30-40%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (25-30%) (Fig. 24D). 

Operculum: Ovate, with terminal nucleus in 
lower right (Fig. 1A). Free surface with con- 
centric growth lines; new growth often par- 
tially overlapping previous growth, resulting in 
lamellose surface; attached surface with 
many (about 30-50) fine growth lines follow- 
ing contour of operculum and with very 
heavily callused, glazed rim (about 30-35% 
of opercular width) on left. 

Anatomy (based on living and preserved an- 
imals): Anterior siphon not extending beyond 
mantle edge. Digestive gland and kidney 
green. Accessory boring organ well devel- 
oped, short distance form sole of foot in 
males, combined with well-developed pedal 
gland in females (Fig. 4B). 

Osphradial length slightly less than one- 
third ctenidial length; osphradial width one- 
third to one-half ctenidial width. Osphradium 
symmetrical in shape along lateral and longi- 
tudinal axes. Osphradial lamellae attached 
along small portion of their base. 

Anteriormost portion of ctenidium straight, 
usually extending farther anteriorly than os- 
phradium. Anterior and posterior ctenidial 
lamellae much wider than deep. Lateral and 
ventral edge of ctenidial lamellae varying from 
concave to convex. Distal tips of ctenidial 



support rods extending beyond lateral edge 
as papillalike projections. 

Vaginal opening a slit situated on distal por- 
tion of tubular extension of palliai gonoduct 
and located directly below anal opening. 
Bursa copulatrix as large diverticulum. Ven- 
tral flange long anteriorly, originating from left 
lobe of capsule gland, and minute posteriorly. 
Large ingesting gland on left side of posterior 
portion of capsule gland extending to albu- 
men gland and consisting of many small 
chambers filled with black granular material. 
Albumen gland a large, single-chambered di- 
verticulum. 

Penis large, elongate, gradually tapering, 
occasionally lightly recurved, pigmented uni- 
form black. Penial vas deferens as well-de- 
veloped duct, semi-closed by epithelium with 
interlocking, lateral ridges (Fig. 5A). Cephalic 
vas deferens well developed. Prostate small, 
posteriorly open to mantle cavity. Seminal 
vesicles brown, well developed, occupying 
large surface area. Testis orange. 

Right accessory salivary gland poorly de- 
veloped, very small, somewhat club-shaped. 
Left accessory salivary gland absent. Paired 
salivary glands large, located on left and right 
sides of valve of Leiblein. Salivary ducts at- 
tached to anterior portion of esophagus at 
base of valve of Leiblein. Valve of Leiblein 
elongate, adjacent to nerve ring. Portion of 
mid-esophagus with glandular folds short; 
folds very well developed, wedged into most 
anterior fold of spiral gland of Leiblein. Gland 
of Leiblein long, spiral, forming two folds, 
long, of hard consistency, with thick strawlike 
external membrane. Duct between mid- 
esophagus and gland of Leiblein short, poorly 
developed. Posterior blind duct of gland of 
Leiblein long, more than half as long as gland 
of Leiblein, and with terminal ampulla located 
in dorsal branch of afferent renal vein. 

Stomach with large, triangular posterior 
mixing area, with many small folds oriented 
towards stomach center. Stomach typhlosole 
poorly developed, intestinal typhlosole thin. 
Two digestive diverticula present. Rectum 
large, embedded in grey opaque connective 
tissue. Anal opening small but distinct with 
small papilla, about equal to size of opening 
and occasionally partially closing it. 

Radula: Ribbon length about 20-25% of 
shell height (Fig. 24E). Rachidian with thin 
central cusp; small lateral denticle separate 
from base of lateral cusps; inner edge of lat- 
eral cusps smooth, convex; outer edge con- 



PHYLOGENY OF RAPANINAE 



225 



cave, with faint, small folds at base, and 
deeply sloping towards edge of rachidian 
tooth. Lateral teeth long, curved, thin, smooth, 
simple, about equal in length to rachidian 
width. 

Egg Capsules: Large, elongate, vase- 
shaped, about 16 mm in height, with concave 
and convex sides. One suture along lateral 
edges and continuing across flattened or con- 
cave apical plate but interrupted by small, 
oval, transparent exit hole in center. Between 
1 ,300 and 1 ,500 embryos per capsule, hatch- 
ing as veligers (D'Asaro, 1986). 

Rapana rapiformis (Born, 1778) 
(Fig. 25A-F) 

Shell: Protoconch (Fig. 25B) tall, conical, of 
3-3.25 adpressed whorls, with minute subsu- 
tural plicae and microscopic pustules on last 
whorls, and with outward-flaring lip and si- 
nusigeral notch. Teleoconch (Fig. 25A) very 
wide, bulbous, of 7-8 whorls, with canalicu- 
late suture, and with moderately long, wide 
siphonal canal. Adult shell up to about 125 
mm in height, 100 mm in width. Body whorl 
bulboso, about 90% of shell height (siphonal 
canal included), sculptured with fine, spiral 
grooves and with three spiral rows of low, 
aligned, blunt, partially open knobs; lower two 
rows of knobs weaker than upper two or ab- 
sent. Apertural opening very wide, oval, about 
80-85% of shell height. Apertural lip semi- 
circular, thin, with faint riblets extending in- 
ward, corresponding to external groove pat- 
tern. Anterior siphonal canal moderately long, 
wide, deep, open, about 20% of shell height; 
posterior siphonal canal poorly developed or 
absent. Columella rounded and slightly con- 
cave, with little callus deposition. Siphonal 
fasciole composed of partially overlapping dis- 
tal ends of siphonal canals from previous 
growth stages. Shell with cream to brown spi- 
rally and/or axially continuous or discontinu- 
ous bands or blotches; columella and interior 
of aperture white to orange. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (20-25%); aragonitic layer with 
crystal planes oriented parallel to growing 
edge (30-40%); aragonitic layer with crystal 
planes oriented perpendicular to growing 
edge (15-25%); calcific layer (10-15%) (Fig. 
25D). 

Operculum: Inverted tear-shaped, with lat- 
eral nucleus in lower right (Fig. IB). Free sur- 
face with staff-shaped growth lines; attached 



surface with about 3-4 bracket-shaped 
growth lines and with callused, dull rim (about 
35% of opercular width) on left. 

Anatomy (based on preserved animals only): 
Head-foot, including long cephalic tentacles 
and anterior siphon, dark brown to black. 
Mantle edge simple, straight, following aper- 
ture contour, or irregular; anterior siphon ex- 
tending slightly beyond mantle edge. Acces- 
sory boring organ (Fig. 25F, abo), large, 
dorsal to well-developed pedal gland in fe- 
males (Fig. 25F, pg). 

Osphradial length slightly less than one- 
half ctenidial length; osphradium and ctenid- 
ium equal in width or osphradial width slightly 
more than ctenidial width. Osphradium sym- 
metrical in shape along lateral and longitudi- 
nal axes, occasionally with posterior portion 
more tapered. Osphradial lamellae attached 
along small portion of their base. 

Anteriormost portion of ctenidium bending 
slightly towards osphradium and extending 
slightly farther anteriorly than osphradium. 
Anterior ctenidial lamellae much wider than 
deep; posterior lamellae about as deep as 
wide. Lateral and ventral edges of lamellae 
varying from straight to slightly concave. Dis- 
tal tips of ctenidial support rods extending be- 
yond lateral edge as papillalike projections. 

Vagina large, situated on distal end of par- 
tially detached tubular extension of pallia! 
gonoduct and located below and slightly an- 
tehor to anal opening. Bursa copulatrix as 
dorso-ventral slit, continuous with ventral 
channel and capsule gland. Ventral flange in 
anterior portion of capsule gland large, 
curved, originating from ventral epithelium, lo- 
cated under small ventral lobe; flange becom- 
ing more reduced posteriorly, located under 
left and right lobe. Albumen gland omega- 
shaped with seminal receptacles on dorsal 
and anterior periphery. 

Penis large, strongly recurved, with short, 
flagelliform tip. Penial vas deferens as duct- 
within-a-duct system occupying about one- 
fourth of penial width. Cephalic vas deferens 
poorly developed. Prostate small, orange, 
with no obvious duct. Seminal vesicles well 
developed, pale yellow to golden orange. 
Testis yellowish. 

Proboscis large, brown, equal in width to 
gland of Leiblein. Paired accessory salivary 
glands about one-third to one-half of shell 
height; right gland located on right anterior 
side of buccal cavity separate from right sal- 
ivary gland, left one sometimes much smaller 



226 



KOOL 




FIG. 25. Rapana rapiformis. A, shell (63 mm), apertural view. B, protoconch, side view, SEM (bar = 0.20 
mm). C, radula, SEM (bar = 0.10 mm). D, shell ultrastructure, SEM (bar = 75 (xm). E, radula, rachidian row, 
SEM (bar = 30 p.m). F, sagittal cross section through anterior foot of female viewed from right side, showing 
accessory boring organ (abo), ventral pedal gland (pg), and transverse furrow (tf), SEM (bar = 0.50 mm). 



PHYLOGENY OF RAPANINAE 



227 



than right and embedded in left salivary 
gland. Salivary glands separate, large; right 
gland ventral to right side of proboscis, left 
one adjacent to anterior side of gland of 
Leiblein and posterior proboscis. Salivary 
ducts attached at varying distance from valve 
of Leiblein. Valve of Leiblein short, sur- 
rounded by salivary glands, and adjacent to 
nerve ring. Portion of mid-esophagus with 
glandular folds long. Duct between esopha- 
gus and gland of Leiblein thin, poorly devel- 
oped. Gland of Leiblein spiral, of hard consis- 
tency, large, usually with external strawlike 
membrane (thickest in older specimens). 
Posterior blind duct longer than gland of 
Leiblein itself. 

Stomach with large posterior mixing area 
extending far posteriorly. Five to fifteen folds 
of different sizes on stomach wall. Stomach 
typhlosole very well developed, partially ex- 
tending posteriorly. Intestinal typhlosole nar- 
row and poorly developed. Several thin folds 
in intestinal groove. Two digestive diverticula 
present. Rectum large in diameter, thin- 
walled. Rectal gland not apparent. Anal open- 
ing wide. 

Radula: Rachidian with thin central cusp 
(Fig. 25C, E); lateral cusps nearly equal in 
length to central cusp, with serrated edges; 
outside of lateral cusp steeply sloping down to 
edge of rachidian. Lateral teeth broad at 
base, simple, smooth, about as long as, 
rachidian width. 

Egg Capsules: Unknown. 

Forreria belcheri (Hinds, 1844) 
(Fig. 26A-F) 

Shell: Protoconch (Fig. 26B, C) tall, conical, 
of about two smooth whorls, and with im- 
pressed suture; transition with teleoconch 
smooth. Teleoconch (Fig. 26A) very large, 
wide, elongate, fusiform, of 6-7 whorls, and 
with slightly impressed suture. Adult shell up 
to about 150 mm in height, 95 mm in width, 
and with long, well-developed siphonal canal. 
Body whorl (siphonal canal included) about 
85% of shell height, with 10-1 1 varices over- 
hanging new growth; body whorl sculptured 
with axial growth lines. Large, spinelike knobs 
on upper corner of square shoulder; moder- 
ately deep, wide canal below lower angle of 
shoulder. Apertural opening wide, oval, about 
75% of shell height (siphonal canal included). 
Apertural lip semi-circular, or semi-hexago- 
nal, thin (even where enforced by varix) to 



moderately thick; pronounced labial spine on 
lower lip; interior of aperture smooth and 
shiny. Anterior siphonal canal long (about 
25% of shell height), wide, deep, straight, 
open; posterior siphonal canal absent. Col- 
umella round, moderately curved, with narrow 
parietal region; moderate callus layer partially 
detached at margin. Siphonal fasciole well 
developed, spiny in appearance due to earlier 
anterior siphonal canals. Wide, concave sur- 
face forming umbilicus between siphonal ca- 
nal (opening) and margin of siphonal fasciole. 
Shell with faint bands of cream to light brown; 
columella, interior of aperture and anterior si- 
phonal canal white. 

Shell Ultrastructure: Aragonitic layer with 
crystal planes oriented perpendicular to grow- 
ing edge (5-10%); aragonitic layer with crys- 
tal planes oriented parallel to growing edge 
(10-20%); calcific layer (70-80%) (Figure 
26F). 

Operculum: D-shaped, upper end rounded, 
with lateral nucleus in lower right (Fig. ID). 
Free surface with staff-shaped, growth lines; 
attached surface with about 7-10 arch- and 
bracket-shaped growth lines and with cal- 
lused, glazed rim (about 30-35% of opercular 
width) on left. 

Anatomy (based on preserved animals only): 
Head-foot, including sole, and short, cephalic 
tentacles greyish. Mantle edge folded. Ante- 
rior siphon not extending beyond mantle 
edge. Accessory boring organ adjacent to 
pedal gland in females (Fig. 4A). Digestive 
gland dark brown. 

Osphradial length one-fourth to one-third 
ctenidial length; osphradial width less than 
one-third ctenidial width. Osphradium sym- 
metrical in shape along lateral and longitudi- 
nal axes, occasionally wider anteriorly, and 
occasionally with right pectin occasionally 
slightly wider than left one. Osphradial lamel- 
lae attached along varying portions of their 
base. 

Anteriormost portion of ctenidium straight, 
extending farther anteriorly than osphradium. 
Anterior and posterior lamellae more than 
twice as wide as deep (widest and shallowest 
lamellae located anteriorly). Lateral and ven- 
tral edge of ctenidial lamellae varying from 
straight to concave. 

Vaginal opening large, simple, formed from 
mantle and tubular anterior portion of palliai 
gonoduct and located below and slightly pos- 
terior to anal opening. Bursa copulatrix as 



228 



KOOL 




FIG. 26. Forreria belcheri. A, shell (1 14 mm), apertural view. B, protoconch, side view, SEM (bar = 80 м.т). 
С, protoconch, apical view, SEM (bar = 80 (xm). D, radula, SEM (bar = 50 |xm). E, radula, rachidian row, 
SEM (bar = 25 ц.т). F, shell ultrastructure, SEM (bar = 0.10 mm). 



PHYLOGENY OF RAPANINAE 



229 



large, separate diverticulum. Ventral channel 
formed by very small flange originating from 
left capsule gland lobe. Ventral lobe present 
only in anterior portion of capsule gland. In- 
gesting gland partially to right of posterior por- 
tion of capsule gland, consisting of one large 
and many smaller chambers, all filled with 
dark brown granular material. Albumen gland 
arch-shaped, nearly square in side view, 
lower ends slightly invaginated. Ovary beige 
to orange. 

Penis elongate, gradually tapering, with mi- 
croscopic pustules on dorsal side. Penial vas 
deferens as well-developed duct, semi-closed 
by epithelium with small, lateral interlocking 
ridges (Fig. 5A). Cephalic vas deferens well 
developed. Prostate large, grey to orange 
brown, composed of two lobes with yellowish 
longitudinal ridges, and with duct as dorso- 
ventral slit, open ventrally to mantle cavity. 

Paired accessory salivary glands extremely 
long, about one-half of shell height; right 
gland separate from salivary gland, left gland 
intertwined with salivary gland. Salivary 
glands adjacent to left side of proboscis and 
equal in size to accessory salivary glands. 
Salivary ducts attached to anterior portion of 
esophagus at short distance from valve of 
Leiblein. Valve of Leiblein elongate, with cap 
structure on anterior end, and surrounded by 
salivary gland lobes and lying adjacent to 
nerve ring. Portion of mid-esophagus with 
glandular folds short; folds very well devel- 
oped, directly attached to gland of Leiblein. 
Gland of Leiblein large, spiral, elongate, of 
hard consistency, lacking strawlike mem- 
brane. Posterior esophagus horseshoe- 
shaped, lying against left side of gland of 
Leiblein. Posterior blind duct of gland of 
Leiblein short, less than one-half length of 
gland of Leiblein. 

Stomach with large posterior mixing area 
and many fine folds oriented towards center 
of stomach. Small smooth area prior to intes- 
tinal area. Stomach typhlosole well devel- 
oped, intestinal typhlosole thin. Two digestive 
diverticula present. Rectum moderately wide. 
Anal opening very small. Anal papilla occa- 
sionally formed from anteriorly extended dor- 
sal wall of rectum. 

Radula: Ribbon length about 15% of shell 
height (Fig. 26D, E). Rachidian with thin, nee- 
dle-shaped central cusp; lateral cusps with 
3-4 inner denticles and serrated outer edge 
with 1-2 faint outer denticles on base; base of 
outer edge of lateral cusps adjacent to base 



of inner edge of large marginal cusp; marginal 
cusps in different plane than lateral cusps 
(about 75° angle) and parallel to elongate lat- 
eral extension at base of rachidian tooth, re- 
sulting in bifid rachidian edge (compare Fig. 
15E). Lateral teeth broad, smooth, simple, 
equal in length to rachidian width. 

Descriptions of Taxa Used to Test 
Robustness of Synapomorphies 

The species Acanthina monodon and Tro- 
chia cingulata were only examined on few 
features after initial cladistic analyses had re- 
vealed synapomorphies for a clade consisting 
of Nucella and Forreria. These two species, 
suspected of being closely allied to Nucella 
and Forreria, were tested for having the same 
synapomorphies as found for the Nucella- 
Forreria clade. The two taxa were usually in- 
cluded in Thaididae/nae of authors. 

Acanthina monodon (Pallas, 1 774) 
(Fig. 27A-D) 

Anatomical data for Acanthina monodon 
were obtained from Wu (1985); this species 
has a bursa copulatrix that is separate from the 
lumen of the capsule gland, very long acces- 
sory salivary glands, a lightly curved penis with 
pseudo-papilla, an accessory boring organ 
separate from the ventral pedal gland (in fe- 
males; Fig. 4A), and a D-shaped operculum 
with its upper end rounded and with a lateral 
nucleus in the lower right (compare Fig. ID). 
Scanning electron micrographs of the shell ul- 
trastructure were not available at the time of 
the cladistic analysis, but from light micros- 
copy it was obvious that an inner aragonitic 
layer with the crystal planes oriented in a 45° 
angle to the growing edge is absent. The pro- 
toconch (Fig. 270, D) is smooth, paucispiral 
(about 1 .5 whorls), and lacks an outward-flar- 
ing lip. 

Trochia cingulata (Linnaeus, 1758) 
(Fig. 28A-E) 

Scanning electron micrographs of the pro- 
toconch and the shell ultrastructure revealed 
a smooth, paucispiral protoconch of about 
1.5 whorls, lacking an outward-flaring lip 
(Fig. 280, D), and a shell ultrastructure con- 
sisting of an aragonitic layer with crystal 
planes oriented perpendicular to growing edge 
(10-30%), an aragonitic layer with crystal 
planes oriented parallel to growing edge (25- 



230 



KOOL 




FIG. 27. A-D, Acanthina monodon. A, shell (46 mm), apertural view. B, shell (46 mm), abapertural view. C, 
protoconch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E-G, 
Urosalpinx cinerea. E, protoconch, side view, SEM (bar = 0.10 mm). F, radula, SEM (bar = 10 ц.т). G, 
protoconch, apical view, SEM (bar = 0.10 mm). 



PHYLOGENY OF RAPANINAE 



231 




FIG. 28. Trochia cingulata. A, shell (40 mm), apertural view. B, shell (40 mm), abapertural view. C, proto- 
conch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, shell 
ultrastructure, SEM (bar = 50 jim). 



40%), and a calcitic layer (30-65%) (Fig. 
28E). 



Phylogenetic Analysis 

Figure 30 shows a consensus tree of 6,288 
trees obtained with all multistate characters 
(Table 3) scored as unordered and using the 
rigorous "mh* bb*" command. The consis- 
tency index of each of the trees is 0.86; the 
consistency index of the consensus tree is 
0.77. 



DISCUSSION AND CONCLUSIONS 

Phylogenetic Analysis 

It is obvious that the Thaididae/nae of au- 
thors, which prior to now usually included all 
taxa used in this study except Muricanthus, 
Rapana, and (usually) Forreria, can be di- 
vided into two monophyletic groups and that 
para- and polyphyly was present in previous 
taxonomic arrangements both at the generic 
and (sub)familial levels. For example, the 
type species of Nucella (often referred to in 
the literature as "Thais" lapillus or "Purpura" 



232 



KOOL 




FIG. 29. Ecphora cf. quadricostata. A, shell (71 mm), apertura! view. B, shell (71 mm), abapertural view. C, 
protoconch, side view, SEM (bar = 0.15 mm). D, protoconch, apical view, SEM (bar = 0.15 mm). E, shell 
ultrastructure, SEM (bar = 0.30 mm). 



lapillus), is excluded from the taxon name to 
be used for Clade С (Fig. 30), based on a 
wide variety of characters, many of which It 
shares as synapomorphies with Porrería 
beleben, the type species of Porrería, which 
was previously grouped within the Rapaninae 
as well as Thaidinae. 

The high number of trees is partially due to 
the lack of data for two of the species of Clade 



В {Acantbina monodon and Trocbia cingu- 
lata). This resulted in a multitude of resolu- 
tions for this clade and thus increased the to- 
tal number of equally parsimonious trees. 

The number of convergences and parallel- 
isms among the two main clades (e.g. a sep- 
arate pedal gland and accessory boring organ 
in Nucella and Cymia) and the outgroup, in- 
dicate that boundaries among these three 



PHYLOGENY OF RAPANINAE 

Ocenebrinae Rapaninae 



233 



X Z l-li- < 



£ «0 Q. с 



2 $ s с E :^ 



>ч Í= <б О _ _ 

UCOCCUÛ>Zû.ÛÛ. 



Ü > F Q- 2 zi- 



B 



G 



I 

FIG. 30. Consensus cladogram with taxonomic groupings superimposed. Mur = Muricanthus; Hau = 
Haustrum; Nue = Nucella; Tro = Trochia; For = Forreria; Аса = Acanthina; Сугл = Cymia; Stra = 
Stramonita; Rap = Rapana; Con = Concholepas; Die = Dicathais; Vex = Vexilla; Nas = Nassa; Pin = 
P/nax/a; Dru = Drupa; Plie = Plicopurpura; Mor = Morula; Cro = Cronia; Vas = Vasu/a; Tha = Thais; Pur 
= Purpura; Man = Mancinella; Neo = Neorapana; Trib = Tribulus. 



groups are not sufficiently clear-cut to justify 
familial ranking for all three clades. I suggest 
that these clades merely be ranked as sub- 
families. 

The taxa on Clade A form a distinct, cohe- 
sive clade, despite the limited data available 
for two of its taxa. Previously, the genera 
Haustrum, Acanthina, Nucella, Trochia, and 
Forreria, had been included in Thaididae/nae 
of authors, although Forreria has also been 
allocated to Rapaninae of authors. However, 
the five species in Clade В show no more 
resemblance with members of Clade С than 
they do with Muricanthus (Muricinae). As 
stated earlier, studies of Ocenebra s.s. (Kool, 
1993) revealed close phylogenetic relation- 
ship among Ocenebrinae and the taxa of 
Clade A. 



The consensus tree shows that including 
only Rapana in Rapaninae would result in 
paraphyly. Cymia can be considered as an 
atypical member of Rapaninae (see below), 
but providing it with separate subfamilial sta- 
tus appears unjustified. All taxa of Clade С 
should be included in Rapaninae. Perhaps fu- 
ture studies will reveal that Rapaninae should 
be further subdivided into two or more sub- 
families. For example, in some previous anal- 
yses Cronia and Morula grouped at the base 
of Clade С (Kool, 1 989); either these two gen- 
era are very highly derived members of Clade 
C, or their placement in Clade С should be 
subjected to further examination, which may 
show that they are better placed in Ergalatax- 
inae Kuroda & Habe, 1971. The present 
study, however, indicates that all taxa of 



234 



KOOL 



TABLE 3. Characters and character states. Nurnbers and letters correspond to those in text. 



Character 


1 


2 


3 


4 


5 


6 


7 


8 


9 


10 


11 


12 


13 


14 


15 


16 


17 


18 


Taxon 






































Muricanthus 


a 


a 


a 


а 


a 


a 


a 


a 


a 


a 


а 


a 


a 


a 


a 


а 


а 


а 


Porrería 


b 


b 


b 


а 


b 


a 


b 


a 


a 


a 


b 


a 


a 


a 


b 


? 


b 


с 


Nucella 


b 


b 


b 


а 


b 


a 


b 


a 


a 


a 


b 


b 


с 


a 


b 


b 


b 


с 


Haustrum 


b 


? 


с 


а 


b 


a 


a 


a 


a 


a 


b 


b 


b 


a 


e 


b 


а 


b 


Morula 


a 


a 


a 


а 


с 


b 


a 


b 


b 


b 


с 


с 


d 


b 


с 


а 


а 


е 


Cronia 


с 


a 


a 


а 


с 


b 


a 


b 


b 


b 


d 


с 


d 


b 


с 


а 


а 


е 


Rapana 


a 


a 


с 


а 


d 


a 


a 


a 


b 


с 


d 


e 


d 


b 


b 


а 


а 


f 


Cymia 


? 


? 


с 


а 


e 


a 


b 


a 


b 


a 


? 


d 


d 


b 


b 


а 


а 


d 


Stramonita 


a 


a 


b 


а 


e 


a 


a 


a 


b 


с 


d 


e 


d 


b 


b 


а 


а 


g 


Concholepas 


a 


a 


b 


а 


e 


a 


a 


a 


b 


с 


d 


e 


d 


b 


b 


а 


а 


g 


Dicathais 


с 


a 


b 


а 


e 


a 


a 


b 


b 


с 


? 


e 


d 


b 


с 


а 


а 


g 


Vasula 


? 


? 


7 


b 


e 


a 


a 


b 


b 


с 


d 


e 


d 


b 


с 


? 


а 


j 


Vexilla 


d 


a 


a 


а 


f 


a 


с 


b 


b 


с 


d 


f 


d 


b 


d 


а 


а 


? 


Nassa 


a 


a 


a 


а 


e 


a 


a 


b 


b 


с 


d 


f 


d 


b 


с 


а 


а 


? 


P'maxia 


a 


a 


a 


а 


e 


a 


a 


b 


b 


с 


d 


f 


d 


b 


с 


а 


а 


h 


Drupa 


? 


? 


? 


а 


e 


a 


a 


b 


b 


с 


d 


e 


d 


b 


d 


а 


а 


h 


Plicopurpura 


a 


a 


? 


а 


e 


a 


a 


b 


b 


с 


d 


e 


d 


b 


с 


а 


а 


? 


Thais 


? 


a 


с 


b 


e 


a 


a 


b 


b 


с 


d 


e 


d 


b 


с 


а 


а 


J 


Purpura 


? 


a 


с 


b 


e 


a 


a 


b 


b 


с 


d 


e 


d 


b 


с 


а 


а 


h 


Mancinella 


? 


? 


с 


b 


e 


a 


a 


b 


b 


с 


? 


e 


d 


b 


с 


а 


а 


i 


Neorapana 


? 


a 


с 


b 


e 


a 


a 


b 


b 


с 


d 


e 


d 


b 


с 


а 


а 


j 


Tribulus 


9 


a 


с 


b 


e 


? 


a 


b 


? 


? 


? 


e 


d 


? 


? 


? 


а 


j 


Acanthina 


b 


b 


? 


a 


b 


? 


b 


? 


a 


? 


? 


b 


? 


? 


? 


? 


? 


? 


Trochia 


b 


b 


b 


a 


? 


? 


? 


? 


? 


? 


? 


? 


? 


? 


? 


? 


? 


? 



Clade С are to be included in one subfamily, 
of which Rapana is the provider of the subfa- 
milial name. Thaidinae becomes a subjective 
junior synonym of Rapaninae, by priority. 

A discussion of the relationships among the 
taxa of the main clades of the consensus cla- 
dogram (Fig. 30) follows. 

Clade A: Haustrum haustorium is more 
closely allied with the species of Clade В than 
it is with any of the species of Clade С Two of 
the taxa of Clade В {Acanthina and Trochiia) 
were not examined in detail for this study, but 
they grouped unambiguously with Nucella 
and Porrería based on the data available. 
Nevertheless, the hiatus of character states of 
these two taxa resulted in a large number of 
variations in the resolution of Clade B, con- 
tributing to the high number of trees obtained 
from the analysis. 

Clade С (individual clades treated sepa- 
rately): Although Cymia is included in Clade 
C, it shares a synapomorphy with the species 
of Clade В (accessory boring organ and ven- 
tral pedal gland [females] with separate duct) 
and lacks, as do all members of Clade A, a 
synapomorphy found in all other members of 
Clade В (posterior seminal receptacles [fe- 
males]). However, Cymia shares several sy- 



napomorphies with all other taxa of Clade В 
(bursa copulatrix continuous with capsule 
gland [females], strongly recurved penis, 
closed prostate, penial vas deferens a duct- 
within-a-duct [males]). Further detailed stud- 
ies may determine whether the placement of 
this atypical, perhaps primitive, species in Ra- 
paninae is justified. 

The radular morphology of Cymia tecta re- 
veals a possibly closer relationship with 
Haustrum than the tree topology indicates. To 
a posteriori test for homology (Patterson, 
1982) in the radular morphology, the radular 
characters (17 and 18, Table 3) of Cymia 
were alternatively scored identical to those in 
Haustrum, because the superficial resem- 
blance may be indicative of homology. How- 
ever, this did not alter the tree topology; other 
characters overrode this "attempted" switch 
of Cymia to Clade A, and the original place- 
ment prevailed. 

Clades D, E, F, G: Clades D and E have suf- 
fered significant loss of resolution compared 
to the individual trees from which the consen- 
sus tree was obtained. However, several dis- 
tinct and stable clades can be found higher up 
the tree. Clade G consists of the taxa Vasula, 
Thais, Purpura, Mancinella, Neorapana, and 



PHYLOGENY OF RAPANINAE 



235 



Tribulus. The similarity in radular morphology 
among the taxa Thais, Tribulus, Neorapana, 
and Vasula suggests that at these four genera 
are only distinct at the subgeneric level; I con- 
sider Tribulus, Neorapana, and Vasula sub- 
genera of Thais, the oldest available name. 
Mancinella and Purpura are sufficiently differ- 
ent in radular morphology from one another 
and from the other four genera in Clade G to 
justify separate generic status for these two 
taxa. This separation at the generic level is 
further supported by the topologies of many of 
the obtained trees. Clade F, consisting of 
Morula and Cronia, is also very stable. 

The low resolution among the taxa Rapana, 
Stramonita, and Concholepas of Clade D, 
and of Dicathais, Vexilla, Nassa, Pinaxia, 
Drupa, and Plicopurpura of Clade E, can be 
attributed to several factors. The characters 
and character states used are adequate to 
identify major groups, but are not sufficiently 
robust to yield only one most parsimonious, 
highly resolved tree. At the lower taxonomic 
levels, convergence and parallelism appear 
to be more common, thus increasing the num- 
ber of equally parsimonious branching pat- 
terns. This low resolution could furthermore 
be attributed to close phylogenetic relation- 
ship. I propose that a combination of these 
factors is the cause for a low resolution in 
Clades D and E, as well as in Clades В and G. 
It should be noted that low resolution by itself 
does not provide a strong argument for syn- 
onymization of any of the genera in these 
clades; autapomorphies for the type species 
of a genus most likely become synapomor- 
phies for almost all species within that genus 
when more species are added to the analysis. 

Character State Transformations 
on Cladogram 

The topology of the cladogram (Fig. 30) 
supports a single hypothesis for character- 
state evolution in 13 characters. More than 
one (and equally parsimonious) transforma- 
tion series are possible for the remaining five 
(3,5,11,1 2, and 1 8). I chose for the scheme 
which would place character-state changes 
as high on the tree as possible; this reasoning 
prevents placement of less informative syn- 
apomorphies to be placed in basal positions. 
For example, if state (a) occurred in the out- 
group, (b) in Clade A (Fig. 30), and (c) in 
Clade C, I would choose a scheme whereby 
both (b) and (c) evolved from (a), although it 
would be equally parsimonious to assume a 



linear transformation series [(a) -^ (b) -^ (c) 
or (a) - (c) -^ (b)]. 

The hypotheses about character state ev- 
olution and possible causal schemes are dis- 
cussed below. The numbers and letters as- 
signed to, respectively, the characters and 
character states correspond to the numbers 
and letters in Table 3 and to those in the list of 
characters in MATERIALS AND METHODS. 

Protoconch: — Number of whorls and sculp- 
ture (1). From a multispiral, sculptured condi- 
tion (a) (e.g. Fig. 24C) evolved three other 
conditions: a paucispiral, smooth condition (b) 
(e.g. Fig. 15C); a multispiral, smooth condi- 
tion (c) (e.g. Fig. 9C); and a paucispiral, 
sculptured condition (d) (e.g. Fig. 23D). 

— Transition into teleoconch (2). The apo- 
morphic condition is the absence of an out- 
ward-flaring lip and sinusigeral notch (b) (e.g. 
Fig. 15C). In most of the studied taxa, these 
features are present (a) (e.g. Fig. 13D). The 
absence of the outward-flaring lip and si- 
nusigeral notch correlates with the mode of 
development; species with direct develop- 
ment lack these features, whereas it is 
present in taxa with a planktonic larval stage. 
The tree topology suggests that the direct 
mode of development evolved from a free- 
swimming mode of development. 

Shell Ultrastructure:— Ca\c\\\c outer layer (3). 
Absence of calcite is the plesiomorphic con- 
dition (a); presence of calcite is the derived 
condition. The presence of calcite is arbitrarily 
quantified into the states "thick" (> 25% of 
total shell thickness) (b) (e.g. Fig. 15G), and 
"thin" (< 20% of total shell thickness) (c) (e.g. 
Fig. 20E). A thick layer probably evolved from 
a thin layer. 

It is difficult to determine whether calcite is 
present in Drupa, Vasula and Plicopurpura. 
Crystallographic (e.g. X-ray diffraction) tech- 
niques should be used to determine whether 
calcite is present in those taxa scored with "?" 
for this character in Table 3. The lacking data 
and low resolution of the cladogram does not 
allow for speculation on evolutionary trends 
for this character, other than that the lack of 
calcite is the plesiomorphic condition found in 
the outgroup, some members of the Rapani- 
nae, and in other neogastropods (Buccinidae, 
Volutidae, etc.) (Harasewych & Kool, in prep- 
aration). 

— 45° innermost aragonitic layer (4). Ab- 
sence of this inner layer of aragonite, the 
crystal planes of which are oriented in a 45° 



236 



KOOL 



angle to the growing edge, is the plesiomor- 
phic condition (a); presence of this layer is the 
derived state (b) (e.g. Fig. 20E). This layer not 
only adds thickness to the shell, but presum- 
ably also gives more strength to it, which may 
serve as defense to prédation. 

Operculum: — Morphology of operculum (5). 
The opercular shape in the outgroup is oval, 
with a terminal nucleus in the lower right, and 
with concentric growth lines (a) (Fig. 1 A). This 
condition gave rise to both a D-shaped oper- 
culum with upper end rounded and with lat- 
eral nucleus in the lower right (b) (e.g. Fig. 
1 D), and a D-shaped operculum with a lateral 
nucleus in the center right (e) (e.g. Fig. 1С). 
From this last condition (e) arose three other 
opercular morphologies: an inverted tear- 
shaped operculum with a rounded upper 
edge, a tapered lower end, and with a lateral 
nucleus in the lower right (d) (e.g. Fig. 1B); a 
D-shaped operculum, tapered at the lower 
end, with an S-shaped left edge (adjacent to 
columella), and with a lateral nucleus in the 
lower right (c) (e.g. Fig. IF); and an ovate- 
elongate operculum, tapered at the lower 
end, and with a lateral nucleus in the upper 
right (f) (Fig. 1E). 

The shape of the operculum is, of course, 
largely dependent on aperture shape; how- 
ever, it is interesting that the operculum of 
Haustrum, a non-rapanine, is very different in 
morphology from that of Purpura or Plicopur- 
pura, whereas these three species have ex- 
tremely similar apertural shapes. It should be 
noted that the operculum of Rapana rapi- 
formis is scored differently from the other ra- 
panines, but that the operculum of other Ra- 
pana species is D-shaped and with a nucleus 
in the center right, as in most other rapanines. 

Taki (1950) provided an evolutionary sce- 
nario for opercular morphologies in which a 
D-shaped operculum with an "extranuclear" 
nucleus (as found in Purpura) evolved from 
an ovate operculum with an "extraeccentric" 
nucleus (as found in Muricanthus). 

— Rodlike structures in hypobranchial 
gland (6). Presence of rodlike structures in 
the hypobranchial gland, oriented perpendic- 
ular to the mantle (b) is the apomorphic con- 
dition (Fig. 2A, B). The function of these struc- 
tures is not known. 

— Ventral pedal gland and accessory bor- 
ing organ (7). In female specimens of the out- 
group and in many of the rapanines, the ac- 
cessory boring organ and ventral pedal gland 
share a common duct to the outside (a) (Fig. 



48). From this condition arose two conditions: 
the development of a ventral pedal gland with 
an opening separate from that of the acces- 
sory boring organ (b) (Fig. 4A); and loss of the 
accessory boring organ (c). 

In the majority of taxa studied herein, a sin- 
gle accessory boring organ duct is responsi- 
ble for the excretion of decalcifying agents 
and for the intake and tanning of egg cap- 
sules. The derived condition of having sepa- 
rate ducts enables the female to specialize 
both structures further and may allow feeding 
during periods between laying eggs. This in- 
crease in flexibility is of more importance to 
snails with seasonal patterns in feeding and 
spawning, than to those that can feed and 
spawn at any time. The most derived condi- 
tion is loss of the accessory boring organ, 
which probably is the result of specialized 
feeding habits. {Vexilla is parasitic on urchins 
[Kay, 1979; Kool, 1987].) 

Mantle Cavity Organs: — Osphradial length 
relative to ctenidial length (8). The plesiomor- 
phic condition is an osphradial length of less 
than one-half the ctenidial length (a). This 
condition gave rise to an osphradial length of 
at least one-half that of the ctenidium (b) (Fig. 
3D). 

Numbers of osphradial lamellae vary from 
about 7-14 per mm; those of the ctenidium 
from 9-22 per mm. It seems probable that a 
relatively larger osphradium facilitates the 
search for food. However, because the os- 
phradium is measured against ctenidium size, 
it may be that the small size of the ctenidium 
only causes the osphradium to appear larger 
than the osphradium in other species. Fur- 
thermore, the density of osphradial lamellae 
may be age and/or size dependent. This char- 
acter thus does not lend itself for adaptationist 
schemes. 

Female Reproductive System: — Bursa copu- 
lathx (9). A sacklike bursa, usually located an- 
terior to the capsule gland, and with its lumen 
separate from that of the capsule gland is the 
plesiomorphic condition (a) (Fig. 4C). From 
this condition evolved a bursa that is merely 
an anteriorly located specialized extension of 
the capsule gland (b) (Fig. 4D). 

— Posterior seminal receptacles on dorsal 
periphery of the albumen gland (1 0). Absence 
of these structures is the plesiomorphic con- 
dition (a) (Fig. 4F, G); from this condition 
evolved a development of specialized struc- 
tures for sperm storage that open into the al- 
bumen gland (c) (Fig. 4H). A situation where 



PHYLOGENY OF RAPANINAE 



237 



two or three seminal receptacles branch off 
the ovi-sperm duct appears to have evolved 
from the latter condition (b) (Fig. 4E). 

Kool {1988a, b) described in detail why the 
posterior seminal receptacles, which open di- 
rectly into the albumen gland, allow a more 
efficient mode of fertilization, and suggested 
that this evolutionary novelty may have trig- 
gered a radiation in rapanines. Presence of a 
specialized receptacle branching off the ovi- 
sperm duct could be interpreted as an inter- 
mediate condition, but the tree topology sug- 
gests it is the most highly derived condition. 

— Morphology of albumen gland (11). The 
ancestral condition of albumen gland mor- 
phology was most likely a dorsally swollen 
oviduct, which then developed into a lobular 
structure (a) (Fig. 4F). Two morphologies 
evolved from this ancestral state. The ventral 
side of the oviduct may have invaginated, re- 
sulting in an arch-shaped tube, appearing like 
a tube coiled onto itself (b) (Fig. 4G), and an 
omega-shaped tube (d) (Fig. 4H). From the 
last condition (d) arose a more asymmetrical, 
staff-shaped albumen gland (c) (Fig. 4E). 

If, indeed, this is the sequence of evolution- 
ary events in the development in this charac- 
ter, it may be hypothesized that albumen 
glands became more efficient in the process 
of coating of albumen due to an increased 
surface area and a longer route for the eggs 
to travel (Kool, 1988a, b). Higher efficiency 
may explain the reduction of the anterior lobe 
of this gland in a highly derived taxon, such as 
Morula. 



Male Reproductive System: — Morphology of 
penis (12). The outgroup has an elongated, 
occasionally lightly curved, gradually tapering 
penis (a) (Fig. 5A). From this shape, several 
different morphologies evolved: a relatively 
short, wide, straight or lightly curved penis 
with a small pseudo-papilla (b) (Fig. 5B); an 
elongate, wide penis, strongly recurved, club- 
shaped, with a slightly swollen distal end (d) 
(Fig. 5F); a consistently strongly recurved pe- 
nis tapering distally into a flagelliform append- 
age of varying length (e) (Fig. 5D). From (e) 
evolved a slightly recurved penis, long and 
gradually tapering distally (f) (Fig. 5C); the 
tree topology furthermore suggests that a pe- 
nis with a large side lobe (c) (Fig. 5E, I, si) 
evolved from (e). The side lobe may have 
some purpose in the copulation process. 

— Morphology of penial vas deferens (13). 
The outgroup has a well-developed duct. 



semi-closed by interlocking lateral ridges (a) 
(Fig. 5A). From (a) evolved three states: an 
open duct, located on the posterior edge of 
the penis (b); a semi-closed condition, similar 
to (a), but with minute duct and without lateral 
ridges, and lying more adjacent to the penial 
posterior edge (c) (Fig. 5B); and a convoluted, 
coiling, meandering tube within a larger cavity 
(duct-within-a-duct system) (d) (Fig. 5D). 

Histological studies may show that the dor- 
sal and ventral flaps of tissue in conditions (a) 
(with lateral ridges) and (c) (without lateral 
ridges) are held together by cilia. Dissections 
of well-preserved specimens of Haustrum will 
determine whether the "open" condition is not 
an artifact of poor preservation. 

— Morphology of prostate duct (palliai vas 
deferens) (14). A prostate duct that is in open 
connection with the mantle cavity (in the pos- 
terior portion) is the plesiomorphic character 
state (a) (Fig. 5H). A duct closed throughout 
the prostate developed from this condition (b) 
(Fig. 5G). 

A prostate with a duct in open connection 
with the mantle cavity may be to some advan- 
tage by allowing for an emergency release for 
sperm in case the snail is forced to withdraw 
into the shell. However, it is doubtful that the 
elasticity of the palliai gonoduct could not ab- 
sorb some extra pressure while the animal is 
withdrawing. Furthermore, loss of sperm 
would be prevented in a closed prostate duct. 



Alimentary System.-^Length of accessory 
salivary glands (15). A very poorly developed, 
almost vestigial, minute right accessory sali- 
vary gland is present in the outgroup (a). 
From this condition arose a pair of very long 
accessory salivary glands (up to over one-half 
of shell height) (b), from which arose two 
other conditions: presence of a very well-de- 
veloped, long (nearly one-half of shell height) 
right accessory salivary gland (e), and a pair 
of glands of short to medium length (less than 
one-fourth of shell height) (c) (Fig. 3F, ra, la). 
From the latter condition evolved loss of both 
the left and the right glands (d). 

— Length of posterior blind duct of gland of 
Leiblein (1 6). The plesiomorphic condition is a 
long duct (> one-half length of gland itself) 
(Fig. 3F, dgL) which reaches into the dorsal 
branch of the afferent renal vein (a). From this 
condition evolved a very short duct (< 1/2 of 
length of gland itself) which empties into the 
posterior portion of the cephalic cavity (b) 
(Fretter & Graham, 1962: fig. 153). 



238 



KOOL 



Radula (Rachidian): — Orientation of marginal 
cusp (17). A marginal cusp in the same plane 
with the lateral cusp is the plesiomorphic con- 
dition (a). From (a) arose a marginal cusp 
which is in a different plane with the lateral 
cusps (b) (e.g. Fig. 15E, F). 

— Morphology of cusps on rachidian tooth 
(18). From a rachidian without a marginal 
area and cusps, with a small, free-standing 
inner lateral denticle, and long lateral cusps 
(a) (Fig. 24E) evolved four morphologies; the 
first, without marginal area and cusps, with 
large, free-standing inner lateral denticle and 
long lateral cusps (b) (Fig. 11D); the second, 
without marginal area, with small marginal 
cusps, one or more inner lateral denticles and 
long lateral cusps (c) (e.g. Fig. 15F); the third, 
without marginal area, with small marginal 
cusps, a small inner lateral denticle and short, 
nearly triangular lateral cusps (d) (Fig. 8H); 
the fourth, without marginal area, with small 
marginal cusps, with one or more inner lateral 
denticles and long lateral cusps (g) (e.g. Fig. 
7F). From (g) arose four other rachidian mor- 
phologies: a wide marginal area, without mar- 
ginal cusps, with free-standing inner lateral 
denticle and short lateral cusps (e) (e.g. Fig. 
8D); one without marginal area and cusps, 
with several faint inner lateral denticles and 
long lateral cusps (f) (Fig. 25C, E); one with 
wide marginal area with many denticles and a 
small marginal cusp, a small inner lateral den- 
ticle and long lateral cusps (h) (e.g. Fig. 18D); 
and one with a short marginal area, with small 
marginal cusps, with or without small inner 
lateral denticle and with long lateral cusps (j) 
(e.g. Fig. 22E). From (j) evolved a rachidian 
without marginal area and cusps, without in- 
ner lateral denticles, and with short lateral 
cusps (i) (Fig. 1 1 1). Three additional morphol- 
ogies (scored with "?") that arose from (g) 
are: similar to (i) but with a free-standing lat- 
eral denticle in some specimens, and with 
short lateral cusps (Fig. 13G); also similar to 
(i), but with slit in central cusp (Fig. 17E); and 
the last situation, also similar to (i) but with the 
base of the central cusp nearly as wide as the 
rachidian itself (Fig. 23C). 

The following are synapomorphies for the 
different clades and taxonomic groups of the 
consensus tree (Fig. 30). 



Clades A, С ("the ingroup"): 

(1) layer of calcite of medium thickness 
(character 3). 

(2) accessory salivary glands very long 



(nearly one-half of shell height) (char- 
acter 15). 

Calcite is absent in several taxa of Clade E, 
whereas a thick layer of calcite is present in 
taxa in Clades В and D (see remarks under 
Clade G). Among taxa of both clades, the ac- 
cessory salivary glands vary from medium in 
size to absent. 

Clade A (Ocenebrinae): 

(1) protoconch paucispiral and smooth 
(Character 1). 

(2) operculum D-shaped, with upper end 
rounded and with lateral nucleus in 
lower right (character 5). 

(3) albumen gland arch-shaped, elongate 
(character 11). 

(4) penis straight or mildly curved with 
pseudo-papilla (character 12). 

(5) short blind duct of gland of Leiblein 
(character 16). 

Clade В (within Ocenebrinae): 

(1) transition from protoconch to teleo- 
conch smooth, outward-flaring lip ab- 
sent (character 2). 

(2) layer of calcite thick (character 3). 

(3) accessory boring organ separate from 
pedal gland (character 7). 

(4) marginal cusp in different plane than 
lateral cusp (character 17). 

(5) rachidian with small marginal cusps, 
one or more small inner lateral denti- 
cles, and with lateral cusps nearly 
equal in length to central cusp (char- 
acter 18). 

A thick calcific layer (2) and separate ducts 
for the accessory boring organ and ventral 
pedal gland (3) are also found in Clade С 
(Cymia) and are probably the result of parallel 
evolution. Absence of an outward-flaring lip 
(1) may become a synapomorphy for Clade 
A, once it is shown that the transition from 
protoconch to teleoconch in Haustrum haus- 
torium is smooth. 

Clade С (Rapaninae): 

(1) operculum D-shaped, with lateral nu- 
cleus in center right (character 5). 

(2) bursa copulatrix continuous with cap- 
sule gland (character 9). 

(3) penial vas deferens as duct-within-a- 
duct (character 13). 

(4) prostate gland closed to mantle cavity 
(character 14). 



PHYLOGENY OF RAPANINAE 



239 



Clade D: 

(1) posterior seminal receptacles on dor- 
sal periphery of albumen gland (char- 
acter 10). 

(2) omega-shaped albumen gland (char- 
acter 11). 

(3) penis strongly recurved, with flagellate 
pseudo-papilla (character 12). 

(4) marginal area absent, marginal cusps 
small; one or more inner lateral denti- 
cles; lateral cusps nearly equal in 
length to central cusp (character 18). 

Clade E: 

(1) layer of calcite absent (reversal; see 
remarks under Clade G) (character 3). 

(2) osphradial length at least one-half 
ctenidial length (character 8). 

(3) accessory salivary glands short to me- 
dium (character 15). 

Clade F: 

(1) operculum D-shaped, with tapered 
lower end, S-shaped left edge, and 
with lateral nucleus in lower right 
(character 5). 

(2) rodlike structures in the hypobranchial 
gland (character 6). 

(3) 1-3 large seminal receptacles lying 
over the dorsal periphery of albumen 
gland, and branching off ovi-sperm 
duct (character 10). 

(4) penis with large side lobe (character 
12). 

(5) rachidian with very wide, smooth mar- 
ginal area, without marginal cusps, 
with small inner lateral denticle free 
from lateral cusp, and with central 
cusp much longer than lateral cusps 
(character 18). 

Clade G: 

(1) layer of calcite thin (character 3). 

(2) innermost aragonitic shell layer with 
crystal planes oriented in 45° angle to 
growing edge (character 4). 

(3) short marginal area with small mar- 
ginal cusps; inner lateral denticle small 
or absent; lateral cusps nearly equal in 
length to central cusp which is wide at 
base (character 18). 

A thin calcitic layer appears to have 
evolved in a parallel manner in one taxon in 
Clade A (Haustrum) and in two taxa within 
Clade С {Cymia, Rapana). This layer is ab- 
sent in many taxa of Clade E (reversal as 
synapomorphy for this Clade) and is present 
again in the taxa of Clade G. This character- 



state distribution suggests that this character 
needs more detailed study and that the pat- 
tern of parallelism, convergence and reversal 
in character 3 may only be the result of inad- 
equate understanding of this character. 

Congruence between Proposed Phylogeny 
and Fossil Record 

There are several reasons for not basing a 
branching sequence on the fossil record of 
rapanines a priori. First, rapanines do not fos- 
silize well in their rocky intertidal environment 
and have a poor, incomplete fossil record. 
Thus, an extant taxon with a short fossil his- 
tory may be part of a primitive lineage with 
fossil members which have either not yet 
been discovered or have not been identified 
as close allies of the extant species. 

The second reason for not using the fossil 
record a priori is the problem of taxon identi- 
fication, especially above the species level, 
which at most may be based on superficial 
shell characters. It is difficult to identify phy- 
logenetic relationships among Recent taxa on 
the basis of external shell morphology alone 
and even more so to determine phylogeny 
from fossil shells. For example, because of 
convergence in shell shape, what may be 
identified as a fossil species of Morula may 
not be related to Recent Morula s.s. species. 

Thirdly, fossil records taken from the litera- 
ture are often unreliable because limits have 
not been set for most rapanine genera. This 
causes the scope of genera to vary widely 
among authors. For example, some of the 
fossil records of so-called 'Thais s.s." may 
not be based on fossils of the type species of 
Thais, which has a very limited geographical 
distribution. Rather, they may be based on 
fossils of the nominal species " haemastoma," 
which many authors have placed under 
Thais, but is herein shown to belong in the 
genus Stramonita. If Stramonita had a longer 
fossil record than Thais s.S., the geological 
record of Thais would be erroneously set 
back to the time Stramonita appeared. 

Finally, it is nearly impossible to determine 
the geological origin of a genus prior to know- 
ing which species should be included in that 
genus; the record of a genus may be based on 
a geologically younger species (e.g. the type), 
while other (older) members of that genus are 
incorrectly allocated to another genus. 

It is clear — to the dismay of many paleon- 
tologists — that the meager fossil record (in 
this case of the Rapaninae), cannot a priori be 
interpreted with any degree of certainty. Nev- 



240 



KOOL 



ertheless, the fossil record is potentially use- 
ful. A phylogenetic tree resulting fronn suites 
of primarily anatomical, radular, shell ultra- 
structural, and protoconch characters can be 
compared to ultrastructural data supplied 
from the fossil record (for example Ecphora). 
Furthermore, congruence between the phylo- 
genetic hypothesis (tree topology) and the 
fossil record can then support a cladogram 
and at least suggest relationships. A detailed 
study of the shell ultrastructure of fossil Ra- 
paninae and closely related taxa may provide 
further insight into evolutionary relationships 
among both extant and fossil taxa. 

Congruence of Proposed Phylogeny with 
Recent Zoogeographical Patterns 

A comprehensive study, ideally of mono- 
graphic nature, based on character suites 
(such as presented in this study), is neces- 
sary prior to determining the zoogeographical 
range of a genus. Only after questions of re- 
lationship among species have been solved, 
distribution patterns for genera may appear 
and can be interpreted. For example, the dis- 
tribution of the genus Nucella is far more ex- 
tensive if some "Thais" species from the 
South African Province are shown to belong 
to Nucella s.s. I predict that many range ex- 
tensions of genera treated herein will be re- 
vised when new limits are set for each genus. 

Preliminary geographical patterns for the 
genera are discussed below, following the 
branching sequence of the consensus cla- 
dogram (Fig. 30). 

Clades A, В (Fig. 30): The genus Nucella oc- 
curs from the eastern Atlantic (northern Eu- 
rope) to the western Atlantic (northeastern 
U.S.) Ocean and in the North Pacific (Cal- 
ifornia to the Aleutians to Japan). Preliminary 
anatomical data (Kool, unpublished data) 
suggest that the South African muricids, 
"Thais" dubia (Krauss, 1848), "T." squamosa 
(Lamarck, 1816), and "7." wahlbergi {Krauss, 
1848), are ocenebrines; further research may 
reveal that these species should be placed in 
Nucella, as suggested by Kilburn & Rippey 
(1 982), thus extending the range of the genus 
Nucella considerably. Porrería is limited to the 
North American West Coast. If future studies 
reveal that this genus is synonymous with 
Chorus Gray, 1847, the range would be ex- 
tended to northwest South America. The ge- 
nus Haustrum is limited in distribution to New 
Zealand (some records from Australia). The 



Recent terminal taxa of Clade A (Fig. 30) live 
in cool to cold water environments. This sim- 
ilarity in habitat may be considered an addi- 
tional synapomorphy of Clade A. 

Clade C: This clade has representatives from 
the Atlantic, eastern Pacific, and Indo-Pacific 
oceans. Only minor patterns can be detected 
in this clade when superimposing geographic 
distribution onto the topology of the tree. Most 
of the genera in the Rapaninae {Rapana, Vex- 
illa, Nassa, Pinaxia, Drupa, Cronia, Purpura, 
and Mancinella) have representatives only in 
the Indian and Pacific oceans. Rapana inhab- 
its the Black Sea in addition, but was intro- 
duced there by man. Nassa comprises at 
most two species, N. serta and N. "fran- 
colina," the former occurring in the Indian 
Ocean, the latter in the central and western 
Pacific Ocean and on the Cocos-Keeling Is- 
lands (Maes, 1967). However, these two taxa 
may be conspecific (see "Remarks" under 
treatment of Nassa). A similar distribution pat- 
tern is found in the genus Drupa: Drupa lo- 
bata (Blainville, 1 832), from the Indian Ocean, 
and D. grossularia, from the Pacific Ocean 
and Cocos-Keeling Islands (Maes, 1967), 
may also be conspecific. Other species of 
Drupa, such as D. morum and D. ricinus, oc- 
cur throughout the Indo-Pacific. Although 
most species of Morula live in the Indo-Pa- 
cific, some representatives inhabit the (sub) 
tropical Atlantic (Kool, unpublished data) and 
eastern Pacific Oceans. 

Cymia tecta, the only living representative 
of the genus Cymia (Clade C, at base, Fig. 
30), is limited to the Panamic Province, as are 
Vasula melones, Neorapana muricata, and 
Tribulus planospira (Clade G). Several spe- 
cies of Stramonita and Thais are known from 
the tropical eastern Pacific as well, but the 
type of Stramonita occurs in the (sub)tropical 
eastern and western Atlantic, and so does the 
type of Thais. I suspect that future studies of 
" Stramonita-Wke" and "7A7a/s-like" taxa from 
the Indo-Pacific may reveal that Stramonita 
and Thais, like Morula, have an almost global 
distribution. 

The monotypic genera Concholepas and 
Dicathais have limited distributions. Conch- 
olepas is found exclusively in western South 
America (Chile), while Dicathais is endemic to 
temperate Australia and New Zealand. Fos- 
sils of what are believed to be representatives 
of Concholepas have been reported from 
Australia (Vokes, 1972: 31) and South Africa 
(Kensley, 1985). 



PHYLOGENY OF RAPANINAE 



241 



Plicopurpura has one representative in the 
Panamic Province, and one in the western 
Atlantic (see "Remarks" under treatment of 
this genus, and Kool, 1988b). Occurrence of 
what appears to be a Plicopurpura species in 
Réunion and Mauritius (Drivas & Jay, 1 987) is 
under investigation. 

Protoconchs: Reproductive Mode and 
Phylogenetic Implications 

Protoconch morphology has been shown to 
be indicative, at least to a degree, of relation- 
ship and modes of development of gastro- 
pods (Shuto, 1974; Jablonski, 1982). A pau- 
cispiral, smooth protoconch, with smooth 
transition from protoconch to teleoconch, is 
usually indicative and typical of species with a 
crawl-away larva. A multispiral protoconch 
with varying degrees of sculpture, outward- 
flaring lip, and sinusigeral notch for accom- 
modation of the velar lobes, is usually indica- 
tive of a planktonic larval phase. 

The species used as outgroup in the cla- 
distic analysis, the muricine, Muricanthus ful- 
vescens, has the greatest number of proto- 
conch whorls (4.5-4.75), and a pattern of 
microscopic pustules on most of its whorls, 
with an outward-flaring lip and sinusigeral 
notch (Fig. 24C, F). The protoconch of Nu- 
cella is smooth, paucispiral (about 1.25 
whorls), and has a smooth transition into the 
teleoconch (Fig. 15C, D). In contrast to Nu- 
cella, all rapanine genera examined have 
multispiral protoconchs, varying from two to at 
least 4.25 whorls (completely intact speci- 
mens of protoconchs may reveal numbers as 
high as 4.75), with outward-flaring lip and si- 
nusigeral notch, and with sculptural patterns 
varying from subsutural plicae to pustulate 
whorls. 

Within Glade D no distinct trend in reduc- 
tion or increase in number of whorls is visible; 
some of the highest numbers of whorls occur 
in Glade F {Morula, Cronia). Most rapanine 
species have three to four protoconch whorls. 
Concholepas, Thais, Plicopurpura, and Vex- 
illa, have a relatively low number of whorls, 
varying from two to about three. 

A certain degree of convergence in proto- 
conch morphology is apparent. Although the 
rapanine protoconch usually has one to three- 
and-a-half more whorls than the protoconch 
of the ocenebrines herein examined, Vexilla 
is an exception in having only two whorls. A 
very high number of whorls is found both in 



the outgroup and in the rapanines. Morula 
and Nassa. 

Despite some degree of convergence in 
protoconch whorl number, the cladogram pro- 
vides great predictive power for missing data 
on protoconch morphology. For example, I 
predict that well-preserved protoconch spec- 
imens of the species of Glade G (Fig. 30) will 
reveal a sculptural pattern as found in most 
members of Glade E (3-4.5 whorls, with sub- 
sutural plicae). The cladogram furthermore 
predicted that Haustrum haustorium has a 
paucispiral, smooth protoconch, which I found 
confirmed in Suter (1913) prior to the final 
computer analysis. Scanning electron micro- 
graphs will reveal if the protoconch of Haus- 
trum haustorium lacks an outward-flaring lip 
and sinusigeral notch, as suggested by the 
cladogram. The protoconch of Cymia is more 
difficult to predict because of its position be- 
tween the ocenebrine clade (Glade A, Fig. 30) 
and the remaining members of the rapanine 
clade (Glade D). 

Evidence obtained from protoconch mor- 
phology indicates that all members of the Ra- 
paninae studied herein (Glade G, Fig. 30) 
probably have planktonic larvae. It has al- 
ways been believed that rapanine ("thaidine") 
gastropods displayed two very different 
modes of development: lecithotrophic (direct) 
and planktotrophic (indirect). For example, 
Nucella, traditionally included in Thaididae/ 
пае of authors, has direct development with 
"crawl-away" hatchlings (Ankel, 1937; Spight, 
1979) and lays egg capsules containing nurse 
eggs (Spight, 1979). However, as shown pre- 
viously (Kool, 1993), Nucella is to be ex- 
cluded from Rapaninae and to be included in 
Ocenebrinae. It is now clear that a planktonic 
larval stage is typical for Rapaninae and that 
the direct mode of development is a synapo- 
morphy for Glade В (Fig. 30) and, perhaps, for 
Glade A if Haustrum is revealed to be leci- 
thotrophic. 

It should be noted that although one basic 
protoconch type is present in the Rapaninae 
(multispiral and [usually] sculptured), and an- 
other in the Ocenebrinae (paucispiral and 
smooth), protoconch morphology varies 
greatly within the Muricinae. Therefore, de- 
pending on which muricine species is used as 
outgroup, the character state "multispiral" is 
either the apomorphic or the plesiomorphic 
condition. Perhaps the muricine outgroup 
should be coded "either multispiral, sculp- 
tured or paucispiral, smooth" in future analy- 
ses. 



242 



KOOL 



Phylogenetic Relationships Between 
Rapaninae and Other Muricid Taxa 

In this study two taxa were examined in 
less detail {Acanthina and Trochia). Some of 
the data on these lesser-understood taxa in- 
dicate or, at least, suggest their relationships 
with the taxa studied in detail. An "incom- 
plete" and sometimes scattered data base 
based on anatomical, radular, protoconch, 
opercular, and shell ultrastructural charac- 
ters, yielded several conclusions about phy- 
logenetic relationships between taxa studied 
in detail and those within the Muricidae. 

For example, a few anatomical, proto- 
conch, and shell ultrastructural data suggest 
that Acanthina is very closely related to Nu- 
cella and should also be excluded from Ra- 
paninae. Nucella and Acanthina both ap- 
peared in the Miocene, and Acanthina also 
occurs in cold to temperate waters (Califor- 
nia — North Mexico, Chile), and overlaps in 
geographic range with the range of Nucella 
emarginata (Deshayes, 1839). 

The monotypic genus Trochia from South 
Africa, with a paucispiral protoconch of about 
1 .5 whorls (Fig. 28C, D), and similar to Nu- 
cella in shell ultrastructure (Fig. 15C, D), 
should also be excluded from Rapaninae. Re- 
sults from future anatomical studies may re- 
veal justification for synonymization of Tro- 
chia with Nucella. Kilburn & Rippey (1982) 
referred the nominal species, cingulata, to 
Nucella instead of Trochia. Egg capsule mor- 
phology, however, differs greatly among Tro- 
chia cingulata and members of Nucella 
(Kilburn & Rippey, 1982; D'Asaro, 1991). 

Porrería (Fig. 26A-F) may be closely re- 
lated to the genus Chorus, an eastern Pacific 
genus from the Chilean waters. Future stud- 
ies may show that Chorus and Porrería are 
merely synonyms. Both genera have a labial 
tooth (a structure also found in Acanthina), 
and have a very similar, distinct shell shape. 

The fossil genus Ecphora (Fig. 29A-E), has 
been allocated to different muricid fami- 
lies [Rapanidae (Wenz, 1941); Thaididae 
(Petuch, 1988, in Ecphorinae Petuch); Muri- 
cidae (Ward & Gilinsky, 1988)]. The proto- 
conch of Ecphora cf. quadricostata (Say, 
1824) (Fig. 29C, D) is multispiral and counts 
about three smooth whorls, similar to Cronia 
and Dicathais, but lacks an outward-flaring lip 
and sinusigeral notch as does, for example, 
Nucella. Based on these criteria it could be- 
long to either the Ocenebrinae or the Rapani- 
nae. The shell ultrastructure consists of an 



aragonitic layer with crystal planes oriented 
perpendicular to growing edge (15-30%), an 
aragonitic layer with crystal planes oriented 
parallel to growing edge (25-35%), and a cal- 
cific layer (45-55%) (Fig. 29E). This type of 
shell ultrastructure is found in Nucella and re- 
lated taxa, such as Trochia and Porrería, but 
also in Concholepas and Dicathais. The shell 
of Ecphora (Fig. 29A, B) bears resemblance 
to both the ocenebrine Trochia (Fig. 28A, B) 
and the rapanines Dicathais (Fig. 9A, B) and 
Rapana (Fig. 25A). However, based on the 
absence of an outward-flaring lip and sinusig- 
eral notch, I place Ecphora provisionally in the 
Ocenebrinae. 

The protoconch and radula of Urosalpinx 
cinerea (Say, 1822) (Fig. 27E-G) are very 
similar to those of Nucella (Fig. 15C-F). Fur- 
ther studies of Urosalpinx species are likely to 
confirm a close tie with Nucella. Although 
Urosalpinx lacks a calcific outer layer 
(Petitjean, 1 965), it may belong in a clade with 
Nucella, Acanthina, Trochia, and Porrería. 

Radular Evolution in the Rapaninae 

Patterns of rapanine radular morphology 
are not usually congruent with present taxo- 
nomic classifications of rapanines and closely 
allied muricids (Bändel, 1984; Fujioka, 1985; 
Kool, 1987), because these classifications 
are based solely on shell morphology and are 
thus unreliable (see INTRODUCTION). Now 
that monophyly has been established for the 
Rapaninae, patterns in radular morphology 
can be discussed against a phylogenetic 
background. Comparisons between findings 
presented here and reports from the literature 
are discussed below in an order reflective of 
the branching sequence in the cladogram 
(Fig. 30). 

Clade A: Troschel (1866-1893) included 
Haustrum haustorium in the genus Polytropa 
( = Nucella), based on the width of the rachid- 
ian tooth. Cooke (1919) pointed out that the 
rachidian tooth in Haustrum (Fig. 1 1 D) is very 
different from the rachidian found in Nucella 
(Fig. 15F) and Porrería (Fig. 26E), and sug- 
gested that either Haustrum was the "progen- 
itor" of the Thais and Nucella groups (making 
a clear distinction between the "Nucella" 
group and the "Thais" group [pp. 103, 109]), 
or was derived from one of them. Later in the 
same paper, he stated that Haustrum is prim- 
itive. Troschel (1866-1893) suspected a 
close tie between Nucella and Acanthina but 



PHYLOGENY OF RAPANINAE 



243 



proclaimed separate generic status for both 
taxa. The position of Nucella, Acanthina and 
Haustrum on the cladogram (Fig. 30) is 
largely congruent with both Troschel's and 
Cooke's conclusions. 

According to Cooke (1919) and Wu (1968) 
there are some similarities between the bases 
of the rachidian teeth of Morula and Nucella, 
suggesting a relatively close tie between 
these two genera. Bändel (1984) noted close 
similarity between the radula of Ocenebra er- 
inacea and a Morula radula depicted by Cer- 
nohorsky (1969). These conclusions are not 
supported by the branching pattern in the cla- 
dogram. Kool (1993) has shown the high de- 
gree of similarity in radular morphology be- 
tween Ocenebra and Nucella. 

Clade C: Cymia (Fig. 8H) is considered a 
"link between Morula and Thais" by Cooke 
(1919) who based this conclusion on radular 
resemblances among these three genera. 
Cymia has a radular morphology somewhat 
atypical of rapanines and, derived from the 
cladogram, is the most primitive member of 
the rapanines examined herein. 

Tanaka (1958) deemed the rachidian tooth 
of Rapana (Fig. 25C) to be very similar to that 
of Purpura (Fig. 18D). I do not agree; the 
rachidian of Rapana has three large cusps 
and no marginal area, or marginal cusp, 
whereas Purpura has a wide marginal area 
with well-developed denticles and a pro- 
nounced marginal cusp. 

Clade D: Troschel (1866-1893) placed 
Nassa (Fig. 13G) close to Plicopurpura (as 
'Tatellipurpura") (Fig. 17E), based on rachid- 
ian tooth morphology. Cooke (1919) dis- 
agreed, placing Nassa close to Vexilla (Fig. 
23C). Furthermore, Cooke (1919) placed the 
genera Rapana, Concholepas, Pinaxia, and 
Drupa close to Thais. I agree with Cooke on 
the close evolutionary relationship between 
Nassa and Vexilla, and the close ties among 
the other four taxa, although Rapana and 
Concholepas are located at the base of Clade 
D. 

Cooke (1919) considered the morphology 
of the rachidian tooth in the genus Plicopur- 
pura (Fig. 17E) distinct enough to justify sep- 
aration of this genus (as "Patellipurpura Dall") 
from Thais (Fig. 20F) (and, presumably, from 
Purpura). My conclusions are in agreement 
with those of Cooke (Kool, 1988b). Cooke 
also stated that the rachidian tooth morphol- 
ogy must be primitive, based on the distribu- 
tion of this genus (occurring on both sides of 



the Panamic Isthmus). I do not agree with this 
statement; the rachidian tooth morphology of 
Plicopurpura is unique and should be consid- 
ered as derived. 

Clade F: Authors generally agree that the 
rachidian teeth of Cronia (Fig. 8D) and Morula 
(Fig. 12G) are extremely similar (Cooke, 
1919), and that Morula and Drupa (Fig. IOC) 
are more distantly related than their shell mor- 
phologies suggest (Cooke, 1919; Emerson & 
Cernohorsky, 1973). The tree (Fig. 30) and 
data presented by Kool (1987) show that 
Drupa and Morula are not sister taxa. 

Clade G: Arakawa (1962) allotted full generic 
status to Mancinella, based on the morphol- 
ogy of the rachidian tooth (Fig. 111). I agree 
and recognize Mancinella as a full genus. 
Cooke (1918) proposed the subgenus Neora- 
pana under Acanthina for Acanthina muri- 
cata. He considered Neorapana to be a close, 
New World relative of Rapana based on rad- 
ular and shell morphology. (Note: his drawing 
of a Neorapana muricata rachidian tooth does 
not resemble that of Neorapana muricata.) 

Fujioka (1985a) suggested from ontoge- 
netic data that a complex pentacuspid 
("comb-" or "sawlike") rachidian tooth may be 
a primitive condition in Thaidinae of authors, 
whereas a simple monocuspid rachidian tooth 
may represent a derived condition. He pre- 
sented a pattern of transformations in radular 
morphology for several genera and species 
(including Nucella and other non-rapanines). 
The major drawback of using terms such as 
"comblike" or "sawlike" or as "pentacuspid" 
or "tricuspid" is that a division in these cate- 
gories is artificial and may not reflect homol- 
ogy. Furthermore, they are too general and 
allow for different interpretations. For exam- 
ple, I would interpret the "sawlike" condition 
in Drupa as more comblike and homologous 
with the comblike condition in Purpura; addi- 
tionally, I consider the "sawlike" condition in 
Drupa as being very different from the sawlike 
condition in Nucella, or in Concholepas. 

The cladogram (Fig. 30) is, however, con- 
gruent in some aspects with the pattern dis- 
cussed by Fujioka (1985a). "Sawlike" radula 
are found in several taxa at the bases of 
Clades D and E (Fig. 30) {Rapana, Stra- 
monita, Concholepas, and Dicathais), as well 
as in the taxa Nucella and Porrería (Clade B; 
non-rapanines). Some of the other taxa on 
Clades E and G have relatively narrow, tricus- 
pid rachidians {Nassa, Mancinella), several of 
which have only small lateral cusps {Neora- 



244 



KOOL 



pana, Vexilla, Plicopurpura). Haustrum, a 
non-rapanine, clearly has a wide, pentacus- 
pld, but not comblike, rachidian tooth. A more 
or less comblike condition occurs only in more 
derived rapanlnes, such as Drupa, Purpura, 
and Pinaxia, and appears to be the derived 
condition. Morula and Cronia both have a 
wide rachidian due to the wide marginal area, 
but only the central cusp is well developed in 
these taxa. 

Several other authors have attempted to 
group muricids on the basis of rachidian cusp 
number (tricuspid and pentacuspid [Arakawa, 
1962; Wu, 1965b, 1967, 1973]). However, as 
is clear from this paper, divisions in Muricidae 
based on this character, result in para- and 
polyphyletic groups. Only after monophyly 
has been established can this character be 
used to provide a basis for further resolution 
within clades. 

Evolution in Egg Capsule Morphology 

Patterns in egg capsule morphology are 
not obvious. The egg capsules of Haustrum 
haustorium, a non-rapanine, resemble those 
of the rapanine Purpura pérsica, and the egg 
capsules of Nucella spp. are also similar to 
those of certain rapanines. 

Habe (1960) recognized two different types 
of egg capsules in muricids: (1) vase-shaped 
or pillar-shaped, with a short stalk (e.g. Fig. 
6A), and (2) lenticular, with a broad base. He 
included several species from the Muricinae, 
Thaidinae (of authors), and two species of the 
Rapaninae (of authors) in the first category, 
other muricids (trophonines etc.) in the 
second. This division is too simplistic, and nu- 
merous exceptions can be found (for exam- 
ple. Purpura bufo and Thais deltoidea have 
egg capsules with broad bases and lack a 
stalk). 

Bändel (1976) provided a phylogenetic hy- 
pothesis for evolution of egg capsule mor- 
phology, after recognizing different "Formen- 
gruppe." He placed members of Nucella, 
Thais, Stramonita (as "Thais"), and Rapana 
together into one of these categories, exclu- 
sive of Thais deltoidea, which he placed into a 
category with members of Coralliophila. This 
indicates a case of convergence in egg cap- 
sule morphology. 

When the egg capsule morphologies of 
more rapanine type species, some of which 
were recently described and illustrated by 
D'Asaro (1991), become known, a search for 



overall patterns in egg capsule morphology 
may reveal certain evolutionary trends. 

Systematic Conclusions and New 
Taxonomic Arrangement 

The cladogram (Fig. 30) indicates that 
Thaididae/nae of authors is paraphyletic and 
consists of two taxonomic groups: Clade A, 
comprising Haustrum, Nucella, Porrería, 
Acanthina, and Trochia; and Clade C, com- 
prising Cymia, Rapana, Stramonita, Conch- 
olepas, Dicathais, Vasula, Thais, Tribulus, 
Neorapana, Purpura, Mancinella, Drupa, Pli- 
copurpura, Pinaxia, Vexilla, Nassa, Morula, 
and Cronia. However, a clear cut-off point for 
either group is not obvious; some parallelism 
is evident in several character states found in 
members of Clade A and in taxa at the base 
of Clade С (long accessory salivary glands, 
separate ventral pedal gland [females] and 
boring organ, very thick outer calcific layer, 
lack of posterior seminal receptacles [fe- 
males]). Furthermore, the tree topology re- 
veals a parallelism in the morphology of the 
prostate duct [males] (not in open connection 
to mantle cavity) between Haustrum and the 
members of Clade С These taxon groups are 
not sufficiently distinct from one another, nor 
are they sufficiently distinct from Muricinae to 
warrant family status for either Clade A or С 
I therefore agree with Ponder (1973) that the 
family Muricidae contains several subfami- 
lies, and that Muricoidea includes, amongst 
other groups, the Buccinidae and Muricidae. 

The taxonomic revision of the Thaididae/ 
пае of authors (Clades A and C, Fig. 30) has 
important nomenclatural consequences. 
First, the taxa on Clade A are placed in the 
Ocenebrinae (Kool, 1993) rather than Thaid- 
inae. Secondly, the higher category name of 
the taxa in Clade С (the remains of Thaididae/ 
пае of authors) needs to be reevaluated. Be- 
cause Rapana is monophyletic with the other 
taxa in Clade С (Fig. 30) the name for this 
natural group becomes Rapaninae Gray, 
1853, which has priority over Thaidinae Jous- 
seaume, 1888, rendering Thaidinae a junior 
subjective synonym of Rapaninae. 

The high degree of similarity in radular 
morphology among Tribulus, Neorapana, and 
Vasula of unresolved Clade G (Fig. 30), and 
the fact that two of these taxa are monotypic. 
suggests that these taxa should be allotted 
subgeneric status under Thais. Perhaps 
further studies will justify synonymization 
of these genera with Thais. Mancinella and 



PHYLOGENY OF RAPANINAE 



245 



Purpura, however, are sufficiently different 
from the other four taxa and from one another 
to be conserved as separate genera. In the 
more resolved output trees, the latter two taxa 
are separate from the other four, which often 
form a polytomy in many of the trees. 

The polytomous Clade В (Fig. 30) suggests 
a close relationship among Acanthina, Tro- 
chia, and Nucella, but the low resolution is 
most likely the result of the lack of morpho- 
logical data for the former two taxa. Data on 
the egg capsule morphology of Trochia 
(Kilburn & Rippey, 1982) support separate 
generic status for this monotypic taxon, but 
anatomical and/or molecular studies of the 
South African Nucella-Wke species are neces- 
sary before any conclusions can be drawn. 

The newly proposed classification for the 
taxa examined in this study is as follows: 



MURICOIDEA Rafinesque, 1815 
Muricidae Rafinesque, 1815 

Rapaninae Gray, 1853 

[+ Thaidinae Jousseaume, 1 
Concholepas Lamarck, 1801 
Cronia H. & A. Adams, 1853 
Cymia Mörch, 1860 
Dicathais I rédale, 1936 
Drupa Röding, 1798 
Mancinella Link, 1807 
Morula Schumacher, 1817 
Nassa Röding, 1 798 
Pinaxia H. & A. Adams, 1853 
Plicopurpura Cossmann, 1903 
Purpura Bruguière, 1 789 
Rapana Schumacher, 1817 
Stramonita Schumacher, 1817 
Thais Röding, 1798 
Л/еогарала Cooke, 1918 
Tribulus Sowerby, 1 839 
Vasula Mörch, 1860 
Vexilla Swainson, 1840 

Ocenebrinae Cossmann, 1903 

[+ Ecphohnae, Petuch, 1988 

+ Nucellinae Kozloff, 1987] 

Acanthina Fischer von Waldheim, 1 807 
Ecphora Conrad, 1843 
Ролгел/а Jousseaume, 1880 
Haustrum Perry, 1811 
Nucella Röding, 1798 
Trochia Swainson, 1840 



ACKNOWLEDGMENTS 

I wish to express my gratitude to Dr. Rich- 
ard S. Houbrick, Department of Invertebrate 
Zoology, National Museum of Natural History, 
Smithsonian Institution, for overseeing the 
progress of this study and for his assistance, 
comments and suggestions. I thank Drs. M. 
G. Harasewych and R. Hershler, from the 
same institution, for valuable comments and 
criticisms. Dr. Diana Lipscomb of The George 
Washington University shared her insights 
about phylogenetic systematics and was of 
great help in the cladistic analyses. Thanks 
are also due Dr. Robert E. Knowlton, who pro- 
vided many valuable suggestions for this 
manuscript. 

I thank Mrs. Susann G. Braden, Mr. Walter 
R. Brown and Mr. Brian E. Kahn of the Scan- 
ning Electron Microscopy Laboratory at the 
USNM. I also am grateful to Dr. Mary F. Mick- 
evich. Associate, Maryland Center for Sys- 
tematic Entomology, University of Maryland, 
and of the Smithsonian Institution, and the 
Systematic Entomology Laboratory, U.S. De- 
partment of Agriculture, for access to 
PHYSYS. Mr. J. Michael Brittsan of the Ma- 
rine Systems Laboratory, Smithsonian Institu- 
tion, kindly provided specimens of Nucella 
lapillus. Dr. Eugene V. Coan provided several 
papers which assisted in solving some taxo- 
nomic problems. I wish to extend a special 
word of thanks to Mr. Richard E. Petit, Re- 
search Associate at the Division of Mollusks 
at the National Museum of Natural History, for 
his support during the first year of my gradu- 
ate studies. 

I am further indebted to Dr. Mary E. Rice, 
Chief Scientist, and her staff at the Smithso- 
nian Marine Station, Link Port, Florida. This is 
Contribution No. 279 of the Smithsonian Ma- 
rine Station, at Ft. Pierce, Florida. 

I gratefully acknowledge the support of the 
Smithsonian's Caribbean Coral Reef Ecosys- 
tems Program, and thank Dr. Klaus Rützler 
for funding my stay at the National Museum of 
Natural History's Field Laboratory on Carrie 
Bow Cay, Belize. This is Contribution No. 338 
of the Caribbean Coral Reef Ecosystems Pro- 
gram, Carrie Bow, Belize, partly supported by 
the Exxon Corporation. 

I thank those who have assisted me dunng 
visits to their institutions; Dr. James H. 
McLean and Mr. С Clifton Coney of the Los 
Angeles County Museum; Dr. William K. Em- 
erson and Mr. Walter Sage III of the American 
Museum of Natural History; Dr. Lucius El- 



246 



KOOL 



dredge of the Marine Laboratory of the Uni- 
versity of Guam; Dr. Michael Hadfield of the 
Pacific Biomedical Marine Laboratory, Uni- 
versity of Hawaii; Dr. Winston F. Ponder of 
the Australian Museum, Sydney; Mr. and Mrs. 
Jon and Gillianne Brodie of the Institute of 
Natural Resources (University of the South 
Pacific), Suva, Fiji; Dr. Rick Steiger of the 
Gump Marine Station (University of Califor- 
nia, Berkeley), Moorea, French Polynesia; 
and Dr. Timothy M. Collins of the Smithsonian 
Tropical Research Institute, Naos, Panama, 
and his assistant, Mrs. Maria del Carmen Car- 
les, who has since become my wife. 

The following names are acknowledged for 
kindly providing room and board during my 
travels: Mr. Brian Parkinson, Viti Levu, Fiji; Dr. 
Gustav Paulay and Mrs. Bernadette Paulay- 
Holthuis, Niue; Mr. and Mrs. Gerald McCor- 
mack, Rarotonga, Cook Islands; and Drs. 
Timothy M. Collins and Laurel S. Collins, Bal- 
boa, Panama. 

I wish to thank my parents for providing me 
the opportunity to commence and complete 
my studies in the United States. Thanks and 
respect are due Ms. Robin E. Milman for pro- 
viding emotional support and for her under- 
standing and patience during my last three 
years in Graduate School. 

Financial support came from The George 
Washington University, the Lerner Fund for 
Marine Research, the Hawaiian Shell Club, 
and the National Capital Shell Club. I am 
grateful for having received a Smithsonian 
Predoctoral Fellowship, as well as funds to 
visit the Smithsonian Marine Station at Link 
Port, Ft. Pierce, and the National Museum of 
Natural History's Field Laboratory on Carrie 
Bow Cay, Belize. 

I thank Drs. Frederick M. Bayer, Winston F. 
Ponder and Gary Rosenberg for critically re- 
viewing an earlier draft of this manuscript and 
providing many helpful comments and sug- 
gestions. Drs. Alan R. Kabat, Kenneth J. 
Boss, and Mr. Richard I. Johnson assisted 
with some nomenclatorial problems. 



APPENDIX 1 

Species Examined Thaididae/nae of au- 
thors: 
Concholepas concholepas (Bruguière, 

1789) 
Cronia amygdala (Kiener, 1835) 
Cymia tecta (Wood, 1828) 
Dicattiais órbita (Gmelin, 1791) 



Drupa morum Röding, 1798 
IHaustrum fiaustorium (Gmelin, 1791) 
Mancinella alouina (Röding, 1798) 
Morula uva (Röding, 1798) 
Nassa serta (Bruguière, 1 789) 
Neorapana muricata (Broderip, 1832) *1 
Nucella lapillus (Linnaeus, 1 758) 
Pinaxia versicolor {Gray, 1839) 
Plicopurpura patula (Linnaeus, 1758) *2 
Purpura pérsica (Linnaeus, 1758) 
Stramonita haemastoma (Linnaeus, 1767) 
Ttiais nodosa (Linnaeus, 1 758) 
Tribulus planospira (Lamarck, 1822) 
Vasula melones (Duelos, 1832) 
Vexilla vexilla (Gmelin, 1791) 
Acanthina monodon (Pallas, 1774) *3 
Troctiia cingulata (Linnaeus, 1771) *3 
Ecphora cf. quadricostata (Say, 1824) *3 

Rapaninae, of authors: 
Porrería belcherí (Hinds, 1844) 
Rapana rapiformis (Born, 1778) *4 

Muhcinae: 
Muricantlius fulvescens (Sowerby, 1841) 
*5 

*1 Specimens of the type species of Neora- 
pana were typical "Neorapana tuberculata" 
(Sowerby, 1835) morphs; it appears that N. 
tuberculata and N. muricata are synonyms. 
Neorapana muricata (Broderip, 1832) is the 
senior synonym of Neorapana tuberculata 
(Sowerby, 1835) (see "Remarks" under ij 
Neorapana). ' 

*2 The type species of Plicopurpura {Plicopur- 
pura columellaris Lamarck, 1816) was not ex- 
amined, but was substituted by Its very similar 
congener Plicopurpura patula (Linnaeus, 
1758) because well-preserved anatomical 
material of this species was available (Kool, 
1988b). 

*3 These taxa were examined to test if syn- il 
apomorphies present in some taxa could be | 
recognized in these, facilitating taxonomic al- if 
location. Therefore they were only examined 
for synapomorphic (diagnostic) cfiaracters. 

*4 Rapana rapiformis (Born, 1 778) is a typical 
rapanine, but it is not the type of Rapana; it 
was included In this study because well-pre- 
served specimens were available. t 

*5 Muricanthus fulvescens (Sowerby, 1841) 
was chosen to represent the Muhcinae as an 
outgroup in the cladistic analysis, because 
many living and well-preserved specimens 
were available. 



PHYLOGENY OF RAPANINAE 

APPENDIX 2 

List of abbreviations used in text. 

AMS: Australian Museum, Sydney. 

ANSP: Academy of Natural Sciences, Philadelphia. 

LACM: Los Angeles County Museum. 

MGH: Myroslaw George Harasewych. 

SEM: Scanning electron micrograph. 

SPK: Silvard Paul Kool. 

USNM: United States National Museum. 

ZMA: Zoologisch Museum, Amsterdam. 



247 



APPENDIX 3 



Voucher numbers 

Concholepas concholepas 

USNM 706703 

AMNH 132968 

NMNH 857055 

USNM 518777 

USNM 706703 

Cronia amygdala 

USNM 836880 

USNM 836880 

USNM 836880 

USNM 795252 

Cymia tecta 

ANSP 355766 

MCZ 302757 

ANSP 355766 

USNM 589636 

USNM 216294 

Dicathais órbita 

USNM 836862 

USNM 681578 

USNM 836862 

USNM 836862 

USNM 618246 

Drupa morum 

USNM 857059 

USNM 720340 

USNM 857059 

USNM 857059 

USNM 6721 11 

Haustrum haustorium 

AMS no number 

AMS no number 

USNM 531495 

USNM 531495 

USNM 76300 

Mancinella alouina 

AMS no number 

AMS no number 

AMS no number 

USNM 669734 

Morula uva 

USNM 857058 

USNM 587364 

USNM 857058 

USNM 685003 

USNM 684893 



Anatomy: Playa Caleta, Chile 

Protoconch: Catrihue, Tierra del Fuego, Chile 

Radula: Valparaiso, Chile 

Ultrastructure: Antofagasta, Chile 

Shell: Playa Caleta, Chile 

Anatomy: Magnetic Island, Queensland, Australia 
Radula: Magnetic Island, Queensland, Australia 
Ultrastructure: Magnetic Island, Queensland, Australia 
Shell: Collaroy, New South Wales, Australia 

Anatomy: Vera Cruz, Panama 

Anatomy: Punta Guanico, Panama 

Radula: Vera Cruz, Panama 

Ultrastructure: Venado Beach, Ft. Knobbe, Canal Zone, Panama 

Shell: Panama City, Panama 

Anatomy: Botany Bay, New South Wales, Australia 
Protoconch: Omapere, Hokianga Harbour, New Zealand 
Radula: Botany Bay, New South Wales, Australia 
Ultrastructure: Botany Bay, New South Wales, Australia 
Shell: Ulladulla Harbour, New South Wales, Australia 

Anatomy: Pago Bay, Guam, U.S.A. 

Protoconch (D. grossularia): Garumaoa Island, Tuamotu Islands 

Radula: Pago Bay, Guam, U.S.A. 

Ultrastructure: Pago Bay, Guam, U.S.A. 

Shell: Tongatapu, Tonga Islands 

Anatomy: Titirangi Bay, New Zealand 
Radula: Titirangi Bay, New Zealand 
Ultrastructure: Rangitoto Island, New Zealand 
Shell: Rangitoto Island, New Zealand 
Shell: New Zealand 

Anatomy: Lizard Island, Queensland, Australia 
Radula: Lizard Island, Queensland, Australia 
Ultrastructure: Lizard Island, Queensland, Australia 
Shell: Pescadores Islands, China Sea 

Anatomy: Pago Bay, Guam, U.S.A. 
Protoconch: Kwajalein Atoll, Marshall Islands 
Radula: Pago Bay, Guam, U.S.A. 
Ultrastructure: Motu Akaiami, Aitutaki, Cook Islands 
Shell: Aitutaki, Cook Islands 

(continued) 



248 



KOOL 



Nassa serta 
USNM no number 
USNM 719808 
ANSP 269309 
USNM no number 
USNM 631480 
USNM 89600 
USNM 618429 
Neorapana muricata 
USNM 836661 
USNM 60718 
USNM 836661 
USNM 836661 
USNM 749212 
Nucella lapillus 
USNM 857053 
USNM 416825 
USNM 857053 
USNM 857053 
USNM 191106 
USNM 191094 
Pinaxia versicolor 
USNM 262193 
USNM 709294 
ANSP 262193 
ANSP 262193 
USNM 673781 
Plicopurpura patula 
USNM 857056 
USNM 734594 
USNM 857056 
USNM 736748 
USNM 662235 
Purpura pérsica 
ZMA no number 
MNHL no number 
ZMA no number 
ZMA no number 
USNM 700108 
Stramonita tiaernastoma 
USNM 857063 
USNM 597536 
USNM 857063 
USNM 857063 
USNM 597536 
Thais nodosa 
USNM no number 
AMNH 5172 
USNM no number 
USNM no number 
USNM 767917 
Tribulus planospira 
LACM no number 
USNM 708234 
LACM no number 
USNM 558161 
USNM 678916 
Vasula melones 
USNM 664731 
USNM 796187 
USNM 664731 
USNM 732982 



Anatomy: Pago Bay, Guam, U.S.A. 

Protoconch (Л/. "francolina"): Nossi Be, Madagascar 

Larval shell: Gatope island. New Caledonia 

Raduia: Pago Bay, Guam, U.S.A. 

Ultrastructure: Gigmoto, Catanduanes Islands, Philippine Islands 

Shell: Samoa Islands 

Shell: Low Wooded Island, N. Queensland, Australia 

Anatomy: Puerto Peñasco, Sonora, Mexico 
Protoconch: Acapuico, Mexico 
Raduia: Puerto Peñasco, Sonora, Mexico 
Ultrastructure: Puerto Peñasco, Sonora, Mexico 
Shell: San Carlos, Sonora, Mexico 

Anatomy: Kittery, Maine, U.S.A. 

Protoconch: Manchester, Massachusetts, U.S.A. 

Raduia: Kittery, Maine, U.S.A. 

Ultrastructure: Kittery, Maine, U.S.A. 

Shell: Shetland Islands, Scotland 

Shell: Balta Sound, Shetland Islands, Scotland 

Anatomy: Ambatoloaka, Madagascar 
Protoconch: Kuri Island, Hawaii, U.S.A. 
Raduia: Ambatoloaka, Madagascar 
Ultrastructure: Ambatoloaka, Madagascar 
Shell: Mogadishu, Somalia 

Anatomy: South Miami Beach, Florida, U.S.A. 
Protoconch: San Bias Islands, Panama 
Raduia: South Miami Beach, Florida, U.S.A. 
Ultrastructure: Cozumel Island, Mexico 
Shell: Mujeres Island, Mexico 

Anatomy: Krakatoa, Indonesia 

Protoconch: Tjoba, Tidore, Indonesia 

Raduia: Krakatoa, Indonesia 

Ultrastructure: Krakatoa, Indonesia 

Shell: Taiohae Bay, Nukuhiva, Marquesas Islands 

Anatomy: Sebastian, Florida, U.S.A. 
Protoconch: Cocoa Beach, Florida, U.S.A. 
Raduia: Sebastian, Florida, U.S.A. 
Ultrastructure: Sebastian, Florida, U.S.A. 
Shell: Cocoa Beach, Florida, U.S.A. 

Anatomy: Ascension Island 
Protoconch: Cape Verde Islands 
Raduia: Monrovia, Liberia 
Ultrastructure: Ascension Island 
Shell: Monrovia, Liberia 

Anatomy: Galápagos Islands, Ecuador 

Protoconch: Malpelo Island, Colombia 

Raduia: Galápagos Islands, Ecuador 

Ultrastructure: Ensenada de los Muertos, Mexico 

Shell: Academy Bay, Isla Santa Cruz, Galápagos Islands 

Anatomy: Palo Seco, Panama 

Raduia: Marchena, Punta Estego, Galápagos Islands 

Ultrastructure: Palo Seco, Panama 

Shell: Stony Point, Ft. Amador, Panama 



PHYLOGENY OF RAPANINAE 



249 



Vexilla vexillum 
USNM 836956 
USNM 718391 
USNM 836956 
USNM 836956 
USNM 622852 
Forreria belcheri 
USNM no number 
USNM no number 
USNM 169034 
Collection MGH 
Rapana rapiformis 
BMNH no number 
USNM 655026 
BMNH no number 
BMNH no number 
BMNH no number 
Muricanthus fulvescens 
USNM 857064 
USNM 621380 
USNM 857064 
USNM 857064 
Collection SPK 
Acanthina monodon 
USNM 2778 
USNM 131004 
Trochia cingulata 
AMNH 128952 
AMNH 128952 
USNM 2752 
Urosalpinx cinerea 
USNM no number 
USNM no number 
Ecphora cf. quadhcostata 
USNM no number 
USNM no number 
MCZ 263350 



Anatomy: Pupukea Beach, Oahu, Hawaii, U.S.A. 
Protoconch: Tulear, Madagascar 
Radula: Pupukea Beach, Oahu, Hawaii, U.S.A. 
Ultrastructure: Pupukea Beach, Oahu, Hawaii, U.S.A. 
Shell: Mauke, Cook Islands 

Anatomy: Off San Francisco, California, U.S.A. 
Radula: Off San Francisco, California, U.S.A. 
Ultrastructure: San Pedrao, California, U.S.A. 
Shell: Catalina Island, California, U.S.A. 

Anatomy: Ause Major, Mahe, Seychelles 
Protoconch: South Pagi Island, Indonesia 
Radula: Ause Major, Mahe, Seychelles 
Ultrastructure: Ause Major, Mahe, Seychelles 
Shell: Ause Major, Mahe, Seychelles 

Anatomy: off Cape Canaveral, Florida, U.S.A. 
Protoconch: 30°18'N, 88°34'W, Gulf of Mexico 
Radula: off Cape Canaveral, Florida, U.S.A. 
Ultrastructure: off Cape Canaveral, Florida, U.S.A. 
Shell: off Cape Canaveral, Florida, U.S.A. 

Protoconch: Valparaiso, Chile 
Shell: Valparaiso, Chile 

Protoconch: Sea Point, Cape Town, South Africa 
Ultrastructure: Sea Point, Cape Town, South Africa 
Shell: Cape Good Hope, South Africa 

Protoconch: Ft. Pierce, Florida, U.S.A. 
Radula: Ft. Pierce, Florida, U.S.A. 

Protoconch: St. Mary's Co., Maryland, U.S.A. 
Ultrastructure: St. Mary's Co., Maryland, U.S.A. 
Shell: Chancellor Pt., St. Mary's Co., Maryland, U.S.A. 



LITERATURE CITED 



ABBOTT, R. T., 1974, American seashells, 2nd. ed. 
Van Nostrand Reinhold Company. New York, 
663 pp. 

ABBOTT, R. T. & S. P. DANCE, 1982, Compen- 
dium of seasfiells. Dutton, New York, 41 1 pp. 

ABE, N., 1983. Breeding of Thais clavigera (Küster) 
and prédation of its eggs by Cronia margariticola 
(Broderip). Pp. 381-392, in: B. Morton & D. Dud- 
geon, eds., 7776 malacofauna of Hong Kong and 
southern China. II., Vol. 1 Hong Kong University 
Press, Hong Kong, viii + 361 pp. 

ADAMS, H. & A. ADAMS, 1853, The genera of Re- 
cent Mollusca, Vol. 1 . Van Voorst, London, 256 

PP- 

ADANSON, M., 1757, Histoire naturelle de Séné- 
gal, coquillages. Paris, 275 pp., 19 pis. 

AGERSBORG, H. P K., 1929, Factors in the evo- 
lution of the prosobranchiate mollusc, Thais lapil- 
lus. The Nautilus, 43: 45-49. 

AMIO, M., 1957, Studies on the eggs and larvae of 



marine gastropods — I. Journal of the Shi- 
monoseki College of Fisheries, 7: 107-1 16. 

ANKEL, W. E., 1937, Der feinere Bau des Kokons 
der Purpurschnecke Nucella lapillus (L.) und 
seine Bedeutung für das Laichleben. Verhand- 
lungen der Deutschen Zoologischen Gesell- 
schan, Supplement, 10: 77-86. 

ARAKAWA, K. Y., 1957, On the remarkable sexual 
dimorphism of the radula of Drupella. Venus, 19: 
52-58. 

ARAKAWA, K. Y., 1962, A study on the radulae of 
the Japanese Muricidae. (1) The genera Pur- 
pura, Thais and Mancinella. Venus, 22: 70-78. 

ARAKAWA, K. Y., 1964, A study on the Japanese 
Muricidae. (2) The genera Vexilla, Nassa, Ra- 
pana, Murex, Chicoreus and Homalocantha. Ve- 
nus, 22: 355-364. 

ARAKAWA, K. Y., 1965, A study on the Japanese 
Muricidae. (3) The genera Drupa, Drupina, 
Drupella, Cronia, Morula, Morulina, Phrygio- 
murex, Cymia and Tenguella gen. nov. Venus, 
24: 113-126. 

ATAPATTU, D. H., 1972, The distribution of mol- 



250 



KOOL 



luscs on littoral rocks in Ceylon, with notes on 
their ecology. Marine Biology, 16: 150-164. 

BAKER, F. C, 1895, Preliminary outline of a new 
classification of the family Muricidae. Bulletin of 
the Chicago Academy of Sciences, 2: 169-189. 

BALAPARAMESWARA RAO, M. & P. V. BHA- 
VANARAYANA, 1976, Environment and shell 
variation in relation to distribution of a tropical 
marine snail. Drupa tuberculata (Blainville). Jour- 
nal of Molluscan Studies, 42: 235-242. 

BÄNDEL, К., 1976. Morphologie der Gelege und 
ökologische Beobachtungen an Muriciden (Gas- 
tropoda) aus der südlichen Karibischen See. Ver- 
handlungen der Naturforschung Gesellschaft, 
Basel, 85: 1-32. 

BÄNDEL, K., 1984, The radulae of Caribbean and 
other Mesogastropoda and Neogastropoda. Zo- 
ologische Verhandelingen, 214: 176 pp. 

BÄNDEL, К., 1987, Hydroid, amphineuran and gas- 
tropod zonation in the littoral of the Caribbean 
Sea, Colombia. Senckenbergiana Maritima, 19: 
1-129. 

BARNARD, K. H., 1959, Contributions to the knowl- 
edge of South African Mollusca. Part II. Gas- 
tropoda: Prosobranchiata: Rhachiglossa. Annals 
of the South African Museum, 45: 1-256. 

BELLARDi, L., 1882, 1 Molluschi del Terreni Terziari 
del Piemonte e delta Liguria, Vol. 3. Stamperia 
Reale, Torino, 253 pp. 

BERNARD, P. A., 1984, Coquillages du Gabon. Li- 
breville, Gabon, 140 pp., 75 pis. 

BERNSTEIN, A. S., 1970, Notes on the ecology of 
Drupa morum 1 798 [sic] in Hawaii. The Biology of 
Molluscs. Hawaii Institute of Marine Biology, Uni- 
versity of Hawaii, Honolulu, Hawaii, Technical 
Report 18: 4 (abstract). 

BERRY, R. J. & J. H. CROTHERS, 1968, Stabiliz- 
ing selection in the dog whelk (Nucella lapillus). 
Journal of Zoology, London, 155: 5-17. 

BERRY, R. J. & J. H. CROTHERS, 1970, Geno- 
typic stability and physiological tolerance in the 
dog whelk {Nucella lapillus (L.)). Journal of Zool- 
ogy, London, 162: 293-302. 

BED, A. G., 1970, Taxonomic position of Leppistes 
pehuensis Marwick, with a review of the species 
of Concholepas (Gastropoda, Muricidae). Jour- 
nal of the Malacological Society of Australia, 2: 
39-46. 

BLAINVILLE, H. DE, 1829, Cours de physiologie 
générale et comparée. Paris, 3 vols. 

BLAINVILLE, H. DE, 1832, Disposition méthodique 
des espèces Récentes et fossiles des genres 
pourpre, ricinule, licorne et concholepas de M. de 
Lamarck. 75 pp. 

BOGI, С & I. NOFRONI, 1984, A new Coralliophil- 
idae from the Bay of Vigo. La Conchiglia, 15: 
4-5. 

BOONE, С е., 1984, Search and seizure. More on 
Thais haemastoma. Texas Conchologist, 21 : 1-6. 

BORN, I. VON, 1778, Index rerum naturalium 
Musei Caesarei Vindobonensis. Officina Krausi- 
ana, Vienna, 458 pp. 

BRADLEY, J. С & К. V. W. PALMER, 1963, The 



cases of Purpura and Ceratostoma. Z.N.(S.) 
1088. Bulletin of Zoological Nomenclature, 20: 
251-254. 

BRAZIER, J., 1889, Mollusca. In: E. P. Ramsay, ed., 
Lord Howe Island, its zoology, geology, and phys- 
ical characters. Memoirs of The Australian Mu- 
seum, Sydney, 2: [10] + 132 pp., 7 pis., 4 maps. 

BRIGHT, D. A. & D. V. ELLIS, 1990, A comparative 
survey of imposex in northeast Pacific neogas- 
tropods (Prosobranchia) related to tributyltin con- 
tamination, and choice of a suitable bioindicator. 
Canadian Journal of Zoology, 68: 1915-1924. 

BRITTON, J. С & В. MORTON, 1989, Shore ecol- 
ogy of the Gulf of Mexico. University of Texas, 
Austin, Texas, 387 pp. 

BRODERIP, W. J., 1832, On new species of Ovu- 
lum, Murex, Typhis, Ranella, etc. Proceedings of 
the Zoological Society, London, 20: 173-179. 

BRODERIP, W. J., 1839, Malacology. The Penny 
Cyclopaedia ... 14: 314-325. 

BRUGUIÈRE, J. G., 1789, Encyclopédie méthod- 
ique. Histoire naturelle des vers. Paris, 1 , i-xvii 
+ 344 pp. 

BRUGUIÈRE, J. G., 1792, Description de deux co- 
quilles, des genres de l'Oscabrion et de la Pour- 
pre. Journal d'Histoire Naturelle, 1 : 20-33. 

BRYAN, G. W., P. E. GIBBS, L. G. HUMMER- 
STONE & G. R. BURT, 1986, The decline of the 
gastropod Nucella lapillus around South-West 
England: evidence for the effect of tributyltin from 
antifouling paints. Journal of the Marine Biologi- 
cal Association of the United Kingdom, 66: 61 1- 
640. 

BRYAN, G. W., P. E. GIBBS, G. R. BURT. & L G. 
HUMMERSTONE, 1987, The effects of tributyltin 
(TBT) accumulation on adult dogwhelks, Nucella 
lapillus: long-term field and laboratory experi- 
ments. Journal of the Mahne Biological Associa- 
tion of the United Kingdom, 67: 525-544. 

BURKENROAD, M. D., 1931, Notes on the Louisi- 
ana conch, Thais haemastoma Linn., in its rela- 
tion to the oyster, Ostrea virginica. Ecology, 12: 
656-664. 

CAIN, A. J., 1981, Possible ecological significance 
of variation of Cerion shells with age. Journal of 
Conchology, 30: 305-315. 

CAKE, E. W., 1983, Symbiotic associations involv- 
ing the southern oyster drill Thais haemastoma 
floridana (Conrad) and macrocrustaceans in Mis- 
sissippi waters. Journal of Shellfish Research, 3: 
117-128. 

CARRIKER, M. R., 1943, On the structure of and 
function of the proboscis in the common oyster 
drill, Urosalpinx cinerea Say. Journal of Morphol- 
ogy, 73: 441-506. 

CARRIKER, M. R., 1955, Critical review of biology 
and control of oyster drills Urosalpinx and Eu- 
pleura. V. S. Fish and Wildlife Service, Special 
Scientific Report — Fisheries, 148: 150 pp. 

CARRIKER, M. R., 1981, Shell penetration and 
feeding by naticacean and muricacean predatory 
gastropods: a synthesis. Malacologia. 20: 403- 
4?2. 



PHYLOGENY OF RAPANINAE 



251 



CARRIKER, M. P., P. PERSON, R. LIBBIN & D. 
VAN ZANDT, 1972, Regeneration of the probos- 
cis of muricid gastropods after amputation, with 
emphasis on the radula and cartilages. Biological 
Bulletin, 143: 317-331. 

CARRIKER, M. R., L. G. WILLIAMS & D. VAN 
ZANDT, 1978, Preliminary characterization of the 
secretion of the accessory boring organ of the 
shell-penetrating muricid gastropod Urosalpinx 
cinerea. Malacologia, 17: 125-142. 

CASTILLA, J. С & J. CANCINO, 1976, Spawning 
behaviour and egg capsules of Concholepas 
concholepas (Mollusca: Gastropoda: Muricidae). 
Manne Biology, 37: 255-263. 

CASTILLA, J. С & L R. DURAN, 1985, Human 
exclusion from the rocky intertidal zone of central 
Chile: the effects on Concholepas concholepas 
(Gastropoda). Oikos, 45: 391-399. 

CERNOHORSKY, W. O., 1969, The Muricidae of 
Fiji — Part II. Subfamily Thaidinae. The Veliger, 
11:293-315. 

CERNOHORSKY, W. O., 1972, Marine shells of the 
Pacific, Vol. II. Pacific Publications, Sydney, 411 
pp., 68 pis. 

CERNOHORSKY, W. O., 1980, Thaididae (Gas- 
tropoda): proposed amendment of entry in the 
Official List of Family-Group Names in Zoology. 
Z.N.(S.). Bulletin of Zoological Nomenclature, 37: 
148. 

CERNOHORSKY, W. O., 1982, The taxonomy of 
some Indo-Pacific Mollusca. Part 10. ¡Records of 
the Auckland Institute and Museum, 19: 
125-147. 

CERNOHORSKY, W. O., 1983, The taxonomy of 
some Indo-Pacific Mollusca. Part 1 1 . Records of 
the Auckland Institute and Museum, 20: 185-202. 

CHENU, J. C, 1859, Manuel de conchyliologie et 
de paléontologie conchy liologique. Paris, 508 pp. 

CHILDREN, J. G., 1823, Lamarck's genera of 
shells. The Quarterly Journal of Science, Litera- 
ture and the Arts, 16: iv + 409 pp. 

CHUKHCHIN, V. D., 1970, Functional morphology 
of Rapana. Academia Dumka, Kiev, 138 pp. (In 
Russian). 

CLENCH, W. J., 1927, A new subspecies of Thais 
from Louisiana. The Nautilus, 41 : 6-8. 

CLENCH, W. J., 1947, The genera Purpura and 
Thais in the western Atlantic. Johnsonia, 2: 61- 
75. 

CLENCH, W. J. 1948. A new Thais from Angola 
and notes on Thais haemastoma Linné. Ameri- 
can Museum Novitates, 1374: 4 pp. 

COEN, G. S., 1946, Di una nuova forma di Stra- 
monita. Atti delta Societa Italiana di Scienze Nat- 
urali, 85: 38-39. 

COLTON, H. S., 1916, On some variations of Thais 
lapillus in the Mount Desert Region, a study of 
individual ecology. Proceedings of the Academy 
of Natural Sciences of Philadelphia, 68: 
440-454. 

COLTON, H. S., 1922, Variation in the dog whelk, 
Thais {Purpura auct.) lapillus. Ecology, 3: 146- 
157. 



CONNELL, J. H., 1970, A predator-prey system in 
the marine intertidal region. 1 . Balanus glándula 
and several predator species of Thais. Ecological 
Monographs, 40: 49-78. 

CONRAD, T. A., 1837, Descriptions of new marine 
shells, from upper California. Collected by Tho- 
mas Nuttall, Esq. Journal of the Philadelphia 
Academy of Natural Sciences, 7: 227-268. 

CONRAD, T. A., 1843, Descriptions of a new ge- 
nus, and of twenty-nine new Miocene, and one 
Eocene fossil shells of the United States. Pro- 
ceedings of the Academy of Natural Sciences of 
Philadelphia, 1 : 305-31 1 . 

COOKE, A. H., 1895, Molluscs. In: S. F. Harmer & 
A. E. Shipley, eds.. The Cambridge natural his- 
tory. Vol. 3 Macmillan Press, London, [i-v], vi-xi, 
[xii-xiv] + pp. [1], 2-459 (with Index [pp. 513- 
535] for whole volume). 

COOKE, A. H., 1915, The geographical distribution 
of Purpura lapillus (L.) Part I: in palaearctic wa- 
ters. Proceedings of the Malacological Society, 
11:1 92-209. 

COOKE, A. H., 1918, On the radula of the genus 
Acanthina, G. Fischer. Proceedings of the Mala- 
cological Society of London, 13:6-11. 

COOKE, A. H., 1919, The radula in Thais, Drupa, 
Morula, Concholepas, Cronia, lopas, and the al- 
lied genera. Proceedings of the Malacological 
Society of London, 13: 90-1 1 0. 

COOMANS, H. E., 1962, The marine mollusk fauna 
of the Virginian area as a basis for defining zoo- 
geographical provinces. Beaufortia, 9(98): 
83-104. 

COSSMANN, A. E. M., 1903, Essais de paléocon- 
chologie comparée. Vol. 5. Paris, 215 pp. 

COWELL, E. B. & J. H. CROTHERS, 1970, On the 
occurrence of multiple rows of "teeth" in the shell 
of the dog-whelk Nucella lapillus. Journal of the 
Marine Biological Association of the United King- 
dom, 50: 1101-1111. 

CROTHERS, J. H., 1973, On variation in Nucella 
lapillus (L.): shell shape in populations from Pem- 
brokeshire, South Wales. Proceedings of the Ma- 
lacological Society of London, 40: 318-327. 

CROTHERS, J. H., 1974, On variation in the shell 
of the dog-whelk, Nucella lapillus (L.). 1 . Pem- 
brokeshire. Field Studies, 4: 39-60. 

CROTHERS, J. H., 1985, Dog-whelks: an introduc- 
tion to the biology of Nucella lapillus. Field Stud- 
ies, 6: 291-360. 

DALL, H. W., 1871 , Descriptions of sixty new forms 
of mollusks from the west coast of North America 
and the North Pacific Ocean, with notes on oth- 
ers already described. American Journal of 
Conchology, 7: 93-160, pis. 13-16. 

DALL, H. W., 1905, Note on the earliest use of the 
generic name Purpura in binomial nomenclature. 
Proceedings of the Biological Society of Wash- 
ington, 18: 189. 

DALL, H. W., 1909, The Miocene of Astoria and 
Coos Bay, Oregon. U. S. Geological Survey Pro- 
fessional Paper 59: 278 pp. 

DALL, H. W., 1923, Notes on Drupa and Morula. 



252 



KOOL 



Proceedings of the Academy of Natural Sciences 
of Philadelphia, 75: 303-306. 

D'ASARO, С N-, 1966, The egg capsules, embryo- 
genesis, and early organogenesis of a common 
oyster predator, Thais haemastoma floridana 
(Gastropoda: Prosobranchia). Bulletin of Marine 
Science, 16: 884-914. 

D'ASARO, С N., 1970a, Egg capsules of proso- 
branch mollusks from south Florida and the Ba- 
hamas and notes on spawning in the laboratory. 
Bulletin of Marine Science, 20: 414-440. 

D'ASARO, С N., 1970b, Egg capsules of some 
prosobranchs from the Pacific coast of Panama. 
The Veliger, 13: 37-43. 

D'ASARO, С N., 1986, Egg capsules of eleven ma- 
rine prosobranchs from northwest Florida. Bulle- 
tin of Marine Science, 39: 76-91 . 

D'ASARO, С N., 1991, Gunnar Thorson's world- 
wide collection of prosobranch egg capsules: 
Muhcidae. Ophelia, 35: 1-101. 

DAVIS, G. M., 1979, The origin and evolution of the 
gastropod family Pomatiopsidae, with emphasis 
on the Mekong River Thculinae. Monograph of 
the Academy of Natural Sciences of Philadel- 
phia, 20: 1-120. 

DAY, A. J., 1990, Microgeographic variation in al- 
lozyme frequencies in relation to the degree of 
exposure to wave action in the dogwhelk Nucella 
lapillus (L.) (Prosobranchia: Muhcacea). Biologi- 
cal Journal of the Linnean Society of London, 40: 
245-261 . 

DEMOND, J., 1957, f\/licronesian reef-associated 
gastropods. Pacific Science, 1 1 : 275-341 . 

DESMAYES, G. P., 1830, Encyclopédie méthod- 
ique. Historie naturelle des vers, Vol. 2, Part 1 , 
pp. i-vii, 1-256. 

DESMAYES, G. P., 1839, Nouvelles espèces de 
Mollusques, provenant des côtes de la Cali- 
fornie, du Mexique, du Kamtschatka et de la 
Nouvelle-Zélande, décrites par M. Deshayes. 
Revue Zoologique par La Société Cuvierienne, 2. 
356-361 . 

DESHAYES, G. P., 1844, Histoire naturelle des an- 
imaux sans vertèbres, . . . 2nd. ed., Vol. 10. Paris, 
638 pp + Index [1 p.] 

DISALVO, L. M., 1988, Observations on the larval 
and postmetamorphic life of Concholepas conch- 
olepas (Bruguiére, 1798) in laboratory culture. 
The Veliger, 30: 358-368. 

DODGE, H., 1956, A historical review of the mol- 
lusks of Linnaeus. Part 4. The genera Buccinum 
and Strombus of the Class Gastropoda. Bulletin 
of the American Museum of Natural History, 1 1 1 : 
153-312. 

DRIVAS, J. & M. JAY, 1987, Coquillages de La Ré- 
union et de l'Ile Maurice. Delachaux and Niestlé, 
Neuchâtel, Switzerland, 159 pp., 58 pis. 

DUBOIS, R., J. С CASTILLA & R. CACCIOLATTO, 
1 980, Sublittoral observations of behaviour in the 
Chilean 'loco' Concholepas concholepas (Mol- 
lusca: Gastropoda: Muhcidae). The Veliger, 23: 
83-92. 

DUCLOS, M., 1832, Description de quelques espè- 



ces de pourpres servant de type à six sections 
établies dans ce genre. Annales des Sciences 
Naturelles, 26: 1-11. 

DUMÉRIL, A. M. C, 1806, Zoologie analytique, . . . 
Paris, xxxii + 344 pp. 

DUNKER, G., 1853, Index molluscorum, . . . Cas- 
sellis Cattorum, Fischer, 74 pp. 

DUNKER, W., 1861, Mollusca Japónica descripta 
et tabulis tribus iconum. Schweizerbart, Stuttgar- 
tiae, 36 pp. 

EISENBERG, J. M., 1981, A collector's guide to 
seashells of the world. McGraw-Mill, New York, 
237 [+ 2] pp., 158 pis. 

EMERSON, W. K. & W. O. CERNOHORSKY, 1973, 
The genus Drupa in the Indo-Pacific. Indo-Pacific 
Mollusca, 3: 1-40. 

EMERSON, W. К. & А. D'ATTILIO, 1981, Remarks 
on Muricodrupa Iredale, 1918 (Muhcidae: Thaid- 
inae), with the description of a new species. The 
Nautilus, 95: 77-82. 

EMLEN, J. M., 1966, Time, energy and risk in two 
species of carnivorous gastropods. Ph.D. disser- 
tation. University of Washington, Seattle, Wash- 
ington, U.S.A., 138 pp. 

ETTER, R. J., 1987, The effect of wave action on 
the biology of the intertidal snail Nucella lapillus. 
Ph.D. dissertation, Marvard University, Cam- 
bridge, Massachusetts, i-ix + 198 pp. 

FAIRWEATHER, P. G., 1988, Movements of inter- 
tidal whelks (Morula marginalba and Thais órbita) 
in relation to availability of prey and shelter. Ma- 
rine Biology, 100: 63-68. 

FARRIS, J. S., 1 979, The information content of the 
phylogenetic system. Systematic Zoology, 28: 
483-519. 

FARRIS, J. S., 1982, Outgroups and parsimony. 
Systematic Zoology, 31 : 314-320. 

FERAL, C, 1976, Répartition géographique des 
femelles à tractus génital mâle externe chez 
Ocenebra erinacea (L.), espèce gonochorique. 
Haliotis, 7: 29-30. 

FISCMER VON WALDMEIM, G., 1807, Catalogue 
systématique et raisonné des curiosités de la na- 
ture et de l'art. Vol. 3, Végétaux et animaux. Mu- 
séum Demidoff, Moscow, i-ix -i- 330 pp. 

FISCMER, P., 1884, Manuel de conchyliologie et de 
paléontologie conchyliologique ou histoire na- 
turelle des mollusques vivants et fossiles. Part 7, 
pp. 609-688. 

FRETTER, V., 1 941 , The genital ducts of some Brit- 
ish stenoglossan prosobranchs. Journal of the 
Marine Biological Association of the United King- 
dom, 25: 173-211. 

FRETTER, V. & A. GRAMAM, 1962, British proso- 
branch molluscs. Ray Society, London, 755 pp. 

FUJIOKA, Y., 1982, On the secondary sexual char- 
acters found in the dimorphic radula of Drupella 
(Gastropoda: Muricidae) with reference to its tax- 
onomic revision. Venus, 40: 203-223. 

FUJIOKA, Y., 1984, Sexually dimorphic radulae in 
Cronia margariticola and Morula musiva (Gas- 
tropoda: Muricidae). Venus, 43: 315-330. 

FUJIOKA, Y., 1985a, Systematic evaluation of rad- 



PHYLOGENY OF RAPANINAE 



253 



ular characters in Thaidinae (Gastropoda: Muri- 
cidae). Journal of Science of ttie Hiroshima Uni- 
versity, Ser. В, Div. 1 (Zoology), 31 : 235-287. 

FUJIOKA, v., 1985b, Seasonal aberrant radular 
formation in Tfiais bronni (Dunker) and T. clavig- 
era (Küster) (Gastropoda: Muricidae). Journal of 
Experimental l\/larine Biology and Ecology, 90: 
43-54. 

GALLARDO, С. S., 1973, Desarrollo intracapsular 
de Concholepas conctiolepas (Bruguière) (Gas- 
tropoda IVluhcidae). Museo Nacional de ¡Historia 
Natural, Publicación Ocasional, 16: 3-16. 

GALLARDO, С S., 1979, El ciclo vital del muri- 
cidae Conctiolepas conctiolepas y consid- 
eraciones sobre sus primeras fases de vida en el 
bentos. Biología Pesquera Chile, 12: 79-89. 

GALLARDO, O. S., 1980, Adaptaciones reproduc- 
tivas en gastrópodos muricaceos de Chile; 
conocimiento actual y perspectivas. Investiga- 
ciones Marinas, 8: 115-128. 

GALLARDO, С S. & F. F. PERRON, 1982, Evolu- 
tionary ecology of reproduction in marine benthic 
molluscs. Malacologia, 22: 109-114. 

GALLARDO, С. S. & О. А. GARRIDO, 1989, Sper- 
miogenesis and sperm morphology in the marine 
gastropod Nucella crassilabrum with an account 
of morphometric patterns of spermatozoa varia- 
tion in the family N/luricidae. Invertebrate Repro- 
duction and Development, 15: 163-170. 

GANAROS, A. E., 1958, On development of early 
stages of Urosalpinx cinerea (Say) at constant 
temperatures and their tolerance to low temper- 
atures. Biological Bulletin, 114: 188-195. 

GIBBS, P. E. & G. W. BRYAN, 1986, Reproductive 
failure in populations of the dogwhelk, Nucella 
lapillus, caused by imposex induced by thbutyltin 
(TBT) contamination from antifouling paints. 
Journal of the Marine Biological Association of 
the United Kingdom, 66: 767-777. 

GIBBS, P. E., G. W. BRYAN, P. L. PASCOE & G. R. 
BURT, 1987, The use of the dogwhelk, Nucella 
lapillus, as an indicator of tributyltin (TBT) con- 
tamination. Journal of the Marine Biological As- 
sociation of the United Kingdom, 67: 507-523. 

GMELIN, J. F., 1791 , Systema naturae, ed. 13, Vol. 
1, Pars VI, pp. 3021-3909. 

GOHAR, H.A.F. & A. M. EISA WAY, 1967, The egg 
masses and development of five rachiglossan 
prosobranchs. Publications of the Marine Biolog- 
ical Station Al-Ghardaqa (Red Sea), 14: 215- 
268. 

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

GOULD, A. A., 1853, Descriptions of shells from 
the Gulf of California and the Pacific Coasts of 
Mexico and California. Journal of the Boston So- 
ciety of Natural History, 6: 1-35. 

GOULD, A. A., 1855, Catalogue of shells collected 
in California by W. P. Blake, with descriptions of 
the new species. Pp. 22-28, In: W. P. Blake, Pre- 
liminary Geological Report (appendix), U. S. Pa- 
cific Railroad Exploring Expedition, July 1 855. 



GOULD, S. J., 1971, Environmental control of form 
in land snails — a case of unusual precision. The 
Nautilus, 84: 86-93. 

GRAHAM, A., 1941 , The oesophagus of the steno- 
glossan prosobranchs. Proceedings of the Royal 
Society of Edinburgh, Section В (Biology), 61: 
1-23. 

GRAHAM, A., 1949, The molluscan stomach. 
Transactions of the Royal Society of Edinburgh, 
27, 61 : 737-778. 

GRAHAM, D. H., 1941, Breeding habits of twenty- 
two species of marine Mollusca. Transactions 
and Proceedings of the Royal Society of New 
Zealand, 71: 152-159. 

GRAY, J. E., 1839, Molluscous animals and their 
shells [part]. Pp. 101-155, in: F. W. Beechey, The 
zoology of Capt Beechey 's voyage . . . in His 
Majesty's Ship Blossom, London. 

GRAY, J. E., 1847, A list of the genera of Recent 
Mollusca, their synonyma and types. Proceed- 
ings of the Zoological Society of London, 15: 
129-219. 

GRAY, J. E., 1853, On the division of ctenobranch- 
ous gasteropodous Mollusca into larger groups 
and families. The Annals and Magazine of Natu- 
ral History, (2) 11: 124-133. 

GUNTER, G., 1979, Studies of the southern oyster 
borer, Thais haemastoma. Gulf Research Re- 
ports, 6: 249-260. 

GUPPY, R. J. L., 1869, Notice on some new marine 
shells found on the shores of Trinidad. Proceed- 
ings of the Scientific Association of Trinidad, 
June 1869:366-369. 

HABE, T., 1960, Egg masses and egg capsules of 
some Japanese marine prosobranchiate gastro- 
pod [sic]. Bulletin of the Marine Biological Station 
of Asamushi, 10: 121-126. 

HABE, T., 1964, Shells of the western Pacific in 
color. Vol. 2. Hoikusha, 233 pp. 

HABE, T. & S. KOSUGE, 1966, Shells of the world 
in colour. Vol. 2. Osaka, 193 pp. 

HALL, J. G. & S. Y. FENG, 1976, Genital variation 
among Connecticut populations of the oyster 
drill, Urosalpinx cinerea Say (Prosobranchia: Mu- 
ricidae). The Veliger, 18: 318-321. 

HALLAM, A., 1965, Environmental causes of stunt- 
ing in living and fossil marine benthonic inverte- 
brates. Palaeontology, 8: 132-155. 

HALLER, В., 1888, Die Morphologie der Proso- 
ЬгапсЫег. MorphologischesJahrbuch, 1 4:54-1 69. 

HARASEWYCH, G. M., 1983, A review of the 
Columbariinae (Gastropoda: Turbinellidae) of the 
western Atlantic with notes on the anatomy and 
systematic relationships of the subfamily. Nem- 
ouria, 27: 1-42. 

HARASEWYCH, G. M., 1984, Comparative anat- 
omy of four primitive muhcacean gastropods: im- 
plications of trophonine phylogeny. American 
Malacological Bulletin, 3: 11-26. 

HEDLEY C, 1902, Studies on Australian Mollusca, 
Part VII. Proceedings of the Linnean Society of 
New South Wales, 4: 596-619. 

HEDLEY, C, 1 905, Studies on Australian Mollusca, 



254 



KOOL 



Part IX. Proceedings of the Linnean Society of 
New Southi Wales, 4: 520-546. 

HEDLEY, C, 1 91 8. A check-list of the marine fauna 
of New South Wales. Part I. Mollusca. Journal 
and Proceedings of the Royal Society of New 
South Wales, 51: 1-120. 

HENNIG, W., 1966, Phylogenetic systematics. Uni- 
versity of Illinois Press, Urbana, 263 pp. (trans- 
lated from German). 

HERRMANNSEN, A. N., 1847, Indicis generum 
malacozoorum primordia, Vol. 1 . Cassellis, 637 
pp. 

HERTLEIN, L G., 1960, The subfamily Drupinae 
(Gastropoda) in the eastern Pacific. The Veliger, 
3: 7-8. 

HINDS, R. В., 1844, Descriptions of new species of 
Scalaria and ¡\/lurex, from the collection of Sir Ed- 
ward Belcher, C.B. Proceedings of the Zoological 
Society of London, 11:1 27-1 29. 

HOMBRON, J. B. & С H. JAQUINOT, 1852, Atlas 
for Dumont d'Urville: Voyage au Pole Sud. Zool- 
ogie. Paris, 143 pis. 

HOUBRICK, R. S., 1978, The family Cerithiidae in 
the Indo-Pacific. Part 1 : The genera Rhinoclavis, 
Pseudovertagus, Longicerithium and Clavocer- 
ithium. Monographs of l\Aarine Mollusca, 1 : 1 30 
pp. 

HOUSTON, R. S., 1976, The structure and function 
of neogastropod reproductive systems: with spe- 
cial reference to Columbella fuscata Sowerby, 
1832. The Veliger, 19: 27-46. 

HOXMARK, R. C, 1970, The chromosome dimor- 
phism of Nucella lapillus (Prosobranchia) in rela- 
tion to the wave exposure. Nytt Magasin for 
Zoologi, 18: 145-148. 

HOXMARK, R. C, 1971, Shell variation of Nucella 
lapillus in relation to environmental and genetic 
factors. Norwegian Journal of Zoology, 19: 145- 
148. 

HUANG, С L. & G. N. MIR, 1972, Pharmacological 
investigation of salivary gland of Thais haemas- 
toma (Clench). Toxicon, 10: 111-117. 

HUMPHREY, G., 1797, Specification of the vahous 
articles . . . consisting of . . . subjects in . . . 
Conchology, . . . Part 1, in С A. de Calonne, 
Museum Calonnianum. London, i-viii + 84 pp. 

HUTTON, F. W., 1884, Revision of the Recent 
Rhachiglossate Mollusca of New Zealand. Trans- 
actions and Proceedings of the New Zealand In- 
stitute, 16: 216-233. 

IREDALE, T., 1912, New generic names and new 
species of marine Mollusca. Proceedings of the 
Malacological Society of London, 1 0: 21 7-228. 

IREDALE, T., 1915, A commentary on Suter's 
"Manual of the New Zealand Mollusca." Trans- 
actions and Proceedings of the New Zealand In- 
stitute, 47:417-497. 

IREDALE, T., 1936, Australian molluscan notes. 
Number 2. Records of the Australian Museum, 
19:267-353. 

IREDALE, T., 1937, Mollusca. Pp. 232-261, in: The 
Middleton and Elizabeth Reefs, South Pacific 
Ocean. The Australian Zoologist, 8: 199-268. 



IREDALE, T. & D. F. MCMICHAEL, 1962, A refer- 
ence list of the marine Mollusca of New South 
Wales. The Australian Museum, Sydney, Mem- 
oir, II: 109 pp. 

JABLONSKl, D., 1982, Evolutionary rates and 
modes in Late Cretaceous gastropods: role of 
larval ecology. Proceedings of the Third North 
American Paleontological Convention, 1: 
257-262. 

JOUSSEAUME, F., 1880, Révision méthodique de 
la famille des purpurides. Le Naturaliste, 2: 335- 
336. 

JOUSSEAUME, F., 1888, Descriptions des mol- 
lusques recueillis par M. le Dr. Faurot dans la 
Mer Rouge et du Golfe d'Aden. Mémoires de la 
Société Zoologique de France, 1 : 12-223. 

JUNG, P., 1969, Miocene and Pliocene mollusks 
from Trinidad. Bulletins of American Paleontol- 
ogy 55: 289-657. 

KAY, E. A., 1971, The littoral molluscs of Fanning 
Island. Pacific Science, 25: 260-281. 

KAY, E. A., 1979, Hawaiian marine shells. Bishop 
Museum Press, Honolulu, 652 pp. 

KEEN, A. M., 1964, Purpura, Ocenebra, and, Mu- 
ricanthus (Gastropoda): request for clarification 
of status. Z.N.(S.) 1621. Bulletin of Zoological 
Nomenclature, 21 : 235-239. 

KEEN, A. M., 1971a, A review of the Muricacea. 
Western Society of Malacologists, Echo 4: 35- 
36. 

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

KENSLEY, В., 1985, The fossil occurrence in 
southern Africa of the South American intertidal 
mollusc Concholepas concholepas. Annals of the 
South African Museum, 97: 1-7. 

KENSLEY, B. & J. PETHER, 1986, Late Tertiary 
and early Quaternary fossil Mollusca of the Hon- 
deklip area. Cape Province, South Africa. Annals 
of the South African Museum, 97: 141-225. 

KIENER, L. C, 1835, Species général et iconogra- 
phie des coquilles vivantes, . . . voyageurs. 
Genre pourpre. Baillière, Paris, 151 pp. 

KILBURN, R. & E. RIPPEY, 1982, Sea shells of 
southern Africa. South China Printing Co., Hong 
Kong, 249 pp. 

KINCAID, T., 1957, Local races and dines in the 
marine gastropod Thais lamellosa Gmelin. A 
population study. Calliostoma, Seattle, 75 pp. 

KITCHING, J. A., L MUNTZ & F. J. EBLING, 1966, 
The ecology of Lough Ine XV: The ecological sig- 
nificance of shell and body forms in Nucella. 
Journal of Animal Ecology, 35: 1 13-126. 

KOOL, S. P., 1986, Radular convergence due to 
diet: an overestimated phenomenon? American 
Malacological Bulletin, 4: 233 (abstract). 

KOOL, S. P., 1987, Significance of radular charac- 
ters for reconstruction of thaidid phylogeny (Neo- 
gastropoda: Muricacea). The Nautilus, 101: 117- 
131. 

KOOL, S. P., 1988a, Functional morphology of the 
reproductive system of Plicopurpura patula 
(Linné, 1758) (Thaidinae: Neogastropoda): phy- 



PHYLOGENY OF RAPANINAE 



255 



logenetic implications for thaidid gastropods. 
American Zoologist, 27: 60A (abstract). 

KOOL, S. P., 1988b, Aspects of the anatorлy of 
Plicopurpura patula (Prosobranchia: Muricoidea: 
Tliaidinae), new combination, with emphasis on 
the reproductive system. Malacologia, 29: 373- 
382. 

KOOL, S. P., 1989, Piiylogenetic analysis of the 
subfamily Ttiaidinae. Ph.D. dissertation, The 
George Washington University, Washington, 
D.C., U.S.A., xiii + 342 pp. 

KOOL, S. P., 1993, The systematic position of 
the genus Nucella (Prosobranchia: Muricidae: 
Ocenebrinae). The Nautilus, 107: 43-57. 

KOOL, S. P. & K. J. BOSS, 1992, Nucella Röding, 
1798 (Gastropoda: Muricidae): type species. The 
Nautilus, 106:21-23. 

KOROBKOV, LA., 1955, Spravochnik i method- 
icheskoe rukovodstvo po Tretichnym molli- 
uskam. Leningrad, 795 pp. 

KOZLOFF, E. N., 1987, Marine invertebrates of the 
Pacific Northwest. University of Washington, Se- 
attle & London, X + 51 1 pp. 

KRAUSS, F., 1848, Südafrikanischen Mollusken. 
Stuttgart, [2] + 140 pp., 6 pis. 

KURODA, T., 1930, New Japanese shells. (2). Ve- 
nus, 2: 1-2. 

KURODA, T., & T. HABE, 1971, In: T. Kuroda, T. 
Habe & К. Oyama, eds. The sea shells of Sagami 
Bay. Maruzen Company, Ltd., Tokyo, 489 pp. 
(English part) + Index. 

LAMARCK, J. B. P. A. DE M. DE, 1799, Prodrome 
d'une nouvelle classification des coquilles. Mé- 
moires Société Histoire Naturelle, Paris, 1 : 63- 
91. 

LAMARCK, J. B. P. A. DE M. DE, 1801, Système 
des animaux sans vertèbres, . . . Paris, viii + 432 
pp. 

LAMARCK, J. B. P. A. DE M. DE, 1816, Tableau 
encyclopédique et méthodique des trois règnes 
de la nature, Pt. 23. mollusques et polypes 
divers. Paris, 16 pp., pis. 391-488. 

LAMARCK, J. B. P. A. DE M. DE, 1822, Histoire 
naturelle des animaux sans vertèbres, ... Vol. 6, 
part 2. Paris, 71 1 pp. 

LAMY, E., 1928, La ponte chez les gastéropodes 
prosobranches. Journal de Conchyliologie, 72: 
25-52. 80-126, 161-196. 

LARGEN, M. J., 1967, The diet of the dog-whelk, 
Nucella lapillus (Gastropoda Prososbranchia). 
Journal of Zoology, 151:1 23-1 27. 

LARGEN, M. J., 1971, Genetic and environmental 
influences upon the expression of shell sculptur- 
ing in the dog-whelk {Nucella lapillus). Proceed- 
ings of the Malacological Society of London, 39: 
383-388. 

LEACH, W. E., 1852, A synopsis of the Mollusca of 
Great Britain, John van Voorst, London, 376 pp., 
13 pis. 

LEBOUR, M. v., 1936, Notes on eggs and larvae of 
some Plymouth prosobranchs. Journal of the 
Marine Biological Association of the United King- 
dom, 20: 547-565. 



LEBOUR, M. v., 1 945, The eggs and larvae of some 
prosobranchs from Bermuda. Proceedings of the 
Zoological Society of London, (4)144: 462-489. 

LEHTINEN, P. T., 1985, Thaididae Jousseaume, 
1888 (Mollusca, Gastropoda) and Thaididae 
Lehtinen, 1967 (Arachnidae, Araneae): propos- 
als to remove the homonymy. Z.N.(S.) 2307. Bul- 
letin of Zoological Nomenclature, 42: 389-390. 

LEWIS, J. В., 1960, The fauna of rocky shores of 
Barbados, West Indies. Canadian Journal of Zo- 
ology 38: 391-435. 

LINK, D. H. F., 1807, Beschreibung der Naturalien 
Sammlung der Universität zu Rostock, Part 3: 
101-165. 

LINNAEUS, С, 1758, Systema naturae, 10th ed. 
Stockholm, 824 pp. 

LINNAEUS, С, 1767, Vermes Testacea. In: Sys- 
tema naturae, 12th ed., 1: 1106-1269. 

LINNAEUS, C, 1771, Mantissa plantarum . . . 
— Regni animalis . . . — Appendix. Stockholm, 
i-iv + pp. 143-510. 

LIPSCOMB, D. L., 1984, Methods of systematic 
analysis: the relative superiority of phylogenetic 
systematics. Origins of Life, 13: 235-248. 

LIU, L L., D. W. FOLTZ & W. B. STICKLE, 1991, 
Genetic population structure of the southern oys- 
ter drill Stramonita ( = Thais) haemastoma. Ma- 
rine Biology, 1 1 1 : 71-79. 

LOCAR D, A., 1886, Prodrome de malacologie 
Française. Catalogue général des mollusques 
vivants de France — Mollusques marins Paris, 
778 pp. 

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

MAES, V. O., 1967, The littoral marine mollusks of 
Cocos-Keeling Islands (Indian Ocean). Proceed- 
ings of the Academy of Natural Sciences of Phil- 
adelphia, 119: 93-217. 

MARTINI, F. H. W., 1777, Neues systematisches 
Conchyliencabinet. Vol. 3, Nürnberg, 434 pp, pis. 
66-121. 

MARTYN, T., 1784, The universal conchologist. 
Vols. 1 , 2. London. 

MENKE, С T., 1828, Synopsis methodica mollus- 
corum generum omnium et species earum, quae 
in Museo Menkeano adservantur . . . Pyrmont, xii 
+ 91 pp. 

MEUSCHEN, F. C, 1787, Museum Geversianum. 
P. & J. Holsteyn, Rotterodam, 659 pp. 

MILLER, A. C, 1970, Observations on the distribu- 
tion and feeding of Morula uva (Bolten) and 
Morula granulata (Duelos) (Gastropoda: Thai- 
sidae) in Hawaii. The Biology of Mollusks. Uni- 
versity of Hawaii, Hawaii Institute of Marine Biol- 
ogy, Technical Report, 18: 17. 

MÖRCH, O. A. L., 1852, Catalogus conchyliorum. 
Ludovici Kleini, Hafniae, 74 [ + 2] pp. 

MÖRCH, O. A. L., 1860, Beitrage zur Mollusken- 
fauna Central-Amerika's. Malakozoologische 
Blatter, 7:66-106. 

MONTFORT, P. D. DE, 1810, Conchyliologie sys- 
tématique, II. Paris, 676 pp. 



256 



KOOL 



MOORE, H. В., 1936, The biology of Purpura lapil- 
lus. I. Shell variation in relation to environment. 
Journal of the Marine Biological Association of 
the United Kingdom, 21 : 61-89. 

MOORE, H. В., 1938, The biology of Nucella lapil- 
lus. III. Life history and relation to environmental 
factors. Journal of the Marine Biological Associ- 
ation of the United Kingdom, 23: 67-74. 

MURDOCH, W. W., 1969, Switching in general 
predators: experiments on predator specificity 
and stability of prey populations. Ecological 
Monographs, 39: 335-354. 

NORDSIECK, F., 1968, Die europäischen Meeres- 
Gehäuseschnecken. Fischer, Stuttgart, 273 pp. 

NORDSIECK, F., 1982, Die europäischen Meeres- 
Gehäuseschnecken. Fischer, Stuttgart, 539 pp. 

OLD, W. E., 1964, Comments on Thais planospira. 
Annual Reports of The American Malacological 
Union, for 1964: 47-48. 

PAETEL, F, 1 875, Die bisher veröffentlichten Fam- 
ilien- und Gattungsnamen der Mollusken. Ge- 
bruder Paetel, Berlin, 229 pp. 

PALLAS, P. S., 1774, Spicilegia zooligica. Lange, 
Berolini. Vol. 1, Part 10, 41 pp., 4 pis. 

PALMER, A. R., 1979, Fish prédation and the 
evolution of gastropod shell form: experimental 
and geographic evidence. Evolution, 33: 697- 
713. 

PALMER, A. R., 1984, Species cohesiveness and 
genetic control of shell color and form in Thais 
emarginata (Prosobranchia, Muricacea): Prelim- 
inary results. Malacologia, 25: 477-491 . 

PALMER, A. R., 1985, Genetic basis of shell vari- 
ation in Thais emarginata (Prosobranchia, Muri- 
cacea). I. Banding in populations from Vancouver 
Island. Biological Bulletin, 169: 638-651. 

PATTERSON, С 1982. Morphological characters 
and homology. Pp. 21-74; in: K. A. Joysey & A. 
E. Friday, eds., Problems of phylogenetic recon- 
struction. Academic Press, New York. 

PCHELINTSEV, V. F. & I. A. KOROBKOV. 1960. 
Molliuski-Briukhonogie. In: Y. A. Orlov, ed., Os- 
novy paleóntologa. Moskva, 359 pp. 

PEASE, W. H., 1868, Descriptions of marine Gas- 
teropoda inhabiting Polynesia. American Journal 
of Conchology, 4: 71-80, 91-102. 

PERRIER, E-, 1897, Mollusques. Traité de Zoolo- 
gie 2: 1929-2140. 

PERRY, G., 1811, Conchology, or the natural his- 
tory of shells. London, 4 pp., 61 pis., index [1 p., 
unnumbered]. 

PETITJEAN, M., 1965, Structures microscopiques, 
nature minéralogique et composition chimique de 
la coquille des muricidés (gastéropodes proso- 
branches). Importance systématique de ces ca- 
ractères. Ph.D. dissertation. University of Paris, 
Paris, France, 131 pp. 

PETUCH, E. J., 1982, Geographical heterochrony: 
contemporaneous coexistence of Neogene and 
Recent molluscan faunas in the Americas. Palae- 
ogeography, Palaeoclimatology, Palaeoecology, 
37: 277-312. 

PETUCH, E. J., 1988, New species of Ecphora and 



ecphorine thaidids from the Miocene of Chesa- 
peake Bay, Maryland, U.S.A. Bulletin of Paleo- 
malacology, 1: 1-16. 

PHILIPPI, R. A., 1849, Centuria quarta Testaceo- 
rum novorum. Zeitschrift für Malakozoologie, 6: 
27-32. 

PHILLIPS, B. F., 1969, The population ecology of 
the whelk Dicathais aegrota in western Australia. 
Australian Journal of Marine and Freshwater Re- 
search, 20: 225-265. 

PHILLIPS, B. F., N. A. CAMPBELL & B. R. WIL- 
SON, 1973, A multivariate study of geographic 
variation in the whelk Dicathais. Journal of Ex- 
perimental Mahne Biology and Ecology, 1 1 : 27- 
69. 

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

PONDER, W. F. & A. WAREN, 1988, Classification 
of the Caenogastropoda and Heterostropha — a 
list of the family-group names and higher taxa; 
Appendix. In: W. F. Ponder, D. J. Eernisse & J. 
H. Waterhouse, eds., Prosobranch phylogeny 
Malacological Review, Supplement, 4: 288-326. 

POWELL, A. W. В., 1961, Shells of New Zealand. 
Whitcombe & Tombs Ltd., Auckland, 203 pp. 

POWELL, A. W. В., 1 964, The family Turridae in the 
Indo-Pacific. Indo-Pacific Mollusca, 1 : 227-346. 

POWELL, A. W. В., 1979, Shells of New Zealand. 
Collins, London, 500 pp. 

POWYS, W. L. & G. B. SOWERBY, 1835, On new 
species of Pandora, Buccinum, Nassa, and Pur- 
pura. Proceedings of the Zoological Society of 
London, 1835: 93-96. 

QUOY, J. R. С & J. P. GAIMARD, 1833, Voyage de 
découvertes de L Astrolabe, Part 2, 686 pp. 

RADWIN, G. E. & A. D'ATTILIO, 1971, Muricacean 
supraspecific taxonomy based on the shell and 
the radula. Western Society of Malacologists, 
Echo, 4: 55-67. 

RADWIN, G. E. & A. D'ATTILIO, 1972, The sys- 
tematics of some new world muricid species 
(Mollusca, Gastropoda), with descriptions of two 
new genera and two new species. Proceedings 
of the Biological Society of Washington, 85: 323- 
352. 

RADWIN, G. E. & A. D'ATTILIO, 1976, Murex 
shells of the world. Stanford University Press, 
Stanford, 284 pp. 

RAFINESQUE, С S., 1815, Analyses de la nature 
ou tableau du univers et des corps organises. 
Barravecchia, Palermo, p. 5-6, 136-149, 218- 
223. 

RAJALAKSHMI BHANU, R. C, K. SHYAMA- 
SUNDARI & K. HANUMANTHA RAO, 1980, His- 
tochemistry of the mucous cells in the gut wall of 
the snail Thais bufo (Gastropoda: Prosobran- 
chia). Proceedings of the National Academy of 
Sciences of India, (B) 50: 38-42. 

RAJALAKSHMI BHANU, R. C, K. SHYAMA- 
SUNDARI & K. HANUMANTHA RAO, 1981a, 
Histological and histochemical studies on the sal- 
ivary glands of Thais bufo (Lamarck). Monitore 
Zoológico Italiano, 15: 239-247. 



PHYLOGENY OF RAPANINAE 



257 



RAJALAKSHMI BHANU, R. C, K. SHYAMA- 
SUNDARI & K. HANUMANTHA RAO, 1981b, 
Studies on the alimentary canal of Thais bufo 
(Lamarck): histology and histochemistry of the 
foregut and midgut glands. Acta Histochemica et 
Cytochemica, 14: 516-523. 

REEVE, L. A., 1846, Conchologia Iconica, Volume 
3, Purpura, Ricin ula, Monoceros, and Conchole- 
pas; Buccinum. 

RENDER, H. A., 1962, The status of Nucella. The 
Nautilus, 75: 109-111. 

RENDER, H. A., 1980, The marine mollusks of 
Easter Island (Isla de Pascua) and Sala y Go- 
mez. Smithsonian Contributions to Zoology, 289: 
167 pp. 

RENDER, N. A. & N. S. LADD, 1973, Deep and 
shallow-water mollusks from the Central Pacific. 
Science Reports of the Tohoku University, Sen- 
dai, Japan, (2-Geology), Special Vol. (6): 37-49. 

RIGNI, G., 1964, Sobre о estómago de Thais hae- 
mastoma. Anais Academia Brasileira de Cien- 
cias, 36: 189-191. 

RÍOS, E. С, 1970, Coastal Brazilian seashells. Mu- 
seu Oceanógrafico de Rio Grande, Rio Grande, 
Brazil, 255 pp., 4 maps, 60 pis. 

RÖDING, P. F., 1798, Museum Boltenianum. Part 
1. Hamburg, 156 pp.; Part 2. Hamburg, i-viii + 
199 pp. 

ROSEWATER, J., 1975, An annotated list of the 
marine mollusks of Ascension Island, South At- 
lantic Ocean. Smithsonian Contributions to Zool- 
ogy ^ 89: 41 pp. 

ROVERETO, G., 1899, Prime ricerche sinonimiche 
sui generi dei gasteropodi. Atti delta Societa Li- 
gustica di Scienze Natur all e Geografiche, 10: 
101-110. 

SABELLI, B. & S. TOfVlMASINI, 1979, Osservazioni 
sulla radula di aicuni Muricacea delle Galapagos. 
Bollettino Malacologico, 15: 19-28. 

SALVAT, B. & С RIVES, 1975, Coquillages de 
Polynésie. Les Editions du Pacifique, Papeete, 
Tahiti, 393 pp. 

SAY, T., 1822, An account of some of the Marine 
Shells of the United States. Journal of the Natural 
Academy of Natural Sciences of Philadelphia, 2: 
221-248. 

SAY, T., 1824, An account of some of the fossil 
shells of the Academy of Natural Sciences of 
Philadelphia. Proceedings of the Academy of 
Natural Sciences of Philadelphia, 4: 124-155. 

SCNAUFUSS, L. W., 1869, l^olluscorum systema 
et catalogus. System und Aufzählung sämmtli- 
cher Conchylien der Sammlung von Fr. Paetel. 
Dresden, [4] + XIV + 1-119 pp. 

SCHUMACHER, С F., 1817, Essay d'un nouveau 
système des habitations des vers testacés. 
Copenhagen, 287 pp. 

SETTEPASSI, F., 1971, Af/an/e malacologico mol- 
luschi marini viventi nel i\Aediterraneo, Vol. 2. 
Roma, unpaginated. 

SNUTO, T., 1974, Larval ecology of prosobranch 
gastropods and its bearing on biogeography and 
paleontology. Lethaia, 17: 239-256. 



SHYAMASUNDARI, K, R. С RAJALAKSHMI 
BHANU & K. HANUMANTHA RAO, 1985, Obser- 
vations on the histology of the alimentary tract of 
Thais bufo (Lamarck) (Neogastropoda: Muri- 
cidae). Folia t^orphologica, 33: 116-124. 

SIGNOR, P. W., 1982, Influence of shell shape on 
burrowing rates in infaunal turritelliform snails. 
Proceedings of the Third North American Pale- 
ontological Convention, 2: 483-487. 

SMITH, E. A., 1913, Note on Murex mancinella, 
Linn. Proceedings of the Malacological Society of 
London, 10: 287-289. 

SMITH, E. H., 1967, The neogastropod stomach, 
with notes on the digestive diverticula and intes- 
tine. Transactions of the Royal Society of Edin- 
burgh, 67: 23-42. 

SOWERBY, G. В., 1835, New species of shells col- 
lected by Mr. Cuming. Proceedings of the Zoo- 
logical Society of London, 2: 4-7, 21-23, 49-51, 
84-85, 109-110. 

SOWERBY, G. В., 1839, A conchological manual. 
Odell, London, i-v + [2] + 130 pp., 24 pis. 

SOWERBY, G. В., 1841, Conchological illustra- 
tions. Part 11. Murex. A catalogue of Recent spe- 
cies. 9 pp. + Index, pp. 1-22. 

SPIGHT, T. M., 1972, Patterns of change in adja- 
cent populations of an intertidal snail, Thais 
lamellosa. Ph.D. dissertation. University of 
Washington, published by University Microfilms, 
Ann Arbor, Michigan, 308 pp. 

SPIGHT, T. M., 1973, Ontogeny, environment and 
shape of a marine snail. Journal of Experimental 
Marine Biology and Ecology, 13: 215-228. 

SPIGHT, T. M., 1976, Colors and patterns of an 
intertidal snail, Thais lamellosa. Researches on 
Population Ecology, 17: 176-190. 

SPIGHT, T. M., 1979, Environment and life history: 
the case of two marine snails. The Belle W. 
Baruch Library in Marine Science, 9: 135-143. 

SPIGHT, T. M., 1982, Population sizes of two ma- 
rine snails with a changing food supply. Journal 
of Experimental Marine Biology and Ecology, 57: 
195-217. 

SRILAKSHMI, G., 1991, Histological and his- 
tochemical studies on the female reproductive 
system of Morula granulata (Duelos) (Prosobran- 
chia: Neogastropoda). Zoologische Anzeiger, 
226: 71-87. 

STEPHENSON, L. W., 1923, The Cretaceous For- 
mations of North Carolina. Part I. Invertebrate fos- 
sils of the Upper Cretaceous Formations. North 
Carolina Geological Survey, 5: 1-604, 102 pis. 

STEPHENSON, L. W., 1941, The larger inverte- 
brate fossils of the Navarro Group of Texas. Uni- 
versity of Texas Publication, 4101: 641 pp., 95 
pis. 

STEWART, R. В., 1927, Gabb's California fossil 
type gastropods. Proceedings of the Academy of 
Natural Sciences, Philadelphia, 78: 287-447, 
pis. 20-32. 

STIMPSON, W., 1865, On certain genera and fam- 
ilies of zoophagous gasteropods. American Jour- 
nal of Conchology, 1 : 55-64. 



258 



KOOL 



STRAUSZ, L., 1966, Die Miozän-Mediterranean 
Gastropoden Ungarns. Akademiai Kiado, Buda- 
pest, 692 pp., 221 figs., 79 pis. 

SUTER, H., 1909, The Mollusca of the Subantarctic 
Islands of New Zealand. In: С Chilton, ed., Sub- 
antarctic islands of New Zealand, Vol. 1 . Govern- 
ment Printer, Wellington, New Zealand, xxxv + 
848 pp., 18 pis. 

SUTER, H., 1913, Manual of the New Zealand Mol- 
lusca. Wellington, 1120 pp. 

SWAINSON, W. A., 1835, The elements of modern 
conchology; briefly and plainly stated. Baldwin & 
Cradock, London, vii + 61 pp. 

SWAINSON, W. A., 1840, Treatise on malacology 
or shells and shell-fish. London, 419 pp. 

ТАК!, I., 1950, Morphological observations on the 
gastropod operculum. Venus, 16: 32-48. 

TANAKA, Y., 1958, Radula of Rapana thomasiana. 
Venus, 20: 128-130. 

TAYLOR, D. W. & N. F. SOHL, 1962, An outline of 
gastropod classification. Malacologia, 1 : 7-32. 

TAYLOR, J. D., 1971, Reef associated molluscan 
assemblages in the Western Indian Ocean. Zoo- 
logical Society of London, Symposium, 28: 501- 
534. 

TAYLOR, J. D., 1976, Habitats, abundance and diet 
of muricacean gastropods at Aldabra Atoll. Zoo- 
logical Journal of the Linnean Society, 59: 155- 
193. 

TAYLOR, J. D., 1983, The food of coral-reef Drupa 
(Gastropoda). Zoological Journal of the Linnean 
Society, 78:299-316. 

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

THIELE, J., 1925, Gastropoda der Deutschen Tief- 
see-Expedition. Berlin, 382 pp. 

THIELE, J., 1929, Handbuch der systematischen 
Weichtierkunde. Part 1 . Fischer, Jena, 376 pp. 

THOMAS, F. I. M. & A. J. KOHN, 1985, Trophic 
roles of the tropical limpet-like predatory gastro- 
pod, Drupa. American Zoologist, 25: 88. 

THORNLEY, G., 1952, A new Tha/s found on a log 
at Port Stephens. Proceedings of the Royal Zoo- 
logical Society of New South Wales, 1951-1952: 
44-45. 

TIRMIZI, N. M. & I. ZEHRA, 1983, Study of the 
eggs of six common proobranchs [sic] of the Pa- 
kistani coast. Pakistan Journal of Zoology, 15: 
39-43. 

TOMLIN, J. R., 1928, Reports on the marine Mol- 
lusca in the collection of the South African Mu- 
seum. Annals of the South African Museum, 25: 
313-335. 

TROSCHEL, F. H., 1866-1893, Das Gebiss der 
Schnecken zur Begründung einer natürlichen 
Classification, Vol. 2. Berlin, 409 pp. 

VERMEIJ, G. J., 1975, Marine faunal dominance 
and molluscan shell form. Evolution, 28: 656- 
664. 



VERMEIJ, G. J., 1979, The architectural geography 
of some gastropods. Pp. 428-433 in: J. Gray & 
A. J. BoucoT, eds.. Historical biogeography, plate 
tectonics, and the changing environment. Ore- 
gon State University Press, Oregon. 

VERMEIJ, G. J. 1982. Phenotypic evolution in a 
poorly dispersing snail after arrival of a predator. 
Nature, 229: 349-350. 

VERMEIJ, G. J. & J. D. CURREY, 1980, Geograph- 
ical variation in the strength of thaidid snail shells. 
Biological Bulletin, 158: 383-389. 

VERMEIJ, G. J. & E. ZIPSER, 1986, Burrowing per- 
formance of some tropical Pacific gastropods. 
The Veliger, 29: 200-206. 

VOKES, E. H., 1970, On Cernohorsky's designa- 
tion of a lectotype for Murex mancinella Lin- 
naeus. The Veliger, 12: 368-370. 

VOKES, E. H., 1972, Notes on the fauna of the 
Chipóla Formation VII. On the occurrence of the 
genus Concholepas (Gastropoda: Thaididae), 
with the description of a new species. Tulane 
Studies in Geology and Paleontology, 1 0: 31-33. 

WARD, L. W. & N. L. GILINSKY, 1988, Ecphora 
(Gastropoda: Muricidae) from the Chesapeake 
Group of Maryland and Virginia. Notulae Natu- 
rae, 469: 1-21, 5 pis. 

WELLINGTON, G. M. & A. M. KURIS, 1983, 
Growth and shell variation in the tropical eastern 
Pacific intertidal gastropod genus Purpura: eco- 
logical and evolutionary implications. Biological 
Bulletin, 164: 518-535. 

WENZ, W., 1941, Prosobranchia. Pp. 961-1200 in: 
O. H. ScHiNDEwoLF, ed., Handbuch der Paläozo- 
ologie. Vol. 6, part 5 Gebrüder Borntraeger, Ber- 
lin. 

WILBUR, К. M. & G. OWEN, 1964, Growth. Pp. 
211-242, in: K. M. Wilbur & С M. Yonge, eds.. 
Physiology of Mollusca, Vol. 1, Academic Press. 

WINCKWORTH, R., 1945, The types of the Bolte- 
nian genera. Proceedings of the Malacological 
Society of London, 26: 136-148. 

WOOD, W., 1828, Index testaceologicus; or a cat- 
alogue of shells, British and foreign. London, viii 
+ 188 pp., 8 pis., index, errata. 

WOODRING, W. P., 1959, Geology and paleontol- 
ogy of Canal Zone and adjoining parts of Pan- 
ama. Description of Tertiary mollusks (Gastro- 
pods: Vermetidae to Thaididae). United States 
Geological Survey Professional Paper, 306-B: 
193-202. 

WU, S. K., 1965a, Comparative functional studies 
of the digestive system of the muricid gastropods 
Drupa ricina and Morula granulata. Malacologia, 
3:211-233. 

WU, S. K., 1965b, Studies of the radulae of Taiwan 
muricid gastropods. Bulletin of the Institute of Zo- 
ology, Academia Sínica, 4: 95-106. 

WU, S. К., 1967, Studies of the radulae of Taiwan 
muricid gastropods. Annual Reports of the Amer- 
ican Malacological Union, for 1967: 46 (abstract). 

WU, S. K., 1968, On some radulae of the muricid 
gastropods. Venus, 27: 89-94. 



PHYLOGENY OF RAPANINAE 259 

WU, S. K., 1973, Comparative studies on the di- tropoda: Muricacea) in West America. Special 

gestive and reproductive systems of some muri- Publications of the Mukaishima Marine Biological 

cid gastropods. Bulletin of the American Malaco- Station, Special Contribution, 236: 45-66. 
logical Union, for 1972, p. 18 (abstract). 

WU, S. K., 1985, The genus Acanthina (Gas- Revised Ms. accepted 4 January 1 993 



MALACOLOGIA, 1993, 35(2): 261-313 

PHYLOGENETIC RELATIONSHIPS AND GENERIC REVIEW OF THE BITTIINAE 
(PROSOBRANCHIA: CERITHIOIDEA) 

Richard S. Houbrick 

Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian 
Institution, Washington, D.C. 20560, U.S.A. 

ABSTRACT 

The anatomy of seven members of the Bittium group is described, clarifying the status of the 
genus-level taxa comprising it. Bittium reticuiatum, the type species of Bittium Gray, is described 
in depth, thereby establishing criteria for comparisons with other taixa of Bittiinae. The type 
species of Stylidium Dall and Lirobittium Bartsch, and representatives of Bittiolum Cossmann 
and Cacozeliana Strand are examined and compared with Bittium, s.S. Results of anatomical 
studies and a phylogenetic analysis using the Hennig86 and CLADOS programs, with Cerithium 
as an outgroup, establish monophyly for Bittiinae Cossmann and reveal six different genus-level 
taxa. A new genus, Ittibittium, from the Indo-Pacific, is proposed. Synonymies of each genus- 
level taxon and representative species examined are presented. Brief accounts of the ecology 
and zoogeography of each taxon are given. Two taxa formerly attributed to the Bittium-group are 
herein excluded from it and referred to Cerittiium Bruguière. These are Cerithium zebrum 
Kiener, 1 841 , and Cerithium boeticum Pease, 1 861 . The subfamily Bittiinae Cossmann, 1 906, is 
thought to comprise nine genera (four of which were not included in phylogenetic analyses) : 
Bittium Gray, 1847; Bittiolum Cossmann, 1906; Ittibittium gen. п., Stylidium Dall, 1907; Lirobit- 
tium Bartsch, 191 1 ; Cacozeliana Strand, 1928; Argyropeza Melvill & Standen, 1901 ; Varicopeza 
Gründe!, 1976; Zebittium Finlay, 1927. The genus Cassiella Gofas, 1987, of uncertain place- 
ment, is included as a possible member of the group. 

Key words: Bittiinae, Bittium, Cerithioidea, anatomy, taxonomy, phylogenetic analysis. 



INTRODUCTION 

Shells of most small-sized cerithiids are no- 
tably difficult to classify, even to familial and 
generic levels. There has been much confu- 
sion and disagreement among malacologists 
as to the limits and subdivisions of genus- 
level taxa, because most genera have been 
defined or based upon convergent shell fea- 
tures alone. Reflective of this unstable taxon- 
omy, unreliable curatorial systems exist in 
most museums, where many lots of small- 
sized cerithiid taxa are randomly intermixed 
with each other and with immature specimens 
of larger-shelled genera, such as Cerithium. 
These mixed lots frequently are assigned to 
the convenient "trash basket" category Bit- 
tium. 

The genus Bittium Gray, 1847, sensu lato, 
comprises many poorly understood species 
placed in the family Cerithiidae Bruguière, 
1789. The concept of Bittium has been gen- 
erally broad, encompassing many other di- 
verse genera, and opinions on the relation- 
ships of the genus with other small-shelled 
cerithiid groups have also been varied. For 
these reasons and due to the lack of good 



anatomical characters, most of the small- 
sized cerithioideans were left out of my anal- 
ysis of cerithioidean phylogeny (Houbrick, 
1988). 

The most recent revision of the Bittium 
group was published by Gründel (1976), who 
based his taxonomy and phylogeny of the 
group on sculptural characters of the proto- 
conch (embryonic spiral formation), ontoge- 
netic sculptural development of the teleo- 
conch, and overall shell form. Gründel (1976) 
included many fossil and extinct taxa in his 
revision, but did not consider radular, opercu- 
lar, and anatomical characters of Recent 
taxa. Although he noted the similarities of Bit- 
tium and Cerithium Bruguière, 1 789, he indi- 
cated that Cerithium differs considerably from 
Bittium in shell form, sculpture, aperture, and 
especially in ontogenetic sculptural develop- 
ment. On the basis of the ontogeny of early 
spiral shell sculpture, Gründel (1976: 38) be- 
lieved that genera in the Bittium group {Bit- 
tium, Lirobittium, Bittiolum, Semibittium) were 
descendents of the Jurassic genus Procerith- 
ium Cossmann, 1902, of the family Procerithi- 
idae Cossmann, 1906. Indeed, he remarked 
that Bittium and Procerithium shared greater 



261 



262 



HOUBRICK 



TABLE 1. Bittium-group genera and species used for anatomical studies (asterisk indicates 
type species). 



Genus 



Species 



Geographic Region 



Bittium * reticulatum (DaCosXa, 1778) 

Bittium impendens (Hedley, 1 899) 

Bittiolum var/um (Pfeiffer, 1840) 

Bittiolum alternatum (Say, 1 822) 

Ittibittium parcu/77 (Gould, 1861) 

Lirobittium subplanatum Bartsch, 1911 

Lirobittium attenuatum (Carpenter, 1 864) 

Stylidium "eschrichtii (Middendorf , 1 849) 

Cacozeliana * granaría (K\ener, 1842) 



Sao Miguel, Azores 
Honolulu, Hawaii 
Ft. Pierce, Florida 
Provincetown, Massachusetts 
Honolulu, Hawaii 
Palos Verdes, California 
Catalina Id., California 
Carmel, California 
Albany, Western Australia 



similarities In ontogenetic sculptural develop- 
ment and overall shell morphology than did 
Bittium and Cerithium. Gründel (1976: 40) 
noted that the genera Argyropeza Melvill & 
Standen, 1901, and Varicopeza Gründel, 
1 976, usually placed near Bittium, were strik- 
ingly similar in their ontogenetic sculptural de- 
velopment and morphologies to species of 
the Jurassic genus Cryptaulax Tate, 1869 
(Procerithiidae), and stated that he consid- 
ered Argyropeza and Varicopeza to be 
Recent members of Procerithiidae. Under 
Procerithiidae, he assigned the Argyropeza- 
Cryptaulax group to the subfamily Cryptaul- 
axinae Gründel, 1976, which he believed 
showed many of the "ancient characteristics" 
of the family, and the Bittium-Procerithium 
group to the subfamily Procerithiinae Coss- 
mann, 1902. Gründel (1976) considered both 
subfamilies to have developed independently 
of one another and to have been separate 
since the Dogger (Middle Jurassic). 

Houbrick (1 977) discussed the status of Bit- 
tium Gray, 1 847, and included a historical re- 
view, extensive synonymy, and a concholog- 
ical redescription of the genus. This paper 
noted that most of the supraspecific taxa as- 
sociated with the Bittium group are parochial 
In conception and scope, based on specific 
rather than generic characters, and convey 
little or misleading phylogenetic information 
about the group. In the Interest of pragmatism 
and taxonomic parsimony, it was suggested 
that many of the generic and subgeneric 
names be abandoned or synonymized with 
Bittium, sensu lato, until the entire group was 
properly evaluated on the basis of more than 
shell characters. 

Since Gründers (1976) work and my paper 
on Bittium (Houbrick, 1977), studies on a 
number of Bittium-Wke genera and other 
small-shelled cerithioidean taxa have been 



published: Dahlal<ia (Houbrick, 1978), Argyro- 
peza (Houbrick, 1980a), Varicopeza (Hou- 
brick, 1980b, 1987a), Glyptozaria (Houbrick, 
1981a), Alaba and Litiopa (Kosuge, 1964; 
Houbrick, 1987b; Luque et al., 1988), Colina 
(Houbrick, 1990a), Plesiotrochus (Houbrick, 
1990b), and Diala (Ponder, 1991). Many of 
these papers include anatomical data that 
have helped partially to untangle the confus- 
ing mixture of cerithiid genera of similar small- 
shelled morphology. 

The relationships of small-shelled species 
of the family Obtortionidae Thiele, 1925, 
which are very similar to those of members of 
the Bittiinae, remain uncertain because ana- 
tomical characters are unknown. It is unclear 
if Obtortionidae constitutes a separate family 
or should be included under Bittiinae. 



MATERIALS AND METHODS 

The goals of this study are threefold: first, to 
examine the anatomy of Bittium reticulatum 
(DaCosta, 1778), the type species of the ge- 
nus, thus setting the limits of the genus with a 
description of distinctive anatomical charac- 
ters; second, to study the anatomy of a num- 
ber of other "Bittium" species, thereby estab- 
lishing the validity or arlificiality of other 
component groups or closely related higher 
taxa; and third, to make a phylogenetic anal- 
ysis of the group based on a morphological 
data set that includes more than shell char- 
acters. 

This revision is based primarily on collec- 
tions of preserved material in the USNM and 
on living material studied in the field. Fossils 
representing extinct genera and species were 
not considered, although a brief survey of ex- 
tinct forms and their possible relationships to 
living members of the Bittium-group is in- 



GENERIC REVIEW OF BITTIINAE 



263 



eluded. The great number of species and 
higher category groups traditionally included 
under Bittium, sensu lato, and the difficulties 
of obtaining good anatomical material pre- 
cluded an exhaustive, comprehensive ana- 
tomical study of all members the group. In- 
stead, I decided to look at representative taxa 
of genera assigned to the Bittium-group com- 
prising species having diverse shells from 
widely different geographic regions. A total of 
seven Bittium-group species representing five 
higher taxa (genera) from different localities 
were examined by dissecting live-collected 
material and by studying living populations in 
situ, where possible. These species are listed 
below in Table 1 and include the type species 
of Bittium Gray, 1847, Stylidium Dall, 1907, 
and Cacozeliana Strand, 1928, and represen- 
tative species of Bittiolum Cossmann, 1906, 
Lirobittium Bartsch, 1911, and a new genus, 
described herein. Two other species, each 
having a distinctive shell morphology, and 
considered as putative genera formerly attrib- 
uted to "Bittium," S.I., were also studied in the 
field: "Bittium" zebrum (Kiener, 1841) from 
Pago Bay, Guam, and Enewetak Atoll, Mar- 
shall Islands; and "Bittium" boeticum (Pease, 
1861), from Honolulu, Hawaii. When the soft 
parts of these two species were examined, 
they were found to lack an epipodial skirt, and 
the ciliated ridge tract and spermatophore 
bursa in the lateral lamina of the palliai ovi- 
duct, characters distinctive of members of the 
Bittium-group. Therefore, both species were 
excluded from the ß/Wtvm-group and as- 
signed to Cerithium Bruguière. Due to the cur- 
rent alpha-level taxonomic disarray of the Bit- 
tium-group, I have attempted to present a 
comprehensive, annotated synonymy and 
have illustrated the shells of the species stud- 
ied in this review. I hope that this will give 
other workers an unequivocal idea about the 
species and genera they represent. 

All specimens were dissected under water 
in wax-filled petri dishes using a Wild M-5 dis- 
secting microscope. Methylene blue was 
used to enhance anatomical features during 
dissection. Sections were made at 5 |xm and 
stained with Hematoxolin and Eosin. Photo- 
micrography was done using a Zeiss Photo- 
microscope III. 

The emphasis of this study is on the anat- 
omy of Bittium reticulatum, the type species of 
Bittium, s.S., which is the criterion against 
which other Bittium-group genera are de- 
scribed and compared in this paper. Descrip- 
tions of Bittiolum, Cacozeliana, Stylidium, Li- 



robittium, and a new genus described herein, 
are less detailed and emphasize the anatom- 
ical differences from Bittium reticulatum. 

The anatomy of the genera Argyropeza and 
Varicopeza is only superficially understood. 
Anatomical knowledge about Zebittium Fin- 
lay, 1927, and Cassiella remains unknown, 
because I was unable to obtain preserved 
material of species representing them; conse- 
quently, only the shells are considered in this 
review. 

Phylogenetic Analysis 

The guiding principles of this study are those 
of phylogenetic systematics (Hennig, 1966; 
Wiley, 1981). The Hennig86 computer pack- 
age, version 1 .5, ie and bb options (copyright 
James S. Farris, 1988) and CLADOS, version 
1.2 program (copynght Kevin С Nixon, 1988, 
1991, 1992) were used to analyse data and 
construct trees. 

Phylogenetic analysis of six genus-group 
taxa of the Bittiinae {Bittium, Ittibittium, Bitti- 
olum, Lirobittium, Stylidium, and Cacozeli- 
ana) was undertaken using 21 morphological 
characters compnsing 51 character states de- 
rived from the shell, operculum, radula, and 
soft anatomy of the taxa listed in Table 1 . Ini- 
tially, there were 30 characters, but these 
were reduced to 21 . Seven of the 21 charac- 
ters were multi-state characters. Autapomor- 
phies defining terminal branches, which were 
not part of multiState series, were not included 
in the analysis, but were retained for the di- 
agnosis of each genus-group taxon. Multi- 
state characters were unordered. 

Genus-Group Taxa Analysed 

Six genus-group taxa were included: Caco- 
zeliana, Lirobittium, Stylidium, Bittium, Ittibit- 
tium, and Bittiolum (Table 1). The phyloge- 
netic analysis excluded poorly known genera 
that have been assigned without justification 
to Bittiinae, such as Zebittium and Cassiella. 
Although the shell morphologies, opercular 
and radular characters of Argyropeza and 
Varicopeza have been well studied (Houbrick, 
1980a, 1980b), these genera also were left 
out of the analysis because of lack of anatom- 
ical data. 

Outgroup Selection 

The genus Cerithium Bruguière, family Cer- 
ithiidae Férussac, 1819, was selected as the 



264 



HOUBRICK 



TABLE 2. Comparison of dentition of radular teeth among genera (C = central or main cusp; numbers 
signify no. of denticles). 



Taxon 



Rachidian 



Lateral 



Inner Marginal 



Outer Marginal 



Bittium 

Bittiolum 

Ittibittium 

Lirobittium 

Stylidium 

Cacozeliana 

Argyropeza 

Varicopeza 



2-3+C+2-3 
3 + C + 3 
2 + C + 2 
6 + C + 6 
2 + C + 2 
2 + C + 2 
2-3+C+2-3 
3-4+C+3-4 



1+C+3-6 

2+C+3-4 

1+C+3-4 

6 + C + 15-17 

1+C+3-4 

1+C+3-4 

1+C+5-6 

1+C+5-6 



3-4+C+4 
3-4+C+2-3 
2 + C + 3 
15-19 + C + 5-6 
4-5+C+3 
5-6+C+3-4 
5-6+C+4-5 
3-4+C+3 



3-4+C+O 
6 + C + O 
5 + C + O 
15-19 + C + O 
4 + C + O 
4 + C + O 

5-6+C+O 
3 + C + O 



outgroup to root the trees generated by the 
analyses. The Bittium-group traditionally has 
been considered as a subfamily (Bittiinae) of 
Cerithiidae by various authors (see below, for 
history). Cerithium, subfamily Cerithiinae, is 
the most appropriate group to use for out- 
group comparison, because it is the closest 
sister group that is well known anatomically. 
The anatomy of Cerithium species has been 
described by Houbrick (1971, 1978, 1992) 
and is very similar to that of Bittiinae mem- 
bers, However, Cerithium species have more 
generalized and less complex external fea- 
tures. Several external anatomical features of 
members of the Bittium-group, such as a 
metapodial mucus gland, and the epipodial 
skirt and associated papillae, are lacking in 
Cerithium. The anatomy of such small-sized 
snails as Bittium may be highly derived and/or 
modified due to their reduction in size. Cerith- 
ium species are generally much larger ani- 
mals than "Bittium" species, but a number of 
species are very small and often are confused 
with "Bittium" species. 

Among small-shelled cerithioideans, Litiopa 
and Alaba, family Litiopidae, were considered 
as possible outgroup candidates. These small 
snails have external features, such as an 
epipodial skirt and epipodial tentacles, similar 
to those seen among members of the Bittii- 
nae, and are well known anatomically; how- 
ever, they differ from bittiid species in internal 
anatomy (Kosuge, 1964; Houbrick, 1987b; 
Luque et al., 1988). Phylogenetically, Litiop- 
idae is far removed from the family Cerithiidae 
(Houbrick, 1988: 114), and is therefore re- 
jected as a suitable outgroup. 

Another group of small-shelled species, the 
Dialidae, was also considered as a possible 
outgroup. However, only one species is 
known anatomically (Ponder, 1991), and 
Healy (1 986) has shown that the parasperma- 
tozoa of Diala are unique and highly derived 



among cerithioideans. Render's (1991) phy- 
logenetic analysis showed that dialids were 
closely related to litiopids and far removed 
from Cerithiidae (Ponder, 1991: 514). Diala 
was therefore rejected as an outgroup. 

Characters 

The characters listed below comprise three 
categories: shell characters (1-5), anatomical 
characters (6-19), reproductive characters 
(20-21). Radular characters were eliminated 
from the final analysis because of their au- 
tapomorphic condition. Nevertheless, radular 
characters are important diagnostic charac- 
ters of genera and are summarized in Table 2. 

Because the polarities of multistate charac- 
ters were largely speculative, all character 
states were left unordered; i.e., the integer 
assignment was arbitrary. The coding of 
these characters and their states are pre- 
sented in Table 3. An annotated list of the 
morphological characters and character 
states used in the phylogenetic analysis is 
presented below: 

Shell Characters: 1. Shell sculpture — = 
spiral; 1 = cancellate. Most members of the 
subfamily are characterized by a markedly 
cancellate shell sculpture, in contrast to Cer- 
ithium species where spiral elements domi- 
nate sculptural patterns (Houbrick, 1992). Ex- 
ceptions are species of the genera Stylidium 
and Ittibittium, where spiral sculpture domi- 
nates and axial ribs are either lacking or 
poorly developed. 

2. Anal canal — = well developed; 1 = 
weakly developed or missing. A well-devel- 
oped anal canal is present in Cerithium mem- 
bers (the outgroup), but occurs only in two 
genera of the Bittium-group, Cacozeliana and 
Varicopeza, and is exceptionally well devel- 
oped in the latter genus (Houbrick, 1980b). 



GENERIC REVIEW OF BITTIINAE 



265 



TABLE 3. Data matrix derived from morphological characters of species representing six genus-group 
taxa of Bittiinae. Cerithium is the outgroup. 



Character 


Taxon 


1 


2 


3 


4 


5 


6 


7 


8 


9 


10 


11 


12 


13 


14 


15 


16 


17 


18 


19 


20 


21 


Outgroup 

































































Bittium 


1 


1 





1 





1 


1 











2 





1 


2 


1 


1 


1 


1 


1 








Ittibittium 





1 





1 








1 








1 


2 


1 





1 


1 











1 


1 


1 


Stylidium 





1 


1 


1 


2 








1 


1 





1 





2 


1 


1 


1 


1 


1 


1 





1 


Cacozeliana 


1 





1 





2 


2 


2 











1 














1 


1 


1 











Bittiolum 


1 


1 





1 


1 





1 








1 


3 





1 


1 


1 





1 


1 


1 








Lirobittium 


1 


1 


1 


1 





1 





1 


1 





1 


1 


2 


1 


1 





1 


1 


1 


2 


1 



3. Varices — = present; 1 = absent. Va- 
rices, thickened, fornner growth lines, are a 
common feature of most cerithiids and occur 
among members of Bittiinae with the excep- 
tion of Lirobittium and Stylidium. 

4. Anterior canal — = well developed; 1 
= weakly developed. The anterior siphonal 
canal is a strong feature on most cerithiids, 
but in smaller-shelled taxa frequently is poorly 
developed (most Bittiinae) or absent {Cass- 
iella, Cerithidium). 

5. Protoconch sculpture — = two spiral 
lirae; 1 = one spiral lira; 2 = entirely smooth. 
Most outgroup species have strong spiral 
sculptural elements on their protoconchs 
(Houbrick, 1992). Bittiinae genera range from 
species with spiral sculpture to those having 
only one weak spiral lira or no sculpture, but 
this is probably reflective of the type of devel- 
opment. 

Anatomical Characters: 6. Opercular mor- 
phology — = ovate shape; 1 = round, cir- 
cular shape; 2 = round shape with fringed 
spiral edges. Cerithium species have oper- 
cula with an ovate shape (Houbrick, 1992), 
and it is thought herein that the more circular 
shape observed among several Bittium-group 
taxa are modifications due to size reduction, 
although this is not always the case (excep- 
tions in Ittibittium and Bittiolum, both small 
shelled genera). The spirally fringed condition 
seen in Cacozeliana departs from the norm 
and is probably derived. 

7. Snout shape — = wide; 1 = narrow, 
elongate; 2 = short, narrow. This character is 
a variable feature among cerithiids. Cerithium 
species usually have large, wide, muscular 
snouts (Houbrick, 1992), whereas they tend 
to be narrow and elongate in members of the 
Bittiinae, especially among taxa of the Bittium 
clade {Bittium, s.S., Ittibittium, Bittiolum). 



8. Cephalic tentacle length — = elongate; 
1 = short. Among cerithiids and the Bittiinae, 
cephalic tentacles are usually elongate and 
much longer than the snout, but in the eastern 
Pacific genera Lirobittium and Stylidium, the 
tentacles are much shorter than the length of 
the snout. 

9. Eye size — = normal; 1 = small; 2 = 
large. Most cerithiids have eyes of normal size, 
but in such deep-water species as Argyropeza 
and Varicopeza, the eyes are very large, pos- 
sibly an adaptation to water depth and poor 
light. In contrast, the eyes of Styliodium and 
Lirobittium species are exceptionally small. 

10. Metapodial mucus gland — = absent; 
1 = present. Although this structure is absent 
in the outgroup. It does occur among a few 
other cerithioidean groups (Litiopidae [Alaba, 
Litiopa], Cerithiidae [Colina]; Houbrick, 
1987b, 1990a, respectively). This gland may 
be an adaptation to an algal and/or high en- 
ergy habitats. Species having a metapodial 
gland are known to use the mucus thread se- 
creted by the gland to anchor themselves 
while they climb about the algal fronds 
(Houbrick, 1987b, 1990a). 

1 1 . Epipodial skirt — = rudimentary; 1 = 
well developed, smooth; 2 = well developed, 
papillate along edges; 3 = well developed, 
scalloped. Cerithium species have a weak 
operculigerous lobe on the rear of the foot, 
which is here interpreted as a rudimentary 
posterior epipodial skirt. In Bittiinae species, 
the skirt extends forward along the sides of 
the foot to form a fully developed epipodial 
skirt. An epipodial skirt occurs also among 
small-shelled members of the Litiopidae (Ko- 
suge, 1964; Houbrick, 1987b; Luque et al., 
1988) and the Dialidae (Ponder, 1991). Al- 
though this character is homoplastic among 
cerithioideans, an epipodial skirt is character- 
istic of Bittiinae. 



266 HOUBRICK 

TABLE 4. Comparison of developmental features among Bittiinae genera and species. 





Max. 


Shell 


Protoconch 


Developmental 




Taxon 


Length 


Sculpture 


Type 


Egg Size 


Bittium 












reticulatum 


15 


mm 


2 spirals 


planktonic 


0.1 mm 


Ittibittium 












parcum 


6 


mm 


2 spirals 


direct 


0.2 mm 


Bittiolum 












varium 


7 


mm 


1 spiral 


planktonic 


0.1 mm 


Lirobittium 












subplanatum 


10 


mm 


2 spirals 


direct 


0.5 mm 


Stylidium 












eschrichtii 


17.5 


mm 


smooth 


direct 


0.2 mm 


Cacozeliana 












granaría 


24 


mm 


smooth 


planktonic 


0.1 mm 


Argyropeza 












divina 


7.6 


mm 


2 spirals 


planktonic 


? 


Varícopeza 












varícopeza 


10 


mm 


1 spiral 


planktonic 


? 



12. Ovipositor — = present; 1 = absent. 
This gland, although common among cerlthio- 
ideans, Is absent In some taxa, such as those 
having Internal brooding (Houbrick, 1987c). 
The absence of an ovipositor in females may 
be falsely scored, as it Is thought that its pres- 
ence can be easily ascertained only during 
breeding season; moreover, this gland Is also 
difficult to detect in some preserved speci- 
mens. Among Bittiinae, the ovipositor is ab- 
sent only in Ittibittium and Lirobittium. 

13. Osphradial morphology — = bipecti- 
nate; 1 = monopectinate; 2 = vermiform. 
This character varies greatly among Bittiinae 
genera. Although the osphradium in Ceritti- 
ium species is bipectlnate, it is vermiform 
among most other cerlthloidean families, 
such as the estuarine Potamldidae and fresh- 
water families Thiarldae and Pachychilidae 
(Houbrick, 1988, 1991). 

14. Osphradial length — = equal to 
ctenidial length; 1 = a little less than ctenidial 
length; 2 = one-half the ctenidial length. This 
Is a highly variable character, but often diag- 
nostic of some taxa. No overlap among char- 
acter states was detected in the species stud- 
ied. 

15. Zygoneurous nervous system — = 
absent; 1 = present. Bouvier (1887) docu- 
mented a zygoneurous condition among 
some cerithiids, and this was summarized by 
Houbrick (1988). Zygoneury is absent in Cer- 
ithium, and In all Bittiinae except for Bittiolum. 

16. Common opening to sperm pouch and 
seminal receptacle openings — = close to- 
gether; 1 = far apart. In Stylidium and Liro- 



bittium, the openings have a wide separation, 
whereas in Bittium they are not as far apart. In 
other bittiids and in most other cerithiids, the 
openings are close together. 

17. Spermatophore bursa location — = 
located in medial lamina; 1 = located in lat- 
eral lamina. The spermatophore bursa Is 
found in the lateral lamina in most members 
of the Bittium-group, but in Ittibittium and in all 
other known cerithiids, it occurs in the medial 
lamina (Houbrick, 1988). 

18. Ciliated ridge tract — = absent; 1 = 
present. This structure, one of the synapo- 
morphies defining Bittiinae, is lacking in Ittibit- 
tium members and in most other cerithiids. 

19. Seminal receptacle with grape-like mor- 
phology — = present; 1 = absent. This 
grape-like configuration may not represent a 
distinct morphology, but may be due to the 
highly filled condition of the receptacle. This 
condition occurs only in Cacozeliana. 

Reproductive Characters: 20. Spawn mor- 
phology — = formed into gelatinous string 
wound into mass; 1 = short gelatinous tube; 
2 = balloon-like cluster. A gelatinous string 
mass is the common spawn morphology seen 
among cerlthloidean taxa and within Bittiinae. 
The balloon-like cluster of eggs in members 
of Lirobittium is unique, whereas a short ge- 
latinous tube morphology is seen only in It- 
tibittium: both taxa have few, large eggs and 
undergo direct development (Table 4). 

21 . Type of development— = planktonic; 
1 = lecithotrophic (demersal/direct). Most 
members of the outgroup have a planktonic 



GENERIC REVIEW OF BITTIINAE 



267 



Cacozeliana Lirobittium 



Styljdjum Bittium 



Ittibittium 



BittJolum 




FIG. 1 . Cladogram showing relationships among six genera of Bittiinae, using Cerithium as the outgroup (L 
= 41 ; CI = 70; Rl = 53; trees two. Numbers to left of black bars indicate characters: those to right of bars 
represent character states. Only characters with a CI of 100 are shown). 



larval phase in their development. It is thought 
that planktotrophy can evolve to lecithotrophy 
but not vice-versa (Strathmann, 1978). Direct 
developers have larger, fewer eggs per 
spawn mass (Table 4). 

RESULTS 

Phylogenetic analysis resulted in two 
equally parsimonious trees, each with a 
length of 41 steps, a consistency index of 70, 
and a retention index of 53 (Fig. 1). The num- 
ber of steps and the consistency indices of 
each character used in the construction of the 
cladogram are shown in Table 5. The support- 
ing branches of both cladograms had identi- 
cal tree topologies except for the clade sup- 
porting Bittium, Ittibittium, and Bittiolum. In the 
first tree, illustrated herein (Fig. 1), Ittibittium 
and Bittiolum are sister groups of Bittium, 
while in the second tree, Bittium and Bittiolum 
are sister groups of Ittibittium. Both analyses 



strongly support the recognition of six genus- 
level taxa. The monophyly of Bittiinae is es- 
tablished by three synapomorphies (11[1], 
18[1], 20[0]) and one homoplastic character 
(17[1]). The layout of the palliai oviduct, dis- 
cussed in greater detail below, is the source 
of two good synapomorphous characters: a 
ciliated ridge tract and a spermatophore 
bursa in the medial lamina. An epipodial skirt, 
while distinctive of the Bittium-group, is plesi- 
omorphic, because it occurs also in other cer- 
ithioidean groups. 

Cacozeliana stands apart at the base of the 
cladogram from the other taxa and is closest 
to Cerithium, the outgroup. Cacozeliana is de- 
fined by two autapomorphous characters 
(6[2], 7[2]) and by two homoplastic characters 
(5[2], 16[1]). Cacozeliana is well separated 
from all other genera of Bittiinae higher on the 
tree by five synapomorphies (2[1], 4[1], 14[1], 
15[1], 19[1]) and with one homoplastic char- 
acter (13[1]). 

The Lirobittium-Stylidium clade, which is 



268 HOUBRICK 

TABLE 5. List of steps and consistency indices of characters used in construction of cladogram. 



Character 

Steps 

C.L 

Character 

Steps 

C.I. 



1 

3 

33 

12 

2 

50 



2 

1 

100 

13 
3 

66 



3 

2 

50 

14 

2 

100 



4 

1 

100 

15 

1 

100 



5 

3 

66 

16 

3 

33 



6 

3 

66 

17 

2 

50 



7 

2 

100 

18 

2 

50 



8 

1 

100 

19 

1 

100 



9 

1 

100 

20 

2 

100 



10 

1 

100 

21 

2 

50 



11 

3 

100 



geographically confined to the west coast of 
North America, is supported by two synapo- 
morphies (8[1], 9[1]), and two homoplastic 
characters (13[2], 21 [1]) In this clade, Stylid- 
ium is poorly defined by three homoplastic 
characters (1[0], 5[2], 16[1]), whereas Lirobit- 
tium is better founded on one autapomorphy 
(20[2]) and three homoplastic characters 
(6[1], 12[1], 16[0]). 

The Bittium clade is supported by one sy- 
napomorphy (7[1]) and two homoplastic char- 
acters (3[0], 13[1]). Bittium, s.S., is defined by 
one autapomorphy (14[2]) and three ho- 
moplastic characters (2[0], 12[1], 18[1]). It- 
tibittium and Bittiolum, the sister taxa to Bit- 
tium, are separated from it by one 
synapomorphy 10[1]). Bittiolum is supported 
by two autapomorphies (5[1], 11[3]) and two 
homoplastic characters (11[3], 16[0]). A sin- 
gle autapomorphy (20[1]) and six homoplastic 
characters (1[0], 12[1], 13[0], 16[0], 17[0], 
18[0], 21 [1]) define Ittibittium. The characters 
listed above are those derived only from the 
data matrix (Table 3) used in the construction 
of the cladogram (Fig. 1). Other autapomor- 
phies defining terminal branches but not part 
of multiState series were not included in the 
data matrix. These characters are given un- 
der the diagnosis of each genus In the sys- 
tematic portion of this paper. 



DISCUSSION 

The phylogenetic analysis of morphological 
characters of the species in Table 1 resulted 
in recognition of six different morphological 
groups (Fig. 1), which are herein interpreted 
as genus-group taxa under the subfamily Bit- 
tilnae Cossmann, 1906. Generic names al- 
ready exist for five of these groups: Bittium 
Gray, 1847; Bittiolum Cossmann, 1906; Ca- 
cozeliana Strand, 1928; Stylidium Dall, 1907; 
and Lirobittium Bartsch, 1911. A new genus, 
from the Indo-Pacific, is described herein. All 
of the above genera, with the exception of 
Stylidium, are defined by autapomorphous 



characters. If the cladogram shown in Figure 
1 is interpreted strictly, Ittibittium and Bittiolum 
may be regarded as subgenera of Bittium; 
however, because this is a preliminary revi- 
sion of the Bittium-group, based on only a few 
representatives of each genus, and not in- 
cluding other poorly known taxa, it is best not 
to assign differential rank to genus-group taxa 
at this stage. Therefore, I have decided to 
treat all terminal nomina as full genera. 

As noted in an earlier paper (Houbrick, 
1977), other genus-level taxa have been pro- 
posed under the Bittium-group or are thought 
to be linked closely to it. Many of these taxa 
are synonyms of Bittium-group genera de- 
scribed herein or have been proposed for fos- 
sils. The subfamily Bittiinae, as understood in 
this paper, is thought herein to comprise nine, 
possibly ten. Recent genus-group taxa: Bit- 
tium Gray, 1847; Bittiolum Cossmann, 1906; 
Ittibittium gen. п.; Stylidium Dall, 1907; Liro- 
bittium Bartsch, 1911; Cacozeliana Strand, 
1928; Argyropeza Melvill & Standen, 1901; 
and Varicopeza Gründel, 1976. The genera 
Zebittium Finlay, 1927, and Cassiella Gofas, 
1987, are provisionally referred to Bittiinae 
until more information is available. 

Argyropeza and Varicopeza have been 
treated previously by Houbrick (1980a, 
1980b, 1987a), but their anatomy remains 
poorly known and they are not described in 
great detail here. An epipodial skirt has been 
recorded in Varicopeza crystallina (Houbrick, 
1987a: 80), but due to poorly preserved ana- 
tomical material, this structure could not be 
ascertained in Argyropeza species; however, 
the radula of Argyropeza species (Houbrick, 
1980a) is similar to those of members of the 
Bittium-group. 

Anatomical knowledge about potential Bit- 
tium-group species as yet unstudied, such as 
Cassiella from the eastern Atlantic, Zebittium 
from New Zealand, and the many species of 
small-shelled, Bittium-Wke cerithioldeans from 
the Indo-Pacific, may reveal even more new 
genus-level taxa to be included under Bittii- 
nae. 



GENERIC REVIEW OF BITTIINAE 



269 



SYSTEMATIC TREATMENT OF BITTIINAE 

The species studied have been placed into 
groups (genera) according to the above phy- 
logenetic analysis. The type- or representative 
species of each genus is described, and notes 
on reproductive biology and ecology are in- 
cluded, when possible. Shell-length measure- 
ments for each species represent the largest 
specimen observed. Representatives of other 
genera for which anatomical material was 
lacking are described from shell morphology 
and radular morphology, if available. 

BITTIINAE COSSMANN, 1906 

Bittiinae Cossmann, 1906: 61. 
Procerithiinae Cossmann, 1906, sensu Grün- 
del, 1976 (in part). 

Diagnosis 

Shell small, turreted, narrowly elongate to 
pupate, with moderate spiral and axial sculp- 
ture frequently cancellate and/or beaded. Ap- 
erture with short but distinct anterior canal. 
Spiral sculpture usually 4-5 spiral cords per 
whorl. Animal with epipodial skirt, opercular 
lobe, and palliai oviducts comprising large 
sperm bursa and seminal receptacle in pos- 
terior part of medial lamina, and spermato- 
phore bursa and ciliated ridge tract in poste- 
rior lateral lamina. Ciliated gutter leading from 
oviduct down right side of foot in females. 
Glandular ovipositor at base of right side of 
foot in most species. Nervous system dialy- 
neurous. Spawn consisting of gelatinous, 
winding strings. 

Taxonomic Remarks 

The Bittium-group (Bittiinae Cossmann, 
1906) has been placed under Cerithiidae by 
nearly all authors (Cossmann, 1906; Thiele, 
1929; Wenz, 1938; Golikov & Starabogatov, 
1975; Ponder & Waren, 1988), except Grün- 
det (1976), who assigned the group to the Ju- 
rassic family Procerithiidae Cossmann, 1906 
(erroneously cited by Cossmann as 1 905). He 
allocated 12 genus-group taxa to the subfam- 
ily Procerithiinae (= Bittiinae). Of these, Bit- 
tium, Bittiolum, Semibittium and Procerithium 
were treated as full genera; Cerithidium Mon- 
terosato, 1884, Rasbittium Grünäe\, 1976, Li- 
robittium Bartsch, 1911, Cacozeliana Strand, 
1928, and Stylidium Dall, 1907, were consid- 
ered to be subgenera of Bittium. The extinct 



taxa Cosmocerithium Cossmann, 1906, In- 
fracerithium Gründel, 1974, and Rhabdocol- 
pus Cossmann, 1906, were treated as sub- 
genera of Procerithium. Gründel (1976) also 
included Argyropeza Melvill & Standen, 1 901 , 
Varicopeza Gründel, 1976, and the extinct 
genus Cryptaulax Gründel, 1976, with sub- 
genera Pseudocerithium Cossmann, 1884, 
and Xystrella Cossmann, 1 906, in the Bittium 
group under the subfamily Cryptaulaxinae 
Gründel, 1976. Excluding the Jurassic taxa, 
the Recent genera Argyropeza and Varico- 
peza should probably be included in the Bit- 
tiinae, because the few morphological and 
anatomical characters known about these 
taxa strongly suggest affinity to this subfamily. 
The other extinct genus-group taxa and Pro- 
cerithium should be excluded from Bittiinae, 
because the evidence supporting a relation- 
ship of these taxa with the Bittium-group is 
based solely on the ontogenesis of spiral 
sculpture as seen on the early shell spire, a 
character which is, at best, tenuous: more 
characters are needed to lend credence for 
such a relationship. While Gründel's (1976) 
hypothesis poses interesting questions, it is 
founded mostly on shell sculpture, which is 
taxonomically informative but potentially phy- 
logenetically misleading. Considering the Ju- 
rassic age of the Procerithium group and the 
great likelihood of homoplasy in shell mor- 
phology, the belief that the Bittium- and Pro- 
cerithium- groups are of the same lineage is 
largely speculative, cannot be falsified, and 
should not be accepted as evidence for a phy- 
togeny (Houbrick, 1988). 

The name Elassum Woodring, Bramlette & 
Kew, 1946, has been traditionally associated 
with the Bittium-group in the literature, and 
was proposed by Woodring et al. (1946: 68) 
for Pleistocene and Recent material from 
southern California previously named Bittium 
californicum Dall & Bartsch, 1901, and origi- 
nally assigned to the subgenus Elachista Dall 
& Bartsch, 1 901 . Bittium californicum is the 
type species of Elachista by monotypy. How- 
ever, as Elachista is preoccupied, a new 
name, Alabina Dall, 1902, was proposed to 
replace it. Woodring et al. (1946) did not be- 
lieve the taxon californicum Dall & Bartsch, 
1901, was an Alabina and thus proposed 
Elassum to accomodate it, noting that the 
species was more Bittium-Wke than Alabina- 
like. Because Elachista, Elassum, and Alab- 
ina have the same type species, Elassum be- 
comes a junior synonym of Alabina. The shell 
of the type species somewhat resembles 



270 



HOUBRICK 



those of members of the Bittium-group, and I 
concur with Woodring et al. (1946) that it pos- 
sibly should be included as a component ge- 
nus of the Bittium-group; however, as there is 
no preserved material of living animals of this 
taxon to confirm this supposition, Alabina [ = 
Elassium] is not further treated herein. 

Houbrick (1977: 103) initially placed 13 
nomina into the synonymy of Bittium, sensu 
lato. Subsequent studies on the Bittium-group 
and evidence derived from anatomical char- 
acters presented herein now allow exclusion 
of six genera originally included in that syn- 
onymy and a more focused diagnosis of Bit- 
tium, s.S. An annotated list of taxa previously 
included in the Bittium-group, but now ex- 
cluded, is presented below (Jurassic genera 
not included): 

1 . Bittinella Dall, 1 924 (type species: Bittium 
hiiloense Pilsbry & Vanatta, 1908). The type 
species of this genus is a rissoid of the genus 
/sse//e//a Weinkauff, 1881, subfamily Rissoin- 
inae (Ponder, 1985: 95; Kay, 1979: 80). Bit- 
tium parcum Gould, 1861, has been errone- 
ously assigned to Bittinella (see below). 

2. Bittiscalia Finlay & Мапл /ick, 1 937 (type 
species: Bittium simplex Marshall, 1917). It is 
unclear to which group this extinct species 
should be assigned. Although Finlay & Mar- 
wick (1937: 44) placed it under Cerithiidae, 
they noted its similarity to Zeacumantus Fin- 
lay, a batillahid (Houbrick, pers. obser.). Their 
drawing of the type species (Finlay & Marwick, 
1937: pi. 5, fig. 20) shows a shell with an an- 
terior canal that is a wide shallow notch, similar 
to poorly developed anterior canals seen in 
some Bittium and Alabina species. Because 
this is a fossil, we may never know with cer- 
tainty the correct family assignment. Although 
the authors placed it under Cerithiidae, they 
were obviously equivocal about this assign- 
ment. It is best to leave Bittiscalia under the 
broader category of Cerithiidae and to exclude 
it from the more narrow assignment of Bittinae. 

3. Brachybittium Weisbord, 1962 (type spe- 
cies: Bittium (Brachybittium) caraboboense 
Weisbord, 1962). The type species, a fossil, 
appears to be an immature or fragmentary 
Cerittiium species, judging from its illustration 
(Weisbord, 1962: pi. 15, figs. 5-6). 

4. Cerithidium Monterosato, 1884 (type 
species: Cerittiium submamillatum Rayneval 
& Ponzi, in Rayneval et al., 1854). Cerittiidium 
was introduced by Monterosato (1884) who 
noted that it was characterized by a rounded 
aperture and lack of an anterior canal. Mon- 
terosato listed a single species under the ge- 



nus, Cerittiium submamillatum Rayneval & 
Ponzi, 1854, which he considered a synonym 
of Turritella pusilla Jeffreys, 1860. As Gofas 
(1987: 110) remarked, the former name was 
originally given to a Pleistocene fossil which is 
not conspecific with the Recent species. Go- 
fas (1987) remarked that the designation of 
Cerittiium submamillatum as the type species 
of Cerithidium by Cossmann (1906) should 
prevail over that of Turritella pusilla by Wenz 
(1940). I agree with Gofas (1987: 109-110) 
that both species are congeneric and have 
sculpture similar to Bittium reticulatum; how- 
ever, in a Cerithidium species examined by 
Ponder (Ponder, in litt.), the female palliai ovi- 
duct was closed, which is very different from 
the open systems known in all other members 
of Bittiinae. A closed pallia! oviduct has not 
yet been demonstrated in the type species of 
Cerithidium, but on the basis of the closed 
system noted by Ponder, Cerithidium is ex- 
cluded provisionally from Bittiinae. 

5. Dahlakia Biggs, 1971 (type species: 
Dahlakia leilae Biggs, 1971). The type spe- 
cies is a junior synonym of Cerithium proteum 
Jousseaume, 1930 (Houbrick, 1978), and I 
believe both names are probable synonyms 
of Cerithium scabridum Philippi, 1848. 

6. Eubittium Cotton, 1937 (type species: 
Bittium lawleyanum Crosse, 1863) [not Eubit- 
tium Cossmann, 1902]. The syntypes of the 
type species of this genus (MNHN, Paris) are 
Batillariella estuarina (Tate, 1893), which is a 
batillahid (family Batillariidae), and not closely 
related to Cerithiidae. In any case, the name 
Eubittium Cotton is a secondary homonym. 

7. Paracerithium Cotton, 1932 (type spe- 
cies: Bittium lawleyanum Crosse, 1863) [not 
Paracerithium Cossmann, 1902]. This taxon 
is a secondary homonym and has the same 
type species as the previous taxon, which is a 
batillariid. 

8. Sundabittium Shuto, 1978 (type species: 
Cerithium fritschi Boettger, 1883). It is highly 
unlikely that this fossil genus is related to the 
Bittium group. Shuto himself (1978: 152) was 
equivocal in assigning it to Bittium. The fig- 
ures of С fr/fsc/7/ depicted by Martin (1914: pi. 
5, figs. 132-134) suggest an Abyssochrysos 
species, but this assignment needs confirma- 
tion by examination of the type material. 

Discussion 

The subfamily Bittiinae is characterized by 
small-shelled species generally having can- 
cellate sculpture and short canals. Monophyly 



GENERIC REVIEW OF BITTIINAE 



271 



for Bittiinae is tentatively established by the 
synapomorphous layout of the palliai oviduct 
(see description under Bittium reticulatum; 
Fig. 6C); i.e., the presence of three sperm 
chambers: a large bursa (1), and smaller 
seminal receptacle (2) in the posterior half of 
the medial lamina, and a spermatophore 
bursa (3) in the posterior lateral lamina. The 
position of the spermatophore bursa in the 
lateral lamina appears to be a unique synapo- 
morphy defining Bittiinae, but this needs to be 
confirmed by observation of spermatophores 
in the bursa in other members of the subfam- 
ily. This character does not occur in Ittibittium, 
a new genus described herein; thus, it had a 
CI of 50 in the analysis. The ciliated ridge tract 
(Fig. 6B, C, ctr) on the lateral lamina epithe- 
lium leading into the spermatophore bursa is 
also a synapomorphy defining Bittiinae. This 
is an uncommon feature among cerithioide- 
ans, and is unusually long. Some plesiomor- 
phic characters, such as the well-developed 
epipodial skirt and epipodial tentacles, occur 
in other cerithioidean groups, but in combina- 
tion with the above synapomorphous fea- 
tures, are characteristic of the Bittiinae. Ittibit- 
tium, new genus, deviates from other 
members of the subfamily in having the albu- 
men gland protrude beyond the posterior 
mantle cavity into the visceral coil. In other 
respects, it generally agrees with the remain- 
ing genera of the Bittiinae. 

The Recent genera treated herein are each 
characterized by external anatomical charac- 
ters (Fig. 2), which allow easy classification of 
living animals. Two genera of the subfamily 
{Bittiolum and Ittibittium, gen. п., have a large 
metapodial mucus gland marked by an elon- 
gate slit in the middle of the sole (Fig. 2), lead- 
ing deep into the center of the foot. While the 
epipodial skirt and opercular lobe are charac- 
teristic of Bittiinae, these characters and the 
metapodial mucus gland also occur in spe- 
cies of Alaba H. Adams & A. Adams, 1854, 
and Litiopa Rang, 1829 (Litiopidae Fischer, 
1885), in members of Colina H. Adams & A. 
Adams, 1854 (Cerithiidae Férussac, 1819), 
and in species of Plesiotrochus Fischer, 1878 
(Plesiotrochidae Houbrick, 1990b) (Kosuge, 
1964; Houbrick, 1987b; Luque et al., 1988; 
Houbrick, 1990a, 1990b, respectively). I have 
previously pointed out the anatomical fea- 
tures shared by Colina with members of the 
Bittiinae (Houbrick, 1990a: 50-51). Species 
of Plesiotrochus Fischer, 1878, also have a 
papillate epipodial skirt and an elongate 
metapodial slit leading into a large metapodial 



mucus gland, but differ considerably from 
members of the Bittium-group in other ana- 
tomical characters (Houbrick, 1990b: 247- 
248), and are an unusual family. 

The relationship of the Bittium-group to 
other small-shelled cerithioidean genera such 
as Scaliola A. Adams, 1860, and Finella A. 
Adams, 1860, remains unclear because the 
anatomy of these taxa is still unknown. Ponder 
(1991) recently described the anatomy of a 
species of Diala A. Adams, 1861, which re- 
sulted in his recognition of a separate family, 
Dialidae Ludbrook, 1 941 . According to Ponder 
(1991 : 504-506), Diala species have a weak 
epipodial fold (epipodial skirt), a pair of lateral 
opercular lobes, and a posterior opercular 
flap, which appear to be homologous with the 
epipodial skirt and opercular lobe described in 
the Bittiinae members above. However, unlike 
the situation in Bittiinae, Diala species lack the 
metapodial mucus gland and the glandular 
ovipositor on the right side of the foot in fe- 
males. Additionally in Diala species, the lateral 
lamina of the pallia! oviduct does not have a 
sperm pouch and the paraspermatozoa are 
unique among Cerithioidea (Healy, 1986). 

The rachidian radular tooth of most mem- 
bers of the Bittium-group is characterized by 
being wider than tall and usually has a basal 
plate with concave sides. This differs from the 
hour-glass shape of the rachidian tooth found 
in small-sized species of Diala, Litiopa, Alaba, 
and Varicopeza (Ponder, 1991: fig. 3F, G; 
Houbrick, 1987a: figs. 14, 19; 1987b: figs. 9, 
10), taxa frequently confused with Bittium- 
group members. For dental cusp patterns 
among Bittiinae taxa, see Table 2. 

Although members of Bittiinae are primarily 
grazers of epiphytic microalgae, many species 
appear to feed on particulate matter gathered 
by cilia and mucus on the anterior ctenidial 
filaments when the animal is stationary. 

The ultrastructure of the sensory epithelium 
of the osphradia of members of the Bittium- 
group is typical of Cerithioidea, and Haszpru- 
nar (1 985: 479) has shown that the osphradial 
cells bear paddle cilia. The osphradial classi- 
fication of Bittiinae species falls under Hasz- 
prunar's (1985) group "Si2." Haszprunar 
(1985) repeated the Fretter & Graham (1962: 
367) statement that the osphradium is a "sim- 
ple brown ridge," but this is not concordant 
with my observations of the pectinate condi- 
tion in many taxa of the group. 

The phylogeny and relationship of mem- 
bers of the Bittium-group will remain unclear 
until the anatomy of other cerithioidean taxa is 



272 



HOUBRICK 





BITTIUM 





ITTIBITTIUM 





BITTIOLUM 





STYLIDIUM 




LIROBITTIUM 





CACOZELIANA 

FIG. 2. External anatomical characters of five genera of the Bittium-group. Figures to left represent right 
lateral views of headfoot, showing mantle edge, ciliated gutter, ovipositor and epipodial skirt configuration; 
figures to left show sole of foot, anterior mucus gland, metapodial mucus gland (when present) and con- 
figuration of epipodial skirt. 



GENERIC REVIEW OF BITTMNAE 



273 



better understood and a phylogenetic analy- 
sis can be accomplished. 



BITTI UM GRAY, 1847 

BiWum Gray, 1847a (Oct.): 270 (Type species 
by subsequent designation, Gray, 1847b: 
Strombiformis reticulatus DaCosta, 
1778). Thiele, 1929: 211; Wenz, 1940: 
755; Nordsieck, 1968: 68; Houbrick, 
1977: 103. 

Cerithiolum Tiberi, 1869: 263 (Type species 
by original designation, Strombiformis re- 
ticulatus DaCosta, 1778). 

Manobittium Monterosato, 1917: 20 (Type 
species by monotypy, Cerithium latreillei 
Payraudeau, 1826, = S. reticulatus). 
Thiele, 1929:212. 

Inobittium Monterosato, 1917: 20 (Type spe- 
cies by monotypy, Cerithium lacteum 
Philippi, 1836, = S. reticulatus). Thiele, 
1929:212; Wenz, 1940: 757. 

Rasbittium Gründel, 1976: 53 (Type species 
by original designation, Cerithium latreil- 
lei Payrauäeau, 1826, = S. reticulatus). 

Diagnosis 

Shell small, elongate, with short anterior ca- 
nal and sculptured with 4-5 spiral cords with 
many aligned small beads formed where axial 
riblets are crossed by spirals. Operculum cir- 
cular, paucispira! with subcentral nucleus. Epi- 
podial skirt with many small, short papillae. 
Opercular lobe with small pointed papillae. 
Well-developed ovipositor comprising parallel 
glandular ridges and bisected by egg-laying 
gutter on right side of foot near edge of epi- 
podial skirt. Osphradium ridge-like, weakly 
monopectinate, one-half the ctenidial length. 
Openings to sperm bursa well separated from 
opening to seminal receptacle. 

Remarks 

Bittium Gray, 1847a, was first proposed in 
manuscript by Leach in 1818 for a classifica- 
tion of British Mollusca, and it was subse- 
quently made available by Gray (1847a). 
Leach's list referred Bittium and several other 
diverse genera to Purpuridae and under the 
65th entry listed Murex reticulatum, M. tuber- 
culare, M. adversum, M. elegantissimum, 
and M. spenceri, consecutively, under Bit- 
tium. Besides Bittium reticulatum, the other 
species listed by Leach represent two gen- 



era, Triphora Blainville, 1 828, and Cerithiop- 
sis Forbes & Hanley, 1851 . Neither a descrip- 
tion of Bittium nor a type species were given. 
Three months later. Gray (1847b) cited only 
Bittium reticulatum (Da Costa, 1778) under 
Bittium, and this citation is a subsequent des- 
ignation. (Gray's system is explained in his 
introduction, pp. 129-130, and the species so 
listed are to be taken as type designations). 
The earliest diagnosis of Bittium is that of H. 
Adams & A. Adams (1854) who besides de- 
scribing shell characters, noted the opercu- 
lum, epipodial skirt, and opercular lobe. 

My original paper on Bittium (Houbrick, 
1977) reviewed the nomenclatural history of 
the genus, and should be consulted for de- 
tailed information about the confusion and 
taxonomic problems between Bittium and 
other taxa of small-shelled cerithioideans. 
Subsequent to that review, there have been 
many changes and the synonymy of Bittium 
originally published (Houbrick (1977: 103) 
has been modified herein: some taxa have 
been excluded, and genera not originally in- 
cluded have been added. A commentary on 
the present synonymy follows: Cerithiolum is 
an objective junior synonym of Bittium: both 
genera share the same type species, Bittium 
reticulatum. Gründel (1976) regarded Cer- 
ithidium and Rasbittium Gründel, 1976, as 
subgenera under Bittium, s.S., but as shown 
before, Cerithidium is excluded from Bittiinae. 
Rasbittium is a primary objective synonym of 
Manobittium as seen in the synonymy above. 
Manobittium and Rasbittium are considered 
subjective junior synonyms of Bittium be- 
cause both share the same type species, Cer- 
ithium latreillei, which is considered by me 
and a number of authors to be conspecific or 
subspecific with Bittium reticulatum (see Ver- 
duin, 1976). The eastern Atlantic species, 
Cerithium lacteum, which is the type species 
of Inobittium, also is considered herein to be 
conspecific with Bittium reticulatum. Wenz 
(1940: 757) regarded Inobittium as a syn- 
onym of Lirobittium, but I see no close resem- 
blance between the shells of the two. Should 
Cerithium lacteum be a distinct species, as 
thought by Verduin (1976), the differences 
are certainly not of generic weight; conse- 
quently, Inobittium is regarded as a subjective 
junior synonym. 

Discussion 

The genus Bittium is characterized by a can- 
cellate, beaded shell sculpture formed by 4-5 
dominant spiral cords and numerous axial rib- 



274 



HOUBRICK 



lets (Fig. 3A-E), a circular operculum with sub- 
centric nucleus (Fig. 3F), and by the snnall 
papillae along the edge of the epipodial skirt 
and opercular lobe (Fig. 2). The ovipositor in 
fernales is a highly developed, raised glandu- 
lar lump at the base of the foot near the sole 
edge, forming a series of parallel, glandular 
ridges bisected by the deep ciliated egg-laying 
groove (Fig. 4B, ovp). The ridge-like monopec- 
tinate osphradium is unusual in having the 
pectins on its right side. It is half the length of 
the ctenidium. The openings to the sperm 
bursa and seminal receptacle in the lateral 
lamina of the palliai oviduct (Fig. 68, C, osr, 
osp) are well separated from each other in 
contrast to most other members of the Bittium- 
group. 

The shells of small-sized Cerithium species 
frequently are erroneously misclassified as 
Bittium species. Gründel (1976) presented 
several conchological features that he be- 
lieved separated the two genera. He stated 
that Cerithium differs from Bittium in having a 
more complex aperture, but this is only true 
for larger Cerithium species: some small spe- 
cies, such as Cerithium atromarginatum, Cer- 
ithium egenum, and Cerithium zebrum, have 
apertures like those of Bittium (Houbrick, 
1978). Gründel (1976) further indicated that 
ontogenetic sculptural development in Cerith- 
ium begins with a single primary spiral cord 
that becomes stronger and more prominent, 
forming a keel that is not integrated with the 
weaker axial riblets; moreover, there are 
many fine spiral threads of varying strength. 
In Bittium, whorl sculpture begins with two 
spirals that quickly become four primary spiral 
cords forming a network with sharply defined 
axial riblets. The so called "definitive" shell 
characters proposed by Gründel (1976) are 
unreliable, because the more species that are 
examined, the more exceptions and ambigu- 
ities one encounters. 

Marcus & Marcus (1963) cited the pres- 
ence of a metapodial mucus gland In Bittium 
reticulatum, crediting this information to Frot- 
ter (1948). However, no such gland was ob- 
served in living or preserved, sectioned spec- 
imens from the Azores; furthermore. Ponder 
(in litt.) did not note this structure on speci- 
mens of Bittium reticulatum from the western 
coast of Sweden. Fretter's (1948: 628) paper 
merely cites the presence of this gland in 
such small gastropods as Bittium, Cerithiop- 
sis, and Triphora, but as she mentioned only 
generic names, it is unclear what "Bittium" 
species she actually observed. 



All living, observed members of the Bittii- 
nae appear to be feeders of epiphytic microal- 
gae, such as diatoms, which occur commonly 
on sea grasses. Most species occur in large 
populations and are highly gregarious. 

Species of the genus Bittium appear to be 
primarily concentrated in the eastern Atlantic: 
the Bittium reticulatum complex and species 
closely related to it are commonly found 
throughout the Mediterranean, north African, 
and western European regions, and appear 
to be adapted to temperate and cold waters. 
Bittium impendens from the Indo-Pacific, 
which differs from the Atlantic Bittium species 
only in lacking a monopectinate osphradium, 
is herein included under the genus Bittium. If 
this species truly belongs in Bittium s.S., and if 
other anatomically unknown Indo-Pacific spe- 
cies are examined, the geographic distribu- 
tion of the genus Bittium may be far wider 
than is now thought. 

Bittium reticulatum (Da Costa, 1778) 
(Figs. 3-6) 

Strombiformis reticulatus Da Costa, 1778: 

117, pi. 8, fig. 13. 
Murex rei/cu/afus (Da Costa). Montagu, 1803: 

272. 
Cerithium latreillei Payraudeau, 1826: 143. 
Cerithium lacteum Philippi, 1836: 195. 
Cerithium reticulatum, Risso, 1826: 157; G. B. 

Sowerby, 1855: pi. 15, fig. 8; Jeffreys, 

1867: 258; 1869: pi. 80, fig. 4; 1885: 57. 
Bittium reticulatum, Watson, 1886: 540; Buc- 

quoy et al., 1884: 212-215, pi. 25, figs. 

3-9; Tryon, 1887: 150-151, pi. 29, figs. 

78-83; Dautzenberg, 1889: 40-41. 

Description 

Shell {F\g. 3A-H): Shell elongate, reaching 
15 mm in length, comprising 9-1 moderately 
inflated whorls. Protoconch (Fig. 3G) com- 
prising two weakly sculptured whorls. Early 
whorls beginning with two spiral cords and 
broad subsutural ramp (Fig. 3H). Adult whorls 
sculptured with 4-5 spiral cords beaded 
where many small axial riblets cross over 
them, creating cancellate sculpture. Suture 
deeply impressed. Body whorl a little under 
one-third shell length, having weak basal con- 
striction and small anterior canal weakly re- 
flexed to left. Body whorl sculptured with five 
major spiral cords and 5-6 weaker cords on 
its base. Aperture ovate, a little over one-third 
shell length, with concave columella having 



GENERIC REVIEW OF BITTIINAE 



275 




FIG. 3. Representatives of genus Bittium: А-И, B. reticulatum; l-N, B. impendens. A-C, SEM micrographs 
of a reticulatum from Sao Miguel, Azores (USNM 878030), 6 mm length; D, E, B. reticulatum from Tunisia 
(USNM 754051), 1 1 mm length; F, SEM micrograph of operculum of B. reticulatum, bar = 0.5 mm; H, SEM 
micrograph of immature shell of B. reticulatum, bar = 0.5 mm; l-L, SEM micrographs of shell of B. 
impendens from Honolulu, Hawaii (USNM 857098), 5 mm length; M, SEM micrograph of operculum of B. 
impendens, bar = 0.5 mm; N, SEM micrograph of protoconch of B. impendens, bar = 150 ixm. 



276 



HOUBRICK 



slight columellar callus; anterior canal short, 
shallow; anal canal very small; outer lip 
rounded, weakly crenulate. Periostracum 
thin, light tan. 

Animal (Figs. 4-6): Head-foot of animal pig- 
mented light yellowish-brown overlain by 
large dark brown blotches and small white 
spots. Visceral mass with 8 visceral whorls, 
comprising mostly digestive gland and over- 
lying gonads. Ovary white; testis dirty yellow. 
Stomach about one whorl in length. Kidney 
large, light tan, about two-thirds whorl in 
length. Columellar muscle white, broad, short, 
about one-half length of palliai cavity. Head 
(Fig. 4A) with elongate, narrow snout (Fig. 4B, 
sn), flattened dorso-ventrally, expanded at bi- 
lobed tip, with bright yellow, oval-shaped oral 
pad at antero-ventral end (Fig. 4A, C, 1). 
Cephalic tentacles (Fig. 4A, t) elongate, nar- 
row, with broad peduncular bases each with 
large dark eye. Foot narrow, elongate, cres- 
cent shaped anteriorly. Deep transverse slit 
(Fig. 4C, amg) between epipodial lips marks 
entrance to large ovate anterior mucus gland 
extending via central duct deep into anterior 
foot. Epipodium separated from lower foot 
and densely ciliated sole by deep, laterally 
placed groove (Fig. 48, epg) forming broad 
epipodial skirt (Fig. 48, C, eps) extending 
posteriorly on each side of foot from corners 
of anterior epipodial lips of anterior mucus 
gland around entire foot base, joining behind 
and below opercular lobe. Lateral epipodial 
skirt scalloped along edges of each side of 
median and posterior parts of epipodium, 
having small papillae (Fig. 48, C, ep); epipo- 
dial skirt forming long opercular lobe (Fig. 48, 
C, opi). Sole of foot (Fig. 4C, s) indistinctly 
divided into two parallel axial parts, forming 
anterior longitudinal fold. No metapodial mu- 
cus gland. Operculum (Fig. 3F) corneous, 
tan, circular, paucispiral with subcentral nu- 
cleus and with thin, transparant border. Cili- 
ated gutter (Fig. 48, C, eg) emerging from 
right side of mantle cavity (Fig. 4C, ex) and 
running down right side of foot; ciliated gutter 
leads to large glandular ovipositor (Fig. 48, C, 
ovp) and egg-laying pit at base of epipodium 
in females. Ovipositor oval-shaped, com- 
prised of glandular, transparant white tissue 
formed into many parallel pleats divided 
transversely by deep central slit. Mantle bi- 
lobed at edge, having smooth outer lobe and 
inner lobe with many small papillae, becom- 
ing smooth ventrally. Mantle papillae (Fig. 48, 
C, mp) slender, darkly pigmented, each with 



white spot. Mantle edge thickened at inhalant 
(Fig. 4C, inh) and exhalant siphons. 

Palliai Cavity: Palliai cavity deep, comprising 
about two whorls. Osphradium olive colored, 
ridge-like, pectinate on right side only, bor- 
dered on each side by narrow ciliated strip. 
Osphradium wide, about one-half ctenidial 
length, beginning close behind inhalant si- 
phon and extending length of ctenidium. 
Ctenidium bluish-gray, comprising numerous 
finger-like, triangular filaments with narrow 
bases. Hypobranchial gland narrow, glandu- 
lar comprising several kinds of large gland 
cells that stain dark blue. Rectal tube dis- 
tended, filled with elongate, ovoid-shaped fe- 
cal pellets. Palliai gonoducts open, beginning 
behind mantle edge and extending posteriorly 
as far as kidney. 

Reno-pericardial System: Kidney large, about 
two-thirds whorl in length, beginning at ante- 
rior end of style sac, extending anteriorly well 
into mantle cavity roof, lying over one-third of 
posterior palliai gonoduct. Kidney with simple 
kidney opening, but no renopericardial duct. 
Pericardium typically monotocardian, lying ad- 
jacent to posterior wall of mantle cavity. 

Alimentary System: Mouth (Fig. 4A, m) lying 
antero-ventral ly on snout, opening into oral 
cavity between two semicircular lips (Fig. 4A, 
C, 1). Buccal mass (Fig. 4D, bm) relatively 
small, about one-third snout length, loosely 
attached to snout wall by numerous thin mus- 
cle strands. Jaw tan, semicircular, comprised 
of cuticular cones and lying on either side of 
entrance to anterior buccal cavity. Radular 
ribbon (Fig. 5A; Table 2) folded beneath buc- 
cal mass and radula sac emerging behind it. 
Rachidian tooth (Fig. 5C) with dorso-ventrally 
compressed basal plate with concave sides 
rounded base and with V-shaped base but- 
tressed on each side with a basal lateral ex- 
tension; rachidian broader above than below, 
having cutting edge with slightly concave top, 
and comprising large, spade-shaped central 
cusp flanked on each side by 2-3 small, 
pointed denticles. Lateral tooth (Fig. 58) with 
broad basal plate comprising long, ventrally 
extending, central pillar having small pustule 
on its face, and with moderately long lateral 
extension; cutting edge comprising very large 
spade-shaped cusp with one inner denticle 
and 3-6 outer denticles. Marginal teeth (Fig. 
5A) curved, elongate, with broad, swollen 
shafts, narrowing and becoming spatulate at 
tips; inner marginal tooth with tip having long 



GENERIC REVIEW OF BITTIINAE 



277 



amg 




FIG. 4. Anatomical representations of Bittium reticulatum. A, head and snout; B, lateral view of headfoot; C, 
head and sole of foot; D, anterior alimentary system exposed by dorsal longitudinal cut through wall of buccal 
cavity, aes = anterior esophagus; amg = anterior mucus gland; beg = subesophageal gland; bg = buccal 
ganglion; bm = buccal mass; с = ciliated strip; eg = ciliated gutter; eg = esophageal gland; ep = epipodial 
papilla; epg = epipodial groove; eps = epipodial skirt; ex = exhalant siphon; inh = inhalant siphon; I = lip; 
leg = left cerebral ganglion; Ipg = left pleural ganglion; Isg = left salivary gland; m = mouth; mp = mantle 
papilla; op = operculum; opi = opercular lobe; ovp = ovipositor; pes = posterior esophagus; rcg = right 
cerebral ganglion; rpg = right pleural ganglion; rsg = right salivary gland; s = sole;seg = supraesophageal 
ganglion; sn = snout; t = tentacle. 



278 



HOUBRICK 




FIG. 5. Scanning electron micrographs of radula of Bittium reticulatum from Sao Miguel, Azores (USNM 
878030). A, half row with marginal teeth folded back, bar = 19 |xm; B, rachidian and lateral teeth, bar = 15 
fjim; C, detail of rachidian teeth, bar = 4 (im. 



central cusp, 3-4 inner denticles, 4 outer 
denticles; outer marginal tooth sanne, but 
lacking outer denticles. Salivary glands (Fig. 
4D, rsg, Isg) comprising pair of narrow, un- 
coiled, shiny tubes, beginning behind nerve 
ring, extending through it anteriorly, opening 
into far anterior portion of buccal cavity. Buc- 
cal cavity opening and enlarging immediately 
behind nerve ring, having pair of prominent 
dorsal folds and smaller pair of smaller ventral 
folds. Interior mid-esophageal walls highly 



folded, forming large, olive-brown esophageal 
gland (Fig. 4D, eg). Internal epithelium of 
esophageal gland (Fig. 7A, B, eg) forming nu- 
merous transverse folds or lamellae, staining 
dark blue with Methylene blue. Posterior 
esophagus (Fig. 4D, pes) narrow and straight, 
running on top of columellar muscle, entering 
into left side of stomach. Stomach large, com- 
prising about one whorl of visceral mass, in- 
cluding style sac. Esophageal opening into 
median ventral part of stomach floor. Large 



GENERIC REVIEW OF BITTIINAE 



279 



sorting field with many fine folds adjacent to 
right side of esophageal opening. Minor 
typhlosole bordering right side of esophageal 
opening. Large central elevated pad in center 
of stomach adjacent to single duct to diges- 
tive gland lying short distance below esoph- 
ageal opening. Digestive gland comprising 
single brown lobe consisting of digestive cells 
and secretory cells with dark brown granules. 
Gastric shield on right side of stomach having 
cuticular lining with protruding, toothed edge. 
Depressed epithelial pocket on floor of stom- 
ach adjacent to posterior part of gastric 
shield. Style sac short, about one-third the 
stomach length, nearly spherical, and con- 
taining crystalline style. Style sac adjacent to 
but separate from intestine opening, except 
for limited connection where both enter stom- 
ach. Anterior part of stomach with many par- 
allel ciliated folds and closed off from style 
sac by major typhlosole. Internal intestinal 
walls with many fine folds where exiting stom- 
ach. Intestine curves around style sac, turns 
to right, and runs straight forward. Rectum 
with thin muscular wall, terminating in anal- 
bearing papilla. 

Nervous System: Nervous system epiath- 
roid, dialyneurous. Nerve ring comprised of 
large ganglia. Pleural ganglia (Fig. 4D, rpg, 
Ipg) close to cerebral ganglia (Fig. 4D, rcg, 
leg). Cerebral connective equalling length of 
cerebral ganglion. Buccal ganglia (Fig. 4D, 
bg) small, lying at posterior edge of buccal 
mass. Subesophageal ganglion (Fig. 4D, 
beg) very close to left pleural ganglion (Fig. 
4D, Ipg). Supraesophageal connective mod- 
erately long, about twice length of right pleural 
ganglion; dialyneury between left palliai nerve 
and nerve emerging from supraesophageal 
ganglion (Fig. 4D, seg). Visceral ganglion lo- 
cated in floor of posterior mantle cavity. 

Reproductive System: Testis creamy yellow, 
overlying dark brown digestive gland, extend- 
ing anteriorly about five whorls, ending one- 
half whorl before stomach. Testicular ducts 
on inner side of visceral coil, joining to form 
spermatic duct, enlarging anteriorly, becom- 
ing seminal vesicle and containing two kinds 
of spermatozoa: euspermatozoan with single 
long flagellum and paraspermatozoan with 
[four ?] flagellae. Males aphallate. Male palliai 
gonoduct (Fig. 6A) open, comprising two thin 
walled laminae (Fig. 6A, 11, ml) with thicker 
transverse glandular folds at their attached 
bases bordering gonaductal groove (Fig. 6A, 
gd). Posterior half of male gonoduct thick. 



glandular, comprising prostate gland (Fig. 6A, 
pg). Anterior half of male gonoduct glandular, 
not as thick, putative spermatophore-forming 
organ (Fig. 6A, so). 

Ovary opaque white, thin-walled, overlying 
digestive gland, extending anteriorly, ending 
about one-half whorl before stomach. Coelo- 
mic oviduct (Fig. 68, C, cod) short tube, highly 
ciliated within, beginning anterior to stomach 
with duct wall lying against pericardium (no 
connection), ending at posterior mantle cavity 
where circular sphincter muscle separates it 
from palliai oviduct. Female palliai oviduct 
(Fig. 68, C) large, comprising two laminae, 
enlarged and glandular at their bases, at- 
tached basally to each other and to mantle 
floor, forming ciliated oviductal groove (Fig. 
68, C, ovg). Posterior end of palliai oviduct 
closed. Medial, free lamina with wide anterior 
ciliated sperm gutter (Fig. 68, C, sg) along its 
edge leading to two, well-separated, pocket- 
like openings. First opening (Fig. 68, C, osp) 
leading into large, deep bursa having smooth 
inner epithelium and containing large num- 
bers of non-directed spermatozoa (Fig. 7C, D, 
sp); ciliated gutter continuing posteriorly to 
open (Fig. 7C, osr) into pouch-like, muscular 
seminal receptacle (Fig. 6C, 8 sr; 8C, D, sr) 
containing oriented euspermatozoa with 
heads embedded in receptacle walls. Lateral 
lamina attached to palliai wall, having anterior 
ciliated tract comprising many parallel elon- 
gate, fine ciliated folds (Fig. 68, C, ctr; 7A, 8, 
ctr) running posterior to open into thin-walled 
tube leading into posterior pouch-like bursa 
having highly vacuolated epithelium and func- 
tioning as spermatophore bursa (Fig. 68, C, 
sb). Ciliated tract and folds opening to semi- 
nal receptacle on lateral lamina located oppo- 
site sperm gutter and opening to seminal re- 
ceptacle of medial lamina, both edges 
interdigitating to form closed system. Poste- 
rior half of glandular portion of both laminae 
opaque white color, comprising albumen 
gland (Fig. 68, C, ag; 7C, D, ag); anterior half 
dirty white, comprising capsule gland (Fig. 
68, C, eg; 7A, 8, eg). 

Spawn comprising thin gelatinous string 
(about 25 mm length, uncoiled) tightly coiled 
clockwise or irregularly folded on itself and 
attached to substrate. Jelly string containing 
many small opaque eggs (0.65 fxm diameter) 
each within thin, transparent hyaline capsule 
(110 Jim diameter). Entire spawn mass con- 
tains about 800 eggs. Free swimming bilobed 
planktotrophic veliger larval stage present. 
Larval shell ranging from 170-330 jjim, de- 



280 



HOUBRICK 




ovg 




eg „ 



0.2mm 




ovg 



FIG. 6. Representation of palliai gonoducts of Bittium reticulatum. A, male palliai gonoduct, showing section 
through mid-duct beneath, represented by dotted line; B, palliai oviduct showing three cross sections of duct 
represented by dotted arrows and sections to right; C, reconstruction of palliai oviduct showing configuration 
of ducts and glands (anterior to right), ag = albumen gland; ant = anterior; eg = capsule gland; cod = 
coelomic oviduct; ctr = ciliated ridge tract; gd = gonaductal groove; II = lateral lamina; ml = medial lamina; 
osb = opening to spermatophore bursa; osp = opening to sperm bursa; osr = opening to seminal 
receptacle; ovg = oviductal groove; po = closed portion of palliai oviduct; sb = spermatophore bursa; sg 
= sperm gutter; sp = sperm bursa; sr = seminal receptacle; so = spermatophore-forming organ. 



pending upon age. Larval shell with rounded, 
nearly smooth whorls having thin spiral thread 
forming weak keel and with deep sinusigeral 
notch (Thorson, 1946: 192, fig. 109). 



Discussion 

The status of the many specific and sub- 
specific names comprising the Bittium reticu- 



GENERIC REVIEW OF BITTIINAE 



281 




FIG. 7. Successive sections, anterior to posterior, through palliai oviduct of Bittium reticulatum. A, anterior 
of pallia! oviduct showing relationship of nnantle cavity organs to oviduct, bar = 0.25 тгл; В, mid-section 
showing ciliated ridge tract and opening to sperm bursa, bar = 0.25 mm; C, section through enlarged sperm 
bursa in posterior palliai oviduct, bar = 0.25 mm; D, section through closed posterior of palliai oviduct, bar 
= 0.25 mm. ag = albumin gland; eg = capsule gland; ct = ctenidium; ctr = ciliated ridge tract; eg = 
esophageal gland; hg = hypobranchial gland; os = osphradium; osp = opening to sperm bursa; ovg = 
oviductal groove; r = rectum; sb = spermatophore bursa; sg = sperm gutter; sp = sperm bursa; sr = 
seminal receptacle. 



latum complex is controversial (Verduin, 
1976). It is not my intention to address alpha- 
level problems in this generic review, but the 
Azorean population used for the anatomical 
study herein is considered by some as a sub- 



species or a closely related species of the 
Bittium reticulatum complex. Bittium reticula- 
tum is exceedingly variable in shell sculpture 
throughout its range (compare Figs. 2A, C, 
D), but this is not unusual among cerithioide- 



282 



HOUBRICK 



ans. The palliai oviduct described by Johans- 
son (1947) and notes and sketches made by 
Ponder (Ponder, In litt.) on the anatomy of 
specimens from western Sweden agree sub- 
stantially with my observations of Azorian 
specimens. For the purposes of this study, 
the Bittium reticulatum complex is regarded in 
the broad sense {sensu lato), as a single spe- 
cies. 

The epipodial skirt, characteristic of mem- 
bers of the Bittium-group, forms a highly cili- 
ated lateral groove where it overhangs the 
foot, and carries detrital particles posteriorly 
to the back of the foot where they are dis- 
carded. 

The posterior roof of the pallia! cavity is 
covered by the anterior extension of the renal 
organ, which overlays the posterior palliai 
gonoduct. The renal organ opens via a mus- 
cular sphincter, the renal opening, into the 
posterior palliai cavity. 

The ridge-like osphradium of Bittium retic- 
ulatum is unusual in being pectinate on its 
right side. Although these pectins are small, 
they are clearly visible and very unlike simple 
nonpectinate osphradia of closely related 
taxa. 

The rachidian tooth of the radula of Bittium 
reticulatum is similar to those of members of 
other genera in the group, but unlike that of 
Cacozeliana (see below). Table 2 gives the 
comparative dentition of the radular teeth. 

Bittium reticulatum has three sperm stor- 
age spaces, two connected to the ciliated 
groove of the non-glandular portion of the me- 
dial free lamina, and one in the posterior part 
of the non-glanduiar attached lateral lamina 
(Fig. 6B, 11). It is not entirely clear how these 
three bursae function. Of the two bursae in 
the medial lamina, the smaller one is clearly 
the seminal receptacle, because oriented eu- 
spermatozoa are found in it, exclusively (Fig. 
7C, D, sr). The larger bursa (Fig. 68, sp) con- 
tains considerable numbers of unoriented 
sperm, and much nondescript material (pre- 
sumably disintegrating paraspermatozoa and 
degenerating spermatophores), although 
some euspermatozoa occur with heads ori- 
ented on the inner wall epithelium, especially 
near the opening to the sperm gutter (Fig. 
7D). Although this large bursa in the medial 
lamina contains spermatophores in most cer- 
ithiids, this is not the case in members of the 
Bittium-group, where it appears to function as 
a sperm storage and ingesting area. It is in- 
ferred that the pouch in the posterior of the 
lateral lamina (Fig. 6C, sb, Fig. 7C, D, sb) 



functions as a spermatophore bursa in Bittium 
reticulatum and probably in most other mem- 
bers of the Bittium-group, because Marcus & 
Marcus (1963) found spermatophores in this 
structure in the western Atlantic Bittiolum var- 
ium. I was unsuccessful in finding spermato- 
phores in either structure in specimens of 
Bittiolum varium from Florida. A new genus 
from the Indo-Pacific, Ittibittium, described 
herein, deviates from the typical palliai ovi- 
duct layout in lacking the spermatophore 
bursa in the lateral lamina and in having the 
albumen gland protrude posteriorly beyond 
the back of the palliai cavity into the visceral 
coil. 

The spawn of Bittium reticulatum was first 
described and figured by Meyer & Möbius 
(1872), and the spawn and larvae described 
by Lebour (1937) and Graham (1988). 
Spawn, larvae, veliger, protoconchs, and ju- 
venile shells of this species were described 
and well illustrated by Thorson (1946: 192, 
fig. 109). Other depictions of the larval shell of 
this species are those of Fretter & Pilkington 
(1970: 10-11, fig. 6) and Richter & Thorson 
(1975: pi. 3, figs. 16-17). According to Gra- 
ham (1988), British Bittium reticulatum is a 
summer breeder and attaches its spawn to 
shells, stones or weeds. Spawn comprises a 
cylindrical nbbon about 3 mm in diameter, 
having a total length of 25 mm, and coiled in 
tight spirals. A spawn mass contains about 
1000 eggs, which develop to veliger larvae. 

The geographic range of the Bittium reticu- 
latum complex is broad, comprising western 
Europe, the Azores, North Africa, and the 
Mediterranean. 

Bittium impendens (Hedley, 1899) 
(Fig. 3, l-N) 

Cerithium impendens Hedley, 1899: 434- 
435, fig. 23 (Holotype: AMS C5944; type 
locality: Funafuti Atoll, Ellice Islands); 
Kay, 1979: 118, 120, fig. 45A. 

Description 

Shell: (Fig. 31-N). Shell short, stout, with 
wide base, reaching 7 mm length and com- 
prising 8-9 convex whorls. Protoconch (Fig. 
3N) comprising 2.5 whorls; protoconch 1 
smooth; protoconch 2 sculptured with thin 
central, spiral keel and weak presutural spiral 
thread; lower part of each whorl with micro- 
scopic pustules. Whorls slightly pendant 
abapically, constricted at suture. Adult shell 
sculptured with 3-4 major spiral cords inter- 



GENERIC REVIEW OF BITTIINAE 



283 



spersed with spiral threads. Spiral cords 
weakly beaded and beads aligned to form ax- 
ial riblets. Suture well defined. Weak varices 
randomly distributed. Body whorl very broad, 
about one-half the shell length, with promi- 
nent wide, dorsal varix (Fig. 3J, L); body whorl 
sculptured with about 14 spiral cords and 
strongly constricted at base. Aperture a little 
over twice shell length, broadly ovate, with 
short, wide, shallow anterior canal and 
smooth outer lip extending widely at shell 
base (Fig. 31). 

Animal: Headfoot pinkish white, blotched 
with brown, covered with white spots and with 
chestnut stripes. Kidney bright pink. Right 
side of foot in females with ciliated gutter end- 
ing in small ovipositor at edge of lateral 
groove. Epipodial skirt having very small pus- 
tules or protuberances along lateral edges on 
each side of foot; opercular lobe scalloped 
and pointed at end. Sole of foot pink, without 
metapodial mucus gland. Mantle edge fringed 
dorsally with papillae; underside of inhalant 
siphon with three large papillae. Marginal 
teeth of radula having three inner denticles. 
Osphradium a thin brown ridge, non-pecti- 
nate. Openings to sperm pouch and seminal 
receptacle in medial lamina close to each 
other, situated within common aperture at end 
of sperm gutter in edge of anterior third of 
medial lamina adjacent to opening of sper- 
matophore bursa of lateral lamina. No ciliated 
tract leading to spermatophore bursa. 

Discussion 

Examination of the type lot (holotype and 7 
paratypes) of Cerithium impendens confirms 
that the Hawaiian specimens studied herein 
are conspecific with this taxon. This species 
has not been cited frequently in the literature. 

The assignment herein of Bittium impen- 
dens to the genus Bittium is made with some 
doubt. The shell morphology of this wide- 
spread Indo-Pacific species is quite different 
from that of the type species of Bittium, Bit- 
tium reticulatum (compare Fig. 3A-E and 
31-L), and unlike the shells of other eastern 
Atlantic Bittium species. In addition, the os- 
phradium is ridge-like rather than mono- 
pectinate, and there does not appear to be a 
ciliated tract associated with the spermato- 
phore bursa on the lateral lamina. Instead, the 
opening to the spermatophore bursa is adja- 
cent to the two openings of the bursae in the 
medial lamina. The radula of Bittium impen- 



dens is very similar to that of Bittium reticula- 
tum except that the marginal teeth have fewer 
outer and inner denticles. Aside from these 
differences, the animal shares most of the an- 
atomical features of Bittium reticulatum. Al- 
though an argument could be made that this 
species represents yet another new genus, I 
have conservatively placed Bittium impen- 
dens under Bittium, s.s, with a query, be- 
cause it does have many characters in com- 
mon with the type species of Bittium. 

The shell of Bittium impendens differs from 
other Bittium-group genera by its fir-tree out- 
line and wide body whorl with prominent dor- 
sal varix (Fig. 31-L). The protoconch (Fig. 3N) 
is smooth except for a thin spiral thread and a 
deep sinusigeral notch, indicative of a plank- 
tonic larval phase. Judging from specimens 
from other regions that appear to be concho- 
logically conspecific, this species has a wide 
Indo-Pacific distribution, occurring from cen- 
tral Pacific islands throughout the Indo-West- 
Pacific to east Africa. 

ITTIBITTIUM, New Genus 

Diagnosis 

Shell small, reaching 6 mm length, with in- 
flated whorts and dominant spiral sculpture of 
4-5 cords. Protoconch with depressed, con- 
cave apex, broad sutural ramp, sculptured 
with minute axial striae and two strong spiral 
cords. Operculum ovate, paucispiral with ec- 
centric nucleus. Each side of propodium with 
elongate papilla. Epipodial skirt laterally 
fringed with slender papillae. Large opercular 
lobe having elongate papillae. No ovipositor 
in females. Sole of foot with long, central lon- 
gitudinal slit marking entrance into large 
metapodial mucus gland. Osphradium weakly 
bipectinate. Albumen gland extending past 
posterior of palliai cavity into visceral coil. No 
spermatophore bursa in lateral lamina of pal- 
liai oviduct. Spawn comprising short gelati- 
nous tube. 

Type Species: Bittium parcum Gould, 1861. 

Etymology: A compound of "itti," American 
vernacular prefex for very small, and Bittium. 

Remarks 

This genus is perhaps one of the most dis- 
tinctive of the Bittium group, in terms of its 
unusual protoconch and anatomical features. 



284 



HOUBRICK 



The protoconch with depressed apex and 
broad sutural ramp (Fig. 8!) is unique among 
the Bittium -дюир. The distinctive propodial 
and epipodlal papillae, well-developed epipo- 
dial skirt, and long metapodial mucus gland 
are conspicuous autapomorphiic characters 
in living specimens (Fig. 2). The lack of a 
spermatophore bursa in the lateral lamina of 
the palliai oviduct and the protrusion of the 
albumen gland through the posterior palliai 
cavity into the visceral coil are highly unusual 
autapomorphies, and set Ittibittium, gen. п., 
apart from the rest of the Bittiinae. The place- 
ment of the spermatophore bursa in the lat- 
eral lamina is one of the synapomorphous 
character used in this review to define the 
subfamily Bittiinae; therefore, it is noteworthy 
that Ittibittium, gen. п., has lost this feature. 
The spawn mass of Ittibittium, gen. п., is also 
unusual in being a simple, short tube. 

In some museum collections, Bittium par- 
cum and species similar to it are incorrectly 
assigned to Bittinella Dall, 1924, a genus 
based on Bittium hiloense Pilsbry & Vanatta, 
1 908, which has been shown to be a rissoid of 
the genus Isselia (Ponder, 1985: 95; Kay, 
1979:80). 

Ittibittium parcum (Gould, 1861) 
(Figs. 8-11) 

Bittium parcum Gould, 1861 : 387 (Lectotype, 
R. Johnson, 1964, USNM 2040; type lo- 
cality Okinawa, Ryukyu Islands); G. B. 
Sowerby, 1866: pi. 18, fig. 125; Tryon, 
1887: 155, pi. 30, fig. 20; R. Johnson, 
1964: 122, pi. 12, fig 14; Kay, 1979: 120, 
figs. 22D, 45D, E. 

Cerithium hawaiensis Pilsbry & Vanatta, 
1905: 576 (Holotype ANSP; type locality: 
Hilo, Hawaii). 

Description 

Shell (Fig. 8): Shell small, pupate-elongate, 
comprising about 8 inflated, angulate whorls 
and reaching 5.8 mm length. Protoconch (Fig. 
8F-I) comprising two concave whorls, con- 
cavely flattened apex, very broad sutural 
ramp sculptured with minute axial striae (Fig. 
8F); protoconch whorls sculptured with two 
strong, keel-like spiral cords, with central spi- 
ral cord becoming dominant one. Early whorls 
sharply angulate (Fig. 81); first post-larval 
whorl with keel-like median spiral cord; sec- 
ond whorl with another spiral cord above keel 
and third whorl having 3 spiral cords above 



keel. Adult whorls angulate, sculptured with 
keel-like median cord, 7-8 minor spiral cords, 
each cord abapically overlapped by succes- 
sive one. Eight to nine weak to strong axial 
ribs occasionally on whorls, especially on up- 
per ones (Fig. 8J). Varices randomly placed. 
Suture moderately impressed. Body whorl 
(Fig. 8L) slightly constricted at base, compris- 
ing a little less than half shell length, sculp- 
tured with 15-19 weak flattened spiral cords, 
occasional weak axial ribs and with broad 
varix. Aperture about one-third shell length, 
ovate with smooth outer lip and short broad 
anterior canal. Slight columellar callus 
present. Periostracum thin, nearly transpar- 
ent. 

Animal: Animal pigmentation highly variable, 
ranging from greenish-yellow to pink and 
brown and covered with white blotches. 
Cephalic tentacles wide at bases, elongate, 
twice snout length. Snout elongate, narrow, 
bilobed at tip. Operculum (Fig. 8K) thin, cor- 
neous, tan, circular-ovate, paucispiral with 
subcentral nucleus. Anterior part of foot cres- 
cent-shaped, cowl-like, having single long pa- 
pilla on each side (Fig. 2). Narrow transverse 
slit at edge of propodium leading into large, 
spherical anterior mucus gland, staining deep 
purple in toluidine blue. Lateral epipodial skirt 
with about 10 small, slender papillae along 
edges (Fig. 2) on each side of foot, extending 
posteriorly to large opercular lobe having long 
papillae along its edges; papillae show 
through edges of opercular border. Sole of 
elongate, narrow foot having deep, centrally 
placed, narrow longitudinal slit (Fig. 2) begin- 
ning behind anterior mucus gland slit (Fig. 2) 
and extending posteriorly to back of foot; slit 
leading by way of ciliated duct into deep, mas- 
sive, metapodial mucus gland, staining deep 
purple in toluidine blue. Males with ciliated 
strip on right side of foot, emerging from right 
side of mantle cavity and extending down to 
edge of sole. Ciliated gutter on right side of 
foot in females deep, running down side of 
foot and extending through lateral epipodial 
groove (Fig. 2). No ovipositor present. Mantle 
edge dorsally fringed with many small papil- 
lae. 

Palliai Cavity: Osphradium a little less long 
than ctenidium, broad, about one-third ctenid- 
ial width, dark brown, weakly bipectinate with 
small pectins on each side but unconnected 
dorsally; osphradium becoming monopecti- 
nate at inhalant siphon. Ctenidium narrow, 
extending length of palliai cavity, comprising 



GENERIC REVIEW OF BITTIINAE 



285 




FIG. 8. SEM micrographs of Ittibittium parcum Uom Honolulu, Hawaii (USNM 857100). A, B, apertural and 
lateral views of shell, 3.6 mm length; C-E, apertural, lateral and dorsal views of shell, 3.6 mm length; F, 
newly hatched larval shell showing protoconch and details of whorl sculpture, bar = 63 у.т; G, H, embryonic 
shells removed from eg capsule, bar = 23 (xm; I, larval and early whorls of shell, bar = 0.4 mm; J, shell with 
strong axial ribs, 5.3 mm length; K, operculum, bar = 0.2 mm; L, detail of penultimate and body whorl, 
showing details of sculpture and aperture, bar = 0.6 mm; M, apertural view of shell, 3.6 mm length. 



HOUBRICK 




FIG. 9. SEM micrographs of radula of Ittibittium parcum Uom Honolulu, Hawaii (USNM 857100). A, middle 
of radular ribbon with right marginal teeth folded back, bar = 30 |xm; B, detail of rachidian and lateral teeth, 
bar = 8 p,m. 



long, finger-like, triangular filaments. Hypo- 
branchial gland partially overlaying rectum, 
well developed, composed of several large, 
dark-staining glandular cells. 

Reno-pericardial System: Pericardium lying 
adjacent to posterior pallia! wall. Kidney large, 
extending from anterior of style sac forward, 
into roof of posterior palliai cavity. 

Alimentary System: Snout tip and lips of 
mouth yellow. Buccal mass large, about two- 
thirds snout length. Radula (Fig. 9A) short, 
about one-tenth shell length. Rachidian tooth 
having weak hour-glass shape and cutting 
edge with large central cusp flanked by 2 den- 
ticles on each side. Lateral tooth (Fig. 9B) 
having cutting edge with large pointed cusp, 
one inner denticle, 3-4 outer denticles. Inner 
marginal tooth with 2 inner denticles, large 
elongate major cusp and 3 outer denticles; 
outer marginal tooth with 5 inner denticles. 
Salivary glands paired, comprising tangled 
mass behind nerve ring, extending through it 
anteriorly as slender tubes. Esophagus be- 
coming wide behind nerve ring, developing 
lateral glandular pouches with many small 
transverse internal folds, comprising short 
esophageal gland. Stomach large, about one 
whorl in length, having single opening to di- 
gestive gland, central raised pad, gastric 
shield, short crystalline style and style sac. 



about two-thirds the stomach length. Intestine 
leaving stomach looping dorsally and across 
anterior style sac, turning sharply, running an- 
teriorly, adjacent to right side of kidney and 
albumen gland. Rectum slightly wavy, wide, 
containing large ovoid fecal pellets. 

Nervous System: Cerebral ganglia very 
large, twice size of pleural ganglia. Sube- 
sophageal ganglion very close to left pleural 
ganglion. Supraesophageal ganglion sepa- 
rated from right pleural ganglion by connec- 
tive two-thirds ganglion length. 

Reproductive System: Testis white, overlay- 
ing brown digestive gland. Males aphallate 
with open palliai gonoducts. Palliai oviduct 
open, with large albumen gland extending 
through posterior of mantle cavity mantle cav- 
ity, protruding into visceral coil. Albumen 
gland staining cream-green in toluidine blue. 
Capsule gland very large, swollen, staining 
dark blue in toluidine blue. Large spermato- 
phore bursa in posterior medial lamina. No 
ciliated ridge tract or seminal receptacle in lat- 
eral lamina. Spawn mass comprising wide ge- 
latinous tube covered with thin membrane 
forming compact, short tube about 2 mm long, 
and 1.2 mm wide, containing large opaque, 
compacted eggs each 0.2 mm in diameter. 
Eggs arranged in short jelly tube about 3-4 



GENERIC REVIEW OF BITTIINAE 



287 



deep. Development direct with young snails 
hatching from eggs. 

Discussion 

"Bittium" parcum has not been cited com- 
monly in the literature, and due to great inter- 
specific variability in shell sculpture and color, 
is frequently misclassified or unidentified in 
museum collections. Shell shape can vary 
from slender, elongate (Fig. 8J) to shorter, 
more inflated (Fig. BC-E), and shell sculpture 
is highly variable: the axial ribs seen in some 
specimens may be entirely lacking in others. 
The protoconch with its flattened apex, broad 
sutural ramp and concave whorls is highly 
distinctive and unusual (Fig. 8F-H). However, 
Ittibittium parcum is readily distinguished from 
by several external anatomical features: (1) 
the epipodial skirt and opercular lobe are 
fringed with well-developed papillae; (2) a pair 
of long epithelial extensions (papillae) of the 
front of the foot (propodium); (3) the longitu- 
dinal slit marking the entrance to the metapo- 
dial mucus gland is very long. Ittibittium par- 
cum has an unusual palliai oviduct in that the 
albumen gland projects posteriorly past the 
posterior end of the mantle cavity into the vis- 
ceral coil, and there is no seminal receptacle 
in the lateral lamina of the palliai oviduct. 

Living snails are quick, active crawlers, and 
even when removed from their shells showed 
a great deal of movement. 

The operculum in this species tends to be 
more ovate than circular: in most other spe- 
cies of the Bittium -дюир, the operculum is cir- 
cular. The opercular lobe papillae show 
through the transparent edges of the opercu- 
lum. 

This species undergoes direct develop- 
ment. The embryos pass through a veliger 
stage and hatch out as juvenile snails after 
losing the velar lobes. Direct development, 
while also occurring in Stylidium, is not the 
common mode of development among mem- 
bers of the Bittium-group. The comparatively 
large eggs of Ittibittium parcum are each en- 
closed within individual hyaline capsules 
about 0.2 mm diameter, and the egg capsules 
are stacked within a short, wide gelatinous 
tube and deposited on the substrate in an ir- 
regular mass. Here they undergo develop- 
ment, passing through a modified veliger 
stage and producing a well-developed embry- 
onic shell (Fig. 8F-H), after which they 
emerge as small snails. 

Ittibittium parcum is common in shallow wa- 



ter throughout the Hawaiian chain, and also 
occurs in French Polynesia (Naim, 1982) 
where it is very abundant in some localities. 
Naim (1982) found that this species repre- 
sented 89% of the molluscan fauna associ- 
ated with algae in Tiahura Lagoon in French 
Polynesia. 

A species from Western Australia, very 
similar to the type species, recently has been 
described in great detail (Ponder, in press), 
and appears to be closely related to Ittibittium 
parcum. 



BITTIOLUM COSSMANN, 1906 

Bittiolum Cossmann, 1906: 139. (Type spe- 
cies by original designation: Bittium pod- 
agrinum Dall, 1892). Wenz, 1940: 755; 
Olsson & Harbison, 1953: 289-290. 

Diagnosis 

Shell small, turreted, stout, sculptured with 
4 spiral cords and many axial ribs, and occa- 
sional weak varices. Protoconch with one spi- 
ral lira. Whorls presuturally constricted, body 
whorl elongate, narrow at aperture and con- 
stricted at base, having less width than pen- 
ultimate whorl. Operculum ovoid-circular, 
paucispiral and with subcentral nucleus. An- 
terior canal weakly defined, short. Mantle 
edge smooth, epipodial skirt scalloped. Foot 
elongated anteriorly and having median lon- 
gitudinal slit in posterior part of sole, leading 
into large metapodial mucus gland. Ovipositor 
small. Osphradium bipectinate, wide, one- 
third ctenidial length. Nervous system with 
right zygoneury and with short supraesoph- 
ageal connective. 

Remarks 

Bittiolum species have small shells (Table 
3) and are distinctive in having the body whorl 
elongated and constricted basally so that the 
aperture width is less than that of the penul- 
timate whorl. The smooth mantle edge, nar- 
row elongate anterior foot, right zygoneury 
and short supraesophageal connective are 
autoapomorphous characters of this genus. 

The type species of this genus is a Neo- 
gene fossil from Florida that has a shell mor- 
phology very similar to that of living Bittiolum 
varium and Bittiolum alternatum. As the fossil 
species occurs in mid- to late-Neogene strata, 
and in the same geographic area as Recent 



288 



HOUBRICK 



Bittiolum varium, it is not unreasonable to in- 
fer that the two species belong to the same 
clade, and the living species is considered to 
be congeneric with Bittium podagrinum. 
Cossrnann (1906: 140) pointed out that Bitti- 
olum varium (Pfeiffer) (cited as Cerithium) oc- 
curred from the Pleistocene of Florida and ex- 
tended into the Recent. He further noted the 
superficial resemblance of Bittiolum varium to 
some fossils of Aneurychilus Cossmann, 
1889, which he placed in the Diastomatidae 
(as Diastomidae, Cossmann, 1906: 174). 

Dall (1889) was the first author to confuse 
American members of Bittiolum with the ge- 
nus Diastoma Deshayes, 1850, when he re- 
ferred Bittiolum varium to that genus. Abbott 
(1974), probably following this cue, later re- 
ferred western Atlantic species of Bittium, s.l., 
to Diastoma Deshayes, 1850, but this subse- 
quently has been shown to be incorrect 
(Houbrick, 1977: 102, 1981b), as the latter 
genus belongs to the Diastomatidae Coss- 
mann, 1894, a totally different lineage repre- 
sented by individuals of much larger size and 
different anatomy that are not closely related 
to the Bittium-group (Houbrick, 1981b). 

The anatomy of "Bittium" alternatum, from 
the northeastern coast of North America, is 
identical to that of its southeastern, Carib- 
bean Province congener, Bittiolum varium. 
Thus, these two species and probably all 
other American western Atlantic species be- 
long in the genus Bittiolum, which is also rep- 
resented by several eastern Pacific species, 
such as Bittiolum fastigiatum (Carpenter, 
1864). 

Because the two Bittiolum representatives 
studied, B. varium and S. alternatum, are so 
alike, they are treated jointly in the section 
below. 

Bittiolum varium (Pfeiffer, 1840) 
(Figs. 10-11) 

Cerithium varium Pfeiffer, 1 840: 256. 

Cerithium columellare Orbigny, 1842: pi. 23, 
figs. 13-15; 1845: 244 (in part; syntypes 
BMNH). 

Cerithium gibberulum С В. Adams, 1845: 5 
(Lectotype MCZ 186078, type locality Ja- 
maica). 

Bittium varium (Pfeiffer). Tryon, 1887: 152, pi. 
29, fig. 86; Perry, 1940: 134, pi. 28, fig. 
202. 

Cerithium (Bittium) gibberulum (C. B. Ad- 
ams). Kobelt, 1898: 245-246, pi. 43, 
fig. 1. 



Diastoma varium (Pfeiffer). Abbott, 1974: 
107, fig. 1037. 

Description 

Shell (Fig. 10): Shell turreted, pendent- 
shaped, comprising about 10 flat-sided 
whorls and reaching 7 mm length. Protoconch 
(Fig. 101) comprising 2.5 whorls; protoconch 1 
smooth, protoconch 2 with central keel-like 
spiral lira and microscopic pustules on abapi- 
cal part of whorl. Early whorls (Fig. 10H) with 
two weak spiral lirae, and sculptured with 
dominant suprasutural spiral cord and two 
weaker spiral cords above it, and with weak 
axial ribs. Adult whorls sculptured with 4 spiral 
cords and 14 strong axial ribs forming small 
beads at crossover points and producing can- 
cellate pattern. Body whorl elongate, more 
than one-third shell length, constricted at ap- 
erture and more at siphon; body whorl sculp- 
tured with about 10 flattened spiral cords and 
14 weak axial ribs. Aperture ovate, con- 
stricted, not as wide as width of body whorl, 
narrowing posteriorly and having short, dis- 
tinct siphonal canal. Columella concave with 
slight callus. Outer lip of aperture smooth, 
rounded, thin and pendant, extending beyond 
siphonal canal. Periostracum thin, tan. 

Animal: Snout, cephalic tentacles, and neck 
slender, extremely long and extensible. Snout 
bilobed at tip. Foot narrow, extremely elon- 
gate anteriorly, three times snout length, and 
with crescent-shaped propodium (Fig. 2). 
Deep crescentic transverse slit formed by two 
lips in anterior foot and leading via a central 
duct into large anterior mucus gland (Fig. 
IIA, amg). Corners of anterior pedal lips ex- 
tending laterally and posteriorly forming uncil- 
iated undulating epipodial skirt (Fig. 11A-B, 
es) delineating lateral groove between epipo- 
dium and sole; epipodial skirt weakly scal- 
loped posteriorly (Fig. 2), forming lanceolate 
opercular lobe, scalloped around edges. Cili- 
ated gutter (Fig. IIB, eg) in both sexes 
emerging from floor of right side of palliai cav- 
ity, running down right side of foot leading into 
epipodial groove. Ciliated gutter terminating 
in small glandular ovipositor (Fig. 1 1 B, ovp) at 
edge of foot in females. Posterior third of sole 
with median longitudinal slit leading into mas- 
sive mesopodial mucus gland (Fig. 11 A, 
mmg), extending deeply into head foot up to 
nerve ring and cephalic hemocoel. Opercu- 
lum (Fig. 10F, G) corneous, light tan, circular- 
ovate, paucispiral with subcentric nucleus. 
Mantle edge (Fig. IIB, me) bilobed, smooth. 



GENERIC REVIEW OF BITTIINAE 



289 




FIG. 10. SEM micrographs of Bittiolum varium from Ft. Pierce, Florida (USNM 77639). A, B, D, E, two shells 
showing sculptural variation and shell shape; length 3.2 mm; C, immature shell, length 2.8 mm; F, G, 
operculum, bar = 0.2 mm; H, sculpture of early whorls, bar = 0.3 mm; I, protoconch, bar = 88 |xm. 



without papillae, slightly scalloped, Iridescent 
at edges. 

Palliai Cavity: Osphradium wide, one-third 
ctenidial length, weakly monopectinate, com- 
prising small, dorsally placed pectins, flanked 
on each side by weak ciliated strip. Ctenidium 
comprising long, triangular filaments with soft 
rods and mucus glands. 

Alimentary System: Radula (Fig. 1 1 C) short. 
Rachidian tooth (Fig. 1 1 D) with cutting edge 
of 3 small denticles on each side of central 
cusp. Lateral tooth (Fig. 1 1 D) with two outer 
and 3-4 inner denticles. Inner marginal tooth 
with 3-4 inner and 2-3 outer denticles. Outer 



marginal tooth with 6 small inner denticles. 
Midesophagus with wide ciliated dorsal food 
groove; posterior esophagus narrow. 

Nervous System: Cerebral ganglia slightly 
larger than pedal ganglia and with short con- 
nective (about one-third cerebral ganglion 
length). Pedal ganglia nearly fused at connec- 
tive, each with posterior statocyst; two pairs of 
accessory pedal ganglia present: pair of small 
propodial ganglia, and larger pair of metapo- 
dial ganglia. Subesophageal connective be- 
tween subesophageal ganglion and left pleu- 
ral ganglion equal in length to left pleural 
ganglion; supraesophageal connective about 
equal in length to subesophageal connective. 



290 



HOUBRICK 




FIG. 11. SEM micrographs of Bittiolum varium from Ft. Pierce, Florida (USNM 776639). A, B, critical point 
dried specimens showing external anatomical features of headfoot, bar = 0.2 mm; C, mid-section of radula, 
bar = 21 ixm; D, detail of rachidian and lateral teeth, bar = 7 ^JLm. amg = anterior mucus gland; eg = 
ciliated groove; eps = epipodial skirt; I = lip of mouth; mmg = metapodial mucus gland; op = operculum; 
ovp = ovipositor. 



Right zygoneury between subesophageal 
and right pleural ganglion. 

Reproductive System: Ducts of testicular fol- 
licles joining to form spermatic duct, moving 



anterior as seminal vesicle, containing dimor- 
phic sperm. Males producing crescent- 
shaped spermatophore with flared bifurcate 
end and pointed, filamentous tip. Spermato- 
phores containing both eu- and parasperma- 



GENERIC REVIEW OF BITTIINAE 



291 



tozoa. Ovary cream colored, overlying brown 
digestive gland, extending forward to stom- 
ach. Palliai oviduct open, but closed in far 
posterior portion. Common aperture to open- 
ing of spermatophore bursa in lateral lamina 
anterior to opening of sperm pouch and open- 
ing of seminal receptacle located on edge of 
medial lamina one-third from posterior of lam- 
ina. Opening to spermatophore bursa not ad- 
jacent to opening on medial lamina, but lo- 
cated one-third back from anterior of lateral 
lamina. Spermatophore bursa comprising cil- 
iated and high vacuolated epithelial cells. 
Spawn mass composed of spirally wound thin 
jelly string containing many small eggs 100- 
120 jxm In diameter, hatching as vellger lar- 
vae, becoming planktotrophic. 

Bittiolum alternatum (Say, 1822) 

Turritella altérnala Say, 1822: 243. 
Pasithea nigra Totten, 1834: 369, figs. 7a, b. 
Bittium nigrum (Totten), Gould, 1870: 321 , fig. 

590. 
Bittium alternatum (Say), С W. Johnson, 

1915: 127. 
Diastema alternata (Say), Abbott, 1974: 107, 

fig. 1037. 

Description 

This species Is essentially the same as 
Bittiolum varium, described above, although 
the shell differs slightly in being more pupoid 
and less narrowly elongate. 

Remarks 

Marcus & Marcus (1963) thoroughly de- 
scribed the anatomy of Bittiolum varium in 
Brazil. My work on populations of this species 
from Florida basically confirms their detailed 
observations. In addition, the basic anatomy 
of the Brazilian and Florida specimens is very 
similar to that of Bittiolum alternatum from the 
American northeastern coast, suggesting that 
the latter is probably a sister taxon of Bittiolum 
varium. 

Bittiolum is the only genus studied in which 
the mantle edge is smooth, with no trace of 
papillae, a character noted by Marcus & Mar- 
cus (1963). A wavy epipodlal skirt and nar- 
rowly elongate anterior foot are also distinc- 
tive external features (Fig. 2) of both 
examined Bittiolum species. The ovipositor 



(Fig. 1 1 B, ovp) is barely visible only during the 
breeding season, but is basically the same as 
that observed in Bittium. The massive 
metapodial mucus gland located in the pos- 
terior part of the sole differs from that seen in 
Ittibittium species, in which the slit is much 
longer. This gland secretes a string of mucus 
by which the animal can suspend itself In the 
algae, but the thread does not have the ten- 
sile strength of the mucous threads produced 
by members of the Litiopidae (Houbrick, 
1987b). Except for major differences in exter- 
nal features, the radula and internal anatomy 
of Bittiolum varium is quite similar to that of 
Bittium reticulatum. The radula differs only mi- 
nor details (Table 2). Although Bittiolum var- 
ium primarily is a grazer of epiphytic microal- 
gae, Marcus & Marcus (1963: 79) have 
shown that the snail can use its anterior 
ctenidial filaments for particle feeding while 
stationary. 

Marcus & Marcus (1963: 88-89) found four 
spindle-shaped spermatophores, each 1 mm 
long and 0.06 mm wide, in the bursa of the 
lateral lamina in Bittiolum varium, and noted 
that the spermatophores dissolve in this 
bursa. The location of the spermatophore 
bursa in the lateral lamina is a unique feature 
among cerithioidean taxa, and this layout is 
probably the same among other members of 
the Bittium-group, in which the bursa in the 
lateral lamina has been confirmed. However, 
spermatophores have not been observed in 
this bursa in any other species. 

Bittiolum varium lays its eggs mostly on 
seagrasses. In the Indian River, Florida, I ob- 
served numerous irregular egg masses com- 
prising strands of eggs embedded in a loose 
jelly matrix deposited on Halodule grass 
blades and on ramose algae. In the spring, 
nearly all adults were ripe and egg laying con- 
tinued through the summer months tapering 
off in September. 

Bittiolum varium has been the subject of a 
number of ecological investigations. Virnstein 
& Curran (1986) measured the colonization 
time of this species in seagrasses in the In- 
dian River, Florida. Hardison & Kitting (1985) 
found that Bittiolum varium fed primarily on 
diatoms and coralline algae in seagrass 
meadows of the northwest Gulf of Mexico. 
Despite the high population densities of this 
snail (3,000/m^), little impact on its food could 
be detected. In Chesapeake Bay, Van Mont- 
frans et al. (1982) found that the grazing ac- 
tivities of Bittiolum varium, which selectively 
eats diatoms from blades of manne grasses. 



292 



HOUBRICK 



could have important implications for the 
abundance and distribution of Zosters. 

Bittiolum varium has a wide range in the 
western Atlantic, occurring from Chespa- 
peake Bay south to Florida and the Gulf of 
Mexico, throughout the Caribbean, and south 
to Brazil. 

STYUDIUM DALL, 1907 

Stylidium Dall, 1907: 178 (Type species by 
original designation: Bittium eschrichtii 
Middendorf, 1849). Thiele, 1929: 211; 
Wenz, 1940: 757; Abbott, 1974: 106. 

Diagnosis 

Shell relatively large, dirty chalky white, 
smooth, weakly sculptured with four broad 
spiral cords defined by incised lines. Proto- 
conch unsculptured. Snout twice length of 
cephalic tentacles. Epipodial skirt poorly de- 
veloped, smooth along edges, but opercular 
lobe with small, pointed papillae. No metapo- 
dial mucus gland. Osphradium non-pectinate. 
Common aperture to sperm bursa and semi- 
nal receptacle in edge of anterior third of me- 
dial lamina of palliai oviduct. Openings to 
sperm bursa and seminal receptacle well- 
separated. Long ciliated ridge tract in lateral 
lamina of palliai oviduct. Development direct. 

Remarks 

This genus is represented by species living 
in cold-water habitats from California north to 
Alaska. The shell is dull and chalky under the 
periostracum. Shell length can be quite large 
(Table 3) for a member of the Bittiinae, and 
the large smooth protoconch, without sinusig- 
eral notch, is indicative of direct development. 

At first glance, the shell of Stylidium does 
not appear to fit the Bittium-group mold. How- 
ever, anatomical features, such as the epipo- 
dial skirt, large opercular lobe (Fig. 2) and pal- 
liai gonoduct configuration unmistakably 
place it into the Bittiinae. The common aper- 
ture to sperm pouch and seminal receptacle 
is unusual in being located in the far anterior 
edge of the medial lamina of the palial ovi- 
duct, and not adjacent to the opening of the 
spermatophore bursa of the lateral lamina. 
The length of the ciliated ridge tract of the 
lateral lamina is also atypical. 

Stylidium eschrichtii (Middendorff, 1849) 
(Figs. 12-14) 

Turritella eschrichtii Middendorf, 1849: 396- 
397, pi. 11, fig. 1 (Holotype, Zoological 



Institute, St. Petersburg; type locality, 

Sitka, Alaska). 
Bittium (Stylidium) eschrichtii icelum Bartsch, 

1907: 178 (Holotype USNM 15209a; type 

locality, Neah Bay, Washington); 1911: 

388, pi. 57, fig. 3; Ruhoff, 1973: 81. 
Bittium eschrichtii (Middendorf). Oldroyd, 

1927: 18-19, pi. 79, fig. 4. 
Bittium (Stylidium) eschrichtii (Middendorf). 

Abbott, 1974: 106, fig. 1010. 

Description 

Shell (Fig. 12): Shell large, turreted, reaching 
17.5 mm in length, comprising 9-11 convex 
whorls. Protoconch (Fig. 12G) has two 
smooth whorls. Early whorls (Fig. 12E-G) 
sculptured with three spiral bands. Adult 
whorls sculptured with 4 weak, widely flat- 
tened spiral bands separated from one an- 
other by deep incised spiral grooves. Penul- 
timate whorls with 5 wide, spiral, weak bands. 
Suture well defined, slightly counter-sunk into 
each abapical whorl. Body whorl a little less 
than one-third shell length, sculptured with 
about 8 broad spiral cords and incised lines. 
Shell base weakly constricted at base; ante- 
rior siphon broad and shallow. Aperture ovate 
having concave columella with weak callus; 
outer lip of aperture circular, crimped where 
spiral grooves end. Shell color chalky white- 
gray, covered by thin tan periostracum. 

Animal: Base color dirty white with trans- 
verse black stripes on snout, head, and epi- 
podium (Fig. 14A). Ciliated epithelial strip run- 
ning from mantle cavity floor on each side of 
headfoot and ending beneath peduncle of 
each cephalic tentacle. Ciliated gutter on right 
side of foot in females ending in small pink, 
glandular ovipositor at foot edge. Snout very 
long, twice length of cephalic tentacles, wide, 
bilobed at tip. Eyes very small. Lateral epipo- 
dial skirt with minute pointed papillae along 
edge of posterior third of foot; opercular lobe 
long, pointed posteriorly, darkly pigmented 
and with small pointed papillae along edge 
(Fig. 2). Anterior foot crescent-shaped with 
long slit along edge leading into centrally 
placed, ovate mucus gland deep within propo- 
dium. No metapodial mucus gland. Opercu- 
lum (Fig. 12H, I) thick, ovate, paucispiral, with 
eccentric nucleus. Mantle edge bilobed, with 
small papillae, and with slightly elongate ex- 
halant siphon. Mantle roof folded longitudi- 
nally over exhalant siphon forming dorsal, 
posteriorly extending ridge. 



GENERIC REVIEW OF BITTIINAE 



293 




FIG. 12. Stylidium eschrichtii Uom Carmel, California. A-D, two shells showing sculptural variation (USNM 
804376), 22.4 and 20.2 mm length, respectively; E, F, SEM micrographs of immature shells showing early 
sculptural patterns, bar = 0.5 mm; G, SEM micrograph of protoconch and early whorls, bar = 0.3 mm; H, 
I, SEM micrographs of operculum, showing eccentric nucleus and attachment scar, 2.4 mm length. 



Palliai Cavity: Osphradium tan, vermiform, 
non-pectinate, extending length of palliai cav- 
ity, but slightly shorter than ctenidium. Ctenid- 
ium pink, comprising long, finger-like fila- 
ments twice length of their attached bases. 



Alimentary System: Radular ribbon (Fig. 
13A) short. Lateral tooth (Fig. 13B) with long 
lateral basal extension and cutting edge with 



3 inner denticles, and 3-5 outer denticles; in- 
ner marginal tooth with 4-5 inner and 3 outer 
denticles. Paired salivary glands vermiform, 
loosely compacted, lying mostly anterior to 
nerve ring, but beginning behind it as thick 
swellings, and passing through as thin tubes. 
Stomach large, about one whorl in length; in- 
ternally with large sorting area and roundish 
central pad; single opening to digestive gland 
on right side of pad; 6-7 large transverse ribs 



294 



HOUBRICK 




FIG. 1 3. SEM micrographs of radula of Stylidium eschrichtii (USNM 804376); A, section of mid-radular ribbon 
with marginal teeth folded back, bar = 38 ixm; B, detail of rachidian and lateral teeth, bar = 12 |xm. 



on left side of pad, posterior to cuticular gas- 
tric shield; short, wide style sac one-half stonn- 
ach length, separate fronn intestinal opening. 
Intestine opening separated from lumen of 
style sac by typhlosole ridge. 

Nervous System (Fig. 14): Nerve ring large 
with thick commissure connecting cerebral 
ganglia. Dialyneury (Fig. 14B, d) between left 
pallia! nerve and nerve arising from supra- 
esophageal ganglion. Supraesophageal con- 
nective (Fig. 14A, sec) twice length of right 
pleural ganglion. Subesophageal ganglion 
(Fig. 14A, sbe) closely adjacent to left pleural 
ganglion. 

Reproductive System Posterior half of palliai 
oviduct with thick, white, opaque albumen 
gland comprising flocculant transverse glan- 
dular ridges; mid-section of palliai oviduct with 
thin, weak glandular transparent walls; very 
thick, opaque transverse glandular ridges 
present in anterior third of palliai oviduct, 
comprising capsule gland. Sperm gutter in 
anterior edge of medial lamina having elon- 
gate common aperture to spermatophore 
bursa and seminal receptacle. Openings to 
sperm pouch and seminal receptacle within 
common aperture well separated. Long tube 
within edge of medial lamina leading to pos- 
teriorly placed pouch-like seminal receptacle. 
Large sperm pouch with internal transverse 
epithelial folds, occupying posterior third of 



medial lamina. Very long ciliated ridge tract 
beginning in anterior part of lateral lamina, 
leading into posterior spermatophore bursa. 
Spawn comprising thin gelatinous string 
wound into irregular mass. Eggs 0.2 mm in 
diameter. Development direct. 

Remarks 

Several subspecific taxa have been de- 
scribed, but it is debatable if all of these nom- 
inal taxa are good subspecies or merely cli- 
nal/ecophenotypic varieties of Stylidium 
eschrichtii. Abbott (1974) synonymized the 
subspecies icelum Bartsch with S. eschrichtii. 

Stylidium eschrichtii is characterized by its 
chalky gray, smooth shell sculptured with 
broad flattened spiral cords. The protoconch 
is large, unsculptured, and lacks a sinusigeral 
notch (Fig. 12G). The ovate operculum (Fig. 
12H, I) with eccentric nucleus is a departure 
from a more circular operculum with subcen- 
tral nucleus, as seen in other bittiid species. 
Shell length seems to vary greatly among 
populations, but some individuals can be very 
large, approaching 18 mm length (Table 3). 
Large shell size appears to be more common 
in northern populations. 

This species lives on intertidal to subtidal 
rubble in cool waters of the northeastern Pa- 
cific. I observed a large intertidal population 
living among the intertices of gravel and algae 



GENERIC REVIEW OF BITTIINAE 



295 




В 



ses she Isg 



sec 




FIG. 14. Anatomical features of Stylidium eschrich- 
tii. A, head and anterior foot, showing pigment pat- 
tern; B, position of salivary glands relative to nerve 
ring, d = left dialaneury; leg = left cerebral gan- 
glion; Ipg = left pleural ganglion; Isg = left salivary 
gland; rcg = right cerebral ganglion; rsg = right 
salivary gland; rpg = right pleural ganglion; sbe = 
subesophageal ganglion; sec = supraesophageal 
connective; seg = supraesophageal ganglion. 



at Carmel, California. According to Strath- 
mann (1987), Stylidium eschrichtii has direct 
development. Spawn is deposited on the sub- 
strate in gelatinous masses (presumably 
comprising coiled strings) containing egg cap- 
sules measuring 0.2 ixm diameter in which 
the embryos undergo direct development, 
passing through the veliger stage and hatch- 
ing as small snails. 

UROBITTIUM BARTSCH, 191 1 

Lirobittium Bartsch, 191 1 : 384 (Type species 
by original designation, Bittium catalinen- 
sis Bartsch, 1907). Thiele, 1929: 211; 
Wenz, 1940: 757; Abbott, 1974: 106; 
Gründe!, 1976: 54. 



Diagnosis 

Shell turreted, elongate, sculptured with ax- 
ial riblets and spiral beaded cords. Proto- 
conch with two spiral lirae. Varices not 
present on adult whorls. Operculum circular. 
Radular ribbon very small; radular teeth with 
many small denticles. Snout long; head with 
small cephalic tentacles and small eyes. Ovi- 
positor and ciliated groove on right side of foot 
absent. Mantle edge with long papillae. Epi- 
podial skirt very weakly developed. Osphra- 
dium vermiform, wide. Spawn comprising 
large egg capsules, each attached to long 
stalk and anchored together. Development di- 
rect. 

Remarks 

Bartsch (1911) divided Bittium-group spe- 
cies from the American west coast into four 
genera: Bittium, Lirobittium, Semibittium, and 
Stylidium. His groups were defined only on 
superficial shell characters, such as the pres- 
ence or absence of varices, protoconch 
sculpture, and axial and spiral sculpture. 
Many of the species Bartsch (1911) included 
under his generic scheme have been ignored 
or referred by subsequent authors to different 
generic taxa. 

The genus Lirobittium Bartsch, from the 
temperate eastern Pacific, was based on mi- 
nor shell sculptural characters: Bartsch 
(1 91 1 : 384) noted that the defining characters 
of Lirobittium were a protoconch with two spi- 
ral lirae and the absence of varices from the 
adult whorls. These features were also men- 
tioned by Gründel (1976: 54), who addition- 
ally noted that of the two primary spiral cords, 
the abapical one was inserted a little later. 
Gründel (1976: 54-56) assigned Cacozeli- 
ana and Stylidium (with a query) as subgen- 
era of Lirobittium. He indicated that Cacoze- 
liana differed from Lirobittium by the formation 
of varices, and Stylidium by the suppression 
or complete absence of axial ribs. It has been 
shown herein that the Cacozeliana is sepa- 
rated from Lirobittium by many significant 
characters. 

The above history of Lirobittium shows that 
much of the confusion regarding the place- 
ment of the numerous California species 
stems from the original superficial generic de- 
scriptions based solely on shell morphology. 
It is obvious that the characters derived by 
these authors from minor sculptural details 
hardly seem to be of generic weight and have 



296 



HOUBRICK 



resulted in poorly defined, ambiguous genera 
with broad or discordant limits, and that have 
been used in varying combinations. Although 
shell sculpture may have some value at the 
specific level, it is generally not useful at the 
generic level, especially in cerithiids. Not a 
single author has included radular or opercu- 
lar characters and no mention is made of an- 
atomical features in the definition of genera. 
Abbott (1974: 106) considered both BiWum 
catalinense and B. subplanatum to be syn- 
onyms of Lirobittium afíenuatum Bartsch, 
1911, but gave no reasons for this decision. 
Hertz (1 981 : 40) showed that Lirobittium sub- 
planatum (cited as Bittium) was a valid spe- 
cies. I have examined two species of Lirobit- 
tium: L. catalinense. {one dried specimen) and 
well-preserved material of L. subplanatum. 
Observations on the poorly preserved, dried 
animal of L. catalinense are included because 
it is the type species of the genus, but the bulk 
of the descriptive anatomical characters of Li- 
robittium are derived from study of L. sub- 
planatum. The two species are anatomically 
very similar, have similar radulae, and are un- 
doubtedly congeneric. The above diagnosis 
and following specific descriptions represent 
an integrated analysis of generic characters, 
based on these two species. 

Lirobittium catalinense Bartsch, 1907 

Bittium catalinensis Bartsch, 1907: 28, pi. 57, 
fig. 13 (Holotype: USNM 165232, type lo- 
cality: Santa Barbara, California [Pleis- 
tocene]); Abbott, 1974: 106, fig. 1013. 

Bittium (Lirobittium) catalinense Bartsch, 
1911:402-403, pi. 51, fig. 1. 

Remarks 

The type species of this genus is a Pleis- 
tocene fossil, but Bartsch (1911) described 
many subspecies, some of which are Recent. 
Bittium cataliense is now regarded as a syn- 
onym of "Bittium" attenuatum Carpenter, 1864 
(Abbott, 1974: 106). 

Examination of a reconstituted, dried spec- 
imen of the type species of Lirobittium, Bittium 
catalinense (= Bittium attenuatum), showed 
that the animal is basically the same as Liro- 
bittium subplanatum. It is relatively unpig- 
mented, has a large, broad snout, bilobed at 
the anterior end and short cephalic tentacles, 
about half the snout length. The mantle edge 
has many long papillae along its dorsal and 
lateral sides, while the mantle edge forming 



the inhalant siphon has large paddle-shaped 
papillae. The buccal mass is small, and the 
radula minute, about one-thirteenth the shell 
length. The rachidian tooth has a triangular 
basal plate with a long glabrella and is as 
wide as tall; there is a deep concave inden- 
tation and a cutting edge with a long pointed 
central cusp flanked on each side by 4-5 
small denticles. The lateral teeth are deeply 
concave on the top, have a wide basal plate 
with a large central buttress, and have numer- 
ous small denticles. The marginal teeth are 
slender, and serrated along their tips with 
many small pointed denticles (Fig. 15). 

Lirobittium subplanatum (Bartsch, 1911) 
(Figs. 15-17) 

Bittium (Semibittium) subplanatum Bartsch, 
1911: 395-396, pi. 57, fig. 5 (Holotype, 
USNM 160076; type locality, Catalina Id., 
California); Oldroyd, 1927: 23: Ruhoff, 
1973: 130. 

Bittium subplanatum Bartsch. Dali, 1921 : 146; 
Hertz, 1981:40, figs. 23-27. 

Bittium subplanatum Bartsch. Oldroyd, 1927: 
23. 

Bittium (Lirobittium) subplanatum (Batsch). 
Abbott, 1974: 106. 

Description 

Stielt (Fig. 15): Shell elongate, turreted, com- 
prising 8-9 moderately inflated whorls. Pro- 
toconch (Fig. 15) about 1.5 whorls, well 
rounded, smooth. Early whorls sculptured 
with two major spiral lirae, soon crossing over 
axial riblets (Fig. 15). Adult whorls sculptured 
with three major spiral cords crossed over by 
numerous thin axial ribs (24-26), forming 
cancellate appearance; small beads occur- 
ring at crossover points. Body whorl (Fig. 15) 
sculptured with four major spiral cords and 
numerous axial ribs; moderately constricted 
at base. Shell base with about 7 spiral cords. 
Aperture ovate with oblique columella and 
curved, thin outer lip. Anterior canal moder- 
ately developed; anal canal weak. Shell color 
white, covered with brown periostracum. 

Animal (Fig. 16A, B): Animal pure white with 
pink buccal mass showing through snout. 
Head large with very large, wide, extensible 
snout, dorso-ventrally flattened, bilobed at tip; 
cephalic tentacles small, a little less than one- 
third snout length, each with small black eye 
adjacent to opaque white spot at tentacular 
peduncular base. Snout ringed with many 



GENERIC REVIEW OF BITTIINAE 



297 





FIG. 15. SEM micrographs of shells of Lirobittium 
subplanatum from Palos Verdes, California (USNM 
881021). A, bar = 1.8 mm; B, detail of protoconch 
and early teleoconch sculpture, bar = 0.6 mm; C, 
bar = 1.8 mm. 

deep, transverse epithelial folds (Fig. 16B). 
Foot with very weak epipodial skirt and with- 
out papillae or distinctive operculiferous lobe. 
No ciliated groove on right side of foot; no 
ovipositor. Anterior of sole crescent shaped 
with deep transverse slit marking entrance to 
anterior mucus gland. No metapodial mucus 
gland. Mantle edge bilobed, fringed with 
many papillae emerging from ventral side of 
mantle edge. 

Palliai Cavity: Osphradium brown, vermi- 
form, without pectins, wide, about one-third 
the ctenidial width, nearly equaling ctenidial 
length. Ctenidium extending length of palliai 
cavity. Hypobranchial gland thick, comprising 
transversely ridged glandular tissue. 

Alimentary System: Mouth at tip of snout, de- 
fined by pair of fleshy pads. Buccal mass (Fig. 
16B, bm) pink, small, about one-third snout 
length. 

Radular ribbon (Fig. 17) small, about 
one-ninth shell length. Rachidian tooth (Fig 
17C) with large glabrella, long serrated cen- 
tral cusp and 6 small denticles on each side. 
Lateral tooth (Fig. 17 B,C) with broad basal 
plate; cutting edge has large denticle with 6 
inner denticles and 15-17 outer denticles. 
Marginal teeth (Fig. 17D) long, curving; inner 
marginal tooth with 15-19 inner denticles, 
large central cusp and 5-6 outer denticles; 
outer marginal tooth same, but lacking outer 
denticles. 



Stomach with central pad, gastric shield, 
short style sac and crystalline style; one 
opening to digestive gland. 

Nervous System: Cerebral ganglia joined by 
short connective. Pleural ganglia close to ce- 
rebral ganglia; left pleural ganglion connected 
to subesophageal by very short connective. 
Supraesophageal connective about two- 
thirds length of right pleural ganglion. 

Reproductive System (Fig. 16A): Testis 
white, producing dimorphic sperm; ovary 
cream-yellow containing large ova, 0.5 mm in 
diameter. Glandular portion of female palliai 
oviduct comprising many transverse folds, 
posterior opaque white portion comprising al- 
bumen gland (Fig. 16A, ag), and anterior, 
transparent greyish-white portion comprising 
capsule gland (Fig. 16A, eg). Anterior two- 
thirds of edge of medial lamina with large 
sperm gutter (Fig. 16A, sg) leading into deep 
slit containing two openings: anterior opening 
(Fig. 16A, osp) into large sperm bursa and 
posterior opening (Fig. 16A, osr) into small 
tubular sac-like seminal receptacle (Fig. 16A, 
sr). Lateral lamina less glandular than medial 
lamina and with short ciliated ridge tract (Fig. 
16A, crt) leading into opening of spermato- 
phore bursa (Fig. 16A, osb), adjacent to 
openings on medial lamina. Spermatophore 
bursa (Fig. 16A, sb) small, elongate, sac-like. 

Discussion 

Bartsch (1911) assigned this species to the 
subgenus Semibittium, and his assignment 
was followed by Dall (1921), Oldroyd (1927), 
and Hertz (1981). Semibittium is shown 
herein to comprise a group of Eocene fossils 
probably related to the extant Australian mo- 
notypic genus Cacozeliana, which differs con- 
siderably in anatomy from the California spe- 
cies. Abbott (1974) transferred this species, 
which he considered a synonym of Bittium at- 
tenuatum Carpenter, 1864, to Lirobittium, but 
gave no reasons for doing so. 

The shell is of moderate size (Table 3) and 
has a large protoconch sculptured with two 
spiral lirae and lacking a sinusigeral notch. 
Although the shell of Lirobittium subplanatum 
does not resemble that of Stylidium es- 
chrichtii, the anatomical features of the two 
species are quite similar. As far as can be 
seen in preserved material, Lirobittium sub- 
planatum appears to have a very weak epi- 
podial skirt, but closer examination of living 
animals may show that this character is com- 



298 



HOUBRICK 




FIG. 16. Lirobittium subplanatum. A, palliai oviduct, spread open to reveal details; B, head, showing broad 
snout, short cephalic tentacles and small buccal mass; C, dorsal view of attached spawn mass, showing 
individual capsules with enclosed embryos and attachment strands, ag = albumen gland; ant = anterior of 
palliai oviduct; bm = buccal mass; eg = capsule gland; cod = coelomic oviduct; crt = ciliated ridge tract; 
osp = opening to sperm pouch; osb = opening to spermatophore bursa; ovg = oviductal groove; sb = 
spermatophore bursa; sg = sperm groove; sp = sperm pouch; sr = seminal receptacle. 



pletely absent. The operculum also differs in 
being more typically rounded than that of Sty- 
lidium. 

The radula of Lirobittium subplanatum (Fig. 
16) is very similar to that of Lirobittium atten- 
uatum, but differs in having many more den- 
ticles on the teeth. The exact dentition for- 
mula is given in Table 2. 

There has apparently been some difficulty 
in identifying this species, as it has been con- 
sidered synonymous with a number of other 
sympatric species, but Hertz (1981) has 
shown that it is a distinct, valid species. As 
mentioned above, the radula is distinct. 



Lirobittium subplanatum lives offshore on 
sandy-rubble bottoms. The shell is frequently 
severly eroded and abralded. 

Spawn morphology of Lirobittium is unique 
among Bittiinae (Fig. 17C) and is deposited 
on pieces of rubble or empty shells. It com- 
prises clusters of large egg capsules, each 
about 0.5 mm in diameter and containing one 
embryo. Each egg capsule is connected by a 
strand to a central attachment point so that 
the spawn mass looks like a group of small 
balloons with their strings attached together. 
Embryos revolve slowly with their capsules, 
where they pass through the veliger stage. 



GENERIC REVIEW OF BITTIINAE 



299 




FIG. 17. SEM micrographs of radula of Lirobittium subplanatum (USNM 881021). A, radular ribbon with 
marginal teeth spread open, bar = 35 ц,т; В, half row showing rachidian and lateral teeth, bar = 19 ^.m; 
C, detail of dentition of rachidian and lateral teeth, bar = 10 (xm; D, detail of dentition of marginal teeth, bar 
= 12 |j.m. 



finally hatching out as small snails. Develop- 
ment is direct (pers. obs.). 

CACOZEUANA STRAND, 1928 

? Semibittium Cossmann, 1896: 29 (Type 
species by original designation: Cerith- 
ium cancellatum Lamarck, 1804; not 
Semibittium Bronn, 1831 ; nor Lea, 1842; 



norTuomey, 1848; nor J. de С. Sowerby, 
in Dixon, 1850). Thiele, 1929: 211; 
Wenz, 1940: 756; Gründel, 1976: 56-57. 
Cacozelia Iredale, 1924: 246 (Type species 
by monotype: Cerithium lacertinum 
Gould, 1861); not Cacozelia Grote, 1878 
[Lepidoptera]. Thiele, 1929: 21 1 ; Murray, 
1969: 111. 



300 



HOUBRICK 



Cacozeliana Strand, 1928: 66 (new name for 
Cacozelia Iredale, 1924). Wenz, 1940: 
756. 

Lirobittium {Cacozeliana) Strand. Gründet, 
1976: 54-55. 

Diagnosis 

Shell large, elongate with many weakly in- 
flated whorls, sculptured with four beaded 
spiral cords per whorl and having overall pus- 
tulose appearance. Protoconch unsculptured 
except for microscopic subsutural pustules, 
but large sinusigeral notch present (Fig. 18F). 
Operculum circular-ovate, paucispiral with 
subcentric nucleus and fringed edges. Epipo- 
dial skirt with smooth edges. Snout short, nar- 
row. Opercular lobe lanceolate and with lon- 
gitudinal median groove. Large ovipositor 
gland on right side of foot. Osphradium bipec- 
tinate. Salivary glands anterior to nerve ring. 
Rachidian tooth without glabrella. Openings 
to sperm bursa and seminal receptacle well 
separated. Seminal receptacle comprising 
several grape-like lobes. 

Remarks 

The genus Cacozelia was proposed by Ire- 
dale for Cerithium lacertinum Gould, a sub- 
jective synonym of Cerithium granarium 
Kiener. The living Australian species is 
thought to be congeneric with the Paris Basin 
Eocene species Cerithium cancellatum La- 
marck, which is the type species of Semibit- 
tium Cossmann; however, as Cacozelia is a 
junior homonym, the name Cacozeliana was 
subsequently proposed by Strand (1928) as a 
replacement. The allocation of Cacozeliana 
as a subgenus of Liocerithium by Gründel 
(1976) was made on the observation that in 
Cacozeliana, the fourth primary spiral cord is 
initially weaker than the three formed earlier, 
whereas in Liocerithium all four are equally 
strong. Gründel (1976) also pointed out that 
varices are present in the subgenus, whereas 
they are absent in Lirobittium. These minor 
sculptural differences hardly seem appropri- 
ate as generic-level characters; furthermore, 
radular and anatomical characters of Cacoz- 
eliana show that it is far-removed from Liro- 
bittium. 

The type species of Semibittium, which is 
placed into synonymy with Cacozeliana with a 
query, is an Eocene fossil from the Paris Ba- 
sin, Cerithium cancellatum Lamarck. This fos- 
sil species is conchologically very close to 



Cerithium granarium Kiener, the living type 
species of Cacozeliana from southern Austra- 
lia redescribed herein; however, because the 
anatomy of the fossil is unknown, it is impos- 
sible to declare with confidence that the two 
species are congeneric. Gründel (1976: 56) 
considered the Eocene genus Semibittium to 
be separate from Cacozeliana. He noted that 
the shell of Semibittium species has a slight 
varix on the lip of the protoconch followed by 
an almost simultaneous insertion of the three 
primary spiral cords. The name Cerithium 
cancellatum Lamarck is preoccupied, and 
needs a replacement name. Moreover, the 
name Semibittium cannot be used because it 
is thrice preoccupied. The possibility that Ca- 
cozeliana granarla is a living survivor of the 
Eocene genus Semibittium represented by 
Cerithium cancellatum should be considered, 
because several other Tethyan Eocene cer- 
ithioidean genera survive among the living 
Australian molluscan fauna; e.g.. Diastema 
Deshayes, 1850; Gourmya Fischer, 1884; 
Campanile Fischer, 1884; and Plesiotrochus 
Fischer, 1878 (Houbrick, 1981b, 1981c, 
1981d, 1990b, respectively). It is also notable 
that Cacozeliana falls out at the base of the 
cladogram (Fig. 1) as the closest taxon to the 
outgroup. Moreover, Cacozeliana is sepa- 
rated from all other ß/Wum-group genera by 
five non-homoplastic synapomorphies (Fig. 
1), further demonstrating its distinctiveness. 
Gründers (1976: 56-57) separation of Semi- 
bittium from Cacozeliana was based on the 
order of the insertion of spiral lirae on the 
early whorls, but this character has not been 
shown to be of generic weight, and therefore 
is not seriously considered herein. If Cacoze- 
liana is truly congeneric with Semibittium, the 
genus would date from the Eocene, when the 
latter was common in the Paris Basin fauna 
(Cossmann, 1906: 138). Cacoze//ana is today 
monotypic and confined to the temperate wa- 
ters of southern Australia. The type species, 
Cacozeliana granarla (Kiener), undoubtedly 
has the largest shell of any representative of 
the subfamily Bittiinae and differs from other 
species of the group in several ways: 

1 . The short narrow snout (Fig. 20A) is dis- 
tinctive, as is the fringed operculum (Fig. 
18G). 

2. The rachidian tooth of Cacozeliana gra- 
narla is unique, differing from other Bittiinae 
members in lacking a glabrella on the basal 
plate. Additionally, the rachidian tooth lacks 
concave sides and a strong pair of basal but- 
tresses (Fig. 19B). Moreover, the lateral basal 



GENERIC REVIEW OF BITTIINAE 



301 



extensions of the basal plate are nearly ab- 
sent. 

3. The paillai oviduct of Cacozeliana grana- 
ría (Fig. 20C), while having a typical layout, is 
unique annong known palliai oviducts in the 
Bittium-group in having the seminal recepta- 
cle divided into several grape-like lobes (Fig. 
20C, sr) and in having a highly developed, 
swollen anterior capsule gland (Fig. 20C, eg). 
As pointed out earlier, a grape-like seminal 
receptacle also occurs in some species of 
Ceríthium Bruguière, 1789, Rhinoclavis 
Swainson, 1840, and in DialaA. Adams, 1861 
(Houbrick, 1971, 1978, 1992, pers. obser.; 
Ponder, 1991), although this structure in Diala 
is not proven to be a seminal receptacle. This 
kind of seminal receptacle does not necessar- 
ily indicate relatedness among these groups: 
the bulging, grape-like morphology may be 
due to the swollen state of the filled seminal 
receptacle and may represent sexual "ripe- 
ness" rather than a distinct morphological 
character state of the seminal receptacle. 

Cacozeliana granaría (Kiener, 1842) 
(Figs. 18-20) 

Ceríthium granarium Kiener, 1842: 72-73, pi. 
19, fig. 3 (Holotype MNHNP; type local- 
ity, "les côtes de Timor," in error, here 
corrected and restricted to Albany, West- 
ern Australia); G. B. Sowerby, 1855: 879, 
pi. 184, figs. 225-227; 1865: pi. 19, fig. 
135; Kobelt, 1898: 249, pi. 23, fig. 9. 

Cerittiium lacertinum Gould, 1861: 368 (Ho- 
lotype USNM 16571; type locality Syd- 
ney Harbor, New South Wales, Austra- 
lia); 1862: 141; G. B. Sowerby, 1866: pi. 
18, fig. 128; Tryon, 1884: 155, pi. 30, fig. 
100; R. Johnson, 1964: 96, pi. 11, fig. 4. 

Bittium granaríum (Kiener). Tryon, 1887: 155, 
pi. 30, fig. 98; Weils, 1984: 30-31. 

Synonymic Remarks 

Kiener's (1842) name, granarium, predates 
Gould's (1861) /ace/t/num. Examination of the 
holotypes of both taxa leaves no doubt that 
the two are conspecific. 

Description 

Shell (Fig. 18): Shell large, elongate, tur- 
reted, reaching 24 mm in length comprising 
12-13 nearly flat-sided whorls sculptured 
with four beaded spiral cords. Protoconch 
(Fig. 18F) comprising two smooth whorls with 



weak, microscopic subsutural pustules, no 
spiral lirae, and with deep sinusigeral notch. 
Early whorls (Fig. 18H) sculptured with 3 spi- 
nosely beaded spiral cords alined to form 
about 12-13 axial riblets. Adult whorls slightly 
beveled abapically, defining weak suture. 
Body whorl one-third shell length, having 6 
spiral beaded cords and weakly constricted 
base. Aperture ovate, small, about one-fifth 
shell length. Columella concave with weak 
columellar callus and smooth, rounded outer 
lip. Anterior canal short, narrow, well defined. 
Shell color white to tan, blotched with pink to 
reddish brown and having brown spiral bands 
with white flecks (Fig. 18C, D). Beads some- 
times white (Fig. 18A B). Periostracum light 
tan, thin. 

Animal (Fig. 20): Head, snout and epipodium 
pigmented tan with chocolate blotches, tiny 
white spots, and irridescent green. Cephalic 
tentacles darkly pigmented, having many 
black spots, slender, elongate, aboui twice 
snout length. Snout narrow, short (Fig. 20A, 
sn) with flared bilobed tip. Mantle edge 
fringed with very small papillae each bearing 
white spot. Pair of ciliated strips emerging 
from mantle floor and running to base of 
cephalic tentacles on each side of headfoot. 
Deep ciliated groove running down right side 
of foot to edge, ending in small flap in males. 
Ciliated groove in females having thick glan- 
dular strips on each side of groove, compris- 
ing ovipositor. Epipodial skirt poorly devel- 
oped, smooth along edge, forming short 
lanceolate opercular lobe with dorsal longitu- 
dinal furrow and without papillae along edge. 
Crescent-shaped propodial slit at edge of an- 
terior foot leading into deep oval anterior mu- 
cus gland (Fig. 20A, amg). Longitudinal fold in 
middle of sole, but no metapodial mucus 
gland present. Operculum (Fig. 18G) circular- 
ovate, paucispiral, with subcentral nucleus. 
Opercular spiral fringed with thin lamella (Fig. 
18G). 

Palliai cavity: Osphradium bipectinate, with 
weak pectins. Osphradium equaling ctenidial 
length. Ctenidium comprising light tan elon- 
gate, triangular filaments. Hypobranchial 
gland thick, comprising irregular transverse 
glandular folds, secreting large amounts of 
mucus. 

Alimentary system (Fig. 19B): Buccal mass 
large, filling snout cavity, having small jaws 
and short radula (Fig. 19A). Rachidian tooth 
(Fig. 19B) with rectangular basal plate lacking 



302 



HOUBRICK 




FIG 1 8 Cacozeliana granaría from King George Sound, Western Australia (USNM 858551 ). A-D, two shells 
showing variation in color pattern and sculpture, length 22.4 глт and 20.2 mm, respectively; E, SEM 
micrograph of immature shell, bar = 0.6 mm; F, SEM micrograph of protoconch, bar = 16 м -m; G, SEM 
micrograph of operculum, bar = 0.8 mm; H, SEM micrograph showing early sculpture, bar = 0.8 mm. 



Strong basal lateral buttresses, with straight 
base and equal in length to top of tooth; cut- 
ting edge with small central cusp flanked by 
two denticles on each side. Lateral tooth (Fig. 
19B) with one inner denticle and 3-4 outer 
denticles. Inner marginal tooth with 5-6 inner 
denticles and 3-4 outer denticles. Outer mar- 
ginal tooth (Fig. 19A) with 4 inner denticles. 
Salivary glands (Fig. 208, Isg, rsg) paired, 
vermiform, coiled, lying anterior to nerve ring. 
Midesophagus expanded laterally having 



many transverse internal epithelial folds com- 
prising esophageal gland. Stomach with one 
digestive gland opening to left of large central 
pad dividing left sorting area from right gastric 
shield complex. Style sac separated from in- 
testinal opening by large typlosole fold. 

Nervous System (Fig. 20, B): Cerebral gan- 
glia joined by short connective, one-third the 
ganglion length. Subesophageal ganglion 
very close to left pleural ganglion. 



GENERIC REVIEW OF BITTIINAE 



303 




FIG. 19. Radula of Cacozeliana granaría from King George Sound, Western Australia (USNM 858551 ¡ 
mid-section of radula, bar = 60 |xm; B, details of rachidian and lateral teeth, bar = 15 |xm. 



A, 



sr osr cod 




0.25mm 



FIG. 20. Anatomical features of Cacozeliana granaría. A, head and foot anterior, showing narrow snout; B, 
position of salivary glands anterior to nerve ring; C, palliai oviduct, spread open to reveal interior details, a 
= anterior end of palliai oviduct; ag = albumen gland; eg = capsule gland; cod = coelomic oviduct; ctr = 
ciliated ridge tract; Isg = left salivary gland; osb = opening to spermatophore bursa; osp = opening to 
sperm pouch; osr = opening to seminal receptacle; rpg = right pleural ganglion; rsg = right salivary gland; 
sb = spermatophore bursa; sg = sperm groove; sp = sperm pouch; sr = seminal receptacle. 



Reproductive System: Male palliai gonoduct 
thick, glandular, having wide transverse folds 
forming spermatophore organ in posterior 



half; anterior half of male palliai gonoduct less 
glandular, white but not opaque. Female pal- 
liai oviduct (Fig. 20C) having seminal recep- 



304 



HOUBRICK 



tacle comprising several grape-like lobes in 
medial lamina (Fig. 20C, sr). Openings to the 
sperm pouch (Fig. 20C, osp) and seminal re- 
ceptacle (Fig. 20C, osr) separated by long cil- 
iated groove. Ciliated ridge tract (Fig. 20C, 
ctr) beginning behind anterior capsule gland 
(Fig. 20C, eg) comprising many swollen trans- 
verse elements. Opening to spermatophore 
bursa (Fig. 200, osb) in lateral lamina adja- 
cent to opening of sperm pouch in medial 
lamina. Spawn mass comprising a jelly string 
containing many encapsulated eggs, 
0.1-0.13 mm diameter, wound into flattened 
coil about 20 mm wide. Eggs opaque, white, 
each within hyaline capsule. Development in- 
direct with free swimming veliger stage. 

Discussion 

Although the shell of Cacozeliana granaría 
(Fig. 18) looks very much like those of some 
Centhium species, the weak epipodial skirt, 
palliai oviduct, and other anatomical features 
are very typical of members of the Bittiinae. 
The protoconch, as indicated by Gründel 
(1976), differs from those of most other gen- 
era in being nearly smooth, and in lacking any 
spiral threads (Fig. 18F; Table 3), but it does 
have a deep sinusigeral notch, indicative of 
planktotrophy. Stylidium species also have a 
smooth protoconch. The operculum of Caco- 
zeliana is unusual in having a thin lamellar- 
like fringe along its spiral (Fig. 1 80). The shell 
of this species is undoubtedly the largest of 
any member of the Bittium-group (Table 3), 
but the aperture is very small in relation to the 
shell length. There is much color variation 
within populations. 

The early life history of this species has 
been described by Murray (1969), who illus- 
trated the spawn (1969: pi. 17). The spawn 
comprises a coiled gelatinous thread contain- 
ing encapsulated eggs that hatch as plank- 
totrophic veligers. Murray (1969) stated that 
8-9 days after deposition, veliger-stage em- 
bryos hatched out and were maintained in 
sea water containers for up to 10 weeks. 

Cacozeliana granaría is found in the shal- 
low subtidal, temperate waters of southern 
Australia where it is common among Posi- 
donia, Zostera, and other sea grasses. It also 
occurs on moderately exposed and sheltered 
shores, on sandy-muddy bottoms, under 
stones, and on rocky areas. I observed large 
populations of this species living on algal 
mats and on Posidonia grass blades in King 
George Sound, Western Australia, and in 




FIG. 21 . SEM micrographs of shell of Argyropeza 
divina Melvill & Standen, from Refugio Id., Tanon 
Str., Philippines (USNM 302513); A, B, apertural 
and dorsal views of adult shell, 6.3 mm length; C, 
protoconch showing sculpture and sinusigeral 
notch, bar = 1 mm. 



similar habitats in Sydney Harbor and Botany 
Bay, New South Wales. 

ARGYROPEZA MELVILL & 
STANDEN, 1901 

Argyropeza Melvill & Standen, 1 901 : 371-372 
(Type species by original designation, Ar- 
gyropeza divina Melvill & Standen, 1 901 ). 
Thiele, 1929: 212; Wenz, 1940: 757; 
Oründel, 1976: 44; Houbrick, 1980a: 2. 

Diagnosis 

Shell small, turreted, thin and vitreous, 
sculptured with axial and spiral elements, va- 
rices, and with many small nodules. Proto- 
conch comprising three and a half whorls with 
deep sinusigeral notch; sculptured with two 



GENERIC REVIEW OF BITTIINAE 



305 




FIG. 22. SEM micrographs of radula of Argyropeza divina (USNM 302513), A, radular ribbon with marginal 
teeth spread open, bar = 100 jim; B, half row, bar = 50 |xm. 



spiral cords and nnany minute subsutural 
folds. Aperture ovate with well-developed, 
short anterior canal. Operculum corneous, 
subcircular, paucispiral, with subcentral nu- 
cleus. Snout broad with large cephalic tenta- 
cles and large eyes. Foot with anterior mucus 
gland. Mantle edge papillate. Palliai gono- 
ducts open. Radula taenioglossate; rachidian 
tooth wider than tall; lateral tooth with trans- 
verse ridge on basal plate; marginal teeth 
slender, scythe-shaped. 

Remarks 

An alpha-level review of Argyropeza has 
been published by Houbrick (1980a), which 
should be consulted for details about taxon- 
omy, morphology and geographic distribution. 
The genus comprises five described species 
and several undescribed ones (pers. obser.). 
Members of this genus live on fine-grained 
substrates of deep water shelves and slopes, 
and not much is known about their biology. All 
examined species have small shells and pro- 
toconchs sculptured with two spiral lirae, sub- 
sutural pleats, and a deep sinusigeral notch 
(Fig. 21 C; Table 3) indicative of a plank- 
totrophic larval stage. The anatomy of Argy- 
ropeza species is virtually unknown except for 
superficial observations made from reconsti- 



tuted, dried specimens. The shell and radula 
of the type species, Argyropeza divina Melvill 
& Standen, 1901, are shown in Figures 21 
and 22. I do not agree with Powell's (1979) 
suggestion that Tasmalira Dell, 1956, may be 
closely related to Argyropeza, because the 
shell morphology does not appear to fit the 
limits of the genus. Argyropeza is tentatively 
assigned to the Bittiinae until more complete 
anatomical information is available. 

VARICOPEZA GRÜNDEL, 1976 

Varicopeza Gründel, 1976: 46 (Type species 
by tautonomy, Varicopeza varicopeza 
Gründel, 1976). Houbrick, 1980b: 525; 
1987: 85. 

Diagnosis 

Shell small, slender, turreted, vitreous, hav- 
ing impressed suture, and sculptured with 
strong spiral cords, weaker axial elements, 
and many nodules. Protoconch having three 
and one-half smooth whorls, with weak, me- 
dian spiral cord, minute subsutural pustules, 
and sinusigeral notch. Aperture ovate with 
short, well-developed anal and anterior ca- 
nals. Operculum corneous, ovate, paucispi- 
ral, with subcentral nucleus. Radula taenio- 



306 



HOUBRICK 




FIG. 23. SEM micrographs of shell of Varicopeza 
pauxilla (A. Adams, 1854) from Nagubat Id., E. Min- 
danao, Philippines (USNM 276898). A, B, apertural 
and side views of adult shell, 8.1 mm length; C, 
protoconch, bar = 100 |xm. 



glossate with hourglass-shaped rachidlan 
tooth; lateral tooth with transverse ridge on 
basal plate; marginal teeth elongate, slender 
with denticulate sickle-shaped tips. Animal 
with large headfoot, elongate, wide snout, 
long cephalic tentacles and very large eyes. 
Deep ciliated groove on right side of foot. 
Mantle edge having short, thick papillae. 

Remarks 

The two known species of Varicopeza have 
been thoroughly described by Houbrick 
(1980b, 1987a). These publications should be 
consulted for specific information about tax- 
onomy and a detailed description of the type 
species. The shell is of moderate length (Ta- 
ble 3) and has a protoconch sculptured with 
one spiral lira and a shallow sinusigeral notch 
(Fig. 23C). Although the shell and radula (Fig. 



24) are well described, only a few external 
anatomical features are known. Varicopeza 
species occur at moderate subtidal depths on 
fine-grained substrates in the tropical Atlantic 
and Pacific. The shell sculpture of Varicopeza 
(Fig. 23A, B) is similar to that of Argyropeza 
species, differing chiefly in protoconch mor- 
phology. The aperture (Fig. 23A, B) is distinc- 
tive in having a large, flaring anal sinus. The 
radula (Fig. 24) has more denticles on the 
marginal teeth than in Argyropeza (Table 2). 
Gründel (1976) suggested that Varicopeza 
was closely related to the extinct Jurassic ge- 
nus Cryptaulax and considered it to be a Re- 
cent representative of the of the extinct family 
Procerithiidae Cossmann, 1905. The shell 
and radula of Varicopeza pauxilla (A. Adams, 
1854) is shown in Figures 23 and 24. This 
genus is tentatively assigned to the Bittium- 
group until more complete anatomical infor- 
mation is available. 

ZEß/rr/L/M FINLAY, 1927 

Zebittium Finlay, 1927: 381 (Type species by 
original designation, Cerithium exilis Hut- 
ton, 1873); Wenz, 1940: 756; fig. 2191; 
Powell, 1979: 132, fig. 32:1. 

Diagnosis 

Shell very small, turreted, sculptured with 
beaded spiral cords, and weak axial riblets, 
having impressed suture. Aperture ovate with 
weak notch-like anterior canal. Protoconch 
two and a half whorls, bluntly rounded, un- 
sculptured. 

Remarks 

This genus was proposed without any de- 
fining characters, and was apparently intro- 
duced only to accomodate the New Zealand 
species, Bittium exile Hutton and Bittium vit- 
reum Suter. The shell of Zebittium exile (Hut- 
ton, 1873) is shown in Figure 25. Zebittium 
was assigned as a subgenus of Bittium by 
Wenz (1940), who noted that the genus oc- 
cured from tfie Miocene to the Recent of New 
Zealand. The shell of the type species closely 
resembles those of Bittium and Bittiolum spe- 
cies and does not appear to have any distin- 
guishing features of generic significance. The 
unsculptured protoconch (Fig. 25D) appears 
to indicate lecithotrophic development. No 
preserved material of this species was avail- 



GENERIC REVIEW OF BITTIINAE 



307 




FIG. 24. SEM micrographs of radula of Varicopeza pauxilla. A, section of ribbon with some marginal teeth 
spread open, bar = 50 |xm; B, detail of rachidian and lateral teeth, bar = 25 \xm. 



able for study; therefore, the genus Zebittium 
is included in this review only tentatively. 

CASSIELLA GOFAS, 1976 

Cassiella Gofas, 1987: 109 (Type species by 
original designation, Cassiella abylensis 
Gofas, 1987). 

Diagnosis 

Shell small, slender, turrited, sculptured 
with spiral cords, without varices and with im- 
pressed suture. Aperture ovate, without ante- 
rior canal and simple outer lip. Operculum 
corneous, ovate, paucispiral, with subcentral 
nucleus. Animal with bllobed snout and two 
elongate cephalic tentacles. Foot short and 
broad without ovipositor or ciliated groove on 
right side, and with large opercular lobe. Rad- 
ula taenioglossate; rachidian tooth with 
squarish basal plate, moderately concave on 
each side with small median glabrella, and 
having cutting edge with large central cusp 
flanked by 3 smaller denticles on each side. 
Lateral tooth with large triangular cusp with 
one small inner denticle and 7-8 outer denti- 
cles. Marginal teeth elongate, spatulate with 
curved tips; inner marginal teeth denticulate 
on both sides; outer marginal teeth lacking 
outer denticles. 



Remarks 

This monotypic genus was recently pro- 
posed and described by Gofas (1987), and 
his publication should be consulted for de- 
scriptive details of the genus and figures of 
the type species. Cassiella abylensis does 
not fit easily into the Bittium-group, although 
there are some resemblances. The shell of 
Cassiella abylensis (Fig. 26) varies highly in 
color pattern and in spiral sculpture (Gofas, 
1987: 111). The shell morphology is unlike 
those of other members of the Bittium -дюир. 
No vestige of an anterior canal is present, and 
the shell morphology strongly resembles 
those of some rissoids. The absence of an 
anterior canal is also a feature of Cerithidium 
Monerosato, a taxon I have excluded from 
Bittiinae. 

The external anatomy of Cassiella abylen- 
sis was depicted by Gofas (1987: figs. 10, 14, 
15). The animal does not have epipodial ten- 
tacles, although there is an inconspicuous 
groove around the foot, just above the edge of 
the sole, which may be homologous with the 
epipodial skirt found in members of Bittiinae. 
The opercular lobes are said to be "massive" 
(Gofas, 1987: 111), but they are not depicted 
or labeled in the figures of the external anat- 
omy. The headfoot, operculum, and radula 
are not unlike those observed in other species 



308 



HOUBRICK 




FIG. 25. SEM micrographs of shell of Zebittium ex- 
ile (Hutton, 1873) from Long Bay, Auckland, New 
Zealand (USNM 681043); A, apertural view of adult 
shell, 4.7 mm length; B, dorsal view, 4.6 mm length; 
C, immature shell, 4.4 mm length; D, protoconch, 
bar = 0.25 mm. 



of Bittiinae. There is no metapodial mucus 
gland, no ovipositor is indicated, and males 
are aphallate (Gofas, 1987: 111). 

Pending further anatomical studies, the 
eastern Atlantic taxon Cassiella is tentatively 
assigned to Bittiinae with doubt. 



ACKNOWLEDGEMENTS 

This study was accomplished in many di- 
verse places and with the help of many col- 
leagues and friends. I wish to thank Dr. Anto- 
nio Frias Martins, of the University of the 
Azores, for sponsoring me at the First Inter- 
national Workshop of Malacology, held at Sao 
Miguel, Azores. This part of my study was 
supported by a grant of the Portuguese Uni- 
versity of the Azores and the Sociedade de 




FIG. 26. SEM micrographs of shell of Cassiella 
abylensis Gofas, 1976, from Ceuta, Spain (USNM 
869532); A, apertural view of shell, 2.3 mm length; 
B, dorsal view of shell, 2.5 mm length. 



Estudos Açorianos "Alfonso Chaves." Work 
on the Western Atlantic species was done at 
the Smithsonian Marine Station, Link Port, 
Florida. I am grateful to Dr. Mary Rice and the 
staff of the marine station for their assistance 
throughout this project. This paper is Smith- 
sonian Marine Station contribution No. 272. 
Work in Hawaii and Guam was supported by 
two grants from the Smithsonian Secretary's 
Research Opportunity Fund. I am grateful to 
the University of Guam for laboratory space, 
equipment and logistic support. I thank Dr. 
Michael Hadfield, of the University of Hawaii, 
for providing laboratory space at the Pacific 
Biomedical Research Laboratory, and for his 
assistance with field work. A grant from the 
Smithsonian Secretary's Research Opportu- 
nity Fund supported field and laboratory stud- 
ies and attendance at the Workshop on Ma- 
nne Biology at Albany, Western Australia. I 
am indebted to Dr. Fred Wells, Western Aus- 
tralian Museum, Perth, for his assistance in 
the field. Dr. Henry Chaney, Mrs. Barbara 
Chaney, and Mr. Paul Scott of the Santa Bar- 
bara Museum of Natural History, provided lo- 
gistic and field assistance in an heroic, alas 
unsuccessful, attempt to find living Lirobittium 
specimens. I thank Don Cadien for sending 
me live specimens of "Semibittium" sub- 
planatum Bartsch from off Palos Verdes, Cal- 
ifornia, and am grateful to Serge Gofas, Nat- 
ural History Museum, Paris, for sending shells 



GENERIC REVIEW OF BITTIINAE 



309 



of Cassiella abylensis. For technical assis- 
tance (proofreading and SEM, and computer 
macro design) I tfiank Shelley Greenhouse, 
National Museum of Natural History, Smith- 
sonian Institution. Susanne Braden, National 
Museum of Natural History, Smithsonian In- 
stitution, provided technical assistance with 
SEM operation. John Wise provided valuable 
assistance in learning various aspects of the 
Hennig86 and CLADOS programs. Finally I 
am grateful to Dr. Winston F. Ponder for crit- 
ically reading a draft of this paper and for 
stimulating discussions and exchanges of 
data about anatomy and evolution of small- 
sized cerithioidean taxa. 



LITERATURE CITED 

ABBOTT, R. T., 1974, American seashells, 2nd ed.: 
663 pp., illus. New York, Van Nostrand. 

ADAMS, A., 1860. Mollusca Japónica: new species 
of Aclis, Ebala, Dunkeria, etc. Annals and Mag- 
azine of Natural History, series 3, 6: 1 18-121 . 

ADAMS, A., 1861, On some new genera and spe- 
cies of Mollusca from the north of China and Ja- 
pan. Annals and Magazine of Natural History, se- 
ries 3, 8: 239-246. 

ADAMS, С В., 1845, Specierum novarum con- 
chyliorum in Jamaica repertorum synopsis. Pro- 
ceedings of thie Boston Society of Natural His- 
tory, 2: 1-17. 

ADAMS, H. & A. ADAMS, 1853-1858, The genera 
of Recent Mollusca; arranged according to thieir 
organization. 2 volumes. 1(1-8): 1-256, plates 
1-32 (1853); 1(9-15): 257-484, 2: 1-92, plates 
33-72 (1854); 2(19-24): 93-284, plates 73-96 
(1855); 2: 285-412, plates 97-112 (1856); 2: 
413-540, plates 113-128 (1857); 2: 541-660, 
plates 1 29-1 38 (1 858). London: John Van Voorst. 

BARTSCH, P., 1907, New marine mollusks from 
the west coast of America. Proceedings of the 
United States National Museum, 33: 177-183. 

BARTSCH, P., 1911, The Recent and fossil mol- 
lusks of the genus Bittium from the West coast of 
America. Proceedings of the United States Na- 
tional Museum, 40: 383-414, pis. 51-58. 

BIGGS, H. E. J., 1971, On a proposed new genus 
of cerithid Mollusca from the Dahlak Islands, Red 
Sea. Journal of Conchology, 27: 221-223, pi. 7. 

BLAINVILLE, H. M., 1816-1830. Vers et Zoo- 
phytes, in: Dictionnaire des sciences naturelles. 
Pt. 2. Règne organisé. Paris. 60 vols. + atlas. 

BOETTGER, O., 1883. Die Tertiärformation von 
Sumatra und ihre Thierreste, 2, Anhang S., 137, 
pi. 11, fig. 14. 

BUCQUOY, E., P. DAUTZENBERG & G. DOLFUS, 
1882-1886, Les Mollusques marins du Roussil- 
lon. Vol. 1, Gastéropodes: 570 pp., Atlas, 66 pis. 

BRONN, H. G., 1831, Italiens Tertiär-Gebilde und 



deren organische Einschlüsse, xii + 176 pp. 
Heidelberg. 

BRUGUIÈRE, J. G., 1789-1792, Encyclopédie 
méthodique: Histoire naturelle des vers. Paris: 
Panckoucke. 1(1): 1-344 (1789); 1(2): 345-758 
(1792). 

CARPENTER, P. P., 1864, Supplementary report 
on the present state of our knowledge with regard 
to the Mollusca of the west coast of North Amer- 
ica. Report of the British Association for the Ad- 
vancement of Science, for 1863: 517-686. 

COSSMANN, M., 1884. Étude paléontologique et 
stratigraphigue sur le terrain oligocène marin aux 
environs d'Étampes. Mémoires de la Société Gé- 
ologique de France, (3)3(1): 1-187, pis. 1-6 
[With J. Lambert]. 

COSSMANN, M., 1889, Catalogue illustré des co- 
quilles fossiles de l'Éocène des environs de 
Paris. Quatrième fascicule. Annales de la So- 
ciété Royale Malacologique de Belgique, 24: 
3-381, pis. 1-12. 

COSSMANN, M., 1896, Appendice No. 2 au Cata- 
logue illustré des coquilles fossiles de l'Éocène 
des environs de Paris. Annales de la Société 
Royale Malacologique de Belgique, 31: 3-94, 
pis. 1-3. 

COSSMANN, M., 1902, Note sur l'Infralias de la 
Vendée et spécialement sur un gisement situé 
dans la commune de Simon-la-Vineuse 
(Vendée). Bulletin de la Société Géologique de 
France, (4) 2: 163-203, pis. 3-4. 

COSSMANN, M., 1906, Essais de paléoconcholo- 
gie comparée. Paris, F.R. de Rudival, 7: 261 pp.; 
14 pis., illustrated. 

COTTON, В. С, 1932, Notes on Australian Mol- 
lusca, with descriptions of new genera and new 
species. Records of the South Australian Mu- 
seum, Adelaide, 4: 537-547. 

COTTON, B. C, 1937, Nomenclatural note. The 
South Australian Naturalist, 18: 2. 

CROSSE, M. H., 1863, Description d'espèces nou- 
velles d'Australie. Journal de Conchyliologie, 1 1 : 
84-90, pi. 1. 

DACOSTA, E. M., 1778, Historia naturalis Testace- 
orum Britanniae, or the British Conchology. Lon- 
don, xii + 254 pp., 17 pis. 

DALL, W. H., 1889, Reports on the results of dredg- 
ing, under the supervision of Alexander Agassiz, 
in the Gulf of Mexico (1877-78) and in the Car- 
ibbean Sea (1879-80), by the U.S. Coast Survey 
steamer Blake (etc.) XXIX. Report on the Mol- 
lusca. Part II. Gastropoda and Scaphopoda. Bul- 
letin of the Museum of Comparative Zoology, 
Harvard 18: 492 pp., 40 pis. 

DALL, W. H., 1892, Contributions to the Tertiary 
fauna of Florida with special reference to the Mi- 
ocene Silex-Beds of Tampa. Part 2. Streptodont 
and other gastropods, concluded. Transactions 
of the Wagner Free Institute of Science of Phila- 
delphia, 3: 201-448, pis. 13-22. 

DALL, W. H., 1902, Note on the names Elachista 
and Pleurotomaria. The Nautilus, 15: 127. 

DALL, W. H., 1907, New marine mollusks from the 



310 



HOUBRICK 



west coast of America. Proceedings of the United 
States National Museum, 33: 177-183. 

DALL, W. H., 1921, Summary of the marine shell- 
bearing mollusks of the northwest coast of Amer- 
ica, from San Diego, California, to the Polar Sea, 
mostly contained in the collection of the United 
States National Museum, with illustrations of hith- 
erto unfigured species. United States National 
Museum Bulletin ^^2■. 1-217, pis. 1-22. 

DALL, W. H., 1924, Notes on molluscan nomencla- 
ture. Proceedings of thie Biological Society of 
Washington, 37: 87-90. 

DALL, W. H. & P. BARTSCH, 1901 . A new Califor- 
nian Bittium. The Nautilus, 15: 58-59. 

DELL, R. K., 1956, Some new off-shore Mollusca 
from New Zealand. Records of the Dominion Mu- 
seum, 3: 27-59. 

DAUTZENBERG, P., 1889, Contribution a la Faune 
Malacologique des îles Acores. Résultats des 
Dragages effectives par le Yacht l'Hirondelle 
pendant sa Campagne scientifique de 1887. 1 1 2 
pp., pis. 1-4. Monaco. 

DESMAYES, G. P., 1850, Traité élémentaire de 
conchyliologie, Atlas, 80 pp., 132 pis. V. Massen, 
Paris. 

FÉRUSSAC, J. B. L d'A. DE, 1819, Histoire na- 
turelle général et particulière des mollusques ter- 
restres et fluviátiles. Paris, 1: 128 pp., 162 pis. 

FINLAY, H. J., 1 927, A further commentary on New 
Zealand molluscan systematics. Transactions of 
the New Zealand Institute, 57: 320-484, pis. 1 8- 
23. 

FINLAY, H. J. & J. MARWICK, 1937, The Wanga- 
loan and associated molluscan faunas of Kaitan- 
gata-Green Island Subdivision, New Zealand 
Geological Survey Branch. Paleontological Bul- 
letin No. 15. Loney, Wellington. 140 pp., 17 pis. 

FISCHER, P., 1878, Diagnosis molluscorum novo- 
rum. Journal de Conchyliologie. 26: 21 1-212. 

FISCHER, P., 1880-1887 (1883-1884), Manuel de 
conchyliolgie et paléontologie conchyliologique 
(ou histoire naturelle des mollusques vivants et 
fossiles). Paris: Librairie F. Savy, xxiv + 1369 
pp., 23 pis. (fascicule 6, pp. 513-608, 1883; fas- 
cicule 7, pp. 609-688, 1884). 

FORBES, E. & S. HANLEY, 1850-1851. A history 
of British Mollusca and their shells, vol. 3. Van 
Voorst, London. 616 pp. 

FRETTER, v., 1948, The structure and life history 
of some minute prosobranchs of rock pools: Ske- 
neopsis planorbis (Fabricius), Omalogyra ato- 
mus (Philippi), Rissoella diaphana (Alder) and 
Rissoella opalina (Jeffreys). Journal of the Ma- 
rine Biological Association of the United King- 
dom, 27: 597-632. 

FRETTER, v., & A. GRAHAM, 1962, British proso- 
branch molluscs. Ray Society, London. 755 pp.; 
317 figs. 

FRETTER, V. & M. PILKINGTON, 1970, Proso- 
branchia. Veliger larvae of Taenioglossa and 
Stenoglossa. Conseil International pour l'Ex- 
ploration de la Mer, Zooplankton, sheets 129- 
132:3-26. 



GOFAS, S., 1987, Cassiella nov. gen., a cerithi- 
acean endemic to the Strait of Gibralter. Basteria, 
51: 109-119. 

GOLIKOV, A. N., & Y. I. STAROBOGATOV, 1975, 
Systematics of prosobranch Gastropoda. Mala- 
cologia, 15: 185-232. 

GOULD, A. A., 1861, On the specific distribution of 
faunae far removed from one another. Proceed- 
ings of the Boston Society of Natural History. 7: 
98. 

GOULD, A. A., 1870, Reporten the Invertebrata of 
Massachusetts. Boston, Wright and Potter, 2nd. 
ed., ii + 427 pp., pis. 16-27. 

GRAHAM, A., 1988. Mollusks: prosobranch and 
pyramidellid gastropods. Keys and notes for the 
identification of the species, 662 pp., 276 figs. In: 
D. M. Kermack & R. S. K. Barnes; eds.. Synopsis 
of the British fauna (New Series) No. 2. Leiden, 
Brill, 2nd ed. 

GRAY, J. E., 1847a, The classification of the British 
Mollusca by J. E. Leach, M.D.. Annals and Mag- 
azine of Natural History, 20: 267-273 (Septem- 
ber). 

GRAY, J. E., 1847b, A list of the genera of Recent 
Mollusca, their synonyma and types. Proceed- 
ings of the Zoological Society of London (for 
1847) 15(178): 129-219 (November). 

GROTE, A. R., 1878, On the pyralid genus Epipas- 
chia of Clemens, and allied forms. Proceedings 
of the Boston Society of Natural History, 1 9: 262- 
267. 

GRÜNDEL, J-, 1974, Bemerkungen zur Fassung 
der Gattungen Procerithium Cossmann, 1902 
und Cryptaulax Täte, 1 869 (Gastropoda, Cerithi- 
acea) im Jura. Zeitschrift für geologische Wis- 
senschaft, 2: 729-733. 

GRÜNDEL, J., 1976, ZurTaxonomie und Phyloge- 
nie der Biftium-Gruppe. Malakologische Abhand- 
lungen, 5: 33-59. 

HARDISON, L. K., & С L KITTING, 1985, Epi- 
phytic algal browsing by Biftium varium (Gas- 
tropoda) among Thalassia testuidinum turtle- 
grass. Journal of Psychology, 21 : 13. 

HASZPRUNAR, G., 1985. The fine morphology of 
the esophageal sense organs of the Mollusca. 1 . 
Gastropoda, Prosobranchia. Philosophical 
Transactions of the Royal Society of London, B, 
103:457-496. 

HEALY, J. M., 1986. Ultrastructure of parasperma- 
tozoa of cerithiacean gastropods (Prosobran- 
chia: Mesogastropoda). Helgoländer Meere- 
suntersuchungen, 40: 177-199. 

HEDLEY, С, 1899, The Mollusca of Funafuti. Part 
1 .-Gastropoda. Memoirs of the Australian Mu- 
seum. 3: 397-567, 80 figs. 

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

HERTZ, J., 1981. A review of several eastern Pa- 
cific Biftium species (Gastropoda: Cerithiidae). 
The Festivus, 13: 25-44. 

HOUBRICK, R. S., 1971. Some aspects of the 
anatomy, reproduction and early development of 



GENERIC REVIEW OF BITTIINAE 



311 



Cerithium nodulosum (Bruguière) [Gastropoda: 
Prosobranchia]. Pacific Science, 25: 560-565. 

HOUBRICK, R. S., 1977, Réévaluation and new 
description of the genus Bittium (Cerithiidae). 
The Veliger, 20: 101-106. 

HOUBRICK, R. S., 1978, Redescription of Bittium 
proteum (Jousseaume, 1 930) with comments on 
its generic placement. Ttie Nautilus, 92: 9-1 1 . 

HOUBRICK, R. S., 1980a, Review of the deep-sea 
genus Argyropeza (Gastropoda: Prosobranchia: 
Cerithiidae). Smittisonian Contributions to Zool- 
ogy, 321 : 30 pp., 1 1 figs. 

HOUBRICK, R. S., 1980b, Reappraisal of the gas- 
tropod genus Varicopeza Gründel (Cerithiidae; 
Prosobranchia). Proceedings of tfie Biological 
Society of Wasfiington, 93: 525-535. 

HOUBRICK, R. S., 1981a, Systematic position of 
the genus Glyptozaria Iredale (Prosobranchia: 
Gastropoda). Proceedings of the Biological Soci- 
ety of Washington, 94: 838-847. 

HOUBRICK, R. S., 1981b, Anatomy of Diastema 
melanoides (Reeve, 1849) with remarks on the 
systematic position of the family Diastomatidae 
(Prosobranchia: Gastropoda). Proceedings of the 
Biological Society of Washington, 94: 598-621 . 

HOUBRICK, R. S., 1981c, Anatomy and systemat- 
ics of Gourmya gourmyi (Prosobranchia: Cerithi- 
idae), a Tethyan relict from the southwest Pacific. 
The Nautilus, 95: 2-1 1 . 

HOUBRICK, R. S., 1981d, Anatomy, biology and 
systematics of Campanile symbolicum with ref- 
erence to adaptive radiation of the Cerithiacea 
(Gastropoda: Prosobranchia). f^alacologia, 21: 
263-289. 

HOUBRICK, R. S., 1987a, Transfer of Cerithiopsis 
crystallina Dali to the genus Varicopeza Gründel, 
family Cerithiidae (Prosobranchia: Gastropoda). 
The Nautilus, 101: 80-85. 

HOUBRICK, R. S., 1987b, Anatomy of Alaba and 
Litiopa (Prosobranchia: Litiopidae). The Nautilus, 
101:9-18. 

HOUBRICK, R. S., 1988, Cerithioidean phylogeny. 
in: W. F. Ponder, ed., Prosobranch phylogeny. 
Proceedings of a symposium held at the 9th in- 
ternational malacological congress, Edinburgh, 
1986. I^alacological Review, Supplement 4: 88- 
128. 

HOUBRICK, R. S., 1990a, Review of the genus 
Colina H. and A. Adams, 1854 (Cerithiidae: 
Prosobranchia). The Nautilus, 104: 35-52. 

HOUBRICK, R. S., 1990b. Aspects of the anatomy 
of Plesiotrochus (Plesiotrochidae, fam. n.) and its 
systematic position in Cerithioidea (Prosobran- 
chia, Caenogastropoda). Pp. 237-249, in: F. E. 
Wells, D. I. Walker, H. Kirkman, & R. Leth- 
BRiDGE, eds.. Proceedings of the Third Interna- 
tional f^arine Biological Workshop: the marine 
flora and fauna of Albany, Western Australia. 
Perth, Western Australian Museum, vol. 1 . 

HOUBRICK, R. S., 1992, Monograph of the genus 
Cerithium Bruguière in the Indo-Pacific (Cerithi- 
idae: Prosobranchia). Smithsonian Contributions 
to Zoology, 510: iv -i- 211 pp., 145 figs. 



HUTTON, F. W., 1873. Catalogue of the marine 
mollusks of New Zealand. Wellington, 1873. xx 
+ 116 pp. 

IREDALE, T., 1924, Results from Roy Bell's mol- 
luscan collections. Proceedings of the Linnean 
Society of New South Wales, 49: 179-278, pis. 
33-36. 

JEFFREYS, J. G., 1867-1869, British conchology, 
or an account of the Mollusca which now inhabit 
the British Isles and the surrounding seas. Lon- 
don, van Voorst, vol. 4, 1867; vol. 5, 1869. 

JEFFREYS, J. G., 1885, On the Mollusca procured 
during the Lightning and Porcupine expedition, 
1868-70 (part 9). Proceedings of the Zoological 
Society of London, for 1885: 27-63, pis. 4-6. 

JOHANSSON, J., 1947, Über den offenen Uterus 
bei einigen Monotocardiern ohne Kopulationsor- 
gan. Zoologiska Bidrag fràn Uppsala, 25: 102- 
110. 

JOHNSON, С W., 1915, Fauna of New England, 
13. List of the Mollusca. Occasional Papers of the 
Boston Society of Natural History, 7: 231 pp. 

JOHNSON, R., 1964. The Recent Mollusca of Au- 
gustus Addison Gould. Bulletin of the United 
States National Museum, 239: 182 pp., 45 pis. 

JOUSSEAUME, F., 1930, Cerithiidae de la Mer 
Rouge. Journal de Conchyliologie. 74: 270-296. 

KAY, E. A., 1979, Hawaiian marine shells. Bishop 
Museum Press, Special Publication 64(4), Hono- 
lulu, Hawaii. 653 pp. 

KIENER, L C, 1841-1842, Spéies général et ico- 
nographie des Coquilles vivantes . . . genre cé- 
rite. Paris: Rousseau, vol. 5: 1-104, pis. 1-32 
[pis., 1841; text, 1841-1842]. 

KOBELT, W., 1888-1898, Die Gattung Cerithium, 
297 pp., 47 pis., in F. H. W. Martini & J. H. Chem- 
nitz, Neues systematisches Conchylien- 
Cabinet. . . . Nürnberg, Bauer & Raspe, 1(26). 

KOSUGE, S., 1964. Anatomical study of Diala go- 
niochila (A. Adams) (Gastropoda). Bulletin of the 
National Science Museum, Tokyo, 7: 33-36. 

LAMARCK, J. B. P. A. de, 1804, Suite des mé- 
moires sur les fossiles des environs de Paris. 
Annales du Muséum d'Histoire Naturelle, Paris. 
3:436-441. 

LEA, H. С, 1842, Descriptions of eight new species 
of shells, native to the United States. American 
Journal of Science and Arts, 42: 1 06-1 1 2. 

LEBOUR, M., 1937, The eggs and larvae of the 
British prosobranchs with special reference to 
those living in the plankton. Journal of the Marine 
Biological Association of the United Kingdom, 22: 
105-166. 

LUDBROOK, N. H., 1941 . Gastropods from the Ab- 
attoirs Bore, Adelaide, South Australia together 
with a list of some miscellaneous fossils from the 
bore. Transactions of the Royal Society of South 
Australia, 65: 79-102. 

LUQUE, A. A., J. TEMPLADO & L. P. BURNAY, 
1988. On the systematic position of the genera 
Litiopa Rang, 1 829 and Alaba A. Adams, 1 853. In 
W. F. Ponder, ed., Prosobranch phylogeny: pro- 
ceedings of a symposium held at the 9th Inter- 



312 



HOUBRICK 



national Malacological Congress, Edinburgh, 
1986. Malacological Review, Supplement, 4: 
180-193. 

MARCUS, E. & E. MARCUS, 1963. Mesogastropo- 
den von der Küste Sao Paulos. Abhandlungen 
der Mathematisch-Natunfi/issenchaftlichen 

Klasse Jahrgang 1963, Akademie der Wissen- 
schaften und der Literatur, Wiesbaden, 1: 1-105. 

MARSHALL, W. В., 1917, The Wangaloa beds. 
Transactions of the New Zealand Institute, Well- 
ington, 49: 450-460, pis. 34-37. 

MARTIN, K., 1914, Die Fauna des Obereocäns von 
Nanggulan, auf Java. Sammlungen des Geolo- 
gischen Reichs-Museums in Leiden, 2: 107-178, 
pis. 1-6. 

MELVILL, J. С & R. STANDEN, 1901, The Mol- 
lusca of the Persian Gulf, Gulf of Oman and Ara- 
bian Sea, as evidenced mainly through the col- 
lections of Mr. F. W. Townsend. 1893-1900; with 
descriptions of new species I: Cephalopoda, 
Gastropoda, Scaphopoda. Proceedings of the 
Zoological Society of London, for 1901, 2: 327- 
460. 

MEYER, H. A. & K. MÖBIUS, 1872, Die Prosobran- 
chia und Lamellibranchia nebst einem Supple- 
ment zu den Opisthobranchia. Fauna der Kieler 
Bucht, Leipzig, vol. 2: xxiv + 139 pp. 

MIDDENDORFF, A. T. von, 1849, Beiträge zu einer 
Malacozoologia Rossica. Mémoires sciences na- 
turelles de lAcadémie Impériale des Sciences, 
St. Petersburg, 6(2): 187 pp., 10 pis. 

MONTAGU, G., 1803, Testacea Brittannica, or nat- 
ural history of British shells. London, xxxviii -i- 
606 pp., 16 pis. 

MONTEROSATO, T., 1884, Nomenclatura genér- 
ica e specif ica di alcune conchiglie Méditerranée. 
Virzi, Palermo, 152 pp. 

MONTEROSATO, T., 1917. Molluschi viventi e 
quaternarii raccolti lungo le coste délia Tripolita- 
nia. Bulletino délia Società Zoológica Italiana, 
(3)4: 1-28, 1 pi. 

MURRAY, F. v., 1969, The spawn and early life 
history of Cacozeliana granaría (Kiener, 1842) 
(Gastropoda: Cerithiidae). National Museum 
Memoirs, Victoria, 29: 111-113. 

NAIM, O., 1982, Bilan qualitatif et quantitatif de la 
fauna malacologique mobile associée aux algues 
du lagon de Tiahura (île de Moorea, Polynésie 
Française). Malacologia, 22: 547-551. 

NORDSIECK, F., 1968, Die europäischen Meeres- 
Gehàuseschnecken (Prosobranchia). Gustav 
Fischer Verlag, Stuttgart, 273 pp. 

OLDROYD, I. S., 1927, Marine shells of the west 
coast of North America. Stanford University 
Press, Stanford, 2(3): 339 pp., pis. 73-108. 

OLSSON A. A. & A. HARBISON, 1953, Pliocene 
Mollusca of Southern Florida with special refer- 
ence to those from North Saint Petersburg. The 
Academy of Natural Sciences of Philadelphia. 
Monograph 8: v + 457 pp., 65 pis. 

ORBIGNY, A. de, 1 841-1 846. Mollusques, in: R. de 
LA Sagra, ed.. Histoire physique, politique, et na- 
turelle de l'île de Cu