Full text of "Malacologia"

See other formats

red = Abr bed я À > Ale У LE i. as eh Зы и
Peer Е : ne ee a ’ A Let. y unse à - a
> sans

Env

PPT
atm wel
MI le A ee rere ET
WON tL je nd RT ET tes mis by
E AUF ET
er
vs

Am aan

: 4 Mens 40

Е В N: x aaa an

LA | Kr VA AY AJ NE ad
va

fed ei
d ol he A BEA Me nn A Le dde heiten
nan peda Bite aile Las du sae ae пре Nal me
= Ballen a toute a un ме AR de di in 1.2

z инь;
2b AGN мини атм
TEE

wh strlen che
DRE

nt mo ant a ne a tn GA mete ii

CET PRES

ne Oba А Soke hen
FOC NA

ne Ran À a an I À J > A ve
Da Bonn cta у г к Y , \ A nen A pha A
nein N $ - in a en : a h A . vi ae

eV IED AN AN ee?

HARVARD UNIVERSITY
e
Library of the

Museum of

Comparative Zoology

iA) Ni |

Ñ UY р

ÿ у

ii i | ni]
MR me de 7

ARNO:
namen arm ale

Br E

wo 0 ap me — a EEE Du eae

VOL. 35 1993

E LS

MALACOLOGIA

International Journal of Malacology
Revista Internacional de Malacologia
| Journal International de Malacologie
/ y
Международный Журнал Малакологии

Internationale Malakologische Zeitschrift

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

Publication dates

28, No.
29, No.
29, No.
80, No.
31, No.
31, No.
32, No.
33, No.
34, No.
85, No.

19 January 1988
28 June 1988
16 Dec. 1988

1 Aug. 1989
29 Dec. 1989
28 May 1990

7 June 1991

6 Sep. 1991

9 Sep. 1992
14 July 1993

а.
| LIBRARY a
JUL 21 19%

HARVARD
UNIVERSITY

if A i a
eon me 19 le Wi ь
itional Journal of Malacology =
vista Internacional de Malacologia

International de Malacologie

73 ` €
igo. Ne ASE,
Journ
be I m ПВ . is т. te j y j а “ee hs vy
x p mr Г 4 ^ x} Y 9
(| | р Е

: Lea NR \ y de | ФА. 2 N > Ñ E и awh | u hs
Международный Журнал Малакологии

РА

ar
PU
Г ~
EA h, ie a
Re ER A TVA f À Г / Ti |
a у 2 N A À ES à » + ra
as % DES y y \ { $ 4 I |
Ся С Mco: : Sí

“Internationale Malakologische Zeitschrift
: > I 8 Mea te | A р р ‘ г Je % +; N 3 je ; т Le. S т ; | д : | у é CE

DM : AE 4,

у у Г у x

i =

_ MALACOLOGIA

Editor-in-Chief:
GEORGE М. DAVIS |

Editorial and Subscneñon Offices:

Department of Malacology AI

‚Тре Academy of Natural Sciences of Philadelphia paño

_ 1900 Benjamin Franklin Parkway : 7
Re Pennsylvania 19103- > U.S.A. WG

| | Associate Editors: Baar
JOHN-B>BURCH а ed ANNE GISMANN
University of Michigan “= : A Maadi
Ann Arbor HS | Egypt - Wr

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

KENNETH J. BOSS Jr.) р; JAMES NYBAKKEN _ у. ух

Museum of Comparative Zoology Moss Landing Marine Laboratory _
Cambridge, Massachusetts : de California |
JOHN BURCH, President = = CLYDEF.E. ROPER.
MELBOURNE R. CARRIKER US = Smithsonian Institution -

Washington, D.C.

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

_ University of Delaware, Lewes

GEORGE M. DAVIS.
Secretary and Treasurer
- SHI- KUEI WU

. CAROLE $. HICKMAN - =
University of California, Berkeley. EICH Е Di na Museum, Bo

President- -Elect

| SP anopaling Members _ ES
EDMUND GITTENBERGER _ JACKIE L. VAN GOETHEM _

Secretary, UNITAS MALACOLOGICA | - Treasurer, UNITAS MALACOLOGICA is r
Rijksmuseum van Natuurlijke | = Koninklijk Belgisch Instituut = y) “ae
Historie © voor Natuurwetenschappen
Leiden, Netherlands RER N PNY Brussel, Bebe | вк

SEN Wy, Y I

; | Emeritus Members à
J. FRANCIS ALLEN, Emerita ` : ROBERT ROBERTSON

Environmental Protection Agency The Academy of Natural Sciences |
Washington, D.C. _ Philadelphia, Pennsylvania Ve
ELMER G. BERRY, к | NORMAN F. SOHL их

Germantown, Maryland | > _U.S. Geological Survey
| | ; _ | Reston, Virginia

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

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

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

A. J. CAIN
University of Liverpool
United Kingdom

P. CALOW
University of Sheffield
United Kingdom

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

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

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

B. C. CLARKE
University of Nottingham
United Kingdom

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

C. J. DUNCAN
University of Liverpool
United Kingdom

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

V. FRETTER
University of Reading
United Kingdom

1993
EDITORIAL BOARD

E. GITTENBERGER
Rijksmuseum van Natuurlijke Historie
Leiden, Netherlands

F. GIUSTI
Universita di Siena, Italy

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

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

A. V. GROSSU
Universitatea Bucuresti
Romania

T. HABE
Tokai University
Shimizu, Japan

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

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

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

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

B. HUBENDICK
Naturhistoriska Museet
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
Facults des Sciences
Brest, France

C. MEIER-BROOK
Tropenmedizinisches Institut
Tübingen, Germany

H. K. MIENIS
Hebrew University of Jerusalem
Israel

J. E. MORTON
The University
Auckland, New Zealand

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

R. NATARAJAN
Marine Biological Station
Porto Novo, India

J. OKLAND
University of Oslo
Norway

T. OKUTANI
University of Fisheries
Tokyo, Japan

W. L. PARAENSE
Instituto Oswaldo Cruz, Rio de Janeiro
Brazil

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

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

W. F. PONDER
Australian Museum
Sydney

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

QUIZE
Academia Sinica
Qingdao, People's Republic of China

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

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

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

A. STANCZYKOWSKA
Siedlce, Poland

Е. STARMÜHLNER
Zoologisches Institut der Universitát
Wien, Austria

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

W. STREIFF
Université de Caen
France

J. STUARDO
Universidad de Chile
Valparaiso

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

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

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

J. А. VAN EEDEN
Potchefstroom University
South Africa

М. Н. 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 (Ariophantidae), found т
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.

2 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 т 5 ст x 30 ст 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. 1B, symmetrical
rise and fall of flash) or asymmetrical (Fig.

A
ОЛА Л И

ln N
E F G

Dis A

——
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).

1C), and modulated flashes (with more than
one peak). In modulated flashes, an intensity
modulation produced a pulsation of light (Fig.
1D-G). Sometimes, the pulsation could be re-
solved into two discrete flashes (Fig. 1G).
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 3

A SIMPLE FLASHES
1 2 3

30 30 30
se 20 20 20
= т
= 10 10
=
2.0 4.0 6.0 8.0 2.0 4.0 6.0 8.0 2.0 4.0 6.0 8.0
и.
о
m MODULATED FLASHES
2 2 3
= 20 ! 20 20
z

10 10 10

2.0 4.0 6.0 8.0 2.0 4.0 6.0 8.0 2.0 4.0 6.0 8.0
FLASH DURATION
(SECONDS)
B
1

45 56 M:NUYES

5
u 5 10 15 20 25 30 35 40 45 50 55 60 65 70 78 80
T
es 2
< 15 63 MINUTES
u 10
и.
O 5
5
5 5 10 18 20 26 30 35 40 45 50 55 60 65 70 75 80
2

aS 31 MINUTES

$ 10 15 20 25 30 35 40 45 50 SS 60 66 70 75 80

INTERFLASH INTERVAL
(SECONDS)

FIG. 2. Flash duration for simple (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 3B 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-
опа$ (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 5
SIMPLE FLASHES

2 3
A Zi 20 70
15 15 6
(dp) 10 10 10
Ww
5 5 5 5
< o 0
a ¡LOS A Oso VOTA OA ESO о ig ao So 140 Be
u MODULATED FLASHES
O soy 1 2 3
E 3
Ww 30 30
(se)
5 25 25 25
=) |
zZ 20 20 20
15 15 15
10 10 10
5 5 5
0 - - - 5 5
© 20, 59 JO > Be 10 20 30 40 50 IAE
FLASH DURATION
(SECONDS)
B 1
= 23 MINUTES
40
35
30
25
20
15
10
5
0
o 10 20 30 40 50 60
ШУ ROS
T je 30 MINUTES
2 40
3 | 35
ww 30
te 25
20
O 15
GE 10
WwW 5
fea) 0
= 0 10 20 30 40 50 60
z ne 0
45 18 MINUTES
40
35
30
25
20
15
10
5
o
0 10 20 30 40 50 60
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 A3 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.

6 COPELAND & DASTON

snails, could not find evidence of a luciferin-
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. /n 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. B., 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., 1961, 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, 1963, Control of flashing
in fireflies. Il. 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. & М. М. DASTON, 1989, Biolumi-

eE АИ

ADULT AND JUVENILE FLASHES IN DYAKIA 76

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 (Quantula) striata. Society for Neuro-
science Abstracts, 10: 396.

COUNSILMAN, J. J., D. LOH, $. 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, М. & W. D. McELROY, 1974, Kinetics of
the firefly luciferase catalysed reactions. Bio-
chemistry, 13: 921-925.

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

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

HANEDA, Y., 1981, Luminous activity of the land
snail Quantula striata. Pp. 257-265, in М. А. DE-
LUCA & W. D. MCELROY, eds., Bioluminescence and
chemiluminescence. Academic Press, New York.

HARVEY, Е. 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. y

MARTOJA, М. & J. М. BASSOT, 1970, Etude his-
tologique de complexe glandulaire pedieux de
Dyakia strata, Goodwin et Austin, gastéropode
pulmoné données sur l'organe lumineux. Vie et
Millieu, Serie A: Biologie Marine, XXI, Fasc. 2-A:
395-452.

MASON, J. 8 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 (Coleoptera). Journal of Cell Bi-
ology 16: 323-359.

Revised Ms. accepted 20 April 1992

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: Dyakia, 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—16 mm;
N = 4) had small, poorly developed repro-
ductive systems when compared to the large
snails (shell diameter = 20 mm, М = 9). Тур-
ical small snail and large snail reproductive
systems are shown in Figure 1A and in Figure
1B, 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 1B, C. 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. 1B). 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. 1C). 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 М = 2 with “non-visible” luminescent
organ) contained mature spermatozoa. Ma-
ture sperm were identified by the appearance
of the axoneme of the flagellum in cross sec-

LUMINESCENT ORGAN AND SEXUAL MATURITY

ww | = WO | :э[е9$ ‘элодоцоб uouuo9 ‘|| ‘pue|5 yep jo jonp
‘OL ‘puel6 pep ‘6 ‘oes мер ‘8 ‘ешбел ‘7 ‘eseueuneds jo jonp ‘9 ‘eseuyeuueds ‘g ‘snyesedde jeluad ‘p опр wuads “eg ‘Yonpınowueds ‘г ‘рие!б цэшпае ‘|
‚зиоцелелаау ‘иебло S2U89SEUILUNI , 8IQISIA-UOU,, цим (WW O'Zz 1э}эшер пэцз) ¡reus a6187 ‘9 ‘иебло зизозэашишп| „algısıa,, чим (ww с-6е лээшер Jays)
leus эблел ‘а ‘иебло зиэозэциши| „algısın,, чим (шш 0'91 JeJaweıp |эц$) ¡reus jjews “y ‘вещз ‘а элщеш pue эпиэлп! jo шэ}5А$ элцопролаэн ‘| "HI

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 um (Tompa, 1984). No cells of
that size were found in the ovotestis.

Luminescent Organ: Organ of Haneda: 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, 1970). 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 um + 0.02 S.D. (N = 15) (Fig. 3C) and
2.4 uM + 0.56 S.D. (М = 15) for large snails
with a “non-visible” luminescent organ (Fig.
3D). For small snails, the average droplet size
was 5.8 um + 2.15 5.0. (М = 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 (Tauc, 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 дай 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 TEM, 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). В. High
magnification view of sperm tails in cross-section showing the axoneme (arrow) (54,300 x ).

Cellular Structure and Function of the Organ lium, a ventral simple squamous epithelium,
of Haneda and large granular photocytes surrounded by

The organ of Haneda is discus-shaped and connective fibers (Figs. 3—5). This confirms
yellow. It consists of a dorsal ciliated epithe- the morphology described by Bassot & Mar-

14 DASTON & COPELAND

FIG. 3. Luminescent organ of adult snails. A. Ciliated epithelial cells; с, cilia (35,700). В. 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). В, С, 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;
с, cilia (17,900 x). В. Border between ciliated epithelium (ep) and photocytes (sd, secretory droplets
(4,000 x ). С, D, material within the photocytes (С = 15,000 x ; D = 13,000 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 &

16 DASTON & COPELAND

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 ). В. Beneath the ciliated epithelium,
collagen fibers are seen (72,000 x ). Abbreviations: c, cilia, co, collagen fibers.

LUMINESCENT ОАСАМ AND SEXUAL MATURITY 1174

Pre-buccal Canal

Floor of
Pre-buccal canal

Foot Muscle

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

organ of Haneda = 1 mm.

Martoja, 1968; Martoja & Bassot, 1970;
Haneda, 1963, 1981). What is known is that
the luminescent substance in D. striata tests
negatively to a luciferin-luciferase reaction
(Haneda, 1963) and, from spectrophotometric
evidence that used extracted luminescent or-
gans, that the luminescent substance of D.
striata is different from firefly luciferin (Isobe
et al., 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, 1970). 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 !ocalized 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 8 Martoja, 1968; Martoja 8 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 8 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,
1984). 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 т D. striata. As yet, the behavioral sig-
nificance of the flash of D. striata remains
enigmatic.

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

ВАЗЗОТ, 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’Academie 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 Limax 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. У. CHAN, W. Н.
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 M. DELUCA
& W. MCELROY, eds., Bioluminescence and
chemiluminescence. Academic Press, New York.

HEUSER, J. E. & T. $. 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. M. 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. Etude his-
tologique du complexe glandulaire pedieux de
Dyakia striata, Godwin et Austin, gasteropode
pulmone donnees sur Гогдапе lumineux. Vie et
Millieu, Serie A: Biologie Marine, XXI, Fasc. 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, Observa-
tions produced by the Malayan gastropod Dyakia
striata. Malayan Nature Journal, 28: 173-180.

PORTER, К. & М. 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-
woop, 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 1992

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

А POPULATION STUDY OF THE BIVALVE /DAS ARGENTEUS JEFFREYS, 1876,
(BIVALVIA: MYTILIDAE) RECOVERED FROM А 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 /das argenteus was recovered
from a wood block submerged for 11 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 I.
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
I. 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. /das 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, I. 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 (with 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 &
Hinga, 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-

22 ОЕАМ

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 flocs 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 рай 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 /das 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 (1.0. number
N-17) 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 т a
plastic mesh bag to hold together the crum-
bling wood during recovery. Block N-17 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 10 x
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 11—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 50x each unit of measure
was equal to 19.4 um). 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 /das 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 from 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 А layer
of gold-palladium, and viewed using an AMR-
1000 electron microscope. The analysis of

24 ОЕАМ

)

т|

0.5mm

FIG. 3. Camera lucida drawing of an acetate peel taken from the polished surface of the valve of /das
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:

@ = (S,-S,)/S,,

where S, = Shell length at the beginning of
time interval T, and S, = 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
um long and 1.7 рт 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 pm 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 А DEEP-SEA BIVALVE 25

FIGS. 4, 5. Scanning electron micrograph of a cross section of the shell of /das 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 /das argenteus. Arrows indicate two
growth lines. ol = outer shell layer; ml = middle shell layer; il = inner shell layer; ит! = innermost growth
layer. Scale bars = 10 um.

26 ОЕАМ

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.

Shell length (mm)

Number of growth lines Minimum

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 /. argenteus 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 11 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 predation by
epifaunal organisms on the less protected
wood panels (Williams & Turner, 1986).

To determine the reproductive strategy of I.

Maximum Number of specimens
1.26 6
1.70 10
1.90 9
2.62 12
ES 15
4.29 15
5.88 24
6.85 10
7.15 1

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 пре 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 I. 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. (1990) reported that the simple prismatic
outer shell layer of this species was calcitic
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, 1980: fig. 5). These authors report that

POPULATION STUDY OF A DEEP-SEA BIVALVE 27

200

> Y

150

= D

Frequency
Pi
=
S

0 Ab
0.0

N X

119 0.93
1384 32
2126 LS
1877 2.2)
1538 2.83

608 35)

199 4.60
OZ
6.87

—
UY O0

Bees AE

1.07 2.0. 3.0, 4:0, 5.0 76.0. 7.0

Length (mm)

FIG. 6. Size frequency analysis of the wood block population of /das argenteus. 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) derived from NORMSEP. X = mean valve length of each age/growth line

class.

tropical ог warm-water mytilids generally pos-
sess shells composed entirely of aragonite.
Idas argenteus, 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 aragonitic
inner shell layer (Fig. 4, iml) 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.
А concentration of less soluble organic matrix
would occur in the region between two depo-
sitional periods resulting in what would then
be recognized as a growth line.

The innermost growth band of /. argenteus,
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 ОЕАМ

TABLE 2. Results of the growth line analysis Нот 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 /das argenteus recovered from each panel are

also included. *, See text.

Length (mm)

Number of lines

Length (mm) Number of lines

Panel N-37 11 months N = 6 Panel N-76 35 months М = 1577
1.05* 1 2.16 3
2.23 3
Panel N-39 23 months N = 221 2.23 3
2.25 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 85 months N = 363
2.18 3
1.47 1 2.33 2
1.82 2.47 3
2.61 3
2.66 2
2.86 3
2.90 3
Panel N-93 25 months N = 71
1.29 1
12452 2
lol” 2
Panel N-55 47 months N = 2068
Panel N-82 35 months N = 79 2.47 3
2.48 3
1.59* 2 2.57 3
1.82 2 2.58 2
1.84 2 2.67 3
2.72 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

| | 5 з®еюм |

Female Hermaphrodite Unripe

Poe ton | I
| © |
— © O © BR ND PB BR 01

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-

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

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 /. argenteus (Fig. 5), and this aspect
ofthe 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 I. argenteus, 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 (Linne) (Panella & Mac-
Clintock, 1968), Spisula solidissima Dillwyn
(Jones et al., 1978), Anadara granosa (Linné)
(Richardson, 1987), Mya arenaria Linné
(MacDonald & Thomas, 1980), Mytilus edulis
(Linne) (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 I.
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 I. 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. /das 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 пит-
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 I. 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 I. 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
predation 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 ОЕАМ

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

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 Arctica islandica 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 /. argenteus.

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 /das 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 & Cooday, 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., 1981).
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 enrich 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 I. 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,
1974), 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
& 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

Distance along transect (mm)

30 er)

«< Growth Line

<——— Growth Line

45 25,000 OA

Crystal size (1 unit = 0.96 um)

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, 1980). 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 А 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-
Ну 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 а 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 ripe
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 predation 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 ОЕАМ

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 8 Bull, 1977), and
protandry would not necessarily be the opti-
тит strategy in а! 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 I. 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 I. 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). /das 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 (В? = .914), Logistic (В? = .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 I. 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 McCuaig & 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 I. 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 А DEEP-SEA BIVALVE

= я arc mlidastareenteus

= RCIP DIMM Eds Е Lampsilis radiata
= ee Anodonta grandis
Е 2 5 $7 et Cerastoderma edule
< ии à à + Modiolis modiolis
© O === Spisula solidissima
qa Me

$ DA

(ab)
=
A

Oetinger:

A das o

Annual Growth Period

FIG. 8. Size specific growth rates of /das argenteus (solid line) and five species of marine shallow-water and
freshwater bivalves (dotted lines). Lampsilis radiata Gmelin and Anodonta grandis Say from data in McCuaig

8 Green (1983); Cerastoderma edule (Linné) and Modiolus modiolus (Linné) from data in Seed 8 Brown
(1978); Spisula solidissima Dillwyn from data in Jones et al. (1978).

years and a resultant very slow growth rate of
0.084 mm/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 8 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
growth 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. (1982) carried out in situ mea-
surements of growth for specimens of the
large mussel, Bathymodiolus thermophilus
Kenk 8 Wilson, 1985, 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 ОЕАМ

the hydrothermal vents are most likely the соп-
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 /. argenteus taken from the
wood block at DOS 2. Apparently, these high
size-specific growth rates for /. argenteus 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 /.
argenteus. 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 /. argenteus 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 {Пе 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 /. argenteus 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
|. argenteus possesses both of these at-
tributes.

The small prodissoconch | (length = 110
um) of I. argenteus indicates an egg size as-
sociated with bivalves possessing plank-
totrophic larvae, and the well-developed pro-
dissoconch Il (approximately 500 um) is an
indication of an appreciable free-swimming
phase (Turner & 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, /. argenteus 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, /. argenteus
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
(1987) also reported large numbers of juvenile
specimens of /. argenteus in the organically
enriched sediments surrounding the wood is-
land at both DOS 1 and DOS 2. Whereas the
larvae of I. argenteus 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 /. argenteus ensures dense settlement of
the wood island area by those larvae remain-

POPULATION STUDY OF А 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 /.
argenteus has a high reproductive potential
and is capable of rapid population increase by
dense larval settlement of an established site,
itlacks 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 I. argenteus 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-
yeres, 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, Grassie & 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 /. argenteus, 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 (WHO!).
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 рго-
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
Al 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, Bathymetric 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. А. & Н. L. SANDERS, 1973, Studies on
the deep-sea Protobranchia (Bivalvia): the fami-
lies Siliculidae and Lametilidae. Bulletin of the
Museum of Comparative Zoology 145: 263-310.

BILLETT, О. 5$. M., В. $. LAMPITT, A. L. RICE & В.
Е. С. MANTOURA, 1983, Зеазопа! 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 Marine and Freshwater Research, 25: 231—
237.

BURKE, В. 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 ОЕАМ

cal controls of bivalve shell mineralogy and mi-
crostructure. Рр. 69-113, 627-643, in: D. С.
RHOADS, & В. 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. 391-411, 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.

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

DEBRUYERES, D., Е. GAILL, L. LAUBIER & Y.
FOUGUET, 1985, Polychaetous annelids from
hydrothermal vent ecosystems: an ecological
overview. In: М. L. Jones, ed., The hydrothermal
vents of the eastern Pacific; an overview. Bulletin
of the Biological Society of Washington, 6: 103—
116.

DEBRUYERES, D. & L. LAUBIER, 1988, Exploita-
tion d’une source de matiere organique concen-
tree dans l'océan profond: intervention d'une an-
nèlide polychète nouvelle. Comptes Rendus de
l'Académie des Sciences, Paris, 30(): 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, 1927 and /da-
sola lredale, 1915. National Museum of New
Zealand Records, 3(3): 17-36.

DEUSER, W. G., Е. H. ROSS & В. Е. 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, В. R., W. J. GOULD, Р. A. GURBUTT &
Р. D. KILLWORTH, 1982, А 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-METIVIONI, A., C. METIVIER, A. HERRY &
M. LE PENNEC, 1986, 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 Malacological Soci-
ety of London, 41: 571-588.

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

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

САСЕ, J. D., 1985, The analysis of population dy-
namics in deep-sea benthos. Pp. 201-212, in:
Р.Е. GIBBS, ed., Proceedings of the 19th European
Marine 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: 3707-3711.

GALLUCCI, V. F. 8 В. В. 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. C. & J. S. PEARSE, 1974, Introduction:
general principles. Pp. 1-49, in: A. С. GIESE & J.
$. 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, А. J., 1988, A response by benthic For-
aminifera to the deposition of phytodetritus in the
deep sea. Nature, 332: 70-73.

GOODAY, A. J. & Р. 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. & C. 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. Е. & 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., & |. $. MORSE-PORTEOUS,
1987, Macrofaunal colonization of disturbed

POPULATION STUDY OF А 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, 17: 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. & В. Е. HAEDRICH, 1984,
Growth and population structure in two species of
bivalves (Nuculanidae) from the deep sea. Ma-
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,
1983, 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, Р. А. 8 Е. О. 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, K. 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
О. С. RHoaDs & В. A. Lutz, eds., Skeletal growth
of aquatic organisms.

KENK, V. C. & B. WILSON, 1985, A new mussel
(Bivalvia, Mytilidae) from hydrothermal vents in
the Galapagos Rift zone. Malacologia, 26: 253—
2:

KNOWLTON, М. & В. 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, Р. J. D. & А. 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. В. 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. C. RHOADS, 1978, Shell struc-
ture of the Atlantic ribbed mussel, Geukensia de-
missa (Dillwyn): a reevaluation. Bulletin of the
American Malacological Union, for 1978: 13-17.

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

LUTZ, R.A., L. W. FRITZ& В. М. CERRATO, 1988,
A comparison of bivalve (Calyptogena magnifica)
growth attwo deep-sea hydrothermal vents in the
eastern Pacific. Deep-Sea Research, 35: 1793-
1810.

LUTZ, В. A., L. W. FRITZ & D. C. RHOADS, 1985,
Molluscan growth at deep-sea hydrothermal
vents. In: М. 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. & В. С. SUMMERFELT, 1978, Eval-

40 ОЕАМ

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

PEARSON, Т. Н. & В. 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 & В. D.
TURNER, 1979, Reproductive pattern in the
abyssal snail Benthonella tenella (Jeffreys). Pp.
173-188, in: S. Е. 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., В. A. LUTZ, В. М. CERRATO 4
Е. С. REVELAS, 1982, Growth and predation 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, C. 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, 111: 77-98.

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

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

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

ROWE, С. 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 Marine Research, 37:
581-600.

SAILA, 5. B., С. W. RECKSIEK & M. H. PRAGER,
1988, Basic fishery science program. À compen-
dium of microcomputer programs and manual of
operation. Developments in aquaculture and fish-
ery science, 18. Amsterdam, Elsevier, 230 pp.

SANDERS, H. L., В. В. HESSLER & G. В. НАМР-
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, В. 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, В. & В. А. BROWN, 1978, Growth as а
strategy for survival in two marine bivalves
Cerastoderma edule and Modiolus modiolus.
Journal of Animal Ecology, 47: 283-292.

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

SMITH, К. L., Jr. & К. В. 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, В. 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, 1., D. $. JONES & D. DREIBELBIS,
1980, Annual internal growth banding and life
history of the ocean quahog Arctica islandica
(Mollusca: Bivalvia). Marine Biology, 57: 25-34.

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

TUREKIAN, К. K., J. К. COCHRAN, D. P.
KHARKAR, В. М. CERRATO, J. В. 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, К. K., J. К. COCHRAN, Y. NOZAKI, I.
THOMPSON & D. $. JONES, 1982, Determina-
tion of shell deposition rates of Arctica islandica
from the New York Bight using natural 228-Ra
and 228-Th and bomb-produced 14-C. Limnol-
оду and Oceanography, 27: 737-741.

TURNER, В. 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, В. 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 А 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. А. & 5. 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. Il. 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, Р. & M. А. KENDALL, 1981, Рори-
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: 1011-1026.

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 18 August 1992

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

EVOLUTIONARY RELATIONSHIPS AND EXTREME GENITAL VARIATION IN A

CLOSELY RELATED GROUP OF PARTULA

Michael $. 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 уапа-
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
hybrids between P. otaheitana and P. jackieburchi have genitalia with characteristics 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 bring 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, 1980; Johnson et al., 1986a;
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), Р. jackiebur-
chi (Kondo, 1980), and Р. 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, Р. о. affinis, the most distinc-
tive of the “subspecies,” is widely sympatric
with P. о. 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. lt 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.
2Department of Biology, University of Virginia, Charlottesville, Virginia 22901, U.S.A.
3Department of Genetics, School of Biological Sciences, Queens Medical Centre, Nottingham NG7 2UH, United Kingdom.

44 JOHNSON, MURRAY & CLARKE

striking that Р. 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 al.
1986с). Indeed, P. otaheitana, P. jackieburchi,
and P. affinis cannot be distinguished by their
allozymes or their mitochondrial DNA (Murray
et al., 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 Р. otaheitana and Р. jackieburchi. We
also compare the Partula species with two
species of Samoana.

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 (1979) and Kondo (1980). The samples
contain 24 obvious P. otaheitana, 22 P. jack-
ieburchi, 17 Р. affinis, and 45 specimens of

uncertain placement (Table 1). The sampling
localities are concentrated in the eastern half
of Tahiti Nui, the region where Р. otaheitana
and P. affinis are sympatric. All the securely
identified P. otaheitana are Р. o. rubescens,
except those from Sample 801 (Р. о. crassa
“Pease” Garrett, 1884) and Sample 778 (Р. о.
amabilis Pfeiffer, 1846). All are sinistral, ex-
cept two dextrals in Sample 778. Allthe 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, 1864) (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 upto
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 $ис-
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 ofthe anatomical traits, but two
mature offspring of that mating were dis-
sected.

Measurements

Seventeen anatomical characters were
measured in each snail (Fig. 2): length of vas
deferens (LVD), coded as 0 (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

—+ =

EA
5km

TAIARAPU

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 Раагита! 7, 4

776 Tiarei 2 1

784 Tiarei 3

786 Tiarei 1

742 Tiarei 1 9

780 north Mahaena 9 1
791 south Mahaena 7

792 south Mahaena 4 25
793 south Mahaena 3
774 Faone 3

813 Faone 3

TOTAL 24 22 Ue 45

46 JOHNSON, MURRAY & CLARKE

LVAG

LFSP

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
(АРАМС); 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 г? 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. otaheit-

48 JOHNSON, MURRAY & CLARKE

PC2

conspecifics of readily identifiable individuals. Circles

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
WPENS give P. jackieburchi negative scores,
which reflect the distal insertion of the vas
deferens into the relatively narrow penis. Pop-

PCI

FIG. 3. Results of the principal components analysis of variation in genital morphology. Polygons enclose

= Р. otaheitana; open triangles = P. jackieburchi;

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 Р. 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 РС1 РС2

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, Р.
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, WPENS, 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. jackieburchi and Р. 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. jackieburchi, and 17 P. affinis were ana-
lyzed, and the discriminant analysis тсог-
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 (г = —0.74, P<0.001) and for the
“unplaced” snails (г = —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 Р. 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

-6 -4 -2 0

DF 1

FIG. 4. Discriminant scores from the analysis of genital morphology. Symbols as in Fig. 3. Additional
symbols: P = parents for mating MJ430; Н = 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 DF1 DF2

SHLEN 0.857 —0.451
SPWID —0.039 —0.485
APWID —0.251 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-

ANATOMICAL VARIATION IN PARTULA 51

P. otaheitana

P. jackieburchi

Hybrid Hybrid

P. affinis

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 7791 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. The 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. otaheitana but genitalia of intermediate
character. Within the group of sinistrals,

scores on the first discriminant functions for
genitalia and shells are significantly corre-
lated (г = —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 Р. otaheitana.
They have large, yellow shells with a pink
apex, typical of Р. о. rubescens or P. jackie-
burchi. 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. jackieburchi. The snails
with low scores tend to have shells with some

52 JOHNSON, MURRAY & CLARKE

DF1 for SHELLS

- 6 A 0

2 = 6 8

DF1 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 813, 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 Р. 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. otaheitana,
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. otaheit-
ana, and possibly P. jackieburchi. 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

SCORE ON DF2

RE UE ее
SCORE ON DFI

CAE DD NE AD CAE

Fic. 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, Р.
jackieburchi, and Р. 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 $. 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 Р. jackieburchi and Р. 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 Р. jackieburchi and Р. affinis
(Fig. 9). The trait contributing the most to that
separation is the relative width of the proximal

54 JOHNSON, MURRAY & CLARKE

PCI

Fic. 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 5. attenuata and $. 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 Kondo’s (1980) 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 Р.
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

DF1

Fi. 9. Discriminant scores for the analysis of genital morphology in P. otaheitana (circles), P. jackieburchi
(open triangles), Р. affinis (filled triangles), and $. attenuata and $. 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 тат
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 Moorean 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. А similar situation occurs оп
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. jackieburchi 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, asample of 17 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 t0 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 from hybrids of
P. otaheitana and P. affinis. т Sample 794
from Papenoo, for example, the snails vary
from obvious P. affinis, with small, brown,
dextral shells, to Р. otaheitana, with large,
pink or yellow, sinistral shells. All the individ-
uals with intermediate genital morphologies,
however, have shells like Р о. rubescens,
with no sign of introgression from P. affinis.
Since typical Р. 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 Р. 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 Р. 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 Р. 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 Р. 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. otaheitana 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. otaheitana 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.
Philosophical Transactions of the Royal Society,
Ser. B., 335: 167-191.

CRAMPTON, H. E., 1916, Studies on the variation,
distribution, and evolution of the genus Partula.

The species inhabiting Tahiti. Carnegie Institu-
tion of Washington Publications, 228: 1-311.

CRAMPTON, Н. E., 1932, Studies on the variation,
distribution, and evolution of the genus Partula.
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 Partula otaheit-
ana. Malacologia, 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
Partula on Moorea. Heredity, 56: 319-327.

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

JOHNSON, М. 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. Malacologica, 27:
97-106.

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

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

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

KONDO, Y. & J. В. BURCH, 1983, Two amend-
ments to Crampton’s monograph on Tahitian
Partulidae. Malacological Review, 12: 79-84.

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

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

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

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

MURRAY, J., O. C. STINE & M. S. JOHNSON,
1991, The evolution of mitochondrial DNA in
Partula. 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 1992

981
6/1
611
LES
191
991
sg
591
vl
L61
861
441
DA
191
961
cv!
851
66

861
Stk
06

6rL
191
06

¿6

817
vel
Led
cOc
gel
ZH
851
vile
041
LOC
El
903

JOHNSON, MURRAY & CLARKE

835
61
953
6p!
801
951
Oc!
161
841
cal
Ole
691
vl
903
091
vcd
815
c0s
¡YA
Lvl
856
Lec
Ges
981
881
981
556
706
gel
SOC
LLC
coc
Ges
561
ce
661
061

c9
vl
v6
6v
6€
8€
ve
LE
cL
ZS
c9
05
cS
v8
99
94
05
98
09
SE
cl
08
LS
08
c9
85
SZ
29
59
09
58
99
LL
€Z
Lv
PL
LS

€9
ZL
78
05
9c
ce
61
ce
9
99
59
vv
Sv
ys
8€
GZ
es
CL
05
Oc
69
22
eS
18
Lv
09
82
19
09
6S
1974
9S
09
LS
cg
55
vy

Sp
55
99
Lv
ce
61
ec
ce
9€
Lv
Lv
05
LE
05
Le
er
vv
89
iv
Oc
8p
05
8t
c9
Ov
LE
05
СУ
vr
Lv
es
6v
95
LS
ov
97
67

21€
88c
896
OS!
c6l
291
H61
841
£6c
El?
99
183
505
ALTA
193
PAT
Сус
761
OS!
[51
Lve
056
61<
1474
Lye
083
9v€
69
817
955
pce
vis
LLE
ove
94
603
695

85
54
€l
[ra
81
SL
ke
91
87
eZ
16
v8
SP
19
95
ce
95
ce
SL
cl
Le
ve
95
Le
Ge
55
82
Lv
Ze
GZ
94
65
55
ve
Ov
Ze
Cv

Se
Sp
8
61
ge
Oc
Le
61
ov
vv
LS
er
55
Sp
Lp
8c
Se
Se
Zt
Sl
55
ce
6c
8c
Le
0€
55
Le
сс
85
8c
6c
9c
Le
05
55
ec

vs
OZ
CL
91
LE
el
Oc
Ss!
85
vl
16
52
05
59
9S
05
GE
ce
Sl
cl
Le
ce
95
65
6c
55
54
LS
Ov
82
CL
Le
95
55
Lv
95
Ov

SVE
953
Lye
pri
0/3
913
Sec
ecc
Sle
(144
991
LOC
vec
857
012
581
941
cel
Op!
061
817
941
61
gel
rel
Lee
685
961
056
183
vce
80c
041
Ebc
Ole
LZA!
+61

615
9!
9!
001
94
9S
09
19
ect
¿el
eb!
Ott
ec!
501
Lech
sol
96
26
Sl
58
bol
Olt
56
06
901
ZH
Sec
LL
LEE
091
c6l
LOL
92
90!
ZL
18
08

cee
815
965
651
ect
AA
cel
6cl
60€
€Sc
1274
015
LLE
705
LEE
953
LLC
v8c
261
811
£6c
L9¢
182
20€
19
Lec
LIE
cac
vec
LLE
885
13:74
Le
£ec
Sve
Es
89

851
gel
Lvl
Sl
ce
LL
GE
cl
82
19
18
c8
LL
94
84
ys
69
78
8?
cl
eZ
es
es
c9
82
c8
811
08
58
96
6c!
ZS
c8
98
gg
Ir
16

06
Ss!
Ss!
с
Sl
sol
Sl
gel
SL
SL
SI
Sl
Sl
Sl
Sl
Oc!
sol
Sl
gel
081
OSI
gel
gel
05
OSI
Sl
05
06
SE
06
Oc!
Oc!
0€
SL
06
Oc!
0€

Sol
gel
sel
sel
081
081
081
081
081
691
081
OS
Sol
051
591
691
691
Sol
591
961
021
051
081
Sel
sel
051
Sol
961
051
Sel
081
041
081
081
081
041
о

957
657
865
851
615
081
Lie
161
ple
886
SSE
19€
9/€
595
LSE
Cle
915
Ele
Loc
191
507
866
086
6/5
ГАЯ
895
[87
Lve
LLE
8
857
955
SOV
0743
057
89
ver

— AS NSS ee er NN Se SN TV ey Er Zr Se Ze Ze Zr Ze

ео

yun
yun
yun
yun

ye
yun
yun
yun
yun
yun
ео
ejo
ejo
ejo
ejo
ejo
ejo
yun
ео
yun
ejo
yun

VDOoNMovoor- яч юго — ON M OR © O
Le]
N
N

ЭТУ SST EdSM

LdSM CdSM GAH ENIAM

IN3AM GAdM DVA1 4537 dS1 13H0A GAY 138V N3d1 GAT ds

leus SS

‘SajqeueA JO иоцеие|аха 10} }ха} aas ‘dnouf вивиэцео 4 ayy ш Abojoydıow ¡eyuab jo sısAjeue 10) взеа “KIGNAddV

ANATOMICAL VARIATION IN PARTULA

(рапициоэ)
L6L Elk
el 071
181 651
091 ZO!
Lhe GC
pel cok
881 881
vec 821
163 evi
9pc AA
GOC Sc!
+61 GGc
Sc! Er
701 891
121 871
+61 971
Led c6l
ysl OZ!
961 Op!
Sve 911
Lo? 761
591 803
glg gel
903 081
161 orl
28! cel
cat 191
8 911
89c 011
val 161
903 vol
ELL 151
бус 003
cle 581

Ze
Lv
Lv
ve
€9
Cv
oS
95
87
vs
Sp
c8
Lv
Sy
LE
ZS
cS
05
v9
05
89
L6
6c
SY
05
Lv
ce
Or
es
Cv
ZS
Se
v9
LZ

8c
Se
er
65
95
8c
ve
8c
9
Le
9
es
9
Ov
Ov
Ov
ov
Le
Ov
Ov
LS
89
9
ve
6c
9€
ve
Lv
ve
81
05
ce
82
19

Le
vc
9
85
LS
03
49
Ge
[ra
Le
6c
05
[ra
cv
Le
0€
Le
9
LE
05
9
ZS
8
vv
6c
ve
Ge
vv
Le
8c
55
Ge
Ov
er

+61
506
941
191
9cc
161
¿61
791
181
981
$41
Сус
181
c0€
665
19
056
Ele
7:4
Zl?
505
902
Lye
9bc
ges
681
12-74
061
Se
281
c6l
895
cove
cee

ce
Le
Se
8!
€c
0€
8c
ve
ve
LE
Ge
сс
9
[ra
8
9
Se
Le
Le
05
cr
Sv
Lv
gg
6€
Ze
LS
SS
67
85
ce
55
ZS
9€

LLL
561
ces
GLE
605
291
Vi
181
TAA
856
591
LOC
Ove
096
$05
89c
Lyc
Lve
ect
461
[65
96

581
851
IS}
ist
vec
007
Lev
961
ges
903
vec
815

05
69
LEE
es
Oc!
LZ
29
v8
58
98
18
Sc!
v8
rel
891
151
18
col
09
68
606
805
6/
CL
58
921
vi
IS!
cel
18
eZ
96
901
92!

LLL
691
083
cl!
961
691
506
¿81
961
791
691
996
LLC
E85
86°
595
89
993
91
817
586
90€
681
Lob
Lhe
gel
953
6r

261
68

9+1
LLL
065
615

Oc!

gel

gel
081

Oc!
Oc!
091

gel
gel
gel

081
081
081
081
081
gel
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
gel
081
081
081
081
OSI
081
OS!
OS!
OS!
081
Oct!
sg,

Lve
ve?
ers
LOG
vsc
Sve
Lie
003
¿Lc
v9c
Ove
60€
ATA
555
616
06c
Esc
gcc
SOE
ecc
9565
965
656
¿Sc
8vc
1474
$96
Ole
857
561
Ole
c6c
Sy
vor

Se a a NTI ON i I N ID I OS

ye
yun
yun
yun
yun
yun
yun
yun
yun
yun
yun
yun
yun
oe!
oel
oe!
oel
oel
oe!
oef
oel
210
oel
oe!
oe!
oel
oel
oel
oel
oel
ое!
oel
ео
ejo

0€

NOrrNTNOOR DDO
=

c6L
c6/
c6L
c6L
66/1
661
664
664
661
c6Z
661
661
c6L
084
082
082
084
082
082
084
082
674
674
672
672
674
674
672
674
674
671
982
14:74
782

JOHNSON, MURRAY & CLARKE

Ole
581
6vt
681
651
991
9vc
61
9vc
sol
001
951
661
951
991
srl
OS!
041
cet
817
991
LZ
bob
051
691
961
[61
LOL

g1V1 951 €dSM IdSM @dSM GAH ENÍAM IN3dM GAdM DVA1 dS31 «$7 13нал GAV 13HV N3d41 GAT ds

951
cl
AA
vel
Ott
el
091
891
291
6/1
946
046
551
001
Olt
ec!
Blt
921
vel
sol
921
951
OL
851
A4!
9p!
91
rel

05
55
9
9
65
Ov
Ov
LE
0€
8S
09
19
Se
ce
55
kr
6c
[43
8
9
05
Le
ce
ge
05
05
8c
Se

6c
ce
ve
Oc
ce
9
Sy
8c
ve
8S
04
19
03
€c
Le
ce
65
6c
Ge
Se
6c
сс
9c
Le
ce
67
Se
9

gel
cv!
eck
vel
142:
Soc
LLC
181
92!
975
LEE
Cle
961
v8l
891
861
061
SIC
LOC
ces
LOL
961
sol
081
vol!
981
14074
LLC

Le
55
ce
55
ce
Lv
L

vi
Oc
9
9
€c
81
el
сс
81
SL
61
8c
el
2!
6c
Le
ce
05
Le
8c

55
Ze
65
Le
ve
05
Lv
se
LE
€c
Le
05
8
ec
ve
Oc
Lt
Oc
Le
8c
Oc
[ra
9
85
ce
9%
6c
8c

ce
9
€c
61
el
el
Sl
LL
St
LE
81
vl
vi
6c
vi
Sc
Le
81
61

0593
191
vic
981
901
est
881
056
er
20€
09
9565
v8l
915
6/1
581
rel
OSI
LOC
92!
981
OS!
Zul
Lvl
Loe
66

Ove
sz!

6c
9€
09
97
9
eS
Sc!
201
CL
vs!
Spy!
691
58
16
OZ
v9
LZ
c9
69
58
OZ
ZL
29
ol
OZ
SS
v6
CL

vec
gl
611
981
cl!
Lob
951
95
Op!
Lee
VLE
946
981
581
181
8cc
Ole
615
561
661
0593
841
Op!
961
16

orl
806
vec

LZ
6c
55
8S
9

OS!
gel
OSI
OS!
gel
081
06
081
081
05
06
Oc!
Oc!
sat
sg
06
sg
sol
Oc!
06
06
sg
05
gel
gel
06
06
gel

gel
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081
081

c9
6/1
181
LOC
041
056
6/13
181
941
607
СР
716
Lee
612
Lee
Sed
Occ
Lve
995
066
42
1444
cv!
996
vec
Lle
€9¢
Сус

| ye 2 L6Z
| ye Ss 162
| ye + 162
| ye” TE 162
IE E 162
ет L6Z
PE 2 772
zer ey
Вр УД
L yun € 564
р ЖП © 664
Г yun | 56/
L yun 16 262
L yun €G 262
L yun 26 264
L yun 15 262
р yun 09 264
L Yun У 262
р м Er 262
р yun It 262
L yun 68 262
Е ЖП 96 262
Z un Ge 264
L yun pe 262
| ye 05 262
се № 9 ez
L yun 12 264
L ye 0 с

EUS aus

(рэепициоЭ) XIGNAddV

ANATOMICAL VARIATION IN PARTULA

956
[972
191
951
603
861
c6l

22!
68c
581
9cc
AA!
El?
Oc!

961
461
60
923
60
69}
191

Lye
€S¢
871
IS}
803
991
01

95
ZS
c9
09
59
69
LE

6S
28
er
vv
er
9
49

09
8p
es
gg
6€
05
Lv

87
vs
6c
ce
Le
95
6c

9
er
Ov
er
65
vv
St

vv
Se
62
ce
05
ve
ec

116
vs!
661
505
€cc
Сус
191

9741
806
181
LZL
Lhe
041
col

er
[ra
Lv
Sy
€v
55
6c

ec
Se
ce
81
vy
8€
Se

ec!
611
981
681
1425
8rL
6rL

Lvl
003
gel
val
08c
LES
603

LS
Lv
18
68
55
79
58

v9
89
LS
ZS
€S
59
ec

9€€
Sec
ccc
v9
015
c6c
041

+86
555
rel
gel
961
661
93

06
Gt
Stl
StL
Ol
Sl
gel

gel
St

Oc!
051
Oc!
gel
gel

081
081
081
OZ!
081
041
081

OSI
081
081
Oc!
081
081
sat

[2745
vie
546
695
6/6
Lee
59+

cle
2956
913
cal
LLE
781
891

NA mm Ts

Ce rr

El
ld
15
15
LA
210
fe) =|

с
|
€
с
|

LE
Ol

Lev
Lev
DErWN
057И
OE VÍA

108

082

збицеш Алозелоаел

ejo
yun
yun
yun
yun
yun
ye

el
cl
ZI
91
LE
6

761
76/1
518
518
518
084
16Z

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

GENITAL MORPHOLOGY OF CARACOLLINA LENTICULA (MICHAUD, 1831),
WITH А NEW PROPOSAL OF CLASSIFICATION OF HELICODONTOID GENERA

(PULMONATA: HYGROMIOIDEA)

Carlos E. Prieto, Ana I. Puente, Kepa Altonaga 8 Benjamin J. Gomez

Department of Animal Biology and Genetics, Faculty of Sciences, University of the Basque

Country, Р. О. Box 644, 48080-BILBAO, SPAIN

ABSTRACT

The genital system of Caracollina lenticula (Michaud, 1831) has been studied in many Бепап
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 (1918). 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, Trissexodontidae, Caracollina, anatomy, morphology, classifica-

tion.

INTRODUCTION

Caracollina Beck, 1837, is a typical Medi-
terranean genus; its unique species, С. 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 Oestophorini
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

64 PRIETO ET AL.

FIGS. 1, 2. Shell microsculpture of Caracollina lenticula. (1) Protoconch; scale, 100 am. (2) Protoconch and

first whorl of the teloconch; scale, 50 pm.

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 С. 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 Jerica (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 jm 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; pl.
15, figs. 15-17); Moquin-Tandon (1855, t. Il:
109; Atlas: pl. 10, figs. 15, 16); Haas (1929:
241, fig. 74); Germain (1930: 236; pl. 3, figs.

69-71; pl. 12, figs. 355, 356); Nobre (1941: 85;
pl. 15, fig. 9; pl. 16, figs. 4-6); Zilch (1960: 693,
fig. 2418); Gasull (1965: 59); Backhuys (1975:
223; р. 27, figs. 79-80); Gasull (1975: 103; р.
3, fig. 31bis); Mateo (1978: 13; fot. 14); Ker-
ney, in Kerney et al. (1983: 304 + fig.).

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 whorl and is characteristically
sculptured by small, regularly 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: 49);
Giusti (1970: 102; pl. 14, figs. 1-3).

Genital System

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

MORPHOLOGY OF CARACOLLINA 65

Ortiz de Zärate & Ortiz de Zärate (1961: fig.
3); Giusti (1970: fig. 20); Nordsieck (1987: 30;
fig. 22); Schileyko (1991: 208; fig. 8—XVIII).

Description (Figs. 3-7, 12): Right ommato-
phore 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 U 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 Morphologies (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 (Caceres
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 “ар-
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.

23
Zz
CA
ES

NAS

FIGS. 3-7. Genital system of Caracollina lenticula. (3) Dalías (Almería, WF1174). (4) Tavira (Algarve,
PB2011). (5) El Villar (Huelva, PB9974). (6) Dart from a specimen of Jerez de la Frontera (Cádiz, QA5163).
(7) Scheme of the stimulatory organ. Abbreviations: as, accessory sac, b, bursa copulatrix; bd, bursa

copulatrix duct; а, dart; ds, дай 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. 8-11. Defective genital systems of Caracollina lenticula. (8) Porcuna-Bujalance (Jaén, VG9492),
without mucous gland. (9) La Guardia de Jaén (Jaén, VG3977), without mucous gland or accessory sac. (10)
Plasenzuela (Câceres, 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.

в
E

FIG. 12. Two histological sections of the genital atrium and penial distal region of a Caracollina lenticula
specimen from Jérica (Valencia, YKO620) (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 um.

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 1$ 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. 13d-f); close to the junction
of both ducts (approximately, 25 um 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. 139). 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 C. lenticula

As it has been stated above, the genital
system of C. 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 дай 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 1$ that de-
picted by Moquin-Tandon (1855, t. Il: 109):
“Point de poche a dart. Une seule vésicule
muqueuse simple, vermiforme, flexueuse, a
peine renflée au sommet (...). Vagin assez
developpé, 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 multi-
fide simple, vermiforme, flexueuse (...); pas
du sac du dard.” Although Moquin-Tandon
stated that there is no dart sac, he mentioned

MORPHOLOGY OF CARACOLLINA 69

FIG. 13. Microscopical sections of the vaginal structures of Caracollina lenticula 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. (9—1) 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 pm.

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 РНЕТО ET AL.

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

Appendix Mucous gland

PRESENT BIFURCATE
SIMPLE

ABSENT BIFURCATE
SIMPLE

References and searched localities

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
O. Zärate & O. Zärate (1961): La Räbida (Huelva)
Soos (1933)(+): Maltese Islands

Moquin-Tandon (1855): S-France

Germain (1930)(*): S-France

(+) taken from Ortiz de Zärate & Ortiz de Zärate (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 allthe 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 Jaen 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 по

accessory sac near the dart sac, although
there is a dart sac appendix. Finally, Schi-
leyko (1991) emphasized that Caracollina
“posseses а 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; Gömez, 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 ОЕ 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 Ропо, 1836; and Mastigophal-
lus Hesse, 1918), as well as genera lacking a
dart sac (Helicodonta Férussac, 1819; Cana-
riella Hesse, 1918; Caracollina; Soosia Hesse,
1918; and Trissexodon Pilsbry, 1895), plus
some incertae sedis (Helix buvignieri Mi-
chaud, H. hispanica 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 (1931),
and that Drepanostoma and Lindholmiola
Hesse, 1931, do not have а 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, Atenia Gitten-
berger, 1968, Soosia, and perhaps Caracol-
lina would constitute the second.

Schileyko (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 Schi-
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 а 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, 1949
(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
Schileyko: the “Ciliellinae” of Nordsieck
(1987), minus Ciliella, should be named Tris-
sexodontinae, and the “Ciliellidae” of Schi-
leyko (1991), minus Ciliella, should be named
Halolimnohelicidae.

72 PRIETO ET AL.

Nevertheless, we disagree with Nordsieck’s
diagnosis for Oestophorini and Caracollinini.
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); Caracollinini, 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 Caracollinini agree
with the one for Trissexodontini and, thus, Мог-
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-!Ш and
5-Ill, and to the columellar muscle in his figs.
8-| and 9-1. Moreover, the penial retractor mus-
cle reverses once more to appear attached to
the diaphragm in his figs. 8-П (scheme of ev-
olution of the Ciliellidae) and 9-II (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, for Atenia; Prieto, 1986: fig. 7B,
Gittenberger et al., 1970: fig. 183, and Nord-
sieck, 1989, for Helicodonta; Schileyko, 1971:
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; Falkneria 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 CARACOLLINA 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 зас; 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 зас (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: \berian Peninsula,
northwest Africa and ?Macaronese Islands.

74 РНЕТО ЕТ 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, 1961;
Gasulliella 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 (1991), 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 п = 28-29 and the
Monadeniinae (Xanthonychidae) has n = 29.
The most common number appears to ben =

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 п = 23-26 (Trichiinae and Eu-
omphaliinae) and п = 21, 23-26 (Hygromii-
nae) (Patterson & Burch, 1978; Арапсю,
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, /berus
Montfort, 1810 (Garcia San Nicoläs, 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
(Caracollina, 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. |. Puente,
and by the “Fauna Ibérica Il” 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 aspersa injects a pheromone
that decreases courtship duration. The Journal of
Experimental Zoology, 255: 80-87.

APARICIO, М. Т., 1983, Estudio morfolégico y ci-
totaxonómico de algunos Helícidos 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 8 Meesters, Amsterdam, 350
pp., 97 map., 32 pls.

CULLING, C. F. A., 1974, Handbook of histological
and histochemical techniques. Butterworths,
London, 712 pp.

76 РНЕТО ЕТ AL.

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

GARCIA SAN NICOLÄS, E., 1957, Estudios sobre
la biologia, la anatomia y la sistemätica del
género /berus Montfort, 1810. Boletin de la Real
Sociedad Española de Historia Natural (Biolo-
gia), 55(2/3): 199-390 + 29 läm.

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

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

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

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 & Т.Е. RIP-
KEN, 1970, De Landslakken van Nederland.
Koninklijke Nederlandse Natuurhistorische Ve-
reniging, Amsterdam, 177 pp.

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

GÖMEZ, B. J., 1991, Morphological and histologi-
cal study of the genital ducts of Cryptazeca mo-
nodonta (Pulmonata, 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 malacolögica terrestre y de
agua dulce de Cataluna. Trabajos del Museo de
Ciencias Naturales de Barcelona, 13: 1-491.

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

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

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

KERNEY, М. P., В. А. О. CAMERON & J. H. JUNG-
BLUTH, 1983, Die Landschnecken Nord- und
Mitteleuropas, P. Parey, Hamburg und Berlin,
384 pp., 24 pls.

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 4 Е. GUISTI,
1989, New data on the systematics of two Sicilian
land snails, Helix parlatoris Bivona, 1839 and He-
Их гетае L. Pfeiffer, 1856 and description of
Schileykiella n. gen. (Pulmonata: Hygromiidae).
Journal of Conchology, 33: 141-156.

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

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

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

MOQUIN-TANDON, A., 1855, Histoire naturelle
des mollusques terrestres et fluviatiles de
France. Т.И + Atlas. J.B. Baillière, Paris, 648 pp.
+ 92 pp., 54 pls.

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

NORDSIECK, H., 1986, Das System der tertiáren
Helicoidae Mittel- und Westeuropas (Gas-
tropoda: Stylommatophora). Heldia, 4(1): 109—
120, pls. 15-17.

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

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

ODHNER, N. H., 1931, Beitráge zur Malakozoolo-
gie der Kanarischen Inseln, Lamellibranchien,
Caphalopoden, Gastropoden. Arkiv for Zoologi,
23A(14): 52-115. р

ORTIZ DE ZARATE, А. & А. ORTIZ DE ZARATE,
1961, Moluscos terrestres recogidos en la pro-
vincia de Huelva. Boletin de la Real Sociedad
Española de Historia Natural (Biología), 59: 169—
190.

OUTEIRO, А. M., 1988, Gasteröpodos de О Courel
(Lugo). Tesis Doctoral. Universidad de Santiago,
Santiago de Compostela, 627 pp., 1 läm.

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

PRIETO, C. E., 1986, Estudio sistemätico 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 CARACOLLINA UL

Cryptosaccus asturiensis n. gen., n. sp. Archiv
fúr Molluskenkunde, 123.

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

PUENTE, А. 1., С. E. PRIETO & К. ALTONAGA,
1990, Nuevos datos sobre la distribuciön 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, М. A. & М. 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, 117: 384 pp. Lenin-
grad.

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

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

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

VERDCOURT, B., 1981, Contributions to the
knowledge of the Helicidae-Bradybaeninae of
Zaire (Mollusca, Gasteropoda). Revue de Zoolo-
gie Africaine, 95(3): 525-556, pls. 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 HELICELLA CANDICANS (GASTROPODA,
HELICIDAE) AND ITS POSSIBLE ADAPTIVE SIGNIFICANCE

Alois Hon&k

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

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 тт). In Bo-
hemia, western Czechoslovakia, it inhabits
dry steppes on calcium-rich soils, particularly
on the southern slopes of hills along the Ohre
(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. | have developed
a system that enables the degree of melan-
ism of the shell to be classified. | 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, Н. candicans was collected
at 184 sites т 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-ventrally 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
ofthe 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 ofthe dorsum and with a lateral belt
consisting of four or five fused bands were
termed “dark” forms. Individuals having a thin
dorsal band оту were termed “light” forms.

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

80 НОМЕК

FIG. 1. Variation in shell banding pattern in H. can-
dicans. 1-2, light and dark shells viewed dorsally.
3-7, shells with different numbers of lateral bands.
8-12, 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 melanisation.” This index was
calculated as follows. The dorsal band width
was scored аз: < 0.15 тт, 0.15—0.39 mm,
0.40—0.69 mm, 0.70-1.00 mm, ог > 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 10 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 Sladek (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 authors
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

млин

MELANISM IN THE LAND SNAIL 81

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, TiSice, Brandys nad Labem, Lysá. Insert: Beroun, Kladno).

(3) The areas of frequent occurrence of fogs (shaded).

factory in Krälüv Dvür 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 (IOM < 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 IOM, 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. | 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 (IOM = 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

НОМЕК

LEl
O gu

Sel

bel SOU

0.99%

MELANISM IN THE LAND SNAIL 83

“i 60 W o 200 W

: 0009009
000
00000 „0 909
= 1000009899 20 90°
> | RE EE ат
Lid
œ Foo
a 2 000
= O
г” 0° 0,000” 0%e
ne 12°C
1 One®
O
200888880800 | ¿*
IA] Ф

FIG. 4. Vegetation cover and shell melanisation. Top: Density of plant cover DEN and index of melanisation
IOM, regression y = 0.413x + 7.84, t = 2.767, р < 0.01, coefficient of determination г? = 4.0%, p<0.05.
Bottom: Average height of the plant stand and proportion of dark individuals MEL in populations, regression
у = 0.015х + 0.807, t = 1.764, coefficient of determination Г? = 1.7%, n.s. Symbols: O 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 (IOM) 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: IOM <8.9, with central spot: 9.0 <IOM
<10.9, solid: IOM >11.0. Localities included: 1. РИМаку, 2-3. Stroupet, 4. Zatec, 5. Lenesice, 6. Mila, 7-9.
Rana, 10-11. Chraberce, 12. Chozov, 13-15. Dobroméfice, 16. Zidovice, 17. KoSetice, 18-21. Kfesin, 22.
Dubany, 23-25. Libochovice, 26-27. Klapy, 28. Radovesice, 29. Zabovfesky nad Ohri, 30. Brezany nad
Ohfi, 31-34. Doksany, 35-37. Libochovany, 38-39. Velké Zernoseky, 40. Zalhostice, 41-44. Litoméfice,
45. Velky Ujezd, 46. KfeSice, 47. Encovany, 48. Polepy, 49-51. Vrutice, 52. Ho$t'ka, 53. Brzänky, 54.
Kochovice, 55-59. Steti, 60-61, Вадоий, 62. Cakovice, 63. Stra&i, 64-66. Pocepice, 67. JeSovice, 68—69.
Libéchov, 70. Vehlovice, 71. Melnickä Vrutice, 72. Мау Újezd, 73. Vavïineë, 74-75. Kelské Vinice, 76.
Tuhañ, 77-80. Типайзке Vétrusice, 81-83. Cervená Piska, 84-86. Privory, 87-88. Nedomice, 89-91.
Drísy, 92. BySice, 93. Себейсе, 94. Konétopy, 95-97. Sudovo Hlavno, 98-100. Kostelní Hlavno, 101. Krpy,
102. Skorkov, 103. Тийсе, 104. Pferov nad Labem, 105-109. Semice, 110. Roudnice, 111. Ctinéves, 112.
Kostomlaty pod Ripem, 113-115. Libkovice pod Ripem, 116-117. Nové Ouholice, 118. Micechvosty, 119.
Uzice, 120. Velika Ves, 121-122. Praha, 123. Slaviky, 124-128. Suchomasty, 129-132. Vinafice, 133-137.
VSeradice, 138. Liteñ, 139-140. Korno, 141-145. Méñany, 146-151. Tobolka, 152-155. Koledník, 156.
Jarov, 157-159. Tetín, 160-167. Beroun, 168—174. Srbsko, 175. KarlStejn, 176-177. Hlásná Trebáñ, 178.
Мойпка, 179. Мойпа, 180. Bubovice, 181. Lodénice, 182. Vrbice, 183. Vikov pod OSkobrhem, 184.
Hrad£any. The localities are designated with names of the nearest village.

84 НОМЕК

o o AS e see anes =
10 e 2 ses ee ее
= o oe ооо о Be
N оо ae EE
2 -000 8 ТТ ez e » a
оо ee 00 ee 9688 Oo =
= 90900 Oo 80 6
pe E oe ® E e a
3 ®
A A рвы Wr A O A AA
1 2 3 4 5 6 1
15 DEN
a
10
o O i
=] > =
= .
5 . = . .
e e
o o s O e o
ye ove oi à =
o ® Фо Eee e B Bm
e o POR &
0 я B 13 = B ee Ss [|

0 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) lamp, at 26°C (above) and 12°C (below) ambient temperatures. The circles indicate differences in
within-shelltemperature read every 30 s from the start ofthe 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 E
White background surface
12.1°C 7.9 5.3 2.1 1.5
+0.5 +0.4 +0.1 +02
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 a +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 т-
ferred from analogy with other helicids (e.g.
Wolda, 1969), and here | 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. | 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 Stfedohorí
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 inthe 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 рор-
ulations near the cement factory in Krälüv
Dvur (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 т-
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,
1981a; Livshits, 1981; Nevo et al., 1981; Em-
berton, 1982; Nevo et al., 1982; Heller & Ga-
dot, 1984; Ramos, 1984, 1985; Sacchi, 1984;
Vicario et al., 1988; Hazel & Johnson, 1990).
| 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, particularly 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
particularly sensitive to overheating and des-

86 НОМЕК

iccation when active, and there is а 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. | have no data
on mortality, but | 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 |
suppose that a great proportion of variation is
random.

ACKNOWLEDGMENTS

| thank Prof. А. J. Cain of the University of
Liverpool, and two anonymous reviewers for
critical reading and valuable comments on the
MS, and Martin Vakar 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. Biological Journal of the Lin-
nean Society, 14: 335-358.

CAMERON, В. А. 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 Theba pisana (Pulmo-
nata: Helicidae), Heredity, 65: 123-126.

COWIE, В. Н. & J. $. 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 Partula 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. М. & М. $. JOHNSON, 1990, Microhab-
itat choice and polymorphism in the land snail
Theba pisana (Müller). Heredity, 65: 449—454.

HELLER, J. & M. GADOT, 1984, Shell polymor-
phism of Theba 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 the 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., В. 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, КИС Ceskoslovenskych mékkysu
[Key to Czechoslovak Mollusca]. Vydavatelstvo
Slovenskej Akademie Мед. Bratislava.

NEVO, E., C. BAR-EL & A. BEILES, 1981, Genetic
structure and climatic correlates of desert land-
snails. Oecologia, 48: 199-208.

NEVO, E., C. BAR-EL, A. BEILES & Y. YOM-TOV,
1982, Adaptive microgeographic differentiation of
allozyme polymorphism in landsnails. Genetica,
59: 61-67.

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

RAMOS, М. A., 1985, Shell polymorphism in a
southern peripheral population of Cepaea nem-
oralis (L.) (Pulmonata: Helicidae) in Spain. Bio-
logical Journal of the Linnean Society, 25: 197—
208.

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

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

SLADEK, I., 1977, Studium geografického
rozlozeni potenciálu znecisténi ovzdusí na üzemi
CSR. [Geographic 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, A., L. I. МАХОМ, A. AGUIRRE, A. ES- oslovakia]. Ustredni spräva geodezie a kar-
TOMBA & C. 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. Genetica, 40: 475—
217-227. 502.

VESELY, A. ed., 1953, Atlas podnebí 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 ARIANTA 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 0 to 4.44 т (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
0 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

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-
toperiod 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; B. 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, Switzerland (47°28'N, 7°34'E; altitude
360 m a.s.l.). А 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) А. 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 11 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 (г = 0.998,
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 fairly
evenly over the period and occurred on 10 of
the 16 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 В. Baur (1986, site A).

DAILY MOVEMENTS AND DISPERSAL IN А LAND SNAIL 91

In September 1981, 60 А. 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 1m? units (11
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 т 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 variates generated from the ex-

ponential distribution of daily distances cov-
ered (Fig.1a) were assigned to a random di-
rection (derived with an accuracy of 1° 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 (c.f. B. 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. 1a), 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,
1962; B. Baur, 1984, 1986; 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 0 to 4.44 m (median value: 0.58
m), and their frequency distribution fitted a
function with exponential decay (Fig. 1a). A
proportion of the snails (28.6%, Fig. 1a) 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

30 A 30 B
РО М = 119 20 N=45
>
O
=
D
=
о
= 10 10

O STO 207725035 O 0.5 ВОВЕ

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

ence

The mean distance covered per day (all snails
considered) was positively correlated with
daily minimum temperature (r = 0.59, n =
16, p = 0.016), and negatively correlated with
the number of sunshine hours (г = —0.76, п
= 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-

= 0.79, t = 6.74, df = 12) р < 0.001; (b) у = 55.060 е- 027,7 = 0.88, + — 6 00nd tas:
р < 0.01; x = distance in cm and у = frequency (%).

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

In the vegetation belt, the recovery rate of
marked А. 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 0 to
1.57 m (median = 0.40 m) (Fig. 1b). 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, п = 45, р > 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 А LAND SNAIL 93

IKKKKKkKkKkkkkKKK
==

ААА

FIG. 2. Representative movement tracks of individuals of A. arbustorum in (a) а clearing and (b) а vegetation
beit (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 А. arbustorum is influenced by the
form of the habitat.

In the second experiment performed in the
vegetation belt, 13 out of the 92 marked А.
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 120 days), (3) the structure ofthe 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

а
) Two-dimensional
IS habitat
Linear habitats:
10 (8 m)
(1 m)
ES 5
Е
3
9) 0
5 0 20 210 60 = 0" “100 ED
©.
un
mi
"9
D
о b) |
= Two-dimensonal
S 10 habitat
Y
A Linear habitats:
(8 m)
(1 m)
5
0
0 20 40 60 80 100 | 120
Time (days)
FIG. 3. Simulated dispersal of snails in habitats of different form for periods of 10, 20,...., 110, 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 А LAND SNAIL

“aul e ul paseajas = 7 ‘эи$ EUIIO Je paseajaı AENPIAIPUI = © ‘JUIOd |едизо e ye paseajas = 9,

youp e биое
Арп}$ juasaud O 89 (LFL)EL [si] a9 syyuow OL | ‘ainysed auidjeqns pueyaz}ims wnsojsnqie “Y
(e:93)6p [< 62 syjuou € эбедлэц yBnoi
(0661) пез 3 ıneg O SS (0've)99 [oi] ez yjuow | 95 ‘эблэл epispeoi uapems winojsnqie “y
(6 1£)691 = [pl] 22 syjuow € abequay ybno,
(0661) ¡neg 3 ıneg O 193 (9'/e)661 [8] 5+ you | с ‘эблэл apispeos уэрэм$ шпаодпаие “y
pueg ээ/5$ e
(9861 ‘7861) ıneg O 6S (s'p1)8z [91] 0°2 JeaÁ | | Buoje ejyeu jo Jaq риерэ2им$ winsojsnque “y
(S'Ly)es [Zt] Lip weak |
(ors)eoı Liilee SHOOM El мореэш pazeıb
(2961) WeYypooy 7 02-01 (s'ag)ger [2] 02 SHOOM y 8 ‘yueq e jo adojs puejBug MENO)
:SJe]Iqeu Jeauı
ebequay ybno,
Арп}$ juasaud O e~ (0'Sz)9 lol re yyuou | — “Buleaj9 pueyaz}IMS winsojsnqie “y
(9'82)vz [LL] 6+ syyuow € эбедлэц uybnoi
(0661) ıneg $ ıneg O e Le (Z91)b1 [LL] ee you | — ‘бимеэр цэрэм$ шпаодзпаие “y
jeusjew 991095 YM
(9861 'v861) 1neg O v7 (2:81)81 [ez] 021 1еэ^ | = ‘puejsseuß эм@е риерэгим$ WNJOJSNGUE вивиу
(9261) ‘Je ja uosjbuag 9 = (0'61)8p [s’zJı 2 syuow Z == мореэш quay pue]89] sısuayoy “Y
(7/61) poomusaly эщеи pue
Aq payo ‘(2961) Áeuny O yl — OL weak | — sse6 био] jo eave ривбиз 5иелошаи ‘7
saysng paayeos
(O'ELEL [29] ı ze sıeak с = ‘мореэш
(1561) sayauyos 5) L> (OELEL [9p] 8'52г syjuou 9 — payeanjnoun Аиешиэ9 5/елошаи ‘7
(1561) эцошел 9 — (0'SL)0£ [oz] L'8 SUJUOW G — мореэш quey ээие1- 5/елошаи “9
(1561) эцошел 9 oz (O'v)8 [62] /'6 sieak г — uspeb ээие14 si/esowau eaeday
:SJe]Iqeu |PUOISUBLUIP-OM |
80JNOS .0S29/91 („Ww/synpe рэллАеээл [шпиихеш] aseajal (ш) цоцезэбэл Ayıle9o7] salsods
jo adÂj Jo ‘ou) SIIBUS jo уеэш (Lu) Joye U}PIM yeyiqey

Áysuag (%) JequinN = jesuedsiq owl} зенаен

“syeyiqey Jesu pue [еиовиэилр-ом} UI SIIEUS рюнац jo saivads ээлц} ul jesuadsıp jo Алешштс ‘| AIGVL

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 ехреп-
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; В. 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 ге-
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 А 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, С. H., 19886, The dispersal of Cernuella
virgata (Mollusca: Helicidae). Australian Journal
of Zoology, 36: 513-520.

BAKER, С. Н. & В. G. HAWKE, 1990, Life history
and population dynamics of Theba 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, B., 1984, Dispersion, Bestandesdichte und
Diffusion bei Arianta arbustorum (L.) (Mollusca,
Pulmonata). Ph.D. Thesis, University of Zürich.

BAUR, B., 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, B., 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, B. & C. 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. Ill. 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, В. 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, В. А. 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
pulmonate slug Limax 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, В. H., 1954, The activity of slugs. |. The
induction of activity by changing temperatures.
Journal of Experimental Biology, 31: 165-187.

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 Muller. Ph.D. Thesis, University
of Manchester.

EDELSTAM, C. & C. PALMER, 1950, Homing be-
haviour in gastropods. Oikos, 2: 259-270.

ENDLER, J. A., 1977, Geographic variation, speci-
ation and clines. Princeton University Press,
Princeton.

FORD, D. J. G. & A. COOK, 1987, The effects of
temperature and light on the circadian activity of
the pulmonate slug Limax pseudoflavus Evans.
Animal Behaviour, 35: 1754-1765.

FROMMING, E., 1954, Biologie der mitteleuropäis-
chen Landgastropoden. Duncker & Humblot,
Berlin.

GOODHART, С. B., 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, М. S. 8 В. BLACK, 1979, The distribu-
tion of Theba pisana on Rottnest Island. Western
Australian Naturalist 14: 140-144.

KERNEY, M. P. 8 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. 4 S. E. В. BAILEY, 1989, The
effects of environmental factors on slug behav-
iour. In 1. HENDERSON, ed., Slugs and snails т
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.). Netherlands 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, C. D., 1982, The regulation of activity in
populations of the terrestrial slug Limax maximus
(Gastropoda: Limacidae). Research in Popula-
tion Ecology, Kyoto, 24: 1-32.

SCHNETTER, M., 1951, Veranderungen der ge-

netischen Konstitution in natürlichen Popula-
tionen der polymorphen Bänderschnecken. Zoo-
logischer Anzeiger, Supplement 15: 192-206.

SPEISER, В. & 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 Monadenia 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-ókologische Mitteilungen, 4: 283-298.

Revised Ms. accepted 28 October 1992

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 (L.A. INRA) Université de Rennes |,
Campus de Beaulieu Av. du General Leclerc, 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 Reunion, 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-
04$ 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 (heritabilities and

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 primarily 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 Balearics, and
after the winter activity for snails from Algeria;
the annual activity rhythm of snails from La
Reunion 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), Surgeres (central-west),
Toulouse (southwest), Belmont (east), Lyon
(central-east), Avignon (southeast), and Bas-
tia (Corsica). Colonies from Lorient, Belmont
and St. Denis de La Reunion 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 Reunion, a vol-
canic island of the Mascarene Archipelago
(Indian Ocean), a sample from Palma de
Mallorca (Balearics), 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;

‘J. С. Bonnet, Institut National de la Recherche Agronomique, Domaine du Magneraud, Surgéres.

GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS ОЕ HELIX ASPERSA 101

123456789 101112

Surgères
(01-04-1983)

123456789 101112

Belmont
(02-05-1983)
(29-05-1985)

123458 789101112

Lyon
19-05-1985 100P
( У )

Avignon
(0904-1983)
0

1234567 8 9101112

Bastia
19-04-1983)
TI 120P
|

st. с
(15-05-1985)

1234568789 101112

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;

—F1 generation: offspring of random
crosses between individuals of the AS gener-
ation;

—F2 generation: offspring of random
crosses between individuals of the F1 gener-
ation.

In the laboratory, snails of the AS generation
remained into an artificial hibernation (5+1°C;
60+5% R.H.; OL:24D light cycle) for one week,

102 MADEC & DAGUZAN

except for samples from La Reunion 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 18 x 7 cm) with a biomass
density per cage of approximatively 18 kg/m?
(13-15 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. Surgéres or Lorient (Da-
guzan, 1981; Le Guhennec 8 Daguzan, 1983).
For snails from Algeria, which are larger, the
density was 30 kg/m? (8-10 individuals т
small boxes). Atleast 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+1*C; R.H.=100%; 12L:12D).

For the JS, F1, 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°С; 60% R.H.; OL:24D light
cycle). Revival was triggered in a room at
12°C, 80% В.Н. 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,
F1, 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 arıd 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/Surgeres; 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 а|-
ways low, but this sample was relatively close
to the group Lorient/Surgeres. 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 < Surgeres, 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—
13 weeks) were shorter than in others (14—16
weeks).

Over three generations in the laboratory
(JS, F1, 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 Reunion
AS-F2, P = 0.05, N.S. for the other compar-
isons). Frequency distributions of matings
and layings per week in the F1, 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 Clutches:
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, Surgeres; 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

40 Lorient 40 Surgères
30 30
% 20
10 10
0 0

a
tp
©

2456-67. 1819510111213 8145151617 18 19 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
40 Belmont 40 Lyon
30
% % 20
10
| |
1203 4115 6: 71 8119 10 1213) 1481S 716/17 18019 1 23450678 9101 1213 16 15 16 по
40 Toulouse 40 Avignon
30 30
% 20 % 20
10 10
0 0
I 34 Si 67684910211 1243. 141516717) 718219 1234506078910 12 13 14 15 1691819
40 Bastia 40 Palma
30 30
% 20 % 20
10 10
0 0
142345. 6:7) 18: 910 110123014152 1617/1819 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
40 St-Denis 40 Alger
30 30
% 20 % 20
10 10 |
0 0

1-23-54 SG) J 185901011 11213214 15/16 1718519) 1 2 23.1455 22 23 24 25 26 27 28

Time (weeks) 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

Palma

BR
=, Toulouse
A
И.
Lorient
40 AR
% / a
a
0 Belmont
IZ 30450 6 Zus 9: 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 Р < 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 F1, 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 ifthe
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 F1 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 ofthese 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 12 weeks of repro-
duction, by about 60 clutches for 45 individu-
als, corresponding to 1.3 clutches per snail.

MADEC & DAGUZAN

106

passaidxa) sjeus jo uIBuo eu} 0} бирлоээе ‘зиоцелэиеб 24 pue ‘| 4 ‘Sf ul (payoyeyssos

(syoom) JUN L (syaom) JUN L

ISIDORO ЕО WEIL LOVE

TA-UOLUNIY e] os ан ua
ИИС 3 EI SEE TE Ито Gy ЕС
%

СЯ-э5 по по, 0s TA-3SNO[NO J
ОиСТ Об 3) JL OS У Is 7G COVE EV GET VOMIG AS LAOS С

ZA-JUILIO"] 0S

ТЯ -3 4914071

‘(uoneieue цоеэ ицум зиеиз Jo злэашпи |е}о} ay) jo % se
9) uonisodino pue (pios) бицеш jo зиоцеиел Амеэм ‘+ "HI

(51994) au],
| ЕЛЬ Добре AT

%
Sf-uorunyy e7 os
I SUP DEV CMLDOING) оси CZ AT
0 TE 0
ТИ
от y y OI
р 7 И
07 HEN 07
|
0€ % 4 05 %
Or Or
0s Sf-asnojnoy 0s
I SHAMAN TALC LOL Oe Gh CURE TA
0 0
Ol 01
07 07
05 % 05 %
Ov Or

0S S['-3U31.10] os

107

GEOGRAPHIC VARIATION IN REPRODUCTIVE ТВАП$ OF HELIX ASPERSA

622 9'8 ¿St 192 LAN 89 Erz [а er 0'0€ L 22 (%)
Ayyeyow ynpy
L'68 L'98 c 68 8°88 co €6 G62 5'28 8'98 7'06 S'98 8'06 (%)
sseoons
Buiyoyey
85 +711 6€ + [69 Er +968 OF +908 с9+ 898 [9+ 158 сч сб Кб = ДУСЕ 6S+1CEL 85 = 0'911 Er + £66 9ZIS YOIN|ID
L9+89€ 8'6 = 98€ 18 #219 CB = 019 9€l +996 S6 + 8'721 COL + E'OEL 9bl = CGI Bri + SSZL Viel = ©4561 26 + E38 ‘pul
Jad s6Be jo
Jaquinu ueoW
ЕО + 9'0 COFY0O 10+80 10+2Z0 hO+LEL LOGE roO+Frl LO+EL LO+EL LO+c!H LO+60 ‘pul Jod
seu9}n|9 JO
Jaquinu ueoW
6 cr У 629 0'09 ГА 298 9°88 0'08 9°82 Er 0'09 (%) SBuiAel
JO э}е1 UPOW
LO+LL £O+80O0 LOFrI 10+60 kO+G?C 10+6С КО = ZE КО + 9< co+ Ar cO+EC cos tE ‘pui Jod
эбицеш jo
Jaquinu ueoW
6 c9 0'09 vl 9°89 L'26 9°86 9°86 0'06 1`96 0'08 0'08 (9%) збицеш
Jo syeı ueayy
КО #042 20 #29 CO+E6C SO +986 ZOFEESE SO + 8'05 2OFOIE COFPEE 2O*+9EE COF+T6CE 20 +GOE (ши)
yıpeaug |ец$
GE 02 OZ OZ GE OZ OZ OZ OZ 04 OZ ezis ajdwes
(9861) (5861) (5861) (5861) (5861) (S861) (5861) (5861) (5861) (5861) (5861) (1284 бинашез)
siuag 1S зшэа IS ewied enseg uoA] juowjeg juoueg UuOUBIAY asnojno, sa1a6ins }u21107 uIBUO

‘(10118 p1epue]s + X) зиошриоэ
Asoyesoge| циолип Japun рэ!рп}$ suoieindod enjeu auiu шодц esyadse хуэн Jo Ашенош pue azis ¡joys ‘злэоелецо anmonpolday “| 37191

108

MADEC & DAGUZAN

60 LORIENT
50
40
30
20
10
0

0 1 2 3 4 5 6

60 BELMONT
50
40
30
20
10
0

60 TOULOUSE
50
40
30
20
10
0

1 2 3 4 5 6
0 1 2 3 4 5 6

0 1 2 3 4 5 6

ST.-DENIS

60 BASTIA
il %

60 LYON
50
40
30
20
10
0

0 2 3 4 5 6
0 2 3 4 5 6
3 4 5 6

60 AVIGNON
50
40
30
20
10
0

60 SURGERES
50
40
30
20
10
0
2

1
1
1

60 PALMA

0 1 2 3 4 5 6
Numbers of matings and clutches per snail

0 1 2 3 4 5 6

Numbers of matings and clutches per 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

% 60 LORIENT 60 TOULOUSE 60 ST. DENIS
E 50 50
40 40 40
JS 30 30 30
20 20 20
10 10 10
0 o 0
NA бт RES CRE Cut ВЕ AA
60 60 60
50 50 50
40 40 40
F1 30 30 30
20 20 20
10 10 10
0 0 0
UNO sitio; OF 028 142 3 APS бот биз lis ova
60 60 60
50 50 50
40 40 40
F2 30 30 30
20 20 20
10 10 10
0 0 0
DAT DNS CRC, OD O A AO SA RS бат

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.

MADEC & DAGUZAN

110

ВЕ eel 0'0€ DER 68 0`01 gel ggL rel (%)

Анерош упру
p28 ÿ'06 L'88 5'88 8°98 698 L'E8 2 16 0'68 (%)
ssaoons BulyoJeH
b + c'c6 he+cve IJEFEL 96+ 196 У + 0'/8 LE +896 LE + 8c6 6'5 = 9'98 8€ +/7/6 9215 Y9NID
ое L+pvOEl SIL+9GCI |6 + 9`/8 6'01 + 8561 6'8 + 496 GEL + 6103 ZIL = 8°/ер LOL+tZ6 901 = 8201 ‘pul sad $бба

Jo лэашпи ueayy

rO*+Frl КО+ Е cO+t tL КО ЕЕ LO+LEL c0 +07 roO=+rl ОЕ КО ЕЕ ‘pul Jod sayoynjo
jo Jaquinu ueey;y

298 S'v8 [81572 6 88 G'v8 0'S6 6'88 0'08 5'94 (%) sbuihe|
jo 9/21 иеэи\

L'O + 9'L c0 +67 c0+91 c0 +07 cose 60 +97 cO+EC 0 +77 AMARA ‘pul sad збицеш
Jo лэашпи ueayy

8216 001 L'96 001 001 g'16 001 001 206 (%) збицеш
yo эе1 veaW

810: DCE MCD 00 1.0922 20 ice CD PIE 60-915 60-026 60-е SOx coe (ши)
уреэла ||эч$
Gt Sr 05 Sy Gt Or Sr Sr 85 921$ ajdwes
23 15 sr 23 15 sr 23 15 sr UOI}PIOUEE)
sıusq ‘IS asno¡no] 1U91107 ulbuO

"4018 рлерие}$ + X) sjeus
eu} jo иоцелэиэб pue шбио o] бирлоээе ‘зиошриоэ ¡ele лэрип Bsiedse xılayy jo Ayıleuow pue azis ¡pays 'siajoejeyo элцопролАэн ‘2 AIGVL

GEOGRAPHIC VARIATION IN REPRODUCTIVE TRAITS OF HELIX ASPERSA 111

80 AS

0
ON IS O CES SE 6:27
Numbers of matings and clutches per individual

80 F6

60
%
40

O eg.
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 т-
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 (Р < 0.001) and therefore,
the population classification according to М.
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, F1, 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 М1 (dependent variable) and shell breadth (in mm x
10) in Helix aspersa from ten natural populations. P: level of significance of r.

Origin N Slope
Lorient 40 0.88
Surgères 51 0.74
Toulouse 55 0.84
Avignon 56 0.82
Lyon 27 1.30
Belmont 60 0.90
Bastia 41 0.82
Palma 44 0.65
St. Denis 29 0.98
Alger Y 0.15

Intercept r Р
—164.6 0.60 тия
= 115.5 0.44 “=
— 131.2 0.51 ==
— 141.5 0.46 =
-291.2 0.68 sd
—179.9 0.66 La
— 149.3 0.60 ыы

—95.3 0.63 DE
— 182.4 0.74 Gi
+121.5 0.10 NS

TABLE 4. Classification of natural populations according to shell breadth and first clutch size М1 ($.М.К.

test; P < 0.05)

Shell breadth classification

Terms used Means SNK test
Toulouse 33:5 А

Avignon 33.3 A

Surgeres 32.4 A

Belmont 30.8 B

Lorient 30.5 B

Lyon 29.4 B С
Palma 29.3 B C
Bastia 28.4 C

La Réunion 26.3 D

Clutch size classification

Terms used Means SNK test
Toulouse 150.2 A

Avignon 132.8 A

Surgeres 124.3 A

Lorient 104.9 B

Belmont 97.3 B С
Palma 95.1 B C D
Lyon 91.1 B C D
Bastia 83.8 C D
La Réunion 74.1 D

TABLE 5. Classification of AS, JS, F1 and F2 samples according to shell breadth and first clutch size

(S.N.K. test; Р < 0.05)

Shell breadth classification

Terms used Means SNK test
AS-Toulouse 33.3 A

F2-Toulouse 32.6 A B

F2-Lorient 32.0 A B С
F2-La Réunion 31.9 A B C
JS-Toulouse 31.5 B C
F1-Toulouse Silks B С
F1-Lorient Silat B С
F1-La Réunion 30.8 B C
JS-Lorient 30.6 B C
AS-Lorient 30.4 C
JS-La Réunion РЕ D
AS-La Réunion 26.4 D

Clutch size classification

Terms used Means SNK test
AS-Toulouse 145.9 A
F2-La Réunion 102.8 B
F1-La Réunion 101.2 B
F2-Lorient 100.5 B
JS-Toulouse 100.3 B
F2-Toulouse 100.1 B
AS-Lorient 100.1 B
JS-Lorient 99.7 B
F1-Toulouse 91.2 B С
F1-Lorient 90.7 B С
JS-La Réunion 80.8 © 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 mortality of Algerian Helix aspersa from 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 ЗО
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 (тд) 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 т F1 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 during 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
characterized 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 (H. a. maxima) seem to
have developed a specific combination of re-
productive traits. Egg weight (or size; ry, =
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 Heli-
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 115

TABLE 7. Summary of life-history traits observed in Helix aspersa from La Reunion and Algeria.

Population from Algeria

e Thicker shells

eAdult size larger

eLater maturity

eLonger length of life

eMore offspring, smaller/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 intraspecific 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 predation (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). Ona
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

Population from St. Denis

e Thinner shells

e Adult size smaller

eEarlier maturity

eShorter length of life

eFewer offspring, larger/parent size

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 (H. 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, 1988). 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
H. 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 shel! in adults as in juveniles 15 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,
Н. 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 8 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 & №е! (1990) show that in unstable
populations of Helix texta, a low population
density, caused by a massive predation of
adults, allows a very rapid growth of young.
An other density-dependent mechanism, also
related to predation 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
seetzeni (Yom-Tov, 1972).

Finally, a valid comparison with the predic-
tions of life-history models requires a field
study on tactics used by Helix aspersa 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 Н. а. 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, 5. А. & В. CHASE, 1988, Courtship and
copulation т 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 landsnail
Helix aspersa Müller. Journal of Molluscan Stud-
ies, Supp. 12A: 2-5.

BAUR, B. & C. RABOUD, 1988, Life-history of the
landsnail Arianta arbustorum along an altitudinal
gradient. Journal of Animal Ecology, 57: 71-87.

ВЕСОМ, M. J. L. HARPER 4 С. В. TOWNSEND,
1987, Life-history variation. Pp. 501-538 in: м.
ВЕСОМ, J. L. HARPER & С. В. TOWNSEND, eds., Ecol-

ogy: Individuals, Populations and Communities.
Blackwell Scientific Publications, Oxford.

BROWN, К. 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, К. М., D. В. DEVRIES & В. К. 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. Рр. 649—677 in: W. D. RUSSELL-HUNTER, ed.,
The Mollusca, Vol. 6: Ecology. Academic Press,
London.

CAMERON, В. А. D. & M. А. 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 commercialises en Europe occidentale. Pp.
21-35 in: J. DAGUZAN, ed., L’escargot et l'Hélicicul-
ture. Informations techniques des services vete-
rinaires, Paris.

CLARKE, B., 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, eds., Pulmonates, Vol. 2A. Ac-
ademic Press, London.

COWIE, В. H., 1984, The life-cycle and productivity
of the land snail Theba pisana (Mollusca: Heli-
cidae). Journal of Animal Ecology, 53: 311-325.

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. |-Re-
production et éclosion des jeunes en batiment 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. II-Evo-
lution de la population juvénile de l'éclosion a
раде de 12 semaines en bâtiment et en condi-
tions d'élevage contrôlées. Annales de Zootech-
nie, 31: 87-110.

DAN, М. & S. E. В. 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, С. 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 117

in Israël (Pulmonata, Helicidae). Journal of Mol-
luscan Studies, 56: 189-204.

LE CALVE, D., 1988, Influence de Гаде sur les
comportements d’accouplement et de ponte
chez l’escargot petit-gris Helix aspersa Müller. Mé-
moire, 1.S.P.A., Rennes.

LE GUHENNEC, М. Е. & J. DAGUZAN, 1983, Rôle
de la lumière sur la reproduction de l'escargot
petit-gris, Helix aspersa Muller. 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 Molluscan Studies, 51: 105-115.

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’Helix aspersa
Müller. Evolution de ces caractéres au labora-
toire. Bulletin de la Société Zoologique de
France, 114: 85-100.

MADEC, L., 1989b, Etude de la differenciation de
quelques populations géographiquement sépa-
rées de l'espèce Helix aspersa Müller: aspects
morphologiques, écophysiologiques et biochi-
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’Helix aspersa Muller 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). Netherland Journal of Zool-
ogy, 27: 1-132.

PANELLA, F., 1982, Effect of one cycle of divergent
selection for shell length in Helix aspersa Miller.
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, E., 1975, Aspects of the ecology of He-
Их ротайа 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, C. F., 1971, Ecologie comparée des gas-
téropodes pulmonés des dunes Méditerra-
néennes et Atlantiques. Natura, 62(3): 277-358.

SCHERRER, B., 1984, Biostatistique. G. MORIN, ed.
Chicoutimi, 850 pp.

STAIKOU, A., М. LAZARIDOU-DIMITRIADOU & М.
FARMAKIS, 1988, Aspects of the life cycle, pop-
ulation dynamics, growth and secondary produc-
tion of the edible snail Helix /исогит 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. (Зее 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. Il. Production and survival
of eggs and juveniles. Netherland Journal of Zo-
ology, 23: 168-188.

YOM-TOV, Y., 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 17 November 1992

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
the laboratory, Boonea impressa parasitized both Crassostrea virginica (Gmelin, 1791) and
Geukensia 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. impressa’s 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. (1987) have suggested that
B. impressa may be a vector for the oyster
pathogen Perkinsus marinus.

119

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. impressa’s 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-

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 um 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-DO micro-
scope.

Snails for histochemical studies were de-
calcified prior to fixation in B-4 (consisting of
0.1% glutaraldehyde, 6% HgC,,, and 1% so-
dium acetate) for 5 h. Tissue was treated as
described above. Once sections were cut (3—
5 вт) they were deparaffinized, dezinkarized
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

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 ОЕ ВООМЕА 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
pallial kidney, a simple apectinate osphra-
dium, and a pigmented mantle organ (Fig.
2А).

The epidermis of the ащепог 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 База!
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. 3A). 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. 3A). 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, 3A). 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. 2B), 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. 2B). 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.
3B, 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. 3B; 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
pustulose, with the greatest concentration of
papillae anterior to the tips of the tentacles
(Figs. 2C, 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. 4B). 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

FIG. 2. А. Generalized representation of pallial complex. Mantle skirt cut on left side and reflected to the
right. B. Schematic of Boonea impressa in the non-feeding posture, with proboscis retracted. Mantle re-
moved and cephalic hemocoel opened to expose alimentary structures in “natural position,” with exception
of salivary glands. Salivary glands shown upright to reveal location to right of buccal pump Il. С. Schematic
of partially protracted proboscis, with buccal pump | uncoiling as it is pulled forward. Note new position of the
buccal sac, now lying just anterior to head. A = anus, Вр! = buccal pump |; ВрИ = buccal pump |; BS =

buccal sac; DCS = dorsal ciliated strip; H = heart; K = kidney; MO
esophagus; PMO = pigment mantle organ, SGL = salivary gland; VCS

ин

mouth; Р = proboscis; Е =
ventral ciliated strip.

FEEDING STRUCTURE MORPHOLOGY ОЕ ВООМЕА 123

FIG. 3. A. Section through head-foot, mantle, and cephalic hemocoel. В. 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. C. Longitudinal section of inverted
proboscis. BL = basal lamina; CH = cephalic hemocoel; СМ = 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 primary retractor muscle, the base of C). The stylet bulb, extending posteriorly,
which is attached to the columellar muscle, curves dorsally to lie beneath the most ante-
extends into the cephalic hemocoel to insert rior portion of the buccal pump. Within the
on ether side of the sucker (Fig. 5A). posterior portion of the stylet bulb is a cres-

The buccal sac has two major components: cent-shaped lumen, surrounded by the mus-
the stylet bulb and the buccal sucker (Fig. 5B, cles of the stylet bulb (Fig. 5A). The stylet

FIG. 4. А. Scanning electron microphotograph of partially extended proboscis. В. Tumid papillae on external
surface of the proboscis, each with apical tuft of cilia. C. 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 а 9 + 2 microtubule arrangement. С = cilia; CC =
central cell; CEM = 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 ВООМЕА 125

FIG. 5. A. Histological section through buccal зас. В. Scanning electron microphotograph of Бисса! зас;
portion of buccal sac surrounding stylet and cuticular enclosure removed. Stylet bulb intact. C. 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). Е. Cross-sectional view of stylet. Вр! = buccal
pump I; С = cilia; CS = cuticular sheath; L = lumen; LA = labium; MO = mouth; OT = 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 |) 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 matrix 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 | 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 Il) 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 | (Fig.
6D). Buccal pump II is similar to the buccal
pump | 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 |
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 | and buccal Il 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 Il 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 |, just anterior of the buc-
cal pump Nbuccal pump И 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 ОЕ ВООМЕА 127

FIG. 6. Feeding structures. A. Cross sections of buccal pump | lying to one side of larger buccal pump Il. В.
Scanning electron microphotograph of buccal pump | in cross section. Etched outer covering encloses
internal layer of muscle fibers extending to the lumen. С. Transmission electron microphotograph of buccal
pump | 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 Il covered by a thin layer of
epithelium, comprised of myofibrils. E. Transmission electron microphotograph of buccal pump И in cross
section. Note circular muscle, longitudinal muscle, and muscle perpendicular to the organ’s axis. Bpl =
buccal pump I; ВрИ = buccal pump Il; CRM = circular muscle; L = lumen; M = muscle; SD = salivary
ducts.

FIG. 7. А. Histological section of the esophagus and a single salivary gland. В. 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; Е = 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 ofthe
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.
Рог В. 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. | 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 ВООМЕА 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. 5Е).
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 | by the oral tube (Fig.
5A, C). Contractions of only buccal pump Il
draw host hemolymph into the alimentary ca-
nal. Located at the junction of buccal pump |
and buccal pump Il is a ring of muscle that
closes this passageway when contracted,
thereby forcing host hemolymph into the
esophagus once the lumen of buccal pump Il
is compressed. Elongate cilia, present at the
junction of the buccal pump Il 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. 8B), 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 8C. 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

РЕМ

| 1.0mm

FIG. 8. Retractor muscles of the proboscis. A. Schematic 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. C. 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 131

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 apically (redrawn from Fretter & Graham, 1949). B.
Schematic of internal proboscial arrangement of Odostomia eulimoides. Papillae comprised of large-celled
epithelium; note large gland cell дис! extending between papillae to open externally (redrawn from Maas,
1965). С. Schematic of feeding structures of O. eulimoides (redrawn from Maas, 1965). BS = buccal sac;
BPI = buccal pump I; ВРИ = buccal pump Il; 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

(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
Вр! twice the length of Bpll.

(3) Salivary ducts:
enter buccal sac & then
stylet bulb, without exiting
alimentary canal.

*Maas; Ankel

(a) papillae composed of large-

celled epithelium.

(b) internal to papillae a layer of

circular muscle, above a
layer of longitudinal muscle.

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.

**Fretter & Graham

(a) papillae composed of only 3

)

—

or 4 cells.

beneath papillae a of layer
gland cells with ducts that
extend to exterior via the
center of the papillae.
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, О. plicata, and Liostomia clavula, with cursory attention given to Odostomia

rissoides, Chrysallida spiralis, and C. obtusa.

**Examined in detail Odostomia unidentata, О. plicata, and О. lukisii, with some attention given to O. scalaris, (= О.

rissoides), O. trifida, and Chrysallida spiralis.

(1949) (e.g., Odostomia unidentata), роз-
sessed papillae comprised of only three to
four cells, containing large basal nuclei, side
by side within 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 а 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 о 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 (30m) 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. 3B, 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. /ukisii, 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 ОЕ ВООМЕА 133

two buccal pumps (Вр | and Вр И respec-
tively) that are histologically discrete (Fig.
9C). The first buccal pump (like the buccal
pump | of В. 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 Bpl of Boonea
impressa is very elongate and twice the
length of the ВрИ, whereas in all examined
European odostomians, the buccal pump is
divided into approximately equal sections
(Fig. 9C). The Bpl of Boonea impressa is a
well-developed structure comprised chiefly of
muscle cells that surround а triangularly
shaped lumen (Fig. 6A, 6C). This is contrary
to White et al. (1985), who determined that
this portion of the feeding apparatus of B. im-
pressa was a роопу 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. 9C). 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.

ACKNOWLEDGEMENTS

This paper is part of a master’s thesis com-
pleted at the College of Charleston and is
contribution no. 105 from the Grice Marine
Biological Laboratory, College of Charleston,
Charleston, South Carolina. | 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. | thank Karen Swan-
son and Richard Houbrick for assistance with
the illustrations and Bob and Jan Ashcraft for
their help with the SEM and TEM procedures.
Richard Houbrick, Jerry Harasewych, and
Robert Hershler critically reviewed the first
draft of this manuscript and offered many
helpful suggestions for its betterment. | 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, В. T., 1974, American sea shells. О. 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. Verhandlungen 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-
Uke

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 Morphologic 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,
1987a, 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,
1987b, 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, В. & 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, А. H., 1965, Two gastropod hosts of
the pyramidellid gastropod Odostomia bisutura-
lis. The Nautilus, 79: 7-10.

ЗНЕЕМАМ, D. С. & В. В. НВАРСНАК, 1980, The-
огу 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. & C. LANGDON, 1986, Effects of the
ectoparasitic Воопеа (= 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., Е. М. 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 & Е. М. POWELL,
1985, Aspects of reproduction, larval develop-
ment, and morphometrics in the pyramdellid Boo-
nea impressa (= Odostomia impressa) (Gas-
tropoda: Opisthobranchia). The Veliger, 28(1):
37-51.

WHITE, М. Е., Е. М. POWELL, $. M. RAY & Е. А.
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., Е. М. POWELL, $. M. RAY, Е. А.
WILSON & С. Е. 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, М. E., Е. М. POWELL 4 $. M. RAY, 1988b,
Effect of parasitism by the pyramidellid gastropod
Boonea impressa on the net productivity of oys-
ters (Crassostrea virginica). Estuarine, Coastal,
and Shelf Science, 25: 359-377.

Revised MS accepted 22 November 1992

MALACOLOGIA, 1993, 35(1): 135-140

INFLUENCIA AMBIENTAL ЗОВВЕ 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 Cientificas (CONICET), C. C. 92 (9410), Ushuaia,
Tierra del Fuego, Argentina

ABSTRACT

Environmental influence on shell allometric growth in Nacella (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: morfologia, alometria, influencia ambiental, intermareal, lapas,

Nacella, prosobranquios.

INTRODUCCION

Los gasteröpodos presentan valvas que
varian su morfometria general y las propor-
ciones entre los distintos parametros estruc-
turales de la valva en relaciön con las varia-
ciones del ambiente, generando alometras
en el crecimiento. Los factores ambientales
que producirian cambios mas marcados so-
bre la morfometria valvar serian el oleaje o
corrientes intensas у la exposiciön a la dese-
caciön (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 morfologia valvar.

Nacella (Patinigera) deaurata (Gmelin,
1791) habita el intertidal inferior quedando
expuesta a la desecaciön solamente en las
mareas de sicigia. Por ello las variaciones
morfolögicas que presenta pueden correla-
cionarse fundamentalmente con el grado de
exposiciön al oleaje. El propösito de esta in-
vestigaciön fue comparar los diferentes pa-
rämetros estructurales en Nacella (P.) deau-
rata, colectada en dos localidades con
diferente grado de exposiciön.

135

MATERIAL Y METODOS

Los muestreos se realizaron en dos loca-
lidades (Fig. 1): (a) Punta Occidental (PO)
(54°50'$., 68°20’W.: área expuesta a los
vientos dominantes del SO, fuerte oleaje, de-
clive suave, con abundancia de coralinaceas
incrustantes como Pseudolitophillum sp. y
Synartrophitum sp. (Mendoza, 1988) y ejem-
plares aislados de Macrocystis pirifera. (b)
Bahia Lapataia (BL) (54°52’S., 68°35’W.):
costa orientada hacia el norte, protegida de
los vientos dominantes, con fuerte pendiente
у denso cinturön de Macrocystis pirifera. Las
lapas fueron extraidas por buceo autönomo.
Se separaron las partes blandas de las val-
vas, las que se secaron al aire durante varios
dias hasta que el peso no varió.

Las caracteristicas de las valvas que se
consideraron fueron las siguientes (Fig. 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 micras determinándose el

136 MORRICONI & CALVO

BAHIA
USHUAIA

BAHIA

ARGENTINA A

FIG. 1: Ubicación de las localidades de muestreo. A: Punta Occidental (54°50'$, 68°20’W) у Bahia Lapataia
(54°52'S, 68°35’W) área sombreada. В: Punta Occidental, zona expuesta (area sombreada). La flecha
señala los vientos dominantes. С: Bahia Lapataia, zona protegida (area sombreada). La flecha señala los

vientos dominantes.

peso de la misma. Luego se pesé 1 cm? de
arena, calculandose el volumen correspon-
diente a cada valva. La transformaciön peso
de arena a volumen se realizó promediando el
peso de diez réplicas de 1 cm? de arena.
Las valvas utilizadas fueron seleccionadas
empleando numeros al azar de la colecciôn
total de valvas (PO: 662 ejemplares; BL: 628
ejemplares). Las valvas dañadas о con epi-
biontes fueron descartadas del muestreo.

El rango de LT considerado сотргепаю
valvas de 13 a65 mm estableciéndose clases
de 5 mm. En una primera selecciön se
tomaron diez valvas para cada clase en am-
bas localidades; posteriormente y a los efec-
tos de disminuir la dispersiön de las variables
dependientes se aumentó а 20 por clase el
numero de valvas de las clases mayores de
36 mm. Las variables (AT, PV y VI) fueron
tomadas como dependientes del LT, calcu-

CRECIMIENTO ALOMETRICO VALVAR EN N. (Р.) DEAURATA 137

Pta.Occidental

Lapataia

FIG. 2: Vista lateral de las valvas de Nacella (Patinigera) deaurata provenientes de ambas localidades de

muestreo.

TABLA 1: Regresiôn de AT, PV y VI sobre LT para zonas expuestas (PO) y protegidas (BL).

Localidad и ат ох r P N

(A) PO АТ = —5.32 (0/52 ET) 0.96 < 0.001 161
(В) BL АТ = —2.52 + (0.37 * LT) 0.96 < 0.001 167
Localidad 1g Y = a + (6 * lg X) r P N

(C) PO 1g PV = -5.43 + (3.68 * Ig LT) 0.98 < 0.001 161
(D) BL 1g PV = —4.68 + (3.11 * Ig LT) 0.98 < 0.001 167
(E) PO 1g 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. Test de homogeneidad de las pendientes (Ну : b, = b,) siendo b, la pendiente de la
ecuaciön (A) y b, la pendiente de la ecuaciön (В) de la Tabla 1.

Fuente de Suma de Grados de Cuadrado

variacion cuadrados libertad Medio Е Р

Localidad 64.552 1 64.552 22.646 <0.000
LT 12599.998 1 12599.998 4420.336 <0.000

Localidad * LT 354.555 1 354.555 124.385 <0.000
Error 923.549 324 2.85

lándose las ecuaciones de regresión corre-
spondientes. Cuando fue necesario se realizó
la transformación logarítmica de los datos a
fin de ajustarlos 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 regresiön entre AT/LT, PV/LT у

VILT para Punta Occidental (
pataia (- - - -).

) y Bahia La-

(Tabla 1). La сотрагасюп entre las pen-
dientes de las rectas de regresiön de ambas
localidades muestra diferencias significativas

(Tabla 2). A igual LT las valvas de Punta Oc-
cidental son таз altas que las de Lapataia
(Fig. 3).

Relaciön LT—PV

Esta relaciön se ajusta a una curva poten-
cial tanto en Punta Occidental como en La-
pataia por lo que se realizó la transformación
logaritmica de la misma (Tabla 1). La com-
paraciön 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 transformaciön logaritmica correspon-
diente (Tabla 1), comparändose las dos гес-
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).

DISCUSION

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
experimentalmente 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
adaptativo. 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 (1971) 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. (Р.) DEAURATA 139

TABLA 3: LT/PV. Test de homogeneidad de las pendientes (Но : b, = bs) siendo b, la pendiente de la
ecuaciön (C) y b, la pendiente de la ecuaciön (D) de la Tabla 1.

Fuente de Suma de Grados de
variaciön cuadrados libertad
Localidad 0.51 1
IgLT 108.086 1
Localidad *IgLT 0.742 1
Error 2.948 324

Cuadrado
Medio F E
0.51 56.098 <0.000
108.086 11879.686 <0.000
0.742 81.584 =0.000
0.009

TABLA 4: LT/VI. Test de homogeneidad de las pendientes (H, : b; = b,) siendo b, la pendiente de la
ecuación (E) y b, la pendiente de la ecuación (>) de la Tabla 1.

Fuente de Suma de Grados de
variación cuadrados libertad
Localidad 0.027 1
IgLT 114.137 1
Localidad *1gLT 0.072 1
Error 1.424 324

Cuadrado
medio F P
0.027 6.046 <0.014
114.137 25969.286 <0.000
0.072 16.357 <0.000
0.004

Walker (1972) en Patinigera polaris y Simp-
son (1985) en Nacella macquarensis relacio-
nan la intensidad alométrica 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 mareales, 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 Р. 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 C. 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 intertidal superior, pero no ha-
llan diferencias en zonas con distinta exposi-
ción.

En N. (Р.) 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 (desecaciön,
exposiciön al oleaje) es encontrada por Orton
(1932). Este autor argumenta que los estimu-
los para mantener la valva fuertemente ad-
herida al sustrato ocasionan la retracciön del
borde del manto. De esa manera disminuiria
el crecimiento periferico y por lo tanto aumen-
taria el crecimiento en altura de la valva.
Kopp (1980) establece una relaciön analoga
entre la forma de la valva у la extensiön о
retracciön del borde del manto, apoyändose
en pruebas experimentales.

Se considera que un proceso similar daria
lugar а un тауог engrosamiento de la valva
que conduciria a un aumento de su peso. El
incremento en altura sin cambio en la super-
ficie о perimetro de la base aumentaria el vo-
lumen interno.

AGRADECIMIENTOS

Los autores desean expresar su agradeci-
miento a Gustavo Suarez y Regina Silva por
su colaboraciön en la mediciön de las valvas,
a Lucas Ramos por su ayuda en el procesa-
miento de los datos, a Pedro Medina y Rafael
Pastorino por su participaciön en la recolec-
ciön de las muestras, y a Miguel Barbagallo
por la confecciön de los dibujos y gräficos.
Esta investigaciön es parte del Proyecto de
Investigaciön у Desarrollo (PID № 266): Bio-
logia reproductiva de moluscos у equinoideos
del Canal Beagle. Implicancias ecolögicas y
fisiolögicas, financiado por el Consejo Nacio-
nal de Investigaciones Cientificas у Тестсаз,
Argentina.

LITERATURA CITADA

BALAPARAMESWARA ВАО, В. у Р. 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, Allometric 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, С. 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. C. 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. B., 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, М. L., 1988, Consideracines biológicas
у biogeogräficas de las Corallinaceae (Rho-
dophyta) de las costas de la Isla Grande de Tie-
rra del Fuego. Gayana Botanica, 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, В. D., 1985, Relationship between allo-
metric growth, with respect to shell height, and
habitats for two patellid limpets, Nacella (Patini-
gera) macquariensis Finlay, 1927, and Cellana
tramoserica (Holten, 1802). The Veliger, 28:18—
27.

VERMEIJ, С. J., 1973, Morphological patterns in
high-intertidal gastropods: adaptative strategies
and their limitations. Marine Biology, 20:319—
346.

VERMEUW, С. J., 1978, Biogeography and adapta-
tions: patterns of marine life. Harvard University
Press: Cambridge, Mass. 352 pp.

VERMEIJ, С. J., 1980, Gastropod shell growth rate,
allometry, and adult size: environmental implica-
tions. Pp. 379-394 in D. C. RHOADS & R. A. LUTZ,
eds., Skeletal growth 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. British Antarctic Survey Bulle-
tin, 28:49-69.

Revised Ms. accepted 17 December 1992

MALACOLOGIA, 1993, 35(1): 141-151

А 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 ОЕС', United
Kingdom, and Woods Hole Oceanographic 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 а depth of 2000 m, adja-
cent to a position where hydrothermal fluid at
between 270-314°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-
пит 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 ofthe 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).

| 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,

“Address for correspondence.

and to the director and staff of tne Woods
Hole Oceanographic Institution for their help
over many years.

DESCRIPTION

Genus Nuculana Link 1807
Type species (OD):
Arca rostrata Brugiere, 1789,
ex Chemnitz MS, = Arca pernula
Müller, 1779.

Shell robust, moderately and posteriorly
elongate; rostrum truncate, usually bicarinate,
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 specimens selected
No. 859481 by J. A. A. from the type
locality.

1 specimen

Named in honour of Dr. J. F. Grassle, friend
and colleague of many deep-sea voyages
and participant in the Guaymas Expedition.

142 ALLEN

Material
Specimens (Number
Dive No. Depth (m) Examined Collected)
Alvin 1168 2003 25 (50)
3 (3)

Alvin 1169 1998 8 (16)
Alvin 1170 2019 — (7)
Alvin 1174 2011 — (1)
Alvin 1175 1997 — (1)
Alvin 1176 2022 4 (152)
Alvin 1607 2012 4 (4)
Alvin 1608 2002 1 (1)
Alvin 1614 2004 2 (2)
Alvin 1628 2000 — (5 postlarva)

(1-5 above bottom)
Alvin 1629 2000 — (1 postlarva)

(3—4 above bottom)

BC—Box Core (225 ст? area sampled)

TC—Tube Core (35 cm? area sampled)

SS—Scoop Sample (

PT—Plankton Tow

Position Equipment Date
27°03'N, 111°23’W Ss 10-1-82
TC
27°03'N, 111°25’W BC 11-1-82
27°01'N, 111°25’W BC 12-1-82
27°01'N, 111°24'W BC 17-1-82
27°03’N, 111°23’W BC 18-1-82
27°01'N, 111°25'W ТС 19-1-82
27°05'N, 111°24.5'W TC 29-7-85
27°07'N, 111°24.4'W TC 31-7-85
27°07'N, 111°24.4'W BC 6-8-85
27°00'N, 111°24.5'W PAT 23-8-85
27°00’N, 111°25.5’W PAL 23—8-85

63 тт mesh Бад over metal frame) non-quantitative
(0.4 m?, 183 u mesh) non-quantitative

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 posterior
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, 11 anterior and 15 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 ¡um 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-
Ыу 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

FIG. 2. Мисшапа 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, with “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 150
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 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 proboscides 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

vG Gl KI HT ES
PA РА“ \ \ | | ST HG
VE
GA N 07. ‘ PE
De Po EE СР

pp of tens

(ts aw La :

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; CP, “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; KI, 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 AE

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

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). AA, 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 „m 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 ¡um 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

5 10

ALLEN

15 20 25

Length (mm)

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
muscle 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 М.
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, 1779) from
shallow Arctic seas is similar in size, as too is
N. taphria Dall, 1897, 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 М. 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 um a figure that is twice
that of N. minuta (Muller, 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 overlying 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 predation, 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 М. 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 Ca/yptogena 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 + num. (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

АВВОТТ, В. Т., 1974, 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. А. & Е. J. HANNAH, 1989, Studies оп
the deep-sea Protobranchia. The subfamily Le-
dellinae (Nuculanidae). Bulletin of the British Mu-
seum (Natural History), Zoology, 55: 123-171.

BERNARD, Е. 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, C. J. and C. 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. Н., 1890, Scientific results of explorations
by the U.S Fish Commission steamer “Albatross.”
VII. Preliminary герой on the collection of Mol-
lusca and Brachiopoda obtained т 1887-1888.
Bulletin of the U. $. National Museum, 12: 219—
362.

DALL, W. H., 1896, Diagnoses of new mollusks
from the west coast of America. Proceedings of
the U. $. 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. 8 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,
8 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. А. TYLER, 1991, Deep-sea biol-
ogy: a natural history of organisms at the deep-
sea floor. Cambridge University Press, 504 pp.

GRASSLE, J: Е., №. 5. BROWNILEGER IE
MORSE-PORTEOUS, R. PETRECCA, & |.
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 |. Zoologica, 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 WITTENBERG, J. B., 1985, Oxygen supply to in-

species of Nuculanidae. Nautilus, 49: 13-14. tracellular bacterial symbionts. In м. к. JONES, ed.,
SIMONEIT, В. В. T. 8 Р.Е. LONSDALE, 1982, Hy- The hydrothermal vents of the eastern Pacific: an

drothermal petroleum in mineralized mounds at overview. Bulletin of the Biological Society of

the seabed of Guaymas Basin. Nature, 295: Washington, 6: 301-310.

198-202. YONGE, C. M., 1939, The protobranchiate Mol-
STANLEY, S. M., 1970, Relation of shell form to life lusca: a functional interpretation of their structure

habits of the Bivalvia (Mollusca). Geological So- and evolution. Transactions of the Royal Society

ciety of America Memoir, 125: 296 pp. of London, В, 230: 79-147.

WILLETT, G., 1944, New species of mollusks from
Redondo, California. Bulletin of the Southern
Californian Academy of Sciences, 43: 71-73. Revised Ms. accepted 29 April 1992

2.728 ATAN
= Са Thy

hs CASITA
| PACS A
sd “en ИА В de tFians

hi 6 SA Ve

e 7 De A bd Ba,
PUR diy reel м
A L
1 $.’ мА
| a
4

MALACOLOGIA, 1993, 35(1): 153—154

LETTERS TO THE EDITOR

REPLY TO “SUPRASPECIFIC NAMES OF MOLLUSCS:
А QUANTITATIVE REVIEW”

М. A. Edwards! & М. 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 covers 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 YO1 1LF, England.

154 EDWARDS & THORNE

is “supposedly the most complete indexing
system,” “а nomenclator 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 nomenclator 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. Otherwise 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. Malacologia, 34:75—86.

The editor-in-chief of Malacologia welcomes let-
ters that comment on vital issues of general im-
portance to the field of Malacology, or that com-
ment on the content of the journal. Publication is
dependent on discretion, space available and, in
some cases, review. Address letters to: Letter to
the Editor, Malacologia, care of the Department of
Malacology, Academy of Natural Sciences, 19th
and the Parkway, Philadelphia, PA 19103.

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 АТ
The Academy of Natural Sciences of Philadelphia

The Academy of Natural Sciences of Philadelphia, through its Jessup and
McHenry funds, makes available each year a limited number of awards to support
students pursuing natural history studies at the Academy. These awards are pri-
marily intended to assist predoctoral and immediate postdoctoral students. Awards
usually include a stipend to help defray living expenses, and support for travel to and
from the Academy. Application deadlines are 1 March and 1 October each year.
Further information may be obtained by writing to: Chairman, Jessup-McHenry
Award Committee, Academy of Natural Sciences of Philadelphia, 1900 Benjamin
Franklin Parkway, Philadelphia, Pennsylvania 19103-1195, U.S.A.

WHY МОТ SUBSCRIBE ТО 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):

INSTRUCTIONS FOR AUTHORS

1. MALACOLOGIA publishes original re-
search on the Mollusca that is of high quality
and of broad international interest. Papers
combining synthesis with innovation are par-
ticularly desired. While publishing symposia
from time to time, MALACOLOGIA encour-
ages submission of single manuscripts on
diverse topics. Papers of local geographical
or systematic interest should be submitted
elsewhere, as should papers whose primary
thrust is physiology or biochemistry. Nearly all
branches of malacology are represented on
the pages of MALACOLOGIA.

2. Manuscripts submitted for publication
are received with the tacit understanding that
they have not been submitted or published
elsewhere in whole or in part.

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

4. Unless indicated otherwise below, соп-
tributors should follow the recommendations
in the Council of Biology Editors (СВЕ) Style
Manual (ed. 5, 1983) available for U.S.
$24.00 Нот CBE, 9650 Rockville Pike,
Bethesda, MD 20814, U.S.A.

5. Be brief.

6. Manuscripts must be typed on one side
of good quality white paper, double-spaced
throughout (including the references, tables
and figure captions), and with ample margins.
Tables and figure captions should be typed on
separate pages and put at the end of the
manuscript. Make the hierarchy of headings
within the text simple and consistent. Avoid
internal page references (which have to be
added in page proof).

7. Choose a running title (a shortened
version of the main title) of fewer than 50
letters and spaces.

MALACOLOGIA

1993

8. Provide a concise and informative Ab-
stract summarizing not only contents but re-
sults. A separate summary generally is super-
fluous.

9. Supply between five and eight key
(topic) words to go at the end of the Abstract.

10. Use the metric system throughout. Mi-
cron should be abbreviated рт.

11. Illustrations are printed either in one
column or the full width of a page of the
journal, so plan accordingly. The maximum
size of a printed figure is 13.5 x 20.0 ст
(preferably not as tall as this so that the cap-
tion does not have to be on the opposite
page).

12. Drawings and lettering must be dark
black on white, blue tracing, or blue-lined
paper. Lines, stippling, letters and numbers
should be thick enough to allow reduction by
Ye or Ys. Letters and numbers should be at
least 3mm high after reduction. Several
drawings or photographs may be grouped
together to fit a page. Photographs are to be
high contrast. High contrast is especially im-
portant for histological photographs.

13. All illustrations are to be numbered
sequentially as figures (not grouped as plates
or as lettered subseries), and are to be ar-
ranged as closely as possible to the order in
which they are first cited in the text. Each
figure must be cited in the text.

14. Scale lines are required for all nondi-
agrammatic figures, and should be conve-
nient lengths (e.g., “200 um,” not “163 um”).
Magnifications in captions are not accept-
able.

15. All illustrations should be mounted,
numbered, labeled or lettered, i.e. ready for
the printer.

16. A caption should summarize what is
shown in an illustration, and should not dupli-
cate information given in the text. Each let-
tered abbreviation labeling an individual fea-
ture in a figure must either be explained in
each caption (listed alphabetically), or be
grouped in one alphabetic sequence after the
Methods section. Use the latter method if
many abbreviations are repeated on different
figures.

TE 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 bl to vs Subscription
| Office.

\ a, a et
N Е We ыы
| { Hy ea ed J к 3,8 “=
TNA CA à fr >
р ик" & ae y A
¢ roe
VO 35 NOL rd MALACOLOGIA Wr: i
CONTENTS. I

В JONATHAN COPELAND & MARYELLEN MANERI DASTON \
_ Adult and Juvenile Flashes in the Terrestrial Snail Dyakia striata .

MARYELLEN MANERI DASTON & JONATHAN COPELAND - de $)
D The Luminescent Organ. ‚and Sexual Maturity i in er striata ie a y,
HARLAN К. DEAN | \ 2
| A Population: Study of the Bivalve Idas argenteus Jeffreys, 1876, (Biv:
_ Mytilidae) Recovered from a he Wood Block in the Dee

$e Atlantic OCBAN Re ato 2 И в а ооо РИ Ne:

MICHAEL S. JOHNSON, JAMES MURRAY & BRYAN CLARKE - ose LER A
Evolutionary Relationships and Extreme Genital Variation in a N A

Related Group of Partula а — 43

CARLOS Е. PRIETO, АМА 1. PUENTE, KEPA ALTONAGA & BENJAMIN J. GOMEZ _ cay ;
_ Genital Morphology of Caracollina lenticula (Michaud, 1831), with we <

-New Proposal of Classification. of: Helicodontoid Genera ‚ (Pulmonata: "A

; : Hygromioidea) BS PR Ne ee er CRE CR TT o
-ALOIS НОМЁК = : м, | AT: Se
_ Melanism in the Land Snail Helicella candicans (Gastrapoda, Helicid

‚ and its Possible Adaptive Significance иене инь а if)
ANETTE BAUR & BRUNO BAUR А м À LL? a VARIE
“Daily Movement Patterns and Dispersal in the Land Snail м 3

Arianta arbustorum ............... емо. be emanan a LA р
_ LUC MADEC 8 JACQUES DAGUZAN — | se
Geographic Variation in Reproductive Traits of Нейх м: Müller st
_ under Laboratory Conditions .. dd habeas SEE. Bas
JOHN В. WISE ' } ый

. Anatomy and Functional оу of the Feeding Structures of the Et
parasitic Gastropod Boonea impressa ИЕ es bee lg: 13%

ELBA MORRICONI Y JORGE CALVO
Influencia Ambiental Sobre el Crecimento Alométrico de la Valva en cola ane

г $

’ (Patinigera) deaurata (Gmelin, 1791) del Fan) Beagle, Argentina ЗА: г )
J. À. ALLEN Pa N os
ANew Deep-Water Hydrothermal Species of маты (Bivalvia: Protobran- — ret $
| chia) from the Guaymas Basin ..... ан В Free sit | 141
М. А. EDWARDS & М. J. THORNE ET Sa I NN au
Taten to the Eon. ина REN RR FR BE, A Nr О
4 RES ; a Я rl р ki = ом
? у és y i ss à
: | rab
? | A ¡La

и В, IBRARY

оо 199

в - HARVARD
в > _ UNIVERSITY +!

mational Journal of Malacology aa
: |

Г ve

Pa ast

z ¥ a | К
| Internationale Malakologische Zeitschrif

MALACOLOGIA Mr: od an
_ Editor-in-Chief:
GEORGE M. DAVIS
Editorial and Subscription Offices: -

| Department of Malacology | be > Ed E.
The Academy of Natural Sciences of Philadelphia DE EE

- 1900 Benjamin Franklin Parkway ls À 1

Philadelphia, Pennsylvania 19103-1195, U.S.A. _ 2 EN AI

| | Co-Editors: ER ]
EUGENE COAN | | | CAROL JONES —
California Academy of Sciences á Denver, co ae

San Francisco, CA |
: Assistant Managing Editor: ei À 1
CARYL HESTERMAN Bart: “q ¿LA
Associate Editors: | > AA

JOHN B. BURCH y : Si > ae
University of Michigan ) vu wpe GISMANN - A
Ann Arbor Egypt

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

KENNETH J. BOSS JAMES NYBAKKEN гой “à 121
Museum of Comparative Zoology Moss Landing Marine Laboratory ‘1
Cambridge, Massachusetts | х California Y E eae

- JOHN BURCH, President Î CLYDE F. E. ROPER

Smithsonian Institution —
Washington, D.C.

W. D. RUSSELL-HUNTER
Syracuse University, New hos:

SHI-KUEI WU
University of Colorado Museum, Boulder

MELBOURNE R. CARRIKER
University of Delaware, Lewes

GEORGE M. DAVIS
Secretary and Treasurer —

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

Participating Members -

EDMUND GITTENBERGER Fa HE JACKIE L. VAN СОЕТНЕМ EU.
Secretary, UNITAS MALACOLOGICA Treasurer, UNITAS MALACOLOGICA - ¿Na
Rijksmuseum van Natuurlijke Koninklijk Belgisch Instituut =
Historie voor Не uN
Leiden, Netherlands 2 | Brussel, Belgium у NOIRS
Emeritus Members - x 5, VON
J. FRANCIS ALLEN, Emerita ROBERT ROBERTSON y\ qe $
Environmental Protection Agency The Academy of Natural Sciences Fok
Washington, D.C. | _ Philadelphia, Pennsylvania —
ELMER G. BERRY, - NORMAN Е. ЗОНЕ FR +5 À 3

Germantown, Maryland - | U.S. Geological Survey
Reston, Virginia

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

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

Е. Е. 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

1993
EDITORIAL BOARD

E. GITTENBERGER
Rijksmuseum van Natuurlijke Historie
Leiden, Netherlands

F. GIUSTI
Università di Siena, ltaly

A. N. GOLIKOV
Zoological Institute
St. Petersburg, Russia

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

A. V. GROSSU
Universitatea Bucuresti
Romania

T. HABE
Tokai University
Shimizu, Japan

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

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

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

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

B. HUBENDICK
Naturhistoriska Museet
Góteborg, Sweden

S. HUNT
Lancashire
United Kingdom

R. JANSSEN
Forschungsinstitut Senckenberg,
Frankfurt am Main, Germany

R. N. KILBURN

Natal Museum
Pietermaritzburg, South Africa

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

J. KNUDSEN
Zoologisk Institut & Museum
Kobenhavn, Denmark

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

A. LUCAS
Faculté des Sciences
Brest, France

C. MEIER-BROOK
Tropenmedizinisches Institut
Tubingen, Germany

H. K. MIENIS
Hebrew University of Jerusalem
Israel

J. Е. MORTON
The University
Auckland, New Zealand

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

R. NATARAJAN

Marine Biological Station
Porto Novo, India

J. OKLAND
University of Oslo
Norway

T. OKUTANI

University of Fisheries

Tokyo, Japan

W. L. PARAENSE

Instituto Oswaldo Cruz, Rio de Janeiro
Brazil

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

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

W. F. PONDER
Australian Museum
Sydney

R. D. PURCHON

Chelsea College of Science & Technology
London, United Kingdom

QUZ Y:
Academia Sinica
Qingdao, People's Republic of China

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

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

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

A. STANCZYKOWSKA
Siedlce, Poland

Е. STARMÜHLNER
Zoologisches Institut der Universitát
Wien, Austria

Y. |. STAROBOGATOV
Zoological Institute
St. Petersburg, Russia

W. STREIFF
Université de Caen
France

J. STUARDO
Universidad de Chile
Valparaiso

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

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

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

J. А. VAN EEDEN
Potchefstroom University
South Africa

М. Н. 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 депега Cymia Mörch, 1860, Rapana Schumacher,
1817, Stramonita Schumacher, 1817, Concholepas Lamarck, 1801, Dicathais lredale, 1936,
Drupa Röding, 1798, Plicopurpura Cossmann, 1903, Pinaxia H. & A. Adams, 1853, Nassa
Röding, 1798, Vexilla Swainson, 1840, Сгота Н. & A. Adams, 1853, Morula Schumacher, 1817,
Thais Röding, 1798, Purpura Bruguiere, 1789, and Mancinella Link, 1807, are placed in Ra-
paninae. The taxa Vasula Mörch, 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!”

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 Muricoidea (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

155

Anatole France, Thais

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 muricid groups.

156 KOOL

Taxonomic History

Traditionally, the superfamily Muricoidea
Rafinesque (sensu Thiele [as Мипсасеа])
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 Muricidae, sensu Pon-
der. Unless noted otherwise, Muricidae will
herein be equivalent to Muricoidea, sensu
Thiele.

Members of the Muricidae have an often
spiny shell, usually bearing a distinct, some-
times long, anterior siphonal canal. An ana-
tomical feature shared by most Muricidae is
the accessory boring organ, located in the
foot, and used for chemically dissolving shell
material. Naticids have an accessory boring
organ as well, but this structure apparently
has arisen independently in these distinct
groups. Most Muricidae have a long radular
ribbon with a row of tri- or pentacuspid rachid-
ian (central) teeth, each of which is flanked by
a lateral tooth. The tri- 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 centuries. Taxonomic problems
within the Muricidae as a whole impede our
understanding of all groups within this taxon.
For example, due to the vague boundaries of
many higher muricid 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-
ricidae are not always clear-cut, . . .” This
taxonomic confusion results in a lack of un-
derstanding of the phylogeny of all muricid
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: Ocenebrinae [authors and dates of
taxa given in Table 1], Muricinae, Trophoni-
nae, Typhinae, and Rapaninae; he included
the members of the Thaididae/nae of authors
in the Purpuridae as a separate family. Thiele
(1929) included two families, Muricidae and
Magilidae, and did not list subfamilies. Wenz
(1941) included the same two families, but
subdivided the Muricidae into the subfamilies
Muricinae, Rapaninae, Columbariinae, and
Drupinae (Thaidinae of authors). Keen

(1971a) recognized the families Muricidae,
Columbariidae, Sarganidae, Coralliophilidae,
Moreidae, and Thaididae; she subdivided
the Thaididae into the subfamilies Thaidinae,
Rapaninae, and Drupinae. Radwin & D’Attilio
(1971) subdivided the Muricidae into the
families Muricidae, Columbariidae, 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 Muricidae.
Harasewych (1983) showed that the Colum-
bariinae do not belong within the Muricidae
but instead in the Turbinellidae. Ponder &
М/агеп (1988) include in Muricidae the sub-
families Muricinae, Thaidinae (with Rapani-
nae in synonymy), Coralliophilinae, Sargani-
nae, and Moreinae.

Of the subgroups of the Muricidae, the
group formerly known as Thaidinae (or as
Thaididae) Jousseaume, 1888 (original 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 generic allotment for the
many rapanine species is highly suspect, as
generic 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 Purpurinae. Broderip (1839) ranked this
group as a family (Purpuridae). 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 & Warén, 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 “Ригриппае,” “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 muricids (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. (1966) 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. Bandel
(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 PECTINIBRANCHIATA
MURICIDAE Rafinesque, 1815
CORALLIOPHILIDAE Chenu, 1859
Cossmann, 1903 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 5...)
CORALLIOPHILIDAE Chenu, 1859
Thiele, 1929 MURICACEA Rafinesque, 1815
MURICIDAE Rafinesque, 1815
MAGILIDAE Thiele, 1925
Wenz, 1941 MURICACEA Rafinesque, 1815
MURICIDAE Rafinesque, 1815
RAPANINAE Gray, 1853
(incl. Forreria)
COLUMBARIINAE Tomlin, 1928
MURICINAE Rafinesque, 1815
DRUPINAE Wenz, 1941
(incl. thaidines s./.)
MAGILIDAE Thiele, 1925
(incl. Coralliophila)
Radwin & D’Attilio, 1971 MURICACEA Rafinesque, 1815
COLUMBARIIDAE Tomlin, 1928
RAPANIDAE Gray, 1853
CORALLIOPHILIDAE Chenu, 1859
THAIDIDAE Jousseaume, 1888
MURICIDAE Rafinesque, 1815
(7 subfamilies)
Keen, 1971a 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
Ponder, 1973 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).
Golikov & Starobogatov, 1975 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 & Warén, 1988

MURICOIDEA Rafinesque, 1815

MURICIDAE Rafinesque, 1815

MURICINAE Rafinesque, 1815

(incl. Trophoninae, Ocenebrinae, etc.)
THAIDINAE Jousseaume, 1888

(incl. Rapaninae)
CORALLIOPHILINAE Chenu, 1859
MOREINAE Stephenson, 1941
?SARGANINAE Stephenson, 1923

TABLE 2. Ranking of thaidine higher taxa since Thaididae, Jousseaume, 1888, by a selection of authors.

Family Rank
Thaididae: Hedley, 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; Hoxmark, 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; Signor, 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 Thaididae/nae of au-
thors and other muricid 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; Bandel, 1984; Harasewych, 1984; Fu-
jioka, 1985а). Troschel (1866-1893) used
radular characters as the sole basis for his
classification.

Although radular characters in Thaididae/
nae 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, 1964 [Ra-
paninae, sensu Kool]; Wu, 1965a [Rapani-
nae, sensu Kool]; Powell, 1964 [Turridae];
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), Houbrick (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; Srilakshmi, 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 (Féral, 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; Bandel,
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 orbita (Gmelin,
1791), Drupa morum Róding, 1798, Haustrum
haustorium (Gmelin, 1791), Mancinella
alouina (Róding, 1798), Morula uva (Róding,
1798), Nassa serta (Вгидшеге, 1789), Neora-
pana muricata (Broderip, 1832), Nucella lapil-
lus (Linnaeus, 1758), Pinaxia versicolor (Gray,
1839), Purpura persica (Linnaeus, 1758),
Stramonita haemastoma (Linnaeus, 1767),
Thais nodosa (Linnaeus, 1758), Tribulus
planospira (Lamarck, 1822), Vasula melones
(Duclos, 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, Forreria 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 рег-
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.C., 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
Il) scanning electron microscopes at the U.S.
National Museum of Natural History. Pallial
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, |
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 Шиз-
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).

| 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. | 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-
ori to 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. А
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, Ригрига; 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. Calcitic 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. 8G, 25D, 18F, e)
4. 45° innermost aragonitic layer
(a) absent (e.g. Fig. 25D)
(b) present (e.g. Figs. 14E, 11С, Н, 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 ofthe 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. 1D)

(с) operculum D-shaped, tapered at
lower end, and with S-shaped left
(adjacent to columella) edge; lateral
nucleus in lower right (Fig. 12)

(d) operculum inverted tear-shaped; lat-
eral nucleus in lower right (Fig. 1B)

(e) operculum D-shaped; lateral nucleus
in center right (Fig. 1C)

(f) operculum ovate-elongate, tapered
at lower end; lateral nucleus in upper
right (Fig. 1E)

Foot and Mantle Cavity: The anatomical de-
scriptions are given as follows. In a first para-
graph, most ofthe external characteristics are
listed (coloration and morphology of tentacles
[e.g. Fig. 3B, t], head-foot region, kidney [e.g.
Fig. ЗВ, С, К], hypobranchial gland [e.g. Fig.
3B, С, hg], nephridial 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

FIG. 1. Morphologies of muricid opercula, showing free surface (facing to the outside) and attached surface
(facing inside), respectively. A, Muricanthus fulvescens. В, Rapana rapiformis. С, Thais nodosa. D, Forreria
belcheri. E, Vexilla vexillum. Е, Cronia amygdala; gr, growth lines; nu, nucleus; ri, rim of callus.

is measured from the posteriormost 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

ОСЬ EEE OO

PHYLOGENY ОЕ RAPANINAE 165

FIG. 2. Rod structures located in hypobranchial gland of Morula nodulosa. À, surface of hypobranchial gland
with rod structure in center (arrow), SEM (bar = 20 um). B, cross section through rod structure, SEM (bar
= 2 um).

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
(6) present (Fig. 2A, В)
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 pallial 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, bc), capsule gland
with left and right lobes (Figs. 3E, cg, 4C, Ic,
rc), ventral channel (Fig. 4C, vc), ovi-sperm
duct (connecting capsule gland with albumen
gland; Fig. 4Е-Н, 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, bc)
(b) continuous with capsule gland (Fig.
4D, bc)
10. Posterior seminal receptacles around al-
bumen gland
(a) absent (Fig. 4F, G)
(b) 1-3 with duct branching off ovi-
sperm duct (Fig. 4E, psr)
(c) many (usually at least 7 or 8) (Fig.
4H, psr)
11. 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. ЗВ, С, р, 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. 3B, C 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 рета! 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)
(Е) 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
(pallial 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 (154), 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 (x 10). С, Morula uva, male, right side (x 10). 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. С,
Nucella lapillus. H, Muricanthus fulvescens; ag, albumen gland; ant, anterior end; cg, 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; k
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; pef, 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, vesicula seminalis.

168 KOOL

abo abo

osd

FIG. 4. Morphologies of muricid female reproductive structures. А, В, sagittal cross sections through anterior
foot of female, viewed from right. А, ventral pedal gland and accessory boring огдап separate (e.g. Nucella
lapillus). В, ventral pedal gland and accessory boring organ combined (e.g. Thais nodosa). С, schematic
representation of anterior pallial gonoduct of female non-thaidine muricid (e.g. Nucella lapillus), viewed from
left, with cross section. D, schematic representation of anterior pallial 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; bc, bursa copulatrix; Ic, left lobe
of capsule gland; od, oviduct; osd, ovi-sperm duct; pg, ventral pedal gland; psr, posterior seminal recepta-
cles; rc, 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, |, penial morphologies in Muricidae. А,
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. |, Cronia amygdala. G-H, schematic 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; sl, 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:
15. 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)

(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:
17. 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 2, 269% Е)

(4) 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 сизр; 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 107

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; т-
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. 17Е).

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 Muricanthus
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

FIG. 6. A, egg capsule of Cymia tecta, apical view. В, 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/
nae 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,
Forreria belcheri, a taxon incertae sedis, and
Rapana rapiformis. | 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 (Bruguiére, 1789); synonym:
Buccinum concholepas Вгидшеге, 1789.

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 Вгидшеге.
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, C. concholepas, has priority. The
genus has one living and several fossil repre-
sentatives (Vokes, 1972; Kensley, 1985).
Haller (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 oriented parallel to growing
edge (15-20%); calcitic layer (60-70%) (Fig.
ТЕ).

Operculum: D-shaped (about one-third size
of aperture), with lateral nucleus in center
right (compare Fig. 1C). 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.

Anteriormost 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 pallial oviduct and located directly
beneath anal opening. Bursa copulatrix 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 um). Е, radula, SEM.

PHYLOGENY OF RAPANINAE 175

open chamber in interior vagina and open to
anterior portion of capsule gland. Posterior
part of pallial 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 pallial 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-
lial gonoduct.

Radula: Central cusp on rachidian 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 um 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 America (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.
(1980) 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 Н. & А. Adams, 1853
(Fig. 8A-D)

Cronia H. & A. Adams, 1853: 128 (as a sub-
genus of Purpura).

Type Species: Purpura amygdala Kiener,
1835, Бу monotypy, = Cronia amygdala
(Kiener, 1835); synonyms: ?Buccinum avel-
lana Reeve, 1846; ?Purpura aurantiaca Hom-
bron & Jacquinot, 1852; ?Ригрига pseu-
damygdala Hedley, 1902.

Remarks: The taxon Cronia was introduced
Бу Н. & A. Adams (1853: 128) as a subgenus
of Purpura “Aldrovandus” [correct author:
Bruguiere, 1789), with one species listed.
Cossmann (1903: 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 Hedley’s (1902)
figures of Purpura pseudamygdala. Kiener's
figures of Purpura amygdala bear more re-
semblance to the figures of Hedley's Purpura
pseudamygdala than to Hombron & Jacqui-
not's figures of Purpura aurantiaca, which 1$
most likely conspecific with Buccinum avel-
lana Reeve, 1846. | strongly suspect all four
“species” to be geographical or ecopheno-
typic variants of the same species. Cooke
(1919: 107) explained that Hedley 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 (Hedley,
1902: pl. 29, figs. 4-5). Teleoconch (Fig. 8A,
В) of 6-7 adpressed, high-spired, fusiform
whorls. Adult shell up to about 30 mm (includ-
ing 3 mm siphonal canal) in height and 15 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-
ole 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.

Operculum: D-shaped, with S-shaped left
edge, tapered at lower end, with lateral nu-
cleus in lower right (compare Fig. 1F). 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): Head-foot and siphon brown with
green, yellow and white specks, cephalic ten-
tacles long. Mantle edge smooth, following
aperture contour; incurrent siphon long. Hy-
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.

SN Г

FIG. 8. A-D, Сгота amygdala. A, shell (28 mm), apertural view. В, shell (28 mm), abapertural view. С, shell
ultrastructure, SEM (bar = 0.10 mm). D, radula, SEM (bar = 30 um). 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, ЗЕМ (bar = 45 um).

178 KOOL

than one-half ctenidial length; osphradium
and ctenidium 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, sl) pointing toward head and tenta-
cles. Penial duct as duct-within-a-duct system
(compare Fig. 5D, pvd) occupyina 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 pallial 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 Mórch, 1860
(Fig. 8E-H)

Cuma Humphrey, 1797 (rejected work).

Cuma Swainson, 1840: 87 (non Milne-Ed-
wards, 1828) [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: Висстит tectum Wood,
1828; Purpura angulifera Duclos, 1832.

Remarks: Swainson (1840: 87) placed Cuma
in the subfamily Pyrulinae, family Turbinell-
idae, and included only one species, Cuma
sulcata. Mórch introduced Cymia as a re-
placement name for Cuma Swainson, which
was pre-occupied, and placed it under Ra-
pana. 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 fasciole 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%); calcitic 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. 1C). 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
denticulation; 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 lredale, 1936
(Fig. ЭА-Р)

Dicathais lredale, 1936: 325.

Type Species: Висстит orbita Gmelin,
1791, by original designation, = Dicathais or-
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, 1952.

Remarks: lredale (1936: 325) removed suc-
cincta from the genus Neothias Iredale, 1912
(type: N. smithi Brazier, 1889, by original des-
ignation; emended [unjustified] by Iredale to
Neothais [1915: 473]), recognized orbita
Gmelin as its valid name and designated Di-
cathais orbita 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 (= orbita)” 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 orbita.

Shell: Protoconch (Fig. 9C, D) low, smooth,
of about four adpressed whorls, with outward-
flaring lip and sinusigeral notch. Teleoconch
(Fig. 9A, В) 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-
опа! canal а short but deep notch; posterior
siphonal canal absent. Siphonal fasciole
curved, about equally, or slightly more ele-

PHYLOGENY OF RAPANINAE 181

FIG. 9. Dicathais orbita. A, shell (58 mm), apertural view. В, shell (58 mm), abapertural view. С, protoconch,
side view, SEM (bar = 0.20 тт). D, protoconch, apical view, SEM (bar = 0.20 mm). E, shell ultrastructure,
SEM (bar = 30 um). Е, radula, SEM (bar = 40 pm).

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%); calcitic layer (20-55%)
(most pronounced at ribs) (Fig. 9E).

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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 smali 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
pallial 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 arid 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 pallial 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 ОЕ 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 orbita 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) (Fairweather, 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 | collected in Botany Bay (New South
Wales, Australia), which had made these in-
dividuals sterile. Phillips (1969) also found
trematodes 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., 1973; Powell,
1979).

Genus Drupa Röding, 1798
(Fig. 10A-E)

Drupa Röding, 1798: 55.

Canrena Link, 1807: 126 [type: Murex neritoi-
deus Linnaeus, 1767 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, 1758, =
Drupa ricinus (Linnaeus, 1758)].

Ricinula Lamarck, 1816: 1, pl. 395 [type:
Ricinula horrida Lamarck, 1816, by sub-
sequent designation, Children, 1823: 56
(as Ricinula horida), = Drupa тогит
Röding, 1798].

Ricinulus Lamarck; Chenu, 1859: 174 (invalid
emendation for Ricinula Lamarck).

Ricimula А. А. Gould, 1855: 263 (error for
Ricinula Lamarck).

Ricinella Schumacher, 1817: 240 [type: Ri-
cinella purpurata Schumacher, 1817, by
subsequent designation, Iredale, 1937:

256, = Drupa rubusidaeus Röding,
1798].
Pentadactylus Mörch, 1852: 87 [поп

Schultze, 1760, nec Gray, 1840] [type:
Murex ricinus Linnaeus, 1758, by subse-
quent designation, Baker, 1895: 186, =
Drupa ricinus (Linnaeus, 1758)].
Drupina Dall, 1923: 303 [type: Ricinula digi-
tata Lamarck, 1816, by original designa-
tion, = Drupa grossularia Röding, 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 horr-
ida Lamarck, 1816; Ricinella violacea Schu-
macher, 1817; Ricinula horida Lamarck, Chil-
dren, 1823 (error for horrida).

Remarks: Cossmann (1903: 68) considered
Ricinula (= Drupa) a full genus. Thiele (1929:
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 Огиртае. 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. À, shell (35 mm), apertural view. B, shell (33 mm), abapertural view. C, radula,
SEM (bar = 25 um). 0-Е, Drupa grossularia. D, protoconch, side view, SEM (bar = 0.10 тт). 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
calcitic layer questionable.

PHYLOGENY OF RAPANINAE 185

Operculum: D-shaped, tapered at lower end,
with lateral nucleus in center right (compare
Fig. 1C). 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 pallial 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. 10C). 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 Огира. 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 complanata (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. | 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.
| have collected Drupa grossularia at 10 т
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, pl. 44.

Lepsia Hutton, 1884: 222 [type: Висстит
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 (Gmelin,
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. 11A, 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).

Орегсшит: D-shaped, upper end rounded,
with lateral nucleus in lower right (compare
Fig. 1D). 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 ctenidium 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 ctenidium 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 ОЕ RAPANINAE 187

FIG. 11. A-D, Haustrum haustorium. A, shell (48 mm), apertural view. В, shell (48 mm), abapertural view.
С, shell ultrastructure, SEM (bar = 0.10 mm). D, radula, SEM (bar = 25 вт). E-I, Mancinella alouina. Е,
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). |, radula, ЗЕМ (bar = 40 um).

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;
?Volema 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
(1941: 1118) 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 ОЕ RAPANINAE 189

(Lamarck, 1816) (= М. aculeata Link) be des-
ignated as the type of Mancinella. The ICZN
ruled (Opinion 911, 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. 11E, 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%);
calcitic layer (4-6%) (Fig. 11G, H).

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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 anteriorly 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 nearly 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 nearly 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. 111). 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 Duclos, 1832, by original des-
ignation, = Morula granulata (Duclos,
1832)].

Type Species: Morula papillosa Schuma-
cher, 1817 (non Philippi, 1849), by monotypy,
= Morula uva (Röding, 1798); synonyms:
Drupa uva Röding, 1798; Ricinula nodus La-
marck, 1816; Ricinula aspera Lamarck, 1816;
Ricinula morus Lamarck, 1822; Purpura
sphaeridia Duclos, 1832; Ricinula alba
Mörch, 1852; ?Sistrum striatum Pease, 1868;
?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 fasciole 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. 1F). 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 рт). D, protoconch, apical view, SEM (bar = 60 um). E, penis, viewed
postero-anteriorly, SEM (bar = 0.20 mm). F, shell ultrastructure, SEM (bar = 0.10 mm). G, radula, SEM (bar
= 10 um).

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
сопсауе; 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 pallial 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 “7”
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. 1ЗА-С)

Nassa Röding, 1798: 132 (non Lamarck,
1799, = Nassarius Duméril, 1806).
lopas H. & A. Adams, 1853: 128 [type: Buc-

cinum sertum Bruguiére, 1789, by sub-

PHYLOGENY OF RAPANINAE 193

FIG. 13. A-C, F-G, Nassa serta: À, shell (40 mm), apertural view. B, shell (44 mm), abapertural view. C,
larval shell, side view, SEM (bar = 25 um). Е, shell ultrastructure, SEM (bar = 0.10 тт). С, radula, SEM,
(bar = 25 рт). D-E, Nassa “francolina” D, protoconch, side view, SEM (bar = 80 um). E, protoconch,
apical view, SEM (bar = 80 um).

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 Гат” as
type of Nassa, which was not one of the spe-
cies included by Röding and is therefore un-
available. | can find no valid subsequent des-
ignation and here designate the type species
as Nassa picta Roding, 1798, = Nassa serta
(Bruguiére, 1789); synonyms: Buccinum ser-
tum Bruguiére, 1789; Buccinum coronatum
Gmelin, 1791; ?Stramonita hederacea Schu-
macher, 1817; ?Buccinum francolinus Bru-
guiére, 1789; Buccinum situla Reeve, 1846.

Remarks: Cossmann (1903: 68) considered
Nassa a full genus (as lopas), and included,
besides /ора$ s.s, Taurasia Bellardi, 1882.
Thiele (1929: 296) used Jopas and included
the subgenera Jopas (= Nassa) and Vexilla.
Wenz (1941: 1116) 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. |
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.

Shell 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. 1C). 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 pallial
gonoduct. Rectal gland dark green, thin,
alons entire capsule or prostate.

Нааша: 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. В, shell (45 mm), abapertural view. С,
protoconch, side view, ЗЕМ, (bar = 0.20 mm). D, protoconch, apical view, ЗЕМ (bar = 0.10 mm). E, shell
ultrastructure, ЗЕМ (Баг = 0.20 тт). Е, radula, SEM (bar = 35 рт).

ES CU ee

PHYLOGENY OF RAPANINAE 197

Type Species: Purpura muricata Broderip,
1832, by original designation, = Neorapana
muricata [Broderip, 1832]; synonyms: Pur-
pura truncata Duclos, 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 Mazatlan, Mexico (Keen, 1971b), thus
partially overlapping in range with N. muri-
cata. | 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, | 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 рго-
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 fasciole 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%); calcitic layer (8-15%) (Fig.
14Е).

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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 сопсауе; 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 pallial 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-
магу 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 Leiblein.
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. | 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 Réding, 1798
(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 ВАРАММАЕ 199

FIG. 15. Nucella lapillus. À, 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). Е, radula, ЗЕМ (bar = 20 um). С, radula, side view, SEM (bar = 10 pm).

200 KOOL

Type Species: Buccinum filosum Gmelin,
1791, by subsequent designation, Stewart,
1927: 386 (footnote 260), = Nucella lapillus
(Linnaeus, 1758); зупопутз: Висстит lapil-
lus Linnaeus, 1758: 739; Nucella theobroma
Röding, 1798; Purpura imbricata Lamarck,
1822; Purpura bizonalis Lamarck, 1822; Pur-
pura buccinoidea Blainville, 1829; Purpura
celtica Locard, 1886; Coralliophila rolani Bogi
& 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 fasciole
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%); calcitic layer (40—
85%) (Fig. 15G).

Operculum: D-shaped, upper end rounded,
with lateral nucleus in lower right (compare
Fig. 1D). 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 тю-
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 (Emlen, 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 (1938) 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. 16А-Е)

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. | 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), apertural view. В, shell (17 mm), abapertural view. С, proto-
conch, apical view, SEM (bar = 0.10 mm). D, protoconch, side view, SEM (bar = 0.10 mm). E, radula, SEM

(bar = 10 pm).

cream brown periostracum. 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 тт)
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 fasciole forming thin,
slightly elevated ridge adjacent to callus on
lower columella. Shell yellow to orange with
10—11 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. 1C). Free side with
bracket-shaped growth rings; attached side
without or with 1-2 bracket-shaped growth

ИН

PHYLOGENY ОЕ RAPANINAE 203

lines and with callused, glazed пт (about 30 —
45% of opercular width) on left.

Anatomy (based оп 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, 1987) to Japan (Abbott & Dance,
1982).

Genus Plicopurpura Cossmann, 1903
(Fig. 17A-F)

Plicopurpura Cossmann, 1903: 69 (as section
of Ригрига).

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

FIG. 17. Plicopurpura patula. A, shell (53 mm), apertural view. В, shell (53 mm), abapertural view. С,
protoconch, side view, SEM (bar = 70 um). D, protoconch, apical view, SEM (bar = 0.10 mm). Е, radula,
SEM (bar = 20 um). Е, shell ultrastructure, SEM (bar = 0.15 mm).

PHYLOGENY OF RAPANINAE 205

Purpurella Dall, 1871: 110 (non Robineau-
Desvoidy, 1853, nec Bellardi, 1883; as
subgenus of Purpura) [type: Purpura col-
umellaris Lamarck, 1816, by original des-
ignation, = Plicopurpura columellaris
(Lamarck, 1816)].

Microstoma Paetel, 1875: 126 (error for Mi-
crotoma Swainson).

Patellipurpura Dall, 1909: 50 [type: Висстит
patulum Linnaeus, 1758, by monotypy,
= Plicopurpura раша (Linnaeus, 1758);
as section of Thais].

Patellapurpura Abbott, 1974: 180 (error for
Patellipurpura Dall).

Type Species: Purpura columellaris Lama-
гск, 1816, by original designation, = Pli-
copurpura columellaris (Lamarck, 1816); syn-
onyms: ?Buccinum patulum Linnaeus, 1758;
Haustrum dentex Perry, 1811 [nomen obli-
tum; ICZN, Opiniori 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 Р/-
copurpura and Patellipurpura as subgenera.
Keen (1971b: 552) indicated that Plicopur-
pura is perhaps a nodose subgenus of Pur-
pura. Kool (1988b) 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, Р. раша, 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, | 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 & Kuris (1983) provided ev-
idence for conspecificity of these two nominal
species. | 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. 17А, В) large, oval, of 5—6 adpressed
whorls, and with high whorl-expansion 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 fasciole 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 (80-35%); 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-
citic layer questionable; scored with “?” in cla-
distic analysis.

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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. 3A): 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 pallial 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-
пог 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 columellaris. 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 ОЕ 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; Bandel, 1987; Kool, 1987) and
nerites (Britton & Morton, 1989), and also on
barnacles (Lewis, 1960; Kool, 1987). As de-
scribed by Bandel (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. Bandel noted that Р/-
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, 1974). Occurrence of a
Plicopurpura-like shell on Mauritius (Drivas &
Jay, 1987) needs further investigation.

Genus Purpura Bruguiere, 1789
(Fig. 18A-G)

Purpura Bruguière, 1789: 15 (non Röding,
1798, nec Lamarck, 1799).

Type Species: Buccinum persicum Lin-
naeus, 1758, by subsequent designation,
ICZN, Opinion 886, 1969: 128, = Purpura
persica (Linnaeus, 1758); synonym: ? Purpura
inerma Reeve, 1846.

Remarks: The generic name “Ригрига” was
first used by Martini (1777) and subsequently
by Martyn (1784) and Meuschen (1787), all of
which are non-binominal works. Bruguiere
formally introduced Purpura as a genus in
1789, but did not mention any species. Three
years later, Bruguiere (1792) included the
nominal species Purpura tubifer Bruguiere,
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. persica as the
sole species in the genus, which did not result
in P. persica 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 persica be designated type species
of Purpura. Purpura persica 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 persica as the
sole example of the genus Purpura. Thiele
(1929: 296) incorrectly cited Purpura patula
as type of Purpura, and synonymized Patel-
Пригрига 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 muricine “Purpura” foliata. Keen
(1971b: 552) synonymized the genera Pli-
copurpura and Patellipurpura with Purpura.
Kool (1988b) 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
7-15 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

SEITE

FIG. 18. Purpura persica. 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 um). 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 ОЕ 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, с) (35—55%);
aragonitic layer with crystal planes oriented
perpendicular to growing edge (Fig. 18F, d)
(5-15%); calcitic layer (5-10%) (Fig. 18F, e).

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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-
попу. Osphradial lamellae attached along
small portion of their base.

Anteriormost portion of ctenidium straight,
equidistant from mantle edge with osphra-
Чит. 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
pallial gonoduct and located directly below
anal opening. Small bursa copulatrix 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, 1987) to Marquesas Islands (Sal-
vat & Rives, 1975).

Genus Stramonita Schumacher, 1817
(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 grisea
Röding, 1798; Thais metallica Röding, 1798;
Thais nebulosa Röding, 1798; Thais stellata
Röding, 1798; Purpura floridana Conrad,
1837; Purpura consul Reeve, 1846; Purpura
forbesii Dunker, 1853; Thais floridana haysae
Clench, 1927; Thais (Stramonita) hidalgoi
Coen, 1946; ?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, Bandel (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-
tural 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 ОЕ RAPANINAE 211

FIG. 19. Stramonita haemastoma. А, shell (33 mm), apertural view. В, shell (33 mm), abapertural view. С,
protoconch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, radula,
SEM (Баг = 25 um). Е, Shell ultrastructure, fracture surface, SEM (bar = 0.15 mm).

212 KOOL

lus. Siphonal fasciole 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 оп-
ented parallel to growing edge (30-40%);
calcitic layer (40-60%) (Fig. 19F).

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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 pallial
gonoduct (in typical S. haemastoma morphs;
in rounded morphs, vagina more elongate)
and located below and slightly anterior to anal
opening. Bursa copulatrix 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. Mid-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 ОЕ 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
(О’Азаго, 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
(Bandel, 1976), barnacles (Cake, 1983), and
polychaetes (Phragmatopoma sp.) (Kool,
1987). A variety of ecological topics was
treated by Gunter (1979). | 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.

?Thalessa H. & A. Adams, 1853: 127 [type:
Murex hippocastanum Linnaeus, 1758,
by subsequent designation, F. C. Baker,
1895: 183 (Suppressed by ICZN, Opin-
ion 911, 1970: 20), = Thais aculeata
(Deshayes, 1844)].

?Menathais Iredale, 1937: 256 [type: Purpura

pica Blainville, 1832, by original designa-
tion, = Thais tuberosa (Röding, 1798)].

?Thaisella Clench, 1947: 69 [type: Purpura
trinitatensis Guppy, 1869, by original
designation, = Thais trinitatensis
(Guppy, 1869)].

?Reishia 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 Quoy & 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
Сте!”) in a synopsis of Dall’s (1909) work.
Stewart (1927: 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 (р. 256) Iredale listed
“... Thais lena Bolten [sic] = Murex fucus
Gmelin, . . .” as the type species, with this
type species fixed as Murex fucus Gmelin,
1791, 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. В, shell (25 mm), abapertural view. С, 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). Е, radula, SEM (bar = 25 рт).

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 meretricula 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 T.
nodosa form meretricula) 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. 1C). 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 pallial 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, 1839
(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-
сте!а 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. 21C, 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. 21A, 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-
опа! 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

UN
\

ARO

en sl

isis

FIG. 21. Tribulus planospira. A, shell (50 mm), apertural view. В, shell (50 mm), abapertural view. С,
protoconch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, radula,

SEM (bar = 35 um).

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. 1C). Free surface
with bracket-shaped growth 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. 21E). 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
Galäpagos Islands (Забей & 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 Duclos,
1832, by monotypy, = Vasula melones (Du-
clos, 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
fasciole 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 calcitic layer
questionable.

Operculum: D-shaped, with lateral nucleus in
center right (compare Fig. 1C). 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. À, shell (45 mm), apertural view. B, shell (45 mm), abapertural view. C, shell
ultrastructure, polished fracture surface, SEM (Баг = 0.20 тт). D, radula, SEM (bar = 35 um). E, radula,

rachidian row, SEM (bar = 20 um).

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.

Vaginal opening enlarged, protruding from
short, tubular extension of pallial 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 posteriorly. 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 оп-
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. 220, Е); 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 ргопоипсеа, 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 Galäpagos 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 (1941: 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
Ир smooth, polished, with crenulations on
edge continuing inward as small ridges for
short distance. Anterior siphonal canal a
poorly developed notch. Posterior siphonal
canal flanked on left by small protrusion of
columellar cailus. Columella rounded to flat,
with little callus, curving inward at lower por-
tion. Siphonal fasciole 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. 1E). Free surface without distinct growth

PHYLOGENY OF RAPANINAE 221

Ken x 74

FIG. 23. Vexilla vexillum. A, shell (14 mm), apertural view. B, shell (14 mm), abapertural view. C, radula,
SEM (bar = 20 um). D, protoconch, apical view, SEM (bar = 50 um). E, protoconch, side view, SEM (bar

= 50 um).

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. Pallial 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 For-
reria belcheri 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 whorls smooth, later whorls with micro-
scopic pustules. Protoconch | nearly as wide
as first whorl of Protoconch Il. 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 whorl
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. В, shell (136 mm), abapertural view. С,
protoconch, side view, SEM (bar = 0.25 тт). D, shell ultrastructure, fracture surface, SEM (x 35). E,
radula, SEM (bar = 50 um). Е, protoconch, apical view, SEM (bar = 0.10 тт).

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 fasciole 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. 240).

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 pallial 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 ОЕ 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
bulbose, 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%); calcitic layer (10-15%) (Fig.
25D).

Operculum: Inverted tear-shaped, with lat-
eral nucleus in lower right (Fig. 1B). 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 pallial
gonoduct and located below and slightly an-
terior 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. В, protoconch, side view, SEM (bar = 0.20
mm). С, radula, SEM (bar = 0.10 mm). D, shell ultrastructure, SEM (bar = 75 um). E, radula, rachidian row,
SEM (bar = 30 pm). Е, sagittal cross section through anterior foot of female viewed from right side, showing
accessory boring огдап (abo), ventral pedal gland (pg), and transverse furrow (tf), SEM (bar = 0.50 mm).

PHYLOGENY OF ВАРАММАЕ 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 паг-
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 10-11 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%); calcitic layer (70-80%) (Figure
26F).

Operculum: D-shaped, upper end rounded,
with lateral nucleus in lower right (Fig. 1D).
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 pallial
gonoduct and located below and slightly pos-
terior to anal opening. Bursa copulatrix as

228

звать HOLE тт

de de 1> In 303

SR Fehr

h, side view, SEM (Баг = 80 um).

‚ radula, SEM (bar = 50 um). E, radula, гас ап row,

‚ apertural view. В, protoconc
SEM (bar = 25 um). Е, shell ultrastructure, SEM (bar = 0.10 mm).

С, protoconch, apical view, SEM (bar = 80 pm). D

FIG. 26. Forreria belcheri. À, shell (114 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-
за! 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, 1774)
(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. 1D).
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. 27C, D) is smooth, paucispiral
(about 1.5 whorls), and lacks an outward-flar-

ing lip.

Trochia cingulata (Linnaeus, 1758)
(Fig. 28А-Е)

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. 28C, 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. А, shell (46 mm), apertural view. В, shell (46 mm), abapertural view. С,
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). Е, radula, SEM (bar = 10 um). С,
protoconch, apical view, SEM (bar = 0.10 mm).

PHYLOGENY ОЕ ВАРАММАЕ 231

FIG. 28. Trochia cingulata. A, shell (40 mm), apertural view. В, shell (40 mm), abapertural view. С, proto-
conch, side view, SEM (bar = 0.10 mm). D, protoconch, apical view, SEM (bar = 0.10 mm). E, shell
ultrastructure, SEM (bar = 50 um).

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), apertural view. В, shell (71 mm), abapertural view. С,
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 C (Fig. 30), based on a
wide variety of characters, many of which it
shares as synapomorphies with Forreria
belcheri, the type species of Forreria, 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 (Acanthina monodon and Trochia 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 233

Ocenebrinae
ic Pe 28] АНИ

= о a £ soc
в 5955 о
г ZEIZ 55686

FIG. 30. Consensus cladogram with taxonomic groupings superimposed. Mur = Muricanthus; Hau
Haustrum; Мис = Nucella; Tro = Trochia; For =

Forreria; Aca =

Rapaninae

nn ne

Acanthina; Cym = Cymia; Stra

Stramonita; Rap = Rapana; Con = Concholepas; Dic = Dicathais; Vex = Vexilla; Nas = Nassa; Pin =

Pinaxia; Dru = Огира; Рис = Plicopurpura; Mor =

Morula; Cro = Cronia; Маз = Vasula; Tha = Thais; Pur

= Purpura; Man = Mancinella; Neo = Меогарапа; Trib = Tribulus.

groups are not sufficiently clear-cut to justify
familial ranking for all three clades. | 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 B show no more
resemblance with members of Clade C 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 C
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, 1989); either these two gen-
era are very highly derived members of Clade
C, or their placement in Clade C 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. Numbers and letters correspond to those in text.

Character U 2 eth от 8

Taxon
Muricanthus
Forreria
Nucella
Haustrum
Morula
Cronia
Rapana
Cymia
Stramonita
Concholepas
Dicathais
Vasula
Vexilla
Nassa
Pinaxia
Drupa
Plicopurpura
Thais
Purpura
Mancinella
Neorapana
Tribulus
Acanthina
Trochia

TOV VV D D D VA DADOD D YH ONO DS OT OT m
© OM D VA AMA VA DMA рю VA D D VOOM
TYIO OO 0Q oO NVM AA лоб бсбоорросс»р
9 9 OO OO OU M D D MHA ONMMA nm M m m m m m m
OO OP D D D D D D D —# D D D D D ANA NA CO OU
Y) Y DY D M M M M D D D D py pp pp) mm m
-) D M M M M M M M M M Oo M M M © co op NMA TOM
) 9 O O OO OO OO OO OO D DAH TON HD DM m

Clade C 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 B than
it is with any о the species of Clade С. Two of
the taxa of Clade B (Acanthina and Trochia)
were not examined in detail for this study, but
they grouped unambiguously with Nucella
and Forreria 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 C (individual clades treated sepa-
rately): Although Cymia is included in Clade
C, it shares a synapomorphy with the species
of Clade B (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 B (posterior seminal receptacles [fe-
males]). However, Cymia shares several sy-

D 9 TO OO OO OU OO OO OO OO OO O OS ppp

0’ 1! 12 13 14 a TG 17 16
а а а а а а а а а
а b a a a b ? b c
a b b с а b b b с
а b b b a e b a b
b с © d b с а а е
b d (© d b E a a e
с d e d b b a a f
a ? d d b b a a d
© d e d b b a a g
с d e d b b a a g
с i? e d b с а а 9
с d e d b Cc ? а ]
C d f d b d a a ?
с а f d b © а а 2
(© а f d b € a a h
C d e d b d a a h
с а е а b с а а ?
€ d e d b lo a a j
с а е d b с а а h
C Y e d b (© а а i
с d e d b с а а ]
? ? е а ? 2 Y a j
2 ? b 2, q ? Y 2 ?
? Y 2 2 2 2 2 2 2

napomorphies with all other taxa of Clade B
(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 С 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, Меогарапа,
and Vasula suggests that at these four genera
are only distinct at the subgeneric level; | 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. | propose that a combination of these
factors is the cause for a low resolution in
Clades D and Е, as well as in Clades B and G.
К 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, 12, and 18). | 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 С, | would choose a scheme whereby
both (b) and (с) evolved from (a), although it
would be equally parsimonious to assume a

linear transformation series [(а) — (b) — (с)
or (а) — (с) > (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 (с) (e.g. Fig. 9С); 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:—Calcitic 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. 20Е). This layer not
only adds thickness to the shell, but presum-
ably also gives more strength to it, which may
serve as defense to predation.

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. 1A). 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.
1D), and a D-shaped operculum with a lateral
nucleus in the center right (e) (e.g. Fig. 1C).
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 (с) (e.g. Fig. 12); 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.

4B). 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 а
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-
latrix (9). А 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 (10). 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 (с) (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, |, sl)
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 (pallial 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 pallial 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 (16). 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). À 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 (|)
(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. 111). 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 А, 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 B 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 B (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).
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 calcitic layer (2) and separate ducts
for the accessory boring organ and ventral
pedal gland (3) are also found in Clade C
(Cymia) and are probably the result of parallel
evolution. Absence of an outward-flaring lip
(1) may become a synapomorphy for Clade
А, once it is shown that the transition from
protoconch to teleoconch in Haustrum haus-
torium is smooth.

Clade C (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

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 Iying
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 C (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-

239

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.

К is clear—to the dismay of many paleon-
tologists—that the meager fossil record (in
this case of the Rapaninae), cannot а priori be
interpreted with any degree of certainty. Nev-

240 KOOL

ertheless, the fossil record is potentially use-
ful. А phylogenetic tree resulting from 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

А comprehensive study, ideally of топо-
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. | 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, B (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 “T.” wahlbergi (Krauss,
1848), are ocenebrines; further research may
reveal that these species should be placed in
Nucella, as suggested by Kilburn & Rippey
(1982), thus extending the range of the genus
Nucella considerably. Forreria 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, 1832), 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, ос-
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. | suspect that future studies of
“Stramonita-like” and “Thais-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
Reunion and Mauritius (Drivas & Jay, 1987) 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 Clade D no distinct trend in reduc-
tion or increase in number of whorls is visible;
some of the highest numbers of whorls occur
in Clade 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, |
predict that well-preserved protoconch spec-
imens of the species of Clade G (Fig. 30) will
reveal a sculptural pattern as found in most
members of Clade Е (3—4.5 whorls, with sub-
sutural plicae). The cladogram furthermore
predicted that Haustrum haustorium has a
paucispiral, smooth protoconch, which | 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 (Clade A, Fig. 30)
and the remaining members of the rapanine
clade (Clade D).

Evidence obtained from protoconch mor-
phology indicates that all members of the Ra-
paninae studied herein (Clade C, 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/
nae 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 Clade B (Fig. 30) and, perhaps, for
Clade 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 сюзеу related to Nu-
cella and should also be excluded from Ва-
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 7ro-
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; О’Азаго, 1991).

Forreria (Fig. 26A-F) may be ciosely re-
lated to the genus Chorus, an eastern Pacific
genus from the Chilean waters. Future stud-
ies may show that Chorus and Forreria 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. 29А-Е), 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 а cal-
citic layer (45-55%) (Fig. 29E). This type of
shell ultrastructure is found in Nucella and re-
lated taxa, such as Trochia and Forreria, 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, | 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 calcitic outer layer
(Petitjean, 1965), it may belong in а clade with
Nucella, Acanthina, Trochia, and Forreria.

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 (Bandel, 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. 11D) is very
different from the rachidian found in Nucella
(Fig. 15F) and Forreria (Fig. 26E), and sug-
gested that either Haustrum was the “ргодеп-
itor’ of the Thais and Nucella groups (making
a clear distinction between the “Nucella”
group and the “Thais” group [рр. 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 ОЕ 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. Bandel (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). | 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 а pro-
nounced marginal cusp.

Clade D: Troschel (1866-1893) placed
Nassa (Fig. 13G) close to Plicopurpura (as
“Patellipurpura”) (Fig. 17E), based on rachid-
ап 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. | 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-
рига (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). | do not agree with this
statement; the rachidian tooth morphology of
Plicopurpura is unique and should be consid-
ered as derived.

Clade Е: 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. 10C)
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-
оду of the rachidian tooth (Fig. 111). | 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 “зам Же”) 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-
ре, | would interpret the “sawlike” condition
in Drupa as more comblike and homologous
with the comblike condition in Ригрига; addi-
tionally, | 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 Forreria (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-
ра, but not comblike, rachidian tooth. А more
or less comblike condition occurs only in more
derived rapanines, 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 persica, 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).

Bandel (1976) provided a phylogenetic hy-
pothesis for evolution of egg capsule тог-
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, Forreria,
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 C (long accessory salivary glands,
separate ventral pedal gland [females] and
boring organ, very thick outer calcitic 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 C. 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 C.
| 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/
nae 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 C (the remains of Thaididae/
nae of authors) needs to be reevaluated. Be-
cause Rapana is monophyletic with the other
taxa in Clade C (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 ОЕ 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 B (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-like 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, 1888]

Concholepas Lamarck, 1801

Cronia H. & A. Adams, 1853

Cymia Mörch, 1860

Dicathais lredale, 1936

Drupa Röding, 1798

Mancinella Link, 1807

Morula Schumacher, 1817

Nassa Röding, 1798

Pinaxia H. & A. Adams, 1853

Plicopurpura Cossmann, 1903

Purpura Bruguiere, 1789

Rapana Schumacher, 1817

Stramonita Schumacher, 1817

Thais Röding, 1798
Neorapana Cooke, 1918
Tribulus Sowerby, 1839
Vasula Mörch, 1860

Vexilla Swainson, 1840

Ocenebrinae Cossmann, 1903
[+ Ecphorinae, Petuch, 1988
+ Nucellinae Kozloff, 1987]

Acanthina Fischer von Waldheim, 1807
Ecphora Conrad, 1843

Forreria Jousseaume, 1880

Haustrum Perry, 1811

Nucella Röding, 1798

Trochia Swainson, 1840

ACKNOWLEDGMENTS

| 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. | 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 Е. Knowlton, who рго-
vided many valuable suggestions for this
manuscript.

| thank Mrs. Susann G. Braden, Mr. Walter
R. Brown and Mr. Brian E. Kahn of the Scan-
ning Electron Microscopy Laboratory at the
USNM. | 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. | 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.

| 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.

| gratefully acknowledge the support of the
Smithsonian’s Caribbean Coral Reef Ecosys-
tems Program, and thank Dr. Klaus Rützler
for funding my stay atthe 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.

| thank those who have assisted me during
visits to their institutions; Dr. James H.
McLean and Mr. C. 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.

| 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. | 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.

| 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 |. Johnson assisted
with some nomenclatorial problems.

APPENDIX 1

Species Examined Thaididae/nae of au-
thors:
Concholepas
1789)
Cronia amygdala (Kiener, 1835)
Cymia tecta (Wood, 1828)
Dicathais orbita (Gmelin, 1791)

concholepas (Bruguiere,

Drupa morum Röding, 1798

Haustrum haustorium (Gmelin, 1791)

Mancinella alouina (Röding, 1798)

Morula uva (Röding, 1798)

Nassa serta (Bruguiere, 1789)

Neorapana muricata (Broderip, 1832) *1

Nucella lapillus (Linnaeus, 1758)

Pinaxia versicolor (Gray, 1839)

Plicopurpura patula (Linnaeus, 1758) *2

Purpura persica (Linnaeus, 1758)

Stramonita haemastoma (Linnaeus, 1767)

Thais nodosa (Linnaeus, 1758)

Tribulus planospira (Lamarck, 1822)

Vasula melones (Duclos, 1832)

Vexilla vexilla (Gmelin, 1791)

Acanthina monodon (Pallas, 1774) *3

Trochia cingulata (Linnaeus, 1771) *3

Ecphora cf. quadricostata (Say, 1824) *3
Rapaninae, of authors:

Forreria belcheri (Hinds, 1844)

Rapana rapiformis (Born, 1778) *4
Muricinae:

Muricanthus fulvescens (Sowerby, 1841)

*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
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-
apomorphies present in some taxa could be
recognized in these, facilitating taxonomic al-
location. Therefore they were only examined
for synapomorphic (diagnostic) characters.

*4 Rapana rapiformis (Born, 1778) 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.

*5 Muricanthus fulvescens (Sowerby, 1841)
was chosen to represent the Muricinae as an
outgroup in the cladistic analysis, because
many living and well-preserved specimens
were available.

AMS:

ANSP:
LACM:

МСН:
SEM:
SPK:

USNM:

ZMA:

PHYLOGENY OF RAPANINAE 247

APPENDIX 2

List of abbreviations used in text.
Australian Museum, Sydney.

Academy of Natural Sciences, Philadelphia.
Los Angeles County Museum.

Myroslaw George Harasewych.

Scanning electron micrograph.

Silvard Paul Kool.

United States National Museum.
Zoologisch Museum, Amsterdam.

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 orbita
USNM 836862
USNM 681578
USNM 836862
USNM 836862
USNM 618246
Drupa morum
USNM 857059
USNM 720340
USNM 857059
USNM 857059
USNM 672111

Haustrum haustorium

AMS no number
AMS no number
USNM 531495
USNM 531495
USNM 76300

Mancinella alouina

AMS по 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

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 persica
СМА no number
MNHL no number
ZMA no number
ZMA no number
USNM 700108

Stramonita haemastoma

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

KOOL

Anatomy: Pago Bay, Guam, U.S.A.

Protoconch (N. “francolina”): Nossi Be, Madagascar

Larval shell: Gatope Island, New Caledonia
Radula: Pago Bay, Guam, U.S.A.

Ultrastructure: Gigmoto, Catanduanes Islands, Philippine Islands

Shell: Samoa Islands

Shell: Low Wooded Island, N. Queensland, Australia

Anatomy: Puerto Penasco, Sonora, Mexico
Protoconch: Acapulco, Mexico

Radula: Puerto Peñasco, Sonora, Mexico
Ultrastructure: Puerto Penasco, Sonora, Mexico
Shell: San Carlos, Sonora, Mexico

Anatomy: Kittery, Maine, U.S.A.

Protoconch: Manchester, Massachusetts, U.S.A.
Radula: 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.
Radula: Ambatoloaka, Madagascar
Ultrastructure: Ambatoloaka, Madagascar
Shell: Mogadishu, Somalia

Anatomy: South Miami Beach, Florida, U.S.A.
Protoconch: San Blas Islands, Panama
Radula: South Miami Beach, Florida, U.S.A.
Ultrastructure: Cozumel Island, Mexico

Shell: Mujeres Island, Mexico

Anatomy: Krakatoa, Indonesia

Protoconch: Tjoba, Tidore, Indonesia

Radula: Krakatoa, Indonesia

Ultrastructure: Krakatoa, Indonesia

Shell: Taiohae Bay, Nukuhiva, Marquesas Islands

Anatomy: Sebastian, Florida, U.S.A.
Protoconch: Cocoa Beach, Florida, U.S.A.
Radula: 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
Radula: Monrovia, Liberia
Ultrastructure: Ascension Island
Shell: Monrovia, Liberia

Anatomy: Galäpagos Islands, Ecuador
Protoconch: Malpelo Island, Colombia

Radula: Galäpagos Islands, Ecuador
Ultrastructure: Ensenada de los Muertos, Mexico

Shell: Academy Bay, Isla Santa Cruz, Galapagos Islands

Anatomy: Palo Seco, Panama

Radula: Marchena, Punta Estego, Galäpagos 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 по number
USNM по 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. quadricostata
USNM no number
USNM no number
MCZ 263350

LITERATURE CITED

АВВОТТ, В. Т., 1974, American seashells, 2nd. ed.
Van Nostrand Reinhold Company. New York,
663 pp.

ABBOTT, R. T. & S. P. DANCE, 1982, Compen-
dium of seashells. Dutton, New York, 411 pp.
ABE, N., 1983. Breeding of Thais clavigera (Kuster)
and predation of its eggs by Cronia margariticola
(Broderip). Рр. 381-392, in: В. MORTON & D. Duo-
GEON, eds., The malacofauna of Hong Kong and
southern China. 1., 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 pls.

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

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.

marine gastropods—I. Journal of the Shi-
monoseki College of Fisheries, 7: 107—116.

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-
schaft, 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, К. 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, Е. С., 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.

BANDEL, K., 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.

BANDEL, K., 1984, The radulae of Caribbean and
other Mesogastropoda and Neogastropoda. Zo-
ologische Verhandelingen, 214: 176 pp.

BANDEL, K., 1987, Hydroid, amphineuran and gas-
tropod zonation in the littoral of the Caribbean
Sea, Colombia. Senckenbergiana Maritima, 19:
1-129.

BARNARD, К. H., 1959, Contributions to the knowl-
edge of South African Mollusca. Part Il. Gas-
tropoda: Prosobranchiata: Rhachiglossa. Annals
of the South African Museum, 45: 1-256.

BELLARDI, L., 1882, / Molluschi dei Terreni Terziari
del Piemonte e della Liguria, Vol. 3. Stamperia
Reale, Torino, 253 pp.

BERNARD, P. A., 1984, Coquillages du Gabon. Li-
breville, Gabon, 140 pp., 75 pls.

BERNSTEIN, А. S., 1970, Notes on the ecology of
Drupa morum 1798 [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.

BEU, 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, А new Coralliophil-
idae from the Bay of Vigo. La Conchiglia, 15:
4-5.

BOONE, С. E., 1984, Search and seizure. More on
Thais haemastoma. Texas Conchologist, 21:1-6.

BORN, 1. VON, 1778, Index rerum naturalium
Musei Caesarei Vindobonensis. Officina Krausi-
ana, Vienna, 458 pp.

BRADLEY, J. C. & K. 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: Е. 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 pls., 4 maps.

BRIGHT, D. A. & D. V. ELLIS, 1990, А 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.

BRUGUIERE, J. G., 1789, Encyclopédie method-
ique. Histoire naturelle des vers. Paris, 1, i-xvii
+ 344 pp.

BRUGUIERE, 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: 611—
640.

BRYAN, С. W., Р. E. GIBBS, G. В. BURT. 4 L. С.
HUMMERSTONE, 1987, The effects of tributyltin
(TBT) accumulation on adult dogwhelks, Nucella
lapillus: long-term field and laboratory experi-
ments. Journal of the Marine 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, А. 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, М. 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: а synthesis. Malacologia, 20: 403-
422.

PHYLOGENY ОЕ RAPANINAE 251

CARRIKER, М. R., Р. PERSON, В. 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. C. & J. CANCINO, 1976, Spawning
behaviour and egg capsules of Concholepas
concholepas (Mollusca: Gastropoda: Muricidae).
Marine Biology, 37: 255-263.

CASTILLA, J. C. & 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 Il. Subfamily Thaidinae. The Veliger,
11: 293-315.

CERNOHORSKY, W. O., 1972, Marine shells of the
Pacific, Vol. И. Pacific Publications, Sydney, 411
рр., 68 pls.

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. О., 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 11. Records of
the Auckland Institute and Museum, 20: 185-202.

CHENU, J. C., 1859, Manuel de conchyliologie et
de paléontologie conchyliologique. 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—
Ya!

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 della 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—
1572

CONNELL, J. H., 1970, А predator-prey system in
the marine intertidal region. 1. Balanus glandula
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: 305-311.

COOKE, А. H., 1895, Molluscs. In: $. Е. HARMER &
А. Е. 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).

СООКЕ, А. Н., 1915, The geographical distribution
of Purpura lapillus (L.) Part 1: in palaearctic wa-
ters. Proceedings of the Malacological Society,
11: 192-209.

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, А. H., 1919, The radula in Thais, Огира,
Morula, Concholepas, Cronia, lopas, and the al-
lied genera. Proceedings of the Malacological
Society of London, 13: 90-110.

COOMANS, Н. E., 1962, The marine mollusk fauna
of the Virginian area as a basis for defining zoo-
geographical provinces. Beaufortia, 9(98):
83-104.

COSSMANN, А. Е. M., 1903, Essais de paléocon-
chologie comparée, Vol. 5. Paris, 215 pp.

COWELL, Е. В. & 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, pls. 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. М., 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., 19706, Egg capsules of some
prosobranchs from the Pacific coast of Panama.
The Veliger, 13: 37—43.

D’ASARO, С. М., 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:
Muricidae. Ophelia, 35: 1-101.

DAVIS, G. M., 1979, The origin and evolution of the
gastropod family Pomatiopsidae, with emphasis
on the Mekong River Triculinae. Monograph of
the Academy of Natural Sciences of 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: Muricacea). Biologi-
cal Journal of the Linnean Society of London, 40:
245-261.

DEMOND, J., 1957, Micronesian reef-associated
gastropods. Pacific Science, 11: 275-341.

DESHAYES, G. P., 1830, Encyclopédie méthoa-
ique. Historie naturelle des vers, Vol. 2, Part 1,
pp. i-vii, 1-256.

DESHAYES, С. 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, С. P., 1844, Histoire naturelle des an-
imaux sans vertèbres, . .. 2nd. ed., Vol. 10. Paris,
638 pp + Index [1 p.]

DISALVO, L. H., 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, 111:
153-312.

DRIVAS, J. & М. JAY, 1987, Coquillages de La Re-
union et de l'Ile Maurice. Delachaux and Niestlé,
Neuchätel, Switzerland, 159 pp., 58 pls.

DUBOIS, R., J. C. CASTILLA & R. CACCIOLATTO,
1980, Sublittoral observations of behaviour in the
Chilean ‘loco’ Concholepas concholepas (Mol-
lusca: Gastropoda: Muricidae). The Veliger, 23:
83-92.

DUCLOS, M., 1832, Description de quelques espè-

ces de pourpres servant de type а six sections
etablies dans ce genre. Annales des Sciences
Naturelles, 26: 1-11.

DUMERIL, А. М. C., 1806, Zoologie analytique, ...
Paris, xxxii + 344 pp.

DUNKER, G., 1853, Index molluscorum, . . . Cas-
sellis Cattorum, Fischer, 74 pp.

DUNKER, W., 1861, Mollusca Japonica descripta
et tabulis tribus iconum. Schweizerbart, Stuttgar-
tiae, 36 рр.

EISENBERG, J. M., 1981, A collector's guide to
seashells of the world. McGraw-Hill, New York,
237 [+ 2] pp., 158 pls.

EMERSON, W. K. & W. О. CERNOHORSKY, 1973,
The genus Drupa in the Indo-Pacific. Indo-Pacific
Mollusca, 3: 1-40.

EMERSON, W. K. & A. D’ATTILIO, 1981, Remarks
on Muricodrupa lredale, 1918 (Muricidae: 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, Harvard University, Cam-
bridge, Massachusetts, i-ix + 198 pp.

FAIRWEATHER, P. G., 1988, Movements of inter-
tidal whelks (Morula marginalba and Thais orbita)
in relation to availability of prey and shelter. Ma-
rine Biology, 100: 63-68.

FARRIS, J. S., 1979, 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.

FISCHER VON WALDHEIM, 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 + 330 pp.

FISCHER, P., 1884, Manuel de conchyliologie et de
paléontologie conchyliologique ou histoire na-
turelle des mollusques vivants et fossiles. Part 7,
рр. 609-688.

FRETTER, V., 1941, 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. GRAHAM, 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.

РОЛОКА, 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 the Hiroshima Uni-
versity, Ser. В, Div. 1 (Zoology), 31: 235-287.

FUJIOKA, Y., 19856, Seasonal aberrant radular
formation in Thais bronni (Dunker) and T. clavig-
era (Kuster) (Gastropoda: Muricidae). Journal of
Experimental Marine Biology and Ecology, 90:
43-54.

GALLARDO, C. S., 1973, Desarrollo intracapsular
de Concholepas concholepas (Bruguiere) (Gas-
tropoda Muricidae). Museo Nacional de Historia
Natural, Publicaciön Ocasional, 16: 3-16.

GALLARDO, С. S., 1979, El ciclo vital del muri-
cidae Concholepas concholepas у consid-
eraciones sobre sus primeras fases de vida en el
bentos. Biologia Pesquera Chile, 12: 79-89.

GALLARDO, C. S., 1980, Adaptaciones reproduc-
tivas en gastropodos muricaceos de Chile;
conocimiento actual y perspectivas. Investiga-
ciones Marinas, 8: 115-128.

GALLARDO, С. S. & Е. E. 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 Muricidae. /nvertebrate 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, Р. Е. & G. W. BRYAN, 1986, Reproductive
failure in populations of the dogwhelk, Nucella
lapillus, caused by imposex induced by tributyltin
(ТВТ) contamination from antifouling paints.
Journal of the Marine Biological Association of
the United Kingdom, 66: 767-777.

GIBBS, P. E., С. W. BRYAN, Р. L. PASCOE & С. В.
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. EISAWAY, 1967, The egg
masses and development of five rachiglossan
prosobranchs. Publications of the Marine Biolog-
ical Station Al-Ghardaga (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 1855.

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 B (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, В. 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.

НАВЕ, T., 1964, Shells of the western Pacific т
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. 8 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, B., 1888, Die Morphologie der Proso-
branchier.MorphologischesJahrbuch, 14:54—169.

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-
оту of four primitive muricacean 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., 1905, Studies on Australian Mollusca,

254 KOOL

Part IX. Proceedings of the Linnean Society of
New South Wales, 4: 520—546.

HEDLEY, C., 1918. А check-list of the marine fauna
of New South Wales. Part 1. 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, А. 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. B., 1844, Descriptions of new species of
Scalaria and Murex, from the collection of Sir Ed-
ward Belcher, C.B. Proceedings of the Zoological
Society of London, 11: 127-129.

HOMBRON, J. B. & C. H. JAQUINOT, 1852, Atlas
for Dumont d’Urville: Voyage au Pole Sud. Zool-
ogie. Paris, 143 pls.

HOUBRICK, R. S., 1978, The family Cerithiidae in
the Indo-Pacific. Part 1: The genera Rhinoclavis,
Pseudovertagus, Longicerithium and Clavocer-
ithium. Monographs of Marine Mollusca, 1: 130

pp.

HOUSTON, В. 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, БВ. C., 1971, Shell variation of Nucella
lapillus in relation to environmental and genetic
factors. Norwegian Journal of Zoology, 19: 145—
148.

HUANG, C. 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 various
articles . . . consisting of . . . subjects т...
Conchology, . . . Part 1, т 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, 10: 217-228.

IREDALE, Т., 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. Рр. 232-261, т: The
Middleton and Elizabeth Reefs, South Pacific
Ocean. The Australian Zoologist, 8: 199-268.

IREDALE, T. & D. Е. MCMICHAEL, 1962, A refer-
ence list of the marine Mollusca of New South
Wales. The Australian Museum, Sydney, Мет-
oir, Il: 109 pp.

JABLONSKI, 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, Revision methodique 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. Memoires 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, Е. A., 1971, The littoral molluscs of Fanning
Island. Pacific Science, 25: 260-281.

KAY, Е. 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. М., 1971b, Sea shells of tropical west
America. Stanford, 1064 pp.

KENSLEY, B., 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, Т., 1957, Local races and clines in the
marine gastropod Thais lamellosa Gmelin. A
population study. Calliostoma, Seattle, 75 pp.

KITCHING, J. A., L. MUNTZ & Е. 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: 113-126.

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 раша
(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 anatomy of
Plicopurpura patula (Prosobranchia: Muricoidea:
Thaidinae), new combination, with emphasis оп
the reproductive system. Malacologia, 29: 373—
382.

KOOL, S. P., 1989, Phylogenetic analysis of the
subfamily Thaidinae. 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, $. P. & К. J. BOSS, 1992, Nucella Röding,
1798 (Gastropoda: Muricidae): type species. The
Nautilus, 106: 21-23.

KOROBKOV, I. A., 1955, Spravochnik i method-
icheskoe rukovodstvo po Tretichnym molli-
uskam. Leningrad, 795 pp.

KOZLOFF, Е. N., 1987, Marine invertebrates of the
Pacific Northwest. University of Washington, Se-
attle & London, x + 511 pp.

KRAUSS, F., 1848, Südafrikanischen Mollusken.
Stuttgart, [2] + 140 pp., 6 pls.

KURODA, T., 1930, New Japanese shells. (2). Ve-
nus, 2: 1-2.

KURODA, Т., & Т. НАВЕ, 1971, In: T. KURODA, Т.
НАВЕ & К. Oyama, eds, The sea shells of Sagami
Вау. Maruzen Company, Ltd., Tokyo, 489 pp.
(English part) + Index.

LAMARCK, J. В. Р. А. DE M. DE, 1799, Prodrome
d'une nouvelle classification des coquilles. Mé-
moires Société Histoire Naturelle, Paris, 1: 63—
91.

LAMARCK, J. В. Р. А. DE M. DE, 1801, Système
des animaux sans vertèbres, ... Paris, viii + 432

PP-

LAMARCK, J. В. Р. А. DE М. DE, 1816, Tableau
encyclopédique et méthodique des trois règnes
de la nature, Pt. 23. mollusques et polypes
divers. Paris, 16 pp., pls. 391—488.

LAMARCK, J. В. P. А. DE M. DE, 1822, Histoire
naturelle des animaux sans vertèbres, ... Vol. 6,
part 2. Paris, 711 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: 123-127.

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. J

LEACH, W. E., 1852, A synopsis of the Mollusca of
Great Britain, John van Voorst, London, 376 pp.,
13 pls.

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., 1945, 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. B., 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 Universitat zu Rostock, Part 3:
101-165.

LINNAEUS, C., 1758, Systema naturae, 10th ed.
Stockholm, 824 pp.

LINNAEUS, C., 1767, Vermes Testacea. In: Sys-
tema naturae, 12th ed., 1: 1106-1269.

LINNAEUS, C., 1771, Mantissa plantarum. . .
—Regni animalis . . . —Appendix. Stockholm,
Нм + 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, 111: 71-79.

LOCARD, A., 1886, Prodrome de malacologie
Frangaise. Catalogue general des mollusques
vivants de France—Mollusques marins Paris,
778 pp.

MAES, V. O., 1966, Sexual dimorphism in the rad-
ula ofthe 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, pls.
66-121.

MARTYN, T., 1784, The universal conchologist,
Vols. 1, 2. London.

MENKE, C. 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 (Duclos) (Gastropoda: Thai-
sidae) in Hawaii. The Biology of Mollusks. Uni-
versity of Hawaii, Hawaii Institute of Marine Biol-
ogy, Technical Report, 18: 17.

MORCH, O. A. L., 1852, Catalogus conchyliorum.
Ludovici Kleini, Hafniae, 74 [+ 2] pp.

MORCH, О. A. L., 1860, Вейгаде zur Mollusken-
fauna Central-Amerika’s. Malakozoologische
Blatter, 7: 66-106.

MONTFORT, P. D. DE, 1810, Conchyliologie sys-
tématique, Il. Paris, 676 pp.

256 KOOL

MOORE, Н. B., 1936, The biology of Purpura lapil-
lus. |. Shell variation in relation to environment.
Journal of the Marine Biological Association of
the United Kingdom, 21: 61-89.

MOORE, Н. B., 1938, The biology of Nucella lapil-
lus. Ill. 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 europdischen 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, Е, 1875, Die bisher veröffentlichten Fam-
Шеп- 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 pls.

PALMER, A. R., 1979, Fish predation and the
evolution of gastropod shell form: experimental
and geographic evidence. Evolution, 33: 697-
7.13:

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, C. 1982. Morphological characters
and homology. Pp. 21-74; in: К. A. Joysey & A.
E. Fripay, eds., Problems of phylogenetic recon-
struction. Academic Press, New York.

PCHELINTSEV, V. Е. & I. A. KOROBKOV. 1960.
Molliuski-Briukhonogie. In: Y. A. OrLov, ed., Os-
novy paleontologii. 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 pls., 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-
ractéres. 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, В. A., 1849, Centuria quarta Testaceo-
rum novorum. Zeitschrift für Malakozoologie, 6:
27-32.

PHILLIPS, В. 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 Marine Biology and Ecology, 11: 27-
69.

PONDER, W. F., 1973, The origin and evolution of
the Neogastropoda. Malacologia, 12: 295-338.
PONDER, W. Е. & A. WAREN, 1988, Classification
of the Caenogastropoda and Heterostropha—a
list of the family-group names and higher taxa;
Appendix. In: W. Е. PONDER, D. J. EERNISSE & J.
H. WATERHOUSE, eds., Prosobranch phylogeny
Malacological Review, Supplement, 4: 288-326.

POWELL, A. W. B., 1961, Shells of New Zealand.
Whitcombe & Tombs Ltd., Auckland, 203 pp.

POWELL, А. W. B., 1964, The family Turridae in the
Indo-Pacific. Indo-Pacific Mollusca, 1: 227-346.

POWELL, А. W. B., 1979, Shells of New Zealand.
Collins, London, 500 pp.

POWYS, W. L. & С. В. 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. C. 4 J. P. GAIMARD, 1833, Voyage de
découvertes de L’Astrolabe, Part 2, 686 pp.

RADWIN, С. Е. & A. D’ATTILIO, 1971, Muricacean
supraspecific taxonomy based on the shell and
the radula. Western Society of Malacologists,
Echo, 4: 55-67.

RADWIN, С. E. 4 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, С. E. & A. D'ATTILIO, 1976, Мигех
shells of the world. Stanford University Press,
Stanford, 284 pp.

RAFINESQUE, C. S., 1815, Analyses de la nature
ou tableau du univers et des corps organises.
Barravecchia, Palermo, р. 5-6, 136-149, 218—
223.

RAJALAKSHMI BHANU, R. C., K. SHYAMA-
SUNDARI 8 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 8 K. HANUMANTHA RAO, 1981a,
Histological and histochemical studies on the sal-
ivary glands of Thais bufo (Lamarck). Monitore
Zoologico Italiano, 15: 239—247.

PHYLOGENY OF RAPANINAE 257

RAJALAKSHMI BHANU, В. С., К. SHYAMA-
SUNDARI & К. HANUMANTHA ВАО, 1981Ь,
Studies on the alimentary canal of Thais bufo
(Lamarck): histology and histochemistry of the
foregut and midgut glands. Acta Histochemica et
Cytochemica, 14: 516-528.

REEVE, L. A., 1846, Conchologia Iconica, Volume
3, Purpura, Ricinula, Monoceros, and Conchole-
pas; Buccinum.

REHDER, H. A., 1962, The status of Nucella. The
Nautilus, 75: 109-111.

REHDER, H. A., 1980, The marine mollusks of
Easter Island (Isla de Pascua) and Sala y Go-
mez. Smithsonian Contributions to Zoology, 289:
167 pp.

REHDER, H. A. & H. 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.

RIGHI, G., 1964, Söbre o estömago de Thais hae-
mastoma. Anais Academia Brasileira de Cien-
cias, 36: 189-191.

RIOS, Е. C., 1970, Coastal Brazilian seashells. Mu-
seu Oceanogräfico de Rio Grande, Rio Grande,
Brazil, 255 pp., 4 maps, 60 pls.

RODING, 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 189: 41 pp.

ROVERETO, G., 1899, Prime ricerche sinonimiche
sui generi dei gasteropodi. Atti della Societa Li-
gustica di Scienze Naturali e Geografiche, 10:
101-110.

SABELLI, B. & S. TOMMASINI, 1979, Osservazioni
sulla radula di alcuni Muricacea delle Galapagos.
Bollettino Malacologico, 15: 19-28.

SALVAT, B. & C. RIVES, 1975, Coquillages de
Polynesie. 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.

SCHAUFUSS, |. W., 1869, Molluscorum systema
et catalogus. System und Aufzählung sámmtli-
cher Conchylien der Sammlung von Fr. Paetel.
Dresden, [4] + XIV + 1-119 pp.

SCHUMACHER, С. F., 1817, Essay d'un nouveau
systeme des habitations des vers testacés.
Copenhagen, 287 pp.

SETTEPASSI, F., 1971, Atlante malacologico mol-
luschi marini viventi nel Mediterraneo, Vol. 2.
Roma, unpaginated.

SHUTO, 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 Morphologica, 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, Е. 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, С. B., 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. B., 1839, A conchological manual.
Odell, London. i-v + [2] + 130 pp., 24 pls.

SOWERBY, G. B., 1841, Conchological illustra-
tions, Part Il. 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 оп
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 (Duclos) (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 pls.

STEPHENSON, L. W., 1941, The larger inverte-
brate fossils of the Navarro Group of Texas. Uni-
versity of Texas Publication, 4101: 641 pp., 95
pls.

STEWART, R. B., 1927, Gabb’s California fossil
type gastropods. Proceedings of the Academy of
Natural Sciences, Philadelphia, 78: 287—447,
pls. 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 pls.

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 pls.

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. А., 1840, Treatise on malacology
or shells and shell-fish. London, 419 pp.

TAKI, |., 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. & М. Е. ЗОНЕ, 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, А 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. 8 А. J. KOHN, 1985, Trophic
roles of the tropical limpet-like predatory gastro-

pod, Drupa. American Zoologist, 25: 88.

THORNLEY, G., 1952, A new Thais found on a log
at Port Stephens. Proceedings of the Royal Zoo-
logical Society of New South Wales, 1951—1952:
44—45.

TIRMIZI, М. М. & 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 оп 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, С. J., 1975, Marine faunal dominance
and molluscan shell form. Evolution, 28: 656—
664.

\УЕВМЕМ., С. J., 1979, The architectural geography
of some gastropods. Pp. 428—433 in: J. GRAY &
A. J. Boucor, eds., Historical biogeography, plate
tectonics, and the changing environment. Ore-
gon State University Press, Oregon.

VERMEIJ, С. J. 1982. Phenotypic evolution in a
poorly dispersing snail after arrival of a predator.
Nature, 229: 349-350.

VERMEIJ, С. J. & J. D. CURREY, 1980, Geograph-
ical variation in the strength of thaidid snail shells.
Biological Bulletin, 158: 383-389.

VERMEIJ, С. J. & Е. 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 тапсте!а Lin-
naeus. The Veliger, 12: 368-370.

VOKES, E. H., 1972, Notes on the fauna of the
Chipola Formation VII. On the occurrence of the
genus Concholepas (Gastropoda: Thaididae),
with the description of a new species. Tulane
Studies in Geology and Paleontology, 10: 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 pls.

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:
О. H. SCHINDEWOLF, ed., Handbuch der Paläozo-
ologie, Vol. 6, part 5 Gebrüder Borntraeger, Ber-
lin.

WILBUR, K. M. & G. OWEN, 1964, Growth. Pp.
211-242, in: K. M. WiLBUR & С. М. 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; ог a cat-
alogue of shells, British and foreign. London, viii
+ 188 pp., 8 pls., 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 Sinica, 4: 95-106.

WU, S. K., 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 ОЕ 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, $. K., 1985, The genus Acanthina (Gas- Revised Ms. accepted 4 January 1993

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 Вит group is described, clarifying the status of the
genus-level taxa comprising it. Bittium reticulatum, the type species of Bittium Gray, is described
in depth, thereby establishing criteria for comparisons with other taxa of Bittiinae. The type
species of Stylidium Ва! 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, /ttibittium, 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 Cerithium Bruguière. These are Cerithium zebrum
Kiener, 1841, and Cerithium boeticum Pease, 1861. The subfamily Bittiinae Cossmann, 1906, is
thought to comprise nine genera (four of which were not included in phylogenetic analyses) :
Bittium Gray, 1847; Bittiolum Cossmann, 1906; Ittibittium gen. n., Stylidium Dall, 1907; Lirobit-
tium Bartsch, 1911; Cacozeliana Strand, 1928; Argyropeza Melvill & Standen, 1901; Varicopeza
Gründel, 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

261

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 Вгидшеге, 1789, 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, Сгапае! (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

262

HOUBRICK

TABLE 1. Bittium-group депега and species used for anatomical studies (asterisk indicates

type species).

Genus Species

Bittium *reticulatum (DaCosta, 1778)
Bittium impendens (Hedley, 1899)
Bittiolum varium (Pfeiffer, 1840)
Bittiolum alternatum (Say, 1822)
Ittibittium parcum (Gould, 1861)
Lirobittium subplanatum Bartsch, 1911
Lirobittium attenuatum (Carpenter, 1864)
Stylidium *eschrichtii (Middendorf, 1849)
Cacozeliana *granaria (Kiener, 1842)

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,
1976, 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 (1977) discussed the status of Bit-
tium Gray, 1847, 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 Gründel’s (1976) work and my paper
on Bittium (Houbrick, 1977), studies on a
number of Bittium-like genera and other
small-shelled cerithioidean taxa have been

Geographic Region
Sao Miguel, Azores
Honolulu, Hawaii
Ft. Pierce, Florida
Provincetown, Massachusetts
Honolulu, Hawaii
Palos Verdes, California
Catalina Id., California
Carmel, California
Albany, Western Australia

published: Dahlakia (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 artificiality 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 ОЕ BITTIINAE 263

cluded. 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, | decided to look at representative taxa
of genera assigned to the Bittium-group сот-
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,” $.1., 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 pallial ovi-
duct, characters distinctive of members of the
Bittium-group. Therefore, both species were
excluded from the Bittium-group and as-
signed to Cerithium Bruguiere. Due to the cur-
rent alpha-level taxonomic disarray of the Bit-
tium-group, | have attempted to present a
comprehensive, annotated synonymy and
have illustrated the shells of the species stud-
ied in this review. | 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 а Wild M-5 dis-
secting microscope. Methylene blue was
used to enhance anatomical features during
dissection. Sections were made at 5 um and
stained with Hematoxolin and Eosin. Photo-
micrography was done using a Zeiss Photo-
microscope Ill.

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 ofthe genera Argyropeza and
Varicopeza is only superficially understood.
Anatomical knowledge about Zebittium Fin-
lay, 1927, and Cassiella remains unknown,
because | 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 (copyright Kevin C. Nixon, 1988,
1991, 1992) were used to analyse data and
construct trees.

Phylogenetic analysis of six genus-group
taxa of the Bittiinae (Bittium, lttibittium, Bitti-
olum, Lirobittium, Stylidium, and Cacozeli-
ana) was undertaken using 21 morphological
characters comprising 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 роопу 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 Bruguiere, family Cer-
ithiidae Férussac, 1819, was selected as the

264 HOUBRICK

TABLE 2. Comparison of dentition of radular teeth among депега (С = central ог main cusp; numbers
signify no. of denticles).

Taxon Rachidian Lateral Inner Marginal Outer Marginal
Bittium 2-3+C+2-3 1+C+3-6 3-4+C+4 3—4+С+0
Bittiolum 3+C+3 2+C+3-4 3-4+C+2-3 6+C+0
Ittibittium 2+C+2 1+C+3-4 2+C+3 5+C+0
Lirobittium 6+C+6 6+C+15-17 15-19+C+5-6 15-19+C+0
Stylidium 2+C+2 1+C+3-4 4-5+C+3 4+C+0
Cacozeliana 2+C+2 1+C+3-4 5-6+C+3-4 4+C+0
Argyropeza 2 3162.83 176756 5-67 Я 5-6+C+0
Varicopeza 3-44+C+3-4 1--C+5—6 3-4+C+3 3+C+0

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 smail and often are confused
with “Bittium” species.

Among small-shelled cerithioideans, [Пора
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 1$
known anatomically (Ponder, 1991), and
Healy (1986) has shown that the parasperma-
tozoa of Diala are unique and highly derived

among cerithioideans. Ponder’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—0 =
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 lttibittium, where spiral sculpture domi-
nates and axial ribs are either lacking or
poorly developed.

2. Anal canal—O = 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 Bittinae. Cerithium is the outgroup.

Character
Taxon 238 4 ога & Tl) sh We de Ye aby м MEA я
Outgroup oO OO оо ооо о оо © © © @ WM @ @ @ ©
Bittium VO 1707000020 1 ? 4 1 1 1 1 OO
Ittibittium оО т ar 05% 1 OOO 1 1
Stylidium od 1 ft 2-0 @ J ad т (ее 1 1 1 1 1 OAI
Gacozeliana 1 0 1 0 2 2 2 0 0 O 1 oO @ ©. © 1 1 1 д 0
Bittiolum 1 +t @ TY @ tT т 37 @ 1 1 1 0 1 1 1 OO
Lirobittium ele a О OA 1 2a el 1 CRE 1 1 D 4

3. Varices—0 = present; 1 = absent. Va-
rices, thickened, former growth lines, are a
common feature of most cerithiids and occur
among members of Bittinae with the excep-
tion of Lirobittium and Stylidium.

4. Anterior canal—0 = 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—0 = 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—0 = 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 /ttibittium and Bittiolum, both small
shelled genera). The spirally fringed condition
seen in Cacozeliana departs from the norm
and is probably derived.

7. Snout shape—0 = 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—0 = 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—0 = 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, роз-
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—0 = absent;
1 = present. Although this structure is absent
in the outgroup, it does occur among a few
other cerithioidean groups (Litiopidae [Alaba,
[Пора], Cerithidae [Сота]; 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).

11. Epipodial skit—0 = 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 Bittinae 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 U mm 1 spiral planktonic 0.1 mm
Lirobittium

subplanatum 10 mm 2 spirals direct 0.5 mm
Stylidium

eschrichtii 17.5 тт smooth direct 0.2 mm
Cacozeliana

granaria 24 mm smooth planktonic 0.1 mm
Argyropeza

divina 7.6 mm 2 spirals planktonic ?
Уапсорега

уапсорега 10 mm 1 spiral planktonic ?

12. Ovipositor—0 = present; 1 = absent.
This gland, although common among cerithio-
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 /ttibittium and Lirobittium.

13. Osphradial morphology—0 = bipecti-
nate; 1 = monopectinate; 2 = vermiform.
This character varies greatly among Bittiinae
genera. Although the osphradium in Cerith-
ит species is bipectinate, it is vermiform
among most other cerithioidean families,
such as the estuarine Potamididae and fresh-
water families Thiaridae and Pachychilidae
(Houbrick, 1988, 1991).

14. Osphradial length—0 = 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—0 =
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—O = 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—O =
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 /ttibittium and in all
other known cerithiids, it occurs in the medial
lamina (Houbrick, 1988).

18. Ciliated ridge tract—0 = absent; 1 =
present. This structure, one of the synapo-
morphies defining Bittiinae, is lacking in /ftibit-
tium members and in most other cerithiids.

19. Seminal receptacle with grape-like mor-
phology—0 = 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—0 = 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 cerithioidean 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—0 = planktonic;
1 = lecithotrophic (demersal/direct). Most
members of the outgroup have a planktonic

GENERIC REVIEW OF BITTIINAE

Cacozeliana Lirobittium Stylidium
20 2
6 2
7 2 8 1
9
outgroup

267
Bittium Ittibittium Bittiolum
5 1
20 1
11 3
14 2
10 1

FIG. 1. Cladogram showing relationships among six genera of Bittiinae, using Cerithium as the outgroup (Е
= 41; CI = 70; RI = 53; trees two. Numbers to left of black bars indicate characters: those to right of bars
represent character states. Only characters with a Cl 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), /ttibittium
and Bittiolum are sister groups of Bittium,
while in the second tree, Bittium and Bittiolum
are sister groups of /ftibittium. 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 pallial 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 1 2 3 4
Steps 3 1 2 1
СЛ. 33 100 50 100 66
Character 12 13 14 15 16
Steps 2 3 2 1
Gill 50 66 100 100 33

6 74 8 9 10 11

3 2 1 1 1 3
66 100 100 100 100 100
17 18 19 20 21

2 2 1 2 2

50 50 100 100 50

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]) п 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]). /t-
tibittium and Bittiolum, the sister taxa to Bit-
tum, 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-
tinae 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, /ttibittium 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, | 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. n.; 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-like cerithioideans 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

Bittinae 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 pallial 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 & Warén, 1988), except Grün-
del (1976), who assigned the group to the Ju-
rassic family Procerithiidae Cossmann, 1906
(erroneously cited by Cossmann as 1905). 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 Gründel, 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, /n-
fracerithium Gründel, 1974, and Rhabdocol-
pus Cossmann, 1906, were treated as sub-
genera of Procerithium. Gründel (1976) also
included Argyropeza Melvill & Standen, 1901,
Varicopeza Gründel, 1976, and the extinct
genus Cryptaulax Gründel, 1976, with sub-
genera Pseudocerithium Cossmann, 1884,
and Xystrella Cossmann, 1906, 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-
tinae, 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-
logeny (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, 1901. 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-like 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 |
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, $.5. 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, 1924 (type species: Bittium
hiloense Pilsbry & Vanatta, 1908). The type
species of this genus is a rissoid of the genus
Isseliella 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 & Marwick, 1937 (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 batillariid (Houbrick, pers. obser.). Their
drawing ofthe type species (Finlay & Marwick,
1937: pl. 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
Cerithium species, judging from its illustration
(Weisbord, 1962: pl. 15, figs. 5—6).

4. Cerithidium Monterosato, 1884 (type
species: Cerithium submamillatum Rayneval
& Ponzi, in Rayneval et al., 1854). Cerithidium
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, Cerithium 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
Cerithium submamillatum as the type species
of Cerithidium by Cossmann (1906) should
prevail over that of Turritella pusilla by Wenz
(1940). | 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 pallial ovi-
duct was closed, which is very different from
the open systems known in all other members
of Bittiinae. A closed pallial 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 а junior synonym of Cerithium proteum
Jousseaume, 1930 (Houbrick, 1978), and |
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
batillariid (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. fritschi depicted by Martin (1914: pl.
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 pallial 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
Cl 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. /ttibit-
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. n., 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). | 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 А. 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, 1941. 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 pallial 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 (1985: 479) has shown that the osphradial
cells bear paddle cilia. The osphradial classi-
fication of Bittiinae species falls under Hasz-
ргипаг$ (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

HOUBRICK

FL Se

BITTIUM

35 «<=

ITTIBITTIUM

BITTIOLUM

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 BITTIINAE 273

better understood and a phylogenetic analy-
sis can be accomplished.

BITTIUM GRAY, 1847

Bittium 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 Payraudeau, 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, paucispiral 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 М. spenceri, consecutively, under Bit-
tium. Besides Bittium reticulatum, the other
species listed by Leach represent two gen-

era, Triphora Blainville, 1828, 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 /nobittium, also is considered herein to be
conspecific with Bittium reticulatum. Wenz
(1940: 757) regarded /nobittium as a syn-
onym of Lirobittium, but | 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, /nobittium is regarded as a subjective
junior synonym.

Discussion

The genus Bittiumis 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 small
papillae along the edge of the epipodial skirt
and opercular lobe (Fig. 2). The ovipositor in
females 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 pallial oviduct (Fig. 6B, 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-
jum 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 Fret-
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, ps8; fig. 13:

Murex reticulatus (Da Costa). Montagu, 1803:
272:

Cerithium latreillei Payraudeau, 1826: 143.

Cerithium lacteum Philippi, 1836: 195.

Cerithium reticulatum, Risso, 1826: 157; С. В.
Sowerby, 1855: pl. 15, fig. 8; Jeffreys,
1867: 258; 1869: pl. 80, fig. 4; 1885: 57.

Bittium reticulatum, Watson, 1886: 540; Buc-
quoy et al., 1884: 212-215, pl. 25, figs.
3-9; Tryon, 1887: 150-151, pl. 29, figs.
78-83; Dautzenberg, 1889: 40-41.

Description

Shell (Fig. 3A-H): Shell elongate, reaching
15 mm in length, comprising 9-10 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-Н, В. reticulatum; I-N, В. impendens. A-C, ЗЕМ micrographs
of В. reticulatum from Säo Miguel, Azores (USNM 878030), 6 mm length; D, Е, В. reticulatum from Tunisia
(USNM 754051), 11 mm length; F, SEM micrograph of operculum of B. reticulatum, bar = 0.5 mm: H, SEM
micrograph of immature shell of В. reticulatum, bar = 0.5 тт; I-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 В. impendens, bar = 150 шт.

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 pallial 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. 4B, epg) forming broad
epipodial skirt (Fig. 4B, 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. 4B, C, ep); epipo-
dial skirt forming long opercular lobe (Fig. 4B,
C, opl). 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. 4B, C, cg) 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. 4B, 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. 4B,
C, mp) slender, darkly pigmented, each with

white spot. Mantle edge thickened at inhalant
(Fig. 4C, inh) and exhalant siphons.

Pallial Cavity: Pallial 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. Pallial 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 pallial 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-ventrally 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. 5B) 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

FIG. 4. Anatomical representations of Bittium reticulatum. À, 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. аез = anterior esophagus; amg = anterior mucus gland; beg = subesophageal gland; bg = buccal
ganglion; bm = buccal mass; с = ciliated strip; cg = ciliated gutter; eg = esophageal gland; ep = epipodial
papilla; epg = epipodial groove; eps = epipodial skirt; ex = exhalant siphon; inh = inhalant siphon; | = lip;
lcg = left cerebral ganglion; Ipg = left pleural ganglion; 159 = left salivary gland; m = mouth; тр = mantle
papilla; ор = operculum; ор! = opercular lobe; ovp = ovipositor; pes = posterior esophagus; rcg = right
cerebral ganglion; rpg = right pleural ganglion; rsg = right salivary gland; $ = sole; seg = supraesophageal
ganglion; sn = snout; t = tentacle.

278 HOUBRICK

FIG. 5. Scanning electron micrographs of radula of Bittium reticulatum from Säo Miguel, Azores (USNM
878030). A, half row with marginal teeth folded back, bar = 19 рт; В, rachidian and lateral teeth, bar = 15

шт; С, detail of rachidian teeth, bar = 4 um.

central cusp, 3—4 inner denticles, 4 outer
denticles; outer marginal tooth same, but
lacking outer denticles. Salivary glands (Fig.
40, rsg, 159) 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,
Icg). 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, lpg). Supraesophageal connective mod-
erately long, about twice length of right pleural
ganglion; dialyneury between left pallial 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 pallial
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. 6B, 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 pallial oviduct. Female pallial oviduct
(Fig. 6B, 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.
6B, C, ovg). Posterior end of pallial oviduct
closed. Medial, free lamina with wide anterior
ciliated sperm gutter (Fig. 6B, C, sg) along its
edge leading to two, well-separated, pocket-
like openings. First opening (Fig. 6B, 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, B sr; 8C, D, sr)
containing oriented euspermatozoa with
heads embedded in receptacle walls. Lateral
lamina attached to pallial wall, having anterior
ciliated tract comprising many parallel elon-
gate, fine ciliated folds (Fig. 6B, C, ctr; 7A, B,
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. 6B, 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. 6B, C, ag; 7C, D, ag); anterior half
dirty white, comprising capsule gland (Fig.
6BNCy cgi7/A; В! сд):

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 ит diameter)
each within thin, transparent hyaline capsule
(110 рт diameter). Entire spawn mass con-
tains about 800 eggs. Free swimming bilobed
planktotrophic veliger larval stage present.
Larval shell ranging from 170-330 um, de-

280 HOUBRICK

FIG. 6. Representation of pallial gonoducts of Bittium reticulatum. À, male pallial gonoduct, showing section
through mid-duct beneath, represented by dotted line; B, pallial oviduct showing three cross sections of duct
represented by dotted arrows and sections to right; C, reconstruction of pallial oviduct showing configuration
of ducts and glands (anterior to right). ag = albumen gland; ant = anterior; cg = capsule gland; cod =
coelomic oviduct; ctr = ciliated ridge tract; gd = gonaductal groove; Il = 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 pallial oviduct; sb = spermatophore bursa; sg
= sperm gutter; sp = sperm bursa; sr = seminal receptacle; so = spermatophore-forming organ.

pending upon age. Larval shell with rounded, Discussion
nearly smooth whorls having thin spiral thread
forming weak keel and with deep sinusigeral The status of the many specific and sub-

notch (Thorson, 1946: 192, fig. 109). specific names comprising the Bittium reticu-

GENERIC REVIEW ОЕ BITTIINAE 281

FIG. 7. Successive sections, anterior to posterior, through pallial oviduct of Bittium reticulatum. À, anterior
of pallial oviduct showing relationship of mantle cavity organs to oviduct, bar = 0.25 тт; В, mid-section
showing ciliated ridge tract and opening to sperm bursa, Баг = 0.25 mm; С, section through enlarged sperm
bursa in posterior pallial oviduct, bar = 0.25 mm; D, section through closed posterior of pallial oviduct, bar
= 0.25 mm. ag = albumin gland; cg = capsule gland; ct = ctenidium; ctr = ciliated ridge tract; eg =
esophageal gland; hg = hypobranchial gland; os = osphradium; osp = opening to sperm bursa; ovg =
oviductal groove; г = rectum; sb = spermatophore bursa; sg = sperm gutter; sp = sperm bursa; sr =
seminal receptacle.

latum complex is controversial (Verduin, species or a closely related species of the
1976). It is not my intention to address alpha- Bittium reticulatum complex. Bittium reticula-
level problems in this generic review, but the tum is exceedingly variable in shell sculpture
Azorean population used for the anatomical throughout its range (compare Figs. 2A, C,
study herein is considered by some as а sub- D), but this is not unusual among cerithioide-

282 HOUBRICK

ans. The pallial oviduct described by Johans-
son (1947) and notes and sketches made by
Ропаег (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 pallial cavity is
covered by the anterior extension of the renal
organ, which overlays the posterior pallial
gonoduct. The renal organ opens via a mus-
cular sphincter, the renal opening, into the
posterior pallial 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. 6B, 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. | was unsuccessful in finding spermato-
phores in either structure in specimens of
Bittiolum varium from Florida. A new genus
from the Indo-Pacific, /ttibittium, described
herein, deviates from the typical pallial 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 pallial cavity into the visceral
coil.

The spawn of Bittium reticulatum was first
described and figured by Meyer & Mobius
(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: pl. 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 ribbon 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, I-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. 3I-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 ОЕ 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. ЗА-Е and
3I-L), and unlike the shells of other eastern
Atlantic Bittium species. п 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. А|-
though an argument could be made that this
species represents yet another new genus, |
have conservatively placed Bittium impen-
dens under Bittium, s.s, with a query, be-
cause it does have many characters т 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-
за! varix (Fig. 3I-L). The protoconch (Fig. ЗМ)
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 whorls 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 pallial cavity into visceral coil. No
spermatophore bursa in lateral lamina of pal-
lial 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. 81) is unique among
the Bittium-group. The distinctive propodial
and epipodial 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 pallial oviduct and the protrusion of the
albumen gland through the posterior pallial
cavity into the visceral coil are highly unusual
autapomorphies, and set Ittibittium, gen. n.,
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 /ttibittium, gen. n., has lost this feature.
The spawn mass of /ftibittium, gen. n., 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,
1908, which has been shown to Бе a rissoid of
the genus /sselia (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: pl. 18, fig. 125; Tryon,
1887: 155, pl. 30, fig. 20; R. Johnson,
1964: 122, pl. 12, fig 14; Kay, 1979: 120,
figs. 220, 450, Е.

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.
8Е-1) 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.

Pallial 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 pallial cavity, comprising

GENERIC REVIEW ОЕ ВП ТИМАЕ 285

FIG. 8. ЗЕМ micrographs of Ittibittium рагсит from Honolulu, Hawaii (USNM 857100). А, В, 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 шт; G, H, embryonic
Shells removed from eg capsule, bar = 23 um; I, larval and early whorls of shell, bar = 0.4 mm: J, shell with
Strong axial ribs, 5.3 mm length; К, operculum, bar = 0.2 тт; L, detail of penultimate and body whorl,
Showing details of sculpture and aperture, Баг = 0.6 тт; M, apertural view of shell, 3.6 mm length.

286 HOUBRICK

mee №

FIG. 9. SEM micrographs of radula of /ttibittium parcum from Honolulu, Hawaii (USNM 857100). A, middle
of radular ribbon with right marginal teeth folded back, bar = 30 um; В, detail of rachidian and lateral teeth,

bar = 8 um.

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 pallial wall. Kidney large,
extending from anterior of style sac forward,
into roof of posterior pallial 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 ощег 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 pallial gonoducts. Pallial 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 ОЕ BITTIINAE 287

deep. Development direct with young snails
hatching from eggs.

Discussion

“Bittium” parcum has not been cited com-
тоту 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. 8C-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. /ttibittium par-
cum has an unusual pallial 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 pallial 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-group, 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 еп-
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 /ttibittium
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.
Cossmann (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 уапит to that genus. Abbott
(1974), probably following this cue, later re-
ferred western Atlantic species of Bittium, $.1.,
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 B. alternatum, are so
alike, they are treated jointly in the section
below.

Bittiolum varium (Pfeiffer, 1840)
(Figs. 10-11)

Cerithium varium Pfeiffer, 1840: 256.

Cerithium columellare Orbigny, 1842: pl. 23,
figs. 13-15; 1845: 244 (in part; syntypes
BMNH).

Cerithium gibberulum C. B. Adams, 1845: 5
(Lectotype MCZ 186078, type locality Ja-
maica).

Bittium varium (Pfeiffer). Tryon, 1887: 152, pl.
29, fig. 86; Perry, 1940: 134, pl. 28, fig.
202.

Cerithium (Bittium) gibberulum (C. B. Ad-
ams). Kobelt, 1898: 245-246, pl. 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.
11A, 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. 11B, cg) in both sexes
emerging from floor of right side of pallial cav-
ity, running down right side of foot leading into
epipodial groove. Ciliated gutter terminating
in small glandular ovipositor (Fig. 11B, ovp) at
edge of foot in females. Posterior third of sole
with median longitudinal slit leading into mas-
sive mesopodial mucus gland (Fig. 11A,
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. 11B, me) bilobed, smooth,

GENERIC REVIEW ОЕ BITTIINAE 289

FIG. 10. SEM micrographs of Bittiolum varium from Ft. Pierce, Florida (USNM 77639). А, В, D, E, two shells
showing sculptural variation and shell shape; length 3.2 тт; С, immature shell, length 2.8 тт; Е, С,
operculum, bar = 0.2 тт; H, sculpture of early whorls, bar = 0.3 mm; |, protoconch, Баг = 88 um.

without papillae, slightly scalloped, iridescent
at edges.

Pallial 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. 11C) short.
Rachidian tooth (Fig. 11D) with cutting edge
of 3 small denticles on each side of central
cusp. Lateral tooth (Fig. 11D) 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). А, В, critical point
dried specimens showing external anatomical features of headfoot, bar = 0.2 mm; C, mid-section of radula,
bar = 21 um; D, detail of rachidian and lateral teeth, bar = 7 рт. amg = anterior mucus gland; cg =
ciliated groove; eps = epipodial skirt; | = lip of mouth; mmg = metapodial mucus gland; op = operculum;
оур = ovipositor.

Right zygoneury between subesophageal anterior as seminal vesicle, containing dimor-

and right pleural ganglion. phic sperm. Males producing crescent-
shaped spermatophore with flared bifurcate
Reproductive System: Ducts of testicular fol- end and pointed, filamentous tip. Spermato-

licles joining to form spermatic duct, moving phores containing both eu- and parasperma-

GENERIC REVIEW OF BITTIINAE 291

tozoa. Ovary cream colored, overlying brown
digestive gland, extending forward to stom-
ach. Pallial 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 um in diameter, hatching as veliger lar-
vae, becoming planktotrophic.

Bittiolum alternatum (Say, 1822)

Turritella alternata Say, 1822: 243.

Pasithea nigra Totten, 1834: 369, figs. 7a, b.

Bittium nigrum (Totten), Gould, 1870: 321, fig.
590.

Bittium alternatum (Say), C. W. Johnson,
1915: 127.

Diastoma 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 epipodial skirt and nar-
rowly elongate anterior foot are also distinc-
tive external features (Fig. 2) of both
examined Bittiolum species. The ovipositor

(Fig. 11B, 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-
¡um 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, | 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 marine grasses,

292 HOUBRICK

could have important implications for the
abundance and distribution of Zostera.

Bittiolum уапит 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.

STYLIDIUM 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 роопу 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 pallial oviduct. Openings to
sperm bursa and seminal receptacle well-
separated. Long ciliated ridge tract in lateral
lamina of pallial 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-
lial 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, pl. 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, pl. 57, fig. 3; Ruhoff, 1973: 81.

Bittium eschrichtii (Middendorf). Oldroyd,
1927: 18-19, pl. 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, 1) 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 ОЕ BITTIINAE 293

FIG. 12. Stylidium eschrichtii from 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 тт; G, SEM micrograph of protoconch and early whorls, Баг = 0.3 mm: H,
|, SEM micrographs of operculum, showing eccentric nucleus and attachment scar, 2.4 тт length.

Pallial Cavity: Osphradium tan, vermiform,
non-pectinate, extending length of pallial 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. 13. SEM micrographs of radula of Stylidium eschrichtii (USNM 804376); А, section of mid-radular ribbon
with marginal teeth folded back, bar = 38 um; В, detail of rachidian and lateral teeth, bar = 12 um.

on left side of pad, posterior to cuticular gas-
tric shield; short, wide style sac one-half stom-
ach length, separate from 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
pallial 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 pallial
oviduct with thick, white, opaque albumen
gland comprising flocculant transverse glan-
dular ridges; mid-section of pallial oviduct with
thin, weak glandular transparent walls; very
thick, opaque transverse glandular ridges
present in anterior third of pallial 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 $. 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, 1) 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. | observed a large intertidal population
living among the intertices of gravel and algae

GENERIC REVIEW OF BITTIINAE 295

FIG. 14. Anatomical features of Stylidium eschrich-
tii. A, head and anterior foot, showing pigment pat-
tern; В, position of salivary glands relative to nerve
ring. d = left dialaneury; Icg = left cerebral gan-
glion; Ipg = left pleural ganglion; 159 = left salivary
gland; rcg = right cerebral ganglion; rsg = right
зайуагу 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 um diameter in which
the embryos undergo direct development,
passing through the veliger stage and hatch-
ing as small snails.

LIROBITTIUM BARTSCH, 1911

Lirobittium Bartsch, 1911: 384 (Type species
by original designation, Bittium catalinen-
sis Bartsch, 1907). Thiele, 1929: 211;
Wenz, 1940: 757; Abbott, 1974: 106;
Gründel, 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
(1911: 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.
К 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 Bittium
catalinense and B. subplanatum to be syn-
onyms of Lirobittium attenuatum Bartsch,
1911, but gave no reasons for this decision.
Hertz (1981: 40) showed that Lirobittium sub-
planatum (cited as Bittium) was a valid spe-
cies. | 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, pl. 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, pl. 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 питег-
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, pl. 57, fig. 5 (Holotype,
USNM 160076; type locality, Catalina Id.,
California); Oldroyd, 1927: 23: Ruhoff,
1973: 130.

Bittium subplanatum Bartsch. Dall, 1921: 146;
Hertz, 1981: 40, figs. 23-27.

Bittium subplanatum Bartsch. Oldroyd, 1927:
23.

Bittium (Lirobittium) subplanatum (Batsch).
Abbott, 1974: 106.

Description

Shell (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 ВП ТИМАЕ 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.

Pallial Cavity: Osphradium brown, vermi-
form, without pectins, wide, about one-third
the ctenidial width, nearly equaling ctenidial
length. Ctenidium extending length of pallial
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 pallial
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, cg). 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

ant

A 0.3mm

HOUBRICK

С : [AA

FIG. 16. Lirobittium subplanatum. A, pallial 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
pallial oviduct; bm = buccal mass; cg = 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 1$ distinct.

Lirobittium subplanatum lives offshore on
sandy-rubble bottoms. The shell is frequently
severly eroded and abraided.

Spawn morphology of Lirobittium 1$ 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 ОЕ BITTIINAE 299

FIG. 17. SEM micrographs of radula of Lirobittium subplanatum (USNM 881021). A, radular ribbon with
marginal teeth spread open, Баг = 35 um; В, half row showing rachidian and lateral teeth, bar = 19 um;
С, detail of dentition of rachidian and lateral teeth, Баг = 10 рт; D, detail of dentition of marginal teeth, bar
= 12 um.

finally hatching out as small snails. Develop- nor Tuomey, 1848; nor J. de C. Sowerby,
ment is direct (pers. obs.). in Dixon, 1850). Thiele, 1929: 211;
Wenz, 1940: 756; Grúndel, 1976: 56-57.

CACOZELIANA STRAND, 1928 Cacozelia lredale, 1924: 246 (Type species

? Semibittium Cossmann, 1896: 29 (Type by monotype: Cerithium lacertinum
species by original designation: Cerith- Gould, 1861); not Cacozelia Grote, 1878
¡um cancellatum Lamarck, 1804; not [Lepidoptera]. Thiele, 1929: 211; Murray,

Semibittium Bronn, 1831; nor Lea, 1842; 1969: 111.

300 HOUBRICK

Cacozeliana Strand, 1928: 66 (new name for
Cacozelia lredale, 1924). Wenz, 1940:
756.

Lirobittium (Cacozeliana) Strand. Gründel,
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 а
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 арргорп-
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 granaria 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., Diastoma
Deshayes, 1850; Gourmya Fischer, 1884;
Campanile Fischer, 1884; and Plesiotrochus
Fischer, 1878 (Houbrick, 19816, 1981с,
19814, 19905, 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 Bittium-group genera by
five non-homoplastic synapomorphies (Fig.
1), further demonstrating its distinctiveness.
Gründel’s (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). Cacozeliana is today
monotypic and confined to the temperate wa-
ters of southern Australia. The type species,
Cacozeliana granaria (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-
naria 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

extensions of the basal plate are nearly ab-
sent.

3. The pallial oviduct of Cacozeliana grana-
ria (Fig. 20C), while having a typical layout, is
unique among known pallial 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, cg).
As pointed out earlier, a grape-like seminal
receptacle also occurs in some species of
Cerithium Bruguière, 1789, Rhinoclavis
Swainson, 1840, and т Diala А. Adams, 1861
(Houbrick, 1971, 1978, 1992, pers. obser.;
Ponder, 1991), although this structure т 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 granaria (Kiener, 1842)
(Figs. 18—20)

Cerithium granarium Kiener, 1842: 72-73, pl.
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,
pl. 184, figs. 225-227; 1865: pl. 19, fig.
135; Kobelt, 1898: 249, pl. 23, fig. 9.

Cerithium lacertinum Gould, 1861: 368 (Ho-
lotype USNM 16571; type locality Syd-
ney Harbor, New South Wales, Austra-
lia); 1862: 141; G. B. Sowerby, 1866: pl.
18, fig. 128; Tryon, 1884: 155, pl. 30, fig.
100; R. Johnson, 1964: 96, pl. 11, fig. 4.

Bittium granarium (Kiener). Tryon, 1887: 155,
pl. 30, fig. 98; Wells, 1984: 30-31.

Synonymic Remarks

Kiener's (1842) name, granarium, predates
Gould’s (1861) /acertinum. Examination of the
holotypes of both taxa leaves по 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

301

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, about twice
snout length. Snout narrow, short (Fig. 20А,
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).

Pallial 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. 18. Cacozeliana granaria from King George Sound, Western Australia (USNM 858551). A-D, two shells
showing variation in color pattern and sculpture, length 22.4 mm and 20.2 mm, respectively; E, SEM
micrograph of immature shell, bar = 0.6 mm; F, SEM micrograph of protoconch, bar = 16 um; ©, 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. 20B, Isg, rsg) paired,
vermiform, coiled, Iying 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 ОЕ BITTIINAE 303

FIG. 19. Radula of Cacozeliana дгапапа from King George Sound, Western Australia (USNM 858551). А,
mid-section of radula, bar = 60 рт; В, details of rachidian and lateral teeth, bar = 15 um.

0.25mm

FIG. 20. Anatomical features of Cacozeliana granaria. A, head and foot anterior, showing narrow snout; B,
position of salivary glands anterior to nerve ring; C, pallial oviduct, spread open to reveal interior details. a
= anterior end of pallial oviduct; ag = albumen gland; cg = 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 pallial gonoduct half; anterior half of male pallial gonoduct less
thick, glandular, having wide transverse folds glandular, white but not opaque. Female pal-
forming spermatophore organ in posterior lial 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, cg) comprising many swollen trans-
verse elements. Opening to spermatophore
bursa (Fig. 20C, 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 granaria
(Fig. 18) looks very much like those of some
Cerithium species, the weak epipodial skirt,
pallial 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. 18G). 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: pl. 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 granaria 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. | 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 8 Standen, 1901: 371-372
(Type species by original designation, Ar-
gyropeza divina Melvill & Standen, 1901).
Thiele, 1929: 212; Wenz, 1940: 757;
Grúndel, 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), А, radular ribbon with marginal
teeth spread open, bar = 100 um; В, half row, Баг = 50 um.

spiral cords and many 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. Pallial 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. 21C; 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
ofthe type species, Argyropeza divina Melvill
& Standen, 1901, are shown in Figures 21
and 22. | 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., Е. Min-
danao, Philippines (USNM 276898). A, B, apertural
and side views of adult shell, 8.1 mm length; C,
protoconch, bar = 100 pm.

glossate with hourglass-shaped rachidian
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.

ZEBITTIUM 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 the 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 ОЕ BITTIINAE 307

FIG. 24. SEM micrographs of radula of Varicopeza pauxilla. À, section of ribbon with some marginal teeth
spread open, bar = 50 um; В, detail of rachidian and lateral teeth, bar = 25 um.

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 bilobed 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-group.
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 | 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 Bittinae. 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. | wish to thank Dr. Antö-
nio Frias Martins, of the University of the
Azores, for sponsoring me at the First Inter-
national Workshop of Malacology, held at Säo
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 Acorianos “Alfonso Chaves.” Work
on the Western Atlantic species was done at
the Smithsonian Marine Station, Link Port,
Florida. | 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. | am grateful to
the University of Guam for laboratory space,
equipment and logistic support. | 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-
rine Biology at Albany, Western Australia. |
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. | 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 ОЕ BITTIINAE 309

of Cassiella abylensis. For technical assis-
tance (proofreading and SEM, and computer
macro design) | thank 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 |
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

АВВОТТ, В. Т., 1974, American seashells, 2nd ed.:
663 pp. illus. New York, Van Nostrand.

ADAMS, A., 1860. Mollusca Japonica: new species
of Aclis, Ebala, Dunkeria, etc. Annals and Mag-
azine of Natural History, series 3, 6: 118-121.

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, C. B., 1845, Specierum novarum con-
chyliorum in Jamaica repertorum synopsis. Pro-
ceedings of the Boston Society of Natural His-
tory, 2: 1-17.

ADAMS, Н. & A. ADAMS, 1853-1858, The genera
of Recent Mollusca; arranged according to their
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 129-138 (1858). 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, pls. 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, pl. 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,
pl. 11, fig. 14.

BUCQUOY, E., Р. DAUTZENBERG & G. DOLFUS,
1882-1886, Les Mollusques marins du Roussil-
lon, Vol. 1, Gasteropodes: 570 pp., Atlas, 66 pls.

BRONN, Н. G., 1831, Italiens Tertiär-Gebilde und

deren organische Einschlüsse. xii + 176 pp.
Heidelberg.

BRUGUIERE, 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. Etude paléontologique et
stratigraphique sur le terrain oligocène marin aux
environs d’Etampes. Mémoires de la Société Gé-
ologique de France, (3)3(1): 1-187, pls. 1-6
[With J. Lambert].

COSSMANN, M., 1889, Catalogue illustré des co-
quilles fossiles de l'Eocéne des environs de
Paris. Quatrième fascicule. Annales de la So-
ciété Royale Malacologique de Belgigue, 24:
3-381, pls. 1-12.

COSSMANN, M., 1896, Appendice No. 2 au Cata-
logue illustré des coquilles fossiles de ГЕосёпе
des environs de Paris. Annales de la Société
Royale Malacologique de Belgique, 31: 3-94,
pls. 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 Geologique de
France, (4) 2: 163-203, pls. 3—4.

COSSMANN, M., 1906, Essais de paléoconcholo-
gie comparée. Paris, Е.В. de Rudival, 7: 261 pp.;
14 pls., illustrated.

COTTON, B. C., 1932, Notes on Australian Mol-
lusca, with descriptions of new genera and new
species. Records of the South Australian Ми-
seum, Adelaide, 4: 537-547.

COTTON, B. C., 1937, Nomenclatural note. The
South Australian Naturalist, 18: 2.

CROSSE, М. H., 1863, Description d'espèces nou-
velles d'Australie. Journal de Conchyliologie, 11:
84—90, pl. 1.

DACOSTA, E. M. 1778, Historia naturalis Testace-
orum Britanniae, or the British Conchology. Lon-
don, xii + 254 pp., 17 pls.

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 Il. Gastropoda and Scaphopoda. Bul-
letin of the Museum of Comparative Zoology,
Harvard 18: 492 pp., 40 pls.

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, pls. 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. Н., 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 112: 1-217, pls. 1-22.

DALL, W. Н., 1924, Notes on molluscan nomencla-
ture. Proceedings of the Biological Society of
Washington, 37: 87-90.

DALL, W. H. & P. BARTSCH, 1901. A new Califor-
nian Bittium. The Nautilus, 15: 58-59.

DELL, В. 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 Açores. Résultats des
Dragages effectives par le Yacht l'Hirondelle
pendant sa Campagne scientifique de 1887. 112
рр., pls. 1-4. Monaco.

DESHAYES, С. P., 1850, Traité élémentaire de
conchyliologie, Atlas, 80 pp., 132 pls. V. Masson,
Paris.

FÉRUSSAC, J. В. L. ФА. DE, 1819, Histoire na-
turelle general et particuliere des mollusques ter-
restres et fluviatiles. Paris, 1: 128 pp., 162 pls.

FINLAY, H. J., 1927, A further commentary on New
Zealand molluscan systematics. Transactions of
the New Zealand Institute, 57: 320-484, pls. 18—
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 pls.

FISCHER, P., 1878, Diagnosis molluscorum novo-
rum. Journal de Conchyliologie. 26: 211-212.

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 pls. (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 рр.;
317 figs.

FRETTER, V. & M. PILKINGTON, 1970, Proso-
branchia. Veliger larvae of Taenioglossa and
Stenoglossa. Conseil International pour ГЕх-
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, А. N., 8 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, Report on the Invertebrata of
Massachusetts. Boston, Wright and Potter, 2nd.
ed., ii + 427 pp., pls. 16-27.

GRAHAM, A., 1988. Mollusks: prosobranch and
pyramidellid gastropods. Keys and notes for the
identification of the species, 662 pp., 276 figs. In:
О. М. KERMACK & В. $. К. 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. Е. 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, 19: 262—
267.

GRUNDEL, J., 1974, Bemerkungen zur Fassung
der Gattungen Procerithium Cossmann, 1902
und Cryptaulax Tate, 1869 (Gastropoda, Cerithi-
acea) im Jura. Zeitschrift fur geologische Wis-
senschaft, 2: 729-733.

GRÜNDEL, J., 1976, Zur Taxonomie und Phyloge-
nie der Bittium-Gruppe. Malakologische Abhand-
lungen, 5: 33-59. `

HARDISON, L. K., & C. L. KITTING, 1985, Epi-
phytic algal browsing by Bittium 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). Helgolander 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. А review of several eastern Pa-
cific Bittium species (Gastropoda: Cerithiidae).
The Festivus, 13: 25—44.

HOUBRICK, R. S., 1971. Some aspects of the
anatomy, reproduction and early development of

GENERIC REVIEW ОЕ BITTIINAE 311

Cerithium nodulosum (Bruguière) [Gastropoda:
Prosobranchia]. Pacific Science, 25: 560-565.
HOUBRICK, R. S., 1977, Reevaluation and new
description of the genus Bittium (Cerithiidae).

The Veliger, 20: 101-106.

HOUBRICK, R. S., 1978, Redescription of Bittium
proteum (Jousseaume, 1930) with comments оп
its generic placement. The Nautilus, 92: 9-11.

HOUBRICK, В. S., 1980a, Review of the deep-sea
genus Argyropeza (Gastropoda: Prosobranchia:
Cerithiidae). Smithsonian Contributions to Zool-
ogy, 321: 30 pp., 11 figs.

HOUBRICK, R. S., 1980b, Reappraisal of the gas-
tropod genus Varicopeza Gründel (Cerithiidae;
Prosobranchia). Proceedings of the Biological
Society of Washington, 93: 525-535.

HOUBRICK, R. S., 1981a, Systematic position of
the genus Glyptozaria lredale (Prosobranchia:
Gastropoda). Proceedings of the Biological Soci-
ety of Washington, 94: 838—847.

HOUBRICK, В. S., 19816, Anatomy of Diastoma
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: 2-11.

HOUBRICK, В. S., 1981d, Anatomy, biology and
systematics of Campanile symbolicum with ref-
erence to adaptive radiation of the Cerithiacea
(Gastropoda: Prosobranchia). Malacologia, 21:
263-289.

HOUBRICK, R. S., 1987a, Transfer of Cerithiopsis
crystallina Dall 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, В. 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. Malacological 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. Е.
WELLS, D. |. WALKER, H. KIRKMAN, € В. LETH-
BRIDGE, eds., Proceedings of the Third Interna-
tional Marine 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 Bruguiére in the Indo-Pacific (Cerithi-
idae: Prosobranchia). Smithsonian Contributions
to Zoology, 510: iv + 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, pls.
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, pls. 4-6.

JOHANSSON, J., 1947, Uber den offenen Uterus
bei einigen Monotocardiern ohne Kopulationsor-
gan. Zoologiska Bidrag fran Uppsala, 25: 102-
110.

JOHNSON, C. 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 pls.

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, pls. 1-32
[pls., 1841; text, 1841-1842].

KOBELT, W., 1888-1898, Die Gattung Cerithium,
297 pp., 47 pls., in F. H. W. MARTINI & J. Н. CHEM-
мт7, 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. C., 1842, Descriptions of eight new species
of shells, native to the United States. American
Journal of Science and Arts, 42: 106-112.

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, М. 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, 1829 and Alaba A. Adams, 1853. In
W. Е. 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, Е. & Е. MARCUS, 1963. Mesogastropo-
den von der Küste Sao Paulos. Abhandlungen
der Mathematisch-Naturwissenchaftlichen
Klasse Jahrgang 1963, Akademie der Wissen-
schaften und der Literatur, Wiesbaden, 1: 1-105.

MARSHALL, W. B., 1917, The Wangaloa beds.
Transactions of the New Zealand Institute, Well-
ington, 49: 450-460, pls. 34-37.

MARTIN, K., 1914, Die Fauna des Obereocäns von
Nanggulan, auf Java. Sammlungen des Geolo-
gischen Reichs-Museums in Leiden, 2: 107-178,
pls. 1-6.

MELVILL, J. C. & 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 |: Cephalopoda,
Gastropoda, Scaphopoda. Proceedings of the
Zoological Society of London, for 1901, 2: 327-
460.

MEYER, H. А. & К. MOBIUS, 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. Memoires sciences na-
turelles de l'Académie Impériale des Sciences,
St. Petersburg, 6(2): 187 pp., 10 pls.

MONTAGU, G., 1803, Testacea Brittannica, or nat-
ural history of British shells. London. xxxviii +
606 pp., 16 pls.

MONTEROSATO, T., 1884, Nomenclatura gener-
ica e specifica di alcune conchiglie Mediterranee.
Virzi, Palermo, 152 pp.

MONTEROSATO, T., 1917. Molluschi viventi e
quaternarii raccolti lungo le coste della Tripolita-
nia. Bulletino della Società Zoologica ltaliana,
(3)4: 1-28, 1 pl.

MURRAY, F. V., 1969, The spawn and early life
history of Cacozeliana granaria (Kiener, 1842)
(Gastropoda: Cerithiidae). National Museum
Memoirs, Victoria, 29: 111-113.

NAIM, O., 1982, Bilan qualitatif et quantitatif de la
fauna malacologique mobile associeé aux algues
du lagon de Tiahura (lle de Moorea, Polynesie
Française). Malacologia, 22: 547-551.

NORDSIECK, F., 1968, Die europäischen Meeres-
Gehäuseschnecken (Prosobranchia). Gustav
Fischer Verlag, Stuttgart, 273 pp.

OLDROYD, |. S., 1927, Marine shells of the west
coast of North America. Stanford University
Press, Stanford, 2(3): 339 pp., pls. 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 pls.

ORBIGNY, A. de, 1841-1846. Mollusques. in: R. DE
LA SAGRA, ed., Histoire physique, politique, et na-
turelle de l'île de Cuba. 1: 1-208 (1841), 209—

264 (1842), 2: 1-112 (1842), 113-380 (“1846”
[1853?]), Atlas, 1845.

PAYRAUDEAU, B. C., 1826, Catalogue descriptif
et methodique des annelides et des mollusques,
a l'Ile de Corse. Academie des Sciences. Béchet,
Paris, 1826. 218 pp.

PEASE, W. H, 1861, Descriptions of forty-seven
new species of marine shells from the Sandwich
Islands, in the collection of Hugh Cuming. Pro-
ceedings of the Zoological Society of London, for
1860: 431—438.

PERRY L. M., 1940, Marine shells of the western
coast of Florida. Bulletins of American Paleontol-
ogy, 26: 260 pp., 39 pls.

PFEIFFER, L., 1840, Übersicht der im Januar, Feb-
ruar, und Marz 1839 auf Cuba gesammelten Mol-
lusken. Archiv für Naturgeschichte, 6: 250-261.

PHILIPPI, R. A., 1836, Enumeratio molluscorum Si-
ciliae, cum viventium tum in tellure Teriaria fos-
silium quae in itinere suo observavit. S. Schrop-
pii, Berlin. 1: 268 pp., 12 pls.

PHILIPPI, R. A., 1848, Centuria tertia testaceorum.
Zeitschrift für Malakozoologie, 11: 161-176.

PILSBRY, H. A. & E. G. VANATTA, 1905, On two
Намайап Cerithiidae. Proceedings of the Acad-
emy of Natural Sciences, Philadelphia, 57: 576.

PILSBRY, Н. A. 8 Е. G. VANATTA, 1908, Descrip-
tions of new Намайап marine shells. The Nau-
tilus, 22: 56-58, figs. 1-3.

PONDER, W. F., 1985, A review of the genera of
the Rissoidae (Mollusca: Mesogastropoda: Ris-
soacea). Records of the Australian Museum,
supplement 4: 1-221.

PONDER, W. F., 1991, The anatomy of Diala, with
an assessment of its taxonomic position (Mol-
lusca: Cerithioidea) pp. 499-519 in, Е. WELLS, D.
WALKER, H. KIRKMAN & В. LETHBRIDGE, eds., Pro-
ceedings of the third international marine biolog-
ical workshop: the marine flora and fauna of Al-
bany, Western Australia. Western Australian
Museum, Perth, vol. 1.

PONDER, W. Е. (in press). A new cerithiid from
South Western Australia (Mollusca: Gastropoda:
Cerithiidae). z

PONDER, W. Е. & А. WAREN, 1988. A systematic
list of the family-group names and higher taxa in
the Caenogastropoda and Heterostropha. in:
Ponder, W.F. (ed.) Prosobranch phylogeny. Pro-
ceedings of a symposium held at the 9th interna-
tional malacological congress, Edinburgh, 1986.
Malacological Review. Supplement 4: 288-328.

POWELL, A. W. B., 1979. New Zealand Mollusca.
Land, marine and freshwater shells. Collins, Syd-
ney. 500 pp., 82 pls.

RANG, P. S., 1829, Notice sur le Litiope, Litiopa,
genré nouveau de Mollusques gastropodes. An-
nales des Sciences Naturelles, 16(3): 303-307.

RAYNEVAL, V. D. H. de, V. DEN HECKE & G.
PONZI, 1854. Catalogue des fossiles du Monte
Mario. Versailles. 25 pp., 4 pls.

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

GENERIC REVIEW OF BITTIINAE 313

RISSO, A., 1826, Histoire naturelle des principales
productiones de l'Europe méridionale et particu-
lièrement de celles des environs de Nice et des
Alpes Maritimes, Paris, vol. 1, 448 pp; vol. 4, 444

PP-

RUHOFF, Е. A., 1973. Bibliography and zoological
taxa of Paul Bartsch. Smithsonian Contributions
to Zoology, 143: v + 166.

SAY, T., 1822. An account of some of the marine
shells of the United States. Journal of the Acad-
emy of Natural Sciences, Philadelphia, 2: 221-
276.

SHUTO, T., 1978, Contribution to the geology and
paleontology of South East Asia. CXCIX. Notes
on Indonesian Tertiary and Quaternary gastro-
pods mainly described by the late Professor К.
Martin. Il. Potamididae and Cerithiidae. Geology
and Paleontology of South East Asia, 19: 113—
160, pls. 15-18.

SOWERBY, С. B., 1855. Monograph of the genus
Cerithium Adanson. Thesaurus Conchyliorum, or
monographs of genera of shells. London. 2(16):
847-899, pls. 176-186.

SOWERBY, С. B., 1866. Cerithium, in: L. A. REEVE,
ed., Conchologia Iconica: or illustrations of the
shells of molluscous animals. 15 [atlas]: 20 pls.
+ index. London.

SOWERBY, J. DE C., 1850, in F. Dixon, The geol-
ogy and fossils of the Tertiary and Cretaceous
formations of Sussex. London, Longman, Brown,
Green and Longmans, 422 pp., pls.

STRAND, E., 1928, Miscellanea nomenclatorica
zoologica et palaeontologica, I-II. Archiv für
Naturgeschichte, (А) 92: 30-75.

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

STRATHMANN, M. F., 1987, Reproduction and de-
velopment of marine invertebrates of the north-
ern Pacific coast: data and methods for the study
of eggs, embryos, and larvae. Seattle, University
of Washington Press, xii + 670 pp, illustrated.

SWAINSON, W., 1840, A treatise on malacology or
the natural history of shells and shell-fish. Long-
mans, London. viii + 419 pages.

TATE, R., 1869, Contributions to Jurassic palaeon-
tology. 1. Cryptaulax, a new genus of Cerithiadae
(sic). Annals and Magazine of Natural History, 4:
417-419.

TATE, R., 1893. On some new species of Austra-
lian marine gastropods. Transactions of the
Royal Society of South Australia, 70: 189-197,
pl. 1.

THIELE, J., 1925. Gastropoda der Deutschen Tief-
see-Expedition. Il. Teil. Wissenschaftlische
Ergebnisse der deutschen Tiefsee-Expedition
auf dem Dampfer “Valdivia” 1898-1899, 17(2):
35-382 [1-348], pls. 13-46 [1-34], 31 text figs.
Gustav Fischer, Jena.

THIELE, J., 1929-1931. Handbuch der systematis-
chen Weichtierkunde, Jena, Gustav Fischer,
1(1): iv + 376 pp. (1929), (2): 402 pp. (1931),
783 figs.

THORSON, G., 1946, Reproduction and larval de-
velopment of Danish marine bottom inverte-
brates. Meddelelser Fra Kommissionen for Dan-
marks Fiskeri- og Havundersogelser, Serie:
Plankton, 4: 523 pp.

TIBERI, N., 1869, Spigolamenti nella conchiliologia
Mediterranea. Bullettino Malacologico, 2: 252—
27

TRYON, G. W., 1887, Cerithiidae. In, Manual of
conchology; structural and systematic with illus-
trations of the species. Philadelphia. Series 1, 9:
112-228, pls. 19-39.

TOTTEN, J. G., 1834, Descriptions of some new
shells, belonging to the coast of New England.
American Journal of Science, 26: 366-369.

TUOMEY, M., 1848, Report on the geology of
South Carolina, by M. Tuomey [geologist to the
State]. A. S. Johnson, Columbia, South Carolina,
iv + 293 pp, 2 maps.

VAN MONTFRANS, J., В. J. ORTH & S. А. VAY,
1982, Preliminary studies of grazing by Bittium
varium on eelgrass periphyton. Aquatic Botany,
14: 75-89.

VERDUIN, A., 1976. On characters, variability, and
distribution of the European marine gastropods
Bittium latreillii (Payraudeau) and Bittium lacteum
(Philippi). Basteria, 40: 133-142.

VIRNSTEIN, R. W. & M. C. CURRAN, 1986, Colin-
ization of artificial seagrass versus time and dis-
tance from source. Marine Ecology, 29: 279-
288.

WATSON, В. B., 1884-1886, Report on the Sca-
phopoda and Gasteropoda collected by H.M.S.
Challenger during the years 1873-1876 in Re-
port on the scientific results of the voyage of
H.M.S. Challenger during the years 1873-76.
15(42): 756 pp., 53 pls.

WEINKAUFF, H. C., 1881, Die Gattungen Rissoina
und Rissoa. Pp. 41-80, in F. H. W. MARTINI & J.
Н. CHEMNITZ, Neues Systematisches Conch-
ylien-Cabinet. . . . Nürnberg, Bauer & Raspe,
1(22).

WEISBORD, N. E., 1962, Late Cenozoic gastro-
pods from northern Venezuela. Bulletins of
American Paleontology, 42: 672 pp., 48 pls.

WELLS, F. E., 1984, A guide to the common mol-
lusks of South-Western Australian estuaries.
Western Australian Museum, Perth. 112 pp.

WENZ, W., 1938-1944, Gastropoda, 1: Allge-
meiner Teil und Prosobranchia In, O. H. SCHINDE-
woL, ed., Handbuch der Paláozoologie. Berlin,
Gebrüder Borntrager, 6: vi + 1,639 pp., 4211
figs.

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

WOODRING, W. P., M. N. BRAMLETTE & W. S.
KEW, 1946, Geology and paleontology of Palos
Verdes Hills, California. U.S. Geological Survey
Professional Paper 207: v + 145 pp., pls. 1-37.

Revised Ms. accepted 18 February 1993

\
o
|

MALACOLOGIA, 1993, 35(2): 315-342

SOME ASPECTS OF THE FUNCTIONAL MORPHOLOGY OF THE SHELL
OF INFAUNAL BIVALVES (MOLLUSCA)

G. Thomas Watters

Aquatic Ecology Laboratory, Ohio State University, Columbus, Ohio 43212, USA

ABSTRACT

Measures of streamlining, gapage, umbo position, and pallial sinus depth were taken from 632
species of Bivalvia in 13 families. Two types of gapage were measured: exchangeable gapage
due to the rocking motion of the shells along a dynamic dorso-ventral axis, and permanent
gapage, that portion of the gapage that could not be closed by the rocking movement. Models
were given to predict changes in shell shape as an adaptation to infaunal life. Three stages occur
in the sequence of shell shapes from shallow to deep infaunal dwellers for the families studied.
The first stage is represented by unstreamlined, often sculptured shells with complete valve
closure. The second, or intermediate stage, consists of an increase in streamlining, a loss of
sculpture, central placement of the umbo, and temporary gapes in the shell for pedal and
siphonal outlets. These gapes may be opened and closed by rocking the shells along a dorso-
ventral axis (exchangeable gapage). Two paths are evident out of the intermediate stage into the
third. The myid path results in unstreamlined shells with central umbos. The solenid path results
in streamlined shells with a variable umbo position. Some families, such as the Mactridae, have
members along both paths.

The entry into this sequence requires a particular set of pre-existing morphological conditions.
The lack of these conditions in most species studied has resulted in a bottleneck, with few
species in the deep infaunal zone. The constraints of bivalve shell geometry have limited the

success of that group in otherwise favorable habitats.
Key words: Bivalvia, morphospace, functional morphology, ecology, phylogeny.

INTRODUCTION

The class Bivalvia of the phylum Mollusca

is the most diverse group of organisms extant

that principally have radiated into the deep
infaunal zone. Nevertheless, the fossil record
shows that this colonization required nearly
200 million years to become widespread, al-
though the earliest known representatives of
this class may have been shallow infaunal
burrowers (Pojeta et al., 1973; Jell, 1980; but
see Yochelson, 1981).

The deep infaunal habitat has several po-
tentially positive adaptive features. Predation
is reduced because of the general lack of bur-
rowing molluscivores. The sediment acts as a
buffer, ameliorating thermal, salinity, pH, and
other environmental extremes. Desiccation is
minimized. For these reasons, this habitat is
advantageous to an organism associated with
this niche.

Therefore why did so few members of the
Bivalvia colonize the deep infaunal zone? It is
probable that the changes required in evolv-
| ng into the deep infaunal zone involve such
considerable morphological modifications
that members of few lineages have survived

315

or ever began the transition. Burrowing in the
substrate to greater depths must have oc-
curred by degrees, where each modification
was either adaptively or neutrally selective.
Such intermediate morphological steps would
have had their own immediate selective ad-
vantage.

The acquisition of shell structures and be-
haviors associated with deep burrowing has
occurred in relatively few members of the bi-
valve families. This implies that characteris-
tics that made for survival in this habitat
served another function in another habitat,
and that these particular characteristics were
selected upon by natural factors or processes
that resulted in deep burial. Members of lin-
eages lacking these prerequisite characteris-
tics could not attain a deep infaunal exis-
tence. These characteristics include the
anatomy of the living individual, behavior, and
the shape of the shell. This study is limited to
a consideration of the shell.

Shell Shape

К is here hypothesized that bivalves asso-
ciated with the deep infaunal habitat should

316 WATTERS

have a similar shell shape if there exists a
suite of characteristics necessary to achieve
this type of existence. The presence of ho-
meoplasy (similar shell shapes by conver-
gence, parallelism, or iteration) by individuals
of deep infaunal species across suprageneric
taxonomic levels would support this hypothe-
sis. This study proposes to obtain measures
of shell shape describing differences that may
arise in a transition from a shallow to a deep
infaunal existence. These measures are:

(1) degree of streamlining. This is а mea-
sure of the amount of surface area of the shell
that is oriented perpendicular to the long axis
of shell.

(2) relative position of the umbo. The place-
ment of the umbo on the shell, standardized
to remove size effects.

(3) relative depth of the pallial sinus. The
depth of the pallial sinus, standardized to ге-
move size effects.

(4) amount of permanent gape. Some bi-
valve shells do not close completely, leaving
gapes anteriorly and posteriorly. These shells
may open and close along a dorso-ventral
axis to close much of the gape, but some por-
tion may remain open. These are permanent
gapes. The amount of permanent gapage is
the sum of the anterior and posterior gapes in
the commissure of the shell that cannot be
closed by rocking the shells along a dorso-
ventral axis (Fig. 1:9, + 9»).

(5) amount of exchangeable gape. The
amount of gape created by rocking the shells
along a dorso-ventral axis minus the amount
of permanent даре (Fig. 1: pg + sg - 9, —
92).

These parameters are discussed т detail
under “Methods.”

Shell shapes form a predictable sequence
among individuals that inhabit the shallow to
deep infaunal habitats because a necessary
suite of shell characteristics is needed to suc-
ceed in a deep infaunal habitat. This se-
quence is defined by the distribution of each
measurement specified for representatives of
the species in this study. The existence of a
sequence could explain the rarity of deep
infaunal bivalves and the degree of homeo-
plasy present in burrowing bivalves in gen-
eral. In may be that few Recent representa-
tives of bivalve lineages are deep infaunal
burrowers because ancestral members of the
|пеаде lacked the shell characteristics nec-
essary to enter the sequence.

The sequence may be divided into three
phases. The shallow infaunal phase contains

bivalves that do not have exchangeable
gapage. The deep infaunal phase contains
forms with permanent gapage. These individ-
uals often are deep burrowing or sedentary
forms. The intermediate phase connects
these two phases and contains forms having
exchangeable gapage. Homeoplasy would be
the expected result if only a few sequences of
shell shape morphologies existed among
those individuals that occur in these phases.

К has long been known that there is con-
vergence in shell characteristics in bivalves.
Seed (1980b: 32) stated that “perhaps one of
the most striking features concerning the ev-
olution of such a diverse group as the bi-
valves has been the repeated appearance of
a comparatively restricted number of very
successful shell morphologies.” Linnaeus,
Cuvier, Bruguière, and Lamarck placed bi-
valves in only a few genera. They based their
criteria for classification primarily upon shell
form and a consideration of hinge dentition,
but little internal anatomy. This is in contrast
to a recent classification (Vaught, 1989) that
lists nearly 1,000 genera. Таха not known to
be related may possess similar shells when
internal anatomy, dentition, and larval types
are also examined. This has been a major
obstacle to the study of fossil forms.

Two hypotheses may be formed to explain
this convergence, and they are not mutually
exclusive. The first states that similar shells
have arisen in response to similar environ-
mental pressures. Convergence has occurred
because of natural selection “favoring” a spe-
cific shell shape. However, evolution may
only act upon available morphological mate-
rial. Pre-existing structures may be co-opted
for a different use or an improved original
function if the genetic program can be modi-
fied in such a fashion. This is the basis behind
the second hypothesis of convergence т
shell shapes: bivalve shells may be similar
because there is only a limited range of val-
ues for shell geometric parameters that occur
in nature. Convergence may be expected be-
cause of this restriction if there are few viable
alternative shell shapes.

The results of this study suggest that the
cause of convergence in bivalve shell shape
may be explained as the consequence of a
sequence of morphologies. This sequence
represents a compromise between natural
selection and morphological constraints. Ev-
olution is conditional and the changes at any
step in a phylogeny depend upon the charac-
teristics of the previous step. Such “trends”

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS 317

FIG. 1. Rocking of shells along а dorso-ventral axis. Heavy line: Axis. x: Fixed pivot at cardinal teeth. Top:
Shells rocked forwards to open siphonal gape (sg). Middle: Shells at neutral position. Bottom: Shells rocked
backwards to open pedal gape (pg). g,: Non-closable permanent anterior даре. g,: Non-closable permanent

posterior gape.

have been modeled satisfactorily by a Markov
process or random walk (Bookstein, 1987).
As an example, Cope’s Law of Phyletic Size
Increase has been shown to be stochastic
(Stanley, 1973). The convergence of bivalve
shell shapes may be such a stochastic pro-
cess.

The molluscan shell has long been recog-
nized as a geometric form, at least in the ar-

tistic sense. Examples of this geometry, as a
by-product or necessity of biological design,
were not popularized until Thompson (1942)
published On Growth and Form. The further
study of shell geometry did not progress past
this recognition stage for many years. The
computations were time consuming and the
results difficult to visualize as three-dimen-
sional shapes. Recently, geometric studies of

318 WATTERS

this type have been facilitated by computers.
Raup (1961, 1962, 1963, 1966, 1967) identi-
fied the basic parameters of spiral coiling and
generated simulations of molluscan shells by
computer emulation. He demonstrated that a
simple gastropod or cephalopod shell design
could be modeled with few variables. Savazzi
(1987) produced an even more realistic com-
puter generated model, and the recent work
of Fowler et al. (in press) has produced amaz-
ing simulations. The science of “theoretical
morphology” (Raup & Michelson, 1965) and,
more specifically, “conchyliometry” (coined
by Naumann, 1840), became a discipline be-
longing as much, if not more, to computer pro-
grammers and mathematicians as to biolo-
gists. The course of these studies culminated
in Bayer’s (1978) and Шег’$ (1992) purely
mathematical analyses of shell shape using
morphogenetic programs. The emphasis of
these studies had shifted from the biological
aspects of shell geometry to a consideration
of the biometrics as the sole purpose of the
investigation.

In 1970, Stanley published a study on ma-
rine bivalves that marked a turning point in
molluscan morphometrics. He presented a
synthesis of shell geometry, systematics,
ecology, and field observation. For the first
time, on a comprehensive scale, explanations
were advanced for why shells were shaped
like they were, rather than how they were
shaped. Following the studies of Trueman et
al. (1966a) and Nair & Ansell (1968) on the
dynamics of bivalve burrowing, Stanley's
work showed that members of such diverse
groups as the solecurtines, the solenids, the
cardiids, and the mactrids had highly conver-
gent shells because of similar habitats. From
his results, | have inferred the possibility of
analogous, predictable shell shapes in equiv-
alent niches despite phylogenetic position.

Stanley (1969, 1970, 1972, 1975, 1977b,
1981) documented the probable function of
many types of marine bivalve sculpture. | be-
lieve that the single most important conclu-
sion of these works was the concept of “com-
posite sculpture,” the exaptation (sensu
Gould & Vrba, 1982) of pre-existing sculpture
for vicarious multiple tasks. Gould and Vrba
coined this term for previous adaptations or
nonadaptations that have been co-opted for a
new function. For example, radial ribs may
have originated as sculpture strengthening
the shell in individuals of the Cardiidae (Stan-
ley, 1981). That sculpture has been exapted
to function as a burrowing device in many

members of the trachycardiinine cockles. As
aspects of the function of shell sculpture have
been discussed elsewhere, they generally will
not be addressed in this study.

Of central importance to this analysis is the
concept of the theoretical morphospace: the
array of potential shapes that an organism
may possess. This space usually is limited to
a few parameters, such as size, coiling rate,
or color, for experimental studies and repre-
sents the possible range of values of that pa-
rameter. The theoretical morphospace may
be contrasted with the actual morphospace.
The actual morphospace is the observed val-
ues of that parameter, or in a broader sense,
the form in which the organism is found in
nature. The actual morphospace is always a
subset of the theoretical morphospace. In its
simplest form, this methodology addresses
the question: why are things shaped the way
they are? Or conversely, why aren't they
shaped like something else? It is the latter
question that may be the most insightful, for it
implies a limitation of form and a constraint on
possible morphologies. The cause of this con-
straint may be fundamental to understanding
the organism in question. The idea of the the-
oretical morphospace has been applied to the
morphological features of several groups,
most notably coiling in cephalopods (Raup,
1967).

Convergence is most apparent in a mor-
phospace scenario. Phylogenetically unre-
lated groups that consistently occupy the
same morphospace have converged toward
the same values of the morphospace param-
eters. In this study, the sum of overlapping
regions is shown to lie along a sequence of
shell shapes.

Rudwick (1965) is usually given credit for
advancing the use of the paradigm approach
in biology, although this method of analysis
may have been in use for many years. The
term is from the Greek paradeigma, meaning
“example” or “model.” The methodology al-
lows the worker to form hypotheses concern-
ing the potential characteristics of an organ-
ism possessing a certain life style or behavior,
given information on the necessities of the or-
ganism’s life and its general morphology. For
example, given the morphological character-
istics of a small dinosaur, what changes are
necessary to metamorphose it into a bird?
The result is a model having parameters de-
scribing the organism in that life style as dic-
tated by the logic of the investigator and the
presumed efficiency of those characteristics.

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS

ARK
EI

FIGS. 2-8. Models of shell shape. 2. Model for interaction between permanent gapage (P) and streamlining
(S). 3. Model for interaction between permanent gapage (P) and exchangeable gapage (E). 4. Model for
interaction between permanent gapage (P) and position of umbo (U). 5. Model for interaction between
permanent gapage (P) and depth of sinus (N). 6. Model for interaction between exchangeable gapage (E)
and streamlining ($). 7. Model for interaction between streamlining ($) and position of umbo (U). Fig. 8.
Model for interaction between streamlining (S) and depth of sinus (N).

The value of the model is in its degree of re-
semblance to the actual organism. What are
the discrepancies, if any, and how are they
significant?

The paradigm model is similar to the theo-
retical morphospace. Both analyses compare
actual and hypothetical characteristics of an
organism. The model represents a region of
the theoretical morphospace that has a high
probability of being the actual morphospace,
based on outside inferences. Both form a
consistent pattern against which to compare
the results of analyses.

Models of Shell Shape

К is possible to predict sequences in the
values of shell shape parameters using the
paradigm methodology. These parameters
may be taken as a whole to describe the over-
all shell shape. The models are understood
most easily as pairwise comparisons of the
parameters.

Permanent Gapage
Streamlining would be expected first to in-
crease into the intermediate phase with in-
creasing depth of burrowing, and then
decrease as permanent gapage becomes
pronounced (Fig. 2). Increased streamlining
occurs as bivalves become more suited to
burrowing in the shallow infaunal zone. At a

and Streamlining:

critical depth, which varies from sediment to
sediment and depends upon the size of the
bivalve, the weight of the substrate limits the
depth of burial. Deeper burrowing can occur
in a lineage only by the formation of ex-
changeable gapage. This is the beginning of
the intermediate phase. The increasing de-
gree of exchangeable gapage should begin to
diminish the amount of streamlining. As ex-
changeable gapage is modified into perma-
nent gapage, streamlining should decrease
continuously as the life style shifts from effi-
ciently moving in the shallow substrate to a
deeply buried sedentary existence.

Permanent Gapage and Exchangeable
Gapage: As with streamlining, levels of ex-
changeable gapage should rise and then fall
with increasing permanent gapage and
deeper infaunal existence (Fig. 3). The peak
of exchangeable gapage lies within the inter-
mediate phase. Streamlining is modified into
exchangeable gapage, which in turn is mod-
ified into permanent gapage.

Permanent Gapage and Relative Position of
Umbo. The model suggests that the umbo,
as a relative measure of the position of the
cardinal teeth, should become centralized to
allow maximum exchangeable gapage as a
lineage enters the intermediate phase (Fig.
4). The position of the umbo in individuals

320 WATTERS

past the intermediate phase may depend
upon the type of life style. The location of the
umbo may be unimportant in sedentary forms
that lack both a functional foot and rocking of
the shell along a dorso-ventral axis. The
umbo may become placed anteriorly in tube-
dwelling forms, which have large muscular
feet, because of its associated pedal muscle
insertions. Thus two paths are expected out
of the intermediate phase.

Permanent Gapage and Relative Depth of Si-
nus. Аз burrowing depth increases, so must
the length of the siphons in non-tube dwelling
forms. This entails an increase in sinus depth
to accommodate them. The depth of the sinus
will be high within the intermediate phase
(Fig. 5). Two paths are predicted as the lin-
eage passes into permanent gapage. Si-
phons in tube-dwelling species do not in-
crease if they remain permanently exterior to
the shell, as in members of the solenaceans.
Siphons may remain retractile in other forms,
requiring a deep pallial sinus.

Exchangeable Gapage and Streamlining.
Streamlining is expected to increase into
the intermediate phase until exchangeable
gapage becomes more evident (Fig. 6). As
exchangeable gapage is modified into perma-
nent gapage, both exchangeable gapage and
streamlining should decrease. Thus, there
should be both a path out and in along the
exchangeable gapage axis.

Streamlining and Relative Position of Um-
bo. The relative position of the umbo should
become centralized for maximum exchange-
able gapage as streamlining passes into the
intermediate phase (Fig. 7). As previously
mentioned, the fate of the position of the
umbo depends upon factors not accounted
for in this model, and two paths are expected
out of the intermediate phase.

Streamlining and Relative Depth of Sinus.
With increasing streamlining, the relative
depth of the sinus should increase into the
intermediate phase (Fig. 8). Past this point
the sinus depth may remain constant or de-
crease.

METHODS AND MATERIALS
Taxa Used in the Study

Representatives of 632 Recent species
and subspecies of bivalves were used in this

study. Specimens were acquired from the fol-
lowing repositories and collections: Museum
of Comparative Zoology, Cambridge, Massa-
chusetts; National Museum of Natural His-
tory, Washington, D. C.; Ohio State University
Museum of Zoology, Columbus, Ohio; and
the author’s private collection. The identifica-
tion of museum specimens was taken from
collection records, with the following excep-
tions at Ohio State University. Individuals of
southeastern United States in the genus El-
liptio, and a few members of other genera
from that region, were identified by the author,
as were all marine species from that collec-
tion. These identifications may not reflect the
views of systematists at that institution. The
higher systematic levels are taken from
Vaught (1989).

Members of 15 families were selected for
study, representing most of the living infaunal
bivalve groups. These families, and the num-
ber of species or subspecies used in this
study for each in parentheses, are: Mactridae
(41), Cardiidae (56), Myidae (6), Psammobi-
idae (25), Solenidae (8), Cultellidae (9), Tell-
inidae (42), Semelidae (7), Donacidae (18),
Veneridae (103), Petricolidae (1), Unionidae
(276), Hyriidae (16), Mycetopodidae (13), and
Mutelidae (11). Many families were chosen
because they displayed a wide range of shell
forms: streamlined vs. rotund, sculptured vs.
unsculptured, etc. Others, such as the Solen-
acea, were chosen because their unique
forms offered insight into this study. Some
families subsequently were divided into sub-
families, and others grouped into orders bet-
ter to indicate functionally alike groups. The
Unionaceans, which have been omitted from
most studies of this sort, were represented by
the most taxa. They were included because
no other group of Recent bivalves encom-
passes such a wide range of shell shapes.
Other infaunal bivalve groups were not in-
cluded, for the following reasons. Individuals
of the anomalodesmaceans generally are too
rare to obtain a reasonable sample. The Ar-
cidae, Mytilidae, and Pinnidae have infaunal
members, but most are sessile and byssate,
and thus different from the free living infaunal
groups chosen for study (Newell, 1969; Rose-
water, 1961; Soot-Ryen, 1955, 1969). Mem-
bers of other groups, such as the Astartidae,
are too homogeneous to warrant repetitive
measurements. Individuals of the Lucinidae
are infaunal and have a wide range of shell
shapes, and members of many species are
common. However, the mode of circulating

FUNCTIONAL MORPHOLOGY ОЕ BIVALVE SHELLS 321

water of the lucinids is quite different from the
groups included here (Allen, 1958). The dif-
ferences are sufficient to eliminate it from this
study of infaunal groups. Because this study
deals only with Recent species, otherwise in-
teresting groups such as the largely extinct
Trigoniacea were excluded.

Measurements and Derived Values

The following measurements, all in mm,
were taken on individuals for each of the 632
species.

Length—the greatest length along an ante-
rior-posterior line (Fig. 9a). This line usually
was parallel to the hinge axis.

Height—the greatest dorsal-ventral height,
perpendicular to the line for length (Fig. 9d).
This line often extended through the umbo.

Width—the greatest lateral width, with both
valves closed (Fig. 9c).

Position of umbo—the distance from the
anterior margin to the umbo, along the length
line (Fig. 9b).

Depth of pallial sinus—maximum depth of
the sinus measured out to a curve that follows
the pallial line (Fig. 9e).

Anterior permanent gape—the maximum
width of any anterior space between the
valves when the valves are closed and rocked
forward, if possible (Fig. 9f). All measure-
ments of gape were made on dry shells with
separated ligaments and no commisural pe-
riostracum. The values obtained therefore
may be overestimated uniformly to some de-
gree.

Posterior permanent gape—the maximum
width of any posterior space between the
valves when the valves are closed and rocked
backwards, if possible (Fig. 9h).

Anterior exchangeable gape—the total an-
terior gape is the maximum width of any
space created anteriorly between the valves
when the valves are rocked backwards (Fig.
9g). The anterior exchangeable gape is the
total minus the permanent anterior gape.

Posterior exchangeable gape—the total
posterior gape is the maximum width of any
space created posteriorly between the valves
when the valves are rocked forwards (Fig. 9i).
The posterior exchangeable gape is the total
minus the permanent posterior gape.

The following derived values were calcu-
lated from the above measurements.

Streamlining (S)—a univariate estimate of
the relative amount of surface area exposed
- perpendicular to the direction of maximum
_ length. The algorithm was devised for this

study to permit the simple quantification of a
parameter that has been expressed histori-
cally as a multivariate construction. The met-
ric is dimensionless, independent of size, and
has a finite range of values. Its derivation,
characteristics, and application will be treated
in detail.

Workers in bivalve morphometrics have re-
alized that some shells are more elongate
than others and should offer less resistance
to the substrate in burrowing activities. Stan-
ley (1970) and subsequent authors (notably
Morton, 1976) have attempted to illustrate this
shape by graphing ratios of shell measure-
ments against one another and delineating a
region of the theoretical morphospace as
“streamlined.” The difficulty with this ap-
proach is that it requires two dimensions to
describe elongation. If one wishes to investi-
gate the relationships between elongation
and any other parameter, one must use mul-
tivariate correlations (at least three variables).
This has not been attempted, except in the
study of Thomas (1975) on glycymerid bi-
valves.

Streamlining in a different sense has been
mathematically defined and quantified by en-
gineers working with fluid and aerodynamics,
and several attempts have been made to treat
organisms in the same manner as ships and
planes. These studies generally focus on op-
timum shapes for maximum speed, or the re-
verse, maximum speeds given a certain
shape. One study calculated swimming
speeds of extinct marine reptiles (Massare,
1988). She calculated the total drag on rep-
tiles using an estimate of surface area, water
velocity, density of the medium, and the Rey-
nolds number (a function of body shape in
lamellar or turbulent flow). Such an analysis is
not applicable to bivalves burrowing through a
mixed substrate.

It must be emphasized that the use of the
term “streamlined” by malacologists working
with bivalves is not that of Massare. That ex-
pression is used here as a descriptive vari-
able, crudely measuring only the relative
amount of surface area normal to the long
axis of the shell, generally coinciding with the
direction of burrowing. It carries no connota-
tion of, or resemblance to, fluid dynamic the-
ory. Neither is it a dynamic value dependent
on burrowing speed, current velocity, or sub-
strate type. Although univariate, the quantifi-
cation of streamlining put forth in this study is
identical with the sense of that term used in
describing bivalve shell shape by Trueman et

322 WATTERS

er BEN

FIG. 9. Measurements used in study. a: Length. b: Distance of umbo from anterior margin. c: Width. d:
Height. e: Depth of pallial sinus. f: Permanent anterior gape. g: Total anterior gape. h: Permanent posterior
gape. i: Total posterior gape.

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS 323

al. (1966b), Stanley (1970), Alexander
(1974), Eagar (1974, 1978), Thomas (1975),
Morton (1976), and Seed (1980a, b).

The calculation of streamlining (S) in this
study estimates the shell shape as a rectan-
gular solid of dimensions Length x Width x
Height. The value of S lies between two hy-
pothetical limits, interpreted as the minimum
and maximum amount of streamlining for the
rectangular model. At the theoretical mini-
mum, Height and Width equal a unit measure
(Height and Width = 1), and Length = 0.
Movement is in the direction of Length, per-
pendicular to Height and Width. The model
resembles a sheet of paper moving perpen-
dicular to the face of the page. This is the
minimum amount of streamlining. The theo-
retical maximum is achieved when Length =
1 and Height and Width both = 0. This model
resembles a line of no thickness moving par-
allel to itself. Bivalves lie between the two ex-
tremes. The calculation is dependent on the
relationship between Length and the remain-
ing descriptors. This has the effect of stan-
dardizing data by size by removing any influ-
ence of Length. The equation can be written
as:

S = (Width/Length)(Height/Length)
(Length/Length) (1)

When Height and/or Width is very small rel-
ative to Length, S approaches 0. Conversely,
when Length is very small relative to Height
and/or Width, S approaches infinity (~). It is
possible to limit these theoretical boundaries
by raising the natural logarithm (e) to the ex-
ponent S and taking the inverse. Removing
the cancelled expression (Length/Length),
and raising e to the remaining parameters
yields the equation:

S = e((Height/Length)
(Width/Length)) (2)

Now аз Length/Height ог Length/Width >
0, $ > ~, and as Height/Length or Width/
Length > 0, $ — 1. Taking the inverse of the
function has the following effect. As Length/
Height or Length/Width — 0, $ — 0; as
Height/Length or Width/Length — 0, $ — 1.
The equation has the final form:

S = 1/(e((Height x Width)/
(Length)?)) (3)

The resulting parameter is independent of
original size, unitless, and ranges from a

value of 0 for no streamlining to a value of 1
for maximum streamlining. Although the val-
ues resemble percentages, they are not. As $
is univariate, it may be compared with other
morphometric parameters without the neces-
sity of multivariate analysis. The function is
nearly rectilinear within the biological range of
its values. In this study, a maximum S of 0.99
was encountered in several members of the
solenid genus Ensis; a minimum of 0.01 was
found in individuals of the epifaunal cardiid
Corculum cardissa (Linnaeus, 1758).

The choice of length as the direction of mo-
tion was necessitated by the lack of knowl-
edge of the actual life positions of most bi-
valves used in this study (Stanley, 1970). The
use of this metric is considered a normalizing
method. Arguments may be raised against its
use based upon the well-known fact that max-
imum length does not always correspond to
burrowing direction. This particularly is true of
such groups as the lucinids not treated here
(Allen, 1958). This discrepancy between
length and direction of movement exists pri-
marily in individuals of very shallow infaunal
species, having a low $ and no gapage. It can
be shown that as S increases, the angle of
offset diminishes, for the few species for
which data are available (Fig. 10). Most ofthe
species discussed here have an S value >
0.8. Thus, for most the forms covered, the
incongruity between length and direction of
movement is small. Even at large offset an-
gles the discrepancy is overestimated. The
species at this level of S are generally circular
in outline, or nearly so. The line of greatest
length is a secant through the shell outline, as
would be the direction of movement. Both ap-
proximately would be equal in length. Height
would differ little between the two lines, and
Width not at all. The calculation of S may
therefore be accurate even at low levels of S.

Relative position of umbo (U)—the mea-
surement of the position of the umbo was di-
vided by total length to standardize this vari-
able. The metric is a percentage of the total
length.

Relative depth of pallial sinus (N)—calcu-
lated as for U, using depth of pallial sinus.

Relative permanent gape (P)—standard-
ized with the formula:

(anterior permanent gape + posterior
permanent gape)/(2 x width) (4)

Relative exchangeable gape (E)—stan-
dardized with the formula:

324 WATTERS

LU
oO

offset angle

0.3 0.4 0.5 0.6

0.7 0.8 0.9 1.0

streamlining

FIG. 10. Offset angle (burrowing angle relative to greatest length) vs. streamlining.

(anterior exchangeable gape + posterior
exchangeable gape)/(2 x width) (5)

RESULTS

Comparison of Shell Shapes With Models

Permanent Gapage and Streamlining: A
comparison with the results reveals that al-
though streamlining initially does increase as
permanent gapage increases, past the inter-
mediate phase the degree of streamlining be-
comes constant rather than decreases in
many individuals (Fig. 11). There are two
paths out of the intermediate phase, although
the numbers of individuals in that region are
so few that it is difficult t0 make such a claim
with certainty. Individuals of the Tellinidae and
Myidae conform to the predicted model given
above. Deep burrowing forms have lost
streamlining and may be sedentary as adults.
Members of the solenaceans and some sole-
curtine psammobiids have maintained high
levels of streamlining despite pronounced
permanent gapage. This is due in large part to
the ability of many of these forms to construct
tubes in which they move (Drew, 1907, 1908).
The highest degree of streamlining is found in
the tube-dwelling members of Solen. Levels
of permanent gapage and streamlining are
both high in these forms because these bi-
valves no longer burrow through the sub-

strate, but rather move within water filled
tubes.

Permanent Gapage and Exchangeable
Gapage: The results support the model, but
two paths are suggested (Fig. 12). Members
ofthe solenaceans and some solecurtines oc-
cupy one path, but the individuals of the My-
idae and other members of the Solecurtinae
occur on the other path. The first path con-
tains forms having high levels of exchange-
able gapage and permanent gapage as the
result of their tube-dwelling behavior. It is im-
portant to note that members of the Solecur-
tinae have participated in both paths, and that
forms of the mactrids also are diverging. This
suggests that members of a single family may
not follow a single morphological path. This
result occurs in several families.

Permanent Gapage and Relative Position of
Umbo: Two paths are evident out of the in-
termediate phase (Fig. 13). The model pre-
dicts 0.5 for maximum exchangeable gapage,
but most bivalves have the umbos placed
slightly anterior to act as a source of attach-
ment and a buttress for pedal muscles. The
intermediate phase average relative position
of the umbo is approximately 0.4. From that
point (and perhaps before), the umbo may be
placed either anteriorly or slightly posteriorly.
The forms with anteriorly positioned umbos
are those that use the foot either as an anchor

FUNCTIONAL MORPHOLOGY ОЕ BIVALVE SHELLS 325

streamlining

0.2 0.4 0.6 0.8

permanent gapage

FIG. 11. Permanent gapage vs. streamlining. a: All data. b: Hypothesized paths. c: Cardiidae. d: Donacidae.
e: Mactridae. f: Solenidae. 9: Unionoida. h: Tellinidae, Semelidae. i: Cultellidae. |: Myidae. к: Psammobiidae.
|: Veneridae, Petricolidae. Shaded area: Actual morphospace. m: Myid path. s: Solenid path.

326 WATTERS

exchangeable gapage

02 0.4 0.6 0.8

permanent gapage
FIG. 12. Permanent gapage vs. exchangeable gapage. See Fig. 11 for details.

(Unionoida) or a wedge within a burrow ($0- the theoretical value of 0.5, indicating the em-
lenaceans, cultellids), not as a device for ac- phasis on active burrowing and exchangeable
tive burrowing. The second path tends toward gapage in most of its members (Tellinidae,

FUNCTIONAL MORPHOLOGY ОЕ BIVALVE SHELLS 327

position of umbo

0.2 0.4 0.6 0.8 1
permanent gapage
FIG. 13. Permanent gapage vs. position of umbo. See Fig. 11 for details.

Solecurtinae, and others). The families Car- Permanent Gapage and Relative Depth of Si-
diidae and Mactridae have morphologies nus: The results seem to suggest two ill-
tending toward both directions. defined routes away from the intermediate

328 WATTERS

phase. One toward slightly increased sinus
depth and the other toward greatly reduced
depth (Fig. 14). Within members of a family,
both paths may be found (Solecurtinae, Tell-
inidae, and Mactridae). Members of the so-
lenaceans have reduced the sinus to a mini-
mum despite their deep infaunal habitat. This
is due to a reduction in siphon length. Individ-
uals of solenaceans live in water-filled tubes
and may dwell at the surface, only retreating
to the bottom of the burrow to escape.

Exchangeable Gapage and Streamlining:
Streamlining is expected to increase into the
intermediate phase as exchangeable gapage
becomes more evident. The results support
this prediction (Fig. 15). Members of all fam-
ilies lie upon a fairly narrow region of the the-
oretical morphospace. This is unexpected in
view of the original prediction: as exchange-
able gapage is exapted into permanent
gapage, both exchangeable gapage and
streamlining should decrease. Thus, there
should be a path out and in. However, the
parameters used could not differentiate these
paths.

Streamlining and Relative Position of Umbo:
Two paths are apparent out of the intermedi-
ate phase (Fig. 16). The first is toward a
slightly more posterior position and contains
members of the Tellinidae, Donacidae, Sole-
curtinae, and Myidae. The second, toward a
more anterior placement, contains forms of
the solenaceans and the Unionoida. The
Mactridae and Veneridae have members in
both paths.

Streamlining And Relative Depth Of Sinus:
The relative depth of the sinus is predicted to
increase into the intermediate phase. Two
paths are possible beyond the intermediate
phase and this pattern is supported by the
results (Fig. 17), along with an unexpected
result. Members of the order Unionoida do not
participate in this path but reach a high level
of streamlining with no appreciable sinus (or
siphons). The presence of individuals of the
Myidae so far back on the path suggests that
the sequence is reversible along its path.

DISCUSSION
Family Accounts

Cardiidae. The cockles are a large family of
shallow infaunal dwellers with heavy compos-

ite sculpture. Anti-scouring, anchoring, and
burrowing sculptures may exist in the same
species (Stanley, 1981). These sculptural de-
vices are suited particularly to a shallow in-
faunal existence. Few members have colo-
nized the deeper infaunal zone.

However, three of the five subfamilies have
members that have entered the intermediate
phase. None have evolved beyond it. In the
Protocardiinae, containing the most primitive
living cockles, members of the genus Lopho-
cardium are in the intermediate phase. This is
a rarely encountered group of perhaps three
species. The Laevicardiinae contains the in-
termediate phase members in the genus Ful-
via. This genus also is composed of very few
species. The Trachycardiinae includes the
genus Papyridea, containing seven or eight
species.

The premier example of a group in the in-
termediate phase is members of the cardiid
genus Papyridea. One must know something
about their ancestral stock to appreciate their
remarkable modifications. Papyridea is a ge-
nus of the trachycardiinine cockles, which is a
widespread group of tropical and sub-temper-
ate species. The members of the subfamily
are characterized by: (1) strongly, radially
ribbed shells, ornamented with complex com-
posite sculptures used for burrowing and anti-
scouring (Stanley, 1981); (2) short siphons,
limiting them to a shallow infaunal existence;
(3) central, or nearly so, umbos; and (4) a
short hinge plate with simple interlocking lat-
eral teeth and centrally located cardinals. The
pronounced ribs apparently act as strength-
ening devices and on the shell margin tend to
interdigitate to form a “ventral hinge” (Carter,
1968).

Members of Papyridea have these shell
characteristics modified into features pre-
dicted for exchangeable gapage. The dorso-
ventral axis of shell rocking employs the fol-
lowing changes: (1) the central umbo and
cardinal teeth become the static dorsal pivot;
(2) the interdigitation of the ribs on the ventral
margin becomes a dynamic pivot as the
sculpture functions like the teeth on two inter-
meshed gears; and (3) the lateral teeth dis-
engage in the resting position, but alternately
mesh as the shells are rocked along the
dorso-ventral axis forward or backward. The
shell has become more streamlined (S =
0.74) than most other cockle shells. The
ribbed sculpture is minimized on the disc of
the shell, although the composite sculpture is
retained. The ligament is shortened and po-

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS 329

depth of sinus

04 06 08

permanent gapage
FIG. 14. Permanent gapage vs. depth of sinus. See Fig. 11 for details.

sitioned near the umbo where it does not in- ley, 1970). Unlike the shallow infaunal habitat
terfere with the rocking movements. The short of other members of the Trachycardiinae,
siphons have become more elongate (Stan- members of Papyridea are known to burrow

330 WATTERS

streamlining

exchangeable gapage

FIG. 15. Exchangeable gapage vs. streamlining. See Fig. 11 for details.

to approximately one half their length and are formis (Bruguière, 1789) “has longer siphons
moderately rapid burrowers. Stanley (1970: and lives at a greater depth than other cardi-
158) stated that an individual of P. soleni- ids studied.”

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS 331

position of umbo

0.2 0.4 0.6

streamlining
FIG. 16. Streamlining vs. position of umbo. See Fig. 11 for details.
Members of the Papyridea lineage are in the intermediate phase. Most bivalves are ei-

the process of colonizing the deeper infaunal ther bottlenecked behind this position (includ-
habitat. It is one of the few modern groups in ing most of the members of the Cardiidae), or

332 WATTERS

depth of sinus

streamlining

FIG. 17. Streamlining vs. depth of sinus. See Fig. 11 for details.

have advanced into the permanent gapage Members of Papyridea stand out from the
phase (members of the solenids, cultellids, few groups in the same level of transition be-
and solecurtines). cause of their high degree of modification of

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS 333

pre-existing shell characteristics. The central
position of the umbo, the short central liga-
ment, and the simple lateral teeth all are pre-
requisite to enter the intermediate phase. It
must be emphasized that entry into this phase
depends upon the chance alignment of sev-
eral shell characteristics, therefore the great
number of shallow infaunal species bottle-
necked behind this morphological barrier.

Veneracea (Veneridae and Petricolidae).
The true, or venus, clams comprise the largest
single family of living bivalves. Ansell (1961)
categorized individuals of this family as soft
substrate-dwelling with few burrowing modifi-
cations. They successfully have exploited the
shallow infaunal zone with little invasion of the
deeper infaunal zone. None have achieved a
streamlining coefficient greater than 0.9 or a
permanent gapage of greater than 0.15. None
have entered the intermediate phase. This is
because venerids have not achieved the suite
of characteristics necessary to enter that part
of the sequence. Ansell (1961: 514) remarked
that “[in members of the genus Petricola], well
developed hinge teeth and the long ligament
make rocking movements of the shell valves
... impossible.” Yet the members of the group
have already begun to diverge along the
streamlining/relative position of the umbo
paths (Fig. 16). Members of the Meretricinae
are tending toward a more central umbonal
position. Individuals of the Tapetinae and
some elements of the Pitarinae (forms in Mac-
rocallista) and the Chioninae (members of
Protothaca) are on the path toward an ante-
riorly positioned umbo.

Mactridae. The surf clams encompass
more morphological forms than any other
family in this study. The group contains ven-
erid-like shallow infaunal forms as well as
deep infaunal dwelling individuals reminiscent
of some members of the solenaceans. Stan-
ley (1972, 1977a) has pointed out the conver-
gence in morphology of mactrids with that of
individuals of such other families as the My-
idae, Veneridae, and Tellinidae.

A singular shell design is prevalent in this
family and has been modified for the interme-
diate phase. The ligament has been partially
internalized and positioned beneath the umbo
in a resilifer, where it serves as a fulcrum dur-
ing rocking as well as providing the opening
moment of the valves (Yonge, 1982). The re-
sult is a central ligament independent of
streamlining (Seilacher, 1984) and offering lit-
tle resistance to exchangeable gapage.

Two paths may be taken out of the inter-

mediate phase. Members of four genera have
entered the intermediate phase and/or ex-
ceeded it into the area of permanent gapage.
As in the Cardiidae, the species within each
genus are very few. These groups are mem-
bers of the lutrariinine genera Lutraria and
Psammophila, both of European seas, and the
Indo-Pacific zenatiinine genera Zenatia and
Resania [Beu (1966) places the latter in its
own subfamily, the Resaniinae]. Members of
Resania tend toward the path to a centrally
located umbo. Members of the other three all
lie on a path toward an anterior umbo (Fig. 13).
For the relative depth of the sinus, members of
Lutraria and Psammophila are tending toward
a deep sinus, whereas those of Resania and
Zenatia are approaching a very shallow sinus
reminiscent of that found in members of the
solenaceans. For exchangeable gapage, in-
dividuals of Psammophilia are on the path of
the myids, whereas the members of the re-
maining three genera are on the solenacean
path.

Individuals of Lutraria and Tresus have a
reduced foot as adults (Yonge & Allen, 1985),
indicative of diminished burrowing ability.
Members of Tresus may live at substrate
depths of 50 cm, where they are sedentary as
adults (Yonge, 1982). Cotton (1961: 297)
gave this account of an individual of Lutraria
rhynchaena Jonas, 1844, a species in inter-
mediate phase (note the modifications for ex-
changeable gapage):

[It] burrows deeply in sandy mud . . . siphons reach-
ing upwards to the surface. . . . The short ligament
allows considerable movement at the ends without
opening the shell throughout. With the valves in
their ordinary positions the shell gapes equally at
each end, but the arrangement of teeth and liga-
ment is such that the front of the shell may be en-
tirely closed.

That members of Lutraria lie on the solen-
acean path is not surprising. Beu (1966) de-
scribed their life habits as tube dwelling in the
manner of individuals of Solen.

Beu (1966) also noted the exchangeable
gapage of members of Resania and Zenatia.
He believed the former to be an active bur-
rower in sand in the wave zone, and the latter
to be a sedentary burrower offshore.

Lineages of the mactrids are evolving (in
the sense of the variables studied here) in
diverse directions, more so than any other
family covered in this study. The family has
members in all possible paths and in all three
morphological phases.

334 WATTERS

Tellinacea (Tellinidae and Semelidae). The
tellins and semelids are large groups of ac-
tive, streamlined, shallow to moderate depth
burrowing bivalves. Most are unsculptured,
and the few groups that are (some members
of Scutarcopagia and Strigilla, for example)
have composite burrowing sculptures. They
are within the intermediate phase and are on
the path of the myids. They have extensive
siphons and a pronounced sinus, also a cen-
tral umbo, and the shell of many forms has
some degree of exchangeable gapage. Mem-
bers of a few species can burrow to moderate
depths (Hughes, 1969).

Yonge (1949) believed that forms of the
Tellinidae, Solecurtinae, and Donacidae were
derived independently from members of the
Psammobiinae resembling individuals of
Gari. Pohlo (1982) offered a different phylog-
eny, making members of the Tellinidae the
end of the sequence Donacidae — Solecur-
tinae — Psammobiinae — Tellinidae. The
present study does not support this conten-
tion, and suggests a phylogeny more similar
to that of Yonge. Members of the donacids
may be an offshoot of the tellins specialized to
the high-energy environment of the sandy in-
tertidal zone.

Most, if not all, tellins, also some forms of
the psammobiids, have a unique “X”-shaped
muscle, the cruciform muscle, connecting the
ventral margins of the shells. Yonge (1949)
noted that this muscle occurs at the ventral
base of the siphonal attachment and believed
that it functioned to anchor the siphons at this
margin during protraction and retraction. This
muscle group also could serve as a ventral
connection during a rocking motion, limiting
the ventral pivot to a specific point. This dif-
fers from the dynamic ventral pivot of most
other groups in the intermediate phase.

Psammobiinae and Sanguinolariinae
(Psammobiidae). Members of these subfam-
ilies are the morphological precursors of the
solecurtine psammobiids, and occupy the in-
termediate phase for this family. They are
morphologically the analog of the tellins. But
unlike them, members of the Psammobiidae
have a permanent gapage group, the Sole-
curtinae. Members of the family lie upon the
myid path.

Solecurtinae (Psammobiidae). Individuals
of this subfamily are a fairly small group that
resemble the razor clams in many shell char-
acteristics. Members of the Solecurtinae, ex-
cept forms of Tagelus, do not construct tube-
like burrows, and have extensive siphons

(and deep sinuses). The members of Tagelus
are similar ecologically and behaviorally to
those of the solenaceans (Stanley, 1970).
They occupy many of the same paths as that
group. The major difference is the position of
the umbos, which are central in members of
Tagelus and anterior in solenaceans. Other
groups of solecurtines are on different mor-
phological paths.

Solenaceans. The razor clams have di-
verged from most infaunal bivalves in behav-
ior and habitat. They construct tube burrows
in which they move horizontally. This habit
has produced a distinct alternative path out of
the intermediate phase. Siphons and sinus
may be greatly reduced because the animal
may dwell at the surface, becoming deep in-
faunal in the sense of this study only to avoid
danger. Because they can retreat into the
deep substrate, permanent gapage is avail-
able. As tube dwellers, the highly streamlined
shape is retained at maximum permanent
gapage. This combination of characteristics
has led to two paths out of the post-interme-
diate phase morphologies. Yonge (1951c:
429) recognized the important principle that
shell and anatomy are separate entities:
“There is the fundamental, though largely un-
recognized, fact that throughout the Mollusca
the growth of the body and the growth of the
shell must be considered separately.”

Myidae. The myids are few in species num-
ber but quite variable in morphology and ecol-
ogy. Members of the genus Cryptomya live at
depths of up to 50 cm, have only short si-
phons, and “tap” into the water filled cavities
of burrowing crustaceans and echinoderms
(Yonge, 1951a). Members of Platyodon bore
into soft stone (Yonge, 1951b). These spe-
cializations aside, the members of the genus
Mya illustrate the expected result of the mod-
eled path. All exchangeable gapage has been
modified into permanent gapage, streamlin-
ing is reduced, teeth are non-functional, and
the sinus is shallow as the siphons become
increasingly non-retractable. Like forms in the
Mactridae, the myids have a central, internal-
ized ligament carried within a resilifer (Yonge,
1982). Analogs in the Hiatellidae (not in-
cluded in this study), are individuals of Pano-
pea, the geoduck clams.

Order Unionoida. Members of the four fam-
ilies of the freshwater unionoids participate in
few of the paths discussed here. This seems
attributable to their lack of fused mantle tis-
sue, necessary to form siphons. Without si-
phons, deep burrowing is not obtainable un-

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS 335

less tubes are constructed, as in the
solenaceans, a behavior unknown in mem-
bers of the Unionoida. Although the members
of the unionoids achieve a high level of
streamlining, this type of shell form appears to
function in quick reburial rather than in effi-
cient movement while buried (Watters, in
prep). Individuals of the unionoids lie upon the
solenacean path rather than upon the path of
the other groups studied for streamlining and
the relative position of the umbo. This is not to
imply that unionoids are following the solena-
ceans in morphology. Individuals of unionoids
have no true siphons (with the possible ex-
ception of members of Leila), usually cannot
burrow far below the substrate/water level,
and do not construct burrows.

Pholadacea. Although not used in this
study, the shipworms and relatives briefly are
discussed here because of their novel use of
exchangeable gapage. The antero-posterior
rocking motion of the shells is used not only to
protrude foot and siphons, but as a mechan-
ical rasping device to excavate burrows in
wood, shell, and stone. The shell and muscu-
lature have been reorganized to maximize
this movement. These innovations have been
discussed by Röder (1977) and Hoagland &
Turner (1981). A recent study (Fuller & Cast-
agna, 1989) also documents the complicated
ontogeny of individuals of one species of this

group.
Underlying Assumptions and Paradigms

The fundamental assumption of this study
is that there is a definite selective advantage
to becoming deep infaunal. The underlying
question, then, is why aren't there more deep
infaunal bivalves? The reason is related to the
possible ways that a bivalve shell can be
modified for this habitat. These modifications
require a particular suite of characteristics,
and only bivalves having this prerequisite
suite can colonize the deep infaunal region. If
the morphology of the lineage cannot be mod-
ified, that group cannot succeed in that habi-
tat. Entry into this sequence would be rare if
there was little or no adaptive significance to
the lineage possessing the suite, or if another
suite had high selective value. In the former
case, the acquisition of the suite would de-
pend on random fluctuations in the character-
istics of the morphology. In the latter, there
may be no impetus to move from one adap-
tive peak to another. A paucicity of deep-

dwelling forms would be the expected result if
either of these factors occurred in the evolu-
tion of the bivalves. Convergence also would
be the expected result if only a few viable
sequences of morphologies were available.

These constraints are due in part to the in-
teractions between sediment and shell with
increasing depth of burial. For simplicity, | will
consider the substrate to be homogeneous.
The addition of heterogeneous and stratified
sediment variables, while a much more real-
istic. scenario, cannot adequately be ac-
counted for in this model. It is suggested that
the simpler model may be extrapolated to the
more complex.

The mechanics of burrowing in shallow in-
faunal bivalves have been documented by
Trueman (1966), Trueman et al. (1966a), and
Stanley (1970, 1975). However, the members
of all groups studied, such as Mercenaria
mercenaria (Linnaeus, 1758) in Stanley
(1975), have low S values, no exchangeable
gape, and no permanent gape. The steps in
burrowing in such forms may be given briefly:

(1) The foot probes the substrate.

(2) The siphons are closed.

(3) Adductor muscles close the valves, rais-
ing pressure in the haemocoel, which is trans-
ferred to the foot, forming an anchor.

(4) Simultaneously, water is ejected from
the mantle cavity, which momentarily loosens
the immediately surrounding substrate.

(5) The anterior pedal retractor contracts,
pulling the animal forward against the an-
chored foot.

(6) The posterior pedal retractor contracts,
returning the shell to the original burrowing
position.

(7) The adductor muscles relax, diminish-
ing haemocoel pressure and redirecting fluid
out of the anchored foot. The siphons are
opened.

This process continues until the animal is
buried. Other factors also may be involved.
Sculpture may assist burrowing, as may the
presence of a prosogyre shape and a lunule
(Stanley, 1969, 1975, 1981). But the focus of
this study is deep-dwelling bivalves. The bur-
rowing model given above may work for only
a few of the groups in this study. The rocking
motion around a dorso-ventral axis becomes
impossible to accomplish as shells become
more elongate (S increasing; Stanley, 1970).
The foot must protrude from the anterior gape
and is often as large in cross-section as the
shell in streamlined forms. It appears, by its
larger size, to be much stronger than the foot

336 WATTERS

of shallow infaunal burrowers of the same
shell size. Eagar (1978) reported that the
force of the pedal retractors may be equal to
100 times the weight of the shell in water in
individuals of deep dwelling Ensis, but equal
to only one-quarter the weight in members of
shallow infaunal Mercenaria. These factors
may be necessary in these groups to offset
the lack of burrowing assistance that is found
in shallow-dwelling forms afforded by the bur-
rowing movement, shell sculpture, and lunule.
Expulsion of water to loosen sediment ap-
pears still to be important. Many deep-dwell-
ing forms have ventrally fused mantle tissue
that presumably directs water forward during
a burrowing cycle.

The ability to enter efficiently the substrate
is a function of shell shape. Nair & Ansell
(1968) found that elongate shells offer the
least resistance to burrowing. In this study,
the design most suited to burrowing is found
in the entity having the highest S value, all
other factors being equal. This often takes the
form of a laterally compressed, antero-poste-
rior elongated blade-like shape. Sculpture
typically is lost, and Stanley (1970) has
shown that coarse-sculptured species are
slow burrowers. In members of a species that
have both infaunal and epifaunal individuals,
the infaunal morphs are more elongate
(Seed, 1980a). Within the same genus, deep-
burrowing members are more streamlined
than are shallow-burrowing ones (Alexander,
1974; Eagar, 1974), although Agrell (1949)
made a correlation between shell morphology
and the trophic level of the water body.

The sediment load pressure increases with
increasing depth of burial (Nair & Ansell,
1968). The animal must exert a force to open
and maintain open the shells (Stanley, 1970).
In bivalves this is accomplished typically by
the ligament and/or haemocoel pressure. The
shells must be opened to allow protrusion of
the foot and siphons. Trueman et al. (1966a,
b) have shown that the sediment pressure
may exceed the opening moment of the liga-
ment at critical depths, effectively limiting
burial depth. One solution to this problem is
the incorporation of permanent shell gapes
into the morphology. The foot and siphons
may be protruded through these openings or
permanently left exposed. But the primary
function of the shell is defense, and therefore
the vast majority of epifaunal or shallow infau-
nal forms have complete closure of the
valves. But a selective advantage is to be
gained by penetrating the substrate further,

including a concomitant decrease in preda-
tion and an increase in habitat stability.

A solution to this problem requires having
the shells retain their function as protective
devices, while allowing the foot and siphons to
protrude in a manner independent of the lig-
amental opening moment. Such a suite of
characteristics does exist, and apparently rep-
resents the only compromise found in living
bivalves. | have termed this unique morphol-
ogy the intermediate phase, between the shal-
low and deep infaunal existence phases. It has
a suite of predictable and testable character-
istics that may be compared with actual forms.

The key innovation is exchangeable gapage
(Fig. 1). The shells rotate along a dynamic
dorso-ventral axis rather than along the dorsal
hinge axis. Movement is effected by the ad-
ductor muscles rather than by the weaker lig-
ament or haemocoel pressure. Contraction of
the anterior adductor muscle closes the ante-
rior (pedal) gape and opens the posterior (si-
phonal) gape. Contraction of the posterior ad-
ductor muscle has the opposite effect. Several
important morphological requirements must
be met for this mechanism to work.

First, the umbo must be approximately cen-
tral. This orientation allows the maximum
amount of exchangeable gapage at both
ends. Second, the ligament also must be cen-
tral and reduced. A long opisthodetic ligament
would not allow rocking along a dorso-ventral
axis. Third, cardinal teeth must be retained to
act as the dorsal pivot of the axis. Lateral
teeth may or may not be present, but if
present, they must be able to disengage
smoothly as the rocking movement takes
place. Forth, the valve commissure must be
open anteriorly and posteriorly, creating a
gape when the shells are rocked.

This morphology may have an adverse side
effect. The simultaneous contraction of the
adductor muscles may split the valves at
the umbo along a line of structural weak-
ness if the shell is sufficiently thin. This is
known to happen in all members of the
anomalodesmacean genus Laternula and
some Periploma (Morton, 1976). Individuals
of other species, all within or past the inter-
mediate phase, may have an internal rib or
buttress at this position to counteract the
stress: Nuculites (Nuculidae); Capistrocardia
(Saxicavidae); Cleidophorus (Ledidae); Sili-
qua, Cultellus, and Phaxus (Solenacea);
Sanguinolaria, Nuttallia, Solecurtus, and
Tagelus (Psammobiidae); among others (Gill
& Darragh, 1964, and this study). In other

FUNCTIONAL MORPHOLOGY OF BIVALVE SHELLS

FIG. 18. Opening moment of movement around
hinge axis (ha). 1: Anterior torque arm. 2: Posterior
torque arm. Magnitudes of torque arms do not
change during movement.

species, additional
present.

The presence and the position of these but-
tresses are not simply the result of adductor
muscles contracting within a shell with ante-
rior and posterior gapes during normal clo-
sure (around the hinge axis). Factors influ-
encing the disposition of internal buttresses
are tied to the mechanics of exchangeable
gapage. In most shells, the valves rotate
along an axis determined by the hinge line,
particularly the line through the ligament. The
insertions of the adductor muscles on the
valves remain the same distance from that
axis throughout contraction and the adductor
muscles work in concert (Fig. 18). The situa-
tion is different during the process of ex-
changeable gapage. The dorsal pivot of the
axis is anchored, usually by the cardinal
teeth. But the ventral pivot moves along the
ventral margin of the shell, sweeping out an
angle defined by the anterior-and posterior-
most positions of the axis (Fig. 19). The dis-
tance from the adductor muscles to this dy-
namic axis changes in a linear fashion during
this rocking motion. The adductor muscles
are antagonistic during this motion.

Thomas (1975) estimated the amount of
force generated during valve closure, the ad-
ductor moment, by:

buttresses may be

(cross-sectional area of adductor)
х (distance to axis) (6)

The cross-sectional area is an estimate of
force. The distance to the axis represents the
torque arm. In his calculations, which involved
no exchangeable gapage, the adductor mo-
ments are constant during closure. The mo-

337

FIG. 19. Opening moment of movement around dy-
namic dorso-ventral axis. x: Fixed pivot at cardinal
teeth. 1: Posterior torque arm at minimum posterior
closure with axis along ab. 2: Posterior torque arm
at maximum posterior closure with axis along cd.
Magnitude of torque arm changes during move-
ment. Anterior torque arm would behave in the op-
posite manner.

ments during exchangeable gapage are not
(Fig. 20). The lines of adductor moments may
or may not cross, depending on the location
of the adductor muscles and the shape of the
shell. If the shell is thin, a buttress generally
will occur near the angle at which the mo-
ments are equal. This angle represents the
point during an exchangeable gapage rocking
motion that the anterior and posterior adduc-
tor forces are equal, thereby placing maxi-
mum strain on the shell between them if they
are contracted simultaneously (Fig. 21). The
buttress reinforces this region. Buttresses
also may occur at the beginning and end of
the exchangeable gapage angle. These may
counteract the forces generated by the ad-
ductor muscles attempting to contract beyond
the limit of the allowable angle. The central
buttress may be placed at the bisection of the
angle, but other evidence suggests that it is
dependent on the point of equal moments.
For the individual in Figure 22, the lines do not
cross and the central buttress is absent, al-
though the two flanking ones limiting the an-
gle are prominent. Figure 23 illustrates the
moment lines for a form in which the lines
cross only at the end of the angle. The forma-
tion of internal buttresses is a modification for
forces generated on the shell by the adductor
muscles during exchangeable gapage.

Past the intermediate phase, the deeply bur-
ied bivalve may take on equally predictable
characteristics. Movement within the sub-

338 WATTERS

0 10 20 30 40 50

5
5
$
S
>
o
3
E

3000

0 30 60 90 120

6000

degrees

FIG. 20. Anterior (aam) and posterior (pam) adductor moments for Tresus nuttali (Conrad, 1837) through

entire angle of exchangeable gape.

FIG. 21. Anterior (aam) and posterior (pam) adductor moments for Tagelus divisus (Spengler, 1794), through
entire angle of exchangeable gape. Dotted lines indicate angles at which buttresses are positioned.
FIG. 22. Anterior (aam) and posterior (pam) adductor moments for Resania lanceolata Gray, 1862, through
entire angle of exchangeable gape. Dotted lines indicate angles at which buttresses are positioned.
FIG. 23. Anterior (aam) and posterior (pam) adductor moments for Siliqua patula (Dixon, 1789), through
entire angle of exchangeable gape. Dotted lines indicate angles at which buttresses are positioned.

strate is minimized as exchangeable gapage
is modified into less streamlined permanent
gapage. Shell shape may return to a non-
streamlined form reminiscent of the shallow
infaunal stage. Sculpture, lost in the transition,
remains absent as the substrate becomes the
primary protective device (Stanley, 1970).
Shell thickness, also originally protective, may
be minimized (Stanley, 1970; Morton, 1976).
The teeth, reduced or weakly meshed in the
intermediate phase, may become rudimentary
as all shell/shell movement is lost (both along
the horizontal hinge line and along the dy-
namic hinge of exchangeable gapage). The
siphons may become partially or wholly non-
retractable, resulting т a decrease of the sinus
depth. Members of some species have been

shown to possess an atrophied foot as an
adult, suggesting a sedentary habit. Individu-
als of Panopea abrupta (Conrad, 1855), a hi-
atellid, may live immotile in burrows 90 cm
deep (Yonge, 1949).

Evolutionary Considerations

Most forms studied are uniform for the cal-
culated parameters. The position of the umbo
is distributed about a mode of 0.3. The depth
of the sinus is generally less than 0.1 (reflect-
ing the large numbers of members of the
Unionoida in the study). Streamlining is quite
high, with a mode of 0.9, indicating that most
bivalves, even shallow infaunal ones, are
somewhat streamlined. But high levels of ex-

FUNCTIONAL MORPHOLOGY ОЕ BIVALVE SHELLS 339

changeable gapage and permanent gapage
are rare. This suggests that most forms are
still in the streamlining phase of the se-
quence. Few have made the transition to the
intermediate phase. Why is this the case?

To enter the intermediate phase requires a
specific set of shell characteristics. The umbo
and cardinal teeth must be central, the laterals
must be able to disengage, and the ligament
must be short and central. Presumably, this
suite of morphological characteristics is not
met in most bivalves. This has resulted in a
bottleneck at the intermediate phase. Species
occurring before this stage are numerous. It is
hypothesized here that the acquisition of the
necessary combination of characteristics
needed to continue in the sequence may be
determined by chance. Like billiard balls
thrown at random on the table, one may drop
in the pocket, but most continue rolling.

Once in the intermediate phase, morpho-
logical change may be rapid. The change
from intermediate phase to exchangeable
gapage phase may be brief on a geological
time scale. Radiation usually is rapid after a
morphological or ecological innovation (Hoag-
land & Turner, 1981). Of the several hundred
species of Mactridae, members of fewer than
a dozen are in the intermediate phase, and
the percentage is less for forms in the Cardi-
idae. Although members of the Mactridae
have been in existence since at least the late
Cretaceous, the groups now in the intermedi-
ate phase are no older than the Miocene. But
within that small group, speciation may be
high. Beu (1966) has recognized three dis-
tinct lineages within the members of the ge-
nus Zenatia.

Geary (1987) found that slow rates of
change in the lineage of species of Pleurocar-
dia are punctuated with quick major changes.
Stanley (1977a) and Stanley & Yang (1987)
also found low levels of phyletic change in
members of the Veneridae and Tellinidae, two
families with members still predominantly in
the streamlining phase. The bottlenecking of
morphologies has created a steady, but low
rate of evolution in these taxa. Even so, as
stated by Stanley (1979: 118), “there is no
evidence that a limit [to diversity] is being
approached even after more than 400 My of
radiation.” But the acquisition of the interme-
diate phase must be seen as a major mor-
phological step opening a new area of the
morphospace.

Within and after the intermediate phase,
members of lineages would be expected to

radiate to fill the new morphospace. As an
example, the Anomalodesmata is a large, di-
verse group, with many of its members tending
toward deep-dwelling, sedentary habits (Mor-
ton, 1977). The Solenacea also is a large
group of species, the members of most in the
permanent gapage phase. They are recogniz-
able as solenaceans as far back as the Cre-
taceous, suggesting that they had passed
through the intermediate phase prior to that
time. Most of the basic adaptive radiation of
the Bivalvia had occurred by the Cretaceous
(Nicol, 1986), though 96% of the species, and
52% of the families became extinct during the
Permo-Triassic extinction (Raup, 1979). This
suggests that the sequence of morphologies
discussed here is an ongoing process, taking
place asynchronously in different lineages as
the necessary morphological prerequisites are
obtained.

No clades have been defined in this study
of Recent species. The phylogeny of most bi-
valves is too insufficiently known to allow the
concepts developed here to be tested by the
fossil record. If the sequences of shell shape
change are reversible, then the precursors of
modern groups may have assumed a wide
variety of forms. While some obvious trends
within clades exist, such as those culminating
in Papyridea, others are too ambiguous. The
trends in shell shape described here are
trends between clades acting simultaneously
on unrelated taxa.

15 the evolution of these groups predict-
able? To a certain extent the answer may be
yes. Н continued studies show that other
groups of bivalves lie along these paths, then
we may assume that bivalve lineages enter-
ing a path may evolve toward the shell
shapes of individuals already on the path. The
great degree of convergence in bivalves sup-
ports this hypothesis. Several groups, such
as the mactrids and venerids, have members
in both the myid and solenacean paths. Mem-
bers of Resania look remarkably like those in
its solenacean counterpart, Phaxus. They oc-
cupy the same place in the path. Will there
eventually be a mactrid counterpart to Solen?
Members of Lutraria already have adopted
the tube dwelling habit of that genus.

SUMMARY

А hypothesis is advanced to explain: (1) the
changes in shell shape in individuals of spe-

340 WATTERS

cies as a continuously deeper infaunal habitat
is colonized; and (2) the degree of conver-
gence in shell shapes among infaunal bi-
valves. À maximum depth of burrowing for
streamlined morphologies will be reached as
sediment weight becomes significant. Up to
this point, forms will adopt streamlined
shapes for more efficient penetration of and
movement in the substrate.

To achieve a deeper infaunal existence re-
quires that the shell possess gapes through
which the foot and siphons may extend. This
would make the animal susceptible to preda-
tion and other immediate environmental dan-
gers because the shell functions as the main
defensive mechanism. Only one morphologi-
cal “solution” has been adopted by the bi-
valves. This entails the antero-posterior rock-
ing of the shell such that a реда! or siphonal
gape alternately may be opened and closed.
Because this action is caused by the adductor
muscles, rather than by the much weaker lig-
amental or haemocoel opening mechanisms,
the probiem of sediment weight has been by-
passed at this depth. The acquisition of ex-
changeable gapage requires several pre-
existing morphological conditions. These
conditions must be modified to new functions
in this stage of development, termed here the
intermediate phase.

The cardinal hinge teeth must still function
as a dorsal pivot, but on a dorso-ventral axis.
These teeth must be located centrally to max-
imize exchangeable gapage. The laterals
must be able to disengage (or no movement
along that axis could take place). The hinge
must be centralized to avoid interference with
the rocking motion of the shells. This may be
accomplished by a shortening of the ligament
or the internalizing of it in a resilifer ventral to
the umbo.

Movement into a deeper infaunal position
may be possible once the intermediate phase
is reached. This entails a further decline in
predation and environmental extremes. At
this point, exchangeable gapage may be
modified into permanent gapage. The animal
may be sedentary, with a reduced foot and
externalized siphons. Shell thickness may de-
crease as the result of the reduced depen-
dency on the shell for defense.

Comparisons between these models and
the actual shell shapes of the individuals
of the species studied show a general agree-
ment. The morphologies are found in the
predicted morphospace. The hypothetical
suite of specialized characteristics does occur

in real species in the intermediate phase.
Members of lineages follow a specific path, a
sequence of body shapes, as they increas-
ingly become infaunal. This results in un-
related species sharing the same general
morphological pattern because they are at
the same point on this path. The constraints
of this sequence are such that some paths
may move in both directions, whereas in oth-
ers a separate course may exist for each di-
rection.

Two paths occur out of the intermediate
phase, termed here the solenacean and the
myid paths after the typical member of each
route. The solenacean path differs because of
the behavior of its members, which construct
tube burrows, allowing the shell to retain its
streamlining along with exchangeable gap-
age. The unionoids appear to lie on this path
but the convergence is superficial. The тет-
bers of that group lack the fused mantle tissue
necessary to form true siphons.

That so few forms exist in the intermediate
phase or in the exchangeable gapage phase
supports the idea that the specific suite of
shell characteristics necessary to enter the in-
termediate phase has not been attained by
most groups. Shallow infaunal species,
though high in diversity, are bottlenecked at
this point. The entry into the intermediate
phase may allow a new morphological radia-
tion. This passage may be quick in geological
time and be largely the product of chance.

ACKNOWLEDGEMENTS

| would like to thank Dr. Ruth Turner, Mu-
seum of Comparative Zoology, Harvard Uni-
versity, and the late Dr. Joseph Rosewater,
National Museum of Natural History, Wash-
ington, D. C., for allowing me to examine the
collections under their care. This study was
conducted as part of the requirements for the
degree of Doctor of Philosophy at the Ohio
State University. | would like to thank my com-
mittee for their support and guidance: Dr. Ab-
bot Gaunt, Dr. David Stansbery, Dr. Barry
Valentine (all Department of Zoology), and Dr.
Walter Sweet (Department of Geology).
Funding for portions of this study were pro-
vided by a scholarship from the National Cap-
ital Shell Club of Washington, D. C. The
manuscript was improved by the valuable
comments of two anonymous reviewers, Dr.
George Davis, and Dr. Eugene Coan.

FUNCTIONAL MORPHOLOGY ОЕ BIVALVE SHELLS 341

LITERATURE CITED

AGRELL, 1., 1949, The shell morphology of some
Swedish unionides as affected by ecological con-
ditions. Arkiv Юг Zoologi, 41(15): 1-30.

ALEXANDER, В. R., 1974, Morphological adapta-
tions of the bivalve Anadara from the Pliocene of
the Kettleman Hills, California. Journal of Pale-
ontology, 48: 633-651.

ALLEN, J. A., 1958, On the basic form and adap-
tations to habitat т the Lucinacea (Eulamellibran-
chiata). Philosophical Transactions of the Royal
Society of London, Series В, 241: 421—484.

ANSELL, А. D., 1961, The functional morphology of
the British species of Veneracea (Eulamellibran-
chiata). Journal of the Marine Biological Associ-
ation of the United Kingdom, 41: 489-517.

BAYER, U., 1978, Morphogenetic programs, insta-
bilities, and evolution—a theoretical study.
Neues Jahrbuch für Geologie und Paläontologie,
Abhandlungen 156: 226-261.

BEU, A. G., 1966, The molluscan genus Lutraria,
Resania and Zenatia in New Zealand. Transac-
tions of the Royal Society of New Zealand, Zool-
оду, 8: 63-91.

BOOKSTEIN, F. L., 1987, Random walk and the
existence of evolutionary rates. Paleobiology, 13:
446-464.

CARTER, R. M., 1968, On the biology and palae-
ontology of some predators of bivalved Mollusca.
Palaeogeography, Palaeoclimatology and Palae-
oecology, 4: 29-65.

COTTON, B. C., 1961, South Australian Mollusca.
Pelecypoda. South Australian Branch of the Brit-
ish Science Guild. 363 pp.

DREW, G. A., 1907, The habits and movements of
the razor-shell claın, Ensis directus, Con. Biolog-
ical Bulletin, 12: 127-140.

DREW, G. A., 1908, The physiology of the nervous
system of the razor-shell clam (Ensis directus,
Con.). Journal of Experimental Zoology, 5: 311-
326.

EAGAR, В. М. C., 1974, Shape of shell of Carboni-
cola in relation to burrowing. Lethaia, 7: 219-
238.

EAGAR, R.M. C., 1978, Shape and function of the
shell: a comparison of some living and fossil bi-
valve molluscs. Biological Reviews of the Cam-
bridge Philosophical Society, 53: 169-210.

FOWLER, D. R., H. MEINHARDT & P. PRUSINK-
IEWICZ, (in press), Modeling seashells. Pro-
ceedings of SIGGRAPH for 1992.

FULLER, S. C. & M. CASTAGNA, 1989, Shell and
pallet morphology in early developmental stages
of Teredo navalis Linne (Bivalvia: Teredinidae).
Nautilus, 103: 24-35.

GEARY, D. H., 1987, Evolutionary tempo and mode
in a sequence of the Upper Cretaceous bivalve
Pleurocardia. Paleobiology, 13: 140-151.

GILL, E. D. & T. A. DARRAGH, 1964, Evolution of
the Zenatiinae (Mactridae: Lamellibranchiata).
Proceedings of the Royal Society of Victoria, 77:
177-190.

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

HOAGLAND, К. E. 4 В. О. TURNER, 1981, Evo-
lution and adaptive radiation of wood-boring bi-
valves (Pholadacea). Malacologia, 21: 111-148.

HUGHES, R. N., 1969, A study in feeding in Scro-
bicularia plana. Journal of the Marine Biological
Association of the United Kingdom, 49: 805-823.

ILLERT, C., 1992, Foundations of theoretical con-
chology. Hadronic Press, Inc., Palm Harbor, Flor-
ida. 104 pp.

JELL, P. A., 1980, Earliest known pelecypod on
Earth—a new Early Cambrian genus from South
Australia. Alcheringa, 4: 233-239.

MASSARE, J. A., 1988, Swimming capabilities of
Mesozoic marine reptiles: implications for
method of predation. Paleobiology, 14: 187-205.

MORTON, B., 1976, The structure, mode of oper-
ation and variation in form of the shell of the Lat-
ernulidae (Bivalvia: Anomalodesmata: Pandora-
cea). Journal of Molluscan Studies, 42: 261-278.

MORTON, B., 1977, Some aspects of the biology
and functional morphology of Myadora striata
(Quoy 4 Gaimard) (Bivalvia: Anomalodesmata:
Pandoracea). Journal of Molluscan Studies, 43:
141-154.

NAIR, N. B. 8 A. D. ANSELL, 1968, Characteristics
of penetration of the substratum by some marine
bivalve molluscs. Proceedings of the Malacolog-
ical Society of London, 38: 179-197.

NAUMANN, C. F., 1840, Winkelmessungen an Ker-
nen versteinerter Schnecken. Neues Jahrbuch
für Mineralogie, Geognosie, Geologie und Petre-
faktenkunde, Stuttgart, 1840: 462-463. [not
seen].

NEWELL, N. D., 1969, Superfamily Arcacea. Pp.
250-262, in: В. С. Moore, ed., Treatise on inver-
tebrate paleontology, N, 1, Mollusca 6. Geologi-
cal Society of America & University of Kansas.

NICOL, D., 1986, The fate of pelecypod families,
subfamilies, and tribes during and after the Cre-
taceous period. Nautilus, 100: 141-143.

POHLO, R., 1982, Evolution of the Tellinacea (Bi-
valvia). Journal of Molluscan Studies, 48: 245—
256.

POJETA, J., B. RUNNEGAR & J. KRIZ, 1973, For-
dilla troyensis Barrande: the oldest known pele-
cypod. Science, 180: 866-868.

RAUP, D. M., 1961, The geometry of coiling in gas-
tropods. Proceedings of the National Academy of
Sciences, 47: 602-609.

RAUP, D. M., 1962, Computer as aid in describing
form in gastropod shells. Science, 138: 150-152.

RAUP, D. M., 1963, Analysis of shell form in gas-
tropods. Special Papers of the Geological Soci-
ety of America, 73: 222.

RAUP, D. M., 1966, Geometric analysis of shell
coiling: general problems. Journal of Paleontol-
ogy, 40: 1178-1190.

RAUP, D. M., 1967, Geometric analysis of shell
coiling: coiling in ammonoids. Journal of Paleon-
tology, 41: 43-65.

342 WATTERS

RAUP, D. M., 1979, Size of the Permo-Triassic bot-
tleneck and its evolutionary implications. Sci-
ence, 206: 217-218.

RAUP, D. M. & А. MICHELSON, 1965, Theoretical
morphology of the coiled shell. Science, 147:
1294-1295.

RÓDER, H., 1977, Zur Beziehung zwischen Kon-
struktion und Substrat bei mechanisch bo-
hrenden Bohrmuscheln (Pholadidae, Teredi-
nidae). Senckenbergiana Maritima, 9: 105-212.

ROSEWATER, J., 1961, The family Pinnidae in the
Indo-Pacific. Indo-Pacific Mollusca, 1: 175-226.

RUDWICK, M. J. S., 1965, The inference of func-
tion from structure in fossils. British Journal for
the Philosophy of Science, 15: 27—40.

SAVAZZI, E., 1987, Geometric and functional con-
straints on bivalve shell morphology. Lethaia, 20:
293-306.

SEED, R., 1980a, A note on the relationship be-
tween shell shape and life habits in Geukensia
demissa and Brachiodontes exustus (Mollusca:
Bivalvia). Journal of Molluscan Studies, 46: 293—
299.

SEED, R., 1980b, Shell growth and form in the Bi-
valvia. Pp. 23-67, in: D. С. RHOADS & В. A. Lutz,
eds., Skeletal growth of aquatic organisms. Ple-
num Press, New York.

SEILACHER, A., 1984, Constructional morphology
of bivalves: evolutionary pathways in primary ver-
sus secondary soft-bottom dwellers. Paleontol-
ogy, 27: 207-237.

SOOT-RYEN, T., 1955, A report on the family Mytil-
idae (Pelecypoda). Allan Hancock Pacific Expe-
ditions, 20(1): 1-175.

SOOT-RYEN, T., 1969, Superfamily Mytilacea. Pp.
271-280, in: В. С. Moore, ed., Treatise on inver-
tebrate paleontology, N, 1, Mollusca 6. Geologi-
cal Society of America & University of Kansas.

STANLEY, 5. M., 1969, Bivalve mollusk burrowing
aided by discordant shell ornamentation. Sci-
ence, 166: 634—635.

STANLEY, S. M., 1970, Relation of shell form to life
habits in the Bivalvia. Memoirs of the Geological
Society of America, 125: 1-296.

STANLEY, S. M., 1972, Functional morphology and
evolution of byssally attached bivalve mollusks.
Journal of Paleontology, 46: 165-212.

STANLEY, S. M., 1973, An explanation for Cope’s
Rule. Evolution, 27: 1-26.

STANLEY, S. M., 1975, Why clams have the shape
they have: an experimental analysis of burrow-
ing. Paleobiology, 1: 48-58.

STANLEY, S. M., 1977a, Trends, rates, and pat-
terns of evolution in the Bivalvia. Pp. 209-250,
in: A. HALLAM, ed., Patterns of evolution as illus-
trated by the fossil record. Elsevier Scientific
Publ. Co., Amsterdam.

STANLEY, S. M., 1977b, Coadaptation in the Trig-
oniidae, a remarkable family of burrowing bi-
valves. Paleontology, 20: 869—899.

STANLEY, S. M., 1979, Macroevolution, pattern
and progress. W. H. Freeman & Co., San Fran-
cisco. xi +332 pp.

STANLEY, 5. M., 1981, Infaunal survival: alterna-
tive functions of shell ornamentation in the Bi-
valvia (Mollusca). Paleobiology, 7: 384—393.

STANLEY, S. M. & X. YANG, 1987, Approximate
evolutionary stasis for bivalve morphology over
millions of years: a multivariate, multilineage
study. Paleobiology, 13: 113-119.

THOMAS, R. D. K., 1975, Functional morphology,
ecology, and evolutionary conservatism in the
Glycymeridae (Bivalvia). Paleontology, 18: 217-
254.

THOMPSON, D’A. W., 1952, On growth and form.
Cambridge University Press (reprint 1971). 346

pp.

TRUEMAN, E. R., 1966, Bivalve mollusks: fluid dy-
namics of burrowing. Science, 152: 523-525.
TRUEMAN, E. R., А. А. BRAND & P. DAVIS,
1966a, The dynamics of burrowing of some com-
mon littoral bivalves. Journal of Experimental Bi-

ology, 44: 469-492.

TRUEMAN, E. R., A. В. BRAND 4 P. DAVIS,
1966b, The effect of substrata and shell shape on
the burrowing of some common bivalves. Pro-
ceedings of the Malacological Society of London,
37: 97-109.

VAUGHT, K. C., 1989, A classification of the living
Mollusca. American Malacologists Inc., Mel-
bourne, Florida. xii + 195 pp.

YOCHELSON, E. L., 1981, Fordilla troyensis Bar-
rande: “the oldest known pelecypod” may not be
a pelecypod. Journal of Paleontology, 55: 113—
125.

YONGE, C. M., 1949, On the structure and adap-
tations of the Tellinacea, deposit-feeding Eu-
lamellibranchia. Philosophical Transactions of
the Royal Society of London, Series B, 234: 29—
76.

YONGE, С. М., 1951a, Studies on Pacific coast
mollusks. |. On the structure and adaptations of
Cryptomya californica (Conrad). University of
California Publications in Zoology, 55: 395—400.

YONGE, С. М., 1951b, Studies on Pacific coast
mollusks. |. Structure and adaptations for rock
boring in Platyodon cancellatus (Conrad). Uni-
versity of California Publications in Zoology,
55(7): 401-408.

YONGE, С. М., 1951c, Studies on Pacific coast
mollusks. Ill. Observations on Siliqua раша
Dixon and on the evolution within the Solenidae.
University of California Publications in Zoology,
55(9): 421-438.

YONGE, C. M., 1982, Ligamental structure in Mac-
tracea and Myacea (Mollusca: Bivalvia). Journal
of the Marine Biological Association of the United
Kingdom, 62: 171-186.

YONGE, С. М. & J. A. ALLEN, 1985, On significant
criteria in establishment of superfamilies in the
Bivalvia: the creation of the superfamily Me-
sodesmatacea. Journal of Molluscan Studies,
51: 345-349.

Revised Ms. accepted 17 February 1993

MALACOLOGIA, 1993, 35(2): 343-349

А CLADISTIC REASSESSMENT ОЕ OCTOPODID CLASSIFICATION

Janet R. Voight

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

ABSTRACT

Octopodid classifications have been traditionally, and are currently, based on a few readily
apparent characters. In this analysis, | examine methods that have contributed to octopodid
classifications from a cladistic perspective that emphasizes the recognition of monophyletic
groups, and | apply parsimony algorithms to the data set reported by Voss (1988a) for the
Octopodidae. | reject current and previous subfamily classifications of the Octopodidae as
having created paraphyletic groups. Use of the category subfamily should be avoided, as it
implies our knowledge of octopodid evolution has reached level that is as yet unattained.

To further our knowledge of octopod phylogeny, we must define primitive and derived char-
acters states by objective criteria, consider only monophyletic species groups in our analyses
and expand the range of characters considered. Analysis of the data set compiled for cladistic
analysis reveals that characters of the radula, anterior digestive system and skin change in
concert. These associated character changes may indicate underlying functional relationships

that have been unsuspected.

Key words: Octopodidae, parsimony analysis, Graneledoninae, Eledoninae, Bathypolypodi-

nae, Octopodinae, systematics, radula.

INTRODUCTION

Taxonomic treatments intended to identify
astonishingly different, or to separate overtly
similar, specimens have produced the current
classification of coleoid cephalopods. This
scheme, similar to Naef’s (1923) reconstruc-
tion of ancestor-descendent relationships,
groups taxa based on morphological similarity,
with primitive characters contributing as much
as derived characters. That comparatively few
characters support subfamily groups in oc-
topodids have been cited as evidence of the
family’s chaotic evolution (Robson, 1932).
Whether these formally recognized morpho-
logically distinct groups constitute monophyl-
etic lineages that share a common evolution-
ary history is unknown.

Phylogenetic reconstruction through phy-
logenetic or cladistic analysis seeks to rec-
ognize only monophyletic groups. The
possession of shared derived characters (sy-
napomorphies) is the criterion on which mono-
phyletic groups are recognized. Neither
shared primitive characters (symplesiomor-
phies) nor character states unique to a single
taxon (autapomorphies) provide information
concerning relationships.

Cladistic analysis considers as many pre-
sumed synapomorphies as possible. Ho-
moplasy (whether due to parallelism, conver-
gence or reversal) affects some character
changes, but these are аззитеа to be fewer

343

than the character changes that reflect
unique modification with descent from a com-
mon ancestor. Cladistics uses the absolute
criterion of parsimony to evaluate alternate
hypotheses of relationships; parsimony dic-
tates that the hypothesized relationship that
requires the fewest number of character
changes is the most likely to reflect history.
In this paper, | test the extent to which oc-
topodid classification is supported by cladistic
analysis. | apply parsimony analysis to the
characters reported by Voss (1988a). My in-
tent is to introduce a cladistic perspective to
octopodid systematics, to examine implicit as-
sumptions that may have affected earlier
treatments of the group and to assess the in-
formation contained in traditional characters.

THE OCTOPODS

Among octopods, the bathypelagic taxa of
the suborder Cirrata are unified by the pres-
ence of fins, cirri and internal shells, all prim-
itive characters (Naef, 1923; Robson, 1932;
Voss, 1988a). Members of the suborder Incir-
rata, which occur throughout the water col-
umn and in benthic habitats, are united by the
absence of these characters, and by egg care
by the female, and associated characters
(Boletzky, 1992). Among the incirrates, male
reproductive characters and pelagic habitats

344

VOIGHT

TABLE 1. Octopodid classifications of Grimpe (1921, 1922), Robson (1932), Thiele (1934) and Voss
(1988a). Listed are the subfamilies and their diagnostic characters; in addition to these characters,
geographic and depth distribution are also cited in subfamily definitions.

Reference Subfamilies Sucker rows Ink sac Other characters
Grimpe Octopodinae 2 + small eggs
(1921, 1922) Eledoninae 1or2 + large eggs
Robson (1932) Octopodinae 1or2 Е typical
Bathypolypodinae 1or2 = reduced crop, gills, radula; large
eggs, spermatophores, squat
body; double funnel organ; narrow
mantle aperture
Thiele (1934) Octopodinae 1or2 + generally small eggs
Bathypolypodinae* 1or2 = reduced crop; large eggs &
spermatophores; short arms;
narrow mantle aperture
Ozaeninae (Eledoninae) 1 SH large eggs
Voss (1988a) Octopodinae 2 +
Bathypolypodinae 2 =
Eledoninae 1 +
Graneledoninae 1 =

*Including Benthoctopus and Teretoctopus, despite the large crop of Teretoctopus.

define membership in the argonauts; multi-
cuspid radular teeth and adaptation to the
mesopelagic zone define members of the
Ctenoglossa. The Octopodidae, with the most
recognized species, contains the benthic oc-
topuses. Prominent among the few charac-
ters that have contributed to octopodid clas-
sification (Table 1) are the number of sucker
rows and the presence or absence of an ink
sac.

Members of the Octopodidae range from
the intertidal zone to over 3500 m depth and
from the equator to the polar ice caps (Voss,
1988b). | follow taxonomic tradition in assum-
ing that the Octopodidae represent a mono-
phyletic group. Although Naef (1923) sug-
gested the pelagic Argonautida are derived
from Octopus s. s., | assume here that the
characters cited as uniting these groups (e.g.
double sucker rows, ink sac) are better attrib-
uted to convergences and symplesiomor-
phies than to synapomorphies (Robson,
1932; Voight, 1990).

Based on similarities in their radulae, the
monotypic taxon, Vitreledonella, has been
suggested to be an octopodid derived for the
mesopelagic habitat (Robson, 1932). Al-
though Vitreledonella lacks the multicuspi-
date radula that has defined the Ctenoglossa
(an apparent clade of the meso- and bathy-
pelagic octopods), this taxon and the cteno-
glossid Amphetritus share a rotated digestive

system unique in the Cephalopoda (Thore,
1949). | tentatively consider Vitreledonella to
be a ctenoglossid (Voight, 1990) and exclude
it from this analysis.

METHODS

Taxa that serve as the operational taxo-
nomic units (OTUs) in this analysis are oc-
topodid genera. Genera that Toll (1991) re-
cently revitalized are not included, pending
complete diagnoses. The characters Voss
(1988a) cited as diagnosing nonoctopodine
genera and his polarity assessments are
summarized on Table 2. For genera not in-
cluded by Voss (1988a), data were gathered
from specimens and literature accounts. Oc-
topodine genera other than Scaeurgus and
Pteroctopus (i.e., Robsonella, Hapaloch-
laena, Cistopus, Enteroctopus, Euaxocto-
pus), however, do not differ from Octopus in
the characters considered (Robson, 1929;
Roper & Hochberg, 1988; Hochberg et al.,
1992). These taxa were excluded, as autapo-
morphies cannot contribute to the analysis.

The data matrix (Appendix 1) was analyzed
by PAUP (Version 3.0) using subtree pruning-
regrafting and the MULPARS option (Swof-
ford, 1989). The specified ancestor (Appendix
1) served to root the analysis. Characters with
polarities defined by Voss (1988a; Table 2)

CLADISTIC REASSESSMENT OF OCTOPODID CLASSIFICATION 345

TABLE 2. Characters, character state definitions, and stated reasoning behind polarity definitions

(Voss, 1988а).

0 = ancestral character state; 1 = derived state.

1. Number of sucker rows: 0 = one; 1 = two (after Naef).

2. Ink sac: 0 = present; 1 = absent (known in fossil cephalopods).

3. Crop: 0 = with diverticulum; 1 = with dilation. (Loss of diverticulum is a modification to small prey.)
4. Posterior salivary glands: 0 = large; 1 = small; 2 = vestigial. (Large is normal in shallow-water

forms.)

5. Rachidian lateral cusps: 0 = present; 1 = absent (commonality).

6. Lateral tooth: 0 = present; 1 = absent (commonality).

7. Marginal plates: 0 = present; 1 = absent (commonality).

8, 9. Funnel organ: 00 = W-shaped; 01 = VV; 10 = Ш! (commonality).

10. Gill lamellae per demibranch: 0 = 9 or more; 1 = less than 9.

(Reduction assumed to be adaptive in the deep sea.)

11. Egg length: 0 = less than 11 mm; 1 = 12-13 mm; 2 = over 15 mm (polarity rationale unclear).
12. Spermatophore size: 0 = small; 1 = medium; 2 = large (commonality, also small in cirrates).
13. Mantle aperture width: 0 = narrow (A or B); 1 = wide (C) (polarity rationale unclear).

14. Skin texture: 0 = smooth; 1 = papillose; 2 = tubercles (polarity rationale unclear).

15. Supra-ocular cirri: 0 = absent; 1 = present (polarity rationale unclear).

were entered as ordered; characters with un-
certain polarities (egg length, mantle aperture
width, skin texture, supra-ocular cirri; Table 2)
were entered unordered.

States of functionally related characters
were examined to assess whether characters
changed independently, or in concert. If as-
sociated changes were identified, characters
were recoded as a single, multistate charac-
ter.

RESULTS

Analysis of the data set (Appendix 1) re-
sulted in at least 1999 equally parsimonious
trees (35 steps, consistency index 0.514).
The strict consensus tree, which depicts
groups supported by all equally parsimonious
trees, revealed two groups, one containing
Pareledone, Eledone, Octopus, Benthocto-
pus and Teretoctopus, and the other contain-
ing the remaining nine genera. None of the
1999 equally parsimonious topologies (Fig. 2)
are consistent with Voss’ evolutionary tree
(Fig. 1). Voss’ tree, when analyzed by cladis-
tic methods requires 49 steps, i.e. 14 steps
(40%) more than the most parsimonious so-
lution.

Relaxation of the strict consensus con-
Straint illustrates relationships supported by
some (in this case by at least 60%) but not all,
of the alternate trees (majority rule consensus
n = 60%). Bathypolypus is suggested to be
more closely related to Graneledone, Thau-
meledone and Bentheledone than to any
taxon with which it shares biserial suckers. Of

Benthoctopus
Teretoctopus
Bathypolypus
Tetracheledone
Velodona
Vosseledone
Eledone
Pareledone
Thaumeledone
Bentheledone
Graneledone

FIG. 1. The evolutionary tree presenting subfamily
and generic relationships of the benthic Octopo-
didae, rooted to the Cirrata, excluding oceanic
forms (after Voss 1988a: 274). 0, Octopodinae; B,
Bathypolypodinae; E, Eledoninae; G, Graneledon-
inae.

Voss’ generic relationships (Fig. 1), close re-
lationships between Benthoctopus-Teretocto-
pus and Thaumeledone-Bentheledone are
supported at the indicated levels. The strict
consensus tree requires the number of sucker
rows to change and the ink sac to be lost at
least twice. The majority rule consensus ar-
rangement requires these changes, and an
additional change in the number of sucker
rows.

346 VOIGHT

= 04
Y VU E
n 3 © ©
о 2a 23238 2888
= a0 Nn $ © зоо
S 2258882383088
2 82.2358 093338585
оао ESS аа
ое ВЕБЕ
A DORA» DROW
92 64
68 89
4
00

FIG. 2. Diagrammatic results of the cladistic analy-
sis of data set in Appendix 1, rooted to the hypo-
thetical ancestor. Numbers at the nodes indicate
the proportion of the 1999 equally parsimonious
trees discovered that support that node. The node
indicated by 100 is the limit of resolution supported
by all equally parsimonious trees.

Examination of the data matrix (Appendix
1) reveals that several functionally related
characters change in concert. All taxa that
lack marginal plates (character 7) also lack
lateral teeth (character 6); all taxa that lack
lateral teeth also have a homodont rachidian
(character 5). These changes in the radula
appear to occur in a cascade pattern. A sim-
ilar suite of changes is also seen in the ante-
rior digestive system (no taxon with small
posterior salivary glands, character 5, has a
crop diverticulum, character 4) and between
skin texture and supraocular cirri (characters
14, 15). Recoding associated characters as
single multistate characters maintains the in-
formation in the original data matrix, reflects
the associated nature of the changes and
condenses the number of characters from 15
to 11 (Appendix 2).

DISCUSSION

Cladistic analysis (Fig. 2) of characters tra-
ditionally used in octopodid classification in-
dicates that the octopodid subfamilies are,
and have been, paraphyletic (Table 1). Al-
though these subfamilies have been defined
on comparatively obvious differences, they

cannot be held to share evolutionary histo-
ries.

The uncertain status of octopodid subfam-
ilies has been a subject of earlier discussion.
In Robson’s original (1928) definition, the Bat-
hypolypodinae (two sucker rows and no ink
sac) included Bathypolypus, Benthoctopus
and Teretoctopus. п 1932, Robson redefined
the group (Table 1) to include Bathypolypus,
Graneledone, Thaumeledone and Benthele-
done, with Benthoctopus and Teretoctopus
assigned to the Octopodinae. Robson (1932:
49—56) apparently recognized that, although
his original definition of Bathypolypodinae
created a morphologically distinctive and co-
hesive group, the presence of multiple char-
acters refuted monophyly of the Eledoninae
and a close relationship between Bathypoly-
pus and Benthoctopus.

Robson stated that his (1932) definition of
the Bathypolypodinae may have made the
Octopodinae paraphyletic; Figure 2 supports
this suggestion. Because Scaeurgus, Pteroc-
topus, Tetracheledone, Vosseledone and Vel-
odona appear to share a more recent com-
mon ancestor with members of the
Bathypolypodinae than do Pareledone, Ele-
done, Octopus, Teretoctopus or Benthocto-
pus (Fig. 2), including them in the Octopodi-
nae creates an unnatural group that exists
only in the classification. Voss (1988a), re-
jected Robson’ subfamilies, in essence, to re-
turn to those erected earlier.

As we appear to be unable to define sub-
families that are even arguably monophyletic,
use of the taxonomic category of subfamily
should be avoided. The presence of an artifi-
cial category implies a level of knowledge that
we have yet to achieve; in doing so, it im-
pedes the discovery of evolutionary histories.
Octopodid groups may best be defined for
discussion by ecological or ontogenetic crite-
ria, for example, holobenthic (Boletzky,
1992).

Among the major problems octopod sys-
tematics faces is how to define ancestral
states. In this analysis, the definition of the
hypothetical ancestor as nearly identical to
shallow-water taxa ensures that deep water
taxa will be found to be derived. This tradi-
tional view (Naef, 1923; Robson, 1925, 1932;
Voss, 1967) may be an artifact of the taxo-
nomic need to distinguish comparatively rare
specimens of deep water taxa from familiar,
normal octopuses.

That the common ancestor of the incirrate
octopods was a benthic octopod, based on

CLADISTIC REASSESSMENT ОЕ OCTOPODID CLASSIFICATION 347

the rationale that the loss of the fins would not
be adaptive in pelagic forms (Boletzky, 1992),
has canalized the way we think of the group.
Young (1977) attributed the absence of the
supra-branchial commissure in the cteno-
glossan Japetella to loss associated with ad-
aptation to a pelagic habitat from a benthic
ancestral state. In that evolutionary scenario,
the possibility that the suprabranchial com-
missure is a synapomorphy shared by oc-
topodids and argonauts is eliminated from
consideration.

To ensure alternate octopodid relationships
are considered, primitive states must be de-
fined by objective criteria such as outgroup
analysis or ontogeny (see discussion by Bry-
ant, 1991). Whether a given character state is
widely distributed, occurs in the most com-
mon species, or characterizes the most di-
verse taxon, does not demonstrate that it is
ancestral.

Systematic studies of octopodids are also
hindered by our inability to define monophyl-
etic species groups. Taxonomy succeeds if
specimens can be assigned to genera; sys-
tematics fails if genera do not share a common
history. Members of the genus Pareledone, for
instance, are separable from those of Eledone
and Graneledone. Whether they represent di-
vergent octopodid lineages that lack the diag-
nostic synapomorphies, or are united by a
unique history is unknown and cannot be dis-
covered with the available characters. The
taxon Eledoninae of Voss (1988a), and the
genus Octopus itself are affected by the same
problem. These taxa are the leftovers afterthe
removal of those with synapomorphies. Incor-
rectly assuming monophyly for species groups
obscures patterns of character change, and
can undermine the analysis.

Too few characters of uncertain (or uncon-
tested) homology also limit phylogenetic re-
construction of the octopodids. Characters of
loss and reduction dominate this data set. Al-
though Begle (1991) showed reductive char-
acters to be as informative as character
gains, and Voss & Voss (1983) found losses
as informative as gains in their cladistic anal-
ysis of the cranchiid squids, in this analysis
too few positive characters are available to
test this statement. Perhaps because taxon-
omy has focused on differences between
| deep-sea and shallow-water octopuses, sev-
| eral of the characters used here (e.g. ink sac,
| crop, posterior salivary glands, gill lamellae,
egg size, mantle aperture) are losses and re-
_ ductions that may be under direct selection in

deep-water habitats (Robson 1925, 1932;
Voss, 1967, 1988a).

Every opportunity must be used to increase
our knowledge of octopod biology. Because
cladistic analysis requires explicit definition of
the characters and character states consid-
ered in the analysis, the data set documents
associated change in characters (Appendices
1, 2). The presence of associated change
may indicate the existence of a functional re-
lationship among characters that might other-
wise be undetected; it can provide insight into
the biology of the animals.

The radular reductions among the octopo-
dids that have been viewed as independent
(characters 5—7, Appendix 1) show unexpect-
edly orderly character change (Appendix 2).
Only taxa in which the rachidian is homodont
lose the first lateral tooth; only taxa without
the first lateral teeth lose the marginal plates.
This sequence suggests that the radulae of
taxa with homodont rachidians differ function-
ally from those with a multicuspid rachidian, in
which the radular teeth may function as a mu-
tually supporting bracing mechanism (Solem
& Roper, 1975). Similar changes in the diges-
tive system, that only taxa without a crop di-
verticulum have small posterior salivary
glands, suggest that these taxa allocate di-
gestive enzymes differently. The changes ap-
pear to be neither independent nor random,
although we must demonstrate that they are
functionally associated. Defining each of
these conditions as separate inflates the
number of characters without increasing the
information entered into the analysis. Eleven
characters cannot resolve relationships
among 14 taxa.

К may be argued that these data were not
intended for parsimony-based methodology,
and that cladistic analysis violates the
premise and rationale behind their collections
and initial analyses. Other, undocumented
characters may have contributed to the rec-
ognition of these taxonomic groups. Group
definitions relying on subtle, inexpressible
similarities, however, only further support that
morphological cohesiveness defines the
groups. Explicit reliance on these few charac-
ters, and on paraphyletic groups they have
created, has limited our knowledge of octo-
pod evolution. We must recognize and elimi-
nate artificial taxonomic divisions to begin
modern systematic treatments of this cosmo-
politan marine group. Shedding preconceived
notions may free us to discover the mono-
phyletic groups that evolution has produced.

348 VOIGHT

ACKNOWLEDGMENTS

| am indebted to the late G. L. Voss for his
encouragement of my evolutionary studies of
octopods. N. Voss, RSMAS, University of Mi-
ami, loaned specimens important to this
study. D. Lindberg made helpful comments
on the manuscript, as did two anonymous re-
viewers. 5. Schwinning and В. Bieler assisted
with translations, and C. Simpson assisted
with figures.

LITERATURE CITED

BEGLE, D. P., 1991, Relationships of the osmeroid
fishes and the use of reductive characters in phy-
logenetic analysis. Systematic Zoology, 40: 33—
53.

BOLETZKY, S. v., 1992, Evolutionary aspects of
development, life style, and reproductive mode in
incirrate octopods (Mollusca, Cephalopoda). Re-
vue Suisse Zoologie, 99: 755-770.

BRYANT, Н. М., 1991, The polarization of character
transformations in phylogenetic systematics: role
of axiomatic and auxiliary assumptions. System-
atic Zoology, 40: 433—445.

GRIMPE, G., 1921, 2. Teuthologische Mitteilungen.
VII. Systematische Ubersicht der Nordseeceph-
alopoden. Zoologischer Anzeiger, 52: 296-304.

GRIMPE, G., 1922, Systematische Übersicht der
europäischen Cephalopoden. Sitzungsberichte
der Naturforschenden Gesellschaft zu Leipzig, 9:
36-52.

HOCHBERG, Е. G., М. МХОМ & В. В. TOLL, 1992,
Order Осюрода Leach, 1818. Рр. 213-279, in:
М. J. Sweeney, С. F. Е. ROPER, К. M. MANGOLD,
М. В. CLARKE & 5. v. BOLETZKY, eds., “Larval”
and juvenile cephalopods: а manual for their
identification. Smithsonian Contributions to Zool-
ogy, 513.

NAEF, A., 1923, Cephalopoda. Part |, Vol. |, Fas-
cicle Il, in Fauna and Flora of the Bay of Naples,
35. Israel Program for Scientific Translations,
Jerusalem, pp. 293-879.

ROBSON, G. C., 1925, The deep sea Octopoda.
Proceedings of the Zoological Society of London,
1925: 1323-1356.

ROBSON, С. C., 1928, Notes on the Cephalopoda.
VI. On Grimpella, a new genus of Octopoda, with
remarks on the classification of the Octopodidae.

Annals and Magazine of Natural History, (10)2:
108—114.

ROBSON, С. С., 1929, А monograph of the Recent
Cephalopoda. Part 1. The Octopodinae. British
Museum of Natural History, 236 pp.

ROBSON, С. C., 1932, А monograph of the Recent
Cephalopoda. Part 2. Octopodidae exclusive of
the Octopodinae. British Museum of Natural His-
tory, 359 pp.

ROPER, С. Е. Е. & Е. G. HOCHBERG, 1988, Be-
havior and systematics of cephalopods from Liz-
ard Island, Australia, based on color and body
patterns. Malacologia, 29: 153-193.

SOLEM, А. & С.Е. Е. ROPER, 1975, Structures of
recent cephalopod radulae. The Veliger, 18:
127-133.

SWOFFORD, D. L., 1989, PAUP phylogenetic
analysis using parsimony. !. Natural History Sur-
vey, Champaign, Ill.

THIELE, J., 1934, Handbuch der systematischen
Weichtierkunde. Verlag von Gustav Fischer,
Jena. 3: 779-1022 pp.

THORE, S., 1949, Investigations of the “Dana” Oc-
topoda. Part |. Bolitaenidae, Amphitretidae, Vit-
reledonellidae, and Alloposidae. Dana Report,
33: 1-85.

TOLL, В. B., 1991, The supraspecific classification
of the Octopodinae (Cephalopoda: Octopoda): a
review. Bulletin of Marine Science, 49: 668.

VOIGHT, J. R., 1990, Population biology of Octo-
pus digueti and the morphology of tropical Amer-
ican octopuses. Ph.D. Dissertation, University of
Arizona, Tucson, 196 pp.

VOSS, G. L., 1967, The biology and bathymetric
distribution of deep-sea cephalopods. Studies in
Tropical Oceanography, 5: 511-535.

VOSS, G. L., 1988a, Evolution and phylogenetic
relationships of deep-sea octopods (Cirrata and
Incirrata). Рр. 253-276, in: M. В. CLARKE & Е. В.
TRUEMAN, eds., The Mollusca Vol. 12. Paleontol-
ogy and neontology of cephalopods, Academic
Press, San Diego.

VOSS, G. L., 1988b, The biogeography of the
deep-sea Octopoda. Malacologia, 29: 295-307.

VOSS, М. А. & В. 5. Voss, 1983, Phylogenetic re-
lationships in the cephalopod family Cranchiidae
(Oegopsida). Malacologia, 23: 397—426.

YOUNG, J. Z., 1977, Brain, behaviour and evolu-
tion of cephalopods. Symposium of the Zoologi-
cal Society of London, 38: 377-434.

Revised As accepted 20 January 1993

CLADISTIC REASSESSMENT OF OCTOPODID CLASSIFICATION 349

APPENDIX 1. Reported are the data matrix, including for each OTU, characters coded as indicated on
Table 2 (9 = character absent, or polymorphic within genus), the total number of characters coded as
derived and the estimated mean depth distribution of each genus (Voss, 1988b).

CHARACTER NUMBER ULA

OTU LESA Roe. O 7.258: Se) Фо 345: > Бер
ANCESTOR 085092 2022072072020, Оооо JOAO 107707 70
OCTOPUS 2075077072207 Or 0 OF о ооо 1 46
ELEDONE OOOO 10) ROMEO: 50. OM OP SION ON INDE] 157
PARELEDONE 0552077207207 ORO 70), 207297 хоро вок соков 481
TERETOCTOPUS I О O 0. O0; » al. COMMON 59259 59500, 907
BENTHOCTOPUS ROM Os 0 20540, 00 529,510, Wie SPO 0 3,0253: 7551060
SCAEURGUS OO 0 ко 044 0:10 7000 HO 1500, Tis 21,234 275
BEIBAGHIEKEBONE © 0 0 ©. ТОТО 1.0,,.1..0 2 1 6 364
PTEROCTOPUS 120575 12220, 2.0440, Oi dd 050,0 Ill 410
VOSSELEDONE ОО Oz 29559 105
VELODONA Y 9 0 Oar OO @O a OS ve a © 588
GRANELEDONE Oeil: 9 9 Oleg О EZ Onli
BATHYPOLYPUS las lle ile 0% ¿0405 10 ei OS ti D 790
THAUMELEDONE A A Об ro NO ely Os 9) ER
BENTHELEDONE Dsl? plot ardid, TRUE Oil 3354

APPENDIX 2. Data matrix recoded to reflect associated (cascading) changes in character states, and
thus the reduction in the number of characters from 15 to 11. Characters defined as in Table 1, except 3,
5 and 14, below.

CHARACTER NUMBER

OTU

OCTOPUS
ELEDONE
PARELEDONE
TERETOCTOPUS
BENTHOCTOPUS
SCAEURGUS
TETRACHELEDONE
PTEROCTOPUS
VOSSELEDONE
VELODONA
GRANELEDONE
BATHYPOLYPUS
THAUMELEDONE
BENTHELEDONE

w
+
a
+

3*. 0 = crop diverticulum; 1 = crop dilation; 2 = crop dilation and posterior salivary gland reduction.

5*. 0 = radula with 7 teeth; rachidian multicuspid; 1 = radula with 7 teeth, rachidian non-cuspid; 2 = rachidian non-cuspid
and lateral teeth absent, 3 = rachidian non-cuspid, lateral teeth and marginal plates absent.

14*. 0 = smooth skin; 1 = papillose skin; 2 = papillose skin with supra-ocular cirri; 3 = tubercles and supra-ocular cirri.

А —

MALACOLOGIA, 1993, 35(2): 351-359

THE ARRANGEMENT OF SUCKERS ОМ OCTOPODID ARMS
AS А CONTINUOUS СНАВАСТЕВ

Janet R. Voight

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

ABSTRACT

Studies of octopodid taxonomy and classification have cited the number of longitudinal sucker
rows on octopus arms as if it were a purely dichotomous character. This character, however, has
been suspected to be continuously distributed and associated with increased sucker density
(Hoyle, 1886; Berry, 1914). This study tests that hypothesis by comparing the relationship
between the mean number of suckers per arm to mean arm length among octopodid genera
occurring above 500 m depth. Specimens of genera typified by a single sucker row but with
suckers arranged in a zigzag pattern are also included.

Most specimens with two sucker rows and with suckers arranged in zigzags have more suckers
at a given arm length than do specimens with suckers arranged in a single row, supporting the
hypothesis. Most specimens with one sucker row are separated from those with two rows by a
curve on the plot of the number of suckers versus arm length, although four specimens of
Pareledone spp., preserved with their arms straightened into a swimming position rather than
recurved, and the holotype of Aphrodoctopus schultzei are exceptional. The number of suckers
on the arms of these specimens predict that they will be arranged in one row. The zigzag
arrangement seen on the specimens may be due to preservation artifact in the case of the
specimens of Pareledone and in A. schultzei by the 6-8 enlarged suckers on each arm. Variation
in the number of suckers within groups defined by the number of sucker rows is greater than that
between groups, suggesting that the number of sucker rows is a continuous character. Evidence

provided here indicates that A. schultzei should be included among the species of Eledone.
Key words: Octopodidae, sucker rows, classification, continuous character, Eledone, Aphro-

doctopus.

INTRODUCTION

Octopodid taxonomy and systematics is
entering a dynamic period; preliminary at-
tempts to reconstruct evolutionary relation-
ships among members of the Octopoda
(Voss, 1988; Voight, 1990) have lead to a re-
assessment of our assumptions about the
group (Voight, 1991, 1993; in press). One such
assumption, expounded by Voss (1988), is
that the number of longitudinal sucker rows
on the oral surface of the octopus arm is a
dichotomous character that accurately re-
flects evolutionary relationships.

Whether suckers on an octopus arm form
one or two longitudinal rows has featured
prominently in diagnoses of octopod families
(Rochebrune, 1884; Joubin, 1918), subfami-
lies (Voss, 1988), and genera (e.g. Robson,
1932; Roper & Mangold, 1991). Statements
such as in young Eledone “as suckers are
added they never form two rows” (Hochberg
et al., 1992: 265; similarly, Rochebrune,
1884) reflect the degree to which the charac-
ter is thought to be dichotomous. Yet, the
arms of specimens of Eledone and Parele-

351

done sometimes carry suckers arranged in
double rows, or in a zigzag pattern where the
number of rows is arguable (Hoyle, 1904;
Joubin, 1905, 1918; Gravely, 1908). Preser-
vation may contribute to the formation of dou-
ble sucker rows in these genera (Guérin,
1908), but live animals also show sucker ar-
rangements considered to be anomalous for
their taxon (Chadwick, cited by Gravely,
1908; Naef, 1923).

Whether the number of sucker rows on an
octopus arm is a valuable character for recon-
structing phylogenies has been questioned
(Owen, 1881; Hoyle, 1886; Berry, 1914; Naef,
1923). Based on his discovery of only slight
differences in the sucker musculature be-
tween specimens of Octopus, with two sucker
rows, and those of Eledone, with one sucker
row, Guérin (1908) doubted that sucker ar-
rangement was an adequate basis on which
to distinguish the genera. Berry (1914) sug-
gested that octopus suckers are inherently or-
ganized in a single row and that only because
of crowding are suckers displaced alternately
to the side. He felt that this displacement cre-
ated the appearance of a double sucker row.

352 VOIGHT

The biological significance of this character
had yet to be evaluated despite this alternate
hypothesis.

This paper tests the hypothesis that sucker
crowding is associated with the formation of
double sucker rows by examining the relation-
ship between the number of suckers on an
arm and arm length among octopodid genera
typically occurring above 500 m depth. Spec-
imens of taxa characterized by one sucker
row that have suckers in a zigzag arrange-
ment are predicted to show the same pattern
as taxa with two sucker rows. The phyloge-
netic significance of sucker arrangement is
assessed.

MATERIALS AND METHODS

To test the hypothesis that the formation of
double sucker rows is associated with sucker
crowding, the number of suckers on octopus
arms with one sucker row was compared to
that with two sucker rows as a function of arm
length. The hypothesis predicts that more
suckers will occupy arms with two rows than
with one row at the same arm length. Speci-
mens of taxa typified by one row with suckers
arranged in a zigzag pattern will reflect the
pattern shown by specimens with two sucker
rows.

Specimens included in this analysis (n =
142) were from the California Academy of
Sciences, San Francisco; Field Museum of
Natural History, Chicago; Rosenstiel School
of Marine and Atmospheric Science, Univer-
sity of Miami; the United States National Mu-
seum, Washington, D.C.; and University of
California Museum of Paleontology, Berke-
ley. Octopuses with suckers arranged in a
double row were represented by specimens
of Octopus, Hapalochlaena and Macrotrito-
pus and the type specimen of Macrochlaena
(Robson, 1926). Data from Toll (1988) for Cis-
topus, Pteroctopus, Robsonella and Scaeur-
gus and from Roper & Mangold (1991) for
Aphrodoctopus increased the number of gen-
era with two sucker rows included. Data from
Toll (1988) also increased the data available
for species of Octopus.

Representing octopuses with suckers ar-
ranged in a single row were typical specimens
of the genera Eledone, Pareledone, Vossele-
done, and Tetracheledone. To ensure com-
plete and unbiased representation of the taxa,
eight data points for Pareledone were taken
from reports of Joubin (1905), Berry (1917),

Adam (1941), Taki (1961) and Kubodera &
Okutani (1986); seven points for Eledone
were from Massy (1916), Rees (1956) and
Adam (1951, 1984). Three specimens of E.
cirrhosa and data from the type of P. turqueti
(Joubin, 1905), all with suckers in a zigzag
arrangement, were included. Only taxa with
mean depth distributions above 500 m were
included to avoid the effects of decreased
sucker size associated with increased depth
distribution (Voight, in press).

Suckers were counted as described by Toll
(1988), using a combination of macroscopic
and microscopic techniques. Suckers on right
arms I-IV were counted; left arms were used if
the right were damaged. Only normal arms
were used for data analysis; injured arms or
those with incomplete regeneration were ex-
cluded. Hectocotylized arms of males (one of
the third pair of arms specialized for sper-
matophore transfer) were considered sepa-
rately from normal arms.

The analysis requires that each datum be
independent, that is, free of any correlations
or association with other data in the analysis.
Because all non-hectocotylized arms of an in-
dividual specimen are subject to identical ge-
netic and environmental variables or controls,
they are not independent. Statistical tests of
the working null hypothesis, that each normal
arm of an individual specimen has the same
number of suckers, were prohibited by the
small sample size within an individual, inevi-
table errors in counting, and errors in regen-
eration that may have failed to restore all
suckers. This hypothesis was rejected if the
number of suckers on different arm pairs var-
ied consistently in all available specimens of a
given species.

Only male specimens of Eledone caparti
were available, and only in this species was
the null hypothesis rejected, as indicated by
Adam (1950). Typical of Eledone, these
males have sucker-derived modifications at
the arm tips (Haas, 1989: Fig. 2). When the
number of modifications and suckers were
summed, the result was virtually invariant
within an individual (Table 1). Because within
individual specimens of all other species ex-
amined, the number of suckers was essen-
tially equal among the arm pairs, data taken
from only one or two arms were considered
representative and were included.

Despite the anomalous pattern seen on
arms of E. caparti, sucker counts of males
with heteromorphic arm tips were repre-
sented in the analysis by mean sucker num-

SUCKER ARRANGEMENT ОМ OCTOPODID ARMS 353

TABLE 1 Sucker counts, heteromorphic arm tip counts and arm lengths for normal
arms and hectocotylized arms (АЗ) of males of Eledone caparti.

Arm Length
Specimen ARM Suckers Modif. Total (mm)
A. R1 98 35 133 193
R2 97 34 131 143
R3 41 — 41 76
L3 59 82 141 UA
А4 60 73 133 106
В А1 89 36 125 174
R2 72 63 135 115
L3 59 77 136 94
R3 43 — 43 65
А4 57 80 137 95
С R1 85 45 130 179
R2 54 47 101 104
R3 41 — 41 64
L3 REGENERATIN
А4 41 68 109 78

Бег, rather than by the sum of suckers and
modifications. Because the modified suckers
at the arm tips are very strongly reduced in
size, e.g. over 14 can occupy 1 mm in males
of Е. caparti, including them would have bi-
ased the results against the hypothesis being
tested.

The number of suckers on, and the lengths
of, the normal arms of each individual speci-
men were meaned. To compare the number
of suckers on normal arms of octopuses with
one sucker row to those with two sucker rows
independent of differences in size, the mean
number of suckers was plotted versus mean
arm length for each individual.

Using arm length as the univariate proxy of
size carries with it liabilities. Voight (in press a)
hypothesized that the different parts of the
muscular octopus body respond to preserva-
tion equally, allowing measurements within a
preserved specimen to be compared without
net preservation bias, as shown by Voight
(1991). Because preservation-linked changes
affect arm length but not the number of suck-
ers, such biases affect only the x-axis in this
analysis. The arms of flaccid specimens may
appear abnormally long with comparatively
few suckers; contracted arms may appear
short with many suckers. To moderate the ef-
fect of this bias, a large size range of speci-
mens was included. Arm length rather than a
multivariate size measure was used here be-
cause it is easily determined, requires no sta-
tistical expertise, and is a biologically realistic
measure by which to compare the number of
suckers.

Data from hectocotyli were analyzed di-
rectly. The number of suckers versus hecto-
cotylus length was plotted for male speci-
mens of each species.

RESULTS

On the normal arms of the octopuses con-
sidered, virtually all specimens with suckers
in double or zigzag rows have more suckers
at a given arm length than do those with one
row. With few exceptions, points representing
specimens with one sucker row can be sep-
arated from those representing specimens
with two sucker rows by a curve on the plot of
sucker number versus arm length (Fig. 1).
Specimens of Eledone cirrhosa and the type
of Pareledone turqueti, both with suckers ar-
ranged in a zigzag pattern, have more suck-
ers at the same arm length than do conge-
neric specimens of comparable size with
suckers arranged in a single row; they fall on
the two-rowed side of the curve.

Four specimens of Pareledone and the ho-
lotype of Aphrodoctopus schultzei violate this
pattern. Suckers on these five specimens
were arranged in double rows or in zigzags,
despite plotting with specimens with a single
sucker row (Fig. 1).

Most specimens of Pareledone have fewer
than 50 suckers on an arm, however, speci-
mens of P. senoi (Taki, 1961; Kubodera &
Okutani, 1986) diagnosed as the genus
Megaleledone based on their large size, ap-
pear to have up to 65 suckers (Fig. 1). Arms

354 VOIGHT
300

240

200

150

100

Mean Number of Suckers
S
о

>
о

0 60 120 180 240 300 340
Mean Arm Length

FIG. 1. Plotted for the normal arms of each specimen are the mean number of suckers versus the mean arm
length. Upper case letters represent specimens with a double sucker row: A, Octopus bimaculatus; В, О.
briareus; С, Cistopus indicus; E, О. selene; Е, O. fitchi; G, О. chierchiae, O. penicilifer and O. stitiochrus; H,
О. hubbsorum and Hapalochlaena spp., |, О. digueti; L, О. californicus; N, Macrotritopus defilippi/horridus;
O, O. macropus/ornatus; P, Pteroctöpus tetracirrhus; Q, Octopus (Macrochlaena) winckworthi: В, Rob-
sonella fontanianus; $, Scaeurgus unicirrhus/patagiatus; U, O. bimaculoides; V, O. vulgaris; X, O. filosus; Y,
О. burryi; ? Aphrodoctopus schultzei. Lower case letters represent specimens of taxa with a single sucker
row: a, Eledone caparti, с, Pareledone charcoti; e, Tetracheledone spinicirrus; д, Е. gaucha; m, Е. moschata;
р, Р. polymorpha; г, E. cirrhosa; $, P. (Megaleledone) senoi; 1, P. turqueti; у, Vosseledone charrua: x, P.
adelieana, P. aurorae P. harrissoni and P. nigra (one specimen each); y, E. massyae. The curve, which was
fitted by eye, generally separates specimens with a single sucker row (below) from those with two sucker
rows and suckers in a zigzag arrangement (above). The points within circles represent specimens of
Pareledone with suckers in zigzags below the curve.

SUCKER ARRANGEMENT ON OCTOPODID ARMS 355

of specimens of Eledone can carry at least
135 suckers; specimens of Octopus can have
up to 300 suckers on an arm. The number of
suckers on an arm of E. cirrhosa and E. mo-
schata approaches that of some specimens
with two sucker rows. The number of suckers
on the arms of the type of P. turqueti (Joubin,
1905) cannot be distinguished from that of oc-
topuses of equal size with two sucker rows.

Although most octopuses with one sucker
row are separated from those with two sucker
rows by a very narrow margin (Fig. 1), within
each group the average number of suckers
borne on an arm of a given length varies con-
siderably. At arm lengths near 200 mm, spec-
imens with one sucker row average from 46
(P. senoi) to 112 (E. moschata) suckers on an
arm, specimens with two sucker rows aver-
age from 135 (in Cistopus indicus) to 247 (in
Macrotritopus spp.) suckers on an arm. Liter-
ature-based and specimen-based data report
a comparable number of suckers on arms of
similar length within a taxon.

On the plot of the number of suckers on the
hectocotylus versus hectocotylus length (Fig.
2), most males of taxa typified by a single
sucker row have fewer suckers on the hecto-
cotylus than do specimens with two sucker
rows. On the hectocotyli of two males of E.
cirrhosa, one with one sucker row and one
with zigzag sucker arrangement, however,
the number of suckers equals or exceeds that
on hectocotyli of octopuses with two rows.
The male type of A. schultzei with two sucker
rows, has as few suckers on the hectocotylus
as do males with one sucker row. Hectocotyli
with one sucker row, other than those of Ele-
done, always plot beneath the curve that sep-
arates normal arms with one from those with
two sucker rows; hectocotyli with two sucker
rows plot on both sides of the curve.

DISCUSSION

The hypothesis that sucker crowding is as-
sociated with the formation of double sucker
rows is supported. In most of the octopus
specimens considered, if the number of suck-
ers exceeds a critical limit dependent on arm
length, the suckers form double rows. The
consistency of this limit, or threshold (Fig. 1),
among the octopuses considered suggests
that a physical constraint affects each of the
taxa considered; the five exceptional speci-
mens reveal the effect of other factors.

In four specimens of Pareledone, the suck-

ers arranged in zigzags despite being few in
number. These specimens may violate the
pattern because their arms were preserved
straight, in a swimming position, as recom-
mended by Roper & Sweeney (1983). The
arms of comparable specimens that are re-
curved in preservation carry a single sucker
row.

In fixation, unrestrained arms recoil, appar-
ently due to contraction of the web. On a re-
curved arm, the oral, suckered surface on the
outer curve of the arm is in tension; the aboral
surface, forming the inner curve, is com-
pressed. Artificially straightened arms are
subject to different forces, which may invali-
date comparisons between straight and re-
coiled arms. When straight arms are flexed
aborally, the space between the suckers in-
creases and their arrangement can approach
a single row.

That a curve rather than a line separates
most taxa with one sucker row from those with
two rows (Fig. 1) illustrates that sucker size
also influences the relationship between suck-
ers. On the short arms of young octopuses
with small suckers, each small sucker at the
arm tip occupies a large proportion of the total
space. On longer arms with larger suckers,
small suckers at the arm tip occupy propor-
tionately less space, the large suckers already
in place dominate. The threshold curves with
increasing size as a result of growth.

Sucker growth may also explain why some
hectocotylized arms violate the pattern seen
in normal arms (Fig. 2). Hectocotyli develop
as normal arms up to a point; if more than the
critical number of suckers recruit, double
sucker rows form. Small hectocotyli plot as
predicted by normal arms (Fig. 2), and they
are directly comparable; the comparison,
however, becomes invalid with growth. The
hectocotylus carries an apparently species-
specific number of suckers, often many fewer
than on normal arms (Toll, 1988; Villanueva et
al., 1991). Although hectocotyli are shorter
with fewer suckers than are other arms, the
arm and suckers continue to grow, as evi-
denced by within species variation in hecto-
cotylus length (Fig. 2; Toll, 1988; Villanueva et
al., 1991). If the suckers on the hectocotylus
become larger than those on normal arms,
their size may maintain the double sucker
rows, despite their reduced number.

На double sucker row is associated with
sucker crowding, and large suckers occupy
more space than small suckers, then a com-
paratively few very large suckers could form

356 VOIGHT

170

160

—

Number of Suckers

180 240 280

120
Length of the Hectocotylized Arm (mm)

FIG. 2. Plotted are the number of suckers on the hectocotylus versus hectocotylus length. Symbols defined
as in Figure 1. The curve separates normal arms with two rows from normal arms with one sucker row.

double rows. This mechanism has been sug-
gested to create double sucker rows in male
specimens of the cirrate octopods Opistho-
teuthis depressa and О. japonica (Sasaki,
1929; Taki, 1963). | suggest that this mecha-
nism also produced the double sucker rows
on the type of А. schultzei. The number of
suckers on the arms of the type predicts that
it will have a single sucker row, but the 6-8
dramatically enlarged suckers on each arm of
Aphrodoctopus schultzei (Roper & Mangold,
1991) may occupy enough space that most
suckers occupy more than one row (Hoyle,
1910: plate Va, fig. 1; Roper & Mangold,
1991: fig. 4).

Sucker number varies more within groups
sharing the same number of sucker rows than
it does between groups. Such groups may
thus be arbitrary units. Three lines of evi-
dence support this statement. First, although
the genera Eledone and Pareledone are de-
fined by having a single sucker row, speci-
mens of both can have suckers arranged in

two rows or in zigzags (Joubin, 1905;
Gravely, 1908). Octopus, defined by having a
double sucker row, contains specimens with
suckers arranged in zigzags or nearly single
rows (Robson, 1932). That exceptions occur
in diverse genera suggest that the character
is artificial.

Second, the muscles attaching the suckers
to the arms are very similar in specimens of
Eledone and Octopus (Guérin, 1908; Kier &
Smith, 1990). Guérin (1908: 59) predicted
that eliminating some of the suckers and elon-
gating the axis of the arm, that is reducing
sucker crowding, would shift the sucker ar-
rangement from two rows to one. The present
results support his prediction and indicate that
these genera differ only superficially in this
character. Detailed studies of other genera
and of developmental series have yet to be
accomplished.

Third, the distribution of points relative to
the critical limit separating specimens with a
single row from those with double suckers

SUCKER ARRANGEMENT ОМ OCTOPODID АВМ$ 357

rows (Fig. 1) reflects the arrangement of suck-
ers on most specimens. Points lying just above
the curve (Fig. 1) represent specimens of Cis-
topus indicus that have suckers arranged di-
agonally, or nearly in a single line (Robson,
1929), as predicted by the plot. Specimens of
Eledone are just below the curve ifthe suckers
form a single rows; specimens of this species
with suckers in a zigzag are just above it. The
continuous distribution of points reflects the
continuous nature of the character.

If, as suggested here, the spatial relation-
ship among the suckers determines their ar-
rangement, different strategies may serve to
influence that relationship. Chief among
these strategies may be differentiation of
sucker sizes along the arms.

If octopuses have dramatically more than
the critical number of suckers required to form
double sucker rows, why do the suckers only
form double rows? Although individuals with
three sucker rows per arm are currently con-
sidered developmental anomalies (Toll & Bin-
ger, 1991), Owen (1881) named the genus
Tritaxeopus for specimens with three sucker
rows. Owen, who suggested that sucker ar-
rangement was continuous among the Ос-
topodidae, stated that because Tritaxeopus
differed as much from Octopus in sucker row
number as did Eledone, it merited equal tax-
onomic recognition. Owen’s (1881) report that
286 suckers occupy the 584 mm-long third
arm of his now missing type specimen is com-
parable to specimens included here with
shorter arms (Fig. 1) and two sucker rows.

The rarity of specimens with multiple
sucker rows may be associated with sucker
size differentiation. In specimens with a single
sucker row, the suckers occupy a compara-
tively narrow size range. Especially in speci-
mens of Pareledone, the terminal suckers are
large compared to those on the tips of arms
with two sucker rows. In shallow-water octo-
puses with two sucker rows, the suckers near
the margin of the web are distinctly the larg-
est; distally, sucker size declines dramatically
but continuously. Because few suckers are
large, the amount of crowding is reduced, as
is the crowding associated with the many
small suckers. By partitioning sucker size, two
discrete sucker rows may be maintained de-
spite the presence of hundreds of suckers.
_ Why multiple sucker rows appear to be
| avoided by octopuses may relate to functional
| difficulties or that increased nervous and
‚ muscular control are required.

That increased sucker density is associ-

ated with double sucker rows is consistent
with data available for specimens of the deep-
water genus Benthoctopus (Voight, unpubl.).
Available specimens and data (Russell, 1922)
for Bathypolypus arcticus and B. faeroensis
show that despite their suckers being few in
number and small in size (Voight, in press)
they also form double rows. If the mechanism
forming double rows can be shown to differ
between Bathypolypus and the octopuses
considered here, double sucker rows would
be shown to be convergent in the Octopo-
didae, as predicted by Robson’s (1932) clas-
sification of the family and my preliminary cla-
dogram (Voight, 1990).

If the number of sucker rows is unreliable
for phylogenetic reconstruction, could the un-
derlying character suite of sucker number and
arm length indicate close evolutionary rela-
tionships, e.g. between Octopus and Ele-
done? Higher order names have been as-
signed, not to reflect relationships, but to
group outwardly similar taxa by readily appar-
ent characters (e.g. Joubin, 1918). Anato-
mists who perhaps believed that the generic
names indicated distinctly different taxa have
compared these genera but have rarely found
significant differences (Girod, 1882; Сиепп,
1908; Kier & Smith, 1990).

Without an independent means of postulat-
ing relationships, and aware that a similarity
in the relationship between sucker number
and arm length can be produced by changes
in either character, conclusions are prema-
ture. The number of suckers in Octopus bi-
maculatus and O. bimaculoides, very similar
species thought to have diverged only re-
cently (Pickford & McConnaughey, 1949), dif-
fer more than among species of Octopus and
Eledone (Fig. 1), suggesting that this charac-
ter does not necessarily reflect evolutionary
history.

Eliminating the number of suckers rows as
a taxonomic character does not affect most
currently recognized genera. The genus
Pareledone should be defined to reference its
few suckers on each arm rather than one
sucker row; its definition, however, may still
be based solely on plesiomorphic, or ances-
tral, characters (Voight, 1993). Eledone
remains as a distinct taxon; its members
share the apparent synapomorphies of male
heteromorphic arm tips formed by the lateral
extension of sucker buds, the reduction or
absence of a calamus, the anterior fusion of
the branchial retractors and, pending more
data, in utero fertilization (Perez et al., 1990).

358 VOIGHT

Whether E. palari Lu & Stranks, 1991, shares
homologous characters is uncertain.

Eledone, however, may not be monophyl-
etic; it appears to share with Aphrodoctopus
several characters that suggest common an-
cestry. А single male specimen was desig-
nated as type of the genus Aphrodoctopus by
virtue of its apparent double sucker rows and
characters unique in Octopus but shared with
species in the genus Eledone. The type spec-
imen, despite the appearance of having two
sucker rows, plots with specimens having one
row (Fig. 1), possibly due to its very large
suckers, as discussed above.

Characters supporting the relationship be-
tween A. schultzei and species in Eledone in-
clude the heteromorphic arm tips of males
and the structure of the ligula. Although Roper
& Mangold (1991) stress the unusual ligula,
the ligulae of males of E. caparti appear to be
very similar (Adam, 1952: fig. 52), as, to a
lesser degree, do those described for E. thys-
anophora by Voss (1962), E. massyae by
Voss (1964), and for Pareledone carlgreni by
Thore (1945).

Because the characters cited here as syn-
apomorphies with Eledone were the basis for
the new genus, and the number of sucker rows
is an artifact of sucker size and density, | sug-
gest that A. schultzei be placed in Eledone.
Features distinguishing it from E. thysano-
phora are yet to be determined. The species
are likely to be closely related to Е. caparti;
they share the structure of the ligula, sucker
size differences, and arm formulae and may
have adjacent geographic distributions. The
species can be distinguished by the spermato-
phores; crochets are present in E. schultzei
and E. thysanophora but absent in E. caparti.

ACKNOWLEDGEMENTS

| thank the Illinois Board of Higher Educa-
tion for support for Research Intern Shillock
Yuan. R. E. Strauss and S. H. Lidgard offered
valuable comments. | thank T. Gosliner, Cal-
ifornia Academy of Sciences; D. Lindberg,
University of California Museum of Paleontol-
ogy; N. A. Voss, Rosenstiel School of Marine
and Atmospheric Science; and C. F. E.
Roper, United States National Museum, for
the opportunity to examine their collections
and for the loan of specimens in their care.
Financial support from the Conchologists of
America and the Hawaiian Shell Club as-
sisted preliminary data collection.

LITERATURE CITED

ADAM, W., 1941, Cephalopoda. Mémoires de
Musée Royal d'Histoire Naturelle de Belgigue,
(2) 21:83-161.

ADAM, W., 1950, Notes sur les Céphalopodes.
XXII. Deux nouvelles espèces de la côte afri-
caine occidentale. Bulletin Institut Royal des Sci-
ences Naturelles de Belgigue, 26(45): 1-9.

ADAM, W., 1951, Les céphalopodes de l'Institut
Français d'Afrique Noire. Bulletin de l'Institut
Français d'Afrique Noire, 13: 771-787.

ADAM, W., 1952, Céphalopodes. Expédition
Océanographique Belge dans les Eaux Côtières
Africaines de l'Atlantique sud (1948-1949), 3(3):
1-142.

ADAM, W., 1984, Cephalopoda from west and
south Africa. Atlantide Report, 13: 151-180.

BERRY, S. S., 1914, The Cephalopoda of the Ha-
waiian Islands. Bulletin of the Bureau of Fish-
eries, 32: 257-362.

BERRY, $. S., 1917, Cephalopoda. Australian Ant-
arctic Expedition 1911-1914. Scientific Reports,
Series C, Zoology & Botany, 4(2): 1-38.

GRAVELY, Е. H., 1908, IV Notes on the spawning
of Eledone and on the occurrence of Eledone
with suckers in double rows. Memoirs and Pro-
ceedings of the Manchester Literature & Philo-
sophical Society, 53(4): 1-14.

GIROD, P., 1882, Recherches sur la poche du noir
des céphalopodes des côtes de France. Archives
de Zoologie Expérimentale et Générale, 10:
1-100.

GUERIN, J., 1908, Contributrion a l'étude des sys-
tèmes cutané, musculaire et nerveux de Гарра-
reil tentaculaire des céphalopodes. Archives de
Zoologie Expérimentale et Générale, (4) 8:
1-178.

HAAS, W., 1989, Suckers and arm hooks in Co-
leoidea (Cephalopoda, Mollusca) and their bear-
ing for phylogenetic systematics. Abhandlungen
des Naturwissenschaftlichen Vereins in Ham-
burg, (NF) 28: 165-185.

HOCHBERG, Е. G., М. МХОМ & В. В. TOLL, 1992,
Octopoda. Smithsonian Contributions to Zool-
ogy, 513: 213-279.

HOYLE, W. Е., 1886, Report on the Cephalopoda
collected by HMS Challenger during the years
1873-1876. Report of the Scientific Results of
the HMS Challenger Zoology, 16(1): 1-245.

HOYLE, W. E., 1904, Reports on the Cephalopoda.
Bulletin of the Museum of Comparative Zoology,
Harvard, 43: 1-71.

HOYLE, W. E., 1910, XV. Mollusca: (A) Ceph-
alopoda. Zoologische und Anthropologische
Ergebnisse einer Forschungsreise im Westlichen
und Zentralen Südafrika, 4: 261-268.

JOUBIN, М. L., 1905, Description de deux elé-
dones provenant de Гехред оп du Dr. Charcot
dans l'Antarctique. Mémoires de la Société
Zoologique de France, 18: 22-31.

JOUBIN, L., 1918, Etudes préliminaires sur les
Céphalopodes recueillis au cours des croisières

SUCKER ARRANGEMENT ОМ OCTOPODID ARMS

de 5. А. $. le Prince de Monaco. 6 Note. Vitrele-
donella richardi Joubin. Bulletin de l'Institut
Océanographique, 340: 1-40.

KIER, W. М. & А. M. SMITH, 1990, The morphology
and mechanics of octopus suckers. Biological
Bulletin, 178: 126-1386.

KUBODERA, T. & T. OKUTANI, 1986, New and
rare cephalopods from the Antarctic waters.
Memoirs of the National Institute of Polar Re-
search Special Issue, 44: 129-143.

LU, C. C. & Т. М. STRANKS, 1991, Eledone palari,
a new species of octopus (Cephalopoda: Oc-
topodidae) from Australia. Bulletin of Marine Sci-
ence, 49: 73-87.

MASSY, A. L., 1916, Mollusca. Part II Ceph-
alopoda. British Antarctic (“Terra Nova”) Expedi-
tion, 1910. Natural History Report. Zoology, 2(7):
141-176.

NAEF, A., 1923, Cephalopoda. Fauna and Flora of
the Bay of Naples, Monograph 35 (Israel Pro-
gram for Scientific Translations 1972).

OWEN, R., 1881, X. Descriptions of some new and
rare Cephalopoda (Part Il.) Transactions of the
Zoological Society of London, 11: 131-168.

PEREZ, J. A. A., M. HAIMOVICI & J. C. BRAHM
COUSIN, 1990, Sperm storage mechanisms and
fertilization in females of two South American ele-
donids (Cephalopoda: Octopoda). Malacologia,
32: 147-154.

PICKFORD, С. Е. & В. H. MCCONNAUGHEY,
1949, The Octopus bimaculatus problem: a study
in sibling species. Bulletin of the Bingham
Oceanographic Collection, 12(4): 1-66.

REES, W. J., 1956, Notes on the European species
of Eledone with especial reference to eggs and
larvae. Bulletin of the British Museum of Natural
History, 3: 283-293.

ROBSON, С. C., 1926, Notes on the Cephalopoda,
|. Description of two new species of Octopus from
S. India and Ceylon. Annals and Magazine of
Natural History, (9)17:159-167.

ROBSON, G. C., 1929, A monograph of Recent
Cephalopoda. Vol. 1 The Octopodinae. British
Museum of Natural History.

ROBSON, G. C., 1932, A monograph of Recent
Cephalopoda. Vol. 2 The Octopodidae exclusive
of the Octopodinae. British Museum of Natural
History.

ROCHEBRUNE, A. T. DE, 1884, Etude mono-
graphique de la famille des Eledonidae. Bulletin
Société Philomathique de Paris, (7)8:152-163.

ROPER, С. Е. Е. & К. М. MANGOLD, 1991, Octo-
pus schultzei (Hoyle, 1910): a redescription with
designation of Aphrodoctopus new genus (Ceph-
alopoda; Octopodinae). Bulletin of Marine Sci-
ence, 49: 57-72.

ROPER, C. F.E. & M. J. SWEENEY, 1983, Tech-
niques for fixation, preservation, and curation of
cephalopods. Memoirs of the National Museum
of Victoria, 44: 29-47.

RUSSELL, Е. S., 1922, Report on the Cephalopoda
collected by the research steamer “Goldseeker”

359

during the years, 1903-1908. Fisheries, Scot-
land, Scientific Investigations, 3: 1-45.

SASAKI, M., 1929, A monograph of the dibranchi-
ate cephalopods of the Japanese and adjacent
waters. Journal of the Faculty of Agriculture,
Hokkaido Imperial University, 20 (supplement)
1-357.

TAKI, I., 1961, On two new eledonid octopods from
the Antarctic Sea. Journal of the Faculty of Fish-
eries and Animal Husbandry Hiroshima Univer-
sity, 3: 297-316.

TAKI, I., 1963, On four newly known species of
Octopoda from Japan. Journal of the Faculty of
Fisheries and Animal Husbandry, Hiroshima Uni-
versity, 5: 57-93.

THORE, S., 1945, On the Cephalopoda of Profes-
sor О. Сайдгеп’$ expedition to South Africa in
1935. Kungl, Fysiografiska Sällskapets I Lund
Förhandlingar, 15: 49-57.

TOLL, R. B., 1988, The use of arm sucker number
in octopodid systematics (Cephalopoda: Oc-
topoda). American Malacological Bulletin, 6:
207-211.

TOLL, В. В. &(. С. BINGER, 1991, Arm anomalies:
cases of supernumerary development and bilat-
eral agenesis of arm pairs in Octopoda (Mol-
lusca, Cephalopoda). Zoomorphology, 110:
313-316. Е

VILLANUEVA, В., Р. SANCHEZ & М. A.
СОМРАСМО ROELEVELD, 1991, Octopus
magnificus (Cephalopoda: Octopodidae), a new
species of large octopod from the southeastern
Atlantic. Bulletin of Marine Science, 49: 39-56.

VOIGHT, J. R., 1990, Population biology of Octo-
pus digueti and the morphology of American trop-
ical octopods. Ph.D. Dissertation. University of
Arizona, Tucson, 196 pp.

VOIGHT, J. R., 1991, Morphological variation in oc-
topod specimens: reassessing the assumption of
preservation-induced deformation. Malacologia,
33: 241-253.

VOIGHT, J. R., 1993, A cladistic reassessment of
the Octopodid subfamilies. Malacologia, 35(2):
343-349.

VOIGHT, J. R., in press, The association between
distribution and octopodid morphology: implica-
tions for classification. Zoological Journal of the
Linnean Society.

VOSS, G. L., 1962, South African cephalopods.
Transactions of the Royal Society of South Af-
rica, 36: 245-272.

VOSS, G. L., 1964, A note on some cephalopods
from Brazil with a description of a new species of
octopod, Eledone massyae. Bulletin of Marine
Science of the Gulf and Caribbean, 14: 511-516.

VOSS, G. L., 1988, Evolution and phylogenetic re-
lationships of deep-sea octopods (Cirrata and In-
cirrata) Pp. 253-276, in: M. В. CLARKE, & Е. В.
TRUEMAN, eds., Paleontology and neontology of
the cephalopods. The Mollusca, Vol. 12. Aca-
demic Press Inc., San Diego.

Revised Ms. accepted 5 May 1993

MALACOLOGIA, 1993, 35(2): 361-369

OVER-REPRESENTATION OF RARE ALLELES IN JUVENILES AND LACK OF
PATTERN IN GEOGRAPHIC DISTRIBUTIONS OF ALLELES IN А LAND SNAIL

Kenneth C. Emberton

Department of Malacology, Academy of Natural Sciences, 19th & The Parkway,
Philadelphia, Pennsylvania 19103, Ц. $. А.

ABSTRACT

Eight populations of Mesodon zaletus (Binney) (Gastropoda: Stylommatophora: Polygyridae),
ranging from West Virginia to Alabama to Missouri and Arkansas, were examined at 16 allozymic
loci, nine of which were variable. Available population samples were generally small (2-17), but
a large sample (140) was taken from Monte Sano, Alabama. Chi-square tests using PGM-1 in
this population showed a fit to Hardy-Weinberg equilibrium, and an over-representation of rare
alleles in juveniles. Among the eight populations, M. zaletus showed substantial geographic
differentiation in allelic frequencies, with no consistent pattern of geographical variation among
loci. These results put important caveats on allozyme systematics of land snails.

Key words: allozymes, Gastropoda, Pulmonata, Polygyridae, Mesodon zaletus.

INTRODUCTION

Mesodon zaletus (Binney, 1837) is a large
(shell diameter 24-31 mm) polygyrid land
snail inhabiting deciduous forests up to an el-
evation of about 1,500 m. This species
ranges from New York to Illinois, south to cen-
tral Alabama, and west through a southern-
Illinoisian constriction to Missouri and Arkan-
sas (Fig. 1). In the course of phylogenetic
Studies on the tribes Triodopsini and Mesod-
ontini (Emberton, 1988, 1991a), allozymic
data (16 loci) were accumulated for eight pop-
ulations of M. zaletus (Fig. 1), including one
large sample (n = 140) with both adults and
juveniles. Here | report unusual results en-
countered in the analysis of these data for
patterns of allelic variation within and among
populations.

MATERIALS AND METHODS

Collection data on the eight populations
(Fig. 1) are as follows; voucher materials are
all in the Field Museum of Natural History,
Chicago (ЕММН); field station numbers are in
the author's “GS” series.

TN BLOUNT. Tennessee: Blount County:
Great Smoky Mountains National Park: White
Oak Sink: limestone bluffs at the north and
west edges of the sink. Adults (17 collected
and electrophoresed) were on or under leaf
litter, juveniles (an unrecorded number col-
lected, and none electrophoresed) were on

361

the rock surfaces of the bluffs. 19 June 1981,
11 a.m.-6 p.m., Ken & Ellen Emberton collec-
tors. Vouchers FMNH 214771 (GS-9; one dis-
sected).

AL MADISON-1. Alabama: Madison
County: Huntsville (east of): Monte Sano State
Park: base of limestone bluffs below scenic
outlook at main picnic area. The bluffs border
a small permanent waterfall and stream.
Adults (31 collected, 26 electrophoresed [all
but numbers 7, 10, 16, 20, and 24]) were most
prevalent on the peripheries of outcrops, on
deep leaf litter. Several mating pairs were
seen. An unrecorded number of juveniles were
also collected, but none were electro-
phoresed. 16 July 1981, 8 p.m.-10:30 p.m.; 17
July 1981, 6:15 a.m.-9:30 a.m.; Ken Emberton
collector. Vouchers FMNH 214772 (GS-20;
none dissected).

AL MADISON-2. Same site as AL MADI-
SON-1. On litter surface and in talus in the
main cove, the day after a rain. Adults (95
collected, all but one [#78] electrophoresed)
more commonly on the litter surface than the
juveniles (an unrecorded number collected,
20 electrophoresed), some of which were on
the cliff face. 30 April 1982, 9 a.m.-10:30 a.m.,
1:15 p.m.-2:45 p.m., Ken Emberton collector.
Vouchers FMNH 214773 (GS-101; three dis-
sected).

AR CRAWFORD. Arkansas: Crawford
County: Devils Den State Park: Self-Guided
Nature Trail. M. zaletus was most common on
talus and deep leaf litter in the lowlands along
the creek at the head of the trail. Conditions

362 ЕМВЕАТОМ

MO BARRY

AR CRAWFORD

AL MADISON

„—— — WV PRESTON

ENS

KY FAYETTE

KY HARLAN

TN BLOUNT

TN FRANKLIN

FIG. 1. The eight sampled populations of Mesodon zaletus within the species’s geographic range in the

eastern United States.

were very wet, due to a recent rain. Collected
eight adults (all electrophoresed) and an un-
recorded number of juveniles (one electro-
phoresed). 25 April 1982, 7 a.m.-10 a.m.; 25
April 1982, 4 a.m.-7:40 a.m.; Ken Emberton
collector. Vouchers FMNH 214787 (GS-90;
one dissected).

MO BARRY. Missouri: Barry County:
Roaring River State Park: 1.1 miles west of
junction with Road F on Missouri Route 112:
wooded ravine at top of bluff overlooking the
park, at the edge of the National Forest. Un-
der logs and litter on scree slopes of chert-like
rock with scattered leaf litter; all logs were
charred by fire (this was the most productive
site, nonetheless, for land snails found within
the park). Two adults collected and electro-
phoresed; number of juveniles unrecorded,
and none electrophoresed. 28 April 1982,
7:30 a.m.-11:00 a.m.; Ken Emberton collec-
tor. Vouchers FMNH 214788 (GS-96; two dis-
sected).

TN FRANKLIN. Tennessee: Franklin
County: 1.5 miles north of Sherwood Post Of-
fice, then a short distance east (along a small
road) from the south side of bridge: wooded
hillside above creek with limestone outcrop-

pings. The area was partially cleared, with a
large trash pile. Six adults collected and elec-
trophoresed; no juveniles (number unre-
corded) electrophoresed. 1 May 1982, 1:30
p.m.-3:30 p.m., Ken & Ellen Emberton collec-
tors. Vouchers FMNH 214774 (GS-104; five
dissected—Emberton, 1991a: figs. 3a-e,
4a-e).

KY FAYETTE. Kentucky: Fayette County:
Grimes Mill Road at Boone Creek: upper
edge of floodplain downstream from parking
lot at crossing. Under logs and leaf litter in oak
forest with limestone outcrops. Collected: four
adults (all electrophoresed) and an unre-
corded number of juveniles (six electro-
phoresed). 7 May 1982; 10 a.m.-2:30 p.m.,
Ken Emberton, John Petranka, and B. Kirk-
patrick collectors; 3:15 p.m.-5:45 p.m., Ken
Emberton and John Kirkpatrick collectors.
Vouchers ЕММН 214775 (GS-112; none dis-
sected).

KY HARLAN. Kentucky: Harlan County:
United States Route 421, 0.1-0.2 miles south
of junction with Kentucky Route 221: oak-for-
ested hillside with sandstone talus overlying
limestone. Of eight adults, seven were elec-
trophoresed; no juveniles (number unre-

ALLELES IN А LAND SNAIL

corded) were electrophoresed. 9 May 1982, 2
%—3 hours in the morning, Ken Emberton
and John Petranka collectors. Vouchers
FMNH 214777 (GS-119; one dissected).

WV PRESTON. West Virginia: Preston
County: Coopers Rock State Forest: along
thin belt of friable limestone about Y of the
way down the west slope of New River Gorge,
just east of main overlook. Under patches of
accumulated leaf litter on very steep slope.
Ten of the 15 collected adults (numbers 1-5,
8, 9, 12, 13, and 15) were electrophoresed.
An unrecorded number of juveniles were col-
lected, none of which were electrophoresed.
14 May 1982, 10 a.m.-1:30 p.m., Ken Ember-
ton collector. Vouchers FMNH 214778 (GS-
126; one dissected).

Thus single collections were made of seven
populations, but AL MADISON was sampled
both in summer of 1981 and in spring of 1982.
The latter collection was the largest, compris-
ing 114 snails, including both juveniles and
adults. Other population samples consisted of
two to 17 adults and various numbers of ju-
veniles, and ranged from northeastern West
Virginia to southwestern Missouri (Fig. 1).

Specimens of Mesodon zaletus were col-
lected into muslin bags. Within one hour after
collection, the bags were placed over ice in a
cooler and held for one-half to five days. Upon
removal, the snails were placed onto a double
layer of dampened paper towels. As each
snail extended from its shell and began to
crawl, the posterior, free portion of its foot was
cut off with an Exacto knife. Each excised
piece of tissue (“snail tail”) was placed into a
screw-top plastic cryogenic vial, which was
dropped into liquid nitrogen contained in a
portable vacuum-walled freezer. Amputated
snails were labelled on their shells using a
Rapidograph; cryogenic vials were labelled
using a black Sharpie.

The amputated snails were drowned over-
night in tap water laced with chloryl hydrate
(one medium-sized crystal per liter), fixed in
95% ethanol (method of A. Solem, personal
communication), and later removed to 70%
ethanol for storage and dissection. One to
three adults were dissected per population.
Adults were detected by their reflected shell
lip (Pilsbry, 1940). Dissections consisted of
removing the reproductive system, slitting
open the uneverted penial tube, and pinning
open the tube to view the functional surface of
the penis (Emberton, 1988: fig. 1). The penial
morphology of M. zaletus is distinctive, is rel-
atively invariant among populations, and thus

363

is reliable for identification
1991a).

Undissected adult M. zaletus were identi-
fied by their conchological features. The only
species in the same geographic range (Fig. 1)
that might be confused for M. zaletus are (1)
M. thyroidus (Say), (2) M. elevatus (Say), and
(3) species of both M. (Akromesodon) and the
Neohelix albolabris (Say) and N. alleni
(Sampson) groups. Adults of these three
groups can be distinguished from adult M. za-
letus by their half-open umbilicus, domed
spire, and lack of parietal denticle, respec-
tively (Burch, 1962; Pilsbry, 1940; Emberton,
1988, 1991a). Juveniles of all these taxa, on
the other hand, are often difficult, and some-
times seemingly impossible, to distinguish by
shells alone. Shells of M. zaletus neoadults
with newly reflected aperatural lips and un-
formed parietal denticles are easily mistaken
for shells of N. albolabris (personal observa-
tions). Field identification of juveniles from AL
MADISON was verified, therefore, using al-
lozymes.

In the laboratory, vials containing tissue
samples were removed from the portable
freezer and sorted in a cold room at 2°C, then
transferred to a —20°C freezer, where they
were stored up to seven weeks until removed
for electrophoresis. One-fifth to all of a given
tissue sample (“пай tail”) was used for each
“run” of four to six electrophoretic gels. Used
samples were placed into alternating wells of
a pre-chilled glazed ceramic depression plate
that was kept on Blue Ice during grinding and
wicking. Grinding of tissues was by one of two
methods, both of which were effective against
the problem of high concentrations of mucus:
(1) coating a large sample with a thin layer of
powdered glass and with an equal volume of
grinding buffer, and grinding slowly (to pre-
vent mucous frothing) with a soft-plastic test
tube, the diameter of which was slightly less
than that of the depression well (tissue and
mucus clings to the roughened bottom of the
test tube when withdrawn, leaving a clear
fluid for wicking); and (2) covering a small tis-
sue sample with an equal volume of ground
glass and three to four times its volume of
grinding buffer, and slowly pulverizing the en-
tire tissue sample, using a small glass test
tube with a frosted bottom. The gummy clots
resulting from this second method were
dragged with forceps to the edge of the of the
well; if insufficient fluid remained in the well,
one or two drops of grinding buffer were
dropped onto the clot, then pressed out of it to

(Emberton,

364

run down the side of the well. Wicks cut from
Whatman #5 filter paper were placed in the
tissue fluid remaining in each well and were
daubed on a KimWipe tissue before being
loaded onto the gels.

Electrophoretic methods were those of Se-
lander et al. (1971) and Shaw & Prasad
(1970), as adapted by Davis et al. (1981) and
Emberton (1988). Sixteen loci were used that
were genetically interpretable, that repre-
sented a wide variety of metabolic pathways,
that included loci of proven heritability (Mc-
Cracken, 1976; McCracken & Brussard,
1980), and that excluded loci of demonstrated
environmental inductability (Oxford, 1973,
1978; Gill, 1978a, b) in land snails. The loci
used were SDH-1, MDH-1, MDH-2, ME, ICD,
PGD, GD-1, GD-2, SOD-1, SOD-2, GOT-1,
GOT-2, PGM-1, LAP-1, MPI, and GPI. All pre-
sumed alleles were tested in side-by-side
comparisons on the same gel. A common al-
lele of each locus was scored as 100, and the
mobilities of other alleles in mm were scored
relative to 100 mm. Details of electrophoretic
procedures are given in Emberton (1988: ap-
pendix A).

Because of generally small sample sizes,
only one enzyme locus in one population
(PGM-1 in AL MADISON) provided reason-
able tests for Hardy-Weinberg equilibrium
and for homogeneity between adults and ju-
veniles. Chi-square tests were used for both,
collapsing the chi-square tables to get rid of
small expectations (Sokal & Rohlf, 1969; El-
ston & Forthofer, 1977).

Geographic variation in allozymes was ex-
amined by the use of pie diagrams of allelic
frequencies, and by two phenetic analyses
(UPGMA and distance-Wagner), each based
on two different indices of genetic distance
(Neïs and Rogers). BIOSYS computer pro-
grams (Swofford & Selander, 1981) were
used for all calculations.

RESULTS

In total, 35 allozymic alleles were detected,
of which seven were from monomorphic and
28 from variable loci. Among the eight popu-
lations, the mean number of alleles per locus
was 1.1 to 1.5, the percentage of loci poly-
morphic was 12%-25%, and mean heterozy-
gosity ranged from 0.04 to 0.08. Allelic fre-
quencies for the nine variable loci are
presented in Table 1.

Hardy-Weinberg equilibrium was strongly

EMBERTON

supported for PGM-1 in the AL MADISON
population (chi square = 0.000, p = 1.00):

Allelic Class Observed Expected
100/100 22 22.0
100/other 67 67.0
other/other 51 51.0

Comparison between 120 adults and 20 ju-
veniles of the AL MADISON population gave
the following allelic frequencies for PGM-1:

Allele Adults Juveniles

103 0.017 0.050

100 0.412 0.300
98 0.154 0.300
95 0.400 0.325
91 0.017 0.025

Collapsing this table for chi-square analysis
and giving allelic counts rather than frequen-
cies yields:

Allele(s) Adults Juveniles Total
100 99 12 112
95 96 13 109
rare _45 15 _60
240 40 280

From this table, chi-square = 7.22, p < 0.05.
This result indicates that rare alleles are sig-
nificantly over-represented among the young.

Allelic geographical distributions are
mapped in Figure 2. The distribution of the 35
alleles of all 16 loci among populations (Table
1, Fig. 2) was bimodal:

# of Populations # of Alleles % of Alleles
1 9 26%
2 6 17%
3 2 6%
4 2 6%
5 0 0%
6 1 3%
7 1 3%
8 14 40%

Thus, alleles predominantly were either local-
ized or widespread geographically among the
sampled populations: 43% occurred in only
one or two populations, and 40% occurred in
all eight populations. This bimodal pattern

ALLELES IN А LAND SNAIL 365

TABLE 1. Allelic frequencies of the nine variable loci for the eight populations of Mesodon zaletus.
Untabulated monomorphic alleles were: MDH-1, MDH-2, ICD, PGD, GD-1, GD-2, and СОТ-2.

Population
Tn AR MO AL TN KY KY WV
Blount Crawford Barry Madison Franklin Fayette Harlan Preston
Locus Allele (n = 17) (n = 9) (n=2) (n= 140) (n=5) (n= 10) (n= 7) (n= 10)
SDH-1 106 0.0 0.0 1.000 0.0 0.0 0.0 0.0 0.0
100 1.000 1.000 0.0 1.000 1.000 1.000 1.000 1.000
ME 100 1.000 1.000 1.000 1.000 1.000 1.000 0.643 1.000
98 0.0 0.0 0.0 0.0 0.0 0.0 0.357 0.0
SOD-1 110 0.0 0.0 0.0 0.0 0.900 0.0 0.0 0.0
100 1.000 1.000 1.000 1.000 0.100 1.000 1.000 1.000
$00-2 104 0.0 0.222 0.0 0.0 0.0 0.0 0.0 0.0
100 1.000 0.778 1.000 1.000 1.000 1.000 1.000 1.000
GOT-1 103 0.0 0.222 0.0 0.018 0.0 0.0 0.0 0.0
100 0.794 0.778 1.000 0.982 1.000 1.000 1.000 1.000
97 0.206 0.0 0.0 0.0 0.0 0.0 0.0 0.850
PGM-1 103 0.0 0.0 0.0 0.021 0.0 0.400 0.857 0.0
102 0.0 0.0 0.250 0.0 0.0 0.0 0.0 0.0
100 0.882 1.000 0.0 0.396 0.100 0.300 0.0 1.000
98 0.118 0.0 0.0 0.0 0.175 0.0 0.0 0.0
97 0.0 0.0 0.0 0.0 0.100 0.300 0.0 0.0
96.5 0.0 0.0 0.750 0.0 0.0 0.0 0.0 0.0
95 0.0 0.0 0.0 0.389 0.800 0.0 0.143 0.0
91 0.0 0.0 0.0 0.018 0.0 0.0 0.0 0.0
LAP-1 104 0.0 0.0 0.0 0.0 0.0 0.200 0.0 0.0
100 0.912 1.000 0.250 0.986 0.500 0.800 1.000 1.000
98 0.088 0.0 0.750 0.007 0.0 0.0 0.0 0.0
96 0.0 0.0 0.0 0.007 0.0 0.0 0.0 0.0
МР! 102 0.0 0.556 0.0 0.0 0.0 0.0 0.0 0.550
100 1.000 0.444 1.000 1.000 1.000 1.000 1.000 0.450
СР! 103 0.0 0.111 0.0 0.0 0.900 0.0 0.714 0.300
100 0.824 0.889 1.000 0.993 0.100 1.000 0.286 0.700
95 0.176 0.0 0.0 0.007 0.0 0.0 0.0 0.0

persisted even after the removal of rare alle-
les with sample frequencies less than 0.02.
Examination of Figure 2 reveals that each
allele, regardless of whether it was localized
or widespread, had a unique distribution
among the eight populations; there was no
obvious geographical correlation among loci.
This generally mosaic geographic distribu-
tion of alleles was further attested by phenetic
analyses. Clustering results (not illustrated)
differed, depending on which genetic similar-

ity or distance measure was used (Ме! vs.

Rogers), and which clustering algorithm was
used (UPGMA vs. Distance Wagner). For ex-
ample, MO BARRY was at the base of the Nei
UPGMA tree, interior to TN FRANKLIN in the
Rogers UPGMA tree, and in the center

| (paired with KY FAYETTE) in the Rogers dis-
_ tance-Wagner tree. Furthermore, patterns of

For example, the distance Wagner tree’s
tightest cluster consisted of AR CRAWFORD,
TN BLOUNT, and WV PRESTON, which
spanned the entire geographic range of sam-

pling (Fig. 1).

DISCUSSION

The important implications of this study are
that in Mesodon zaletus, populations are pan-
mictic, rare alleles are over-represented in ju-
veniles, and geographic differentiation in al-
leles is substantial and without consistent
pattern.

Panmixy is not ubiquitous in polygyrid land-
snail populations, however. Fairbanks &
Miller (1983) found that 12 populations repre-
senting two species of Ashmunella in the

| genetic similarity as revealed by the pheno-
| grams showed no consistent correlation with
geographic proximities among populations.

Huachuca Mountains, Arizona, had signifi-
cant heterozygote deficiencies. This discrep-
ancy is probably due to differences in vagility

366

EMBERTON

FIG. 2. Geographic variation in allelic frequencies of the nine variable loci. Each population (see Fig. 1) is
represented by a pie diagram, the sections of which indicate the frequencies of alleles. А key to alleles is on

the lower right of each map.

between the two депега; Ashmunella are
smaller snails than Mesodon and are re-
stricted to patchily distributed moist microhab-
itats in regions more arid than those inhabited
by Mesodon (Pilsbry, 1940: M. [Mesodon
5.5..]). This view is supported by the allozymic
evidence from other polygyrids. Combining
the present results with those of McCracken &
Brussard (1980) as taxonomically reinter-
preted by Emberton & McCracken (unpub-
lished), the following numbers of natural pop-

ulations of polygyrids conform to Hardy-Wein-
berg expected levels of heterozygosity:

Species # of Populations
Neohelix albolabris (Say)
Neohelix alleni (Sampson)
Neohelix major (Binney)
Neohelix solemi Emberton
Mesodon normalis (Pilsbry)
Mesodon zaletus

NN = = Oo

ALLELES IN А LAND SNAIL 367

Unlike Ashmunella, all of these species are
large, nocturnal and wet-weather foragers on
the leaf-litter or ground surface (Pilsbry, 1940;
McCracken, 1976; Emberton, 1981, 1986,
1991b; Hubricht, 1985; Asami, 1988a, b).
Thus, the ecology of these species correlates
with panmixis. Further tests of the relationship
between ecology and panmixis might be pos-
sible using existing allozyme data on polygy-
rids (Emberton, 1988, 1991a) but are beyond
the scope of this paper.

The preponderance of rare alleles in juve-
nile Mesodon zaletus is intriguing. Possible
explanations include natural selection against
rare alleles and ontogenetic shifts in genetic
expression of alleles. The latter view may be
supported by the mesodontin Patera clarki
(Lea), in which juveniles seem to differ in al-
leles from adults in the MDH-1, GOT-1, and
GOT-2 loci (Emberton, unpublished).

The mosaic, uncorrelated, non-clinal geo-
graphic distributions of alleles among popula-

368 EMBERTON

tions of Mesodon zaletus may find at least
partial explanation in the population biology of
this snail, if it is similar to the population biol-
оду of the conchologically and ecologically
similar Neohelix albolabris. Populations of N.
albolabris are small (estimated at 100 or
fewer individuals), fluctuating in size, geneti-
cally isolated, and probably ephemeral and
are founded by only one or a few individuals
(McCracken, 1976). Thus, geographically
random founder effects could strongly influ-
ence the distributions and frequencies of al-
leles. On the other hand, or in conjunction
with this, allelic distributions in Mesodon zal-
etus may provide hidden clues on glacial
refugia for this species. Certainly the lack of
clinal variation strongly indicates against
post-glacial spread from a single refugium.

The implications of these findings for sys-
tematics are rather important. Allozyme sys-
tematics for Mesodon, and possibly for many
other genera of land snails, ideally should in-
clude both ontogenetic and geographic as-
sessments of variation for as many species
as possible.

ACKNOWLEDGEMENTS

This work was supported in part by NIH Ge-
netics Training Grant GM07197-07 and NSF
Postdoctorai Fellowship BSR-87—00198, and
is a contribution of the Molecular Genetics
Laboratory of the Department of Malacology,
Academy of Natural Sciences. | thank George
Davis and Caryl Hesterman for their help and
encouragement in collecting the electro-
phoretic data.

| am extremely grateful to the anonymous
reviewer who discovered the difference be-
tween adults and juveniles in my data.

This paper is adapted from part of a doc-
toral dissertation for the Committee on Evo-
lutionary Biology, University of Chicago. |
thank the members of my proposal and de-
fense committees: Alan Solem, David Raup,
Michael Wade, Bradley Shaffer, Russell
Lande, Lynn Throckmorton, James Teeri, and
Harold Voris.

For assistance at the Field Museum of Nat-
ural History, Chicago, | am indebted to Alan
Solem, Margaret Baker, Patricia Johnson,
and Lucy Lyon.

For their help in the field, | thank Ellen Em-
berton, John Petranka, and Betsy Kirkpatrick.
Thanks are also extended to the park rangers

and property owners who permitted collec-
tions on lands under their care.

LITERATURE CITED

ASAMI, T., 1988a, Competition and character dis-
placement in the land snails Mesodon normalis
and Triodopsis albolabris. Doctoral Dissertation,
University of Virginia, 206 pp.

ASAMI, T., 1988b, Temporal segregation of two
sympatric species of land snails. Venus, 47: 153—
172!

BURCH, J. B., 1962, How to know the eastern land
snails. William C. Brown Company, Dubuque,
lowa, 214 pp.

CAIN, A. J., 1983, Ecology and ecogenetics of ter-
restrial molluscan populations. Chapter 14, pp.
597-647, in: W. O. RUSSELL-HUNTER, ed. (ed.-in-
chief, K. М. WiLBUR), The Mollusca, Vol. 6. Ecol-
ogy. Academic Press, New York, 695 pp.

CLARKE, B., W. ARTHUR, О. T. HORSLEY, 8 О. T.
PARKIN, 1978, Genetic variation and natural se-
lection in pulmonate molluscs. Chapter 5, pp.
219-270, In: V. FRETTER & J. PEAKE, eds., Pul-
monates. Volume 2A. Systematics, evolution and
ecology. Academic Press, London, 540 pp.

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

ELSTON, В. С. & В. FORTHOFER, 1977, Testing
for Hardy-Weinberg equilibrium in small samples.
Biometrics, 33: 536-542.

EMBERTON, K. C., 1981, Ecological notes on two
sympatric, conchologically convergent polygyrid
snails in Ohio. Bulletin of the American Malaco-
logical Union, 1980: 27-30.

EMBERTON, К. C., 1986, The evolution of multiple
sympatric homeomorphy among three genera of
land snails. Doctoral Dissertation, University of
Chicago, 780 pp.

EMBERTON, К. C., 1988, The genitalic, allozymic,
and conchological evolution of the eastern North
American Triodopsinae (Gastropoda: Pulmo-
nata: Polygyridae). Malacologia, 28: 159—273.

EMBERTON, К. C., 1991a, The genitalic, allozy-
mic, and conchological evolution of the Meso-
dontini, trib. nov. (Gastropoda: Pulmonata: Po-
lygyridae). Malacologia, 33: 71-178.

EMBERTON, K. C., 1991b, Ecology of a shell con-
vergence between subfamilies of polygyrid land
snails. Biological Journal of the Linnean Society,
44: 105-120.

FAIRBANKS, H. L. & W. B. MILLER, 1983, Inbreed-
ing and genetic variation in two species of Ash-
munella (Gastropoda: Pulmonata: Polygyridae)
from the Huachuca Mountains, Arizona. Ameri-
can Malacological Bulletin, 1: 21-26.

GILL, P. D., 1978a, Non-genetic variation in isoen-

ALLELES IN А LAND SNAIL 369

zymes of lactate dehydrogenase of Cepaea ne-
moralis. Comparative Biochemical Physiology,
59B: 271-276.

GILL, P. D., 1978b, Non-genetic variation in isoen-
zymes of acid phosphatase and alpha-glycero-
phosphate dehydrogenase of Cepaea nemoralis.
Comparative Biochemical Physiology, 60B: 365 —
368.

HUBRICHT, L., 1985, The distributions of the na-
tive land mollusks of the eastern United States.
Fieldiana, Zoology, New Series, 24: 1-191

MC CRACKEN, G. F., 1976, The population biology
of the white-lipped snail, Triodoposis albolabris.
Doctoral Dissertation, Cornell University, 136 pp.

MC CRACKEN, С. Е. & Р. Е. BRUSSARD, 1980,
The population biology of the white-lipped land
snail, Triodopsis albolabris: genetic variability.
Evolution, 34: 92-104.

OXFORD, G. S., 1973, The genetics of Cepaea
esterases. |. Cepaea nemoralis. Heredity, 30:
127-139.

OXFORD, С. S., 1978, The nature and distribution
of food-induced esterase in helicid snails. Mala-
cologia, 17: 331-339.

PILSBRY, H. A., 1940, Land Mollusca of North
America (North of Mexico, Volume 1, Part 2.
Academy of Natural Sciences, Philadelphia,
Monographs, 3: 575-994.

SELANDER, В. K., М. H. SMITH, $. У. YONG, W.
E. JOHNSON & J. B. GENTRY, 1971, Biochem-
ical polymorphism and systematics in the genus
Peromyscus. |. Variation in the old-field mouse
(Peromyscus polionotus). University of Texas
Studies in Genetics, 4: 49-90.

SHAW, С. В. & В. PRASAD, 1970, Starch-gel elec-
trophoresis of enzymes—a compilation of reci-
pes. Biochemical Genetics, 4: 297-320.

SOKAL, В. В. & Е. J. ROHLF, 1969, Biometry. W.
H. Freeman and Company, San Francisco, 776

pp.

SWOFFORD, D. L. 4 R. B. SELANDER, 1981,
BIOSYS-1. Release 1. A computer program for
the analysis of allelic variation in genetics. Users
manual. Department of Genetics and Develop-
ment, University of Illinois at Urbana-Champaign,
65 pp.

Revised Ms. accepted 29 April 1993

MALACOLOGIA, 1993, 35(2): 371-388

MORPHOLOGICAL AND ALLOZYMIC POLYMORPHISM AND DIFFERENCES

AMONG LOCAL POPULATIONS IN BRADYBAENA FRUTICUM

(O. Е. MULLER, 1777) (GASTROPODA: STYLOMMATOPHORA: HELICOIDEA)

Andrzej Falniowski', Andrzej Kozik*, Magdalena Szarowska',
Maria Rapata-Kozik? & Izabela Turyna®

ABSTRACT

Morphological variation (shell colour and banding, mantle pigmentation, colour and pigmen-
tation of reproductive organs, external form of the mucous gland) and allozymic polymorphism
at 13 loci (by means of vertical slab polyacrylamide gel electrophoresis) were studied in Brady-
baena fruticum (О. Е. Muller, 1777) from 11 localities in southern Poland and Slovakia. Descrip-
tions and illustrations of variation in all the morphological characters, and frequencies at every
locality are given. Of the 13 loci studied, six were polymorphic. The proportion of polymorphic
loci (15.5-46.1%, mean 36%) was relatively low for a morphologically polymorphic species.
Heterozygote frequencies were as expected from Hardy-Weinberg equilibrium, with the excep-
tion of the CAP, locus, at which a significant heterozygote excess was found. The values of Nei’s
distances between populations (0.01 7—0.282) were relatively high for geographically close con-
specific populations, and often a higher value of genetic distance did not correspond with a
greater geographic distance. For morphological characters and for allozyme frequencies (di-
rectly and after computing Cavalli-Sforza & Edwards’s arc distances) similarity trees were com-

puted for all populations by means of the maximum likelihood and additive tree techniques.
Key words: polymorphism, Bradybaena, land snails, allozymes.

INTRODUCTION

Bradybaena fruticum (О. Е. Müller, 1777) is
one of the most widely distributed land snail
species in Poland (Riedel, 1988), inhabiting all
regions except the higher mountains. The spe-
cies is distributed in Europe from the Urals and
Caucasus to the Balkan Peninsula, southern
Scandinavia, Germany, western France, and
northern Italy (Shileiko, 1978; Kerney et al.
1983; Riedel, 1988). However, the populations
it forms are not as dense as those of, for ex-
ample, Cepaea nemoralis (Linnaeus, 1758). It
inhabits bushes and sunny woodlands, some-
times also grasslands, parks and gardens,
preferring herbs and nettles.

There are a few studies on the biology and
life cycle of B. fruticum, (Zeifert & Shutov,
1979; Baba, 1985; Zeifert, 1987; Staikou et al.,
1990), and its population genetics (e.g.
Khokhutkin & Lazareva, 1975, 1983, 1987;
Khokhutkin, 1979, 1984; Makeeva, 1987; Ma-
keeva & Matiokin, 1987), although the only
genetic characters considered have been
shell colour and banding together with es-

terase pattern. The latter is widely known as
difficult to score or interpret (Richardson et al.,
1986), and in helicoids may vary with feeding
status (Oxford, 1973a-c). Makeeva (1987)
stressed the importance of physical barriers
and the founder effect in determining differ-
ences among local populations. Also,
Khoukhutkin (1979, 1984) and Khokhutkin &
Lazareva (1975, 1983, 1987) pointed out the
importance of semi-isolation of panmictic units
together with a geographic pattern of variabil-
ity including a slight cline from the west to the
east. They have shown that every population
is genetically distinct, more evidently in bio-
chemical characters, and that differences
among nearby populations are smaller than
those among more distant populations.

The aims of the present paper are to de-
scribe both morphological and allozymic vari-
ation in the snail, very poorly known so far,
especially as concerns the soft parts and non-
esterase enzyme systems, and to assess
whether more distant populations differ
among one another more than less distant
ones.

‘Zoological Museum, Institute of Zoology, Jagiellonian University, ul. Ingardena 6, 30-060 Kraków, Poland.
2Department of Biochemistry, Institute of Molecular Biology, Al. Mickiewicza 3, 31-120 Kraków, Poland.

371

372 FALNIOWSKI, КОЙК & SZAROWSKA

FIG. 1. Map of the sampling area. А. Localities 1-7: Shown are rivers, border of the Ojcöw National Park
(dotted line), main groups of rocks, villages (dark shaded), and forest areas (light shaded). В. All localities:
black rectangle = area of A. Shown are rivers, big man-made lakes (shaded), border between Poland and
Slovakia, towns and villages.

MATERIAL AND METHODS
Description of Localities

The material was collected from 11 localities
(Fig. 1). Nine of them are in South Poland: six
(1—6) close to each other and generally similar
in character, lying within the area of the Ojców
National Park, which comprises a complex of

valleys of various sizes, all connected with the
Pradnik River Valley which is the largest. The
predominant exposed formation is Upper Ju-
rassic rocky limestone, which forms the slopes
of the Pradnik River Valley

1. Pieskowa Skala, near a pond.
2. Dolina Zachwytu, a small branch of the
Pradnik River Valley.

POLYMORPHISM IN BRADYBAENA FRUTICUM 373

3. St. John’s Spring, the bottom of the Prad-
nik River Valley.

4. Ruins of the mediaeval castle of Ojcöw.

5. At half the distance from Ojcöw to Вгата
Krakowska, by the road.

6. Brama Krakowska rocks, on the bottom
of the Pradnik River Valley.

7. Dolina Kluczwody, the limestone valley
of the Kluczwoda stream.

8. Skala Kmity, a nature reserve in the Ju-
rassic limestone rocks of the Tenczyn Нитр.

9. Lasek Mogilski, a nature reserve of a for-
est of old trees; situated close to a large steel
mill and heavily polluted.

10. Slovakia, near Nizna, the valley of the
Orawa River in the Skorusina Mountains, a
forest.

11. Slovakia, near Uhliska, SW of Banska
Stiavnica, the Stiavnicke Vrhy Mountains,
banks of the Sikenica stream, a forest.

Collection and Morphological Techniques

The material was collected in July-August
1990, replicate sampling was done in August-
September 1991, and June-July 1992. At
each locality, at least 50 specimens were col-
lected. Only adult snails having a fully devel-
oped lip (Staikou et al., 1990) were taken. The
snails were collected from an area of a few
square metres at each locality. All the speci-
mens not frozen for allozyme study were fixed
and stored in 70% ethanol.

All specimens were classified according to
their shell colour, presence/absence of an
equatorial band on the body whorl, presence/
absence of a lip-adjacent band, presence/ab-
sence of yellow pigment on the mantle, the
pattern and intensity of black pigment on the
mantle. Then the specimens were dissected,
and examined under a stereoscopic micro-
scope, to describe the character states of the
reproductive organ polymorphisms. А! mea-
surements of the reproductive organs were
abandoned because of the observed wide
variability being evidently physiological/arti-
factual in character. Emberton (1989) pointed
out similar problems in camaenid land snails.

Electrophoretic Techniques

Acrylamide, bis-acrylamide, TEMED and
Tris were obtained from Serva (Heidelberg,
Germany), all other chemicals used for elec-
trophoresis were purchased from Sigma (St.
Louis, USA).

Snails were killed by freezing in liquid nitro-

gen and stored in a deep freeze (— 70°C) until
used. To make a suitable homogenate for
electrophoresis, each individual animal was
briefly thawed, put on ice and then the he-
patopancreas was dissected out for electro-
phoresis, taking care to take as little ovotestis
tissue as possible. The shell and all the re-
maining soft parts were then fixed in 70% eth-
anol for further morphological examination.
The homogenization medium was partly as
suggested by Wurzinger (1979) and con-
tained 20 mM Tri-HCI buffer pH 8.0, 1 mM
МАО +, 1 mM МАОР + and 15 mM тегсар-
toethanol in water; 0.3 ml of this solution was
added to each sample in a teflo-glass homog-
enizer. The homogenates were stored frozen
and electrophoresed within several days.

The electrophoretic procedures, buffers
and solutions are detailed in Table 1. Snails
from different populations were run on each
gel to facilitate comparisons. Every popula-
tion was run a minimum of five times, every
time a different group of seven specimens of
the population with a group of seven speci-
mens of another population (each time a dif-
ferent one) being picked, which enabled di-
rect comparisons among six populations to be
made. In any dubious case, additional line-up
gels were run, to enable side-by-side compar-
isons to be made. The line-up gels were pro-
vided to surround unknown mobility states by
known control states. This strategy allowed
exact comparisons of the alleles in all the
populations studied to be made.

Scoring diagrams and photographs of gels
at various stages of staining were taken, to
record the relative mobilities and intensities of
all alleles in the adjacent slots, and the abso-
lute position of each band within each sam-
ple. Loci were numbered and alleles at given
locus were assigned letters a, b, c, in order of
decreasing anodal mobility. The mobilities of
all alleles were determined by measurement
of their distance from the origin. In Table 2,
enzymes assayed, with their E.C. numbers
and staining technique references, are listed.
Zymograms were interpreted following gener-
ally accepted principles (Richardson et al.,
1986) especially the principle of conservativ-
ity, that is, to assume a minimal genetic con-
tribution to overall variation.

Numerical Methods Applied
All the allele frequencies obtained were

tested for homogeneity by means of a chi-
squared test of homogeneity (Richardson et

374

FALNIOWSKI, КОЙК & SZAROWSKA

TABLE 1. Polyacrylamide gel electrophoresis technique applied

Electrophoresis: in slabs (180 X 130 X 0.7 mm) of 7.5% polyacrylamide gel in a discontinuous high-pH

buffer system of B. J. Davis (1964).

Reservoir buffer. Tris-glycine (рн 8.3); 3 g Tris and 14.4 glycine per 1 | water.
Stacking gel buffer: Tris-HCl (pH 6.8); 6 g Tris titrated to pH 6.8 with 1M HCl in 100 т! final volume.
Resolving buffer. Tris-HCI (pH 8.8); 36.3 g Tris and 48 ml 1M НС! mixed and diluted to 100 ml final

volume.

Acrylamide-bisacrylamide solution: 30 д acrylamide and 0.8 д bisacrylamide diluted to 100 ml final

volume and filtered.

Stacking gel: 2.5 ml acrylamide-bisacrylamide solution, 5 ml stacking gel buffer, 2.5 ml 0.004% riboflavin,
10 ml water and 0.015 TEMED mixed and photopolymerized.

Resolving gel (7.5%): 15 ml acrylamide-bisacrylamide solution, 7.5 ml resolving gel buffer, 39.5 ml water
and 0.03 ml TEMED polymerized with 3 ml 1.5% ammonium persulfate as the catalyst.

Runs: Fourteen samples, 20 pl each, applied for a slab; typically, a current 20-30 mA for about 4 hrs
until a marker dye (bromophenol blue) passed all the slab.

TABLE 2. Enzymes assayed by polyacrylamide gel electrophoresis

Symbol Enzyme name

ACP Acid phosphatase

ALP Alkaline phosphatase

AAT Aspartate aminotransferase

“САР” Cytosol aminopeptidase

G3PDH Glycerol-3-phosphate dehydrogenase
HBDH 3-Hydroxybutyrate dehydrogenase
MDH Malate dehydrogenase

PGDH Phosphogluconate dehydrogenase
XO Xanthine oxidase

Enzyme number Staining after

ECIOSEStO Wurzinger (1979)
ECRBASSI Wurzinger (1979)
EC2:6131 Wurzinger (1979)
ЕС. 3:4. 111 Rudolph & Burch (1987)
ECHES Wurzinger (1979)
ЕС 1.1.1:30 Wurzinger (1979)
ЕС. 1.1.1.37 Wurzinger (1979)
ЕС 1.1.1.44 Wurzinger (1979)
ЕС 1.2.3.2 Wurzinger (1979)

Enzyme nomenclature and numbers after: Murphy et al. (1990), ХО after Richardson et al. (1986)

al., 1986). Smith’s H statistic was calculated
for each case in which the lowest allele fre-
quency exceeded 0.2, to test whether a single
panmictic subpopulation was involved (Rich-
ardson et al., 1986). Then, each locus was
tested for independence, using ап т x п chi-
squared test.

Data processing was done using the
PHYLIP package (Felsenstein, 1990). In nu-
merous studies of this kind, different popula-
tions are compared by computing Nei’s dis-
tances (Nei, 1972, 1978), and then the
clustering UPGMA technique is applied. This
is, however, not necessarily the most appro-
priate approach. Nei’s distances are seriously
influenced by numerous assumptions that are
commonly violated (Wright, 1978). Nei’s dis-
tance was originally intended to measure the
number of codon substitutions per locus that
had occurred after divergence between a pair
of populations. However, a rate of gene sub-
stitutions per locus has to be uniform at the
locus in all the populations. Moreover, Nei’s

distance is based on Kimura’s infinite isoal-
leles model of mutation (e.g. Cook, 1991) be-
ing selectively neutral, with each mutant to a
completely new allele (a very unusual phe-
nomenon), a constant rate of mutation for all
loci, and with genetic variability which initially
in a population is at equilibrium between mu-
tation and genetic drift. Nei’s distance is also
heavily influenced by within-population het-
erozygosity (Felsenstein, 1985, 1990; Swof-
ford & Olsen, 1990). Therefore, the applica-
tion of Nei’s distance, even if we accept its
usefulness in general, is dubious in most
cases; in fact, it can hardly be applied in any
comparisons among conspecific populations,
especially if our knowledge of the species’ bi-
ology, genetics, mutation rate, mutations’ se-
lective values, etc., is poor.

Therefore, although we have computed
Nei’s distances to facilitate comparisons with
other studies, we have not used these values
for any further comparisons. Instead, we have
calculated the values of Cavalli-Sforza and

POLYMORPHISM IN BRADYBAENA FRUTICUM

Edwards’s arc distance (Cavalli-Sforza & Ed-
wards, 1967), an index that is not affected by
within-population heterozygosity and that as-
sumes genetic drift as the only source of vari-
ability (Wright, 1978). Then, the values of
Cavalli-Sforza and Edwards’s arc distance
were used to compute a tree of relationships
between the populations, by means of FITCH
of PHYLIP (Felsenstein, 1990), assuming the
error absolute value to be nearly constant. It
is based on the Fitch-Margoliash’s algorithm
(Fitch & Margoliash, 1967), under the “addi-
tive tree model” (Felsenstein, 1984, 1990),
without the dubious assumption of ultra-
metricity, which is necessary when using
UPGMA. The second method applied was
KITSCH from the same package, based also
on the additive tree model, but with an as-
sumption of a molecular clock, and therefore
with an assumption of ultrametricity of the
data. We used it working with the option of the
Cavalli-Sforza & Edwards least squares
method (Edwards & Cavalli-Sforza, 1964), so
the technique was very similar in spirit to the
UPGMA (Felsenstein, 1990). KITSCH can be
considered as a phenetic clustering of the tip
species (Felsenstein, 1990); it is similar to
UPGMA but much better (Felsenstein, 1990;
Weir, 1990).

Gene frequencies have also been used
directly to compute “phylogenetic” (in our
case: phenetic similarity) trees by means
of the CONTML program of the PHYLIP
package (Felsenstein, 1990). This program
applies the restricted maximum likelihood
method based on the Brownian motion
model, and Cavalli-Sforza & Edwards’s model
of evolution (Felsenstein, 1981, 1990; Weir,
1990). The method assumes neither a molec-
ular evolutionary clock nor a new mutation.
The CONTML method has also been applied
to compute phenetic similarity trees based on
morphological character frequencies. In total,
16,965 trees have been analyzed.

RESULTS
Morphological Polymorphism

Frequencies of all morphological polymor-
phisms at all localities together with sample
sizes are given in Table 3. In Bradybaena fru-
ticum, a shell colour-banding pattern poly-
morphism is observed, but simpler and less
clear-cut than that of the well-known Cepaea
nemoralis. In contrast to Cepaea, the shell

375

wall of B. fruticum is much thinner and trans-
lucent: the soft part pigment, therefore, is vis-
ible through it, which makes the pattern vari-
ability observable in a living snail more
complicated than in Cepaea.

The shell (Figs. 2-14) is either light (from
ivory to moderately yellowish) or dark (from
pale brown to brown, with a reddish shade).
The two types always could easily be distin-
guished in shells from one locality, there being
no intermediates, but in some cases a dark
morph from one locality might resemble a light
morph from another one, though in no instance
the two morphs could be confused. In a single
specimen from locality 7, we observed a sharp
ontogenetic change in the shell colour: from
reddish brown to dark yellow; the border be-
tween the two colours was situated at the body
whorl, about 120° from the lip.

In addition to the shell colour polymor-
phism, there is a banding-pattern polymor-
phism (Figs. 5—9), although this is much sim-
рег than in Cepaea. In В. fruticum, usually
only one dark equatorial band occurs along
the body whorl (pattern 00300: Figs. 5, 6),
and/or a pale chestnut band along the lip (e.g.
Fig. 2). The latter does not cover the edge of
the lip (Fig. 3). The dark equatorial band is not
common (Table 3). The dark-lipped shells oc-
curred at each locality in higher proportions
than the banded shells did.

It must be added, however, that exception-
ally the banding pattern may be more compli-
cated. In our material of about 700 speci-
mens, we found two shells with a different
banding pattern: 02300 (Figs. 8-9). One of
them was collected at locality 9, and had on
its dark shell the upper, “accessory” band
broader than the “normal” one, diluted on its
margins and somewhat fused with the other
(Fig. 9). The other specimen had a light shell
and was collected at locality 4: the upper, “ac-
cessory” band was very wide and strongly
marked, with a much weaker and narrower
band in the usual position, fused with the ac-
cessory one (Fig. 8).

Along with the shell colour/banding poly-
morphism, a polymorphism of the soft parts
(especially the mantle) pigmentation was ob-
served (Table 3, Fig. 15). The pattern of the
mantle pigmentation was rather complicated:
composed of yellow and black pigment, more
or less intensive and forming spots of various
kind. The yellow pigment usually accompa-
nied the black one. The black pigment oc-
curred in practically all the specimens, but
showing two different patterns of distribution:

376 FALNIOWSKI, KOZIK & SZAROWSKA

TABLE 3. Frequencies of all morphological polymorphisms

locality
1 2 3 4 5 6 74 8 9 10 11

shell colour

dark 0.800 0.714 0.629 0.429 0.758 0.586 0.571 0.486 1.000 0.850 0.800

light 0.200 0.286 0.371 0.571 0.242 0.414 0.429 0.514 0.000 0.150 0.200
equatorial band

present 0.000 0.000 0.000 0.457 0.257 0.143 0.000 0.014 0.586 0.600 0.050

absent 1.000 1.000 1.000 0.543 0.743 0.857 1.000 0.986 0.414 0.400 0.950
lip band

present 0.857 0.329 0.600 0.671 0.757 0.571 0.714 0.500 0.843 1.000 0.675

absent 0.143 0.671 0.400 0.329 0.243 0.429 0.286 0.500 0.157 0.000 0.325
yellow pigment

present 0.771 0.148 0.829 0.300 0.500 0.286 0.429 0.186 0.029 0.425 0.000

absent 0.229 0.852 0.171 0.700 0.500 0.714 0.571 0.814 0.971 0.575 1.000
black pigmentation

hachured 0.814 0.729 0.414 0.572 0.886 0.586 0.129 0.471 0.957 0.725 0.775

dotted 0.186 0.271 0.586 0.428 0.114 0.414 0.871 0.529 0.043 0.275 0.225
black pigmentation

strong 0.000 0.286 0.000 0.571 0.257 0.300 0.286 0.157 0.571 0.500 0.000

weak 1.000 0.714 1.000 0.429 0.743 0.700 0.714 0.843 0.429 0.500 1.000
reproductive organs

pinkish 0.000 0.171 0.000 0.171 0.129 0.129 0.586 0.209 0.300 0.600 0.025

whitish 1.000 0.829 1.000 0.829 0.871 0.871 0.414 0.971 0.700 0.400 0.975
reproductive organs

pigmented 0.257 0.200 0.400 0.171 0.357 0.271 0.314 0.000 0.157 0.025 0.000

unpigmented 0.743 0.800 0.600 0.829 0.643 0.729 0.686 1.000 0.843 0.975 1.000
mucous gland

lobate 0.986 0.829 0.500 0.414 0.571 0.572 0.529 0.271 0.314 1.000 1.000

unlobate 0.014 0.171 0.500 0.586 0.429 0.428 0.471 0.729 0.686 0.000 0.000
mucous gland outlet

multiple 1.000 0.971 0.771 0.400 0.871 0.557 0.986 0.514 0.671 0.975 1.000

single 0.000 0.029 0.229 0.600 0.129 0.443 0.014 0.486 0.329 0.025 0.000
sample size 70 70 70 70 70 70 70 70 70 40 40

“dotted” (Fig. 15A-F) and “shaded” (Fig.
15G-J). The two patterns never occurred in
one specimen, but both were found at almost
all the localities. Within the two patterns, wide
ranges of continuous variability were ob-
served (Fig. 15). The “shaded” pattern cov-
ered a larger or smaller part of the mantle,
forming irregular, pigmented patches of vari-
ous size or covering almost all the surface.
The “shaded” pigmentation was often inten-
sive or very intensive, covering the major part
of the mantle. Also the “dotted” pattern
showed a wide variability: from minute dots to
big, black spots, which usually were approxi-
mately circular or oval.

In addition to external morphological poly-
morphisms, we have also found polymorphic
characters in the reproductive organs (Fig.
16; Table 3), which have been described and
figured by Shileiko (1978: figs. 52-53, р.
126), although his drawing is not adequately
detailed. The colour of the penis, atrium, dart

sac and oviduct may be whitish or pinkish
(Fig. 16). This colour variation is observed in
mature snails and specimens fixed in ethanol,
frozen in liquid nitrogen, and fresh, indicating
that it is not an artifact of preservation. There
was also a black pigment on the reproductive
organs (Fig. 16); it occurred in grains, more or
less dense and covering a variable part of the
penis and atrium.

The mucous gland of the reproductive or-
gans (Figs. 16, 17) is divided externally into
lobes (Fig. 17B-D, H-K) or not (Fig. 17A,
E-G). Also, the outlet of the gland was vari-
able, consisting of either externally distin-
guishable, separate ducts (Fig. 17E-K), or a
single, fused outlet (Fig. 17A-D).

The frequency distributions of all polymor-
phic characters in the studied populations
were tested for normality. For each pair of
polymorphic characters, Pearson's product-
moment correlation coefficients (Sokal &
Rohlf, 1987) were calculated between the fre-

POLYMORPHISM IN BRADYBAENA FRUTICUM 377

pb: pb

FIGS. 2-14. Shell colour polymorphism in Bradybaena fruticum: 2, 5, 10, 12, 13—dark morph; 4, 6, 7, 11,
14—light morph; 3—band adjacent to lip; 8, 9—atypical double equatorial band; (b—brown, c—chestnut,
i—ivory, pb—pale brown, pc—pale chestnut, y—yellowish); 12, 13—soft parts visible through shell wall.

quencies in all populations. Significant corre-
lations were found only between the equato-
rial band on the shell and the band adjacent to
the lip (r = 0.6149, p < 0.05); the dark shell
and the band adjacent to the lip (r = 0.5364,
p < 0.10); the dark shell and the shaded pig-

mentation of the mantle (r = 0.7296, p <
0.01); the yellow pigment on the mantle and
the black pigment on the reproductive organs
(r = 0.6589, p = 0.02); the lobate mucous
gland and the multiple outlets of the gland (r
= 0.7687, p < 0.005).

FALNIOWSKI, КОЙК & SZAROWSKA

FIG. 15. Mantle pigmentation polymorphism: A-F—dotted black, G-J—shaded black. Black pigment (5)
represented by black, yellow pigment (y) represented by shadings (minute dots).

Enzymatic Polymorphism

For all the individuals studied, the enzyme
ACP separated into three diffuse but well-re-
solved bands. Such a pattern is characteristic
of a dimeric enzyme having two monomorphic
loci, with hybrids as the middle band. This
interpretation is consistent with general com-
ments of Richardson et al. (1986). Similar con-
clusions concern ALP. AAT appeared as a
single diffuse band in all individuals screened
for this enzyme. The G3PDH activity appeared
on gels as multiple, sharp bands concentrated
in a relatively narrow zone (= presumptive
locus), showing no detectable variation among
individuals, so the locus was regarded as
monomorphic. Similar remarks concern
HBDH. For both enzymes, there was some
indication of a second locus but a very low
activity.

For “Cap,” staining with L-leucine-B-naph-
thylamide, two of probably many peptidase
loci were observed. The loci detected are per-
haps related to the human E and S peptidases
(Harris & Hopkinson 1976). According to Ri-
chardson et al. (1986), the PEP-E of verte-
brates is identical with CAP. For both loci, a

monomeric structure of CAP is evident, as in
other snails (Johnson et al., 1977; Rudolph &
Burch, 1987, 1989) in which one locus (Ru-
dolph & Burch 1987; Emberton, 1988; Wood-
ruff et al., 1988), two loci (Ayala et al., 1973;
Selander & Kaufmann, 1975; Johnson et al.,
1977; Kitikoon, 1982; Hoagland, 1984; Brown
& Richardson, 1988) or three loci (G. M. Davis
et al., 1988) have been detected.

MDH separated into two rather diffuse
zones (= presumptive loci) consistent with a
dimeric structure and two loci (Harris & Hop-
kinson, 1976; Wurzinger, 1979; Hoagland,
1984; Richardson et al., 1986; Rudolph &
Burch, 1987; Emberton, 1988; G. M. Davis et
al., 1988; Mulvey et al., 1988; Mimpfoundi &
Greer, 1990a). Weak bands of PGDH activity
were observed but gels were still scorable
and interpretable, showing a single polymor-
phic locus. A dimeric structure of PGDH has
been proposed from studies on vertebrates
(Richardson et al., 1986; Harris & Hopkinson,
1976) and on Stagnicola (Rudolph & Burch,
1987). A single, polymorphic locus of dimeric
XO was found, though the overall activity was
low.

Allele frequencies, sample sizes, mean

POLYMORPHISM IN BRADYBAENA FRUTICUM 379

FIG. 16. Reproductive organs of Bradybaena fruticum (A. А fragment with plural outlets of mucous glands,
divided into four separate lobes; В. Cross section of the penis); at—atrium, bc—bursa copulatrix, bt—dart
sac, dbc—duct of bursa copulatrix, dh—ductus haermaphroditicus, ga—albuminoid gland, gm—mucous
gland (glands), gp—gonoporus, ov—oviduct, p—penis, ut—uterus, v—vagina, vd—vas deferens. Pigmen-
tation of penis and atrium represented by coarse dotting; colour polymorphism represented by w/p (white or

pink)

numbers of alleles per locus, proportions of
polymorphic loci, and proportions of heterozy-
gosities both observed and estimated for all
the studied populations are given in Table 4.
The proportion of polymorphic loci was rela-
tively low (Ртеап = 36.4%) for a polymorphic
helicoid species, and widely variable among
the populations. In several cases, a population
’ was fixed for one allele at a given locality, while

polymorphic at the same locus at another lo-
cality. In all but one observed cases, chi-
squared tests of genotype frequencies pro-
vided no evidence for a significant departure
from random mating expectations (p=0.10). А
significant excess of heterozygotes was found
in the САР, locus (Table 4).

No relation of enzyme polymorphism to any
morphological polymorphism was found.

380

FALNIOWSKI, KOZIK & SZAROWSKA

FIG. 17. Schematic representation of mucous gland polymorphism: А. Gland not lobate, outlet fused; B-D.
Сапа lobate, outlet fused; E-G. Сапа not lobate, outlet divided; Н-К. Gland lobate, outlet divided.

Differences Between Local Populations

To illustrate distances between studied
populations based on morphological polymor-
phism frequencies, the CONTML technique
has been used (Fig. 18). The resulting tree
shows numerous relatively long distances be-
tween closely situated populations. The same
technique has been used for enzyme allele
frequencies (Fig. 19). The resulting grouping is
different, especially in linking populations 10
and 11, but also in this case the distance be-
tween, e.g., populations 1 and 6 (within the
Ojcöw National Park) is not much longer than
the distances between 1 and 10 or 1 and 11.

For each pair of populations, Nei’s dis-
tances and Cavalli-Sforza & Edwards’s arc
distances were calculated (Table 5). The high
Nei’s distance values between populations 10
and 11 and the majority of the others on the
one hand, and the very low value of the dis-
tance between the geographically distant
populations 10 and 11 on the other, are
noteworthy. Cavalli-Sforza & Edwards’s arc
distances were used to compute a Fitch-Mar-
goliash additive tree (Fig. 20) showing a pat-
tern similar to Nei’s distances; the distance
between populations 1 and 6, as well as the
ones between all the Polish populations, were
longer than the distance between populations
10 or 11 and population 7. Finally, a Cavalli-
Sforza & Edwards least square tree with con-
temporary tips (Fig. 21) was computed. It

shows even better the same pattern: popula-
tions 10 and 11 are equally distant from all the
others, while within the Polish group of popu-
lations there is practically no geographic pat-
tern.

For each pair of populations, Spearman’s
rank correlation coefficients between genetic
distances and geographic distances (in km)
were calculated. For Nei’s distance the corre-
lation was not significant, while for Cavalli-
Sforza & Edwards's arc distance the correla-
tion was significant (r = —0.7060, p < 0.001),
but when the most distant populations 10 and
11 were excluded, it was not significant.

DISCUSSION

In Bradybaena fruticum all the three types of
external colouration polymorphism described
by Clarke et al. (1978) (mantle and body, shell
colour, shell banding) can be distinguished. In
another bradybaenid, B. similaris (Férussac,
1821), brown shell colour is dominant to yel-
low, a single banded pattern is dominant to
unbanded, and the two loci are linked (Komai
& Emura, 1955: cited in Clarke et al., 1978).
The dominance of a dark shell and a banded
shell seems common in polymorphic terrestrial
pulmonates (e.g. Clarke et al., 1978; Cain,
1983: the references therein). Khokhutkin
(1979, 1984) and Khokhutkin & Lazareva
(1975, 1983, 1987) considered the single

RE As

POLYMORPHISM IN BRADYBAENA FRUTICUM 381
TABLE 4. Allele frequencies in all polymorphic loci studied
locality
locus/
allele 1 2 3 4 5 6 7 8 9 10 11
CAP, а 0.647 0.343 0.437 0.732 0.036 0.457 0.176 0.167 0.109 0.000 0.000
b 0.353 0.657 0.563 0.268 0.964 0.543 0.824 0.833 0.891 1.000 1.000
CAP, а 0.000 0.157 0.125 0.027 0.196 0.300 0.191 0.183 0.094 0.138 0.129
Ь 0.779 0.629 0.719 0.491 0.340 0.557 0.588 0.567 0.609 0.500 0.532
с 0.221 0.214 0.156 0.482 0.464 0.143 0.221 0.250 0.297 0.362 0.339
MDH, а 0.029 0.929 0.047 0.848 0.268 0.157 0.015 0.517 0.594 1.000 0.661
b 0.000 0.000 0.000 0.000 0.000 0.114 0.000 0.000 0.172 0.000 0.000
с 0.971 0.071 0.953 0.152 0.732 0.729 0.985 0.483 0.234 0.000 0.339
MDH, а 0.000 0.000 0.000 0.009 0.018 0.957 0.000 0.000 0.516 0.000 0.000
b 0.985 0.443 1.000 0.911 0.982 0.043 0.029 0.783 0.484 0.000 0.000
с 0.015 0.557 0.000 0.080 0.000 0.000 0.971 0.217 0.000 1.000 0.726
4 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.274
PGDH а 0.000 0.000 0.000 0.000 0.000 0.186 0.000 0.150 0.063 0.000 0.000
b 1.000 1.000 1.000 1.000 1.000 0.814 1.000 0.850 0.937 1.000 1.000
XO a 0.235 0.486 0.031 0.304 0.500 0.243 0.300 0.367 0.406 0.000 0.000
Ь 0.765 0.228 0.531 0.339 0.286 0.757 0.286 0.283 0.187 0.172 0.000
с 0.000 0.286 0.438 0.357 0.214 0.000 0.414 0.350 0.407 0.828 1.000
Nmin 34 35 32 56 28 35 34 30 32 29 31
en 1.38 1.54 1.46 1.61 1.54 1.61 1.54 1.61 1.69 1.23 1.31
P% 38.5 38.5 30.8 38.5 38.5 46.1 38.5 46.1 46.1 15.4 23.1
SNS) 0.096 0.172 0.119 0.149 0.134 0.174 0.119 0.201 0.195 0.068 0.110
5 0.045 0.067 0.056 0.060 0.065 0.060 0.062 0.067 0.069 0.047 0.057
МС) 0.103 0.182 0.134 0.159 0.134 0.166 0.119 0.205 0.197 0.069 0.111
$ 0.050 0.070 0.065 0.064 0.065 0.057 0.062 0.068 0.069 0.048 0.057
Hear, (e) 0.457 0.451 0.492 0.392 0.069 0.496 0.290 0.278 0.194 0.000 0.000
САР: (о) 0.559 0.571 0.687 0.518 0.071 0.400 0.294 0.333 0.219 0.000 0.000

Nmin Minimum number of specimens screened at given locality; Ame, mean number of alleles per locus (including
monomorphic loci); Р.—ргоройюп of polymorphic loci; Hmean—mean individual heterozygosity (including monomorphic
loci: Hmeancey— expected from Hardy & Weinberg equilibrium, Hmean(oy Observed; Hcap,—heterozygosity in CAP, locus:
Heap, (ey expected, Hcap,(oy— observed; SE—standard error; monomorphic loci: ACP,, АСР, ALP,, ALP,, AAT, G3PDH,
and G3PDH;..

banded pattern to be recessive to unbanded
one. On the other hand, the existence of atyp-
ically banded specimens, as well as the ob-
served change in shell colour from reddish
brown to yellow in one specimen, and wide
ranges of continuous colour variability within
the two morphs distinguished seem to suggest
that the inheritance mechanism of these poly-
morphic characters may be more complicated,
with numerous loci being involved; similar re-
marks concerning Theba pisana (О. Е. Müller,
1774) were given by Cowie (1984). Also en-
vironmental effects on the expression of the
shell colour genes cannot be excluded.

In our populations of B. fruticum, the pro-
portion of polymorphic enzyme loci was
15.4%-46.1%, mean 36.4%. Inthe majority of
marine molluscs, the proportion is between 30
and 50% (Berger, 1983); in the freshwater
Anodonta, 11-36%, depending on species
(Kat, 1983); in the brackish water Hydrobia,
13-23% (G. M. Davis et al., 1988); in fresh-

water gastropods, 14-62% (Brown & Rich-
ardson, 1988; Woodruff et al., 1988). In land
snails, it varies form O to 100% (Nevo, 1978).
For example, in Australian camaenids it
ranges from 19 to 71% (Woodruff & Solem,
1990), from 65 to 80% in Partula (Johnson et
al., 1977), but reaches only about 4% in
Liguus (Hillis et al., 1987). In Cepaea, it is
about 60% (Clarke et al., 1978). Therefore,
the value found in B. fruticum is rather low for
a polymorphic species.

Heterozygosity in B. fruticum in this study
varied from 0.069 to 0.205, mean 0.144. The
values are similar to the ones given by Nevo
(1978) for Theba (0.054—0.165), Brown & Ri-
chardson (1988) for Cepaea nemoralis
(0.134), and by Woodruff & Solem (1990) for
camaenids (0.08-0.24). On the other hand,
in Bradybaena similaris it is lower (0.083:
Brown & Richardson (1988). In land slugs,
average heterozygosity varies among spe-
cies (0-0.19: Foltz et al., 1984), but also

382 FALNIOWSKI, КОЙК & SZAROWSKA

FIG. 18. Distances between populations, based оп morphological character states frequencies, generated
by maximum likelihood method for continuous characters (CONTML). Distances drawn proportionally. Ln
Likelihood = 109.47078; examined 4,770 trees; 1-11, locality numbers, as in text.

0.100

3 q _-- _ ———————— —— E

FIG. 19. Distances between populations, based on allele frequencies, generated by maximum likelihood
method for continuous characters (CONTML). Distances drawn proportionally. Ln Likelihood = 83.33746;

examined 3,724 trees; 1-11, as in Fig. 18.

among conspecific populations from various
parts of the range (0.006—0.19, Milax: Foltz et
al., 1984; means: 0.04-0.19, Oncomelania:
Woodruff et al., 1988). In Partula, it ranges
from 0.13 to 0.17 (Johnson et al., 1977),
whereas in the closely related Samoana, it
does not exceed 0.002 (Johnson et al., 1986).

The heterozygote proportion did not depart
significantly from Hardy-Weinberg equilib-
rium, with the exception of the CAP, locus at
localities 1—4, 6, 8, and 9. At locality 6, there
was a heterozygote deficiency, while at all the
other listed localities, a heterozygote excess
(Table 5). Heterozygote deficiency is com-
monly observed in molluscan populations, es-
pecially in bivalves (Berger, 1983; Zouros &
Foltz, 1984; Hillis et al., 1987; Brown & Rich-
ardson, 1988; Hillis, 1989; Mimpfoundi &
Greer, 1990b, c). Heterozygote excess is
much less common, but has been observed

[Selander & Kaufman, 1975, in introduced
Helix aspersa (O. F. Miller, 1774) popula-
tions; Berger, 1983, in marine molluscs]. Pos-
sible explanations of the observed excess are
assortive mating or heterozygote advantage,
or Wahlund effect. The heterozygosity data
seem to indicate that there is neither self-fer-
tilization nor inbreeding in B. fruticum.

The theoretical background to the dis-
crodance between the patterns of morpholog-
ical variation and enzymatic polymorphism is
clear (Lewontin, 1984; Cheverud, 1988).
Such inconsistency between molecular and
morphological data sets seems common (e.g.
Johnson et al. 1977, 1986; Hillis et al., 1987;
Woodruff & Solem, 1990; Murray et al., 1991).
This is confirmed when comparing the trees
based on morphological (Fig. 18) and enzy-
matic (Figs. 19—21) characters. For example,
Slovakian populations 10 and 11 are close to

POLYMORPHISM IN BRADYBAENA FRUTICUM

383

TABLE 5. Genetic distances between studied populations (below diagonal: Nei distances, above

diagonal: Cavalli-Sforza & Edwards arc distances)

1 2 3 4 5
1 Sr 0.467 0.142 0.294 0.265
2 0.136 ji 0.411 0.118 0.289
3 0.017 0.120 ES 0.256 0.171
4 0.083 0.043 0.081 ie 0.238

5 0.067 0.086 0.045 0.083 a
6 0.101 0.149 0.110 0.159 0.131
if 0.126 0.101 0.100 0.180 0.103
8 0.069 0.034 0.043 0.049 0.019
9 0.121 0.043 0.089 0.068 0.050
10 0.282 0.053 0.215 0.146 0.173
11 0.237 0.069 0.160 0.148 0.135

locality

6 Y 8 9 10 11
0.476 0.482 0.302 0.489 1.192 1.139
0.661 0.323 0.108 0.296 0.307 0.423
0.519 0.409 0.197 0.357 0.923 0.780
0.593 0.514 0.150 0.253 0.611 0.688
0.483 0.425 0.113 0.189 0.790 0.708

oT 0.625 0.510 0.281 1.109 1.116
0.125 и 0.266 0.571 0.498 0.360
0.114 0.080 cn 0.201 0.477 0.473
0.083 0.115 0.027 Fr 0.661 0.657
0.235 0.103 0.099 0.097 Pd 0.158
0.204 0.071 0.081 0.081 0.019 i

FIG. 20. Distances between populations, based on Cavalli-Sforza and Edwards’ arc distances, generated by
Fitch-Margoliash’s method (FITCH). Distances drawn proportionally. Sum of squares = 3.40454; average

percent standard deviation =

each other and distant from the Polish popu-
lations in all the trees based on molecular
data (Figs. 19-21), but not in the tree based
on morphological characters (Fig. 18).

Nei’s distance among populations in B. fru-
ticum ranged form 0.017 to 0.282. The latter
value exceeds the one characteristic of the
subspecies level in the Drosophila willistoni
group (Ayala, 1975). The value of 0.019 be-
tween populations 10 and 11 (150 km away
form each other), compared with values over
0.1 within a few km distance, is noteworthy.
From among the reasons for the observed rel-
atively high values of Nei’s distances, the fix-
ation on one allele at some loci in some pop-
ulations has to be mentioned. Cavalli-Sforza
& Edwards's arc distance shows a similar pic-
ture.

Woodruff et al. (1988) list Nei’s distances

17.75488; 3,242 trees examined; 1-11, as in Fig. 18.

for various molluscan species. They point out
that typically within molluscan species the
value does not exceed 0.1 between local pop-
ulations, whereas interspecific differences for
congeners are within the range 0.2—0.6 (e.g.
for Cerion, Triodopsis). п their study on On-
comelania, the distances between rather
close populations were within the range
0.002—0.104, but between the Philippines
and the Chinese populations reached 0.648,
within the same species. In Biomphalaria,
with growing geographic distance, conspe-
cific populations differed by distance values of
0.00-0.18. In Hydrobia, distances within a
species were not higher than 0.013 (G. M.
Davis et al., 1988). The highest value of in-
terpopulation Nei’s distance within a species
of snail is 0.701 noted in Melanoides tubercu-
lata by Livshits et al. (1984), but between par-

384 FALNIOWSKI, KOZIK & SZAROWSKA
1
3
5
9
2
8
4
7
6
10
11
0.359 0.5 0.2 0.1 0

FIG. 21. Distances between populations, based on Cavalli-Sforza and Edwards's arc distances, generated
by Fitch-Margoliash's method with contemporary tips (KITSCH). Distances drawn proportionally. Sum of
squares = 3.256; 5,229 trees examined; 1-11, as in Fig. 18.

thenogenetically reproducing populations.
The highest intraspecific value reported for
sexually reproducing gastropods (0.63) is the
one between Italian and British populations of
Cepaea nemoralis (Johnson et al., 1984). On
the other hand, the distance found between
two Partula species, about 8,000 km distant
from each other, is 0.125 (Johnson et al.
1977). In Samoana, Nei’s distances between
species varied from 0.004 to 0.602 (Johnson
et al., 1986) and between Cristilabrum spe-
cies (Woodruff & Solem, 1990) from 0.00 (!) to
0.199, but the average distance for five spe-
cies was only 0.081; within camaenids the in-
tergeneric distances were 0.27—0.50 (Wood-
ruff & Solem, 1990). The above data clearly
indicate that there is no general rule concern-
ing genetic distances in snails. The values of
Nei’s distances in B. fruticum are relatively
high for local populations, and in numerous
cases relatively high values of genetic dis-
tance observed correspond with rather low
values of geographic distance.

The allozyme polymorphism shows more
geographic pattern (Figs. 19-21) than the
morphological variation (Fig. 18), which is in
agreement with Makeeva (1987). However,
Makeeva (1987), Makeeva & Matiokin (1987),
Khokhutkin & Lazareva (1975, 1983, 1987)
and Khokhutkin (1984) report a hierarchical
pattern of population structure in B. fruticum.
They describe the species as composed of
semi-isolated, small panmictic colonies (the

latter confirmed by our study). There are
some differences among local panmictic
units, but always less pronounced within a re-
gion than among regions; the main compo-
nent of interpopulation differences is a mac-
rogeographic clinal one. In our study, we have
not observed such a hierarchical structure. Al-
though localities 10 and 11 are both geneti-
cally and geographically distant from the
other localities, they are genetically much
similar to each other, but the geographic dis-
tance between localities 10 and 11 is greater
than the ones between 10 and each of local-
ities 1-9. At the same time, the genetic dis-
tance between populations 1 and 6, which are
situated very closely to each other, is not
much shorter than the genetic distance be-
tween 1 and 10 or 1 and 11. The observed
genetic similarity of the Slovakian populations
10 and 11 is difficult to explain, the more that
B. fruticum is rather uncommon in Slovakia
(Steffek, personal communication). On the
other hand, the relatively high values of ge-
netic distance between each of the two Slo-
vakian populations and each of the Polish
ones can easily be explained. The high Tatra
Mountains at the border between Poland and
Slovakia form an effective barrier for this low-
land snail.

К seems that in В. fruticum, the “stepping
stone” model of Wright (1965) rather than the
“isolation by distance” model (Wright, 1965)
can be applied to describe macrogeographic

——

—

POLYMORPHISM IN BRADYBAENA FRUTICUM

differentiation. The gene and morphological
differentiations we observed were in general
significant among populations, but negligible
among regions. А similar pattern has been
observed in Hydrobia (G. M. Davis et al.,
1988). On the other hand, in what we ob-
served there still was more geographic pat-
tern than was found in parthenogenetic pop-
ulations of Melanoides (Livshits et al., 1984),
in which there was no correlation between the
genetic and geographic distances.

As stressed by Goodhart (1962, 1963) and
Selander & Kaufman (1975) for populations of
Cepaea and Helix respectively, the genetic
structure of a population is a result of the in-
teraction of deterministic and stochastic pro-
cesses. Although in Poland B. fruticum is a
common species in general, it becomes un-
common or rare in the mountainous regions
of South Poland (Riedel, 1988). It is never
found in the mountainous beech forest, which
is a typical natural biotope of the Ojcöw Na-
tional Park. The deforestation caused by man
changed the environment to a one that is
more suitable for the snail. On the other hand,
pieces of arable land are barriers for В. fruti-
cum. The successive deforestation and
changes in agricultural activities have re-
sulted in the observed pattern of small spots
of biotopes inhabited by the species. This,
along with the relatively low densities of the
populations, have resulted in the existence of
several demes which are almost completely
isolated and consist of relatively few individu-
als. Such populations’ genetic structure is the
most dependent on stochastic processes and
this can be an explanation of the observed
high genetic distances between some of the
closely situated Ojcöw populations.

Within the area of the Ojcöw National Park,
there are some barriers to dispersal, such as
streams or roads, or beech forest. For exam-
ple, populations 2 and 4, which are situated
not very closely to each other and are sepa-
rated by a river and a road, are genetically
similar, whereas the very closely situated
populations 3 and 4, separated by a beech
forest, are much different genetically. The
comparison of several genetic distances
within the Ojcöw National Park seems to in-
dicate that for the snail a beech forest is a
much more effective barrier than a river, a
stream or a road. It is not clear in general how
effective such barriers must be to prevent or

| _ Strongly limit gene flow. Even a small river

may be a true barrier for land snails (e.g. Hillis
et al., 1987). On the other hand, Grant & Utter

385

(1988) observed considerable genetic differ-
ences among 12 breeding colonies of the ma-
rine, intertidal whelk Nucella, distributed
within a distance of 100 m of a shore, where
no barrier of any kind was found. They ac-
knowledged that random genetic drift among
small subpopulations was a source of differ-
entiation, and the distinctness of the colonies
had a behavioural background. Especially ju-
venile site fidelity and homing behaviour lim-
ited gene flow among colonies. Little is known
about the behaviour of B. fruticum, but obser-
vations of Zeifert & Shutov (1979) and Zeifert
(1987) suggest both homing and juvenile site
fidelity in the species. These authors also re-
ported variation in mobility: snails inhabiting
microhabitats of milder microclimatic condi-
tions in winter stayed at the same place,
whereas the ones inhabiting less suitable mi-
crohabitats migrated from their winter shelters
to their feeding territories. Such migration
may increase gene flow, resulting in a range
of patterns of microgeographic differentiation.
All the above factors, coupled with a spotty
pattern of distribution and with barriers of var-
ious kind and efficiency may explain the ob-
served pattern.

ACKNOWLEDGMENTS

We are deeply indebted to the Stefan Ba-
tory Trust, Oxford, for a one-month scholar-
ship that enabled the senior author to make a
literature survey, to Dr. Joe Felsenstein for
supplying us with his PHYLIP package, and to
Dr. Jozef Steffek (Banska Stiavnica) for his
assistance in a field work in Slovakia. We are
especially grateful to Dr. G. M. Davis and
three anonymous reviewers for suggestions
and criticism of an earlier version of this
manuscript.

The study was supported by a grant form
the Polish Ministry of Education DNS—P/01/
006/90-2.

LITERATURE CITED

AYALA, F. J., 1975, Genetic differentiation during
the speciation process. Pp. 1-78 in: T. DOBZHAN-
SKY, М. К. HECHT & W. С. STEERE, eds., Evolu-
tionary biology, Vol. 8. Plenum Press, New York
& London.

AYALA, Е. J., D. HEDGECOCK, С. $. ZUMWALT
& J. W. VALENTINE, 1973, Genetic variation in
Tridacna maxima, an ecological analog of some

386

unsuccessful exolutionary lineages. Evolution,
27: 177-191.

BABA, K., 1985, Investigation of the growth rate of
two terrestrial snails: Bradybaena fruticum (О. Е.
Müller) and Euomphalia strigella (Draparnaud).
Soosiana, 13: 79-88.

BERGER, E. M., 1983, Population genetics of ma-
rine gastropods and bivalves. Pp. 563-595, in:
W. D. RUSSELL-HUNTER, ed., The Mollusca, Vol. 6.
Academic Press, New York.

BROWN, К. М. & Т. D. RICHARDSON, 1988, Ge-
netic polymorphism in gastropods: a comparison
of methods and habitat scales. American Mala-
cological Bulletin, 6: 9-17.

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.
Academic Press, New York.

CAVALLI-SFORZA, L. & A. W. F. EDWARDS,
1967, Phylogenetic analysis: models and estima-
tion procedures. Evolution, 21: 550-570.

CHEVERUD, J. M., 1988, A comparison of genetic
and phenotypic correlations. Evolution, 42: 958—
968.

CLARKE, B., W. ARTHUR, T. HORSLEY & D. T.
PARKIN, 1978, Genetic variation and natural se-
lection in pulmonate molluscs. Pp. 219-270, in:
\. FRETTER & J. PEAKE, eds., Pulmonates, Vol. 2a.
Academic Press, New York.

COOK, L. M., 1991, Genetic and ecological diver-
sity: the sport of nature. Chapman & Hall, Lon-
don, etc.

COWIE, R. H., 1984, Ecogenetics of Theba pisana
(Pulmonata: Helicidae) at the northern edge of its
range. Malacologia, 25: 361-380.

DAVIS, В. J., 1964, Disc electrophoresis—ll.
Method and application to human serum pro-
teins. Annals of New York Academy of Sciences,
121: 404—427.

DAVIS, G. M., У. FORBES 4 G. LOPEZ, 1988,
Species status of northeastern American Hydro-
bia (Gastropoda: Prosobranchia): ecology, mor-
phology and molecular genetics. Proceedings of
the Academy of Natural Sciences of Philadel-
phia, 140: 191-246.

EDWARDS, A. W. F. 8 L. L. CAVALLI-SFORZA,
1964, Reconstruction of evolutionary trees. In: V.
H. Heywoo 8 J. MCNEIL, eds., Phenetic and
phylogenetic classification. Systematics Associa-
tion Volume No. 6, Systematics Association, Lon-
don.

EMBERTON, К. C., 1988, The genitalic, allozymic,
and conchological evolution of the eastern North
American Triodopsinae (Gastropoda: Pulmo-
nata: Polygyridae). Malacologia, 28: 159-273.

EMBERTON, K. C., 1989, Retraction/extension and
measurement error in a land snail: effects on sys-
tematic characters. Malacologia, 31: 157-174.

FELSENSTEIN, J., 1981, Evolutionary trees from
gene frequencies and quantitative characters:
finding maximum likelihood estimates. Evolution,
35: 1229-1242.

FELSENSTEIN, J., 1984, Distance methods for in-

FALNIOWSKI, KOZIK & SZAROWSKA

ferring phylogenies: a justification. Evolution, 38:
16-24.

FELSENSTEIN, J., 1985, Phylogenies from gene
frequencies: a statistical problem. Systematic Zo-
ology, 34: 300-311.

FELSENSTEIN, J., 1990, PHYLIP Manual Version
3.3. University Herbarium, University of Califor-
nia, Berkeley, California.

FITCH, W.M. & E. MARGOLIASH, 1967, Construc-
tion of phylogenetic trees. Science, 155: 270-
284.

FOLTZ, D. W., H. OCHMAN & R. K. SELANDER,
1984, Genetic diversity and breeding systems in
terrestrial slugs of the families Limacidae and Ari-
onidae. Malacologia, 25: 593—605.

GOODHART, С. B., 1962, Variation in a colony of
the snail Cepaea nemoralis L. Journal of Animal
Ecology, 31: 207-237.

GOODHART, С. B., 1963, ‘Area effects’ and non-
adaptive variation between populations of Ce-
paea (Mollusca). Heredity, 18: 459-465.

GRANT, W. 5. & Е. М. UTTER, 1988, Genetic het-
erogeneity on different geographic scales in Nu-
cella lamellosa (Prosobranchia, Thaididae). Ma-
lacologia, 28: 275-287.

HARRIS, Н. & D. A. HOPKINSON, 1976, Hand-
book of enzyme electrophoresis in human genet-
ics. Amsterdam, North Holland.

HILLIS, D. M., 1989, Genetic consequences of par-
tial self-fertilization on populations of Liguus fas-
ciatus (Mollusca,: Pulmonata: Bulimulidae).
American Malacological Bulletin, 7: 7-12.

HILLIS, D. M., D. S. ROSENFELD & M. SANCHEZ,
Jr., 1987, Allozymic variability and heterozygote
deficiency within and among morphologically
polymorphic populations of Liguus fasciatus
(Mollusca: Pulmonata: Bulimulidae). American
Malacological Bulletin, 5: 153-157.

HOAGLAND, K. E., 1984, Use of molecular genet-
ics to distinguish species of the gastropod genus
Crepidula (Prosobranchia: Calyptraeidae). Mala-
cologia, 25: 607-628.

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

JOHNSON, М. S., J. MURRAY & В. CLARKE,
1986, High genetic similarities and low heterozy-
gosities in land snails of the genus Samoana
from the Society Islands. Malacologia, 27: 97-
106.

JOHNSON, М. S., О. С. STINE & J. MURRAY,
1984, Reproductive compatibility despite large
scale genetic divergence in Cepaea nemoralis.
Heredity, 53: 655-665.

KAT, P. W., 1983, Genetic and morphological di-
vergence among nominal species of North Amer-
ican Anodonta (Bivalvia: Unionidae). Malacolo-
gia, 23: 361-374.

KERNEY, М. P., В. А. D. CAMERON & J. H. JUNG-
BLUTH, 1983, Die Landschnecken Nord- und
Mitteleuropas. Verlag Paul Parey, Hamburg und
Berlin.

KHOKHUTKIN, I. M., 1979, Inheritance of banding

POLYMORPHISM IN BRADYBAENA FRUTICUM

in natural populations of the land snail Brady-
baena fruticum (Mull.). Genetika (Moscow), 15:
868-871 [in Russian, English abstract].

KHOKHUTKIN, I. M., 1984, Organization and vari-
ation of the polymorphic structure of species and
land molluscs. Journal of General Zoology, 45:
615—623.

KHOKHUTKIN, |. М. 8 А. I. LAZAREVA, 1975,
Polymorphism and concealing coloration of land
mollusc populations. Zhurnal Obshchei Biologii,
36: 863—868 [in Russian, English abstract].

KHOKHUTKIN, I. M. & A. I. LAZAREVA, 1983,
Biotopic and geographic changes in the polymor-
phic structure of Bradybaena fruticum (Mull). Pp.
62—63, in: Ya. |. STAROBOGATOV, А. N. GOLIKOV &
|. М. LIKHAREvV, eds., Mollusca Systematics, Ecol-
ogy and Bionomic distribution. Seventh meeting
on the Investigations of Mollusks. Academiia
Nauk USSR, Zoologicheskii Institut, Leningrad.

KHOKHUTKIN, |. М. & A. 1., LAZAREVA, 1987,
Variability structure in land snails: adaptation on
populational and superspecific levels. Pp. 441-
442, in: YA. |. STAROBOGATOV, A. N. Gotikov, & I.
М. LIKHAREV, eds., Molluscs, Results and Per-
spectives of Investigation. Eighth meeting on the
Investigations of Molluscs, Abstracts of Commu-
nications, Academiia Nauk USSR Zoologicheskii
Institut, Leningrad.

KITIKOON, V., 1982, Studies on Tricula aperta and
related taxa, the snail intermediate hosts of
Schistosoma mekongi. V. Electrophoretic stud-
ies. Malacological Review, 15: 43-57.

LEWONTIN, R. C., 1984, Detecting population dif-
ferences in quantitative characters as opposed to
gene frequencies. American Naturalist, 123:
115-124.

LIVSHITS, G., L. FISHELSON, & G. S. WISE,
1984, Genetic similarity and diversity of parthe-
nogenetic and bisexual populations of the fresh-
water snail Melanoides tuberculata (Gastropoda:
Prosobranchia). Biological Journal of the Lin-
nean Society, 23: 41-54.

MAKEYEVA, V. M., 1987, The role of the anthro-
pogene barriers in the formation of the population
structure of Bradybaena fruticum (Mull.) in the
Moscow region. Pp. 463—464, in: Ya. |. STAR-
OBOGATOV, А. М. GOLikov € |. М. LIKHAREV, eds.,
Molluscs, Results and Perspectives of Investiga-
tion. Eighth meeting on the Investigations of Mol-
luscs, Abstracts of Communications. Akademiia
Nauk USSR Zoologisheskii Institut, Leningrad.

MAKEYEVA, V. M. & P. V. MATIOKIN, 1987, The
role of the natural selection-in the genetic popu-
lation structure of Bradybaena fruticum (Mull.) in
Moscow region. Pp. 475-476, in: Ya. |. Star-
OBOGATOV, А. N. GoLIKOV & I. M. LIKHAREV, eds.,
Molluscs, Results and Perspectives of Investiga-
tion. Eighth meeting on the Investigations of Mol-
luscs, Abstracts of Communications. Akademiia
Nauk, USSR, Zoologischeskii Institut, Leningrad.

MIMPFOUNDI, R. & G. J. GREER, 1990a, Allo-
zyme comparisons and ploidy levels among spe-
cies of the Bulinus truncatus/tropicus complex

387

(Gastropoda: Planorbidae) in Cameroon. Journal
of Molluscan Studies, 56: 63-68.

MIMPFOUNDI, R. & G. J. GREER, 1990b, Allo-
zyme variation among populations of Bulinus for-
Skalii (Ehrenberg, 1831) (Gastropoda: Planor-
bidae) in Cameroon. Journal of Molluscan
Studies, 56: 363-371.

MIMPFOUNDI, R. & G. J. GREER, 1990c, Allo-
zyme Variation among populations of Bi-
omphalaria pfeifferi (Krauss, 1848) (Gastropoda:
Planorbidae) in Cameroon. Journal of Molluscan
Studies, 56: 461—467.

MULVEY, M., М. С. NEWMAN & D. $. WOOD-
RUFF, 1988, Genetic differentiation among West
Indian populations of the schistosome-transmit-
ting snail Biomphalaria glabrata. Malacologia, 29:
309-317.

MURPHY, В. W., J. W. SITES, Jr., D. G. BUTH 4 С.
H. HAUFLER, 1990, Proteins |: Isosyme Electro-
phoresis. Pp. 45-126, in: D. M. Нш$ € С.
Moritz, eds., Molecular systematics. Sinauer
Associates, Inc. Publishers, Sunderland, Massa-
chusetts.

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

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

NEI, M., 1978, Estimation of average heterozygos-
ity and genetic distance from a small number of
individuals. Genetics, 89: 583—590.

NEVO, E., 1978, Genetic Variation in Natural Pop-
ulations: Patterns and Theory. Theoretical Pop-
ulation Biology, 13: 121-177.

OXFORD G. S., 1973a, The genetics of Cepaea
esterases. 1. Cepaea nemoralis. Heredity, 30:
127-139.

OXFORD G. S., 1973b, The biochemical properties
of esterases in Cepaea (Mollusca: Helicidae).
Comparative Biochemistry and Physiology, 45B:
529-538.

OXFORD С. $., 1973c, Molecular weight relation-
ships of the esterases in Cepaea nemoralis and
Cepaea hortensis (Mollusca: Helicidae) and their
genetic implications. Biochemical Genetics,
8: 365-382.

RICHARDSON, В. J., Р. В. BAVERSTOCK & M.
ADAMS, 1986, Allozyme Electrophoresis: a
handbook for animal systematics and population
studies. Academic Press, Sydney.

RIEDEL, A., 1988, Slimaki ladowe—Gastropoda ter-
restria. Catalogus faunae Poloniae, XXXVI, 1, In-
stytut Zoologii PAN, PWN, Warszawa.

RUDOLPH, P. H. & J. B. BURCH, 1987, Inheritance
of alleles at ten enzymatic loci of the freshwater
snail Stagnicola elodes (Basommatophora: Lym-
naeidae). Genetical Research Cambridge, 49:
201-206

RUDOLPH, P. H. & J. В. BURCH, 1989, Electro-
phoretic analysis of enzymes in three species of
Stagnicola (Pulmonata: Lymnaeidae). Journal of
Medical and Applied Malacology, 1: 57-64.

SELANDER, В. К. & D. W. KAUFMAN, 1975, Ge-

388 FALNIOWSKI, КОЙК & SZAROWSKA

netic structure of populations of the brown snail
(Helix aspersa). |. Microgeographic variation. Ev-
olution, 29: 385—401.

SHILEIKO, A. A., 1978, Nazemnye molliuski nad-
semeistva Helicoidea. Fauna SSSR. Molliuski, T.
Ш, 6, Academiia Nauk SSSR, Zoologichestii In-
stitut. (n.s.) 117, Leningrad.

SOKAL, R. R. & F. J. ROHLF, 1987, Introduction to
biostatistics. Second edition. W. H. Freeman and
Co., New York.

STAIKOU, A., M. LAZARIDOU-DIMITRIADOU, &
E. PANA, 1990, The life cycle, population dynam-
ics, growth and secondary production of the snail
Bradybaena fruticum (Muller, 1774) (Gastropoda
Pulmonata) in Northern Greece. Journal of the
Molluscan Studies, 56: 137-146.

SWOFFORD, D. L. & OLSEN, С. J., 1990, Phylog-
eny Reconstruction. Pp. 411-501, in: D. М. HiLLIS
& C. Moritz, eds., Molecular Systematics Sin-
auer Associates, Inc., Sunderland, Massachu-
setts.

WEIR, B. S., 1990, Genetic data analysis. Methods
for discrete population genetic data. Sinauer As-
sociates, Inc., Sunderland, Massachusetts.

WOODRUFF, D. 5. & А. SOLEM, 1990, Allozyme
Variation in the Australian Camaenid Land Snail
Cristilabrum primum: a prolegomenon for a mo-
lecular phylogeny of an extraordinary radiation in
an isolated habitat. Veliger, 33: 129-139.

WOODARUFF EDS К. "С. STAUB, ES: УР-
АТНАМ, V. VIYANANT & Н.-СН. YUAN, 1988,

Genetic variation in Oncomelania hupensis:
Schistosoma japonicum transmitting snails in
China and the Philippines are distinct species.
Malacologia, 29: 347-361.

WRIGHT, S., 1965, Evolution and the genetics of
populations, Vol. 2. The theory of gene frequen-
cies. University of Chicago Press, Chicago.

WRIGHT, S., 1978, Evolution and the genetics of
populations, Vol. 4, Variability within and among
natural populations. University of Chicago Press,
Chicago.

WURZINGER, K.-H., 1979, Allozymes of Ethiopian
Bulinus sericinus and Egyptian Bulinus trunca-
tus. Malacological Review, 12: 51-58.

ZEIFERT, D. V., 1987, The population dynamics of
the land snail Bradybaena fruticum (Muller) in dif-
ferent habitats. Pp. 442-445, in Ya. |. STAR-
OBOGATOV, A. М. GoLIKOV & |. М. LIKHAREV, eds.,
Molluscs, Results and Perspectives of Investiga-
tion. Eight meeting on the Investigations of Mol-
luscs, Abstracts of Communications. Akademiia
Nauk SSSR, Zoologicheskii Institut, Leningrad.

ZEIFERT, D. V. & S. V. SHUTOV, 1979, Role of
certain terrestrial molluscs in the transformation
of leaf litter. Ekologiya, 5: 58—61.

ZOUROS, E. & D. W. FOLTZ, 1984, Possible ex-
planations of heterozygote deficiency in bivalve
molluscs. Malacologia, 25: 583—591.

Revised Ms. accepted 9 February 1993.

MALACOLOGIA, 1993, 35(2): 389-398

THE GENETIC DIFFERENTIATION IN THREE SPECIES OF THE GENUS
HYDROBIA AND SYSTEMATIC IMPLICATIONS
(CAENOGASTROPODA, HYDROBIIDAE)

Martin Haase

Institut für Zoologie der Universität Wien, Althanstrasse 14,
A-1090 Wien, Austria

ABSTRACT

In order to investigate whether the genus Hydrobia should be subdivided, three species
representing the nominal genera involved (Hydrobia, Ventrosia, and Peringia) were compared
on the basis of allozyme data. Based on genetic distances, anatomical and ecological data, as
well as data on reproductive biology, it is argued that (1) there is no reason to split the genus
Hydrobia into different genera, (2) Hydrobia can be subdivided into the subgenera Hydrobia $.
s. and Peringia, and (3) Ventrosia has to be considered synonymous with Hydrobia.

The analysis of the genetic structure of the three populations investigated revealed heterozy-
gote deficiencies in practically all polymorphic loci in one case, and low, respectively complete
lack of variability in the remaining two populations. The deficiencies of heterozygotes are pri-
marily attributed to selection, probably due to a high infection rate with parasites, whereas the

reduced variability is explained by genetic drift following a bottleneck.
Key words: Allozymes, electrophoresis, genetics, systematics, Hydrobiidae, Hydrobia, Perin-

gia, Ventrosia.
INTRODUCTION

The family Hydrobiidae is one of the largest
among gastropods, and its systematics are
one of the most confusing in malacology. A
principal problem is the assessment of the
systematic value of minor differences among
these usually tiny snails, which are poor in
characters and reveal a considerable degree
of convergence. In many issues, there exist
as many opinions as there are authors. Such
a case is the debate whether the genus Ну-
drobia Hartmann, 1821, which defines the
whole family, should be subdivided into sub-
genera or even split into several genera. In
order to evaluate the status of the three nom-
inal genera involved—Hydrobia, Ventrosia
Radoman, 1977, and Peringia Paladilhe,
1874—their type species H. acuta (Drapar-
naud, 1805) (Fig. 2), V. ventrosa (Montagu,
1803) (Fig. 1), and P. ulvae (Pennant, 1777)
(Fig. 3), respectively, were investigated ge-
netically using standard methods of allozyme
electrophoresis.

As to the specific (not generic) designation
of the populations used in this study—topo-

| types were not available—l have followed
| Giusti & Pezzoli (1984) and their suggestion

that within Hydrobia populations with identical

| anatomy belong to a single species despite
| Slight, mainly conchological differences (in

contrast to the views of Radoman, 1973, and

389

Boeters, 1988). This assumption is corrobor-
ated by the morphological, anatomical and
genetic studies of Davis et al. (1988, 1989),
who compared six populations of H. truncata
(Vanatta, 1924) from Massachusetts, New
York and Maryland. But even if this assump-
tion turned out to be unwarranted, the pur-
pose of this paper would not be affected, be-
cause each population can unambiguously be
attributed to one of these nominal genera. Un-
til further notice, the genus Hydrobia is used
for all three species for reasons of simplicity
and clarity.

Nomenclatural History

The genus Hydrobia was introduced by
Hartmann (1821), who included Cyclostoma
acutum Draparnaud, 1805, which later was
designated as type species by Gray (1847).
Radoman (1977) was the first who anatomi-
cally described a Hydrobia from southern
France, the presumptive origin of Drapar-
naud’s specimens, with males possessing a
distally lobed penis (Fig. 2). He ascribed this
anatomy to H. acuta and restricted the type
locality to Palavas, Etang du Prévost. Previ-
ously, Radoman (1974) had introduced the
genus Obrovia Radoman, 1974, for two taxa
with this type of penis from the Adriatic coast.
So, after having identified H. acuta, Obrovia

390 HAASE

E CS ARAS
oe 4: Doge ANR
aes Ve, (as АХ

FIGS. 1-3. Penis. 1. Hydrobia ventrosa; 2. H. acuta; 3. H. ulvae (scale bars = 100 um).

became a synonym of Hydrobia (Radoman,
1977). In the same paper, Radoman de-
scribed the new genus Ventrosia Radoman,
1977, for the species with a slender penis
bearing a pointed lobe on the left side (Fig. 1).
This type of verge has always been associ-
ated with the taxon H. ventrosa (Montagu,
1803) (Robson, 1922; Krull, 1935; Muus,
1963; Bank & Butot, 1984; Giusti & Pezzoli,
1984; Falniowski, 1987). [Radoman (1977)
erroneously used the name Ventrosia stagno-
rum (Gmelin, 1791), which is a Heleobia
Stimpson, 1865, and considered Hydrobia
ventrosa a junior synonym (c.f. Bank & Butot,
1984).

Boeters (1984) found species with both pe-

nial types at Radoman’s restricted type local-
ity of H. acuta and claimed that Draparnaud's
original material of H. acuta also contained
both species. This assumption is based on
the comparison of two syntypes deposited at
the Muséum d’Histoire Naturelle in Paris. One
of these shells has significantly deeper su-
tures, like, according to Boeters (1984), the
species with males possessing the pointed
penis. In order to save the traditional view of
Hydrobia, Boeters (1984) designated this
shell as lectotype of H. acuta and attributed to
it the anatomy of what all authors cited above
thought was H. ventrosa. Thus, Ventrosia
would have to be considered a junior syn-
onym of Hydrobia and H. acuta a junior syn-

СЕМЕТ!С DIFFERENTIATION IN HYDROBIA 391

TABLE 1. Conditions for electrophoresis.

Current/ Run time
Enzyme Buffer System Voltage (hrs) Loci'
AAT Aspartate Amino Transferase ТЕВ 9.1°/TEB 8 (gel/tray) 35 MA 2 2
ACPH Acid Phosphatase TC 35 MA 2 1
AK Adenylate Kinase ой 35 МА 2 1
AO Aldehyde Oxidase TEB 8 35 MA 3.5 1
APH Alkaline Phosphatase TEB 9.1 350 V 4.5 1
CK Creatine Kinase TEB 8 35 MA 3:5 1
EST Carboxyl Esterase TC 8 & TEB 9.1/TEB 8 40 MA/35 МА 3.25/2 0
GDH Glutamate Dehydrogenase TEB 8 35 MA 3.5 1
G6PD Glucose-6-Phosphate Dehydrogenase TEB 9.1 350 V 4.5 1
СР! Glucose-6-Phosphate Isomerase Poulik 350 V 2 1
ISDH Isocitrate Dehydrogenase TEB 8 35 MA 315 2
LAP Leucine Aminopeptidase (= Cytosol TC 7 35 MA 2 0
Aminopeptidase)
LDH Lactate Dehydrogenase TEB 8 35 MA 3.5 1
MDH _ Malate Dehydrogenase TC 8 40 MA 3.25 1
ME Malic Enzyme TC 8 40 MA 3.25 0
MPI Mannose-6-Phosphate Isomerase Poulik 350 V 2 1
NADD Nicotinamide Adenosine TEB 8 35 MA 3.5 1
Dinucleotide Dehydrogenase
OCT Octopine Dehydrogenase TEB 8 35 MA 3.5 1
6PGD 6-Phosphogluconic Dehydrogenase TEB 8 35 MA 3.5 2
PGM Phosphoglucomutase Poulik & TEB 9.1/TEB 8 350 V/35 МА 2/2 2
SDH Sorbitol Dehydrogenase TEB 9.1 350 V 4.5 1
SOD Superoxide Dismutase see text 2
XDH Xanthine Dehydrogenase TEB 8 35 MA 3.5 1

‘Number of loci included in the analysis
2Tris-EDTA-Borate, pH 9.1
STris-Citrate

onym of H. ventrosa. The latter synonymy is
not mentioned by Boeters. He refrained from
discussing any consequences, left the other
species unnamed and did not state its generic
allocation (Boeters, 1984).

Subsequent authors explicitly (Giusti &
Pezzoli, 1984) or implicitly (Davis et al., 1989)
rejected Boeters’ view. To avoid the conse-
quences and further systematic confusion
arising from Boeters’ article, and because
there is no biological reason for Boeters’
purely taxonomic action, as is demonstrated
in this paper, Boeters’ type designation
should be suppressed by the International
Commission of Zoological Nomenclature, and
| am preparing a petition to this effect.

Peringia Paladilhe, 1874, is occasionally
used as a full genus (Kennard & Woodward,
1926; Wenz, 1938—1944; Nordsieck, 1982) or
as a subgenus (Zilch & Jaeckel, 1956; Fretter
& Graham, 1978; Boeters, 1988) for Hydrobia
ulvae (Pennant, 1777) (Fig. 3), although most
authors consider Peringia as a synonym of
Hydrobia (Ehrmann, 1933; Giusti & Pezzoli,
1984; Falniowski, 1987).

MATERIALS AND METHODS

Hydrobia ventrosa and Н. ulvae were col-
lected on the German Baltic island Fehmarn
in August 1991, H. ventrosa from the west
bank of the Burger Binnensee, where it lives
on mud, and H. ulvae from the sandy Süd-
strand. The salinity in both localities was 12%o.
Hydrobia acuta was found т a muddy marsh
(22%) on Torcello, an island in the Gulf of
Venice/Italy, in July 1991. The animals were
taken alive to the University of Vienna. The
specific identity of the samples was deter-
mined by investigating the male copulatory
organ in living specimens under the stereo
microscope. In each sample, only one type of
penis was found, indicating the presence of
only one species per sample. Most of the an-
imals were deep frozen at —70°C in tissue
buffer. The frozen material was carried in liq-
uid nitrogen to the Academy of Natural Sci-
ences in Philadelphia, where electrophoresis
was done. Parts of the samples were fixed in
70% ethanol or BOUIN’s fixative and depos-
ited at the Museum of Natural Histcry

392

TABLE 2. Allele frequencies. N, number of specimens.

Locus
AAT 1

AAT2

ACPH

APH

GDH

G6PD

GPI

ISDH 1

ISDH 2

LDH

MDH

MPI

NADD

OCT

Alleles

рр рр U>DZU>-Z U>Z 0>2 0>2 U>Z РР U>Z 0>2 MOOW»ZW>Z ОШ РЕ 0>2 0U>2Z

HAASE

H. ventrosa

38
0.684
0.316

22
4

—

—
+00 00 00000

A
o

GENETIC DIFFERENTIATION IN HYDROBIA 393

TABLE 2. (Continued)

Locus Alleles H. ventrosa H. acuta H. ulvae
6PGD 1 N 40 30 15

A 1 1 1
6PGD 2 М 35 30 15

А 1 1 1
РСМ 1 М 39 27 20

А 0.744 0 0

В 0.256 1 0

E 0 0 0.925

D 0 0 0.075
РСМ 2 N 28 27 10

A 0.482 1 0

B 0 0 1

D 0.286 0 0

E 0.089 0 0
SDH N 20 26 10

A 1 1 0

B 0 0 1
SOD 1 N 15 10 15

A 1 1 0

B 0 0 1
SOD 2 N 5 40 15

A 1 1 0

B 0 0 1
XDH N 40 30 25

A 1 1 0

В 0 0 1

(NHMW) under the following collection пит-
bers: H. ventrosa (NHMW 86801), H. acuta
(NHMW 86802), H. ulvae (NHMW 86803).

Horizontal starch-gel electrophoresis was
carried out following Davis et al. (1988). In-
stead of tris-citrate (TC) buffer with pH 6, TC
pH 7, was used (Shaw & Prasad, 1970). Ta-
ble 1 lists the 22 enzymes stained for and the
conditions for electrophoresis. Superoxide
dismutase was scored on gel slices stained
for a dehydrogenase. The data were ana-
lyzed using the computer program BIOSYS-1
release 1.7 by Swofford & Selander (1981).
Nei’s standard genetic distance (Nei, 1972)
and unbiased genetic distance (Nei, 1978)
and Cavalli-Sforza & Edwards’s arc and
chord distances (Cavalli-Sforza & Edwards,
1967) were calculated, and cluster analysis
based on Nei’s unbiased distance and Cav-
alli-Sforza & Edwards’s arc distance using
UPGMA were performed.

RESULTS

The enzymes LAP and ME were hardly de-
tectable. The esterases were extremely poly-

morphic and therefore not interpretable.
Thus, these enzymes had to be excluded
from the analysis. Allele frequencies for the
remaining 25 loci with 57 alleles are given in
Table 2. Hydrobia ulvae is characterized by
19 and H. ventrosa by seven unique alleles.
Hydrobia acuta shares all alleles with at least
one of the other two species. The genetic
variability of the three populations is зитта-
rized in Table 3. In H. ventrosa, eight loci are
polymorphic; seven of these are not in Hardy-
Weinberg equilibrium (Table 4). Hydrobia
acuta is remarkably uniform, with only one
polymorphic locus (Table 5). The variability of
H. ulvae lies between the other two species,
but is still very low. Only four loci have more
than one allele (Table 6). The MDH is 100%
heterozygous. Tables 7 and 8 give the genetic
distances between the three species. Hydro-
bia ventrosa and H. acuta are obviously very
closely related. The remarkably and unex-
pectedly large distance of H. ulvae from the
other two species is also depicted in the phe-
nograms of Figures 4 and 5. The cophenetic
correlation is 0.998 for the cluster analysis
based on Nei’s unbiased distance and 0.999

394 HAASE

TABLE 3. Genetic variability. Standard errors in parentheses.

Mean Sample Mean No Percentage
Size Per of Alleles of Loci
Locus Per Locus Polymorphic’
H. ventrosa 30.0 1.5 32.0
(1.9) (0.2)
H. acuta 27.8 1.0 4.0
(1.7) (0.0)
H. ulvae 172 1.2 16.0

ТА locus is considered polymorphic if more than one allele was detected.
2Unbiased estimate (see МЕ!, 1978).

Mean Heterozygosity

Direct HdyWbg
Count Expected?
0.043 0.136
(0.016) (0.045)
0.002 0.002
(0.002) (0.002)
0.070 0.062
(0.043) (0.031)

TABLE 4. Chi-square test for deviations from Hardy-Weinberg equilibrium in H. ventrosa.

Observed Expected
Locus Genotype Frequency Frequency
AAT 1 A-A 25 17.680
A-B 2 16.640
B-B 11 3.680
AK A-A 11 7.077
A-B 2 9.846
B-B 7 3.077
APH A-A 12 7.692
A-B 0 2.564
А-С 0 6.410
A-D 1 0.641
В-В 0 0.154
B-C 4 1.026
B-D 0 0.103
C-C 3 1.154
C-D 0 0.256
0-0 0 0.000
СР! А-А 37 37.000
А-В 1 1.000
В-В 0 0.000
MDH A-A 18 155122
A-B 3 8.755
B-B 4 12122
MPI A-A 18 13.800
A-B 10 18.400
B-B 10 5.800
PGM 1 A-A 26 21.468
A-B 6 15.065
B-B 7 2.468
PGM 2 A-A 13 6.382
A-B 0 3.927
A-C 0 7.855
A-D 1 2.455
B-B 3 0.509
B-C 0 2.327
B-D 2 0.727
C-C 8 2.182
C-D 0 1.455
0-0 1 0.182

x? DF Р
30.471 1 0
13.429 1 0
23.680 6 0.001

0 1 1
11.708 1 0.001

8.154 1 0.004
14.737 1 0
56.901 6 0

GENETIC DIFFERENTIATION IN HYDROBIA

TABLE 5. Chi-square test for deviation from
Hardy-Weinberg equilibrium in H. асша.

Geno- Observed Expected
Locus type Frequency Frequency х? DF P
APH A-A 25 25.000

A-D 1 1.000

D-D 0 0000. 5041. 1

for the analysis based on Cavalli-Sforza 8 Ed-
wards's arc distance, respectively.

DISCUSSION

All but one polymorphic loci of H. ventrosa
significantly lack heterozygotes. That one,
GPI, is polymorphic due only to a rare allele.
Under the frequently applied 95% criterion (a
locus is considered polymorphic if the fre-
quency of the most common allele does not
exceed 95%), the GPI locus would be consid-
ered monomorphic. The theoretically possible
reasons for heterozygote deficiencies are: (1)
inbreeding, (2) the Wahlund effect, (3) biased
sampling of homozygotes due to genetic
patchiness caused by ecological or behav-
ioural factors across a population's habitat, (4)
scoring bias for homozygotes, (5) differential
survival of homozygotes following collection,
(6) location of the locus on a sex chromosome,
(7) assortative mating, (8) presence of null
alleles, and (9) selection against heterozy-
gotes (Crouau-Roy, 1988; Staub et al., 1990).

Because practically all polymorphic loci are
deficient in heterozygotes, it is tempting to as-
sume a single explanation. Inbreeding or the

395

Wahlund effect would affect the allele fre-
quencies of all loci. Both hypotheses, how-
ever, are rejected for the following reasons.
The population is very big and the habitat very
uniform, so that there are no constraints for
inbreeding. The Wahlund effect can be ex-
cluded, because the sample stems from a ho-
mogeneous area of less than Y m?, so it
seems very unlikely that the sample con-
tained members of two or more subpopula-
tions. The remaining causes are more likely to
affect a single locus rather than the whole ge-
nome. Thus, probably a combination of fac-
tors accounts for the heterozygote deficien-
cies. However, three more of the above-listed
points can be excluded. The habitat of the
population is too homogeneous to establish
genetic patchiness, so that there is certainly
no sampling bias. The staining patterns were
easily and unambiguously interpretable.
Thus, a scoring bias can be excluded, as can
the differential survival of homozygotes fol-
lowing collection, because the sample was
frozen less than one week after collection,
and few snails had died during that time. It
cannot be estimated to which degree location
of polymorphic loci on a sex chromosome and
assortative mating are involved, because
nothing is known about the determination of
sex and the choice of mates in Hydrobia. The
presence of null alleles cannot be excluded.
The most probable explanation is selection
against heterozygotes. The population is
highly infected with trematode sporocysts and
rediae, which might cause a considerable se-
lective pressure. Four alleles each were de-
tected in APH and PGM 2. For these two loci,
the small sample sizes (20 and 28, respec-

TABLE 6. Chi-square test for deviation from Hardy-Weinberg equilibrium т H. ulvae.

Observed

Locus Genotype Frequency
AK A-A 5
A-C 4
C-C 1
G6PD A-A 6
A-B 2
B-B 2
MDH A-A 0
А-В 20
В-В 0
РСМ 1 C-C 17
C-D 3
D-D 0

Expected

Frequency Ya DF Р

4.789
4.421
0.789

4.789
4.421

0.105 1 0.745

3.488 1

0.062

19.000 1 0

0.086 1 0.770

396 HAASE

TABLE 7. Matrix of Nei’s genetic distances.
Above the diagonal: Neïs (1972) standard
distance; below: Nei’s (1978) unbiased distance.

Н. acuta H. ulvae

H. ventrosa 0.111 1.648
H. acuta 0.110 — 1.753
H. ulvae 1.645 1.751 —

H. ventrosa

TABLE 8. Matrix of Cavalli-Sforza & Edwards's
(1967) distances. Above the diagonal: chord
distance; below: arc distance.

H. ventrosa Н. acuta H. ulvae
H. ventrosa — 0.306 0.790
Н. acuta 0.323 — 0.814
H. ulvae 0.873 0.903 —

tively) alone might account for the deviations
from Hardy-Weinberg equilibrium.

The 100% heterozygosity of the MDH in H.
ulvae is probably due to selection against ho-
mozygotes, which means the remarkable loss
of 50% of the offspring.

Lack of genetic variation as in H. асша,
which has no polymorphic locus under the
95% criterion (the polymorphism of the APH
locus is again due to a rare allele), is usually
explained by the assumption of genetic drift
following a bottleneck in the population’s past
(Nei et al., 1975).

Nei’s commonly used distances were cho-
sen for reasons of comparability, although
these measures are nonmetric (Wright, 1978)
and the constant substitution of amino-acids,
on which Nei based his model (Nei, 1972), is
hardly, if ever, met (Hillis, 1984). Cavalli-
Sforza 8 Edwards's arc distance is, according
to Wright (1978), superior to all other distance
coefficients due to its geometrical clarity. But
the validity of Cavalli-Sforza 8 Edwards's dis-
tances 15 restricted in that only random ge-
netic drift and selection are considered
causes for divergence between populations
(Cavalli-Sforza & Edwards, 1967). More com-
prehensive presentations of the strengths and
limitations of the various distance measures
can be found in Wright (1978), Davis et al.
(1988), and Swofford & Olsen (1990). How-
ever, the cophenetic correlations (cc) of the
phenograms of Figures 4 and 5 (cc = 0.998
and 0.999, respectively) indicate that in the
present case both distance measures applied
yield equivalent results.

Nei’s (1972) genetic distance D between

congeneric species of molluscs is typically in
the range from 0.20-0.60 (Woodruff et al.,
1988). In a survey on distance data, Thorpe
(1983) found D values larger than 1.05 in only
15% of approximately 900 estimates of inter-
specific distances of congeners of various eu-
karyotes. This value was exceeded in 80% of
about 160 comparisons between confamilial
genera. Davis et al. (1989) compared six pop-
ulations of the North American H. truncata
(Vanatta, 1924). The highest distance value
(Nei’s unbiased distance, 1978) was 0.018.
However, one has to be careful drawing tax-
onomic conclusions from distance data only.
Certain ranges of genetic distance do not
have simple correspondence to taxonomic
levels (Hoagland & Davis, 1987). Based on
the genetic distances in Tables 7 and 8, one
could conclude that H. ventrosa and H. acuta
were conspecific populations or very closely
related species, whereas H. ulvae belonged
to another genus. Taking anatomical (Krull,
1935; Giusti & Pezzoli, 1984; Falniowski,
1987; personal observations) and cytological
(Butot & Kiauta, 1966) data into account, it
becomes clear that H. ventrosa and H. acuta
are distinct species and that there is no char-
acter that would separate H. ulvae from the
other two species on a higher level. [The duct
connecting the prostate with the mantle cavity
described by Johansson (1948) for H. ulvae
has also been found in H. acuta and H. ven-
trosa (personal observations).] However, the
large distance values between H. ulvae and
the other two species correspond with eco-
logical differences and differences in repro-
ductive biology. Hydrobia ventrosa and H.
acuta prefer sheltered bays, whereas H. ul-
vae also tolerates higher water movement
(Fretter & Graham, 1978; Falniowski, 1987;
personal observations). Hydrobia ulvae has
free swimming veligers (Fish & Fish, 1977),
whereas in H. ventrosa the whole veliger
stage is intracapsular (Thorson, 1946). For H.
acuta there is only indirect evidence for the
same mode of reproduction as in H. ventrosa.
The animals reproduced in an aquarium
equipped with pump and filter (personal ob-
servations). Planktonic larvae would not have
survived these conditions.

In this study, only a single population of
each species could be investigated, and the
following systematic conclusions should be
taken with some reservation. However, be-
cause the genetic distances correspond with
ecological and developmental data, it can well
be assumed that the results obtained from

СЕМЕТ!С DIFFERENTIATION IN HYDROBIA 397

1.08 90

Distance

Н. ventrosa
H. acuta

Н. ulvae

72 54 36 18 00

FIG. 4. UPGMA phenogram based on Nei’s (1978) unbiased genetic distance.

H. ventrosa
H. acuta

H. ulvae

a IO — — je;

1.00 .90 .80 .70 .60 .50

Distance

.40 .30 .20 10 .00

FIG. 5. UPGMA phenogram based on Cavalli-Sforza & Edwards’s (1967) arc distance.

these three populations reflect the true rela-
tionships between the three species. Thus, a
separation of H. ulvae from the other two spe-
cies based on allozymes, ecological and de-
velopmental data can well be justified. Be-
cause the general anatomical organization of
all three species is practically identical, a sep-
aration beyond the subgenus level would be
unwarranted. Consequently, the genus Hy-
drobia Hartmann, 1821, can be subdivided
into the subgenera Hydrobia s. $. and Perin-
gia Paladilhe, 1874, and Ventrosia Radoman,
1977, has to be considered synonymous with
Hydrobia. This synonymy is based on natural
arguments, which demonstrate that Boeters’s
(1984) purely taxonomic attempt discussed in
the introduction is unnecessary and also
therefore to be rejected.

ACKNOWLEDGEMENTS

| am grateful to Dr. George M. Davis for his
invitation to the Academy of Natural Sciences
of Philadelphia, where | applied electro-
phoretic techniques under his and Caryl Hes-
terman’s patient guidance. Dr. L. Salvini-Pla-
wen, Dr. E. Wawra and two anonymous
reviewers made helpful comments on the
manuscript. The Academy’s Jessup Fund and
the Austrian Bundesministerium für Wissen-
schaft und Forschung provided financial sup-
port for my work in Philadelphia.

LITERATURE CITED

BANK, R. A. & L. J. M. BUTOT, 1984, Some more
data on Hydrobia ventrosa (Montagu, 1803) and
“Hydrobia stagnorum (Gmelin, 1791) with re-
marks on the genus Semisalsa Radoman, 1974.
Malakologische Abhandlungen, Staatliches Mu-
seum fur Tierkunde, 10: 5-15.

BOETERS, H. D., 1984, Zur Indentitat des Hydro-
bia-Typus (Prosobranchia: Hydrobiidae). Heldia,
1: 3-5.

BOETERS, Н. D., 1988, Westeuropäische Moitess-
ieriidae, 2 und westeuropáische Hydrobiidae, 7.
Moitessieriidae und Hydrobiidae in Spanien und
Portugal (Gastropoda: Prosobranchia). Archiv fur
Molluskenkunde, 118: 181-261.

BUTOT, L. J. M. & B. KIAUTA, 1966, Notes on the
cytology of Rissoacea. |. Cytotaxonomical condi-
tions in some Hydrobiidae and Assimineidae
(Gastropoda, Streptoneura). Basteria, 30: 21-35.

CAVALLI-SFORZA, L. & A. W. F. EDWARDS,
1967, Phylogenetic analysis: models and estima-
tion procedures. Evolution, 21: 550-570.

CROUAU-ROY, B., 1988, Genetic structure of
cave-dwelling beetles populations: significant de-
ficiencies of heterozygotes. Heredity, 60: 321-
327.

DAVIS, G. M., V. FORBES & G. LOPEZ, 1988,
Species status of northeastern American Hydro-
bia (Gastropoda: Prosobranchia): ecology, mor-
phology and molecular genetics. Proceedings of
the Academy of Natural Sciences of Philadel-
рта, 140: 191-246.

DAVIS, С. М., М. МСКЕЕ & С. LOPEZ, 1989, The
identity of Hydrobia truncata (Gastropoda: Ну-
drobiinae): comparative anatomy, molecular ge-
netics, ecology. Proceedings of the Academy of
Natural Sciences of Philadelphia, 141: 333-359.

398 HAASE

EHRMANN, P., 1933, Mollusken. In: P. BROHMER, P.
EHRMANN & G. ULMER, Die Tierwelt Mitteleuropas,
И (1): 264 pp. Leipzig.

FALNIOWSKI, A., 1987, Hydrobioidea of Poland
(Prosobranchia: Gastropoda). Scientific Bulletins
of the Stanislaw Staszic Academy of Mining and
Metallurgy, no. 1096, Folia Malacologica, Bulletin,
1: 1-122.

FISH, J. D. & S. FISH, 1977, The veliger larva of
Hydrobia ulvae with observations on the veliger
of Littorina littorea (Mollusca: Prosobranchia).
Journal of Zoology, London, 182: 495-503.

FRETTER, V. & A. GRAHAM, 1978, The proso-
branch molluscs of Britain and Denmark, part
3—Neritacea, Viviparacea, Valvatacea, terres-
trial and freshwater Littorinacea and Rissoacea.
Journal of Molluscan Studies, Supplement 5:
101-152.

GIUSTI, F. & E. PEZZOLI, 1984, Notulae Malaco-
logicae, XXIX—Gli Hydrobiidae salmastri delle
acque costiere Italiane: primi cenni sulla sistemat-
ica del gruppo e sui caratteri distintivi delle singole
morfospecie. Lavori della Societa Italiana di Ma-
lacologia, 21: 117-148.

GRAY, J. E., 1847, A list of the genera of recent
mollusca, theirsynonyma and types. Proceedings
of the Zoological Society of London, 15: 129-219.

HARTMANN, J. D. W., 1821, System der Erd- und
Flußschnecken der Schweiz. Mit vergleichender
Aufzaehlung aller auch in den benachbarten
Laendern, Deutschland, Frankreich und in Italien
sich vorfindenden Arten. Neue Alpina, 1: 194—
268, 2 plates.

HILLIS, D. M., 1984, Misuse and modification of
Neïs genetic distance. Systematic Zoology, 33:
238-240.

HOAGLAND, К. Е. & С. М. DAVIS, 1987, The suc-
cineid snail fauna of Chittenango Falls, New
York: taxonomic status with comparisons to other
relevant taxa. Proceedings of the Academy of
Natural Sciences of Philadelphia, 139: 465-526.

JOHANSSON, J., 1948, Uber die Geschlechtsor-
gane der Hydrobiiden und Rissoiden und den ur-
sprünglichen Hermaphroditismus der Proso-
branchier. Arkiv för Zoologi, 40A: 1-13.

KENNARD, А. 5. & В. В. WOODWARD, 1926, Syn-
опуту of the British non-marine Mollusca (Re-
cent and post-Tertiary), 447 pp. Trustees of the
British Museum, London.

KRULL, H., 1935, Anatomische Untersuchungen
an einheimischen Prosobranchiern und Вейгаде
zur Phylogenie der Gastropoden. Zoologisches
Jahrbuch, Abteilung für Anatomie und Ontogen-
ese, 60: 399—464.

MUUS, В. J., 1963, Some Danish Hydrobiidae with
the description of a new species, Hydrobia ne-
glecta. Proceedings of the Malacological Society
of London, 35: 131-138.

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

NEI, M., 1978, Estimation of average heterozygos-
ity and genetic distance from a small number of
individuals. Genetics, 89: 583-590.

NEI, M., T. MARUYAMA & R. CHAKRABORTY,
1975, The bottleneck effect and genetic variabil-
ity in populations. Evolution, 29: 1-10.

NORDSIECK, F., 1982, Die europäischen Meeres-
Gehäuseschnecken, 539 pp. Stuttgart, New York.

RADOMAN, P., 1973, New classification of fresh
and brakish (sic!) water prosobranchia from the
Balkans and Asia Minor. Posebna Izdanja,
Prirodnjacki Muzej u Beogradu, 32: 2-30.

RADOMAN, P., 1974, Some new gastropod repre-
sentatives from the brackish waters of the Adri-
atic and Aegean seasides. Veliger, 16: 283-288.

RADOMAN, P., 1977, Hydrobiidae auf der Balkan-
halbinsel und in Kleinasien. Archiv fur Mollusken-
kunde, 107: 203-223.

ROBSON, G. C., 1922, On the anatomy and affin-
ities of Paludestrina ventrosa, Montague. The
Quarterly Journal of Microscopical Sciences, 66:
159-185.

SHAW, С. & D. PRASAD, 1970, Starch gel electro-
phoresis of enzymes—a compilation of recipes.
Biochemical Genetics, 4: 297-320.

STAUB, К. C., D. $. WOODRUFF, Е. $. UPATHAM
& V. VIYANANT, 1990, Genetic variation in Neo-
tricula aperta, the intermediate host of Schisto-
soma mekongi: allozyme differences reveal a
group of sibling species. American Malacological
Bulletin, 7: 93-103.

SWOFFORD, D. L. & С. J. OLSEN, 1990, Phylog-
eny reconstruction. Pp. 411-501, in: D. M.
HILLIS & C. MORITZ, Molecular systematics,
Sunderland, Massachusetts.

SWOFFORD, D. L. & В. В. SELANDER, 1981,
BIOSYS-1: a FORTRAN program for the com-
prehensive analysis of electrophoretic data in
population genetics and systematics. Journal of
Heredity, 72: 281-283.

THROPE, J. P., 1983, Enzyme variation, genetic
distance, and evolutionary divergence in relation
to levels of taxonomic separation. Pp. 131-152,
in: G. $. OxForD € D. ROLLINSON, Protein poly-
morphism: adaptive and taxonomic significance.
London.

THORSON, G., 1946, Reproduction and larval de-
velopment of Danish bottom invertebrates. Med-
delelser fra Kommissionen for Danmarks Fiskeri
og Havundersogelser, Serie Plankton, 4 (1): 523

PP-

WENZ, W., 1938-1944, Gastropoda. In: О. H.
SCHINDEWOLF, Handbuch der Paläozoologie, 6:
1639 pp.

WOODRUFF, D. S., К. С. STAUB, Е. S., UP-
АТНАМ, V. VIYANANT & Н. С. УЧАМ, 1988, Ge-
netic variation in Oncomelania hupensis: Schis-
tosoma japonicum transmitting snails in China
and the Philippines are distinct species. Malaco-
logia, 29: 347-361.

WRIGHT, S., 1978, Evolution and genetics of pop-
ulations, 4. Variability within and among natural
populations, 580 pp. Chicago.

ZILCH, A., & $. С. А. JAECKEL, 1956, Mollusken.
In: P. BROHMER, P. EHRMANN & G. ULMER, Die Tier-
welt Mitteleuropas, || (1) Ergänzung: 294 pp.
Leipzig.

Revised Ms. accepted 27 May 1993.

MALACOLOGIA, 1993, 35(2): 399—406

DIVERGENCE OF ACTIVITY PATTERNS IN COEXISTING
SPECIES OF LAND SNAILS

Takahiro Азат!

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

ABSTRACT

The activity patterns of the land snails Mesodon normalis and Triodopsis albolabris were
examined. These species share microhabitats (leaf litter) and food (fungi) in the Appalachians.
Their patterns of daily activity showed striking dissimilarities in both natural and laboratory
conditions of light and temperature. The activity of M. normalis was more or less crepuscular,
whereas T. albolabris was strictly nocturnal. These distinctive patterns were maintained whether
the two species were kept together or in isolation. Thus, the differences are not due to their direct
interaction; the activity patterns have diverged evolutionarily. The temporal separation of the two
species previously demonstrated in the wild results from this divergence of activity patterns.

Key words: activity pattern, temporal separation, Mesodon normalis, Triodopsis albolabris,

Neohelix, Polygyridae, Pulmonata

INTRODUCTION

In terrestrial communities, molluscan guilds
are mostly composed of pulmonates, which
show a great deal of ecological diversity co-
existing in a large variety of habitats (Machin,
1975; Riddle, 1983). Although niche differen-
tiation in general can be realized in its funda-
mental dimensions, such as food, space, and
time (Hutchinson, 1957), relatively few stud-
ies have documented temporal separation of
coexisting molluscs on land.

In pulmonates, some daytime activities may
be found in the field (Ingram, 1940; Blinn,
1963), initiated by the changes of physical
conditions, such as temperature, humidity,
and precipitation (Dainton, 1954a,b; Karlin,
1961; Webley, 1964; Dainton & Wright, 1985;
Rollo, 1991). In many species, however, the
regular patterns of daily activities have been
shown to be nocturnal; the slugs Arion (Lewis,
1969a), Deroceras (Newell, 1966; Morton,
1979), Limax (Rollo, 1982; Ford & Cook,
1987), and Milax (Barnes & Weil, 1945), and
the snails Arianta (Abdel-Rehim, 1983), Ce-
paea (Cameron, 1970), Helix (Bailey, 1975;
Gelderloos, 1979), and Monadenia (Szlavecz,
1986), and Triodopsis (Henne, 1963). Several
of the above species evidently possess en-
dogenous rhythms of activities (Lewis, 1969b;
Sokolove et al., 1977; Morton, 1979; Bailey,
1981; Ford & Cook, 1987).

On the other hand, only a few studies have
addressed the question of interspecific diver-

sity of activity patterns in pulmonates. Barnes
& Weil (1942, 1945) noted differences of ac-
tivity times in slugs. Cameron (1970) docu-
mented the variation of activity patterns
among the sympatric snails, A. arbustorum
(L.), С. nemoralis (L.), and С. hortensis
(Muller) in the laboratory. Daily activities of
these species commonly showed unimodal
distributions but differed in the degree of noc-
turnality. Cepaea nemoralis and C. hortensis
show different patterns of activity in field en-
closures (Tilling, 1986).

Mesodon normalis (Pilsbry) and Triodopsis
albolabris (Say) are sympatric in many places
in the southern Appalachian Mountains
(Hubricht, 1985). These mycophagous snails
share food and microhabitat on the forest
floor, and show striking similarity in shell
morphology (Pilsbry, 1940, Asami, 1988).
Among coexisting molluscs, these species
are distinctively abundant and large in body
size (approximately 30 mm in diameter). In
mark-recapture experiments in sympatric
populations, M. normalis is captured on the
forest litter more frequently than Т. albolabris
in the daytime, whereas this relationship is
reversed at night (Asami, 1988), suggesting
that the two species appear and forage on the
litter at different times of the day. | conducted
the present study to examine the daily pat-
terns of activities of M. normalis and T. albo-
labris and to test whether their different activ-
ity patterns bring about temporal separation
in the wild.

‘Present address: Division of Biology, Tachikawa College of Tokyo, Azuma-cho, Akishima-shi, Tokyo 196, Japan.

400 ASAMI

MATERIALS AND METHODS
Taxonomy

Because of extreme conchological similar-
ities, the taxonomy of the current species and
related taxa has been often confused (Pilsbry,
1940; Solem, 1976; McCracken & Brussard,
1980; Emberton, 1988, 1991). Mesodon
Rafinesque and Triodopsis Rafinesque are in
separate subfamilies, the Polygyrinae and
Triodopsinae, respectively, ofthe Polygyridae
on the basis of penial structure (Pilsbry,
1940). Examination of shells, genitalia, and
allozymes suggest that the conchological
similarities between Mesodon and Triodopsis
are due to convergence (Pilsbry, 1940; Em-
berton, 1988, 1991). In the revision of the Tri-
odopsinae, Emberton (1988) has raised the
subgenus Neohelix to generic rank. Mesodon
normalis and T. albolabris are one of a num-
ber of species pairs in these subfamilies that
show striking similarity in shell morphology in
spite of their taxonomic positions. Voucher
specimens of the taxa studied here are de-
posited in the Academy of Natural Sciences
of Philadelphia (catalog nos. 369306 and
A12179 for M. normalis, and A12094 for T.
albolabris).

Study Site and Sample Maintenance

These experiments were conducted at the
Mountain Lake Biological Station, 1167 m in
elevation, Giles County, Virginia, USA. Adults
of M. normalis and T. albolabris were collected
from an area of 200 x 10 m, 0.5 km west of
the station (approximately 37°22"N, 80°31"W).
The collected snails of each species were
maintained separately in field enclosures (12
mm metal mesh, 46 cm diameter and 23 cm
height, approximately ten animals per enclo-
sure), established in a deciduous forest near
the collection site, for about two months prior
to the experiment.

Activity Recording

Except in those experiments examining in-
dividual interactions, experimental animals
were individually isolated in plastic containers
(84 mm diameter, 37 mm height) and fed oat-
meal with powdered natural chalk on moist
paper towels. Humidity inside the containers
was close to 100% for the whole period. Con-
tainers were horizontally arranged on a plat-

form 0.8 m above the substratum or floor.
Each animal was transferred to a clean con-
tainer with new food every other day, and lo-
cations of the containers were randomized at
this time. During the complete course of the
experiments, a 40 w red bulb 1.2 m above the
animals was kept on, enabling night observa-
tions. Prior to recording activity patterns, the
animals were conditioned to the experimental
treatment for 5 days. Each individual was
then scored for activity every hour for 24 h
beginning 2 h after the routine maintenance,
unless indicated otherwise below. Activity
was defined as moving the head with ex-
tended antennae, creeping, feeding, or clean-
ing the shell as reported by Ingram (1944).

Experiments were conducted under both
natural outdoor conditions and controlled lab-
oratory conditions. For the former, the con-
tainers were shaded by a shelter and experi-
enced natural changes of temperature and
light (Fig. 1A). In the five-day conditioning pe-
riod, the air temperature changed daily in a
clear cycle (9 to 22 °C). Daylight lasted from
4:30 a.m. through 6:30 p.m. including dawn
and dusk. On the recording day, however, the
weather was overcast, resulting in a rather
obscure pattern of temperature change.

For the indoor experiments, the animals
were conditioned to the day-length and tem-
perature cycle typical for July atthe study site,
light from 4:30 a.m. to 7:30 p.m. and temper-
ature ranging from 18 to 25°C daily (Fig. 1B).
No natural light was admitted to the experi-
mental area. Two fluorescent lamps (34 w
each, placed 1.2 m above the samples) were
used to produce the light phase. There was
no dawn or dusk. To create a daily cycle of
temperature similar to the natural one, an
electric heater was turned on at 6:00 a.m. and
off at 1:00 p.m.

Test of Interaction Between Individuals

In order to test the effects of interactions
between conspecific individuals and between
individuals of different species, the activity
patterns of paired animals were examined,
with those of single animals as controls. All
the animals were conditioned to the same lab-
oratory conditions described above. To pro-
vide them with the same amount of space per
individual, each pair was maintained т a con-
tainer twice as large as that used for single
individuals. Scoring was carried out as de-
scribed above.

ACTIVITY PATTERNS OF LAND SNAILS 401

№
о

=
о

Temperature (°c)
a

о
nN
о
o
>

25
al
20 A

Time

FIG. 1. Temperature and light conditions during the
experiments. A. Outdoor experiment. B. Indoor ex-
periment. Solid line: the temperature pattern on re-
cording day. Interrupted line: the pattern of mean
temperature in the conditioning period. In the indoor
experiment, the same temperature pattern was re-
peated on the recording day as in the entraining
period. The straight bars indicate the light condi-
tions; Open bar: daytime or light phase; hatched
bar: dusk or dawn; filled bar: nighttime or dark
phase.

Statistics

Analyses were designed to test the differ-
ences in the degree of nocturnality, which
was defined as the proportion of nocturnal ac-
tivity in each 24-h period. Values of nocturnal-
ity were calculated by dividing the total scores
for the dark phase by those for 24 h. The
Mann-Whitney test or the Kruskal-Wallis two-
way test was used in each test of the homo-
geneity of the mean nocturnalities and total
scores for 24 h between treatments. For test-
ing interactions between conspecifics and be-
tween species on nocturnality or the total
score, the mean of the two individuals was
used as an independent observation for each
pair.

RESULTS
Interspecific Variation of Activity Pattern

Under natural conditions of light and tem-
perature Mesodon normalis and Triodopsis
albolabris showed notable differences in their
daily patterns of activity (Fig. 2A). The pattern
of T. albolabris was strongly nocturnal,
whereas that of M. normalis was nearly crep-
uscular, showing no activity at 11 p.m. Al-
though 7. albolabris showed high activity at 6
p.m., this was reduced immediately thereaf-
ter, and most of its activity was confined to the
night. The pattern of М. normalis differed from
that of Т. albolabris. After the first peak
around dusk, activity steadily diminished until
11 p.m. and then increased to form the morn-
ing peak. The same individuals of M. normalis
were often active in both periods in a single
24-h cycle; there were not two behavioral
types of individuals corresponding to the two
peaks.

The results of the outdoor experiment were
corroborated by the indoor experiment (Fig.
2B). As in natural light and temperature, 7.
albolabris was strictly nocturnal. In contrast,
M. normalis showed a drastic reduction in ac-
tivity in the middle of the dark phase when Т.
albolabris was most active.

In both species, there were some differ-
ences in activity patterns between the out-
door and indoor experiments. Indoors, 7. al-
bolabris showed an increase of activity
towards the end of the dark phase. In M. nor-
malis, the relatively large activity was ob-
served at dusk outdoors, but in the early
morning indoors. For statistical evaluation of
the differences in activity time between spe-
cies and between the outdoor and indoor con-
ditions, the distributions of individual noctur-
nalities were compared (Fig. 3). In both indoor
and outdoor experiments, nocturnality of M.
normalis was significantly less than that of 7.
albolabris (Kruskal-Wallis two-way test, P <
0.0001), and the results were consistent be-
tween experiments (Р > 0.25). It was con-
cluded, therefore, that Т. albolabris and М.
normalis have distinct patterns of daily activ-
ities and that M. normalis is substantially less
nocturnal than 7. albolabris.

Effects of Individual Interactions
There was no significant difference in mean

nocturnality between isolated and paired con-
specifics of either species (Fig. 4A; P > 0.7

402 АЗАМ!

Proportion of daily activity (%)

"16 18 20 22 0

Proportion of daily activity (%)

16 18 20 220 2 4 6 8 10 12 14

Time

) м

e М = 45

2 12,

8 | 4 LE

Do LR

12 14 16 18 20 22 0 6 8 10

N = 26

12-

Proportion of daily activity (%)

12114 16 1820 22: 01 (2) абон НЯ

Time

FIG. 2. Activity patterns of T. albolabris (upper) and М. normalis (lower). A. Outdoor experiment. В. Indoor
experiment. Each bar shows the mean hourly percentage of the 24-h activities. Black bar: nighttime. Open
bar: daytime. Hatched bar: dusk or dawn. N: sample size.

for M. normalis, P > 0.5 for T. albolabris).
Because coexistence of conspecifics might
affect overall activity levels, the total scores
for 24 h were compared between the treat-
ments in each species. As shown in Figure
4B, paired individuals of M. normalis were
more active than isolated individuals (P <
0.003), whereas there was no difference for T.
albolabris (P > 0.7). The 24-h activity of
paired individuals was higher in M. normalis
than in T. albolabris (P < 0.005), but there
was no difference between species in the ac-
tivity of single individuals (P > 0.2). These
results suggest an interaction between indi-
viduals of M. normalis that causes increased
activity.

In the experiment to test for interspecific
interaction, there were no significant differ-
ences in nocturnality between single and
paired individuals in either species (Fig. 5A; P
> 0.2 for M. normalis, P > 0.2 for T. albola-
bris). In addition, neither species showed any
effect of treatment on overall 24-h activity
(Fig. 5B; P > 0.7 for M. normalis, P > 0.5 for
T. albolabris). These results of pairing exper-
iments indicate that M. normalis and T. albo-
labris retain their distinct nocturnalities, even

when allowed to encounter conspecifics or
other species as they would in the wild. Also,
their coexistence does not lead to direct inhi-
bition or enhancement of the activity of either
species.

DISCUSSION
Evolutionary Divergence of Activity Patterns

Pulmonates are considered to be nocturnal
in general to avoid high daytime temperature
and reduced humidity, which may cause
problems with body-water retention and os-
moregulation (Cameron, 1970; Schmidt-
Nielsen et al., 1972; Machin, 1975; Ford &
Cook, 1987). The present study has demon-
strated, however, that Mesodon normalis and
Triodopsis albolabris have distinct patterns of
daily activities. Mesodon normalis has two ac-
tivity peaks in the daytime, near dawn and
dusk, whereas T. albolabris shows strong
nocturnality, with unimodal distribution of ac-
tivity, the pattern usually considered typical
for pulmonates.

The slight differences between activity pat-

ACTIVITY PATTERNS OF LAND SNAILS 403

— = =
|
50- |
40: ||
Sn IR
> 30- | Е
м |
® |
30 |
ic |
10. | |
| | | I | _/ T. albolabris
et 2 — - M. normalis
0<20 <40 <60 <80 0 < 100
Nocturnality (%)
50-
| 7
401 |
>
= 30!
5 ==
=
В. |
y 20- | |. —
© | | ee ore
Ш | | => | a -
101 | pu] | || [| | | i =
) Е | 1 | №. = Т. albolabris
si \ = M. normalis

0=20 <40 <60 <80 0<100

Nocturnality (%)

FIG. 3. Distributions of individual nocturnalities in
М. normalis and T. albolabris. A. Outdoor experi-
ment. B. Indoor experiment. The vertical axis indi-
cates the frequency of individual nocturnality.

terns in the outdoor and indoor experiments
may be related to the limitations of simulating
natural conditions in the laboratory. For in-
stance, indoors there was no gradual change
of light intensity, while the animals outdoors
experienced dawn and dusk. Outdoors both
species showed high activity at 6 p.m. In-
doors, however, M. normalis was most active
at 5 a.m., and T. albolabris showed nearly
10% of its total activities at the same time. In
the field, T. albolabris burrows under litter just
before dawn. It is possible that T. albolabris
showed an increase of activity after light-on
because no shelter was provided in the ex-
periments. This type of post-dark activity in
artificial light cycles has been found in other
pulmonates (Sokolove et al., 1977; Gelder-
loos, 1979; Wareing & Bailey, 1985; Ford &
Cook, 1987). Except for these differences,
equivalent results were obtained outside and
inside the laboratory. Therefore, the present
results show that the activity patterns of M.
normalis and T. albolabris are distinct, espe-
cially in the degree of nocturnality.
Interspecific separation in activity time can-
not be due to direct reactions between the two

A
(10) | =
М. normalis

(8)

(30)
T. albolabris

0 20 Lao 60
Proportion of nocturnal activities (%)
(10) |
M. normalis
(8)
(30)
T. albolabris

(10) E

Total scores for 24 hr.

| | Single

FIG. 4. Test of the effect of conspecific interaction
on nocturnality and activity. А. Mean nocturnalities
and standard errors. B. Mean total activity scores
for 24 h and standard errors. Number of replicates
is indicated in parenthesis.

species. The samples of M. normalis and T.
albolabris were maintained in separate enclo-
sures for two months prior to the experiments.
They were then individually isolated for the
entraining periods and experiments. These
molluscs are not likely to have determined ac-
tivity patterns so rigidly by learning or habitu-
ation prior to collection from the wild that they
could retain those patterns through these ex-
perimental periods. Besides, the interspecific
pairing experiments showed no effect on noc-
turnality or the total activity of either species.
Therefore, the present results strongly sug-
gest that the divergence of activity patterns
between the two species is evolutionary, as in
Cepaea (Cameron, 1970; Tilling, 1986; Cowie
& Jones, 1987).

Mesodon normalis is usually abundant in
mountainous areas in the southern Appala-
chians, whereas T. albolabris is much more
widely distributed at lower altitudes as well as
in sympatry with M. normalis (Hubricht, 1985).
Cameron (1970) suggested that hot and dry
habitats are occupied by more nocturnal spe-
cies. Low nocturnality of populations that in-
habit high altitudes, where the climate is

404 ASAMI

A
(7) |

2 Ee

(10) E ae _ : Es

Т. albolabris а ” — |
(8) o

M. normalis

O)
M. normalis

Т. albolabris

FIG. 5. Test of the effect of interspecific interaction
on nocturnality and activity. A. Mean nocturnalities
and standard errors. B. Mean activity scores for 24
h and standard errors. Number of replicates in in-
dicated in parenthesis.

cooler and wetter, could be predicted by this
hypothesis. This does not explain, however,
why M. normalis shows a reduction of activity
at night, in contrast to 7. albolabris, in areas
where they occur in the same microhabitats
(Asami, 1988). Moreover, М. normalis is
much inferior in water retention and survival
of juveniles in low humidity to 7. albolabris
(Asami, in press). Their adults similarly show
clear differences in desiccation tolerance
(Asami, in preparation). Thus, the diurnal ac-
tivity of M. normalis is not explicable by a rel-
atively large tolerance of dry and warm day-
time conditions.

Evaluation by Field Experiments

In repeated searches for snails on the for-
est litter, 75% of the animals captured at night
were T. albolabris, whereas 78% of those in
the daytime were M. normalis (Fig. 6). As
these ground-dwelling snails are likely to ap-
pear on the litter for foraging or mating, the
ratio of animals captured per search between
nighttime and daytime would correspond to

Daytime 1985

1984

1986

1984
Night 1985

1986

4 4
0 25 50 75 100

Proportion in captured individuals (%)

FIG. 6. The relative discovery rates (proportions in
yearly captures) of M. normalis (hatched) and 7.
albolabris (black) in natural habitats. Number of
yearly captures 15 given in parenthesis (after Asami,
1988).

the daily proportion of nocturnal activity (noc-
turnality) in the field. Thus, by comparing the
interspecific ratios of nocturnalities between
the field observation and present experi-
ments, it can be examined whether their dif-
ference in activity patterns explain temporal
separation of the two species in the wild. The
mean nocturnalities of field captures in three
years were 48% т М. normalis and 91% in T.
albolabris, a ratio of 0.53 (Asami, 1988). The
interspecific ratios of nocturnalities observed
in the present outdoor and indoor experi-
ments (0.56 and 0.45, respectively) are
closely comparable to the ratio from the wild,
indicating that the present results are a good
representation of the relative activity patterns
of the two species in nature.

Nevertheless, comparison of nocturnalities
in the wild and in the present experiments
suggests that both species tend to be more
nocturnal in the wild. This difference could be
due to the high humidity maintained in the
experiments. In the natural habitats, the hu-
midity is typically 100% from midnight to noon
in summer, whereas it was kept at that level
inside the containers for the entire experimen-
tal periods. The daily change of temperature
was often larger outdoors than in the labora-
tory. The daily shifts of the physical conditions
were, accordingly, greater in the field than in
the experiments. Hence, their nocturnalities
may well be higher in the wild on fine days
than those observed in this study.

In the treatment of pairing individuals, the
density of animals needs to be considered as
existence of one individual could have an ef-
fect on another. In the present paring treat-
ments, the density was higher than in nature.

ACTIVITY PATTERNS OF LAND SNAILS 405

Thus, the effect of individual interaction would
be enhanced if it exists. There was, however,
no significant difference in either nocturnality
or the total activity between paired and iso-
lated snails, except for M. normalis, which
showed higher activity in conspecific pairs.
Mesodon normalis might be more sensitive to
high density or might tend to respond to con-
specifics more promptly than 7. albolabris, al-
though no courtship was observed in these
experiments. The absence of pairing effects,
between and within species, on nocturnality
also indicates that individual isolation in the
experiments did not cause significant artifacts
in activity patterns, relative to the field situa-
tion where snails could encounter each other.
This study has demonstrated substantial
separation of activity times in coexisting spe-
cies of land snails. Further studies are
needed to understand the ecological and ev-
olutionary causes of this divergence.

ACKNOWLEDGEMENTS

| am grateful to Jim Murray, Gene Block,
Diane Campbell, Blain Cole, Ray Dueser, and
Ken Emberton for stimulating discussion and
criticisms of this study. | also thank J. Murray
and two anonymous reviewers for critical
comments on the manuscript and Martha
Dahlen for laboratory assistance. The re-
search was supported by USA National Cap-
ital Shell Club Scholarship, Sigma-Xi Grants-
in-Aid of Research, and fellowships from
University of Virginia.

LITERATURE CITED

ABDEL-REHIM, A. H., 1983, The effects of temper-
ature and humidity on the nocturnal activity of
different shell colour morphs of the land snail Ari-
anta arbustorum. Biological Journal of the Lin-
nean Society, 20: 385-395.

ASAMI, T., 1988, Temporal segregation of two
sympatric species of land snails. Venus, 47:
153-172.

ASAMI, T. (in press), Interspecific differences in
desiccation tolerance of juvenile land snails.
Functional Ecology, 7(5).

BAILEY, S. E. R., 1981, Circannual and circadian
rhythms in the snail Helix aspersa Muller and the

photoperiodic control of annual activity and re-
production. Journal of Comparative Physiology,
142: 89-94.

BARNES, H. F. & J. W. WEIL, 1942, Baiting slugs
using metaldehyde mixed with various sub-
stances. Annals of Applied Biology, 29: 56-68.

BARNES, H. F. & J. W. WEIL, 1945, Slugs in gar-
dens: their numbers, activities and distribution.
Part 2. Journal of Animal Ecology, 14: 71-105.

BLINN, W. C., 1963, Ecology of the land snails Me-
sodon thyroidus and Allogona profunda. Ecology,
44: 498-505.

CAMERON, В. А. D., 1970, The Effect of temper-
ature on the activity of three species of helicid
snail (Mollusca: Gastropoda). Journal of Zool-
ogy, 162: 303-315.

COWIE, R. H. & J. S. JONES, 1987, Ecological
interactions between Cepaea nemoralis and Ce-
paea hortensis; competition, invasion but no niche
displacement. Functional Ecology, 1: 91-97.

DAINTON, В. H., 1954a, The activity of slugs. I.
The induction of activity by changing tempera-
tures. Journal of Experimental Biology, 31: 165-
187.

DAINTON, B. H., 1954b, The activity of slugs. Il.
The effect of light and air currents. Journal of
Experimental Biology, 31: 188-197.

DAINTON, В. H. & J. WRIGHT, 1985, Falling tem-
perature stimulates activity in the slug Arion ater.
Journal of Experimental Biology, 118: 439—443.

EMBERTON, К. C., 1988, The genitalic, allozymic,
and conchological evolution of the eastern North
American Triodopsinae (Gastropoda: Pulmo-
nata: Polygyridae). Malacologia, 28: 159-273.

EMBERTON, K. C., 1991, The genetic, allozymic,
and conchological evolution of the tribe Mesod-
ontini (Pulmonata: Stylomatophora: Polygy-
ridae). Malacologia, 33: 71-178.

FORD, D. Ч. С. 4 A. COOK, 1987, The effect of
temperature and light on the circadian activity of
the pulmonate slug, Limax pseudoflavus Evans.
Animal Behaviour, 35:1754-1765.

GELDERLOOS, О. С., 1979, Locomotor activity
patterns of the Roman garden snail (Helix poma-
tia L.) in light and dark conditions. The American
Midland Naturalist, 101: 218-222.

HENNE, F. C., 1963, Effect of light and temperature
on locomotory activity of Polygyra albolabris
(Say). Bios, 34: 129-133.

HUBRICHT, L., 1985, The distribution of the native
land mollusks of the eastern United States. Fiel-
diana, Zoology, New Series, 24: 1-191.

HUTCHINSON, G. E., 1957, Concluding remarks.
Cold Spring Harbor Symposia on Quantitative Bi-
ology, 22: 415-427.

INGRAM, W. M., 1940, Daylight activity of land mol-
lusks. Nautilus, 54: 87-90.

INGRAM, W. M., 1944, Shell cleaning and epi-
phragm removal by Triodopsis albolabris (Say).
Nautilus, 57: 138-141.

KARLIN, E. J., 1961, Temperature and light as fac-
tors affecting the locomotor activity of slugs. Nau-
tilus, 74: 125-130.

406 ASAMI

LEWIS, В. D., 1969a, Studies on the locomotor ac-
tivity of the slug Arion ater (Linnaeus). |. Humid-
ity, temperature and light reactions. Malacologia,
7: 295-306.

LEWIS, В. D., 1969b, Studies on the locomotor ac-
tivity of the slug Arion ater (Linnaeus). Il. Loco-
motor activity rhythms. Malacologia, 7: 307-312.

MACHIN, J., 1975, Water relationships. Pp. 105—
163, in: V. FRETTER & J. PEAKE, eds., Pulmonates.
Volume 1. Academic Press, London.

MCCRACKEN, G. F. & P. F. BRUSSARD, 1980,
The population biology of the white-lipped land
snail, Triodopsis albolabris: genetic variability.
Evolution, 34: 92-104.

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

NEWELL, P. F., 1966, The nocturnal behavior of
slugs. Medical and Biological Illustration, 16:
146-159.

PILSBRY, H. A., 1940, Land mollusca of North
America (north of Mexico), Volume 1, Part 2. The
Academy of Natural Sciences of Philadelphia,
Monographs, Number 3: 575-994.

RIDDLE, W. A., 1983, Physiological ecology of land
snails and slugs. Pp. 431-461, in: W. D. RUSSELL-
HUNTER, ed., The Mollusca. Volume 6. Academic
Press, Orlando.

ROLLO, C. D., 1982, The regulation of activity in
populations of the terrestrial slug Limax maximus
(Gastropoda; Limacidae). Researches on Popu-
lation Ecology, 24: 1-32.

ROLLO, C. D., 1991, Endogenous and exogenous
regulation of activity in Deroceras reticulatum, a
weather-sensitive terrestrial slug. Malacologia,
33: 199-220.

SCHMIDT-NIELSEN, К., С. В. TAYLOR & А.
SHKOLNIK, 1972, Desert snails: Problems of
survival. Symposia of the Zoological Society of
London, 31: 1-13.

SOKOLOVE, P. G., C. M. BEISWANGER, D. J.
PRIOR & A. GELPERIN, 1977, A circadian
rhythm in the locomotor behaviour of the giant
garden slug Limax maximus. Journal of Experi-
mental Biology, 66: 47-64.

SOLEM, A., 1976, Comments on eastern North
American Polygyridae. Nautilus, 90: 25-36.

SZLAVECZ K., 1986, Food selection and nocturnal
behavior of the land snail Monadenia hillebrandi
mariposa A. G. Smith (Pulmonata: Helmintho-
glyptidae). Veliger 29: 183-190.

TILLING, S. M., 1986, Activity and climbing behav-
iour: a comparison between two closely related
landsnails species, Cepaea nemoralis (L.) and C.
hortensis (Müll.). Journal of Molluscan Studies,
52: 1-5.

WAREING, D. В. & 5. Е. В. BAILEY, 1985, The
effects of steady and cycling temperatures on the
activity of the slug Deroceras reticulatus. Journal
of Molluscan Studies, 51: 257-266.

WEBLEY, D., 1964, Slug activity in relation to
weather. Annals of Applied Biology, 53:
407-414.

Revised Ms. accepted 8 June 1993

MALACOLOGIA, 1993, 35(2): 407—410

THE LOTTERY OF BIBLIOGRAPHICAL DATABASES:
À REPLY TO EDWARDS & THORNE

Philippe Bouchet and Jean-Pierre Rocroi

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

We understand that our finding that over
20% of new molluscan genus-group names
are omitted by Zoological Record (ZR) has
caused surprise to the editors of this journal,
as it has surprised us and many of our col-
leagues, all of us being regular users of this
inescapable and valuable bibliographical tool.
Indeed, the prevailing intuitive opinion is that
approximately 5-7% of the names are отй-
ted. In our answer to Edwards & Thorne (Ma-
lacologia, 35: 153-154), we want to empha-
size two points:

(1) Omission affects all sorts of journals,
including high-profile international journals;

(2) the high rate of omission and other no-
menclatural defects highlight the risks and
problems of establishing a “List of Available
Generic Names in Zoology’ based on
Neave’s Nomenclator Zoologicus and ZR, as
is being currently considered by the Interna-
tional Commission on Zoological Nomencla-
ture (ICZN).

Which Names Get Omitted?

Most of the taxonomists with whom we have
discussed this question seemed to believe that
names that escape ZR were originally pub-
lished in very obscure sources. In discussions,
these scientists often frankly suggest that
omission from ZR and nomenclatural oblivion
are, after all, probably deserved and that the
authors of such names are themselves re-
sponsible for the ill fate of their names. These
beliefs are wrong, as we show below.

To supplement the data provided in our pa-
per (Malacologia, 34: 75-86), we have ana-
lyzed the place of publication of 370 genus-
group names omitted from Nomenclator
Zoologicus and ZR. These are names starting
with the letters A-K, mostly published be-
tween 1940 and 1975. Places of publication
were divided into three categories:

(a) books and collections of books,

(b) main-stream scientific journals,

Admittedly, it is a matter of personnal opin-
ion whether one should rank a journal as
main-stream or little-known. Our appreciation

407

is best explained by a series of examples.
Among main-stream journals, we have
ranked: Memoir of the Geological Society of
America; Mémoires de l'Institut Royal des
Sciences Naturelles de Belgique; Trudy Zoo-
logicheskogo Instituta; Sarsia; Malacologia
(the very journal where this paper is pub-
lished! Omitted names: Globicarina Water-
house, 1965, Malacologia, 3(3): 374; Caliba-
sis and Idabasis Taylor, 1966, Malacologia, 4:
41, 42); Oceanologia et Limnologia Sinica;
etc.

Among little-known and obscure journals,
we have ranked: Science Reports of the To-
hoku University; Bulletin du Muséum d’His-
toire Naturelle de Marseille; Vestnik Mos-
kovskogo Universiteta; Gastropodia; Leaflets
in Malacology; Bulletin of the Department of
Geological Sciences, University of California
Publications; etc.

The first category, books and collections
of books, is itself a mixed category, and in-
cludes both main-stream and little-known ti-
tles. Examples are Habe, 1961, Coloured il-
lustrations of the shells of Japan; Fossils
of central southern China; Nordsieck, 1972,
Die miozäne Molluskenfauna von Miste-
Winterswijk; Das Tierreich; Galathea Report;
Starobogatov, 1970, Fauna molliuskov i гоо-
geografiskoe rajonirovanie kontinentalnits vo-
doemov zemnogo shara; etc.

Our results show that 141 omitted names
(38%) were published in books and series;
153 names (41%) were published in main-
stream journals; 76 names (21%) were pub-
lished in little-known or obscure journals. If
country of publication is considered, 98
names (26%) were published in books or jour-
nals of the former USSR; 91 names (25%) in
USA; 48 names (13%) in Japan. Only 9
names (2%) were published in China, but this
is because Chinese output did not start until
after the end of the cultural revolution (1976),
that is, later than the time span of our study.
These results demonstrate that there is no
correlation between omission from Nomen-
clator Zoologicus and circulation of a journal
or book.

408 BOUCHET & ROCROI

Most, if not all, of the journals, main-stream
or obscure, cited above are in principle cov-
ered by ZR. Hence, obscurity is not the main
reason for omission. On a number of occa-
sions, we have found that four new names
published in a paper are correctly gazetted in
ZR, whereas a fifth name published in the
same paper has been omitted. Or a whole
paper published in a погтайу recorded jour-
nal has been omitted. We regret to say that
carelessness seems to be a not infrequent
source of omission. As our results show,
names published in books constitute a major
proportion of omitted names. Taxonomists
have also noticed that many books and irreg-
ular series are recorded in ZR only several
years after their publication, when most prac-
ticing scientists know of these books within a
few weeks or months after their publication;
often, book reviews have also been published
in main-stream journals. We believe that the
main reason for this regrettable situation is
that there are few, if any, personal contacts
between the recorders and bibliographers, on
one side, and the people that write the books
and monographs, that is, the malacologists
and taxonomists, on the other side. In our era
of frequent and easy travel, we regret that, to
our knowledge, the staff of ZR has attended
only once (in Edinburgh, 1986) an Interna-
tional Malacological Congress, which are held
every three years in Europe. We believe that
attendance of such and other similar con-
gresses in USA and Russia would дгеайу en-
hance the efficiency of ZR, when malacolo-
gists could identify the Mollusca section of ZR
with the face of a person whom they have
personally met.

The Risks of a “List of Available Generic
Names in Zoology” Based on Nomenclator
Zoologicus and ZR

Stability of names has become a much de-
bated topic, both in botanical (Hawksworth,
1991) and zoological (Ng, 1991) nomencla-
ture. The International Commission on Zoo-
logical Nomenclature resolved at its Univer-
sity of Maryland meeting “to enter into
negotiations with Biosis with a view to devel-
oping a data base of generic names as a list
of available names” (Anonymous, 1990). The
report (Anonymous, 1991) of the Amsterdam
meeting of the Commission further stated that
“Biosis has made good progress in the prep-
aration of a draft list of generic names pub-
lished between 1758 and 1990, based on

Neave’s Nomenclator Zoologicus and Zoo-
logical Record.”

Indeed, all taxonomists including ourselves
would dearly like to have a complete nomen-
clator of generic names under a single cover,
and it was precisely for lack of such a cata-
logue that we started compiling our own, al-
beit limited to Mollusca. However, we seri-
ously question the value of the Biosis
nomenclator when up to 23% of recently pub-
lished names are omitted.

More importantly, we want to stress the
risks of making this list a formal List of Avail-
able Generic Names in Zoology. In an unoffi-
cial report of the 1990 ICZN meeting, Savage
(1990) suggested that “most importantly, only
the generic names on this list would be avail-
able for use. Any other names, subsequently
discovered or not, would not exist for nomen-
clatural purposes.” This is what Savage calls
“the statute of limitations for the resurrection
of old names.” We call it a receipe for injustice
and chaos. By using such expressions as
“resurrection of old names,” Savage tends to
suggest that names omitted by Nomenclator
Zoologicus and ZR belong to the very ob-
scure Category, and that those zoologists dis-
covering them are merely book archeologists
that disrupt the work of real taxonomists. We
have amply demonstrated that in malacology,
and probably many branches of invertebrate
(paleo)zoology and vertebrate paleozoology
as well, there are literally thousands of no-
menclaturally available names that get omit-
ted by ZAR. When, for example, Aliomactra
Stephenson, 1952 (U. S. Geological Survey
Professional Paper, 242: 125) or Dancea
Zilch, 1960 (Handbuch der Paläozoologie,
6(2): 730) are omitted by ZR, should Stephen-
son and Zilch be blamed for that? Should А!-
iomactra and Dancea be deemed not to exist
for nomenclatural purposes? We strongly re-
ject this idea, as we reject the idea that an
Official List of Available Generic Names in Zo-
ology should be compiled on a commercial
basis.

In his report of the ICZN 1990 meeting,
Savage (1990) further suggested that “at the
time of publication (e.g., 1996), the dates in
the list (regardless of any subsequent find-
ings) would be the final determinants of prior-
ity.” Again, the rationale behind this point is
probably to avoid changes of names as a re-
sult of bibliographical subtleties, an opinion
that many taxonomists would defend. How-
ever, there are again hidden sides that have
apparently been overlooked: we refer to the

LEMBRITO THE EDITOR 409

listing of generic names by Nomenclator Zoo-
logicus whereas some of these names are
unavailable under ICZN Art. 13, which rules
that genus-group names published after 1930
must have a diagnosis and a type species.

We will give two examples:

Hennocquia is listed in Neave and credited
to Haber, 1932, Fossilium Catalogus, pars 53,
220. However, Haber designated a type spe-
cies but omitted a diagnosis, and Hennocquia
is unavailable under Ап. 13a(i). Wenz, 1938,
Handbuch der Paláozoologie, 6(1): 219, first
provided a diagnosis and type species. Hen-
nocquia should thus be credited to Wenz
(1938). Pseudohelenoconcha is listed in
Neave and credited to Germain, 1932, C.R.
Congr. Soc. Sav. Sci., 1929, 7 (sic, should be
6). Germain failed to designate a type spe-
cies, and the name is unavailable under Art.
13b. Zilch, 1959, Handbuch der Paläozoolo-
gie, 6(2): 215, provided a diagnosis and type
species designation, and is the author of
Pseudohelenoconcha. These examples
should suffice to demonstrate the risks of
binding too strongly the Code of Zoological
Nomenclature and the databases operated
and marketed by Biosis.

In conclusion, we want to outline briefly our
suggestion to introduce a mandatory system
of registration of new zoological names. We
have proposed that the next edition of the
Code adds a new article:

“A copy of every work containing the intro-
duction of a new zoological name must be
sent, by its author or publisher, to the Inter-
national Commission on Zoological Nomen-
clature. Receipt of the publication by the Sec-
retariat of ICZN is necessary to validate a new
name.

When all other relevant provisions of the
Code are satisfied, the date of validity of a
new name is the date (day, month, year)
when the publication containing its introduc-
tion is formally received by the Commission.

The International Commission on Zoologi-

cal Nomenclature publishes every year a list
of the new taxa received at its offices, to-
gether with complete bibliographical refer-
ence, and the date (day, month, year) of their
availability.”

When this proposal was submitted to ICZN
by the senior author, he recommended that
the Zoological Record/Nomenclator Zoologi-
cus be associated with the compilation of
these annual lists, which do not duplicate the
current contents of the Zoological Record.

This proposal would clearly benefit the
more professional journals over the more lo-
cally produced, unedited ones, or with editors
not even aware that there exists a Code of
Nomenclature. With time the authors will
know their interest is to seek publication in
those professional journals that offer a better
service with regard to this provision of the
Code. Those that do not comply can be ig-
nored. But this will be the result of their own
carelessness, not the result of the lottery of
bibliographical databases.

LITERATURE CITED

ANONYMOUS, 1990, International Commission on
Zoological Nomenclature. General session of the
Commission, University of Maryland, 4 July
1990. Bulletin of Zoological Nomenclature, 47:
246-249.

ANONYMOUS, 1991, International Commission on
Zoological Nomenclature. General session of the
Commission, Amsterdam, 2-6 September 1991.
Bulletin of Zoological Nomenclature, 48:
286-292.

HAWKSWORTH, D. L., ed., 1991, Improving the
stability of names: needs and options. Regnum
Vegetabile, 123.

МС, Р. К. L., 1991, How conservative should no-
menclature be? Comments on the principle of pri-
ority. Bulletin of Zoological Nomenclature, 48:
87-91.

SAVAGE, J. M., 1990, Meetings of the International
Commission on Zoological Nomenclature. Sys-
tematic Zoology, 39: 424—425.

410 BOUCHET & ВОСВО!
NOTE BY A CO-EDITOR

While | find convincing the argument by
Bouchet & Rocroi for a greatly improved sys-
tem to capture and publicly recognize valid
new taxa, | find one aspect of the methodol-
ogy they propose for doing so particularly
troubling.

At the same time they call for improvement
of the capture of new taxa by the laudable
expedient of having journal editors (and au-
thors) send all of them to a central repository
for official recognition, they surrender the en-
tire process and leave the dates of recogni-
tion in the hands of the most inefficient and
frequently careless bureaucracy in the world,
the postal authorities.

A name would, they propose, only be offi-
cial when it arrives in the hands of the central
authority and is suitably blessed. Even as-
suming that the costs of sending all issues of
all journals and other works to the central au-
thority is borne by the publishers and editors
and the costs thus saved put into staff to do
the extraction and blessing, this is bound to
be a time-consuming and tedious task that
will require much staff time. Perhaps this is a
price that must be paid.

The weak link, however, is the postal sys-
tem. | estimate that about a third of the things
mailed from Latin America never reach their
destinations. An “airmail” package from
South America can take two months. | esti-

mate based on recent experience that well
over half of the materials going back and forth
between the Far East of the Soviet Union and
the West never reach their destinations at all.
So, must we leave important taxonomic deci-
sions and the all-important dating of taxo-
nomic works in the hands of the Russian, Ital-
ian, or Colombian postal authorities? Are
these bureaucrats to be new arbiters of prior-
ity and validity?

| would suggest that instead we rely upon
the real dates of publication, and then make
every effort to get the materials into the hands
of the Commission of whatever central repos-
itory is chosen.

Eugene Coan, Co-Editor

The editor-in-chief of Malacologia welcomes let-
ters that comment on vital issues of general im-
portance to the field of Malacology, or that com-
ment on the content of the journal. Publication is
dependent on discretion, space available and, in
some cases, review. Address letters to: Letter to
the Editor, Malacologia, care of the Department
of Malacology, Academy of Natural Sciences,
19th and the Parkway, Philadelphia, PA 19103.

MALACOLOGIA, 1993, 35(1-2): 411-420

INDEX

Taxa in bold are new; page numbers in
bold indicate pages on which new taxa
are described; pages in /ta/ics indicate
figures of taxa.

abrupta, Panopea 338

Abyssochrysos 270

Acanthina 160, 195, 197, 233, 234,
242-245

Acanthina monodon 161, 172, 229, 230,
231, 246, 249
muricata 243

aculeata, Mancinella 188, 189

aculeata, Thais 213

acuta, Hydrobia 389-398; 390

acutum, Cyclostoma 389

adelieana, Pareledone 354

adversum, Murex 273

aegrota, Dicathais 183

aegrota, Thais 180

affinis, Partula 43-61

affinis, Partula otaheitana 43

Agnewia 213

Alabina 269, 270

Alba 262, 271

alba, Ricinula 190

albolabris, Neohelix 363, 366-368

albolabris, Triodopsis 399-406

album, Sistrum 183

allenı, Neohelix 363, 366, 367

alouina, Mancinella 161, 187, 188, 190,
246, 247

alouina, Vitreledonella 344

alternata, Diastoma 291

alternata, Turritella 291

alternatum, Bittiolum 262, 287, 288, 291

amabilis, Partula otaheitana 44

Amphetritus 344

amygdala, Cronia 161, 164, 169, 176,
177, 178, 246, 247

amygdala, Ригрига 176

Anadara granosa 29, 30

Anadonta grandis 34, 35

Aneurychilus 288

angulifera, Purpura 179

Anodonta 381

Aphrodoctopus 351-359

Aphrodoctopus schultzei 353-356, 358

arbustorum, Arianta 89-98

Arca pernula 141
rostrata 141

Arcidae 320

Arctica islandica 30

arcticus, Bathypolypus 357

arenaria, Mya 29

argenteus, Idas 21-41; 23, 25

Argonautida 344

Argyropeza 262-265, 268, 269, 304-306
divina 266, 304, 305

Arianta 399
arbustorum 89-98

Ariantinae 74

ascensionis, Purpura 213
Ashmunella 365, 365
aspersa, Helix 99-117, 382
aspersa, Helix aspersa 100, 114-116
aspersa, Ricinula 190
Astarte elliptica 30
Astartidae 320
Atenia 72
atromarginatum, Cerithium 274
attenuata, Samoana 44, 53, 55
attenuatum, Bittium 296, 297
attenuatum, Lirobittium 262, 298
aurantia, Partula 55
aurantiaca, Purpura 176
aurorae, Pareledone 354
avellana, Buccinum 176
avellana, Cronia 183
Bahlakia leilae 270
barbula, Osteophora 72
Bathymodiolus thermophilus 35, 149
Bathypolypodinae 344
Bathypolypus 345, 346, 349
arcticus 357
faeroensis 357
Batillariella estuarina 270
Batillariidae 270
belcheri, Forreria 161, 164, 172, 222,
227, 228, 232, 246, 249
Benthelodone 345, 346, 349
Benthoctopus 344-346, 349
bimaculatus, Octopus 354, 357
bimaculoides, Octopus 354, 357
Bitinella 270
Bittiinae 261-313
Bittiolum 261, 263-269, 272, 287, 306
alternatum 262, 287, 288, 291
fastigiatum 288
varium 262, 266, 282, 287, 288, 289,
2907291
Bittiscalia 270
Bittium 261, 262, 264-271, 272, 273,
274, 283, 287, 291, 295, 300, 304,
306, 307
attenuatum 296, 297
boeticum 262
californicum 269
catalinensis 295, 296
eschrichtii 292
exile 306
granarium 301
hiloense 270
impendens 262, 275, 282, 283
lawleyanum 270
nigrum 291
parcum 270, 283, 284, 287
podagrinum 287, 288
reticulatum 262, 266, 270, 271, 273,
ZU BUS, BUI Bikes, 230728172828
283
simplex 270
subplanatum 296
vitreum 306

411

412

zebrum 262

(Brachybittium) caraboboense 270
(Lirobittium) catalinense 296
(Lirobittium) subplanatum 296
(Semibittium) subplanatum 296
(Stylidium) eschrichtii 292

(Stylidium) eschrichtii icelum 292, 294

Bivalvia 315-342

bizonalis, Purpura 200

boeticum, Bittium 262

Boonea impressa 119-134; 120, 122-
125, 127 020, 130. 131

Brachybittium 270

Bradybaena fruticum 371-388; 377-380

similaris 380, 381
Bradybaenidae 74
briareus, Octopus 354
brightoniana, Cymia 179
bronni, Purpura 213
bronni, Thais 213
Buccinidae 156, 158
buccinoidea, Purpura 200
Buccinum avellana 176
concholepas 173
coronatum 194
filosum 200
francolinus 194
haemastoma 210
haustorium 186
haustrum [non-binomial] 186
lapillus 198, 200
orbita 180
patulum 203
persicum 207
sertum 192, 194
situla 194
succinctum [non-binomial] 180
tectum 179
bufo, Purpura 244
burryi, Octopus 354
buvignieri, Helix 71
Cacozelia 299, 300
Cacozeliana 262-264, 267, 268, 272,
282, 295, 300, 301
granaria 262, 266, 300, 301, 302,
303
caerulea, Patinigera 139
californianus, Mytilus 139
californicum, Bittium 269
californicus, Octopus 354
callistiformis, Tindaria 34
Calyptogena magnifica 149
Camaenoidea 75
Canariella 71, 72, 74
cancellatum, Cerithium 299, 300
candicans, Helicella 79-87; 80
Canrena 183
caparti, Eledone 352-354, 358
Capistrocardia 336
caraboboense, Bittium (Brachybittium)
270
Caracollina 71, 73-75
lenticula 63-77; 64, 66-69
Caracollinae 63
Caracollinini 63, 71, 72
Cardiidae 320, 325-332, 339
cardissa, Corculum 323

INDEX

carlgreni, Pareledone 358

Cassiella 262, 263, 265, 268, 307
abylensis 307, 308, 309

catalinense, Bittium (Lirobittium) 296

catalinense, Lirobittium 296
catalinensis, Bittium 295, 296
Cellana radiata 139

celtica, Purpura 200

Cepaea 385, 399
hortensis 95, 399
nemoralis 95, 114, 371, 375, 381,

384

Cephalopoda 344

Cerastoderma edule 35

Cerion 383

Cerithidium 265, 269, 270

Cerithiidae 261, 263, 264, 270, 271

Cerithioidea 271

Cerithiolum 273

Cerithiopsis 273, 274

Cerithium 261-264, 301
(Bittium) gibberulum 288
atromarginatum 274
cancellatum 299, 300
columellare 288
egenum 274
exilis 306
fritschi 270
gibberulum 288
granarium 300, 301
hawalensis 284
impendens 282, 283
lacertinum 299-301
lacteum 273, 274
latreillei 273, 274
proteum 270
reticulatum 274
scabridum 270
submamillatum 270
varıum 288
zebrum 274

Cernuella virgata 89

charcoti, Pareledone 354

charrua, Vosseledone 354

chierchiae, Octopus 354

Chioninae 333

Chorus 240, 242

Chrysallida obtusa 132
spiralis 132

Ciliella 71-74

Ciliellidae 71, 74

Ciliellinae 63, 71, 72, 74

Ciliellini 71

cinculata, Trochia 229, 161, 172, 229,
231, 231, 242, 246, 249

cinerea, Urosalpinx 230, 249

Cirrata 343

cirrhosa, Eledone 352-355

Cistoctopus 352

Cistopsus 344
indicus 354, 355, 357

citrina, Conothais 201

clavigera, Thais 178

clavula, Liostomia 132

Cleidophorus 336

Colina 271

Columbariidae 156, 158

INDEX 413

Columbariinae 156, 158
columellare, Cerithium 288
columellaris, Plicopurpura 205, 206
columellaris, Purpura 205, 207
complanata, Eurythoe 185
Conchlolepas 244
Concholepa 173
Concholepadidae 156
Concholepas 173, 233, 234, 235, 240-
243, 245
concholepas 160, 161, 173, 174,
175, 246, 247
peruviana 173
concholepas, Buccinum 173
concholepas, Concholepas 160, 161,
178147741175; 246) 247
Conchopatella 173
Conothais 201
citrina 201
consul, Purpura 210
Coralliophila 158, 244
rolani 200
Coralliophilidae 156, 158
Coralliophilinae 156, 158
Corbicula fluminea 29
Corculum cardissa 323
coronata, Pinaxia 201
coronatum, Buccinum 194
Cosmocerithium 269
crassa, Partula otaheitana 43
crassa, Purpura 218
Crassostrea virginica 119
Cristilabrum 384
Cronia 160, 176, 183, 233-235, 240,
244, 245
amygdala 161, 164, 169, 176, 177,
178, 246, 247
avellana 183
margariticola 178
Cryptaulax 262,269, 306
Cryptaulaxinae 269
Cryptomya 334
Cryptosaccus 71,75
crystallina, Varicopeza 268
Ctenoglossa 344
Cultellidae 320, 325-332
Cultellus 336
Cuma 178, 179
sulcata 178, 179
Cumopsis 178, 179
Cyclostoma acutum 389
Cyma 178
Супиа 178, 213, 231, 233, 234, 238-
240, 243-245
brightoniana 179
теста 1611172, 177, 178, 179, 240;
246, 247
Cynthia praeputialis 183
dactylus, Pholas 18
Dahlakia 262, 270
deaurata, Nacella (Patinigera) 135-140
defilippi, Macrotritopus 354
deltoidea, Thais 244
demissa, Geukensia 30
Dendropoma gregaria 185
depressa, Opisthoteuthis 356
Deroceras 399

Diala 262, 264, 271, 301
Dialidae 271
diaphana, Samoana 44, 53, 55
Diastoma 300
alternata 291
varium 288
Diastomatidae 288
Diastomidae 288
Dicathais 180, 233-235, 240, 242, 244,
245
aegrota 183
orbita= 161, 180, 787, 183, 246.247
digitata, Ricinula 183
digueti, Octopus 354
divina, Argyropeza 266, 304, 305
divisus, Tagelus 338
Donacidae 320, 325-332, 334
Drepanostoma 71-73
Drosophila willistoni 383
Drupa 176, 183, 233-235, 240, 244,
245
grossularia 183, 184, 186, 240
lobata 240
morum 161, 183, 184, 185, 186,
240, 246, 247
ricinus 183, 185, 240
rubusidaeus 183
tuberculata 157
uva 190
Drupella 160, 183
Drupina 183
Drupinae 156, 158
dubia, Thais 240
Dyakia striata 1-7, 9-19; 11-17
echinulata, Mancinella 190
Ecphora 240, 242, 245
Rn 161252242246;
49
Ecphorinae 242, 245
edule, Cerastoderma 35
edulis, Mytilus 29, 31
egenum, Cerithium 274
Elachista 269
Elassium 269, 270
Eledone 345-347, 349, 351-359
caparti 352-354, 358
cirrhosa 352-355
gaucha 354
massyae 354, 358
moschata 354, 355
palari 358
thysanophora 358
Eledoninae 344, 347
elegantissimum, Murex 273
elevatus, Mesodon 363
elliptica, Astarte 30
Elliptio 320
Eloninae 71, 73
emarginata, Nucella 242
EnsIS2323"335
Enteroctopus 344
Ergalataxinae 233
erinacea, Ocenebra 243
eschrichtii, Bittium 292
eschrichtii, Bittium (Stylidium) 292
eschrichtii, Stylidium 262, 266, 292,
293, 294, 295

414

eschrichtii, Turritella 292

estuarina, Batillariella 270

Euaxoctopus 344

Eubittium 270

eulimoides, Odostomia 131, 132

Euomphalia 71

Euomphaliinae 73, 75

Euparyphinae 74

Eurythoe complanata 185

exile, Bittium 306

exile, Zebittium 306, 308

exilis, Cerithium 306

faeroensis, Bathypolypus 357

Falkneria 72, 73

fastigiatum, Bittiolum 288

filosum, Висстит 200

filosus, Octopus 354

fitchi, Octopus 354

floridana, Purpura 210

floridana, Stramonita 210

floridana, Stramonita haemastoma 157

fluminea, Corbicula 29

foliata, Purpura 207

fontanianus, Robsonella 354

forbesii, Purpura 210

Forreria 158, 229, 231, 233, 234, 242-
245

Forreria belcheri 161, 164, 172, 222,
227. 228) 232.246, 249

francolina, Nassa 193, 194, 195, 240,
248

francolinus, Buccinum 194

fritschi, Cerithium 270

fruticum, Bradybaena 371-388; 377-380

fucus, Murex 213

fulvescens, Hexapiex 171

fulvescens, Murex 171

fulvescens, Muricanthus 161, 164, 166,
167 160, 169, 171172222225;
241, 246, 249

Fulvia 328

Gari 334

Gasuliella 75

Gasullia 74

Gasulliella 73-75

gaucha, Eledone 354

gemmulata, Mancinella 188

gemmulata, Ригрига 188

Geukensia demissa 30

gibberulum, Cerithium 288

gibberulum, Cerithium (Bittium) 288

Gittenbergeria 75
turriplana 72

Gittenbergia 73, 74

glacialis, Volema 188

Glyptozaria 262

Gourmya 300

granaria, Cacozeliana 262, 266, 300,
301, 302, 303

granarium, Bittium 301

granarium, Cerithium 300, 301

grandis, Anadonta 34, 35

Graneledone 345-347, 349

Graneledoninae 344

granosa, Anadara 29, 30

granulata, Morula 190, 192

granulata, Purpura 190

INDEX

grasslei, Мисшапа 141-150; 143-148
gregaria, Dendropoma 185
grisea, Thais 210
grossularia, Drupa 183, 184, 186, 240
haemastoma, Buccinum 210
haemastoma, Stramonita 157, 161, 168,
РАО Aili, AVA, Caton Zaks
Halolimnohelicidae 71, 74
Halolimnohelix sericata 74
Hapalochlaena 344, 352, 354
harrissoni, Pareledone 354
haustorium, Buccinum 186
haustorium, Haustrum 161, 186, 787,
234, 238, 241, 242, 244, 246, 247
haustrum, Buccinum [non-binomial] 186
Haustrum 186, 233, 234, 236, 239,
237, 240, 243-245
haustorium 161, 186, 787, 234, 238,
241, 242, 244, 246, 247
pictum 216
zealandicum 186
hawaiensis, Cerithium 284
haysae, Thais floridana 210
hederacea, Stramonita 194
Heleobia 390
Helicella candicans 79-87; 80
Helicidae 71, 74, 75
Helicinae 74
Helicodonta 71-73, 75
Helicodontidae 63, 71-74
Helicodontinae 63, 71-73
Helicoidea 71, 73, 75
Helix 385, 399
aspersa 99-117, 382
aspersa aspersa 100, 114-1
aspersa maxima 100, 114-1
buvignieri 71
hispanica 71
lucorum 115
pomatia 89
texta 115, 116
turriplana 71
Hennocquia 409
Hexaplex fulvescens 171
Hiatellidae 334
hidalgoi, Thais (Stramonita) 210
hiloense, Bittium 270
hippocastanum, Murex 213
hispanica, Helix 71
horida, Ricinula 183
horrida, Ricinula 183
horridus, Macrotritopus 354
hortensis, Cepaea 95, 399
hubbsorum, Octopus 354
Hydrobia 381, 383, 385, 389-398
acuta 389-398; 390
ulvae 389-398; 390
ventrosa 389-398; 390
Hygromia 71
Hygromiidae 71, 73-75
Hygromiinae 73-75
Hygromioidea 72-75
Hyriidae 320
lberus 75
icelum, Bittium (Stylidium) eschrichtii
292, 294 |

INDEX

Idas argenteus 21-41; 23, 25
washingtonius 30

imbricata, Ригрига 200

impendens, Bittium 262, 275, 282, 283

impendens, Cerithium 282, 283

impressa, Boonea 119-134; 720, 122-
(DQ VAIS YOR 1508181

impressa, Odostomia 119

indicus, Cistopus 354, 355, 357

inerma, Purpura 207

Infracerithium 269

Inobittium 273

lopas 192, 201

islandica, Arctica 30

Isseliella 270

Ittibittium 263-268, 271, 272, 282, 283-
284
parcum 262, 266, 284, 285, 286, 287

jackieburchi, Partula 43-61

jackieburchi, Samoana 54

Japetella 346

Japonica, Opisthoteuthis 356

Jopas 194, 220

kivuensis, Vicarithelix 74

lacertinum, Cerithium 299-301

lacteum, Cerithium 273, 274

Laevicardiinae 328

Lampsilis radiata 34, 35

lanceolata, Resania 338

langi, Thais (Stramonita) 210

lapillus, Buccinum 198, 200

lapillus, Nucella 157, 161, 166, 167,
168, 169, 198, 199, 200, 201, 246,
248

lapillus, Purpura 231

lapillus, Thais 231

Laternula 336

Latia neritoides 18

latreillei, Cerithium 273, 274

lawleyanum, Bittium 270

Ledidae 336

Leila 335

leilae, Bahlakia 270

lenticula, Caracollina 63-77; 64, 66-69

Lepidodonotus 185

Lepsia 186

Liguus 381

Limax 399

Lindholmiola 71-73

Lindholmiolinae 73

lineata, Purpura 216

Liocerithium 300

Liostomia clavula 132

Lirobittium 261, 262-268, 272, 295,
297, 300, 308
attenuatum 262, 298
catalinense 296
subplanatum 262, 266, 296, 297-299

Litiopa 262, 271

Litiopidae 264, 271, 291

lobata, Drupa 240

Lophocardium 328

Lucinidae 320

lucorum, Helix 115

lukisii, Odostomia 132

lusitanica, Patinigera 139

415

Lutraria 333, 339
rhynchaena 333
macquarensis, Nacella 139
Macrocallista 333
Macrochlaena 352
macropus, Octopus 354
Macrotritopus 352, 355
defilippi 354
horridus 354
Mactridae 320, 325-334, 339
Maculitriton 183
Magilidae 158
magnifica, Calyptogena 149
major, Neohelix 366, 367
Mancinella 188, 213, 216, 233, 234,
235, 240, 243-245
aculeata 188, 189
alouina 161, 787, 188, 190, 246, 247
echinulata 190
gemmulata 188
mancinella 188
mancinella, Mancinella 188
mancinella, Murex 188
margariticola, Cronia 178
marinus, Perkinsus 119
massyae, Eledone 354, 358
Mastigophallus 71, 73-75
maxima, Helix aspersa 100, 114-116
Megaleledone 353
Melanoides 385
tuberculata 383
melones, Purpura 218
melones, Vasula 161, 218, 279, 240,
246, 248
melones, Vexilla 246
Menathais 213
Mengoana 71
Mercenaria 335
mercenaria 29, 335
mercenaria, Mercenaria 29, 335
Meretricinae 333
meretricula, Thais nodosa 213, 215
Mesodon 400
elevatus 363
normalis 366, 367, 399-406
thyroidus 363
zaletus 361-369
Mesodon (Akromesodon) 363
Mesodontoidea 75
Metafruticicolinae 73
metallica, Thais 210
metricula, Thais 213
Microstoma 205
Microtoma 203
Milax 382, 399
Modiolus modiolus 34, 35
modiolus, Modiolus 34, 35
Monadenia 399
Monadeniinae 74
Monobittium 273
Monoceros tuberculatum 197
monodon, Acanthina 161, 172, 229,
230251246249
Moreidae 156, 158
Moreinae 156, 158
Morula 160, 183, 190, 233, 234, 235,
237, 239, 241, 243-245

416 INDEX

granulata 190, 192
nodilifera 190
nodosa 165, 168
papillosa 190
uva 161, 166, 167. 769, 1907797,
246, 247
Morulina 183
тогит, Огира 161, 183, 184, 185, 186,
240, 246, 247
morus, Ricinula 190
moschata, Eledone 354, 355
Murex adversum 273
elegantissimum 273
fucus 213
fulvescens 171
hippocastanum 213
mancinella 188
neritoides 213
neritoideus 183, 213
reticulatum 273
reticulatus 274
ricinus 183
spenceri 273
tuberculare 273
Muricacea 156, 158
Muricanthus 231, 233, 234, 236
fulvescens 161, 164, 166, 167, 168,
169, A71,01172,222} 223) 241246;
249
muricata, Acanthina 243
muricata, Neorapana 161, 196, 197,
198, 240, 243, 246, 248
muricata, Purpura 197
Muricidae 158, 242, 155-259
Muricinae 156, 158, 161, 222, 241, 246
Muricodrupa 183
Muricoidea 156, 158
Mutelidae 320
Mya 334
arenaria 29
Mycetopodidae 320
Myidae 320, 324, 325-334
Mytilidae 320
Mytilus californianus 139
edulis 29, 31
Nacella macquarensis 139
(Patinigera) deaurata 135-140
Nassa 160, 1192, 201-233-235, 240;

241, 244, 245

francolina 193, 194, 195, 240, 248

picta 194

serta 161, 169, 193, 194, 195, 240,
246, 248

Nassa (Jopas) 220

Nassarius 192

Nassinae 220

nebulosa, Thais 210

nemoralis, Cepaea 95, 114, 371, 375,
381, 384

Neohelix albolabris 363, 366-368
alleni 363, 366, 367
major 366, 367
solemi 366, 367

Neorapana 195, 213, 233-235, 243-245
muricata 161, 196, 197, 198, 240,

243, 246, 248

tuberculata 197

Neothais 180
smithi 180
Nerita nodosa 213
neritoides, Latia 18
neritoides, Murex 213
neritoideus, Murex 183, 213
nigra, Pareledone 354
nigra, Pasithea 291
nigrum, Bittium 291
nodilifera, Morula 190
nodosa, Morula 165, 168
nodosa, Nerita 213
nodosa, Thais 161, 164, 169, 213, 214,
246, 248
nodosa, Thais nodosa 213, 215
nodus, Ricinula 190
normalis, Mesodon 366, 367, 399-406
Nucella 160, 198, 200, 229, 231, 233,
234, 240, 242-245, 385
emarginata 242
lapillus 157, 161, 166, 167. 168. 68
198, 199, 200, 201, 246, 248
theobroma 200
Nucellinae 157, 245
Nucula pernula 149
sulcata 149
taphria 149
Nuculana 141-151
grasslei 141-150; 143-148
pernula 33
Nuculidae 336
Nuculites 336
nuttali, Tresus 338
Nuttallia 336
Obrovia 389
obtusa, Chrysallida 132
Ocenebra 233
erinacea 243
Ocenebrinae 156, 158, 233, 238, 241,
242, 245
Octopodidae 343-349, 351-359
Octopodinae 344
Octopus 344, 346, 347, 349, 352, 355-
358
bimaculatus 354, 357
bimaculoides 354, 357
briareus 354
burryi 354
californicus 354
chierchiae 354
digueti 354
filosus 354
fitchi 354
hubbsorum 354
macropus 354
ornatus 354
penicilifer 354
selene 354
stitiochrus 354
vulgaris 354
(Macrochlaena) winckworthi 354
Odostomia eulimoides 131, 132
impressa 119
lukisii 132
plicata 129, 132
rissoides 132
scalaris 132

INDEX

trifida 132
unidentata 128, 132
Oestophora 71-75
Oestophorella 74
Oestophorini 63, 71, 72
Oncomelania 382, 383
Opisthoteuthis depressa 356
japonica 356
orbita, Buccinum 180
orbita, Dicathais 161, 180, 187, 183,
246, 247
orbita, Thais 180
Orbitioniidae 262
ornatus, Octopus 354
Osteophora barbula 72
otaheitana, Partula 43-61
Ozaeninae 344
Pachychilidae 266
palari, Eledone 358
Panopea 334
abrupta 338
pansa, Plicopurpura patula 205
papillosa, Morula 190
Papyridea 328, 329, 331, 332, 339
soleniformis 330
Paracerithium 270
parcum, Bittium 270, 283, 284, 287
parcum, Ittibittium 262, 266, 284, 285,
286, 287
Pareledone 345-347, 349, 351, 352,
355-357
adelieana 354
aurorae 354
carlgreni 358
charcoti 354
harrissoni 354
nigra 354
polymorpha 354
senoi 353, 355
turqueti 353-355
(Megaleledone) senoi 354
Partula 43-61, 381, 382, 384
affinis 43-61
affinis producta 52, 57
aurantia 55
jackieburchi 43-61
otaheitana 43-61
otaheitana affinis 43
otaheitana amabilis 44
otaheitana crassa 43
otaheitana rubescens 43, 51, 54
otaheitana sinistrorsa 52, 57
suturalis 55
Pasithea nigra 291
patagiatus, Scaeurgus 354
Patellapurpura 205
Patellipurpura 205, 207, 243
Patinigera caerulea 139
lusitanica 139
polaris 139
vulgata 138, 139
patula, Plicopurpura 161, 166, 167, 168,
203, 204, 205, 205, 207, 246, 248
patula, Siliqua 338
patulum, Buccinum 203
pauxilla, Varicopeza 306, 307
Pectinibranchiata 158

417

penicilifer, Octopus 354
Pentadactylus 183
Perinereis 185
Peringia 389, 391, 397
Periploma 336
Perkinsus marinus 119
pernula, Arca 141
pernula, Nucula 149
pernula, Nuculana 33
persica, Purpura 161, 207, 208, 244,
246, 248
persicum, Buccinum 207
peruviana, Concholepas 173
Petricolidae 320, 325-332
Phaxus 336, 339
Pholadacea 335
Pholas dactylus 18
Phrygiomurex 183
pica, Purpura 213
picta, Nassa 194
picta, Vexilla 220
pictum, Haustrum 216
Pinaxia 201, 213, 233-235, 240, 243-
245
coronata 201
versicolor 161, 201, 203, 246, 248
Pinnidae 320
pisana, Theba 89, 381
Pitarinae 333
Planithais 216
planospira, Purpura 216
planospira, Tribulus 161, 216, 277, 218,
240, 246, 248
Platyodon 334
Plesiotrochidae 271
Plesiotrochus 262, 271, 300
Pleurocardia 339
plicata, Odostomia 129, 132
Plicopurpura 203, 207, 233-236, 241,
241, 243-245
columellaris 205, 206
patula 161, 7166, 167, 168, 203, 204,
205, 205, 207, 246, 248
patula pansa 205
podagrinum, Bittium 287, 288
polaris, Patinigera 139
Polygyridae 400
Polygyrinae 400
polymorpha, Pareledone 354
Polytropa 198, 200, 242
Polytropalicus 176, 198, 200
pomatia, Helix 89
Ponentina 74
Potamididae 266
praeputialis, Cynthia 183
Procerithiidae 261, 262, 269, 306
Procerithiinae 262, 328
Procerithium 261, 262, 269
producta, Partula affinis 52, 57
proteum, Cerithium 270
Protothaca 333
Provexillum 220
Psammobiidae 320, 325-332, 336
Psammobiinae 334
Psammophila 333
pseudamygdala, Purpura 176
Pseudocerithium 269

418 INDEX

Pseudohelenoconcha 409
Pteroctopus 344, 346, 349, 352
tetracirrhus 354

Purpuidae 157

Purpura” 157, 176, 178, 1797 188, 198;
200, 207, 218) 233-236, 240), 243-
245
amygdala 176
angulifera 179
ascensionis 213
aurantiaca 1/76
bizonalis 200
bronni 213
buccinoidea 200
bufo 244
celtica 200
columellaris 205, 207
consul 210
crassa 218
floridana 210
foliata 207
forbesii 210
gemmulata 188
granulata 190
imbricata 200
inerma 207
lapillus 231
lineata 216
melones 218
muricata 197
persica 161, 207, 208, 244, 246, 248
pica 213
planospira 216
pseudamygdala 176
scalaris 180
sertum 194
sphaeridia 190
succincta 183
taeniata 220
textilosa 180
trinitatensis 213
truncata 197
tubifer 207

Purpuracea 157

Purpuradae 156

purpurata, Ricinella 183

Purpurella 205

Purpuridae 158, 273

Purpurinae 157

pusilla, Turritella 270

Pyrula versicolor 201, 202

quadricostata, Ecphora 161, 232, 242,
246, 249

radiata, Lampsilis 34, 35

Rapana 160, 231, 233, 234, 239, 240,
243-245
rapiformis 161, 164, 172, 222, 225,

226, 236, 246, 249

Rapanidae 158, 242

Rapanina 155

Rapaninae 155-259

rapiformis, Rapana 161, 164, 172, 222,
22592267283622467249

Rasbittium 269, 273

Reishia 213

Resania 333, 339
lanceolata 338

Resaniinae 333
reticulatum, Bittium 262, 266, 270, 271,
2183, 274, 275,277, 278, 28072808
282, 283
reticulatum, Cerithium 274
reticulatum, Murex 273
reticulatus, Murex 274
reticulatus, Strombiformis 273, 274
Rhabdocolpus 269
Rhachiglossa 158
Rhinoclavis 301
rhynchaena, Lutraria 333
Ricimula 183
Ricinella 183
purpurata 183
Ricinula 183, 183
alba 190
aspersa 190
digitata 183
horida 183
horrida 183
morus 190
nodus 190
Ricinulus 183
ricinus, Drupa 183, 185, 240
ricinus, Murex 183
rissoides, Odostomia 132
Rissoininae 270
Robsonella 344, 352
fontanianus 354
rolani, Coralliophila 200
rostrata, Arca 141
rubescens, Partula otaheitana 43, 51, 54
rubusidaeus, Drupa 183
Samoana 384
attenuata 44, 53, 55
diaphana 44, 53, 55
jackieburchi 54
Sanguinolaria 336
Sanguinolariinae 334
Sarganidae 156, 158
Sarganinae 156, 158
Saxicavidae 336
scabridum, Cerithium 270
Scaeurgus 344, 346, 349, 352
patagiatus 354
unicirrhus 354
scalaris, Odostomia 132
scalaris, Purpura 180
scalaris, Thais 180
schultzei, Aphrodoctopus 353-356, 358
Scutarcopagia 334
seetzeni, Trochoidea 116
selene, Octopus 354
Semelidae 320, 325-332
Semibittium 261, 269, 295, 297, 299,
300
subplanatum 308
senoi, Pareledone 353, 355
senoi, Pareledone (Megaleledone) 354
sericata, Halolimnohelix 74
serta, Nassa 161, 169, 193, 194, 195,
240, 246, 248
sertum, Buccinum 192, 194
sertum, Purpura 194
Siliqua 336
patula 338

INDEX

similaris, Bradybaena 380, 381
simplex, Bittium 270
sinistrorsa, Partula otaheitana 52, 57
Sistrum 183
album 183
striatum 190
situla, Buccinum 194
smithi, Neothais 180
Solecurtinae 327, 328, 334
Solecurtus 336
solemi, Neohelix 366, 367
Solen 324, 339
Solenacea 334, 336
Solenidae 320, 325-332
soleniformis, Papyridea 330
solidissima, Spisula 29, 34, 35
Soosia 71-73
spenceri, Murex 273
sphaeridia, Purpura 190
spinicirrus, Tetracheledone 354
spiralis, Chrysallida 132
Spirorbula 74
Spisula solidissima 29, 34, 35
squamosa, Thais 240
stagnorum, Ventrosia 390
stellata, Thais 210
stitiochrus, Octopus 354
Stramonita 205, 210, 213, 233-235,
239, 240, 244, 245
floridana 210
haemastoma 157, 161, 168, 210,
271, 212, 246, 248
haemastoma floridana 157
hederacea 194
striata, Dyakia 1-7, 9-19; 11-17
striatum, Sistrum 190
Strigilla 334
Strombiformis reticulatus 273, 274
Strombus vexillum 220
Stylidium 262-269, 272, 287, 292, 295,
297, 304
eschrichtii 262, 266, 292, 293, 294,
295
submamillatum, Cerithium 270
Suboestophora 74
subplanatum, Bittium 296
subplanatum, Bittium (Lirobittium) 296
subplanatum, Bittium (Semibittium) 296
subplanatum, Lirobittium 262, 266, 296,
297-299
subplanatum, Semibittium 308
succincta, Purpura 183
succincta, Thais 180
succinctum, Buccinum [non-binomial]
sulcata, Cuma 178, 179
sulcata, Nucula 149
Sundabittium 270
suturalis, Partula 55
taeniata, Purpura 220
Tagelus 334, 336
divisus 338
Tapetinae 333
taphria, Nucula 149
Tasmalira 305
Taurasia 194
tecta, Cymia 161, 172, 177, 178, 179,
240, 246, 247

180

419

tectum, Buccinum 179
Tellinacea 334
Tellinidae 320, 324-334, 339
Tenguella 190
Teretoctopus 344-346, 349
Tetracheledone 345, 346, 349, 352
spinicirrus 354
tetracirrhus, Pteroctopus 354
texta, Helix 115, 116
textilosa, Purpura 180
Thaida 157
Thaidae 156
Thaididae 155-158, 161, 229, 231, 242,
246
Thaidiidae 156
Thaidinae 155, 156, 158, 161, 229,
ZSilpe2Z S443 24 5824-6
Thais 176, 200, 213, 233-235, 239,
240, 241, 243-245
aculeata 213
aegrota 180
bronni 213
clavigera 178
deltoidea 244
dubia 240
floridana haysae 210
grisea 210
lapillus 231
metallica 210
metricula 213
nebulosa 210
nodosa 161, 164, 169, 213, 274,
246, 248
nodosa meretricula 213, 215
nodosa nodosa 213, 215
orbita 180
scalaris 180
Squamosa 240
stellata 210
succincta 180
trinitatensis 213
tuberosa 213
vector 180
wahlbergi 240
(Stramonita) hidalgoi 210
(Stramonita) langi 210
(Thais) 201
Thaisella 213
Thaisidae 156, 157
Thaisidinae 156
Thalessa 213
Thaumelodone 345, 349
Theba pisana 89, 381
theobroma, Nucella 200
thermophilus, Bathymodiolus 35, 149
Thiaridae 266
thraciaeformis, Yoldia 29, 30, 33, 35
Thyphinae 156
thyroidus, Mesodon 363
thysanophora, Eledone 358
Tindaria callistiformis 34
Trachycardiinae 328, 329
Tresus 333
nuttali 338
Triaxeopus 357

420

Tribulus 213, 216, 233-235, 244, 245
planospira 161, 216, 217, 218, 240,
246, 248
Trichiinae 73, 75
trifida, Odostomia 132
Trigoniacea 320
trinitatensis, Purpura 213
trinitatensis, Thais 213
Triodopsinae 400
Triodopsis 383, 399, 400
albolabris 399-406
Triphora 273, 274
Trissexodon 71, 72, 74
Trissexodontidae 74, 75
Trissexodontini 63, 71, 72
Trochia 213, 233, 234, 242, 244, 245
cmeulata 229.161. 172, 229: 237,
231, 242, 246, 249
Trochoidea seetzeni 116
Trophon 160
Trophoninae 158
truncata, Purpura 197
tuberculare, Murex 273
tuberculata, Drupa 157
tuberculata, Melanoides 383
tuberculata, Neorapana 197
tuberculatum, Monoceros 197
tuberosa, Thais 213
tubifer, Purpura 207
Turbinellidae 156
turqueti, Pareledone 353-355
Turridae 160
turriplana, Gittenbergeria 72
turriplana, Helix 71
Turritella alternata 291
eschrichtii 292
pusilla 270
Typhinae 158, 207
Typhis 207
ulvae, Hydrobia 389-398; 390
unicirrhus, Scaeurgus 354
unidentata, Odostomia 128, 132
Unionidae 320
Unionoida 325-332, 334, 335, 338
Urosalpinx 160, 242
cinerea 230, 249
Usilla 183
uva, Drupa 190
uva, Morula 161, 166, 167, 169, 190,
191, 246, 247
varicopeza, Varicopeza 266, 305
Varicopeza 262-265, 268, 269, 271,
305
crystallina 268
pauxilla 306, 307
varicopeza 266, 305
varium, Bittiolum 262, 266, 282, 287,
288,209,250, 20
varium, Cerithium 288
varium, Diastoma 288
Vascula 218
Vasidae 158
Vasula 218, 233-235, 244, 245
melones 161, 218, 279, 240, 246,
248
vector, Thais 180
Velodona 345, 346, 349

INDEX

Veneracea 333
Veneridae 320, 325-333, 339
ventrosa, Hydrobia 389-398; 390
Ventrosia 389, 390, 397
stagnorum 390
versicolor, Pinaxia 161, 201, 203, 246,
248
versicolor, Pyrula 201, 202
vexilla, Vexilla 161
Vexilla 194, 220, 233-235, 236, 240,
241, 244, 245
melones 246
picta 220
vexilla 161
vexillum 164, 220, 227, 246, 249
vexillum, Strombus 220
vexillum, Vexilla 164, 220, 221, 246,
249
Vicariihelicinae 74
Vicariihelix kivuensis 74
virgata, Cernuella 89
virginica, Crassostrea 119
Vitreledonella 344
vitreum, Bittium 306
Volema 188, 189
glacialis 188
Vosselodone 345, 346, 349, 352
charrua 354
vulgaris, Octopus 354
vulgata, Patinigera 138, 139
wahlbergi, Thais 240
washingtonius, [das 30
willistoni, Drosophila 383
winckworthi, Octopus (Macrochlaena)
354
Xanthonychidae 71, 73, 74
Xanthonychoidea 74
Xystrella 269
Yoldia thraciaeformis 29, 30, 33, 35
zaletus, Mesodon 361-369
Zeacumantus 270
zealandicum, Haustrum 186
Zebittium 262, 263, 268, 306
exile 306, 308
zebrum, Bittium 262
zebrum, Cerithium 274
Zenatia 333, 339

MALACOLOGIA, VOL. 35

CONTENTS

J. A. ALLEN

A New Deep-Water Hydrothermal Species of Nuculana (Bivalvia: Protobran-

chia)strom the: Guaymas: Basin samen Ka te Re во
TAKAHIRO ASAMI

Divergence of Activity Patterns in Coexisting Species of Land Snails .....
ANETTE BAUR & BRUNO BAUR

Daily Movement Patterns and Dispersal in the Land Snail

AMENA DUO et baad AGAR be Botan nas Ort as Renn Serene cc
PHILIPPE BOUCHET AND JEAN-PIERRE ROCROI

The Lottery of Bibliographical Databases: А Reply to Edwards

Scones ete sete elek aed. ton de O ÍA a
JONATHAN COPELAND & MARYELLEN MANERI DASTON

Adult and Juvenile Flashes in the Terrestrial Snail Dyakia striata ..........
HARLAN K. DEAN

A Population Study of the Bivalve /das argenteus Jeffreys, 1876, (Bivalvia:

Mytilidae) Recovered from a Submerged Wood Block in the Deep North

PUL AMUCIOCS ANN RE ee
M. A. EDWARDS & M. J. THORNE

LR ое Edi хо ne Re ee
КЕММЕТН С. EMBERTON

Over-Representation of Rare Alleles in Juveniles and Lack of Pattern in

Geographic Distributions of Alleles in a Land Snail ........................
ANDRZEJ FALNIOWSKI, ANDRZEJ KOZIK, MAGDALENA SZAROWSKA,
MARIA RAPALA-KOZIK, & IZABELA TURYNA

Morphological and Allozymic Polymorphism and Differences Among Local

Populations in Bradybaena fruticum (О. Е. Müller, 1777) (Gastropoda:

sStylommalophora-Hellcoidea)) - 1. ee ne een
MARTIN HAASE

The Genetic Differentiation in Three Species of the Genus Hydrobia and

Systematic Implications (Caenogastropoda, Hydrobiidae) .................
ALOIS HONEK

Melanism in the Land Snail Helicella candicans (Gastropoda, Helicidae) and

t5iRossible Adaptive Signilicancen 2... a coo ee es
RICHARD S. HOUBRICK

Phylogenetic Relationships and Generic Review of the Bittinae (Prosobran-

chia4@erithioidea).-. oe y ee en.
MICHAEL S. JOHNSON, JAMES MURRAY & BRYAN CLARKE

Evolutionary Relationships and Extreme Genital Variation in a Closely

RelatediiGroupion Рама are ee ee ee ee
SILVARD P. KOOL

Phylogenetic Analysis of the Rapaninae (Neogastropoda: Muricidae) .....
LUC MADEC & JACQUES DAGUZAN

Geographic Variation in Reproductive Traits of Helix aspersa Müller Studied

undernlaboraton Conditions ee ern
MARYELLEN MANERI DASTON & JONATHAN COPELAND

The Luminescent Organ and Sexual Maturity in Dyakia striata ............

141

399

407

361

371

389

261

ELBA MORRICONI У JORGE CALVO
Influencia Ambiental Sobre el Crecimento Alométrico de la Valva en Nacella
(Patinigera) deaurata (Gmelin, 1791) del Canal Beagle, Argentina ........
CARLOS Е. PRIETO, ANA 1. PUENTE, KEPA ALTONAGA & BENJAMIN J. GOMEZ
Genital Morphology of Caracollina lenticula (Michaud, 1831), with a
New Proposal of Classification of Helicodontoid Genera (Pulmonata:
HYgromibidea) =... ое ee ee ee ВОИ
JANET R. VOIGHT
A Cladistic Reassessment of Octopodid Classification ....................

JANET R. VOIGHT
The Arrangement of Suckers on Octopodid Arms as a Continuous
Character. а. вое Е
С. THOMAS WATTERS
Some Aspects of the Functional Morphology of the Shell of Infaunal Bivalves
(MolluSca) 0: 22.2: er ааа OR
JOHN В. WISE
Anatomy and Functional Morphology of the Feeding Structures of the Ecto-
parasitic Gastropod Boonea impressa (Pyramidellidae) ...................

135

343

351

315

WHY NOT SUBSCRIBE ТО 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.

AWARDS FOR STUDY АТ
The Academy of Natural Sciences of Philadelphia

The Academy of Natural Sciences of Philadelphia, through its Jessup and
McHenry funds, makes available each year a limited number of awards to support
Students pursuing natural history studies at the Academy. These awards are pri-
marily intended to assist predoctoral and immediate postdoctoral students. Awards
usually include a stipend to help defray living expenses, and support for travel to and
from the Academy. Application deadlines are 1 March and 1 October each year.
Further information may be obtained by writing to: Chairman, Jessup-McHenry
Award Committee, Academy of Natural Sciences of Philadelphia, 1900 Benjamin
Franklin Parkway, Philadelphia, Pennsylvania 19103-1195, U.S.A.

MALACOLOGIA, 1993, 35(2)

35(2)

INSTRUCTIONS FOR AUTHORS

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.

4. Unless indicated otherwise below, соп-
tributors should follow the recommendations
in the Council of Biology Editors (СВЕ) 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.

7. Choose a running title (a shortened
version of the main title) of fewer than 50
letters and spaces.

MALACOLOGIA

1993

8. Provide a concise and informative Ab-
stract summarizing not only contents but re-
sults. А separate summary generally is super-
fluous.

9. Supply between five and eight key
(topic) words to go at the end of the Abstract.

10. Use the metric system throughout. Mi-
cron should be abbreviated рт.

11. Illustrations are printed either in one
column or the full width of a page of the
journal, so plan accordingly. The maximum
size of a printed figure is 13.5 x 20.0cm
(preferably not as tall as this so that the cap-
tion does not have to be on the opposite
page).

12. Drawings and lettering must be dark
black on white, blue tracing, or blue-lined
paper. Lines, stippling, letters and numbers
should be thick enough to allow reduction by
Y. or Уз. Letters and numbers should be at
least 3mm high after reduction. Several
drawings or photographs may be grouped
together to fit a page. Photographs are to be
high contrast. High contrast is especially im-
portant for histological photographs.

14. Scale lines are required for all nondi-
agrammatic figures, and should be conve-
nient lengths (e.g., “200 um,” not “163 um”).
Magnifications in captions are not accept-
able.

15. All illustrations should be mounted,
numbered, labeled or lettered, i.e. ready for
the printer.

—

mene Ne

Зет

> зу ia. case al Ba re а:
{ - 4 é \ À
DFE y > y
|

RS um =
4

17. Tables are to be used panal, 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 т 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 SRE MUR

Office.
5307 075

VOL. 35, NO. 2 2 "+ MALAGOLOGIA (<< 407 O
| CONTENTS UN

SILVARD P. KOOL
| Phylogenetic HAN of the Rapaninae (Neogastropoda: Muricidae) .

RICHARD S. HOUBRICK y de
Phylogenetic Relationships and Generic Review of the Bittinae (Prosobran- ng
chia: Cerithioidea) ......... Prt EEE Me A + 261
G. THOMAS WATTERS | DS ic Pr
Some Aspects of the Functional | Morphology of the Shell of Infaunal Bivalves - + a г.
(Mollusca) ............................................ pts ak tte brad tee Soy:
JANET R. VOIGHT e KU
A Cladistic Reassessment of 'Octopodid Classification. sn RE ve où Se AS
JANET R. VOIGHT f ae
The Arrangement of Suckers on RE D Arms ‚as a Continuous “oi
Character ico ihe ee A O tar. nn: ¿518
KENNETH C. EMBERTON AT.
Over-Representation of Rare Alleles in Juveniles and Lack of Pattern in
Geographic Distributions of Alleles in a Land Snail .............. Kuren а
ANDRZEJ FALNIOWSKI, ANDRZEJ KOZIK, MAGDALENA SZAROWSKA, = 8
MARIA RAPALA-KOZIK, & IZABELA TURYNA |
Morphological and Allozymic Polymorphism and Differences Among Local _
‘Populations in Bradybaena fruticum (9: F. Müller, 1777) (Gastropoda: |

Stylommatophora: Helicoidea) ........... A D DE ee PUR ni 9
MARTIN HAASE | м.
_ The Genetic Differentiation in Three Species of the Genie: aa jand- >
Systematic Implications RG TOR Hydrobiidae) AS ae bale E 3
TAKAHIRO ASAMI BRAK
Divergence of Activity Patte. in Coexisting Species of Land Snails . 399
PHILIPPE BOUCHET AND JEAN-PIERRE ROCROI 3
| Тре Lottery: of Bibliographical Databases: A Reply to. Edwards r
S\THOINB A DA AE AUS ESS ARE A Mate Jane JA

НН
ЕН i PT
PNEUS GE A AA DRASS A EE GE AT
EE an

“pt PERIANA II IIA ие

VEIA wm,
LI dran NAHE UE TAN

DATE TETE
bn M MR NY ARE ER

PTT

ns
wre

я un
aan мии с arten
TES чи
UPA RES DE à AS
PTE
VE LME un

m.
ran
rey
CONTENT

YAM Uf ый ER Ar ar Yan LH wi

CET ih oe

Ия
A UBER ар AAA

ЛЯ
A
CLONE .
EN
(ey , sop
CHIENS ws
ee ed amd EEE
Ce

wer

RO II RU вр KU зб

WWF KA ay a И

И 4

DUT

ить
an

Parker
we Ve Aa a AO pan eu

Aia tdo
OA

PI
A tan à PA

BL EEE RTL A
AU pgs A ye away НР
COPA ENTER тов чья
DEAN ns

CARTER PEN rit
м

WPL CS TRE
OMAN рр учла

QE VE тем RS NE LD RTS

A A EN om
даче: A

CR лети
rs

NETT 8
o

A en An re AL
LEE MANES ay PE MA ge es AC ou Ry
$“
ETES
RTL a en
PANERA Tar AES Ry WER RET

ри Aa URE MER VA A
A ata ag y

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "Malacologia"

See other formats