Al al A512 AMERICAN MOL.L- MALACOLOGICAL BULLETIN ~ Journal of the American Malacological Society http://erato.acnatsci.org/ams/publications/amb.html VOLUME 22 26 MARCH 2007 NUMBER 1/2 Reproductive performance of Helix pomatia (Gastropoda: Pulmonata: Helicidae) and survival of its hatchlings under farm conditions. MACIEJ LIGASZEWSKI, ANDRZEJ LYSAK, and aes MACH-PATOSZRIEWICZ ... 2... cece eee ce ete eee eee seen eeneeee 1 Determinate growth and variable size at maturity in the marine gastropod Amphissa columbiana. EET cece ce cee eee eee eee ence eee enee Oates Smear 7 Land snail diversity in subtropical rainforest mountains (Yungas) of Tucuman, northwestern Argentina. EUGENIA SALAS ORONO, MARIA GABRIELA CUEZZO, and FATIMA ROMERO Sere eee cote sees aise eP onsale ees Fhe fetenis (ec tus ueMene ters vaNeia te Brake ents fe aaa ened eT aoe ea aes 17 Out of Australia: Belloliva (Neogastropoda: Olividae) in the Coral Sea and New Caledonia. VUREIESFANTOR-and PHILIPPE BOUCHED i.i4 ce ieee ees et pea Pee ale ese pete oe 27 Epibionts on Flexopecten felipponei (Dall, 1922), an uncommon scallop from Argentina. LAURA SCHEJTER and CLAUDIA S. BREMEC ............. 00.0 c eee eee eee eS Partulids on Tahiti: Differential persistence of a minority of endemic taxa among fenchpopplations: @REVOR COOLE. 61.4443 ene eee? pene een es ee een ee os 83 Phyllidiidae (Opisthobranchia: Nudibranchia) from Papua New Guinea with the description of a new species of Phyllidiella. MARTA DOMINGUEZ, PATRICIA QUINTAS, and JESUS San ON GO SO certs rete era tenets eins An ahs wane ee ee Meee ees 89 continued on back cover Cover photo: Radula of Belloliva simplex from Kantor & Bouchet AMERICAN MALACOLOGICAL BULLETIN BOARD OF EDITORS Janice Voltzow, Editor-in Chief Department of Biology University of Scranton Scranton, Pennsylvania 18510-4625 USA Robert H. Cowie Center for Conservation Research and Training University of Hawaii 3050 Maile Way, Gilmore 408 Honolulu, Hawaii 96822-2231 USA Carole S. Hickman University of California Berkeley Department of Integrative Biology 3060 VLSB #3140 Berkeley, California 94720 USA Timothy A. Pearce Carnegie Museum of Natural History 4400 Forbes Avenue Pittsburgh, Pennsylvania 15213-4007 USA Angel Valdés, Managing Editor Natural History Museum of Los Angeles County 900 Exposition Boulevard Los Angeles, California 90007-4057 USA Alan J. Kohn Department of Zoology Box 351800 University of Washington Seattle, Washington 98195 USA Dianna Padilla Department of Ecology and Evolution Stony Brook University Stony Brook, New York 11749-5245 USA Roland C. Anderson The Seattle Aquarium 1483 Alaskan Way Seattle, Washington 98101 USA The American Malacological Bulletin is the scientific journal of the American Malacological Society, an international society of professional, student, and amateur malacologists. Complete information about the Society and its publications can be found on the Society’s website: http:// www. malacological.org AMERICAN MALACOLOGICAL SOCIETY MEMBERSHIP MEMBERSHIP INFORMATION: Individuals are invited to com- plete the membership application available at the end of this issue. SUBSCRIPTION INFORMATION: Institutional subscriptions are available at a cost of $65 plus postage for addresses outside the USA. Further information on dues, postage fees (for members outside the U.S.) and payment options can be found on the Membership Application at the end of this issue. ALL MEMBERSHIP APPLICATIONS, SUBSCRIPTION ORDERS, AND PAYMENTS should be sent to the Society Treasurer: Susan B. Cook 4201 Wilson Blvd. STE 110-455 Arlington, Virginia 22203 USA E-mail: scook@coreocean.org CHANGE OF ADDRESS INFORMATION should be sent to the Society Secretary: Paul Callomon Department of Malacology The Academy of Natural Sciences of Philadelphia 1900 Benjamin Franklin Parkway Philadelphia, Pennsylvania 19103-1195 USA INFORMATION FOR CONTRIBUTIONS is available on-line and appears at the end of this issue. MANUSCRIPT SUBMISSION, CLAIMS, AND PERMISSIONS TO REPRINT JOURNAL MATERIAL should be sent to the Editor-in-Chief: Janice Voltzow, Editor-in-Chief Department of Biology University of Scranton Scranton, Pennsylvania 18510-4625 USA Voice: 570-941-4378 * Fax: 570-941-7572 E-mail: voltzowj2@scranton.edu AMERICAN MALACOLOGICAL BULLETIN 22(1/2) AMER. MALAC. BULL. ISSN 0740-2783 Copyright © 2007 by the American Malacological Society AMERICAN MALACOLOGICAL BULLETIN CONTENTS VOLUME 22 | NUMBER 1/2 Reproductive performance of Helix pomatia (Gastropoda: Pulmonata: Helicidae) and survival of its hatchlings under farm conditions. MACIEJ LIGASZEWSKI, ANDRZEJ LYSAK, and ZOFIAMAGH=PALUSZKIEWIGZ: ecient cede wb sig din eine ee arenes ghy eon Den ee Saeed a eae ] Determinate growth and variable size at maturity in the marine gastropod Amphissa columbiana. BRUNO BERNE Lge aa ohh ee eee ee ite ee Ris Sar ae wy wae e cetera ae be « ve, Ea f, Land snail diversity in subtropical rainforest mountains (Yungas) of Tucuman, northwestern Argentina. EUGENIA SALAS ORONO, MARIA GABRIELA CUEZZO, and EAMES RO NEE RON siotaci tein esate Sir is dat, a outa nahi Sernc tiets be ead want nated Ree R 17 Out of Australia: Belloliva (Neogastropoda: Olividae) in the Coral Sea and New Caledonia. YURI I. KANTOR and PHILIPPE BOUCHET ................00 0... ccc eee 27 Epibionts on Flexopecten felipponei (Dall, 1922), an uncommon scallop from Argentina. LAURA SCHEJTER and CLAUDIA S. BREMEC ........... 00.0. cece cee eee 75 Partulids on Tahiti: Differential persistence of a minority of endemic taxa among relict populations. TREVOR COOTE 94.0345 sacias sasdeiee de Suu 8 6 he kaw eo oeew ed 4s eke 83 Phyllidiidae (Opisthobranchia: Nudibranchia) from Papua New Guinea with the description of a new species of Phyllidiella. MARTA DOMINGUEZ, PATRICIA QUINTAS, and JESUS SHIRONCOSO cpus tatu netics alo, Ge el tent ees nest coment ow ive. 89 Five new species of aeolid nudibranchs (Mollusca, Opisthobranchia) from the tropical eastern Pacific. ALICIA HERMOSILLO and ANGEL VALDES ...............00000200 00 0ee 119 The concentration of calcium carbonate in shells of freshwater snails. MEREDITH M. WHITE, MICHAEL CHEJLAVA, BERNARD FRIED, and JOSEPH SHERMA ...................... 139 Larval settlement and recruitment of a brackish water clam, Corbicula japonica, in the Kiso estuaries, central Japan. RYOGEN NANBU, ESTUKO YOKOYAMA, TOMOMI MIZUNO, ang. HEDEO SEKIGUCHI Nici iin TES Sah) aa nleis ot Sashes Fe eee Dies Somaed « oe ree bee ene 143 Investigation in the laboratory of mucous trail detection in the terrestrial pulmonate snail Mesodon thyroidus (Say, 1817) (Mollusca: Gastropoda: Polygyridae). ELIZA BE PEGs DAVIS 4.55 - te S eed cet ns ase MRE Se ES aes Soe wee ea satay eeeeees 157 RecedrCneNOlem naa ae Ne Wade gah Ca nian hae Roun Ruane Reba aed ena ease eee 165 BOOKBRE VIC Wisin aed. Neeeee Oe eam REC Ene ee OMEN oe aM OSM eg aks Yo daet aa nae ea agente ke itn dB FRA Ota e 169 IN MEMORIAM Wayne Grimm Robert M. Linsley Paul W. Parmalee Ellis L. Yochelson Twila Bratcher-Critchlow A MESSAGE FROM THE EDITOR Dear readers, With the publication of this volume, my term as editor-in-chief of the American Malacological Bulletin comes to an end. I would like to thank all of the authors and reviewers who have contributed their work. I am grateful for their patience and cooperation. I feel especially fortunate to have worked with Angel Valdés of the Natural History Museum of Los Angeles County, who has served as managing editor for the past five years. He is responsible for the new cover and design of the Bulletin and has performed the painstaking task of getting the issues into shape for publication. It has been a complete joy to work with him; I feel fortunate to have him as a colleague. Ken Brown has been elected as the new editor-in-chief. I am delighted that he has agreed to assume this position and am certain he will do an outstanding job. Contact information for Dr. Brown appears below. We are working to make the transition as smooth as possible. I am also very happy that Cynthia Trowbridge has agreed to assume the responsibility of managing editor. I wish them both the very best! Janice Voltzow Editor-in-chief As of January 2007 all enquiries should be sent to: Dr. Kenneth M. Brown Professor Department of Biological Sciences Louisiana State University Baton Rouge, LA 70803 kmbrown@Isu.edu 225-578-1740 ill THE MOLLUSKS: & A GUIDE TO THEIR STUDY, COLLECTION, AND PRESERVATION Edited by C. F. STURM, T. A. PEARCE, and A. VALDES PUBLICATION OF THE AMERICAN MALACOLOGICAL SOCIETY The Mollusks: A Guide to Their Study, Collection, and Preservation C. Sturm, T. A. Pearce, and A. Valdés, editors A new publication from the American Malacological Society Have you ever wondered about collecting snails with a leaf blower? How about the ins and outs of preserving a giant squid? Do you know what a bail, grab, or box corer are? Maybe you have pondered what types of plastics are safe to use for storing specimens or how to use an optical scanner to image shells. If questions like these arise from time to time, you want a copy of the American Malacological Soci- ety’s latest publication, The Mollusks: A Guide to Their Study, Collection, and Preservation. The American Malacological Society, founded in 1931 as the American Malacological Union, is an organization that brings together folks interested in mollusks. In 1942, papers presented at the annual meeting in Maine dealt with studying and collecting shells. These papers were published in the Annual Report of 1942 and were reprinted in 1955, 1966, and 1974. With each reprinting, a few more papers from other publications were added. The 1974 booklet, en- titled How to Study and Collect Shells, was 107 pages in length and had two illustrations. Now, The Mollusks: A Guide to Their Study, Collection, and Preservations is the first update of the 1974 booklet in 32 years. If you are looking for a book full of glossy photos, this book is not for you. If you want a book giving the latest information on all modern classes of mollusks and the best methods to study, collect, and preserve them, look no further. The Mollusks, 445 pages long, with 31 chapters, 101 figures, and 28 tables, is a completely new book. The book was edited by Charlie Sturm, Tim Pearce, and Angel Valdés. An international team of 29 individuals contributed to these chapters. The Mollusks differs in several significant ways from its predecessors. While the former books were compendia of articles, The Mollusks consists of chapters, each covering a specific topic. Some chapters deal with collecting and preserving mollusks, both the shells and soft parts, remote bottom collecting, and SCUBA diving. Other chapters cover archival practices, writ- ing taxonomic papers, the International Code of Zoological Nomenclature, constructing databases, digital imaging, and film photography. Chapter 9 lists over 750 books, mono- graphs, and papers on mollusks indexed by biogeographic region and taxonomic group. If you collected land snails in southern Africa, go to the “Ethiopian (Afrotropic)- terrestrial” listing and you will find a list of 20 books to help you with your material. All modern classes of mollusks are treated in The Mol- lusks. The Aplacophora, Monoplacophora, Polyplacophora, Scaphopoda, and Cephalopoda have their own chapters. The Bivalvia are covered in three chapters while the Gastropoda are covered in four chapters. There is even a chapter on fossil mollusks. These chapters cover the biology and ecology of these groups, where to find these organisms, and how to collect them. Each chapter has a list of cited references for further information. The last four chapters of the book cover a variety of topics. Two chapters deal with conservation, one with freshwater mollusks, and the other with marine mollusks. One chapter discusses maintaining a marine aquarium. The fourth chapter is on non-molluscan marine organisms that have calcareous structures and might be mistaken for mollusks. The Mollusks is a soft covered book retailing for $35.95. For more information and where to order it go to (http:// www.malacological.org/publications/molluskguide.html). At this site you will find a link to the publisher’s website, here you can read the first chapter of The Mollusks. This chapter is a detailed introduction to the rest of the book. Questions about the book can be sent to the editors at doc.fossil@ gmail.com. Amer. Malac. Bull. 22: 1-6 Reproductive performance of Helix pomatia (Gastropoda: Pulmonata: Helicidae) and survival of its hatchlings under farm conditions Maciej Ligaszewski, Andrzej Lysak, and Zofia Mach-Paluszkiewicz National Research Institute of Animal Production, Department of Technology, Ecology and Economics of Animal Production, Sarego 2 Street, 31-047 Krakow, Poland, mligasze@izoo.krakow.pl Abstract: The reproductive ability of 1254 breeding individuals of Helix pomatia originating from a local wild population were studied. Reproduction was carried out in a greenhouse at a stocking density of 51.2 breeding snails per m°. The reproductive season was 83 days long. From 30 May to 21 August 2003 almost all the snails laid eggs at least one time, but 25.1% of the snails laid eggs twice, and 5.2% laid eggs three times. The mean number of eggs per clutch for all the breeding snails was 41.7. Because of the multiple laying of eggs, the number of eggs laid in the 2003 season averaged 61.5 per breeding snail, and the eggs’ total biomass constituted 38.5% of the biomass of all breeding individuals. A significant (P < 0.05) increase in egg laying was found from 30 May to 21 July. This period was followed by a rapid decline in reproductive intensity, until egg laying ceased on 21 August. Breeding snails laid 77100 eggs, out of which 40000 eggs hatched. From the hatched eggs 27000 two-to-three-week-old hatchlings were obtained. Of the obtained hatchlings 15000 were released into a greenhouse; 32.0% of these individuals survived winter hibernation in a pen. In May of the next year, the 4800 hatchlings were released into a field pen, at a density of 260 specimens per m°. They reached a mean body mass of 19.1 g and mean shell diameter of 30.1 mm in July. The rapid rate of growth observed under farm conditions allows us to propose a two-year farming cycle for this species, from hatching to the stage of sexual maturity. Key words: snail breeding, snail growing, snail protection, snail farming In Poland there are still abundant live natural popu- lations of the Roman snail (Helix pomatia Linneus 1758) (Stepczak 1976, Dyduch-Falniowska et al. 2001a, 2001b). As it has already happened with Helix aspersa aspersa (Miller 1774) and Helix aspersa maxima (Taylor 1883), it is likely that specimen numbers in these populations will be in danger of being reduced because of increasing exploitation by exporters for snail meat to other European markets (Lysak 1999). However, over the last 20 years, develop- ment of intensive farm-rearing of Helix aspersa has been developed, made possible by the snail’s high fertility and the recognition of physiological factors and farm- ing technology (Lucarz 1984, Daguzan 1989, Gomot and Gomot 1989, Gomot et al. 1989, Gomot and Deray 1990, Lazaridou-Dimitriadou and Bailey 1991). Due to its slower rate of growth, lower fertility, and difficulty with early spring hatching, Helix pomatia is regarded as a difficult species to breed compared to Helix aspersa. The develop- ment of technology for farm-rearing of this species has therefore become not only an economical concern but also an essential matter for environmental protection. Studies on the physiology of the reproductive biology of this species (JJeppensen 1976, Lysak et al. 2002) provide the basic back- ground for its intensive cultivation. The aim of the present study was to develop the technology to farm the Roman snail, based on its life cycle and to enhance its rate of reproduction. MATERIAL AND METHODS On 25 April 2003, a group of 1254 adult Roman snails, Helix pomatia, were harvested from a natural population in a park surrounding the Radziwill family residence in Balice, now belonging to the National Research Institute of Animal Production in Krakow (Poland). Snails were placed in a pen in an unheated greenhouse planted with white clover and grass. A hardened and turned-out aperture lip was used as a sign of sexual maturity. The mean body mass of full-grown snails was 21.9 g (SD 3.79, range from 13.88 g to 34.09 g) and the shell diameter was of 34.07 mm (SD 1.77, range from 29.1 mm to 39.2 mm). Stocking density was 51.2 snails per m’. Inside the pen, vegetation-free strips of land were left to facilitate the ob- servation of snails laying eggs. In the pen, feed was placed on wooden pallets. A sprinkling system spread water each morning and afternoon. Snails were given extruded veg- etable-mineral feed designed for breeding individuals of He- lix aspersa and produced by the Farming Cooperative in Lubnica (Wielkopolska Province, Poland). The feed was 16.0% soya-bean protein and 12.4% calcium (Ca) in the form of chalk. A detailed composition of the feed was re- served by the manufacturer. Egg laying was monitored every morning. Observations were made on breeding snails that worked their way into the soil to lay eggs. First clutches were laid on 30 May, that is, 35 days after the snails were placed 2 AMERICAN MALACOLOGICAL BULLETIN — 22° 1/2 * 2007 into the greenhouse pen. The last clutches were laid on 21 August. To prevent their escape, laying snails were covered with upturned flower pots. The next day, after egg laying was completed, egg clutches were collected from their hole in the soil. Egg clutches were incubated in soil in plastic trays. Approximately 15000 two-week-old hatchlings were released into a greenhouse pen with a density of 600 specimens per m and fed until late autumn. During the winter, hatchlings hibernated in an unheated greenhouse pen covered with Styrofoam and a gardening fabric used to protect crops from ground frost. In mid-March of the following year, the hatchlings became active and were given snail feed. In mid- April, the protective fabrics were removed to insert wooden feeders. Then, when the spring frosts passed in mid-May, the 4,800 young snails were transferred from the greenhouse to a field pen with a density of 260 specimens per m’. Breeding snails that laid eggs were weighed, their shell diameters were measured, and numbers were painted on their shells. Snails were marked to permit further observa- tions of additional egg-laying by marked specimens in the same season. Egg clutches from the marked breeding snails were weighed and egg numbers were counted. From the mass of the clutch and the number of eggs in the clutch, the mean mass of an egg was calculated. These data were used to determine the values of specific reproductive parameters. The increase of body mass and shell diameter of growing snails was meassured in September 2003, and May and July 2004. Each time, random samples of 150 specimens were collected for measurements. Temperature and relative hu- midity of the air in the greenhouse were measured 20 cm above the surface of the pen. Measurements were taken on workdays at 7:00 and 14:00 hours. The results of reproduc- tion were analyzed using analysis of variance and one-way regression using Statgraphics software. RESULTS Microclimatic conditions in the greenhouse pen Mean air temperature at 7:00 hours decreased from 20.1°C in June to 18.4°C in August, and the temperature at 14 hours ranged from 26.6°C in July to 30.0°C in August. On some days, the afternoon temperature reached 34.3°C in June and 35.9°C in August. There were no significant dif- ferences between all mean morning, afternoon, and mean of day monthly temperatures, respectively. Mean humidity of the air in the morning ranged from 77.9% in June to 86.3% in July. In the afternoon, the mean humidity of the air was 65.7% in July, but was only 54.9% and 47.2% in June and August, respectively. There were significant differences (P < 0.05) between each mean morning monthly humidity, and highly significant differences (P < 0.01) between all mean afternoon and mean daily humidity, respectively. Snails in the greenhouse were active at least until midday, copulating and laying eggs. Snails were raised under conditions of the natural pho- toperiod. The natural daylight in June lasted for 16.5-17.0 hours, during July it decreased from 16.5 hours to 15.5, and in August it decreased from 15.5 to 14.0 hours. On 21 Au- gust, when the last egg clutches of the season were found, the day length was 14 hours. Reproductive parameters in the reproductive season Significantly more snails laid eggs in the period of June- July than in August (P < 0.01), when reproductive intensity dropped by 65.7%. (Table 1). An increase in egg laying was found from the start of the reproductive season (30 May) to 21 July. This period was followed by a rapid decline in re- productive intensity, until egg laying ceased on 21 August. In June, the mean number of clutches and eggs laid during 24 hours per m° was almost the same as July, while in August this parameter dropped significantly (P < 0.05) by 65.5% and 76.2%, respectively. Between June and August, the num- ber of eggs per body mass of parent decreased very signifi- cantly (P < 0.01) by 38.4%. In successive months, the mean number of eggs per clutch decreased rapidly, and in August the mean number of eggs per clutch was 41.6% significantly lower (P < 0.01) than in June. Mean mass per one egg decreased significantly (P < 0.01) between June and the pe- riod of July-August by 5.0%. The mean mass of clutches declined very significantly (P < 0.01) between June, July, and August, by 44.7% during the whole period. The relative mass of a clutch, expressed as a percentage of parental mass, also decreased significantly (P < 0.01) by 40.5%. Significant (P < 0.05) and highly significant (P < 0.01) negative correlation coefficients were found for the relation- ship between the egg-laying intensity per 1 day per 1 m*, and the average body mass and shell diameter of Roman snails laying eggs on a particular day (Table 2). Positive, highly significant correlations were found for the relationship be- tween the individual body mass of a snail and the mean egg mass in the clutch. The same correlation was found for the diameter of the shell in relation to the average mass of a single egg and the mass of a clutch. Multiple egg laying by the snails during the same reproductive season Some snails began laying a second clutch of eggs by the middle of June and some began to lay eggs for the third time in late June and in July. In August, only eggs laid by snails laying for the third time in that season were found. During the three months of observation, 25.1% snails laid eggs twice and 5.2% laid three times. The mean interval between the first and the second clutches was 21.7 days; the mean interval REPRODUCTIVE PERFORMANCE OF HELIX POMATIA UNDER FARM CONDITIONS ie.) Table 1. Reproduction of Helix pomatia housed in a greenhouse pen. Parameter Mean percent of individuals laying eggs per 24 h Mean number of clutches laid per 24 h per m* of pen surface area; stocking density was 51.2 snails per m* Mean number of eggs laid per 24 h per m° of pen surface area; stocking density was 51.2 snails per m* Mean number of eggs laid per 1 g of body mass Mean number of eggs per clutch Mean mass of one egg in clutch (mg) Mean mass per clutch (g) Mean mass of clutch per mass of parent snail (%) Month Mean SE SD Range June 1.99% 8.61 1.74 0.16-6.06 July 252" 7.85 1.19 0.64-4.78 August 0.79" 5.48 0.39 0.24-1.36 June 1.02” 0.22 0.89 0.08-3.10 July 1.29° 0.19 0.61 0.33-2.45 August 0.40° 0.08 0.20 0.12-0.69 June 48.81° 11.62 46.48 3.42-163.61 July 46.52” 8.49 26.85 9.91-102.90 August 11.33" 2.29 6.62 3.10-19.64 June 2.29° 0.05 0.81 0.56-6.11 July 1.78® 0.06 0.82 0.61-5.63 August 4 0.07 0.46 0).26-2.40 June 48.1‘ 0.88 14.82 14-89 July 35.9" 1.05 15.47 11-88 August 28.9 1.45 10.46 5-67 June 138.7" 1.33 22.11 84-226 July 133.1° 1.38 19.39 92-200 August 130.3° 3:53 22.78 85-179 June 6.57% 0.12 2.07 2.23-13.60 July 4.72" 0.13 1.93 1.72-13.02 August 3.63" 0.18 1.20 0.71-6.20 June 30.37" 0.57 9.60 8.09-60.11 July 23:26" 0.65 9.11 9.28-52.23 August 18.05 0.81 5.50 6.11-28.19 a, b, c = significant differences (P < 0.05). A, B, C = highly significant differences (P < 0.01). between the second and the third clutches was 23.2 days. The mean total number of eggs from the three clutches was 117 (range 84-184). All reproductive parameters were higher for the first clutch than for the second. For the mean mass of clutch, differences from first to second and third clutch were highly significant (P < 0.01). The same highly significant differences (P < 0.01) were observed for mean percentage mass of clutch per mass of parent, and for the number of eggs per g mass of clutch (Table 3). Reproduction for the entire period of observation In the entire reproduction season, the mean number of eggs per clutch was 41.7, but the mean number of eggs laid during the entire reproductive period of 3 months was 61.5 per reproductive snail, because 30.3% of the snails laid eggs two or three times during the season. The mean mass of a clutch was 5.6 g, the mean mass of one egg was 132.1 mg, and the mean number of eggs per | g of snail biomass was 2.0. Total egg biomass was 38.5% of the total mass of the reproductive adults. In total, the observed snails laid in total 77100 eggs over 83 days, yielding 3145.5 eggs per m° of pen. Rearing performance Breeding snails laid 77100 eggs, out of which 40000 eggs were hatched. From the hatched eggs 27000 two-to-three- week-old hatchlings were obtained. Of the obtained hatchlings 15000 were released into the greenhouse, from which 7823 specimens or 52.2% of the initial population survived until 15 September. A total of 4864 individuals released into the greenhouse, 32.4% of the initial population, survived winter hibernation. In May of the next year, the 4800 hatchlings were released into a field pen with a density of 260 individuals per m°. By September, the mean diameter of the shells of these snails had reached 14.2 mm and the body mass reached 2.5 g. After the winter hibernation, by July 2004, the mean diameter of the shells had reached the minimum of 30.1 mm required in Poland for commercial snails to be collected from natural 4 AMERICAN MALACOLOGICAL BULLETIN — 22° 1/2 * 2007 Table 2. Results of one-way regression analysis for egg-laying traits of Helix pomatia, 1 June to 21 July 2003. Correlation Parameter | Parameter I] coefficient (1) Significance Number of clutches per m* of pen area per day, Mean mass of egg-laying snails —0.68 P < 0.05 from 30 May to 21 August Mean diameter of parent shell —0.71 P< 0.01 Mean mass of egg —0.71 P< 0.01 Number of eggs per m° of pen area per day, Mean mass of egg-laying snails —0.67 P < 0.05 from 30 May to 21 August Mean diameter of parent shell —0.67 P < 0.05 Mean mass of one egg —0.73 P< 0.05 Mass of clutch, from 30 May to 21 August Number of eggs per clutch 0.87 P< 0.01 Mass of egg-laying snails Mass of one egg 0.44 P< 0.01 Mass of clutch — not significant Number of eggs per clutch — not significant Diameter of parent shell Mass of one egg 0.48 P< 0.01 Mass of clutch 0.39 P< 0.01 Number of eggs per clutch — not significant Table 3. Reproductive performance of Helix pomatia snails which laid eggs 3 times per season. Mean mass of clutch Number of eggs Clutch Mean number Mean mass per mass of the Mean mass of one laid per 1 g of number of egg per clutch of clutch (g) parent snail (%) egg in clutch (mg) body mass I 49.6 6.8" a25° 138.6 2a I 35.2 4.5" 222" 131.0 17 IU 32-1 4.08 92" 125.0 1.6" A, B = highly significant differences (P < 0.01). populations. Also at that date, the mean body mass increased _ peratures in the mornings of June-July 2003, which averaged to 19.1 g (Figure 1). In April 2004, the coefficients of varia- approximately 20°C compensated for the influence of re- tion for the body mass and shell diameter were high (80.0% duced daylight. In August, however, the natural photoperiod and 26.4%, respectively), due to the two-month differences declined below 15 hours, which seemed to cause a rapid in age of the hatchlings, which hatched from June to August decrease in reproductive intensity and termination of egg- of the previous year. In July, the coefficients of variation laying by snails, despite of the still-high air temperature. decreased to 33.3% and 11.8%, respectively, because the | Another potential cause of sustained high reproductive in- snails grew. tensity in July was the statistically significant higher relative humidity in June than in August. Relative humidity in June was close to the optimal humidity of 75-85% recommended DISCUSSION for breading in Helix aspersa. Microclimatic conditions Reproductive parameters in the reproductive season Gomot (1990) found that under laboratory conditions, Roman snails maintained in pens in the field in Poland Roman snails have the highest reproductive activity when during June-August 2004 peaked in their reproductive out- photoperiods last 18 hours. He suggested a the temperature put in June (Chmielewski 2005). The greenhouse pen used of 20°C partly compensates for the effect of shorter photo- in the current study probably provided better microclimatic periods on reproduction. A similar effect of higher tempera- and feeding conditions for reproduction than was possible in ture at shorter photoperiods was found by Jess and Marks the field pen, and permitted the reproductive intensity in (1998) for Helix aspersa maxima, and by Gomot et al. (1989) July to remain at the same high level as in June. The repro- for Helix aspersa. Similarly, we suspect that greenhouse tem- ductive output of the snails in the current greenhouse study REPRODUCTIVE PERFORMANCE OF HELIX POMATIA UNDER FARM CONDITIONS 40 40 36 airs 36 32 baa ae 32 CLE LIA 28 28 24 =) 24 Mass of juveniles = e iS) oo Nn an o oo — _ iJ i n o Shell diameter September Apnil May July 2003 2004 2004 2004 Figure 1. Mean masses (open boxes, in g) and shell diameters (hatched boxes, in mm) of juveniles of Helix pomatia hatching from eggs laid in the greenhouse pen in 2003. Box represents + standard error around mean; whiskers indicate maximum and minimum values. is similar to that for the month of June of snails maintained under natural conditions (Lysak et al. 2001). Multiple egg laying by the snails during the same reproductive season In the current study, some individuals laid second and third clutches of eggs in July and in August. Gomot (1990) found that under experimental conditions the frequency of egg-laying by the same snails was higher during a consistent 18-hour photoperiod than during a shorter 8-hour one. Therefore, the repetitive laying of eggs may be a function of both the photoperiod as well as the length of the reproduc- tive season. The first clutches of eggs laid by Roman snails in the season were significantly heavier (P < 0.01) and propor- tionally greater in relation to parental body mass. Clutches of other Helix species living in natural conditions and laid at the beginning of the reproductive season also contain more eggs than those laid later in the season (Lazaridou- Dimitriadou and Bailey 1991). However, the clutches of eggs laid later in the season may have been the second and third clutches of the same snails, which may explain why they contained fewer eggs. Reproduction for the entire period of observation and rearing performance Our results from rearing older hatchlings, before and after hibernation in the winter of 2003/2004 in a greenhouse pen, and later moving them to a field pen, provides a basis for developing a technology for farming Roman snails over a two-year cycle. Over half of the specimens survived until the time of winter hibernation from the group of two-to- three-week old Roman snails raised in the greenhouse pen 2003, which correspondes to the survival rate required by the (on technologies for the raising of Helix aspersa in its first year of life. We obtained a satisfactory survival rate of 32.0% for the Roman snail hatchlings, calculated from the moment they were placed in the greenhouse pen in the summer of 2003 to the moment they were moved to the field pen in May 2004. The survival rate through July 2004 was lower than the 50% specified by breeding techniques for Helix aspersa, but the breeding cycle of the Roman snail is one season longer than that of Helix aspersa and it is separated by a period of winter torpor that is physiologically difficult for juvenile snails. The rapid rate of growth observed under farm conditions leads us to propose a two-year farming cycle for this species, from hatching to maturity. ACKNOWLEDGMENTS This paper has been supported by grant No. 6 PO6 2003 C/06249. LITERATURE CITED Chmielewski, M. 2005. Rozrod slimaka winniczka (Helix pomatia) w warunkach sztucznych [Reproduction of the Roman snail (Helix pomatia) in artificial conditions]. Report of the 21" Polish Malacological Seminar, April 2005. Association of Pol- ish Malacologists, Torun-Ciechocinek. Pp. 15-16 [In Polish]. Daguzan, J. 1989. Snail rearing or heliciculture of Helix aspersa Miller. In: I. F. Henderson, ed., Slugs and Snails in World Agriculture. British Crop Protection Council Monograph 41. Thornton Heath, London. Pp. 3-10. Dyduch-Falniowska, A., M. Makomaska-Juchniewicz, J. Perza- nowska-Sucharska, S. Tworek, and K. Zajac. 2001. Roman snail (Helix pomatia L.): Conservation and management in the Malopolska region (southern Poland). Ekologia 20: 265-283. Dyduch-Falniowska, A., G. Cierlik, M. Makomaska-Juchiewicz, W. Mroz, J. Perzanowski, S. Tworek, and K. Zajac. 2001b. Body size of Helix pomatia in the natural and synanthropic habitats. In: L. Salvini-Plawen, J. Voltzow, H. Sattman, and G. Steiner, eds., Abstracts of the World Congress of Malacology 2001, Vi- enna, Austria. Unitas Malacologica, Vienna. P. 89. Gomot, A. 1990. Photoperiod and temperature interaction in the determination of reproduction of the edible snail, Helix po- matia. Journal of Reproduction and Fertility 90: 581-585. Gomot, P., and A. Deray. 1990. The length of hibernation affects temperature-induced (25°C) spermatogenic multiplication in Helix aspersa Miller. Experientia 46: 684-686. Gomot, P. and L. Gomot 1989. Inhibition of temperature-induced spermatogenic proliferation by a brain factor in hibernating Helix aspersa (Mollusca). Experientia 45: 349-351. Gomot, P., L. Gomot, and B. Griffond. 1989. Evidence for a light compensation of the inhibition of reproduction by low tem- peratures in the snail Helix aspersa. Ovotestis and albumen 6 AMERICAN MALACOLOGICAL BULLETIN gland responsiveness to different conditions of photoperiods and temperatures. Biology of Reproduction 40: 1237-1245. Jeppensen, L. L. 1976. The control of mating behaviour in Helix pomatia L. (Gastropoda: Pulmonata). Animal Behaviour 24: 275-290. Jess, S. and R. J. Marks. 1998. Effect of temperature and photope- riod on growth and reproduction of Helix aspersa var. maxima. Journal of Agricultural Science 130: 367-372. Klein-Rollais, D. and J. Daguzan. 1988. Oral water consumpticn in Helix aspersa Miller (Gastropoda: Mollusca: Stylommato- phora) according to age, reproductive activity and food sup- ply. Comparative Biochemistry and Physiology 89A: 351-357. Lazaridou-Dimitriadou, M. and S. E. R. Bailey. 1991. Growth, re- production and activity rhythms in two species of Roman snails, Helix aspersa and Helix Iucorum, in non 24-hour light cycles. Journal of Zoology 225: 381-391. Lazaridou-Dimitriadou, M., E. Alpoyanni, M. Baka, T. H. Brouzi- otis, N. Kifonidis, E. Mihaloudi, D. Sioula, and G. Vellis. 1998. Growth, mortality and fecundity in successive generations of Helix aspersa Miiller cultured indoors and crowding effects on fast-, medium- and slow-growing snails of the same clutch. Journal of Molluscan Studies 64: 67-74. Lucarz, A. 1984. Etude expérimentale de l’effet du groupement sur la ponte d’Helix aspersa Miller. International Journal of Inver- tebrate Reproduction and Development 7: 185-192. Lysak, A. 1999. Significance of Helix farming for protection of Helix pomatia. Report of the 15" Polish Malacological Seminar, September 1999, Association of Polish Malacologists, Lodz. P. 260. Lysak, A., M. Ligaszewski, and Z. Mach-Paluszkiewicz. 2002. Farm- ing and histological effects of gonadotropin stimulation in Roman snails of the Helix genus. Annals of Animal Sciences 2: 87-96. Lysak, A., Z. Mach-Paluszkiewicz, and M. Ligaszewski 2001. Influ- ence of Roman snail (Helix pomatia L.) farm rearing upon its reproduction and growth rate. Annals of Animal Sciences 1: 63-74. Madec, L., A. Guiller, M.-A. Coutellec-Vreto, and C. Desbuquois. 1998. Size-fecundity relationships in the land snail Helix as- persa: Preliminary results on a form outside the norm. Inver- tebrate Reproduction and Development 34: 83-90. Stepczak, K. 1976. Wystepowanie, zasoby, uzyskiwanie i ochrona Slimaka winniczka (Helix pomatia L.) w Polsce. [Occurrence, abundance, obtainability and protection of the Garden snail (Helix pomatia L.)|. Uniwersytet im. Adama Mickiewicza w Poznaniu. Seria Zoologia 3: 60-68 [In Polish]. Accepted: 12 April 2006 22° 1/2 + 2007 Amer. Malac. Bull. 22: 7-15 Determinate growth and variable size at maturity in the marine gastropod Amphissa columbiana Bruno Pernet Department of Biological Sciences, California State University, Long Beach, 1250 Bellflower Blvd., Long Beach, California 90840-3702, U.S.A., bpernet@csulb.edu Abstract: Individuals of Amphissa columbiana from the intertidal zone of San Juan Island, Washington, U.S.A., typically have either shells with very thin, delicate apertural lips, or shells with thick, robust lips. In laboratory observations, thin-lipped snails grew rapidly but were not sexually mature, while thick-lipped snails grew very slowly or not at all and were sexually mature. These observations are consistent with the hypothesis that A. columbiana displays determinate growth, as has been inferred for many columbellids on the basis of intraspecific variation in shell form. Sizes of mature snails were very variable, with the largest individuals weighing 4.5 times more than the smallest (wet weight, excluding shell). I tested the hypothesis that maturation and associated shell thickening are phenotypically plastic responses to the presence of predators. Exposure to effluent from the predatory crab Cancer productus in the laboratory had no effect on shell form or relative shell weight (an index of shell thickness), suggesting that this is not the case. Key words: Columbellidae, shell form, growth, maturation, reproduction Determinate growth, a pattern of ontogeny in which sexual maturation is accompanied by the cessation of growth, is common among many lineages of animals. In shelled gastropods that display determinate growth, matu- ration may also be accompanied by changes in the form of the shell aperture (e.g., thickening of the outer lip and for- mation of a callus on the inner lip; Vermeij and Signor 1992). Many members of the gastropod family Columbelli- dae display such variation in apertural form within popula- tions, which suggests that they display determinate growth (McLean 1978, Jung 1989, Vermeij and Signor 1992). How- ever, in the absence of data on relationships between repro- ductive maturity, growth rate, and shell form in any colum- bellid, other interpretations of variable shell form are plausible. For example, shell form might vary in response to environmental cues like wave force, crowding, food avail- ability, or the presence of predators, and so might not be directly related to maturation (e.g., Wellington and Kuris 1983, Kemp and Bertness 1984, Appleton and Palmer 1988, Trussell 1996). Here I provide data on reproduction and growth that are consistent with the hypothesis of determi- nate growth in an intertidal population of the columbellid Amphissa columbiana Dall, 1916. Shell form thus appears to be a reliable indicator of reproductive status in this species. The sizes of mature Amphissa columbiana in this popu- lation were very variable, with the largest mature snails hav- ing 4.5 times the body wet weight (excluding shell) of the smallest mature snails. Maturation at small size likely im- poses a cost in terms of fecundity per reproductive bout in these snails, and thus requires explanation. I present the results of a laboratory experiment aimed at testing the hy- pothesis that maturation and associated shell thickening is a phenotypically plastic response to the presence of predators. In this case, the potential reduction in fecundity associated with maturation at small size might be balanced by increased resistance to predation. Exposure to effluent from the preda- tory crab Cancer productus had no effect on shell form or relative shell weight (an index of shell thickness) in A. co- lumbiana, suggesting that this is not the case. Determining the causes of variable size at maturity in gastropods with determinate growth, like A. columbiana, remains an impor- tant problem whose solution will be useful in exploring hy- potheses on life history evolution, as well as in interpreting ecological and evolutionary patterns in body size in both fossil and recent assemblages (e.g., Budd and Johnson 1991, Roy 2002). MATERIALS AND METHODS Shell form I studied a population of Amphissa columbiana from the intertidal zone of Deadman’s Bay, on the west side of San Juan Island, Washington, U.S.A. Identifications were verified by comparison with specimens of Amphissa spp. in the col- lections of the Natural History Museum of Los Angeles County in February 2003, with the assistance of J. McLean. Shell height (from the apex to the tip of the siphonal canal) was measured with calipers to 0.1 mm. Shell and body weights were estimated in living snails using the method of Palmer (1982). Living snails were weighed while immersed in seawater, blotted dry, and then weighed in air. Twelve 8 AMERICAN MALACOLOGICAL BULLETIN thin-lipped and 12 thick-lipped individuals of a wide range of shell heights were measured and weighed as above, then frozen and dissected to separate shell from body tissues. Shell and body tissues were rinsed in fresh water, dried at 65°C for three days, and weighed. Shell weight was related to submerged weight using the ordinary least squares regres- sion: shell weight (g)=1.494* submerged weight (g) —0.002 (r°=0.999). This regression was used to predict shell weight throughout this study. Body wet weight was calculated by subtracting estimated shell weight from the weight of a snail in air. For this sample of 24 individuals, body wet weight was well correlated with body dry weight (dry weight [g]=0.19* estimated wet weight [g] + 0.001, r°=0.945), suggesting that body wet weight calculated in this way is a good estimator of body mass. Growth of thin-lipped and thick-lipped snails Growth was examined in two sets of laboratory obser- vations made in the winter (during the reproductive season) and spring (after the reproductive season) of 2003. In the first, started in January 2003, 6 thin-lipped (heights 13.2- 16.4 mm) and 10 thick-lipped (14.5-16.6 mm) snails were marked by attaching numbered beetags to their shells with cyanoacrylate glue. The snails were measured and weighed as described above, then placed in a plastic container (15 x 15 x 4 cm) with mesh sides submerged in flowing seawater (8-10°C) in a laboratory seatable. The snails were fed muscle tissue from scallops (Chlamys spp.) weekly for 10 weeks, after which they were measured and weighed again. The second set of growth observations, started in April 2003, included 11 thin-lipped (heights 12.3-16.4 mm) and 9 thick- lipped (11.1-16.0 mm) snails. After being marked, measured, and weighed, the snails were placed in a mesh-sided con- tainer (30 x 18 x 10 cm) submerged in flowing seawater (10-14°C). Snails were fed muscle from Chlamys spp. or Nuttalia obscurata (Reeve, 1857) 1-2 times weekly for the next 11 weeks, after which they were measured and weighed again. In both sets of observations, food was always present in excess. Sexual maturity Maturity of field-collected snails was assessed in two ways. First, | compared lengths of the penises of thin-lipped and thick-lipped snails in a collection of snails made in December 2002 and January 2003 (during the reproductive season). Snails were relaxed in a mixture of equal volumes of seawater and 7.5% MgCl,-6H,O, then fixed in 10% formalin in seawater. Their shells were later removed. Penises were removed from six thick-lipped males, pinned out straight, and measured to the nearest 0.5 mm with a ruler. Penises of the five thin-lipped snails examined were too small to pin 22 * 1/2 * 2007 out, and were measured without removing them from adults. Second, I looked for evidence of deposition of egg cap- sules by snails maintained in the laboratory during the re- productive season. I collected 17 thin-lipped snails and 26 thick-lipped snails in Dec 2002 and sexed them by holding them off the substratum by their shells with forceps and looking for a penis as they extended their bodies from their shells. Most individuals of both groups lacked penises and were assumed to be females. I divided the thin-lipped snails into two separate mesh-walled containers (making sure to include several males in each container), and did the same for the thick-lipped snails. All four containers were sub- merged in flowing seawater and the snails were fed scallop muscle weekly until March 2003. Containers were examined for the presence of egg capsules at every feeding. Causes of variation in size at maturity In a laboratory experiment, I tested the hypothesis that odor cues associated with the presence of crushing predators induce changes in shell form associated with maturation. Because food level has been linked to changes in shell thick- ness in several snails (e.g., Kemp and Bertness 1984, Bould- ing and Hay 1993), I also manipulated this variable. In May 1999, I collected 50 thin-lipped Amphissa columbiana (shell height 10.5-18.8 mm) from the intertidal zone about 1 km south of Deadman’s Bay. These were marked, measured, and weighed. Groups of four or five snails selected to span the size range of collected snails were placed into each of 12 small, mesh-sided containers. Two of these containers were placed in each of six plastic aquaria (20 x 13 x 13 cm). Each aquarium had a separate source of inflowing seawater. Three of the aquaria, “crab” treatments, contained a single indi- vidual of Cancer productus Randall, 1839 (59-66 mm cara- pace width). The crab was restricted to the bottom half of the aquarium (away from the snail containers) with rigid plastic mesh. Thus, in the three “crab” aquaria, snails shared a common pool of water with a potential predator. The remaining three “no crab” aquaria were identical to “crab” aquaria except that no crab was included. In each of the six aquaria, snails in one container received food (1/4 of the adductor muscle from a Chlamys) weekly (fed treatment); snails in the other container received no food (starved treat- ment). Containers were cleaned at each feeding. Each crab was fed a single individual of Chlamys spp. weekly. When crabs molted, they were replaced with newly collected crabs 55-66 mm in carapace width. Snails were maintained in this experiment for two months, after which they were measured and weighed. During the experiment 8 snails (of the total of 50) died, all of these in fed treatments; these were excluded from analyses. Because of this mortality, the number of snails in each container varied from 3-5, except for one DETERMINATE GROWTH IN AMPHISSA COLUMBIANA 9 container in which all the snails died. Effects of predator (crab vs. no crab) and food level (starved vs. fed) on relative shell weight (shell dry weight/total weight of the snail) were assessed in a factorial ANOVA. RESULTS Shell form Most individuals of Amphissa columbiana exhibited one of two distinct shell morphologies, with a few individuals displaying intermediate forms. In thin-lipped individuals (Fig. 1A), the outer lip of the aperture was extremely thin (50-70 4m) and very delicate, frequently breaking when snails were handled. The outer lip of the aperture was con- tinuously curved, with no straight portions. No denticles were present on either the columellar or outer lips of the aperture, and there was no callus on the columellar lip of the aperture. The remainder of the shell was usually relatively free of epibionts, and the apical whorls were not eroded. Thin-lipped snails were most common deep in crevices and overhangs, and especially under large cobbles at low tide levels. In these habitats, they often occurred in aggregations of up to 20-30 individuals of a variety of sizes. On occasion they were also found singly on exposed rock surfaces. In contrast, in thick-lipped individuals (Fig. 1C) the outer lip of the aperture was 750-1000 um thick and very robust. Further, a segment of the outer lip of the aperture— from about 1/3 of the aperture back from the anterior end to about 1/4 of the aperture forward from the posterior end of the aperture—was usually straight and approximately par- allel to the coiling axis of the shell. The outer apertural lip bore 15-20 short denticles on its inner surface and several denticles were usually found on the columellar apertural lip as well. A raised callus was present on the columellar lip of the aperture, although it was frequently obscured by en- crusting epibionts. At Deadman’s Bay, thick-lipped snails were often found in crevices and overhangs and were also relatively common on exposed rocks at low tide. Thick- lipped snails were usually found singly, not in aggregations. Snails of intermediate shell form (Fig. 1B) were much less common than either thin-lipped or thick-lipped snails. Intermediates had apertures of intermediate thickness, with or without a straight portion of the outer lip of the aperture. Figure 1. Shells of Amphissa columbiana from the intertidal zone of Deadman’s Bay, San Juan Island, Washington, U.S.A. A, Thin-lipped individual. Note the continuously curved outer lip of the aperture and the lack of apertural teeth. B, Individual with intermediate shell form. Note the presence of a callus on the columellar lip of the aperture. C, Thick-lipped individual. Note the well-developed teeth, especially on the outer lip of the aperture, the callus on the columellar lip of the aperture, and the greatly thickened outer apertural lip (compare to A). Scale bar=5 mm. 10 AMERICAN MALACOLOGICAL BULLETIN A Othin A + intermediate A thick BE. 08 shell dry weight (g) NORD ODNO 6 8 10 12 74 16 18 20: 22 relative shell weight 6. © 10 12 14 16 18-20. 22 45 C ne body wet weight (g) 6 8 10 12 shell height (mm) 14 16 18 20 22 Figure 2. Relationships of shell height with (A) shell dry weight, (B) relative shell weight, and (C) body wet weight for thin-lipped, intermediate, and thick-lipped individuals of Amphissa columbiana. Ordinary least squares regressions of log-transformed data showed the following relationships: In (A), for n=32 thin-lipped snails, log (dry shell weight) =2.454 (+0.106) * log (shell height) —3.626 (+0.118), °=.947; for n=32 thick-lipped snails, log (dry shell weight) =2.653 (+0.173) * log (shell height) —3.591 (+0.207), r-=.887 (adjusted means significantly different by ANCOVA, 22 * 1/2 * 2007 A raised callus was present on the columellar lip of the aperture and apertural denticles were sometimes present (but weakly developed). The shell was usually neither very eroded nor covered with epibionts. Snails classified according to these criteria also differed quantitatively in allocation to shell and body weight. This is illustrated by data on 74 snails collected in January 2003 (snails collected in 1999 and 2002 showed very similar pat- terns). Thin-lipped and thick-lipped snails overlapped broadly in shell height (thin 8-18 mm, thick 12-20 mm). ANCOVA of log-transformed data showed that thick-lipped snails had significantly higher shell weight than did thin- lipped snails when shell height was considered as a covariate (Fig 2A; see figure legend for regression parameters). Snails of intermediate shell form fell between thin-lipped and thick-lipped snails. Relative shell weight (shell dry weight/ total weight of the snail) was much higher in thick-lipped snails (67%+2.6% standard deviation) than in thin-lipped snails (52%+4.5%), with snails of intermediate shell form falling in between (62%+6.4%; differences among all three classes of snails significant by ANOVA followed by Fisher’s PLSD post-hoc tests, p<0.001). Relative shell weight was unrelated to shell height in intermediate and thick-lipped snails, but negatively correlated with shell height in thin- lipped snails (Fig. 2B). Thus, in addition to qualitative char- acters, a quantitative index (relative shell weight) could be used to distinguish thin-lipped and thick-lipped snails. Thin-lipped and thick-lipped snails also differed in re- lationships between body weight and shell height. ANCOVA of log-transformed data showed that thin-lipped snails had very slightly (but significantly) higher body wet weight than did thick-lipped forms (Fig. 2C). Body weight of the largest thick-lipped snails (0.41 g) was about 4.5 times that of the smallest thick-lipped snails (0.09 g). Growth of thin and thick-lipped snails Thick-lipped snails added very little shell or tissue on average (Table 1). Because results from the two observation periods were very similar, I pooled data for thick-lipped snails to test for differences from hypothesized means of zero. In thick-lipped snails, the mean growth increment in shell height was not significantly different from zero (t-test, p<0.0001). In (B), for n=32 thin-lipped snails, log (tissue wet weight) =2.818 (+0.069) * log (shell height) —4.064 (+0.077), r°=,.983; for n=32 thick-lipped snails, log (tissue wet weight) =2.945 (+0.134) * log (shell height) —4.247 (+0.161), =.941 (adjusted means significantly different by ANCOVA, p=0.0038). In (C), re- gression of relative shell weight on shell height is not significant for intermediate and thick-lipped snails (p=0.54 and 0.06, respec- tively), but it is significant for thin-lipped snails (n=32, relative shell weight =0.614 —0.007 * height, r°=.266). DETERMINATE GROWTH IN AMPHISSA COLUMBIANA 1] Table 1. Initial sizes and changes in size of thin-lipped and thick-lipped individuals of Amphissa columbiana held in a common laboratory environment. Two sets of observations of growth increment were made, the first starting in January 2003 (during the reproductive season) and the second in April 2003 (after the reproductive season). Changes in size are reported as means (standard deviation). January-March observations (10 weeks) Shell form n Initial height (mm) Change in height (mm) Change in shell weight (g) Change in body weight (g) Thin-lipped 6 Thick-lipped 10 14.48 (1.187) 15.47 (0.785) 4.8 (1.58) 0.06 (0.18) April-July observations (11 weeks) 0.128 (0.021) 0.004 (0.004) 0.225 (0.093) 0.007 (0.007) Change in shell weight (g) Change in body weight (g) Shell form n Initial height (mm) Change in height (mm) Thin-lipped 1113.49 (0.686) 5.9 (1.87 Thick-lipped 9 14.32 (1.672) 0.0 (0.15) 0.197 (0.043) 0.007 (0.007) 0.238 (0.105) 0.009 (0.008) p=0.16), but mean growth increments for shell mass and body mass growth were very slightly greater than zero (t- tests, p<0.001). Thin-lipped snails grew very rapidly (Table 1), with many individuals adding more than a whorl of new shell during these observations. There were no obvious dif- ferences between growth increments of thin-lipped snails measured during the reproductive season versus those mea- sured after the reproductive season. Growth increments for all three parameters were significantly greater in thin-lipped snails than in thick-lipped snails in both sets of observations (t-tests, p<0.005). Sexual maturity Penises of six thick-lipped snails (shell heights 19.0-22.1 mm) ranged from 10-12 mm in length (mean 11.2 mm), while penises of five thin-lipped snails (shell heights 14.1- 20.3 mm) ranged from 1-3 mm in length (mean 1.7 mm). Thus, penises of thick-lipped snails were on average about seven fold longer than those of thin-lipped snails (difference significant by t-test, p<0.001). None of the 17 thin-lipped snails observed in the labo- ratory during the reproductive season of 2002-2003 depos- ited egg capsules. In contrast, many egg capsules (indicating deposition by multiple females) were observed in both of the chambers containing thick-lipped snails. Causes of variation in size at maturity Pood availability had clear effects on growth (expressed as percent change) in shell height, shell weight, and body weight (Fig. 3A). These effects were in the expected di- rection—fed snails grew much more than did starved snails. In contrast, presence or absence of crabs had no obvious effects on snail growth. By the qualitative criteria described above, shell form in experimental snails did not change during the experiment— no snails obviously switched from the thin-lipped to the thick-lipped morph. Predator treatment had a nearly signifi- cant effect on an index of shell thickness, relative shell weight, when raw data were used in the analysis (F=3.84, degrees of freedom 1, 38; p=0.057; Fig. 3B). When container means were used in place of raw data (a modification of the analysis that reduces its power, but also reduces the chances of Type I error), predator treatment had no significant effect on relative shell weight (F=1.46, d.f. 1, 7; p=0.267). Food availability had a significant effect on relative shell weight regardless of whether raw data or container means were used in the analysis (raw data F=27.06, df. 1, 38; p<0.0001; container means F=14.08, d.f. 1, 7; p=0.007). At the beginning of the experiment, for all snails, shell com- prised a mean of 50.3+5.0% of total snail weight. At the end of the experiment, starved snails had a higher mean relative shell weight (54.2+4.5%) than fed snails (47.4+4.5%; Fish- er’s PLSD post-hoc test for container mean data, p=0.008; Fig. 3B). DISCUSSION Determinate growth explains striking variation in shell form in the intertidal gastropod Amphissa columbiana (Fig. 1). Young snails build thin-lipped shells and grow rapidly, but eventually they alter the shape and thicken the apertures of their shells and grow only very slowly, if at all (Table 1). These differences in shell form and growth rate are corre- lated with differences in sexual maturity. Male thick-lipped snails have large penises, while the penises of thin-lipped snails are so minute that they are likely not functional in copulation. Thin-lipped snails maintained in the laboratory did not deposit egg capsules, while thick-lipped snails de- posited many egg capsules, suggesting that female thin- lipped snails are not sexually mature. However, in these experiments, female thin-lipped snails were only offered thin-lipped males (which have tiny, probably nonfunctional penises) to mate with, so an alternative hypothesis is that female thin-lipped snails were sperm-limited. Although I studied patterns of shell form, growth rate, and maturity in 12 AMERICAN MALACOLOGICAL BULLETIN — 22 ¢ 1/2 + 2007 fed XC unfed percent change crab no crab shell height crab shell weight crab no crab body weight shell/total weight * 100 crab no crab Figure 3. Percent change in (A) shell height, shell weight, and body weight and (B) relative shell weight (shell weight/total weight) in individuals of Amphissa columbiana after two months of rearing under different conditions in the laboratory. Means (+ standard error) are shown, with n=8-14 replicate snails for each bar. The dashed line in (B) represents the mean relative shell weight of all snails at the beginning of the experiment. only one intertidal population, both thin-lipped and thick- lipped snails are also present in intertidal populations else- where in Washington State and in subtidal populations in the San Juan Islands (personal observation). In these popu- lations as well, similar differences in shell form are likely attributable primarily to determinate growth. To my knowl- edge, these are the first data on growth and reproduction applied to testing the hypothesis of determinate growth in columbellids. Determinate growth in the family has previ- ously been inferred solely from variation in shell form (McLean 1978, Vermeij and Signor 1992). In a popular guide to shells, White (1976) mentioned variation in shell thickness in intertidal populations of Am- phissa columbiana, but interpreted it with the hypothesis that A. columbiana “develops thicker shells where exposed to wave action.” The variability he mentioned (but did not describe further) may simply have been normal ontogenetic variation in shell form and thickness like that described here. It is possible that shell form in A. columbiana varies both ontogenetically and with environmental conditions like wave action, but further studies comparing shell form among populations exposed to different environmental con- ditions are needed to clarify this issue. Tupen (1999) de- scribed substantial site-associated quantitative variation in adult shells of another columbellid, Alia carinata. A plausible explanation for such variation is phenotypic plasticity, simi- lar to that suggested by White (1976). It is interesting to note that collections of post-mortem shells of Amphissa columbiana may, depending on their provenance, represent strongly biased samples of life history stages and shell sizes. For example, assemblages of post- mortem shells of A. columbiana present on the shore at Deadman’s Bay (almost all occupied by hermit crabs) are mostly those of thick-lipped snails greater than 13 mm in height (personal observation). The rarity of thin-lipped shells of juvenile snails in these assemblages may be a result of low mortality among juvenile snails (e.g., because they tend to inhabit protected microhabitats), greater vulnerabil- ity of juvenile shells to shell-destroying predators, or rapid post-mortem degradation of thin-lipped shells. Collections made from natural assemblages of dead shells are very likely to reflect a bias towards the shells of larger, mature snails. For similar reasons, the delicate juvenile shells of A. colum- biana (and perhaps other determinately-growing columbel- lids) may also be less likely to be preserved as fossils than are robust adult shells. Such potential sampling and taphonomic biases should be taken into consideration when making in- ferences from shell form in determinately-growing snails. DETERMINATE GROWTH IN AMPHISSA COLUMBIANA 13 If, as argued above, thickening of the shell aperture and cessation of growth is correlated with maturation in Am- phissa columbiana, then size at maturity varied over a range of about 1.7-fold as shell height and 4.5-fold as wet body weight in the Deadman’s Bay population. This wide range in size within a single population of a single species represents about 22% of the total range of adult sizes of 144 species of columbellids studied by Roy (2002). (Roy used the geomet- ric mean of shell height and width as an index of size, and I estimated this index for small and large adult individuals of A. columbiana for comparison with his data.) Such great variability in adult size appears to be fairly common among determinately-growing gastropods (Vermeij) 1980). Al- though no data are available on the relationship between body size and fecundity in A. columbiana, by analogy with other gastropods (e.g., Iyengar 2002, Angeloni 2003) the two are very likely correlated. The fecundity of small A. colum- biana is thus probably limited relative to that of large indi- viduals. What causes many snails in this population to ma- ture and stop growing at small sizes? Some component of the observed variation in size at maturity is likely to be genetic (e.g., Richards and Merrit 1975, Brown et al. 1985) and might be identified in breeding studies. Size at maturity might also vary as a plastic response to several environmental variables, such as food quantity or quality, or risk of predation or parasitism (e.g., Brown 1985, Brown et al. 1985, Lafferty 1993). I examined whether one of these environmental factors, the presence of odor cues as- sociated with a crushing predator, affects the timing of maturation and shell thickening in Amphissa columbiana. High risk of predation might induce early maturation in Amphissa columbiana because this would improve the chances of reproduction before death, but also because shell thickening, which is associated with maturation, is expected to render shells more resistant to attack by shell-crushing predators (e.g., Palmer 1985, Vermeij and Signor 1992, Trus- sell 2000). My results suggest that odor cues associated with a potential crushing predator (the crab Cancer productus) do not induce maturation and shell thickening in Amphissa co- lumbiana (Fig. 3). There were no qualitative changes in the form of thin-lipped snails raised in the presence of crabs. Whether or not C. productus was present, thin-lipped fed snails increased in shell weight by a mean of about 55% of their initial shell weight over 8 weeks (Fig. 3A). Because fed snails added even more body weight over the course of the experiment (roughly 75% of initial body weight, Fig. 3A), relative shell weight, an index of shell thickness, decreased slightly over the course of the experiment (Fig. 3B). This index was expected to increase if snails thickened their shells in response to crab-associated odor cues. Addition of 55% of initial shell weight, if allocated primarily to thickening the shell instead of to continued spiral growth, would be nearly sufficient to fully convert a thin-lipped to a thick-lipped snail (Fig. 2). This result has many possible interpretations. One is that odor cues associated with predators genuinely do not affect the timing of maturation in Amphissa columbiana. Alternatively, predator cues may have been too weak to elicit a response in experimental snails; snails may respond to predators other than small Cancer productus; or snails may respond to a correlate of predator presence (e.g., the smell of crushed conspecifics) instead of to predator odor itself. However, C. productus is known as an important crushing predator of intertidal snails of a wide range of sizes (A. columbiana falls within that size range: Palmer 1985, Bould- ing et al. 1999) in similar habitats in the northeast Pacific. Nucella lamellosa (Gmelin, 1791) reared under similar con- ditions to those described here alter shell form in the pres- ence of odor cues from C. productus within 2.5 months (Appleton and Palmer 1988), suggesting that sufficient predator cue and time was allowed in my experiments to elicit a response if it existed. In N. lamellosa, the effect of crab odor is enhanced when the C. productus are fed con- specific snails, but is detectable no matter what the crabs are fed (Appleton and Palmer 1988). These considerations lead me to favor the first interpretation, that size at maturation in A. columbiana is genuinely not affected by the presence of cues associated with predators. Regardless of the presence or absence of predator odor cues, food level had a small but significant effect on relative shell weight in Amphissa columbiana. Both fed and starved snails grew during these experiments, but in fed snails, more of the increase in total weight was allocated to the body tissues than the shell, leading to a decline in relative shell weight. In starved snails (which grew much less than fed snails), more of the increase in total weight was due to increased shell weight, leading to a slight increase in relative shell weight (Fig. 3). It seems likely that this increase in allocation to shell is not associated with maturation, but instead is related to slow growth (e.g, Kemp and Bertness 1984; Boulding and Hay 1993). However, habitat quality (of which food quantity and quality is a major part) has been associated with age and size at maturation in several other gastropods (e.g., Vermeij 1980, Brown 1985, Lafferty 1993) and with other changes in shell form in other species (e.g., Appleton and Palmer 1988). Another way of identifying environmental cues that might affect size at maturation is to examine variation in adult size among habitats. Tupen (1999), for example, found significant differences in shell dimensions among popula- tions of another columbellid, Alia carinata (Hinds, 1844), from several different habitats. He argued that these differ- ences might have resulted from phenotypic plasticity or 14 AMERICAN MALACOLOGICAL BULLETIN post-settlement selection on the basis of habitat-specific dif- ferences in predation or wave exposure. No comparable studies have been carried out for Amphissa columbiana. However, individuals of A. columbiana from some subtidal habitats in the San Juan Islands may reach much larger adult sizes than those in the intertidal zone at Deadman’s Bay. At the latter site, I have not seen snails larger than 21 mm shell height in five years of observations, but among the 7 sub- tidally collected specimens in the Friday Harbor Laborato- ries Synoptic Collections, 4 have shell heights greater than 25 mm, with one snail whose apertural lip is of intermediate thickness measuring 28.7 mm. A. columbiana have plank- tonic, feeding larvae that spend at least two weeks in the plankton (personal observation) so subtidal and intertidal populations are likely well-mixed genetically. These data suggest that environmental differences between intertidal and subtidal habitats affect size at maturation in A. columbiana. Individuals of Amphissa columbiana that were kept in the laboratory for two months without food grew substan- tially, adding on average about 20% to their initial shell weight and about 5% to initial body weight (Fig. 3). It is not clear what fueled the growth of starved snails. Shell growth was not occurring at the expense of body weight, as both increased over the course of the experiment. Similar in- creases in shell weight in the face of starvation have been recorded previously in A. columbiana (and other snails: Palmer 1983), although in that study body weight is not reported. Hatfield (1979) found that another columbellid, Anachis avara (Say, 1822), increased in shell height by about 10% over six weeks of starvation, and suggested that par- ticulate or dissolved organic matter present in the seawater fueled this growth. This may be the case for A. columbiana as well. My data (Fig. 3) suggest a trend of higher overall growth for snails raised in the presence of crabs versus those raised without crabs. It is possible that some bivalve tissue was torn into small bits by crabs (all of which were fed) and was carried by water into snail containers. However, starved snails reared in the absence of crabs (and in the absence of crab food) also increased in both shell and body weight. ACKNOWLEDGMENTS I thank G. Bergsma and L. Tretter for help collecting snails, and J. McLean for access to museum specimens and help in identification. Comments by M. deMaintenon, G. Vermeij, J. Voltzow, and an anonymous reviewer improved the manuscript. This work was carried out at the Friday Harbor Laboratories of the University of Washington, and I thank the director and staff of that lab for providing facilities and support. 22 ° 1/2 * 2007 LITERATURE CITED Angeloni, L. 2003. Sexual selection in a simultaneous hermaphro- dite with hypodermic insemination: Body size, allocation to sexual roles and paternity. Animal Behaviour 66: 417-426. Appleton, R. D. and A. R. Palmer. 1988. Water-borne stimuli re- leased by predatory crabs and damaged prey induce more predator-resistant shells in a marine gastropod. Proceedings of the National Academy of Sciences 85: 4387-4391. Boulding, E. G. and T. K. Hay. 1993. Quantitative genetics of shell form of an intertidal snail: Constraints on short-term response to selection. Evolution 47: 576-592. Boulding, E. G., M. Holst, and V. Pilon. 1999. Changes in selection on gastropod shell size and thickness with wave exposure on Northeastern Pacific shores. Journal of Experimental Marine Biology and Ecology 232: 217-239. Brown, K. M. 1985. Intraspecific life-history variation in a pond snail: The roles of population divergence and phenotypic plas- ticity. Evolution 39: 387-395. Brown, K. M., D. R. DeVries, and B. K. Leathers. 1985. Causes of life-history variation in the freshwater snail Lymnaea elodes. Malacologia 26: 191-200. Budd, A. F. and K. G. Johnson. 1991. Size-related evolutionary patterns among species and subgenera in the Strombina group (Gastropoda: Columbellidae). Journal of Paleontology 65: 417- 434. Hatfield, E. B. 1979. Food sources for Anachis avara (Columbelli- dae) and a discussion of feeding in the family. The Nautilus 93: 40-43. Iyengar, E. V. 2002. Sneaky snails and wasted worms: Kleptopara- sitism by Trichotropis cancellata on Serpula columbiana. Ma- rine Ecology Progress Series 244: 153-162. Jung, P. 1989. Revision of the Strombina-group (Gastropoda, Co- lumbellidae), fossil and living. Schweizerische Paldontologische Abhandlungen 111: 1-298. Kemp, P. and M. D. Bertness. 1984. Snail shape and growth rates: evidence for plastic shell allometry in Littorina littorea. Pro- ceedings of the National Academy of Sciences 81: 811-813. Lafferty, K. D. 1993. The marine snail, Cerithidea californica, ma- tures at smaller sizes where parasitism is high. Oikos 68: 3-11. McLean, J. 1978. Marine Shells of Southern California. Natural H1s- tory Museum of Los Angeles County Science Series 24: 1-104 [Revised edition]. Palmer, A. R. 1982. Growth in marine gastropods: A non- destructive technique for independently measuring shell and body weight. Malacologia 23: 63-73. Palmer, A. R. 1983. Relative cost of producing skeletal organic matrix versus calcification: Evidence from marine gastropods. Marine Biology 75: 287-292. Palmer, A. R. 1985. Adaptive value of shell variation in Thais lam- ellosa: Effect of thick shells on vulnerability to and preference by crabs. The Veliger 27: 349-356. Richards, C. S. and J. W. Merrit. 1975. Variation in size of Biom- phalaria glabrata at maturity. The Veliger 17: 393-395. Roy, K. 2002. Bathymetry and body size in marine gastropods: A DETERMINATE GROWTH IN AMPHISSA COLUMBIANA shallow water perspective. Marine Ecology Progress Series 237: 143-149. Trussell, G. C. 1996. Phenotypic plasticity in an intertidal snail: The role of a common crab predator. Evolution 50: 448-454. Trussell, G. C. 2000. Predator-induced plasticity and morphologi- cal trade-offs in latitudinally separated populations of Litto- rina obtusata. Evolutionary Ecology Research 2: 803-822. Tupen, J. W. 1999. Shell form and color variability in Alia carinata (Neogastropoda: Columbellidae). The Veliger 42: 249-259. Vermeij, G. J. 1980. Gastropod shell growth rate, allometry, and adult size: Environmental implications. In: D. C. Rhoads and R. A. Lutz, eds., Skeletal Growth of Aquatic Organisms. Plenum Press, New York. Pp. 379-394. Vermeij, G. J. and P. W. Signor. 1992. The geographic, taxonomic and temporal distribution of determinate growth in marine gastropods. Biological Journal of the Linnean Society 47: 233- 247. Wellington, G. M. and A. M. Kuris. 1983. Growth and shell varia- tion in the tropical eastern Pacific intertidal gastropod genus Purpura: Ecological and evolutionary implications. Biological Bulletin 164: 518-535. White, J. S. 1976. Seashells of the Pacific Northwest. Binford and Mort, Portland. Accepted: 31 August 2006 i Amer. Malac. Bull. 22: 17-26 Land snail diversity in subtropical rainforest mountains (Yungas) of Tucuman, northwestern Argentina Eugenia Salas Orofio', Maria Gabriela Cuezzo', and Fatima Romero” ' CONICET—Facultad de Ciencias Naturales, Universidad Nacional de Tucuman, Miguel Lillo 205, 4000 Tucuman, Argentina, mcuezzo@unt.edu.ar * Fundacion Miguel Lillo, Miguel Lillo 251, 4000 Tucuman, Argentina Abstract: A survey of the micro-land snails in the mountain rainforest or “Yungas” of Tucuman Province, Northwestern Argentina (26°20'-27°30'S) was carried out. A total of 75 samples were processed from 25 stations of 10 x 10 m each. We identified the snails collected to species level and built a species per station data matrix to analyze patterns of diversity. Non-parametric estimators (ICE and Chao2) using EstimateS 5.0.1 were used to estimate the true diversity of the area, the degree of undersampling, and spatial aggregation in the data. Other diversity measurements such as Shannon and Whittaker indices were also calculated. Our study estimated the micro-mollusc species richness of the Yungas of Tucuman to be 21 species distributed among 9 families, with a notably high number of specimens collected (7741). The most speciose family was Charopidae with a total of seven species identified. However, the most abundant families were Diplomma- tinidae and Systrophiidae. Adelopoma tucma and Wayampia trochilioneides were the most abundant species, the least frequently found species was Lilloiconcha tucumana. The non-parametric estimators showed that our survey was complete and that patchiness did not affect diversity estimations. San Javier and Escaba stations had the highest species richness, the most diverse station was Escaba. Most previous studies in other places in the world show high species richness and low density of specimens. In Tucuman, on the contrary, the absolute abundance of specimens was high with a low species richness. Key words: Species richness, micro-molluscs, taxonomy, Adelopoma, South America The subtropical mountain rainforest or “Yungas” in luscs (de Winter and Gittenberger 1998, Myers et al. 2000). northwestern Argentina ranges from 300 to 2000 m above The micro-snails are poorly studied in most forest regions of sea level, in areas where it rains more than 1000 mm annu- South America. This situation is worsened because specimen ally (Brown and Grau 1995). Grassland areas distributed representation of the local fauna in malacological collections above 1800-2000 m are often included in this ecoregion, is also inadequate, due to the limited funds available for although they have a completely different kind of vegetation taxonomic studies (Wheeler 2004). Limited taxonomic and from that characteristic of the Yungas. The Yungas, together _ ecological information of most land snail groups distributed with the rainforest of the Paranaense region (Northeastern in South America make it almost impossible to incorporate Argentina), represents less than 2% of the surface of Argen- this group of invertebrates into plans of conservation (Myers tina but contains more than 50% of the biodiversity present et al. 2000, Lydeard et al. 2004). One of the present preju- in this country. The Yungas, also known in Argentina as dices is the statement that land snails are neither diverse nor “Tucumano-Boliviano” rainforest, are latitudinally distrib- abundant in ecosystems like tropical and subtropical rain- uted along a narrow area in the northwestern part of the forests due to their type of soil and weather (Solem 1984). country from the boundary with Bolivia (S 22°) to southern — Solem stated that the lack of nutrients and litter and the Catamarca province (S 28°). One of the reasons for the — abundance of predators on the molluscan fauna make the comparatively high diversity of flora and fauna reported in tropical rainforest an unfavorable habitat for snails. On the the Yungas is the marked altitudinal variation that occurs in contrary, other environments with stable temperatures, this region. The flora comprises three well differentiated al- moderate moisture, and rich soil litter are habitats with the titudinal zones having distinct physiognomies and floras. highest diverse land snail faunas. Several studies around the Huge areas of the Tucumano-Boliviano rainforest were — world (in Madagascar [Emberton 1995, Emberton ef al. already degraded or profoundly altered before basic infor- 1996, 1999], Mexico [Naranjo and Garcia 1997], French mation on biodiversity, including a complete inventory of | Guyana [Gargominy and Ripken 1998], Cameroon [de Win- molluscan species, could be obtained. Detailed data on di- ter and Gittenberger 1998], Kenya [Tattersfield et al. 2001], versity of undisturbed forest faunas are therefore urgently — and Venezuela [Martinez 2003]) have tried to test Solem’s needed. Our lack of knowledge on biodiversity is particularly — hypothesis with respect to the environmental conditions. apparent concerning invertebrate taxa, especially land mol- Most of them concluded that in fact tropical rainforests can 17 18 AMERICAN MALACOLOGICAL BULLETIN contain very speciose gastropod faunas. Nevertheless, very little information for comparisons exists with respect to sub- tropical rainforest areas, where the floral composition, weather conditions, seasonality, and type of soil differs con- siderably from tropical areas. Moreover, a progressive de- cline in richness and abundance of characteristic tropical land snail groups inhabiting northern and central South America (e.g., Systrophiidae, Streptaxidae, Neocyclotidae) towards lower latitudes and the eventual replacement of these groups by others inhabiting southern subtropical rain- forest habitats is a process not clearly documented nor un- derstood. It is also not clear if the decline of species richness along a latitudinal gradient proposed for vertebrates is also valid for most invertebrates including land molluscs but ex- cluding arthropods (Longino et al. 2002). Even if this lati- tudinal gradient in South America would also be valid for land snails, the causes in the changes of species richness and species turnover have not yet been hypothesized. This study aimed to: (1) Determine the taxonomic com- position of the micro-molluscan fauna from the Yungas in Tucuman, (2) Establish the species richness (gamma diver- sity) of this ecoregion, which is the result of the alpha (within community) and beta (variation of species compo- sition among communities) diversities, and (3) Compare these results with the species richnesses and abundances re- ported for other parts of the world, especially the southern hemisphere. MATERIALS AND METHODS Study area Our field research was carried out in the biogeographic region of the Yungas or Tucumano-Boliviano rainforest in Tucuman Province, northwestern Argentina (26°15'- 27°40'S). This area corresponds to the southernmost exten- sion of the Andean Neotropical Montane Forest (Cabrera 1971, Cabrera and Willink 1973). Precipitation in the Yun- gas is characterized by a monsoonal regime with rainfall concentrated in the summer and early autumn months (November-April) (Grau and Veblen 2000). The Yungas ecoregion is usually divided into three altitudinal vegeta- tion zones (Brown and Grau 1995, Valdora and Soria 1999). The Piedmont Forest located from 400 to 700 m above sea level consists of the basal portion of the Yungas, with an annual precipitation of about 600 to 1000 mm. It is also called “Transition rainforest” (Cabrera 1971, Prado 1995). It is mainly characterized by deciduous trees and a low abun- dance of epiphytic vegetation. Some of the most com- mon trees are Tipuana tipu (Benth, 1898) (common name: “tipa”), Enterolobium contortisiliqum (Vellozo, 1892) » (“pacara”), Phoebe porphyria (Grisebach, 1889) (“laurel”). 22° 1/2 + 2007 The degradation process (e.g., over explotation, fires, crops) produced great structural transformation of the piedmont rainforest into xerophilic woods (Brown and Grau 1995). This basal zone of the Yungas is the most affected by human activity in Tucuman. The Lower Montane Forest extends from 700 to 1500 m, with an annual precipitation between 1500 and 3000 mm. Some of the most characteristic trees of this altitudinal zone are Tipuana tipu (Beth, 1898) (“tipa”), Jacaranda mimosifolia Don, 1822 (“tarco”), Phoebe porphyria (Grisebach, 1889) (“laurel”), Myrcianthes uniflora (Linné, 1773) (“arrayan”) and Blepharocalyx gigantea Lillo, 1911(“horco molle”). Above 1500 m is the Upper Montane Forest, with an annual precipitation between 900 and 1300 mm. This zone is characterized by the presence of Podo- carpus parlatorei Pilger, 1903 (“pino del cerro”), and Alnus acuminata (Kunth, 1904) (“aliso”). Sampling The material used in the present study consisted of adult terrestrial micro-molluscs from different taxonomic groups. Micro-molluscs or micro-snails are defined as those species that as adult have shells not larger than 5 mm maximum dimensions. Molluscs larger than 5 mm, informally called macro-molluscs, were not considered in the present study because they were too low in abundance and patchily dis- tributed to be adequately sampled by our methods. Thus we excluded Epiphragmophora Doering, 1874, Scutalus Albers, 1850, Drymaeus Albers, 1850, and specimens of Veronicel- lidae. All of the specimens of micro-snails collected were deposited at the Fundacion Miguel Lillo Malacological Col- lection, Tucuman, Argentina. The methodology used in the present study was adapted from those of Emberton et al. (1996) and de Winter and Gittenberger (1998). We sampled 25 stations consisting of a 10 x 10 m patch during the sum- mer season when land snails are more active in Tucuman Province (Fig. 1, Table 1). Within each station we qualita- tively searched for micro-snails for half an hour in selected microhabitats that seemed the most favorable for them, such as between exposed roots of trees where dead organic ma- terial usually accumulates and under and between dead tree trunks lying on the forest floor. In each station we also took three samples of 50 x 50 cm quadrats of leaf litter plus 2. cm of topsoil from moist, sheltered microhabitats so that a total of 75 samples were processed from the total of 25 stations. Because land snails are generally distributed in patches, it seemed more appropriate to take samples from places most suitable for them. For each station we recorded the altitude (Thommen altimeter), latitude and longitude (Garmin GPS), general topography, and kind of vegetation. Leaf litter plus topsoil samples were placed in plastic bags and kept as cool as possible (10-15°C) in the laboratory no longer than one week until processing. Each bag was opened daily for DIVERSITY OF LAND SNAILS IN NORTHWESTERN ARGENTINA 19 28° 66° 45 30 15 65° 45° Figure 1. Map of Tucuman province indicating the 25 sample stations located in 15 geographic localities. Station numbers cor- respond to those listed in Table 1. aeration. The qualitative search provided more living speci- mens in most of the stations than did the soil plus leaf litter samples. All snails found alive were relaxed in deoxygenated water for 24 hours and then preserved in 90% ethanol. Soil plus leaf litter samples were dry-sieved through three de- creasing mesh widths (3 mm, 1.5 mm, and 0.5 mm) in the laboratory. The three samples from the same station search were treated together for the statistical analysis because there were no differences in the species composition nor in the abundance of specimens among them. After the process of separation of soil and snails, shells were sorted and identified using a Leica MZ6 stereoscopic microscope. An altitudinal transect from 800 to 1460 m was carried out in the Sierra de San Javier locality (Stations 1-5). In this transect the stations (five in total) were sampled every 100 m of elevation to estimate the possible altitudinal variation of the community in the different vegetation zones of Yungas. Taxonomic identification Specimens were identified based on shell characters only, except in the case of Wayampia trochilioneides Table 1. Stations (Sn) sampled for micro-molluscs in Tucuman. T. = total number of species collected per station. T = total number of ind specimens collected per station. PF = Piedmont Forest, LM = Lower Montane Forest, TF = Transition Forest, UM = Upper Montane Forest. Dn Z He a KF COO OND OF WHF NO Re RR Re RS i SOON DD WT SW LY 21 Name Raco, La Hoyada El Cadillal, Aguas Chiquitas El Cadillal, Aguas Chiquitas Tafi del Valle Tafi del Valle Villa Nougues Timbo Viejo, Burruyacu Dept. Hualinchay, Trancas Dept. Rearte, Trancas Elevation Rainfall (m) Latitude S_ Longitude Ws (mm/year) = =T, = 1,4 ~~ Type of vegetation Nina’s road, San Javier Mountain 920 26°43’ 11" 65°17'35" 1505 14 565 PF Nina’s road, San Javier Mountain 1020 26°43'07" 65°18'02" 1505 14 289 LM Nina’s road, San Javier Mountain 1120 26°43'11" 65°18'17" 1505 14. 688 LM Nina’s road, San Javier Mountain 1210 26°43'09" 65°18'33" 1505 14 386 LM Nina’s road, San Javier Mountain 1460 26°42'54” 65°18'47" 1505 7 15 UM Horco Molle, Yerba Buena Dept. 940 26°51'32” 65°25'32" 1500-2000 1 6241 PF Estancia San Javier, Tafi Viejo Dept. 1050 26°46' 26" 65°23'24" 1173 = 173 LM Potrero de Las Tablas, Lules Dept. 750 26°46’ 16" 65°25'34" 1083 11 809 PF Potrero de Las Tablas, Lules Dept. 800 26°46'16" 65°25'34" 1083 9)" \ 115 PF San Miguel de Tucuman, Capital 450 26°48'26" 65°14'58" 986 | 82 PF 880 26°46'16" 65°28'04" 694 1] 270 LM 515 26°36'48" 65° 11/11" 1096 13 352 PF 560 26°36'51" 65° 1111" 1096 I 362 PF 550 27°05'43" 65°37'10" 900 vi 138 PE 900 27°03'22” 65°40'20" 800-1000 10 148 LM 1200 — = La73 id. 55 LM Medinas Mountain, Buruyacu Dept. 1200 — — — 4 14] LM 26°26'28” 64°59'43" — 7 245 TF Gonzalo, San Pedro de Colalao 1250 26°18'40" 65°31'45" 600-800 9 445 LM 1360 26°19'35" 65°2'21" 700-900 10 687 LM 1370 26°21'05" 65°32'25" 5 16 TR Escaba abajo, Juan Bautista Alberdi 600 27°38'57" 65°44'46" 1119 10 88 PF Escaba abajo, Juan Bautista Alberdi 650 27°40'02" 65°45'40" 1119 13 681 PF to LM Escaba arriba, Juan Bautista Alberdi 700 27°39'14" 65°46'11" 1119 12 491 PF to LM 1235 27°19'41” 65°55'33" 1400-1600 10 253 LM Cochuna (@Orbigny, 1835), for which anatomical dissections were also used. Only specimens with complete shells were con- sidered. Some shells from species difficult to identify were mounted on stubs, sputter coated, and observed with a Jeol 35CF scanning electron microscope. We identified the ma- terial collected to species level and built a species per station data matrix to analyze patterns of diversity. For the taxo- nomic identification we used the available regional iden- tification keys (Fernandez and Castellanos 1973) plus the original species descriptions. We followed the general clas- sification for Neotropical micro-molluscs of Muller da Fonseca and Thome (1993) and a general classification of Gastropoda by Wade et al. (2001). Diversity measurements To estimate the true number of species in the commu- nity, we used non-parametric estimators of diversity calcu- lated with EstimateS, version 5.0.1 (Colwell 1997). EstimateS uses the relative abundance of rare species to estimate the number of species not seen. EstimateS was also employed to assess the degree of undersampling and spatial aggregation in the data. This program computes randomized species ac- cumulation curves and reports the means of various statistics based on those curves (Heyer et al. 1999). A species- accumulation curve is a plot of the accumulated number of species found with respect to the number of units of effort expended. In the present study, the effort is of a discrete-type (samples). The species-accumulation curve for a highly un- dersampled fauna will appear nearly linear, with each sample adding many new species to the inventory. On the contrary, the curve for a thoroughly sampled fauna will reach a pla- teau, with few or no species being added with additional sampling (Longino 2000). We treated the data as presence and absence scores of species by samples. The estimator chosen in the present study for non-parametric richness es- timation were the Incidence-based Coverage Estimator (ICE) (Lee and Chao 1994) and the Chao2 estimator (Chao 1987). Relying on the concept that rare species carry the most information about the number of missing ones, Chao uses only singletons and doubletons to estimate the number of missing species. When the abundance-based coverage es- timators (ACE) were used, transforming the former matrix by adding the abundance of each taxon per sample, similar curves were obtained. We used the default values for number of randomizations (50) and cutoff values for coverage-based estimators (10). Both estimators, ICE and Chao2, augment the negatively biased observed richness by a factor that de- pends on the presence of and distribution within samples of “rare” taxa. For the estimator used, the “rare” species are those observed in only one or two samples (ICE is more complex but the logic is the same). When all species in the data matrix had been observed multiple times, the inventory AMERICAN MALACOLOGICAL BULLETIN 22 ° 1/2 * 2007 ap 25.00 | Z = 20.00 4 tees. A Zz ——Sobs 2) 7 = 15.00 4 Uniques Ps | —— Duplicates 7 —e ICE —*— Chao2 aaccin| me = = Cole 0.00 TIT 3.4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 STATIONS Figure 2. Species richness estimators and patchiness indicators for the micro-snails of the Yungas in Tucuman. Species accumulation curves and Coleman curves obtained with EstimateS 5.0.1. is complete but when inventories are replete with “rare” species, true richness is underestimated. EstimateS 5.0.1 in- clude techniques to assess graphically the degree of spatial aggregation in the data. For this purpose a Coleman curve (rarefaction curve) was also calculated (Fig. 2). Coleman curves are not estimators of species richness in the same sense as the other estimators. While Chao2 and ICE estimate total species richness, rarefaction curves estimate sample species richness from the pooled total species richness (Col- well 1997). Rarefaction curves represent the means of re- peated re-sampling of all pooled samples. Other diversity indices used in this work were the Shan- non Index, Evenness Index and the Whittaker Index. The Shannon Index (H’) determines the alpha (a) diversity ob- tained through the equation: H'= ->, p; log, P; Where p= proportion of individuals in the i" species (ni/N) (Magurran 1989). Evenness (E) was calculated using the formula: R=H' 7H max Its value fluctuated between 0 and 1. An E=1 means that all of the species present at the station are equally abundant (Magurran 1989). To estimate beta ({§) diversity we used the Whittaker index (J): IT=(S/a)-1 which consist of the total number of species (S) divided by the mean number of species per station (a) (Whittaker 1975, Magurran 1989, de Winter and Gittenberger 1998). The Whittaker index provides a measurement of the variability DIVERSITY OF LAND SNAILS IN NORTHWESTERN ARGENTINA 21 among sites (= stations). The community that contributes with fewer species will have the highest B diversity. If I is equal to 1, then the stations have the same or identical faunas and higher values indicate increasing differentiation (de Winter and Gittemberg 1998). The Whittaker index does not take into account the distribution of species on spatial or environmental gradients, so this index is not intended to be used to measure species turnover (Vellend 2001). The total number of specimens recovered per station (each station of 100 m*) was used as an estimate of abundance. RESULTS Faunal composition In the present study, we found 7741 specimens repre- senting 21 species in 15 genera and 9 families, one of them was a “prosobranch” and the remaining were pulmonates (Table 2). The total species per stations data matrix is given in Table 3. The most speciose family was Charopidae, with a total of seven species recorded. Following Charopidae, were Pupillidae (3 species) and Systrophiidae (3 species). The three families together represented 62% of the total number of species. The rest of the families were represented by lower numbers of species, generally one or two. Of the total num- ber of families represented, only one is carnivorous, the rest are herbivorous or detritivorous. The species richness per station ranged from 4 to 14 species (Table 1). The richest stations, with 14 species recorded, were between 1000 m and 1200 m of elevation in the San Javier mountains (Tables 1, 3). These stations were located in the biological reserve of the National University of Tucuman, a protected area called Parque Sierra de San Javier. Density fluctuated between 16 and 809 specimens per 100 m* (Table 3). The densest station was Potrero de Las Tablas with 809 specimens in 100 m° located at 750 m. Abundance of specimens was low in those stations that had a substrate with sand, or with a high proportion of small stones and that were more exposed to sunlight. The most common and abundant taxa were Adelopoma tucma Doer- ing, 1884 and Wayampia trochilioneides (d’Orbigny, 1835). Some species of the genus Radiodiscus Pilsbry, 1906 were also very abundant, especially in stations from San Javier Mountain (Tables 1, 3). The micro-molluscan fauna was dominated by two families, Systrophiidae and Diplomma- tinidae, that were present in 24 out of 25 stations sampled, although the number of genera represented in each of these families was low in the study region. Individuals of W. tro- chilioneides were absent only in stations 10 and 17 corre- sponding to San Miguel de Tucuman city and Medinas (Tables 1, 3). Station 10 had an environment modified by human activities, where invasive species (Zonitoides arbore- Table 2. Systematic classification of the species collected at the 25 stations sampled in the Yungas of Tucuman. Classification systems according to Wade et al. (2001) and Muller da Fonseca and Thome (1993). CLASS GASTROPODA SUBCLASS PULMONATA ORDER EUPULMONATA SUBORDER STYLOMMATOPHORA INFRAORDER ORTHURETHRA FAMILY PUPILLIDAE Pupisoma latens Hylton Scott, 1960 Pupisoma dioscoricola (Adams, 1845) Gastrocopta pulvinata Hylton Scott, 1948 INFRAORDER SIGMURETHRA FAMILY CHAROPIDAE SUBFAMILY ROTADISCINAE Radioconus pilsbryi (Hylton Scott, 1957) Radioconus crenulatus (Hylton Scott, 1963) Radiodiscus wygodzinskyi Weyrauch, 1965 Radiodiscus katiae Hylton Scott, 1948 Radiodiscus golbachi Weyrauch, 1965 Trochogyra gorduraensis (Thiele, 1927) SUBFAMILY AMPHIDOXINAE Ptychodon amancaezensis (Hidalgo, 1869) FAMILY HELICODISCIDAE Lilloiconcha tucumana (Hylton Scott, 1963) FAMILY FERUSACCHDAE Caecilioides consobrina (dV Orbigny, 1835) FAMILY SYSTROPHIIDAE Wayampia trochilioneides (VOrbigny, 1835) Drepanostomella tucma Hylton Scott, 1948 Miradiscops sp. FAMILY ZONITIDAE Zonitoides arboreus (Say, 1916) Zonitoides nitidus (Miller, 1774) FAMILY EUCONULIDAE Guppya lilloana Hylton Scott, 1948 Guppya aenea Hylton Scott, 1948 FAMILY SUBULINIDAE Opeas pumilum (Pfeiffer, 1840) SUBCLASS “PROSOBRANCHIA” ORDER CAENOGASTROPODA SUBORDER ARCHITAENIOGLOSSA SUPERFAMILY CYCLOPHOROIDEA FAMILY DIPLOMMATINIDAE Adelopoma tucma Doering, 1884 ous [Say, 1816], Opeas pumilum [Pfeiffer, 1840]) were more frequent than native ones. Station 17 was a dry transition forest habitat with a low frequency of occurrence of several species that typically inhabit humid forests. Individuals of A. tucma were very abundant in humid habitats of low mon- tane forest, especially when those habitats were well pre- served, as at stations 1-4. When the conditions of the forests 22 AMERICAN MALACOLOGICAL BULLETIN — 22 1/2 + 2007 showed some alteration or had secondary vegetation, A. loiconcha tucumana Hylton Scott, 1963 (12%) found only in tucma was notably less common or was completely absent. piedmont forest and in low montane forest of San Javier. It The species Guppya aenea (Hylton Scott, 1948) had a high ~— was absent in all the other locations. Opeas pumilum frequency of occurrence (96%) in the stations, being absent —_—_ (Pfeiffer, 1840), Z. arboreous and Zonitoides nitidus (Miiller, only in Medinas (station 17), but the congeneric Guppya 1774) were found in localities with high human activities, lilloana (Hylton Scott, 1948) occurred with less frequency such as Tucuman City, Villa Nougues, and El Cadillal (84%). Among the less frequently occurring taxa was Lil- stations. Table 3. Data matrix of the species collected and the stations sampled and used in the analysis. SPECIES/ STATIONS 1 i) es) nS nn 6 7 8 9 10 11 12 #13 = «14 15 16 17 18 19 20 21 22 23 24 25 Radioconus pilsbryi 0 l O 16 14 0 0 0 0 0 0 0 0 0) 0 0 15 0 0 0 0 0 0 0 0 Radiodiscus crenulatus 2 0 3 2. 11 0 0 2 0 0 0 1 0 0 0 0 0 0 oO 0 0 0 10 8 3 Radiodiscus wygodzinskyi 55 86 240 146 O ] 0 37 4 0 3: 30 9 0 14 O 0 19 0 0 0 0 0 0 4 Trochogyra gorduraensis 8 1 79 16 O 18 0 0 18 0 0 1 0 0 0 4 6 0 0 0 0 0 14 2 4 Ptychodon amancaezensis 1 6 5 2 0 28 5 (0) 0 0 ] 0 6 6 0 O 7 4 88 138 4 0 30.40 2 Radiodiscus golbachi 0 0 0 2 O 0 0 0 ] 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 0 Lilloiconcha tucumana 0 1 ] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0) 0) 3 Radiodiscus katiae 20 15 13 16 5 4 0) 0 0 0) 3 10 26 0 6 0 0 0 189 139 5 14 40 35 0 Cecilioides consobrina 7 13 0) 0 O 0 4 17 0 0 9 55 30 5 8 0 0 0 0 0 0 10 37 * 1) 0 Zonitoides arboreus 1 0 0 0 0 0 13 0 0 ] O 44 0 11 0: 1 113 0 0 0. 0 0 0 0 0 Zonitoides nitidus 0 0 0 0 0 0 2 1 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 0 0 Pupisoma latens 7 5 5 2 O 1 ] 3. 0) 1 12 14 0 2. 1 0 0 19 31 O 14 52 31 15 Pupisoma dioscoricola 0 0) 0 0 0) 0 0) 0 0) 0 O 14 0 5 +0 0 13 0 0 O ) 19 0 0 Gastrocopta pulvinata 0 0 0 0 0 0 0 5 0 0 9 6 V4 0 0 0 0 0 12 Il O 2. 1 -.33 0 Wayampia trochilioneides 104 46 105 31 9 57 42 249 25 0 67 55 92 63 26 29 0 146 102 288 5 5 50 67 63 Miradiscops Sp. 32. 70 26 21 5 20 28 3110 0 21 9 0 1 3. 0 0 0 4 9 0 8 16 10. 52 Dreponostomella tucma i 2 2 6 0 20 3 0 0 0 0 oO O 0 0 0 0 0 2 2 0 0 0 oO O Guppya lilloana 10 2 18 5 4 23 2 10 8 0 1 11 50 10 6.0 0 1 4 8 1 8 8 8 0 Guppya aenea 27 12 60 20 3 35 4 34 32 ll 9 31 109 42 4 3 0 27 25°. 27 1 14 40 55 2 Adelopoma tucma 290 29 131 101 O 4 67 420 6 15 146 87 2 0 74 16 0 35 0 34 0 10 364 191 105 Opeas pumilium 0 0 0 0 0 0 2 TOTAL 565 289 688 386 51 211. 173 809 115 82 270 352 362 = 138 148 55 141 245 445 687 16 88 681 491 253 Mean number of specimens per station 27 «14 «#33 «18 «624 = «10 8.2 39 54 39 13 17 17 6.5 7 2.6 6.7 12 21 33 O07 41 32 23 12 DIVERSITY OF LAND SNAILS IN NORTHWESTERN ARGENTINA 23 Table 4. Calculated values of the Shannon index (H’), Whittaker index (J), and Evenness for each sampled station. STATIONS H’ I Evenness 1 2.3097 0.50 0.6066 2, 2.7856 0.50 0.7316 3 2.6649 0.50 0.7202 4 2.6817 0.50 0.7043 5 2.6163 2.00 0.932 6 2.9126 0.90 0.8419 vi 2.4355 0.75 0.704 8 1.883 0.90 0.5668 9 2.7591 1.33 0.8704 10 1.175 4.25 0.7413 11 2.3091 0.90 0.6441 12 3.0736 0.61 0.8306 13 2.7911 0.75 0.7786 14 2.0256 2.00 0.7215 15 2.3689 1.10 0.7131 16 1.8242 2.00 0.6498 17 1.0877 4.25 0.5044 18 1.8369 2.00 0.6543 19 2.1997 1.30 0.6939 20 2.334 1.10 0.7026 21 2.0488 3.20 0.8824 22 3.1332 1.10 0.9432 23 25152 0.61 0.6797 24 2.8186 0.75 0.7862 25 2.1881 1.10 0.6587 In the altitudinal transect between 800 and 1460 m of elevation in the San Javier Mountains, the species richness and composition of the micro-snail community between piedmont and low montane forest were the same. A sharp decrease in the number of species (from 14 to 7) was de- tected between low montane and upper montane forest. Density was also low in the upper montane forest. Diversity estimators and diversity index The behavior of ICE and Chao2 estimators with respect to the accumulation species curve (Sobs) is shown in Fig. 2. The gap between richness estimators and the observed spe- cies accumulation curve was very narrow. The curves con- verged, becoming flat with increasing numbers of samples. Both Chao2 and ICE estimators estimated 21 species, which closely approximated the observed species accumulation curve at the sample number 20. The behavior of the “rare” species curves (uniques and duplicates) was typical of a com- plete inventory (Heyer et al. 1999). When the number of stations was low the curve for the unique species rose faster than did the curve for duplicate species. With increasing numbers of stations, both declined, crossing and tending towards zero when no species remained to be discovered and almost all species were known from two or more locali- ties. In the present analysis the calculated Coleman curve nearly followed the observed curve, showing no evidence of patchiness. The diversity values obtained using the Shannon index fluctuated from 1.08-3.13 (Table 4). The most diverse sta- tions were in the Escaba area (H’ = 3.13; E = 0.94) (station 22) and El Cadillal (H’ = 3.07; E = 0.83) (station 12). The stations with lowest diversities were Medinas (H' = 1; E = 0.50) (station 17) and San Miguel de Tucuman city (H’ = 1.17; E = 0.74) (station 10). The values calculated for the Evenness Index (E) ranged from 0.56 to 0.94 (Table 4). These high values close to 1 mean that those stations were not very different in species abundance. The stations richest in species (stations 2-4, each with 14 species recorded) had similar high values of evenness (0.70-0.73), meaning that in each of these stations the spe- cies were close to be equally abundant. The Whittaker index (J) fluctuated between 0.50 and 4.25 with an average of 1.46 (Table 4). This high value in- dicated a substantial degree of beta diversity or differentia- tion among stations. DISCUSSION We recorded the presence of 21 species of micro- molluscs from stations in the Yungas area in northwestern Argentina. Comparisons of species richness with previous works worldwide are difficult to make because most of them had considered macro- and micro-molluscs together and applied different methodologies. Some of the studies carried out in the southern hemisphere reported the highest levels of species richness in the world. Barker and Mayhill (1999) reported the presence of 105 species of terrestrial molluscs within the Pukeamaru District in New Zealand, de Winter and Gittenberger (1998) reported 97 species of land snails in a square kilometer of rainforest in Cameroon and in Mada- gascan rainforests, 80 species of micromolluscs in 20 x 20 m patches were found by Emberton et al. (1996), and 61 species were found in a square kilometer in Malaysian Borneo (Schilthuizen and Rutjes 2001). The North American land snail fauna is relatively well studied (Lydeard et al. 2004). Emberton (1995) reported that the most diverse site is lo- cated on Pine Mountain, Kentucky, U.S.A., which has yielded 44 species of land gastropods in four hectares. In Central and South America, several lists of taxa from re- stricted geographical localities are available in some coun- tries but research on diversity of land gastropods are very scarce. In consequence, diversity is probably seriously un- derestimated. In Mexico, a total of 52 species have been reported from tropical rain forest leaf litter in southern Ve- 24 AMERICAN MALACOLOGICAL BULLETIN racruz (Naranjo-Garcia 1997), 34 species from a km° of forest in French Guyana (Gargominy and Ripken 1998), and in the Camisea region, Peru, 49 species have been also found (Ramirez et al. 1999). Unfortunately, countries like Bolivia or Colombia with a high biodiversity have been poorly in- ventoried for land snails. Although the species richness recorded for Tucuman in the present study is low (21 species of micro-snails) in com- parison with other previously studied sites in the southern hemisphere, the total number of specimens collected was very high (7741 micro-molluscs specimens). Density in our study (specimens in a 10 x 10m station) ranged from 16 to 809 and varied altitudinally. In comparison, Emberton et al. (1999), working in southeast areas (total of 0.04 ha) of Madagascar, found a total of 2430 specimens from 80 micro- molluscan species. Densities (specimens in a 20 x 20 m plot) in their study ranged from 20 to 104. In the present study, species richness and density declined with increasing altitude along a transect carried out in San Javier Mountains. The highest density in Tucuman was found in Potrero de Las Tablas, a piedmont forest station. The most speciose family of the micro-mollusc fauna in Tucuman is Charopidae with 7 species represented in the area from a total of 26 species of the same family reported for Argentina (Fernandez 1973). Charopidae is the domi- nant and most speciose family of other faunas of the south- ern hemisphere, such as the Pukeamaru fauna in New Zea- land. Barker and Mayhill (1999) found 56 species of Charopidae in that area, a notably higher species richness compared to Tucuman. One of the dominant families in our study was Systrophiidae, with three species represented in Tucuman out of 8 species of this family reported for Argen- tina (Fernandez 1973). In northern areas of South America, the Systrophiidae were found to be one of the predominant families, as is the case in Camisea, Peru, with 14 species reported in that study area (Ramirez et al. 1999) from a total of 55 species cited for the whole country (Ramirez et al. 2003). On the contrary, the family Charopidae (25 species reported for Argentina [Fernandez 1973]) seems to be more abundant in southern South America than in Peru (13 spe- cies reported [Ramirez et al. 2003]). Studies on distribution, species richness, and systematic classification of Charopidae from South America are still scarce. Many families that are common and abundant in northern South America are not present or are poorly represented in southern South America. Similarly, another carnivorous family, the Strep- taxidae, which is very abundant and diverse in northern South America, is scarce in southern areas of South America. Adelopoma tucma Doering, 1884, one of the most com- mon species in the Yungas, was very abundant in the richest stations of this study, with specimens numbers ranging from 105 to 420 in stations with 10 to 14 species (the highest tO i) “1 ies) * 2007 species number recorded in a station). This species is very abundant in well preserved forests between 600 and 1200 m of elevation that had low human impact. This species was less common or is completely absent in places where the substratum and vegetation were altered by human activity. This was apparent in stations 1-4 from the San Javier pro- tected area that has a low human impact compared with the other stations. A second highly abundant species was Wayampa trochilioneides (@Orbigny, 1835), whose number of specimens was high at the same stations as A. tucma. In areas of piedmont forest with anthropogenic alterations, some non-indigenous species from the genera Opeas Albers, 1860 and Zonitoides Lehmann, 1862 were present. Of the indigenous fauna, the less frequently found species were Lil- loiconcha tucumana Hylton Scott, 1963 (Helicodiscidae), which was found in only two stations, and Radiodiscus gol- bachi Weyrauch, 1965 (Charopidae), found in three stations. On the contrary, the most frequently found species of the indigenous fauna was Guppya aenea (Hylton Scott, 1948), which was found in 24 stations. According to our study, the inventory was found to be complete by the non-parametric estimators of diversity used, which are particularly suitable for small samples (Colwell and Coddington 1994). These results indicated that with further collecting effort in this area no other species would probably be found. Historical records for Tucuman provy- ince, mainly from malacological collections, show the pres- ence of additional species of micromolluscs, such as Zilcho- gyra hyltonscottae Weyrauch, 1965 and Radiodiscus thomei Weyrauch, 1965 (Charopidae). These two species were not found in our study area, perhaps because species of Cha- ropidae inhabiting South America are not well described. Taxonomic revisions are urgently needed, in part to test the validity of nominal species. For instance, Radiodiscus thomei Weyrauch, 1965 and Radiodiscus katiae Hylton Scott, 1948, are extremely similar and original descriptions do not pro- vide unique characters for each species. A revision of the genus is necessary to resolve these taxonomic issues. The Coleman curve closely followed the observed spe- cies-accumulation curve, indicating that aggregation in the data was not affecting estimates of species richness. The true diversity of micro-snails in the Yungas of Tucuman was estimated to be 21 species, with a high abundance of specimens. The 8 diversity calculated with the Whittaker index had a high mean value of 1.46. This suggests substantial differ- entiation in species composition among the stations (Barker and Mayhill 1999). The richest stations, such as those in San Javier (1-4 in Fig. 1, Table 1) had similar Whittaker indices (0.5), but station 5 from the upper montane forest of San Javier that had a low species richness, had a high value of the Whittaker index (2), meaning that there was difference in DIVERSITY OF LAND SNAILS IN NORTHWESTERN ARGENTINA diversity of the fauna present in this last station in compari- son with the others. Stations with high Whittaker indices, have low species richness (Magurran 1989) and, in this case, were the ones with more human pressure (e.g., stations 10, 17). The a diversity, the diversity in each station/com- munity, calculated with the Shannon index, showed the highest values at El Cadillal (station 12) and Escaba (station 22). These values were similar to the ones obtained in some stations of the Pukeamaru area in New Zealand (Barker and Mayhill 1999). However, values obtained with the Shannon index are difficult to compare with other places because of the different methodology employed and area considered. Land snail faunas of tropical rainforest tend to be quite diverse (Emberton ef al. 1996). Much of this diversity is a consequence of the micro-molluscan fauna. Because most South American rainforests are largely undercollected for micro-land snails and are undergoing significant defor- estation, there is an urgent need to collect and study the molluscan fauna. In solving the present biodiversity crisis, taxonomic work remains an essential tool. Revisionary tax- onomy is frequently dismissed as merely descriptive, which belies its strong intellectual content and hypothesis-driven nature (Wheeler 2004). Diversity studies on South American molluscan fauna as well as systematic revisions of land mol- luscs are urgently needed and must be developed, especially in countries where high biodiversity occurs. ACKNOWLEDGMENTS Special gratitude is expressed to the Argentinean Na- tional Park Service Agency, especially to the Northwestern regional office, which provided several working permits to carry on research in the National Parks areas. Thanks are also extended to the San Javier Park from the Tucuman National University for allowing us to work and collect in their preserved area. We thank H. R. Fernandez and E. Dominguez for comments, suggestions, and critical reviews of the manuscript. We are also grateful to R. K. Colwell and T. A. Pearce, reviewers of the manuscript, who made useful suggestions that improved the final version. This research was funded by grant PIP 0554/98 awarded to M. G. Cuezzo by the Argentine National Council of Scientific Research (CONICET). LITERATURE CITED Barker, G. M. and P. C. Mayhill. 1999. Patterns of diversity and habitat relationships in terrestrial mollusc communities of the i) Nn Pukeamaru Ecological District, Northeastern New Zealand. Journal of Biogeography 26: 215-238. Brown, A. and H. Grau. 1995. Investigacion, Conservacion y Desar- rollo en Selvas Subtropicales de Montana. Laboratorio de In- vestigaciones Ecologicas De Las Yungas, Universidad Nacional de Tucuman, Argentina. Cabrera A. L. 1971. Fitogeografia de la Republica Argentina. Boletin de La Sociedad Argentina de Botanica 14: 1-42. Cabrera, A. L. and A. Willink. 1973. Biogeografia de America Latina. Programa Regional de Desarrollo Cientifico y Tecnologico, Departamento de Asuntos Cientificos, Secretaria de la Orga- nizacion de los Estados Americanos, Serie Biologica 13, Wash- ington D.C, Chao, A. recapture data with unequal catchability. Biometrics 43: 783- 791, Colwell, R. K. 1997. EstimateS: Statistical estimation of species 1987. Estimating the population size for capture- richness and shared species from samples. Version 5.0.1. User’s guide and application. Available at: htpp://viceroy.eeb.uconn .edu/estimates 20 December 2004. Colwell, R. K. and. J. A. Coddington.1994. Estimating terrestrial biodiversity through extrapolation. In: D. L. Hawsworth, ed., Biodiversity Measurement and its Estimation. Chapman and Hall, London. Pp. 101-118. de Winter, A. J. and E. Gittenberger. 1998. The land snail fauna of a square kilometer patch of rainforest in southwestern Cam- eroon: High species richness, low abundance and seasonal fluctuations. Malacologia 40: 231-250. Emberton, K. C. 1995. Land-snail community morphologies of the highest- diversity sites of Madagascar, North America, and New Zealand, with recommended alternatives to height- diameter plots. Malacologia 36: 43-66. Emberton, Kk. C., T. A. Pearce, and R. Randalana. 1996. Quantita- tively sampling land-snail species richness in Madagascan rainforests. Malacologia 38: 203-213. Emberton, K. C., T. A. Pearce, and R. Randalana. 1999. Molluscan diversity in the unconserved Vohimena and the conserved Anosy mountain chains, Southeast Madagascar. Biological Conservation 89: 183-188. Fernandez, D. 1973. Catalogo de la malacofauna terrestre argen- tina. Comision de Investigaciones Cientificas de la Provincia de Buenos Aires 4: 1-197. Fernandez, D. and Z. Castellanos. 1973. Clave genérica de la mala- cofauna terrestre Argentina. Revista del Museo de La Plata XI, Zoologia 107: 265-285. Gargominy, O. and T. E. J. Ripken. 1998. Micro-pulmonates in tropical rainforest litter: A new bio-jewel. In: R. Bieler and P. M. Mikkelsen, eds., Abstracts of the World Congress of Mala- cology, Washington DC, USA. Field Museum of Natural His- tory, Chicago. P. 116. Grau, H. R. and T. T. Veblen. 2000. Rainfall variability, fire and vegetation dynamics in neotropical montane ecosystems in north-western Argentina. Journal of Biogeography 27: 1107- 1121. Heyer, W. R., J. Coddington, W. J. Kress, P. Acevedo, D. Cole, T. L. Erwin, B. J. Meggers, M. G. Pogue, R. W. Thorington, 26 AMERICAN MALACOLOGICAL BULLETIN R. P. Vari, M. J. Weitzman, and S. H. Weitzman. 1999. Ama- zonian biotic data and conservation decisions. Ciencia e Cul- tura: Journal of the Brazilian Association for the Advancement of Science 51: 372-385. Lee, $. M. and A. Chao. 1994. Estimating population size via sample coverage for closed capture-recapture models. Biometrics 50: 88-97. Longino, J. T. 2000. What to do with the data. In: D. Agosti, J. D. Majer, L. E. Alonso, and T. R. Schults, eds., Ants. Standard Methods for Measuring and Monitoring Biodiversity. Smithso- nian Institution Press, Washington, D.C. Pp. 186-203. Longino, J. T., J. Coddington, and R. K. Colwell. 2002. The ant fauna of a tropical rain forest: Estimating species richness in three different ways. Ecology 83: 689-702. Lydeard, C., R. Cowie, W. F. Ponder, A. E. Bogan, P. Bouchet, S. A. Clark, K. S. Cummings, T. J. Frest, O. Gargominy, D. G. Herbert, R. Hershler, K. E. Perez, B. Roth, M. Seddon, E. E. Strong, and F. G. Thompson. 2004. The global decline of nonmarine mollusks. BioScience 54: 321-330. Magurran, A. E. 1989. Diversidad Ecologica y su Medicion. Ediciones Vedra, Barcelona. Martinez, R. 2003. Moluscos. In: M. Aguilera, A. Azocar, and E. Gonzalez Jiménez, eds., Biodiversidad en Venezuela. Fun- dacion Polar, Fondo Nacional de Ciencia, Tecnologia e Inno- vacion, Caracas. Pp. 488-513. Muller da Fonseca, A. L. and J. W. Thomé.1993. Descricao de Glabogyra Subgen. N., Recaracterizagao de Austrodiscus twomeyt (Parodiz, 1954) e reclassificao das espécies sudameri- canas dos géneros Austrodiscus Parodiz, 1957, Radioconus Baker, 1927, Radiodomus Baker, 1930, Trochogyra Weyrauch, 1965 (Charopidae) e Zilchogyra Weyrauch, 1965 (Helico- discidae) (Gastropoda, Stylommatophora, Endodontoidea). Theringia Série. Zoologica 75: 97-105. Myers, N., R. A. Mittermeier, C. G. Mittermeier, G. da Fonseca, and J. Kent. 2000. Biodiversity hotspots for conservation pri- orities. Nature 403: 853-858. Naranjo-Garcia, E. 1997. Terrestrial gastropods from tropical rain- forest leaf litter, southern Veracruz, Mexico. Western Society of Malacologists, Annual Report 30: 40-46. Prado, D. E. 1995. Selva pedemontana: Contexto regional y lista floristica de un ecosistema en peligro. In: A. D. Brown and H. R. Grau, eds., Investigacién, Conservacion y Desarrollo en Selvas Subtropicales de Montafia. Laboratorio de Investiga- ciones Ecologicas de Las Yungas, Facultad de Ciencias Natu- rales e Instituto Miguel Lillo, Universidad Nacional de Tu- cuman, Horco Molle, Tucuman, Argentina. Pp. 19-52. Ramirez, R., S. Cordova, and K. Caro. 1999. Composicion y diver- sidad morfologica de la fauna malacologica terrestre de la region de Camisea (Cuzco, Pert). Im: C. Guisado, ed., Ab- stracts del IV Congreso Latinoamericano de Malacologia. Uni- versidad Catolica del Norte, Sede Coquimbo. P. 31. Ramirez, R., C. Paredes, and J. Arenas. 2003. Moluscos del Pert. In: Z. Barrientos and J. Monge-Najera, eds, Malacologia Latino- americana. Revista de Biologia Tropical 51: 225-284. Schilthuizen, M. and H. A. Rutjes. 2001. Land snail diversity in a 22° 1/2 * 2007 square kilometer of tropical rainforest in Sabah, Malaysian Borneo. Journal of Molluscan Studies 67: 417-423. Solem, A. 1984. Endodontid Land Snails from Pacific Islands (Mol- lusca: Pulmonata: Sigmurethra), Part Il: Families Punctidae and Charopidae, Zoogeography. Field Museum of Natural History, Chicago. Tattersfield, P., C. M. Warui, M. B. Seddon, and J. W. Kiringe. 2001. Land-snail faunas of afromontane forest of Mount Kenya, Kenya: Ecology, diversity and distribution patterns. Journal of Biogeography 28: 843-861. Valdora, E. E. and M. B. Soria. 1999. Arboles de Interés Forestal y Ornamental para el Noroeste Argentino. Laboratorio de Inves- tigaciones Ecologicas de las Yungas. Universidad Nacional de Tucuman Press, Tucuman, Argentina. Vellend, M. 2001. Do commonly used indices of {-diversity mea- sure species turnover? Journal of Vegetation Science 12: 545- 552. Wade, C., P. B. Mordan, and B. Clarke. 2001. A phylogeny of land snails (Gastropoda: Pulmonata). Proceedings of the Royal So- ciety of London 268: 413-422. Wheeler, Q. D. 2004. Taxonomic triage and the poverty of phy- logeny. Philosophical Transactions of the Royal Society of Lon- don 359: 571-583. Whittaker, R. H. 1975. Evolution of species diversities in land com- munities. Evolutionary Biology 10: 1-68. Accepted: | September 2006 Amer. Malac. Bull. 22: 27-73 Out of Australia: Belloliva (Neogastropoda: Olividae) in the Coral Sea and New Caledonia Yuri I. Kantor’ and Philippe Bouchet” "ALN. Severtsov Institute of Ecology and Evolution of Russian Academy of Sciences, Leninski prosp. 33, Moscow 119071, Russia, kantor@malaco-sevin.msk.ru * Muséum National d'Histoire Naturelle, 55 Rue Buffon, 75005 Paris, France, pbouchet@mnhn.fr Abstract: The genus Belloliva (Gastropoda: Olividae) consists of small (<15 mm) operculate species and was hitherto thought to be essentially confined to coastal waters of southern and eastern Australia. We report a small radiation from deep water (100-1000 m) in the Coral Sea and New Caledonia, consisting essentially of undescribed species. The new genus Calyptoliva, which differs from Belloliva by the absence of a mantle filament and the presence of a mantle lobe, is also represented in the same area by new species. Based on correlation with shell characters, we suggest that the olivid mantle lobe is responsible for secreting the primary spire callus that overlies the suture, rather than producing the columellar callus as was previously believed (Marcus and Marcus 1959). Belloliva and Calyptoliva combine a suite of shell, radular, and anatomical characters that is shared with either the Olivinae or the Ancillariinae. This raises the question of the distinctiveness of the two classically recognized subfamilies within the family Olividae. All species have paucispiral protoconchs with inferred limited larval dispersal, and many have extremely narrow distributions, sometimes endemic to a single guyot, or they show discrete geographical differentiation. New species: Belloliva iota sp. nov., Belloliva alaos sp. nov., Belloliva apoma sp. nov., Belloliva ellenae sp. nov., Belloliva obeon sp. nov., Belloliva dorcas sp. nov., Calyptoliva bolis gen. nov., sp. nov., Calyptoliva amblys sp. nov., Calyptoliva tatyanae sp. nov. Key words: Anatomy, classification, new species, endemism The neogastropod family Olividae stands out as a minor component in offshore and deep-water molluscan faunas. Only 15 species have been recorded worldwide from depths below 400 m, all belonging to subfamily Ancillariinae (Kan- tor and Bouchet 1999), with only four of these (belonging to genera Amalda, Ancilla, and Baryspira) reaching below 1000 m. Ongoing exploration of deep-water benthos in the South West Pacific confirms this unspectacular diversification of Olividae in the local molluscan fauna (Kilburn and Bouchet 1988, Bouchet and Kilburn 1991), although species of Amalda can be locally common on seamounts of the Norfolk Ridge (Y. Kantor and P. Bouchet pers. obs.). The discrete radiation of species from the Coral Sea and New Caledonia described in the present paper is thus remarkable because of its magnitude: 11 species, 8 of which reach or normally occur below 400 meters, and all of them undescribed. Six species are placed in the genus Belloliva Iredale in Peile, 1922, and three in the new genus Calyptoliva. The genus Belloliva Iredale in Peile, 1922 was established for Olivella brazieri Angas, 1877, and a second species from Australian coastal waters, Olivella pardalis A. Adams and Angas, 1864, was also originally included in the genus. Peile (1922: 18) highlighted that the two species “have a tricuspid rachidian, similar to that of Oliva but with minute additional cusp outside each of the lateral cusps;” this was the basis for the establishment of Belloliva. Since Peile (1922), no ana- tomical data nor additional data on radulae have been pub- lished, and the genus has remained little known. Based on radular morphology, Olsson (1956) allocated Belloliva to the subfamily Olivinae, whereas Wilson (1994) without any dis- cussion included the genus in the subfamily Olivellinae, an opinion that was followed by Tursch and Greifeneder (2001). Currently, Belloliva includes four Australian coastal species; a fifth one from the Caribbean has also controver- sially been referred to it (see Discussion below). In the present paper, we redefine Belloliva based on the anatomy of the Australian species, including the type species; we describe the Coral Sea and New Caledonian species at- tributable to it; we describe the new genus Calyptoliva that superficially resembles Belloliva; and finally we discuss the position of Belloliva in the family Olividae. MATERIAL AND METHODS The present paper is based on the extensive material collected by recent expeditions exploring the Coral Sea and New Caledonia area (Richer de Forges 1990, 1993, Richer de Forges and Chevillon 1996) and housed in Muséum national d'Histoire naturelle, Paris (MNHN). The material is not individually catalogued but is unambiguously referred to by the acronym of the cruise (e.g, MUSORSTOM 4, BATHUS 28 AMERICAN MALACOLOGICAL BULLETIN 1) and the station number. In lists of material examined, “lv” refers to live-taken specimens and “dd” to empty shells. The following standard shell measurements were made: shell length (SL); last whorl length (BWL); aperture length (AL); shell width (SW). For the purposes of species discrimi- nation, we used a number of protoconch and teleoconch measurements following those defined and described in de- tail by Tursch and Germain (1985, 1986), and demonstrated by us to be operational (Bouchet and Kantor 2004). The method used for counting protoconch whorls or measuring protoconch is usually not specified in the literature, making comparisons difficult. We counted protoconch whorls from the origin of the suture (Fig. 1B). The number of protoconch and teleoconch whorls was counted with an accuracy of 0.125 whorl. Because measurements taken from camera lucida draw- ings are more accurate than those made directly on the shell with the aid of an ocular micrometer, protoconchs were drawn in standard position, i.e., with the protoconch- teleoconch transition facing the observer (Fig. 1), and mea- surements were made from the drawings. Instead of proto- conch diameter, we used “D1” which equals protoconch diameter + 0.25 of the first teleoconch whorl. “PRE” repre- sents the exposed height of the protoconch and is referred to in the text as protoconch elevation. “TL1” is the diameter of the first 1.25 teleoconch whorls (this measurement can easily be done on the drawings of “standard” protoconch position). Measurements “TL2” and “TL3” are defined on Fig. 1A. Radulae were studied with scanning electron micros- copy (SEM). After being cleaned in diluted bleach, air-dried, Figure 1. Standard orientation of protoconch and measurements made. A, Shell in the standard position with protoconch-teleoconch transition marked by dotted line; measure- ments taken on protoconch: D1, PRE; measurements taken on teleoconch: TL1 to TL3 (see Material and Methods for description and references). B, Count of whorl number. 22 * 1/2 * 2007 mounted on glass slides, and coated with gold-palladium they were investigated with a JEOL JSM 840A Scanning Microscope. We studied histology of some organs of Belloliva (ante- rior foregut and mantle). The tissues were dehydrated and embedded in paraffin and serially sectioned at 10 um. Sec- tions were stained with Masson’s triple stain. The terminology of the shell base of oliviform gastro- pods has not yet been fully standardized and is sometimes “taxonomy-dependent” (e.g., “ancillid band” and “ancillid groove”). We follow Tursch and Greifeneder (2001), who discussed the terminology used in descriptions of olivid gen- era and suggested homologies. The shell base of Belloliva is rather simplified in comparison with other olivid genera. The anterior band (Fig. 2B) is usually covered with incon- spicuous axial striations. It is elevated over the surface of the remaining part of the last whorl (cloak, Fig. 2A) and delim- ited by a rather sharp step (sometimes referred to as “groove,” or “ancillid groove”: see Kilburn 1977). The plication plate (Fig. 2C) in turn is raised over both the cloak and the anterior band. It can be divided (by the posterior limit of the anterior band, or ancillid groove) into parietal plate and anterior plating. The parietal plate is a slightly thickened area, actually formed by the pari- etal callus. Its surface can differ from the cloak, being very finely shagreened. The anterior plating is sharply delimited from the anterior band on the ventral shell surface, but this border becomes more obscured on the lowest part of the shell. The plication plate can be smooth, but usually has several plicae on the anterior plating and a few on the pa- rietal plate. RESULTS Olividae Latreille, 1825 Belloliva Iredale in Peile, 1922 Type species Olivella braziert Angas, 1877 (original designation). Remarks Four Australian species have tra- ditionally been included in Belloliva (Kaicher 1987, Wilson 1994) (for more details see Discussion). Here we de- scribe in details the anatomy of the type species, Belloliva brazieri, and provide comparative remarks on the second studied Australian species, Bel- loliva leucozona. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 29 (za (___) body f— whorl cloak > band adapical limit of anterior band or "ancillid groove" anterior | j filament channel parietal / plate B plication anterior plating Figure 2. Terminology used in shell descriptions, with emphasis on the shell base. Belloliva brazieri (Angas, 1877) (Figs. 3, 4A-D, 32A-B) Olivella brazieri Angas 1877: 172, pl. 26, fig. 6. Type material Not traced. Type locality Newcastle Beach, New South Wales, Australia. Material examined Australia, New South Wales, 2 km E of Long Bay, Syd- ney, 33°58.8'S, 151°17.0'E, 66 m, AMS C388726. Dissected specimen male: SL 12.9, BWL 9.75, AL 7.4 mm, SW 4.8 mm. After dissection, the radula was removed and the anterior foregut was serially sectioned. Anatomy General morphology.—Body consisting of nearly 5 whorls, mantle cavity spanning about 0.5 whorl, nephridium more than 0.3 whorls, digestive gland about 1.25 whorls, testis about 3 upper whorls (Fig. 3A-B). Nephridium with transparent walls, excretory lamellae very distinct in poste- rior part and more closely spaced anteriorly, 15 in total. Ne- phridial gland narrow. Anterior lobe of digestive gland small, spanning about 0.3 whorls and separated from posterior lobe by the stomach, which is oriented obliquely with regard to columellar axis (Fig. 3A). Posterior lobe spanning slightly less than one whorl, occupying the entire width of the whorl immediately posterior to the stomach and more posteriorly has the shape of narrow band occupying the upper part of the whorl above testis. Foot thick, metapodium broadly tri- angular-oval, propodium small in comparison with crescent shaped propodium typical for Olividae, subdivided longitu- dinally on the ventral surface (Fig. 3C) and separated from metapodium by thin but distinct furrow on both the dorsal and ventral sides. Parapodia of medium size. Operculum completely transparent, yellowish, very thin, elongate-oval. Operculum attached along narrow oval area (about 0.5 of operculum width) to opercular pad. In its posterior third, part of the opercular pad is free from the dorsal side of the foot, forming a tongue-like projection. Head rather large, with two separate vertical flaps (Fig. 3C, tn) separated by a long, deep furrow. No eyes. Mantle cavity (Fig. 3F).—Mantle edge even and slightly thickened. Mantle thin, osphradium, ctenidium, and hypo- branchial gland visible by transparency. Siphon with thick and contracted walls, long, extending considerably beyond 30 AMERICAN MALACOLOGICAL BULLETIN — 22 1/2 * 2007 Figure 3. Anatomy of Belloliva brazieri (Angas, 1877). A, B, Dorsal and ventral views, respectively, of the body removed from the shell. C, Head-foot, antero-dorsal view, mantle removed. D, E, Right and left views, respectively, of the anterior foregut. F, Mantle complex. Abbreviations: ag, anal (rectal) gland; aldg, anterior lobe of digestive gland; bm, buccal mass; cme, cut mantle edge; ct, ctenidium; gL, gland of Leiblein; hg, hypobranchial gland; Isg, left salivary gland; mf, mantle filament; ne, nephridium; nr, circumesophageal nerve ring; odr, odontophoral retractor; oe, esophagus; op, operculum; os, osphradium; p, penis; par, parapodium; pldg, posterior lobe of digestive gland; pr, proboscis; prp, propodium; prr, proboscis retractor; re, rectum; rsg, right salivary gland; s, siphon; sl, sole of the foot; st, stomach; te, testis; tn, cephalic tentacles; vL, valve of Leiblein; vs, seminal vesicle. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 3] mantle edge. Ctenidium large, occupying about 0.8 of mantle length, consisting of simple triangular lamellae. Os- phradium as wide as ctenidium and 0.75 of its length, sym- metrical. Mantle filament (Fig. 3A, F, mf), rather short when contracted. Mantle lobe and anterior mantle tentacles ab- sent. Hypobranchial gland very distinct, narrow, brown, lacking folds. Rectum narrow, very thin in posterior third, gradually narrowing towards the anal opening. Rectal gland distinctly visible through the mantle wall as a narrow red sinuose line (Fig. 3A, ag), extending along most of rectum length. Alimentary system (Fig. 3D-E).—Rhynchostome asym- metrical, situated below the right head flap. Proboscis short in contracted state (1.4 mm, or 0.19 AL), occupying nearly the entire rhynchocoel length, rhynchodeum semitranspar- ent. Proboscis walls and rhynchodeum very thin, about 30 um (7% of proboscis diameter), covered by cuticularized cuboidal epithelium in histological section. Walls with a very thin outer layer of circular muscle fibers and an inner layer of longitudinal fibers, constituting about half of the wall thickness. Mouth opening very broad compared to proboscis diameter. Buccal tube long (about 0.3 of proboscis length) and broad, lined with thick cuticle, leading to buccal cavity. Radular diverticulum long and narrow, extending at least 0.5 of proboscis length. Odontophoral retractor large, flattened (Fig. 3D, odr), extending posteriorly from the proboscis, running anteriorly along ventral side of rhynchodeum and bypassing the nerve ring, following to the ventral side of cephalic hemocoel, its edges thickened, the muscle itself be- ing rather thin and transparent. Esophagus rather broad posterior to proboscis and forming a very short loop. Several very thin retractor muscles attached to the rhynchodeum (wall of the proboscis sheath) in its mid-length. Odonto- phore protruding significantly behind the proboscis edge. Radular sac nearly as long as odontophore. Radula (Fig. 4A-D) comprising about 80 teeth rows, membrane width about 120 um (0.93% SL, 1.62% AL). Rachidian tooth with 3 main cusps, central cusp about 1.5 times narrower and shorter than the lateral cusps, and an additional small but distinct cusp abutting each side of the main lateral cusps. Small and shallow depressions on dorsal side of main lateral cusps, corresponding to cusps of preceding row. Anterior profile of the rachidian slightly concave. Lateral sides of basal plate gradually embedded into the membrane without dis- tinct border. Lateral teeth (Fig. 4D) with subrectangular base and long, curved, hook-like cusp. Valve of Leiblein large, pyriform, well demarcated from the esophagus, with ciliary cone. Esophagus very narrow immediately posterior to the valve and passing through the nerve ring. Circumesophageal nerve ring comparatively very large, nearly as long as re- tracted proboscis. Posterior esophagus (posterior to the opening of the duct of the gland of Leiblein) significantly widening as approaching the stomach. Gland of Leiblein large, colorless in preserved condition, bulky anteriorly, and tapering posteriorly, opening into esophagus by very short and constricted duct close to the nerve ring. Gland of Leiblein with broad internal cavity, separated by tall and rather narrow folds. Salivary glands medium-sized, separate, loose, ramified-tubular, typical for Olivoidea (Fig. 3D-E, rsg, Isg). Wall of each tube composed of a single layer of large, irregularly angular, glandular cells with granulated cyto- plasm, at least some of them ciliated. Glands situated on both sides of posterior part of rhynchodeum and anterior esophagus and tightly attached to them by connective tissue fibers. Salivary ducts rather thick, passing into the tubules of the salivary glands without obvious border. Ducts fused with the walls of esophagus at posterior end of proboscis. Acces- sory salivary glands absent. Stomach badly damaged while extracting the body, but in general appearance similar to that of Belloliva leucozona (see below). Reproductive system.—Testis large, occupying 3 upper whorls, situated ventral to the digestive gland in the anterior part, contrary to other neogastropods. Seminal vesicle poorly differentiated from the testis, consisting of few loops and situated at the lower part of the whorl, in close prox- imity to the stomach. Within the mantle cavity, seminal duct and prostate gland forming several large tight loops and then extending to the base of the penis. Penis as long as the mantle cavity, of even diameter along its length, tip rounded, without seminal papilla. Belloliva leucozona (A. Adams and Angas, 1864) (Figs. 4E-H, 5, 32C-D) Olivella leucozona A. Adams and Angas 1864: 422, pl. 37, fig. 23. Type material Four syntypes BMNH 1870.10.26.93. Type locality Port Jackson, New South Wales, Australia, 6 fathoms (11 m). Material examined Australia, New South Wales, 2.3 km E of Malabar, 33°59.5'S, 151°16.8’E, 66-73 m, AMS C330373. Dissected specimen, male: SL 12.25, BWL 8.89, AL 7.13, SW 4.38; after dissection, mantle serially sectioned. Anatomy Body morphology (Fig. 5A-B) is similar to that of Bel- loliva brazieri, except that testis spans 3.5 whorls instead of 3 in B. brazieri. Nephridium has less pronounced excretory lamellae, 9 in total as seen through nephridial wall. Mantle cavity.—Mantle cavity in all details similar to that of Belloliva braziert. The main difference is the absence 32 AMERICAN MALACOLOGICAL BULLETIN — 22° 1/2 + 2007 hia ae Figure 4. Scanning electron micrographs of the radulae of Belloliva brazieri (A-D) and Belloliva leucozona (E-H). A, E, Dorsal view of the central part of the radular membrane. B, F, Enlarged rachidian teeth. C, G, Left lateral view of the rachidian teeth. D, H, Left lateral view of the lateral teeth. Scale bars = 10 um. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA Ow QW Figure 5. Anatomy of Bello- liva leucozona (A. Adams and Angas, 1864). A, B, Ventral and dorsal views, respectively, of the body removed from the shell. C, Head-foot, dorsal view, mantle and visceral mass removed. D, E, Left and right views, respectively, of the anterior foregut. F, Exter- nal view of the stomach and part of the visceral mass. G, Posterior whorl of the visceral mass, view from inside, whorls slightly uncoiled, ne- phridium removed. Abbre- viations: aldg, anterior lobe of digestive gland; aoe, anterior esophagus; att, attachment of the opercular pad to the oper- culum; bm, buccal mass; cm, columellar muscle; cme, cut mantle edge; ct, ctenidium; gL, gland of Leiblein; hg, hy- pobranchial gland; int, intes- tine; mf, mantle filament; ne, nephridium; nr, circum- esophageal nerve ring; odr, odontophoral retractor; op, operculum; os, osphradium; p, penis; par, parapodium; pldg, posterior lobe of diges- tive gland; poe, posterior esophagus; pr, proboscis; prp, propodium; rs, radular sac; s, siphon; sem.d, seminal duct; sg, salivary gland; sl, sole of the foot; st, stomach; te, testis; tn, cephalic tentacles; vL, valve of Leiblein; vs, seminal vesicle. of the rectal gland. Filament channel on upper shell whorls Alimentary system.—Anatomy of the alimentary system occluded with solid debris and obviously filament is not able _ is similar to Belloliva braziert. The minor differences include: to extend further than about last 2.5 whorls. Mantle filament shorter proboscis in contracted state (0.85 mm, or 0.12 AL) (serially sectioned) composed mostly of muscular elements, (Fig. 5D, E, pr); odontophore and radular sac strongly pro- both longitudinal and helical, with few connective tissue — truding posterior to the proboscis (Fig. 5E, bm, rs); radular cells. Innervation poor. sac slightly longer than odontophore; relatively larger and 34 AMERICAN MALACOLOGICAL BULLETIN more rounded valve of Leiblein (Fig. 5D, vL); position of the valve more anterior to the circumesophageal nerve ring; larger fused salivary glands (Fig. 5D, E, sg), situated poste- riorly to the retracted proboscis and surrounding anterior esophagus and valve of Leiblein; more coiled gland of Leiblein (Fig. 5D, E, gL). Radula (Fig. 4E-H) comprised of about 80 teeth rows, membrane width about 130 um (1.1% SL, 1.82% AL). An- terior profile of rachidian straight. Reproductive system. Seminal vesicle situated at the ven- tral border between gonad and posterior lobe of digestive gland, comprising a very thickened, wide, short vesicle, and the much narrower duct (Fig. 5G, vs). Seminal duct entering the mantle cavity where it forms numerous very wide loops i) in) ay i) * 2007 on the right side of the mantle cavity and continuing to the base of the penis (Fig. 5C, sem.d), nearly straight within penis. Belloliva alaos Kantor and Bouchet sp. nov. (Figs. 6, 7, 8A-C, 9) Type material Holotype (Moll 9469) and 4 paratypes (Moll 9470) in MNHN. Material examined North of New Caledonia. MUSORSTOM 4, st. DW156, 18°54'S, 163°19'E, 525 m (2 dd); st. DW159, 18°46’S, Figure 6. Belloliva alaos sp. nov. A-C, Holotype. D, E, Paratype (st. DW918, SL 12.1 mm). F, G, Paratype (st. DW918, SL 8.2 mm). H, St. DW916, SL 7.2 mm. All shells illustrated at the same scale. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 35 163°16’E, 585 m (11 dd); st. DW160, 18°42’S, 163°13’E, 668 m (5 dd, 1 lv [anatomy and radula]). BATHUS 4, st. DW914, 18°49'S, 163°15’E, 600-616 m (1 dd); st. DW916, 18°53’S, 163°20’E, 518-570 m (1 dd); st. DW917, 18°47’'S, 163°14’E, 397-400 m (27 dd); st. DW918, 18°49’'S, 163°16’E, 613-647 m (22 dd—holotype and 4 paratypes); st. DW919, 18°50'S, 163°17'E, 610-660 m (1 dd). Type locality North of New Caledonia, 18°49'S, 163°16’E, 613-647 m (BATHUS 4, st. DW918). Description (holotype) Shell solid, glossy, oval (BWL/SL = 0.76, AL/SL = 0.60, D/SL = 0.46), with moderately wide aperture and elevated, somewhat turreted spire, consisting of about 1.0 protoconch and almost 3 teleoconch whorls. Protoconch large, evenly rounded, diameter 1650 ttm, exposed height 1140 ym, smooth, protoconch-teleoconch transition distinctly marked by onset of filament channel. Profile of whorls evenly rounded, with very obtuse shoulder. Filament channel com- pletely open. Aperture lanceolate-oval, gradually narrowing abapically. Outer lip slightly convex, nearly straight in most adapical part, straight in middle portion and evenly rounded abapically. Parietal plate narrow, slightly thickened, anterior plating having one inconspicuous plica. Color uniformly white, upper teleoconch whorls and protoconch translucent. Dimensions (holotype): SL 10.8 mm, SW 5.0 mm, BWL 8.2 mm, AL 6.5 mm. Largest specimen (paratype): SL 12.1 mm, SW 5.3 mm, BWL 8.6 mm, AL 6.5 mm. Anatomy The anatomy of the single live-taken female (SL 11.7 mm, BWL 9.6 mm, AL 7.5 mm, SW 6.0 mm) has been studied (Fig. 7). Body in alcohol uniformly pale yellow, lack- ing pigmentation. General morphology.—Body consisting of nearly 4 whorls, mantle cavity spanning about 0.5 whorls, ne- phridium 0.3 whorls, digestive gland about 1 whorl. Ovary occupying upper 2 whorls of the visceral hump, its border with posterior digestive gland forming a wavy line across the whorl. Nephridium with transparent walls, with 8 main ex- cretory lamellae (Fig. 7B, ne). Nephridial gland narrow, with nearly smooth walls (Fig. 7B, ng). Anterior digestive gland small, spanning about 0.25 whorls and completely separated from posterior one by obliquely situated stomach (Fig. 7A, C). Foot thick, strongly contracted during fixation, folded transversely, metapodium broadly triangular-oval, propo- dium small in comparison to metapodium, typically cres- cent-shaped, subdivided longitudinally (Fig. 7C). Opercu- lum transparent, very thin, elongate, constricted in adapical part, and slightly thickened along low inner edge. Opercu- lum attached to opercular pad along long narrow area (less than 0.3 of operculum width) (Fig. 7B, att). About 1/5 of most posterior part of the pad detached from dorsal surface of the foot, forming a tongue-like extension. Head weakly distinguished from the body, with two separate small vertical flaps (Fig. 7C, tn). No eyes. Mantle cavity—Mantle edge even. Mantle thin, and os- phradium and ctenidium are seen through it. Siphon short, rather thin-walled, slightly extending beyond mantle edge, with smooth edges. Osphradium bipectinate, nearly sym- metrical, with very narrow axis, very broad, slightly exceed- ing the maximal width of large, deeply pendant ctenidium (Fig. 7G, os). The inner row of osphradial lamellae overhang the ctenidium; when viewing the mantle from the inside, the osphradial maximal width appears to be 1.5 the ctenidial width. The length of osphradium nearly equals the length of ctenidium. Ctenidium occupies nearly entire mantle length, formed of very tall triangular lamellae. Hypobranchial gland moderately glandular, although not forming distinct folds (Fig. 7G, hg). Mantle filament not long, with folded walls, indicating significant state of contraction. Posterior mantle tentacle and mantle lobe absent. Female pallial gonoduct large, swollen. Bursa copulatrix rather large, long, subcylin- drical. Female genital opening situated close to anus. Alimentary system.—Rhynchostome asymmetrical, situ- ated below the right tentacle. Proboscis short in contracted state (about 1.7 mm, or 0.3 AL) (Fig. 7D, E, pr), thin (about 0.5 mm in diameter), occupying nearly the entire rhyncho- coel length. Rhynchodeum (Fig. 7E, rnh) thin-walled, semi- transparent. Proboscis wall very thin, about 20 um (4% of the proboscis diameter), lined with low cuboidal epithelium (12 ttm), underlain by single layers of circular and longitu- dinal muscle fibers (8 um thick overall). Mouth opening rather wide. Muscular buccal tube, not less than 0.5 of pro- boscis length, leading from mouth to buccal cavity. Wall of buccal tube, in contrast to proboscis wall, thick, about 130 uum in total, lined with thick cuticle (20 ttm) and cubical epithelium (12 «tm), and underlain by very thick layer of circular muscles (about 90 um). Proboscis lumen filled with oval cells with slightly granular cytoplasm. Several very thin retractor muscles attached to median part of rhynchodeum (wall of the proboscis sheath) when proboscis is retracted. From posterior to the proboscis, esophagus rather narrow and forming a long loop when proboscis is retracted. Large odontophoral retractor leaving proboscis from the posterior, extending anteriorly along ventral side of rhynchodeum and bypassing the nerve ring, attached to ventral side of cephalic hemocoel (Fig. 7E, odr). Radula (Fig. 8A-C) about 145 um wide (1.24% SL, 1.93% AL), consisting of 70 rows of teeth. Rachidian with 3 main cusps, central cusp having the same width as the lateral and about 1.5 times shorter than the lateral cusps, and a secondary, very small, indistinct cusp on 36 AMERICAN MALACOLOGICAL BULLETIN — 22* 1/2 * 2007 Figure 7. Anatomy of Belloliva alaos sp. nov. A, B, Ventral and dorsal views, respectively, of the body removed from the shell. C, Head-foot, dorsal view, mantle and visceral mass removed. D, E, Right and left views, respectively, of the anterior foregut. F, External view of the stomach and part of the visceral mass. G, Mantle complex. Abbreviations: ag, anal (rectal) gland; aldg, anterior lobe of digestive gland; aoe, anterior esophagus; att, attachment of the opercular pad to the operculum; bc, bursa copulatrix; cm, columellar muscle; cme, cut mantle edge; ct, ctenidium; gL, gland of Leiblein; hg, hypobranchial gland; Isg, left salivary gland; mf, mantle filament; ne, nephridium; ng, nephridial gland; nr, circumesophageal nerve ring; odr, odontophoral retractor; op, operculum; os, osphradium; ov, ovary; par, parapo- dium; per, pericardium; pg, pallial gonoduct; pldg, posterior lobe of digestive gland; pma, posterior mixing area; poe, posterior esophagus; pr, proboscis; prp, propodium; prr, proboscis retractor; re, rectum; rnh, rhynchodeum (=proboscis sheath); rsg, right salivary gland; s, siphon; st, stomach; tn, cephalic tentacles; vL, valve of Leiblein. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 37 each side of the main lateral cusps; secondary cusps most visible on lateral view of the rachidians (Fig. 8C, indicated by black arrows). Rachidians rather widely spaced, cusps not abutting the next teeth. Anterior profile of the rachidian slightly convex, nearly straight. Lateral sides of the basal plate gradually embedded in the membrane without distinct border. Lateral teeth with subtriangular bases and long, curved, hook-like cusps. Valve of Leiblein large, pyriform, well distinguished from esophagus, which becomes very nar- row immediately after the valve and passes through the nerve ring. Circumesophageal nerve ring comparatively very large, with enlarged pedal and buccal ganglia. Posterior esophagus significantly widening posteriorly towards the stomach. Gland of Leiblein large, very light brown, tubular, and coiled, bulky anteriorly, opening into esophagus by very narrow constricted duct abutting the nerve ring posteriorly. Salivary glands medium-sized, ramified-tubular, left one slightly smaller than right, situated on either side of esopha- gus and nearly fused around valve of Leiblein, in retracted position of proboscis situated mostly on right side of rhyn- chodeum, rather loose in appearance with several blind tu- bules extending from the main part of the gland. Salivary ducts poorly differentiated from the glands, appearing like short extensions of the tubules. They enter the esophageal wall anterior to the valve of Leiblein and pass towards their openings the lateral folds of esophagus. Accessory salivary glands absent. Stomach large, with very long posterior mix- ing area (Fig. 7F, pma) that spans more than 0.5 whorl. Stomach anatomy not investigated due to poor fixation. Rectal gland a simple, blind, rather long tube (Fig. 7G, ag), colorless. Distribution North of New Caledonia, shells in 400-668 m, alive in 668 m. Remarks Belloliva alaos sp. nov. is conchologically most similar to Belloliva apoma sp. nov., and they could easily be mistaken as variations of one another unless their anatomy is exam- ined. However, B. alaos is distinguished by its significantly larger adult size, 12.1 versus 7.6 mm, and its slightly larger protoconch (Fig. 9). Anatomically, B. alaos is readily distin- guished by the presence of a large operculum and the ab- sence of eyes. The radulae also differ markedly in the shape of the rachidian: in B. alaos, the basal plate of the rachidian is shorter than in B. apoma and the anterior profile is nearly straight; in B. apoma, the basal plate is longer and the an- terior profile, which coincides with the anterior edge of the basal plate, is clearly convex. In addition, in B. apoma the central cusp is relatively much narrower and shorter than in B. alaos. Etymology From the Greek alaos, blind. Belloliva apoma Kantor and Bouchet sp.nov. (Pigs. 8D-F, 9, 10) Type material Holotype (Moll 9471) and 3 paratypes (Moll 9472) in MNHN. Material examined North of New Caledonia. BATHUS 4, st. DW923, 18°52’S, 163°24'E, 470-502 m (15 dd) (co-occurring with Belloliva exquisita and Belloliva simplex); st. DW929, 18°52'S, 163°23'E, 502-516 m (7 dd, 2 lv [holotype with dried soft parts and 3 paratypes]). LAGON, st. 475, 18°36'S, 163°11’E, 415-460 m (16 dd). MUSORSTOM 4, st. DW197, 18°51’S, 163°21"E,. 550: m (1 dd}. Type locality North of New Caledonia, 18°52'S, 163°23'E, 502-516 m (BATHUS 4, st. DW929),. Description (holotype) Shell solid, glossy, oval (BWL/SL = 0.80, AL/SL = 0.66, D/SL = 0.51), with moderately narrow aperture and el- evated, somewhat turreted spire, consisting of approximately 0.5 protoconch and 2.5 teleoconch whorls. Protoconch large, evenly rounded, diameter 1330 tm, exposed height 770 jim, smooth, protoconch-teleoconch transition distinctly marked by onset of filament channel. Whorls moderately convex, evenly rounded, poorly shouldered. Filament channel com- pletely open. Aperture lanceolate-oval, gradually narrowing adapically. Outer lip rather evenly convex in most adapical part, nearly straight in median part and evenly rounded abapically. Parietal plate narrow, very thin, anterior plating broadening in abapical part of aperture, appearing nearly smooth in front view, but several very weak oblique plicae visible when the shell is slightly rotated clockwise (Fig. 10D). Color uniformly off-white. Dimensions (holotype): SL 6.8 mm, SW 3.5 mm, BWL 5.4mm, AL 4.5 mm. Largest specimen (LAGON, st. 475): SL 7.6 mm, SW 3.9 mm, BWL 6.1 mm, AL 5.1 mm. Eyes large. Operculum absent. Radular width about 95 wm (1.45% SL, 2.16% AL), consisting of about 55 rows of teeth. Rachidian with 3 main cusps, central cusp more than twice narrower than lateral and about 1.5 times shorter than the lateral cusps, and a very small, indistinct secondary cusp on each side of main lateral cusps (Fig. 8E, F indicated by arrow). Rachidians rather widely spaced, cusps not abutting the next teeth. Anterior profile of the rachidian clearly convex and coinciding with the anterior edge of the basal plate of the tooth. Lateral sides 38 AMERICAN MALACOLOGICAL BULLETIN = 22° 1/2 + 2007 Figure 8. Scanning electron micrographs of the radulae of Belloliva alaos sp. nov. (A-C) (MUSORSTOM 4, st. DW160) and Belloliva apoma sp. nov. (holotype) (D-F) (BATHUS 4, st. DW929). A, D, Dorsal view of the central portion of the radular ribbon. B, E, Enlarged rachidian tooth. C, F, Left lateral view of the radular ribbon. Arrows on C, E, F indicate secondary cusps of the rachidians. Scale bars = 50 um (A, D), 10 um (B, C, E, F). of the basal plate gradually embedded in the membrane without distinct border. Lateral teeth are with subtriangular bases and long, curved, hook-like cusps. Distribution North of New Caledonia, shells in 460-550 m, alive in 502-516 m. Remarks Specimens of Belloliva apoma may be off-white or may have very faint yellow broad spiral bands (one subsutural and one on shell base above periphery), and/or inconspicu- ous yellowish-brown spots on the rim bordering the fila- ment channel, and/or also inconspicuous zigzag axial lines on last whorl (Fig. 10H-I). BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 39 2.0 = Belloliva alaos sp. nov. © Belloliva apoma sp. nov. Vee 1 06 0.9 PRE, mm Figure 9. Morphometric comparison of protoconch dimensions of Belloliva alaos sp. nov. and Belloliva apoma sp. nov. For comparison with Belloliva alaos sp. nov., see under that species. Etymology From the Greek poma, operculum, and prefix a-, with- out; used as a noun in apposition. Belloliva simplex (Pease, 1868) (Figs. 11, 12A-D) Olivella (Callianax) simplex Pease 1868: 281-282, pl. 23, fig. 24. Type material Lectotype designated by Johnson (1994: 24), Academy of Natural Sciences of Philadelphia, ANSP 28969 (Fig. 11A-C). Material examined North of New Caledonia. BATHUS 4, st. DW923, 18°52'S, 163°24’E, 470-502 m (1 dd) (co-occurring with Belloliva exquisita and Belloliva apoma). East coast. LAGON, st. 830, 20°49’S, 165°19’E, 105-110 m (37 dd). BATHUS 1, st. DW692, 20°35’S, 164°59’E, 140-150 m (1 dd). West coast. BATHUS 4, st. DW887, 21°07'S, 164°28'E, 320-344 m (3 dd) (co-occurring with B. exquisita). EXPEDITION MON- TROUZIER, st. 1255, 20°43'S, 165°08’E, 11 m (3 lv); st. 1259, 20°44.6'S, 165°13.7’E, 15-35 m (2 lv, 4 dd); st. 1260, 20°44’S, 165°14’E, 49-59 m (1 dd); st. 1261, 20°46'-20°47’S, 164°15'-164°16.5'E, 45-65 m (1 lv); st. 1269, 20°35.1'S, 165°08’E, 15-20 m (22 lv, 6 dd); st. 1271, 20°52.7’S, 165°19.5'E, 5-25 m (3 dd); st. 1272, 20°49.5'S, 165°19.6’E, 10 m (10 lv); st. 1273, 20°50.4'S, 165°22.8'E, 20 m (4 dd, 10 lv); st. 1275, 20°49'S, 165°17’E, 50-62 m (2 dd); st. 1311, 20°40.4'S, 164°14.9'E, 10-60 m (9 lv) (co-occurring with B. exquisita); st. 1312, 20°40.4’'S, 164°14.9’E, 26-40 m (2 dd, 1 lv) (co-occurring with B. exquisita); st. 1316, 20°40'S, 164°11.2’E, 12 m (1 ly, 1 dd); st. 1318, 20°41.4’S, 164°14.8’E, 20-30 m (17 lv [radula examined]); st. 1319, 20°44.7'S, 164°15.5'E, 15-20 m (4 lv); st. 1322, 20°45.2'S, 164°15.2’E, 53-71 m (1 dd) (co-occurring with B. exquisita); st. 1331, 20°40.6'S, 164°12.1'E, 55-57 m (4 dd) (co-occurring with B. exquisita). Loyalty Islands, Lifou. LIFOU 2000, st. 1423, 20°54.0'S, 167°07:3' E,. 12 ma (2-day st. 1432; 20°53.5'S,' 167°02.7'E; 12-32 m (7 dd); st. 1434, 20°52.5’S, 167°08.1'E, 5-20 m (13 dd); st. 1435, 20°55.2’S, 167°00.7'E, 5-30 m (3 dd); st. 1436, 20°55.5'S, 167°04.2’E, 10-20 m (14 dd); st. 1441, 20°46.4’S, 167°02.0’E, 20 m (1 lv, 5 dd); st. 1442, 20°46.4’S, 167°02.0’E, 47 m (1 dd); st. 1443, 20°53.8'S, 167°07.3'E, 48-52 m (3 dd); st. 1454, 20°56.65’S, 167°02.0'E, 15-18 m (1 lv); st. 1456, 20°49.3'S, 167°10.4’E, 25-30 m (7 dd); st. 1469, 20°54.2’S, 167°00.4’E, 70-130 m (1 dd). Type locality Paumotus Islands [Tuamotu Archipelago], French Polynesia. Description Shell very small, fragile, semitransparent, glossy, oval, with moderately wide aperture and rather low spire, con- sisting of about 0.75 protoconch and 1.75 teleoconch whorls. Protoconch large in comparison with the teleoconch, evenly rounded, diameter around 1000 ttm, smooth, protoconch- teleoconch transition distinctly marked by onset of filament channel. Profile of whorls evenly rounded, last whorl weakly shouldered. Filament channel completely open. Aperture lanceolate-oval, gradually narrowing abapically. Outer lip slightly thickened, almost straight adapically, evenly rounded abapically. Parietal plate narrow, slightly thickened, anterior plating much thicker, broadening on abapical part of aper- ture, without plicae, clearly concave in profile. Color uni- formly off-white. Dimensions: The lectotype, SL 4.2 mm, seems to be the largest specimen. The largest specimen at our disposal (LAGON, st. 830) has SL 3.8 mm, SW 1.9 mm, BWL 3.1 mm, AL 2.5 mm. The morphology of one rehydrated female specimen from Koumac, New Caledonia (EXPEDITION MON- TROUZIER, st. 1318, SL 4.1, AL 2.5, BWL 3.2, SW 2.0 mm) was examined. Outer morphology similar to other studied species. Eyes large, mantle filament comparatively very short and thick, probably due to fixation. Radula (Fig. 12A-D) about 65 um wide (1.59% SL, 2.6% AL), consisting of 85 rows of teeth, including 5-6 nascent; width of rachidian ap- proximately 25 jum (38% of radular width). Rachidian nar- rowly spaced, cusps strongly abutting the next tooth, with 40 AMERICAN MALACOLOGICAL BULLETIN i) tO oa ie) ° 2007 Figure 10. Belloliva apoma sp. nov. A-D, Holotype. E, Paratype (SL 7.0 mm). F, Paratype (SL 6.5 mm). G, LAGON, st. 475, SL 7.6 mm. H, I, LAGON, st. 475, SL 7.4 mm. All shells illustrated at the same scale except D. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 4] Figure 11. Belloliva simplex (Peace, 187X). A-C, Lectotype, ANSP 28969, SL 4.2 mm (courtesy of P. Callomon, ANSP). D-G, New Caledonia, east coast, LAGON, st. 830, 105-110 m. D, (SL 3.6 mm). E, (SL 3.5 mm). F, (SL 3.6 mm). G, (SL 3.5 mm). H, Lifou, st. 1456 m, 25-30 m 2 (SL 3.3 mm). I, New Caledonia, west coast, Expedition Montrouzier, st. 1273, 20 m (SL 3.2 mm). All shells illustrated at the same scale. short lateral flaps, anterior edge straight, lateral sides of basal indistinct cusp on each side of the main lateral cusps; sec- plate gradually embedded in membrane without distinct ondary cusps most distinct in lateral view. Lateral teeth with border; 3 main cusps, central cusp narrower and nearly twice —_ broadened subtriangular bases and long, curved, hook-like as short as the lateral cusps, and a secondary, very small, cusps, and 2-3 very distinct denticles at the base of the cusp. 42 AMERICAN MALACOLOGICAL BULLETIN — 22° 1/2 + 2007 Figure 12. Scanning electron micrographs of the radulae of Belloliva simplex (A-D), New Caledonia, west coast, MONTROUZIER, st. 1318, 20-30 m (SL 4.1 mm), and Belloliva iota sp. nov. (E-F), Coral Sea, Lansdowne Bank, EBISCO, st. DW2631, 372-404 m (SL 7.2 mm). A, E, Dorsal view of the central part of the radular membrane. B, F, Enlarged rachidian teeth. C, D, Left lateral view of the rachidian teeth. Scale bars = 10 pm. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 43 Distribution Earlier known from the Tuamotu Islands, Tonga (Thiele 1929 in 1929-1931), and Western Samoa (Thiele 1929 in 1929-1931; 2 specimens from “Upolu” Samoa—Museum fiir Naturkunde, Humboldt University Berlin, ZMB 18.304; M. Glaubrecht pers. comm.), now recorded from the Loyalty Islands (Lifou) and northern and western New Caledonia, live in 10-45 m, shells down to 110-470 m. Remarks The original description is very brief and the accompa- nying illustration is uninformative. However, the identity of the species is established by the name-bearing type, and is confirmed by our examination of topotypical material from Anaa, Tuamotu Islands, in the collection of Jean Tréndle (La Force, France). Thiele’s illustration (1929 in 1929-1931: fig. 384) of a specimen from western Samoa agrees with this material. Belloliva simplex differs from its congeners by its very small adult size. It superficially resembles juveniles of Bello- liva exquisita, with which it co-occurs at several stations, and from which it is readily distinguished by its smaller proto- conch (1000 jum versus 1240-1580 jum in B. exquisita) and by its smooth, arcuate columella. For comparison with Belloliva iota sp. nov., see under that species. Belloliva iota Kantor and Bouchet sp. nov. (Figs. 12E-B, 13) Type material Holotype (Moll 9473) and 4 paratypes (Moll 9474) in MNHN. Material examined Coral Sea, Lansdowne Bank, MUSORSTOM 5, st. 388, 20°45'S, 160°54’E, 500-510 m (2 dd). EBISCO, st. DW2618, 20°06'S, 160°23’E, 280-304 m (2 dd; co-occurring with Belloliva obeon sp. nov.); st. DW2631, 21°03'S, 160°44’E, 372-404 m (4 lv); st. DW2639, 20°47'S, 161°O1'E, 289-294 m (25 dd). Type locality Coral Sea, Lansdowne Bank, 20°47'S, 161°O1'E, 289-294 m (EBISCO, st. DW2639). Description (holotype) Shell solid, glossy, elongate-oval (BWL/SL = 0.73, AL/SL = 0.59, D/SL = 0.46), with narrow aperture and elevated, somewhat turreted spire, consisting of approximately 0.5 protoconch and nearly 4 teleoconch whorls. Protoconch small, evenly rounded, diameter 820 um, exposed height 580 yum, smooth, protoconch-teleoconch transition distinctly marked by onset of filament channel. Whorls moderately convex, evenly rounded, shoulder not pronounced. Filament channel completely open. Aperture lanceolate-oval, gradu- ally narrowing adapically. Outer lip thickened, evenly convex in most of adapical part, nearly straight in median part and evenly rounded abapically. Parietal plate narrow, thin, ante- rior plating broadening in abapical part of aperture and having 4 plicae, adapicalmost being very weak. Color uni- formly off-white. Dimensions (holotype): SL 5.6 mm, SW 2.6 mm, BWL 4.1 mm, AL 3.3 mm. Largest specimen (MUSORSTOM 5, st. 388): SL 7.6 mm, SW 3.6 mm, BWL 5.5 mm, AL 4.6 mm. One male specimen (EBISCO, st. DW2631, SL 7.2 mm, AL 4.4 mm; Fig. 13G) was dissected. General morphology similar to other studied congeners. Cephalic flaps with rela- tively large eyes. Penis long, exceeding the length of the mantle cavity, of even diameter along its length and obtuse at the tip. Gland of Leiblein narrow, tubular, slightly coiled anteriorly and nearly straight posteriorly, grey, opening into esophagus without constricted duct. Radula (Fig. 12E-F) about 85 wm wide (1.18% SL, 1.93% AL), consisting of 60 rows of teeth, of which 12-13 nascent; width of rachidian approximately 23 jum (27% of radular width). Rachidians narrowly spaced, cusps abutting the next tooth, anterior edge straight, lateral sides of basal plate gradually embedded in membrane without distinct border. Rachidian with short lateral flaps, 3 main cusps, central cusp narrower and 1.5 times shorter than lateral cusps, and a secondary, very small, indistinct cusp on each side of the main lateral cusps. Lateral teeth with broadened subtriangular bases and long, curved, hook-like cusps. Distribution Coral Sea, Lansdowne Bank, alive in 372-404 m, shells in 294-500 m. Remarks Belloliva iota varies only little in the degree of develop- ment of the plicae on anterior plating, which are nevertheless never very conspicuous. One of the specimens from the type locality has 3 extremely faint yellow axial zigzag lines. Bel- loliva iota sp. nov. differs from most congeners by its small adult size. It superficially resembles juveniles of Belloliva exquisita, from which it is readily distinguished by its smaller protoconch (800 ym versus 1240-1580 fm in B. exquisita). Belloliva iota sp. nov. differs from Belloliva simplex by its larger shell with slightly smaller protoconch, by the plicae on the anterior plating, and by its relatively smaller radula, con- sisting of smaller number of rows, with the rachidian teeth having nearly subrectangular lateral flaps (Fig. 12F) versus subtriangular ones in B. simplex (Fig. 12B), and in the ab- sence of denticles at the bases of the cusps on the lateral teeth. 44 AMERICAN MALACOLOGICAL BULLETIN 22° 1/2 * 2007 Figure 13. Belloliva iota sp. nov. A-D, Holotype. D, Detail of columellar region. E, Paratype, SL 5.4 mm. F, Paratype, SL 5.5 mm. G, EBISCO, st. DW2631, SL 7.1 mm. H, MUSORSTOM 5, st. 388, SL 7.6 mm. All shells illustrated at the same scale except D. Etymology Material examined From the Greek iota: very small; used as a noun in Coral Sea, Chesterfield plateau: MUSORSTOM 5, st. apposition. 339, 19°53’S, 158°38’E, 380-395 m (2 dd); st. 361, 19°53'S, 158°38'E, 400 m (4 dd, 4 lv); st. 362, 19°53’S, 158°40’E, 410 m (5 dd); st. 379, 19°53’S, 158°40’E, 370-400 m (3 dd, 2 lv [{holotype]). EBISCO, st. DW2596, 19°43'S, 158°37’E, Type material 382-386 m (1 lv); st. DW2606, 19°36’S, 158°42’E, 442-443 m Holotype (Moll 9475) in MNHN. (3 lv, 3 dd); st. DW2607, 19°33'S, 158°40’E, 400-413 m Belloliva ellenae Kantor and Bouchet sp. nov. (Figs. 14, 15, 16A-D) BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 45 1, SL 8.4 mm (E-F) and 5 mm. J-L, EBISCO, st. Figure 14. Belloliva ellenae sp. nov. A-D, Holotype. D, Detail of columellar region. E-H, MUSORSTOM 5, st. 36 8.6 mm (G-H). I, EBISCO, st. DW2596, intermediate between the “axially striped” and “pale” morphs, SL 6. DW2610, “pale” morph, SL 8.8 mm (J) and 8.5 mm (K-L). All shells illustrated at the same scale except D. 46 AMERICAN MALACOLOGICAL BULLETIN (13 lv); st. DW2610, 19°34'S, 158°41’E, 486-494 m (39 ly, 13 dd). Type locality Coral Sea, Chesterfield plateau, 19°53’S, 158°40’E, 370- 400 m (MUSORSTOM 5, st. 379). Description (holotype) Shell solid, glossy, oval-fusiform (BWL/SL = 0.80, AL/ SL = 0.65, D/SL = 0.49), with moderately wide aperture and elevated spire, width 49% of height, consisting of about 1.0 protoconch and 4.0 teleoconch whorls. Protoconch rather large, evenly rounded, diameter 1170 um, exposed height 920 um, smooth, protoconch-teleoconch transition dis- tinctly marked by onset of filament channel. Profile of whorls nearly straight, very slightly concave below suture, evenly rounded below inconspicuous shoulder. Filament channel completely open. Aperture lanceolate, gradually narrowing towards its tip. Outer lip nearly straight in most adapical 0.3, evenly rounded in lower part. Parietal plate narrow and thin, anterior plating broadening in lower part of aperture and having 7 poorly developed plicae. Back- ground color light yellow. Last whorl and last half of pen- ultimate whorl with distinct, closely spaced, slightly wavy darker yellow axial color lines (27 on last whorl). Near adapical margin of anterior band, lines distinctly opistho- cline and their coaslescence forming a distinct color band. First half of penultimate whorl with gradually fading axial 2.0 «9 pale form » striated form © intermediate specimen 2.0 TL1I, mm 2.5 3:0 D1, mm 22° 1/2 * 2007 lines, early teleoconch whorls with only faint and irregularly spaced spots. Anterior plating with distinct elongated brown spot. Dimensions (holotype): SL 8.2 mm, SW 4.0 mm, BWL 6.6 mm, AL 5.3 mm. Largest specimen (st. 361): SL 8.8 mm, SW 4.6 mm, BWL 7.2 mm, AL 6.2 mm. Distribution Coral Sea, Chesterfield plateau, alive in 386-486 m. Remarks Two distinct forms of Belloliva ellenae sp. nov. can be recognized. The “axially striped” form (Fig. 14A-H) has 26- 46 colored axial lines on the last whorl and a dark spot on the anterior plating. The plicae on the anterior plating vary from nearly completely absent to moderately strong. The “pale” form is similar in shell outline to the typical form, but differs in color: the background is ivory and instead of axial lines there are two narrow light-brown color bands, one at the adapical limit of the anterior band, the other one on the rim of the filament channel; there is no columellar spot. Specimens of this form also differ by their somewhat larger protoconch and first teleoconch whorl (average D1 = 1.51 mim, range 1.45-1.56 mm in the “pale” form versus average D1 = 1.25 mm, range 1.17-1.29 mm in the “axially striped” form). The radulae (Fig. 16) and gross morphology are nearly identical in the two forms. The radula of a specimen of the “axially striped” form (MUSORSTOM 5, st. DW361, 2.0 1.5 1.0 0.6 0.8 1.0 1.2 PRE, mm Figure 15. Comparison of “pale” and “axially striped” forms of Belloliva ellenae sp. nov. Ordinate: protoconch diameter (D1, mm); abscissa: (left) diameter of first teleoconch whorl (TL1) and (right) protoconch elevation (PRE). BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 47 oa | aA Figure 16. Scanning electron micrographs of the radulae of Belloliva ellenae sp. nov. (A-D) and Belloliva dorcas sp. nov. (E-F). A-B, “Axially striated” form (MUSORSTOM 5, st. DW361). C-D, “pale” form (EBISCO, st. DW2607). A, C, E, Dorsal view of the central part of the radular membrane, scale bars = 50 pm. B, D, F, Enlarged rachidian teeth, scale bars = 10 um. 48 AMERICAN MALACOLOGICAL BULLETIN SL 8.6, AL 5.6 mm, male) was about 110 ym wide (1.27% of SL, 1.96% of AL) (Fig. 16A-B), consisting of 65 rows, of which 20 are nascent. Rachidians narrowly spaced, cusps abutting the previous tooth (Fig. 16B), anterior edge very slightly concave in its middle part and rounded at the edges. Lateral sides of basal plate gradually embedded into the membrane without distinct border. Rachidian with 3 broadly spaced main cusps, central cusp about 1.4 times shorter and much narrower than the lateral cusps, and one small, narrow, but distinct, additional cusp on each side of the main lateral cusps. Lateral teeth with subtriangular bases and long curved hook-like cusps bearing 1-2 small distinct denticles at their bases. The radula of the “pale” form (EBISCO, st. DW2607, SL 8.1, AL 5.4 mm, female) (Fig. 16C-D), was also about 110 «um broad (1.25% of SL, 1.96% of AL), con- sisting of 73 rows, of which 20 rows are nascent. Tooth shape was very similar to that of the “axially striped” form. Both forms of Belloliva ellenae sp. nov. occur in a very limited area: the “axially striped” form was found at four stations that straddle only 3 km while the “pale” form was found at 3 stations spanning about 6 km, slightly to the north of the type locality. These two groups of stations are separated by 30 km. This distribution does not appear to be merely a sampling artifact, as hauls made during the EBI- SCO cruise at appropriate depths between the two areas were negative for Belloliva ellenae sp. nov., with the excep- tion of a single specimen from a station (EBISCO, st. DW2596) situated right in the middle of the two clusters of stations. This specimen (Fig. 141) is somewhat intermediate in coloration (axial lines are present but are very pale and the columellar spot is absent) and the dimensions of its proto- conch and first teleoconch whorl are also intermediate (Fig. 15). This intermediate specimen from an intermediate lo- cality is further evidence that the two forms are conspecific. Belloliva ellenae sp. nov. is sympatric with Belloliva obeon sp. nov., Belloliva dorcas sp. nov., and Belloliva ex- quisita. It is readily distinguished from these, and from other congeners, by the combination of small adult size and color pattern (either of axial stripes and brown spot on anterior plating or of two narrow bands over ivory background). Etymology The species is named after our colleague Dr. Ellen E. Strong, curator at the National Museum of Natural History, Smithsonian Institution, Washington, D.C., and companion of the two authors during several field seasons. Belloliva obeon Kantor and Bouchet sp. nov. (Figs. 17,. 18, 19, 20) Type material Holotype (Moll 9476) and 2 paratypes (Moll 9477) in MNHN. 22° 1/2 * 2007 Material examined Coral Sea, Chesterfield Plateau, MUSORSTOM 5, st. 346, 19°40'S, 158°27’E, 245-252 m (3 dd [holotype and paratypes]); st. 388, 20°45’S, 160°54’E, 500-510 m (3 ly). EBISCO, st. DW2608, 19°33’S, 158°40'E, 393-396 m (5 dd, including 2 striped, co-occurring with Belloliva dorcas sp. nov.). Lansdowne Bank. CORAIL 2, st. DE16, 20°48’S, 160°56’E, 500 m (12 dd, 2 Iv). EBISCO, st. DW2617, 20°06'S, 160°22’E, 427-505 m (8 dd, 1 lv striped); st. DW 2618, 20°06'S, 160°23’E, 280-304 m (2 dd, striped; co- occurring with Belloliva iota sp. nov.); st. DW2619, 20°06’S, 160°23’E, 490-550 m (4 ly, striped}s-st. DW2625, 20°04.8’S, 160°20'E, 627-741 m (1 ly, striped); st. DW2629, 21°06’S, 160°46'E, 569-583 m (1 lv, 30 dd, striped). Type locality Coral Sea, Chesterfield Plateau, 19°40'S, 158°27’E, 245- 252 m (MUSORSTOM 5, st. 346). Description (holotype) Shell medium-sized, solid, semitransparent in the cen- tral part of the last whorl, glossy, broadly oval (BWL/SL = 0.86, AL/SL = 0.70, D/SL = 0.52), with wide aperture and low spire, consisting of about 0.5 protoconch and 3.125 teleoconch whorls. Protoconch large, evenly rounded, diam- eter 1770 tm, exposed height 720 tm, smooth, protoconch- teleoconch transition distinctly marked by onset of filament channel. Profile of whorls very slightly concave subsuturally, with rather distinct rounded shoulder. Filament channel completely open. Aperture elongate-oval, gradually narrow- ing and rounded abapically. Outer lip slightly thickened, the edge itself sharp, slightly concave in most adapical part, nearly straight along most of its length and evenly rounded abapically. Parietal plait very narrow, hardly thickened, an- terior plating clearly concave in profile, broadened, bearing ten pronounced plicae, diminishing in size adapically. Color uniformly off-white. Dimensions (holotype largest specimen): SL 14.3 mm, SW 10.1 mm, BWL 12.4 mm, AL 10.1 mm. Anatomy The anatomy of two specimens (MUSORSTOM 5, st. 388, Chesterfield Plateau, SL 13.8, BWL, 12.2, AL 9.8, SW 7.3 mm; EBISCO, st. DW2619, Lansdowne Bank, SL 9.0, BWL 7.9, AL 6.3, SW 5.0 mm) was examined. General morphology.—The body of the first one was badly torn during extraction, but the external morphology is similar to that of Belloliva alaos sp. nov. Foot thick, strongly contracted during fixation, propodium bent ventrally. Meta- podium very broad, triangular-oval. Propodium crescent- BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 49 Figure 17. Belloliva obeon sp. nov. A-D, Holotype. E, Paratype, SL 12.3 mm. F, Paratype, SL 12.6 mm. G-I, CORAIL2, st. DE16, SL 12.0 mm (G), 11.8 mm (H), and 11.3 mm (1). J-N, EBISCO, st. DW2629: “typical,” SL 9.6 mm (J-K), transitional, SL 10.1 mm (L), and slender specimens, SL 10.2 mm (M) and 10.4 mm (N). All shells illustrated at the same scale except D. 50 AMERICAN MALACOLOGICAL BULLETIN — 22° 1/2 + 2007 Figure 18. Scanning electron micrographs of the radula of Belloliva obeon sp. nov. [A-C, Coral Sea, Chesterfield Plateau, MUSORSTOM 5, 500-510 m, st. 388 (SL 13.8 mm); D-F, Coral Sea, Lansdowne Bank, EBISCO, st. DW2619 (SL 9.0)]. A, D, Dorsal views of the central portion of the radular ribbon. B, E, Left lateral view of the radular ribbon. C, Dorsal view of the bending plane of the radular ribbon. E, Dorsal view of enlarged central teeth. Arrow on B indicates additional cusp on the rachidian. Scale bars = 50 pm (A, C, D), 10 wm (B, E; F). BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 5] shaped, subdivided longitudinally by deep furrow and de- limited from metapodium by dorsal and ventral grooves. Parapodia rather short in contracted state. Operculum very large, extremely thin and transparent, attached to opercular disc by narrow oval zone along its left side (about 0.5 of operculum width) (Fig. 19A). About 1/5 of posteriormost part of pad detached from dorsal surface of foot, forming a tongue-like extension. Head well set off from the foot (Fig. 19A), with broadly separated, laterally compressed flaps with large eyes. Columellar muscle thick, splitting into 3 branches in its posterior part. Mantle cavity.—Mantle filament short. Anterior mantle tentacle and mantle lobe absent. Siphon long, narrow, and extending substantially beyond the evenly and slightly thick- ened mantle edge. Mantle itself very thin. Size and shape of osphradium and ctenidium very similar to those of Belloliva alaos sp. nov. Alimentary system.—Organs of anterior foregut strongly contracted during fixation (Fig. 19B). Proboscis short with smooth walls. Salivary glands large, seemingly compact, fused around valve and posteriormost part of rhynchodeum, their structure ramified tubular, similar to that of Belloliva Figure 19. Some details of the morphology of Belloliva obeon sp. nov. A, Head-foot, dorsal view, mantle and visceral mass removed. B, Right view of the anterior foregut. C, External view of the stomach and part of the visceral mass. Abbreviations: att, attachment of the opercular ad to the operculum; cm, columellar muscle; dgd, opening of the digestive gland into stomach; ¢L, gland of Leiblein; nr, circumesophageal § fe) Oo c o fe) fo) co nerve ring; odr, odontophoral retractor; op, operculum; opl, opercular lobe; p, penis; par, parapodium; pldg, posterior lobe of digestive gland; poe, posterior esophagus; pr, proboscis; prp, propodium; rnh, rhynchodeum ( = proboscis sheath); rsg, right salivary gland; sem.d, seminal duct; st, stomach; tn, cephalic tentacle; vL, valve of Leiblein. on i) brazieri. Accessory salivary glands absent. Gland of Leiblein large, massive, not coiled, dark brownish-grey, opening by narrow duct into esophagus. Large odontophoral retractor muscle passing ventrally under very thin-walled rhyncho- deum. Valve of Leiblein large, pyriform. Odontophore large, about 2/3 of proboscis length, deeply withdrawn, as in Amalda (Kantor 1991). Radular diverticulum strongly cu- ticularized. Radula about 170 um wide (1.23% of SL, 1.73% of AL), consisting of 52 rows of teeth. Rachidian about 70 um wide (41% of radular width) with 3 main cusps, central cusp about 1.6 times shorter and narrower than the lateral cusps, and one very small, indistinct additional cusp on each side of the main lateral cusps, best seen in lateral view (Fig. 18B, indicated by arrow). Rachidians rather narrowly spaced, cusps strongly bent in profile, the tips of the lateral cusps resting on the following teeth (Fig. 18B). Anterior profile of the rachidian nearly straight in middle part and rounded at the edges, coinciding with anterior edge of basal plate. Lateral sides of basal plate gradually embedded into the membrane without distinct border. Lateral teeth with subtriangular bases and long curved hook-like cusps. Only part of stomach retrieved, characterized by very long and very narrow posterior mixing area (Fig. 19C). Reproductive system.—Penis very long, flattened, simple, tapering towards the tip. Seminal papilla absent. The specimen from Lansdowne is similar in outer mor- phology. The stomach is slightly larger, with an even longer posterior mixing area. Proboscis length about 2 mm. Radula slightly more than half of proboscis length, about 120 jm wide (1.33% of SL, 1.90% of AL), consisting of about 65 rows. Rachidian about 50 um wide (42% of radular width) with 3 main cusps, central cusp about 1.6 times shorter and narrower than the lateral cusps, and one small, but distinct additional cusp on each side of the main lateral cusps. Ra- chidians rather narrowly spaced, cusps abutting the previous tooth, but tips not resting on it (Fig. I8E) as in the specimen from Chesterfield Plateau (Fig. 18B). Distribution Coral Sea, Chesterfield Plateau and Lansdowne Bank (Fig. 20), alive in 500-627 m, shells from 252 m. Remarks Belloliva obeon sp. nov. is rather variable in terms of shell shape and coloration. The specimens from Chesterfield Plateau are mostly pure off-white (with the exception of one specimen from EBISCO st. DW2608 that has broad, light yellow, nearly axial stripes on the last part of the last whorl) and their shell shape is overall similar to the type material. On Lansdowne Bank, the species is more variable, especially in terms of shell shape. Some specimens are extremely simi- AMERICAN MALACOLOGICAL BULLETIN 22° 1/2 * 2007 lar to those from the Chesterfields, but most have moder- ately to strongly developed axial color stripes, sometimes extending over the whole shell, while the background color may differ from off white to light yellow. Slender specimens (e.g., Fig. 17N) resemble smaller specimens of Belloliva dor- cas sp. nov. but a large sample (31 specimens) from EBISCO st. DW2629 (Fig. 17J-N) contains all transitions to “typical” broad specimens. Radular morphology is also similar be- tween specimens from Chesterfield Plateau and Lansdowne Bank, although slight differences can be observed, especially in the shape of the cusps of the rachidian teeth, which are much more strongly bent in the specimen from Chesterfield (Fig. 18B) than in the specimen from Lansdowne (Fig. 18E). Such differences are smaller than differences with other similar species, especially Belloliva dorcas sp. nov. (Fig. 16E- F), and we consider them intraspecific. The species superficially resembles Belloliva alaos sp. nov. from New Caledonia, differing in the well pronounced plication on the columellar plait, and in the presence of eyes. It also differs from all other species of Belloliva in its straight or sometimes slightly concave outer lip. For comparison with Belloliva dorcas sp. nov., see under that species. Etymology From the Greek for “egg;” used as a noun in apposition. Belloliva dorcas Kantor and Bouchet sp. nov. (Figs. 16E-F, 20, 21, 22) Type material Holotype (Moll 9478) and 1 paratype (Moll 9479) in MNHN. Material examined Coral Sea. Bellona Plateau. MUSORSTOM 5, st. 328, 20°23'S, 158°44’E, 355-340 m- (1 dd); st. 329, 20°23’S, 158°47'E, 320 m (1 dd); EBISCO, st. DW2564, 20°25'S, 158°41'E, 333-386 m (1 lv); st. DW2574, 20°20’S, 158°45’E, 358-374 m (1 ly, radula extracted, co-occurring with Bello- liva exquisita). Chesterfield Plateau. MUSORSTOM 5, st. 347; 19°39'S; 158°28'E, 260 m. (1 dd); st, 375; 19°S2"S; 158°30'E, 300 m (3 dd [holotype and paratype] ); CHALCAL 1984, st. D31, 19°33.5'S, 158°30.5'E, 230 m (1 dd.). EBISCO, st. DW2608, 19°33’S, 158°40’E, 393-396 m (1 dd, co- occurring with Belloliva obeon sp. nov.). Lansdowne Bank. MUSORSTOM 5, st. 389, 20°45’S, 160°54’E, 500 m (4 dd). Type locality Coral Sea, Chesterfield Plateau, 19°52'S, 158°30'E, 300 m (MUSORSTOM 5, st. 375). BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA Monts Norait a Banc de Lansdowne Plateau de | Bellona | *s Vi Np, $22° bay Belloliva obeon sp. nov. & type locality * examined material Belloliva dorcas sp. nov. © type locality © examined material E158° E160° E162° Figure 20. Distributions of Belloliva obeon sp. nov. and Belloliva dorcas sp. nov. nn Ww 54 AMERICAN MALACOLOGICAL BULLETIN 22° 1/2 * 2007 Figure 21. Belloliva dorcas sp. nov. A-D, Holotype, MUSORSTOM 5, sta. 375, SL 13.6 mm. E, Bellona Plateau, EBISCO sta. DW2574 (SL 12.5 mm, radula studied). F, Chesterfield Plateau, MUSORSTOM 5, sta. 328, SL 11.1 mm. G, H, Specimens from Lansdowne Bank (MUSORSTOM. 5, sta. 389). G, (SL 11.1 mm). H, (SL 10.6 mm). I-J, Cheesterfield Plateau. I, CHALCAL, st. D31 (SL 9.6 mm). J, Chesterfield Plateau, MUSORSTOM 5, sta. 347 (SL 11.0 mm). All shells illustrated at the same scale except D. Description (holotype) Shell large, solid, glossy, elongate-oval (BWL/SL = 0.78, AL/SL = 0.69, D/SL = 0.45), with moderately narrow aper- ture and elevated spire, consisting of about 0.75 protoconch and 3 teleoconch whorls. Protoconch large, evenly rounded, diameter 1970 um, exposed height 1070 um, smooth, pro- toconch-teleoconch transition distinctly marked by onset of filament channel. Profile of whorls moderately convex, BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA evenly rounded. Filament channel completely open. Aper- ture lanceolate-oval, gradually tapering adapically. Outer lip slightly convex, nearly straight in adapical-most part, evenly rounded abapically. Parietal plate very narrow, not visible in strictly ventral view, very thin, anterior plating thickening and broadening in abapical part of aperture, bearing 6 weak plicae (Fig. 21D). Background color creamy yellow. Last and penultimate whorls (except anterior plating) covered by dis- tinct, rather broad, and irregularly shaped zigzag brown lines, well pronounced on the anterior band. Rim of filament channel marked by row of irregularly spaced brown spots or dashes. Dimensions (holotype): SL 13.6 mm, SW 6.1 mm, BWL 10.6 mm, AL 9.4 mm. Largest specimen (paratype): SL 14.0 mm, SW 6.3 mm, BWL 11.1 mm, AL 9.6 mm. Anatomy Part of the body was retrieved from one specimen (EBISCO sta. 2574, SL 12.5 mm, AL 8.5 mm - shell see Fig. 21E). External morphology very similar to that of Belloliva obeon, including the presence of eyes. Penis differing from that of B. obeon in being of nearly even diameter along its length, obtuse at the tip, and lacking the attenuated tip. Gland of Leiblein long, tubular, slightly coiled, with strong transverse folds visible through the gland wall. Radula (Fig. 16E-F) about 140 um wide (1.12% of SL, 1.64% of AL), consisting of about 85 rows of teeth, including 6 nascent. Rachidians rather narrowly spaced, cusps strongly abutting previous teeth (Fig. 16F). Anterior edge of rachidian slightly convex and rounded at the edges. Lateral sides of basal plate gradually embedded into the membrane without distinct border. Rachidian about 68 um wide (49% of radula width) with 3 main, closely spaced cusps and broad lateral flaps, central cusp about 1.3 times shorter and narrower than lat- eral cusps, and one small, but distinct, additional cusp on each side of main lateral cusps. Lateral teeth with subtrian- gular bases and long, curved, hook-like cusps. Distribution Coral Sea, northern Bellona Plateau, Chesterfield Pla- teau, and Lansdowne Bank (Fig. 20), alive in 358-374 m, shells from 230 m. Remarks Belloliva dorcas sp. nov. is variable in terms of shell shape and coloration. The holotype has the most pro- nounced zigzag lines; in other specimens, these are either rather inconspicuous (Fig. 21G) or completely absent, the shell then being nearly white with only a row of small light brown dots at the rim of the filament channel (Fig. 21F). On the Chesterfield Plateau, Belloliva dorcas sp. nov. and Belloliva obeon sp. nov. are easily recognized on the basis of wm Nn specimens from Chesterfield-Bellona Plateaus in D1, mm \ © Belloliva dorcas sp, nov / ‘ Kelso Bank Capel Bank 8 = Belloliva exquisita (Angas, 1871) Chesterfield Plateau ) os 0.6 0.9 PRE, mm 1.2 1.5 Figure 22. Morphometric comparison of protoconch dimensions of Belloliva dorcas sp. noy. and different populations of Belloliva exquisita (Angas, 1871) from the Coral Sea. shell shape, shell color, and radular morphology. In B. dorcas sp. noy., the shape of the rachidian is rather distinct in having much broader lateral flaps compared to B. obeon sp. nov. and the other species studied here. The radular mem- brane is also somewhat narrower in B. dorcas sp. nov. (1.12% of SL and 1.64% of AL versus 1.23-1.33% of SL, and 1.73%- 1.90% of AL in B. obeon). Their co-occurrence at one station (EBISCO, st. DW2608) is additional evidence that they are distinct species and not variants. On Lansdowne Bank, the identity of a population that we attribute to Belloliva dorcas sp. nov. (Fig. 21G-H) is problematic and requires anatomi- cal or molecular confirmation. Sympatric, but not syntopic, specimens of B. obeon sp. nov. (Fig. 17M-N), superficially ressemble it by their broad, rather straight, axial stripes (rather than the narrow, zigzag, chevron lines of B. dorcas); the characteristic row of irregularly spaced brown spots or dashes on the rim of the filament channel are another reason why we attribute this lot to B. dorcas. Belloliva exquisita bears a rather strong resemblance to Belloliva dorcas, of which it superficially seems to be a di- minutive form. The two species differ in protoconch mor- phometrics, with a significant gap in their diameters (Fig. 22). The two species are sympatric on the northern Bellona and Chesterfield Plateaus, and even syntopic at one station (EBISCO st. DW2574), thus leaving no doubt that two dif- ferent species are involved. The COI sequence was obtained for Belloliva dorcas (voucher specimen MNHN Moll 9484—Fig. 21E), GenBank accession no. DQ780463. 56 AMERICAN MALACOLOGICAL BULLETIN Etymology From the Greek noun dorcas, designating a kind of ga- zelle, with reference to the elegant colour pattern; used as a noun in apposition. Belloliva exquisita (Angas, 1871) (Figs. 22, 23, 24, 25, 26) Olivella exquisuta Angas 1871: 13, pl. 1, fig. 2. Type material Holotype BMNH 1871.7.5.5 (Fig. 23A-C). Material examined Surprise Atoll: LAGON, st. 444, 18°15'S, 162°59’E, 300- 350 m (1 dd); st. 502, 19°08’S, 163°30'E, 190 m (1 dd). PALEOSUPRISE, st. CP1391, 18°29.8'S, 163°02’E, 365 m (2 dd); st. CP1392, 18°29.8'S, 163°02.7’E, 370 m (1 dd). North of New Caledonia: MUSORSTOM 4, st. DW142, 18°35'S,. 163°10"E,. 525 m (1 dd); st. DW149,.19°508'S, 163°23"E, 155m (5 dd); st. DW150, 19°07"S, 163°22’E, 110 m (6 dd); st. DW151, 19°07'S, 163°22'E, 200 m (4 dd); st. DW162, 18°35’S, 163°10’E, 525 m (1 dd); st. DW164, 18°33'S, 163°13"E, 255 m (1 dd); st. DW184, 19°04’S, 163°27'E, 260 m (28 dd, 3 lv [1 specimen dissected]). BATHUS 4, st. DW923, 18°52’S, 163°24’E, 470-502 m (5 dd [co-occurring with Belloliva simplex and Belloliva apoma]); st. DW926, 18°57’S, 163°25’E, 325-330 m (1 dd); st. DW927, 18°56'S, 163°22'E, 444-452 m (1 dd); st. DW940, 19°00’S, 163°26'E, 305 m (4 dd, 1 lv); st. DW941, 19°02’S, 163°27’E, 270 m (1 dd); st. DW942, 19°04'S, 163°27'E, 264-270 m (11 dd). New Caledonia proper: BATHUS 4, st. DW887, 21°07'S, 164°28'E, 320-344 m (2 dd [co-occurring with Belloliva sim- plex]). BATHUS 1, st. DW 688, 20°33'S, 165°00’E, 270-282 m (1 dd). West coast. EXPEDITION MONTROUZIER, st. 1304, 20°38.6'S, 164°13.2'E, 12-15 m (1 dd); st. 1311, 20°40.4'S, 164°14.9'E, 10-60 m (14 lv and dd) (co-occurring with B. simplex); st. 1312, 20°40.4'S, 164°14.9'E, 26-40 m (36 lv and dd) (co-occurring with B. simplex); st. 1319, 20°44.7'S, 164°15.5’E, 15-20 m (1 lv); st. 1321, 20°40.7’S, 164°14.9'E, 90-115 m (1 lv); st. 1322, 20°44.2’S, 164°15.2’E, 53-71 m (1 dd) (co-occurring with B. simplex); st. 1323, 20°40.9'S, 164°14.8'E, 82-120 m (4 dd); st. 1331, 20°40.0’- 20°40.6’S, 164°11.2'-164°12.1’E, 55-57 m (4 lv) (co- occurring with B. simplex). South of New Caledonia/Norfolk Ridge: MUSORSTOM 4, st. DW210, 22°44'S, 167°09’E, 340-345 m (1 dd, 1 lv). BIOCAL, st. DW41, 22°45’S, 167°12'E, 380-410 m (1 dd). BERYX 11, st. CH41, 23°39’S, 168°00’E, 230-360 m (1 dd). SMIB: 5, st. DW79, 23°41'S, 168°01’E, 285 m (2 dd); st. DW80, 23°42'S, 168°00'E, 300 m (2 dd). East Jumeau Bank SMIB 8, st. DW170-172, 23°41’S, 168°00’-168°01’E, 230-290 m (4 dd); st. DW176, 23°42’S, 168°01'E, 283-290 m (1 dd); 22 * 1/2 * 2007 NORFOLK 1, st. DW1674, 23°40’S, 168°00’E, 245-253 m (1 dd). Antigonia Bank, NORFOLK 1, st. DW1712, 23°23'S, 168°02’E, 180-250 m (1 dd); st. DW1717, 23°23'S, 168°02'E, 250-312 m (1 dd). Banc P, NORFOLK 1, st. DW1723, 23°18'S, 168°15'E, 266-267 m (1 dd); st. DW1724, 23°17'S, 168°14'E, 200-291 m (1 dd); st. DW1726, 23°18'S, 168°15’E, 185-207 m (5 dd); st. DW1728, 23°19'S, 168°15’E, 207-276 m (4 dd). Coral Sea, Capel Bank: MUSORSTOM 55, st. 256, 25°18'S, 159°53’E, 290-300 m: (1 dd.)s st: 22585625233" S, 159°46’E, 300 m (5 dd.), st. 263, 25°21'S, 159°46’E, 225-150 m (21 dd, 1 lv {radula and external morphology]}); st. 265, 25°21'S, 159°45’E, 190-260 m (7 dd); st.-266, 25°20‘S, 159°46'E, 240 m (3 dd, 2 lv); st. 270, 24°49’S, 159°34’E, 223 m (2 dd); st. 273, 24°43’S, 159°43’E, 290 m (3 dd); st. 274, 24°45'S, 159°41'E, 285 m (9 dd, 1 lv). Argo Bank: MUSOR- STOM 5, st. 298, 22°44’S, 159°22’E, 320 m (2 dd); st. 299, 22°48'S, 159°24’E, 360-390 m (1 dd). Nova Bank: MUSOR- STOM 5, st. 303, 22°12’S, 159°23’E, 332 m (1 dd). CHAL- CAL, st. D63, 22°11'S, 159°14’E, 305 m (4 dd). EBISCO, st. DW2522, 22°46'S, 159°21’E, 310-318 m (5 dd); st. DW2538, 22°20'S, 159°25'E, 318-323 m (5 dd). Kelso Bank: MUSOR- STOM 5, st. 277, 24°11°S, 159°35"E, 270ani (10 dd}.1 hy); st: 280, 24°10’S, 159°36’E, 270 m (1 dd). EBISCO, st. DW2509, 24°08'S, 159°35'E, 265 m (1 dd); st. DW2514, 24°06’S, 159°41’E, 295-310 m (1 dd); st. DW2515, 24°04'S, 159°41’E, 330-370 m (1 dd). Bellona Plateau: EBISCO, st. DW 2547, 21°06'S, 158°36’E, 356-438 m (1 dd); st. DW2574, 20°20’S, 158°45'E, 358-374 m (1 dd, co-occurring with Belloliva dor- cas sp. nov.); Chesterfield Plateau: EBISCO, st. DW 2603, 19°36'S, 158°43’E, 570-568 m (8 dd). Type locality Coogee Bay, New South Wales, Australia. Description Shell solid, relatively thin, glossy, with moderately nar- row aperture and elevated spire. Protoconch large, evenly rounded, diameter 1200-1600 pm, smooth. Profile of whorls moderately convex, evenly rounded, very inconspicuously shouldered. Filament channel open. Aperture lanceolate- oval, gradually tapering adapically. Outer lip slightly convex, nearly straight in most adapical part, evenly rounded abapi- cally. Anterior plating bearing 5-7 rather weak, but distinct plicae. Dimensions: largest specimen (Capel Bank, MUSORSTOM 5, st. 258) SL 12.5 mm, SW 5.2 mm, BWL 10.2 mm, AL 8.1 mm. Distribution Southeastern Australia, New South Wales, Queensland; Coral Sea guyots, New Caledonia including its continuation BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 57 Figure 23. Belloliva exquisita (Angas, 1871). A-C, Holotype, BMNH 1871.7.5.5 (SL 7.7 mm). D, Northern New Caledonia, MUSORSTOM 4, st. DW184 (SL 9.2 mm). E, Northern New Caledonia, MUSORSTOM 4, st. DW149, SL 6.4 mm. F, G, Northern New Caledonia, MUSORSTOM 4, st. DW149, SL 7.6 and 8.4 mm. H, Northern New Caledonia, PALEO-SURPRISE, st. DW1391, SL 6.7 mm. I, Koumac, western New Caledonia, st. 1311 (SL 6.5 mm). J, Southern New Caledonia, SMIB5, st. DW80 (SL 9.2 mm). K, Capel Bank, MUSORSTOM 5, st. 258 (SL 11.2 mm). L, Capel Bank, MUSORSTOM 5, st. 265, SL 8.9 mm. M, Kelso Bank, MUSORSTOM 5, st. 277 (SL 9.0 mm). N, Kelso Bank, MUSORSTOM 5, st. 277 (SL 8.7 mm). O, Bank Nova, CHALCAL, st. D63 (SL 8.8 mm). P, Argo Bank, MUSORSTOM 5, st. 299 (SL 9.8 mm). All shells illustrated at the same scale. 58 AMERICAN MALACOLOGICAL BULLETIN — 22 ¢ 1/2 * 2007 Figure 24, Scanning electron micrographs of the radula of Belloliva exquisita (Angas, 1871). A-D, Coral Sea, MUSORSTOM 5, sta. 263. E-H, Northern New Caledonia, MUSORSTOM 4, sta. DW 184. A, E, Dorsal view of the central part of the radular membrane. B, F, Enlarged rachidian teeth. C, G, Left and right lateral views of the rachidian teeth, respectively. D, H, Left and right lateral views of the lateral teeth, respectively. Scale bars = 10 um. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 59 to Surprise Atoll in the North and Norfolk Ridge in the South, alive 26-345 m, shells in 12-525 m. Remarks We are treating Belloliva exquisita as a single, highly variable species, both in terms of geographical and bathy- metric variation. Within-population variation concerns col- oration, with very pale to practically white or semitranspar- ent specimens not uncommon in all parts of the distribution. Most populations from off the north, south and west coasts of New Caledonia, including Norfolk Ridge, are very similar and there is no doubt that they represent a single species. Conchologically specimens from these populations are most similar to those from New South Wales, including the holotype (Fig. 23A-C). These individuals are character- ized by rather slender shells (D/SL 0.45, n = 6, range 0.43- 0.46) with elevated spires (BWL/SL 0.77, n = 6, range 0.75- 0.78), and a color pattern typically of narrow, irregular zigzag brown lines on a creamy yellow background, and two spiral rows of unevenly spaced, spirally elongated, brown spots, one on rim of filament channel, the other on whorl periphery. Coloration is more pronounced in southern populations (Fig. 23J), although pale to nearly white speci- mens are also found there, but in lesser proportion than in the northern populations (Fig. 23H). A little more problematic is a series of populations from a rather narrow depth range (110-190 m) in a restricted area extending for about 13 km (MUSORSTOM 4, st. DW 149, 150, 151) that differ from the “typical” form in having a larger, much broader shell (average D/SL = 0.50, n = 6, range 0.49-0.51, Fig. 23F, G; versus D/SL = 0.45, n = 6, range 0.43-0.46 in “typical” form) and sharper and more numer- ous plicae on the anterior plating. At one station (MUSOR- STOM 4, st. DW 149) the broad and narrow (Fig. 23E) forms co-occur, which might indicate that they represent two different species. However, as only four empty shells were taken at that station, the evidence to treat them as two species is very weak and, in the absence of further data (radular morphology, anatomy), we prefer to hypothesize that the broad form represents a local variant of Belloliva exquisita. Ideally, this should be tested with molecular data. Variation in the Coral Sea is more classically geographi- cal, with each of the guyots having its own recognizable population, described below from south to north. However, the morphometries of specimens from different banks over- lap and form a continuum with the “typical” form from New Caledonia (Fig. 25); we believe that this geographical varia- tion reflects limited genetic exchange between banks, but not isolation. Capel Bank (Fig. 23K, L). Shell to 12.5 mm in length, rather slender (D/SL 0.42, n = 7, range 0.39-0.46), with lower spire (BWL/SL 0.80, n = 7, range 0.77-0.81), usually Argo Bank 1.5 = = =) New Caledonia Kelso Bank 0.8 PRE, mm 1.0 09 Capel Bank Argo Bank Zz ‘a y =e = gree 5 08 Kelso Bank jaa] Nova Bank 0.4 D/SL 0.5 Figure 25. Morphometric comparisons of some protoconch and shell measurements of different populations of Belloliva exquisita (Angas, 1871) to illustrate the overlap of characters. Different populations are marked by different symbols. pale, zigzag lines from very pale to absent, brown spots at channel rim present at least on parts of shell. Protoconch on average slightly larger (D1 = 1.37, n = 7, range 1.26-1.42) than in a “typical” form (D1 = 1.24, n = 6, range 1.21+1.30), although both forms overlap significantly (Fig. 25). Kelso Bank (Fig. 23M, N). Shells to 9.0 mm in length, differing from previous in their broader, more oval shell (D/SL in average 0.46, n = 5, range 0.46-0.47) with shorter spire, generally light in background color, although with better pronounced brown zigzag lines. Protoconch dimen- sions slightly smaller than in specimens from Capel Bank, but overlapping with “typical” form. Argo Bank (Fig. 23N). Shells more similar in shape to those from Capel Bank, but characterized by the largest pro- toconch dimensions. Coloration light, zigzag lines pale, but well defined. Nova Bank (Fig. 230). Shells to 11.6 mm, characterized by rather slender shells (D/SL in average 0.39, n = 4, range 0.38-0.40). Protoconch dimensions completely overlaping 60 AMERICAN MALACOLOGICAL BULLETIN with those from Capel Bank. Coloration light, some speci- mens with very pale zigzag lines. Our material from Bellona and Chesterfield Plateaus is in rather poor condition and does not allow a detailed de- scription, but is sufficient to record Belloliva exquisita in sympatry with Belloliva dorcas and to highlight the fact that the protoconch of these specimens is the smallest in all populations of B. exquisita examined (Fig. 22). The anatomy of three specimens was examined, one from Capel Bank (MUSORSTOM 5, st. 263, SL 9.8, BWL 8.1, AL 6.2, SW 4.1), one from North of New Caledonia (MUSORSTOM 4, st. DW184, SL 9.4, BWL 7.2, AL 6.2, SW 4.2 mm), and one dried, rehydrated, specimen from Kou- mac (MONTROUZIER, st. 1312, SL 6.2, BWL 4.8, AL 4.3, SW 3.1 mm). The specimen from the Coral Sea had the body strongly contracted, mantle cavity spanning about 0.5 whorls, ne- phridium 0.3 whorls, digestive gland with gonad about 3 whorls. Body in alcohol uniformly pale yellow, lacking pig- mentation. Nephridium with transparent walls. Anterior lobe of digestive gland small, spanning about 0.3 whorls and completely separated from posterior lobe by stomach which is oriented obliquely with regard to columellar axis. Foot thick, strongly contracted during fixation, transversely folded, metapodium broadly triangular-oval, propodium small in comparison with metapodium, typically crescent- shaped, subdivided longitudinally. Operculum transparent, very thin, elongate, constricted abapically, slightly thickened along low inner edge. Head well distinguished from the rest of the body, with two large vertical flaps, bearing large eyes. Mantle cavity.—Mantle strongly contracted, edge straight and thickened. Mantle thin, osphradium and ctenidium showing through it. Siphon very thin-walled, slightly extending beyond mantle edge, with smooth edges. General arrangement and proportions of organs in the mantle complex similar to that of Belloliva alaos. Alimentary system.—Anterior foregut very similar to that of Belloliva alaos. Salivary glands apparently ramified- tubular. Accessory salivary glands absent. Radula consisting of about 75 rows of teeth, membrane width about 90 um (0.92% SL, 1.45% AL). Rachidian with 3 main cusps, central cusp about twice as narrow and 1.5 times shorter than lateral cusps; an additional small but distinct cusp on each side of the main lateral cusps. Dorsal grooves on the main lateral cusps shallow and broad, best seen in lateral view (Figs. 24C, G). Anterior profile of rachidian nearly straight. Posterior edge of basal plate slightly convex. Sides of basal plate gradu- ally embedded in the membrane without distinct border. Lateral teeth subtriangular with curved hook-like tips (Figs. 24D, H). Stomach large, spanning more than 0.5 whorl, with very long posterior mixing area. Stomach anatomy not ex- amined in detail due to poor fixation. Rectal gland not found 22 ° 1/2 * 2007 during dissection, probably obscured by thick mucous layer produced by hypobranchial gland. Reproductive system.—Specimen a mature male, penis long, smooth, narrowing sharply towards its tip, similar to Belloliva alaos. The external morphology of the two specimens from New Caledonia is very similar. Stomach similar in shape but much shorter. Digestive gland small, anterior spanning about 1/6 whorl, posterior about 4 whorl. Upper 3 whorls occupied by testis. Seminal vesicle situated at lower corner of junction of digestive gland and testis, making several loops. Penis as long as mantle cavity, smooth, oval in section, end- ing in a small, nearly transparent, seminal papilla. Radula of both specimens identical and very similar to that of the Coral Sea specimen (Fig. 24E-H), differing only in the rela- tively broader, more triangular central cusp of the rachidian, as well as slightly wider radular membrane (membrane width 100 versus 95 tm, 1.e., 1.06% versus 1.53% of SL and 1.61% versus 2.20% of AL). Calyptoliva Kantor and Bouchet gen. nov. Diagnosis Shell small, 7-15 mm, narrowly elongate-oval, with at- tenuated spire. Suture narrowly channeled, overlaid by thin primary callus. Protoconch paucispiral, consisting of about one whorl, smooth, evenly rounded, large, diameter 1300- 1650 tum, protoconch-teleoconch transition not clear. Aper- ture narrow, elongate. Parietal plate narrow, anterior plating smooth. Foot with well developed parapodia and crescent- shaped propodium. Operculum present. Mantle without filament, mantle lobe well developed. Head consisting of two separate vertical flaps, separated by furrow; eyes present or absent. Rhynchostome opening situated below the right flap. Proboscis short. Salivary glands paired, accessory salivary gland absent, valve of Leiblein large, gland of Leiblein nar- row tubular, stomach with long posterior mixing area. Ra- chidian radular teeth with 3 main cusps, central cusp nar- rower and shorter, additional smaller cusp abutting each side of main lateral cusps. Lateral teeth with subtriangular bases and long curved hook-like cusps. Type species Calyptoliva bolis Kantor and Bouchet sp. nov. Remarks Calyptoliva gen. nov. bears a strong overall resemblance to Belloliva, both in protoconch and teleoconch shape, with similar soft body gross morphology and similar radula. Ca- lyptoliva differs from Belloliva in the absence of open fila- ment channel of the shell, and, correspondingly, of the BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 61 E160° E158° ChdMerfield * Plateau KE ellona E156° E158° E160° E162° E162" E166° E168° | | Esperitu Santo E164° E166° Figure 26. Distribution of different forms of Belloliva exquisita (Angas, 1871) in the Coral Sea and New Caledonia. mantle filament; it also has a mantle lobe that is not present in Belloliva. The gland of Leiblein is massive in Belloliva, tubular in Calyptoliva. The mophology of the suture requires special com- ments. In all but one specimen studied, the suture is con- cealed by a very thin callus, extending slightly adapically onto the preceding whorl, and there is no filament channel. However, a specimen with corroded suture area (Fig. 27G) shows that a filament channel remains present below the callus. Etymology From the Greek kalypto, meaning to cover, to conceal, referring to the concealed filament channel of the shell. 62 AMERICAN MALACOLOGICAL BULLETIN Calyptoliva bolis Kantor and Bouchet sp. nov. (Figs. 27A-C, 28, 29) Type material Holotype (Moll 9480) in MNHN. Material examined Coral Sea, Lansdowne-Fairway Bank, CORAIL 2, st. DE14, 21°01'S, 160°57’E, 650-660 m (1 dd). MUSORSTOM 5, st. 390, 21°01’S, 160°50’E, 745-825 m (1 lv). Type locality Coral Sea, Lansdowne-Fairway Bank, 21°01'S, 160°57’E, 650-660 m (CORAIL 2, st. DE14). Description (holotype) Shell solid, glossy, elongate-oval (BWL/SL = 0.71, AL/SL = 0.54, D/SL = 0.38), with narrow aperture and high spire, consisting of just over one protoconch and 3.75 teleo- conch whorls. Protoconch large, low, evenly rounded, diam- eter 1650 um, exposed height 780 tm, smooth, protoconch- teleoconch transition indistinctly marked by the appearance of the callus overlapping the suture on teleoconch whorls. Profile of whorls moderately convex, evenly rounded. Suture shallowly impressed and overlain by very narrow and thin smooth callus, extending slightly adapically. Filament chan- nel (seen by transparency through callus) narrow, closed by overlaid callus but not filled. Aperture narrow, gradually tapering adapically. Outer lip evenly and slightly convex. Parietal plate narrow, microshagreened, broadening and thickening in its abapical part prior to anterior band. Ante- rior plating without any plicae, similarly microshagreened except for most abapical part. Color very light yellow, ante- rior band white. Dimensions (holotype largest specimen): SL 12.9 mm, SW 4.9 mm, BWL 9.2 mm, AL 7.0 mm. Anatomy General morphology.—Body only partially retrieved from shell, in alcohol uniformly off-white, lacking pigmen- tation. Mantle cavity spanning about 0.3 whorls, nephridium 0.25 whorls with transparent walls, with 8 low excretory lamellae. Anterior lobe of digestive gland small, spanning about “41 whorl and completely separated from posterior lobe by stomach, which is oriented obliquely with regard to columellar axis. Foot thick, strongly contracted, trans- versely folded, metapodium broadly oval, propodium small in comparison with metapodium, typically crescent-shaped, subdivided longitudinally. Operculum transparent, very thin, elongate, constricted in upper part, without pro- nounced growth lines. Head poorly distinguished from the 22 * 1/2 * 2007 rest of the body, with two small vertical flaps (Fig. 28C). No eyes. Mantle cavity—Mantle edge even, forming a long, rather muscular and thicker extension on the right side, terminating with a medium-sized posterior mantle lobe (Fig. 28B, E, ml). Mantle thin, osphradium and ctenidium show- ing through. Siphon short, thin-walled, slightly extending beyond mantle edge (Fig. 28E), with smooth edges. Osphra- dium bipectinate, broad, exceeding the width of the ctenidium and about 0.8 of its length. Osphradium nearly bilaterally symmetrical, with very narrow axis. Ctenidium occupying about 0.8 of mantle length, consisting of tall tri- angular lamellae, similar in shape and size along most of its length, except near mantle edge where ctenidium sharply narrows and lamellae become much lower. Hypobranchial gland moderately glandular, although not forming distinct folds. Mantle filament and posterior mantle tentacle absent. Female pallial gonoduct large, swollen, not studied in detail due to poor fixation. Female genital opening situated close to anus. Alimentary system.—Rhynchostome asymmetrical, situ- ated below the right cephalic flap (Fig. 28C, rns). Organs of the anterior hemocoel very contracted, compact and situated nearly at right angle with regard to pedal axis. Proboscis short in contracted state (Fig. 28D, pr), about 1.2 mm in length (0.16 AL), thick (L/W approximately 2), occupying nearly the entire rhynchocoel length, rhynchodeum thin- walled, semi-transparent. Pair of thin retractor muscles at- tached to median part of the rhynchodeum (wall of probos- cis sheath) in retracted condition. Large odontophoral retractor extends from proboscis posteriorly, then follows anteriorly along ventral side of rhynchodeum and, bypassing the nerve ring, attached to the ventral side of cephalic he- mocoel. Esophagus, posterior to proboscis, rather broad and not forming a loop. Odontophore rather broad, occupying nearly entire volume of proboscis, about the same length as proboscis and not protruding from the rear of it. Subradular cartilages large, fused antero-ventrally by rather thin inter- connection. Radular sac slightly longer than the odonto- phore. Radula (Fig. 29) consisting of about 45 rows of teeth, width of the membrane about 155 um (1.23% SL, 2.02% AL). Rachidian with 3 main cusps, central cusp about three times as narrow and 1.5 times shorter than lateral cusps, and an additional very small, but distinct, cusp abutting outer side of main lateral cusps. Dorsal grooves on main lateral cusps shallow and broad, best seen in lateral view (Fig. 29C). Anterior profile of rachidian nearly straight, very slightly concave in the middle. Posterior edge of basal plate very slightly convex. Sides of basal plate gradually embedded in the membrane without distinct border. Lateral teeth subtri- angular with curved hook-like tips (Fig. 29D). Valve of Leiblein very large, pyriform, well distinguished from BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 63 Figure 27. Calyptoliva bolis sp. nov. (A-C). A-B, Holotype, SL 12.9 mm. C, MUSORSTOM 5, st. 390, SL 12.6 mm. Calyptoliva tatyanae sp. nov. (D-G). D-E, Holotype, SL 13.1 mm. F, Paratype, SL 12.8 mm. G, Upper whorls of paratype, SL 13.0 mm, illustrating the open filament channel, covered by the callus; the areas with intact callus are indicated by white arrows. Calyptoliva amblys sp. nov. (H-J). H-I, Holotype, SL 9.3 mm. J, Paratype, SL 9.2 mm. 64 AMERICAN MALACOLOGICAL BULLETIN — 22+ 1/2 + 2007 aldg 1 mm Figure 28. Morphology of Calyptoliva bolis sp. nov. (for shell see Fig. 26D). A, B, Dorsal and ventral views of the body removed from the shell, respectively. C, Head-foot, anterior view, mantle and visceral mass removed. D, Right view of the anterior foregut. E, Mantle complex. Abbreviations: aldg, anterior lobe of digestive gland; cm, columellar muscle; cme, cut mantle edge; ct, ctenidium; fgo, female genital opening; gL, gland of Leiblein; hg, hypobranchial gland; ml, mantle lobe; ne, nephridium; nr, circumesophageal nerve ring; odr, odonto- phoral retractor; op, operculum; os, osphradium; par, parapodium; pg, pallial gonoduct; poe, posterior esophagus; pr, proboscis; prp, propodium; prr, proboscis retractor; re, rectum; rnh, rhynchodeum (= proboscis sheath); rns, rhynchostome; rsg, right salivary gland; s, siphon; sd, salivary duct; st, stomach; tn, cephalic tentacles; vL, valve of Leiblein. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 65 Figure 29. Scanning electron micrographs of the radula of Calyptoliva bolis sp. nov. A, Dorsal view of the central portion of the radular ribbon. B, Enlarged dorsal view of the rachidian teeth. C, Left lateral view of the rachidian teeth. D, Left lateral view of the lateral teeth. Scale bars = 10 jm. esophagus (Fig. 28D, vL), which becomes very narrow im- mediately posterior to the valve and passes through the nerve ring. Circumesophageal nerve ring comparatively large. Gland of Leiblein small, colorless, narrow, tubular. Opening of the duct into esophagus not traced during dissection. Salivary glands medium-sized, apparently (but not certainly due to small size) broad-tubular, left one slightly smaller than right, situated on either side of esophagus posterior to the proboscis and not fusing. Salivary ducts rather thick, entering the esophageal walls shortly after leaving the glands and passing towards their opening along esophagus dorsal side. Accessory salivary glands not found. Stomach large, spanning about 0.3 whorls, with long and narrow posterior mixing area (Fig. 28B, st). Stomach anatomy not studied in detail due to poor fixation. Remarks The second specimen is very similar to the holotype, except that its apertural lip is not thickened and its color is pure white. Etymology From the Greek bolis, a missile, with reference to the general shell profile; used as a noun in apposition. Calyptoliva tatyanae Kantor and Bouchet sp. nov. (Fig. 27D-G) Type material Holotype (Moll 9481) and 2 paratypes (Moll 9482) in MNHN. Material examined Coral Sea, Fairway Bank. EBISCO, st. CP2647, 21°32’S, 162°27'E, 737 m (3 dd). Type locality Coral Sea, southeastern part of Fairway Bank, 21°32'S, 162°27'E,. 737 m (EBISCO, st..CP2647). Description (holotype) Shell solid, glossy, elongate, nearly biconic (BWL/SL = 0.61, AL/SL = 0.50, D/SL = 0.35), with narrow aperture and 66 AMERICAN MALACOLOGICAL BULLETIN high spire, consisting of about 0.8 protoconch and 4.75 te- leoconch whorls. Protoconch very large, tall, globular evenly rounded, diameter 2340 um, exposed height 1640 pm, smooth, protoconch-teleoconch transition indistinctly marked by the appearance of the callus overlapping the su- ture on teleoconch whorls. Profile of whorls weakly convex, evenly rounded. Suture shallowly impressed and overlain by very narrow and thin smooth callus, extending slightly adapically. Filament channel (seen by transparency through callus) narrow, closed by overlaid callus but not filled. Ap- erture narrow, tapering adapically. Outer lip slightly convex adapically and nearly straight along most of its length. Pa- rietal plate narrow, clearly microshagreened, broadening and thickening in its abapical part prior to anterior band. Ante- rior plating without any plicae, similarly microshagreened. Color uniformly off-white, sutures slightly darker than the rest of the shell surface. Dimensions (holotype, largest specimen): SL 13.1 mm, SW 4.6 mm, BWL 8.1 mm, AL 6.6 mm. Remarks Calyptoliva tatyanae sp. nov. differs from Calyptoliva bolis sp. nov. by its narrower shell with taller spire, less convex whorls, narrower aperture, and much larger proto- conch (diameter 2340 tum versus 1650 ttm in C. bolis). All three specimens are very similar in shape, with some vari- ance in the convexity of the whorls. Etymology The species is named after the biologist and wife of the senior author, Tatiana Steyker, from the P.P. Shirshov In- stitute of Oceanology of the Russian Academy of Sciences. Calyptoliva amblys Kantor and Bouchet sp. nov. (Fig. 27H-J) Type material Holotype in AMS, 2 paratypes in MNHN. Material examined Coral Sea, CORAIL2, Mellish Reef, st. DW172, 18°26’S, 155°12’E, 1100 m (4 dd). Type locality Coral Sea, Mellish Reef, 18°26'S, 155°12’E, 1100 m (CORAIL 2, st. DW172). Description (holotype) Shell solid, glossy, elongate-oval, white (BWL/SL = 0.78, AL/SL = 0.54, D/SL = 0.46), with medium-wide aperture and low spire. Shell consists of 0.75 whorl of protoconch and 2.75 teleoconch whorls. Protoconch large, low, evenly 22° 1/2 * 2007 rounded, diameter 1330 pm, exposed height 590 um, smooth, protoconch-teleoconch transition is indistinctly marked by the appearance of the callus overlapping the su- ture on teleoconch whorls. Profile of whorls moderately con- vex, evenly rounded. Suture is shallowly impressed and overlain by very nar- row, thin smooth callus, extending slightly adsuturally. Fila- ment channel (seen by transparency through callus) narrow, closed by overlaid callus but not filled. Aperture medium wide, obtuse adapically. Outer lip thickened, convex adapi- cally, nearly straight along most of the length and rounded abapically. Parietal plate narrow, slightly broadens in its abapical part prior to anterior band, microshagreened. An- terior plating smooth, similarly microshagreened. Dimensions (holotype largest specimen): SL 9.2 mm, SW 4.2 mm, BWL 7.2 mm, AL 5.0 mm. Animal unknown. Remarks Calyptoliva amblys differs from Calyptoliva bolis sp. nov. by its smaller adult size with more convex whorls, wider aperture with straight lip in its middle part, and smaller protoconch. All four known specimens are very similar in shape, with some variance in the convexity of the whorls. Etymology From the Greek amblys, obtuse, with reference to the shell shape. DISCUSSION Revised diagnosis of Belloliva Iredale (1924) described Gemmoliva as a subgenus of Belloliva, with Olivella pardalis A. Adams and Angas, 1864 as type species, based on minor differences in radulae (Peile 1922) namely the additional marginal cusps on the rachidian tooth were said to be “very small and apparently sometimes missing.” We found this character to be of not more than specific value in the species of Belloliva we studied and, the shell and anatomical characters being otherwise equal, we agree with Wilson (1994) in synonymizing Gemmoliva with Belloliva. Thiele (1929 in 1929-1931) also described Olivel- lopsis as a subgenus of Belloliva, with Olivella simplex Pease, 1868 as type species based on subtle shell differences (essen- tially suture more appressed and columellar callus without lower fold). However, its radula and gross morphology do not differ significantly from other species of Belloliva, in- cluding its type species, and thus do not support segregation of O. simplex in a separate genus-group taxon. As a result of the new data presented in this paper and BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 67 comparison with other taxa, the following is a revised diag- nosis of Belloliva. Shell less than 15 mm long, olivelliform, elongate-oval, with attenuated spire, and open filament channel. Proto- conch paucispiral, consisting of about one whorl or less, smooth, evenly rounded, large in comparison with the te- leoconch, diameter 1000-1800 jum, protoconch-teleoconch transition marked by the onset of filament channel. Aperture elongate or lanceolate-oval, gradually narrowing abapically. Parietal plate narrow, slightly thickened, anterior plating smooth or plicate. Foot with well developed parapodia and crescent-shaped propodium. Operculum usually present, narrow. Mantle with mantle filament, without mantle lobe. Head with two separate vertical flaps, separated by furrow. Rhynchostome opening below the right flap. Proboscis short or of medium length when retracted. Salivary glands paired, ramified tubular, accessory salivary glands absent, valve of Leiblein medium-sized to large, gland of Leiblein bulky, stomach with long posterior mixing area. Rachidian radular teeth with 3 main cusps, central cusp narrower and shorter than the lateral ones, and additional small, but usually dis- tinct, cusps abutting each side of the main lateral cusps. Lateral teeth with subrectangular or subtriangular bases and long curved hook-like cusps. Composition of the genera Four Australian species have traditionally been included in Belloliva (Kaicher 1987, Wilson 1994, see below). In ad- dition, a couple of other taxa have been, at one time or another, allocated to the genus and need to be discussed separately. (1) Based on his examination of the radula of Olivella tabulata Dall, 1889 from off Cuba, Olsson (1956) tentatively included it in Belloliva, and this placement was followed by Kaicher (1987). We concur with Olsson that Olivella tabu- lata bears an overall resemblance to Belloliva, especially in the large size of the protoconch. However, Olsson described the radula with a rachidian bearing 3 cusps, of which the central one is the largest, whereas in Belloliva from Australia (including the type species) and the South-West Pacific, the central cusp is the smallest. In addition, O. tabulata lacks an operculum (Dall 1889), a character admittedly shared with Belloliva apoma but with no other species of Belloliva. The diverging distributions and the radular differences suggest that O. tabulata is probably not congeneric with the Austra- lia-South-West Pacific clade, but probably belongs to some other, possibly still unnamed, genus of the subfamily Olivinae. (2) Hunon (2000) attributed Oliva lacanientai Greite- neder and Blécher, 1985 to Belloliva based on the presence of two fasciolar bands and protoconch shape; Hunon also stated he had found remains of an operculum. We have examined material of O. lacanientai and find Hunon’s state- ment misleading, since shell shape and protoconch (mul- tispiral, consisting of about 4.25 whorls) are typical for Oliva. Examination of the anatomy and radula (Fig. 30A) confirm a placement in Oliva. We did not find an operculum and we suggest that Hunon mistakenly interpreted the pres- ence of an operculum from the presumably rotten and dried (as this was dealer's material from tangle nets) soft parts of his specimen. The present study thus brings to 10 the number of spe- cies included in Belloliva, and three in Calyptoliva, and high- lights the Coral Sea as their center of diversity: In fact, our material still contains one additional undescribed species from Lansdowne-Fairway Banks (represented by a single empty juvenile shell), suggesting that additional findings of new species of Belloliva in the Coral Sea are still possible. By contrast, it should be emphasized that cruises conducted since 1992 in other South Pacific archipelagoes (Solomons, Vanuatu, Fiji, Wallis and Futuna, Tonga, Marquesas) did not yield any deep-water material of Belloliva or Calyptoliva. Thus the following species are currently recognized: Bel- loliva alaos sp. nov., northern New Caledonia, alive in 668 m, shells in 397-660 m (Fig. 31), Belloliva apoma sp. nov., northern New Caledonia, alive in 502-516 m, shells in 470- 550 m (Fig. 31), Belloliva brazieri (Angas, 1877), type species of the genus, coastal waters of south-eastern Australia (New South Wales, Victoria, and Tasmania) (Fig. 32A-B), Belloliva dorcas sp. noy., Coral Sea, Chesterfield Plateau and Lans- downe-Fairway Bank, shells in 230-355 m (Fig. 31), Belloliva ellenae sp. nov., Coral Sea, Chesterfield plateau, alive in 386- 486 m (Fig. 31), Belloliva exquisita (Angas, 1871), coastal waters of eastern Australia, Coral Sea and New Caledonia, alive in 26-345 m (Fig. 31), Belloliva leucozona (A. Adams and Angas, 1864), coastal waters of the eastern seaboard of Australia from Caloundra, Queensland to Lorne, Victoria (Fig. 32C-D), Belloliva obeon sp. nov., Coral Sea, Chesterfield Plateau and Lansdowne-Fairway Bank, alive in 500-672 m, shells from 252 m (Fig. 31), Belloliva simplex (Pease, 1868), coastal waters of Tuamotu Island (French Polynesia), Sa- moa, Tonga, Loyalty Islands, New Caledonia, and Vanuatu, alive in 10-45 m (Fig 31), Belloliva triticea (Duclos, 1835) (= O. pardalis A. Adams and Angas, 1864), coastal waters of southern Australia (New South Wales to Albany, Western Australia) (Fig. 32E-F), Calyptoliva amblys sp. nov., Coral Sea, Mellish Reef, shells in 1100 m, Calyptoliva bolis sp. nov., Coral Sea, Lansdowne-Fairway Bank, alive in 745-825 m, shells from 650-660 m (Fig. 31), Calyptoliva tatyanae sp. nov., Coral Sea, Fairway Bank, shells in 737 m (Fig. 31). All species of Belloliva and Calyptoliva have a paucispiral protoconch indicating non-planktotrophic development, and therefore inferred limited larval dispersal, which prob- ably account for restricted ranges and multiple speciation 68 AMERICAN MALACOLOGICAL BULLETIN 22° 1/2 * 2007 Figure 30. Scanning electron micrographs of the radulae of different members of Olividae. A, Oliva lacanientai Greifeneder and Blocher, 1985 (Coral Sea) MUSORSTOM 5, sta. 380). B, Ancila cinnamomea Lamarck, 1801 (Southern India, Rameshwaran). C, Ancillina sp. (northern New Caledonia, BATHUS 4, st. DW914). D, Amalda fuscolingua Kilburn and Bouchet, 1988 (northern New Caledonia, BATHUS 2, st. DW729). E, Turrancila sp. (Indonesia, KARUBAR, st. CP71). F, Belloliva alaos sp. nov. (northern New Caledonia, MUSORSTOM 4, st. DW160). Scale bars = 10 pm (A, C, E, F), 100 um (B, D). events on the isolated seamounts and banks in the middle of the Coral Sea. However, it is difficult to ascertain whether the very narrow ranges of some of the new species are real or represent sampling artefacts. For instance, Belloliva ellenae is known from 8 stations that straddle only 30 km on the Chesterfield Plateau (Fig. 31), and Belloliva apoma is known from 4 stations over a distance of 35 km. Conversely, Belloliva exquisita ranges from Australia to New Caledonia, including isolated guyots and banks in the Coral Sea. Lim- ited sampling in the Coral Sea may explain the ap- parently extremely narrow ranges, such as that of B. ellenae, but the more than one thousand hauls taken in New Caledonia suggest that the very limited range of B. apoma is real. A similar pattern of narrow endemism has been de- BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 69 obeon bolis 116° alaos aonee® * anoer® ho ra apoma Plateau a ___ Bank Kelso me Bank Capel dorcas Figure 31. General distributions of Coral Sea and New Caledonian species of Belloliva and Calyptoliva. Holotypes illustrated when named in this paper, all at the same scale. 70 AMERICAN MALACOLOGICAL BULLETIN scribed in the family Volutomitridae (Bouchet and Kantor 2004). Position of Belloliva and Calyptoliva in the family Olividae Although the genus Belloliva has been referred to Oli- vellidae (or Olivelliinae) in the recent literature (e.g., by Wilson 1994, Tursch and Greifeneder 2001, Sterba 2003), this is not supported by its radular morphology, a point already made by Olsson (1956), who referred the genus to the Olivinae. The current consensus (Bouchet and Rocroi 2005) on the composition and classification of the family Olividae is to recognize two subfamilies, the nominal sub- family Olivinae (synonyms Agaroniinae and Olivancillari- nae) and the Ancillariinae (= Ancillinae). Surprisingly few anatomical data are available, essentially only for various species of Oliva (Marcus and Marcus 1959, Kantor 1991, Kantor and Tursch 2001) and two of Olivancillaria (Marcus and Marcus 1959); for Ancillariinae, the information is re- stricted to just two species of Amalda (Marcus and Marcus 1968, Kantor 1991). One of the characteristic traits of the family Olividae is the presence of mantle appendages. The most complex as- semblage is present in the Olivinae, which have anterior i) ie) 1/ ie) * 2007 mantle tentacle, a mantle filament, and a mantle lobe. The mantle filament is a muscular, contractile, and mobile organ which originates on the right side of the mantle edge, ex- tending through the aperture adapically and positioning in the filament channel. It is present in all oliviform gastropods (including the Olivellidae) that have channelled sutures, and its function remains unknown. The anterior mantle tentacle is situated near the siphon and, when the snail is crawling, it passes through the siphonal canal and rests on the dorsal side of the shell. Its function is not known either. The small posterior mantle lobe is situated at the base of the mantle filament (when present). In Belloliva, only a rather short (in preserved animals) mantle filament is present; neither an anterior mantle tentacle nor a mantle lobe were found in dissections. By contrast, in Calyptoliva, the mantle filament is absent but the mantle lobe is well-developed. Judging from the differences in shell morphology, it may be inferred that the mantle lobe is responsible for secreting the primary spire callus (following the terminology of Kilburn 1977) that overlies the suture, rather than producing the columellar callus (as suggested by Marcus and Marcus 1959), which is equally developed in the two genera. With the Olivinae Belloliva shares a canaliculate suture Figure 32. A-B, Belloliva brazieri (Angas, 1877), AMS C388726, dissected specimen (SL 12.9 mm). C-D, Belloliva leucozona (A. Adams and Angas, 1864), probably illustrated syntype, BMNH 1870.10.26.93 (SL 13.8 mm). E-F, Belloliva triticea (Duclos, 1835), illustrated syntype, MNHN 1273 (SL 10.6 mm), photo by D. Brabant. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA 7\ and, correspondingly, the presence of a mantle filament, a foot that is rounded posteriorly, a stomach that has a long posterior mixing area, and the radula type. The olivine radula is rather uniform (Troschel 1866, Marcus and Marcus 1959, Kantor and Tursch 2001), with rachidians with 3 non- serrated cusps and laterals that are leaf-shaped, concave pos- teriorly, and convex anteriorly with long curved hook-like tips (Fig. 30A). However, Belloliva (and Calyptoliva) differs from all Olivinae in having an operculum; it also differs from Oliva in the absence of the anterior mantle tentacle (which is also absent in Olivancillaria) and mantle lobe. The absence of tentacles on the vertical flaps on the head of Belloliva and Calyptoliva is a character shared with Olivan- cillaria, but not with Oliva. Calyptoliva differs from Olivinae in its open filament channel overlain by thin callus; super- ficial examination of two ancillariines (Entomoliva mirabilis Bouchet and Kilburn, 1991 and Amalda aureomarginata Kil- burn and Bouchet, 1988) revealed complex multi-layered structure of the shell but no sign of the filament channel. At this moment we do not know whether the peculiar suture of Calyptoliva should be regarded as ancestral (representing an intermediate stage between Olivinae and Ancillariinae), or an autapomorphy of Calyptoliva. With the Ancillariinae Belloliva and Calyptoliva share the presence of an operculum and the head morphology. Calyptoliva also shares with the Ancillariinae the suture cov- ered by a thin callus. Belloliva differs from Ancillariinae in having a channelled suture and a mantle filament. Both Bel- loliva and Calyptoliva differ from Ancillariinae in a stomach with a long posterior mixing area (the only genus of Ancil- lariinae studied in this respect, Amalda, has a narrow tubular U-shaped stomach without posterior mixing area), and a rounded versus posteriorly deeply notched foot. Ancillariine radulae (Fig. 30B-E) are much more variable than those of the Olivinae and essentially follow a genus-specific pattern; rachidians are multicuspid in Turrancilla Martens, 1903 (Fig. 30E) and Ancillina Bellardi, 1882 (= Gracilancilla Thiele, 1925) (Fig. 30C), or have three major cusps and numerous Table 1. Summary of major characters of Olivinae, Ancillariinae and Belloliva and Calyptoliva. Olivinae Ancillariinae Character Oliva* Olivancillaria** Belloliva Calyptoliva Amalda*** Suture of the shell Canaliculate Canaliculate Canaliculate Non-canaliculate, Non-canaliculate, overlaid by callus overlaid by callus Operculum Absent Absent Present Present Present Foot Rounded posteriorly Rounded posteriorly Rounded posteriorly Rounded posteriorly | Notched posteriorly Head morphology Vertical flaps with Vertical flaps tentacles without tentacles Anterior mantle Present Absent tentacle Mantle filament Present Present Mantle lobe Present Present Stomach With long posterior ? mixing area Vertical flaps without tentacles Vertical flaps Dorso-ventrally without tentacles compressed flaps without tentacles Absent Absent Absent Present Absent Absent Absent Present Present With long posterior With long posterior U-shaped, without mixing area mixing area posterior mixing area Rachidian of the radula Lateral teeth of the radula Tricuspid, with non-serrated CuSps Leaf-shaped, concave posteriorly, and convex anteriorly with long curved hook-like tips (Se) main non-serrated cusps and additional cusps abutting each side of the main lateral cusps Leaf-shaped, concave posteriorly, and convex anteriorly with long curved hook-like tips 3 main non-serrated cusps and additional cusps abutting each side of the main lateral cusps Leaf-shaped, concave posteriorly, and convex anteriorly with long curved hook-like tips main non-serrated (Sty cusps and additional cusps abutting each side of the main lateral cusps Leaf-shaped, concave posteriorly, and convex anteriorly with long curved hook-like tips 3 main serrated cusps Nearly flat, without complex, bent tips * Data based on Marcus and Marcus (1959), Kantor (1991), and Kantor and Tursch (2001). ** Data based on Marcus and Marcus (1959). *** Data based on Marcus and Marcus (1968), and Kantor (1991). 72 AMERICAN MALACOLOGICAL BULLETIN smaller denticles or serrations on the rachidian in Amalda (Fig. 30B, D). In addition, the lateral teeth of the Ancillar- linae are nearly flat, without complex, bent tips. The radulae of Belloliva and Calyptoliva are therefore much closer to that of Olivinae than to that of Ancillariinae. In conclusion, Belloliva and Calyptoliva share morpho- logical and conchological characters with both Olivinae and Ancillariinae (Table 1). The general similarity between Bel- loliva and Calyptoliva in most shell characters, external anatomy, anatomy of the digestive system, and radula is remarkable, and is not likely to be the result of convergence. Thus, their differing in the presence/absence of an open canaliculate suture puts into question the validity of this character, which was hitherto considered to be a fundamen- tal difference between, respectively, the Olivinae and Ancil- lariinae. Clearly, the validity of the two subfamilies requires examination of the anatomy of additional genera. ACKNOWLEDGEMENTS We thank Ian Loch, malacology collection superviser, Australian Museum, Sydney, for providing the anatomical material of Australian species of Belloliva, as well as infor- mation on holdings of Belloliva in the collection of the Aus- tralian Museum. Dr. Richard Kilburn assisted with the initial stage of this research and Jean Trondle shared with us his knowledge of Belloliva simplex. Nicolas Puillandre was re- sponsible for the COI sequence of Belloliva dorcas. The work was conducted during visiting curatorships of the senior author in Muséum national d'Histoire naturelle, Paris; he expresses his thanks to the staff of the malacology group, in particular Virginie Héros, Philippe Maestrati, and Dr. Pierre Lozouet for on-going assistance during his stays in Paris. We thank Dr. Alan Kohn and an anonymous referee for their time and efforts in improving the first version of the manuscript. LITERATURE CITED Adams, A. and G. F. Angas. 1864. Descriptions of new species of shells from the Australian Seas, in the Collection of George French Angas. Proceedings of the Zoological Society of London for 1863: 418-428. Angas, G. F. 1871 [December 5, 1870]. Descriptions of thirty-four new species of shells from Australia. Proceedings of the Zoo- logical Society of London for 1871: 13-21. Angas, G. F. 1877. Description of one genus and twenty-five species of marine shells from New South Wales. Proceedings of the Zoological Society of London for 1877: 171-177. Bouchet, P. and Yu. I. Kantor. 2004. New Caledonia: The major centre of biodiversity for volutomitrid molluscs (Mollusca: Neogastropoda: Volutomitridae). Systematics and Biodiversity 1: 467-502. 22 ° 1/2 * 2007 Bouchet, P. and R. Kilburn. 1991. A new genus of Ancillinae from New Caledonia, with the description of two new species. Bul- letin du Museum National d’Histoire Naturelle (4)A 12: 531- 539. Bouchet, P. and J.-P. Rocroi. 2005. Classification and nomenclator of gastropod families. Malacologia 47: 1-397. Dall, W. H. 1889. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877- 78) and in the Caribbean Sea (1879-80), by the U.S. Coast Survey Steamer “Blake,” Lietenant-Commander C. D. Sigsbee, U.S.N., and Commander J. R. Bartlett, U.S.N., commanding. Report on the Mollusca, Pt. 2: Gastropoda and Scaphopoda. Bulletin of the Museum of Comparative Zoology 18: 1-492, pls. 10-40. Duclos, P. L. 1835. Histoire naturelle générale et particuliére de tous les genres de coquilles univalves marines a l'état vivant et fossile publiée par monographies. Genre Olive. Didot, Paris. Hunon, C. 2000. Notes sur le genre et la forme adulte d'un Olividae récent du circalittoral philippin. Xenophora 91: 10-11. Iredale, T. 1924. Results from Roy Bell’s molluscan collections. Proceedings of the Linnean Society of New South Wales 49: 179-278. Johnson, R. I. 1994. Types of the shelled Indo-Pacific mollusks described by William Harper Pease (1824-71). Bulletin of the Museum of Comparative Zoology 154: 1-61. Kaicher, S. 1987. Card Catalogue of World-wide Shells. Pack # 49. Olividae. Part II. Published by the author, St. Petersburg, Florida. Kantor, Yu. I. 1991. On the morphology and relationships of some oliviform gastropods. Ruthenica, Russian Malacological Journal 1; 17-52. Kantor Yu. I. and P. Bouchet. 1999. A deep-sea Amalda (Gas- tropoda: Olividae) in the north-eastern Atlantic. Journal of Conchology 36: 11-16. Kantor, Yu. I. and B. Tursch. 2001. Morphology of Oliva. In: B. Tursch and D. Greifeneder, eds., The genus Oliva and the species problem. L’Informatore Piceno and Bosque BMT, S.A., Ancona. Pp. 75-102. Kilburn, R. N. 1977. Descriptions of new species of Amalda and Chiloptygma (Gastropoda: Olividae: Ancillinae) with a note on the systematics of Amalda, Ancillus and Ancillista. Annals of the Natal Museum 23: 13-21. Kilburn, R. N. and P. Bouchet. 1988. The genus Amalda in New Caledonia (Olividae, Ancillinae). Bulletin du Museum National ad Histoire Naturelle (4)A 10: 277-300. Marcus, E. and E. Marcus. 1959. Studies on “Olividae.” Boletins da Faculdade de Filosofia, Ciencias e Letras. Universidade de Sao Paulo 232 (Zoologia 22): 99-188. Marcus, E. and E. Marcus. 1968. On the prosobranchs Ancilla dimidiata and Marginella fraterculus. Proceedings of the Mala- cological Society of London 38: 55-69. Olsson A. A. 1956. Studies of the genus Olivella. Proceedings of the Academy of Natural Sciences of Philadelphia 108: 155-225. Pease, W. H. 1868. Descriptions of sixty-five new species of marine Gastropodae, inhabiting Polynesia. American Journal of Con- chology 3: 271-297. BELLOLIVA FROM THE CORAL SEA AND NEW CALEDONIA Peile, A. J. 1922. Some notes on radulae. Proceedings of the Mala- cological Society of London 15: 13-18. Richer de Forges, B. 1990. Explorations for bathyal fauna in the New Caledonian economic zone. In: A. Crosnier, ed., Résultats des Campagnes MUSORSTOM, Vol. 6. Mémoires du Museum National d’Histoire Naturelle (A) 145: 9-54. Richer de Forges, B. 1993. Campagnes d’exploration de la faune bathyale faites depuis mai 1989 dans la zone économique de la Nouvelle-Calédonie. Listes des stations. In: A. Crosnier, ed., Résultats des Campagnes MUSORSTOM, Vol. 10. Mémoires du Muséum National d@ Histoire Naturelle 156: 27-32. Richer de Forges, B. and C. Chevillon, 1996. Les campagnes d’echantillonnage du benthos bathyal en Nouvelle-Calédonie, en 1993 et 1994 (BATHUS 1 a 4, SMIB 8 et HALIPRO 1). In: A. Crosnier, ed., Résultats des Campagnes MUSORSTOM, Vol. 15. Mémoires du Muséum National d'Histoire Naturelle 168: 33-53. Sterba, G. H. W. 2003. Olividae: Fibel der Schalen (Mollusca, Neo- gastropoda). Selbstverlag Prof. Dr. Dr. Giinther H. W. Sterba, Kiel. Troschel, F. H. 1866. Das Gebiss der Schnecken zur Begriindung einer natiirlichen Classification, 2. Nicolaische Verlagbuchhandlung, Berlin. Thiele, J. 1929-1931. Handbuch der Systematischen Weictierkunde. Bd. 1. Gustav Fischer, Jena. Tursch, B. and L. Germain. 1985. Studies on Olividae. I. A mor- phometric approach to the Oliva problem. Indo-Malayan Zo- ology 1: 331-352. Tursch, B. and L. Germain. 1986. Studies on Olividae. II. Further protoconch morphometrical data for Oliva taxonomy. Apex 1: 39-45, Tursch, B. and D. Greifeneder. 2001. The genus Oliva and the Spe- cies Problem. L’Informatore Piceno and Bosque BMT, S.A., Ancona. Wilson, B. 1994. Australian Marine Shells, Vol. 2—Prosobranch Gastropods. Part two (Neogastropoda). Odyssey Publishing, Kallaroo, Australia. Accepted: 20 November 2006 Oe i : : Amer. Malac. Bull. 22: 75-82 Epibionts on Flexopecten felipponei (Dall, 1922), an uncommon scallop from Argentina Laura Schejter and Claudia S. Bremec CONICET and INIDEP, Paseo Victoria Ocampo 1, 7600, Mar del Plata, Argentina, schejter@inidep.edu.ar Abstract: Flexopecten felipponei (Dall, 1922) is a non-commercial, seldom reported pectinid from the SW Atlantic Ocean. In this contribution we review its taxonomy, describe epifaunal species and their levels of encrustation, and discuss the composition of the macrobenthic assemblage where this scallop lives. Eighteen epibiont taxa were observed to live on the valves of these scallops. The most frequent and abundant epibionts on F. felipponei were serpulids, barnacles, and oysters. Although both valves were encrusted, the left valves had higher percentages of coverage. The benthic community contained 69 invertebrate taxa that generally characterize other mid-shelf bottoms between 37°S and 39°S. Eight pea crabs of the species Tumidotheres maculatus (Say, 1818) were found inside eight individuals of F. felipponei. Two other scallops had burrows of Polydora webster Hartman, 1943. These were the first observations of these infestations on F. felipponei. Key words: Epibiosis, Pectinidae, SW Atlantic Ocean Scallops are distributed worldwide and support impor- tant commercial fisheries and mariculture efforts. They are one of the best known groups of bivalves. Numerous studies on the biology, anatomy, physiology, genetics, population dynamics, fishery, and aquaculture of commercial pectinids have been carried out (see Shumway and Parsons 1991). In Argentina, the commercial pectinids include Aequipecten te- huelchus (dOrbigny, 1846) and Zygochlamys patagonica (King and Broderip, 1832) (Ciocco et al. 2006), target species of a local fishery in the gulfs of northern Patagonia (Lasta ef al. 1998, Ciocco et al. 1998) and of a fishery that started in 1996 (Lasta and Bremec 1998), respectively. During the course of cruises conducted in 2002 to locate new commercial beds of Aequipecten tehuelchus in the coastal shelf waters of Buenos Aires, we observed the pres- ence of the non-commercial pectinid Flexopecten felipponei (Dall, 1922) as part of the benthic community. The species has rarely been recorded from the SW Atlantic Ocean (Waller 1991). It is distributed from 36°S to San Matias and Nuevo Gulfs (43°S), and has been collected in rocky and sandy bottoms from the lower tidal fringe (Castellanos 1970, 1971) and between 40 to 50 m depth (Rios 1994, Nunez Cortés and Narosky 1997). The only biological study on F. felipponei indicates that it is a simultaneous hermaphrodite (Penchaszadeh and Giménez 2001). The availability of a suitable substratum is one of the critical factors for the colonization of sessile species. Mol- luscs, decapod carapaces, and the spines of sea urchins are frequently used as hard substrata available for attachment of sessile organisms in soft bottoms, together with many other organisms such as ascidians, corals, gorgonians, and sea pens that are also used as surfaces for settlement by invertebrate larvae (Abello et al. 1990, Davis and White 1994, Gutt and Schickan 1998). Epibiosis is the association between epibi- onts (organisms growing attached to a living surface) and basibionts (organisms that provide substrate to the epibi- onts). This association creates a complex network of benefits and disadvantages for both organisms (Wahl 1989). Bivalves are often associated with encrusting epibionts (see Feifarek 1987, Vance 1978, Keough 1984). Epizoic organisms can be very diverse, especially on scallops (i.e, Waloszek 1991, Rosso and Sanfilippo 1994, Fuller et al. 1998, Bremec and Lasta 2002). Studies examining the epibiosis between scal- lops and other organisms include foraminiferans (Alexander and Delaca 1987), sponges (Evans 1969, Bloom 1975, For- ester 1979, Chernoff 1987, Burns and Bingham 2002, Dono- van et al. 2002), hydroids (Getchell 1991), polychaetes (Blake and Evans 1973, Bergman ef al. 1982, Mori et al. 1985, Ci- occo 1990, Sanfilippo 1994), crustaceans (Donovan ef al. 2003), bryozoans (Ward and Thorpe 1991), and ascidians (see Uribe et al. 2001 and references therein). In this contribution we give new information about Flexopecten felipponei from coastal shelf waters of Buenos Aires, Argentina. We review its taxonomy, describe epifaunal species and their levels of encrustation, and discuss the com- position of the macrobenthic assemblage where this scallop lives. MATERIALS AND METHODS Sampling was conducted with commercial otter trawls (51 hauls) by the scallopers Atlantic Surf I, Erin Bruce, and Mr. Big, between 40-50 m depth and between 39°00'- 76 AMERICAN MALACOLOGICAL BULLETIN 39°37'S and 60°21'- 58°47'W in February, July, August, and September 2002, and with a dredge by the research vessel Capitan Canepa (INIDEP) at 38°26'S and 57°40’W in Janu- ary 2004 (Fig. 1). Samples of the macrobenthic community were frozen on board. The species of macroinvertebrates comprising this community were identified to the lowest possible level in the laboratory using the available literature (Bernasconi 1964, 1973, Castellanos 1970, Orensanz 1975, Fauchald 1977, Bernasconi and D’Agostino 1977, Boschi et al. 1992, Lana and Bremec 1994, Roux and Bremec 1996, Pérez 1999, and Forcelli 2000). The identification of ascid- ians was made by Dr. Marcos Tatian. From a total of 95 specimens of Flexopecten felipponei identified in 26 hauls, we preserved 30 available specimens in 5% buffered formalin solution in seawater. Presence-absence and quantitative data of epibionts were recorded for right and left valves. A Wilcoxon matched paired test (Steel and Torrie 1985) was used to establish the significance in differ- ences between total abundances of epibionts on each valve. To quantify the level of encrustation, each valve was arbi- trarily divided into seven regions (Fig. 2), roughly following the procedures of Ward and Thorpe (1991) and Sanfilippo (1994). The percentage of coverage of each species of epibi- ont was estimated by eye as either <10%, 10-30%, or >30% of the surface for each region of each valve. Maximum shell width was measured to the nearest mm with calipers (Fig. 2). RESULTS Taxonomy Our study material agreed well with the original de- scription of Flexopecten felipponei (Dall, 1922). The genus belongs to the Decatopecten group (Waller 1991) and is characterized by plain shells with 5-8 ribs that are sometimes 41 Figure 1. Map of the sampling area. Triangles mark sampling sites. 22° 1/2 * 2007 Figure 2. Diagram showing division of each valve into 7 arbitrary regions and the measurement of maximum shell width. inconspicuous. Flexopecten felipponei is an uncommon spe- cies from the Argentine Sea. It has been synonymized as: Flexopecten felipponei (Dall, 1922) Pecten felipponei: Dall 1922; Carcelles 1944 Chlamys felipponet: Castellanos 1970,1971; Waloszek 1984; Rombouts 1991; Rios 1994; Nunez Cortés and Narosky 1997; Forcelli 2000 Aequipecten felipponei: Nunez Cortés and Narosky 1997; Penchaszadeh and Giménez 2001 Flexopecten felipponei: Waller 1991; Pena 2001 We follow the nomenclature proposed by Waller (1991), who assigns the species to the genus Flexopecten based on the external morphology of the shell. Epibiosis Epibionts were present on both valves of all studied individuals of Flexopecten felipponei. Tube-building poly- chaetes, barnacles, and oysters were the most frequent epi- zoic organisms on both valves (Fig. 3). Significant differ- ences were found in total number of epibionts recorded between both valves (Z = 3.3629, p < 0.001); the highest number of organisms was always found on the upper (left) valve. Serpulid tubes were present on 90% of the sampled scallops and were found on both valves (Fig. 4). These tubes were found in all 7 regions on both valves, with variable EPIBIONTS ON FLEXOPECTEN FELIPPONEI NI | Figure 3. Epizoic organisms on Flexopecten felipponei from the coastal shelf waters of Buenos Aires. A, Spirorbid polychaetes. B, Several individuals of Ostrea puelchana d’Orbigny, 1841, ascidians, and some serpulid tubes. C, Balanus cf. amphitrite. D, Balanus cf. amphitrite, ascidians, serpulid tubes, and individuals of Ostrea puelchana. E, Idanthyrsus armatus Kinberg, 1867 (Sabellariidae) and serpulid tubes. F, Ostrea puelchana, some with serpulid tubes on them. Abbreviations: As, ascidian; Bal, Balanus cf. amphitrite; la, Idanthyrsus armatus; Op, Ostrea puelchana; Ser, Serpulid polychaetes; Spd, Spirorbid polychaetes. percentages of covered surface in each region (Fig. 5A-B). In some cases the complete valve was covered, while in others only a few tubes were found (Fig. 3). The number of tubes found on a single valve varied between 1 and 65. Serpulids were also epibionts of other epibionts of Flexopecten felip- ponei. For example, they also occurred on epizoic individuals of Ostrea puelchana @Orbigny, 1842 (Fig. 3D-F). Only 3 small individuals (<26 mm maximum height) of F. felippo- net had valves that lacked serpulid tubes. Tubes of the poly- chaete Phyllochaetopterus sp. were found on both valves (left: 60%; right: 46.7%) (Fig. 4). These tubes were small and consequently covered small surfaces (<10%). They were found on both valves, more abundantly on the left (2-16 tubes) than on the right (1-5 tubes). Tubes of Idanthyrsus armatus Kinberg, 1867 were found only on left valves in 33.3% of the sampled scallops (Fig. 4). These tubes were found on the left valves in any of the 7 regions, but there was generally only one tube per valve. In some cases, the open region of the tube extended over the valve (Fig. 3E). Spiror- bid tubes were very abundant on 3 (10%) scallops belonging to a particular sample (Fig. 4). In any case, they were not frequent epibionts on F. felipponei; only 3 scallops had be- tween 47 and 224 tubes per valve, which were homoge- neously distributed on both valves (Fig. 3A). Members of the Eunicidae (Eunice magellanica Mc Intosh, 1885 and Eunice argentinensis |Treadwell, 1929]) were recorded on only a few scallops (Fig. 4). A few burrows of the parasitic polychaete Polydora websteri Hartman, 1943 were found on 2 left (up- per) valves (6.67%). Barnacles (Balanus cf. amphitrite) were found on 26 78 AMERICAN MALACOLOGICAL BULLETIN 22° 1/2 * 2007 D LEFT valve @ RIGHT valve 100 > 90 4 80 4 8 704 by = 60 4 3 50- ° 5 40 & 304 10 4 a LL [i Oe Cm cae Ce oo om |[(E : 3 . 3 g g iF - 3 S 2 o & 3b s S ra 2 §$ 38s 8 g S @ 9 2g 2 2 io} 2 9 g 2 g x so S£8 $8 so s 2 2 g B 8 & 2 E = 8 8 8 — 3s CE aa oa Ww — os od 72) eres 2 fie @ G18 € 3 z|* 2 f° D g£ Fg S S = s = oS | a = Ss Ss 3 & & & * Polychaetes Other Bivalves Crustaceans Other Epibionts Figure 4. Frequency of occurrence of epibionts on Flexopecten felipponei (based on presence-absence data, N = 30). (86.7%) left valves, but only on 5 (16.7%) of the right ones (Fig. 4). They were found most frequently on the left (upper) valves, with high values of coverage especially at regions 1, 2, 3, and 6 (Fig. 5C-D). The left valves had between 1 and 36 individual barnacles, and the right valves between 1 and 4 individuals. Region 1 of the valve was conspicuously pre- ferred as a settlement surface; in some cases it was com- pletely covered by barnacles (Fig. 3C). Oysters (Ostrea puelchana) were found on 43.3% and 33.3% of the left and right valves, respectively (Fig. 4). They were found on both valves and in all regions (Fig. 5E-F). In many cases, the oysters completely covered the auricular areas or extended over the edges of the scallops (Fig. 3F). Numbers of epizoic oysters varied between 1 and 9. Serpu- lids and individuals of Phyllochaetopterus sp. also encrusted epizoic oysters (Fig. 3B, D, F). Recruits of the mussel Mytilus edulis d Orbigny, 1846 and other unidentified small bivalves were recorded as epibionts on a few scallops (Fig. 4). Solitary ascidians were found on 16.7% and 26.7% of the left and right scallop valves, respectively (Fig. 4). These organisms were observed in low numbers (1-2), on both valves, and in all regions (7.e., Fig. 3B-D). Epizoic organisms such as bryozoan colonies, small isopods, and amphipods were also observed on a few scallops (Fig. 4). The crusta- ceans were free-living between the crevices in the association of epibionts. There were no shells without epibionts; even small in- dividuals had epizoic organisms on their valves. Macrobenthic assemblage Individuals of Flexopecten felipponei (between 16 and 90 mm maximum shell width) were primarily found associated with the tehuelche scallop Aequipecten tehuelchus, as part of the by-catch of the fishery. Other organisms of commercial importance found in the benthic community were the com- mon mussel Mytilus edulis, the oyster Ostrea puelchana, and the mussel Atrina seminuda (d’Orbigny, 1846). A total of 69 invertebrate taxa were recorded from the study area (Table 1). Infestation Seven females (between 6.6 and 9.6 mm carapace length) and one male (3.8 mm carapace length) of the pea crab Tumidotheres maculatus (Say, 1818) were found inside eight different specimens of Flexopecten felipponei. Two left scallop valves were burrowed into by the parasitic polychaete Polydora websteri. DISCUSSION We found 18 epizoic taxa on the valves of Flexopecten felipponei from the sublittoral of Buenos Aires. The most frequent and abundant epibionts on Flexopecten felipponei were serpulids, barnacles, and oysters. Additional organisms such as the gastropods Calliostoma sp., Crepidula spp. and Calyptraea sp. occured as part of the fauna closely related with the epibiont association. Previously, only the presence of bryozoans and polychaetes on five specimens of this scal- lop was mentioned by Castellanos (1971). The number of associated species greatly varies in dif- ferent species of scallops recorded from different habitats. Eleven species were found on cultured Euvola ziczac (Lin- naeus, 1758) and Nodipecten nodosus (Linnaeus, 1758) in Cariaco Gulf, Venezuela, but in Santa Catarina, Brazil, 16 EPIBIONTS ON FLEXOPECTEN FELIPPONEI 79 B 3 4 5 6 ig Cc D 15 15 12 9 N | 64 l a eae oO 1 2 3 4 5 6 7 F 15 4 12 q4 N Dn wal i oO om 1 2 3 4 5 6 rf RIGHT VALVE REGION LEFT VALVE REGION O<10% 10-30% @>30% O<10% 10-30% @>30% Figure 5. Portion of the surface (%) of each region of each valve of Flexopecten felipponei covered by: A-B, serpulids; C-D, barnacles; E-F, oysters. species were found on N. nodosus. In Magdalena Bay and Bahia de la Paz, Mexico, 36 species were found inside or outside the valves of Argopecten ventricosus (Sowerby II, 1842) and Nodipecten subnodosus (Sowerby, 1835). In Ton- goy, Guanaqueros, and Inglesa Bays, Chile, a total of 63 species were recorded associated with the valves of Argopec- ten purpuratus (Lamarck, 1819) (Uribe et al. 2001). Forty nine epizoic taxa were associated with the non-cultured spe- cies Placopecten magellanicus (Gmelin, 1791) in the Bay of Fundy, Canada (Fuller et al. 1998) and 19 sessile epizoic species were recorded on Zygochlamys patagonica in com- mercial beds of Argentina (Bremec and Lasta 2002, Bremec et al. 2003). Although both valves were encrusted, the left (upper) valves had higher percentages of coverage. Except for ascid- ians, the epibiont species were more abundant on the left valve and occured on all regions. Flexopecten felipponei is a non-sedentary species that should have limited swimming capacity (Stanley 1970), which would permit it to escape from predators, as observed in other scallops (see Wilkens 1991). Epibionts can settle on both valves, depending on the living position adopted by the scallop, which seems to be more frequently with the right valve in contact with the substrate. Surprisingly, no sponges were found on our specimens of Flexopecten felipponei. The symbiotic relationship between scallops and sponges has been studied worldwide (Bloom 1975, Forester 1979, Chernoff 1987, Burns and Bingham 2002, Donovan et al. 2002). Cover by sponges is believed to protect pectinids by camouflaging the shell and to reduce predation by asteroids by altering the surface texture of shells. Although we found sponges in the study area, they were encrusting other invertebrates, mainly crustaceans and eunicid tubes. The majority of the epibionts on Flexopecten felipponei were sessile suspension-feeders. They created additional sur- faces and crevices where other small free-living individuals, such as isopods and amphipods, could live. We found only a small number of free-living organisms inhabiting the epizoic association. However, we consider that the number of vagile species associated with this scallop is higher and underestimated due to the limitations of our sampling procedure. Many of the macroinvertebrates that were part of the benthic assemblage associated with Flexopecten felipponei were also recorded from other middle shelf bottoms between 37°S and 39°S where the mussel Mytilus edulis was domi- nant. The coastal area of Buenos Aires is highly heteroge- neous, with patches of different types of substrates, and the most diversified benthic assemblages are usually dominated by bivalves (Bremec and Roux 1997, Schejter and Bremec 2003). The settlement substrate and microhabitats provided by bivalves and associated epibionts greatly influence com- munity structure by increasing the species richness of the benthic assemblages. In our study area, where the soft bot- toms are subjected to hydrodynamic conditions that remove sediments, the scallops provided substrate for the settlement of encrusting filter feeders and permitted the colonization of coastal environments. This is the first record of Flexopecten felipponei as a host of the spionid polychaete Polydora websteri and the pea crab Tumidotheres maculatus. Species of the genus Polydora are reported to be a pest of bivalves (Getchell 1991). They have been found in many commercial species such as Placopecten magellanicus (Bergman et al. 1982), Argopecten purpuratus (Basilio et al. 1995), Patinopecten yessoensis (Jay, 1857) (Mori et al. 1985), Pecten maximus (Linnaeus, 1758) (Mortensen et al. 2000), and Aequipecten tehuelchus (Ciocco 1990). Pea crabs cause slight irritation to severe structural alterations and pathology in their scallop hosts (Kruckzynski 1972, Getchell 1991, Bologna and Heck 2000, Narvarte and Saiz 2004). Tumidotheres maculatus was also reported inside sev- 80 AMERICAN MALACOLOGICAL BULLETIN — 22° 1/2 * 2007 Table 1. Invertebrates recorded from the study area. PORIFERA Porifera unidentified CNIDARIA Tripalea clavaria (Studer, 1878) Actinaria unidentified Hydrozoa ANNELIDA Aphroditidae Eunice magellanica McIntosh, 1885 Chaetopterus variopedatus (Ranier, 1807) Phyllochaetopterus sp. Idanthyrsus armatus Kinberg, 1867 Polydora webstert Hartman, 1943 Spirorbidae Serpulidae Maldanidae Polychaeta unidentified MOLLUSCA Aequipecten tehuelchus (dOrbigny, 1846) Flexopecten felipponei (Dall, 1922) Ostrea puelchana VOrbigny, 1841 Pododesmus rudis (Broderip, 1834) Mytilus edulis @Orbigny, 1846 Atrina seminuda (dOrbigny, 1846) Panopea abbreviata (Valenciennes, 1839) Pitar rostrata (Koch, 1844) Bivalve unidentified Calyptraea sp. Crepidula spp. Calliostoma sp. Zidona dufresnei (Donovan, 1823) Fissurellidea megatrema @Orbigny, 1841 Nudibranchia Octopus tehuelchus d Orbigny, 1834 ARTHROPODA Peltarion spinosulum (White, 1843) Platyxanthus patagonicus A. Milne Edwards, 1879 Coenophthalmus tridentatus A. Milne Edwards, 1879 Rochinia gracilipes A. Milne Edwards, 1875 Collodes rostratus A. Milne Edward, 1878 Pilumnoides hasslert A. Milne Edward, 1880 Leurocyclus tuberculosus (H. Milne Edwards and Lucas, 1843) Libinia spinosa H. Milne Edwards, 1834 Pelia rotunda A. Milne Edwards, 1875 Leucipa pentagona H. Milne Edwards, 1833 Propagurus gaudichaudi (H. Milne Edwards, 1836) Pagurus sp. Pinnotheridae Tumidotheres maculatus (Say, 1818) Pinnixa brevipollex Rathbun, 1896 Balancus cf. amphitrite Lepadomorpha Amphipoda unidentified Isopoda unidentified Table 1. (Continued) ECHINODERMATA Arbacia dufresnei (Blainville, 1825) Pseudechinus magellanicus (Philippi, 1857) Astropecten brasiliensis Miller and Troschel, 1842 Luidia sp. Pterasteridae Asteroidea | Asteroidea 2 Asteroidea 3 Asteroidea 4 Ophioplocus januari (Liitken, 1856) Ophiactis asperula (Philippi, 1858) Ophiacanta vivipara Ljungman, 1870 BRACHIOPODA Magellania venosa (Solander, 1816) BRYOZOA Colonial Bryozoa unidentified CHORDATA Paramolgula gregaria (Lesson, 1830) Cnemidocarpa robinsont Hartmeyer, 1916 Pyura legumen (Lesson, 1830) Ascidiella aspersa (Miller, 1776) Sycozoa sigillinoides Lesson, 1830 Colonial Ascidacea unidentified eral mollusks, in the tunicate Molgula sp., inside tubes of the polychaete Chaetopterus variopedatus (Renier, 1804), and on the asteroid Asterias vulgaris Verril, 1866 (Fenucci 1975). ACKNOWLEDGMENTS We are grateful to Angel Marecos and Monica Di Pace for valuable help in the sampling work, to Dr. Enrique Bos- chi for helping in the identification of the pea crab, to Dr. Marcos Tatian for the identification of the ascidians, to Dr. Diego Zelaya and to INIDEP Librarians for helping with the bibliography, to Dr. Eduardo Spirak for the barnacle iden- tification and to Daniel Hernandez for statistical supervi- sion. The manuscript greatly benefited from comments and suggestions by Dr. Sandra Shumway, Dr. Janice Voltzow, and an anonymous reviewer. This paper was funded by PEI 6026, PICT 01-15080 and Antorchas 13900-13. This is INIDEP Contribution N° 1356. LITERATURE CITED Abelld, P., R. Villanueva, and J. M. Gili. 1990. Epibiosis in deep-sea crab populations as indicator of biological and behavioural characteristics of the host. Journal of the Marine Biological Association of the United Kingdom 70: 687-695. Alexander, S. P. and T. E. Delaca. 1987. Feeding adaptations of the EPIBIONTS ON FLEXOPECTEN FELIPPONEI 81 foraminiferan Cibides refulgens living epizoically and parasiti- cally on the Antarctic scallop Adamussium colbecky. Biological Bulletin 173: 136-159. Basilio, C., J. I. Canete, and N. Rozbaczylo 1995. Polydora sp. (Spi- onidae), un poliqueto perforador de las valvas del ostion Ar- gopecten purpuratus (Bivalvia: Pectinidae) en Bahia Tongoy, Chile. Revista de Biologia Marina de Valparaiso 30: 71-77. Bergman, K. M, R. W. Elner, and M. J. Risk. 1982. The influence of Polydora websteri borings on the strength of the shell of the sea scallop, Placopecten magellanicus. Canadian Journal of Zoology 60: 2551-2556. Bernasconi, I. 1964. Asteroideos argentinos. Claves para los ordenes, familias, subfamilias, y géneros. Physis 24: 241-277. Bernasconi, I. 1973. Los equinodermos colectados por el “Walter Herwig” en el Atlantico Sudoeste. Hidrobiologia 3: 287-334. Bernasconi, I. and M. M. D’Agostino. 1977. Ofiuroideos del mar epicontinental Argentino. Revista del Museo Argentino de Ciencias Naturales “Bernardino Rivadavia” e Instituto Nacio- nal de Investigaciones de las Ciencias Naturales. Hidrobiologia 5: 65-114. Blake, J. A. and J. W. Evans. 1973. Polydora and related genera as borers in mollusk shells and other calcareous substrates (Polychaeta: Spionidae). Veliger 15: 235-249. Bloom, S. A. 1975. The motile escape response of a sessile prey: A sponge-scallop mutualism. Journal of Experimental Marine Bi- ology and Ecology 17: 311-321. Bologna, P. A. and K. L. Heck, Jr. 2000. Relationship between pea crab (Pinnotheres maculatus) parasitism and gonad mass of the bay scallop (Argopecten irradians). Gulf and Caribbean Re- search 12: 43-46. Boschi, E. E., K. Fischbach, and M. I. Lorio. 1992. Catalogo ilus- trado de los crustaceos estomatopodos y decapodos marinos de Argentina. Frente Maritimo 10: 1-94. Bremec, C. S. and M. L. Lasta. 2002. Epibenthic assemblage asso- ciated with scallop (Zygochlamys patagonica) beds in the Ar- gentine shelf. Bulletin of Marine Science 70: 89-105. Bremec, C. and A. Roux. 1997. Resultados del analisis de una campana de investigacion pesquera sobre comunidades bentonicas asociadas a bancos de mejillon (Mytilus edulis pla- tensis D’Orb.) en costas de Buenos Aires, Argentina. Revista de Investigacion y Desarrollo Pesquero 11: 153-166. Bremec, C., A. Marecos, L. Schejter, and M. Lasta. 2003. Guia Técnica para la Identificacion de Invertebrados Epibentonicos Asociados a los Bancos de Vieira Patagoénica (Zygochlamys patagonica) en el Mar Argentino, Publicaciones Especiales INIDEP. Mar del Plata. Burns, D. O. and B. L. Bingham. 2002. Epibiotic sponges on the scallops Chlamys hastata and Chlamys rubida: Increased sur- vival in a high-sediment environment. Journal of the Marine Biological Association of the United Kingdom 82: 961-966. Carcelles, A. 1944. Catalogo de los moluscos marinos de Puerto Quequén. Revista del Museo de La Plata (nueva serie), Zoologia 23: 233-309. Castellanos, Z. A. J. bonaerenses. Anales de la Comision de Investigaciones Cienti- ficas de Buenos Aires 8: 1-365. 1970. Catalogo de los moluscos marinos Castellanos, Z. A. J. 1971. Los Chlamys mas comunes del Mar Argentino. Neotropica 17: 55-66. Chernoff, H. 1987. Factors affecting mortality of the scallop Chla- mys asperrima (Lamarck) and its epizoic sponges in South Australian waters. Journal of Experimental Marine Biology and Ecology 109: 155-172. Ciocco, N. F. 1990. Infestacion de la vieyra tehuelche (Chlamys tehuelcha (d‘Orbigny)) por Polydora websteri Hartman (Polychaeta: Spionidae) en el golfo San José (Chubut: Argen- tina): Un enfoque cuantitativo. Biologia Pesquera 19: 9-18. Ciocco, N. F., M. L. Lasta, and C. S. Bremec. 1998. Pesquerias de bivalvos: Mejillon, vieiras (tehuelche y patagonica) y otras especies. El Mar Argentino y sus Recursos Pesqueros 2: 143-166. Ciocco, N., M. L. Lasta, M. Narvarte, C. Bremec, E. Bogazzi, J. Valero, and J. M. Orensanz. 2006. Argentina. In: S. E. Shum- way and G.J. Parsons, eds., Scallops: Biology, Ecology and Aquaculture, 2"! Edition. Elsevier, Amsterdam. Dall, W. H. 1922. Two new bivalves from Argentina. The Nautilus 36: 58-59. Davis, A. R. and G. A. White. 1994. Epibiosis in a guild of sessile subtidal invertebrates in south-eastern Australia: A quantita- tive survey. Journal of Experimental Marine Biology and Ecology 177: 1-14. Donovan, D. A., B. L. Bingham, H. M. Farren, R. Gallardo, and V. L. Vigilant. 2002. Effects of sponge encrustation on the swim- ming behaviour, energetics and morphometry of the scallop Chlamys hastata. Journal of the Marine Biological Association of the United Kingdom 82: 469-476. Donovan, D. A., B. L. Bingham, M. From, A. F. Fleisch, and E. S. Loomis. 2003. Effects of barnacle encrustation on the swim- ming behaviour, energetics, morphometry, and drag coeffi- cient of the scallop Chlamys hastata. Journal of the Marine Biological Association of the United Kingdom 83: 813-819. Evans, J. W. 1969. Borers in the shell of the sea scallop, Placopecten magellanicus. American Zoologist 9: 775-782. Fauchald, K. 1977. The polychaete worms: Definitions and key to the orders, families and genera. Natural History Museum of Los Angeles County Science Series 28: 1-190. Feifarek, B. P. 1987. Spines and epibionts as antipredator defenses in the thorny oyster Spondylus americanus Hermann. Journal of Experimental Marine Biology and Ecology 105: 39-56. Fenucci J. L. 1975. Los cangrejos de la familia Pinnotheridae del litoral argentino (Crustacea, Decapoda, Brachyura). Physis (A)34: 165-184. Forcelli, D. O. 2000. Moluscos Magallanicos. Guia de Moluscos de Patagonia y Sur de Chile, Vol. 2. Vazques Mazzini, Buenos Aires. Forester, A. J. 1979. The association between the sponge Halichon- dria panicea (Pallas) and scallop Chlamys varia (L. ): A com- mensal protective mutualism. Journal of Experimental Marine Biology and Ecology 36: 1-10. Fuller, S., E. Kenchington, D. Davis, and M. Butler. 1998. Associated Fauna of Commercial Scallop Grounds in the Lower Bay of Fundy. Marine Issues Committee, Special Publication 2. Ecol- ogy Action Centre, Halifax. Getchell, R. G. 1991. Diseases and parasites of scallops. In: S. E. 82 AMERICAN MALACOLOGICAL BULLETIN Shumway, ed., Scallops: Biology, Ecology and Aquaculture, Elsevier, Amsterdam. Pp. 471-494. Gutt, G. and T. Schickan. 1998. Epibiotic relationships in the Ant- arctic benthos. Antarctic Science 10: 398-405. Keough, M. J. 1984. Dynamics of the epifauna of the bivalve Pinna bicolor: Interactions among recruitment, predation and com- petition. Ecology 65: 677-688. Kruczynski, W. L. 1972. The effect of the pea crab, Pinnotheres maculatus Say, on the growth of the bay scallop, Argopecten irradians concentricus Say. Chesapeake Science 13: 218-220. Lana, P. C. and C. S. Bremec. 1994. Sabellariidae (Annelida: Polychaeta) from South America. In: J. C. Dauvin, L. Laubier, and D. J. Reish, eds., Actes de la 4°" Conférence Internationale des Polychétes. Mémoires du Muséum National d’Histoire Na- turelle 162: 211-222. Lasta, M. L. and C. Bremec. 1998. Zygochlamys patagonica in the Argentine Sea: A new scallop fishery. Journal of Shellfish Re- search 17: 103-111. Lasta, M. L., N. F. Ciocco, C. S. Bremec, and A. M. Roux. 1998. Moluscos bivalvos y gaster6podos. El Mar Argentino y sus Recursos Pesqueros 2: 115-142. Mori, K., W. Sato, T. Nomura, and M. Imajima. 1985. Infestation of the Japanese scallop Patinopecten yessoensis by boring polychaetes, Polydora, on the Okhotsk Sea coast of Hokkaido, especially in Abashiri waters. Bulletin of the Japanese Society of Science and Fisheries 51: 371-380. Mortensen, S., T. Van der Meeren, A. Fosshage, I. Hernar, L. Hake- stad, L. Torkildsen, and O. Bergh. 2000. Mortality of scallop spat cultivation, infested with tube dwelling bristle worms, Polydora sp. Aquaculture International 8: 267-271. Narvarte, M. A. and M. N. Saiz. 2004. Effects of the pinnotherid crab Tumidotheres maculatus on the tehuelche scallop Aequipecten tehuelchus in the San Matias Gulf, Argentina. Fish- eries Research 67: 207-214. Nunez Cortés, C. and T. Narosky. 1997. Cien caracoles argentinos. Albatros, Buenos Aires. Orensanz, J. M. 1975. Los anélidos poliquetos de la Provincia Bio- geografica Argentina. VII. Eunicidae y Lysaretidae. Physis 34: 85-111. Pena, J. B. 2001. Taxonomia, morfologia, distribucion, y habitat de los pectinidos Iberoamericanos. In: A. N. Maeda-Martinez, ed., Los Moluscos Pectinidos de Iberoamérica: Ciencia y Acut- cultura. McGraw-Hill, México. Pp. 1-25. Penchaszadeh, P. E. and J. Giménez. 2001. Sexuality in three South American Atlantic species of the genus Aequipecten. In: J. E. Illanes, ed., Book of Abstracts of the 13" International Pectinid Workshop. Universidad Catolica del Norte, Coquimbo, Chile. Pp. 135-136. Pérez, C. D. 1999. Taxonomia, Distribucion y Diversidad de los Pennatulacea, Gorgonacea y Alcyonacea del Mar Epicontinental Argentino y Zonas de Influencia. Ph.D. Dissertation, Universi- dad Nacional de Mar del Plata, Argentina. Rios, E. 1994. Seashells of Brazil. Fundacion de la Universidad de Rio Grande do Sul, Rio Grande, Brazil. Rombouts, A. 1991. Guidebook to Pecten Shells. Recent Pectinidae 22° 1/2 + 2007 and Propeamusstidae of the World. W. Backhuys, Universal Book Services, Oegstgeest, Netherlands. Rosso, A. and R. Sanfilippo. 1994. Epibionts distribution pattern of Chlamys patagonica (King and Broderip) of the Magellan Strait. Memorie di Biologia Marina e di Oceanografia 19: 237- 240. Roux, A. and C. Bremec. 1996. Brachiopoda collected in the west- ern South Atlantic by R/V Shinkai Maru Cruises (1978-1979). Revista de Investigacion y Desarrollo Pesquero 10: 109-114. Sanfilippo, R. 1994. Polychaete distribution patterns on Chlamys patagonica of the Magellan Strait. In: J. C. Dauvin, I. Laubier, and D. J Reish, eds., Actes de la 4°”" Conférence Internationale des Polychétes. Mémoires du Muséum National d Histoire Na- turelle 162: 535-540. Schejter, L. and C. Bremec. 2003. Fauna bentonica de plataforma media bonaerense: Prospeccion de especies de interés comer- cial. In: E. E. Boschi, ed., Libro de Restimenes de las V Jornadas Nacionales de Ciencias del Mar. Universidad Nacional de Mar del Plata, Mar del Plata, Argentina. P. 169. Shumway S. E., ed. 1991. Scallops: Biology, Ecology and Aquaculture. Elsevier, Amsterdam. Stanley, S. M. 1970. Relation of shell form to life habits in the Bivalvia (Mollusca). Memoirs of the Geological Society of America 125: 1-296. Steel, R. G. D. and J. H. Torrie. 1985. Bioestadistica: Principios y procedimientos, 2°" Ed. McGraw-Hill, Bogota. Uribe, E., C. Lodeiros, E. Félix-Pico, and I. Etchepare. 2001. Epibi- ontes en pectinidos de Iberoamérica. In: A. N. Maeda- Martinez, ed., Los Moluscos Pectinidos de Iberoamérica: Ciencia y Acuicultura. McGraw-Hill, México. Pp. 249-266. Vance, R. R. 1978. A mutualistic interaction between a sessile ma- rine clam and its epibionts. Ecology 59: 679-685. Wahl, M. 1989. Marine epibiosis. I. Fouling and antifouling: Some basic aspects. Marine Ecology Progress Series 58: 175-189. Waller, T. R. 1991. Evolutionary relationships among commercial scallops (Mollusca: Bivalvia: Pectinidae). In: S. E. Shumway, ed., Scallops: Biology, Ecology and Aquaculture. Elsevier, Am- sterdam. Pp. 1-73. Walossek, D. 1991. Chlamys patagonica (King and Broderip, 1832), a long “neglected” species from the shelf off the Patagonia coast. In: S. E. Shumway and P. A. Sandifer, eds., World Aqua- culture Workshops I. An International Compendium of Scallop Biology and Culture. The World Aquaculture Society, Baton Rouge. Pp. 256-263. Waloszek, D. 1984. Variabilitaét, taxonomie und Verbreitung von Chlamys patagonica (King and Broderip, 1832) und An- merkungen zu weiteren Chlamys-Arten von der Siidspitze Siid-Amerikas (Mollusca, Bivalvia, Pectinidae). Verhandlun- gen des naturwissenschaftlichen Vereins Hamburg 27: 207-276. Ward, M. A. and J. P. Thorpe. 1991. Distribution of encrusting bryozoans and other epifauna on the subtidal bivalve Chlamys opercularis. Marine Biology 110: 253-259. Wilkens, L. A. 1991. Neurobiology and behaviour of the scallop. In: S. E. Shumway, ed., Scallops: Biology, Ecology and Aquaculture, Elsevier, Amsterdam. Pp. 429-470. Accepted: 3 February 2006 Amer. Malac. Bull. 22: 83-87 Partulids on Tahiti: Differential persistence of a minority of endemic taxa among relict populations* Trevor Coote BP 2407, Papeete 98713, Tahiti, French Polynesia, partula2003@yahoo.co.uk Abstract: The extinction of many species of endemic land snails in French Polynesia because of the introduction of the carnivorous snail Euglandina rosea is a salutary lesson in hasty biological control undertaken without adequate scientific field trials. Fewer than 20 of the original 70+ nominal species of the family Partulidae in French Polynesia survive. In 2004 surveys were carried out in nearly 70 of the valleys of Tahiti. All of the populations found were of Partula hyalina, the closely related Partula clara, or Samoana attenuata. No individuals of the Partula otaheitana/Partula affinis complex were found, yet P. otaheitana, together with Samoana burchi, still survive in many montane forest areas (over 1000 m altitude), while P. affinis persists on the Peninsula of Tahiti. Partula nodosa, with a previous distribution of just 7 valleys, has most likely been extirpated in the wild but persists well in captive populations. The species Partula filosa, Partula producta, and Partula cytherea (each previously inhabiting a single valley) are almost certainly extinct, as is Samoana jackieburchi. Key words: Extinctions, Partulidae, biological control, Tahiti, Euglandina The endemic land snails of French Polynesia constitute a significant component of its biodiversity. Their polymor- phism in shell color, banding patterns, and coiling has been the focus of classic studies in evolutionary and ecological genetics (Crampton 1916, 1932, Johnson et al. 1993). In the past their shells have been collected by local artisans on some islands for making shell jewelry (E. Loeve, pers. comm.). The mass extinction of many of the endemic species is an ex- ample of a disastrous attempt at biological control without adequate field trials (Clarke et al. 1984, Cowie 1992). The giant African snail Lissachatina fulica (Bowdich, 1822) was introduced as a food resource in 1967, but spread rapidly and destructively. The solution at the time was wrongly per- ceived to be the introduction of the carnivorous snail Eug- landina rosea (Férussac, 1821), which took place on Tahiti in 1974, on Moorea in 1977, and on other islands in the 1980s and 1990s. The predators preferentially attacked the smaller endemic species, notably members of the family Partulidae. Over half of the 120 known species of the family Partulidae were native to French Polynesia. Sadly now most are extinct (Murray et al. 1988). An international captive breeding effort—the only one in the world for an invertebrate family—has continued to maintain a number of species of Partula that no longer exist in the wild (Pearce-Kelly et al. 1997). An ad hoc program of field surveys has been carried out in the Society Islands of * From the symposium “Pacific Island Land Snail Diversity: Origins and Conservation” presented at the annual meeting of the Ameri- can Malacological Society, held 26-30 June 2005 at Asilomar, Pa- cific Grove, California, U.S.A., and supported by the National Sci- ence Foundation. 83 French Polynesia over the last few years to establish the exact status of remaining species in their natural habitat and to locate any relict populations that may have survived the rav- ages of the last 20 years. These surveys have concluded that there is a high probability of virtually all species being extinct on the islands of Bora Bora, Huahine, Raiatea, and Tahaa, and that only Moorea and Tahiti still support remnant populations (Coote and Loeéve 2003), though there have been recent discoveries of a few partulid individuals on the highest peaks of Huahine and Raiatea (J.-Y. Meyer, pers. comm.). This paper concentrates on a major effort to survey Tahiti Nui, which comprises the bulk of the island of Ta- hiti—the peninsula of Tahiti Iti has yet to be surveyed. Tahiti is by far the largest island in French Polynesia. The American biologist, H. E. Crampton (1916) made an extensive study-collection of the genus Partula from over 60 valleys (50 on Tahiti Nui and 12 on Tahiti Iti) between 1906 and 1909. The malacologists Y. Kondo and J. B. Burch col- lected from 33 valleys in 1970 (Anonymous 2004). With the introductions of Euglandina rosea to Tahiti in 1974, the first such introduction in French Polynesia, focus changed from pure research to conservation. While undertaking large-scale surveys and emergency collections on Moorea, Murray et al. (1988) also undertook smaller surveys of valleys on Tahiti. They searched 11 valleys on visits between 1980 and 1987. Extrapolating from the situation on Moorea, they believed that all species from Tahiti would be extinct within a few years, as many of the valleys were found to be empty already. However, in 1995 a visiting team of biologists from the U.K. and U.S.A., acting on local advice, found thriving popula- tions on Mt. Marau and in the valleys of Te Pari (“the cliffs”) on Tahiti Iti. Since that year, a number of isolated popula- 84 AMERICAN MALACOLOGICAL BULLETIN — 22* 1/2 * 2007 tions have been discovered at different locations on Tahiti, some of which have since disappeared (Coote et al. 1999). METHODS Although the nature of the habitat and terrain would have changed almost beyond recognition since his time, the information in Crampton (1916) has still formed the basis of the surveys reported here. Details were refined on the ground, with 1:20,000 scale maps and advice from local people. The majority of the surveys took place between Janu- ary and September 2004, in the hotter rainy season and the cooler drier season. Each survey of a valley was restricted to a single day. Because of the extreme rarity (and possible non-existence) of partulids in the valleys, simple searches for their presence or absence were carried out in habitats that experience suggested were most amenable to their survival and which were accessible. Using existing forest trails where available and continuing as deep as possible into the valleys, the areas searched were usually patches of around 5 m* in size. Where populations were found, all snails seen within the immediate patch were counted inside 20 minutes and descriptive details recorded. Dead shells were collected; those of Euglandina rosea, being large and conspicuous, were easily seen. RESULTS Seventy-six valleys on Tahiti Nui were identified and 69 of them were surveyed (including 22 on two or more occa- sions). Access could not be gained to the other 7 valleys. Live populations of partulids were found in 22 valleys (Table 1). There was no obvious geographical pattern to these finds. They occurred in the north, east, west, and south, in Table 1. Valleys on Tahiti Nui where partulids survive. No. of Individuals counted in Valley Administrative commune — Species found populations 20 minute search Valley type Fautaua-Faaiti Papeete Partula hyalina l 10 Small, dry Pirae Arue Partula hyalina l * Large, dry Ahonu Mahina Partula hyalina ] 5 Medium, wet Puhi Hitia’a O Te Ra, Papenoo — Partula clara l 2 Small, wet Faarapa Hitia’a O Te Ra, Papenoo — Partula hyalina, Partula clara Several 5 (hy) 3 (hy) 4 (cl) Small, wet Vaipu (Faaromai) Hitia’a O Te Ra, Tiarei Partula hyalina 1 2 Small, wet Haapoponi Hitia’a O Te Ra, Tiarei Partula hyalina, Partula clara, Several 9 (hy) 5 (hy) 3 (cl) Small, wet Samoana attenuata 8 (hy) 5 (hy) 4 (cl) Onohea-Faaiti Hitia’a O Te Ra, Tiarei Partula hyalina, Partula clara 2 12 (hy) 17 (cl) Small, wet Tahaute Hitia’a O Te Ra, Mahaena = Partula hyalina 1 2 Large, wet Faaiti Hitia’a O Te Ra Hitia’a Partula hyalina 1 2 Small, wet Vaiiha (Papeiha) Hitia’a O Te Ra, Hitia’a Partula hyalina, Partula clara’ 1 11 (hy) 2 (cl) Large, wet Vaitoare Taiarapu Est, Faaone Partula hyalina, Partula clara Several 1 (hy) 5 (hy) 1 (cl) Small, wet 18 (hy) 2 (cl) Apirimaue* Teva-I-Uta, Papeari Samoana attenuata ] — Large, wet Vaioo Teva-I-Uta, Mataiea Partula clara 2 2 Small, wet 6 Faurahi* Teva-I-Uta, Mataiea Partula clara’ 1 — Medium, wet Taapua Papara Partula clara ] 1 Medium, wet Tereia* Papara Samoana attenuata 1 — Medium, wet Tiapa (Hopuetama) Paea Partula clara, Samoana 1 14 (hy) 1 (at) Medium, dry attenuata Papehue Paea Partula clara, Samoana ] 1 (cl) 1 (at) Medium, dry attenuata Maruapo Punaauia Partula clara 1 4 Small, dry Matatia Punaauia Partula hyalina 1 3 Small, dry Tipaerui Faa’a Samoana attenuata 1 1 Medium, dry * Population found by W. Teamotuaitau. oe ; : : Live Euglandina rosea found in valley. hy = Partula hyalina; cl = Partula clara; at = Samoana attenuata. PARTULIDS ON TAHITI 85 valleys big and small, dry and wet (Table 1), although there was a slight tendency for there to be more living snails in the small, wet valleys of the east coast. On the other hand, there was a striking pattern in the persisting species. Apart from a few individuals of the rare Samoana attenuata (Pease, 1864), the only snails discovered were of Partula hyalina (Broderip, 1832) and the very similar Partula clara (Pease, 1864) (Fig. 1). Nearly 76% of the populations of P. hyalina and 40 % of P. clara were found on one native plant species, the wild red ginger Etlingera cevuga (opuhi maohi). Live individuals of the carnivore Euglandina rosea were found in just 10 valleys, including 3 adjacent valleys in Pa- peari. It does not appear to be an immediate threat to any of the surviving valley populations of partulids on Tahiti Nui. In no valley did I find more than one individual of E. rosea. In addition to the valleys, five mountain tracks on Tahiti Nui and one above Taravao Plateau on Tahiti Iti were sur- veyed. The conditions for snails are very different above 1000 m elevation compared to the lowland areas of dis- turbed habitat (Fosberg 1992). In contrast to the valleys, the dominant partulid species surviving in montane forests is Partula otaheitana (Bruguiere, 1792). The largest populations still survive above 1000 m on Mt. Marau. These consist principally of different forms of Partula otaheitana, but also Samoana burchi (Kondo, 1973), a species believed extinct (Murray et al. 1988) but recently confirmed by molecular analysis (T. Lee, pers. comm.). Sev- eral populations of more than 30 individuals still survive in patches along the trail to Mt. Aorai, although individuals of Euglandina rosea are often seen at around 800 m altitude and also at 1200 m encroaching into partulid populations. No live partulids were found on the route to Mt. Mauru (where they were last seen in 2002) or from the Sentier de Milles Source to the summit of Mt. Pihaiateta (600 m to 1400 m), although they have since been seen on the latter trail (B. S. Holland, pers. comm.). In neither place did it appear that the populations had fallen victim to E. rosea. In contrast, along the trail from Pic Rouge to Masif du Pic Vert, there were live individuals of E. rosea and many empty shells of partulids. Above Plateau Taravao on Tahiti Iti, populations of partulids persist, even though E. rosea is also present. DISCUSSION The discovery in 1995 of apparently thriving popula- tions of species of partulids previously believed extinct was a surprise (Murray et al. 1988). This led to a renewed search for endemic snails in 2001. Most of these searches took place at high altitudes (over 1000 m) where surveys and informa- tion from local biologists and mountain guides confirmed the presence of small isolated populations on the mountains Mauru, Tahiti, Aorai, and Atara and on the plateaus Faufiru and Terepo. The presence of the majority of endemic plant species at high altitude meant that these areas are regularly visited by botanists and good information was available. Be- cause shells of Euglandina rosea but no live animals had been found among the partulid populations of Mt. Marau, it ap- peared that the predatory snail had reached the area but had not thrived there. This led to the suggestion that an altitu- dinal ceiling may exist for the principal agent involved in the extinction of Society Island partulids (Gerlach 1994). This did not, however, explain the abundant partulid populations remaining at sea level in the Te Pari district of Taiarapu Peninsula (Tahiti Iti). Whereas in 1995 there had been no evidence of the predator in this small arc of valleys in the extreme southeast, in the following years EF. rosea spread quickly until by 2001 it had reached every valley. An upgrade in the existing level of protective legislation has been proposed in order to safeguard the populations of partulids surviving above 1000 m altitude (Meyer et al. 2005) because the terrain would not be amenable to physical pro- tection in the form of predator-proof protected area, a method being tested at lower elevations (Coote et al. 2004). However, if there were any surviving populations threatened by Euglandina rosea at low altitudes in the valleys of Tahiti, then measures such as reserves could be considered as real- istic options. Because of the size and topographical nature of Tahiti, a systematic search of the valleys required a long time and extensive planning. Until the resources became available in 2003, these barely seemed to be viable options. Given the timescale of the contract, valleys on Tahiti Iti were unable to be included in the surveys. It became clear after the first few discoveries of valley populations that only three species persisted and that all the others had most likely been extirpated. By far the most com- mon species were the universally white Partula hyalina and the very similar Partula clara, which is polymorphic in shell color. The third species, Samoana attenuata, has a distribu- tion and ecology that differs from most of the species in the genus Partula. It has always been a rare species, made elusive by the fact that it favors higher branches in the trees (Crampton 1916). It has also survived on the neighboring island of Moorea (Coote 1999). In the lowland forests of Tahiti it was represented in the current surveys by just a few individuals that were found occasionally. No individuals of Partula otaheitana or Partula affinis (Pease, 1867) were found in any of the valleys of Tahiti Nui, yet in Crampton’s 1906-09 collections P. otaheitana (of which at the time P. affinis was considered a subspecies) formed over 90% of all valley collections, except in the 7 valleys that had Partula nodosa (Pfeiffer, 1851) (Crampton 1916). Partula nodosa is now considered as extinct in the 86 AMERICAN MALACOLOGICAL BULLETIN wild, although it persists well in captivity (P. Pearce-Kelly, pers. comm.). In contrast, Partula hyalina and Partula clara together accounted for just over 5% of those same collec- tions. Ratios of species similar to those reported by Cramp- ton were found by researchers up until the introduction of Euglandina rosea in 1974 (J. B. Burch, pers. comm.). Emerg- ing molecular evidence suggests that P. hyalina and P. clara should be synonymized (D. O Foighil, pers. comm.). Tahiti Nui is divided ecologically into those valleys on the dry leeward side of the island, roughly between Mahina in the north and Pointe Maraa in the southwest, and the wet valleys that constitute the remainder. The flora is quite dif- ferent in the two regions. Most noticeable is the distribution of the dominant alien plant pest species, Miconia calvescens, abundant in the eastern valleys but rarer in the drier western ones. In terms of partulid distribution, Crampton (1916) maintained that Partula hyalina occurred across the whole island, but preferred drier areas, and Partula clara was absent from the driest quadrant. However, there appeared little preference for any type of valley for either species, both being found in dry and wet, large and small. The densest and most widespread populations of both these species were, however, found in the small, wet valleys of the northeast. Not enough is known about the ecology of the surviving species to determine why they should have escaped to some degree the ravages of Euglandina rosea, which has left extinct so many others across the Society Islands (Clarke et al. 1984, Murray et al. 1988, Cowie 1992). Partula hyalina, however, differs from the other species on Tahiti in that it is the only French Polynesian species of Partula that was not a single island endemic, occurring also on four of the Austral Islands and three of the Cook Islands (Crampton 1916). Because of this distribution, it was believed to be an ancient species (Crampton 1916), especially because it was universally white, in contrast to the conspicuous polymorphism of many other partulids in the Society Islands (Crampton 1916, 1932). However, recent genetic analysis confirms that some individuals of P. hyalina from the Austral Islands share mi- tochondrial haplotypes with those on Tahiti, suggesting evo- lutionarily recent among-archipelago dispersal (D. O Foi- ghil, pers. comm.). Partula hyalina was generally the first species to be seen on entering valleys—in other words, it more easily tolerated disturbed (and drier) habitats (Cramp- ton 1916). The finds during the current surveys confirm this ecological distribution. As a result of these surveys the distribution of the re- maining partulid species on Tahiti Nui can best be summa- rized thus: below 250 m altitude, Partula hyalina, Partula clara, and Samoana attenuata; between 250 m and 1000 m, none; above 1000 m, Partula otaheitana and Samoana bur- chi. Partula affinis seems to have disappeared from Tahiti Nui, but still survives on Tahiti Iti. Partula nodosa has been 22° 1/2 * 2007 ~—— Approximate boundary of dry/wet valley types r \ ‘ \ Partula clara Figure 1. Current distributions of Partula hyalina and Partula clara on Tahiti. extirpated from the wild but is maintained in captivity. The two single-valley endemics, Partula filosa (Pfeiffer, 1851) and Partula producta (Pease, 1864), as well as two rare species of unclarified taxonomy, Partula cytherea (Crampton and Cooke, 1930) and Samoana jackieburchi (Kondo, 1981) are almost certainly extinct. The surviving populations of par- tulids in the valleys of Tahiti do not appear to be under any immediate threat apart from that of external forces acting on small population size, generally less than 50 individuals. A number of colonies are being regularly monitored for changes in their populations or their habitats. ACKNOWLEDGMENTS I am extremely grateful to the Direction de ’Environnement de Polynésie frangaise which provided the funds that enabled this work to be carried out. In the field I thank especially Walter Teamotuaitau who accompanied me on several of my surveys, advised me on the flora of Tahiti, and independently discovered populations of partulids. I also thank Eric Lenoble who accompanied me later in the year and Teuira Tepuhiarii and Ioane Toa for two days of their time in Faaiti Valley, Hitia‘a. I thank Diarmaid O Foi- ghil and Taehwan Lee for molecular information, Jack Burch for information on Tahitian partulids, and Bryan Clarke and Jim Murray for advice on the manuscript. I am most grateful to the people of Tahiti who allowed me unrestricted access to their property. PARTULIDS ON TAHITI LITERATURE CITED Anonymous 2004. Endangered Tahitian snails found in Michigan freezer. Nature 428: 687. Clarke, B., J. Murray, and M. S. Johnson. 1984. The extinction of endemic species by a program of biological control. Pacific Science 38: 97-104. Coote, T. and E. Loéve. 2003. From 61 species to five: Endemic tree snails of the Society Islands fall prey to an ill-judged biological control program. Oryx 37: 91-96. Coote, T., D. Clarke, C. S. Hickman, J. Murray, and P. Pearce- Kelly. 2004. Experimental release of endemic Partula species, extinct in the wild, into a protected area of natural habitat on Moorea. Pacific Science 58: 429-434. Coote, T., E. Loéve, J-Y. Meyer, and D. Clarke. 1999. Extant popu- lations of endemic partulids on Tahiti. Oryx 33: 215-222. Coote, T. 1999. The Genetics and Conservation of Polynesian Tree Snails (Family Partulidae). Ph.D. Dissertation, University of London, London. Cowie, R. H. 1992. Evolution and extinction of Partulidae, endemic Pacific island land snails. Philosophical Transactions of the Royal Society of London (B)335: 167-191. Crampton, H. E. 1916. Studies on the variation, distribution and evolution of the genus Partula. The species inhabiting Tahiti. Carnegie Institute of Washington Publication 228: 1-311. Crampton, H. E. 1932. Studies on the variation, distribution, and evolution of the genus Partula. The species inhabiting Moorea. Carnegie Institute of Washington Publication 410: 1-335. Fosberg, F. R. 1992. Vegetation of the Society Islands. Pacific Science 46: 232-250. Gerlach, J. 1994. The Ecology of the Carnivorous Snail, E. rosea. Ph.D. Dissertation, Oxford. University, Oxford. Johnson, M. S., J. Murray, and B. C. Clarke. 1993. The ecological genetics and adaptive radiation of Partula on Moorea. In: D. Futuyma and J. Antonovics, eds., Oxford Studies in Evolution- ary Biology, Vol. 9, Oxford University Press, Oxford. Pp. 167- 238. Meyer, J.-Y., J.-C. Thibault, J.-F. Butaud, J. Florence, T. Coote, and R. Englund. 2005. Sites de conservation importants et priori- taires en Polynésie francaise. Contribution a la Biodiversité de Polynésie francaise 13: 1-35. Murray, J., E. Murray, M. S. Johnson, and B. Clarke. 1988. The Extinction of Partula on Moorea. Pacific Science 42: 150-153. Pearce-Kelly, P., D. Clarke, C. Walker, and P. Atkin. 1997. A con- servation programme for the partulid tree snails of the Pacific region. Memoirs of the Museum of Victoria 56: 431-433. Accepted: 1 December 2006 87 lb ee = id 4 rye asf sp 2a tee) =e) Thi 7 ri F . ean af : A 7 Rar 4 : : i] ® — 7 x é 7 os as ~ i - = > ( es ‘ ip je - v s qo * ' ¢ y wD io 0.05). The shells of wild H. trivolvis showed no significant difference in the concentra- tion of calcium carbonate from the laboratory-reared indi- viduals of H. trivolvis (Student’s t-test, P > 0.05); however, the shells of the wild population of H. trivolvis as a group were significantly smaller in diameter than the shells of the laboratory-reared H. trivolvis (Student’s t-test, P < 0.05). Table 1. Percentage of calcium carbonate by dry weight of snail shell, shell size, and calcium carbonate content (mg/L) of water. Helisoma trivolvis* H. trivolvis° H. trivolis Physa sp.” Wild snails Laboratory-reared snails H. trivolvis (CO) ASW Biomphalaria glabrata’ = ASW Commercially purchased snails —§ Pomacea bridgesii ASW ' ASW = artificial spring water. * Water from the collection sites for wild snails and ASW for all others. * From White et al. (2005). Pond or ASW! Amwell Lake Hoch Pond Delaware Pond Delaware Pond CaCO, content in shells (%) Size of snail (mm) CaCO, content (mean + SE) (mean + SE) in water (mg/L)? 97.6 + 0.27 15.0 + 0.57 141.1 95.2 + 0.4" 10.4 + 0.27 39.9 98.1 + 0.6 10.6 + 0.3” 190.4 97.8 + 0.5 8.4+0.17° 190.4 97.6 + 0.44 13.2 + 0.3” 32.0 98.8 + 0.2 11.1 +:0.37 32.0 98.2 +0.4 36 + 2° 32.0 * One sample was determined by the Q-test (90% confidence interval) to be an outlier and was excluded from the results and all statistical analyses, giving n=9, ° This sample had a mean concentration of calcium carbonate that was significantly lower than the other samples (ANOVA, P < 0.05). ° Maximum length. ” Maximum diameter. CALCIUM CARBONATE IN SHELLS 141 When considering only shells from the wild populations of H. trivolvis, the group collected from Hoch Pond had a significantly lower concentration of calcium carbonate in their shells than did the groups collected from Amwell Lake and Delaware Pond. A correlation between the hardness of the water and the concentration of calcium carbonate in the shells was found only for the wild snails. Hoch Pond had the softest water (although still above the 20 mg/L CaCO; minimum found to limit the number of species of snails that can survive; Boy- cott 1936, Macan 1950). Shells of snails obtained from that pond had the lowest concentrations of calcium carbonate. Delaware Pond had the hardest water and the shells of snails collected from it had the highest concentrations of calcium carbonate. Amwell Lake had a calcium content between that of Hoch Pond and Delaware Pond. Shells of snails from that lake had concentrations of calcium carbonate between those of the snails from Hoch Pond and the snails from Delaware Pond. The ASW had the lowest concentration of calcium car- bonate, yet the shells of the laboratory-reared and commer- cially-purchased snails that were maintained in it had con- centrations of calcium carbonate that were significantly higher than those of the wild snails (Student’s t-test, P<0.05). According to the standard classification for water hardness of the U.S. Geological Survey (2006), the water from Hoch Pond and the ASW were soft, the water from Amwell Lake was hard, and the water from Delaware Pond was very hard. DISCUSSION In spite of considerable variation in the calcium content (mg/L) of the water in which the snails were maintained, all of the snails showed concentrations of calcium carbonate in their shells in the range of 95.2-98.8 % by weight, similar to the range reported by Hare and Abelson (1965). Thus, under conditions of variable concentrations of calcium carbonate in the water, freshwater snails are able to maintain a high concentration of calcium carbonate in their shells. No clear trend between the concentrations of calcium carbonate of the external media and the concentrations of the snail shells was found when both wild and laboratory-reared popula- tions were considered. Freshwater snails obtain their calclum from the surrounding water and their food source (van der Borght and van Puymbroeck 1966, Young 1975), first local- izing the calcium in the mantle before depositing it in the shell (Bevelander 1952). It is surprising that the laboratory- reared snails, which were raised in the softest water, had relatively high concentrations of calcium carbonate in their shells. One important difference between the laboratory- reared snails and the wild snails was that the laboratory- reared snails were fed ad libitum; we do not know how adequate the food supply was for the snails obtained from the wild. Individuals of Lymnaea peregra (Miller, 1774) and Planorbarius corneus (Linnaeus, 1758) reared in calcium-rich water, obtain calcium more from the water than from the food. Individuals reared in calcium-poor water, however, obtain two to four times more calcium from the food than from the water (Young 1975). This relationship between water hardness and the source of calcium may be responsible in part for the results of the present study; however, the relative role of food versus water as a source of calcium for the snails was undetermined in our study. The data showed no direct correlation between size of the shell and concentration of calcium carbonate. Within the wild populations of Helisoma trivolvis, the snails from Am- well Lake were the largest; however, this did not correspond to a lower or higher concentration of calcium carbonate in the shell. The shells from the wild population of H. trivolvis showed no difference in concentration of calcium carbonate from those in the laboratory-reared strain, but the wild population was significantly smaller in diameter. The data supported the claim that shells of freshwater snails are comprised of 95-99.9 % calcium carbonate by weight. ACKNOWLEDGEMENTS We are grateful to Dr. Fred A. Lewis, Head, Schistoso- miasis Laboratory, Biomedical Research Institute, Rockville, Maryland, USA. for supplying individuals of Biomphalaria glabrata used in this work through NIH-NIAID contract NO1-AI-55270. We also thank Dr. John R. Stonesifer, Lafay- ette College Mathematics Department, Easton, Pennsylva- nia, USA, for his help in the statistical analysis. White’s work was supported by a Camille and Henry Dreyfus Foundation Senior Scientist Mentor Initiative Grant awarded to J. Sherma. LITERATURE CITED Bevelander, G. 1952. Calcification in mollusks. HI. Intake and de- position of Ca’? and P*’ in relation to shell formation. Bio- logical Bulletin 102: 9-15. Borght, O. van der and S. van Puymbroek. 1966. Calcium metabo- lism in a freshwater mollusk. Quantitative importance of wa- ter and food as supply for calcium during growth. Nature 210: 791-793. Boycott, A. E. 1936. The habitats of fresh-water Mollusca in Britain. Journal of Animal Ecology 5: 116-186. 142 AMERICAN MALACOLOGICAL BULLETIN — 22* 1/2 * 2007 Fried, B., S. Scheuermann, and J. Moore. 1987. Infectivity of Echi- nostoma revolutum miracidia for laboratory raised pulmonate snails. Journal of Parasitology 73: 1047-1048. Hare, P. E. and P. H. Abelson. 1965. Amino acid composition of some calcified proteins. Carnegie Institution of Washington Yearbook 64: 223-232. Macan, T. T. 1950. Ecology of freshwater Mollusca in the English Lake District. Journal of Animal Ecology 19: 124-146. Marxen, J. C., W. Becker, D. Finke, B. Hasse, and M. Epple. 2003. Early mineralization in Biomphalaria glabrata: Microscopic and structural results. Journal of Molluscan Studies 69: 113- 121. Ulmer, M. J. 1970. Notes on rearing of snails in the laboratory. In: A. J. MacInnis and M. Voge, eds., Experiments and Techniques in Parasitology. W. H. Freeman, San Francisco. Pp.143-144. U.S. Geological Survey. 2006. Explanation of hardness. Available at: http://water.usgs.gov/owq/Explanation.html 7 November 20006. White, M.M., M. Chejlava, B. Fried, and J. Sherma. 2005. Effects of various larval digeneans on the calcium carbonate content of the shells of Helisoma trivolvis, Biomphalaria glabrata, and Physa sp. Parasitology Research 95: 252-255. Young, J.O. 1975. A laboratory study, using ‘Ca tracer, on the source of calcium during growth in two freshwater species of Gastropoda. Proceedings of the Malacological Society of London 41: 439-445. Accepted: 7 November 2006 Amer. Malac. Bull. 22: 143-155 Larval settlement and recruitment of a brackish water clam, Corbicula japonica, the Kiso estuaries, central Japan Ryogen Nanbu, Estuko Yokoyama, Tomomi Mizuno, and Hideo Sekiguchi Faculty of Bioresources, Mie University, 1515 Kamihama-cho, Tsu, Mie 514-8507, Japan, sekiguch@bio.mie-u.ac.jp Abstract: Population dynamics of the brackish water clam Corbicula japonica were examined in the Kiso estuaries (the Ibi-Nagara Estuary and the Kiso Estuary), central Japan, during the process of larval recruitment. Based on temporal variation in densities (sampling every 2 weeks for planktonic larvae, new settlers, and small individuals, sampling every month for large and commercially important individuals) from May 2001 to April 2004, we conclude that densities of large and of commercially important individuals were determined not by larval supply but by benthic processes. Density-dependent processes were detected between densities of new settlers and recruits. These processes, however, were detected for spring-summer cohorts, but not for autumn-winter cohorts. Spatial distributions of each cohort were almost the same within the Ibi-Nagara Estuary and within the Kiso Estuary, although cohorts were collected mainly in the middle to upper regions of the Ibi-Nagara Estuary but were collected in the upper region of the Kiso Estuary. A shift in an ontogenetic habitat within each cohort was detected in the Ibi-Nagara Estuary but not in the Kiso Estuary: New settlers and small individuals were collected in the upper region while large and commercially important individuals were collected in the middle region. The shift may be explained by tidal migration using byssal threads or by site-specific differences in mortality, although it was not clear why the shift was detected in the Ibi-Nagara Estuary but not in the Kiso Estuary. Key words: population dynamics, density-dependent processes, commercially important species The brackish water clam Corbicula japonica (Prime, 1864) is endemic to eastern Asia (Sakai et al. 1994, Harada and Nishino 1995). The clam is commonly found in estua- rine waters throughout Japan except for the Ryukyu Archi- pelago, southern Japan, which geographically belongs to the Subtropical/Tropical Zone. The species is a target for clam fisheries in Japan, especially in the Kiso estuaries (the Ibi- Nagara and the Kiso Rivers), central Japan (see Nakamura 2000). Despite several regulations imposed to manage fish- eries of Corbicula in Japan, the total annual catch yields of Corbicula species (of which approximately 99% is C. ja- ponica) have decreased drastically over the last two or three decades (Mizuno et al. 2005), probably through the progress of eutrophication in the estuarine and coastal waters of Ja- pan. This is true for the yields in the Kiso estuaries as well as in other areas of Japan. Traditionally in Japan, the larger hard shell clam Meretrix lusoria (Réding, 1798) has been commercially more important than C. japonica. A drastic decrease in the yield of M. /usoria in the Kiso estuaries oc- curred in the late 1970s, when the yield of C. japonica abruptly increased (Mizuno et al. 2005). This resulted in a much greater fishing effort for C. japonica. However, the yields of C. japonica in the Kiso estuaries has drastically decreased since early 1980s despite several regulations im- posed on the fishery. The causes or mechanisms by which the drastic decrease in the yield of C. japonica in the Kiso estuaries, as well as in the other areas of Japan, may be driven are not well understood (see Nakamura 2000). Recent studies on marine benthic invertebrates have emphasized the role of larval recruitment in the population dynamics of intertidal and subtidal organisms that have complex life cycles (those that include planktonic and ben- thic phases) (e.g., Roughgarden et al. 1988, Underwood and Keough 2001), although few studies have been made in the marine environment, probably due to difficulties in identi- fying planktonic larvae (e.g., Sakai and Sekiguchi 1992), in examining the coupling of larval transport and dispersal with oceanographic conditions (e.g., Roughgarden et al. 1988), and in discovering larval settlement processes (e.¢., Connell 1985, Gaines and Roughgarden 1985, Gaines ef al. 1985). This situation has been true also for bivalves, includ- ing the clams that are commercially important in Japan (see Miyawaki and Sekiguchi 1999, 2000, Ishi et al. 2001a, 2001b). Unfortunately, there is not sufficient data on larval recruitment of Corbicula japonica, in contrast to the clam Ruditapes philippinarum (Adams and Reeve, 1850), which dominates Japanese tidal flats (Miyawaki and Sekiguchi 1999, 2000, Ishii ef al. 2001a, 2001b). Tolerance to varying salinity by planktonic larvae and benthic stages of Corbicula japonica was examined in the laboratory by Tanaka (1984a, b) and Saito et al. (2002). Tidal transport of the larvae was investigated in relation to the salinity distribution in estuarine waters in the Kiso estuaries by Sekiguchi et al. (1991) and in the laboratory by Kuwabara and Saito (2003). In Lake Shinji where C. japonica is the most important target species for clam fisheries, growth of 144 AMERICAN MALACOLOGICAL BULLETIN benthic stages of the clam was examined by Takada et al. (2001) and Oshima et al. (2004). Takada et al. (2001) also studied the seasonal abundance of spat of the clam, estimat- ing the season of larval settlement. However, larval recruit- ment of the clam has not been studied. Nanbu et al. (2005) examined the spatio-temporal variations in densities of dif- ferent life stages (planktonic larvae, new settlers, and small, large and commercially important individuals) of the clam in the Kiso estuaries, central Japan, to understand larval recruitment, by which benthic populations may be generated and maintained in the estuaries. In contrast to the other common and abundant bivalves (Ruditapes philippinarum, Musculista senhousia {Benson, 1842], and Mactra venerifor- mis Deshayes in Reeve, 1854), of which all life stages were detected around the river mouths, C. japonica showed a marked ontogenetic habitat shift in the Ibi-Nagara Estuary, but not in the Kiso Estuary. The primary habitat for new settlers and small individuals and of large and commercially important individuals of C. japonica was located in the upper and middle regions of the Ibi-Nagara Estuary, while the benthic stages were primarily found in the upper region of the Kiso Estuary. However, it is not clear whether the on- togenetic habitat shift is generated by migration during ben- thic stages or by differences in site-specific mortality of new recruits. To understand the population dynamics of Corbicula japonica in the Kiso estuaries, we examined which life stage may determine the strength of catch yields (or population size) of the clam, and whether the ontogenetic habitat shift of the clam may be generated by migration during benthic stages or by differences in site-specific mortality, using the cohort separation based on three years of data (from May 2001 until April 2004) collected in the Kiso estuaries. Nanbu et al. (2005) examined spatio-temporal distributions of den- sities of each life stage of the clam, using the first two years of data from the present study. METHODS Study area The Kiso Rivers (the Ibi, Nagara, and Kiso Rivers), three of the largest rivers in Japan, flow into Ise Bay on the Pacific coast of central Japan (Fig. 1). The Ibi and the Nagara Rivers join at their lower regions where both rivers are united into the Ibi-Nagara River. The Nagara River, however, has re- cently been closed at a point 5 km upstream by the Nagara Dam (Fig. 1). The Kiso estuaries (the Ibi-Nagara and the Kiso Estuaries), defined as the areas where bottom water has a detectable salinity of 1.0 psu, are 2-10 m in depth and have a maximal tidal range of 3 m, reaching 30 km upstream for the [bi River and 26 km upstream (where there is a dam) for the Kiso Estuary (Japan Society of Oceanography 1985). 22° 1/2 * 2007 ee e Neath « Neb 12 y, \ j 44° | i J AE oe ed sy Aen fe Kiso Estuary pg L- a Ibi-Nagara } \Da b Japan Sea / ? Ca Estuary | Page ‘ Pacific | | oa a . ifi Ib a3 4 pe Ocean | ' en —1 \agara Dam 132° 43° —~E 7 N « NAGOYA Kiso R 35°F ® 4 TSU/ — 1SE BAY Pacific L 137° E Figure 1. Study area and location of sampling stations. Solid circles, sampling stations with 2 or 3 sites (e.g., I-la, I-1b) within each sampling station; bold solid lines seaward from the estuary mouth, protection bars against the tide. Environmental characteristics For the last decade, the Kiso River Management Office of the Ministry of Transport and Infrastructure has continu- ously monitored water temperature and salinity every hour in water 0.5 m above the bottom at a site 0.5 km upstream in the Ibi-Nagara Estuary (Zyonan, see Fig. 1 for location). To examine environmental characteristics of the Kiso estu- aries, we used averages over 24 h of these environmental data obtained from that office (Fig. 2A). Although we have no similar data available for the Kiso Estuary, environmental conditions are similar at the mouths of these estuaries (Mi- zuno et al. 2005). According to Mizuno et al. (2005), the silt-clay fraction of sediments in the Kiso estuaries is usually less than 5.0% in dry weight, being much higher at the mouths and down- stream areas of these estuaries, and lower in the upstream areas. Accordingly, bottom sediments are sandy 5 km or ————__SS_ LARVAL RECRUITMENT OF CORBICULA JAPONICA IN THE KISO ESTUARIES 145 3 Salinity Figure 2. A, Water temperature and = salinity in the Kiso estuaries, central He Japan. B-F, Variation in average 3 I} “ae = 53 5 Es o 2 E E 9 MJJASONDJFMAMJJASONDJIFMAMJIJASONDJIFMA 2001 2002 2003 2004 Months of year more upstream of these estuaries, although an extraordinar- Sampling procedures ily high percentage (50% or more) of the silt-clay fraction Sampling was undertaken in the Kiso estuaries from occurs in the area closest to the Nagara Dam. Muddy sedi- May 2001 until April 2004. The study area, environmental ment was common in troughs, so that the silt-clay fraction characteristics, sampling procedures, and data processing was very different between troughs and ridges, even at the — were described in detail in Nanbu et al. (2005). same distance from the mouths of the estuaries. Planktonic larvae of Corbicula japonica were obtained at 146 AMERICAN MALACOLOGICAL BULLETIN stations located 1 km seaward of each river mouth (I-la and I-lb in the Ibi-Nagara Estuary, K-la and K-lb in the Kiso Estuary) (Fig. 1) using a vertical haul of plankton nets (22 cm in diameter, 133 um mesh openings) from the bottom to the surface every 2 weeks (except for January to March 2002). Larval density was indicated as individuals/m*. Bi- valve larvae were identified to species using a compound microscope according to Sakai and Sekiguchi (1992) and Kimura ef al. (2004). For sampling benthic stages of Corbicula japonica, 8 stations (1-3, I-1, 11, 12, 15, 17, 19, and 112 for the Ibi-Nagara Estuary, K-3, K-1, K1, K3, K5, K7, K9, and K12 for the Kiso Estuary) were located every 2 or 3 km up to 12 km upstream from the river mouth in each estuary and also 2 or 3 sites 1 km and 3 km downstream/seaward from the mouth, respec- tively (Fig. 1). Samples were collected at 2 or 3 sites with different depths within each sampling station area. To sample new settlers and small individuals, one sediment sample was collected with a core sampler (3.1 cm in diam- eter, 1.0 cm depth) from the surface layer of bottom sedi- ments which were obtained at each site within each station using a Smith-MclIntyre grab. For sampling large and com- mercially important individuals, one sediment sample was obtained at each site within each station using a Smith- McIntyre grab. New settlers and small individuals of the clam were collected every 2 weeks while large and commer- cially important individuals were collected every month. Identification of new settlers and small individuals of C. japonica was done following Sakai and Sekiguchi (1992) and Kimura et al. (2004). Density of the benthic stages of the clam was indicated as individuals/0.01m°. The life stages of Corbicula japonica were defined as in Nanbu et al. (2005): “Planktonic larvae” were D-shaped lar- vae (i.e. early-stage veligers); “new settlers” were individuals with shell lengths less than 300 tim; “small individuals” were ones with shell lengths 0.3 mm or more but less than 1.0 mm; “large individuals” were ones with shell lengths of 1.0 mm or more but less than 12.0 mm; “commercial individu- als” were ones with shell lengths of 12.0 mm or more. These definitions are the same as those used to describe bivalves that are common in Japanese tidal flats (e.g., Ruditapes phil- ippinarum) by Sekiguchi and his co-workers (e.g., Miyawaki and Sekiguchi 1999, 2000, Ishii et al. 2001a, b, Nanbu et al. 2005) except for planktonic larvae. We defined “successful recruitment” as new settlers and small individuals that reached average shell lengths of 1.0 mm or more for each cohort. Data analysis Sediment samples were collected at 2 or 3 sites within each station. The density of each life stage of C. japonica was not significantly different between sites within each station 22° 1/2 * 2007 (t-test, p >0.05). We used average densities of each life stage. Based on these averages for the period from May 2001 to April 2004, we examined the differences in density of each life stage between the two estuaries and between sampling year for each estuary, using Mann-Whitney’s U-test (signifi- cance level, @ = 0.05) and Kruskal-Wallis’s H-test (signifi- cance level, a = 0.05), respectively. We used Bonferroni’s method (a’ = 0.05) when significant differences in density were detected between sampling years. To separate each cohort, the shell lengths for the two groups (new settlers/small individual group, large/ commercial individual group) were compiled for all stations of each estuary. However, it was difficult to separate each cohort for commercial individuals due to their small num- bers. Based on these data, cohorts within each group were identified by the method of Akamine (1985), who separated polymodal length distribution into two or more normal dis- tributions. The growth curve of each cohort was estimated based on temporal change of the mean shell length of each normal distribution. Estimation of densities of new settlers and recruits of Corbicula japonica To examine the relationships between the densities of new settlers and recruits (7.e., large individuals) of the clam, using the data for cohorts that were successful in recruit- ment, we estimated densities of new settlers and recruits, respectively, as follows: Based on the growth curve of each successful cohort, we determined the day when average shell lengths of new settlers reached 1.0 mm. Then, assuming larval settlement day as day 0 when new settlers of each successful cohort were first collected, we fitted a regression line to the temporal change of each cohort density using a graph in which the X-axis was days after larval settlement of each successful cohort and the Y-axis was the log- transformed density of each successful cohort. Using the density data estimated for successful cohorts with a signifi- cant (p < 0.05) regression line, we examined the relation- ships between the densities of new settlers and recruits (z.e., based on the data for each estuary and then for each estuary according to season), and then the relationships between the density of new settlers and the ratio of their densities (re- cruits/new settlers) (7.e., based on the data for each estuary and then for each estuary according to season). RESULTS Variation in densities of different life stages of Corbicula japonica The densities of planktonic larvae peaked primarily in May to December every year (Fig. 2B). There was no sig- LARVAL RECRUITMENT OF CORBICULA JAPONICA IN THE KISO ESTUARIES 147 nificant difference in larval density between the Ibi-Nagara and the Kiso Estuaries nor between sampling years for each estuary (Table 1). As seen in Fig. 2B, however, larval density in 2003 appeared to be lower than in the other years, prob- ably due to more days with less than 10 psu in 2003 than in the other years, because salinity preferred by spawning of the clam is in the range of 9.35-21.82 psu (Asahina 1941). Water temperature reached about 30°C in summer and decreased below 10°C in winter (Fig. 2A). On the other hand, although tending to become lower in summer and higher in winter, salinity did not indicate such a clear sea- sonal change but always showed irregular variations (about 15-32 psu) and occasionally marked lowering (down to <1 psu, corresponding to low larval density) owing to freshwa- ter discharge through high rainfall in early summer. Larval densities were low or larvae were completely absent from the water column usually from December to the following April, when the water temperature decreased to below 15°C. Ac- cording to laboratory rearing experiments (Kimura et al. 2004), larvae of the clam reared at 15°C or lower failed to settle and recruit. The larvae from earlier and later portions of a much longer period spawning of the clam every year may not contribute to generating cohorts of new settlers. The densities of new settlers peaked primarily in July to August every year (Fig. 2C). There was no significant differ- ence in density between the two estuaries (except for 2003) nor between sampling years for each estuary (Table 1), al- though the density in 2003 appeared to be lower than in the other years, possibly due to a lower larval density in 2003. Higher densities of small individuals were found for a longer period than new settlers, occurring from August to the following April (Fig. 2D), although the density appeared to be much lower in 2003 than in the other years. In each year, density of small individuals was significantly higher in the Ibi-Nagara Estuary than in the Kiso Estuary (Table 1). The Kiso Estuary had a higher density in 2002; other- wise there was no significant difference in the density of small individuals between sampling year for each estuary (Table 1). Variation in densities of large individuals was very similar between the two estuaries (Fig. 2E). There was no significant difference in density between these estuaries and also between sampling year for each estuary (Table 1). As seen in Fig. 2E, however, the density of large indi- viduals appeared to be higher in 2003 than in the other years, in contrast to larvae and new settlers/small individuals (Table 1). The density of commercially important individuals was Table 1. Differences in densities of different life stages of Corbicula japonica between the Ibi-Nagara and the Kiso estuaries and between sampling years for each estuary. 2001 Ibi-Nagara Kiso Ibi-Nagara Sampling year Larvae - = Larvae 2002 = 2001 = 2003 New settlers = = 2 Small individuals O 4 New settlers 2002 = 2001 = 2003 Large individuals — = : ieee. Bete 2 a Corimereal icin duals e) sé Small individuals ave = 2001 See foe aH dann 2003 = 2001 = 200? 2002 Ibi-Nagara reer Large individuals ate 2001 ee Larvae = _ Commercial individuals 2001 > ule > 2003 New settlers = = Small individuals @ x Kiso Sampling year Large individuals ~ ~ a ee Larvae 2002 = 2001 = 2003 Commercial individuals = = =e : ee New settlers 2002 = 2001 = 2003 2003 Ibi-Nagara Kiso Powe: - | lance PY, = Small individuals 2002 = ~” = 2003 NEW oetners . . Large individuals 2003 = 2002 - 2001 Small individuals O x sie een cad eos ae Large individuals ; 7 7 Commercial individuals 2001 > 2002 > 2003 Commercial individuals _ = Sw Mann-Whitney’s U-test (significance level, « = 0.05). Kruskal-Wallis’ H-test (significance level, a = 0.05). O: significant difference with higher density. x: significant difference with lower density. —: no significant difference. >: significant difference. =: no significant difference. Multiple comparison (a! = 0.05; Bonferroni’s method). 148 AMERICAN MALACOLOGICAL BULLETIN significantly lower in 2003 than in the other years (Table 1) and their density appeared to decline from 2001 to 2004 (Fig. 2F). Cohort separation of Corbicula japonica In the Ibi-Nagara Estuary, 21 cohorts of new settlers and small individuals and 16 cohorts of large individuals were identified (Fig. 3B). Of these cohorts, 7 were found to settle 22 ° 1/2 + 2007 and to succeed in recruitment through the three-year inves- tigation. Larval settlement for 2 cohorts (i, ii) occurred in August to September 2001. For 4 cohorts (iii-vi), the larval settlement season in 2002 varied depending on cohort: Co- horts i, iv, and v settled in January, May to June, and September, respectively. Larval settlement for the remaining cohort occurred in April of 2003. Of these 7 cohorts, sig- nificant regression lines were fitted to the temporal change Figure 3. Cohorts of benthic stages of Corbicula japonica in the Ibi- Nagara Estuary, central Japan. A, Bold solid lines, larval density; thin solid lines, salinity; dotted lines, wa- ter temperature. B, i-vi, cohorts suc- cessful in recruitment; solid or open circles with vertical lines, averages of shell lengths and standard error; horizontal solid line, shell length 1.0 mm, which we defined as the size (nsd) Ayiuyes '(9,) ‘dwey '€ a mo) £ = 8 < a at E E = % 8 3 2 a Cc 100000 10000 ~ ~~ 1000 + wi E 3 mo] £ 100 > 2 a = o =) 10 04 for successful recruitment. C, i-vi, cohorts same as in B. MJJASONDJIFMAMJIJASONDJFMAMJJASONDJIFMA 2001 2002 Months of year 2003 2004 LARVAL RECRUITMENT OF CORBICULA JAPONICA IN THE KISO ESTUARIES 149 of density from larval settlement to recruitment for 6 co- horts (cohort i-vi) (Fig. 3C). These cohorts (except cohort ili) appeared to settle in August to November. In the Kiso Estuary, 12 cohorts of new settlers and small individuals and 13 cohorts of large individuals were identi- fied (Fig. 4B). Of these cohorts, 7 were detected to settle and to succeed in recruitment through the three-year investiga- tion. Larval settlement for cohorts 1 and 2 occurred in August to September 2001. For 3 cohorts (3-5), larval settlement occurred in July to September 2002. Larval settle- ment for the remaining 2 cohorts occurred in Decem- ber 2002 and in July 2003, respectively. Significant regres- sion lines were fitted to the temporal change of density from larval settlement to recruitment for 5 cohorts (1-5) (Fig. 4C). These cohorts appeared to settle in July to September. A 2500 2000 3 Larvae (inds./m ) 1500 1000 + 500 + Figure 4. Cohorts of benthic stages of Corbicula japonica in the Kiso Es- tuary, central Japan. A, Bold solid solid lines, larval density; thin solid lines, salinity; dotted lines, water temperature. B, 1-5, cohorts suc- cessful in recruitment; solid or open (nsd) Ayluyes (O,) dwey circles with vertical lines, averages of shell lengths and standard error; horizontal solid line, shell length 1.0 mm, which we defined as the size MJ JASONDJFMAMJJASONDJFMAMJJASONDJFMA for successful recruitment. G 1-5, cohorts same as in B. Shell length (mm) 0.74 0.6 4 0.54 0.45 0.3 0.25 0.1 0 C 100000 MJJASONDJFMAMJJASONDJFMAMJJASONDJFMA \ AEE MJJASONDJFMAMJIJASONDJIFMAMJSJIASONDJIFMA 2001 2002 Months of year 2004 Kiso estuaries (combined) Kiso estuaries (combined) 5 n= 11 n= 11 r=0.76 A -=093 p = 0.005 5 eSeCUr ve 4_ ‘ 3 7 m1 aq 2 2 2 5 ¥ 5 5 2 8 5 . 9° 6 2 3 4 5 gi . he new settlers (log inds./m ) new settlers (log inds./m ) Ibi-Nagara Estuary Ibi-Nagara Estuary nw o 2 9 Pa = mo} o £ & 2 — 3 wn rs) 8 5 2 ° % 2 3 4 5 new settlers (log inds /m ) new settlers (log indsvin Kiso Estuary Kiso Estuary 7 Es g 5 2 2 S 6 = 20 s T 2 3 4 5 2 F pe: new settlers (log inds./m ) new settlers (log inds./m ) Figure 5. Relationships between densities of new settlers and recruits of Corbicula japonica in the Kiso estuaries, central Japan. Upper figures, data for the two estuaries combined; middle figures, data for the Ibi-Nagara Estuary; lower figures, data for the Kiso Estuary; left figures, density of new settlers vs. density of recruits; right figures, density of new settlers vs. the ratio of their densities (recruits/new settlers); open circles, data for the Ibi-Nagara Estuary; triangles, data for the Kiso Estuary; n=number of cohorts. LARVAL RECRUITMENT OF CORBICULA JAPONICA IN THE KISO ESTUARIES 15] Based on the data for successful cohorts in the Kiso estuaries, it took nearly two years from larval settlement for individuals to reach commercially viable shell lengths. Takada et al. (2001) and Oshima et al. (2004) estimated shell growth of clams using growth rings on the shell surface and found a similar growth rate as in the present study. As indicated in Figs. 3B and 4B, the densities of new settlers in the cohorts identified in 2003, which originated in planktonic larvae and new settlers with much lower densi- ties, were considerably lower than in the other years (Figs. 2B, 2C). Density peaks of new settlers appeared to generate Spring-Summer “E 2 Q _ £ ; Pa) 2 @ = = °° 2 ] I 3 4 5 2 new settlers (log inds./m ) Autumn-Winter “E a me} © ii) 2 @ A= = 12) ig 2 new settlers (log inds./m ) successful cohorts of benthic stages (compare Fig. 4B with Fig. 4A). However, as indicated by comparing Fig. 3B with Fig. 3A, peaks of densities of planktonic larvae did not al- ways contribute to the generation of successful cohorts of benthic stages. The larvae from earlier and later periods of spawning of the clam every year may not contribute to gen- erating cohorts of new settlers. Relationships between densities of new settlers and recruits of Corbicula japonica Based on the data from 11 successful cohorts (6 from Spring-Summer ratio of recruits/new settlers 2 new settlers (log inds./m ) Autumn-Winter 06 n=4 05 | N r=0.95 2 p = 0.05 2 044 /\ w” z A 2 2 03- 5 ® So 02-4 2 g oO ss A o1 | = o—-- 1 0 1 2 3 4 5 9 2 new settlers (log inds./m_ ) Figure 6. Relationships between densities of new settlers and recruits of Corbicula japonica in the Kiso estuaries, central Japan. Upper figures, data for spring-summer cohorts; lower figures, data for autumn-winter cohorts; open circles, data for the Ibi-Nagara Estuary; triangles, data for the Kiso Estuary; n=the number of cohorts. 152 AMERICAN MALACOLOGICAL BULLETIN the Ibi-Nagara Estuary, 5 from the Kiso Estuary), there was a significantly positive correlation between densities of new settlers and recruits in the combined estuaries, but not in each estuary (Fig. 5). There was a significant negative cor- relation between the density of new settlers and the ratio of their densities (recruits/new settlers) for these estuaries com- bined and also for each estuary individually (Fig. 5). The 11 successful cohorts that were divided into two groups, 7 spring-summer cohorts and 4 autumn-winter ones, according to the months with successful recruitment. There was no significant correlation between densities of new settlers and recruits for spring-summer cohorts, whereas there was a significant negative correlation between the density of new settlers and the ratio of their densities (recruits/new settlers) (Fig. 6). The reverse was true for au- tumn-winter cohorts (Fig. 6): There was a significant posi- tive correlation between densities of new settlers and recruits Cohort i 22° 1/2 * 2007 for autumn-winter cohorts, but there was no significant cor- relation between the density of new settlers and the ratio of their densities (recruits/new settlers). Ontogenetic habitat shift of Corbicula japonica Of the cohorts identified in the Ibi-Nagara Estuary, 4 cohorts (i-iv) were successful in reaching shell lengths (12.0 mm) of commercially important individuals (Fig. 3B). Each cohort had a similar spatio-temporal distribution (Fig. 7): new settlers and small individuals were collected mainly in the upper region of the estuary, large and commercially im- portant individuals were found primarily in the central re- gion. Nanbu et al. (2005) found an ontogenetic habitat shift during the benthic stages (from new settlers to commercially important individuals) in mixed cohorts of this clam. An ontogenetic habitat shift was also detected within the same cohort, as indicated in Fig. 7. However, this was not true for Cohort ii 200200 Ls hoe 200 124 100) 94 74 par taraietcy MJJASOND Distance from the river mouth (km) Toe. a 8 jee Sey Sa Sem a, ay aes a DL a ore [ia ir JFMAMJJASONDJFMAMJ JASOND J FMA ST eB a MJJASONDJFMAMJJASONDJFMAMJJASONDJFMA Cohort iil Cohort iv 9) 10 94 | o fe 9+ «Wes NA. “8 ° 7 8 -14 -14 -34 -37 . MJJASOND JFMAMJJASOND JFMAMJJASOND J FMA. MJJASOND UJ FMAMJ JA SONDJ FMAMJJASONDUJFMA 2001 2002 2003 2004 2001 2002 2003 2004 Months of year Figure 7. Habitats of different cohorts of Corbicula japonica in the Ibi-Nagara Estuary, central Japan. Dots, sampling day/location; solid lines, isopleths of densities of new settlers/small individuals or large/commercial individuals; numerals, density in inds./0.01m”; gray to black areas, the highest density areas with 50 inds./0.01m* or more for new settlers/small individuals and 5 inds./0.01m* or more for large individuals. Left and right isopleths in each figure indicate isopleths for new settlers/small individuals and large/commercial individuals, respectively. Open space between isopleths of new settlers/small individuals and large/commercial individuals in each figure is due to the transition between these two groups, as indicated in Figures 3 and 4. SS ae LARVAL RECRUITMENT OF CORBICULA JAPONICA IN THE KISO ESTUARIES 15 cohorts in the Kiso Estuary. In the Kiso Estuary, 4 cohorts (1-4) were successful in reaching shell lengths of commer- cially important individuals (Fig. 4B). All cohorts had simi- lar spatio-temporal distributions (Fig. 8): Benthic stages were collected primarily in the upper region of the estuary. DISCUSSION As summarized for densities of different life stages of Corbicula japonica in Table 1, there was no significant dif- ference in larval densities between the Ibi-Nagara and the Kiso Estuaries through the three-year investigation. This was also true for new settlers (except for 2003) and for large and commercially important individuals (except for 2001). On the other hand, there were significant differences in the den- Cohort 1 eS) sity of small individuals between the two estuaries: A higher density was detected in the Ibi-Nagara Estuary than in the Kiso Estuary. Because the local fishermen’s union (e.g., Aka- suka) gets higher annual catch yields of the clam in the Ibi-Nagara Estuary, putting higher fishing pressure on com- mercially important individuals in the estuary (Mizuno et al. 2005), commercially important individuals may in fact have a much higher density in the Ibi-Nagara Estuary. In each estuary, the density of commercially important individuals was significantly higher in 2002, but there was not a signifi- cant difference in the densities of larvae and new settlers between sampling years (Table 1). This suggests that pro- cesses affecting benthic stages (new settlers and small indi- viduals), not processes affecting larval settlement and larval supply, may contribute to generating the differences in den- sities of small individuals between these estuaries. However, Cohort 2 90 Cohort 3 Distance form the river mouth (km) f | (A MJ JASONDUFMAMJJASOND JFMAMJJASOND J FMA Cohort 4 =34 | f MJJASONDJFMAMJJASONDJFMAMJJASONDJFMA, MJJASONDJFMAMJJASOND JFMAMJJASOND FMA 2001 2002 2003 2004 2001 2002 2003 2004 Months of year Figure 8. Habitats of different cohorts of Corbicula japonica in the Kiso Estuary, central Japan. Dots, sampling day/location; solid lines, isopleths of densities of new settlers/small individuals or large/commercial individuals; numerals, density in inds./0.01m°*; gray to black areas, the highest density areas with 50 inds./0.01m* or more for new settlers/small individuals and 5 inds./0.01m* or more for large individuals. Left and right isopleths in each figure indicate isopleths for new settlers/small individuals and large/commercial individuals, respectively. Open space between isopleths of new settlers/small individuals and large/commercial individuals in each figure is due to the transition between these two groups, as indicated in Figures 3 and 4. 154 AMERICAN MALACOLOGICAL BULLETIN the difference in densities of small individuals between the estuaries may not contribute to generating the differences in the densities of the successive benthic stages (large and com- mercially important individuals) between the estuaries. For populations of Corbicula japonica in the Kiso estu- aries, there were significant correlations between the density of new settlers and the ratio of their densities (recruits/new settlers) for each estuary and also for the two estuaries com- bined (Figs. 5-6). The density of new settlers may have a great influence on the density of recruits. This was also true for spring-summer cohorts, but not for autumn-winter co- horts. According to Mizuno et al. (2005), annual catch yields of C. japonica in the Kiso estuaries are sustained by indi- viduals (new cohorts) reaching commercially viable shell lengths (12.0 mm) of large individuals in spring to summer every year. Densities of these new cohorts drastically de- crease in winter due to high mortality caused by fishing pressure, so that density-dependent processes may not op- erate on autumn-winter cohorts. On the other hand, den- sity-dependent processes may affect spring-summer cohorts because higher densities of new settlers and small individuals were observed in summer to autumn every year. We con- clude that densities of large and commercially important individuals were determined by benthic processes, not by larval supply. It is not immediately apparent why the shift in an on- togenetic habitat of Corbicula japonica was detected only in the Ibi-Nagara Estuary. Nanbu et al. (2005) also reported the occurrence of a similar ontogenetic habitat shift for mixed cohorts in the Ibi-Nagara Estuary. They proposed alternative scenarios to explain this shift: (1) The shift may be generated by tidal/diurnal/seasonal/ontogenetic migration using the byssus or other means, as observed in many common bi- valves (Hamada and Ino 1954, Sigurdsson et al. 1976, Pre- zant and Chalermwat 1984, Lane et al. 1985), particularly in light of the salinity sensitivity of Corbicula japonica during ontogeny (Saito et al. 2002, Kuwabara and Saito 2003), (2) The shift does not occur within the same cohort; habitats may differ depending on cohort (z.e., the site-specific mor- tality may differ depending on benthic stage), so that the shift only appears to occur within the same cohort; and (3) The shift does not occur within the same cohort, and the site-specific mortality may differ depending on benthic stage, so that the shift appears to occur within the same cohort. Because of the cohort separation observed in the present study, indicating the occurrence of an ontogenetic habitat shift within the same cohort and because all benthic cohorts had a similar distribution pattern for each estuary, scenario 2 may be rejected. However, we cannot evaluate scenarios | or 3 unless we measure the site-specific mortality in the Ibi-Nagara Estuary. 22° 1/2 + 2007 ACKNOWLEDGMENTS We express our sincere thanks to Dr. Taeko Kimura of the Faculty of Bioresources of Mie University and the staffs of Mie Prefectural Science and Technology Promotion Cen- ter for their moral and logistic support during the course of the present study. We are grateful to Akasuka Fishermen’s Union located in Kuwana, Mie Prefecture, and the staffs of our laboratory for helping in field sampling on board in the Kiso estuaries. Thanks are due to the staff of the Kiso River Management Office of the Ministry of Transport and Infra- structure for permission to use environmental data of the Kiso estuaries. LITERATURE CITED Akamine, T. 1985. Considerations of BASIC program to analyze the polymodal frequency distribution into normal distribution. Bulletin of the Japan Sea Regional Research Laboratory 35: 129- 160 [In Japanese with English summary]. Asahina, E. 1941. An ecological study of Corbicula japonica group, the brackish water bivalves, with special reference to the en- vironmental factors of its habitat in Hokkaido. Bulletin of the Japanese Society of Scientific Fisheries 10: 143-152 [In Japanese with English summary]. Connell, J. H. 1985. The consequences of variation in initial settle- ment vs. post-settlement mortality in rocky intertidal commu- nities. Journal of Experimental Marine Biology and Ecology 93: 11-45. Gaines, S. D. and J. Roughgarden. 1985. Larval settlement rate: A leading determinant of structure in an ecological community of the marine intertidal zone. Proceeding of National Academy of Science 82: 3707-3711. Gaines, S. D., S. Brown, and J. Roughgarden. 1985. Spatial variation in larval concentrations as a cause of spatial variation in settle- ment for the barnacle, Balanus glandula. Oecologia 67: 267- 272. Hamada, S. and T. Ino. 1954. Studies on the movement of the Japanese hard clam, Meterix meretrix lusoria (Roding)—1. Histological studies on the gland in relation to locomotion. Bulletin of the Japanese Society of Scientific Fisheries 20: 1-3 [In Japanese with English summary]. Harada, E. and M. Nishino. 1995. Differences in inhalant siphonal papillae among the Japanese species of Corbicula (Mollusca: Bivalvia). Publication of the Seto Marine Biological Laboratory 36: 389-408. Ishii, R., S. Kawakami, H. Sekiguchi, Y. Nakahara, and Y. Jinnai. 2001a. Larval recruitment of the mytilid Musculista senhousia in Ariake Sound, southern Japan. Venus 60: 37-55. Ishii, R., H. Sekiguchi, Y. Nakahara, and Y. Jinnai. 2001b. Larval recruitment of the Manila clam Ruditapes philippinarum in Ariake Sound, southern Japan. Fisheries Science 67: 579-591. Japan Society of Oceanography, ed. 1985. Coastal Oceanography of Japan Islands. Tokai University Press, Tokyo [In Japanese]. LARVAL RECRUITMENT OF CORBICULA JAPONICA IN THE KISO ESTUARIES 15 Kimura, T., Y. Soutome, and H. Sekiguchi. 2004. Larval develop- ment of the brackish water clam Corbicula japonica with notes on larval and post-larval shell morphology. Venus 63: 33-48. Kuwabara, H. and H. Saito. 2003. Trajectories of planktonic larvae of Corbicula japonica in the lower part of Hinuma River. Bul- letin of Coastal Engineering of Japan 50: 1106-1110 [In Japa- nese with English summary]. Lane, D. J., A. R. Beamont, and J. R. Hunter. 1985. Byssus drifting and the drifting threads of the young post-larval mussel Myti- lus edulis. Marine Biology 84: 301-308. Miyawaki, D. and H. Sekiguchi. 1999. Interannual variation of bivalve populations on temperate tidal flats. Fisheries Science 65: 817-829. Miyawaki, D. and H. Sekiguchi. 2000. Long-term observations on larval recruitment processes of bivalve assemblages on tem- perate tidal flats. Benthos Research 55: 1-16. Mizuno, T., R. Nanbu, and H. Sekiguchi. 2005. Population dynam- ics of the brackish water clam Corbicula japonica in Kiso es- tuaries, central Japan. Nippon Suisan Gakkaishi 71: 151-160 [In Japanese with English summary]. Nakamura, M. 2000. Ecological characteristics of the brackish water clam Corbicula japonica. In: M. Nakamura, ed., Corbicula Fish- eries in Japan: Present Status and Issues. Tatara Shobo, Matsue, Japan. Pp.1-30. Nanbu, R., E. Yokoyama, T. Mizuno, and H. Sekiguchi. 2005. Spa- tio-temporal variations in density of different life stages of a brackish water clam Corbicula japonica in the Kiso estuaries, central Japan. Journal of Shellfish Research 24: 1067-1078. Oshima, K., N. Suzuki, M. Nakamura, and K. Sakuramoto. 2004. Shell growth and age determination of the brackish water bivalve Corbicula japonica in Lake Shinji, Japan. Fisheries Sci- ence 70: 601-610. Prezant, R. S. and K. Chalermwat. 1984. Floatation of the bivalve Corbicula fluminea as a means of dispersal. Science 225: 1491- 1493, Roughgarden, J., S. Gaines, and H. Possingham. 1988. Recruitment dynamics in complex life cycles. Science 241: 1460-1466. Saito, H., K. Nakayama, J. Watanabe, C. Murakami, T. Koyama, and Y. Nakamura. 2002. Laboratory experiments on the change of salinity selectivity of Corbicula japonica through ontogeny. In: Abstracts of the 16"" Annual Meeting of the Japan Society of Benthology. Tsu, Mie, Japan. P. 17 [In Japanese with English summary]. Sakai, A. and H. Sekiguchi. 1992. Identification of planktonic late- stage larval and settled bivalves in a tidal flat. Bulletin of the Japanese Society of Fisheries Oceanography 56: 410-425 [In Japanese with English summary]. Sakai, H., K. Kamiyama, S. R. Jeon, and M. Amio. 1994. Genetic relationships among three species of freshwater bivalves genus Corbicula (Corbiculidae) in Japan. Nippon Suisan Gakkaishi 60: 605-610 [In Japanese with English summary]. Sekiguchi, H., H. Saito, and H. Nakao. 1991. Spatial and temporal distributions of planktonic and benthic phases of bivalves in a tidal estuary. Benthos Research 40: 11-21. Sigurdsson, J. B., C. W. Titman, and P. A. Davies. 1976. The dis- nn persal of young post-larval bivalve molluscs by byssus threads. Nature 262: 386-387. Takada, Y., T. Sonoda, M. Nakamura, and S. Nakao. 2001. Growth and settlement of the bivalve, Corbicula japonica population in Lake Sinji. Nippon Suisan Gakkaishi 67: 678-686 [In Japanese with English summary]. Tanaka, Y. 1984a. Morphological and physiological characteristics of the post larval stage in Corbicula japonica Prime, reared in the laboratory. Bulletin of the National Research Institute of Aquaculture 6: 23-27 [In Japanese with English summary]. Tanaka, Y. 1984b. Salinity tolerance of the brackish-water clam, Corbicula japonica Prime. Bulletin of the National Research Institute of Aquaculture 6: 29-32 [In Japanese with English summary]. Underwood, A. J. and M. J. Keough. 2001. Supply-side ecology: The nature and consequences of variations in recruitment of in- tertidal organisms. In: M. K. Bertness, S. D. Gaines and M. E. Hay, eds., Marine Community Ecology. Sinauer Associates, Sunderland, Massachusetts. Pp.183-200. Accepted: 5 September 2006 Amer. Malac. Bull, 22: 157-164 Investigation in the laboratory of mucous trail detection in the terrestrial pulmonate snail Mesodon thyroidus (Say, 1817) (Mollusca: Gastropoda: Polygyridae)* Elizabeth C. Davis Columbia College Chicago, Department of Science and Mathematics, Chicago, Hlinois 60605-1996, U.S.A., edavis@colum.edu Abstract: Many gastropods, including limpets, periwinkles, and mud snails, detect and follow mucous trails. The stylommatophoran pulmonate snail Mesodon thyroidus follow conspecific trails in the field, potentially following in the direction they were laid, as has been demonstrated for other pulmonates. This study investigated whether individuals of M. thyroidus directionally follows conspecific trails, and whether substrate type or incline influences trail following. Trail following was quantified on plexiglas and glass surfaces at horizontal, vertical, and 45° inclines. On horizontal plexiglas surfaces, 36% of M. thyroidus followed a marker trail made by a conspecific (n = 11). On horizontal glass surfaces, 45% of the snails followed a marker trail (n = 20). On glass (all inclines combined) 75% of snails that followed a conspecific trail followed it in the same direction it was laid (n = 60). On plexiglas (all inclines combined) 86% of trail-following proceeded in the direction the trail was laid (n = 33). The difference in the results across the two substrates could indicate a behavioral reaction to the chemical difference of the substrates. Preliminary observations of tentacle movements and of the mucous trails indicate that previously laid trails can be detected before the foot of the following snail contacts the marker trail. Key words: trail following, mucus, chemoreception, behavior Moving snails leave mucous trails that can be detected optic tentacles at a characteristic angle, around 90° in the by other organisms, including other snails. Many snails can horizontal plane with respect to each other (Lemaire and detect or follow mucous trails of conspecifics and non- Chase 1998, Davis 2004), which may allow the snail to com- conspecifics (Cook 2001). Mucous trail following is used in pare information from its right side to its left side. This homing, finding food sources and mates, and forming ag- tentacle position may also facilitate trail-following. Trails gregations (Chase 1986, Tankersley 1989, Cook 2001). Gas- contain information that can be detected by chemosensory tropods often move faster on previously-laid mucous trails and mechanosensory receptors (Denny 1989). Trail- than on bare substrate (Wareing 1986, Erlandsson and Ko- following may be a primary mechanism to find mates, given stylev 1995) and can detect and respond to the trail’s age as that most terrestrial pulmonates cannot self-fertilize despite well as the physiological state of the individual that left it, being hermaphroditic (Stanisic 1998). avoiding trails left by stressed individuals (Edwards and Mesodon thyroidus (Say, 1817) is a species of terrestrial Davies 2002). Understanding this behavior in gastropods snail native to the eastern United States (Burch 1962, Hu- could assist with conservation of endangered species and bricht 1985) and is common in Lawrence, Kansas, U.S.A. control of invasive species. (Pilsbry 1940, Leonard 1959). It eats plants and fungus Most pulmonate gastropods have two pairs of tentacles (Burch 1962) and, in Kansas, is often found in large groups that are used in chemoreception and to follow mucous trails (> 100 individuals in 1 m7’). Pearce (1990) showed that (Chase 1986, Lemaire and Chase 1998). Both the oral (an- individuals of M. thyroidus follow conspecifics by marking terior) and optic (posterior) tentacles may be used when the paths of snails in the field (using the “spool and line” following mucous trails, and to detect odors associated with technique), but did not examine mucous trails or the role of the trail (Chase 1986, Stirling and Hamilton 1986). Chase _ trail-following. and Croll (1981) found that the giant African land snail To follow up on Pearce’s (1990) study I asked how often Achatina fulica Bowdich, 1822 primarily uses its oral ten- individuals of Mesodon thyroidus follow conspecific trails tacles for following substrate-bound mucus, while the optic and if they can follow in the direction that the trail was laid. tentacles are used for tracking airborne odors. While track- I quantified trail-following on two different substrates ing airborne odors, snails have been observed to hold their (plexiglas and glass) at three different inclines (horizontal, vertical, and 45°). Snails encounter many inclines in their natural environment, and I expected that substrate incline * Work completed at University of Kansas, Department of Ecology | would not influence trail-following. However, snails often and Evolutionary Biology, Lawrence, Kansas 66045-7534, U.S.A. crawl in straight lines up vertical surfaces, which could 157 158 AMERICAN MALACOLOGICAL BULLETIN change the tortuosity (or curviness) of the trail by incline. I also explored the effects of trail remnants on trail following because plexiglas and glass plates retain mucus differentially after cleaning. The goals of this study were to determine: (1) how often the polygyrid snails of Mesodon thyroidus follow mucous trails made by conspecifics, (2) if the substrate in- cline (horizontal, 45°, and vertical) affects trail following, and (3) the effects of trail remnants on trail following. METHODS Approximately 100 specimens of Mesodon thyroidus col- lected from a compost pile in Lawrence, Kansas, U.S.A., were brought into the laboratory. Sexually mature snails (n = 26), determined by the presence of a lip on the aperture, were kept for at least 7 days in individual boxes with damp paper towels (for moisture), and a container of moist peat moss at least 2 cm deep to provide a substrate for egg-laying. Snails were kept at room temperature (20°C) and in natural light near windows in the laboratory, and were fed carrots or lettuce on alternate days. All snails used in this experiment can be found at the University of Kansas Natural History Museum, catalog numbers 002437- 002462. To test whether a snail could detect a conspecific mu- cous trail, the first snail—designated the “marker” snail— was allowed to crawl on a 20 cm x 20 cm pane of picture glass. In each trial a marker snail was centered at an edge of the pane of glass facing the center and allowed to crawl until it reached an edge. Trials were stopped if a snail did not move for more than 5 min or if the mucous trail was not straight for at least one body length. After the snail reached the edge of the glass it was removed. The glass was then rotated 90° (randomized between clockwise and counter- Cross Follow 22° 1/2 * 2007 clockwise) and a second (experimental) snail was placed at least one body length from the trail at approximately 90° to the marker trail. The experimental snail was allowed to crawl until it reached the edge of the glass or was removed due to lack of movement; the trial was discarded if the snail did not move for at least 5 min. Trials were conducted on horizon- tal, sloped (45° from the horizontal), and vertical panes to test if the incline of the substrate affected trail following (both marker and experimental snails crawled on panes at the same incline). All trials were videotaped to facilitate ob- servation of tentacle movements, their angle, and to confirm the paths of the snails. After the experimental snail was removed, the mucus was allowed to dry on the glass before being stained (see below). To ensure independent trials, each pair of marker and experimental snails was used only once. Individual shells were numbered with a paint pen. The trails were stained by soaking the glass panes for 1-5 min in a suspension of carbon particles (laser printer toner) in distilled water (Karowe et al. 1993). These stained trails were photocopied for record keeping. Afterward, the glass plates were soaked for 5 min in 5% acetic acid, washed with soap and distilled water, rinsed with distilled water, soaked for 20 min in 3% sodium hypochlorite, and rinsed six times with distilled water. This cleaning process removed all traces of the mucus from the glass, as confirmed by exposure to carbon particles. To test for the effects of trail remnants on trail follow- ing, plexiglas substrates were used. For plexiglas substrates (20 cm x 20 cm), the procedure was similar to that used with glass, with two exceptions. Snails were kept in groups in containers rather than individually, and each group of snails was tested against another group. Second, the plexiglas was washed with soap and distilled water, rinsed with distilled water, and then rinsed with 95% ethanol. The porous plexi- Turn Figure 1. Categories used to score snail behavior; marker (first) trail is black, experimental (second) trail is gray. TRAIL DETECTION IN MESODON THYROIDUS 159 ee © snail 1 end average snail body length snail 2 start 2 snail 4 start Figure 2. Example of stained trails showing a cross response. glas substrate retained portions of previous trails. Thus, ex- perimental snails on plexiglas were tested in the presence of remnant trails as well as a marker trail. The presence of remnant trails was confirmed by staining with carbon par- ticles after these trials were conducted. Some of these trials were traced onto acetate sheets after being stained with col- ored chalk before the carbon particle method was perfected. All other procedures were the same. For glass and plexiglas substrates, the behavior of each experimental snail was analyzed by examining its mucous trail. I characterized each trial as belonging in one of three categories: cross, follow, and turn (Fig. 1). A cross was de- | average psnail ! snail 2 start body a : length snail 1 end \ snail 2 end, | snail 1-start ° Figure 3. Example of stained trails showing a follow response. fined as any overlap by an experimental trail of less than one body length of a marker snail trail (Fig. 2). A follow was defined as an overlap of at least one body length of marker and experimental trails (Fig. 3). A turn was defined as a complete turn (180° + 15°) of an experimental trail occur- ring just before or adjacent to a marker trail (Fig. 4). Both turning and following were considered to be responses be- cause the snail appeared to have an obvious behavioral re- sponse to a marker snail’s trail. Although crossing a trail was considered a non-response behavior, it does not imply that the snail did not detect the trail. To quantify how often individuals of Mesodon thyroidus follow mucous trails, the counts of the behavioral results of trail encounters were compared by category to two null models. These results were also used to test for a difference by substrate incline. Fisher’s exact test (Sokal and Rohlf 1995) was used to test the grouped results of response (turn/ follow) and non-response (cross) categories against the null models. Because of low counts the results were grouped as response (turn/follow) and non-response (cross) rather than as the three scored categories. The first null model, which was used on both plexiglas and glass trials, was generated by scoring trails against randomly drawn straight lines rather than against the marker trails. These lines were drawn be- tween two randomly chosen points along the edges of a grid that was size-matched to the substrates. The lines were then scored against an experimental snail’s path to generate ex- pected responses for a “null” trail. I used these scores as the snail2 So start a eo rs i ot average Ne snail snail 4 start body length Figure 4. Example of stained trails showing a turn response. 160 AMERICAN MALACOLOGICAL BULLETIN 12 10 ** Count Follow Turn Response Behaviors 22 * 1/2 * 2007 Plexiglass Substrates Turn Cross Follow Response O Vertical @ 45° B Horizontal Figure 5. Response of experimental snail to marker trail on glass and plexiglas substrates. On both substrates, snails were able to detect trails at all substrate inclines and followed most often on horizontal surfaces. * = Results by incline are significantly different than the straight line null (a@ = 0.05) with straight line nulls. ** = Results by incline are significantly different than the straight line null (a@ = 0.05) with both “marker” trail and straight line nulls. expected values in Fisher’s exact test. The second null model used on both plexiglas and glass trials quantified trail cross- ing, following, and turning for two overlaid marker trails, selected randomly. These trails were overlapped by ran- domly orienting the “null” marker substrate (0°, 90°, 180°, and 270° from its original orientation) and then lining up the corners of the substrates between the two trials. The two marker trails were then scored. The results from the second null model generated an additional set of expected values. These expected results were tested using the same statistics as the “null lines.” The second null model did not work as well for the plexiglas substrate because many trials were recorded on acetate sheets that did not include information on snail orientation or trail location on the substrate. This meant that not all plexiglas trials could be tested the same way. These problems make the second null model much less reliable for the plexiglas substrate than the straight line null model. However, both models show the same trend with the results. To analyze the path of the snail trails, | wrote a program in Visual Basic for Applications (VBA—Microsoft) in Excel (Microsoft) to calculate the length, tortuosity, and angle of approach of the tracker snail to the marker path (Davis 2005). Paths were digitized using Didger (v. 2, Golden Soft- ware, Inc.) and uploaded into the VBA program as two- dimensional points. The total length of the path was calcu- lated by adding successive line segments. Tortuosity, a measure of the curviness of the trail, was calculated by taking the ratio of the total length of the path of the snail to the straight line distance between the first and last points (start and stop points of the trail). High tortuosity in the marker trail could hamper the experimental snail’s trail-following ability. A general linear model (GLM) was used to test the tortuosity ratio (beginning-to-end distance/total distance) against both incline of substrate (vertical, horizontal, 45°) and behavioral response (cross, follow, turn). GLM was also used to test behavioral response of the snail against tortu- osity (Sokal and Rohlf 1995). The angle between paths was determined by calculating the angle of approach of the ex- perimental snail in relation to the marker snail. These angles were divided into two groups—those that were within 45° of ——=—-- - TRAIL DETECTION IN MESODON THYROIDUS 161 Table 1. Expected values for Fisher’s exact tests generated from the two different null models. Trials on glass (single trail) Straight “Marker” trail “Null” Lines Cross Follow Turn as null Cross Follow Turn Incline Incline Vertical 43 6 0 Vertical 56 2 0 Horizontal 41 5 2 Horizontal 57 3 l 45° 47 2) 0 45° 63 4 3 Trials on plexiglas (most recent trail) Straight “Marker” trail “Null” Lines Cross Follow Turn as null Cross Follow Turn Incline Incline Vertical 39 2 0 Vertical 34 5 1 Horizontal 3] 1 1 Horizontal 33 4 l 45° 48 2 0 45° 27 4 ] perpendicular to the marker trail and those that were within 45° of parallel to the marker trail. A G-test was used to determine if the angle of approach of the experimental trail was independent of the behavioral response of the experi- mental snail (Sokal and Rohlf 1995). In addition, I tested if the angle of approach between trails affected the direction- ality of following by using a Chi-squared test (a G-test could not be used because of counts of zero). RESULTS Individuals of Mesodon thyroidus were able to detect and follow trails on both substrates and at all inclines, but trail- following diminished as the inclination approached vertical (Fig. 5). At each inclination, snails followed more trails on glass than on plexiglas. On glass surfaces, snails responded to conspecific trails significantly more often than was predicted by either null model at all inclinations (Fisher’s Exact Test, a = 0.05, Tables 1 and 2), but differences between inclina- tions were significant (Fig. 5). An individual snail did not necessarily show the same response to every trail encoun- tered. Of the snails that followed on glass surfaces, 75% of them followed in the direction in which the marker trail had been laid. On plexiglas substrates, the behavioral responses indicate that snails detected trails at all inclinations, even in the presence of remnant trails. However, responses to marker trails were statistically significant (a@ = 0.05) for hori- zontal surfaces only (Table 2). Eighty-six percent of snails that followed a conspecific trail did so in the same direction in which the trail was laid. The plexiglas null model using superimposed marker trails was problematic because some of these trials were conducted before the carbon particle method was developed and the trails were traced without recording their orientation on the plexiglas. Observations of optic tentacles showed that optic ten- tacles were often held parallel to the substrate and oral ten- tacles alternately touched the substrate while the snail was at rest and when it was moving. A snail was often seen lifting its head from the substrate and moving its head from side to side while its tentacles remained stationary with respect to the head. On glass substrates, the results of the trail analysis (us- ing the VBA program) indicated that there was not a sig- nificant difference between the tortuosity of the trail and the behavioral response (from GLM of tortuosity by behavioral response and snail order: Response [Cross, Follow, Turn]: Table 2. Results of Fisher’s exact test. * = results that are signifi- cantly different than the null (a@ = 0.05), indicating that snails are responding to trails. With straight line “trails” as null Incline Probability Vertical 45° Horizontal 2-tailed, single trail (glass) 0.0076* ~—-0.0059* —0.0002* 2-tailed, most recent trail (plexiglas) 0.0574 0.1455 0.0269* With “marker” trail as null ‘Incline Probability Vertical 45° Horizontal 2-tailed, single trail (glass) 0.0004* 0.0355* 4.553 x 10 ** 2-tailed, most recent trail 0.3848 1.0 (plexiglas) 0.1785 162 AMERICAN MALACOLOGICAL BULLETIN F = 0.30, df = 2 p = 0.744, Snail Order: F = 0.73, df = 1 p = 0.395, Interaction [Snail Order*Response]: F=0.41, df = 2 p = 0.662). On plexiglas, there was a significant difference between the tortuosity of the trail and the behavioral re- sponse (from GLM of tortuosity by behavioral response and snail order: Response [Cross, Follow, Turn]: F = 11.26, df = 2 p = 0.0, Snail Order: F = 2.03, df = 1 p = 0.159, Interaction [Snail Order* Response]: F=3.66, df = 2 p = 0.032). A graphi- cal representation of the tortuosity data from glass can be seen in Fig. 6. One trial (45° glass, cross), however, could not be digitized from the carbon visualization and so could not be included in the trail analysis using the VBA program. Similarly, on glass substrates, there was no significant cor- relation of angle difference (between marker and experimen- tal snail) and behavioral response (cross, response) (G = 0.86, p = 0.35, df = 1) for the angle between the two trails. However, on plexiglas substrates there was a significant cor- relation of angle difference and behavioral response (G = 13.5, p < 0.005, df = 1). All but one of the follow responses i) iS) °1/ bo * 2007 on plexiglas occurred with an angle that was within 45° of parallel to the trail. With the subset of snails that followed on both substrates, there was no significant difference between the angle difference (between marker and experimental snail) and directionality of the follow response (right way, wrong way) (glass x* = 0.042, p = 0.838, df = 1; plexiglas x” = 0.194, p = 0.659, df = 1). Table 3 summarizes the results of the trail analysis. DISCUSSION Individuals of Mesodon thyroidus were able to detect conspecific mucous trails at all three substrate inclines tested in this study (horizontal, 45°, and vertical). The substrate type and incline had the greatest effects on the behavioral responses of the snails. The effect of incline is interesting given that snails encounter all inclines of substrate in their environment. Individuals of M. thyroidus did not respond to Tortuosity of trails on glass 1.5 0.5 Total Distance/Beginning to End Distance Marker snail Cross | Follow| | Experimental snail Figure 6. Results from VBA analysis of snail paths on glass substrates. One outlier data point (experimental cross) was excluded because the trail returned to the same place it started, causing the ratio of total distance/beginning to end distance to be very large. Error bars indicate plus/minus one standard error. TRAIL DETECTION IN MESODON THYROIDUS 163 Table 3. Statistics on the results of quantitative analysis of trails using the VBA program. null (a@ = 0.05) *, results that are significantly different than the Statistical Test Substrate Results General Linear Model (GLM) of Glass Source DF Seq SS Adj SS Adj MS F P tortuosity of trail by behavioral Response 2 22.61 22.61 11.30 0.30 0.744 response and snail order Snail 1 55.27 27.79 27.79 0.73 0.395 Response* Snail 2 31.48 31.48 15.74 0.41 0.662 Error 112 4262.97 4262.97 38.06 Total 117 4372.32 Plexiglas Source DF Seq SS Adj SS Adj MS F P Response Z 34.682 34.682 17.341 11.26 0.000* Snail | 0.217 3.134 3.134 2.03 0.159 Response* Snail 2 11.276 11.276 5.638 3.66 0.032* Error 60 92.414 92.414 1.540 Total 65 138.588 G-test of the angle difference by Glass G = 0.86, p = 0.35, df =1 behavioral response (cross/follow) Glass Plexiglas Chi-squared test of the angle difference by following direction Plexiglas G = 13.5, p < x = 0.042, p xX = 0.194, p 0.005*, df =1 = 0.838, df = 1 = 0.659, df = | every trail encountered in this experiment, as indicated by the cross responses. On glass substrates (single mucous trail) no significant effect of inclination of substrate was seen on behavioral response. On plexiglas substrates (most recent mucous trail) only the horizontal incline showed statistically significant response behavior compared to the straight null lines (which were a better null for these data). Cain and Cowie (1978) found that snails with flatter-spired shells were more likely to be active on horizontal surfaces, which could explain why the horizontal incline showed statistically sig- nificant response behavior in both experiments. However, these snails are often found climbing trees in the southern part of their range (personal observations, Davis et al. 2004). The non-significance of the results on plexiglas could be due to the presence of remnant trails. But it could also be an artifact of the small counts due to sample size (n = 11 for each incline) or a difference in chemical composition of the substrate. I did not observe that snails had any potential difficulties in climbing on plexiglas or any other behavioral reaction to indicate a difference on plexiglas versus glass. The results on plexiglas were important in demonstrating that M. thyroidus can detect the most recent mucous trail when remnant trails exist in the environment. Staining of plexiglas showed physical evidence of remnant trails, which may or may not have been detected. Unfortunately, the pres- ence of remnant trails on what was believed to be “clean” plexiglas was only confirmed by staining with carbon par- ticles after this experiment was conducted. However, these results can be used as preliminary data on the effect of trail remnants on trail following. As expected, this experiment verified the field experiments of Pearce (1990), in which many mucous trails were present in the environment and conspecific following was observed. The mechanisms used by snails to detect trails are not well understood (Stirling and Hamilton 1986, Denny 1989, Erlandsson and Kostylev 1995) but the turn response I ob- served indicates that trails can be detected before the foot of the following snail contacts the marker snail’s trail. The ten- tacles of snails can be used to track odors by both tropotaxis and anemotaxis (Chase and Croll 1981, Lemaire and Chase 1998). In the terrestrial pulmonate Achatina fulica, Chase and Croll (1981) observed that both pairs of tentacles are used to orient to concentration gradients and to mucous trails. | was able to confirm Chase and Croll’s (1981) obser- vation that both pairs of tentacles were moved and seemed to be used when following trails. It is possible that each snail could detect every trail it encountered in the experiment but responded only to some of them, which is why I observed many more cross responses than follow or turn responses. In this study, there were clear differences between the behavioral results obtained across the two tested substrates. All of the snails tested against the glass substrate were indi- vidually housed. However, the snails tested on plexiglas were kept in group containers, and it is possible that recent mat- ing of the marker snail could be assessed through the mu- cous trail and effected the response behavior of the experi- mental snail. Feeding time was consistent across all snails. Other studies with non-pulmonate snails have shown that the physiological state of the individual, such as stress due to starvation, can be assessed (Edwards and Davies 2002) by a 164 AMERICAN MALACOLOGICAL BULLETIN — 22 ¢ 1/2 * 2007 following snail. I do not think that food-stress was a factor in my results. If we understand the mechanisms that are used to detect and follow trails in pulmonate snails then it is possible that we can use this knowledge to aid in conservation. For ex- ample, the carnivorous pulmonate snail Euglandina rosea (Férussac, 1821) uses mucous trails to find its prey (Cook 1985, Davis 2005) and has been implicated in the decline and extinction of many native snails on the islands of Hawaii, Tahiti, and Moorea among others (Cowie and Robinson 2003). Understanding the mechanisms of trail detection could be used to create false trails leading to traps, control- ling the pest species, if we can assume that those mucous trails are followed in the direction that they were laid. ACKNOWLEDGEMENTS I wish to thank the Conchologists of America for fund- ing, Dr. Tim Pearce and an anonymous reviewer for their comments on the manuscript, Dr. Catherine Loudon for advice on the project and manuscript, J. Berg, S. Berke, D. Fautin, I. Khavin, and B. Wilson for comments on the manuscript, J. Botz for collecting assistance, and Dr. Norm Slade for statistical advice. LITERATURE CITED Burch, J. B. 1962. How to Know the Eastern Land Snails. William C. Brown Company publishers, Dubuque, Iowa. Cain, A. J. and R. H. Cowie. 1978. Activity of different species of land-snail on surfaces of different inclinations. Journal of Con- chology 29: 267-272. Chase, R. 1986. Lessons from snail tentacles. Chemical Senses 11: 411-426. Chase, R. and R. P. Croll. 1981. Tentacular function in snail olfac- tory orientation. Journal of Comparative Physiology (A) 143: 357-362. Cook, A. 1985. Functional aspects of trail following by the carnivo- rous snail, Euglandina rosea. Malacologia 26: 173-181. Cook, A. 2001. Behavioural ecology: On doing the right thing, in the right place at the right time. In: G. M. Barker, ed., The Biology of Terrestrial Molluscs. CAB! Publishing, New York. Pp. 447-488. Cowie, R. H. and A. C. Robinson. 2003. The decline of native Pacific island faunas: Changes in status of the land snails of Samoa through the 20th century. Biological Conservation 110: 55-65. Davis, E. C. 2004. Odour tracking to a food source by the gastropod Meridolum gulosum (Gould, 1864) from New South Wales, Australia (Camaenidae: Eupulmonata: Mollusca). Molluscan Research 24: 187-192. Davis, E. C. 2005. Trail Detection and Ecological Niche Modeling in Terrestrial Gastropods. Ph.D. Dissertation. University of Kan- sas, Lawrence, Kansas. Davis, E. C., K. E. Perez, and D. J. Bennett. 2004. Euglandina rosea (Férussac, 1821) is found on the ground and in trees in Florida. The Nautilus 118: 127-128. Denny, M. W. 1989. Invertebrate mucous secretions: Functional alternatives to vertebrate paradigms. Symposium for the Society of Experimental Biology 43: 337-366. Edwards, M. and M. S. Davies. 2002. Functional and ecological aspects of the mucus trails of the intertidal prosobranch gas- tropod Littorina littorea. Marine Ecology Progress Series 239: 129-137. Erlandsson, J. and V. Kostylev. 1995. Trail following, speed and fractal dimension of movement in a marine prosobranch, Lit- torina littorea, during a mating and a non-mating season. Ma- rine Biology 122: 87-94. Hubricht, L. 1985. The distributions of the native land mollusks of the Eastern United States. Fieldiana Zoology 24: 1-191. Karowe, D. N., T. A. Pearce, and W. R. Spaller. 1993. Chemical communication in freshwater snails: Behavioral responses of Physa parkeri to mucous trails of P. parkeri (Gastropoda: Pul- monata) and Campeloma decisum (Gastropoda: Prosobran- chia). Malacological Review 26: 9-14. Lemaire, M. and R. Chase. 1998. Twitching and quivering of the tentacles during snail olfactory orientation. Journal of Com- parative Physiology (A) 182: 81-87. Leonard, A. B. 1959. Handbook of gastropods in Kansas. With the technical assistance of E. J. Roscoe and others. Museum of Natural History Miscellaneous Publications 20: 1-224, 11 pls. Pearce, T. A. 1990. Spool and line technique for tracing field move- ments of terrestrial snails. Walkerana 4: 307-316. Pilsbry, H. A. 1940. Land Mollusca of North America (North of Mexico), Vol. 1, Part 2. The Academy of Natural Science of Philadelphia, Philadelphia. Sokal, R. R. and F. J. Rohlf. 1995. Biometry: The Principles and Practice of Statistics in Biological Research. W. H. Freeman and Company, New York. Stanisic, J. 1998. Pulmonata. In: P. L. Beesley, G. J. B. Ross, and A. Wells, eds., Mollusca: The southern synthesis. Fauna of Austra- lia, Vol. 5, Part B. CSIRO Publishing, Melbourne. Pp. 1037- 1061. Stirling, D. and P. V. Hamilton. 1986. Observations on the mecha- nism of detecting mucous trail polarity in the snail Littorina irrorata. Veliger 29: 31-37. Tankersley, R. A. 1989. The effect of trail following on the loco- motion of the marsh periwinkle Littorina irrorata (Mesogas- tropoda: Littorinidae). Marine Behaviour and Physiology 15: 89-100. Wareing, D. R. 1986. Directional trail following in Deroceras re- ticulatum (Miller). Journal of Molluscan Studies 52: 256-258. Accepted: 5 December 2006 i Amer. Malac. Bull. 22: 165-167 RESEARCH NOTE Confirmed absence of a relict population of Gonidea angulata (Lea, 1838) (Mollusca: Bivalvia: Unionidae) in Colorado James (Jay) R. Cordeiro NatureServe, 11 Avenue de Lafayette, 5‘ Fl., Boston, Massachusetts 02111, U.S.A., and Division of Invertebrate Zoology, American Museum of Natural History, Central Park West at 79'" Street, New York, New York 10024, U.S.A., jay_cordeiro@natureserve.org Abstract: The diversity of freshwater mussels in Colorado is low compared to other regions of the U.S.A., with only three extant species (down from seven a century ago). The western ridged mussel (Gonidea angulata) occurs in Pacific Coast drainages from British Columbia to California and eastward to Idaho and Nevada. It has been reported from Colorado based on a single museum specimen from Clear Lake, representing a significant range expansion for this species. During extensive recent surveys, I was unable to pinpoint this locality and regarded it as questionable. Examination of the specimen and locality data indicated this specimen was collected from Clear Lake, Lake County, California. Key words: freshwater mussels, Unionidae, Colorado, distribution, range Studies of freshwater mussels in the Rocky Mountain region are rare compared to other regions of North America. Historical surveys were conducted by Cockerell (1889, 1927), Ellis (1916), Ellis and Keim (1918), and Henderson (1907a, 1907b, 1912, 1924). More recent surveys include Brandauer and Wu (1978), Herrmann and Fajt (1985), and Wu (1989). Gonidea angulata (Lea, 1838) is a freshwater mussel (family Unionidae) belonging to a monotypic genus con- fined to Pacific drainages of northwestern North America (Graf 2002). Its historical range extended from southern British Columbia to southern California and eastward to Idaho and Nevada with extant populations in southwest Washington, northwest Oregon, continuously from south- west Oregon south to southern California, as well as interior Washington and Oregon, southern Idaho, and northern Ne- vada (Ingram 1948, Taylor 1981, COSEWIC 2003, Nature- Serve 2004). I published a distributional guide to the few remaining freshwater mussels in Colorado (Cordeiro 1999) as well as some short notations on distribution of freshwater bivalves in the state (Cordeiro 1998, Cordeiro and MacWilliams 1999). Distributional data included records from museum collections, published and unpublished literature, and field surveys. Prior surveys revealed six species in the state: Ano- dontoides ferussacianus (Lea, 1834), Lampsilis siliquoidea (Barnes, 1823), Lampsilis teres (Rafinesque, 1820), Pygano- don grandis (Say, 1829), Strophitus undulatus (Say, 1817), and Uniomerus tetralasmus (Say, 1831). A seventh species, Lampsilis ventricosa (Barnes, 1823) was cited by Henderson 165 (1907a) for Lodgepole Creek in the northwest corner of the state but this species has been synonymized under both Lampsilis ovata (Say, 1817) and Lampsilis cardium Rafinesque, 1820. Presently, only A. ferussacianus, P. grandis, and U. tetralasmus are extant in Colorado (Cordeiro 1999). Additionally, | cited an unconfirmed record of Gonidea an- gulata (Lea, 1838), based on a single specimen in the Florida Museum of Natural History (FLMNH 65292) collected from “Clear Lake, Colorado.” I was unable to survey Clear Lake due to time, distance, and some confusion over its exact location. There are three different Clear Lakes in Colorado: in Delta and Gunnison Counties (Colorado River Basin), as well as in San Juan County (San Juan River Basin). None are even remotely close to any current or historical occurrences of freshwater mussels in Colorado nor are they within any Pacific drainages where G. angulata is typically found. Sur- veys in western Colorado for freshwater molluscs (including mussels) in 2003 (Sovell and Guralnick 2004) and 2004 (J. Sovell, pers. comm.) have failed to find this or any other species of freshwater mussel. Other recent surveys have detected Gonidea angulata in the Humboldt River drainage (Lahonta Basin) in northern Nevada (Hovingh 2004). Smaller individuals found down- stream in Carlin (Hovingh 2004) indicate this population may be increasing and may have been overlooked in previ- ous surveys as this area historically contained only Anodonta californiensis Lea, 1852, in 1912 and 1939 (Walker 1916, Jones 1940). Despite early reports by Henderson (1924, 1929, 1936) for Utah and Montana, more recent surveys in these states have failed to find any individuals of G. angulata 166 AMERICAN MALACOLOGICAL BULLETIN Figure 1. Florida Museum of Natural History specimen lot FLMNH 65292 of Gonidea angulata (shell length approximately 70 mm). Photo by John Slapcinscky, FLMNH. (Chamberlin and Jones 1929, Jones 1940, Oliver and Bos- worth 1999, Gangloff and Gustafson 2000, Lippincott and Davis 2000). Clear Lake, occupying several towns in Lake County, California, is the state’s second largest freshwater lake and is likely the proper locality for the FLMNH specimen lot of Gonidea angulata. The presence of G. angulata in Clear Lake is documented by Taylor (1981: 143) and by museum speci- mens from the Florida Museum of Natural History (FLMNH 4234) and United States National Museum (USNM 26094). Examination of FLMNH 65292 (Fig. 1) re- vealed writing on the interior left valve, “N. angulata Clear Lake Cal”, with the “a” not quite making a complete circle and having a very short tail. The specimen was collected by C. Mohr, who deposited other specimens in the Florida Mu- seum of Natural History from California (Monterey, Puris- sima, San Diego, San Francisco, and Tomales Bay) but not from Colorado (J. Slapcinsky, pers. comm.). This confirms the locality as Clear Lake, California, and not Clear Lake, Colorado. ACKNOWLEDGEMENTS Thanks to John Slapcinsky (Florida Museum of Natural History) for providing access and the included image of FLMNH 65292 and to John Sovell (Colorado Natural Heri- tage Program) for access to the Colorado heritage database. LITERATURE CITED Brandauer, N. and S.-K. Wu. 1978. The Bivalvia of Colorado. Part 2. The freshwater mussels. Natural History Inventory of Colo- rado 2: 41-60. 22 * 1/2 * 2007 Chamberlin, R. V. and D. T. Jones. 1929. A descriptive illustrated catalog of the Mollusca of Utah. University of Utah Bulletin 19 [Biological Series 1]: 1-203. Cockerell, T. D. A. 1889. Preliminary remarks on the molluscan fauna of Colorado. Journal of Conchology 6: 60-65. Cockerell, T. D. A. 1927. Zoology of Colorado. University of Colo- rado, Boulder, Colorado. Cordeiro, J. R. 1998. Distribution and habitat of freshwater mussels in Colorado. Triannual Unionid Report 14: 19. Cordeiro, J. R. 1999. Distribution and habitat of freshwater mussels (Bivalvia: Unionoida: Unionidae) in Colorado. Natural His- tory Inventory of Colorado 19: 1-56. Cordeiro, J. R. and S. MacWilliams. 1999. Occurrence of the Asian clam, Corbicula fluminea (Miller, 1774) (Bivalvia; Sphaeri- acea; Corbiculidae) in Colorado. The Veliger 42: 278-288. COSEWIC. 2003. COSEWIC Assessment and Status Report on the Rocky Mountain Ridged Mussel Gonidea angulata in Canada. Committee on the Status of Endangered Wildlife in Canada, Ottawa, Canada. Ellis, M. M. 1916. Anodonta danielsi Lea in Colorado. The Nautilus 29: 116-119. Ellis, M. M. and M. Keim. 1918. Notes on the glochidia of Stro- phitus edentulus pavonius (Lea) from Colorado. The Nautilus 32: 17-18. Gangloff, M. M. and D. L. Gustafson. 2000. The freshwater mussels (Bivalvia: Unionoida) of Montana. Central Plains Archaeology 8: 121-130. Graf, D. L. 2002. Molecular phylogenetic analysis of two problem- atic freshwater mussel genera (Unio and Gonidea) and a re- evaluation of the classification of Nearctic Unionidae (Bival- via: Palaeoheterodonta: Unionoida). Journal of Molluscan Studies 68: 65-71. Henderson, J. 1907a. The Mollusca of Colorado. Part I. University of Colorado Studies 4: 77-96. Henderson, J. 1907b. The Mollusca of Colorado. Part II. University of Colorado Studies 4: 167-185. Henderson, J. 1912. The Mollusca of Colorado. Part HI. University of Colorado Studies 9: 53-64. Henderson, J. 1924. Mollusca of Colorado, Utah, Montana, Idaho, and Wyoming. University of Colorado Studies 13: 65-223. Henderson, J. 1929. The non-marine Mollusca of Oregon and Washington. University of Colorado Studies 17: 47-190. Henderson, J. 1936. The non-marine Mollusca of Oregon and Washington—supplement. University of Colorado Studies 23: 81-145, 251-280. Herrmann, S. J. and J. PF. Fajt. 1985. Additional Colorado records of Anodonta grandis grandis Say (Bivalvia: Unionidae). The Nau- tilus 99: 107-109. Hovingh, P. 2004. Intermountain freshwater mollusks, USA (Mar- garitifera, Anodonta, Gonidea, Valvata, Ferrissia): Geography, conservation, and fish management implications. Monographs of the Western North American Naturalist 2: 109-135. Ingram, W. M. 1948. The larger freshwater clams of California, Oregon, and Washington. Journal of Entomology and Zoology 40: 79-92. Jones, D. T. 1940. Recent collections of Utah Mollusca, with ex- ABSENCE OF A RELICT POPULATION OF GONIDEA ANGULATA tralimital records from certain Utah cabinets. Utah Academy of Science, Arts and Letters 27: 33-45. Lippincott, K. and L. B. Davis. 2000. A prehistoric freshwater mus- sel collection from the Schmitt Chert Mine Site (24BW559) near Three Forks, Montana. Central Plains Archaeology 8: 131- 142. NatureServe. 2004. NatureServe Explorer: An online encyclopedia of life [web application]. Version 4.0. NatureServe, Arlington, Virginia. Available at: http://www.natureserve.org/explorer 26 October 2004. Oliver, G. W. and W. R. Bosworth, III. 1999. Rare, Imperiled, and Recently Extinct or Extirpated Mollusks of Utah. State of Utah Division of Wildlife Resources Publication 99-29, Salt Lake City, Utah. Sovell, J. R. and R. Guralnick. 2004. Montane Mollusk and Crusta- cean Survey of Western Colorado—2003 Annual Report. A re- port to the Colorado Division of Wildlife, Fort Collins, Col- orado. Taylor, D. W. 1981. Freshwater mollusks of California: A distribu- tional checklist. California Fish and Game 67: 140-163. Walker, B. 1916. The Mollusca collected in northeastern Nevada by the Walker-Newcomb expedition of the University of Michi- gan. Occasional Papers of the Museum of Zoology (University of Michigan) 29: 1-8. Wu, S.-K. 1989. Colorado freshwater mollusks. Natural History Inventory of Colorado 11: 1-117. Accepted: 25 January 2005 167 Amer. Malac. Bull. 22: 169-170 BOOK REVIEW The Caribbean Land Snail Family Annulariidae: A revision of the higher taxa and a catalog of the species by G. T. Watters (2006). Backhuys Publishers, Leiden, The Netherlands. vii + 557 + 9 + 4 pp. With 9 black-white-figures and 57 maps. ISBN: 90-5782-155-9. Ira Richling Zoologisches Institut, Christian-Albrechts-Universitat zu Kiel, Olshausenstr. 40, 24098 Kiel, Germany, ira@richling.de Although the Annulariidae immediately attract the at- tention of every naturalist visiting the Caribbean region by their beauty and amazing diversity, the tremendous number of taxa - Watters accepts half of the about 1,400 described names as valid species or subspecies - and the difficult access to reliable information prevent the enthusiast from entering deeper into the subject. It would certainly be too high an expectation to find all these problems answered in this new book, but Watters’ bulky 577-page hardcover contribution presents the most comprehensive compilation on the Annu- lariidae to date, including all traceable taxa from subspecies to family level. It intends “...to bring all available infor- mation on the family together, to provide an overview of the group including their morphology, zoogeography, evolu- tion, and systematics” (p. 1). The aim is certainly met with regard to the annotated taxonomic list, which comprises the lion’s share of the book with about 540 pages. On the other hand, the claim — if carefully read — also makes clear what might not be expected and consequently will not be found: substantial new research results on the topics mentioned. So, except for the historical and nomenclatural section (6 pages), the remaining 13 general pages are not ambitious and give the impression that the book results from diligent library and museum collection work, but lacks the inspira- tion of fieldwork. Who else could overlook to tell about the exceptional “walking” of Annulariidae in a general account and only mention it by chance in a discussion of phyloge- netic relationships? The very little but precious information cited on anatomy remains similarly isolated when, in an- other context, the author seriously discusses the published speculation (“deserves further consideration”) that decolla- tion might provide a breathing device at the very top of the shell by a more porous closure (Rees 1964) which, conclud- ing from general anatomy, would require that the digestive gland or parts of the reproductive system spontaneously take up a function as respiratory epithelium! 169 Apart from some weaknesses in the section of general information, the strength and value of Watters’ contribution consists in the immense bibliographic work, the hunt for type material in museum collections and accompanying his- torical information, and the clarification of entangled, often- nightmarish nomenclatural problems. The species account (with 434 pages about 75% of the book) includes data on the original reference; the type lo- cality and known range, the depository of type material, if located; the current systematic assignment; and widely ac- cepted synonyms. Herein it stays in the correctly given limits of a pure catalog, which is appreciated as the most realistic way to approach a review at the family level, added by the “extra” of etymological information for the reader interested in history of science. The listing of species by country under each genus is a most helpful addition for students of specific geographic areas. Unfortunately, and this being inexcusable for author and publisher, the alphabetical arrangements of genera and species fail to compensate for the lack of an index to the thousands of names included in the text, rendering all syn- onyms, subgenera, and subspecies not searchable. The “Out- line of higher taxa” provides only limited help at the supra- specific level. The lack of figures is most unfortunate. Especially in times of digital photography, it is astonishing that not a single (!) member of the Annulariidae is figured on these 577 pages. Only the character that is least necessary to figure is illustrated, with images of 9 opercula hidden at the end of the book instead of accompanying the respective text, while characters that are difficult to describe — such as hidden breathing devices of astonishing shapes and details of radular teeth — remain to the imagination of the reader. At the very least, illustrations of the type species or typical representa- tives of the 56 genera (and perhaps 31 subgenera) would have greatly improved the contribution, especially because 170 AMERICAN MALACOLOGICAL BULLETIN almost all descriptions and delineations of the supraspecific taxa refer to conchological characters. The “revision” as announced in the misleading title dis- appoints in the sense of a new scientific analysis. Without intention to judge the currently proposed classification of the Annulariidae and despite thorough reading I was unable to find any clear systematic concept on which the separation into three subfamilies is based. Not even the descriptions would allow one to attribute a genus to one of the subfami- lies. The extensive use of conchological characters, many of which are known to be or should be suspected to be the result of convergent evolution, restricts the study to the level of knowledge of earlier contributions. Phrases like “is closely related” often have to be understood as “is most similar [conchologically].” The same applies to the recognition of the genera. The descriptions of the randomly selected Cuban genera Blaesospira Crosse, 1890 (p. 89) and Guajaibona Torre and Bartsch, 1941 (p. 91), for example, are identical to the word, except for the addition in the latter of “Whorls nearly adnate,” a character that is not described for Blaeso- spira. No further remark is given. Another example: the genus Parachondria Dall, 1905 is said to be “similar to Colo- bostylus [which belongs to another subfamily] but usually has a higher spire, a non-sinuate lip, and rarely tufts” (p. 42). To add a last example and hereby provoke future research: I failed to find any argument for why breathing devices should have “inexplicably” evolved several times independently in all three subfamilies in Cuba and in a few annulariids in the Bahamas instead of questioning the reliability of the cur- rently-applied system. As to the general remarks on zoogeography, I am per- haps too curious to find clues to understand what really happened in the geological formation of the Caribbean re- gion and the development of its flora and fauna to be sat- isfied with an account on biogeography and endemism that only tries to match the geologic evidence with the annulariid distribution. While this distribution pattern results from a systematic concept that constantly appears to have incorpo- rated the geological evidence it remains simple story telling instead of searching for independent characters and patterns in the Annulariidae for reasoning and concluding. Despite some drawbacks, the book will inevitably serve as a source for future workers on the subject and will cer- tainly be helpful for students of local faunas as an excellent summary of all nomenclatural issues in Annulariidae. Real progress in understanding the phylogeny and evolution of the family in the Caribbean region has been made elsewhere (e.g., Thompson 1978) and will be greatly stimulated by Watters’ contribution. To end with Watters: “It is hoped that it will be the basis for detailed anatomical and genetic studies” (p. 1). 22 * 1/2 * 2007 LITERATURE CITED Rees, W. J. 1964. A review of breathing devices in land operculate snails. Proceedings of the Malacological Society of London 36: 55-66, pls. 3-5. Thompson, F. G. 1978. A new genus of operculate land snails from Hispaniola with comments on the status of family Annulari- idae. The Nautilus 92: 41-54. Accepted: 21 December 2006 Antwerp, Belgium 15 - 20 July 2007 a Universiteit Antwerpen ll museum WORLD CONGRESS OF MALACOLOGY ANTWERP, BELGIUM, 15-20 JULY 2007 FIRST CIRCULAR: September 20", 2006 The congress will be held on campus «Groenenborger» of the University of Antwerp. It is the 16" International Congress of UNITAS MALACOLOGICA (UM). The congress will also host the 73™ annual meeting of the AMERICAN MALACOLOGICAL SOCIETY (AMS). All payments will be in EUROS (€). The congress is open for all contributions in the field of malacology and will host several exciting, open symposia, including: e «Sexual selection» (organised by R. Chase & J. Koene) e «Micromolluscs» (organised by D. Geiger) e «Molluscs as models in evolutionary biology: From local speciation to global radiation» (organised by M. Glaubrecht & T. von Rintelen) e «Molluscan models: Advancing our understanding of the eye» (organised by J. Serb & L. Robles) e «Inventorying the molluscan fauna of the world: Frontiers and perspectives» (organised by P. Bouchet & S. Panha) e «Neogastropod origins and evolution» (organised by J. Harasewych) Yet, there is still room for further symposium or session proposals... There will also be a contributed papers session and a poster session, with posters on display throughout the conference. The conference will start with an « icebreaker » on Sunday late afternoon, 15 July 2007. The scientific presentations are organised in four parallel sessions on Monday, Tuesday, Thursday and Friday. During the poster presentation on Tuesday evening there will be a reception with wine, typical Belgian degustations, cheese and of course... a selection of Belgian beers. On Thursday evening AMS will host its annual auction of molluscan books and paraphernalia (no specimens) to benefit its student programs. The conference dinner will be on Friday evening (several options are still being considered). Wednesday is a free day during which participants can discover the many historical and beautiful places in Antwerp. They can also join one of the suggested congress activities or do whatever they want, of course! Convenient, though modest accommodation will be available at the university campus (about 200 single and 20 double rooms with lavabo, but toilets and showers are shared [even though cabines are individual of course]; prices: 20€/person or 27€/person per night ; breakfast included). Hotel accommodation will be provided in the city centre of Antwerp, near « Antwerpen Centraal » railway station, the main bus terminals and the shuttle bus from/to Brussels international airport. Prices range from 47.5€ (singles) to about 155€ (4 persons) per room per night (breakfast included). ACCOMMODATION WILL BE PROVIDED ON A FIRST COME FIRST SERVED BASIS Congress registration fees are in € (before/after 30 April 2007): Full registration, UM-members 220/270 Full registration, non- UM-members 280/330 Student, UM-member 110/150 Student, non-UM-member 160/200 Fees include registration, abstract book, icebreaker, lunches, drinks and the wine/beer/degustation poster reception. The congress dinner is not included. There will be several student awards for oral and poster presentations, including six awards presented by UM and the Constance Boone Award presented by AMS. UM will provide Travel Grants. Applicants must be a member of UM or of an affiliated organisation. If not, a three-year UM membership will be deduced from the grant. The maximum amount of any Travel Grant will be 800 € for applicants from outside Europe and 400 € for residents in Europe. Application forms will be sent out with the next circular and will be available from the WCM 2007 website and the UM website (see below). They can also be requested when pre-registering (see below). AMS will also be offering travel grants to its student members - please check the AMS website (see below) for information and application. The congress website will soon be activated. In the meantime additional info can be obtained at : wem(@naturalsciences.be You can also PRE-REGISTER at this E-mail address, indicating: (1) what kind of presentation(s) you would like to give (NOTE: each participant can only act as first author of ONE oral presentation and ONE poster presentation), (2) which accommodation you prefer (campus vs hotel + how many persons/room), (3) what type of registrant you are (UM member, student UM, student non-UM), (4) whether you want to receive a Travel Grant application form, (5) whether you need a congress « invitation » or « acceptance » letter (sometimes needed for certain grant applications) Please provide your contact information. Pre-registration IS NOT A FORMAL BOOKING; it simply implies that you will be put on the congress mailing list, so that you will automatically receive the next circulars (via E-mail, unless explicitly requested otherwise). Useful websites: Website of UNITAS MALACOLOGICA: http://www.ucd.ie/zoology/unitas/ Website of the AMERICAN MALACOLOGICAL SOCIETY: http://www.malacological.org About Antwerp : http://www.antwerpen.be/eCache/BEN/S2.html http://www.aviewoncities.com/antwerp.html http://www.trabel.com/antwerp.htm Website of the University of Antwerp (look for campus « Groenenborger »): http://www.ua.ac.be/main.aspx?c=.ENGLISH&n=25878 Belgian railways : http://www.b-rail.be/main/E/index.php Note : There are good international train connections between Antwerp and several major cities outside Belgium. All train tickets can be purchased on-line using your credit card. International bus connections to Antwerp (Europe only): http://www.eurolines.com/ A route planner for if you come by car : http://www.viamichelin.com/viamichelin/gbr/tpl/hme/MaHomePage.htm Airports : Antwerp airport : http://www.antwerpairport.be/en/index.html Brussels airport : http://www.brusselsairport.be/index.cfm?lang=en Charleroi airport : http://www.charleroi-airport.com/BSCA/siteEN.nsf/.Accueil? Readform VLM Airlines, the Flemisch regional airline, offers daily flights from London, Liverpool, and Manchester to Antwerp Airport, where you can take the bus to « Antwerpen Centraal » railway station (10min; 1.50 €). Website of VLM : http://www.flyvlm.com/emc.asp Brussels (Zaventem) is the main international airport in Belgium (home of SN Brussels Airlines and VirginExpress), with a shuttle bus to the city centre of Antwerp (1 bus/hour; trip takes 45min; 8 €). You can also take the train in Brussels Airport and switch trains in « Brussel Noord » railway station (5 trains/hour; 60-80min; 6.70 €). Website of SN Brussels Airlines: http://www.flysn.be/en%5Fbe/home/default.aspx Website of VirginExpress : http://www.virgin-express.com/ Charleroi (Brussels South) airport is a major hub of Ryanair. From Charleroi you can reach « Antwerpen Centraal » railway station by direct train (2 trains/hour; 90-100min; 12.40 €). Website of Ryanair : http://www.ryanair.com/site/EN/?culture=GB Another convenient possibility is to fly to Amsterdam (Schiphol) and take the train from Schiphol Airport to Antwerp. You can take either the fast trains (Thalys; you have to book in advance and it is more expensive) or the « normal » direct trains (1train/hour; |120min; 26 €). PLEASE NOTE THAT WE (= CONGRESS ORGANISATION) WILL NEITHER PROVIDE TRANSPORTATION, NOR WILL WE PICK UP PEOPLE AT AIRPORTS OR RAILWAY STATIONS. WE COUNT ON YOUR SCOUTING TALENTS! Currency Converter : http://www.oanda.com/convert/classic? user=onlineconversion&lang=en SEE YOU IN ANTWERP ! Thierry Backeljau President of Unitas Malacologica Ligaszewski, M. 22: 1 2 ee Ay 22: | an net, B. 22: 7 Salas Orono, E. 22: 17 Cuezzo, M. G. 22: 17 Romero, F. 22: 17 Kantor, Y. I. 22: 27 Bouchet, P. 22: 27 Schejter, L. 22: 75 INDEX TO VOLUME 22 AUTHOR INDEX Bremec, C. S. 22: 75 Coote, T. 22: 83 Dominguez, M. 22: 89 Quintas, P. 22: 89 Troncoso, J. S. 22 Hermosillo, A. 22: 119 Valdés, A. 22: 119 White, M. M. 22: . 39 Chejlava, M. 22: Fried, B. 22: 139 Sherma, J. 22: 139 Nanbu, R. 22: 143 Yokoyama, E. 22: 143 Mizuno, T. 22: 143 Sekiguchi, H. 22: 143 Davis,-E. C. 22: 157 Cordeiro, J. R. 22: 165 Richling, I. 22: 169 PRIMARY MOLLUSCAN TAXA INDEX [first occurrence in each paper recorded, new taxa in bold face] Achatina 22: 157 Adelopoma 22: 21 aenea, Guppya 22: 22 Aeolidiidae 22: 133 Aequipecten 22: 75 affinis, Partula 22: 85 africana, Cerberilla 22: 136 africana, Herviella 22: 121 Agariniinae 22: 70 Agropecten 22: 79 alaos, Belloliva 22: 34 Amalda 22: 27 amblys, Calyptoliva 22: 66 ambonesis, Cerberilla 22: 136 Amphissa 22: 7 Ancilla 22: 27 Ancillariinae 22: 27 Ancillina 22: 71 Ancillinae 22: 70 angulata, Gonidea 22: 165 Annulariidae 22: 169 annulata, Cerberilla 22: 136 annulata, Phyllidiella 22: 113 Anodontoides 22: 165 apoma, Belloliva 22: 37 arabica, Phyllidia 22: 90 arboreous, Zonitoides 22: 21 asamusiensis, Cerberilla 22: 136 aspersa, Helix 22: 1 aspersa, Helix aspersa 22: 1 Atrina 22: 78 attenuata, Samoana 22: 85 aureomarginata, Amalda 22: 71 backeljaui, Phyllidiella 22: 89 Baryspira 22: 27 behrensi, Cuthona 22: 128 Belloliva 22: 27 bernardettae, Cerberilla 22: 136 Biomphalaria 22: 139 Bivalvia 22: 165 Blaesospira 22: 170 bolis, Calyptoliva 22: 60 braziert, Belloliva 22: 28, 67 brazieri, Olivella 22: 27 bridgesii, Pomacea 22: 139 burchi, Samoana 22: 85 Calliostoma 22: 78 Calyptoliva 22: 27 Calyptoliva 22: 60 Calyptraea 22: 78 cardinalis, Phyllidiopsis 22: 99 cardium, Lampsilis 22: 165 Ceratophyllidia 22: 113 Cerberilla 22: 133 Charopidae 22: 21 chavezi, Cerberilla 22: 133 Chlamys 22: 8 clara, Partula 22: 85 174 cloaca, Herviella 22: 121 cocoachroma, Cuthona 22: 130 coelestis, Phyllidia 22: 93 Colobostylus 22: 170 Columbellidae 22: 7 columbiana, Amphissa 22: 7 concinna, Cuthona 22: 130 cooraburrana, Phyllidiella 22: 113 Corbicula 22: 143 corneous, Planorbarius 22: 141 Crepidula 22: 78 cucullus, Eubranchus 22: 132 Cuthona 22: 121 cytherea, Partula 22: 86 destinyae, Cuthona 22: 121 divae, Cuthona 22: 130 diversicolor, Cuthona 22: 127 dorcas, Belloliva 22: 48 Drymaeus 22: 18 echizenicus, Eubranchus 22: 133 edulis, Mytilus 22: 78 elegans, Phyllidia 22: 93 ellenae, Belloliva 22: 44 Entomoliva 22: 71 Epiphragmophora 22: 18 Eubola 22: 78 Eubranchidae 22: 130 Eubranchus 22: 130 Euglandina 22: 83, 164 exquisita, Belloliva 22: 37 Facelinidae 22: 119 felipponei, Aequipecten 22: 76 felipponei, Chlamys 22: 76 felipponet, Flexopecten 22: 75 ferussacianus, Anodontotdes 22: 165 filosa, Partula 22: 86 Flexopecten 22: 75 Fryeria 22: 97 fulica, Achatina 22: 157 fulica, Lissachatina 22: 83 Gastropoda 22: 1, 89, 157 Gemmoliva 22: 66 glabrata, Biomphalaria 22: 139 golbachi, Radiodiscus 22: 24 Gonidea 22: 165 Gracilanilla 22: 71 grandis, Pyganodon 22: 165 granulata, Phyllidiella 22: 113 Guajaibona 22: 170 Guppya 22: 22 hageni, Phyllidiella 22: 89 hamanni, Cuthona 22: Helicidae 22: 1 Helicodiscidae 22: 24 Helisoma 22: 139 Helix 22: 1 Herviella 22: 119 hyalina, Partula 22: 85 hyltonscottae, Zilchogyra 22: 24 iota, Belloliva 22: 43 jackieburchi, Samoana 22: 86 japonica, Corbicula 22: 143 katiae, Radiodiscus 22: 24 krempfi, Phyllidiopsis 22: 99 lacanientai, Oliva 22: 67 Lampsilis 22: 165 leucozona, Belloliva 22: 28 lilloana, Guppya 22: 22 Lilloiconcha 22: 22 Lissachatina 22: 83 lizae, Cuthona 22: 123 lizae, Phyllidiella 22: 111 longi, Cuthona 22: 123 longicirrha, Cerberilla 22: 136 lusoria, Meretrix 22: 143 Lymnaea 22: 141 Mactra 22: 144 madapanamensis, Eubranchus 22: 132 magallanicus, Placopecten 22: 79 marindica, Phyllidia 22: 97 maxima, Helix aspersa 22: 1 maximus, Pecten 22: 79 meandrina, Phyllidiella 22: 113 melanocera, Phyllidia 22: 103 menindae, Fryeria 22: 97 Meretrix 22: 143 Mesodon 22: 157 millenae, Cuthona 22: 124 mirabilis, Entomoliva 22: 71 misakiensis, Eubranchus 22: 132 moebi, Cerberilla 22: 136 mosslandica, Cerberilla 22: 136 Musculista 22: 144 Mytilus 22: 78 Neocyclotidae 22: 18 Neogastropoda 22: 27 nigra, Phyllidiella 22: 113 nitidus, Zonitoides 22: 22 nobilis, Phyllidia 22: 101 Nodipecten 22: 78 nodosa, Partula 22: 85 nodosus, Nodipecten 22: 78 Nudibranchia 22: 89 Nuttalia 22: 8 obeon, Belloliva 22: 43 obscurata, Nuttalia 22: 8 ocellata, Phyllidia 22: 95 olivaceus, Eubranchus 22: 132 Olivancillaria 22: 70 Olivancillariinae 22: 70 Olivellidae 22: 70 Olividae 22: 27 Olivinae 22: 70 Opeas 22: 21 Opisthobranchia 22: 89, 119 ornata, Cuthona 22: 127 otaheitana, Partula 22: 85 ovata, Lampsilis 22: 165 Parachondria 22: 170 pardalis, Olivella 22: 27 Partula 22: 83 Partulidae 22: 83 patagonica, Zygochlamys 22: 75 Pecten 22: 79 peregra, Lymnaea 22: 141 philippinarum, Ruditapes 22: 143 phoenix, Cuthona 22: 123 Phyllidia 22: 90 Phyllidiella 22: 89 La Phyllidiidae 22: 89 Phyllidiopsis 22: 97 Physa 22: 139 picta, Fryeria 22: 97 picta, Phyllidia 22: 97 pinnifera, Cuthona 22: 124 pipeki, Phyllidiopsis 22: 99 Planorbarius 22: 141 Platinopecten 22: 79 Polygyridae 22: 157 Pomacea 22: 139 pomatia, Helix 22: 1 producta, Partula 22: 86 puelchana, Ostrea 22: 77 Pulmonata 22: 1 pumilium, Opeas 22: 21 pungoarena, Cerberilla 22: punicea, Cuthona 22: 12 Pupullidae 22: 21 purpuratus, Agropecten 22: 79 pustulosa, Phyllidia 22: 103 pustulosa, Phyllidiella 22: 103 Pyganodon 22: 165 Radiodiscus 22: 21 Reticulidia 22: 113 ripium, Eubranchus 22: 132 rosans, Phyllidiella 22: 113 rosea, Euglandina 22: 83, 164 Ruditapes 22: 143 rudmant, Phyllidiella 22: 109 rueppelli, Fryeria 22: 97 Samoana 22: 85 136 sanjuanensis, Eubranchus 22: 133 Scutalus 22: 18 seminuda, Atrina 22: 78 senhousia, Musculista 22: 144 shireenae, Phyllidiopsis 22: 97 siliquoidea, Lampsilis 22: 165 simplex, Belloliva 22: 37 speciosa, Cuthona 22: 127 steinbecki, Eubranchus 22: 133 Streptaxidae 22: 18 Strophitus 22: 165 subnodosus, Nodipecten 22: 79 Systrophiidae 22: 18 tabulata, Olivella 22: 67 tanna, Cerberilla 22: 136 tatyanae, Calyptoliva 22: 65 tehuelchus, Aequipecten 22: 75 teres, Lampsilis 22: 165 Tergipedidae 22: 121 tetralasmus, Uniomerus thomei, Radiodiscus 22: thyroidus, Mesodon 22: trilineata, Phyllidia 22: 90 triticea, Belloliva 22: 67 trivolvis, Helisoma 22: 139 trochilioneides, Wayampia 22: 19 tuberculata, Phyllidia 22: 99 tucma, Adelopoma 22: 21 Turrancilla 22: 71 undulatus, Strophitus 22: 165 Uniomerus 22: 165 Unionidae 22: 165 varicosa, Phyllidia 22: 90 veneriformis, Mactra 22: 144 ventricosa, Lampsilis 22: 165 ventricosus, Agropecten 22: 79 Veronicellidae 22: 18 176 Wayampia 22: 19 yessoensis, Platinopecten 22: 79 yolandae, Eubranchus 22: 130 zeylanica, Phyllidia 22: 105 zeylanica, Phyllidiella 22: 89 ziczac, Eubola 22: 78 Zilchogyra 22: 24 Zonitoides 22: 21 Zygochlamys 22: 75 INFORMATION FOR CONTRIBUTORS The American Malacological Bulletin is the scientific pub- lication of the American Malacological Society and publishes notable contributions in malacological research. Manu- scripts concerning any aspect of original, unpublished re- search, important short reports, and detailed reviews dealing with molluscs will be considered for publication. Each original manuscript and accompanying illustrations must be submitted with two additional copies for review purposes. Text must be printed in 12 pt font on one side of 8-1/2 x 11 inch paper, double-spaced, with all pages num- bered consecutively. Leave ample margins on all sides. The form of the manuscript should follow that outlined in the Council of Biology Editors Style Manual (sixth edition, 1994). This can be purchased from the CBE, 11 S. LaSalle Street, Suite 1400, Chicago, IL 60603, USA. Text should be arranged in sections as follows: 1. Cover page with title, authors, addresses, email addresses, and suggested running title of no more than 50 characters and spaces. Authors should also supply five key words, placed at the base of this page, for indexing purposes. Key words should not duplicate terms already in title. 2. Abstract (less than 5% of manuscript length) 3. Text of manuscript starting with a brief introduction followed by methodology, results, and discussion. Sepa- rate sections of text with centered subtitles in capital letters. Acknowledgments Literature cited Figure legends Tables (each on a separate sheet, headed by a brief legend) SEM ue All binomens, excluding non-molluscan taxa, must in- clude the author and date attributed to the taxon the first time the name appears in the manuscript, such as Crassostrea virginica (Gmelin, 1791). The full generic name and specific epithet should be written out the first time a taxon is referred to in each paragraph. The generic name can be abbreviated in the remainder of the paragraph as follows: C. virginica. References should be cited within text as follows: Hillis (1989) or (Hillis 1989). Dual authorship should be cited as follows: Yonge and Thompson (1976) or (Yonge and Thompson 1976). Multiple authors of a single article should be cited as follows: Beattie et al. (1980) or (Beattie et al. 1980). In the section of literature cited, references should also be typed double-spaced. All authors must be fully identified and listed alphabetically; journal titles should not be abbre- viated. Citations should be formatted as follows: Donovan, D.A., J. P. Danko, and T. H. Carefoot. 1999. Functional significance of shell sculpture in gastropod molluscs: Test of a predator-deterrent hypothesis in Ceratostoma foliatum (Gmelin). Journal of Experimental Marine Biology and Ecology 236: 235-251. Seed, R. 1980. Shell growth and form in the Bivalvia. In: D.C. Rhoads and R. A. Lutz, eds., Skeletal Growth of Aquatic Organisms, Plenum Press, New York. Pp. 23-67. Yonge, C. M. and T. E. Thompson. 1976. Living Marine Mol- luscs. William Collins Son and Co., Ltd., London. Dall, W. H. 1889. Reports on the results of dredging, under the supervision of Alexander Agassiz, in the Gulf of Mexico (1877-78) and in the Caribbean Sea (1879-80), by the United States Coast Survey Steamer “Blake,” Lieu- tenant-Commander C. D. Sigsbee, U.S. N., and Com- mander J. R. Bartlett, U.S. N., commanding. Report on the Mollusca, Pt. 2: Gastropoda and Scaphopoda. Bulle- tin of the Museum of Comparative Zoology 18: 1-492, pls. 10-40. Orbigny, A, d’. 1835-46. Voyage dans lAmérique Meridionale (le Brésil, la République Orientale de PUruguay, la République Argentine, la Patagonie, la Reé- publique du Chili, la République de Bolivia, la Republique du Pérou), exécuté pendant les années 1826, 1827, 1828, 1829, 1830, 1831, 1832 et 1833. Vol. 5, Part 3 (Mol- lusques). Bertrand, Paris. Dates of publication: pp. 1-48, [1835], pp. 49-184 [1836], pp. 185-376 [1837], pp. 377- 408 [1840], pp. 409-488 [1841], pp. 489-758 + pls. 1-85 [1846]. Hurd, J. C. 1974. Systematics and Zoogeography of the Union- acean Mollusks of the Coosa River Drainage of Alabama, Georgia and Tennessee. Ph.D. Dissertation, Auburn Uni- versity, Alabama. U.S. Environmental Protection Agency. 1990. Forest riparian habitat survey. Available at: http://www.epa.gov/waterat- las/geo/iil6_usmap.html 25 January 2003. Illustrations should be clearly detailed and readily repro- ducible. Fine patterns and screens do not reproduce well. All line drawings should be in black, high quality ink. Photo- graphs must be on glossy, high contrast paper. All diagrams must be numbered in the lower right hand corners and adequately labeled with sufficiently large labels to remain readable with reduction by one half. Scale bars must appear on the figure, or the caption must read Horizontal field width = x mm or x pm. All measurements must be in metric units. All illustrations submitted for publication must be fully cropped, mounted on a firm white backing ready for reproduction, and have author’s name, paper title, and fig- ure number on the back. All figures in plates must be con- tiguous. Additional figures submitted for review purposes must be of high quality reproduction. Xerographic repro- duction of photomicrographs or detailed photographs are not acceptable for review. Explanations of abbreviations used in a figure should occur in the figure legend. Indicate in text margins the appropriate location in which figures should appear. Color illustrations can be included at extra cost to the author. Original illustrations will be returned to author if requested. Final submission of accepted, revised manuscripts should include one printed copy of the text, tables, etc. and an addi- tional copy of the text, tables, and illustrations in electronic form on a CD, Zip disk, or 3.5” diskette. Original media will be returned to the author if requested. Text documents and tables should be in MSWord or RTF formats for Macintosh or Windows. All illustration files should be in TIFF or EPS formats. Files for full color images must be in CMYK color space. Do not submit native application formats. AMB quality reproduction will require grayscale and color files at resolu- tions yielding approximately 300 dpi. Bitmapped line art should be submitted at resolutions yielding 600-1200 dpi. These resolutions refer to the output size of the file; if you anticipate that your images will be enlarged or reduced, resolutions should be adjusted accordingly. Each individual figure or graphic must be supplied as a separate, stand-alone file, accompanied by a high quality hard copy. Figure files must be named with their respective numbers and graphic type such as Fig1.tif, Figure2.eps, etc. Long file names are acceptable. When creating figures, use font sizes and line weights that will reproduce clearly and accurately when figures are sized to the appropriate column width. Do not include figure legends in a graphic file. Any manuscript not conforming to AMB format will be returned to the author for revision. New Taxa. The Bulletin welcomes complete descriptions of new molluscan taxa. Establishment of new taxa must conform with the International Code of Zoological Nomenclature (1999). Descriptions of new species-level taxa must include the following information in the order as given: higher taxon designation as needed for clarity; family name with author and date; generic name with author and date; Genus species author sp. nov. followed by numeration of all figures and tables; complete synonymy (if any); listing of type material with holotype and any other type material clearly designated along with complete museum catalog or accession information; listing of all additional non-type material also with full museum deposition information; type locality; diagnosis and full description of material done in telegraphic style including measurements and zoogeographic distribution as necessary; discussion; etymology. Descriptions of new supraspecific taxa should include type species (for new genus) or type genus (for new family), diagnosis and full description done in telegraphic style, and list of included taxa. Proofs. Page proofs will be sent to the author and must be checked for printer’s errors and returned to the managing editor within three days. Significant changes in text, other than printer’s errors, will produce publishing charges that will be billed to the author. Charges. There are no mandatory page costs to authors lack- ing in financial support. Authors with institutional, grant, or other research support will be billed for page charges. The current rate is $35.00 per printed page. Acceptance and ulti- mate publication is in no way based on ability to pay page costs. Reprints. Order forms and reprint cost information will be sent with page proofs. The author receiving the order form is responsible for insuring that orders for any co-authors are also placed at that time. Submission. Submit all manuscripts to Dr. Janice Voltzow, Editor-in-Chief, Department of Biology, University of Scranton, Scranton, PA 18510-4625, USA. Subscription Costs. Institutional subscriptions are available at a cost of $65.00 per volume. Membership in the American Malacological Society, which includes personal subscriptions to the Bulletin, is available for $60.00 ($20.00 for students, $60.00 for affiliated clubs). Outside the U.S. postal zones, add $5.00 surface and $10.00 airmail per volume. All prices quoted are in U.S. funds. For membership information and institutional subscriptions contact Dr. Susan Cook, Trea- surer, American Malacological Society, 4201 Wilson Blvd., STE 110-455, Arlington, VA 22230. For other information, including availability of back issues, contact Dr. Janice Voltzow, Department of Biology, University of Scranton, Scranton, PA 18510-4625, USA. Complete information also available at the AMS website: http://www.malacological.org THE AMERICAN MALACOLOGICAL SOCIETY http://www. malacological.org Dr. Susan B. Cook, Treasurer American Malacological Society, Inc. 4201 Wilson Boulevard, Ste. 110-455 Arlington, VA 22203-1859 USA NEW MEMBERSHIP APPLICATION FORM Please fill out both sides of this form and mail them with payment of dues to the Treasurer at the address above. Membership is by CALENDAR YEAR (January | through December 31). NAME LAST NAME TITLE FIRST NAME MIDDLE INITIAL Mailing address (for Bulletin and annual dues notices): Institutional addresss (for membership directory if different from mailing addresss): Telephone (Office): E-mail: Telephone (Home): FAX: Home Page URL: Special area of Study or Interest (60 characters maximum): STUDENT MEMBERS only: College/University in which you are enrolled: Signature of Advisor SCHEDULE OF ANNUAL DUES AND OTHER CHARGES: Members at non-U.S. addresses, there is an additional fee to cover postage for the Bulletin. Please indicate Surface or Airmail under POSTAGE below and include fee in total payment. MEMBERSHIP CATEGORY (please check box and circle amount paid): L) Regular Member — One year dues (2006) $ 60.00 LJ Regular Member — Two years (2006 & 2007) $ 105.00 L) Regular Member — Three years (2006-2008) $ 145.00 OY Each additional family member per year $ 1.00 J Student Member (requires institution & instructor signature) $ 20.00 L) Sustaining Member — Dues plus $25.00 $ 85.00 L) Affiliate Membership (Shell clubs and other organizations) $ 60.00 POSTAGE: For U.S. addresses, Bulletin is mailed bulk rate at no additional charge. For all countries outside the U.S., please indicate mail category and remit fee: J AIRMAIL $10 LJ SURFACE MAIL $5.00 $ = TAX-DEDUCTIBLE GIFT: To Symposium Endowment Fund $ a L) To Student Research Grant Endowment Fund $ = TOTAL ENCLOSED: $ Because of high bank charges, payment can be made only by checks on a U.S. bank (with proper coding at the bottom), by international money order, or by MasterCard or Visa. Make checks payable to the A.M.S. or AMERICAN MALACOLOGICAL SOCIETY. TO SAVE TIME AND POSTAGE, the Treasurer does not issue dues receipts or confirm membership acceptances. In special cases where documentation is required, members may request an electronic acknowledgment from the Treasurer at scook919@msn.com. If you wish to make payment via MasterCard or Visa, please complete the following: LU) MasterCard LU) VisaCard # Exp: Signature of Cardholder FOR AMS MEMBERSHIP RECORDS Send form and payment to Treasurer. WELCOME TO A.M.S. Thank you for becoming a member! wii eastern Pacific. ALICIA HERMOSILLO and ANGEL VALDES ................0.0-0c0eeeae The concentration of calcium carbonate in shells of freshwater snails. MEREDITH M. WHITE, MICHAEL CHEJLAVA, BERNARD FRIED, and JOSEPH SHERMA ..................005. Larval settlement and recruitment of a brackish water clam, Corbicula japonica, in the Kiso estuaries, central Japan. RYOGEN NANBU, ESTUKO YOKOYAMA, TOMOMI MIZUNO, ad HIDEO SEKIGUCEL | 4 s:is nsies's cee ee see cece sees eee bbe conn wae Saeeeae er erent Investigation in the laboratory of mucous trail detection in the terrestrial pulmonate snail Mesodon thyroidus (Say, 1817) (Mollusca: Gastropoda: Polygyridae). Pee ELM CODAVIS cccccnde che se acada § Ones. sugne case eee ae y Ogle oe ge a eee RESeAT CH OIOLE occ cc ees coe ee eb ea aw hb wlelman ede eae dem gurauee we tele ne ae wane andl ome BOG Re VIEW. Sc eee Uv ee how Wink adobe “Lovie he ew Went pn a ws os cg co ee