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 W А iA) Ni | m0 р Ñ UY р ÿ у 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 al 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 Co-Editors: м Ag et | a 4 EUGENE COAN REN TR AER ACA aS CAROL ones -California Academy of ne N | | Denver, ER: 3 > San Francisco, СА ме y | Assistant Managing Editor: 3 | __ CARYL НЕЗТЕВМАМ - | | | 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 Copyright O 1993 by the Institute of Malacology 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 10 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 5 $ 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 fl 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- 11 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. de LUMINESCENT ОАСАМ AND SEXUAL MATURITY 1174 Pre-buccal Canal Floor of Pre-buccal canal Foot Muscle CE 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 21 “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 © © -J O O1 R CO) D — № > о 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 50 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 © © OU E UN — UY O0 Bees AE NA 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 4 © © —J O O1 HWP 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 20 т 15 10 Distance along transect (mm) 0 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 D a, A (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 op 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 58 ‘SajqeueA JO иоцеие|аха 10} }ха} aas ‘dnouf вивиэцео 4 ayy ш Abojoydıow ¡eyuab jo sısAjeue 10) взеа “KIGNAddV 59 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 sg 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 60 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 95 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 v 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 61 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 8 761 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 63 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 79 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 12 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 82 LEl O gu Sel bel SOU 0.99% ae 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 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 89 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 == AZ 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). 95 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 99 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 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 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 20 0 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. me 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 (b ) — 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 an 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 i ‚ 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 Copyright O 1993 by the Institute of Malacology 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 od 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) (c) both glands short to medium (less than one-quarter of shell height; Fig. 3F, la, ra) (d) both glands absent (e) right gland very long (nearly one-half of shell height); left gland absent 16. Length of posterior blind duct of gland of Leiblein (a) duct at least one-half of length of gland (Fig. 3F, dgL) (b) duct shorter than one-half (usually less than one-fourth) of length of gland Radula: Radulae (2-6 per species) were dis- sected from living and preserved animals, cleaned in potassium hydroxide, and exam- ined using scanning electron microscopy. For the sake of consistency, only scanning elec- tron micrographs were used for analyzing radular structures. Four micrographs were taken of the central portion of each radular ribbon. The first two micrographs (one includ- ing lateral teeth, one excluding lateral teeth) were taken perpendicular to the radular rib- bon. The radula was then tilted laterally to an angle of 40° to obtain a lateral view of the morphology of the cusps and denticles on the rachidian tooth. Finally, the radula was tilted laterally to an angle of about 85° to examine the edge of the rachidian tooth and the an- gles, sizes and locations of its cusps and den- ticles, in an area from which the lateral teeth had been cut away with a surgical knife. The morphology of the radula is described starting with the rachidian tooth (Fig. 6B), fol- lowed by the lateral teeth. The cusps (three or five) on the rachidian are described beginning with the central cusp (Fig. 6B, cc), followed by the inner lateral denticle (ild), lateral cusp (Ic), the marginal area (ma), marginal denticles (d), and marginal cusp (mc). The marginal area is defined as the more or less horizontal area on the outside of the lateral cusp, ex- tending to—if present—the marginal cusp. Size of lateral cusps is given relative to size of central cusp (“nearly equal” translates into 75% or more of central cusp length). The po- sition of the inner denticle(s) is against the base of the inner edge of the lateral cusp, unless noted otherwise. Size of inner lateral denticle is relative to lateral cusp. Size of lat- eral teeth is given relative to rachidian width. An approximate range of the length of the rad- ular ribbon is given, where available, relative to shell height. Characters: 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 А В 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. Shell Ultrastructure: Aragonitic layer with crystal planes oriented perpendicular to grow- ing edge (25-30%); aragonitic layer with crystal planes oriented parallel to growing edge (70-75%) (Fig. 8C). Operculum: D-shaped, with S-shaped left edge, tapered at lower end, with lateral nu- cleus in lower right (compare Fig. 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). oe 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 \ 4 in 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. Anteriormost portion of ctenidium straight, equidistant from mantle edge with osphra- dium. Anterior ctenidial lamellae wider than deep; posterior lamellae deeper than wide. Lateral and ventral ctenidial lamellae con- cave. Vaginal opening enlarged, protruding from short, tubular extension of 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 RL de de 1> In 303 SR Fehr Е 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- 9 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). (5 — 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) 5 *1 Specimens of the type species of Neora- pana were typical “Neorapana tuberculata” (Sowerby, 1835) morphs; it appears that N. tuberculata and N. muricata are synonyms. Neorapana muricata (Broderip, 1832) is the senior synonym of Neorapana tuberculata (Sowerby, 1835) (see “Remarks” under 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 20 267 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 N 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 AA 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 5 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 50 40 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 1 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 С 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 ae 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 + OO TOO 202 OO a a 222 20000022000 |m NN==000000000 VN=0=00=000000 o0o00000-00-000|» RE OO 4) (©) 45) ©) | Ce) O O) NOOO OOO (O OJO SA OS O O (S) ADO Ke) Co) = NNNN==0==00000|m-= o000--00000000|w- 2=N0W=NWNOOVO=O0O|ah= 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 — 20 80 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 TO 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 7 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. 23 5 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 Al 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 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 AK AO APH CK 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 (© 0.143 0 0 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 a 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! > œ > Proportion of daily activity (%) о "16 18 20 22 0 Proportion of daily activity (%) 16 18 20 220 2 4 6 8 10 12 14 Time ) м 5 e М = 45 > = 2 12, о oO > ss 5 [= © 4 A 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 B 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., 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 of London, 41: 415—428. 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, (c) little-known and obscure 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. И 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 89 407 361 371 389 79 261 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 63 343 351 315 13 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 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 from CBE, 9650 Rockville Pike, Bethesda, MD 20814, U.S.A. 5. Be brief. 6. Manuscripts must be typed on one side of good quality white paper, double-spaced throughout (including the references, tables and figure captions), and with ample margins. Tables and figure captions should be typed on separate pages and put at the end of the manuscript. Make the hierarchy of headings within the text simple and consistent. Avoid internal page references (which have to be added in page proof). 7. Choose a running title (a shortened version of the main title) of fewer than 50 letters and spaces. MALACOLOGIA 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. 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. lt — mene Ne Зет > зу ia. case al Ba re а: { - 4 é \ À DFE y > y | g 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 О 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 A 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 Dr CE 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 QE ae ить an Parker we Ve Aa a AO pan eu NT 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