COLA te H en HR а ans Nord H : E : make : = $ E o 5 at > = Пат 2e Ma A + À > ET ; “ ИИ С . 5 i x ; 7 : E A rra aha 2 Fete “* ss ‘ - e y Л sy PRISON EEE aie SES hae ва OA OPEN DS ante ad PTE Я 2A na ям жа ое хо sv. EN CPR A mite! po Им RL HARVARD UNIVERSITY € Library of the Museum of Comparative Zoology и [ar | A | т A NER | a { À x } VOL. 36, NO. 1-2 . MALACOLOGIA | 1995 | | | Aue | Se TUE CONTENTS о MARÍA VILLARROEL Y JOSÉ STUARDO i ey Morfologia del Estomago y Partes Blandas en Mytella strigata (Hanley 1849 Mey (Bivalvia: Mytilidae) ia La 222 2a ns о a A Pet ERBE fi | JOST BORCHERDING К 3 п Laboratory Experiments on the Influence of Food Availability, Temperature and FX } - Photoperiod on Gonad Development in the Freshwater Mussel Dreissena EN „NPalymearpha 2/72 So oe) O A er ee RSR АЕ Е i 15 4 R. ‚ ARAUJO, J. М. ВЕМОМ, D. МОВЕМО & M. А. RAMOS HER - 39 Relaxing Techniques for a! Molluscs: Trials for Evaluation of Different à Méthode Ch Den И O a ee KENNETH C. EMBERTON qlo TEN Land-Snail Community Morphologies of the Highest-Diversity Sites of Ман gascar, North America, and New Zealand, with RECO Alternatives to”. HelghteDiameter Plots N Rune ac eben ben eee eee See > KENNETH C. EMBERTON \ ENT FO | dino PE Distributional Differences Among Acavid Land Snails Around Antalahdy Mada- eee gascar: Inferred Causes and Dangers Of Extinction N RTS ESS ES HA. 4 Ir KATHERINE COSTIL & JACQUES DAGUZAN | ne iR + Effect of Temperature on Reproduction in Planorbañus corneus (L.) and Plan- . I orbis planorbis (L.) Throughout the Life “ad О OS + eats See ИТ L М. COOK 4 J. BRIDLE ewe ye! 7 1 Colour Polymorphism in the a Snail Littoraria intermedia т Sinai..... 91 > MICHAEL G. GARDNER, PETER В. MATHER, IAN WILLIAMSON & JANE M. HUGHES : via р The Relationship Between Shell-Pattern Frequency and Microhabitat Variation nS - in the Intertidal Prosobranch, ‚Clithon oualaniensis (Lesson) .......... Jesse ta MIGUEL, IBÁÑEZ, ELENA PONTE-LIRA 8 MARÍA R. ALONSO < | El Género Canariella Hesse, 1918, y su Posición: en Ja Familie Hygromidae e | | Gastropoda, Pulmonata, Helicoidea) .......... rn IA 2 «le wu; EN "N. ELEUTHERIADIS & М, LAZARIDOU-DIMITRIADOU ~~ Age-Related Differential Catabolism in the Snail Bithynia graeca (Westárluñd, À ay 1879) and its Significance in the Bioenergetics of Sexual Dimorphism “4 №, ee HEINZ BRENDELBERGER | ' Y # Kar Dietary Preference of Three Freshwater Gastropods for Eight Natural Foods of | Different Engraetic ‘Content Lil. a de ee PE VE SEE a ROBERT H. COWIE, GORDON M. NISHIDA, YVES BASSET & SAMUEL M. GON, mo Patterns of Land Snail Distribution 16, a Montane Habitat on the ‘sland oft > Ва [a dis aye a ele ase mia A O AA LER e 15 ALAN Е. STIVEN a ; O E Genetic Heterozygosity and Growth Rate in the Southern Appalachian Land A 4 y Snail Mesodon normalis (Pilsbry 1900): The Effects of реа Stress . Tey | DAVID В. LAWRENCE / IR N N Rr A sats. | Diagnosis of the Genus Crassostrea (Bivalvia, Ostreidae) . ón ot tk an, |: KENNETH С. EMBERTON & SIMON TILLIER | i р 7 o! 14 Clarification and Evaluation of Tillier’s a0 989) Siylommatophora Mono- 1 EA A A A E AS AU RS grap RA Bu ae / RS / i . | | PRE ee APE UE e la Jr ÓN ' | fe A (a | SS , | SNA oe a | L L * 4 x he Г iN À $ — | | МСО... } VOL. 37, NO. 1 LIBRARY 1995 À | 210. 1995 y JER ds | TY MALACOLOGIA \ À International Journal of Malacology Revista Internacional de Malacologia Journal International de Malacologie Международный Журнал. Малакологии Internationale Malakologische Zeitschrift MALACOLOGIA | he Editor-in-Chief: GEORGE M. DAVIS t Editorial and Subscription Offices: Department of Malacology The Academy of Natural Sciences of Philadelphia т 1900 Benjamin Franklin Parkway D Sp I Philadelphia, Pennsylvania 19103-1195, U.S.A. ь Co-Editors: ^ not TIER: EUGENE COAN CAROL JONES KR California Academy of Sciences ~ Denver, CO - A San Francisco; CA ; ` Assistant Managing Editor: Е CARYL HESTERMAN er ` Associate Editors: ” ieee и ANNE GISMANN University of Michigan Ма : ) | | € Ann Arbor ' ‚Egypt, : Er y MALACOLOGIA is published by the INSFITUTE OF MALACOLOGY, the Sponsor Members OF which (also serving as editors) are: | | 4 - RUDIGER BIELER ^^ ALAN KOHN Field Museum, Chicago University, of Washington, Seattle à JOHN BURCH ~ JAMES NYBAKKEN 3 te MELBOURNE R. CARRIKER Moss Landing Marine Laboratory President Elect | California | University of Delaware, Lewes CLYDE FE. E. ROPER + / *] GEORGE M. DAVIS ‘ Smithsonian Institution e Secretary and Treasurer ~ Washington, D.C. > } | CAROLE S. HICKMAN, President SHI-KUEI WU > University of California, Berkeley ) University of Colorado Museum, Boulder | HAE Participating Members, Ke N EDMUND GITTENBERGER JACKIE L. VAN GOETHEM | Secretary, UNITAS MALACOLOGICA | Treasurer, UNITAS MALACOLOGICA | Rijksmuseum van Natuurlijke | Koninklijk Belgisch Instituut Historie | voor Natuurwetenschappen EDS Leiden, Netherlands pe Brussel, Belgium j ee у $. | NM Us aoe * а Emeritus Members bhai A J. FRANCIS ALLEN, Emerita : € ROBERT ROBERTSON | (VERSER vay Environmental Protection Agency The Academy of Natural Sciences. «Л AE Washington, D.C. Lee Philadelphia, Pennsylvania due: № q г y KENNETH J. BOSS. | y Museum of Comparative Zoology - МО, RUSSELL-HUNTER Arbo Massachusetts \ Easton, Maryland | \- \ ГЕ, y y PAT | "Copyright © 1995 by the Institute of Mall ; Y J. A. ALLEN Marine Biological Station Millport, United Kingdom E. E. BINDER Muséum d'Histoire Naturelle Genève, Switzerland А. J. САМ University of Liverpool United Kingdom P. CALOW University of Sheffield United Kingdom J. G. CARTER University of North Carolina Chapel Hill, U.S.A. R. COWIE Bishop Museum Honolulu, HI., U.S.A. А. Н. CLARKE, Jr. Portland, Texas, U.S.A. B. C. CLARKE University of Nottingham United Kingdom R. DILLON College of Charleston SC, U.S.A. C. J. DUNCAN University of Liverpool United Kingdom D. J. EERNISSE California State University Fullerton, U.S.A. E. GITTENBERGER Rijksmuseum van Natuurlijke Historie Leiden, Netherlands Е. GIUSTI Universita di Siena, Italy А. М. GOLIKOV Zoological Institute St. Petersburg, Russia 1995 EDITORIAL BOARD 5. J. GOULD Harvard University Cambridge, Mass., U.S.A. A. V. GROSSU Universitatea Bucuresti Romania T. HABE Tokai University Shimizu, Japan R. HANLON Marine Biomedical Institute Galveston, Texas, U.S.A. J. А. HENDRICKSON, уг. Academy of Natural Sciences Philadelphia, PA, U.S.A. D. M. HILLIS University of Texas Austin, U.S.A. К. Е. HOAGLAND Association of Systematics Collections Washington, DC, U.S.A. B. HUBENDICK Naturhistoriska Museet Göteborg, Sweden S. HUNT Lancashire United Kingdom R. JANSSEN Forschungsinstitut Senckenberg, Frankfurt am Main, Germany В. М. KILBURN Natal Museum Pietermaritzburg, South Africa М. А. KLAPPENBACH Museo Nacional de Historia Natural Montevideo, Uruguay J. KNUDSEN Zoologisk Institut & Museum Kobenhavn, Denmark A. LUCAS Faculté des Sciences Brest, France C. MEIER-BROOK Tropenmedizinisches Institut Túbingen, Germany Н. К. MIENIS Hebrew University of Jerusalem Israel J. E. MORTON The University Auckland, New Zealand J. J. MURRAY, Jr. University of Virginia Charlottesville, U.S.A. R. NATARAJAN Marine Biological Station Porto Novo, India J. OKLAND University of Oslo Norway Т. OKUTANI University of Fisheries Tokyo, Japan W. L. PARAENSE Instituto Oswaldo Cruz, Rio de Janeiro Brazil J. J. PARODIZ Carnegie Museum Pittsburgh, U.S.A. J. P. POINTER Ecole Pratique des Hautes Etudes Perpignan Cedex, France W. Е. PONDER Australian Museum Sydney QUIZE Academia Sinica Qingdao, People's Republic of China D. G. REID The Natural History Museum London, United Kingdom N. W. RUNHAM University College of North Wales Bangor, United Kingdom S. G. SEGERSTRLE Institute of Marine Research Helsinki, Finland А. STANCZYKOWSKA Siedlce, Poland Е. STARMÜHLNER Zoologisches Institut der Universitát Wien, Austria У. |. STAROBOGATOV Zoological Institute St. Petersburg, Russia W. STREIFF Université de Caen France J. STUARDO Universidad de Chile Valparaiso S. TILLIER Muséum National d'Histoire Naturelle Paris, France R. D. TURNER Harvard University Cambridge, Mass., U.S.A. J.A.M. VAN DEN BIGGELAAR University of Utrecht The Netherlands J. А. VAN EEDEN Potchefstroom University South Africa N. H. VERDONK Rijksuniversiteit Utrecht, Netherlands B. R. WILSON Dept. Conservation and Land Management Kallaroo, Western Australia Н. ZEISSLER Leipzig, Germany A. ZILCH Forschungsinstitut Senckenberg Frankfurt am Main, Germany MALACOLOGIA, 1995, 37(1): 1-11 THE LIFE CYCLE, DEMOGRAPHIC ANALYSIS, GROWTH AND SECONDARY PRODUCTION OF THE SNAIL HELICELLA (XEROTHRACIA) PAPPI (SCHÜTT, 1962) (GASTROPODA PULMONATA) IN E. MACEDONIA (GREECE). M. Lazaridou-Dimitriadou Laboratory of Zoology, School of Biology, Faculty of Sciences, Aristotle University of Thessaloniki, 54006 Thessaloniki, Macedonia, Greece ABSTRACT The life cycle, population dynamics, and growth of the pulmonate snail Helicella (Xerothracia) pappi were studied in northern Greece. The spatial distribution of H. рарр! was found to be contagious. Demographic analysis of its population revealed that (a) three cohorts exist throughout the year, (b) increased growth rate is observed during spring, (c) snails attain their maturity 21 months after hatching when their greatest shell diameter reaches 15 mm, and (d) it is an iteroparous species, with egg-laying in autumn. Von Bertalanffy analysis showed that H. рарр! may reach its maximum size (dead shells collected in the field had 25.5 mm maximum shell diameter) in four years. Shell morphology changes when spermatozoa are fully formed in the gonad for the first time. Mortality rate is high just after hatching and also after winter, when tissue degrowth occurs, at which time the snails lose 33% of their biomass. Life expectancy decreases with increasing age. Net reproductive rate (Ro) was 3.025, per capita rate of increase (г) is 0.04, annual production (P) was 5.82 + 0.45 g/(m?-yr) in 1987 and 3.73 + 0.31 g/(m?-yr) in 1988, mean standing crop (В) was 2.89 g/m? in 1987 and 1.81 g/m? т 1988, and annual productivity rate constant (P/B) was 0.2 per year in both 1987 and 1988. Key words: Biology, ecology, growth, production, snail, Helicella pappi, Xerolenta obvia. INTRODUCTION Little is known about the species Helicella (Xerothracia) pappi Schútt, 1962. Wagner (1927) referred population of this species from near our study area in Xanthi, southern Thrace, to Martha filimargo Krynicki, placing it in the genus Helicopsis Fitzinger, 1833. Ur- banski (1960) said that samples from south- ern Thrace of Helicella (Helicopsis) filimargo (Rossmássler) did not differ from samples of Krimea and Odesco. Schútt (1962) described the shell and genital apparatus of samples of Helicella (Xerothracia) pappi from a type lo- cality in Philippi, Kavala, eastern Macedonia; he claimed that this species could not be placed in the genus Helicopsis, as Wagner (1927) had done, because of such shell fea- tures as curvature and the presence of a keel. Hausdorf (1988) claimed that H. рарр! could not be placed in a subgenus of Helicella be- cause it differs in the dart sac and in the nerves coming from the cerebral ganglia. He concluded that it is a subspecies of Xerolenta obvia. In the present study we are using the name Helicella (Xerothracia) pappi following Schütt (1962), because the samples come from the type locality. We have studied the biology and ecology of this species, which 1$ re- stricted to Philippi, eastern Macedonia, in or- der to find out why this species does not have a broader and continuous distribution from Krimea to Greece. METHODS AND MATERIALS Helicella рарр! was studied in the archae- ological site of Philippi, 25 km from Kavala, eastern Macedonia, Greece. Philippi is a limestone area with limited vegetation, т- cluding mosses, lichens, several Taraxacum species, and succulent plants. Several grass species are dominant. Apart from Н. pappi, there are small, sparse populations of Lindol- holmia lens and Helix figulina. Weather con- ditions during the investigation period are shown in Figure 1. Random samples (Lewis 4 Taylor, 1972) were taken every month for three consecutive years between April 1986 and April 1989. No samples were taken in winter (December through February), when the snails are in di- 2 LAZARIDOU-DIMITRIADOU Mean monthly 90 temperature (°С) ==_== Total precipitation 50 (mm) ern 45 A86M J J AS О М ОМ87А M J J AS О NM88A MJ JAS ON DM89A М Months FIG. 1. Ombrothermic curve showing mean monthly temperatures (°С) and total monthly precipitation (mm) at Philippi from April 1986 to Мау 1989 (striated areas represent arid periods of the year). apause and hidden under vegetation. Quadrat sample-size (25 x 25 cm?) was determined by Healy's method (Cancela da Fonseca, 1965). Elliot's (1971) method was used to determine the necessary total number of sampling units (sampling error less than 20%). Sampling was carried out during morning hours in the ab- sence of rain. All snails found in a quadrat were collected, measured and then replaced. The largest diameter of the shell (D) and the peristome diameter (d) were measured with vernier calipers to the nearest 0.1 mm. D was used for size-frequency histograms, using 3-mm class intervals (Cancela da Fonseca, 1965). The cohorts were discriminated using probability paper (Harding, 1949). This method was valid because the modes of the age-classes were separated by at least 2.5 standard deviations (Grant, 1989), except for two cases, one in March 1987 and one in June 1988; although many age classes had less than 50 individuals, the modal values were consistent from month to month. This con- firms that the modes were real and not the result of sampling variation. The same method has been used before for demographic anal- yses of other populations of molluscs (Hughes, 1970; Lévéque, 1972; Daguzan, 1975; Lazaridou-Dimitriadou et Kattoulas, 1991). There were no statistically significant vari- ations in field density over the three years. Life data and rate estimates were based on successive samples over 1987 and 1988. Spatial distribution of the snails in the hab- itat was examined by using Taylor's (1961) power law 6° = ax”, where 6° = variance and x = mean number of snails/0.25 m“. For the study of relative growth, the mor- phometric criteria of shell diameter (D) in re- lation to the peristome diameter (d) were used from all the animals sampled during 1987 (М = 2567). Mayrat's method (1965а, b) was used to compare the growth of D in re- lation to d between immature and mature snails. A logarithmic transformation was ар- plied to the data. Bertalanffy’s (1938) method was used to calculate the theoretical growth curve and life span; this method is widely used (Moreteau, 1987). Life-table and fertility data, as well as an- nual secondary production, were estimated as described for Helix lucorum (Staikou et al., 1988) and Eobania vermiculata (Lazaridou- Dimitriadou 4 Kattoulas, 1991). Annual pro- duction is calculated by the size-frequency method because single cohorts need not be identified. For dry body-weight (Wb) analysis, 100 an- imals comprising five from each size class were individually marked and their greatest shell diameter (D) was measured; they were then dried to constant weight over a period of 36 h in a vacuum at room temperature. Shell organic matter was calculated as the residual weight of dry shell after treatment with 5 m mol/ml HCl, successive washing over a fine filter; the residue was dried in a vacuum. Two size classes were used; one comprised im- BIOLOGY AND ECOLOGY OF HELICELLA PAPPI 3 2 N/2500 ст 30 egglaying 0 A86M J J O М DM87A M JJ ег аут? egglaying О NM88A M JJ N DM89M Months FIG. 2. Density of Helicella (Xerothracia) pappi (number of snails/2500 ст? (mean + SD)) at Philippi from April 1986 to May 1989. Double lines on the x-axis indicate periods during which no samples were taken. mature snails (4 < D < 13 mm) with shell or- ganic matter containing 1.4% of the total shell weight, and the other comprised mature snails (14 < D < 21 mm) with shell organic matter containing 0.19% of the total shell weight. These size classes were chosen as a result of study of the maturation of the geni- talia and gonad. The shell organic matter was then added to the dry body weight of the different size classes and the sum was used (Table 3) in estimating annual secondary pro- duction. RESULTS Aspects of the Biology of Helicella pappi Helicella pappi is an iteroparous species with overlapping generations throughout the year. Rate of growth and reproduction were not constant during the three study years (Figs. 2, 4). The greatest shell diameter of sexually mature snails was 22 mm, and ma- turity was attained 21 months after hatching (Figs. 3, 4) for 98% of the population. The aperture lip started to form 23 to 24 months after hatching. In the third year, the other 2% of the population matured. Egg laying took place at the end of September, October or November (Fig. 2), depending on the prevail- ing climatic conditions, especially precipita- tion (Fig. 1) Measurements of ten egg clutches in the field showed that the mean number of eggs laid was 69 + 4.3 (т + SE), with a range of 42 to 85 eggs, and the mean weight of 70 eggs was 0.03 + 0.004 g. The mean number of eggs laid by older adults, which had already laid eggs in the previous year, was 22 + 1.9 (m + SE) (N = 5), with a range of 18 to 28 eggs. Hatching took place 25 days to one month later, but hatchlings remained in the soil. The greatest shell diam- eter of the newly hatched snails was 1.5 + 0.09 mm. By marking ten different egg clutches, hatching and after-hatching losses were estimated at 55%. After-hatching losses were estimated in the laboratory to be 20%. During winter, from the end of Novem- ber to the end of February, the snails did not really hibernate. Juveniles formed a thin, transparent epiphragm, whereas 30% of the snails with 14 > D > 7 formed a thick epi- phragm. Snails with D > 14 mm did not form an epiphragm, but they diapaused under creeping plants occurring on the soil or stones. No real aestivation took place. During July and August, all snails diapaused during 4 LAZARIDOU-DIMITRIADOU TABLE 1. Estimation of statistical parameters of the population of Helicella pappi (where a, b = constants, г = correlation coefficient, N = number of snails examined, logd + 6,4 and logD + O\go = Means of the greatest shell diameter (mm)(D) and the peristome diameter (mm)(d) + SD)). Entire sample ato, 1.189 + 0.005 DEBIGE 0.302 + 0.003 e 0.954 1099 + Goga 0.603 + 0.153 logD + 6,40 1.019 + 0.182 М 2569 dry weather, as in summer 1986 (Fig. 1), but only the juveniles formed a thin, transparent epiphragm. Snails were active during spring and on humid weather in summer (Fig. 1). Helicella pappi became mature and the genitalia became well formed when D > 14 mm. Histologically, the gonad of these snails showed fully formed spermatozoa during Oc- tober (the reproductive period), although oocytes were not fully grown. There was a positive correlation (r? = 0.918, М = 100, P< 0.001) between (а) the greatest shell diameter (D) and the corresponding dry body weight and (b) the dry body weight and the dry shell weight (r° = 0.742, N = 100, Р < 0.001). Population Dynamics and Spatial Distribution The population fluctuated during the study period (Fig. 2). The mean population density was 18.1 + 3.3 snails/0.25 m° (mean + SE) in 1986, 14.4 + 6.6 in 1987, and 12.7 + 1.5 in 1988. An ANOVA test among the three con- secutive years did not show any statistical differences (F = 1.646, P = 0.2077) in the population densities. Densities of H. рарр! peaked in early spring. The population den- sity was above the mean density for 5-6 months (Fig. 2). The spatial distribution of H. рарр! was found to be contagious because parameter b of Taylor's power law was equal fo) 21043 [© ОТ Demographic Analysis of the Population of Helicella рарр! The analysis of size frequency histograms (Fig. 3, 4) with probability paper showed that: (a) three cohorts existed in the habitat throughout the year; a fourth was added after the reproductive period, though the third co- hort contained adults of different ages (Fig. Juveniles 2) o o | 25/3/87 1 7 14 22 1 — © = m = œ 8/5/86 = № ao (=) o o о = Nm o o 20 10 0 24/10/87 pe 31/8/88 10 0 1 7 14 1 10 2 22 0 S 30/4/87 ” 27/11/87 E 29/9/88 20 8 6 10 4 2 0 0 1 9 19 1 6 12 19 9 18 1 10 20 30 12 40 20 811/88 6/6/86 т Re 20/3/88 30 20 8 6 20 10 10 4 io 2 0 0 0 0 16 12 19 - = 1 9 18 1 10 20 207 2/7/86 12 30/6/87 30 - 30/4/88 20 8/12/88 10 8 20 10 6 10 4 10 2 0 0 0 0 1 7 14 21 1 10 20 1 9 17 1 9 19 20 15 20 20 3/10/86 31/7/87 25/5/88 31/3/89 10 10 10 10 5 0 0 0 0 1 10 20 1 7 14 21 1 9 18 : 9 19 204 15/11/86 12 20 12 10 29/8/87 30/6/88 10 31/5/89 8 8 10 6 10 6 4 4 2 2 0 0 0 0 1 10 20 1 7 14 21 1 9 19 1 9 19 207 41286 = 30/9/87 1 26/7/88 8 10 10 6 4 5 2 0 0 0 1 10 20 1 7 14 21 1 9 19 Dmm FIG. 3. Size-frequency histograms of the population of Helicella pappi at Philippi from April 1986 to May 1989. 28.8[1—е %-0*8+0,313]. According to this method, the greatest shell diameter mea- sured in the field (inferred from dead shells (D = 25.6 mm)) may be attained in 48 months. The above equation was calculated by using the known greatest shell diameter in the field (26 mm), the mean D value of newly hatched snails, which was 1.2 mm, and the mean size of young snails first sampled in the field, which was 1.5 mm (Fig. 3: November 1987). 6 LAZARIDOU-DIMITRIADOU АЗ © М D MESA М т Аз SOsM Ти $ О ММ: 8 А МО Л ВА $ Months FIG. 4. Мода! distribution of Helicella рарр! at Philippi from April 1986 to May 1989. Broken lines indicate periods during which no samples were taken. G,-G, indicate the different generations during the study period. Time breaks denote winter time, during which diapause took place and no samples were taken. Dt is the shell diameter at time t; D,,,., is the diameter at the upper growth asymptote cal- culated according to Ford-Walford equation (Walford, 1946); t is time in months, t, is the hypothetical time when D is equal to zero (mi- nus the egg stage for this paper), and k is the growth rate coefficient. Life and Fertility Table For the construction of the fertility table, we used (a) the numbers of eggs laid by two or three-year old adults, and by adults of more than three years old, known from field observations and (b) egg-hatching and after- hatching losses calculated in the field and the laboratory respectively. From Table 2, the following may be con- cluded: (a) mortality rate (Kx) was high after hatching and then stabilized. It increased af- ter the first winter, before the second winter (in 3-year-old adults snails) and stayed high after the third winter, (b) values of expecta- tion of life (ex) decreased with increasing age, (c) the value of net reproductive rate (Ro) was high (Ro = 3.024), and (d) the per capita rate of increase was greater than zero, with a rate of 0.04 per unit of time. Annual Secondary Production The calculations for Hynes’ size-frequency method are listed in Table 3. The mean bio- mass of each size class was expressed as dry-weight of body plus organic material of the shell. After conversion, using Benke's (1979) correction, annual production (P) was found to be 5.82 + 0.45 g/(m?-yr) in 1987 and 3.73 + 0.31 g/(m*-yr) in 1988. Biomass (В) was 2.89 g/m” т 1987 and 1.81g/m° in 1988. The annual productivity rate constant (P/B) was 0.21 in both 1987 and in 1988. Turnover time (B/P x 365 days) was 1765 days in 1987 and 1775 days in 1988. DISCUSSION The contagious spatial distribution of H. рарр! is similar to the xerothermophilic spe- cies living in similar habitats in Greece (e.g., Xeropicta arenosa and Cernuella virgata (Laz- aridou-Dimitriadou 8 Kattoulas, 1985). Recruitment of newly hatched snails of H. pappi seems to be the main reason for the rise in population density after winter. The low values of population density in November and December are due to the fact that some BIOLOGY AND ECOLOGY OF HELICELLA РАРР! TABLE 2. Life and fertility table of a cohort of Helicella (Xerothracia) pappi starting in April 1986 (Figure 3-G:). Age (months) ly Q, К, e, m, Em; 1 1000 0.27 0.31 12.29 0.00 0.00 2 730 0.00 0.00 15.65 0.00 0.00 3 730 0.00 0.00 14.65 0.00 0.00 4 730 0.00 0.00 13.65 0.00 0.00 5 730 0.18 0.20 12.65 0.00 0.00 6 600 0.00 0.00 14.28 0.00 0.00 ih 600 0.06 0.06 13.28 0.00 0.00 8 565 0.00 0.00 13.07 0.00 0.00 9 565 0.00 0.00 12.07 0.00 0.00 10 565 0.01 0.01 11.07 0.00 0.00 11 560 0.00 0.00 10.14 0.00 0.00 12 560 0.39 0.47 9.14 0.00 0.00 13 350 0.00 0.00 13.32 0.00 0.00 14 350 0.00 0.00 12:32 0.00 0.00 15 350 0.00 0.00 11.32 0.00 0.00 16 350 0.00 0.00 10.32 0.00 0.00 17 350 0.00 0.00 9.32 0.00 0.00 18 350 0.00 0.00 8.32 0.00 0.00 19 350 0.01 0.01 7.32 0.00 0.00 20 347 0.63 0.98 6.39 0.00 0.00 2] 130 0.00 0.00 15.25 0.00 0.00 22 130 0.00 0.00 14.25 0.00 0.00 23 130 0.00 0.00 13.25 18.00 53.75 24 130 0.14 0.15 12.25 0.00 0.00 25 112 0.00 0.00 13.09 0.00 0.00 26 112 0.00 0.00 12.09 0.00 0.00 27 112 0.00 0.00 11.09 0.00 0.00 28 112 0.03 0.03 10.08 0.00 0.00 29 110 0.00 0.00 9.40 0.00 0.00 30 110 0.00 0.00 8.40 0.00 0.00 31 110 0.00 0.00 7.40 0.00 0.00 32 110 0.10 0.10 6.40 0.00 0.00 33 101 0.00 0.00 6.03 0.00 0.00 34 101 0.00 0.00 5.03 0.00 0.00 35 101 0.07 0.07 4.03 7.00 24.07 36 90 0.15 0.17 3.30 0.00 0.00 37. 77 0.27 0.32 2.81 0.00 0.00 38 56 0.19 0.21 2.67 0.00 0.00 39 46 0.38 0.49 2.18 0.00 0.00 40 28 0.38 0.47 2:22 0.00 0.00 41 18 0.40 0.51 2.25 0.00 0.00 42 10 0.33 0.41 2.43 0.00 0.00 43 7 0.00 0.00 2.40 0.00 0.00 44 7 0.50 0.69 1.40 0.00 0.00 45 4 0.50 0.69 1.30 0.00 0.00 46 2 0.60 0.92 1.10 0.00 0.00 47 1 0.50 0.69 1.00 0.00 0.00 48 0 1.00 0.00 0.50 0.00 0.00 Во = УЁ т, = 3.025 г = InRo / Тс = 0.04 |, : Number of animals surviving at the beginning of age-class x (months) out of 1,000 originally hatched. а, : Mortality rate during age interval x (d,/l,, where d, is the number of animals during age interval x). K, : Intensity or rate of mortality: loga, — loga e, : Expectation of life: T,/l, where T, =L, + Ly; ...... L, (L, : is the number of animals alive between age x and х+1: (I, + 1, 1)/2; L, is the total number of animals x age units beyond age x). L,m, : Total number of hatchlings in each age interval (m, : Number of living animals hatched per adult snail). where Ro is net reproductive rate, г is per capita rate of increase, and Tc is generation time (25,7 months). х— 1 8 LAZARIDOU-DIMITRIADOU TABLE 3. Calculation of production of Helicella pappi by the size-frequency method. Annual production based on sets of samples from April 1988 to April 1989 (where п, = number of snails at the size class j in number; Un, = variance of п; W, = mean individual dry body weight + mean dry shell of organic matter (in mg); G, = geometric mean of weight of pairs of successive size classes; В = mean standing crop or population biomass in mg; P = annual production in mg; P/B = annual turnover ratio; a = number of size classes; CPI = cohort production interval: 730 days). Class п, W, (mg) range п, /0.25т? Un; Di + st. error 1-2 0.01 0.0000 —0.13 0.410 + 0.01 2-3 0.15 0.0020 —0.35 0.410 + 0.02 3— 0.50 0.0554 —0.22 1.000 + 0.02 4-5 0.72 0.1764 —0.05 1.000 + 0.03 5-6 0.77 0.0806 —0.25 2.000 + 0.14 6-7 1.02 0.0960 —0.46 5.000 + 0.27 7-8 1.48 0.2454 012 7.000 + 1.00 8-9 1.36 0.2364 0.37 8.000 + 0.44 9-10 0.99 0.0892 0.04 11.000 + 1.00 10-11 0.95 0.0828 0.39 15.000 + 1.00 11-12 0.56 0.0493 —0.06 22.000 + 3.00 12-13 0.61 0.0436 0.07 26.000 + 2.00 13-14 0.54 0.0483 —0.15 27.000 + 4.00 14-15 0.70 0.0616 —0.23 36.000 + 2.00 15-16 0.93 0.0980 0.10 49.000 + 2.00 16-17 0.83 0.0677 —0.07 51.000 + 2.00 17-18 0.90 0.0736 0.37 65.000 + 4.00 18-19 0.53 0.0427 0.37 71.000 + 5.00 19-20 0.16 0.0058 0.15 85.000 + 5.50 20-21 0.01 0.0003 0.01 134.000 + 6.00 13.73 x 4 = 54.92/m? (259 days) 77.39/m* (365 days) (B) Р’ AC, [njWj] (пп. 1)(@,) (WW.,,)°° (mg/0.25 m°) (mg/0.25 m°) 0.41 0.0051 —0.0549 0.64 0.0600 —0.2258 1.00 0.4990 = 0.2233 1.41 0.7223 —0.0738 3.16 1.5489 —0.7781 5.92 5.1025 —2.7369 7.48 10.3819 0.9115 9.38 10.8906 3.5145 12.85 10.8534 0.5313 181 14.1796 7.0363 23.92 12.2754 —1.3548 26.50 15.9802 1.9403 31.18 14.6175 —4.7975 42.00 25.0296 —9.7470 49.99 45.4396 4.9246 57.58 42.2702 —4.2654 67.93 58.6892 25.1060 77.69 37.8676 28.9068 106.72 13.7058 16.0344 134.00 1.4744 1.4744 321.593 x 4 = 66.1227 x 4 = 1286.36 mg/m?” (259d) 264.49 mg/m? (259 а) 1812.82 (365 а) 372.74 (365 d) P = 20 x 365/730 x 372.74 = 3727.4 mg/(m?.yr) or ог 3,727 g/(m?.yr) U(P) = Uñ(G,_G,_,) x a? x (365/730) = 24371.67 ¿UPS = (22371.67)? = 312.22 = 0:31 P = 3.73 + 0.31 g/(m°.yr) Р/В = 372.74/1812.82 = 0.206 Turnover time = B/P x 365 = 1775.2 days snails are already dormant because of the prevailing weather conditions and the fact that old adults and some new adults (D > 15 mm) die after egg laying. The population dy- namics suggests that there is a characteristic annual periodicity, with synchronization of population and life-cycle development. This appears to be a species adapted to recover slowly after an adverse period. The slow re- covery of the population results from the low hatch-rate of eggs deposited just before the adverse period. The low rate of juvenile de- velopment into adult stage during spring, and the rapid decline in population size after the density peak, indicate a reduced reproduc- tive effort of later adult stages. Increased growth took place during spring because temperatures were not exceedingly high (20°C), and total monthly precipitation did not fall below 20-30 mm (Fig. 1). During autumn, growth took place only in juveniles (Fig. 4) when temperatures were around 15°С (in October and November) and only if there was precipitation. Newly hatched snails dur- ing that period of the year remained dormant. The rate of growth, however, was not the same for newly hatched snails and juveniles (Fig. 4). This is also related to differences in temperature [e.g., the 1987 March (5.8°) and April (12.5%) temperatures were lower than in 1986 (8.6 — 15.4”C respectively)]. Addition- ally, the growth rate was not the same for juveniles and mature snails. This was evident from the study of the population analysis of H. pappi and from the comparison of the rate of relative growth of D in relation to d be- tween juveniles and adults. This is a general phenomenon in many Helicidae and it is usu- BIOLOGY AND ECOLOGY OF HELICELLA PAPPI 9 ally due to internal changes in genitalia and gonad maturation (Yom-Tov, 1971; Bonavita, 1972; Williamson, 1976; Lazaridou-Dimitria- dou, 1986; Staikou et al., 1988). Seasonal variation in growth and, more specifically, in- creased growth rate in spring, have been re- ported for other snails in Greece, including Е. vermiculata (Lazaridou-Dimitriadou & Kattou- las, 1985), X. arenosa and C. virgata (Lazari- dou-Dimitriadou, 1986), H. lucorum (Staikou et al., 1988), B. fruticum (Staikou et al., 1990), and М. cartusiana (Staikou & Lazaridou-Dim- itriadou, 1990). Baba (1985) also reports that growth is related to climatic factors and that it increases only before sexual maturity. As for H. lucorum (Staikou et al., 1988), B. fruti- cum (Staikou et al., 1990) and other terrestrial snails, H. pappi continues to increase D even after maturation, because it is heterothermic. There were adverse periods during which ac- tivity, and consequently growth and repro- duction, stopped. These periods coincided mainly with winter and summer drought. Moreover, the values of k showed that winter constitutes the most important environmen- tal stress. Helicella pappi matures only in the second year of its life, with a rate of increase equal to 0.041. Being iteroparous, this snail reproduces again after its first egglaying. However, it seems that high reproductive output on one occasion influences future re- productive output. Although Ro and the turn- over time was high (1,775 days), productivity rate (0.2/year) was low, which may be related to the long life span (4 years) and high mor- tality of this species at subadult and adult stages, especially after winter. Moreover, maintenance costs during dormant periods are increased; 33% of the previously ac- quired biomass 1$ lost. These snails try to ex- ploit favourable conditions, but the absence of a true hibernation or aestivation and a var- ied capacity for dormancy cause the death of much of the population. The annual adult mortality rate was similar to that reported by Osterhoff (1977) and Williamson et al. (1977) for С. nemoralis and by Shachak et al. (1975) for Sphincterochila zonata, which also has a life-span of 4-6 years. Turnover times are ob- viously related to length of life; long turnover times have only been reported for such bi- valves as Anodonta (1,789 days/4.9 years) (Russell-Hunter & Buckley, 1983) and for the terrestrial snail Monacha (1,177 days/2-3 years) (Staikou 8 Lazaridou-Dimitriadou, 1990). Otherwise, terrestrial snails seem to have short turnover times from 50.7 days (Vallonia, Russell-Hunter 8 Buckley, 1983) to 293 days/3 years (Helix lucorum, Staikou et al., 1988). However, the possibility of long turnover times in age-structured populations, especially in relation to interspecific compar- isons, must be treated with caution, because some average standing-crop values in the lit- erature are estimated from an entire popula- tion and others from a model cohort. Gener- ally, turnover times of more than two years appear to be associated with life spans of four years or more. The fact that H. pappi has long turnover times and a low productivity rate may be related to its small size, its long life span, and the fact that it is iteroparous. Published distribution data seem to sug- gest that H. pappi comprises a species with a patchy distribution (Schútt, 1962). Because of its evolutionary origin in colder climates (Ur- banski, 1960), this species manages barely to survive in this region of Greece, with wide daily and seasonal temperature variations. Actually, it is the climatic factors that play an important role in controlling energy flux in H. рарр!. Low rates and efficiencies of growth place severe restraints on the snail's ability to meet the demands of over-winter mainte- nance and reproduction. As a result, repro- duction is delayed until the second autumn. Energy investment is not concentrated on egg production, and there 1$ variation in rates of growth and fecundity according to age. Mor- tality takes place at all stages, but mainly after winter; high mortality after adverse periods of the year is a common characteristic of many helicid snails. Consequently, the demograph- ic characteristics of Helicella conform to an A-selectionist's strategy, as defined by Green- slade (1983); that is, suitable conditions for breeding last for only a short period but occur regularly and predictably, such that the pop- ulation synchronizes with those conditions (Figs. 1, 2). Moreover, interspecific competi- tion is rare, because only very small, sparse populations of Lindolhomia lens and Helix figulina were observed. ACKNOWLEDGEMENTS | would like to thank Dr. E. Gittenberger of The Natural Museum of Leiden for informa- tion on the distribution and systematic posi- tion of Helicella (Xerothracia) pappi. Thanks are also due to Dr. С. В. J. Dussart from Christ Church College, Canterbury, U.K, for his critical remarks, K. Asmi and Dr. A. 10 LAZARIDOU-DIMITRIADOU Staikou for their technical help, and Dr. T. Sofianidou for providing me with samples from the area of Philippi before this study was undertaken. LITERATURE CITED BABA, K., 1985, Investigation of the growth rate of two terrestrial snails: Bradybaena fruticum (О. Е. Müller) and Euomphalia strigella (Draparnaud), Pulmonata. Soosiana, 13: 79-88. BENKE, A. C., 1979, A modification of the Hynes method for estimating secondary production with particular significance for multivoltine pop- ulations. Limnology and Oceanography, 24: 168-171. BERTALANFFY, L. VON, 1938, A quantitative the- ory of organic growth. Human Biology, 10: 181- 213. BONAVITA, D., 1972, Eléments pour une monog- raphie écologique et biologique du gastéropode terrestre Leucochroa candidissima (Drap.). Thèse Doctorat des Sciences Naturels. Univer- sité de Provence, Marseille. CANCELA DA FONCECA, J., 1965, L'outil statis- tique en biologie du sol. |. Distributions de Fré- quences et Tests de Signification. Revue d’Ecolo- gie et de Biologie du Sol, 11 (3): 299-332. DAGUZAN, J., 1975, Contribution à l'écologie des Littorinidae (Mollusques, gastéropodes, proso- branches); recherches écophysiologiques chez quatre espèces: Littorina neritoides (L.), L. saxa- tilis (O.), L. littorea (L.) et L. littoralis (L.). These Doct. d'Etat, Université de Rennes |, 35042 Rennes. ELLIOTT, J. M., 1971, Some methods for the sta- tistical analysis of samples of benthic inverte- brates.—(Ph.D.). Freshwater Biological Associa- tion, 25: 1-144. GRANT, A., 1989, The use of graphical methods to estimate demographic parameters. Journal of the Marine Biological Association of the United Kingdom, 69: 367-371. GREENSLADE, Р. J. М. 1983. Adversity selection and the habitat templet. The American Naturalist, 122 (3): 352-365. HARDING, J. P., 1949, The use of probability paper for the graphical analysis of polymodal fre- quency distribution. Journal of the Marine Bio- logical Association of the United Kingdom, 28: 141-153. HAUSDORF, B., 1988, Zur Kenntnis der system- atischen Beziehungen einiger Taxa der Helicelli- nae lhering 1909. Archiv für Molluskenkunde, 119: 9-37. HUGHES, R. N., 1970, Population dynamics of the bivalve Scrobicularia plana (Da Costa) on an in- tertidal mud-flat in north Wales. Journal of Ani- mal Ecology, 39: 333-356. LAZARIDOU-DIMITRIADOU, М. & М. KATTOU- LAS, 1985, Contribution a l'étude biologique et écologique d'Eobania vermiculata (Gastéro- pode, Helicidae). Biologica Gallohellénica, X: 132-137. LAZARIDOU-DIMITRIADOU, M., 1986, Contribu- tion a l’etude de la croissance des escargots Cernuella virgata (Da Costa) et Xeropicta arenosa (Ziegler) (Gasteropodes, Pulmonés, Stylom- matophores) vivant sur les microdunes de Potidea (Chalkidiki, Grece). Bulletin de la Société Zoologique de France, 111: 89-97. LAZARIDOU-DIMITRIADOU, М. & М. KATTOU- LAS, 1991, Energy flux т a natural population of the land snail Eobania vermiculata (Müller) (Gas- tropoda: Pulmonata: Stylommatophora) in Greece. Canadian Journal of Zoology, 69: 881- 891. LEVEQUE, C., 1972, Mollusques benthiques du Lac Tchad: écologie, production et bilans éner- gétiques. These doctorat des Sciences Naturels Paris, VI, C.N.R.S. LEWIS, T. & L. R. TAYLOR, 1972, Introduction to experimental ecology. Academic Press, Inc. (London). MAYRAT, A., 1965a, Calcul et comparaison des droites d’ allométrie de croissance. Comptes Rendus de la Société de la Biologie, 159: 2139- 2143. MAYRAT, A., 1965b, La loi а’ allométrie et des al- lometries. Comptes Rendus de la Société de la Biologie, 159: 544-548. MORETEAU, J. 1987. Quelques modèles de crois- sance appliqués a des Mollusques. Haliotis, 16: 439-451. ООЗТЕВНОЕЕ, С. M., 1977, Variation in growth rate as an ecological factor in the landsnail Ce- paea nemoralis (L.). Netherlands Journal of Zo- ology, 27: 1-132 RUSSELL-HUNTER, W. D. & D. E. BUCKLEY, 1983. Actuarial Bioenergetics of nonmarine mol- luscan productivity. Pp. 463-503 in The Mol- lusca, Vol. 6 Ecology, ed. by K. M. WILBUR 8 W. D. RUSSELL-HUNTER. Academic Press, New York. SHACHAK, M., Y. ORY & Y. STEINBERG, 1975. Field observations on the natural history of Sphincterochila (S.) zonata (Bourgignat, 1953) (= S. boissieri Charpentier, 1847). Argamon, Israel Journal of Malacology, 5: 20-46. SCHUTT, H., 1962, Eine gekielte Helicella aus Thrazien. Archiv fúr Molluskenkunde, 91: 151- 156. STAIKOU, A. & М. LAZARIDOU-DIMITRIADOU, 1990, Aspects of the life cycle, population dy- namics, growth and secondary production of the snail Monacha cartusiana (Müller, 1774) (Gas- {города Pulmonata) in Greece. Malacologia, 31: 353-362. STAIKOU, A., М. LAZARIDOU-DIMITRIADOU & М. FARMAKIS, 1988, Aspects of the life cycle, pop- ulation dynamics, growth and secondary pro- duction of the edible snail Helix lucorum, Lin- naeus 1758 (Gastropoda Pulmonata) in Greece. Journal of Molluscan Studies, 54: 139-155. STAIKOU, A., М. LAZARIDOU-DIMITRIADOU & E. BIOLOGY AND ECOLOGY OF HELICELLA PAPPI 11 PANA, 1990, The life cycle, population dynam- ics, growth and secondary production of the snail Bradybaena fruticum (Múller, 1774) (Gas- tropoda Pulmonata) in Northern Greece. The Journal of Molluscan Studies, 56: 137-146. TAYLOR, L. R., 1961, Aggregation, variance and the mean. Nature (London), 189: 732-735. URBANSKI, J., 1960, Ein Beitrag zur Kenntnis der Molluskenfauna der Insel Thasos und des maze- donisch-thrazischen Küstenlandes. /svestija na Soologiceskija Institut, 9: 71-105, taf. 1-4. WAGNER, А. J., 1927, Studien zur Molluskenfauna der Balkanhalbinsel mit besonderer Berücksich- tigung Bulgariens und Thraciens, nebst mono- graphischer Bearbeitung einzelner Gruppen. Ann. Zool. Mus. Polon. Hist. Nat., 6: 263-399, taf. 10-23. WALFORD, L. A., 1946, A new graphic method of describing the growth of animals. Biological Bul- letin (Woods Hole), 90: 141-147. WILLIAMSON, P., 1976, Size-weight relationship and field growth rates of the land snail Cepaea nemoralis L. Journal of Animal Ecology, 45: 875- 885. WILLIAMSON, P., В. А. D. CAMERON & М. A. CARTER, 1977, Population dynamics of the landsnail Cepaea nemoralis L.: a six year study. Journal of Animal Ecology, 46: 181-194. YOM-TOV, Y., 1971, The biology of two desert snails Trochoidea (Xerocrass) sectzeni and Sphincterochila boisseiri. Israel Journal of Zool- оду 20: 231-248. Revised ms. accepted 28 November 1994 Be, >| i | | o > = 2. 8 ME = ny Shane ae ecu 414 EN ibe “abr x my pe LUS | de к” | MALACOLOGIA, 1995, 37(1): 13-21 EGG-LAYING AND ASSOCIATED BEHAVIOURAL RESPONSES OF LYMNAEA PEREGRA (MULLER) AND LYMNAEA STAGNALIS (L.) TO CALCIUM IN THEIR ENVIRONMENT Hilary Piggott & Georges Dussart Canterbury Christ Church College, Canterbury, Kent CT1 1QU, United Kingdom ABSTRACT In laboratory trials, specimens of the freshwater pulmonate snail Lymnaea peregra from Ullswater, a soft-water lake in the English Lake District (6.5 mg/l [Са?*], pH 7.1), showed a significant preference for laying eggs on clean, dead conspecific snail shells (y? = 38.91, Р < 0.001 n = 20), thereby confirming anecdotal field observations of this behaviour. A choice chamber was used to investigate to test the hypothesis that Iymnaeid snails might be able to use calcium as a cue for orientation, for example as a stimulus to find target shells for oviposition. Lymnaea stagnalis showed a more positive response to calcium than did L. peregra (t = 4.2, Р < 0.05, п = 137). When snails from soft-water environments were reared in soft water in the laboratory, specimens of L. peregra showed a strong preference for calcium (x? = 19.6, Р < 0.001, п = 202), but specimens reared in hard water (84 mg/l [Ca?*], п = 44)) showed no such preference. The hypothesis that snails could use a calcium cue to select a shell as an oviposition site was supported, and, in addition, the breadth of the chemical niche of L. peregra appears to be wider than that of L. stagnalis. Key words: egg-laying, calcium, preference, ecology, hardness, Lymnaea peregra, Lymnaea stagnalis. INTRODUCTION The importance of calcium in the distribu- tion of freshwater molluscs has been widely reported (Boycott, 1936; Macan, 1950; Okland, 1969; Williams, 1970; McKillop 8 Harrison, 1972). In his qualitative study of the ecology of freshwater molluscs in Britain, Boycott suggested that whereas some spe- cies were restricted to waters with calcium concentration exceeding 20mg/l, other spe- cies were more tolerant and could occur both in high calcium waters and elsewhere in low calcium waters, L. peregra being a typical ex- ample of such tolerance. Boycott (1936) also highlighted the difficulty of separating water chemistry from geographical distribution and physical characters of the habitats. In quan- titative studies of molluscan ecology in rela- tion to water chemistry, Dussart (1976) showed that general mollusc abundance was greater in hard waters ([Ca*] > 40 mg/l), al- though medium waters ([Са?*] 5-40 mg/l) had greater species diversity. As well as af- fecting distribution, calcium concentration also affects such aspects of freshwater mol- lusc biology as shell composition in planorbid snails (Madsen, 1987), freshwater sphaeriid clams (Burky et al., 1979), and the ampullariid 13 snail Marisa cornuarietis (L.) (Meier-Brook, 1978). Other evidence of a direct metabolic response to environmental calcium was demonstrated by Dussart & Kay (1980), who showed that L. peregra reared in waters of different hardness had different respiration rates. Dussart (1979) observed anecdotally that in some soft-water habitats, L. peregra ap- peared to show a preference for laying eggs on the shells of other L. peregra from the same generation, and it was proposed that this behaviour could possibly *'... provide an immediate source of nutrients to the off- spring after the post-egg laying deaths of the parent population.” These observations therefore raised questions about whether snails could behaviourally orientate to envi- ronmental calcium. Thomas (1982) observed that taxes of freshwater molluscs along inor- ganic ion gradients had been little studied. There is precedent for suggesting that fresh- water molluscs exhibit chemoreception, because, for example, the freshwater pulmo- nate Planobarius corneus (L.) can discrimi- nate between amino acids (Lombardo et al., 1991), and Thomas et al. (1980) observed species-specific responses to four amino acids. 14 PIGGOTT & DUSSART Lymnaea peregra 1$ probably the most common freshwater snail in Europe (Fitter 8 Manuel, 1986); it occurs in all types of habi- tats, including water with a calcium concen- tration as low as 1 mg/l. By contrast, L. stag- nalis is a calciphile species comprising larger individuals, often sympatric with L. peregra in harder waters. The objectives of the following experiments were therefore, firstly to investi- gate the hypothesis that soft-water snails might lay eggs preferentially on the shells of other individuals, secondly to investigate whether L. peregra and L. stagnalis might ori- entate to calcium in solution and thirdly, to identify any species-specific differences in such behaviour. MATERIALS AND METHODS Sources and Maintenance Conditions of Snails All samples were taken in June and July 1992, when L. peregra were obtained from the River Stour in Kent (OS ref. TR1785990, 84 mg/l [Са?*], pH 7.4) and L. stagnalis were collected from Monkton Nature Reserve (OS ref. TR657295, 72mg/l [Са?*], pH 7.5). Spec- imens of L. peregra were also collected from Ullswater, a soft-water lake in the Lake Dis- trict (OS ref. NY421205, 6.5 mg/l [Ca**], pH 7.1). In all the choice experiments, snails were young-mature specimens, of about 8-11 mm maximum length for L. peregra and 25-35 mm maximum length for L. stagnalis; because some of the snails came from the field, age could only be estimated. Other- wise, snails were four-six months old. Snails were housed in the laboratory in plastic tubs measuring 16 x 16 x 16 cm, each tub containing one litre of water and ten snails. Where snails from hard water were being cultured in hard water, water from the River Stour was used. The snails were main- tained at room temperature, with continuous aereation, in natural daylight and fed fresh lettuce every three days; ten millilitres of fil- tered pond water were added to each tub to provide additional micro-nutrients. A small quantity of washed, fine sand was provided in each tub to aid digestion, and the water was changed at three-day intervals. In the experiment to investigate substratum choice, two round, flint pebbles (10 mm diameter, the approximate size of an L. peregra shell), and two empty L. peregra shells were added to each tub. Each tub thus comprised a micro- cosm for which both the number of egg cap- sules and the area of the surface types upon which eggs were laid was recorded, includ- ing the submerged walls of the tub. Review of Choice-Chamber Designs for Aquatic Snails Various experimental designs have been employed to examine chemoreception in aquatic molluscs. For example, Uhazy et al. (1978) investigated chemicals attractant to Biomphalaria glabrata (Say) by using a grid of 10 x 5 units marked on a white enamel dish with test material at one end of the dish and control material at the other; Madsen (1992) used a similar design in food location exper- iments for Helisoma duryi (Wetherby) and Bi- omphalaria camerunensis (Boettger). Lom- bardo et al. (1990) used a ‘Y’ shaped maze to investigate Planorbarius corneus, whereas Thomas et al. (1980) used an olefactometer to investigate the response of B. glabrata to amino acids and related compounas. There are however, many problems asso- ciated with the design of such experiments; firstly, diffusion causes dynamic change in ionic concentrations, so that snails might be responding to an ionic flow, as opposed to absolute concentrations. Secondly, the ani- mals might become satiated as they move up the concentration gradient so that their be- haviour changes accordingly. Thirdly, there are problems of deciding when a choice has been made; for example, Lombardo et al. (1991) reported that a time limit was neces- sary due to the slowness and sometimes in- activity of the snails in their experiments. Fourthly, the distinction between olfaction and gustation is not clear in aquatic molluscs (Kohn, 1961). Fifthly, when a liquid medium rather than a solid object is the object of the choice, there are problems of experimental design because seiche phenomena can be entrained at the start of each trial. Given the nature of the animal, it is unlikely that any aquatic choice-chamber design will be wholly satisfactory for aquatic snails. To circumvent these problems, Vareille- Morel (1986) used a flowing water design in studies of the response of Potamopyrgus jenkinsi Smith to a nutrient source, and Dus- sart (1973) used a static chamber with a “starting box”; test specimens introduced into the starting box had an instant choice of waters when the starting gate was removed. RESPONSE OF LYMNAEA TO ENVIRONMENTAL CALCIUM 15 Central baffle Water flow os FIG. 1. Choice chamber used in the present study. Design of the Choice-Chamber Because L. stagnalis and L. peregra can be amphibious, the choice-chamber shown in Figure 1 was used, as it circumvents many of the problems described above. Preliminary dye experiments and water samples taken during the trials showed that there was no significant transfer of chemicals across the central divider; the latter was made of wood (ramin) because streaming was found to oc- cur when plastic dividers were used. Character of Water For all choice experiments, solutes were added to distilled de-ionised water to give the following standard water (mg/l): magne- sium, 15; sodium, 30; potassium, 7; ortho- phosphate, 15; calcium, 0; pH 7.5; (pH ad- justment with 2M НС). By contrast, the test hard water included 50 mg/l calcium. Methods of Testing Preferences Testing one species at a time, up to ten snails were placed on the divider, all facing the same way. They could then move to the left or right, or could remain on the divider. To prevent extraneous bias, the location of choice waters were systematically varied be- tween runs. Because Green et al. (1992) had shown significant anticlockwise-left move- ment of L. peregra when out of water, runs were repeated, with all the snails facing in the opposite direction. There were thus four ori- entations in each run and each run was re- peated, so that approximately 80 snails were involved in each trial. Similar numbers of snails were used by Madsen (1992) in food location experiments. After ten minutes, or sooner if all the snails had moved, the snails were removed from the choice chamber and the dividers cleaned; there was no contact between snails and no evidence of trail fol- lowing. y? tests with Yate's corrections were used to see if there was a significant choice and snails which stayed on the central divider were not included in the analysis. For the final investigation, L. peregra reared from eggs produced by Ullswater snails were used. Two batches of adult snails were separately main- tained in soft water (5 mg/l [Са?*]), and two batches were separately maintained in hard water (84 mg/l [Са?*]), both being kept for 1-3 weeks before the trials. RESULTS Oviposition Choice of L. peregra Besides the walls of the tub, the snails had only two pebbles and two shells upon which to lay, so that oviposition on these surfaces needed a significant positive choice. After correction of the data for available area, soft- water snails imported from Ullswater showed a significant bias towards laying eggs on shells, and this bias was greater for snails which, before the trials, had been kept in the laboratory in soft water (y? = 38.91, P < 0.001) compared with those kept in hard wa- ter (X? = 13.49, Р < 0.01). These results imply that L. peregra from soft water showed a preference for laying eggs on snail shells, particularly having been maintained in soft water. It is a statistical requirement of a chi- square test that no observed values should be less than five; because no eggs were laid on the pebbles, this requirement was not met, and the results should therefore be treated with caution; it is conceivable that eggs might never have been laid on a flint pebble. It should be noted that in other batch cultures however, eggs were intermittently laid on such pebbles. Choice Experiments A number of preliminary tests were con- ducted. Ten L. stagnalis were housed in soft water and used in the following preliminary experiments. When given a choice of sodium and calcium cations but with the same con- centrations of chloride anions, the null hy- pothesis could be refuted at P < 0.05, imply- ing that movement was significantly towards the source of calcium (Table 1-Trial (1). To determine whether this apparent bias was 16 PIGGOTT & DUSSART TABLE 1. Preliminary investigation of choice behaviour. For the purposes of this report, a trial is defined as an opportunity to make a choice by a single individual snail. Because some snails stayed on the central baffle for the duration of an experiment, the number of movements 15 frequently less than the number of trials. In experiments (i)-(iv), each of the ten snails was tested four times (i.e. four runs) in the four orientations, giving 160 trials. Preliminary trials—ten mature specimens of L. stagnalis previously housed in soft water for one week Bias Towards.. vo Significance (i) Comparison of cations 100 mg/l [CI ] cf 100 mg/l [CI ] (as NaCl) (as Ca Cl,.2H,0) Movements 81 56 [Cas] 4.2 Р < 0.05 (ii) Sodium against water 100 mg/l [Na*] ai HEC (as NaCl) Movements 73 50 [Na*] 3.9 Р < 0.05 (iii) Calcium against water 100 mg/l [Са?*] cf H,0* (as Ca CI,.2H,0) Movements 89 53 [Gaz] 8.8 Р < 0.01 Osmotic Potential Trial —40 mature specimens of L. stagnalis previously housed in soft water for one week were tested in each of the four orientations, ¡.e. 160 trials. (iv) [Ca?*] cf [Na*] 500 mg/l 370 mg/l (аз Ca CI,.2H,0) (as NaCl) Movements 92 45 prompted by an aversion for sodium, the snails were given a choice between sodium and de-ionised water in Trial (ii). Again, the null hypothesis could be refuted at P < 0.05, implying that the snails chose the sodium re- gime and aversion had not been a factor in Trial (i). In Trial (iii), there was again an apparent bias towards calcium, implying that the null hypothesis could be refuted at P < 0.01; this could represent an aversion to de-ionised water, possibly due to the effects of osmotic potential. Freshwater gastropods have body fluids that are hyper-osmotic to the external media and therefore have to cope with the continual influx of water. To establish whether the difference in osmotic pressure between the test substances was affecting the responses, in Trial (iv), snails were offered a choice between calcium and sodium but at concentrations that would each exert the same osmotic potential. The result provides evidence to refute the null hypothesis that there is no significant difference in the re- sponse of snails to different substances of the same osmotic pressure at P < 0.01; the snails apparently once again showed a sig- nificant bias towards the water containing calcium. Bias Towards.. Ne Significance [Ca?*] 15.4 Р < 0.001 To investigate whether there might be spe- cies-specific differences in these move- ments, 40 specimens each of L. stagnalis from Monkton and L. peregra from the Stour were maintained in hard and soft water for periods of one to three weeks before inves- tigation. The results of the main trials shown in Table 2 allow comparison between their behaviour. Trials (v) and (ix) were controls showing that when exposed to identical test substances, there was no significant bias in choice for either species. For Trials (vi) to (viii), L. stagnalis displayed a highly significant positive response to cal- cium, but in Trials (x) to (хи), L. peregra showed a lower response (Table 2). A Stu- dent's t-test of these data confirmed this dif- ference; across the three trials, the mean number of L. stagnalis choosing calcium was 42.17 with a comparable mean of 32.42 for L. peregra, thereby indicating that L. stagnalis were orientating more strongly than L. pere- gra (t = 4.33). The responses for both species towards calcium differed, depending on the length of time they had been maintained in hard or soft water before the experiment. For example, the y? value for L. stagnalis that had been kept in soft water for one week was 62.3, compared with 19.5 for snails that had RESPONSE OF LYMNAEA TO ENVIRONMENTAL CALCIUM АГ TABLE 2. Results of trials to investigate choices of forty individuals of each of L. stagnalis and L. peregra. Snails were kept in either hard water for one week (H1), hard water for two weeks (H2), soft water for one week (S1) or soft water for three weeks (S3). All calcium was presented as 100 mg/l calcium as [Ca Cl,.2H,0]. Each snail was tested twice in each of the four orientations, i.e. 320 trials. Trial Choice Available Result L. stagnalis Bias Towards.. x? Significance (v) H1 [Ca?*] cf [Са] Movements 140 135 neither 0.08 N.S. (vi) H2 [Са?*] ef H,O* Movements 155 98 [Са || 12.4 Р < 0.001 (vii) $1 [Ga?*] cf 850% Movements 186 61 [Cart] 62.3 P < 0.001 (viii) $3 [Ca?*] cf H,O* Movements 165 93 [Ga?*] 19.5 Р < 0.001 L. регедга (x) H1 [Gas] cf [Са?*] Movements 135 139 neither 0.08 N.S. (x) H2 [Gas] cf H,O* Movements 124 128 [Ca**] 0 N.S. (xi) $1 [Са?*] cf H,O* Movements 132 95 [Са?*] 5.7 Р < 0.05 (xii) $3 [Ca] et 150: Movements 133 69 Са | 19.6 Р < 0.001 DISCUSSION been kept in soft water for three weeks; by contrast, the equivalent y? values for L. per- egra were 5.7 compared with 19.6. It is possible that the snails were orientat- ing to chloride rather than calcium. However, Trial (x) for L. peregra showed no significant preference between chloride and de-ionised water; and, in fact 34 snails compared with 26 snails actually migrated into the de- ionised water. There is always the possibility, however, that this experiment shows a bal- anced preference/aversion for both chloride and de-ionised water. For example, aversive behavioural and physiological responses to salinity are well documented (Perkins, 1974). The animals may have balanced their aver- sion to salinity with an aversion to the os- motic problems posed by deionised water, and consequently made по significant choice. Those L. peregra that had been raised in hard water showed no calcium preference (Table 3). By contrast, L. peregra raised in soft water showed a significant preference (X? = 8.2, Р < 0.01) for calcium. For example, over four runs, the mean number of snails that had been raised in hard water and chose calcium was 6.5, and the mean number of snails that had been raised in soft water and chose calcium was 8.75, the difference be- tween the means being significant at (t = 2:67). The molluscan shell is formed by the dep- osition of calcium carbonate on a protein ma- trix. For snails in a soft water environment, an immediate source of calcium could be ben- eficial to the developing juveniles and so there is an a priori reason for expecting snails to be able to detect and orientate towards calcium. The possible attractant properties of calcium were demonstrated here by the re- sults of Trial (i) on [. stagnalis. lt appeared that sodium chloride did not act as a repel- lant, because movements away from this compound were not significant in Trial (ii); Madsen (1990) found that such snails as H. duryi and Bulinus truncatus Audouin were not adversely affected by low concentrations of sodium chloride. However, there could have been a significant aversion to the de-ionised water in Trial (ii). Young (1975) reported that in soft water, L. peregra extracted 70% of the calcium re- quirement from lettuce and in hard water, ex- tracted only 46% from lettuce. This was compared with L. stagnalis, which, although it efficiently extracted 95% of the calcium content of the lettuce, usually only took 20% of the total requirement from this source, the remaining 80% being derived directly from the water. Thus, the relationships between environmental Базе-юп concentration and 18 PIGGOTT & DUSSART TABLE 3. Results for eight L. peregra reared in hard water and eight reared in soft water. The snails were tested when they had reached an overall shell length of 6-8 mm. Calcium was presented as 100 mg/l [Ca Cl,.2H,0]. Each specimen was tested twice in each of the four orientations, i.e. 64 trials. Choice Available Snails reared in soft water [Са?*] cf Movements 35 Snails reared in hard water [Ca?*] cf Movements 22 snail biology appear to be complicated. For example, ionic ratios might be involved; Har- rison et al. (1966) invoked the ratio of cal- cium/magnesium as a significant factor in egg production. There are also contrary re- sults; though Harrison et al. (1966) found a curvilinear relationship between egg produc- tion and calcium concentration for В+ omphalaria pfeifferi, Thomas et al. (1974) found a positive linear relationship for Bi- omphalaria glabrata. Nevertheless, it ap- peared that in our experiments, the need to respond to a source of environmental cal- cium was less for L. peregra than for L. stag- nalis, possibly because L. peregra obtains a smaller proportion of its calcium require- ments direct from the environment. This proposition 1$ supported by the results shown in Table 2: a gradually increasing re- sponse to a source of calcium by L. peregra, whereas the corresponding response of L. stagnalis was immediately highly significant. Even when kept, albeit temporarily, in hard water (84 mg/l [Са?*]), it seems that the cal- cium requirement of L. stagnalis was not be- ing met in these relatively small containers. Trials (v) to (xii) were designed to show species-specific differences in behaviour and were not particularly designed to distinguish between the effects of calcium, chloride or possible osmotic potential effects of de-ion- ised water. However, the results of trial (x) for L. peregra suggest that chloride and osmotic potential were not playing respectively at- tractive and aversive roles, because a major- ity of snails chose de-ionised water. There is also circumstantial evidence from Trials (i) to (iv) and from the literature to suggest that cal- cium 1$ a significant factor (e.g., Greenaway, 1971a, b). It is probable that ten L. stagnalis housed in one litre of pond water would re- duce the calcium concentration to such a level that a significant response to a source of calcium would be both essential for the H,0* H,0* Result Bias Towards.. x Significance 14 lea] 8.2 P < 0.01 22 neither 0 N.S snail and observable by the experimenter. For example, in our experiments, L. stagnalis showed a massive response to the source of calcium, having been in soft water for only one week. Out of 247 observed movements, 186 were towards the source of calcium; if osmotic potential were the only factor, it might be expected that the response would not have changed between trials. In addition, at this stage, the shells of L. stagnalis were becoming increasingly fragile. Nduku & Har- rison (1976) suggested that snails cultured in low calcium concentration are physiologi- cally stressed and cannot carry out normal metabolic processes. Bielefeld et al. (1993) have implicated alka- line phosphatase in the mantle epithelium as a factor in shell mineralisation, and Green- away (1971a, b) suggested that there was a net movement of calcium from the environ- ment into the blood, excess calcium being deposited as carbonate in the shell. When Greenaway cultivated snails in soft water, there was a loss of calcium from the blood to the external environment. To compensate, a reverse flow of calcium occurred from the shell to the blood. It is therefore likely that in the later stages of our experiments, L. stag- nalis were physiologically stressed. Given that the test snails of both species were of similar sizes, these results suggest that either L. stagnalis has a greater calcium demand than L. peregra, or that L. peregra has a higher threshold for the inception of calcium-mediated stress at lower concentra- tions than L. stagnalis. The implication is therefore that in terms of meeting calcium re- quirements, the niche breadth of L. peregra is wider than that of L. stagnalis. This hypothe- sis 15 supported by Costil-Fleury (1991), who used factor analysis to show that L. peregra is less confined in terms of habit type than L. stagnalis. However, many other physico-chemical RESPONSE OF LYMNAEA TO ENVIRONMENTAL CALCIUM 19 and biotic factors are as important as cal- cium. Temperature, pH, macro-vegetation, suspended solids and the nature of the allochthonous input to the habitat have all been suggested as crucial factors in mollus- can distribution (e.g., Macan, 1974; Okland, 1983; Pip, 1986), and Dussart (1979) showed that potassium, mud substratum-type and rock substratum-type were major variables in the distribution of Bithynia tentaculata (L.), Gyraulus albus (Müller) and Planorbis planor- bis (L.) respectively; magnesium was a major water chemistry variable for L. peregra. The results of the investigation using L. peregra reared in different environmental conditions tentatively support the suggestion of Dussart (1979) that in soft waters, snails might satisfy a metabolic need for calcium by orientating towards, and laying eggs on shells. There is anecdotal evidence that, es- pecially in areas of base-ion deficiency, aquatic snails will aggregate on shells, bones and other calcium sources; it would be useful to compare the calcium responses of snails originating from soft-water environments with snails of the same species from hard- water environments. If raised in identical con- ditions, any significant differences in re- sponses persisting in the F1 and succeeding generations would indicate the existence of physiological races and plasticity. This in turn might suggest that the process of speciation for water type is under way. The existence of a microhabitat at the sub- stratum surface of, say, a shell or a pebble needs to be recognised. Environmental cal- cium concentration in the microhabitat could be higher at interfaces due to mineralization by bacteria, algae or fungi. Consequently, a variety of factors could act as a cue for ovi- position on shells, examples being biofilms (auchwuss) on the shell surface, phero- mones, amino acids from the shell protein, or even physical contact with the shell as a sub- stratum rather than the calcium itself. In the experiments reported here, it is possible that the differing physico-chemical nature of the plastic, pebble and shell surface had facili- tated the development of biofilms that might have differentially encouraged egg deposi- tion on these surfaces; this possibility needs further investigation. Conversely, if an amino- acid attractant was leaking from the shell sur- face, it could be species-specific, because the concentration of amino acids within shells differs in different species of mollusc and with waters of different hardness (Dus- sart, 1973; 1983). Freshwater snails can show chemoreception in relation to amino acids exuded from a food source (Croll, 1983; Thomas et al., 1980), and the parent snails could be following an amino acid gra- dient to the target shell. In natural conditions, the amino acids could act as a primary stim- ulus and calcium could be secondary. The behavioural response to the amino acids in the shell of L. peregra could be examined by other choice-chamber experiments. How- ever, the possibility of synergism must be recognised; for example, Uhazy et al. (1978) noted a synergistic response of B. glabrata to the amino acids proline and glutamine. Despite the fact that Uhazy et al. (1978) found that B. glabrata would orientate to- wards magnesium but not calcium, our dem- onstration of species-specific differences has implications for control of snail vectors of helminth disease; for example, species within the Bulinus and Biomphalaria genera might show similar variability. As suggested by Thomas et al. (1980), Thomas (1982) and Lombardo et al. (1991), baiting techniques might be used to control helminth vectors, and knowledge of the role of calcium could be an important contributing factor. In conclusion, it seems from this study that there are differences in orientation behaviour that might be symptomatic of the calciphile distribution of L. stagnalis compared with the eurycalcic distribution of L. peregra. In the investigation of oviposition-site behaviour of L. peregra, the bias towards laying eggs on shells may have been due to such factors as leakage of calcium or amino acids from the shells, or mechanical quality of the shell sur- face. Both species might be able to orientate towards sources of calcium, though aversion to a low osmotic potential and attraction to chlorides are alternative, though less likely hypotheses. Also, there appears to be a dif- ference in response within a species when the animals are raised under different envi- ronmental conditions, suggesting that there may be the capacity for physiological plas- ticity within these metabolic requirements. LITERATURE CITED BIELEFELD, U., К. KORTJE, H. RAHMANN & W. BECKER, 1993, The shell forming mantle epithe- lium of Biomphalaria glabrata (Pulmonata): ultra- structure, permiability and cytochemistry. Jour- nal of Molluscan Studies, 59, 323-338. 20 PIGGOTT & DUSSART ВОУСОТТ, А. E., 1936, The habitat of freshwater molluscs in Great Britain. Journal of Animal Ecol- ogy, 5, 116-186. BURKY, A., M. BENJAMIN, D. CATALANO & D. HORNBACH, 1979, The ratio of calcareous and organic shell components of freshwater sphaeriid clams in relation to water hardness and trophic conditions. Journal of Molluscan Studies, 45: 312-321. COSTIL-FLEURY, K., 1991, Données sur les gas- téropodes dulcicoles de l'ouest de la France: mise en évidence de cinq groupes d'especes. Haliotis, 21: 125-134. CROLL, R. B., 1983, Gastropod chemo-reception. Biological Review, 58: 293-319. DUSSART, С. В. J., 1973, The ecology and phys- ¡ology of freshwater molluscs in relation to water chemistry. Ph.D. Thesis, University of Salford. DUSSART, G. B. J., 1976, The ecology of freshwa- ter molluscs in north west England in relation of water chemistry. Journal of Molluscan Studies, 42: 181-198. DUSSART, G. B. J., 1979, Life cycles and distribu- tion of the aquatic gastropod molluscs Bithynia tentaculata (L.), Gyraulus albus (Müller), Planor- bis planorbis (L.) and Lymnaea perega (Múller) in relation to water chemistry. Hydrobiologia, 67: 223-239. DUSSART, G. B. J., 1983, The amino acid compo- sition of freshwater molluscan shells in relation to phylogeny and environment. Journal of Mol- luscan Studies, 49: 213-223 DUSSART, G. B. J. & R. KAY, 1980, Relationships between water chemistry and respiration rate in several populations of Lymnaea peregra (Muller). Hydrobiologia, 69: 57-65. FITTER, R. & R. MANUEL, 1986, Collins guide to freshwater life. Collins, London. GREEN, P., G. B. J. DUSSART, & C. GIBSON, 1992, Surfacing and water leaving behaviour of the freshwater pulmonate snail Lymnaea peregra (Muller), Biomphalaria glabrata (Say) and Bulinus jousseaumei (Dautzenberg). Journal of Mollus- can Studies, 58: 169-179. GREENAWAY, P., 1971a, Calcium regulation in the freshwater mollusc Lymnaea stagnalis (L). |, The effect of internal and external calcium concen- tration. Journal of Experimental Biology, 54: 199-214. GREENAWAY, P., 1971b, Calcium regulation in the freshwater mollusc Lymnaea stagnalis (L). Il, Cal- cium movements between internal calcium com- partments. Journal of Experimental Biology 54: 609-620. HARRISON, A. W. NDUKU & A. HOOPER, 1966, The effects of a high magnesium-to-calcium ra- tio on the egg-laying rate of an aquatic planorbid snail, Biomphalaria pfeifferi. Annals of Tropical Medicine and Parasitology, 60: 212-214. KOHN, A., 1961, Chemo-reception in gastropod molluscs. American Zoologist, 1: 291-308. LOMBARDO, F., R. MARAMALDO, B. FRATELLO & D. SONETTI, 1991, Sequential analysis of chemo-reception in Planorbarius corneus. Com- parative Biochemistry and Physiology, 99: 627- 632. MACAN, T. T., 1950, Ecology of freshwater Mol- lusca in the English Lake District. Journal of An- imal Ecology, 19: 124-146. MACAN, T. T., 1974, Freshwater ecology, 2nd edi- tion. Longman, London. MADSEN, H. 1987, Effect of calcium concentration on growth and egg laying of Helisoma duryi, Bu- linus africanus and Bulinus truncatus (Gas- tropoda: Planorbidae). Journal of Applied Ecol- ogy, 24: 823-836. MADSEN, H., 1990, The effect of sodium chloride concentration on growth and egg laying of Heli- soma duryi, Biomphalaria alexandrina and Buli- nus truncatus (Gastropoda: Planorbidae). Jour- nal of Molluscan Studies, 56: 181-187. MADSEN, H., 1992, A comparative study of the food locating ability of Helisoma duryi, Bi- omphalaria camerunensis and Bulinus truncatus (Pulmonata: Planorbidae). Journal of Applied Ecology, 29: 70-78. MCKILLOP, W. & A. HARRISON, 1972, Distribution of aquatic gastropods across the interface be- tween the Canadian Shield and Limestone for- mations. Canadian Journal of Zoology, 50: 1433-1445. MEIER-BROOK, C., 1978, Calcium uptake by Marisa cornuerietis, a predator of schistosos- ome-bearing snails. Archiv fur Hydrobiologie, 82: 449—464. NDUKU, W. К. & А. E. HARRISON, 1976, Calcium as a limiting factor in the biology of Biomphalaria pfeifferi (Krauss). Hydrobiologia, 49: 143-170. OKLAND, J., 1969, Distribution and ecology of freshwater snails (Gastropoda) of Norway. Mal- acologia, 9: 143-151. @KLAND, J., 1983, Factors regulating the distribu- tion of freshwater snails (Gastropoda) in Norway. Malacologia, 24: 277-288. PERKINS, E. J., 1974, The biology of estuarine and coastal waters. Academic Press, London. PIP, E., 1986, A study of pond colonisation by freshwater molluscs, Journal of Molluscan Stua- les, 52: 214-224. THOMAS, J. D., 1982, Chemical ecology of the snail hosts of schistosomiasis; snail-snail inter- actions and snail-plant interactions. Malacolo- gía, 22: 81-91. THOMAS, J. D., В. ASSEFA, С. COWLEY 4 J. OFOSU-BARKO, 1980, Behavioural responses to amino acids and related compounas including propionic acid by adult Biomphalaria glabrata, a snail host of Schistosoma mansoni. Comparative Biochemistry and Physiology, 66C: 17-27. THOMAS, J. D., M. BENJAMIN, A. LOUGH, 4 В. ARAM, 1974, The effects of calcium in the ex- ternal environment on the growth and natality rates of Biomphalaria glabrata. Journal of Animal Ecology, 43, 839-860. UHAZY, S., В. TANAKA, & A. MACINNIS, 1978, Schistosoma mansoni—ldentification of chemi- RESPONSE OF [УММАЕА ТО ENVIRONMENTAL CALCIUM 21 cals that attract or trap its snail vector Bi- omphalaria glabrata. Science, N.Y. 201: 924-926. VAREILLE-MOREL, C., 1986, Contribution a l'é- tude écologique et écophysiologique du mol- lusque prosobranche Potamopyrgus jenkinsi dans la region Dordogne-Haute-Vienne. PhD Thesis-Université de Limoges. WILLIAMS, N. V., 1970, Studies on aquatic pulmo- nate snails in Central Africa. 1. Field distribution in relation to water chemistry. Malacologia, 10: 153-164, YOUNG, J., 1975, A laboratory study using *°Ca tracer, on the source of calcium during growth in two freshwater species of gastropoda. Proceed- ings of the Malacological Society of London, 41, 439-445. Revised Ms accepted 19 October 1994 > + mi sl om a » & ip "нет | ee DR ir Ar MALACOLOGIA, 1995, 37(1): 23-32 ANATOMICAL STUDY ON TONNA GALEA (ИММЕ, 1758) AND TONNA MACULOSA (DILLWYN, 1817) (MESOGASTROPODA, TONNOIDEA, TONNIDAE) FROM BRAZILIAN REGION. Luiz Ricardo Lopes de Simone Seçäo de Moluscos, Museu de Zoologia da Universidade de Sao Paulo, Caixa Postal 7172, CEP 01064-970, Sáo Paulo, SP, Brazil ABSTRACT Tonna galea and Tonna maculosa, from Brazilian region, are described anatomically. Each character is compared between the species and also with other known Tonnoidea. These species differ anatomically in characters of the mantle color and collar, osphradium, hypobran- chial gland, kidney, proboscis length, radular rachidian, lateral tooth, penis, vas deferens, and female genital opening. Characters of the anterior region of the digestive system, heart, penis and pallial oviduct are of particular interest in tonnoidean systematics. INTRODUCTION Tonna galea (Linné, 1758) is a very wide- ranging species, occurring in the Pacific and Atlantic oceans and in the Mediterranean Sea. Tonna maculosa (Dillwyn, 1817), in con- trast, only occurs in the tropical west Atlantic (Rios, 1985: 70; Matthews et al., 1987: 33). Tonna perdix (Linné, 1758), which is closely related to Т. maculosa, occurs in the Indo- Pacific region, and some authors consider them to be synonyms (e.g., Morretes, 1949). Two questions arise: (1) is “Топпа galea” а single species, and (2) are Т. maculosa and Т. perdix really separate species? These ques- tions have been discussed (e.g., Turner, 1948; Matthews et al., 1987), with arguments based on shell characters, but are still unre- solved. The objective of this paper is not to answer these questions, but to be a step in this di- rection, providing anatomical descriptions of specimens identified as Tonna galea and Т. maculosa from the Brazilian region. These data could be used in comparisons with sup- posed co-specific specimens from other ar- eas to assist in clarifying the systematic problems. Another objective of this paper is _ to contribute to the understanding of system- atic characters in the Tonoidea, identifying some useful characters not previously uti- lized in the systematics of this group. Little has been published on the anatomy of the Tonnoidea, and in particular the Ton- nidae. What has been has mainly concerned the alimentary canal and feeding habits. The 23 following information is available: Tonna galea: Weber (1927) studied the digestive system, and some of his data were repro- duced by Hyman (1967); Turner (1948) described the penis and figured the radula; Matthews et al. (1987) figured the penis, jaw and radular rachidian; and Bentivegna & Toscano (1991: 37) figured an active speci- men preying on Holothuria tubulosa and Н. sanctori. Tonna maculosa: Turner (1948) described the penis and figured the radula; and Matthews et al. (1978) figured jaw and radular rachidian. Other Tonnoidea de- scribed anatomically and used here for com- parision are: Reynell (1905) —description of a male Cassidaria rugosa (Linné) (Ranellidae); Day (1969) — digestive system of Argobuc- cinum argus (Gmelin) (Ranellidae); Houbrick 8 Fretter (1969) — digestive system and other organs of three species of Bursa (Bur- sidae) and four of Cymatium (Ranellidae); Lewis (1972) — anatomy of anterior region of digestive system, head-foot complex and pe- nis of Distorsio perdistorta Fulton (Ranell- idae); and Hughes 8 Hughes (1981) — diges- tive system of Cassis tuberosa (Linné), (Cassidae). MATERIAL AND METHODS The specimens studied belong to malaco- logical collection of the Museu de Zoologia, Universidade de Sáo Paulo (MZUSP). They are preserved in 70% ethanol. All specimens were dissected using stan- dard techniques. The buccal region, region of 24 SIMONE female genital opening and pallial oviduct were extracted, dehydrated in ethanol series, stained by carmine, cleared, and fixed with creosote. Radulae and jaws were examined on slides with Hoyer. All drawings were made using a camera lucida. Anatomical terminology is based on Rey- пе! (1905) and Hughes & Hughes (1981). Conchological description and synonymy are omitted, and can be found mainly in Mat- thews et al. (1987) and Turner (1948). Tonna galea (Linné, 1758) (Figs. 1-19, 37) Synonymy and types material: Turner (1948: 173) and Matthews et al. (1987: 31) Diagnosis Shell of clear, homogeneous color; outline globose; sculptured with strong, spiral, somewhat isometrical ridges. Mantle border thick; hypobranchial gland poorly developed. Kidney large, with complex tissue arrange- ment. Proboscis about half projecting from rhynchodeum in fixed specimens. Central cusp of radular rachidian smooth; main cusp of lateral tooth smooth. Penis with a small pointed papilla; anterior region of vas defer- ens fused with seminal receptaculum. Fe- male genital opening with larger inner divi- sion for the bursa opening. Description Shell: Detailed descriptions of the shell given by Turner (1948: 173 pl. 78) and Matthews et al. (1987: 31, figs. 1, 2). Protoconch brown, of almost four glassy, convex whorls. Head-Foot Complex (Figs. 1, 2): Foot large, solid, rounded posteriorly, notched anteri- orly; propodium narrow, with anterior pedal gland opening to ventral slit (Fig. 1, mp). Operculum lacking in adult, present in young (Rios, 1985). Tentacles long, fairly thick, bluntly pointed. Black eyes on tubercles on outer upper part of tentacle bases. Rhyncho- deum with simple, rounded opening. Probos- cis with about half of length projecting from rhynchodeum in all specimens (Fig. 2). Head- foot structures beige, with dark brown, irreg- ular spots. Pallial Complex (Fig. 12): Mantle edge entire, simple, not reflected, thick, rounded, pale cream in color. Pallial cavity occuping first whorl. Siphon long, well developed, pale- beige, with dark-brown, somewhat longitudi- nal, irregular spots. Osphradium large, bipec- tinate, on pallial roof at base of siphon; osphradium leaflets lamellate, pigmented brown. Ctenidium very large, monopectinate (Fig. 12); leaflets very numerous, triangular, low. Hypobranchial gland not well devel- oped; some specimens with folds of this gland running from the anterior and left re- gion of rectum. Flaccid tissue covering rec- tum (and pallial oviduct in females) on right side of mantle cavity, allowing a well-devel- oped ad-rectal sinus. Excretory-Circulatory Systems: Kidney very- large, on right side of pericardium, immedi- ately behind pallial cavity, from which it is separated by a thin, nearly transparent mem- brane (Fig. 12, km); this membrane with slit- like nephrostome, surrounded by muscular fibres to form a sphincter (Fig. 12, ne). Kidney traversed by intestine, which divides it into two lobes, the largest (Fig. 13) anterior, the smallest posterior. Internally, the kidney is very complex, the outer divisions being formed by green-brown lobes and tubes, its anterior limit bulging the posterior region of pallial oviduct glands of females. Nephridial gland cream-colored, poorly developed, sit- uated above nephrostome (Fig. 13, ng). Heart (Fig. 13) with very thin, transparent, flaccid auricle and a very thick, rounded ventricle (Fig. 13). Ctenidial vein on right margin of gill, entering auricle both anteriorly and posteri- orly (Fig. 13). Central part of auricle inserting directly in gill margin (Fig. 13, au). Ad-rectal sinus very developed, apparently continuous from chamber of kidney into an aperture, with muscular walls, near the anus (Fig. 37, at), similar to an ureter. Digestive System: Similar to that described by Weber (1927), typical of tonnoideans (Hughes & Hughes, 1981). Some details of insertion of proboscis gland duct and oe- sophagial caecum duct in buccal complex are shown in Figure 4; detail of buccal com- plex (sectioned dorsally) shown in Figure 5; transversal section of the mid region of ante- rior oesophagus shown in Figure 6. Salivary glands surrounding duct of proboscis gland (Fig. 3). Duct of each proboscis gland looping anteriorly to nerve ring in all specimens ex- amined (Fig. 11). Radular rachidian tricuspate (Fig. 7; Matthews et al., 1987); lateral teeth with a flattened, irregular base and two cusps, a long, large acuminate cusp and a ANATOMICAL STUDY ОМ ТОММА GALEA 25 sheath exposed, scale = 10 mm; (3) anterior region of digestive system, dorsal view, scale = 10 mm; (4) anterior extremity of opened proboscis, right-dorsal view, scale = 1 mm; (5) buccal complex opened dorsally, inner view, scale = 2 mm; (6) transversal section in mid region of the anterior oesophagus, ventral region down, scale = 1 тт; (7) rachidian tooth of the radula; (8) lateral tooth showing the accessory cusp (ac); (9) inner marginal tooth; (10) outer marginal tooth, scale (Figs. 7-10) = 0.5 mm; (11) dorsal view of the region of nerve ring, scale = 2 mm. 26 SIMONE FIGS. 12-19: Tonna galea: (12) inner view of pallial cavity and viceral mass of a male, scale = 10 mm; (13) detail of opened nephridial and pericardial chambers, scale = 2 mm; (14) viceral mass of a male, mantle partially removed, scale = 2 mm; (15) ventral view of the penis, scale = 2 mm; (16) detail of insertion region of the vas deferens in seminal groove, showing the receptaculum, scale = 2 тт; (17) pallial oviduct, tegument removed, scale = 10 mm; (18) detail of the albumen gland showing the vesicles (ve), scale = 2 mm; (19) detail of female genital pore, tegument removed, scale = 2 mm. ANATOMICAL STUDY ON ТОММА GALEA 27 minute cusp (Fig. 8, ac); base of inner mar- ginal teeth (Fig. 9) a little longer than that of outer marginal teeth (Fig. 10). Stomach poorly developed, with two ducts to digestive glands, without developed style sac, folds or typhlosole. Nervous System: Nerve ring (Fig. 11, nr) with cerebral ganglia turned to left side in all spec- imens; from these, three pairs of nerves run anteriorly, fusing near the proboscis gland duct loops to become only one pair, which lie ventraly to the oesophagus (Fig. 11, np). In mid oesophagial region, this pair of nerves bifurcates, the median nerves (Fig. 6, in) in- nervating radular bulb and the lateral nerves (Fig. 6, on) innervating proboscis wall. Genital System, Male: Testis (Fig. 14, tt) branching into digestive gland, mainly on col- umellar surface of viceral mass. Convoluted seminal vesicle rather spheric (Fig 14, pt), confined to anterior part of viceral mass just anterior to testis. Vas deferens thin walled (Fig. 14), its anterior region to right of the re- ceptaculum, fused with its walls (Fig. 16). Re- ceptaculum a modified, bulging region of the spermatic groove posterior to vas deferens insertion in spermatic groove (Fig. 16). In floor of right margin of pallial cavity, sper- matic groove thick walled near prostate gland (Figs. 1, 14, 16). Penis large, somewhat flattened, with open penial groove, which ter- minates in a small, pointed papilla at central region of penis tip (Fig. 15). Female: Ovary branching into digestive gland. Oviduct slender (Fig. 17), with a small gonopericardial duct. Oviduct opening into a short, thick-walled albumen gland. A series of small, paired vesicles present in ventral side of the albumen gland (Figs. 17, 18, ve). Capsule gland long, curved, thick walled (Fig. 17). Bursa copulatrix long, claviform, sepa- rate and to right of capsule gland. Posterior limit of bursa sacciform, thin walled, the walls gradually becoming thickly muscular anteri- orly (Figs. 17, 19, bc). Genital pore small (Figs. 17, 19, gp), to right, behind anus; pore with two inner divisions: the largest and pos- terior is the end of the bursa, the smallest and _ anterior the end of pallial oviduct (Fig. 19). Measurements Length, width in mm and if mature (m) or immature (i): MZUSP 27967: male, 78.5 by 66.5, m; female, 73.0 by 61.0, 1; male, 69.5 by 55.5, m; MZUSP 27984: male, 74.2 by 60.6, m; MZUSP 27968: female, 102.0 by 85.0, m; MZUSP 27986: male 118.5 by 104.0, т: MZUSP 27969: female 135.0 by 115.0, т; male 98.7 by 71.0, m. Habitat The specimens were obtained by diving, burrowing in sandy sediment, near rocks or reefs. Some specimens were also dredged from about 150 m depth in muddy sediment. Material Examined BRAZIL. Espirito Santo: MZUSP 27970, 1 male and 1 female, Barra do Riacho (8/ix/72). Sao Paulo: MZUSP 27967, 2 males and 1 female, Saco da Ribeira Beach, Ubatuba; MZUSP 27983, 2 females, Enseada Beach, Ubatuba (x/91); MZUSP 27984, 1 male, est. 42, otter traw (22/x/86); MZUSP 27985, 1 male, IOUSP-Veliger, ‘‘rede de pesca fixa 8”; URUGUAY. off Maldonado: MZUSP 27968, 1 female, 35°18’S 52°32’W, “W. Besnard,” station 1920, 150 m deep (30/x/72); MZUSP 27969, 2 males and 1 female, same data; MZUSP 27986, 4 males and 1 female, ““W. Besnard,’ station 1921 OT.9 (20/x/72). Tonna maculosa (Dillwyn, 1817) (Figs 20-36) Synomymy and type material: Turner (1948: 169) and Matthews et al (1987: 37) Diagnosis Shell dark-brown, spotted; outline fusi- form; sculpture of low spiral ridges. Mantle border thin. Hypobranchial gland developed. Kidney with a smooth surface. Proboscis to- tally within rhynchodeal cavity in fixed spec- imens. Radular rachidian with crenulations on base of central cusp; crenulation on main cusp of lateral radular teeth. Penis without papilla, with a flap of the tegument on tip. Anterior region of vas deferens separated from walls of receptaculum. Inner division in female genital opening of capsule gland larger than in 7. galea. Description Shell: Detailed decriptions of shell are given by Turner (1948: 169-172, pl. 75, fig. 2, pl. 76, figs. 1, 2) and Matthews et al. (1987: 37, fig. 28 SIMONE 6). Protoconch of almost four glassy, convex whorls, brown in color (Figs. 20, 21). Head-Foot Complex (Fig. 22): Foot solid, large, rounded posteriorly, notched ащеп- orly; propodium narrow, with anterior pedal opening to ventral slit (Fig. 22, mp). Opercu- lum lacking in adult. Tentacles long, fairly thick, bluntly pointed (Fig. 22). Black eyes on tubercles on outer upper part of tentacle base. Rhynchodeum with simple, rounded opening. Proboscis completely retracted within proboscis sheath in all specimens. Color of all head-foot structures beige, with clear-brown irregular spots. Pallial Complex (Fig. 31): Mantle edge entire, simple, not reflected, thin, flattened, pale cream in color. Pallial cavity occupying first whorl. Siphon long, well developed, pale- beige in color, with clear-brown irregular spots. Osphradium large, bipectinate, pro- portionaly larger than that of Т. galea, situ- ated on pallial roof at base of siphon; osphra- dium leaflets lamellate, pigmented brown. Ctenidium very large, monopectinate, with many low, triangular filaments. Hypobran- chial gland developed, along left side of an- terior region of rectum. Tissue covering rec- tum and pallial oviduct of females less flaccid than that of 7. galea, but allowing a well-de- veloped ad-rectal sinus. Excretory-Circulatory Systems: Kidney mod- erately large (Fig. 30), forming a sac situated like that of Т. galea. Well-developed, slit-like nephrostome, surrounded by muscular fibres to form a sphincter (Figs. 30, 31, ne). Internal structures of kidney similar, but simpler than in 7. galea, with a smooth surface and a cream color; its anterior limit does not bulge with albumen gland of females. Nephridial gland somewhat inconspicuous, situated dorsal to nephostome (Fig. 30, ne). Heart (Figs. 30, 31) with a very thin, transparent, flaccid walled auricle, and a very thick-walled, rounded ventricle. Ctenidial vein and auricle like those of T. galea. Ad-rectal sinus well developed but less than that of 7. galea; as in that species, sinus apparently continuous from kidney chamber into an aperture (Figs 35, 36, at), with mus- cular walls, near anus, like an ureter. Digestive System: Like that of Т. galea (Figs. 23-29). Structures within buccal bulb very similar to those of 7. galea; a pair of ventral jaws (Fig. 24, md), two dorso-lateral folds (If) in buccal cavity, one on either side, with opening of proboscis gland duct median to fold and near its anterior end. Radula with rachidian with a crenulation on base of its central cusp; tip of this cusp slender and smooth (Fig. 26). Lateral teeth with a series of crenulations on main cusp (Fig. 25, cr); small accessory cusp present (Figs. 25, 27, ac). In- ner (Fig. 28) and outer (Fig. 29) marginal teeth similar of those of Т. galea. Oesophagus and its inner ventral folds and glands similar to 7. galea, but much shorter (Figs. 23, 24). Oe- sophageal caecum present, the folds and glands of the oesophagus terminating in os- tium of caecum (Fig. 24, eo). Posterior oe- sophagus without distinct glands or crop. Stomach poorly developed, with two ducts to digestive glands, but without developed style sac, folds, or typhlosole. Inner surface of posterior oesophagus, stomach and intestine with low longitudinal folds. Salivary glands, proboscis glands and their ducts (Figs. 23, 24), similar of those of 7. galea (Weber, 1927). All anterior structures of digestive system maintained in position by a tridimensional net of muscle fibres running to wall of vesopha- gus, body wall and foot. Looping section of ducts of proboscis gland lying anterior to nerve ring, as in 7. galea. Nervous System: Nerve ring (Fig. 23, nr) with cerebral ganglia turned to left side in all spec- imens examined. Ventral pair of nerves of proboscis similar to those of 7. galea. Genital System, Male: Testis branching into digestive gland, concentrated mainly on col- umellar surface of viceral mass. End of vas deferens an enclosed, small, thin-walled tube, lying to right of receptaculum, without fusion except for insertion (Fig. 34). Recep- taculum a modified, bulging region of sper- matic groove, posterior of insertion of vas deferens in floor of pallial cavity (Fig. 34). Spermatic groove in right side of floor of pal- lial cavity, thick walled, by the prostate gland (Figs. 22, 32, 33, 34). Penis very large (Figs. 32, 33), curved backwards, somewhat flat- tened, its free end curving downwards fol- lowing curve of floor of pallial cavity, blunt at the apex. At right side, near apex of penis there is a flap of tissue (fig. 32: pf), under which the penial duct opens; there is no pa- pilla. Female: Ovary branching into digestive gland. Oviduct slender, opening into a short, thick-walled albumen gland (Fig. 35). A series of paired vesicles present in ventral side of ANATOMICAL STUDY ON ТОММА GALEA 29 FIGS. 20-31; Tonna maculosa: (20) protoconch in profile; (21) protoconch, apical view, scale (Figs. 20, 21) = 2 mm; (22) head-foot complex from male, scale = 5 mm; (23) dorsal view of anterior region of the digestive system, proboscis opened, scale = 5 mm; (24) the same, oesophagus opened longitudinally, scale = 5 mm; (25) lateral tooth of radula, showing the crenutation (cr) and the accessory cusp (ac), scale = 0.1 mm; (26) rachidian tooth; (27) lateral tooth; (28) inner marginal tooth; (29) outer marginal tooth, scale (Figs. 26-29) = 0.2 mm; (30) detail of opened nephidial and pericardial champers, scale = 2 mm; (31) pallial cavity of a male, inner view, scale = 5 mm. 30 SIMONE FIGS. 32-36: Tonna maculosa: (32) penis and seminal groove, dorsal view; (33) the same, ventral view, scale = 5 тт; (34) detail of the insertion of vas deferens in seminal groove, showing the receptaculum, scale = 1 mm; (35) pallial oviduct ventral view, tegument partially removed, scale = 5 mm; (36) detail of female genital pore, tegument removed, scale = 2 mm. albumen gland, similar to those of 7. galea, Capsule gland long, сигуеа, thick walled (Fig. 35, cg). Bursa copulatrix long, claviform, slender, separate and to right of capsule gland; posterior end of the bursa sacciform, thin walled, the walls becoming thick and muscular anteriorly (Fig. 35, bc). Small genital opening at right and posterior to anus (Figs. 35, 36, gp). Genital opening with two inner divisions, the smallest and posterior is end of bursa (bc), and the larger and anterior is end of capsule gland (cg) (fig. 36). Measurements MZUSP 27961, female = length 65.0 mm by width 44.0 mm; male = 40.4 mm by 26.5 mm. MZUSP 27962, female = 45.3 mm by 31.0 mm. Habitat The collected specimens were found by diving or at low tide, burrowing on sandy bot- toms near reefs. Material Examined BRAZIL. Bahia: MZUSP 27961 (one male and one female) Цариа Beach, Salvador (7/ vii/71); MZUSP 27962 (one female) Карча, Salvador (29/ix/84). DISCUSSION Tonna galea differs anatomically from 7. maculosa in having (1) a thick mantle border; (2) darker spots on the epidermis; (3) a less- developed hypobranchial gland; (4) a propor- tionally smaller osphradium; (5) a more de- veloped kidney, with more complex internal structure; (6) proboscis extending 50% from rhynchodeum (in 7. maculosa proboscis al- ways completely retracted within proboscis sheath) in fixed specimens; (7) oesophagus and inner oesophagial structures much longer; (8) central cusp of radular rachidian teeth and lateral teeth without crenulations (present in 7. maculosa); (9) penis with a pa- ANATOMICAL STUDY ОМ TONNA GALEA 31 FIG. 37: detail of anal region of Tonna galea, scale = 2 mm. pilla (7. maculosa has a flap, without papilla); (10) end region of vas deferens fused with the receptaculum walls (in 7. maculosa this duct is free); and (11) female genital pore with larger inner opening to the bursa (7. таси- losa has the larger opening leading to the capsule glana). The function of the aperture near the anus (Figs 36, 37, at) in both species is unknown. It probably controls the exit of the inner fluid of the ad-rectal sinus, which 1$ apparently con- tinuous to the kidney chamber. These struc- tures resemble the ureter of the Viviparidae (Hyman, 1967), for example, which have no well-developed nephrostome as in Tonna (Figs. 21, 31, ne). The fortuitous use of the ad-rectal sinus as an ureter merits further study. Some differences between the literature accounts (Turner, 1948; Mattews et al., 1987) and the specimens studied here were: (1) the protoconchs of both species are closely sim- _ ilar, and of almost four whorls, in contrast with the data of Matthews et al. (1987), in which differences in number of whorls was given; (2) the jaw lies ventral to the proboscis; (3) the radular rachidian of 7. maculosa has a crenulation only on the base of the central cusp; the tip of this cusp is smooth and slen- der (Fig. 26, cr); (4) the radular lateral tooth has a small, but conspicuous accessory basal cusp in all specimens examined of both species (Figs 8, 25, 27, ac); (5) Turner (1948: 168) reported an extremely long and flagel- late papilla in the penis of Т. maculosa, dif- ferent from the penis described herein, in which the papilla is lacking (Figs. 32, 33). Fur- ther investigation is need to determine the significance of these differences. The proboscis of Tonna, as in all known Tonnoidea, has a great development of the buccal mass, this region taking most of the proboscis length (Figs. 2, 23). The proposal of the tonoidean proboscis as a distinct type (Day, 1969) is perhaps not justified, but rather it can be regarded as a specialized and mod- ified pleurembolic type. The auricle structure and pallial oviduct may be considered as additional characters of Tonnoidea, in addition to the anterior region of the digestive system. This type of auricle 1$ found in Cassidaria rugosa (Ranellidae) (Rey- nell, 1905). However, no reference to the pal- lial oviduct has been found in the literature except for Bursa cruentata (Houbrick & Fret- ter, 1969: 417), but details are missing that would allow a full comparison. Besides radular aspects, other charac- ters of the Tonna digestive system differing from other Tonnoidea (Reynell, 1902; Day, 1969; Houbrick & Fretter, 1969; Lewis, 1972; Hughes & Hughes, 1981) are (1) the presence of a oesophagial caecum and (2) the absence of a clear oesophagial gland (crop or bulb) in the posterior oesophagus. These are per- haps characters of Tonnidae. LITERATURE CITED BENTIVEGNA, F. 8 A. TOSCANO, 1991, Observa- tion au laboratoire sur le comportement alimen- taire de trois especes de la superfamille Ton- noidea (Mollusca, Gastropoda). Revue Française d’Aquariologie Herpetologie, 18: 33-38 DAY, J. A., 1969, Feeding of the cymatiid gastro- pod, Argobuccinum argus, in relation to the structure of the proboscis and secretions of the proboscis gland. American Zoologist, 9: 909- 916 HOUBRICK, J. R. 8 V. FRETTER, 1969, Some as- pects of the functional anatomy and biology of Cymatium and Bursa. Proceedings of the Mala- cological Society of London, 38: 415-429 HUGHES, В. М. & H. Р. 1. HUGHES, 1981, Mor- phological and behavioural aspects of feeding in the Cassidae (Tonnacea, Mesogastropoda). Ma- lacologia, 20: 385-402 32 SIMONE HYMAN, L. H., 1967, The invertebrates, Volume VI, Mollusca |. McGraw-Hill Book Company. New York, 792 pp. LEWIS, H. 1972, Notes in the genus Distorsio (Cy- matiidae) with descriptions of new species. Nau- tilus, 86: 27-50 MATTHEWS, H. R.; J. Н. М. LEAL 8 A. С. S. CO- ELHO, 1987, Superfamilia Tonnacea no Brasil. Vli—Familia Tonnidae (Mollusca: Gastropoda). Arquivos de Ciéncias do Mar, 26: 29-45 MORRETES, F. L., 1949, Ensaio de catálogo dos moluscos do Brasil. Arquivos do Museu Para- naense, 7: 2-216 REYNELL, A., 1905, Some account of the anatomy of Cassidaria rugosa (Linn.). Proceedings of the Malacological Society of London, 6: 292-299, pl. 6 RIOS, E. C., 1985, Sea shells of Brazil. Fundacáo Universidade de Rio Grande, Fundacáo Cidade de Rio Grande, Museu Oceanográfico. Rio Grande 239 pp., 102 pls. TURNER, R. D., 1948, The family Tonnidae in the western Atlantic. Johnsonia, 2: 165-192 WEBER, H., 1927, Der Darm von Dolium galea L. eine vergleichend anatomische Undersuchung unter besonderer Berücksichtigung der Trito- пит Arten. Zeitschrift für Morphologie und Okologie der Tiere, 8: 663-804 Revised Ms. accepted 26 October 1994 ABBREVIATIONS ac: accessory cusp of lateral tooth ag: albumen gland ao: anterior aorta an: anus at: aperture of ad-rectal sinus au: auricle Be: bursa copulatrix bu: buccal complex CE central fold of buccal complex cm: columellar muscle Cp: capsule gland ch crenulated ridge CV: ctenidial vein dc: duct of oesophagial caecum dg: digestive gland dp: duct of proboscis gland ес: oesophagial caecum eo: ostium of oesophagial caecum floor of pallial cavity foot gonopericardial duct gill female genital opening hypobanchial gland inner proboscis nerve intestine kidney membrane between kidney and pallial cavities lateral fold of buccal complex buccal lips mantle border mandibule (jaw) muscle fibers mid-ventral mucous gland of oesophagus mouth anterior pedal gland nephrostome nephridial gland nerve ring anterior oesophagus oesophagial folds outer proboscis nerve posterior oesophagus osphradium oviduct posterior aorta proboscis pericardial chamber penis penian flap penian seminal groove proboscis nerve proboscis gland penian papilla : convoluted seminal vesicle radular complex proboscis sheath receptaculum seminalis radular nucleus rhynchodeum rectum salivary gland seminal groove siphon cephalic tentacle testis vas deferens vesicles of albumen gland ventricle MALACOLOGIA, 1995, 37(1): 33-40 A TAXONOMIC APPLICATION OF MULTIVARIATE MIXTURE ANALYSIS IN PATELLIDAE J. D. Acuña? & М. A. Muñoz* ABSTRACT Multivariate mixture analysis is a powerful tool that appears to be useful for dealing with situations in which several traits with overlapping variation are used for species discrimination. A multivariate mixture analysis technique is applied to discrimination of two sibling species of the genus Patella (P. aspera Róding and P. caerulea Linnaeus). These species show a sub- stantial overlap in the distribution of maximum shell width and shell height. In spite of this overlap, the results of mixture analysis in a sample of 101 specimens, classified but treated as unclassified for the purpose of the analysis, suggest the existence of two mixed distributions. Moreover, examination of specimen classification derived from mixture analysis reveals that these mixed distributions correspond to P. aspera and P. caerulea. The estimated values of mixture parameters confirm a substantial overlap in the bivariate distribution of maximum shell width and shell height of the two species. We hope that these results will contribute to making multivariate mixture analysis more popular among taxonomists. Key words: multivariate mixture analysis, taxonomy, species discrimination, Gastropoda, Patellidae, Patella aspera, Patella caerulea. INTRODUCTION Individual variation in morphological traits often provides valuable information for spe- cies discrimination. However, the interpreta- tion of variation is sometimes difficult due to polymorphism, polytypy, and similarity be- tween species. A particularly difficult case arises when species overlap in the distribution of contin- uous morphologic traits used for their classi- fication. Overlap in these traits can be mis- takenly interpreted as evidence in favor of interbreeding when, in fact, two or more non- hybridizing species are involved. The solution to this problem requires a de- tailed analysis of trait variation which is fre- quently made resorting to several statistical methods. However, taxonomists have not taken full advantage of recent advances in statistics to deal with the problem of overlap- ping variation. Mixture analysis techniques offer a useful alternative to traditional meth- ods of statistical analysis when overlapping variation is a concern (Everitt 8 Hand, 1981; Titterington et al., 1985; McLachlan 8 Bas- ford, 1988). Although originally developed to deal with problems in the biological realm (Pearson, 1894), mixture analysis has received little at- tention from biologists. This may be in part due to difficulties in the computations re- quired by the method. Recently, several au- thors have confirmed the usefulness of mix- ture analysis for species discrimination (Do & McLachlan, 1984), the resolution of the age- class structure of a population (Equihua, 1988), and the study of sexual dimorphism (Flury et al., 1992). In a previous герой (Muñoz & Acuña, 1994), we were able to discriminate between two sibling species of the genus Patella (P. aspera Róding and P. caerulea Linnaeus) by means of a univariate mixture analysis tech- nique. The specimens included in the sample were actually classified, but for the purpose of the analysis they were treated as unclas- sified. The taxonomic trait that we selected was shell height. Our aim was to show the usefulness of mixture analysis as a tool for species discrimination using a trait with over- lapping variation. Here we attempt the dis- crimination of these two species with the same method but using two conchological traits, which requires the use of multivariate mixture analysis. Multivariate mixture analy- sis provides powerful techniques for dealing with situations in which several traits with “Departamento de Biología Animal, Universidad de Valencia, Dr. Moliner, 50, 46100 Burjasot, Valencia, España. “Departamento de Ciencias Morfolögicas |, Universidad Complutense de Madrid, Arcos de Jalón, s/n, 28037 Madrid, España. 34 ACUÑA & MUÑOZ overlapping variation are used for species discrimination. We hope that our results will contribute to making multivariate mixture analysis more popular among taxonomists. PATELLA ASPERA AND PATELLA CAERULEA Patella aspera Róding (= P. ulyssiponensis Gmelin) and P. caerulea Linnaeus are two very abundant European marine gastropods. The first is found in the Mediterranean and is also widely distributed along the Atlantic coast, whereas the second 1$ restricted to the Mediterranean. The two species live in very similar habitats. Both are found on hard sub- strata in the littoral zone and reach upper subtidal levels, although P. aspera has a lower distribution range in the littoral zone and exhibits a preference for areas exposed to wave action. From a taxonomic stand- point, P. aspera and P. caerulea are consid- ered two separate species, although their specific status was a contentious issue for some time. Part of the protracted debate regarding the taxonomic status of these two species came about because there 1$ substantial overlap in the distribution of traits (shell shape, orna- mentation, and coloration) used for their clas- sification. In the past, taxonomists interpreted this overlap as arising from interbreeding and accordingly rejected a specific distinction or considered this to be an instance of incom- plete speciation (Fischer-Piette, 1935, 1938; Evans, 1953, 1958). Others, however, argued that reproductive features (e.g., timing of the breeding season) could be used as a basis for discriminating between the two species (Fi- scher-Piette, 1948). The controversy was fi- nally settled when Fischer-Piette 8 Gaillard (1959) reported clear-cut species differences in the single cusp lateral teeth of the radula. More recently, analyses of a variety of taxo- nomic traits has confirmed the taxonomic va- lidity of Patella aspera and P. caerulea. Such is the case of studies that relied on caryotypic (Cervella et al., 1988), electrophoretic (Sella et al., 1989; Cretella et al., 1990) and soft-part traits (Cretella et al., 1990). A review of the literature reveals that con- chological traits, when taken altogether, allow for a separation of the two species. Indeed, several authors have relied on conchological traits to separate samples, the species mem- bership of which was later confirmed by dif- ferences in radular (Fischer-Piette & Gaillard, 1959), electrophoretic (Cretella et al., 1990), or soft-part traits (Cretella et al., 1990). Never- theless, taxonomists were reluctant to grant these Patella their current specific status until clear-cut differences were found in non-con- chological traits. This may be partly due to the difficulties involved in analyzing shell traits with overlapping variation. MATERIALS AND METHODS Our study of shell height by means of univariate mixture analysis (Muñoz & Acuña, 1994) used a very large sample. Part of this material was judged suitable for the present study. The complete sample (detailed description in Muñoz & Acuña, 1994) included over а thousand specimens of Patella aspera and P. caerulea obtained at Cabo Oropesa, Castel- lón, España. Specimens of all available sizes were collected randomly in a narrow band of uniform characteristics located at the base of the littoral zone. Sampling took place in May 1989. In the laboratory, shell length (distance between anterior and posterior shell mar- gins), maximum width (maximum distance between lateral shell margins), and height (distance between apex and line between an- terior and posterior shell margins) were mea- sured to the nearest 0.05 mm using calipers on 1025 useful, whole shells. Specimens were also assigned to either species using non-conchological traits, mainly foot mor- phology and color (Cretella et al., 1990). At Cabo Oropesa, Patella aspera can be easily recognized by its pyriform or oval foot, with sole yellow or cream with no dark areas. Pa- tella caerulea, on the other hand, has an oval foot, with sole dark gray or bluish with edge and center cream. Use of these diagnostic characters resulted in 439 specimens being classified as P. aspera and 581 as P. caer- ulea. Five specimens could not be unambig- uosly assigned to either species and were classified as doubtful. Later, the range of shell lengths in the total sample (5.30-35.10 mm) was divided into 31 intervals of 1 mm, which were operationally considered as growth stages. Specimens were, irrespective of their specific identity, grouped into these 31 shell-length class intervals. Because the performance of mixture anal- ysis is greatly improved by a large sample MULTIVARIATE MIXTURE ANALYSIS IN PATELLIDAE 35 size (e.g., Equihua, 1988), only the largest subsample was considered for the analysis performed in the present study. This sub- sample included all specimens with shell length between 14 and 15 mm (n = 103). Ac- curate measurements of maximum shell width and shell height could only be obtained from 101 of the 103 specimens (37 belonging to Patella aspera and 64 to P. caerulea) yield- ing 101 bivariate data that were submitted to the statistical analysis. The subsample was considered a mixture (superposition of density functions) of two components with bivariate normal distribu- tion as to maximum shell width and shell height. Multinormality of the continuous phe- notypic traits in a population is a frequent theoretical assumption (Нат & Clark, 1989). In a real mixture situation, testing for normal- ity would be impossible because group membership of the specimens 1$ not known. In our sample of classified specimens, multi- normality was tested for confirmation. It was done for each species separately using (1) Kolmogorov-Smirnov goodness-of-fit test with Lilliefors correction (Sokal & Rohlf, 1981) to test the univariate normality of the mar- ginal distributions, and (2) Mardia’s test of multinormality (Mardia, 1970; Appendix 1) to evaluate joint normality. Although marginal normality does not imply joint normality, the converse is true. Hence, test (1) is useful for a fast and easy detection of many types of non-normality, whereas test (2) permits eval- uation of multivariate normality when results of the previous test are inconclusive (i.e., fails to reject the hypothesis of normality). In the Patella sample case, results of good- ness of fit testing revealed no significant de- partures from marginal normality and multi- normality (P > 0.05) in the growth stage (shell-length class interval) that is the focus of this report, as well as in most other growth stages of either species. Statistical analysis of the mixture required estimation of 11 parameters, including four means, four variances, two covariances, and one mixing parameter (proportion of either component in the mixture). The estimation was accomplished by means of a maximum- likelihood approach. The EM algorithm (Dempster et al., 1977) was used to compute the maximum-likelihood estimates of the pa- rameters using the equations studied by Wolfe (1970). The procedure has been de- scribed by Everitt & Hand (1981) for the gen- eral case of a mixture of multivariate normal distributions (an arbitrary number of mixed distributions and variables, with all parame- ters unknown) (Appendix 2). This method, although slow, is easily pro- grammable and very stable (i.e., shows little dependency on the initial estimates of the parameters). In the case studied here, includ- ing two mixed distributions and two vari- ables, the programming was particularly sim- ple. Initial estimates of the parameters were computed by the program directly from the data. The variance/covariance matrices of the components were assumed equal to the variance/covariance matrix of the mixture, and the means were calculated by imposing a small deviation (10% of the range of the variables) on both sides of the mixture means. The means calculated in this way were then grouped into vectors taking into account the sign of the correlation of the data in the mixture. Mixing proportions were set at 0.5. This approach has proved successful in several runs with simulated data. In addition, different initial estimates were used in order to confirm that the results of the mixture anal- ysis did not correspond to a local maximum of likelihood (Appendix 2). The convergence criterion of the iterative procedure was spec- ified in terms of the Euclidean distance be- tween successive estimates of the parameter vector. Following Everitt (1984), the criterion value was set at 0.0001. A likelihood ratio test was used to evaluate the goodness-of-fit of the mixture (Hassel- blad, 1969; Everitt & Hand, 1981; Equihua, 1988). The statistic is given by G=2 (Ly — № where L, is the log-likelihood computed un- der the null hypothesis (which assumes only one distribution), and L, is the log-likelihood under the alternative hypothesis (which as- sumes the mixture of two distributions). Sta- tistic G is asymptotically distributed as chi- square with degrees of freedom equal to the difference in the number of parameters be- tween the two hypotheses (six in the present case). This approach, however, is not devoid of criticisms (Everitt & Hand, 1981; Tittering- ton et al., 1985). Under the null hypothesis, the mixing proportions fall in the boundary of the parameter space, so that conditions for G to be asymptotically distributed as chi- square are not fulfilled. However, a satisfac- tory performance of the test was found by Hasselblad (1969) with mixtures of exponen- 36 ACUÑA & MUÑOZ tial, Poisson, and binomial distributions. Based on this evidence, Equihua (1988) has argued that the likelihood ratio test can be of assistance in assessing the number of com- ponents in a mixture. Once the analysis was completed, the probabilities of membership (posterior prob- abilities) of each datum (specimen) to each component (species) in the bivariate mixture were calculated by dividing the component density function weighed by its proportion in the mixture by the mixture density function. The presence of two components in the mixture could be taken as evidence of a spe- cific discrimination. Alternative interpreta- tions based on other phenomena that yield mixed distributions in natural populations ap- pear rather unlikely. For example, the species are hermaphroditic (Bacci, 1947; Fretter & Graham, 1976), thus sexual dimorphism is not a likely explanation for the presence of two components in the mixture. Mendelian segregation in the shell dimensions of these two species has never been reported, and it is likely that, as is the case with other organ- isms, traits related to body size are under additive polygenic control resulting in a nor- mal distribution of maximum shell width and shell height (Falconer, 1989; Hartl 8 Clark, 1989). Also, because the samples were col- lected in a uniform environment, a possible effect of disruptive selection and/or differen- tial reaction seems unlikely. RESULTS Figure 1 shows scatterplot of the two vari- ables measured in the 14-15 mm shell-length growth stage. Inspection does not permit discrimination of two components. However, the results of the mixture analysis (Table 1) suggest the existence of two mixed multivari- ate distributions. This can be inferred from (1) the absence of an empty component in the mixture; (2) the absence of mixing propor- tions suggesting a component with a mar- ginal representation as could arise from a spurious frequency peak in the tails of the distribution, and (3) the results of the likeli- hood ratio test, which allow rejection of the null hypothesis with P < 0.05 (G = 12.72; d.f. = 6). Furthermore, the estimated values of the parameters indicate a substantial overlap in the bivariate distributions for the two compo- nents. The taxonomic interpretation of the results SHELL HEIGHT (mm) 9 10 11 12 13 MAXIMUM SHELL WIDTH (тт) FIG. 1. Scatterplot for the sample used in the present study. Specimens belonging to Patella as- pera are symbolized by closed dots, and those be- longing to P. caerulea by open dots. Superim- posed lines limit the regions with probabilities of membership (posterior probabilities) > 0.95 in the components detected through the use of mixture analysis. TABLE 1. Results of the mixture analysis. Component 1 Component 2 (Patella aspera) (Patella caerulea) Mixing proportion 0.45 0.55 Maximum shell width mean 10.71 11.62 Shell height mean 4.76 3.88 Maximum shell width variance 0.35 0.31 Shell height variance 0.35 0.18 Maximum shell width—Shell height covariance 0.04 0.07 of the mixture analysis is straightforward if one considers simultaneously the probability that each specimens belongs to any one component in the mixture and the results of their specific diagnosis based on non-con- chological traits (Table 2; Fig. 1). When strin- gent probability levels are applied (e.g., 0.99 or 0.95), the classification derived from the mixture analysis is congruent with that based on non-conchological traits. With few excep- tions, the component with the lowest maxi- mum shell width mean and the highest shell height mean (component 1) corresponds to Patella aspera shells, whereas the compo- nent with the highest maximum shell width MULTIVARIATE MIXTURE ANALYSIS IN PATELLIDAE 37 TABLE 2. A comparison of the classifications arising from mixture analysis and from specific diagnosis based on non-conchological traits. PM: probability of membership in a component; A: number of spec- imens ascribed to either component; lA: number of incorrect ascriptions; OIA: observed percentage of incorrect ascriptions; ElA: expected percentage of incorrect ascriptions. PM A IA OIA(%) EIA(%) 0.99 31 0 0.00 1.00 0.95 59 5 8.47 5.00 0.90 70 5 7.14 10.00 0.75 89 8 8.99 25.00 0.50 101 11 10.89 50.00 mean and the lowest shell height mean (com- ponent 2) corresponds to P. caerulea shells. The number of incorrect ascriptions in- creases at low probability levels (e.g., 0.75 or 0.50), but is always under reasonable values. Therefore, it seems safe to conclude that the two components detected through the use of mixture analysis correspond to P. aspera and P. caerulea. Although the following discussion is mainly based on the analysis of specimens between 14-15 mm in shell length, the same proce- dure was also applied to other shell-length class intervals in the sample. Only occasion- ally did mixture analysis reveal two compo- nents using data from a single interval [e.g., 18-19 mm (n = 65)]. However, in some cases discrimination was accomplished after pool- ing data from two successive intervals [e.g., 14-15 mm + 15-16 mm (n = 101 + 87), 17-18 mm + 18-19 mm (n = 68 + 65)]. DISCUSSION The results of the mixture analysis pre- sented here indicate that there are only slight differences between the sibling species Pa- tella aspera and P. caerulea as far as the bi- variate distribution of maximum shell width and shell height. The shell of P. aspera 1$ slightly narrower and taller than that of P. caerulea. It is worth noting that a third vari- able, namely shell length, was bearing on the results of the bivariate analysis, because the specimens were grouped into growth stages according to shell length. Therefore, the spe- cific differences revealed by the analysis con- cern all three variables that determine the general shape of the patelliform shells. The shell of P. aspera is slightly more oval-conic and elevated than that of P. caerulea. This difference has been reported in type-materi- als for a long time (Bucquoy et al., 1886; Christiaens, 1973; Powell, 1973) and is po- tentially interesting, because shell shape is a taxonomic trait that was extensively investi- gated in relation to adaptive value (Segal, 1956; Lowell, 1984). The discrimination between Patella aspera and P. caerulea extends the results of a pre- vious report (Muñoz & Acuña, 1994) and il- lustrates the use of multivariate mixture anal- ysis to deal with situations in which several traits with overlapping variation are used for classification. Multivariate mixture analysis proved capable of discriminating between these two species in spite of a substantial overlap in the mixed distributions and a mod- erate sample size. We hope that these results will encourage the use of multivariate mixture analysis among taxonomists. Moreover, mixture analysis seems prefera- ble in many taxonomic applications to other statistical procedures, such as discriminant analysis and ordinary cluster analysis (Flury et al., 1992). Mixture of distributions arise when a population is subdivided into homo- geneous components, but it is unknown from which of the components any given observa- tion originates. Mixture analysis models this situation, attempting to estimate the statisti- cal parameters of the components and their proportions in the population by means of a sample of unclassified observations. Later, the observations can be classified using the parameter estimates. In discriminant analy- sis, the basic problem 1$ to assign a given observation to one of two or more classes on the basis of the value of this observation. The procedure requires reference samples with known group membership. Ordinary cluster analysis attempts to partition the data into homogeneous subgroups without consider- ing a statistical model and assuming that the subgroups are distinct and do not overlap. Therefore, mixture analysis seems preferable for dealing with taxonomic problems in which (1) the statistical distribution of traits 1$ used for discrimination, (2) overlapping variation is present, and (3) reference samples are not available. In these situations, mixture analy- sis is a reasonable alternative provided that the number of variables (and therefore pa- rameters that need estimation) does not de- mand an inordinate sample size. When the number of variables is high, such methods as principal component analysis can be used to 38 ACUÑA & MUÑOZ reduce the dimensions of the variability to a smaller number of meaningful and indepen- dent variables. However, a general discussion of the methodological implications of mixture anal- ysis is necessary. In particular, application of mixture analysis to the problem of species discrimination entails a risk of misinterpreta- tion. The coexistence of populations leads, indeed, to mixtures of distributions, but other phenomena may be responsible for the pres- ence of mixed distributions in populations. Mendelian segregation, sexual dimorphism, disruptive selection, and multiple environ- mental reaction are examples of potential sources of mixed distributions. Thus, mixture analysis, like other more popular statistical techniques, should be not be considered as an alternative, but rather as an aid to tradi- tional taxonomic methods. ACKNOWLEDGMENTS We are very grateful to J. D. Bermúdez (De- partamento de Estadística e Investigación Operativa, Universidad de Valencia), E. Font (Departamento de Biología Animal, Univer- sidad de Valencia), and M. Sendra (Departa- mento de Estadística e Investigación Opera- tiva, Universidad de Valencia) for providing useful information and discussion of an ear- lier draft of this manuscript. We also thank R. T. Dillon and an anonymous referee for stim- ulating discussion and constructive criticism. LITERATURE CITED BACCI, G., 1947, L'inversion del sesso et il cyclo stagionale della gonade in Patella coerulea. Pub- blicazioni della Stazione Zoologica di Napoli, 21: 183-217. BUCQUOY, E., РН. DAUTZENBERG & G. Е. DOLL- FUS, 1886, Les mollusques marins du Roussil- lon, | (12). Paris. 570 pp. CERVELLA, P., L. RAMELLA, C. A. ROBOTTI 4 G. SELLA, 1988, Chromosome analysis of three species of Patella (Archaeogastropoda). Genet- ica, 77: 97-103. CHRISTIAENS, J., 1973, Révision du genre Patella (Mollusca, Gastropoda). Bulletin du Muséum Na- tional d'Histoire Naturelle, 3e série, 182: 1305- 1392. CRETELLA, M., G. SCILLITANI, Е. ТОЗСАМО, P. TURELLA & О. PICARIELLO, 1990, Comparative morphology of soft parts of Patella L., 1758 from the Bay of Naples (Gastropoda: Patellidae). Bol- letino Malacologico, 26: 205-210. DEMPSTER, А. P., М. M. LAIRD & D. В. RUBIN, 1977, Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Sta- tistical Society, Series B, 39: 1-38. DO, K. & G. J. MCLACHLAN, 1984, Estimation of mixing proportions: a case study. Applied Sta- tistics, 33: 134-140. EQUIHUA, M., 1988, Analysis of finite mixture of distributions: a statistical tool for biological clas- sification problems. Computer Applications in the Biosciences (CABIOS), 4: 435-440. EVANS, R. G., 1953, Studies on the biology of Brit- ish limpets. The genus Patella on the south coast of England. Proceedings of the Zoological Soci- ety of London, 123: 357-376. EVANS, R. G., 1958, The genus Patella on the west coast of France. Journal de Conchyliologie, 98: 126-150. EVERITT, В. S., 1984, Maximum likelihood estima- tion of the parameters in a mixture of two univari- ate normal distributions; a comparison of differ- ent algorithms. The Statistician, 33: 205-215. EVERITT, B. S. & D. J. HAND, 1981, Finite mixture distributions. Chapman & Hall, London. 143 pp. FALCONER, D. S., 1989, Introduction to quantita- tive genetics, 3rd. ed. Longman, London. 438 pp. FISCHER-PIETTE, E., 1935, Systématique et bio- géographie. Les patelles d'Europe et d’Afrique du Nord. Journal de Conchyliologie, 79: 5-66. FISCHER-PIETTE, E., 1938, The concept of spe- cies and geographical isolation in the case of North Atlantic patellas. Proceedings of the Lin- nean Society of London, 150: 268-275. FISCHER-PIETTE, E., 1948, Sur les éléments de prospérité des patelles et sur leur espécificité. Journal de Conchyliologie, 88: 45-96. FISCHER-PIETTE, E. & J. М. GAILLARD, 1959, Les patelles au long des côtes Atlantiques lbériques et Nord-Marocaines. Journal de Conchyliologie, 99: 135-200. FLURY, B. D., J. P. AIROLDI & J. P. BIBER, 1992, Gender identification of water pipits (Anthus spinoletta) using mixtures of distributions. Jour- nal of Theoretical Biology, 158: 465-480. FRETTER, \. & А. СВАНАМ, 1976, The proso- branch molluscs of Britain and Denmark. Part 1 (Pleurotomariacea, Fissurellacea and Patella- cea). Journal of Molluscan Studies, Supplement ITA HARTL, D. L. & A. G. CLARK, 1989, Principles of population genetics, 2nd. ed. Sinauer, Sunder- land. 682 pp. HASSELBLAD, V., 1969, Estimation of finite mix- tures of distributions from the exponential family. Journal of the American Statistical Association, 64: 1459-1471. LOWELL, R. B., 1984, Desiccation of intertidal lim- pets: effects of shell size, fit to substratum, and shape. Journal of Experimental Marine Biology and Ecology, 77: 197-207. MARDIA, K. V., 1970, Measures of multivariate skewness and kurtosis with applications. Bi- ometrika, 57: 519-530. MULTIVARIATE MIXTURE ANALYSIS IN PATELLIDAE 39 MCLACHLAN, С. J. 8 К. E. BASFORD, 1988, Mix- ture models. Marcel Dekker, New York. 253 pp. MUNOZ, M. A. 8 J. D. ACUNA, 1994, On the tax- onomic discrimination between Patella aspera Róding and P. caerulea Linnaeus (Gastropoda: Patellidae) using conchological traits. Journal of Conchology, 35: 37-43. PEARSON, K., 1894, Contribution to the mathe- matical theory of evolution. Philosophical Trans- actions of the Royal Society of London, Series A, 185: 71-110. POWELL, A. W. B., 1973, The patellid limpets of the world (Patellidae). Indo-Pacific Mollusca, 3: 75-206. SEGAL, E., 1956, Adaptive differences in water- holding capacity in an intertidal gastropod. Ecol- оду, 37: 174-178. SELLA, G., С. A. ROBOTTI & V. BIGLIONE, 1989, Evolutionary divergence between sympatric species of Mediterranean Patella: electro- phoretic analysis of proteins. Atti Associazione Genetica Italiana, 35: 333-334. SOKAL, В. В. & Е. J. ROHLF, 1981, Biometry. The principles and practice of statistics in biological research, 2nd. ed. W. H. Freeman and Co., New York. 859 pp. TITTERINGTON, D. M., А. Е. М. SMITH 4 U. E. MAKOV, 1985, Statistical analysis of finite mix- ture distributions. John Wiley & Sons, Chich- ester. 243 pp. WOLFE, J. H., 1970, Pattern clustering by multi- variate mixture analysis. Multivariate Behavioral Research, 5: 329-350. Revised Ms. accepted 17 January 1995 APPENDIX 1 Mardia’s Test of Multinormality Mardia’s test is a simple but useful test of multinormality (Mardia, 1970). If y,,..., y, are a random sample of k-dimensional data from a normal multivariate distribution of y, the sample estimators of skewness and kur- tosis, b, and b,, are n n by = 1/12 Y Y In -w’S (y, - УВ h=1 i=1 =) 62 = 1/n) [y -У)’5 * (y, -УР i=1 where y is the sample estimate of the mean vector and $ is the sample estimate of the variance/covariance matrix. Then, for large samples, we have, asymptotically, A=1/6nb; -xf {= 1/6 k(k + 1)(k + 2) Bee ran [8k(k + 2) /n] Although these distributions are only ap- proximate for moderate sample sizes, they at least indicate whether the data show any marked departure from multivariate normal- ity. APPENDIX 2 Maximum Likelihood Estimation of the Parameters of a Mixture of Multivariate Normal Distributions by Means of the EM Algorithm ИС.,..., Cy are the g components with multivariate normal distribution which contribute to the mixture in proportions ру, -.., Pg respectively, О < pj < 1 9 Ур =1 i=1 and f, (y) is the probability density function of the distribution of y ~ М, (m, S) in С, (i=1, ., 9) given by f(y) =1218,1 1 exp [-1/2(y — т)” s (y — m;)] then, considering a random sample of n k-di- mensional data y;,..., y, the density of y; is given by 9 fly) = Ур) (j=1, ..., n) i=1 and the maximum likelihood estimates of the parameters of the mixture (Everitt & Hand, 1981) are given by n В, = У wi/n dei, 2.9) (1) j=1 n j=1 n Si = > Wi (y; = mi) (y; = m;)’ /n Pi iz where 40 ACUÑA 8 MUÑOZ Equations 1 through 3 are solved itera- tively. Initial estimates of the parameters are used in the right hand side of the equations to obtain other estimates of the parameters op- erating the data. The resulting estimates are then used in the right hand side of the equa- tions to yield new estimates, and so on. The iterative procedure continues until two suc- cessive estimates of the parameters differ only by some arbitrarily small amount. As is the case with other iterative pro- cesses, with the EM algorithm there is no guarantee that the global maximum of the likelihood function will ever be found. It is possible that, depending on the initial esti- mates of the parameters, one or several local maxima might result. In these instances, the solution with the largest likelihood would be accepted. However, EM algorithm shows, compared to other algorithms, little depen- dency on the initial estimates of the parame- ters (Everitt, 1984). Note: A copy of the program written to compute the maximum likelihood estimates of the parameters of a mixture of two bivari- ate normal distributions is available from the authors upon request. This program 1$ writ- ten in Quick Basic and runs on IBM PC or compatible computers. MALACOLOGIA, 1995, 37(1): 41-52 THE LIFE CYCLE, POPULATION DYNAMICS, GROWTH AND SECONDARY PRODUCTION OF THE SNAIL VIVIPARUS CONTECTUS (MILLET) (GASTROPODA: PROSOBRANCHIA) IN THE MARSHES OF THE RIVER STRYMONAS, SERRES, MACEDONIA, NORTHERN GREECE. М. Eleutheriadis & М. Lazaridou-Dimitriadou Section of Zoology, Department of Biology, University of Thessaloniki 54006, Thessaloniki, Greece ABSTRACT The life cycle, population dynamics, growth and secondary production of the prosobranch freshwater snail Viviparus contectus were studied in the marshes of Strymonas River upstream of its entry into the artificial Lake Kerkini, Serres, Macedonia, northern Greece. Demographic analysis of the population of V. contectus revealed that two or more generations existed in the field throughout the year. The sex ratio was 1:1. Reproduction took place in the beginning of spring or at the end of autumn, depending on the water level and water temperature in the study area. Growth of newly born individuals mainly took place during spring and early summer. Von Bertalanffy's method suggested that V. contectus may live up to five years to reach ¡ts maxi- mum possible size (51 mm). Annual secondary production, calculated by Hynes' size frequency method, revealed a mean annual density of three individuals/m?, a mean standing crop (В) of 4.19 g dry body weight/(m”.year), and an annual production (P) of 13.45 + 12.9 д dry body weigh/(m?.year). Annual turnover ratio (Р/В) was 3.21 and turnover time was 113.7 days. Key words: Viviparus contectus, Gastropoda, ecology, population dynamics, production. INTRODUCTION Viviparus contectus is a prosobranch snail found in the marshes of Strymonas River (old bed of Strymonas River) upstream of its entry to Lake Kerkini, which lies 100 km from Thessaloniki in Serres, Macedonia, northern Greece. These marshes were formed in 1982 after the construction of a new bed of Stry- monas River parallel to the old one. The lake is an artificial water storage reservoir, con- structed on the site of a natural marsh and small natural lake. The area of the lake 1$ in- cluded in the list of the internationally impor- tant wetlands, edited by the United Nations at Ramsar in 1971. It supports a numerically rich, diverse wildlife (Pelecanus onocrotalus, P. crispus, Hirudo medicinalis), including V. contectus. Studies on the biology and ecology of fam- ily Viviparidae occurring in freshwater show considerable differences between genus, species and populations (Stanczykowska et al., 1971; Samochwalenco 4 Stanczykowska, 1972; Young, 1975; Bernardi et al., 1976; Browne, 1978; Vail, 1978; Jokinen et al., 1982; ВЫ & Gebhardt, 1986). Studies of these dif- ferences can provide worthwhile insights into the strategies that the species follow (Young, 1975). 41 Although the prosobranch snail Viviparus contectus (Millet) is widespread in Europe (Zhadin, 1952), little is known about its life history, apart from studies done by Samoch- walenco & Stanczykowska (1972) in Poland. К is one of the most important constituents in the food web of lake fauna and also the main intermediate host of trematode parasites. This gastropod is relatively large and is nu- merically and functionally a dominant mem- ber of the second trophic level of the macro- fauna of Lake Kerkini. Consequently, a study on the life cycle, population dynamics, growth, and secondary production of this prosobranch snail was considered valuable. MATERIALS AND METHODS The abiotic characteristics of the study area have been studied in detail, including water level, precipitation, water temperature, pH, dissolved oxygen, chloride, water hard- ness, and PO,-P, in relation to the density and growth of V. contectus (Eleutheriadis & Lazaridou-Dimitriadou, in press.). Presented here are the water level of Lake Kerkini at weekly intervals during the study period (Fig. 1), which affects the marshes around the lake 42 ELEUTHERIADIS & LAZARIDOU-DIMITRIADOU 37 36 35 34 33 Water level 32 31 0 JFMAMJJASONDIJFM AM 1989 JJASONDJFMAMJJASOND 1990 1991 Months FIG. 1. Water level at weekly intervals of artificial Lake Kerkini during the years 1989, 1990 & 1991. where this snail lives, monthly water temper- atures (Fig. 2) at the marshes, and monthly changes of the concentration of dissolved oxygen (Fig. 3), at the marshes, which seem to affect the onset of hibernation and repro- ductive period of this species. The study started in October 1989 and lasted for two years. Data from October 1989 to August 1991 were used for the demo- graphic analysis of the populations of V. con- tectus, which dominate the benthic, littoral and sublittoral fauna. Samples were taken using random coordinates (Lewis & Taylor, 1972) by use of a net “drag” based on the Doulkeite sampler (illustrated in Lammote & Bourlière, 1971) with a 1-m° surface, at ap- proximately monthly intervals throughout the year. Vegetation, debris and snails were hand washed through 1-mm mesh sieves. Snails were sexed (males have an enlarged right tentacle used as a copulatory organ), mea- sured and then returned as near as possible to their initial places. Measurements of the distance between the apex of the shell and the farthest point on the aperture (shell height) were made using vernier calipers. In addition to the shell height (H) of each specimen, the largest shell diam- eter (D) and the perpendicular diameters of the peristome (d & a) were also measured to the nearest 0.1 mm. The total number of samples were deter- mined so that the percentage error (Table |) was not more than 25% (Elliott, 1971). Spatial distribution of the snails in the hab- itat was examined by using Taylor's (1961) power law. The parameter b from Taylor's equation s* = ax” (where a = constant, $2 = variance, х = mean number of snails found in a sample unit, a and b = constants) was used as an index of dispersion. Parameter b is fairly constant and characterizes a species (Southwood, 1966); it is independent of the total number of samples and the total num- ber of animals in the samples but 15 affected by quadrat size (Elliott, 1971). The class interval of the monthly size-fre- quency histograms of shell height (H) was 3 mm, as determined by Goulden's method (Cancela da Fonseca, 1965). Cohorts were separated using probability paper (Harding, 1949). This method was valid because the modes of the age classes were separated by at least 2.5 standard deviations (Grant, 1989). Although some age classes had less than 50 individuals, the modal values were consistent from month to month, which confirms that the modes were real and not the result of sam- pling variation. This method has been used for demographic analyses of the populations of other molluscs (Hughes, 1970; Léveque, 1972; Daguzan, 1975; Lazaridou-Dimitriadou, 1981; Lazaridou-Dimitriadou 8 Kattoulas, 1985; Staikou et al., 1988, 1990, 1991). LIFE CYCLE OF VIVIPARUS CONTECTUS 43 1990 FIG. 2. Monthly water temperatures at the marshes of Strymonas River from July 1989 to April 1992. 20 > E Е > > 10 O 8 > o 2 TD 0 $ с 8 10 1989 FIG. 3. Monthly concentrations of dissolved охудеп (D.O.) at the marshes of Strymonas River from July 1989 to April 1992. For the study of absolute growth, data from the modal distribution of V. contectus were used (e.g., the growth of one age class was followed by the growth of the same age class the following month taking into consid- eration that time intervals always had to be equal). For determination of the theoretical growth curve, Bertalanffy's (1938) equation was employed: H, = Ha, [1 e KE where H, = the largest shell height at age t, Hmax = the asymptotical maximum possible largest shell height, К = growth rate coeffi- cient, t = time in months, and t, = hypothet- ical time when Н is equal to “zero.” The coefficient К and H,„.. were deter- mined according to Walford’s (1946) method. Hmax is the intersection point of the growth curve H,,, = f(H,) and the line drawn at 45° through the zero point. The coefficient К 1$ equal to —loga* 2,30259 (where a = the slope of Walford’s line). For the determination of the date of birth of an age class on the time axis, it is possible to use a secondary origin (Г = 0) corresponding to the smallest snails found in the biotope during the study period of a species (in our case was H = 5.5 mm), assuming that all the small snails of this spe- cies have been captured with the same size and that all age classes follow the same growth laws. Consequently, it is possible to draw the theoretical growth curve of shell height in relation to time from the first capture Hy = Hmax [1 — e 9], If the shell height at the moment of birth is known from labora- tory data (in our case was H = 3.5 mm), the axes may be changed taking as origin birth (zero point) (to = t’-t”). So Bertanlaffy’s equation becomes H, = Hina, [1-е “* ‘J and the life span of the studied species until H max Can be estimated. For the study of relative growth, the mor- phometric criteria of shell diameter (D) in re- lation to the perpendicular diameters of the peristome (d & a) were used from all the an- imals sampled during 1990 (Table |. Mayrat's method (1965 a, b) was used to compare the ’ 44 ELEUTHERIADIS & LAZARIDOU-DIMITRIADOU growth of the shell diameter (D) in relation to the peristome surface (P.) between immature and mature male and female snails. Because the peristome was almost an ellipse, its sur- face area (P.) could be calculated by the for- mula (Rey = 3.14: Ча Annual production in 1990 was calculated by the size-frequency method, because it has an important advantage in that single co- horts need not be identified to calculate pro- duction (Krueger & Martin, 1980), although it may produce an overestimate (Waters 4 Crawford, 1973). This method has been used for the determination of gastropod produc- tion, because it gives similar results to those obtained by Russell-Hunter's (1970) method. Both methods have been compared for B. graeca (Eleutheriadis 8 Lazaridou-Dimitria- dou, submitted). The formula as modified by Benke (1979) and Krueger & Martin (1980) can be written as: Z (nn) (VW Win)" P=a + ia.) - 365/CPI, where Р = the mean annual production in mg, a = number of size classes, п; = number of snails at the size class | in number, Un; = variance of ñ,, W, = mean individual dry body weight+mean dry shell of organic material (in mg), and CPI = cohort production interval in days. For the determination of the variance (U) of P the following formula was used: — 22] С? И. U(P)=a*| GF U(n:)+ Z (G¡-G;¡ +) U(n;) ¡NA j + (W,—Ga-1)2U(na) |: (365/CPI)?, where G = geometric mean of weight of pairs of successive size classes. For the determi- nation of n; and Un;, data from the population dynamics of V. contectus were used and snails collected during 1990 were grouped into 15 size classes of 3-mm intervals. To determine dry body weight and the dry organic shell material, 60 snails representing all size classes were brought to the labora- tory in April 1990, where their shell height (H) was measured in mm and they were dried in vacuo at room temperature; dry body weight (W) in mg was measured one week later. Dry organic shell material was also estimated af- ter successive treatments of immature and mature shells (mature: H > 17 mm and imma- ture: H < 17 mm, according to the study of relative growth) with 5N HCl, as described in Staikou et al (1988). Size classes >18 mm contained two females and two males (ex- cept for the last size class, which contained only females), in the sex ratio 1:1 as in nature, and in the size classes from 3 to 18 mm 20 snails were examined (in total 60 snails). On March 1991, some females were re- turned to the laboratory for observations on birth. There the mean number of new-born snails released per female was determined. We also measured the mean shell height of new-born snails. The differentiation of the genitalia in relation to the age of the snails was also studied under a stereoscope. The maturation of sectioned female and male go- nads in relation to the age of the snails were studied under a light microscope. RESULTS Aspects of the Biology The snails were sexually mature when the largest shell height (H) exceeded 17 mm, as verified by following gonad maturation. Ma- turity was attained three months after birth if the new-born appeared in early spring, but eight months if they appeared in autumn. The sex ratio in the three consecutive years of study (1989, 1990, 1991) was equal to 1:1. In 1989, new-born snails appeared in October when water level was high (Fig. 1) and the water temperature was > 15°C (Fig. 2). In the same season in October 1990, because wa- ter level was low (Fig. 1) and water tempera- ture < 15°C (Fig. 2), females carried fully formed embryos through the winter and gave birth during spring 1991. No females smaller than 16 mm shell height contained develop- ing embryos. The mean number of new-born snails released per female was 16 + 8.8 (n = 8) and the new-born snails had a mean shell height 5.15 + 0.58mm (n = 61). Fully devel- oped young were found throughout the year in the brood sacs of the adult females. Fe- males attained larger sizes because of their greater longevity. The snails in this popula- tion were active during spring, summer and until the end of autumn. At the end of autumn and during winter, snails were buried in the mud of deeper water (1-2 т). The population began to move into shallower waters by early НЕЕ CYCLE OF VIVIPARUS CONTECTUS 45 TABLE 1. Viviparus contectus population density in the marshes of Strymonas River from July 1989 to August 1991 (п: number of samples; X: mean number of animals/m?; s: standard error). 1989 6/7 21/7 2/8 17/8 20/9 2/10 19/10 n 40 31 38 33 22 28 40 x 0:15 1.52 1.26 112 2.45 1253 0.77 $ 0.26 0.38 0.25 0.29 0.50 0.34 0.17 Percentage error D 34.6 25:2 20.0 26.3 20.3 22.1 22.4 1990 22/4 13/5 26/5 10/6 29/6 21/7 4/8 29/8 13/9 п 28 24 24 24 24 20 24 16 16 x 1.64 РУ 215 3.29 2.62 12 2.91 215 2.68 5 0.48 0.45 0.43 0.67 0.61 0.34 0.47 0.6 0.53 Percentage error D 29.7 16.7 Wes 20.1 23.4 28 16.5 21.6 19.7 1990 1991 24/9 22/10 7/11 24/3 21/4 16/5 3/6 26/6 п 16 22 24 14 8 16 15 21 x 2:5 1.59 1.41 4.78 8.75 3 2 2 Ss 0.74 0.48 0:37 1.02 2:23 0.6 0.47 0.32 Percentage error D 28.9 30.5 25.8 21.4 25:5 19.2 22.8 16.1 1991 12/7 1/8 24/8 n abe) 11 14 x 2.4 1.3 2.43 $ 0.52 0.56 0.83 Percentage error D 21.4 32 34 spring and occupied the highest 50 cm of water. Migration to deeper waters occurred again between October and November, so that over-wintering of all members of the population occurred in water of 1-2 m depth. Population Density and Spatial Distribution Population density fluctuated during the study period, either from month to month or from year to year (Table 1). Low values ap- peared in July and in the end of autumn each year, whereas high values were recorded in March and April in 1991. The mean popula- tion density of V. contectus during the study period 1989-1991 was 2.84 + 2 (mean + stan- dard deviation) snails m ?. The spatial distribution of V. contectus was found to be contagious, because parameter b of Taylor’s power law was equal to 1.43 ($2 17189): Demographic Analysis of the Population of V. contectus The analysis of size frequency histograms (Fig. 4) from October 1989 to April 1991 with probability paper showed the following (Fig. 5): (a) Two cohorts were found in the habitat throughout the year; a third was added after the release of newly born individuals. (b) In 1989, the newly born individuals ap- peared in mid-autumn and in 1991 the new- born appeared in the beginning of spring. (c) Increased growth rate for newly born young occurred during spring and in the be- ginning of summer. Growth was continuous until the end of autumn. (d) One year after the new-born appeared, the largest shell height was about 28 тт; two years later, the snails were about 35 mm (March 1991) and the third year about 37-40 mm (Gas). 46 ELEUTHERIADIS 8 LAZARIDOU-DIMITRIADOU 307 % 10/89 20 10 О 6 12 18 24 30 36 42 % 4/90 40 20 0 O 6 12 18 24 30 36 42 205 % 5/90 10 0 0 6 12 18 24 30 36 42 59] 2 6/90 20 10 0 6 12 18 24 30 36 42 30- + 7/90 20 10 0 6 12 18 24 30 36 42 304 + 8/90 20 10 0 6 12 18 24 30 36 42 307 0, 9/90 20 10 0 6 12 18 24 30 36 42 SU ca 11/90 20 10 O 6 12 18 24 30 36 42 407% 3/91 30 20 10 O 6 12 18 24 30 36 42 60 + % 4/91 40 20 O 6 12 18 24 30 36 42 401 % 5/91 20 O 6 12 18 24 30 36 42 SO Tg 6/91 20 10 0 6 12 18 24 30 36 42 407 + 8/91 30 20 10 O 6 12 18 24 30 36 42 D (mm) FIG. 4. Size frequency histograms of Viviparus contectus in the marshes of Strymonas River from October 1989 to August 1991. H(mm) Hibernation Hibernation 10 Months 11 Reproductive period ' 3/91 4 5 6 7 30% 55% FIG. 5. Population analysis of the populations of Viviparus contectus at marshes of Strymonas River from October 1989 to August 1991. Percentages denote the contribution of each cohort to the total population. (G88 to G91 indicate when a generation started and when it ended). Dotted lines represents a decrease in shell height (H) because of the death of the largest individuals. LIFE CYCLE OF VIVIPARUS CONTECTUS 47 TABLE 2. Estimation of statistical parameters of the population of Viviparus contectus [where a,b: constants, r: coefficient correlation, N: number of snails, D: the mean shell diameter, Ps: 1/10 d - а, (а, а = the perpendicular diameters of the peristome, о: standard deviation)] from Teissier's regressions. Data Entire sample a+o 0.511 =0.037 logb +0 0.657 + 0.047 г? 0.957 logD + o 1.304 + 0.137 logPs +o 1.266 + 0.268 М 819 Relative Growth The study of the relative growth of D in relation to Р. (for practical reasons we used P. as 1/10 : d a) showed a positive correlation between D and P, (r? = 0.957, п = 819) (Fig. 6). Knowing that gonad differentiation was com- plete (first appearance of spermatozoa and mature oocytes) when the largest shell diam- eter reached 13.5 mm (corresponding to 17 mm shell height), it was decided to examine whether relative growth rate was the same in the two size groups, that is those with D< 13.5 mm and those with D > 13.5 mm. A logarith- mic transformation was applied to the data because the coefficient of correlation was higher than for raw data (0.957 and 0.950 re- spectively). A statistical difference (P < 0.01) was found between the slopes of the two re- gression lines using Mayrat's method (1965а, b) (Table 2). The intersection point of the two regression lines, corresponded to D = 14.9 mm, near to the size that sexual maturity was attained (Fig. 6). No statistical difference was found between the regression lines of mature female and male snails. Absolute Growth Knowing that after maturity growth rate was the same between male and female V. con- tectus, we decided to study absolute growth in all snails. H,,,,., which represents the inter- section point of Walford equation of a straight line (H,,, = 0.941H, + 3.087) and the diagonal H,=H,,,, was 52.3 тт. By using the slope of the line (a = 0.941), which showed the growth rate of the animals, the coefficient К was са|- culated as 0.06. Knowing the minimum H of the measured snails in the biotope during the study period (H = 5.5 mm), the growth of the snail shell height was calculated by Bertalan- ffy's equation H, = 52.3 [1 — е %:08('+1.84)] Because it was known from laboratory data immature snails mature snails 0.421 + 0.012 0.556 + 0.0055 0.735 + 0.008 0.594 + 0.007 0.937 0.932 1.019 + 0.10 1.333 + 0.103 0.673 + 0.235 1.326 + 0.186 79 744 that the minimum H of new-born snails was about 3.5 mm and their age t”, the age from zero point t, = (Г — t”) was calculated. By starting the curve at birth (Zero point), when snails had their smallest H (equal to 3.5 mm), the theoretical growth curve of H in rela- tion to age was calculated: H, = 52.3 [1-е °`95+1-17)]. From this curve (Fig. 7), it was found that V. contectus may live up to five years before reaching its maximum size ac- cording to Von Bertalanffy’s equation. Secondary Production The calculations of the size-frequency method are listed in Table 3. The mean bio- mass of each size class 15 expressed in dry weight. Applying Benke's correction, values of n (mean annual density), B (mean annual crop) and P (annual production) were calculated to be 3 individuals/m?*, 4.19 g dry body weight/ п?.уеаг and 13.45 + 12.9 g dry body weight/ m*.year respectively. The annual turnover ra- tio P/B was 3.21 and the turnover time was 113.7 days. DISCUSSION Populations of Viviparidae from various habitats appear to differ in a number of life history traits. Intraspecific and interspecific differences exist in size of new-born snails, number of broods/year, size of largest males and females, time of birth, and life span of females (Chaberlain, 1958; Samochwalenko 8 Stanczykowska, 1972; Young, 1975; Ber- nardi et al., 1976; Browne, 1978; Vail, 1978; Jokinen et al., 1982; Buckley, 1986; НЫ & Gebhardt, 1986). Viviparus contectus is iter- oparous and viviparous. Birth begins in mid autumn or in the beginning of spring. Spring 48 ELEUTHERIADIS 8 LAZARIDOU-DIMITRIADOU -.2 .2 .4 .6 .8 1 5 1.2 1.4 1.6 1.8 Ps=10.9 mm FIG. 6. Relative growth of the shell diameter in relation to Ps of shell peristome in the whole population and in mature and immature snails of Viviparus contectus (Teissier's regression lines). 55 4 mm Shell height Months FIG. 7. Theoretical growth curve of Viviparus contectus. reproduction 1$ also reported for other spe- cies ofthe Viviparidae by Van Cleave & Cham- bers (from Vail, 1978), Van Cleave & Altringe (from Vail, 1978), Chaberlain (1958), Fretter & Graham (1962), Young (1975), Bernardi et al. (1976), Browne (1978), Vail (1978) and Jokinen et al. (1982), possibly because in this season the environmental conditions favour survivor- ship and rapid growth of the new-born snails. In contrast to most other molluscs, fecundity of У. contectus is low, reflecting the conse- quences of viviparity, as has also been re- ported for V. georgianus by Browne (1978). The number of new-born snails released per female is variable. This number in the genus Viviparus ranges from 2.5 to 90 snails. In Poland, V. contectus released 4.5-10.2 new-born snails per female (Samochwalenco 8 Stanczykwoska, 1972), whereas in our study, the number of newborn snails was 16 + 8.8 (n = 8). Selection probably drives female toward a larger size than males as a conse- quence of the cost of viviparity. The repro- ductive output of smaller females is not only limited bioenergetically, but smaller size places severe physical constraints on the LIFE CYCLE OF VIVIPARUS CONTECTUS 49 TABLE 3. Calculation of production of Viviparus contectus by the size-frequency method. Annual production based on sets of samples from April 1990 to April 1991 (where п; = number of snails at the size class j in number; Un, = variance of п; W = mean individual dry body weight+mean dry shell of organic matter (in mg); G, = geometric mean of weight of pairs of successive size classes; В = mean standing crop or population biomass in mg; P = annual production in mg; P/B = annual turnover ratio; a = number of size classes; CPI = cohort production interval. Class — range n,/m? Un; N, NA W, (mg) 3-6 0.42 0.0224 — 0.04 0.042 6-9 0.45 0.6475 0.22 0.100 9-12 0.23 3.7573 0.19 0.238 12-15 0.04 0.0157 0.01 0.406 15-18 0.06 0.0023 — 0.06 0.575 18-21 0.12 0.0196 —0.01 0.742 21-24 013 0.0191 — 0.01 0.910 24-27 0.14 0.0047 — 0.02 1.078 27-30 0.16 0.0469 — 0.03 1.376 30-33 0.19 0.0240 = 0.12 1.870 33-36 0:31 0.0975 —0.09 2.381 36-39 0.39 0.0554 0.14 2.780 39-42 0.26 0.0897 0.18 3.390 42-45 0.08 0.0185 0.05 3.970 45-48 0.02 0.0013 0.02 4.490 365 days 3 (B) Pp __G [njWi] (N, ñ,,:)G) (WW...) (mg m?) (mg m?) 64.8 ir — 2.5 154.3 45.4 34.7 310.9 54.6 58.1 483.2 17.3 —7.1 653.2 33.0 — 39.9 821.7 87.9 —12.0 990.4 12.1 Or 1217.9 153:0 — 18.8 1604.1 216.5 —4.9 2087.8 351.3 =250.3 2545.6 717.4 219.1 3069.9 1094.9 420.8 3668.6 870.4 659.1 4222.0 306.0 225.9 4490.0 105.9 105.9 4192.1mg/m? 897 P =а.Р”-365/СР! = 15:897-365/365 = 13455 mgm * or 13.455 gm ? in 365 days. U(P) = UN,(G;G,1)* - (865/CPI)? - a? = 41730400 Confidence limits of P = P + 2-[U(P)9*] = 13.455 + 12.92 P/B = 13.455/4.192 = 3.21 Turnover time = 113.7 days number of embryos that can be maintained in the uterus if discrete embryo size units are to be maintained. By contrast, males following the strategy of attempting to mate with as many females as possible would have only limited time for feeding, resulting in slow growth and a much shorter life span than fe- males. This has also been reported for V. georgianus by Browne (1978). Viviparus contectus in Lake Kerkini under- take distinct seasonal migrations into deeper water in November following the drop in marsh temperatures, and they migrate back into shallow water in early spring. Although other species of Viviparidae appear to mi- grate before a decrease in temperature (Stanczykowska & Magnin, from Jokinen et al., 1982), the decline in temperature appears to trigger a migration of V. contectus into deeper water in the marshes, bringing the snails from summer feeding areas into the hibernation area of 2 m depth. Snails migrate to avoid areas of low temperatures (Skoog, 1971; Horst & Costa, 1975; Vincent et al., 1981) and unfavourable ecological condi- tions in the surface water (Coulet 8 Alfaro- Tejera, 1985). Other factors that provoke mi- gration might be seasonal habitat changes (Lilly, 1953; Jokinen, 1985), changes in water level (Skoog, 1971), food availability (Russell- Hunter, 1953), or water currents that flow faster as vegetation dies back (Lilly, 1953; Fretter 4 Graham, 1962; Boss et al., 1984; Lodge et al., 1987). Additionally, snails mi- grate to areas of macrophytic vegetation as a preferred habitat for the release of new-born snails or egg deposition, as shown for Amni- cola limosa by Horst & Costa (1975). The re- cruitment of newly born snails to the popula- tion during the breeding season in March and April 1991 was the main reason for the rise in population density in these months. By con- trast, the low numbers found at the end of autumn were probably due to migration to the undersampled marsh bottom. The de- crease after the breeding season was prob- ably due to mortality of the new-born snails. According to Taylor's law, the spatial distri- bution of V. contectus was contagious, and this behaviour, as Bovbjerg (1965), Duch (1976) and Brown (1979) have demonstrated, is characteristic of freshwater snails that tend 50 ELEUTHERIADIS & LAZARIDOU-DIMITRIADOU to aggregate on filamentous algae and large diatom clusters. Bithynia graeca in Lake Ker- kini also shows a contagious distribution (Eleutheriadis 8 Lazaridou-Dimitriadou, sub- mitted). Growth of V. contectus stopped at the end of autumn and in winter for about 4 months. Growth pauses have also been re- ported for V. georgianus by Jokinen et al. (1982) and Buckley (1986) in New York. In contrast, Young (1975) for V. viviparus in En- gland, Bernardi et al. (1976) for V. ater in Italy, and Browne (1978) for V. georgianus in the USA noted that growth was continuous throughout the year, although growth rate varied seasonally. Rapid growth during spring may be due to the favourable water temperatures (Fig. 2) and oxygen concentra- tions (Fig. 3) that prevailed during this period in the Strymonas marshes. Water level and water temperature seem to influence the on- set of birth. Fully developed young were found throughout the year in the brood sacs of the adult females, so it appears that adult snails choose the most favourable environ- mental conditions for the release of young snails. The fact that both internal changes in this species concerning the maturation of the gonads correspond to external morphomet- ric changes in the shell agrees with results reported for other prosobranch species of the family Littorinidae (Daguzan, 1975), for Monodonta lineata (Daguzan, 1991), and for B. graeca (Eleutheriadis 4 Lazaridou-Dimitri- adou, submitted). Viviparus 1$ a genus show- ing considerable variation in the duration of life. Viviparus contectus has a multiyear life cycle, and there are some differences in the pattern of the cycle within the genus. Such European investigators as Samochwalenco & Stanczykwoska (1972) reported that V. con- tectus and V. viviparus lives up to 4 years; Young (1975) reported that V. viviparus lives up to 2 years, and Ribi & Gebhardt (1986) reported that V. ater lives 5 to 8 years. A Ca- nadian population of V. malleatus was re- ported to live for 5 years (Stanczykwoska et al., 1971). In the USA, populations of V. geor- gianus were reported to have a 2 to 3 year life-span (Van Cleave & Lederer, from Vail, 1978; Browne, 1978; Jokinen et al., 1982) and 4 years (Buckley, 1986). These studies indicate that Viviparus life span may be a trait determined by habitat factors. The compari- sons in production among freshwater snails were done by use the turnover times, be- cause productivity rates are very sensitive to difficulties of assessing environmental space in calculating densities for biomass (Russell- Hunter & Buckley, 1983). The turnover time for V. contectus is short and reflects relatively high levels of productivity. The value of 113.7 days is low compared with values from other freshwater prosobranchs. For four popula- tions of V. georgianus, the turnover time was 477, 510, 393 and 421 days (Browne, 1978), for Bithynia tentaculata 337 and 314 days for females and males respectively (Tashiro, from Russell-Hunter & Buckley, 1983), and for three population of Leptoxis carinata 372, 311 and 303 days (Aldridge, 1982). All these populations are at least biennial. Low values were reported by Lévéque (1973) in Lake Chad for three annual prosobranchs snails including 74.5-117 days for Melanoides tu- berculata, 101-126 days for Cleopatra buli- moides, 56-70.2 days for Bellamya unicolor, and in Lake Kerkini 77.9 days for an annual population of В. graeca (Eleutheriadis 4 Lazaridou-Dimitriadou, submitted). The dif- ference in turnover time of V. contectus com- pared to other biennial and triennial popula- tions must be due to lack of competition with other freshwater snails; in two years” study, only small numbers of Valvata piscinalis and Lymnaea stagnalis were recorded. There 1$ also little direct human intervention on these marshes. Viviparus contectus, being an A-strategist, seems to be able to profit the favourable con- ditions even when the climate shows a sud- den difference and its parameters do not follow the characteristic cycle of the Mediter- ranean climate at the marshes of Strymonas River. ACKNOWLEDGEMENTS We would like to thank Dr. G. Dussart from Canterbury Christ Church College for his comments on an earlier draft of this paper and the two anonymous referees of this journal. LITERATURE CITED ALDRIDGE, D. W., 1982, Reproduction tactics in relation to life-cycle bioenergetics in three natu- ral populations of the freshwater snail, Leptoxis carinata. Ecology, 63: 196-208. BENKE, A. C., 1979, A modification of the Hynes method for estimating secondary production LIFE CYCLE OF VIVIPARUS CONTECTUS 3 with particular significance for multivoltine pop- ulations. Limnology and Oceanography, 24: 168- AC. BERNARDI, R., K. W. RAVERA & B. OREGIONI, 1976, Demographic structure and biometric characteristics of Viviparus ater Cristofori and Jan (Gastropoda: Prosobranchia) from Lake Al- serio (N. Italy). Journal of Molluscan Studies, 42: 310-318. BERTALANFFY, L. VON, 1938, A quantitative the- ory of organic growth. Human Biology, 10: 181- 213; BOSS, N.G., T. G. LAMAN 4 Н. О. BLANKESPOOR, 1984, Dispersal movements of four species of pulmonate and operculate snails in Douglas Lake, Michigan. The Nautilus, 98: 80-83. BOVBJERG, В. V., 1965, Feeding and dispersal in the snail Stagnicola reflexa (Basommatophora: Lymnaeidae). Malacologia, 2: 199-207. BROWN, K. M., 1979, The adaptive demography of four freshwater pulmonate snails. Evolution, 33: 417-432. BROWNE, R. A., 1978, Growth, mortality, fecun- dity, biomass and productivity of four lake pop- ulations of the prosobranch snail Viviparus geor- gianus. Ecology, 59: 742-750. BUCKLEY, D. E., 1986, Bioenegetics of age-re- lated versus size-related reproductive tactics in female Viviparus georgianus. Biological Journal of the Linnean Society, 27: 293-309. CANCELA DA FONSECA, J., 1965, L’outil statis- tique en biologie du sol. |. Distributions de fre- quences et tests de signification. Revue d’Eco- logie et de Biologie du Sol, 2: 299-332. CHAMBERLAIN, М. A., 1958, Life history studies of Campeloma decisum. The Nautilus, 72: 22-29. COULET, M. & L. ALFARO-TEJERA, 1985, The life cycle and ecology of the freshwater Planorbidae Armiger crista (L.). Journal of Molluscan Studies, 51: 35=51. DAGUZAN, J., 1975, Contribution à l'écologie des Littorinidae (mollusques gastéropodes proso- branches); recherches écophysiologiques chez quatre espèces: Littorina neritoides (L.), Littorina saxatilis (Olivi), Littorina littorea (L.) et Littorina littoralis (L.). Thèse d'Etat Univ. de Rennes Fac. des Sc. Biol. DAGUZAN, J., 1991, Recherhes sur la croissance et l'écologie de Monodonta lineata (da Costa) (Gastropoda, Prosobranchia, Trochidae) vivant sur littoral atlantique armoricain. Cahiers de Bi- ologie Marine, 32: 3-22. DUCH, T. M., 1976, Aspects of the feeding habits of Viviparus georgianus. The Nautilus, 90: 7-10. ELEUTHERIADIS N. & M. LAZARIDOU-DIMITRIA- DOU, in press, Density and growth of freshwater prosobranch snails (Bithynia graeca 8 Viviparus contectus) in relation to water chemistry. Journal of Molluscan Studies. ELEUTHERIADIS, М. & М. LAZARIDOU-DIMITRIA- DOU, submitted, The life cycle, population dy- namics, growth and secondary production of the snail Bithynia graeca (Westerlund) (Gastropoda: Prosobranch) in lake Kerkini (Serres, Macedonia, Northern Greece). ELLIOTT, J. M., 1971, Some methods for the sta- tistical analysis of samples of benthic inverte- brates. (Ph.D.). Freshwater Biological Associa- tion, 25: 1-144. FRETTER, V. & A. GRAHAM, 1962, British proso- branch molluscs; their functional anatomy and ecology. Ray Society London. GRANT, A., 1989, The use of graphical methods to estimate demographic parameters. Journal of the Marine Biological Association of the United Kingdom, 69: 367-371. HARDING, J. P., 1949, The use of probability paper for the graphical analysis of polymodal frequency distribution. Journal of the Marine Biological As- sociation of the United Kingdom, 28: 141-153. HORST, T. J. & R. R. COSTA, 1975, Seasonal mi- gration and density patterns of the freshwater snail, Amnicola limosa. The Nautilus, 89: 56-59. HUGHES, В. N., 1970, Population dynamics of the bivalve Scrobicularia plana (Da Costa) on an in- tertidal wind-flat in north Wales. Journal of Ani- mal Ecology, 39: 333-356. JOKINEN, E. H., 1985, Comparative life history patterns within a littoral zone snail community. Internationale Vereinigung fur Theoretische und Angewandte Limnologie Verhandlungen, 22: 3292-3299. JOKINEN, E. H., J. GUERETTE & R. W. KORT- MANN, 1982, The natural history of an ovovivi- parus snail Viviparus georgianus (Lea), in a soft water eutrophic lake. Freshwater Invertebrate Bi- ology, 1(4): 2-17. KRUEGER, C. C. & F. B. MARTIN, 1980, Compu- tation of confidence intervals for the size-fre- quency (Hynes) method of estimating secondary production. Limnology and Oceanography, 25: 773-777. LAMOTTE, M. & F. BOURLIERE, 1971, Problemes а’ écologie |’ echantillonnage des peuplements animaux des milieux aquatiges. Publication sous les auspices du Comite Francais du Programme Biologique International. LAZARIDOU-DIMITRIADOU, M., 1981, Contribu- tion a l'étude biologique et écologique des es- cargots Cernuella virgata (Da Costa) et Xeropicta arenosa Ziegler (Gastreropoda, Pulmonata) vi- vant sur les microdunes de Potidea, Chalkidiki (Grèce du Nord). У Convegno Nationale della So- cieta Malacologica Italiana, 73-83. LAZARIDOU-DIMITRIADOU, M. & M. KATTOULAS, 1985, Contribution a l’etude biologique et écologique d’ Eobania vermiculata (Gastero- pode, Hélicidae). Biologia Galohellénica, 10: 131- 137. LEWIS, T. & L. В. TAYLOR, 1972, Introduction to experimental ecology. Academic Press Inc. (Lon- don). LÉVÉQUE, C., 1972, Mollusques benthiques du Lac Tchad: écologie, production et bilans éner- gétiques. Thése doctorat des Sciences Natural Paris, VI, C.N.R.S. 52 ELEUTHERIADIS 8 LAZARIDOU-DIMITRIADOU LÉVÊQUE, C., 1973, Billans énergétiques des рор- ulations naturelles de mollusques benthiques du Lac Tchad. Cahiers O.R.S.T.O.M., Sér. Hydro- biol., 7: 151-165. LILLY, М. M., 1953, The mode of life and the struc- ture and functioning of the reproductive ducts of B. tentaculata. M. Sc. Thesis, University of Lon- don. LODGE, №. М, К М. BROWNE} 5. NP: KLOSIEWSKI, В. A. STEIN, А. Р. СОМСН, В. К. LEATHERS & С. BRONMARK, 1987, Distribution of freshwater snails: spatial scale and the relative importance of physicochemical and biotic fac- tors. American Malacological Bulletin, 5: 73-84. MAYRAT, C. F., 1965a, Calcul et comparison des droites d’ allometrie de croissance. Comptes Rendus de la Société de la Biology, 159: 2139- 2143. MAYRAT, С. F., 1965b, La loi d’ allometriete des allometries. Comptes Rendus de la Société de la Biology, 159: 544-548. RIBI, С. & М. GEBHARDT, 1986, Age specific fe- cundity and size of offspring in the prosobranch snail Viviparus ater. Oecologia (Berlin), 71: 18- 24. RUSSELL-HUNTER, W., 1953, On migration of Lymnaea peregra (Müller) on the shores of Loch Lomond. Proceedings of the Royal Society of Edinburgh, 65(7)(В)(1): 84-106. RUSSELL-HUNTER, W. D. & D. E. BUCKLEY, 1983, Actuarial bioenergetics of nonmarine mol- luscan productivity. In the Mollusca, 6: 463-503, ed. by W. D. RUSSELL-HUNTER. Academic Press, Inc., New York. SAMOCHALENKO, T. 8 A. STANCZYKOWSKA, 1972, Fertility differentiation of two species of Viviparidae (Viviparus fasciatus Mull. and V. vivi- parus |.) in some environments. Ecologia Polska, 20: 479-492. SKOOG, G., 1971, Variation in the distribution of Theodoxus fluviatilis on stony localities in the northern Baltic proper. Thalassia Yugoslavia, 7: 363-372. SOUTHWOOD, Т. В. E., 1966. Ecological methods. Chapman and Hall, London. STAIKOU, A. & М. LAZARIDOU-DIMITRIADOU, 1990, Aspects of the life cycle, population dy- namics, growth and secondary production of the snail Monacha cartusiana (Múller, 1774) (Gas- {города Pulmonata) in Greece. Malacologia, 31: 353-362. STAIKOU, A. & М. LAZARIDOU-DIMITRIADOU, 1991, The life cycle, population dynamics, growth and secondary production of the snail Xeropicta arenosa (Múller, 1774) (Gastropoda: Pulmonata) in northern Greece. Zoological Jour- nal of the Linnean Society, 101: 179-188. STAIKOU, A., M. LAZARIDOU-DIMITRIADOU & N. FARMAKIS, 1988, Aspects of the life cycle, pop- ulation dynamics, growth and secondary pro- duction of the edible snail Helix lucorum Lin- naeus, 1758 (Gastropoda, Pulmonata) in Greece. Journal of Molluscan Studies, 54: 139-155. STAIKOU, A., M. LAZARIDOU-DIMITRIADOU 4 E. PANA, 1990, The life cycle, population dynam- ics, growth and secondary production of the snail Bradybaena fruticum Ziegler (Gastropoda: Pulmonata) in northern Greece. Journal of Mol- luscan Studies, 56: 137-146. STANCZYKOWSKA, A., E. MAGNIN £ A. DU- MOUCHEL, 1971, Etude de trois populations de Viviparus malleatus (Reeve) (Gastropoda: Proso- branchia) de la région de Montréal. I. Crois- sance, fécondité, biomasse et production an- nuelle. Canadian Journal of Zoology, 49: 1431- 1441. TAYLOR, L. R., 1961, Aggregation, variance and the mean. Nature, London 189: 732-735. VAIL, V. A., 1978, Seasonal reproductive patterns in 3 viviparid gastropods. Malacologia, 17: 73- 79. VINCENT, B., G. VAILLANCOURT & М. HARVEY, 1981, Cycle de developpement, croissance, ef- fectifs, biomasse et production de В. tentaculata L. (Gastropoda: Prosobranchia) dans le Saint- Laurent (Québec). Canadian Journal of Zoology, 59: 1237-1250. WALFORD, L.A., 1946, A new graphic method of describing the growth of animals. Biological Bul- letin, 90: 141-147. WATERS, Т. Е. & G. W. CRAWFORD, 1973, Annual production of a stream mayfly population. A comparison of methods. Limnology and Ocean- ography, 18: 286-296. YOUNG, М. R., 1975, The life cycles of six species of freshwater molluscs in Worcester-Birming- ham Canal. Proceedings of the Malacological Society of London 41: 533-548. ZHADIN, V. |., 1952, Mollusks of fresh and brackish waters of the U. S. S. R. Akademiia Nauk Soyuza Soviets. Sotsial. Respublic, 46: 80-196. Revised ms accepted 28 Nov 1994 MALACOLOGIA, 1995, 37(1): 53-68 COMPARATIVE LIFE CYCLE AND GROWTH OF TWO FRESHWATER GASTROPOD SPECIES, PLANORBARIUS CORNEUS (L.) AND PLANORBIS PLANORBIS (L.). Katerine Costil & Jacques Daguzan Laboratoire de Zoologie et Ecophysiologie (L. А. INRA & URA 1853), Université de Rennes |, Campus de Beaulieu, Av. du General Leclerc, 35042 Rennes Cedex, France ABSTRACT The shell growth and life cycle of two natural populations of Planorbarius corneus and Planorbis planorbis were studied for 16 and 17 months, respectively. These planorbid popu- lations lived in ponds where they exibited great annual density variations, which could be related to environmental conditions, snail behaviour and, especially, their life cycle. The pop- ulation of Planorbis planorbis showed a life cycle with two breeding periods per year, and consequently, two generations. Whereas both generations were semelparous and lived for 11 months, they grew with different seasonal patterns. But of the two species, Planorbarius cor- neus growth seemed more influenced by the climate. For the population of this species, an annual life cycle with two generations was observed in 1987, whereas only one spring gener- ation seemed to be present in 1986 and 1988. In 1987, the vernal generation, showing an estimated maximum life span of 18-20.5 months, was potentially iteroparous, and the estival- autumnal generation living 15-21 months was basically semelparous. Life-cycle variations of natural populations of freshwater snails are reviewed. Key words: shell growth rate, life cycle, Planorbidae, population dynamics, reproduction. INTRODUCTION Life cycles of freshwater snails are of the highest importance for the study of their bi- ology because they can be viewed as a syn- thesis of the main life-history traits of snail populations. Such studies have been per- formed for planorbid temperate species (Rus- sell-Hunter, 1961; Boerger, 1975; Young, 1975; Eversole, 1978; Alfaro Tejera, 1982; Lodge & Kelly, 1985; Byrne et al., 1989; Ca- quet, 1993). Planorbis planorbis has been re- ported to show an annual life cycle (De Coster 8 Persoone, 1970; Dussart, 1979), whereas the life cycle of P. corneus is almost unknown. In England, a population of P. corneus was studied for only four months (Berrie, 1963). Moreover, freshwater snails present great in- traspecific variation in life-history patterns (Calow, 1978; Brown, 1985), and this plastic- ity has been considered to be of fundamental selective value in their evolution. (Russell- Hunter, 1961). An observed life history is the result of both long-term evolutionary forces, and the more immediate responses of an or- ganism to the environment in which it is and has been living (Begon et al., 1990). The present data constitute part of a set of results on the biology and the ecology of Ar- 53 morican freshwater gastropods (Costil, 1993, 1994; Costil 8 Daguzan, 1994). The aim of the present study was to determine the life cy- cles of the populations of P. corneus and P. planorbis in order to compare them with life- history strategies of other freshwater snail populations. Special attention was given to snail shell growth. MATERIAL AND METHODS Study Sites The population of P. corneus was studied for 17 months in pond Le Boulet, located 35 km north of Rennes, east Brittany (48°20’N, 1°38’W). The population of P. planorbis living in pond La Musse, 35 km _ southeast of Rennes (48°00’N, 1°58’W), was sampled for 16 months. These continuous studies were preceded and followed by instantaneous ob- servations made in 1986 and in 1989. The sampling areas in the large ponds (Le Boulet: 120 ha.; La Musse: 49 ha.) were the shallowest (depth of 20-50 cm) and the most eutrophic parts. Vegetation contained nu- merous aquatic and amphibious macro- phytes (Clement, 1986; Costil, 1993), and the bottom was covered by mud. Table 1 sum- marizes the results of water chemical analy- 54 COSTIL & DAGUZAN TABLE 1. Water chemistry in the two study ponds; the unit is mg/l except for conductivity (uS/cm) and pH (pH unit); S.M. = matter in suspension; the analyses were performed in May 1988. Total S.M. CON СЕР” Mg?* pH Total iron mangenese Le Boulet 80 160 13:8 6.0 ES SU 1.4 La Musse 150 160 12.6 4.3 8.7 MOS ES №№. N-NO, N-NH,* Total P-PO,? Total Or 10m nitrogen (035 phosphorus Le Boulet 1253 1.4 9.0 21 2.1 5.6 107 La Musse 0.39 1.3 8.0 1.86 0.9 2:4 11078 ses. Because of human interferences, the water level was subject to great variations (up to + 180 cm at the study site in La Musse). Planorbarius corneus lived with 10 coexisting gastropod species (4 planorbids, 3 lym- naeids, 1 physid and 2 prosobranchs), whereas P. planorbis shared its habitat with 7 species (3 planorbids, 3 lymnaeids and 1 physid). In both ponds, birds and fishes that include molluscs in their diets were present, e.g. Anas platyrhynchos L. and Aythya fuligula L., Rutilus rutilus L. and Abramis brama L. Sampling Samples were taken fortnightly from May 5, 1987, to October 18, 1988, for P. corneus, and from April 14, 1987, to August 16, 1988, for P. planorbis. Sampling was sometimes impossible because of ice cover or flooding. Random sampling took place in a limited zone of 50 т? surface. In mid September 1987 in pond La Musse, the deterioration of the environment leading to the disappear- ance of the studied population obliged us to transfer the study site from the south to the north bank, a few meters away. The snails, however, from both banks belonged to the same population. The snails were collected using a pond net (1 mm mesh) and a wood frame delimiting a sampling surface of one square meter. This sampling method was chosen especially be- cause it respects the environment and allows long-term study. On every sampling date, we tried to collect at least 100 individuals. The number of sampled т? depended on the snail density and environmental conditions, and fluctuated from 3 to 10 for P. planorbis, and from 3 to 15 for P. corneus. The number of collected snails sometimes reached 380 (P. planorbis) and 389 (P. corneus), but sam- pling difficulties (particularly ice cover or flooding) did not always allow us to collect 100 snails. In the field, living snails and empty shells were counted (рег m*), and measured (maximum shell diameter) using a vernier cal- liper (to the nearest of 0.05 mm). The snails were then returned to the study site, while the empty shells were removed. Egg capsules were sampled to determine periods of repro- duction. Sample Treatment The results were recored in 1-mm size classes. The number of individuals in each size class was expressed as a percentage of total number, establishing size-frequency histograms. The cohorts, represented by gaussian structures, were discriminated us- ing Bhattacharya's (1967) method. On every sampling date, a mean shell diameter (+ stan- dard error) was calculated for each cohort. Each cohort was assumed to be representa- tive of one generation resulting from a spe- cific reproductive period. The empty shells were attributed to a particular cohort if pos- sible. The observed life span of a cohort corre- sponded to the time from recruitment until death. For the “real life span,” we had to add the time necessary for the snails to reach the recruitment size. This time was known if the water temperature in the field and the growth results of the laboratory experiments were taken into consideration (Costil, 1994). Von Bertalanffy’s model was used to de- scribe the growth of the vernal and estival or autumnal cohorts (Von Bertalanffy, 1938). It is given by the equation: D, = De [t= ee where D, is the diameter at time t; О„.„„, the asymptotic maximum diameter; K, a growth max PLANORBID LIFE CYCLES 55 rate coefficient; t,, the hypothetical time when О is equal to “zero” (minus the duration of the egg stage). К and L,,,,, are determined according to Ford-Walford's method by the slope and intercept on the x-axis of the re- gression line of growth rate on size, respec- tively. RESULTS Population Density The mean densities of P. corneus and P. planorbis were respectively 20.0 (S.E. = 3.5) and 37.8 (S.E. = 5.9) individuals/m”, but great annual variations were noted for both species (Fig. 1). The maximum density was the same (127 individuals/m?), but it was observed at different dates (P. planorbis: 07-07-87; P. corneus: 20-10-87). Life Cycle Two reproduction periods were found, a vernal for both species and an autumnal or estival-autumnal (16 weeks) for P. planorbis and P. corneus, respectively (Table 2). For P. planorbis, autumnal reproduction was only observed on two sampling dates. On Octo- ber 13, 1987, the site was inaccessible and perhaps egg capsules were present (repro- duction period for a minimum of two, possi- bly up to four weeks). The percentages of empty shells in relation to sampled live snails varied from 0.4 to 26.8% for P. corneus, and from 0.0 to 32.4% for P. planorbis (Fig. 2). They were especially high in July 1987 for the first species, and in August 1987 and in April 1988 for the latter. For two consecutive years, the maximum shell diameter of the population of P. corneus was observed in May (20.9 mm on 19-05-87 and 18.9 mm on 31-05-88), whereas the min- ‘imum (6.9 mm) was noted in March 1988. For P. planorbis, the minimum and maximum shell diameters were 4.2 mm (23-05-87) and 11.8 mm (26-05-87), respectively. These changes in mean diameter of the population reflected the succession of cohorts that ap- peared, grew and died. Each cohort repre- sented a generation stant after a particular reproductive period and descended from the same parent generation. At the beginning of the study, the popula- tion of P. corneus corresponded to cohort G1, the mean diameter of which was 18.9 mm (Fig. 3). One month later, the diameter reached 23.7 mm and the snails reproduced. Some individuals of the cohort G1 lived until September 22, and some until mid Novem- ber. The very large shells (25-28 mm) were thicker and more eroded than the large shells (22-23 тт). Cohort G2 appeared on June 2, 1987, and the largest snails laid egg capsules in summer 1987. The individuals of G2 over- wintered and, after May, except for observa- tions made on September 20, 1988 (24 indi- viduals in a restricted area), the number of survivors was low. Cohort G3 appeared on August 25, 1987, continued to be alive until the end of the study. Individuals of G3, but also survivors of G2, which had laid egg cap- sules in previous summer, were the parents of cohort G4. This cohort, recruited on May 31, seemed not to reproduce, although three small snails were collected on October 4, probably belonging to cohort G5. In the case of P. planorbis, the population showing a mean diameter of 8.6 mm on April 14, 1987, was mainly made up of one cohort (G1) (Fig. 3). This cohort coexisted with a more aged cohort (GO; mean size of 15.7 mm), which was no longer found in the mid- dle of June. G1 retained a relatively large number of snails until June 9, when the first newly hatched snails of cohort G2 appeared. After ovipositing in autumn (mean diameter of 8.2 mm), the individuals of G2 did not disap- pear, and a lot of them were collected on March 15. Snails of G3 were seen for the first time on October 27, 1987. However, consid- ering their relative large size and the period when egg capsules were observed, it seemed that recruitment had already oc- curred on October 13. (On this date, we could not sample because of flooding.) In spring 1988, individuals of G3 gave the co- hort G4. From July 5, 1988, the snails of co- hort G3 contributed less than 10% to the to- tal population. For P. corneus, the observed life span of cohort G2 was 70 weeks (about 16 months). The number of these snails was sometimes very low or null on certain sampling dates. Moreover, the individuals of cohorts G1 and G2 were very similar in size at the beginning of May for two consecutive years. So, G2 ap- peared to be homologous to G1, and its members could survive six weeks after Oc- tober 4, as was the case of G1 in 1987. The observed life span of G2 was estimated to be from 70 to 76 weeks. If the correction factor and the size of G2 on the recruitment date 56 COSTIL 8 DAGUZAN 180 A 120 Population density (snails/m2) 00 © 8 as = a 8 as as A =F == AS AAs E am as Aa 5-6. 7_8 9 101112 1234 5 6 7 в Study dates (days & months) 180 160 120 Population density (snails/m2) 5 Study dates (days € months) FIG. 1. Changes in population density with time: Mean + S.E.. A: Planorbarius corneus (from April 1987 to October 1988); B: Planorbis planorbis (from May 1987 to August 1988). were taken into consideration, the “real life snails belonging to cohort G3 measured 24 span” was then from 79 to 85 weeks (18 to mm, that is, 2 mm lower than the maximum 20.5 months). At the end of the study, the size noticed for G2. Their growth curve did PLANORBID LIFE CYCLES 57 30 А 25 e a 20 © <= on > 15 dd о. Е 4 æ 10 о o г 5 0 8 as = an an as as as =8 == A Aas = m= ae Ss as = 5. “6 2711869, 10: 11121234 5 6 7 8 9 10 Study dates (days & months) = B No. of empty shells (%) S м Ро. Study dates (days & months) FIG. 2. Variation of the number of collected empty shells in relation to the number of live snails collected. A: Planorbarius corneus; B: Planorbis planorbis. 58 COSTIL & DAGUZAN TABLE 2. Features of reproduction period (В. P.) in Planorbarius corneus and Planorbis planorbis; the water temperatures, recorded on every study date at midday, correspond to minimum and maximum values during the reproduction periods. Water Start of the End of the Duration temperature Rip: R.P. (weeks) (°C) Planorbarius 19-05-87 16-06-87 4 16-20 corneus 28-07-87 17-11-87 16 9.5-28 17-05-88 28-06-88 6 15-22 Planorbis 28-04-87 23-06-87 8 12-22 planorbis 15-09-87 29-09-87 2 14.5-18.5 (13-10-87?) (4?) 11-05-88 19-07-88 10 15-29 not show a plateau, typical of an asymptotic growth. Moreover, the individuals of G3 were not found in samples collected during spring 1989. The observed life span was estimated to be from 60 to 80 weeks. The “real life span” varied then from 67 to 87 weeks (15 to 21 months). For P. planorbis, the cohort G2 showed an observed life span of 44 weeks and a “real life span” of 48 weeks. The population study was stopped in mid-August 1988, when five individuals of G3 were collected. If we con- sider G3 as the homologue of G1, we could say that G3 lived until the beginning of Sep- tember 1988. So, the cohorts G2 and G3 had the same observed (44 weeks) and “real life span” (48 weeks or 11 months). Growth of the Vernal and Estival-Autumnal Cohorts In P. corneus, strongly different growth patterns were noted, as evidenced by the co- horts (Fig. 4). In the first sample, the mean diameter of individuals of cohort G2 was 6.4 mm. After 6 weeks, it reached 53% of the maximum size, that 1$, 13.8 mm, and the maximum growth rate attained 2.1 mm/week (Fig. 5). A second peak (1.5 mm/week) was observed for planorbids one year old. On the other hand, from December 15 to April 19 (age: 29-45 weeks), shell growth gain did not exceed 2.3% (rates below 0.035 mm/week). For the growth of G3 cohort in the field, two periods appeared to be particularly favour- able: at the time of cohort recruitment (rates higher than 1.05 mm/week), and from April 19 to mid-June (age: 29-45 weeks, mean rate of 1.2 mm/week, and so the size changed from 13.6 to 20.1 mm). During the winter and in July and August, shell growth was slow or null. The mean growth rate was slightly higher for G3 than for G2 (0.35; S.E. = 0.09 and 0.29; S.E. = 0.08). Nevertheless, in- dividuals of both cohorts reached a size of about 25 mm at the end of their life. The growth rate coefficient of Von Berta- lanffy's model (К) was equal to 0.046. For the entire population of P. corneus, the shell growth in terms of individual age was given by the relation: О; = 37.31 1 = et SIE time unit of 14 days. The theoretical diameter for maximum age estimated in the field (32.4 for snails 20 months old) was higher than maximum size observed in the field (28 mm) (Table 3). In the case of P. planorbis, the snails of cohort G2 reached 65% of their maximum size eight weeks after recruitment; maximum rates were of 0.85 and 0.80 mm/week (Figs. 4, 6). Thirty-two weeks were necessary to at- tain 89% of their size before growth started again and the death of the cohort. For the cohort G3, a long recruitment period influ- encing the growth rates was observed. The snails reached the two-thirds of the maxi- mum size after 26 weeks, and two peaks (more than 0.7 mm/week) were noted in spring and in summer 1988. The mean growth rates of the cohorts G2 and G3 were 0.21 (S.E. = 0.05) and 0.19 (S:E- = 0:05), mn week, respectively. The theoretical shell growth of the popula- tion in relation to snail age was given by the following relation: D, = 13.4 1 = & о: time unit of 14 days. According to Von Bertalanffy's model, the maximum mean diameters of G2 (11.3 mm) and G3 (11.6 mm) conferred on these cohorts a life span of 11-12 months (Table 3). PLANORBID LIFE CYCLES 59 Egg laying NS Diameter (mm) A 30 25 20 15 10 1987 Study dates (days & months) Egg laying ER Diameter (mm) B 15 10 1987 1988 Study dates (days & months) FIG. 3. Changes in mean shell diameter of Planorbarius corneus (A) and Planorbis planorbis (B) (+ S.E.) with time (from May and April 1987 to October and August 1988, respectively), at the Le Boulet and La Musse ponds respectively. The numerals correspond to percentages denoting the contribution of each cohort to the total population. For P. corneus, G1, G2 and G4 represent the spring generations, and G3 the estival- autumnal one. For P. planorbis, the spring generations are GO, G2 and G4, whereas the autumnal gener- ations are G1 and G3. COSTIL 8 DAGUZAN © o —#— Cohort G2 =--0-- Cohort G3 A A / Pooog “voce ps $ Diameter/maximum diameter (%) in © 0 5 10 15 20 25 30 35 40 45 50 55 6 6. Observed age (weeks) Diameter/maximum diameter (%) 0 5 10 15 20 25 30 33 40 45 Observed age (weeks) FIG. 4. Relation between the snail age and the percentage of maximum diameter attained by the individuals belonging to the vernal (C2) and the estival or automnal (C3) cohorts. A: Planorbarius corneus; B: Planorbis planorbis. PLANORBID LIFE CYCLES 61 Cohort G2 Water temperature Growth rates (°С) (mm/week) 26 Sp Su f Ww Sp Su 22 a PAT а [20 | 1,8 1,6 1,4 152 1,0 0.5 10 15 20 25 30 35 40 45.50 55: 60 65 70 Observed age (weeks) Cohort G3 Water temperature Growth rates (°C) (mm/week) 26 Su f Ww Sp Su f 722 247 % : 2,0 22 a+ gt 1,8 h , у 1,6 1,4 1,2 1,0 RN Осьеэх 19:15 20725.30 355.04 60 Observed age (weeks) 0,8 0,6 0,4 02 0,0 FIG. 5. Variation of the water temperature at Le Boulet pond, and changes in growth rates for the vernal (C2) and the automnal (C3) cohorts of Planorbarius corneus. Sp = spring; Su = summer; F = fall; W = winter. The value corresponding to C3 snails at the age of 13 weeks is put in brackets because it is an abnormal value due to sampling difficulties. 62 COSTIL 8 DAGUZAN TABLE 3. Theoretical relation between the age and the shell diameter (D) of Planorbarius corneus and Planorbis planorbis according to Von Bertalanffy's model, H = hatching. Р. Age (months) H 1 3 6 12 18 20 corneus D (mm) 172 4.0 10.3 11742 26.3 31.2 32.4 22 Age (months) H 1 3 6 10 12 planorbis D (mm) 0.8 РЕЙ 5:5 8.4 10.6 11.4 DISCUSSION nately declines as the water level increases. Density Variations Variation in population structure appears to be a major factor explaining variation in snail density. In our study, the recruitment of newly hatched snails had a strong effect on the den- sity, whereas the cohort disappearance af- fected the density more or less progressively. We do not know exactly how long the resi- dence time of the empty shells in the studied ponds is and when the corresponding snails were dead. Such a topic has not been exam- ined much. Hunter (1990) has reported that adverse water chemistry (low pH and/or low calcium concentration) has a much greater effect on shell dissolution than does presence orabsence of periostracum. Moreover, empty shells do not have an absolute value, because they could accumulate in certain areas. Nev- ertheless, the percentages of empty shells and above all their variation helped us to de- termine the life cycle of both populations, and to explain the variations in their density. For P. planorbis, great numbers of empty shells and low densities could be attributed to death of cohorts. In summer, the disappearance of G1 and G3 took place, whereas in spring the dis- appearance of G2 occurred. On July 14 and 28, 1987, the density of P. corneus was low and the percentage of empty shells high (20- 27%). The reproductive effort of the largest individuals seemed to be responsible for their death, whereas the high mortality of small in- dividuals might be dueto predation. The pred- ators seem to be small invertebrates unable to eat whole shells, such as insects or leeches, because a lot of empty shells were pierced. Nevertheless, the major predation on fresh- water snails 15 exerted by vertebrates (Lodge et al., 1987). According to Eisenberg (1966), the whole predation could lead to the death of 93% to 98% of the young Lymnaea elodes (Say). The snails were sampled using a pond net which respects the environment, but the ef- ficiency of this sampling method unfortu- This is especially illustrated in winter for P. corneus, and on June 7, 1988, when the den- sity of P. planorbis population was less than 6 snails/m” at the flooded study site. On the other hand, during dry periods, snails could be seen on the mud and easily collected. Such was the case in March 1988 for P. planorbis and in September 1988 for P. corneus. Snail behaviour also influenced their density. When the environmental conditions became unfa- vourable in winter, we observed some indi- viduals of P. corneus sinking into substratum crevices or even into mud. It was then very difficult to collect them. Such behaviour has been reported for Lymnaea catascopium (Say) (Pinel Alloul, 1978). At the same time, some individuals of P. corneus were seen moving on the ice cover on December 1 and 15, 1987. Nevertheless, the extreme climatic conditions (cold and hot) appeared particularly unfavour- able for the population density of P. corneus. In summer, lack of dissolved oxygen into wa- ter was a problem, as was the case in late July 1987, when chlorophytes pullulated at the study site and no individual of P. planorbis was sampled. Moreover, Eisenberg (1966) emphasized the importance of food for the regulation of density in a natural population of L. elodes. The death and then the regrowth of submerged macrophytes led to changes in the densities of those snails inhabiting them (Lodge et al., 1987). Lymnaea peregra and Valvata piscinalis (Müller) exhibited low resis- tance to habitat disturbance (decline of sub- mersed macrophytes), but high adjustment following the disturbance. The maximum den- sity for P. corneus (127 snails/m*) was noted three days after a storm occurred in Brittany with winds of 140 km/h. The accumulation of individuals of Anisus rotundatus (Роге) in lim- ited areas attributed to wind was also re- ported by Marazanof (1970). Life Cycle Different reproductive periods were deter- mined according to the cohort to which the PLANORBID LIFE CYCLES 63 Cohort G2 Water temperature Growth rates (°С) (тт/меек) 28 Su f Ww Sp 1,0 0 3 10 15 20: “25 30 35 40 45 Observed age (weeks) Cohort G3 Water temperature Growth rates (°С) (mm/week) 287 № Ww Sp : su 1,0 26 N: 24 A 6 > Pa 0,8 20 | 18 16 | oP 14 12 10 0,4 8 6 0,2 4 /\ 0 0,0 0 5 10 15 20:7 25 307 35 40 45 Observed age (weeks) FIG. 6. Variation of the water temperature at La Musse pond, and changes in growth rates for the vernal (C2) and the automnal (C3) cohorts of Planorbis planorbis. Sp = spring; Su = summer; F = fall; W = winter. 64 COSTIL & DAGUZAN majority of the parents belonged. In the stud- ied population of P. corneus, the egg-laying period began in May and ended in Novem- ber, but we accepted two periods. The esti- val-autumnal reproduction period lasted for 16 weeks (the parents belonged to the cohort G2). However, at the end of this period, we could not exclude an egg-laying by the larg- est individuals of G3 estival cohort (13 mm on November 17, 1987). Moreover, no egg cap- sules were found in August or in October 1987. Perhaps some were in the field, but it was very difficult to see them among the dense vegetation. The length of the egg-lay- ing period was due to the inter-individual growth variation. In L. catascopium, all the snails belonging to spring generation were not mature at the same time, and they con- tinued to reproduce as long as the water tem- perature allowed it (Pinel Alloul 4 Magnin, 1979). According to Berrie (1963), the repro- duction period of P. corneus was short. On the other hand, De Coster & Persoone (1970) sampled very small P. corneus (0-2 mm) from April to September. Concerning P. planorbis, the spring reproduction periods were longer than the autumnal period. In northwest En- gland, P. planorbis also showed two egg-lay- ing periods, one in February and the other in July (Dussart, 1979). According to Precht (1936), P. corneus did not reproduce in autumn or in winter, and an endogenous rhythm had to be suggested, be- cause the snails brought in the laboratory at 23°C did not lay eggs either. This author added that P. corneus began to reproduce when the temperature reached 12°C. The minimum threshold temperature for the re- production of freshwater snails appears to be between 7°C and 12°C (Boerger, 1975; Dun- can, 1975; Eversole, 1978). Three egg cap- sules were laid by P. corneus reared at 8°C (De Wit, 1955). In our study, the reproduction period of both species began when temper- ature reached 15-16°C, which was above the minimum threshold. In spite of suitable tem- peratures, the genital organs could be imma- ture. Such a result was observed by Russell- Hunter (1961) in Gyraulus albus (Müller). Moreover, the freshwater snails could be more sensitive to temperature variations than to absolute temperatures. The main life-cycle patterns encountered in natural populations of freshwater snails have been described by Russell-Hunter (1961, 1978) and reviewed by Calow (1978). During our study, P. planorbis showed a type C as defined by Calow (1978), that is, an annual life cycle with reproduction periods occurring in spring and in autumn with parent generation replacement. The snails of the vernal cohort G2 did not survive until the next reproductive period. However, these individuals coexisted with the snails of the autumnal cohort for 28 months. Both cohorts corresponding to two generations were semelparous, and their life span was estimated at 11 months. A life span of 12-13 months and an annual life cycle have been reported respectively by De Coster & Persoone (1970) and Dussart (1979). How- ever, the second author described a life cycle with egg-laying periods completely brought forward in comparison with our results (re- cruitment of newly hatched snails in February and July instead of early May and September). Such an annual life cycle with two generations per year occurred in the following species: Lymnaea peregra (Müller) (Russell-Hunter, 1961; Lambert, 1990), Lymnaea palustris (Müller) (Hunter, 1975), Physa fontinalis (L.) (De Wit, 1955; Russell-Hunter, 1961), Heli- soma trivolvis (Say) (Eversole, 1978); Bathy- omphalus contortus (L.) (De Coster & Per- soone, 1970), Armiger crista (L.) (Alfaro Tejera, 1982), and A. rotundatus (Marazanof, 1970). For the latter planorbid species, Caquet (1993) observed a semelparous life cycle with a maximum life span apparently reaching 17-18 months. From a four-month study, an annual life cy- cle was attributed to P. corneus (Berrie, 1963). According to different authors, the life span of this species in the field was 13 months (Hilbert, 1911), 2 or 3 years (Boycott, 1936; Boettger, 1944), and 4 years (Frómming, 1956). Russell-Hunter (1978) has reported that biennal life-cycle patterns involve such larger pulmonates as P. corneus and Lym- naea stagnalis (L.) in higher latitudes or more oligotrophic waters (1.е., poorer temperature or trophic conditions). Like for P. planorbis, the life cycle of P. corneus in 1987 was annual showing two generations. The estival-autum- nal generation was not replaced by the new generation (type B). So, other differences were also observed between the two species: longer life span of both generations and an annual life cycle close to a biennial (18-20.5 months for G2; 15-21 months for G3, these life spans could be very similar); a longer sec- ond reproductive period; a spring cohort probably iteroparous to a certain extent and an estival-autumnal cohort probably semel- parous (the second egg laying period proba- bly missing in 1986 and 1988). At the begin- ning of May 1987, only one cohort was PLANORBID LIFE CYCLES 65 observed (G1). Its mean diameter was the homologous of the spring cohort G2 diame- ter at the same date in 1988. In autumn 1988, no egg capsule was present at the sampling site, although three small snails were col- lected on October 4 (mean diameter of 6.1 mm). The second reproduction period might occur every other year or might be far less predictable, depending on annual climatic variations. In the pond Le Boulet in 1988, the water temperature was apparently not as high as in 1987, and shell growth was slower than in 1987. The climatic hypothesis was put forward by Vincent & Harvey (1985) to explain both types of life cycle (short and long) en- countered in a Canadian population of Bithyn- ia tentaculata (L.). The number of genera- tions per year can also be related to the length of dry periods (Duncan, 1959; Ma- razanof, 1970). Lymnaea catascopium exhib- ited an annual life cycle with one generation in hard or medium waters, and two genera- tions in soft waters (Pinel Alloul, 1978). In L. peregra, snails from the exposed habitats matured earlier and put more effort into it than snails from the sheltered habitats; these differences in growth and reproduction could be explained in terms of differences in selec- tion pressure between habitats of varying exposure (Calow, 1981). Individuals of H. trivolvis from the eutrophic sites grew faster and exhibited an annual life cycle, whereas those from the mesotrophic site grew slowly and lived about two years, breeding in both summers (Eversole, 1978). From reciprocal transfer experiments, Brown et al. (1985) concluded: “genetic divergence among pop- ulations of L. elodes explained a compara- tively small proportion of the variation in life histories in comparison with proximal factors like habitat productivity. Nevertheless, snails from the vernal pond always grew more slowly, matured at a smaller shell length, and had higher fecundities than other popula- tions. These smaller differences may still be important over evolutionary time scales.” These examples show that life-cycle varia- tions can be adaptative. The adaptative plas- ticity allows the populations to have maxi- mum productivity in given conditions and then to compensate for bad years (Huben- dick, 1958; Russell-Hunter, 1961). Snail Growth The shell growth of freshwater pulmonates is continuous until death (indeterminate growth). For both studied species, the growth differed according to time when a cohort had been recruited. The shell growth did not only depend on age of the animals, but the ob- served differences could be due to various physiological potentialities that remain to be studied. For B. tentaculata, growth rates fluc- tuated from 0.17 mm/week to 0.80 mm/week with the seasonal trophic conditions (Pinel Alloul 4 Magnin, 1971). In L. peregra and L. palustris, the vast majority of the interpopu- lation variation of shell growth rate appeared to result from non-genetic ecophenotypic en- vironmental influences and in particular from habitat productivity (Byrne et al., 1989). Cli- mate seemed very important for growth, es- pecially in the case of P. corneus. Spring is very favourable to the shell growth of P. corneus and other species (Ma- razanof, 1970; Pinel Alloul, 1978). In June 1987, a marking experiment (with painting) allowed us to observe that some shells had grown from 0.75 of whorl to 1.75 whorls in two weeks (diameters attaining respectively 10.1 and 18.1 mm). In spring, the shell growth of young Lymnaea humilis (Say) was very fast reaching 7% per day (MacCraw, 1961). Dur- ing autumn, the studied snails grew with in- termediate rates. In summer (July-August) and in winter (December to March), a lower rate of shell growth was noticed for L. pere- gra (Lambert, 1990) and P. corneus inhabit- ing Le Boulet pond. In summer, the high tem- peratures (up to 26°С) monitored probably exceeded the optimum, which was close to 20°С for the growth of the two studied spe- cies reared in the laboratory (Costil, 1994). For a population of P. corneus in England, Berrie (1963) recorded a mean size increase of 4% between July and August, and in- creases of 70% and 33% respectively be- tween June-July and August-September. In our study, the growth of individuals belong- ing to cohort G2 slowed down at the same period that the snails reproduced. Concern- ing the homologous cohort G4, no summer reproduction period was observed in 1988, but growth did not slow down. These facts about the cohorts G2 and G4 suggested the energy allocation theory which, on the other hand, was not confirmed by the cohort G3 (growth slowed down after reproduction pe- riod). A slow or null shell growth during winter has been reported for many freshwater snails (Russell-Hunter, 1961; Calow, 1973; Vincent et al., 1981). Individuals of В. contortus sam- pled in winter and then brought to the labo- ratory grew fast again (Calow, 1973). Natural 66 COSTIL 8 DAGUZAN population of aquatic molluscs can exhibit degrowth (decrease in unit mass of structural proteins through time) (Russell-Hunter, 1985); according to this author, degrowth can be associated with reproductive or sea- sonal stress, an example being the overwin- ter starved individuals of Р. corneus, which lost 44% of their tissue biomass over 126 days (with 33% mortality). For the same spe- cies, Emerson (1967) has reported similar losses: 62.3% with 56% mortality after 58 days of starvation; at this time, 95% of the original polysaccharides, 49% of the proteins and 22% of the lipids were metabolized. A very omnivorous diet could be responsible for the winter shell growth observed in P. plan- orbis and also L. peregra (Russell-Hunter, 1961), Physa integra and P. gyrina (Clampitt, 1970). In the diets of both P. corneus and Planorbis carinatus (Müller) (close to P. plan- orbis), detritus formed over 85% of the food ingested and sand grains about 10% (Reav- ell, 1980). In comparison with Р. corneus, Р. planorbis seemed far less affected by the cli- matic variations. The spring cohorts (G2 and С4) grew the fastest in July and August. Their renewal of growth at the end of life could be due to the late death of the largest individu- als. According to Von Bertalanffy's model, the largest individuals of P. corneus in the field were 14 months old. However, this model does not take into account the environmental conditions and their variations. lt appears more suitable for P. planorbis than for P. cor- neus. The first species showed higher growth constant (k = 0.069) than the second (k = 0.046), which illustrates a faster develop- ment. For comparison with other freshwater snails, some values of k are given (same time unit: 14 days): 0.024 for A. crista (Alfaro Te- jera, 1982), 0.057 to 0.485 for snails from Lake Tchad (Lévêque, 1971), 0.095 for L. per- egra (Lambert, 1990). The maximum size reached by animals in- habiting in a given site depends on many fac- tors, such as environmental conditions, but also genetic makeup and parasitic infection. Before death, the cohorts G1 and G2 of P. corneus respectively measured 27.4 or 25.9 mm, and the size of the largest individuals reached 28 mm. These sizes were lower than the value that is usually given for this species (35 mm). According to the cohorts of P. plan- orbis, the mean diameter attained before disappearance fluctuated from 11.3 to 14.3 mm. The maximum diameter recorded was 16.1 mm, whereas the maximum size re- ported for this species is 20 mm. It is difficult to know to what extent the environment con- ditions influence the maximum size of the snails of a given species. In both ponds, the water mineralization was rather low, and flooding and drying occurred quite often, making the environment unstable. According to Byrne et al. (1989), the extended adult sur- vival and oviposition was a life-history trait that allows L. palustris to survive in marginal, unstable habitats. Such an observation could be made for P. corneus showing a potentially itereparous cohort and a very long annual life cycle, but not for P. planorbis. Nevertheless, for both species the spring cohort growth was rapid enough to allow a second repro- duction period per year (every year in the P. planorbis population and in 1987 for P. cor- neus). Moreover, the trophic conditions in both studied ponds appeared favourable for the coexistence of a great number of gastro- pod species. The environmental factors are of the highest importance in the life-history strategies of the freshwater snails. The great range of strategies used by these snails т- habiting various types of environment make them particularly useful for studies about ev- olutionary biology. ACKNOWLEDGMENTS We are grateful to Maria Lazaridou-Dimitri- adou for her comments on early draft of this manuscript. We also thank Stacy Payne for checking the english and Jean Luc Foulon for technical help. LITERATURE CITED ALFARO TEJERA, L., 1982, Dynamique de popu- lation de l'espèce Armiger crista (L.) (Mollusque Planorbidae). These 3°™° cycle, Université de Lyon |, 106 pp. ВЕСОМ, M., J. L. HARPER & С. В. TOWNSEND, 1990, Ecology: individuals, populations and communities. Blackwell Scientific Publications, Oxford, 944 pp. BERRIE, A. D., 1963, Life cycle of Planorbarius cor- neus (L.). Nature, 198: 805-806. BHATTACHARYA, C. G., 1967, A simple method of resolution of a distribution into gaussian compo- nents. Biometrics, 23: 115-135. BOERGER, H., 1975, A comparison of the Ше-су- cle, reproductive ecologies and size weight re- lationships of Helisoma anceps, H. campanula- PLANORBID LIFE CYCLES 67 tum and H. trivolvis (Gastropoda, Planorbidae). Canadian Journal of Zoology, 53: 1812-1824. BOETTGER, C. R., 1944, Basommatophora. Die Tierwelt der Nord and Ostsee, 9: 241-478. BOYCOTT, A. E., 1936, The habitats of the fresh- water Mollusca in Britain. Journal of Animal Ecol- ogy, 5: 116-186. BROWN, К. M., 1985, Intraspecific life history vari- ation in a pond snail: the roles of population di- vergence and phenotypic plasticity. Evolution, 39: 387-395. BROWN, K., D. DE VRIES 4 В. LEATHERS, 1985, Causes of life history variation in the freshwater snail Lymmaea elodes. Malacologia, 26:191-200. BYRNE, R., J. REYNOLDS & R. MAC MAHON, 1989, Shell growth, reproduction and life cycles of Lymnaea peregra and L. palustris (Pulmonata: Basommatophora) in oligotrophic turloughs (temporary lakes) in Ireland. Journal of Zoology, London, 217: 321-339. CALOW, P., 1973, On the regulatory nature of individual growth: some observations from freshwater snails. Journal of Zoology, 170: 415- 428. CALOW, P., 1978, The evolution of life-cycle strat- egies in freshwater gastropods. Malacologia, 17: 351-364. CALOW, P., 1981, Adaptational aspects of growth and reproduction in Lymnaea peregra from ex- posed and sheltered aquatic habitats. Malacolo- gia, 21: 5-13. CAQUET, T., 1993, Comparative life cycles, bio- mass and secondary production of three sym- patric freshwater gastropod species. Journal of Molluscan Studies, 59: 1-6. CLAMPITT, P. T., 1970, Comparative ecology of the snail Physa gyrina and Physa integra. Mala- cologia, 10: 113-151. CLEMENT, B., 1986, Typologie des zones humides de Bretagne. Recherches de bioindicateurs. Uni- versité de Rennes |, Ministère de l'Environne- ment, 129 pp. COSTIL, K., 1993, Contribution à l'écologie des mollusques gasteropodes dulcicoles Armori- cains; recherches bio-écologiques et écophysi- ologiques sur deux espèces de pulmonés Pla- norbidae: Planorbarius corneus (L.) et Planorbis planorbis (L.). Ph.D. Thesis, University of Rennes |, 421 pp. COSTIL, K, 1994, Influence of temperature on sur- vival and growth of two freshwater planorbid species, Planorbarius corneus (L.) and Planorbis planorbis (L.). Journal of Molluscan Studies, 60: 223-235. COSTIL, K 8 J. DAGUZAN, 1994, Effect of temper- ature on reproduction in Planorbarius corneus (L.) and Planorbis planorbis (L.) throughout the life span. Malacologia, 36: 79-89. DE COSTER, W. & G. PERSOONE, 1970, Ecolog- ical study of Gastropoda in a swamp in the neighbourhood of Ghent (Belgium). Hydrobiolo- gia, 36: 65-80. DE WIT, W. F., 1955, The life-cycle and some other biological details of the freshwater snail Physa fontinalis (L.). Basteria, 19: 35-73. DUNCAN, C. J., 1959, The life-cycle and ecology of the freshwater snail Physa fontinalis (L.). Jour- nal of Animal Ecology, 28: 97-117. DUNCAN, C. J., 1975, Reproduction in pulmonates. Pp. 309-366, in: V. FRETTER & J. PEAKE, eds., Pul- monates, Vol. 1, Academic Press, London. DUSSART, G. B., 1979, Life-cycles and distribu- tion of the aquatic gastropods molluscs Bithynia tentaculata (L.), Gyraulus albus (M.), Planorbis pl- anorbis (L.) and Lymnaea peregra (M.) in relation to water chemistry. Hydrobiologia, 67: 223-239. EISENBERG, ВР. M., 1966, The regulation of density in a natural population of the pond snail Lym- naea elodes. Ecology, 47: 889-906. EMERSON, D. N., 1967, Carbohydrate oriented metabolism of Planorbarius corneus (Mollusca, Planorbidae) during starvation. Comparative Biochemistry and Physiology, 22: 571-589. EVERSOLE, A., 1978, Life cycles, growth and pop- ulation bioenergetics in the snail Helisoma triv- olvis (Say). Journal of Molluscan Studies, 44: 209-222. FROMMING, E., 1956, Biologie der mitteleuropäis- chen Sússwasserschnecke. Duncker et Hum- bolt, Berlin: 74-90. HILBERT, R., 1911, Uber die neue Molluskenkunde in Altpreussen. Schriften der Onigichen Physikalisch Okonomischen Gesellschaft zu Konigsberg, 52: 267-273. HUBENDICK, B., 1958, Factors conditioning the habitat of freshwater snails. Bulletin of the World Health Organization, 18: 1072-1080. HUNTER, R. D., 1975, Growth, fecundity and bioenergetics in three populations of Lymnaea palustris in upstate New-York. Ecology, 56: 50- 63. HUNTER, ВР. D., 1990, Sheel dissolution rate in freshwater pulmonate snails: influence of perio- stracum and water chemistry. The Nautilus, 104: 87-91. LAMBERT, M. C., 1990, Contribution a la biologie et a l'écophysiologie d'un Lymnaeidae armoric- ain: Lymnaea peregra (Múller) (Mollusque, Gas- téropode, Pulmoné, Basommatophore). Ph.D. Thesis, University of Rennes |, 317 pp. LÉVÊQUE, C., 1971, Equation de Von Bertalanffy et croissance des mollusques benthiques du Lac Tchad. Cahiers de ГО.В.$.Т.О.М., Hydrobiolo- gie, 5: 263-283. LODGE, D. & P. KELLY, 1985, Habitat disturbance and the stability of freshwater gastropod popu- lations. Oecologia, 62: 111-117. LODGE, D., K. BROWN, P. KLOSIEWSKI, R. STEIN, P. COVICH, В. LEATHERS & С. BRON- MARK, 1987, Distribution of freshwater snails: spatial scale and the relative importance of physiochemical and biotic factors. American Malacological Bulletin, 5: 73-84. MacCRAW, B. M., 1961, Life history and growth of the snail Lymnaea humilis (Say). Transactions of the American Microscopical Society, 80: 16-27. 68 COSTIL 8 DAGUZAN MARAZANOF, F., 1970, Contribution à l’etude écologique des mollusques des eaux douces et saumátres de Camargue. Tome Il: Anisus rotun- datus (Poiret, 1801). Annales de Limnologie, 6: 191-213. PINEL ALLOUL, B., 1978, Ecologie des popula- tions de Lymnaea catascopium catascopium (Mollusque, Gastéropode, Pulmoné) du Lac St Louis, pres de Montréal, Québec. Verhand- lungen Internationale Vereinigung fúr Theoritis- che und Angewandte Limnologie, 20: 2412- 2426. PINEL ALLOUL, В. 8 Е. MAGNIN, 1971, Cycle vital et croissance de Bithynia tentaculata L. du Lac Saint Louis, pres de Montreal. Canadian Journal of Zoology, 49: 759-766. PINEL ALLOUL, B. & E. MAGNIN, 1979, Cycle de developpement, croissance et fecondite de cing populations de Lymnaea catascopium catascopium (Lymnaeidae) au Lac Saint-Loius, Quebec. Malacologia, 19: 87-101. PRECHT, H., 1936, Zur Kopulation und Eiablage einer Planorbiden. Zoologischer Anzeiger, 115: 80-89. REAVELL, P., 1980, A study of the diets of some British freshwater gastropods. Journal of Con- chiology, 30: 253-271. RUSSELL-HUNTER, W., 1961, Life-cycle of four freshwater snails in limited populations of Loch Lomond, with a discussion of intraspecific vari- ation. Proceedings of the Zoological Society of London, 137: 135-171. RUSSELL-HUNTER, W., 1978, Ecology of fresh- water pulmonates. Pp. 355-383, in: V. FRETTER & J. PEAKE, eds., Pulmonates, Vol. 2, Academic Press, London. RUSSELL-HUNTER, W., 1985, Physiological, eco- logical and evolutionary aspects of molluscan tissue degrowth. American Malacological Bulle- tin, 3: 213-221. VINCENT, В & М. HARVEY, 1985, Dynamique de deux populations du gasteropode Bithynia ten- taculata. Verhandlungen Internationale Vereini- gung für Theoritische und Angewandte Limnol- ogie, 22: 3288-3291. VINCENT, B., G. VAILLANCOURT & М. HARVEY, 1981, Cycle de développement, croissance, ef- fectifs, biomasse et production de Bithynia ten- taculata L. (Gastropoda : Prosobranchia) dans le Saint-Laurent. Canadian Journal of Zoology, 59: 1237-1250. VON BERTALANFFY, L., 1938, A quantitative the- ory of organic growth. Human Biology, 10: 181- 213: YOUNG, М. R., 1975, The life-cycles of six species of freshwater Molluscs in the Worcester Birming- ham Canal. Proceedings of the Malacological Society of London, 41: 533-548. Revised Ms. accepted 15 December 1994 MALACOLOGIA, 1995, 37(1): 69-110 WHEN SHELLS DO NOT TELL: 145 MILLION YEARS OF EVOLUTION IN NORTH AMERICA'S POLYGYRID LAND SNAILS, WITH A REVISION AND CONSERVATION PRIORITIES Kenneth C. Emberton Department of Malacology, Academy of Natural Sciences of Philadelphia, 1900 Benjamin Franklin Parkway, Philadelphia, Pennsylvania 19103-1195 U.S.A. ABSTRACT A phylogenetic hypothesis for the 274 known polygyrid species is presented, based on 115 anatomical, behavioral, and shell characters (39 of which are newly discovered or newly as- sessed), as corroborated in part by published allozyme data. The hypothesis differs little from that of Emberton (1994a), except for greatly increased resolution in the tribes Stenotremini and Polygyrini. A corresponding revision is presented that assigns species to 50 subgenera, 24 genera, seven tribes, three infrafamilies, and two subfamilies. All taxa in the revision, as well as unnamed clades, are defined phylogenetically by shared derived characters. Nine new subge- neric names are introduced: Neohelix (Asamiorbis), N. (Solemorbis), Triodopsis (Brooksorbis), T. (Pilsbryorbis), T. (Macmillanorbis), T. (Vagvolgyorbis), Stenotrema (Archerelix), S. (Pils- brelix), and Millerelix (Prattelix). A separate phylogenetic analysis of polygyrid subgenera was conducted based on shell morphology alone, using 71 character states in 14 characters, all new or newly assessed, including ontogenetic characters scored from shell x-rays. The resulting shell-based hypothesis had only one-fourth the phylogenetic resolution of, and showed major topological discrepan- cies from, the anatomy-behavior-shell hypothesis. Thus, at present level of knowledge, poly- gyrid shells are inadequate for reconstructing phylogenetic history, and identifications of pre- Miocene fossils should be considered dubious at best. Remaining to be evaluated, however, are shell-surface microsculptures and ultrastructural layers. Museum and field surveys discovered the closest known polygyrid convergences in sympa- try on flat, umbilicate, and tridentate shell forms: Patera laevior and Xolotrema fosteri at Hawes- ville, Harlan County, Kentucky; Appalachina sayana and Allogona profunda on Pine Moun- tain, Hancock County, Kentucky; and Inflectarius inflectus and Triodopsis fallax in Vinton County, Ohio. These provide starting points for analyses of naturally replicated experiments in evolutionary morphology, such as those already conducted on the globose shell form (Mesodon normalis and Neohelix major at numerous sites in the Southern Appalachian Moun- tains). Adult Triodopsis tridentata from which apertural barriers had been removed lost water 27% faster than controls when retracted, and 9% faster when extended. Controls were 83% more successful than barrierless snails in forming epiphragms. Epiphragms reduced the rate of water loss by 3% in controls and by 38% in barrierless snails. Ramsay's (1935) hypothesis is extended to suggest that barriers and epiphragms slow evaporation not directly, but indi- rectly by interrupting natural convection currents between the aperture and the retracted snail's body. Because 7. tridentata and other triodopsins have larger barriers in more humid climates, barriers should also serve at least one other function, such as impeding invertebrate predators. A new hypothesis is proposed for the function of extremely obstructed shell apertures: to exclude water, allowing the snails to float to safety during floods. Polygyrids have many con- vergences on apertural obstruction by barriers, some extreme examples of which are illustrated together. Based on the phylogenetic hypothesis/revision, the most urgent remaining targets for polygyrid conservation are (1) Colombia's Isla de Providencia, where deforestation threatens the ancient, relic, uniquely live-bearing, endemic Giffordius; and (2) the northern coves of Pine Mountain, Harlan County, Kentucky, which harbor North America's most diverse land-snail communities, including four simultaneous cases of polygyrid convergence in sympatry. Key words: cladistics, phylogenetic taxonomy, shell ontogeny, paleontology, fossils, con- vergence in sympatry, functional morphology, Gastropoda, Pulmonata, Stylommatophora. 69 70 ЕМВЕАТОМ INTRODUCTION The Polygyridae are an endemic North American family of land snails that are remark- able for their shell convergences in sympatry (due to iterative, punctuated evolution cou- pled with phenotypic and genetic, adaptive parallel environmental responses); their 145- million-year biogeographic history (paralleling that of, for example, plethodontid salaman- ders); and their evolution of hermaphroditic, external sperm exchange via several behav- iorally bizarre intermediates that remain ex- tant (Pilsbry, 1940; Webb, 1974; Solem, 1976; Emberton, 1981, 1988a, 1991a,b 1994a,b, 1995a,b; Asami, 1988, 1993). Polygyrids cur- rently comprise 271 nominal and three unde- scribed species (see below), with a maximum sympatric diversity of ten species reported from Pine Mountain, Harlan County, Ken- tucky, U.S.A. (Hubricht, unpublished; Ember- ton, 19950), with many more species yet to be described and discovered in Mexico and Cen- tral America (F. Thompson, pers. commun.), and probably with some undiscovered spe- cies remaining in the Pacific Northwest (T. Frest, pers. commun.). Phylogenetic hypoth- eses for the family as a whole (Webb, 1974; Pratt, 1979; Emberton, 1994a) have been to genus only, and the most recent hypothesis omitted shell characters completely and had major unresolved polytomies. Polygyrid fossil shells of the Pleistocene and Pliocene are nearly all identical with Recent species (Hubricht, 1985, and pers. commun.; Emberton & Bogan, unpublished). Miocene and later fossils are generally as- signable to genus (Pilsbry, 1940; Auffenberg 8 Portell, 1992; Roth 4 Emberton, 1994). Ear- lier polygyrid fossils, however, which are rare but presumably valuable in reconstructing the first 125 million years of polygyrid history, could be difficult to identify because of multi- convergent, punctuated-equilibrium evolu- tion (Emberton, 1994a). One way of testing the value of early fossils is to make an hon- est, concerted attempt to reconstruct poly- gyrid phylogeny based on Recent shells alone. This has not previously been at- tempted. One possible source of new and useful characters for such an analysis—and for fossil identification—is shell ontogeny as viewed in x-rays (Ramirez, 1993). For exam- ple, a recent study discovered, using x-rays, a significant difference in whorl-expansion rate between the polygyrid tribes Triodopsini and Mesodontini (Emberton, 1994a). Thus a more general x-ray survey of polygyrid shell ontogeny could be productive. Non-mimetic sympatric convergences are important as naturally replicated experiments in evolutionary morphology (Emberton, 1995a). Polygyrid non-mimetic convergences in sympatry fall into four distinct shell forms (Pilsbry, 1940) that have been called globose, flat, tridentate, and umbilicate shell-static clades (Fig. 1; Emberton, 1991b, 1994a). The globose clades show the closest conver- gences in sympatry (Solem, 1976; Emberton, 1981; Asami, 1988, 1993), of which the most extreme 1$ between Neohelix major and Ме- sodon normalis. Recent analyses ofthese two species have yielded important new insights into the ecological, genetic, and natural- selective influences on shell morphology (Emberton, 1994b, 1995a). The most extreme cases of sympatric convergence on flat, tri- dentate, and umbilicate shell forms have not been analyzed or even reported, however. Polygyrid shell evolution is further note- worthy for its convergences in extreme aper- tural obstruction by denticles and other shell structures (Fig. 2; Pilsbry, 1940; Zilch, 1959- 60). Such apertural barriers appear in many forms among numerous land-snail clades (Zilch, 1959-60). Four hypotheses have been proposed concerning the function of these barriers: (1) to deter attacks by predatory in- sects (Cook, 1895; Boettger, 1921, 1935; Solem, 1972, 1974; Falkner, 1984); (2) to re- tard evaporative water loss (Boettger, as cited by Goodfriend, 1986; Rees, 1964; Paul, 1974; Christelow, 1992); (3) to strengthen the aperture against accidental breakage (Paul, 1974); and (4) to orient the shell during crawl- ing (Paul, 1974). A fifth hypothesis, proposed here for the first time, is suggested by some polygyrid cases of extreme apertural ob- struction existing in flood-prone environ- ments (Hubricht, 1985; Emberton, 1986): that some barriers function (5) to trap air within the submerged shell, preventing drowning and enhancing gene flow by allowing the snail to float downstream. None of these hy- potheses has been adequately tested using living snails (Goodfriend, 1986; pers. ob- serv.). Also untested is the likely hypothesis that some barriers perform two or more functions simultaneously. In the polygyrid tribe Triod- opsini, there is a correlation in several lin- eages— including that of the common spe- cies Triodopsis tridentata—between greater apertural obstruction and increased environ- 71 POLYGYRID SHELL EVOLUTION ‘oIyO ‘AjUNOD UOJUIA WO4} xe/¡e4 sısdopou] pue sn398/yu/ зпивэаци! pue ‘емерч| ‘UIEYD PUBAE) шо 1/97S0/ ешелоох pue /0/188/ B/987ey ‘Ayonjuey “send oy low врипзола euoboyy pue euefes eunyoejeddy :sassauua| ‘элоЭ Sepe9 wo (зуб) ло/еш хиецоэм pue (ya) seuuou uoposayy :(woyoq о} doy) suo:]93|¡09 tuunasnu UO paseq ‘AneduuAs и! зээцеблелиоэ ||э4$ puABAjod эшедхе 150 INO} S POWAY Y HON WIEISEZ ‘| "914 72 ЕМВЕАТОМ FIG. 2. Convergent shell-apertural obstruction among various polygyrid subgenera. Species are (left to right, top row): Triodopsis (Haroldorbis) henriettae (ANSP 109734), Triiobopsis гореп (150631), Ashmunella mudgii (319743), Stenotrema (Stenotrema) unciferum (171138), Stenotrema (Stenotrema) maxillifer (170141); (bottom row): Millerelix (Millerelix) tamaulipasensis (317942), Millerelix (Prattelix) plicata (143448), Daedalochila (Upsilodon) hippocrepis (84629), Daedalochila (Daedalochila) uvulifera (10990), Inflectarius (Inflectarius) rugeli (116893). mental moisture (Vagvolgyi, 1966; Emberton, 1988a). This implies that if apertural barriers retard evaporative water loss in T. tridentata, then they should also serve at least one ad- ditional function, such as deterring predators or trapping air. Polygyrid conservation is relatively ad- vanced, with at least five rare taxa currently listed as endangered or threatened: Patera clarki nantahalae, Fumonelix archeri, F. or- estes, F. jonesianus, and Triodopsis playsay- oides. Thanks to this recognition, these taxa are receiving the protection they need to sur- vive. Phylogenetic analysis, however, can provide insights into conservation priorities beyond the rarity of individual taxa. Polygyrid conservation has never been considered in a phylogenetic context. The purposes of this paper are (1) to con- duct a phylogenetic analysis of the Polygy- ridae to the species-group/subgeneric level; to use the resulting phylogenetic hypothesis (2) to taxonomically revise the family; (3) to conduct an independent phylogenetic analy- sis based on shells alone, incorporating x-rays; to use the resulting cladogram (4) to assess the reliability of fossils for recon- structing polygyrid phylogeny; (5) to find and report the closest polygyrid convergences in sympatry on the flat, tridentate, and umbili- cate shell forms; (6) to test for retardation of evaporative water loss by shell apertural bar- riers in living Triodopsis tridentata; and (7) to assess remaining conservation priorities for the Polygyridae from a phylogenetic per- spective. MATERIALS AND METHODS General Phylogenetic Analysis Standard phylogenetic methods of charac- ter analysis were used (Hennig, 1966; Wiley, 1981; Wiley et al., 1991; Brooks & McLennan, 1991). The outgroups used were the type genera of the bradybaenids, helminthoglyp- tids, thysanophorids, camaenids, and sag- dids, plus Cepolis to represent the xanthony- cids (Emberton, 1991c). Data sources were from anatomical and behavioral characters compiled by Emberton (1994a: appendix A), with added anatomical character analyses using illustrations of Pilsbry (1940), Archer POLYGYRID SHELL EVOLUTION 73 (1948), Metcalf & Riskind (1979), Pratt (1981a), and Emberton (1988a, 1991a), and shell character analyses using the collection of the Department of Malacology, Academy of Natural Sciences of Philadelphia (ANSP). Shell characters were limited as much possi- ble to those involving complex apertural barriers. No anatomy of Pilsbry’s (1940) “Poly- gyra plicata group” had ever been ade- quately published (W. G. Binney's [1878: plate 15, fig. I] sketch of the genitalia of “Poly- gyra troostiana Lea” is uninformative and un- trustworthy, appearing as it does beside a totally inaccurate depiction of the genitalia of “P. tridentata [Say]’’). Therefore, all the (lim- ited) ANSP alcohol material of “Р. plicata Say” was examined, and one adult was cho- sen for dissection and illustration of the shell and the reproductive system, which were in- cluded in the general character analysis. Based on experience with anatomical-shell comparisons in the context of an anatomical- allozymic phylogenetic hypothesis among all species of both the Triodopsini and the Me- sodontini (Emberton, 1988a, 1991a), anatom- ically unknown polygyrid species were tenta- tively assigned to subgenus Бу shell characters alone. Thus, subgenus was used as the operational taxonomic unit for this phylogenetic analysis, even though any given subgenus may have been represented by up to eleven species that were each scored for each character. For this reason, all '“autapo- morphies” defining subgenera were retained in the data matrix. Polygyrid allozyme data (Emberton, 1988a, 1991a, 1994a) were not used in this analysis because data were lacking for many subgen- era and because existing data were from two separate analyses involving only partially overlapping sets of loci. Instead, relevant portions of this phylogentic hypothesis were visually compared for topological congru- ency with previously published, allozyme- based hypotheses (Emberton, 1988a, 1991a, 1994a). Phylogenetic analysis was conducted by hand and was interactive with construction of the data matrix. The analysis progressed along successively more restricted ingroups, each of which was compared with its closest possible outgroup. Thus, the resulting cla- dogram and its coordinate subgenus-by- character-state matrix were superficially free of certain homoplasies that would have re- sulted from a non-hierarchical, single-out- group analysis. Taxonomic Revision The revision was based on the general phylogenetic hypothesis and followed basic principles of phylogenetic taxonomy (Queiroz & Gauthier, 1990), which are becoming well established among vertebrate systematists and have recently been introduced to mala- cology by Roth (1995). The method, as em- ployed here, designates as taxa clades that are defined by shared derived characters (sy- napomorphies), regardless of subsequent evolutionary modifications of those charac- ters. For a discussion of the naturalness of this method and its great advantages over traditional taxonomy, see Queiroz & Gauthier (1990). Shell-Based Phylogenetic Analysis Previous studies had shown the extents and limits of intraspecific variation in polygy- па shell morphology (Emberton, 1988a, 1988b, 1994a, 1994b), which were consid- ered during character analysis. In general, one intact adult shell of the type species was chosen to represent each subgenus, but for subgenera with extreme shell variation, two or more representative species were se- lected. Outgroups were represented by one shell each of the type species of the type genus of each of the six closest polygyrid outgroup families (Emberton, 1991c), except when the type genus happened to have a highly derived shell morphology. Shells were mounted in the planes simul- taneously of both the rotational axis and the aperture (Emberton, 1988a: fig. 29b, d) on clear acetate sheets using thick rubber ce- ment, then x-rayed over single-coat SR-5 In- dustrex R film. Contact prints were made from the x-ray negatives. Shell “apertures” at half-whorl intervals were drawn, with accom- panying 3-mm scale lines, from the contact prints using a camera lucida mounted on a Wild M-5 dissecting microscope. The draw- ings were inked, reduce-xeroxed until all of about the same size, and mounted in regular array, arranged by a former classification (Webb, 1974; Richardson, 1986) that has since been revised (Emberton, 1994a, this paper). The demounted shells themselves, as well as the x-ray drawings of their ontogenies, were compared in a standard phylogenetic character analysis (Wiley, 1981; Wiley et al., 1991; Brooks & McLennan, 1991: chapter 2). 74 ЕМВЕАТОМ Conchological differences known to occur among closely related species within sub- genera of the Triodopsini and the Mesodon- tini (Emberton, 1988a, 1991a) were dis- counted as characters. The resulting shell-character by subgenus matrix was analyzed phylogenetically using Hennig86 (Farris, 1988), assigning equal weights to all characters. A Nelson semi- strict consensus tree (Farris, 1988) was com- puted from the set of resulting, equally and maximally parsimonious cladograms. Reliability of Fossils To qualitatively assess the reliability of fos- sils for reconstructing polygrid phylogeny, the shell-based and anatomy-behavior-shell- based phylogenetic hypotheses were com- pared for their degrees of phylogenetic res- olution among subgenera. Resolution in both cases was quantified as the number of nodes relative to the maximum number of possible nodes for the given number of taxa. Closest Convergences in Sympatry Closest potential polygyrid shell conver- gences in sympatry on the flat, tridentate, and umbilicate shell forms were searched for us- ing Pilsbry (1940), Hubricht (1985), and the collections of the Field Museum of Natural History, Chicago. Actual sympatry was tested by field work conducted during the spring months of 1979, 1981, 1982, and 1983. Shell Barriers and Water Loss Twenty-five adult Triodopsis tridentata were collected during five field trips in the spring and fall of 1970. Twenty-three were from Athens County, Ohio, and two from Monongahela County, West Virginia. Initial weights ranged from 0.39 to 1.06 gm. From ten of these snails, all three apertural denti- cles were completely removed using a table- mounted dentist's drill. These barrier-less nails were given 20 hours to recuperate from the operation, and seven appeared at that time to be uninjured and normally active. All snails were weighed and placed into in- dividual rubber-stoppered Erlenmeyer flasks in which were suspended 10.0 gm о the des- iccant calcium sulfate in a cheesecloth bag. Temperature was maintained at 14-15°C in a walk-in refrigerator. Each snail was weighed on a tortion balance every hour for 18 hours. Weighing took about two minutes per snail. At each weighing, the snail's activity state was recorded as (a) fully extended, (b) par- tially extended, (c) retracted without epi- phragm, (d) retracted with partial epiphragm, or (e) retracted with complete epiphragm. Pe- riods of individual fully extended activity were also recorded when they occurred between weighings. Each recorded weight was expressed as a percentage of the snail's initial weight, then the percent decrease from the previous hour's weight was calculated for each “snail- hour” that proved usable. Mean percent de- creases were calculated for each of eight ap- erture/activity categories. Differences among categories were expressed as percent changes in rates of evaporative water loss. The study was conducted some 25 years ago, and since that time the raw data were lost, so no statistical analysis could be per- formed. Remaining Conservation Priorities Phylogeny-based conservation priorities were assessed for the categories of (1) radi- ating, endemic clades; (2) extremely autapo- morphic, endemic taxa; (3) relic sister groups to major clades; and (4) sites rich in conver- gences in sympatry (Emberton, 1992). Prior- ities in these categories were judged from the general polygyrid phylogenetic hypothesis/ revision (this paper; Emberton, 1988a, 1991a) and from known species distributions (Hu- bricht, 1985) and sites of known high diver- sity (Solem, 1976; Emberton, 1995c, unpub- lished; Hubricht, unpublished). The current conservation status of each high-priority taxon or site was evaluated based on the U.S. Federal Endangered Species List and the locations of national and regional parks and forest preserves. RESULTS General Phylogenetic Analysis Table 1 defines the 115 characters used for phylogenetic analysis, with references to il- lustrations of the characters. | had not previ- ously used or detected some of these char- acters (39, or 34%) (Emberton, 1988a, 1991a, 1994a), primarily those dealing with (a) penial morphology within the tribe Stenotremini (Fig. 3), (b) reproductive-system morphology POLYGYRID SHELL EVOLUTION 15 TABLE 1. Anatomical, behavioral, and shell characters used for a phylogenetic hypothesis (Fig. 8) and a revision (text) of the Polygyridae. Citations refer to papers of Emberton, unless otherwise indicated. OO —J O O1 R © ND — . Nodulose fertilization pouch-seminal receptacle complex (1994a: fig. 2: character-state 2.2). . Anteriorly united retractor muscles (1994a: fig. 2: character-state 3.2). . Deeply excavated ureteric interramus (1994a: fig. 2: character-state 4.2). . Single, dorsal, penial pilaster (1994a: fig. 2: character-state 5.2). . Proximally swollen, internally lamellate vas deferens (1994a: fig. 2: character-state 8.2). . Terminally papillate penial verge (1988a: figs. 2f, 3b, 4b, 5е, 7b, 7d, 8c). . Two, flat terminal papillae on a terminal penial verge (1988a: fig. ба). . Distinct but pustulate lappets (= transversely, partially fused pustules) on the penial pilaster (1988a: figs. 2b, 5с, 58. . Doubled density of pustulate lappets on the penial pilaster (1988a: figs. 5a, 50). . Smooth lappets (= transversely, completely fused pustules) on the penial pilaster (1988a: fig. 2е). . Halved density of smooth lappets оп the penial pilaster (1988а: figs. 2d, 4a). . Smooth penial-pilastral lappets, plus in adulthood a short vas deferens (only about twice as long as the penis) and a penial-retractor-muscle attachment near the penis (on the vas deferens within one-third penis-length of the apex of the penis) (1988a: table 4). . Greatly enlarged pustules on the penial pilaster (1988a: fig. 7). . Penial-wall columns that merge mid-ventrally into 6-10 U-shapes that are tapered and slightly separated and that bear unequally sized pustules (1988a: fig. 8). . Penial-pilastral pustules forming a single column of abutting cubes (1988a: fig. 8a). . A ventrally subterminal penial verge (1988a: fig. 7). . Penial-pilastral pustules that are knob-like, unfused, and abruptly larger than the penial-wall pustules (1988a: figs. 9a, 9c), or derivatives thereof. . 15-20 penial-wall columns unmerging and radiating directly from the ejaculatory роге (1988a: figs. 9a, 9c), or derivatives thereof. . Club-shaped penis with a ventrally subterminal ejaculatory pore about 1/5-way from the apex and indented into the penial wall (Webb, 1959: figs. 22, 27, 34, 38; 1988a: fig. 9), or derivatives thereof (characters 28-30). . Penial-pilastral pustules fused into two interdigitating columns of rectangular boxes (1988a: fig. 12). . Penial-pilastral polygons 4-10 times the size of penial-wall pustules and armed with pustule- sized knobs, or derivatives thereof (characters 22-24). . Penial-pilastral polygons fused into a single mass or into large, irregular masses (1988a: fig. 11). . Ventral penial-wall columns with pustules indistinct (1988a: fig. 11). . Penial pilaster 3/4 the length of the penis and bearing polygons armed with blunt spurs (1988а: figs. 13a, 14a, 14b, 16a). . Indistinct pustules on the ventral-most radiating penial-wall columns (1988a: fig. 18b). . Penial-wall columns merging mid-ventrally into 5-7 acute, equilateral, widely separated V-shapes bearing equally sized pustules (1988a: figs. 14a, 15b, 16a). . Extremely long, narrow penis (at least 25 times as long as wide) (1988a: fig. 13). . Mace-shaped penis with a ventrally subterminal ejaculatory pore and with a sub-pore region erectile as a fleshy peduncle, and derivations thereof (characters 29, 30). . Ejaculatory-pore position approximately 1/4-way from the penial apex, peduncle small (Webb, 1948: fig. 4; Webb, 1959: figs. 14, 25а, 40, 43; 1988a: figs. 14a-b, 15-17). . Ejaculatory-pore position approximately 2/5-way from the penial apex, peduncle large (Webb, 1959: figs. 12, 13, 15, 41; 1988a: figs. 14c-d, 18а, 180). . A discernible clasping disc during mating (1994a: fig. 2: character-state 1.4), or derivatives thereof. . An unfanned origin of the penial retentor muscle (1994a: fig. 2: character-state 7.3). . An epiphallus and flagellum (1994a: fig. 2: character-state 9.2), or derived loss thereof. . А constriction in the epiphallus from the penial apex part-way toward the flagellum (1994a: fig. 2: character-state 9.3). . А Незпу protuberance near the apex of the penis (Roth 8 Miller, 1992: fig. 6). (continued) 76 EMBERTON TABLE 1. Anatomical, behavioral, and shell characters used for a phylogenetic hypothesis (Fig. 8) and a revision (text) of the Polygyridae. Citations refer to papers of Emberton, unless otherwise indicated. (Continued) 36. Conical, non-papillate verge at the apex of the penis (Pilsbry, 1940: fig. 512Bb; Webb, 1970: plate 35, fig. 5; Roth & Miller, 1993: figs. 19, 24). 37. Clasping disc during mating that is as voluminous as the inserted portion of the penis (1994a: fig. 2: character-state 1.5), or derivatives thereof. 38. Minutely pustulate sculpture on the dorsal penial wall (1994a: fig. 2: character-state 5.4). 39. Paired dorsal penial pilasters (Pilsbry, 1940: fig. 496Aa, Bb’, D), or derivative thereof (character 42). 40. Rugose clasping disk (Webb, 1990: plate 6, figs. 1—4). 41. Nearly effaced paired dorsal pilasters (1994a: appendix B, character state 5.4). 42. Single dorsal pilaster, plus a clasping disc that is twice as voluminous as the inserted portion of the penis (1994a: fig. 2: character-state 1.5). 43. One or more small, pointed, fleshy processes on the clasping disc, and derivatives thereof (characters 44, 45). 44. Asingle large, pointed, fleshy process on the clasping disc (Pilsbry, 1940: fig. 506, 1d-e; Webb, 1948: figs. 3, 3a). 45. Two large, pointed, fleshy processes on the clasping disc (Pilsbry, 1940: fig. 510: 2a, 5a). 46. Clasping disc divided peripherally into two or three broad, unpointed lobes (Webb, 1965: plate 27, figs. 1-4). 47. Epiphallus longer than the prostate-uterus (Pilsbry, 1940: figs. 524, 525). 48. Chitinous, ornate spermatophore (Webb, 1954). 49. No penial insertion during mating (1994a: fig. 2: character-state 1.8). 50. Lateral pilaster(s) on the clasping disc (= basal penis) (1994a: fig. 2: character-state 5.5). 51. A (secondarily) slender spermathecal duct (1994a: fig. 2: character-state 10.4). 52. (Basal) penial lateral pilasters apically modified into two fleshy-walled cups half as long as the penis (Fig. 3E). 53. Shell aperture with a complete basal lamella having a central trough half or more as broad as the lamella, and derivatives thereof. 54. Shell bearing a straight, even-height parietal apertural denticle isolated from both the umbilicus and the aperture (Fig. 6: character-state 5a), and derivatives thereof. 55. During mating, bearing an everted female organ that receives ejaculate from a pocket at tip of the penis (1994a: fig. 2: character-state 1.8). 56. Apertural-basal-lamellar central trough about one-third or less as broad as the lamella (Fig. 6: character-state 4b), and derivatives thereof. 57. Parietal apertural denticle extending from the umbilicus into the aperture (Fig. 6: character-state 5b), and derivatives thereof. 58. A single penial lateral pilaster apically modified into a symmetrical, fleshy-walled cup one-third as long as the penis, with a medial branch leading to a medial fleshy protuberance (Fig. 3C). 59. Lower apertural shell lip joined to the basal body whorl as a thin callus (Fig. 5, character-state 3b), and derivatives thereof. 60. Two large, fleshy penial lateral pilasters, both bearing apical V- or U-shaped structures, one up and one down (Fig. 3A). 61. Two large, fleshy penial lateral pilasters, neither bearing apical V- or U-shaped structures, and one or none bearing an apical, cup-like depression (Fig. 3T, S, P), and derivatives thereof. 62. One of the two large, fleshy penial lateral pilasters bearing an apical, cup-like depression about one-tenth as long as the penis (Fig. 31). 63. The two large, fleshy penial lateral pilasters free of apical, cup-like structures (Fig. 3S). 64. One of the two large, fleshy penial lateral pilasters bearing an apical, cup-like depression one-fifth to one-half as long as the penis (Fig. 3P). 65. Penial-retractor-muscle insertion on or near penial apex (1994a: fig. 2, character-state 7.5-7.8). 66. Penial sheath (secondarily) completely absent (1994a: fig. 2, character-state 7.8). 67. Only a vestigial flagellum near the penial apex (1994a: fig. 2, character-state 9.5), and deriva- tives thereof. 68. Shell bearing a triangular parietal denticle (Fig. 6: character-state 6a), and derivatives thereof. 69. Ovoviviparity (Pilsbry, 1930a). (continued) POLYGYRID SHELL EVOLUTION 1% TABLE 1. Anatomical, behavioral, and shell characters used for a phylogenetic hypothesis (Fig. 8) and a revision (text) of the Polygyridae. Citations refer to papers of Emberton, unless otherwise indicated. (Continued) 70. fale Shell apertural expansion rate abruptly increasing then decreasing such that successive whorls are nearly equal in volume (Fig. 6: character-state 2b), and derivatives thereof. At least one full whorl of growth beyond the expansion-rate increase (Fig. 6: character-state 2c-right). . Shell bearing a V- to U-shaped parietal denticle (Fig. 6: character-state 6b). . Shell with a narrow baso-palatal interdenticular notch (Pratt, 1981a: fig. 3i). . Complete loss of pustulation in the penial apex (1994a: fig. 2, character-state 5.6). . Asmall, sac-like, glandular diverticulum of the lower penis (Fig. 5Ln), and derivatives thereof. . Patches of glandular cells on the penial wall above the diverticulum (Fig. 5Ln). . Penial diverticulum large, one-third to equal the volume of the penis (Fig. 5Lb). . Shell apertural expansion rate (Secondarily) regular throughout ontogeny. . A secondary lobe on the penial diverticulum (Fig. 5Lb). . A (secondarily) globose shell with the aperture entirely free of denticles (Pilsbry, 1940: figs. 425-429). . Adnate penial diverticulum (Fig. 5PrF). . Penial diverticulum long and at least twice the volume of the penis (Fig. 5PrP, PrX). . Shell depressed, broadly umbilicate (Zilch, 1959-1960: fig. 2036). . Spiral, threadlike sculpture on the embryonic shell (Pilsbry, 1940: 689). . A bifurcate or trifurcate penial retractor muscle (Fig. 5PrP). . A vestigial epiphallus without a flagellum (Fig. 5MiM, МР). . А slender penis (width < 0.12 length) with an apical, pendant, conical projection (Fig. 5MiM, MiP), and derivatives thereof. . An extremely long and slender penis (width < 0.06 length) (Fig. SMiM). . A greatly enlarged, muscular, proximal vas deferens (Figs. 7МГР, 13). . Even-diameter vas deferens with no trace of epiphallus (Fig. 500, DU). . A stout penis (length/diameter <3.5) with a straight apex (Fig. 5DU). . Amoderately long penis (4 < length/diameter < 10) with a bent or convoluted apex (Fig. 5DD). . A downward curve on the lower limb of the parietal apertural denticle (Pilsbry, 1940: figs. 384-387). . A raised parietal callus (Pilsbry, 1940: figs. 384-387). . А penial apical chalice formed by the junction of lateral pilasters (1991a: fig. 27). . An even-diameter distal vas deferens with no trace of flagellum or epiphallus (1994a: fig. 2: character-state 9.6). . An arched parietal apertural denticle (Fig. 13: character-state 9b), and derivatives thereof. . Adepressed, hairless shell (height/diameter 0.4-0.6) (1991a: figs. 49, 50), and derivatives thereof. . A regularly oval-shaped aperture with the reflected lip uniform in width throughout its palatal and basal regions, with no basal dentition (1991a: fig. 48a). . A basal apertural lamella (1991a: figs. 46, 47). . A pronounced, blade-like parietal apertural denticle (1991a: figs. 46, 47). . A straight basal region of the aperture with only a vestigial lamella (1991a: fig. 10b). . A barrel-shaped, solid pedestal underlying the penial chalice (1991a: fig. 45c-d). . Asmall, globose or subglobose shell (diameter 8-15 mm, height/diameter 0.6-0.7) (1991a: figs. 35a,b, 40a,b), and derivatives thereof. . A globose, hairless shell (1991a: figs. 35a,b). . A subglobose shell (height/diameter 0.5-0.6) bearing periostracal scales, with the umbilicus broadly covered by an extension of the basal apertural lip, and with palatal and basal apertural denticles (1991a: figs. 40a,b), and derivatives thereof. . A thick-walled, hooded, cup-shaped penial chalice (1991a: figs. 7, 8, 9b). . Penial chalice with a higher left than right wall (1991a: fig. 27, transformation 21). . Shell very broadly umbilicate (1991a: figs. 39a,c). Dorsal penial sculpture (1991a: fig. 28, transformations 31-33). Dorsal penial sculpture consisting of 4-10 cord-like, subparallel, anastomosing ridges, running longitudinally to 30-degrees obliquely (1991a: figs. 4, 6, 16a). (continued) 78 EMBERTON TABLE 1. Anatomical, behavioral, and shell characters used for a phylogenetic hypothesis (Fig. 8) and a revision (text) of the Polygyridae. Citations refer to papers of Emberton, unless otherwise indicated. (Continued) 112. Dorsal penial sculpture consisting of 8-12 cord-like, subparallel, anastomosing ridges, running longitudinally to 30-degrees obliquely, many of which are contiguous with one or both lateral pilasters, and many of which enlarge basally to form a network of large basal bulges (1991a: figs. 2b, 11b,c, 156). 113. Penial chalice a deep, thin-walled scoop, with the left wall much higher than the right (1991a: figs: 2b, bie, 156). 114. Dorsal penial sculpture consisting of about 8-12 thin parallel ridges, equal in diameter, which is constant or gradually increases basally (1991a: figs. 15a,b, 16b,c). 115. Penial chalice a thick-walled, rounded or pointed ear-like flap, flared to the left (1991a: figs. 15a,b, 16b,c). within the tribe Polygyrini (Figs. 4, 5), and (c) shell ontogenetic and apertural morphology within—primarily—the tribes Stenotremini and Polygyrini (Fig. 6). Figure 7 shows the distributions of all 115 characters among polygyrid subgenera (as revised below). Figure 8 gives the maximum-parsimony cladogram of polygyrid subgenera (as ге- vised below), based on the data presented in Figure 7. Each node in the cladogram is sup- ported by one to three characters, numbered as in Table 1. Taxonomic Revision The following revision exactly follows the phylogenetic hypothesis of Figure 8. For brevity's sake, definitions employ character numbers as defined in Table 1. Species with only tentative assignment to subgenus are preceded by a question mark. Family POLYGYRIDAE Pilsbry, 1894 Definition: The first stylommatophoran pul- monate gastropod to possess characters #1, 2, and 3, and all of its descendants. Subfamily TRIODOPSINAE Pilsbry, 1940 Definition: The first Polygyridae to possess characters #4 and 5, and all of its descen- dants. Tribe TRIODOPSINI Definition: as for the subfamily. Unnamed Clade Comprising Webbhelix, Neohelix, and Xolotrema Synonym: “Xolotrema Rafinesque” (Webb, 1952) Definition: The first Triodopsinae to pos- sess character #6, and all of its descendants. Genus Webbhelix Emberton, 1988 Type species: Helix multilineata Say, 1821, by original designation. Definition: The first Triodopsinae to pos- sess character #7, and all of its descendents. Species: W. multilineata (Say, 1821). Genus Neohelix Ihering, 1892 Type species: Helix albolabris Say, “1816” 1817, by subsequent designation (Pilsbry, 1930a). Definition: The first Triodopsinae to pos- sess character #8, and all of its descendents. Subgenus Neohelix (Asamiorbis) subgen. n. Type species: Helix dentifera Binney, 1837. Definition: The first Neohelix to possess character #9, and all of its descendents. Etymology: Dr. Takahiro Asami, land-snail ecologist and geneticist, who has worked extensively with N. (A.) dentifera and other polygyrids in Virginia (Asami, 1988, 1993); or- bis (Latin) “disc” or “coil.” Species: N. (A.) dentifera (Binney, 1837); N. (A.) divesta (Gould, 1851); N. (A.) lioderma (Pilsbry, 1902). Unnamed clade comprising subgenera N. (Neohelix) s.s. and N. (Solemorbis) subgen. n. Definition: The first Neohelix to possess character #10, and all of its descendents. POLYGYRID SHELL EVOLUTION 79 FIG. 3. Character-state analysis of penial-functional-surface anatomy of the Stenotremini; see Table 1, characters #52, 58, 60-64. Characters are delineated by solid lines; arrows are hypothesized transforma- tions among characters. Anatomical figures (at different size scales) are from Archer (1948). A, Stenotrema (Archerelix) subgen. n.; С Stenotrema (Cohutta); E, Euchemotrema; P, Stenotrema (Pilsbrelix) subg. n.; $, Stenotrema (Stenotrema); T, Stenotrema (Toxotrema); aa, $. altispira altispira; ad, $. altispira depilatum; bb, $. barbatum; bd, $. blandianum; bg, $. barbigerum; bv, $. brevipila; cd, $. caddoense; ch, $. cohuttense; dc, S. deceptum; eg, $. edgarianum; ev, $. edvardsi; ex, $. exodon; ext, $. exodon turbinella (= $. turbinella); fl, 5. florida; ft, E. fraternum; h $. hirsutum; |, $. labrosum, ma, E. monodon aliciae; mg, $. magnifumosum; mn, E. monodon, mx, $. maxillatum; pb, $. pilsbryi; pl, $. pilula; sp, $. spinosum; st, $. stenotrema; и, $. unciferum. 80 ЕМВЕАТОМ FIG. 4. Reproductive system (minus ovotestis) and shell (umbilical view) of Millerelix (Prattelix) plicata (ANSP A2423-A) from Knox County, Tennessee. a = atrium, ag = albumen gland, ap = apical, pendant, conical projection of the penis, fpsc = fertilization pouch-seminal receptacle complex (= talon = carrefour), gp = gonopore, hd = hermaphroditic duct, mf = mantle-cavity floor, о = oviduvt, р = penis, pr = penial retractor muscle, pt = prostate, sd = spermathecal duct (= bursa copulatrix duct = gametolytic duct), u = uterus, v = vagina, vd = vas deferens. Subgenus Neohelix (Neohelix) s.s. Synonym: Neohelix albolabris group (Em- berton, 1988). Definition: The first Neohelix to possess character #11, and all of its descendents. Species: N. (N.) albolabris (Say, 1817); N. (N.) major (Binney, 1837). Subgenus Neohelix (Solemorbis) subgen. n. Type species: Neohelix solemi Emberton, 1988. Synonym: Neohelix alleni group (Ember- ton, 1988). Definition: The first Neohelix to possess character #12, and all of its descendents. Etymology: The late Dr. Alan Solem, long POLYGYRID SHELL EVOLUTION 81 МР FIG. 5. Character-state analysis of the lower-reproductive-tract anatomy of the Polygyrini; see Table 1, characters #75-92. Characters are delineated by solid lines; arrows are hypothesized transformations among characters. Anatomical figures (at different size scales) are from Pilsbry (1940), Metcalf 8 Riskind (1979), Pratt (1981a), and this paper (Fig. 4). DD, Daedalochila s.s.; DU, Daedalochila (Upsilodon); G, Giffordius; Lb, Lobosculum; Ln, Linisa; ММ, Millerelix s.s.; МР, Millerelix (Prattelix); Po, Polygyra; PrF, Praticolella (Farragutia); PrP, Practicolella (Praticolella); PrX, Praticolella (Filapex); be, Pr. berlandieriana; bu, D. burlesoni; ch, D. chisosensis; cr, Po. cereolus; df, Mi. doerfeuilliana; hp, D. hippocrepis; Ip, D. leporina; Iw, Pr. lawae; mb, Pr. mobiliana; mp, D. multiplicata; mr, Mi. тоогеапа; pi, G. pinchoti; pl, Mi. plicata; pu, Lb. pustula; sp, Po. septemvolva volvoxis; tm, Ln. tamaulipasensis; tx, Ln. texasiana; uv, D. uvulifera. 82 ЕМВЕАТОМ FIG. 6. Character analysis of shell morphologies of the Stenotremini and the Polygyrini; Table 1, characters #53, 54, 56, 57, 59, 70-72. 83 POLYGYRID SHELL EVOLUTION TTOOO OOTTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 STI =ETT TOTOO TOTOO TOTOO OTTOO 000TO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 OTT -90T 00000 00000 00000 00000 00000 0т000 TTOOO 00TTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 SOT -TOT OOOTT OOOTT OOOTT OOOTT OOOTT OOOTT OOOTT TOTTT TOTIT OTTIT 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00-96 T0000 T0000 T0000 70000 T0000 70000 T0000 T0000 T0000 T0000 OTTTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 S6-T6 T0000 T0000 70000 10000 10000 70000 T0000 T0000 10000 70000 TOOOT TOOOT OTOTT OOTTT 00000 00000 00000 00000 00000 00000 00000 т0000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 06-98 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 TOOTO OTOTO OOTTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 S8-T8 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 TOTTO TOTTO TOTTO TOTTO OTTTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 08-94 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 OTTTO OTTTO OTTTO OTTTO TITIO TITTO TITTO TITTO TITTO TTTTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 SL-TL OOOTT OOOTT OOOTT OOOTT OOOTT OOOTT OOOTT OOOTT O00TT OOOTT TOTTT TOTTT TOTTT TOTTT TOTTT TOTTT TOTTT TOTTT TOTTT TOTTT TOTTT OTTTT 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 02-99 70000 70000 70000 70000 T0000 T0000 70000 T0000 70000 T0000 10000 T0000 T0000 T0000 70000 70000 T0000 70000 70000 70000 70000 70000 OTOOT OOTOT OOOTT 00000 00000 00000 ‘00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 S9-T9 ‘sdno1Byno pue елэцэбап$ puABAjod Buowe (| э|ае1) заэзоелецо вр | JO SUONNQUISIC ‘7 "Ol 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 OTOTT OTOTT OTOTT TTOTT OOTTT 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 09-95 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T 0000T OTTTT OTTTT OTTTT OTTTT OTTTT ETT ET 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 SS-TS TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO 11000 TTOOO TTOOO TTO00 TTO00 TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO TTO00 TTOOO TTOOO TTOOO TTOOO TTOOO TTOOO OOTTO 00000 00000 00000 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 05-97 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00007 00000 00000 000T0 TOTTO OTTTO 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 ЗУ-ТУ 000T0 000T0 000T0 000T0 000T0 000T0 000TO 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 000T0 OTTTO OTTTO TTTTO OOTTO OOTTO OOTTO 0000T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 07-95 OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OTTTT TTITE 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 SE-TE 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 TOTOT OTTOT OOOTT 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 05-95 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 0т000 0т000 0т000 TTOOO OOTTO 00000 00000 00000 00000 00000 00000 00000 00000 00000 ST-Te 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 OOTTO OOTTO OOTTO OOTTO OOTTO TOTTO OTTTO o0000T: 00000 00000 00000 00000 00000 00000 OZ-9T 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 OTTOO TITOO 00000 000T0 0000T 00000 00000 ST-TT 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 00000 0000T 00001 OTTOT TOTOT TOTOT OOOTT 00000 OT-9 OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT OOTTT TEILTE ТЕТЕ ТЕ EE GEE TITTTE TITTT ETT 181919151 1515 575 TUBEE ETTET TEBEE TITIT 00000 S-T (uoposayw) ‘en (uoposeuoryy) “Sw (euoboreuydy) :aw витчоетеаау xTTauoung (snr18309T3UI) ‘I (sntqyotaqny) ‘т (эталоэтерзБеч) ‘ed (елэзеа) ‘ed (eza3ediadsaA) ‘ва (eTreyooTepseg) ‘а (иороттзап) ‘а (хттеззела) ‘ти (хттедеэттти) ‘ти (еттетортзела) ‘ла (хэ4еття) ‘ла (эзпрлепря) "Id (етэпбеллея) ‘ла uwnTnosoqoT eSTUTT едАБАТОа 5птрлоуутэ (xTTS1qsTId) *5 (ешэ130и935) *5 (ews130x0L) *s (хттелэечолу) *5 (23314092) "Ss PUSIJOWSUINT eITeunuysy (ersdeue1ot) ‘2 (xT3seuo3dÂ12) *о (euobordng) ‘2 STsdoqoTTIL (ewopausAg) *Ty (euoboTIv) “TY eTooTtıedsaA snTT9Sb19qU90H (STq10ABTOABEA) “PL (stsdopoT1L) “Pu (зталортоден) “PL (зталоиетттшоеи) “PL (sTGIOA1ASTIA) “PL (sTqitosyoo1g) “PL (stq2opXoJTaus) “PL (ешэдзотох) *х (зталохоэттм) *х (sTqioTuesy) *N (stq2ousTos) *N (хттечоем) "N хттечачем sdno1b3n0 84 ЕМВЕАТОМ В Bradybaena X Cepolis H Helminthoglypta outgroups Т Thysanophora с Pleurodonte Ss Sagda —_—_ + Webbhelix | 10-11—NN Neohelix s.s. 6 8—— 12—NS Neohelix (Solemorbis) | 9—_NA Neohelix (Asamiorbis) 4,5 13,14——15—XW Xolotrema (Wilcoxorbis) Trio- 16—XX Xolotrema s.s. dop- 19——_———————_TdS Triodopsis (Shelfordorbis) sini 17,18420—————————TGB Triodopsis (Brooksorbis) 21-22,23————TdP Triodopsis (Pilsbryorbis) 24-25——————TÓM Triodopsis (Macmillanorbis) 1-3 26-27 TdH Triodopsis (Haroldorbis) 28-29 —TdT Triodopsis s.s. 30—TdV Triodopsis (Vagvolgyorbis) 34————35—H Hochbergellus Vesperi- 36—V___Vespericola colini 40—CB Cryptomastix (Bupiogona) 3 8 ———— 39 —— HCC Cryptomastix s.s. Allo- 31-33 41—CM Cryptomastix (Micranepsia) go- r46 Tb Trilobopsis nini 42—43—44—А1А Allogona s.s. 45—A1D Allogona (Dysmedoma) 47 48—— АЗ Ashmunella Ashmunellini 55——E Euchemotrema 52-544 58————5С Stenotrema (Cohutta) Steno- 37 56/97 60 SA Stenotrema (Archerelix) tremini 59] 64—SP Stenotrema (Pilsbrelix) 61—463-SS Stenotrema s.s. 62—ST Stenotrema (Toxotrema) 49-51 69 ——-—_ TE Giffordius 68 VPO Polygyra 76—————Ln Linisa 70 75— 79———Lb Lobosculum 77,78 81 PrF Praticolella (Farragutia) 80—| 83-PrE Praticolella (Eduardus) Poly- 72-74 82-84-PrX Praticolella (Filapex) gyrini 85-PrP Praticolella s.s. 65-67 87-88-MiM Millerelix s.s. 85 ————] 89-MiP Millerelix (Prattelix) 90-91-DU Daedalochila (Upsilodon) 92-9400 Daedalochila s.s. 98—799——PaV Patera (Vesperpatera) 100-101PaP Раёега s.s. 102,103PaR Patera (Ragsdaleorbis) 95-97———————-1047105—IH Inflectarius (Hubrichtius) | 106— ТТ Inflectarius s.s. Meso- 107————_F Fumonelix dontini 109——App Appalachina 1081 111MeA Mesodon (Aphalogona) 110+23MeK Mesodon (Akromesodon) 1145MeM Mesodon s.s. FIG. 8. Phylogenetic hypothesis for polygyrid subgenera based on the data in Fig. 7. Synapomorphies supporting each node and defining each subgenus are numbered as in Table 1. “23MeK” = 112, 113 MeK. “1145MeM” = 114,115 MeM. POLYGYRID SHELL EVOLUTION 85 one of the world's leading and most prolific specialists on land snails, who did some im- portant work on polygyrids (Solem, 1976); or- bis (Latin) “disc” or “coil.” Species: N. (S.) alleni (Sampson, 1883); N. (S.) solemi Emberton, 1988. Genus Xolotrema Rafinesque, 1819 Type species: Helix denotata Ferussac, 1821 (= Helix notata Deshayes, 1830) by sub- sequent designation (Pilsbry, 1940). Definition: The first Triodopsini to possess characters #13 and 14, and all of its descen- dents. Subgenus Xolotrema (Wilcoxorbis) Webb, 1952 Type species: Polygyra appressa fosteri F. C. Baker, 1932, by original designation. Definition: The first Xolotrema to possess character #15, and all of its descendents. Species: X. (W.) fosteri (F. C. Baker, 1932); X. (W.) occidentalis (Pilsbry & Ferriss, 1907). Subgenus Xolotrema (Xolotrema) s.s. Definition: The first Xolotrema to possess character #16, and all of its descendents. Species: X. (X.) caroliniensis (Lea, 1834); X. (X.) denotata (Ferussac, 1821); X. (X.) ob- stricta (Say, 1821). Genus Triodopsis Rafinesque, 1819 Type species: Helix tridentata Say, “1816” 1817, by original designation. Definition: The first Triodopsinae to pos- sess characters #17 and 18, and all of its descendants. Subgenus Triodopsis (Shelfordorbis) Webb, 1959 Type species: Triodopsis fraudulenta vul- gata Pilsbry, 1940, by original designation. Synonym: species group Triodopsis vul- gata (Emberton, 1988). Definition: The first Triodopsis to possess character #19, and all of its descendents. Species: Т. (S.) claibornensis Lutz, 1950; Т. (S.) fraudulenta (Pilsbry, 1894); Т. (S.) picea Hubricht, 1958; 7. (S.) vulgata Pilsbry, 1940. Subgenus Triodopsis (Brooksorbis) subgen. n. Type species: Polygyra platysayoides Brooks, 1933. Synonym: species group Triodopsis platysayoides (Emberton, 1988). Definition: The first Triodopsis to possess character #20, and all of its descendents. Etymology: The late Dr. Stanley T. Brooks, who described the type species; orbis (Latin) “ISO OF GO Species: Т. (B.) platysayoides (Brooks, 1933). Unnamed Clade Comprising Triodopsis Subgenera 7. (Pilsbryorbis) subgen. n., Т. (Macmillanorbis) subgen. n., Т. (Haroldorbis), T. (Triodopsis), and Т. (Vagvolgyorbis) subgen. п. Definition: The first Triodopsis to possess character 421, and all of its descendents. Subgenus Triodopsis (Pilsbryorbis) subgen. n. Type species: Polygyra tridentata tennes- seensis Walker 8 Pilsbry, 1902. Synonym: species groups Triodopsis ten- nesseensis and Т. burchi (Emberton, 1988a). Definition: The first Triodopsis to possess characters #22 and 23, and all of its descen- dents. Etymology: The late Dr. Henry A. Pilsbry, for some 70 years the world's best known and most productive land-snail specialist, who wrote the definitive monograph on poly- gyrids (Pilsbry, 1940); orbis (Latin) “disc” or “coll.” Species: Т. (P.) burchi Hubricht, 1950; 7. (P.) complanata (Pilsbry, 1898); Т. (P.) tennes- seensis (Walker & Pilsbry, 1902). Unnamed Clade Comprising Triodopsis Subgenera T. (Macmillanorbis) subgen. n., T. (Haroldorbis), T. (Triodopsis), and T. (Vagvolgyorbis) subgen. n. Definition: The first Triodopsis to possess character #24, and all of its descendents. Subgenus Triodopsis (Macmillanorbis) subgen. n. Type species: Triodopsis tridentata rugosa Brooks & MacMillan, 1940. Synonym: species group Triodopsis rug- osa (Emberton, 1988). 86 ЕМВЕАТОМ Definition: The first Triodopsis already hav- ing character #24, to possess character #25, and all of its descendents. Etymology: The late Gordan K. MacMillan of the Carnegie Museum, Pittsburg, who col- lected and coauthored the type species; or- bis (Latin) “disc” or “coil.” Species: Т. (M.) fulciden Hubricht, 1952; Т. (M.) rugosa Brooks & MacMillan, 1940. Unnamed Clade Comprising Triodopsis Subgenera 7. (Haroldorbis), T. (Triodopsis), and 7. (Vagvolgyorbis) subgen. n. Definition: The first Triodopsis already hav- ing Character #24, to possess character #26, and all of its descendents. Subgenus Triodopsis (Haroldorbis) Webb, 1959 Type species: Triodopsis cragini Call, 1886, by original designation. Synonyms: “Triodopsis соре! (Wetherby)” (Vagvolgyi, 1968; species group Triodopsis cragini Call (Emberton, 1988a). Definition: The first Triodopsis already hav- ing characters #24 and 26, to possess char- acter #27, and all of its descendents. Species: Т. (H.) cragini Call, 1886; Т. (H.) henriettae (Mazyck, 1877); T. (H.) vultuosa (Gould, 1848). Unnamed Clade Comprising Triodopsis Subgenera 7. (Triodopsis) and Т. (Vagvolgyorbis) subgen. n. Definition: The first Triodopsis already hav- ing Characters #24 and 26, to possess char- acter #28, and all of its descendents. Subgenus Triodopsis (Triodopsis) s.s. Synonym: species groups Triodopsis tri- dentata (Say) and Т. fallax (Say) (Emberton, 1988a). Definition: The first Triodopsis already hav- ing characters #24, 26, and 28, to possess character #29, and all of its descendents. Species group 7. (T.) tridentata s.s.: Т. (T.) anteridon (Pilsbry, 1940); Т. (T.) tridentata (Say, 1817). Species group Т. (Т.) fallax: Т. (T.) alabam- ensis (Pilsbry, 1902); Т. (T.) fallax (Say, 1825); Т. (T.) hopetonensis (Shuttleworth, 1852); Т. (T.) obsoleta (Pilsbry, 1894); T. (T.) palustris Hubricht, 1958; Т. (T.) soelneri (Henderson, 1907); Т. (T.) vannostrandi (Bland, 1875). Subgenus Triodopsis (Vagvolgyorbis) subgen. n. Type species: Polygyra tridentata juxtidens Pilsbry, 1894b. Synonym: species group Triodopsis juxti- dens (Pilsbry) (Emberton, 1988). Definition: The first Triodopsis already hav- ing Characters #24, 26, and 28, to possess character #30, and all of its descendents. Etymology: Dr. Joseph Vagvolgyi, author of a conchological monograph on the Triod- opsinae and Cryptomastix (Vagvolgyi, 1968); orbis (Latin) “disc” or “coil.” Species: Т. (V.) discoidea (Pilsbry, 1904); Т. (V.) juxtidens (Pilsbry, 1894b); Т. (V.) neglecta (Pilsbry, 1899); T. (V.) pendula Hubricht, 1952. Subfamily POLYGYRINAE s.s. Definition: The first Polygyridae to possess characters 431, 32, and 33, and all of its de- scendents. Tribe VESPERICOLINI Emberton, 1994 Type genus: Vespericola Pilsbry, 1939, by original designation. Definition: The first Polygyrinae to possess character #34, and all of its descendents. Genus Hochbergellus Roth & Miller, 1992 Type species: Hochbergellus hirsutus Roth 8 Miller, 1992, by original designation. Definition: The first Vespericolini to pos- sess character #35, and all of its descen- dents. Species: H. hirsutus Roth & Miller, 1992. Genus Vespericola Pilsbry, 1939 Type species: Polygyra columbiana pilosa Henderson, 1928, by original designation. Definition: The first Vespericolini to pos- sess character #36, and all of its descen- dents. Species: V. armigera (Binney, 1885); V. co- lumbianus (Lea, 1838); У. euthales (Berry, 1939); V. hapla (Berry, 1933); V. karokorum Talmadge, 1962; У. marinensis Roth & Miller, 1993; У. megasoma (Dall, 1905); “М. sp. п. 1” (Roth 8 Miller, 1993); “М. sp. п. 2” (Roth 4 Miller, 1993); V. orius (Berry, 1933); V. pilosus POLYGYRID SHELL EVOLUTION 87 (Henderson, 1928); V. pinicola (Berry, 1916); V. pressleyi Roth, 1985; V. shasta (Berry, 1921); V. sierrana (Berry, 1921). Unnamed Clade Comprising Tribes Allogonini and Ashmunellini and Infrafamily Polygyrinai Definition: The first Polygyrinae to possess character 437, and all of its descendents. Tribe ALLOGONINI Emberton, 1994 Type genus: Allogona Pilsbry, 1939, by original designation. Definition: The first Polygyrinae already having character #37, to possess character #38, and all of its descendents. Genus Cryptomastix Pilsbry, 1939 Type species: Polygyra mullani olneyae Pilsbry, 1928, by original designation. Synonym: Triodopsis Rafinesque (in part) (Pilsbry, 1940; Vagvolgyi, 1968). Definition: The first Allogonini to possess character #39, and all of its descendents. Subgenus C. (Bupiogona) Webb, 1970 Type species: Polygyra mullani hendersoni Pilsbry, 1928, by original designation. Definition: The first Cryptomastix to pos- sess character #40, and all of its descen- dents. Species: С. (В.) hendersoni (Pilsbry, 1928). Subgenus С. (Cryptomastix) s.s. Definition: (as for the genus). Species: C. (C.) devia (Gould, 1846); C. (C.) mullani (Напа & Cooper, 1862); С. (C.) san- burni (Binney, 1886). Subgenus C. (Micranepsia) Pilsbry, 1940 Type species: Helix germana Gould, 1851, by original designation. Definition: The first Cryptomastix to pos- sess character #41, and all of its descen- dents. Species: C. (M.) germana (Gould, 1851). Unnamed Clade Comprising Allogona and Trilobopsis Definition: The first Allogonini to possess character #42, and all of its descendents. Genus Allogona Pilsbry, 1939 Type species: Helix profunda Say, 1821, by original designation. Definition: The first Allogonini already hav- ing Character #42, to possess character #43, and all of its descendents. Subgenus A. (Allogona) s.s. Definition: The first Allogonini to possess character #44, and all of its descendents. Species: A. (A.) profunda (Say, 1821) Subgenus A. (Dysmedoma) Pilsbry, 1939 Type species: Helix townsendiana Lea, 1838, by original designation. Definition: The first Allogonini to possess character #45, and all of its descendents. Species: А. (D.) lombardii Smith, 1943; A. (D.) ptychophora (Brown, 1870); A. (D) townsendiana (Lea, 1838). Genus Trilobopsis Pilsbry, 1939 Type species: Helix loricata Gould, 1846, by original designation. Definition: The first Allogonini already hav- ing Character #42, to possess character #46, and all of its descendents. Species: Т. loricata (Gould, 1846); Т. peni- tens (Hanna 4 Rixford, 1923); T. roperi (Pils- bry, 1889); Т. tehamana (Pilsbry, 1928); Т. tra- chypepla (Berry, 1933). Tribe ASHMUNELLINI Webb, 1954 Type genus: Ashmunella Pilsbry 8 Cocker- ell, 1899, by original designation. Definition: The first Polygyrinae already having character 437, to possess characters #47 and 48, and all of its descendents. Genus Ashmunella Pilsbry & Cockerell, 1899 Type species: Polygyra miorhyssa Dall, 1898 [= Ashmunella rhyssa miorhyssa (Dall, 1898)], by subsequent designation (Pilsbry, 1905). Definition: (as for the tribe). Species (n = 49): A. altissima (Cockerell, 1898); A. angulata Pilsbry, 1905; A. ani- masensis Vagvolgyi, 1974; A. ashmuni (Бай, 1896); A. auriculata (Say, 1818); A. bequaerti Clench & Miller, 1966; A. binneyi Pilsbry & Ferriss, 1917; A. carlbadensis Pilsbry, 1932; 88 ЕМВЕАТОМ А. chiricahuana (Dall, 1896); А. cockerelli Pils- bry 4 Ferriss, 1917; A. danielsi Pilsbry 8 Fer- riss, 1915; A. edithae Pilsbry 8 Cheatum, 1951; A. esuritor Pilsbry, 1905; A. ferrissi Pils- bry, 1905; A. harrisi Metcalf 8 Smartt, 1977; А. hawleyi Metcalf, 1973; A. hebardi Pilsbry & Vanatta, 1923; A. intricata Pilsbry, 1948; A. jamesensis Metcalf, 1973 (fossil); A. juarezen- sis Pilsbry, 1948; A. kochi Clapp, 1908; A. lenticula Gregg, 1953; A. lepidoderma Pilsbry 8 Ferriss, 1910; A. /evettei (Bland, 1881); A. macromphala Vagvolgyi, 1974; A. mearnsi (Dall, 1896); A. mendax Pilsbry 8 Ferriss, 1917; A. meridionalis Pilsbry, 1948; A. mog- ollonensis Pilsbry, 1900; A. montivaga Pils- bry, 1948; A. mudgei Cheatum, 1971; A. or- ganensis Pilsbry, 1936; A. pasonis (Drake, 1951); A. pilsbryana Ferriss, 1914; A. proxima Pilsbry, 1905; A. pseudodonta (Dall, 1897); A. rhyssa (Dall, 1897); A. rileyensis Metcalf & Hurley, 1971; A. ruidosana Metcalf, 1973 (fossil); A. salinasensis Vagvolgyi, 1974; A. sprouli Fullington 4 Fullington, 1978; A. tegil- lum Metcalf, 1973; A. tetrodon Pilsbry & Fer- riss, 1915; A. thomsoniana (Ancey, 1887); A. todseni Metcalf 8 Smartt, 1977; A. tularosana Metcalf, 1973; A. varicifera Ancey, 1901; A. walkeri Ferriss, 1904; A. watleyi Metcalf € Fullington, 1978. Infrafamily POLYGYRINAI s.s. Definition: The first Polygyrinae already having character #37, to possess characters #49, 50, and 51, and all of its descendents. Tribe STENOTREMINI Emberton, 1994 Type genus: 1819. Definition: The first Polygyrinai to possess characters #52, 53, and 54, and all of its de- scendents. Stenotrema Rafinesque, Genus Euchemotrema Archer, 1939 Type species: Helix monodon Rackett, 1821, by subsequent designation (Pilsbry, 1940). Definition: The first Stenotremini to pos- sess character #55, and all of its descen- dents. Species: E. fasciatum (Pilsbry, 1940); E. fraternum (Say, 1824); E. hubrichti (Baker, 1937); E. leai (Binney, 1840); E. monodon (Rackett, 1821); E. occidaneum Roth 8 Em- berton, 1994 (early Miocene fossil); E. wichi- torum Branson, 1972. Genus Stenotrema Rafinesque, 1819 Type species: Stenotrema convexa Rafinesque, 1819 [nomen nudum = Helix stenotrema Pfeiffer, 1842]. Definition: The first Stenotremini to pos- sess characters #56 and 57, and all of its descendents. Subgenus Stenotrema (Cohutta) Archer, 1948 Type species: Polygyra cohuttensis Clapp, 1914, by original designation. Definition: The first Stenotrema to possess character #58, and all of its descendents. Species: S. (Cohutta) cohuttensis Archer, 1948 Unnamed Clade Comprising Stenotrema (Archerelix), S. (Pilsbrelix), S. (Stenotrema) s.s., and S. (Toxotrema) Definition: The first Stenotrema to possess character #59, and all of its descendents. Subgenus Stenotrema (Archerelix) subgen. n. Type species: Helix barbigera Redfield, 1856. Description: The first Stenotrema to pos- sess character #60, and all of its descen- dents. Etymology: The late Dr. Allan F. Archer, who contributed “information, notes, and manuscript”? (Archer, 1948: 8) to Pilsbry’s (1940) monograph on Stenotrema and wrote a revision and ecological manual on the ge- nus (Archer, 1948); helix (Latin) “coil” or “snail.” Species: $. (A) barbigerum (Redfield, 1856); S. (A.) edgarianum (Lea, 1841); S. (A.) edvardsi (Bland, 1858); S. (A.) pilsbryi (Fer- riss, 1900). Unnamed Clade Comprising S. (Pilsbrelix), S. (Stenotrema) s.s., and S. (Toxotrema) Definition: The first Stenotrema to possess character #61, and all of its descendents. POLYGYRID SHELL EVOLUTION 89 Subgenus $. (Toxotrema) Rafinesque, 1819 Type species: Helix hirsuta Say, 1817, by subsequent designation (Pilsbry, 1930). Definition: The first Stenotrema to possess character #62, and all of its descendents. Species: 5. (T.) barbatum (Clapp, 1904); $. (T.) hirsutum (Say, 1817); S. (T.) labrosum (Bland, 1862); ?S. (T.) simile Grimm, 1971. Subgenus Stenotrema (Stenotrema) s.s. Synonyms: Stenotrema (Stenostoma) Rafinesque, 1831 (Archer, 1948); Stenotrema (Maxilliter) Pilsbry, 1940 (Archer, 1948); Stenotrema (Coracollatus) Archer, 1948. Definition: The first Stenotrema to possess character #63, and all of its descendents. Species and subspecies: $. (S.) altispira (Pilsbry, 1894); $. (S.) altispira depilatum (Pilsbry, 1895); ?S. (S.) angellum Hubricht, 1958; $. (S.) brevipila (Clapp, 1907); $. (S.) caddoense (Archer, 1935); 75$. (S.) calvescens Hubricht, 1961; S. (S.) florida Pils- bry, 1940; $. (S.) magnifumosum (Pilsbry, 1900); S. (S.) maxillatum (Gould, 1848); ?S. (S.) morosum Hubricht, 1978 (Pleistocene- Recent fossil); S. (S.) pilula (Pilsbry, 1900); S. (S.) spinosum (Lea, 1831); S. (S.) stenotrema (Pfeiffer, 1842); S. (S.) unciferum (Pilsbry, 1900); ?S. (S.) waldense Archer, 1938. Subgenus Stenotrema (Pilsbrelix) subgen. n. Type species: Polygyra stenotrema exodon Pilsbry, 1900. Description: The first Stenotrema to pos- sess character #64, and all of its descen- dents. Etymology: The late Dr. Henry A. Pilsbry, who wrote the definitive monograph on poly- gyrids (Pilsbry, 1940), and who described the type species; helix (Latin) “coil” or “snail.” Species: S. (P) blandianum (Pilsbry, 1903); $. (P) deceptum (Clapp, 1905); $. (P.) exodon (Pilsbry, 1900); S. (P) turbinella (Clench & Ar- cher, 1933). Unnamed Clade Comprising Polygyrini and Mesodontini Definition: The first Polygyrinai to possess characters #65, 66, and 67, and all of its de- scendents. Tribe POLYGYRINI s.s. Definition: The first Polygyrinai already having characters #65, 66, and 67, to pos- sess character #68, and all of its descen- dents. Genus Giffordius Pilsbry, 1930 Type species: Giffordius pinchoti Pilsbry, 1930, by original designation. Definition: The first Polygyrini to possess character #69, and all of its descendents. Species: G. corneliae Pilsbry, 1930; G. pin- choti Pilsbry, 1930. Unnamed Clade Comprising Polygyra, Linisa, Lobosculum, Praticolella, Millerelix, and Daedalochila Definition: The first Polygyrini to possess character #70, and all of its descendents. Genus Polygyra Say, 1818 Type species: Polygyra septemvolva Say, 1818, by subsequent designation (Herr- mannsen, 1847). Definition: The first Polygyrini already hav- ing character #70, to possess character #71, and all of its descendents. Species: P. caloosaensis Johnson, 1899 (Pliocene fossil); Р. cereolus (Mühlfeld, 1818); P. paludosa (Wiegmann, 1839); P. plana (Dunker, 1843); P. septemvolva Say, 1818. Unnamed Clade Comprising Linisa, Lobosculum, Praticolella, Millerelix, and Daedalochila Definition: The first Polygyrini already hav- ing character #70, to possess characters #72, 73, and 74, and all of its descendents. Unnamed Clade Comprising Linisa, Lobosculum, and Praticolella Definition: The first Polygyrini already hav- ing characters #70, 72, 73, and 74, to pos- sess character #75, and all of its descen- dents. Genus Linisa Pilsbry, 1930 Type species: Helix (Polygyra) anilis Gabb, 1865, by original designation. Synonyms (fide Pratt, 1981a): Polygyra (Daedalochila) texasiana group (in part) (Pils- 90 ЕМВЕАТОМ bry, 1940); Polygyra (Erymodon) Pilsbry, 1956; Polygyra (Monophysis) Pilsbry, 1956; Polygyra (Solidens) Pilsbry, 1956; Polygyra (Linisia) (Pratt, 1981a,b); Daedalochila (in part) (Richardson, 1986). Definition: The first Polygyrini already hav- ing Characters #70, 72, 73, 74, and 75, to possess character #76, and all of its descen- dents. Species: ?L. adamnis (Dall, 1890) (Upper Oligocene fossil); L. albicostulata (Pilsbry, 1896); L. anilis (Gabb, 1865); ?L. aula- comphala (Pilsbry & Hinkley, 1907); L. behri (Gabb, 1865); ?L. bicruris (Pfeiffer, 1857); ?L. cantralli (Solem, 1957); ?L. couloni (Shuttle- worth, 1852); ?L. dissecta (Martens, 1892); ?L. dysoni (Shuttleworth, 1852); ?L. euglypta (Pilsbry, 1896); ?L. hertleini Haas, 1961; ?L. hindsii (Pfeiffer, 1845); ?L. idiogenes (Pilsbry, 1956); ?L. matermontana (Pilsbry, 1896); ?L. nelsoni (Dall, 1897); L. pergrandis (Solem, 1959); ?L. plagioglossa (Pfeiffer, 1859); L. polita (Pilsbry 4 Hinkley, 1907); ?L. ponsonbyi (Pilsbry, 1896); [. richardsoni (Martens, 1892); ?L. suprazonata (Pilsbry, 1900); L. tamaulipasensis (Lea, 1867); L. texasiana (Moricand, 1833); L. ventrosula (Pfeiffer, 1845); ?L. yucatanea (Morelet, 1853). Unnamed Clade Comprising Lobosculum and Praticolella Definition: The first Polygyrini already hav- ing characters #70, 72, 73, 74, 75, and 76, to possess characters #77 and 78, and all of its descendents. Genus Lobosculum Pilsbry, 1930 Type species: Helix pustula Férussac, 1822, by subsequent designation (Pilsbry, 1930b: 320). Definition: The first Polygyrini already hav- ing characters #72, 73, 74, 75, 76, 77, and 78, to possess character #79, and all of its descendents. Species: L. pustula (Férussac, 1822); L. pustuloides (Bland, 1858). Genus Praticolella Martens, 1892 Type species: Praticola ocampi Strebel 8 Pfeffer, 1880 (= Helix ampla Pfeiffer, 1866), by original designation. Definition: The first Polygyrini already hav- ing characters #72, 73, 74, 75, 76, 77, 78, to possess character #80, and all of its descen- dents. Subgenus Praticolella (Farragutia) Vanatta, 1915 Type species: Helix mobiliana Lea, 1841, by original designation. Definition: The first Praticolella to possess character #81, and all of its descendents. Species: P. (F.) mobiliana (Lea, 1841). Unnamed Clade Comprising Praticolella (Eduardus), P. (Filapex), and Р. (Praticolella) s.s. Definition: The first Praticolella to possess character #82, and all of its descendents. Comment: The membership of P. (Eduar- dus) in this clade needs to be tested by dis- section, because Pilsbry (1936) was inexplicit about the size of the appendix. Subgenus Praticolella (Eduardus) Pilsbry, 1930 Type species: Polygyra martensiana Pils- bry, 1907, by original designation. Definition: The first Praticolella already having character #82, to possess character #83, and all of its descendents. Species: P. (E.) martensiana 1907). (Pilsbry, Subgenus Praticolella (Filapex) Pilsbry, 1940 Type species: Helix jejuna Say, 1821, by original designation. Definition: The first Praticolella already having character #82, to possess character #84, and all of its descendents. Species: P. (F.) bakeri (Vanatta, 1915); P. (F.) jejuna (Say, 1821); P. (F.) lawae (Lewis, 1874). Subgenus Praticolella (Praticolella) s.s. Definition: The first Praticolella already having character #82, to possess character #85, and all of its descendents. Species: P. (P.) ampla (Pfeiffer, 1866); P. (P.) berlandieriana (Moricand, 1833); P. (P.) candida Hubricht, 1983; P. (P.) flavescens (Pfeiffer, 1848); P. (P.) griseola (Pfeiffer, 1841); P. (P.) pachyloma (Pfeiffer, 1847); P. (P.) strebeliana Pilsbry, 1899; P. (P.) taeniata Pilsbry, 1940; P. (P.) trimatris Hubricht, 1983. POLYGYRID SHELL EVOLUTION 91 Unnamed Clade Comprising Millerelix and Daedalochila Definition: The first Polygyrini already hav- ing characters #70, 72, 73, and 74, to pos- sess character #86, and all of its descen- dents. Genus Millerelix Pratt, 1981 (see below) Type species: Helix mooreana W. G. Bin- ney, 1857, by original designation. Definition: The first Polygyrini already hav- ing characters #70, 72, 73, 74, and 86, to possess character #87, and all of its descen- dents. Subgenus Millerelix (Millerelix) s.s. Pratt, 1981 Definition: The first Millerelix to possess chiaracter #88, and all of its descendents (Pratt, 19815). Species: M. doerfeulliana (Lea, 1838); M. gracilis (Hubricht, 1961); ?М. implicata (Mar- tens, 1865); M. lithica (Hubricht, 1961); M. mooreana (Binney, 1857); 2M. rhoadsi (Pils- bry, 1900); M. tholus (Binney, 1857). Subgenus Millerelix (Prattelix) subgen. n. Type species: Polygyra plicata Say, 1821. Synonyms: Polygyra plicata group (Pilsbry, 1940); Daedalochila plicata group = unnamed subgenus (Pratt, 1981a). Definition: The first Millerelix to possess character #89, and all of its descendents. Comments: The shell of the type species is sparsely and evenly covered with long, con- spicuous periostracal hairs, seemingly round in cross-section and slightly curved at the tip, that fade out toward the umbilicus (Fig. 13). They are easily broken, hence Pilsbry’s (1940: 626) mildly erroneous, “a few short hairs, usually preserved only in the umbilicus and behind the lip.” The long hairs were found in all four of the Academy of Natural Sciences's alcohol-preserved lots of D. pli- cata: Alabama, Madison County (ANSP A 2423-B); Tennessee, Marion County (A 2423- C); Tennessee, Knox County (A 2423-A, Fig. 13); and Kentucky, Barren County (A 2387-B, most hairs broken). Etymology: The subgenus is named in honor of Will Pratt. Species: deltoidea (Simpson, 1889); fatigi- ata (Say, 1829); jacksoni (Bland, 1866); pere- grina (Rehder, 1932); plicata (Say, 1821); simp- soni (Pilsbry & Ferriss, 1907); troostiana (Lea, 1838). Genus Daedalochila Beck, 1837 Type species: Helix auriculata Say, 1818, by subsequent designation (Herrmannsen, 1847). Definition: The first Polygyrini already hav- ing Characters #70, 72, 73, 74, and 86, to possess character 490, and all of its descen- dents. Subgenus Daedalochila (Upsilodon) Pilsbry, 1930c Type species: Helix hippocrepis Pfeiffer, 1848, by original designation. Synonym: Daedalochila (Acutidens) Pils- bry, 1956. Definition: The first Daedalochila to pos- sess character #91, and all of its descen- dents. Comment: The recurved palatal apertural denticle of D. acutidentata also occurs in the otherwise very different shell of D. poeyi, so does not seem to be a reliable character for defining a clade. Species: ?D. (U.) acutedentata (Binney, 1858); D. (U.) burlesoni (Metcalf & Riskind, 1979); D. (U.) chisosensis (Pilsbry, 1936); D. (U.) да! (Metcalf & Riskind, 1979); D. (U.) hip- pocrepis (Pfeiffer, 1848); D. (U.) leporina (Gould, 1848); D. (U.) multiplicata (Metcalf & Riskind, 1979); ?D. (U.) poeyi (Aguayo 8 Jaume, 1947); О. (U.) sp. п. A (Pratt, 1981 a); О. (U.) sterni (Metcalf & Riskind, 1979). Subgenus Daedalochila (Daedalochila) s.s. Synonym: Polygyra auriculata group (Pils- bry, 1940). Definition: The first Daedalochila to pos- sess characters #92, 93, and 94, and all of its descendents. Comments: The downward curve on the lower limb of the parietal apertural denticle 1$ a newly discovered synapomorphy. Pratt (1981a) stated (without giving evidence) that 2D. ariadne and ?Mi. implicata are members of a new genus; this needs to be investi- gated. Species: 2D. (D.) ariadne (Pfeiffer, 1848); D. (D.) auriculata (Say, 1818); D. (D.) auriformis (Bland, 1862); D. (D.) avara (Say, 1818); D. (D.) delecta (Hubricht, 1976); D. (D.) hausmani (Jackson, 1948); 2D. (D.) oppilata (Morelet, 92 ЕМВЕАТОМ 1849); D. (D.) peninsulae (Pilsbry, 1940); D. (D.) postelliana (Bland, 1862); D. (D.) sub- clausa (Pilsbry, 1899); D. (D.) uvulifera (Shut- tleworth, 1852). Tribe MESODONTINI Emberton, 1991 Type genus: Mesodon Férussac, 1821. Definition: The first Polygyrinai already having characters #65, 66, and 67, to pos- sess characters #95, 96, and 97, and all of its descendents. Genus Patera Albers, 1850 Type species: Helix appressa Say, 1821, by subsequent designation Pilsbry, 1930c: 326). Definition: The first Mesodontini to pos- sess character #98, and all of its descen- dents. Subgenus Patera (Vesperpatera) Emberton, 1991 Type species: Polygyra binneyana Pilsbry, 1899, by original designation. Definition: The first Patera to possess char- acter #99, and all of its descendents. Species: P. (V.) binneyana (Pilsbry, 1899); P. (V.) clenchi (Rehder, 1932); P. (V.) indian- orum (Pilsbry, 1899); P. (V.) kiowaensis (Sim- pson, 1888); P. (V.) leatherwoodi (Pratt, 1971); P. (V.) roemeri (Pfeiffer, 1848). Unnamed Clade Comprising P. (Patera) and P. (Ragsdaleorbis) Definition: The first Patera to possess char- acter #100, and all of its descendents. Subgenus Patera (Patera) s.s. Definition: The first Patera already having character #100, to possess character #101, and all of its descendents. Species: P. (P.) appressa (Say, 1821); P. (P.) clarki (Lea, 1858); P. (P.) laevior (Pilsbry, 1940); P. (P.) panselena (Hubricht, 1976); P. (P.) perigrapta (Pilsbry, 1894b); P. (P.) sargen- tiana (Johnson & Pilsbry, 1892). Subgenus Patera (Ragsdaleorbis) Webb, 1954 Type species: Helix pennsylvanicus Green, 1827, by original designation. Definition: The first Patera already having character #100, to possess characters #102 and 103, and all of its descendents. Species: P. (R.) pennsylvanica (Green, 1827). Genus Inflectarius Pilsbry, 1940 Type species: Helix inflecta Say, 1821, by original designation. Definition: The first Mesodontini to pos- sess character #104, and all of its descen- dents. Subgenus /nflectarius (Hubrichtius) Emberton, 1991 Type species: Mesodon kalmianus Hu- bricht, 1965, by original designation. Definition: The first /nflectarius to possess character #105, and all of its descendents. Species: /. (H.) downieanus (Bland, 1861); /. (H.) kalmianus (Hubricht, 1965) Subgenus /nflectarius (Inflectarius) s.s. Definition: The first /nflectarius to possess character #106, and all of its descendents. Species group I. (l.) edentatus: I. ((.) eden- tatus (Sampson, 1889); I. (.) magazinensis (Pilsbry & Ferriss, 1907). Species group /. ((.) smithi: I. (1.) smithi (Clapp, 1905). Species group /. ((.) inflectus: I. (l.) арргох- imans (Clapp, 1905); I. (l.) inflectus (Say, 1821); /. (1.) rugeli (Shuttleworth, 1852); /. ((.) verus (Hubricht, 1954). Species group: I. (l.) ferrissi: I. (l.) ferrissi (Pilsbry, 1897); I. (1.) subpalliatus (Pilsbry, 1893). Genus Fumonelix Emberton, 1991 Type species: Helix wheatleyi Bland, 1860, by original designation. Definition: The first Mesodontini to pos- sess character #107, and all of its descen- dents. Species: F. archeri (Pilsbry, 1940); F. christyi (Bland, 1860); F. jonesiana (Archer, 1938); F. orestes (Hubricht, 1975); F. weth- erbyi (Bland, 1874); Е. wheatleyi (Bland, 1860). Unnamed Clade Comprising Appalachina and Mesodon Definition: The first Mesodontini to pos- sess character #108, and all of its descen- dents. POLYGYRID SHELL EVOLUTION 93 Genus Appalachina Pilsbry, 1940 Type species: Polygyra sayana Pilsbry, 1906, by original designation. Definition: The first Mesodontini already having character #108, to possess character #109, and all of its descendents. Species: A. chilhoweensis (Lewis, 1870); A. sayanus (Pilsbry, in Pilsbry & Ferriss, 1906). Genus Mesodon Férussac, 1821 Type species: Helix thyroidus Say, 1817, by monotypy. Definition: The first Mesodontini already having character #108, to possess character #110, and all of its descendents. Subgenus Mesodon (Aphalogona) Webb, 1954 Type species: Helix elevata Say, 1821, by original designation. Definition: The first Mesodon to possess character #111, and all of its descendents. Species: M. (Aph.) elevatus (Say, 1821); M. (Aph.) mitchellianus (Lea, 1838); M. (Aph.) za- letus (Binney, 1837). Subgenus Mesodon (Akromesodon) Emberton, 1991 Type species: Polygyra andrewsae norma- lis Pilsbry, 1900, by original designation. Definition: The first Mesodon to possess characters #112 and 113, and all of its de- scendents. Species: M. (Akr.) altivagus (Pilsbry, 1990); M. (Akr.) andrewsae Binney, 1879; M. (Akr.) normalis (Pilsbry, 1900). Subgenus Mesodon (Mesodon) s.s. Definition: The first Mesodon to possess characters #114 and 115, and all of its de- scendents. Species: M. (M.) clausus (Say, 1821); M. (M.) sanus (Clench & Archer, 1933); M. (M.) thyroidus (Say, 1817); M. (M.) trossulus Hu- bricht, 1966. Shell-Based Phylogenetic Analysis Figures 9-11 present x-ray outlines of 57 shells representing polygyrid subgenera and outgroups. These outlines were used for phy- logenetic character analysis. Also included in the character analysis were the x-rayed shells themselves, plus four shells from the ANSP collection representing four additional species: Ashmunella angulata, Stenotrema (Archerelix) barbigerum, S. (Pilsbrelix) ex- odon, and S. (Toxotrema) hirsutum. No shell was available of Hochbergellus hirsutus, but its published description (Roth & Miller, 1992) was used to score as many characters as possible. Thus, a total of 62 species were included, of which only five lacked x-ray data. Table 2 defines the 14 characters and 71 character states scored for phylogenetic analysis, which are illustrated in Figures 12, 13, and 6. All of these characters and char- acters states are new or newly evaluated. Figure 14 shows the distributions of shell characters among subgenera and species. Figure 15 diagrams shell characters that were excluded from phylogenetic analysis because of Known high levels of homoplasy within subgenera of the Triodopsini and Me- sodontini. Figure 16 shows the results of cladistic analysis of the data in Figure 14. In total, Hennig86 generated 1,529+ equally and maximally parsimonious cladograms with a consistency index of 0.34 and a retention in- dex of 0.68, of which Figure 16 is the Nelson consensus tree. Reliability of Fossils Comparison of shell-based and anatomy- behavior-shell-based phylogenetic hypothe- ses revealed conspicuous differences in res- olution and topology. Among 54 ingroup (polygyrid) taxa, with a maximum possible resolution of 53 nodes, the shell hypothesis (Fig. 16) had 10 nodes (19% of maximum) and the anatomy-behavior-shell hypothesis (Fig. 8) had 40 nodes (75% of maximum). Thus, the anatomy-behavior-shell hypothesis had four times the resolution of the shell- based hypothesis. Topologically, the only congruence be- tween the two hypotheses was т the Stenotremini, which both showed as a mono- phyletic clade with Euchemotrema and Stenotrema (Cohutta) at its base. Other clades showed major discrepancies between the two phylogenetic hypotheses. Thus, rel- ative to the anatomy-behavior-shell hypoth- esis/revision, the shell-based hypothesis (a) grouped Vespericola and Appalachina, Dae- dalocheila (Upsilodon) hippocrepis and Lo- bosculum, and Polygyra and Millerelix (Prat- 94 ЕМВЕАТОМ AID IGE 9! FIGS. 9-11. Shell ontogenies, from x-rays of adult shells, of representatives of polygyrid subgenera and of polygyrid closest outgroup families. The representatives are type species unless otherwise indicated. All scale bars = 3 mm. The shells are arranged by a previous classification (Webb, 1974; Richardson, 1986), since revised (Emberton, 1994a, this paper). Dotted lines indicate portions of the x-rays that were difficult to interpret. By chance the x-rayed shell representing Ashmunella (As) was broken internally; all characters could be scored from this drawing, however, so the specimen was not replaced. Specimen abbreviations, defined in alphabetical order with their catalog numbers at the Academy of Natural Sciences of Philadel- phia, are: AIA, Allogona (Allogona) profunda, 77867; AID, Allogona (Dysemdoma) townsendiana, 100390; App, Appalachina sayana, 139140; As, Ashmunella rhyssa, 166077; CB, Cryptomastix (Bupiogona) hena- ersoni, 171267; CC, Cryptomastix (Cryptomastix) mullani, 171245; CM, Cryptomastix (Micranepsia) ger- mana, 11154; DD, Daedalochila (Daedalocheila) auriculata, 57070; DU, Daedalochila (Upsilodon) hippocre- pis, 84629; DUa, D. (U.) acutidentata, 166418; E, Euchemotrema leai, 172539; F, Fumonelix wheatleyi, 169691; G, Giffordius pinchoti, 150735; IH, Inflectarius (Hubrichtius) downieanus (non-type), 91035; Il, Inflectarius (Inflectarius) inflectus, 91616; 11, /. (1.) ferrissi, 98085; Lb, Lobosculum pustula, 86968; Ln, Linisa anilis, 166371; Lnb, Linisa behri, 166487; MeA, Mesodon (Aphalogona) elevatus, 81161; MeK, Mesodon (Akromesodon) normalis, 169640; MeM, Mesodon (Mesodon) thyroidus, 71950; MiM, Millerelix (Millerelix) mooreana, 158375; MiP, Millerelix (Prattelix) plicata, 143448; NA, Neohelix (Asamiorbis) dentifera, 78876; (continued) POLYGYRID SHELL EVOLUTION 95 PrE IG 510: NN, Neohelix (Neohelix) albolabris, 75843; NS, Neohelix (Solemorbis) solemi, 182281; OB, outgroup Brady- baenidae: Bradybaena similaris, 174469; OC, outgroup Camaenidae Pleurodonte lynchnuchus, 32588; OH outgroup Helminthoglyptidae: Helminthoglypta tudiculata, 112911; OS, outgroup Sagdidae: Sagda cooki- ana, 139388; OT, outgroup Thysanophoridae: Thysanophora impura, 177310; OX, outgroup Xanthony- chidae: Cepolis cepa, 33301; PaP, Patera (Patera) appressa, 335137; PaR, Patera (Ragsdaleorbis) penn- sylvanica, 251512; PaV, Patera (Vesperpatera) binneyana, 176765; Po, Polygyra septemvolva, 69117; PrE, Praticolella (Eduardus) martensiana, 98578; PrF, Praticolella (Farragutia) mobiliana, 105999; PrP, Praticolella (Praticolella) ampla, 131749; PrX, Praticolella (Filapex) jejuna, 77035; SC, Stenotrema (Cohutta) cohuttensis, 170118; SS, Stenotrema (Stenotrema) stenotrema, 169148; SSm, S. (S.) maxillatum, 170141; SSp, S. (S.) spinosum, 11383; ST, Stenotrema (Toxotrema) hirsutum, 11396; Tb, Trilobopsis loricata, 11149; TdB, Triodopsis (Brooksorbis) platysayoides, 183201; Тан, Triodopsis (Haroldorbis) cragini, 186723; ТОМ, Tri- odopsis (Macmillanorbis) rugosa, 174909; TAP, Triodopsis (Pilsbryorbis) tennesseensis, 139143; TdS, Tri- odopsis (Shelfordorbis) vulgata, 68807; Тат, Triodopsis (Triodopsis) tridentata, 211921; TdV, Triodopsis (Vagvolgyorbis) juxtidens, 64720; V, Vespericola columbiana, 158355; W, Webbhelix multilineata, 190168; XW, Xolotrema (Wilcoxorbis) fosteri, 157255; XX, Xolotrema (Xolotrema) denotata, 128444. 96 ЕМВЕАТОМ TdB TdM OT E 111 telix), each pair of which has been classified into two widely separated clades; and (b) split Ashmunella rhyssa and A. angulata, Daedalocheila (Upsilodon) hippocrepis and D. (U.) acutidentata, Praticolella and Lobos- culum, Linisa anilis and Linisa behri, Millerelix (Millerelix) and M. (Prattelix), and Polygyra and Giffordius, each pair of which has been classified into a single, closely related clade. Among outgroups, the shell-based hypothe- sis conflicted with a previous, anatomy- based hypothesis (Emberton, 1991c; Fig. 8) by joining Cepolis in a clade with Pleurodonte and Sagda. Closest Convergences in Sympatry Figure 1 shows the closest known, field- validated, polygyrid convergences in sympa- try on the globose, umbilicate, flat, and tri- dentate shell forms. As mentioned in the Introduction, the globose case 1$ the closest convergence and has received detailed POLYGYRID SHELL EVOLUTION 97 TABLE 2. Shell characters used for a separate phylogenetic analysis of the Polygyridae (Fig. 16). 1: 2. 13. 14. Shape of the aperture's generating curve throughout late-juvenile ontogeny (Fig. 12: 1). 1a, “egg,” “pinto bean,” or “lima bean.” 1b, “kidney bean.” 1c, “bulging kidney bean.” Approximate number of complete adult whorls (Fig. 12: 2, which shows the shell apex down). 2a, four. 2b, five. 2c, six. 2d, eight. 2e, nine. Approximate ratio of apertural areas three whorls apart, beginning near the apex and ending at least 3/4-whorl before the aperture (Fig. 12: 3). 3a, 40:1. 3b, 32:1. 3c, 15:1. 3d, 9:1. Зе, 7:1. Columellar surface of the body whorl approximately 1/4-whorl before the adult aperture (Fig. 12: 4). 4a, flat or slightly concave, but convex at the suture. 4b, flat or slightly concave, flat at the suture. 4c, convex throughout. Umbilical shape (Fig. 12: 5, which outlines the x-rayed umbilici). 5a-5h, extremely narrow to broad (some intergradation among categories). Umbilical sutures (Fig. 13: 6). 6a, strongly to weakly shouldered. 6b, rounded, unshoul- dered. Umbilical-wall whorls (Fig. 13: 7). 7a, flat. 7b, slightly flat. 7c, round. Basal denticle(s) (Fig. 13: 8). 8a, absent. 8b, baso-columellar shoulder or knob parallel to apertural plane. 8c, partial lamella parallel to apertural plane. 8d, basal denticle parallel to apertural plane. 8e, denticle or lamella(e) transverse to apertural plane. 8f (= Fig. 6: char- acter 4), complete lamella parallel to apertural plane. Parietal denticle shape (size variable) (Fig. 13: 9). 9a, absent. 9b, linear, curved toward the umbilicus, higher toward the aperture. 9c (= Fig. 6: character 5), linear, straight, even in height. 9d (= Fig. 6: character 6), Triangular to spatulate. Ye, linear, curved away from the umbilicus, higher away from the aperture. Size (Fig. 13: 10). 10a-10i, gigantic to minute (some intergradation among categories). . Gradually increasing outward tilt of the long axis of the aperture (= Fig. 6: character 1). 11a, throughout ontogeny or until tilting upward slightly. 11b, until tilting upward conspicuously. . Whorl expansion rate (= Fig. 6: character 2). 12a, constant. 12b, increasing then decreas- ing, with successive whorls always larger. 12c, increasing then decreasing, such that successive whorls are equal. Lower apertural lip (= Fig. 6: character 3). 13a, separate from basal shell. 13b, joined to basal shell as a thin callus. Palatal denticle (= Fig. 6: character 7). 14а, absent. 146, discrete, parallel to apertural plane. 14c, non-discrete, forming a thin shelf grading into the basal denticle, parallel to apertural plane. 14d, semidiscrete, with bottom tapering and thinning, basally slanted inward into the apertural plane. study, including discoveries of several sites of sympatry (Emberton, 1994b, 1995b). The second closest shell-form conver- gence in sympatry is on the flat form, exhib- ited by the mesodontin Patera (Patera) laevior and the triodopsin Xolotrema (Wilcoxorbis) fosteri. Fieldwork in April 1981 along the lower Ohio River Valley, visiting all stone bluffs reasonably accessible by road, accu- mulated collections at 22 stations (numbered as “H-1” through “Н-22”), all material from which is catalogued at the Field Museum of Natural History. Both species had been re- ported from Grand Chain, Indiana (Pilsbry, 1940), but extensive search failed to find both in 1981, perhaps due to recent floods. Sym- patry between Р. laevior and X. fosteri was discovered at only one station: on a sand- stone wall above the Ohio River at the Fern Cliff estate, Hawesville, Hancock County, Kentucky, 21 April 1981. Figure 17 illustrates the external anato- mies, shells, and dissected reproductive anatomies of representative specimens of Patera laevior and Xolotrema fosteri from the Hawesville, Kentucky, site, and summarizes their mating-behavioral differences accord- ing to Webb (cited in Emberton, 1994a). The shells and external bodies were virtually iden- tical in size and shape (Figs. 1, 17), but inter- nally X. fosteri differed from P. laevior by its penial sheath, penial retentor muscle, retrac- tor-muscle attachment on the vas deferens, and thick gametolytic-gland (= spermathecal) duct, all correlating with its internal (vs. external) sperm exchange (Fig. 17). Ecologi- 98 ЕМВЕАТОМ 00006 PPS BR OOO © ©, lamer: А ND t= С FIGS. 12-13. Shell characters among polygyrid subgenera, п addition to those in Fig. 6. Definitions are given in Table 2. POLYGYRID SHELL EVOLUTION 99 о LS) CD 0 NS US —— с Ted D Te IS e № Ре eCD {

MA “ oe > што она see joo) TOUS aaa © & A U U U E OGGQAHHH HAA D D OT TTIHH aves OOOO0OOOSE2223x“ HH HH HR HTPI YOU Baal MHONNNNNNNYVAHHAHAAAAESZOADDWIAAAHHH RMSE 2, etc.). FIG. 14. Distributions of shell characters among polygyrid subgenera. Characters are numbered as in Table 2, but character-states are converted to letters (a = 1, b POLYGYRID SHELL EVOLUTION 101 de mer oil FIG. 15. Shell variation discounted as characters: a, low-to-high spires are documented within Patera s.s.; b, rounded-to-angulate-to-keeled peripheries occur within Xolotrema s.s., X. (Wilcoxorbis), and Patera s.s.; c, covered-to-open umbilici are known within both Patera (Vesperpatera) and Mesodon s.s.; d, smooth- to-ribbed sculpture appears within Xolotrema s.s., Triodopsis s.s., Г. (Pilsbryorbis), and T. (Haroldorbis); e, bald-to-hirsute sculpture occurs within Xolotrema s.s., Inflectarius (Summinflectarius), and Fumonelix; f, unicolor-to-color-banded shells are known within both Mesodon s.s. and M. (Aphalogona). App. sayana tended to have a smaller shell than Al. profunda. The northern Midwest was not extensively surveyed during 1982 and 1983 fieldwork, however, so closer conver- gences in sympatry on the umbilicate shell form may exist in that region. Least close—but still remarkable, espe- cially in apertural-barrier construction—of the four polygyrid shell-form convergences in sympatry (Fig. 1) is on the tridentate form. The closest field-documented case of triden- tate convergence in sympatry comprises the mesodontin Inflectarius (Inflectarius) inflectus and the triodopsin Triodopsis (Triodopsis) fal- 102 ЕМВЕАТОМ FIG. 16. Shell-based consensus tree of polygyrid subgenera resulting from cladistic analysis of data in Fig. 14. POLYGYRID SHELL EVOLUTION 103 Lengthy courtship and mating Intertwining of penes External deposition of sperm | mass on mates everted penis | 2 < EN K Duct of \ gametolytic gland vas №7 DEFERENS () | ] RETRACTOR PENIS Brief courtshipand mating Insertion of penes Internal deposition of sperm mass in spermathecal duct RETRACTOR RETENTOR MUSCLE FIG. 17. Closest known convergence in sympatry on the flat shell form: Patera laevior (left) and Xolotrema fosteri (right) from Hawesville, Hancock County, Kentucky. Center: external anatomies and shells in two views. Bottom: dissected reproductive anatomies. Sides: mating-behavioral differences according to Webb (cited in Emberton, 1994a). lax, which | collected together under the same log in dense woods, Vinton County, Ohio, in 1979 (vouchers at Field Museum of Natural History, Chicago). The ranges of these two species overlap moderately in the central Midwest, and other cases of sympa- try are documented— sometimes as inadvert- ently mixed lots—in the collection of the Academy of Natural Sciences of Philadelphia (Emberton, unpublished). Shell Barriers and Water Loss Table 3 gives the results of the experiment on Triodopsis tridentata. A total of 280 “snail- hours” were recorded. Snails from which ap- ertural barriers had been removed lost water by evaporation faster than those with barriers left intact: 27% faster when inactive (re- tracted into the shell) and 9% faster when active (extended from shell). 104 ЕМВЕАТОМ Snails with intact barriers were 83% more successful than those from which barriers had been removed in forming complete or partial epiphragms. Epiphragms slightly re- duced the rate of evaporative water loss in snails with barriers intact (by 3% for both partial and complete epiphragms). In snails from which barriers had been removed, how- ever, epiphragms greatly reduced the rate of water loss: by 22% for partial epiphragms and by 38% for complete epiphragms. Remaining Conservation Priorities Table 4 gives highest priorities for polygy- rid conservation for each of four, phylogeny- based categories. In the category of radiat- ing, endemic clades, Fumonelix must rank very high; the genus 1$ restricted to the southern Appalachians, and three of its six species are narrow-range endemics that have been officially designated as endan- gered or threatened: F. archeri, F. jonesianus, and F. orestes. In the same category, al- though Mesodon (Akromesodon) contains the widespread and relatively common spe- cies M. normalis, its other two species both have narrow, high-altitude ranges: M. altiva- gus оп summits of the Great Smoky Moun- tains, and M. andrewsae on the summit of Mount Rogers, Virginia. Four species have been listed as high pri- orities for conservation in the category of ex- tremely autapomorphic endemics. The two- species genus Giffordius, endemic. to Colombia’s tiny Isla de Providencia (off the eastern coast of Nicaragua), is unique within the family for its ovoviviparity; Giffordius pin- choti by far is the rarer and most endangered of the two species (Emberton, unpublished). Inflectarius ferrissi, endemic to high eleva- tions of the Smoky Mountains, represents extreme phylogenetic shifts in both shell morphology (Emberton, 1991a, 1991b) and in penial morphology (Emberton, 1991a). Triod- opsis platysayoides, endemic to a few bluffs along the New River Gorge of northeastern West Virginia, U.S.A., also embodies extreme phylogenetic divergences in both shell and penis (Emberton, 1988a). Mesodon chilhow- eensis is remarkable for its gigantic, broadly umbilicate shell, exhibiting nearly regular, log-spiral growth (Emberton, 1994a: fig. 1) and for its extremely long penis, much longer than the diameter of the shell (Emberton, 1991а). Giffordius also deserves high priority for conservation in another category, as relic sis- ter-group to a major clade. According to the general phylgenetic hypothesis (Fig. 8), Gif- fordius is basal to the Polygyrini, hence is sister-group to the remaining Polygyrini, a highly diverse and speciose clade. The conservation category of localities with diverse sympatric convergences 15 led by the northern slope of Pine Mountain, Har- lan County, Kentucky. This 1$ the only known site where convergences on all four polygyrid shell forms (Fig. 1) coexist. There the globose form is represented by Mesodon zaletus and Neohelix albolabris; the umbilicate form by Appalachina sayana and Allogona profunda (mentioned above); the flat form by Patera appressa and Xolotrema denotata; and the tridentate form by /nflectarius inflectus, Tri- odopsis vulgata, and Т. tridentata (Emberton, 19950). This site is also important as North America's most diverse known locality for land snails (Emberton, 19950). Thus, Table 4 lists seven high priorities (in- cluding a double listing of Giffordius) for poly- gyrid conservation, based on phylogenetic criteria. Five of these priorities are currently under protection. The Smoky Mountains are a TABLE 3. Effect of apertural dentition and epiphragm on the rate of evaporative water loss in Triodopsis tridentata. Category Toothed, no epiphragm Toothed, partial epiphragm Toothed, complete epiphragm Toothless, no epiphragm Toothless, partial epiphragm Toothless, complete epiphragm Toothed, after activity Toothless, after activity Total “Snail-Hours” Number of Mean Rate of “Snail-Hours” Water Loss 70 3.86% 59 3.74% 26 3.73% 63 4.90% Z 3.97% 20 3.06% 25 15.80% 10 17.18% 280 POLYGYRID SHELL EVOLUTION TABLE 4. Conservation high priorities for polygyrids, based on four phylogenetic criteria. Criterion High Priority Fumonelix Mesodon (Akromesodon) Radiating, endemic clade Radiating, endemic clade Extremely autapomorphic endemic Extremely autapomorphic endemic Extremely autapomorphic endemic Triodopsis platysayoides Extremely autapomorphic Mesodon Giffordius pinchoti Inflectarius ferrissi endemic chilhoweensis Relic sister-group to major clade Giffordius Diverse sympatric four shell-form convergences convergences 105 #Spp Locality Protected? 6 Southern Blue Ridge, U.S.A. Yes 3 Southern Blue Ridge, U.S.A. Yes 1 Isla de Providencia, Colombia No 1 High Smoky Mountains, U.S.A. Yes 1 New River Gorge, U.S.A. Yes 1 Smoky Mountains, U.S.A. Yes 2 Isla de Providencia, Colombia No 11 Pine Mountain, Kentucky, U.S.A. No U.S. National Park and International Bio- sphere Reserve, protecting four species of Fumonelix, two species of Mesodon (Akrome- sodon), Inflectarius ferrissi, and Mesodon chilhoweensis. The remaining Fumonelix are protected in U.S. National Forests, and the remaining M. (Akromesodon) is protected by Mount Rogers State Park. Triodopsis platysayoides is somewhat protected in Coo- pers Rock State Forest, West Virginia, U.S.A. Two high-priority sites remain unprotected. There is a “Pine Mountain State Park,” Bell County, Kentucky, but it is about 64 km away from and has a much lower diversity than the Harlan-County Pine Mountain site listed in Ta- ble 4 (Emberton & Petranka, unpublished). Thus, the Pine Mountain site, U.S.A., is un- protected. Also unprotected 1$ the conservationally important, small Isla de Providencia, Colom- bia. In 1987, both species of Giffordius were still surviving on the island in remnant patches of forest, primarily at higher eleva- tions, but deforestation had already de- stroyed the type locality of G. pinchoti and seemed to be rapidly advancing up the cen- tral peak (Emberton, 1992, unpublished). DISCUSSION General Phylogenetic Hypothesis/Revision This hypothesis/revision culminates the author's 15 years of work on polygyrid sys- tematics, and hopefully provides a replicable data set and a fully testable hypothesis upon which future workers may build. For allozyme data and for more detailed phylogenetic hy- potheses on the Triodopsini and the Meso- dontini, see Emberton (1988a, 1991a, 1994). For a polygyrid biogeographic/historical hy- pothesis, see Emberton (1994a). Clearly, this is not the final word; much remains to be learned. Shell-Based Phylogenetic Analysis/Reliability of Fossils Despite the discovery of many new char- acters, shell-based phylogenetic analysis re- sulted in very low resolution. Two possible sources of additional shell characters were neglected, however: shell-surface micro- sculpture (Emberton, 1995b) and shell ultra- structural layers (Boggild, 1930; Wilbur 8 Saleuddin, 1983; Roth, 1987; Watabe, 1988). Both of these should be investigated. Quali- tative character analysis of x-ray data failed to detect a difference in whorl expansion rate between the Triodopsini and the Mesodontini suggested by quantitative analysis (Ember- ton, 1994a). Thus, more subtle analysis of x-rays could yield more characters for phylo- genetic resolution. Based on current data, however, “shells do not tell” the estimated 145 million years of phylogenetic history of the Polygyridae in North America (Emberton, 1994a). This implies that identification of pre- Miocene polygyrid fossils may be very diffi- cult at best. Thus, based on the general phylogenetic hypothesis/revision, convergences in shell morphology were rampant within the Poly- gyridae. “Kidney-bean” generating curves cropped up in one subgenus each of Poly- 106 EMBERTON дуга and Daedalochila. Whorl-counts of over six hypothetically evolved one or more times each in the Triodopsini, Vespericolini, Stenotremini, Polygyrini, and Mesodontini. Extremely low whorl-expansion rates (< 2.7 per 360-degree rotation) seemingly evolved independently in the Triodopsini, Stenotrem- ini, and Polygyrini. Convergences on a col- umellarly flattened body whorl seem to have occurred in all polygyrid tribes except the Ashmunellini. Expanded umbilici (Fig. 12: char 5e-g) appeared in all tribes but the Ves- pericolini. Rounded, unshouldered umbilical sutures seems to be a good synapomorphy of the genus Triodopsis, but nevertheless this character-state was apparently reversed in the nominal subgenus and seems to have been converged upon in the outgroup family Camaenidae. Slightly flattened umbilical-wall whorls seemingly arose convergently within one or more subgenera each of the Triod- opsini, Allogonini, and Mesodontini. Although transverse-to-spiral basal lamellae in Ash- munella, Daedalochila, and Appalachina were hypothetically plesiomorphic and hence not necessarily convergent, all other forms of basal denticles and lamellae seem to have been converged upon repeatedly in the Poly- gyridae, with the single exception о the com- plete basal lamella, which is a hypothetical synapomorphy of the Stenotremini. Two types of parietal denticles—straight and tri- angular-to-spatulate—seem to be good syn- apomorphies, with additional phylogenetic information, for the Stenotremini and the Polygyrini, although the latter was apparently lost secondarily in Practicolella; a down- curved parietal denticle, however, seems to have evolved independently in all other tribes of the Polygyridae (non-type species of Ves- pericola also have it: Pilsbry, 1940). Both ex- tremes of shell size occur among the out- groups, and various intermediate and small sizes recur repeatedly among polygyrid sub- genera, with only very slight tendencies to- ward trends within and among tribes. Rapid shifts in apertural tilt and expansion rate are apparent synapomorphies for a subset of the Polygyrini, yet both were hypothetically re- versed in type species of some genera. An adnate, callus-like apertural basal lip may seem a perfect synapomorphy for four of Stenotrema's five subgenera, but was appar- ently reversed in S. (Archerelix) barbigerum (Pilsbry, 1940). Regarding apertural palatal denticles, discrete denticles (apertural barri- ers) seem to appear sporadically in nearly all polygyrid tribes; and semidiscrete, basally recessed denticles seem to be only an incon- sistent synapomorphy of the Polygyrini; but a non-discrete ‘‘shelf” seems to be a good sy- napomorphy of the Stenotremini. Closest Convergences in Sympatry Identifications and field verifications of the closest convergences in sympatry on the flat, umbilicate, and tridentate polygyrid iterated shell forms (Fig. 1) provide starting points for analyses of these naturally replicated exper- iments in evolutionary morphology, such as those already conducted on the globose shell form (Emberton, 1994b, 1995a). Although these four cases of polygyrid shell conver- gence in sympatry (Fig. 1; Emberton, 1994a) are the most precise known in North Amer- ica, they are representative of numerous less precise cases involving the same four basic shell forms. Among and within these four shell forms, there are other examples of con- vergence (Emberton, 1988a, 1991a, 1994a), the most striking (and informative for polygy- rid evolution) of which is between /nflectarius ferrissi and Neohelix dentifera (Emberton, 1991b). Shell Barriers and Water Loss The experimental results from Triodopsis tridentata suggest that apertural barriers re- duce the rate of evaporative water loss both directly and indirectly, by aiding in the forma- tion of an epiphragm. The manner in which barriers and epiphragms retard water loss may be counter-intuitive, judging from Ram- say's (1935) experiments on evaporation rates from vertical tubes. Ramsay observed faster water loss when the evaporating sur- face was at the bottom of the tube than at the top, but found that the presence of a perme- able occluder tended to diminish the differ- ence. He hypothesized therefore that “when [the evaporating surface] is at the bottom... there is an upward current of moist air which sets up a circulation in the system.” Ram- say’s hypothesis has not been tested, to my knowledge. Extending his hypothesis to land snails, when a snail withdraws deeply (to es- cape predation, for example) into the coiled tube of its shell, its apertural barriers (and the epiphragm they aid in forming) may function not so much to physically block diffusing wa- ter vapor as to interrupt the convection cur- POLYGYRID SHELL EVOLUTION 107 rents that normally result from tubular evap- oration. The trend for apertural barriers in 7. triden- tata and other triodopsins to be larger in moister habitats (Emberton, 1988a) suggests that their demonstrated retardation of evap- orative water loss is not their only—or even primary—function. Other hypothesized func- tions are given in the Introduction. Polygyrids provide some fascinating ex- amples of apertural obstruction that could be used to test these and other hypotheses. Fig- ure 2 presents some extreme examples from several clades, most of which have arisen by evolutionary convergence. The most extreme cases are in the genus Stenotrema; it is a memorable experience to watch one of these snails extend, its body “pouring” like molas- ses through the slit-like, convoluted aperture. This slit seems to be at its most narrow in $. (Pilsbrelix) uncifera, but is augmented in S. (Stenotrema) maxillatum (both in Fig. 2) by recession of the basal lamella, which lacks a notch and which 1$ overlapped by the parietal lamella to form a three-dimensional baffle. Species of Daedalochila have some com- plex apertural conformations, the most bi- zarre of which are the deeply recessed pari- etal scoop in D. (Upsilodon) hippocrepis and the convoluted apertural lip incorporating the parietal scoop and augmented by recessed denticles in D. (Daedalochila) uvulifera (both in Fig. 2). Other notable examples occur in Triodopsis, Lobosculum, Ashmunella, Linisa, Lobosculum, and Inflectarius (some are Шиз- trated in Figs. 1 and 2). Remaining Conservation Priorities The evaluation of conservation priorities presented in this paper is doubtless biased by the author's limited experience with the western and subtropical species. Table 4 15 intended only as a preliminary guideline and as a possible format for future, more enlight- ened assessments. Given these caveats, Table 4 suggests that most phylogenetically important polygyrids are reasonably well protected against extinc- tion due to habitat loss. How well they and other polygyrid species will withstand envi- ronmental degradation due to acid rain (Graveland et al., 1994), local extinctions of forest-canopy tree species (Getz & Uetz, 1994), and global warming, remain to be dis- covered, however. Unfortunately, there is no long-term monitoring study being conducted on any polygyrid population or community, to my knowledge. Two remaining conservation priorities do demand rapid attention, however: Isla de Providencia, Colombia, and Pine Mountain, Harlan County, Kentucky, U.S.A. (Table 4). The high conservation importance of these sites almost assuredly applies to many other, lesser-known, leaf-litter/soil invertebrates that polygyrids represent. ACKNOWLEDGEMENTS Supported in part by National Science Foundation grants BSR-8700198 and DEB- 9201060 and by Academy of Natural Sci- ences of Philadelphia (ANSP) discretionary funds. | am also grateful to S. Schaefer and G. Bohlke for access to the x-ray facility, Ich- thyology Department, ANSP; to E. Gitten- berger for steering me to the Christelow and Falkner papers; and especially to T. Pearce, B. Roth, and an anonymous reviewer for very useful comments on a previous draft of the manuscript. | would like to take this opportu- nity to thank fellow polygyrid workers who have generously helped me with this and other projects during the past 15 years: T. Asami, K. Auffenburg, N. Babrakzai, A. Bogan, J. Burch, R. Caldwell, H. L. Fair- banks, H. Feinberg, Н. В. Foster, В. Fulling- ton, G. Goodfriend, F. W. Grimm, L. Hubricht, E. Keferl, G. Long, C. Mather, R. Maze, G. McCracken, W. Miller, J. Murray, R. Neck, T. Pearce, J. Petranka, W. Pratt, R. Reeder, B. Roth, R. Selander, the late A. Solem, A. Stiven, В. Taylor, Е. Thompson, J. Vagvolgyi, A. Van Devender, W. Van Devender, and G. Webb. LITERATURE CITED ARCHER A. F., 1948, Land snails of the genus Stenotrema in the Alabama region. University of Alabama Museum Paper, 28: 1-85. ASAMI, T., 1988, Temporal segregation of two sympatric species of land snails. Venus, 47: 278-297. ASAMI, T., 1993, Divergence of activity pattern in coexisting species of land snails. Malacologia, 35: 399-406. AUFFENBERG, K. & R. W. PORTELL, 1990, A new fossil land snail (Gastropoda: Pulmonata: Polygyridae) from the Middle Miocene of north- ern Florida. Nautilus, 103: 143-148. BINNEY, W. G., 1878, The terrestrial air-breathing 108 ЕМВЕАТОМ mollusks of the United States and the adjacent territories of North America, vol. V. Bulletin of the Museum of Comparative Zoology, 4: 1-449, plates 1-16. BOETTGER, C. R., 1921, Carabus morbillosus Fabr. und Otala tigri Gerv., eine Studie ge- genseitiger Anpassung. Abhandlungen der Senckenberger Naturforschung Gesellschaft, 37: 321-325. BOETTGER, C. R., 1935, Die Abwehrder Land- schnecken gegen Käfer. Seitberichter der Ge- sellschaft Naturforschender Freundezu Berlin, 1935: 93-102. BOGGILD, ©. B., 1930, The shell structure of the mollusks. D. Konliga Danske Videnskabernes Selskab, Selskabs Skrifter, Naturvidenskabelig og Mathematisk Afdelig, 9: 1-325, plates 1-15. BROOKS, D. R. & D. A. MCLENNAN, 1991, Phy- logeny, ecology and behavior: a research pro- gram in comparative biology. Univ. of Chicago Press, Chicago, 434 pp. CHRISTELOW, A. Q., 1992, The morphology and function of the clausilium of Clausilia bidentata (Ström) (Gastropoda, Pulmonata, Clausiliidae). Pp. 107-113 in: E. GITTENBERGER 4 J. GOUD, eds., Proceedings of the Ninth International Malaco- logical Congress. Unitas Malacologica, National Museum of Natural History, Leiden, The Nether- lands. COOK, A. H., 1895, Molluscs. Pp. 1-459 in: A. T. HARMER & A. E. SHIPLEY, eds., The Cambridge nat- ural history. Volume 3. Macmillan & Co., New York. EMBERTON, K. C., 1981, Ecological notes on two sympatric, conchologically convergent polygyrid snails in Ohio. Bulletin of the American Malaco- logical Union, 1980: 27-30. EMBERTON, K. C., 1986, The evolution of multiple sympatric homeomorphy among three genera of land snails. Unpublished D. Phil. Thesis, Univer- sity of Chicago. EMBERTON, К. C., 1988a, The genitalic, allozymic, and conchological evolution of the eastern North American Triodopsinae (Gastropoda: Pulmo- nata: Polygyridae). Malacologia, 28: 159-273. EMBERTON, К. C., 1988b, Shell variation in a pop- ulation of Polygyra septemvolva (Pulmonata: Polygyridae). Proceedings of the Academy of Natural Sciences of Philadelphia, 140: 285-294. EMBERTON, К. C., 1991a, The genitalic, allozymic, and conchological evolution of the Mesodontini, trib. nov. (Gastropoda: Pulmonata: Polygyridae). Malacologia, 33: 71-178. EMBERTON, К. C., 1991b, Ecology of a shell con- vergence between subfamilies of polygyrid land snails. Biological Journal of the Linnean Society, 44: 105-120. EMBERTON, К. C., 1991c, Polygyrid relations: A phylogenetic analysis of 17 subfamilies of land snails (Mollusca: Gastropoda: Stylommato- phora). Zoological Journal of the Linnean Soci- ety, 103: 207-224. EMBERTON, K. C., 1992, Implications of a polygy- rid genera phylogeny. American Malacological Union, Program and Abstracts, 1992. EMBERTON, K. C., 1994a, Polygyrid land-snail phylogeny: external sperm exchange, early North American biogeography, iterative shell ev- olution. Biological Journal of the Linnean Soci- ety, 52: 241-271. EMBERTON, K. C., 1994b, Partitioning a morphol- ogy among its controlling factors. Biological Journal of the Linnean Society, 53: 353-369. EMBERTON, K. C., 1995a, Sympatric convergence between two land snails and its environmental correlation. Evolution, in press. EMBERTON, K. C., 1995b, Microsculptures of convergent and divergent polygyrid land-snail shells. Malacologia, in press. EMBERTON, K. C., 1995c, Land-snail community morphologies of the highest-diversity sites of Madagascar, North America, and New Zealand, with recommended alternatives to height-diam- eter plots. Malacologia, 36: 43-66. FALKNER, G., 1984, Múndungsbarrieren bei juve- nilen Orculella bulgarica als Schutz vor ráub- erischen Arthropoden—Ein Modell für die Evolu- tion der pentodonten Mündung? Mitteilungen der Deutschen Malakozoologischen Gesell- schaft, 37: 38-41, tafel 1. FARRIS, J. S., 1988, HENNIG86 Reference, Ver- sion 1.5. J. $. Farris, 41 Admiral Street, Port Jef- ferson Station, NY 11776. GETZ, L. L. 4 G. W. UETZ, 1994, Species diversity of terrestrial snails in the southern Appalachian Mountains, U.S.A. Malacological Review, 27: 61-74. GOODFRIEND, G. A., 1986, Variation in land-snail shell form and size and its causes: a review. Sys- tematic Zoology, 35: 204-223. GRAVELAND, J. R., J. H. BALEN & А. J. NOORD- WIJK, 1994, Poor reproduction in forest passe- rines from decline of snail abundance on acidi- fied soils. Nature, 31: 446-448. HUBRICHT, L., 1985, The distributions of the na- tive land mollusks of the eastern United States. Fieldiana, Zoology, New Series, 24: 1-191. METCALF, А. L. & D. H. RISKIND, 1979, Four new species of Polygyra (Gastropoda: Pulmonata: Polygyridae) from Coahuila, Mexico. Proceed- ings of the Biological Society of Washington, 91: 815-827. PAUL, C. R. C., 1974, Azeca in Britain. Journal of Conchology, 28: 155-172. PILSBRY, H. A., 1930a, Results of the Pinchot South Sea Expedition,—l. Land mollusks of the Caribbean Islands, Grand Cayman, Swan, Old Providence and St. Andrew. Proceedings of the Academy of Natural Sciences, 81: 221-261. PILSBRY, H. A., 1930b, Anatomy and relationships of some American Helicidae and Polygyridae. Proceedings of the Academy of Natural Sci- ences, 82: 303-327. PILSBRY, H. A., 1936, Praticolella martensiana. Nautilus, 49: 140. PILSBRY, H. A., 1940, Land Mollusca of North POLYGYRID SHELL EVOLUTION 109 America (North of Mexico), Volume 1, Part 2. Monographs of the Academy of Natural Sci- ences, Philadelphia, 3: 575-994. PRATT, W. L., 1979, A phenetic classification of the land snail family Polygyridae. Western Soci- ety of Malacologists Annual Report, 11: 12. РВАТТ, W. L., 1981a, A revision of the land snail genus Polygyra in Texas. Unpublished D. Phil. Thesis, University of Arizona, Tucson. PRATT, W. L., 1981b, A revision of the land snail genus Polygyra in Texas. Dissertation Abstracts International, 42: 1352-B. QUEIROZ, К. DE & J. GAUTHIER, 1990, Phylogeny as a central principle in Taxonomy: phylogenetic definitions of taxon names. Systematic Zoology, 39: 307-322. RAMIREZ, R. L., 1993, A generic analysis of the family Systrophiidae (Mollusca: Gastropoda): Taxonomy, phylogeny and biogeography. Un- published Masters Thesis, University of Kansas. RAMSAY, J. A., 1935, Methods of measuring the evaporation of water from animals. Journal of Experimental Biology, 12: 355-372. REES, W. J., 1964, A review of breathing devices in land operculate snails. Proceedings of the Mal- acological Society of London, 36: 55-67. RICHARDSON, L., 1986, Polygyracea: catalog of species (parts 1, Polygyridae; 2, Corillidae; 3, Sagdidae). Tryonia, 13: 1-139, 1-40, 1-38. ROTH, B., 1987, Structure of internal varices in the shell of Micrarionta rufocincta and other Helm- inthoglyptidae (Gastropoda: Pulmonata). Bulletin of the Southern California Academy of Sciences, 86: 145-152. ROTH, B., 1995, Homoplastic loss of dart appara- tus, phylogeny of the genera, and a phylogenetic taxonomy of the Helminthoglyptidae (Gas- tropoda: Pulmonata). Cladistics, in press. ROTH, В. & К. С. EMBERTON, 1994, “Extralimital” land mollusks (Gastropoda) from the Deep River Formation: evidence for mesic medial Tertiary climate. Proceedings of the Academy of Natural Sciences of Philadelphia, 145: 93-106. ROTH, B. & W. B. MILLER, 1992, A new genus of polygyrid land snail (Gastropoda: Pulmonata) from Oregon. The Veliger, 35: 22-223. ROTH, B. & W. B. MILLER, 1993, Polygyrid land snails, Vespericola (Gastropoda: Pulmonata), 1. Species and populations formerly referred to Vespericola columbianus (Lea) in California. The Veliger, 36: 134-144. SOLEM, A., 1972, Microarmature and barriers in the apertures of land snails. Veliger, 15: 81-87. SOLEM, A., 1974, The shell makers: introducing mollusks. New York: Wiley-Interscience. SOLEM, A., 1976, Comments on eastern North American Polygyridae. Nautilus, 90: 25-36. VAGVOLGYI, J., 1968, Systematics and evolution of the genus Triodopsis (Mollusca: Pulmonata: Polygyridae). Bulletin of the Museum of Compar- ative Zoology, 136: 145-254. WATABE, N., 1988, Shell structure. Pp. 69-104 in: Е. В. TRUEMAN & М. В. CLARKE, eds., The Mollusca, volume 11, form and function. Academic Press, New York. WEBB, G. R., 1948, Comparative observations on the mating of certain Triodopsinae. Nautilus, 61: 97-103. WEBB, G. R., 1954, The life-history and sexual anatomy data on Ashmunella with a revision of the triodopsin snails. Gastropodia, 1: 13-18. WEBB, G. R., 1959, Pulmonata, Polygyridae: Notes on the sexology of Triodopsis, a new sub- genus, Haroldorbis, and a new section, Shelfor- dorbis. Gastropodia, 1: 23-25. WEBB, G. R., 1965, On the sexology of Trilobopsis loricata sonomaensis (Hemphill). Gastropodia, 1: 58-60. WEBB, G. R., 1970, Fragmentary observations on sexology of Cryptomastix hendersoni Pilsbry and C. magnidentata Pilsbry and a new subgenus (Pulmonata, Polygyridae, Asmunellinae). Gas- tropodia, 1: 77-78. WEBB, G. R., 1974, The sexual evolution of the polygyrid snails. Gastropodia, 1: 85-90. WEBB, С. R., 1990, Photographs of the copulation of Cryptomastix (Bupiogona) hendersoni. Gas- tropodia, 2: 22. WILBUR, К. М. & А. $. М. SALEUDDIN, 1983, Shell formation. Pp. 235-287 in: А. $. М. SALEUDDIN & К. м. WILBUR, eds., The Mollusca, volume 4, Physi- ology, part 1. Academic Press, New York. WILEY, E. O., 1981, Phylogenetics: the theory and practice of phylogenetic systematics. John Wiley and Sons, New York, 439 pp. WILEY, E. O., D. SIEGEL-CAUSEY, D. R. BROOKS 8 V. A. FUNK, 1991, The compleat cladist: a primer of phylogenetic procedures. University of Kansas Museum of Natural History Special Pub- lication No. 19, Lawrence, 158 pp. ZILCH, A., 1959-1960, Gastropoda. Teil 2. Euthy- neura. Band 6, 6. Pp. 1-834 in: O. H. SCHINDE- WOLF, ed., Handbuch der Palaozoologie. Ge- brüder Borntráger, Berlin. Revised MS. accepted 10 January 1995 MALACOLOGIA, 1995, 37(1): 111-122 GEOGRAPHIC DISTRIBUTION AND SHELL COLOUR AND BANDING POLYMORPHISM IN MARGINAL POPULATIONS OF СЕРАЕА NEMORALIS (GASTROPODA, HELICIDAE) A. Honék Department of Entomology, Research Institute of Plant Production, Ruzyné 507, 161 06 Praha 6, Czech Republic ABSTRACT Occurrence and colour and shell banding polymorphism were investigated in populations of Cepaea nemoralis (L.) at the edge of the species’ geographic distribution in the Czech Re- public. Cepaea nemoralis populations were found only at intravillan localities. The great majority of town and village localities were occupied by C. nemoralis in three isolated areas: a northern one integrated with the main geographic area of the species, and two isolated southern insular areas. А few populations were scattered at localities between these areas and further south, where C. nemoralis was replaced by Cepaea hortensis (Múller). The C. nemoralis distribution area and abundance may have increased since 1950. | speculate that human activity encour- aged the spread of С. nemoralis populations, and this species competitively exterminated С. hortensis populations from the intravillan habitats. The frequency of shell colour and banding morphs in local populations was similar to those in oceanic Western Europe (56.2 + 18.6% of pink shell colour), and may be affected by climatic selection and random drift. There exist small areas marked by a high frequency of 00000, 00300, 00345 and 12345 shell banding morphs. Their edges were mostly not concordant with areas of geographic distribution. INTRODUCTION Cepaea nemoralis (L.) is a West-European species distributed from southern Scandina- via, Lithuania and the Ukraine in the east, to Hungary and the northern Balkan Peninsula in the south (Schilder 8 Schilder, 1953, 1957). However, it is absent in the highlands of cen- tral Europe, in the territories of the Czech Re- public and Slovakia. Only Bohemia (the west- ern half of the Czech Republic) is crossed by the edge of the area of species’ continuous distribution (Fig. 1). From this region Lozek (1956) listed a number of localities scattered mostly north of the Labe (Elbe) River where C. nemoralis was found chiefly at intravillan sites. However, the precise borderline of the species distribution is impossible to trace from the published data. The shell colour and banding polymorphism in Bohemian pop- ulations has never been quantitatively inves- tigated despite the fact that variation in marginal populations is of considerable the- oretical interest. Another common west European species is Cepaea hortensis (Múller). It lives in low- land and submontane areas of the whole ter- ritory of the Czech Republic, at both intravil- lan sites and in the open landscape (Lozek, 111 1956). There exist indications (see Discus- sion) that each Cepaea species may exclude the other from occupying the same site. A precise delineation of the geographic distri- bution of intravillan populations of C. horten- sis and С. nemoralis may contribute to test- ing this assumption. A precise study of C. hortensis distribution in Bohemia is also not available. The aim of this study was: (1) mapping the edge of the C. nemoralis distribution, (2) re- cording shell colour and banding polymor- phism in local populations, and (3) investigat- ing the possible interaction of C. nemoralis with C. hortensis populations. Distribution and polymorphism т С. nemoralis have been the subject of numerous studies (Jones et al., 1977; Lamotte, 1988). The present work 1$ justified by the fact that the occurrence and variation were never investigated in the very eastern edge of the С. nemoralis area of dis- tribution. MATERIAL AND METHODS The populations of С. nemoralis and С. hortensis were sampled systematically at in- travillan and open landscape habitats of НОМЕК Lae ‘facing page. A = Ф jo) 2 Ф — © Ф 2 7) 19) © ar — ( \ Figurgl 1 b ee {8 se Оэ ОФ 70. o) o 8 611 © и о N и 9 о © ЗЕ. 58 ze] У3371Г УЛУ ЛА GEOGRAPHIC DISTRIBUTION AND SHELL COLOUR 113 northern Bohemia in 1989-1991. Towns and villages where Cepaea was found are shown in Figure 1. At some localities, the snails were sampled at several sites at least 100 m apart. The shell colour was classified as pink or yellow, although both colours varied largely in saturation and hue. In some populations, a large proportion of shells was very pale. п this case, | lumped all specimens that had even a slight trace of pink colour (usually at the top of the shell). These were classified as “pink,” the rest of the animals was classified as “yellow.” For recording the shell banding polymorphism, | used the commonly ac- cepted notation. Visible bands are indicated from the dorsal side by numbers 1-5, missing bands as 0, and fusions between bands are indicated by brackets (). The sample size varied between 6 and 258. Proportions of col- our and shell banding forms were calculated for populations where at least 15 snails were collected. The rare hyalozonate individuals (with pale blanks instead of black bands), and animals with bands having diluted margins (common at some localities) were classified as the respective band morphs. Evaluating the geographic variation, | compared the pro- portions of 00000, 00300, 00345, and 12345 morphs, regardless of confluences between bands. According to the frequency of a morph the localities were ranked into four categories: (1) zero frequency of the morph, (2) low frequency—quartile 1 of the series of localities where the morph was present ar- ranged in ascending order of the morph per- centage, (3) medium frequency—quartile 2 of the above series, (4) high frequency—quar- tiles 3 and 4 of the above series. Details of the frequencies of colour and shell banding morphs at different localities FIG. 1. The localities of C. nemoralis (solid circles), C. hortensis (open circles), and of mixed populations of both species (divided circles). The areas of “continuous” distribution of С. nemoralis are delimited by heavy lines: above—Liberec LB, left—Litomérice LT, right—Novy Bydzov NB. The area of “scattered” distribution is fenced by a dotted line. Insert: Area shown in Fig. 1 projected onto the map of the Czech Republic. The edge of the C. nemoralis distribution is shown by a solid line (inside the Czech Republic) and a dashed line (after Schilder & Schilder, 1953, 1957). : Localities of С. nemoralis (small figures): 1 Libochovany, 2 Zalhostice, 3 Litoméfice, 4 Tfeboutice, 5 LibéSice, 6 Mlékojedy, 7 Zeletice, 8 Pocaply, 9 Terezín, 10 Nové Kopisty, 11 BohuSovice, 12 Keblice, 13 Doksany, 14 Lobendava, 15 Lipová, 16 Sluknov, 17 Mikulásovicky, 18 MikulaSovice, 19 Brtniky, 20 Rum- burk, 21 Krasná Lipa, 22 Studanka, 23 Dolní Podluzí, 24 Jifetín pod Jedlovou, 25 Rybnisté, 26 Chribska, 27 Jetfichovice, 28 Kamenicky Senov, 29 Volfartice, 30 Novy Bor, 31 Chotovice, 32 Janov, 33 Sloup, 34 Castolovice, 35 ManuSice, 36 Pise@na, 37 Ceska Lipa, 38 Staré Splavy, 39 Doksy, 40 Mafenice, 41 Marenicky, 42 Jablonné у Podjestédí, 43 Mimon, 44 Stráz pod Ralskem, 45 Hamr, 46 Brevnisté, 47 Озебпа, 48 Kfizany, 49 Ves, 50 Andélka, 51 ViSnova, 52 Minkovice, 53 Víska, 54 Kunratice, 55 Srbská, 56 Jindri- chovice pod Smrkem, 57 Nové Mésto pod Smrkem, 58 Dolní Rasnice, 59 Krásny Les, 60 Arnoltice, 61 Frydlant у Cechách, 62 Raspenava, 63 Hejnice, 64 Bily Potok, 65 Chrastava, 66 Liberec, 67 Vratislavice, 68 Janov nad Nisou, 69 Josefüv Dül, 70 Lucany, 71 Tanvald, 72 Velké Hamry, 73 Drzkov, 74 Zásada, 75 Vrkoslavice, 76 Dalesice, 77 Pulecny, 78 Rychnov и Jablonce nad Nisou, 79 Rádlo, 80 Hodkovice nad Mohelkou, 81 Cesky Dub, 82 Miliceves, 83 Slatina, 84 Vrbice, 85 HradiSt’ko, 86 Vysoké Veselí, 87 Cho- mutice, 88 Ostroméf, 89 Нойсе у PodkrkonoSi, 90 Lístkovice, 91 Кпёйсе, 92 Zlunice, 93 Sekeïice, 94 Smidary, 95 Smidarská Lhota, 96 Janovice, 97 Hlusice, 98 Stary Bydzov, 99 Novy BydZov, 100 Vysoéany, 101 Prasek, 102 Mystéves, 103 Petrovice, 104 Suchá, 105 Staré Nechanice, 106 Nechanice, 107 Boharyné, 108 Skochovice, 109 Luzec, 110 Nepolisy, 111 Mlékosrby, 112 Chlumec nad Cidlinou, 113 Mnichovo HradiSté, 114 Sedlist'ka, 115 Turnov, 116 Zelezny Brod, 117 Bozkov, 118 Sobotka, 119 Nova Рака, 120 Vrchlabi, 121 Hostinné, 122 Miletin, 123 Pardubice, 124 Libcany, 125 Hradec Králové, 126 Nedélisté, 127 Ceská Skalice, 128 Olivetin, 129 Broumov. o Localities of С. hortensis (large figures): 1 Dolní Habartice, 2 Benesov nad Ploucnicí, 3 Zandov, 4 Stvolinky, 5 Kravaïe, 6 USték, 7 Zahradky, 8 Jestrebi, 9 Holany, 10 Dfev£ice, 11 Спит, 12 Vrchovany, 13 Dubá, 14 Pavlicky, 15 ZakSin, 16 Polepy, 17 Host'ka, 18 Snédovice, 19 Stéti, 20 Béla pod Bezdézem, 21 Klaster Hradisté, 22 Kosmonosy, 23 Mladá Boleslav, 24 Dobrovice, 25 Dolni Bousov, 26 Liban, 27 Chylice, 28 Kostelec, 29 Kopidino, 30 Cesov, 31 Chroustov, 32 Chotusice, 33 Dymokury, 34 Záhornice, 35 Méstec Králové, 36 Lovüice, 37 Svijansky Ujezd, 38 Hubálov, 39 RadoSovice, 40 Jesenny, 41 Jablonec nad Jizerou, 42 Rovensko pod Troskami, 43 Lomnice nad Popelkou, 44 LibStät, 45 Jilemnice, 46 Horní Branná, 47 Jicín, 48 Sárovcova Lhota, 49 Lázné Bélohrad, 50 Borovnice, 51 Dolní Kalná, 52 Lanzov, 53 Dvür Králové, 54 Horicky, 55 Chvalkovice, 56 ReSetova Lhota, 57 Jaromér, 58 Semonice, 59 Cernozice, 60 Smifice, 61 Probluz, 62 Stézery, 63 Roudnice, 64 Kosicky, 65 Dobfenice, 66 Chyst', 67 Rohovládova Béla, 68 Holice, 69 Cernilov, 70 Jasenná, 71 Оробпо, 72 Tyniste nad Orlici, 73 Vamberk, 74 Praha, 75 KarlStejn. 114 HONEK with an analysis of linkage disequilibria will be published in a separate paper (Honek, in prep.). The data may be also obtained on re- quest from the author. RESULTS The Habitats of Cepaea populations In Bohemia, typical habitats of C. nemora- lis are intravillan areas. Most often the snails were found at ancient or working cemeteries, particularly along the south- and west-facing walls. Typical localities were also the railway stations surrounded by exuberant weedy vegetation. Other sites often populated by C. nemoralis included small gardens, weed stands bordering the margins of quiet streets, and intravillan shrubs on south-fac- ing slopes. Many sites were periodically dis- turbed by human activities: soil cultivation, mowing, herbicide application, transport or building activities. Despite an intensive search, по С. nemoralis populations were found outside the intravillan areas. Populations of С. hortensis in towns and villages (Fig. 1) occurred in habitats similar to those of C. nemoralis. Cepaea hortensis gen- erally preferred less disturbed sites, and its populations were also frequently found in the open landscape (localities not shown in Fig. 1). The preferred rural habitats were hedge- rows and shrubs, south-facing slopes with mixtures of dicotyledonous and grassy veg- etation, and roadside ditches. The distribution of Cepaea populations Cepaea nemoralis was found at 129 town and village localities of northern Bohemia (Fig. 1). Most localities were grouped in three areas where the species occupied nearly all favourable intravillan sites and was present in most towns and villages. These areas | call “areas of continuous distribution” (Fig. 1, de- limited by solid lines), and refer to by the name of the principal included town. The largest area of Liberec (LB, localities no. 14— 81) extends along the northern Bohemian frontier and is probably a southern protrusion of the area of continuous C. nemoralis distri- bution (Fig. 1, insert). Two smaller “island” areas of continuous distribution are the area of Litomérice (LT, localities 1-13), and the area of Novy Bydzov (NB, localities 82-112). The boundaries of the areas of “continu- ous” distribution could be traced with the precision of a few kilometers, that 1$, the dis- tance that divides the neighbouring villages of which one is populated by C. nemoralis, the other by C. hortensis populations. At the edge of distribution of both species, for ex- ample between Ceská Lípa (Fig. 1, locality 37) and Doksy (locality 39) in the LB area, or along the entire west-south-east edge of the NB area, there exist no prominent geo- graphic structures which may cause the sep- aration of the species. Other portions of the boundary are bordered by areas generally unfavourable for Cepaea (e.g., volcanic or sandstone hills, or large forests). This is also the case along the western and eastern sec- tions of the boundary of the LB area, and the northern sections of the LT and NB areas. The LB and NB areas of “continuous” dis- tribution are separated by an area of mosaic distribution of intravillan С. nemoralis and С. hortensis populations, which | refer to as the area of “scattered” С. nemoralis distribution (Fig. 1, dashed line). Sixteen villages and towns with populations of С. nemoralis were found on this territory. In some towns, for ex- ample Pardubice (123) and Hradec Králové (125), а С. nemoralis population was found at one site, whereas many sites were occupied by C. hortensis. In total, | found only 14 sites where popu- lations of С. nemoralis and С. hortensis lived together at one place. Five mixed popula- tions were found at the area of “scattered” distribution, others at the margins of the LB (two populations) and NB (seven populations) areas of “continuous” distribution. Shell Colour and Banding Polymorphism The large variation in the proportions of col- our morphs had a very localised character. Thus, the extent of variation in the frequency of the pink form was only slightly smaller among 24 sites within the small town of Lito- тёйсе (9.7-78.3%) than among 73 popula- tions of the entire LB area (1.2-98.4%). The average proportion of the pink form was 56.2 + 18.3%. The figures for particular areas of “continuous” distribution differed slightly. The highest frequency of pink was in popu- lations from the LB area, followed by the LT and NB areas (Fig. 2). This difference was correlated with greater average altitude (and cooler climate) of the LB area (localities at GEOGRAPHIC DISTRIBUTION AND SHELL COLOUR 115 230-580 m above sea level) than of the LT (160-230 m a.s.l) and NB (190-240 т a.s.l.) areas. However, within the LB area the regression of the proportion of pink individu- als on altitude was not significant (r° = 0.006, p> 0-1) (Fig: 3): The frequency of shell banding 00000, 00300, 00345 and 12345 morphs in local populations (Fig. 2) showed typical area ef- fects. The distribution of frequencies of the 00000 morph was right-skewed and popula- tions with a high frequency of > 17% were found in the LT and LB areas and also at the eastern margin of the area of “scattered” species distribution (Fig. 4). The populations with high frequency > 46.5%, of the 00300 morph were aggregated in the LT and NB areas and also in the north of LB area (Fig. 5). The populations with high frequency of > 6.6% of 00345 morph were aggregated in the LT area and the north LB and west NB areas (Fig. 6). The distribution of the 12345 morph was complementary to distribution of other morphs. Populations with high fre- quency of > 80.0% of this morph were ag- gregated in the south of the LB area. The fraction of fused band phenotypes within the 12345 morph varied among popu- lations. The frequencies of melanic (123)(45) and (12345) morphs varied between 0.0- 45.2% and 0.0-28.4%, respectively, and were not correlated (in LB area) with the alti- tude of the locality. The 00000 phenotype was associated with pink colour, and yellow specimens were very rare. By contrast, 00300, 00345, and 12345 morphs were not associated with any colour phenotype. Details of linkage disequilibria between colour and shell banding morphs will be discussed elsewhere (Honék, in prep.). DISCUSSION Spreading of C. nemoralis Comparing the distribution of C. nemoralis populations from before 1950 (Lozek, 1956) with this study revealed that most localities reported earlier (localities no. 3, 16, 21, 30, 37, 48, 66, 99, 115) were within the areas of “continuous” distribution. The general pat- tern of С. nemoralis distribution apparently has not changed within the last 50 years, but the species became more abundant since Lozek (1956; personal communication) qual- ified its abundance in the 1950's as “rare.” At present, the species also occupies more lo- calities than it did earlier. The difference be- tween abundance estimates from before the 1950’s and in this study cannot be explained by omissions by earlier authors but probably indicates an increase of the area of species distribution. The change involves southward expansion of the LB area, and radial expan- sion of LT and NB areas. On the other hand, | did not find С. nemoralis at three localities where it was established before the 1950's: Karlovy Vary (Karlsbad) and Zerotín (both outside the area shown in Fig. 1) and USt&k (C. hortensis locality no. 6 in Fig. 1). These localities are now populated by C. hortensis. This indicates extinction of local C. nemoralis populations. The causes of extinction are un- clear, because ordinary human activity does not endanger their survival. The plasticity of the distribution of C. nemoralis at the eastern edge of the species” distribution area con- trasts with its constancy in Western Europe, where some populations have persisted, with little variation in morph frequency, since the neolithic period (Cain & Cook, 1989). The factors of fast spreading of an animal with a limited dispersion capacity are of in- terest. | suppose that human activity may be an important factor in species dispersion. This follows from a frequent occurrence of marginal populations at two typical habitats: railway stations and cemeteries. This distri- bution may indicate the accidental passive transport by man. Climbing onto and falling off the railway coaches may disseminate the snails. In fact, some marginal populations of C. nemoralis (e.g. 113, 116, 119, 125) were found at railway stations in towns otherwise populated by С. hortensis. The occurrence of C. nemoralis in cemeteries could be attrib- uted to the popular exchange of potted plants. The potted plants are usually distrib- uted from large gardens to several surround- ing villages. The soil infested with the eggs of garden populations of C. nemoralis may be- come the vehicle of dissemination. Relationship to C. hortensis In fact, С. nemoralis and С. hortensis rarely occurred at the same place. In several towns 118; ВУ. 88; 112. 116.118, 123, 125. 127),€: hortensis was found at several sites, but С. nemoralis inhabited only one place. These towns are at the margins of the areas of 116 НОМЁК 00000 00300 00345 12345 PINK 40 pe i a 40 15 10 20 20 5 5 30 20 30 60 30 40 20 40 20 20 10 20 10 25 60 60 40 15 40 20 20 5 0 40 80 0 40 80 0 40 80 0 40 80 0 40 80 MORPH FREQUENCY a FIG. 2. The frequency of percentages of shell banding morphs 00000, 00300, 00345, 12345 and pink shell coloration in populations of С. nemoralis. A—Pooled sample; B—Litomérice (LT) area of “continuous” distribution; C—Liberec (LB) area of “continuous” distribution; D—Novy Bydzov (NB) area of “continuous” distribution; E—area of ““scattered” distribution. The boundaries of the areas are shown in Fig. 1. сл PERCENT LOCALITIES en 5 a en a à (== [=>] GEOGRAPHIC DISTRIBUTION AND SHELL COLOUR 117 100 . 4 3 я = s = © ña o > о о №. © e O A AR = 50 e e 5 ы ° e u = e ы a e 3 e 1) e a .° e 2 e 0 . ae ALTITUDE FIG. 3. The regression of the percentage of the pink colour form in populations of Liberec (LB) area on the altitude (m) above sea level of the locality. Regression: у = —0.015x + 63.3, г? = 0.006, р >> 0.1. “continuous” distribution or in the area of “scattered” distribution. The reverse situa- tion when С. nemoralis occurred all over the inside of the town and С. hortensis populated a few suburban sites, was established at Novy BydZov (99), in the centre of NB area of “continuous” distribution. Both examples may show stages of invasion of a new locality by C. nemoralis. At first, C. nemoralis is es- tablished at one place and from there it spreads, exterminating C. hortensis popula- tions. At a given site, the transition is perhaps quick, as mixed populations usually contain a majority of one species, either C. hortensis (28, 83, 85, 86) or C. nemoralis (69, 91, 120). _ There exists some experimental evidence for competition superiority of C. nemoralis over C. hortensis (Cameron 8 Carter, 1979, Tilling, 1985a, b), and competitive exclusion has been proposed in explaining the distribu- tion of both species (Boycott, 1934; Cain, 1983). The situation in Bohemia contrasts with that in Western Europe, where mixed populations may coexist for a long time (Cain 8 Currey, 1963a; Cain, 1983). This difference may be due to the fact that in the western part of its distribution area C. nemoralis lives in the extravillan landscape where C. horten- sis may resist its competition. Also, in Bohe- mia C. hortensis populations live outside the towns where intravillan sites are all occupied by C. nemoralis. The examples of invasion of a locality oc- cupied by one Cepaea species by the other species are few. Well documented 1$ a recent (between 1961 and 1985) substitution of C. nemoralis by С. hortensis on severeal sites at Marlborough Downs, England (Cain 8 Cur- rey, 1963b; Cowie 4 Jones, 1987). The trend observed in this non-urban area may be a consequence of the change of the microcli- mate following the change of the vegetation cover. А dense and tall vegetation may favour the occurrence of C. hortensis, which makes better use of solar radiation and is capable of exploiting shade places. The microclimate might favour also the changes in Bohemian populations because intravillan sites where C. nemoralis probably replaced C. hortensis are generally warmer than the sites in the open landscape occupied by C. hortensis. Colour and Shell Banding Polymorphism The study of C. nemoralis variation is a problem that has resisted final solution de- spite the large body of accumulated data НОМЕК 118 ‘ydiow quid jo uoiuodoid 3sayBiy au} pue 15эмо]| ay] зиэзэл4эл yolym $э|лэ um} Aq pajuesaidai эле pajoa]¡o9 элэм чалош quid au} jo suonuodold jua lap YIM suonendod |елэлэ$ aJaym зэциеэол '(%0:/1<) Аэцепбэд ybiy—ajomo pios “(9%0'/1-S'p) Аэцепбэд uunipau—aj9119 papinip ‘(%у’у-‘0) Aouanbay moj—jods ¡erueo YUM 89119 'AÁcuanbax 0187—28[9119 USO ‘з/елошаи “9 jo suoyejndod и! цалош Buipueq ||эц$ 00000 Jo Ácuanbax ay] “y ‘914 119 GEOGRAPHIC DISTRIBUTION AND SHELL COLOUR y ‘Bld ul se ASIMIBUIO '(%9'9$<) Aouanba.y YBIy—ajould pros ‘(%G' 9P-v' pz) AoUeNbe. wNIPSW— 8919 papinip ‘(%5`уг-0' |) АэцэпБэл mo¡—jods eue, YUM ajomo ‘Aouenbes 018Z—819119 UBdO ‘з/елошаи “9 jo зиоцетаоа ul ydıow Buipueq ||э4$ 00$00 JO Ácuanbax эц ‘$ “DI ‘+ ‘64 ui se эзимец `(%9`9<) fouenbeuy uBly—ejouto pos (99'9-9'E) Aouenbey шпреш-—еюло papinip “(9 p"E-p"0) Аоиепед moj—jods едиео цим 89419 ‘Аоцепьед олег—еолю usado ‘зуелошеи “9 jo suoyeindod и! ydiou Bulpueq 194$ SPEOO Jo Ácuanba, ayı “9 “Old HONEK Figure 6 here; 120 GEOGRAPHIC DISTRIBUTION AND SHELL COLOUR 121 (Schilder & Schilder, 1953, 1957; Jones et al., 1977; Cain, 1983; Lamotte, 1988). Typical features of this polymorphism are very local differences in the proportion of morphs (Wolda, 1969a, b), and the “area effects,” that is high frequency of a phenotype in a limited area (Cain 8 Currey, 1963b). Area ef- fects have been observed for both shell co- lour and banding polymorphism (Khemici et al., 1989; Lamotte et al., 1989; Ratel et al., 1989). The factors of differences in morph proportions may be random population pro- cesses (Lamotte, 1951, 1952) or selection by climatic or substrate factors (Currey & Cain, 1968; Cameron et al., 1977; Khemici et al., 1989; Mazon et al., 1988, 1989) or visual predators (Cain & Sheppard, 1950). Random processes have probably influ- enced the composition of some of our iso- lated populations. Morph frequencies may be influenced by genetic drift when populations were established from a small number of founders. An example may be the population at the railway station at Nová Paka (119), which consisted of 95.6% of pink 00300 in- dividuals. Climatic effects contribute to maintaining the frequencies of shell colour forms. Popu- lations from areas with oceanic climate are mostly pink in contradistinction to the mostly yellow populations of areas with Mediterra- nean climate (Mazon et al., 1988; Vicario et al., 1988). Multivariate analysis of climatic and substratum data revealed that these fac- tors contribute also to maintaining small dif- ferences in shell colour frequencies under oceanic climate (Khemici et al. 1989, Ratel et al. 1989). The average proportion of pink individuals in Bohemian populations (56.2%) was similar to that in oceanic Western Europe (about 60% of the pink form) and far greater than in Mediterranean populations (with only 20- 30% of pink form, Lamotte, 1988; Mazon et -al., 1988). Typical for Bohemian populations is also the absence of the yellow 00000 phe- notype, which is the most adapted one to warm conditions. A demonstration of climatic effects on diversification of morph propor- tions among Bohemian populations would require more microclimatic and orographic data. An indication of such effects was the increased frequency of the pink form in hilly LB compared to lowland LT and NB areas. No explanation has been found for the area effects shown by the 00000, 00300, 00345, and 12345 morphs. The role of visual preda- tion on maintenance of differences between populations is difficult to evaluate. The local- ities of C. nemoralis in Bohemia become cov- ered mostly by sparse vegetation (weeds, ornamental plants), which makes a rather uniform optical background for avian preda- tors. Under such conditions, visual predation is not likely to create important differences in morph proportions among local populations of Cepaea. Anyway, the results do not con- tradict the classical theory of the role of visual selection in maintaining Cepaea polymor- phism (Cain & Sheppard, 1950, 1954). The remains of crushed shells were found at only 15 collection sites. This paucity of crushed shells parallels the results of other studies (e.g. Cowie & Jones, 1987) and may caused by preference of avian predators for juveniles (cf. Wolda, 1972; Wolda & Kreulen, 1973). A comparison of morph frequency in adult crushed shells and in living animals could be made only at one locality (Arnoltice 60). There was no significant difference in proportion of colour and banding morphs among living an- imals and dead shells. ACKNOWLEDGEMENTS | thank Prof. A.J. Cain, Prof. M. Lamotte and Dr. V. Lozek for critical reading of my MS and many valuable comments. LITERATURE CITED BOYCOTT, A. E., 1934, The habitats of land Mol- lusca in Britain. Journal of Ecology, 22: 1-38. CAIN, A. J., 1983, Ecology and ecogenetics of ter- restrial molluscan populations. Pp. 597-647, in: W. D. RUSSELL-HUNTER, ed., The Mollusca. Vol. 6. Ecology, Academic Press, London, New York, San Francisco. CAIN, A. J. & L. M. COOK, 1989, Persistence and extinction in some Cepaea populations. Biolog- ical Journal of the Linnean Society, 38: 183-190. CAIN, А. J. & 4. D. CURREY, 1963a, Area effects in Cepaea on the Larkhill artillery ranges, Salisbury Plain. Journal of the Linnean Society (Zoology), 45: 1-15. CAIN, A. J. & J. D. CURREY, 1963b, Area effects in Cepaea. Philosophical Transactions of the Royal Society of London (B), 246: 1-81. CAIN, A. J. & P. M. SHEPPARD, 1950, Selection in the polymorphic land snail Cepaea nemoralis. Heredity, 4: 275-294. CAIN, A. J. & P. M. SHEPPARD, 1954, Natural se- lection in Cepaea. Genetics, 39: 89-116. 122 HONEK CAMERON, В. А. D. & М. A. CARTER, 1979, Intra- and interspecific effects of population density on growth and activity in some helicid land snails (Gastropoda: Pulmonata). Journal of Animal Ecology, 48: 237-246. CAMERON, В. А. D., P. WILLIAMSON & D. I. MOR- GAN-HUWS, 1977, The habiats of the land snail Cepaea nemoralis (L.) on downland and their ecogenetic significance. Biological Journal of the Linnean Society, 9: 231-241. COWIE, R. H. & J. $. JONES, 1987, Ecological interactions between Cepaea nemoralis and Ce- paea hortensis: competition, invasion but no niche displacement. Functional Ecology, 1: 91- 97. CURREY, J. D. & A. J. САМ, 1968, Studies in Cepaea. IV. Climate and selection of banding morphs in Cepaea from the climatic optimum to the present day. Philosophical Transactions of the Royal Society of London, B, 253: 483-498. JONES, J. S., В. H. LEITH & P. RAWLINGS, 1977, Polymorphism in Cepaea: a problem with too many solutions? Annual Review of Ecology and Systematics, 8: 109-143. KHEMICI, E., J. GENERMONT & М. LAMOTTE, 1989, Relations entre polymorphisme et milieu chez Cepaea nemoralis (Mollusques Pulmonés) de Normandie. Bulletin de la Société Zoologique de France, 113: 115-167. LAMOTTE, M., 1951, Recherches sur la structure génétique des populations naturelles de Cepaea nemoralis. Bulletin Biologique de la France et de la Belgique, Supplement, 35: 1-239. LAMOTTE, M., 1952, Le róle des fluctuations for- tuites dans la structure génétique des popula- tions naturelles de Cepaea nemoralis. Heredity, 6: 333-343. LAMOTTE, M., 1988, Facteurs influencant la diver- sité du polymorphisme de la coquille dans les populations naturelles de Cepaea nemoralis (Moll. Pulmonés). Haliotis, 18: 131-157. LAMOTTE, M., I. KASSEM & J. GENERMONT, 1989, Le polymorphisme de la coquille de Ce- раеа nemoralis (L.) dans la vallée d'Ossau (Pyrénées Francaises). Bulletin de la Société Zoologique de France, 113: 53-63. LOZEK, V., 1956, К/с Ceskoslovenskych mékkySu. [Key to Czechoslovak Mollusca] Vydavatelstvo Slovenskej Akademie Vied. Bratislava. MAZON, L. I., М. А. М. DE PANCORBO, A. VI- CARIO, A. 1. AGUIRRE, A. ESTOMBA & С. М. LOSTAO, 1989, Selection in sympatric popula- tions of Cepaea. Génétique Selection et Evolu- tion, 21: 269-281. MAZON, |. 1. A. VICARIO, М. A. М. DE РАМСОВВО, A. I. AGUIRRE, А. ESTOMBA 4 С. M. LOSTAO, 1988, North/south differentiation in the distribution of Cepaea nemoralis in Spain. Heredity, 61: 189-197. RATEL, М. O., J. GENERMONT & М. LAMOTTE, 1989, Relation entre polymorphisme et milieu chez les Cepaea nemoralis (Moll. Pulmonés) de la region Parisienne. Bulletin de la Société Zoologique de France, 113: 145-154. SCHILDER, F. A. 4 M. SCHILDER, 1953, Die Bán- derschnecken. Eine Studie zur Evolution der Tiere. VEB Gustav Fischer Verlag. Jena. SCHILDER, Е. A. & М. SCHILDER, 1957, Die Bán- derschnecken. Eine Studie zur Evolution der Tiere. Schluss: Die Bánderschnecken Europas. VEB Gustav Fischer Verlag. Jena. TILLING, S. M., 1985a, The effects of density and interspecific interaction on mortality in experi- mental populations of adult Cepaea (Held.). Bi- ological Journal of the Linnean Society, 24: 61- 70. TILLING, S. M., 1985b, The effect of interspecific interaction on spatial distribution patterns in ex- perimental populations of Cepaea nemoralis (L.) and C. hortensis (Mull.). Biological Journal of the Linnean Society, 24: 71-81. VICARIO, A., L. I. MAZON, A. AGUIRRE, A. ES- TOMBA & С. LOSTAO, 1988, Variation in popu- lations of Cepaea nemoralis (L.) in North Spain. Biological Journal of the Linnean Society, 35: 217-227. WOLDA, H., 1969a, Fine distribution of morph fre- quencies in the snail, Cepaea nemoralis near Groningen. Journal of Animal Ecology, 38: 305- 327 WOLDA, H., 1969b, Stability of a steep cline in morph frequencies of the snail Cepaea nemoralis (L.). Journal of Animal Ecology, 38: 623-635. WOLDA, H., 1972, Ecology of come experimental populations of the landsnail Cepaea nemoralis (L.). 1. Adult numbers and adult mortality. Neth- erland Journal of Zoology, 22: 428-455. WOLDA, H. 8 D. A. KREULEN, 1973, Ecology of some experimental populations of the landsnail Cepaea nemoralis (L.). Il. Production and survival of eggs and juveniles. Netherland Journal of Zo- ology, 23: 168-188. Revised Ms. accepted 18 October 1994 MALACOLOGIA, 1995, 37(1): 123-132 KARYOTYPE ANALYSIS AND GENOME SIZE IN THREE MEDITERRANEAN SPECIES OF PERIWINKLES (PROSOBRANCHIA: MESOGASTROPODA) В. Vitturi', A. Libertini?, M. Panozzo” & G. Mezzapelle' ABSTRACT The diploid chromosome number 2n = 34 has been determined in early developing embryos of L. saxatilis and male gonads of L. (Melaraphe) punctata both from the Mediterranean Sea. The diploid value 2n = 33 occurred in spermatocytes of L (Melaraphe) neritoides specimens from the Lagoon of Venice. As previously reported for L. neritoides from the Sicilian coast, a male XO sex-determining mechanism seems to operate also in the population of the same species here studied. Variation in the number of NOR-bearing chromosomes per cell has been detected in both L. saxatilis and L (Melaraphe) neritoides. Flow cytometric DNA analysis indi- cates that L. saxatilis and L. (Melaraphe) neritoides are endowed by almost equal genome sizes, whereas L. (Melaraphe) punctata exhibits about 60% of their values. Key words: periwinkles, karyology, genome size, Mediterranean Sea. INTRODUCTION Periwinkles are mesogastropod molluscs belonging to the genus Littorina which inhabit rocky coasts and lagoon brackish waters. п the Mediterranean, this genus is represented by three species: L. saxatilis (Olivi, 1792), L. neritoides (Linnaeus, 1758) and L. punctata (Gmelin, 1791). On the basis of their repro- duction, embryogenesis, and morphology of the radula and penis, the former belongs to the subgenus Littorina, whereas the others are grouped in the subgenus Melaraphe (Nordsieck, 1968; Torelli, 1982). Cytogenetically, two Mediterranean popu- lations of L. neritoides (Thiriot-Quievreux & Ayraud, 1982; Vitturi et al., 1988) and three North Sea populations of L. saxatilis, from western Sweden, northern England (Janson, 1983) and the Barentz Sea (Birstein 4 Mikhai- lova, 1990), have been investigated. Never- theless, the results of these studies do not agree. More precisely, the diploid number of 34 chromosomes was proposed by Thiriot- Quievreux 4 Ayraud (1982) for L. neritoides males from Villefranche-sur-Mer, Provence, France, whereas 2n = 33 was found in male specimens from Palermo, Sicily, Italy (Vitturi et al., 1988). In the latter population, the oc- currence of an unpaired chromosome т spermatocytes, along with n = 17 bivalents in the female, made the authors hypothesise a male XO sex-determining mechanism oper- ating in this species. Moreover, although Janson (1983) and Birstein & Mikhailova (1990) agree upon the chromosome number (34 in the diploid set) of three different populations of L. saxatilis, there are small differences related to the karyotype morphology. In fact, three small chromosome pairs were seen as subtelocen- tric in the Swedish population (Janson, 1983), whereas the same pairs were found to be most likely telocentric in the strain from the Barentz Sea (Birstein & Mlkhailova, 1990). Karyological investigation in the present report includes: (1) analysis of L. neritoides male specimens from the Lagoon of Venice, northeastern Italy, in order to verify if the male XO sex-mechanism also occurs in this geographical location; (2) a comparison among the karyotypes of L. saxatilis from the Mediterranean Sea and from the North Sea previously described (Janson, 1983; Birstein & Mikhailova, 1990); and (3) a preliminary cy- togenetic characterisation of L. punctata, which is still unknown. Moreover, in order to better understand karyological relationships in Littorinidae, nu- cleolar organizer regions (NORs) of L. saxati- lis and L. neritoides as well as genome sizes in the three species have been investigated. MATERIALS AND METHODS Several specimens of the brooding L. sax- atilis and sexually mature L. (Melaraphe) neri- toides were collected along the dock base- ‘Institute of Zoology, University of Palermo, Via Archirafi 18-90123, Palermo, Italy. 2CNR- Institute of Marine Biology, Riva 7 Martiri 1364/A, Venice, Italy. National Institute of Cancer Research, Biotechnological Section, Via Gattamelata 64, Padua, Italy. 124 VITTURI ЕТ AL. TABLE 1. Counts of mitotic spreads in three Mediterranean periwinkles species. SPECIES ORIGIN Littorina saxatilis Littorina (Melaraphe) punctata Gulf of Palermo — — Littorina (Melaraphe) neritoides Lagoon of Venice — — ments, in Venice, Italy, on October 1991 and April 1992, respectively. Sexually mature specimens of L. (Melaraphe) punctata were sampled from July 1991 to August 1993 on rocky shores at Sferracavallo, Palermo Dis- trict, Italy. Identifications were made accord- ing to the guidelines of Parenzan (1970) and Torelli (1982), and voucher shells of five indi- viduals per species were deposited at the Museum of the Institute of Zoology, Univer- sity of Palermo. Mitotic metaphase chromosomes of L. saxatilis were obtained from early developing embryos extracted from 20 different females and treated with 0.025% colchicine in 0.075 M KCI solution for 20 min according to the air-drying technique. The same procedure was applied to the testes of 15 specimens of L. (Melaraphe) neritoides and 30 of L. (Mela- raphe) punctata to obtain diakinetic bivalents and spermatogonial metaphases. Slides were stained т 5% Giemsa-phosphate buffer solution (pH 6.8), observed, then de- stained in ethanol and restained with silver nitrate according to the colloidal 1-step method (Howell & Black, 1980). In order to obtain the karyogram of L. sax- atilis and L. (Melaraphe) neritoides, the chro- mosomes of five plates per species were cut and paired on the basis of decreasing size and centromere position. Chromosomes were measured and classified by arm-ratio (longer/short arm) following the nomencla- ture proposed by Levan et al. (1964). Genome size was evaluated through flow cytometric assay. From dissected mantle of 8-20 specimens per species, a cell зизреп- sion was obtained by a 5 min hypotonic treat- ment with 0.075 М KCI solution. Cells were filtered through a 30-um mesh, then fixed in 70% ethanol and centrifuged twice at 800 g adding new fixative every time. Samples were stored at —20°С. A day before cytofluorimetric analysis, cells were centrifuged again and resus- pended into 1 ml of solution containing 0.12% sodium citrate, 0.005% propidium io- 2п =21 22 23 24 25 26 27 28 29 30 31/ 3223384 TOTAL Lagoon of Venice 1 — =P ======= 1 1,2906 ey nm u Ze SSS 2 dE — 2 dide and 0.1% RNase. RNA digestion was performed at 37°C for 30 min, while staining lasted overnight at 4°C. Gill cells of the blue mussel (Mytilus edulis) were prepared in the same manner and used as standard. Control assigned DNA content was 3.2 pg according to Hinegardner (1974). An EPICS-C flow cy- tometer (Coulter Electronics, Hialeah, Flor- ida) was employed for DNA content mea- surements. A 488-nm argon ion laser was used for ex- citation and total red fluorescence emission was measured. At least three samples per species each containing more than two thou- sand cells were employed. Conditions of analysis were set in order to obtain the modal value of blue mussel at the channel 50 in a 256-channel DNA histogram. The samples of periwinkle species were run both with and without control cells; be- cause data distribution of reference and test- ing nuclei were partially overlapped, only in- dividual histograms are shown. RESULTS The diploid value 2n = 34 has been found in both L. saxatilis and L. (Melaraphe) punc- tata from counts of 32 and 24 metaphases respectively (Table 1). A few aneuploid spreads displaying a chromosome number lower than the mode were also encountered. These might be related to technical short- comings. Littorina saxatilis complement (Figs. 1, 2; Table 2) consisted of all bi-armed pairs (M+SM) except for pairs 4 and 6, which were mono-armed (ST). However, pairs 3, 7, and 12 were characterized by arm ratio on the border line between SM and ST and thus, considering their confidence limits (respec- tively 1.53-2.01, 1.34-2.20 and 2.47-3.29), these chromosome pairs may not be attrib- uted to one or the other chromosome class. NOR location and variation in number of NOR-bearing chromosomes per cell were KARYOTYPES OF PERIWINKLES 125 ПИН: AD DE Be an 44 Au ка x6 £6 В ks BA 8a TE Bk ds 8% а 3 а b a b a a L «+ о с En № а b с 1 +. A À + 4 р e . ¿ А 48 A 141 E | Da > 4 ”’ | + Oe FIG. 1. Karyogram obtained from five mitotic metaphases of L. saxatilis. FIG. 2. Giemsa stained representative karyotype of L. saxatilis. FIG. 3. Different NOR phenotypes (А, В, С, D) of L. saxatilis. FIG. 4. Giemsa stained spermatocyte bivalents at diakinesis of L. (Melaraphe) punctata. ‘6 ” detected in L. saxatilis by silver staining. chromosomes belonging to pairs 4 (“а NORs were located on the short arms in three type), 6 (“‘b” type) and 15 (‘‘c”’ type). In the different types of silver positively stained first case, NOR was in a large-sized subtelo- VITTURI ЕТ AL. 126 «NS и WS N WS и N и N LS W У LS LS N N и Seplo}ueu ‘7 WS W WS WS WS WS W W W WS WS LS WS LS WS W W SIIHEXES "7 NOILISOd 3H3WOH1N39 66 с OL ple ЗЕЕ 0871 Geil 801 ОЕ cel 00 + ¿EL SEL 18'6 OC'E 6€ L 621 ell Seplo}ueu "7 66€ ost LAA 981 ¿Us 88c lv! 63! 9c 89'¢ ДЕ ve € Ка 10'y ¿EL Le! vil SI/Nexes "7 МУЗИ OILVE Nyv e€'0FZ0'L v€'0F89'L be Or9/ | 8e'0FZ6'L Ot'OF/6 L Ob OFOO'? 9+'OF80Z SS'OF8L'Z 9SOFZE'Z SVOFLbZ /90FI92 99'0+69'е /90708 +/'OFOLE 9/'OFSEE PO LFPO + EL LFB + Sapioyuaeu ‘7 Op OFL8'l bb OFZ6'L 19`0+Р0’с 9SOF9LZ У9'0+6е’е 99‘0+6у’е +9 Or8 ce 20692 cC/'0F69 2 88'0F28'2 S8'0F68'Z #0'0+60'5 bL'OFEL'E Z8'OFOY "E OL'OFBS'E Z6'0FLL'y Z8'0FEL'S sınexes ‘7 ‘a's + (WN) HLON3T NVIN Zt 9! SE vi el ct LL OL 6 8 д 9 5 y € rá | dlvd ANOSOWOYHO (омзиеэолое = y “9IJUSIO|9I-QNS = 1$ ‘оэщиээеэщ-ап$ = WS ‘ощиэовеш = JA) “BUIOSOLUOIUO X2S зэзео!ру! Ysuajsy “‘Sep/ojeU (aydelejayy) ‘7 pue SEXES CULIOJIT jo sejejd eseydejau an] jo seuosouwoJyo ey, JO оцел We pue yjbua] чеэи\ ‘г FIGVL KARYOTYPES OF PERIWINKLES 127 centric chromosome and seemed to be ter- minally or sub-terminally located in con- densed (Fig. 3A) and decondensed (Fig. 3B) chromosomes. In the second case it was ter- minal in a medium-large subtelocentric chro- mosome, and finally it resulted to be terminal in а small-sized submetacentric chromo- some. Four NOR phenotypes, differing for the number of chromosomes involved in nu- cleolus organization, were found: one char- acterized by four chromosomes (а, a, b, c in two spreads) (Fig. 3A), two by three chromo- somes (a, a, b and a, b, c, in 7 and 5 plates, respectively) (Fig. 3B, C) and one by two (a, b in six spreads) (Fig. 3D). In L. punctata, notwithstanding several at- tempts during three reproductive seasons (1991-1993), no slides gave spermatogonial metaphases useful to define the karyotype morphology of this species due to the over- condensation of mitotic chromosomes. Overall, the diploid number tendel tube 2n = 34 (Table 1). The haploid number was n = 17 (Fig. 4), and no heteropycnotic elements were identified. Counts of 21 spermatogonial metaphases resulted in the diploid number 2n = 33 in L. (Melaraphe) neritoides from the Lagoon of Venice (Fig. 5). The Кагуодгат (Fig. 6, Table 2) consisted of 16 autosomal pairs, 12 of which were bi-armed and four mono-armed (pairs 4, 5, 6, 8), and one small unpaired sub- metacentric element about 1 um long. After silver staining, spermatogonial meta- phases showed the occurrence of either three or two NOR-bearing chromosomes per plate referred as to two distinct NOR phenotypes. The former consisted of two small-sized me- tacentric chromosomes (pair 16) with terminal NORs and a large subtelocentric (pair 6) with telomeric NORs on the short arms designated “a,” “a” and “b,” respectively (Fig. 7A), and was observed in 15 spreads. The latter in- cluded two small-sized metacentric NOR- bearing chromosomes always similar to those previously identified as “a” (Fig. 7B), and this was found in 12 spreads. Analysis of diakinetic plates showed 16 bivalents plus one small-sized submetacen- tric. The latter being characterized by the same morphology of the unpaired element observed in mitotic metaphases, was inter- preted as “univalent” (Fig. 8, arrow). A typical fluorescence distribution histo- gram in cell samples of the three periwinkle species and the control are given in Fig. 9 (A-D). Each littorinid species is characterized by two peaks. The first one represents diploid cells while the second one—localized at a double distance from the axes origin (tetra- ploid peak)—is probably due to cell aggrega- tion rather than G2-M phase cells. The refer- ence exhibits one peak only. Red fluorescence emission measurements are summarized on Table 3. They indicate that L. saxatilis and L. (Melaraphe) neritoides are endowed by almost equal genome sizes (estimated haploid DNA content of 1.352 and 1.376 picograms, respectively), whereas L. (Melaraphe) punctata exhibits about 60% of these values (0.811 pgs). DISCUSSION Karyological data concerning all periwin- kles of the genus Littorina studied to date are briefly summarized on Table 4. Interestingly, a diploid number 2n = 33 occurs in males of Littorina (Melaraphe) neritoides from the La- goon of Venice. This result, and a karyotype very similar to that observed for L. (Melara- phe) neritoides from the Sicilian coast (Vitturi et al., 1988), supports a male XO sex chro- mosome system presumably operating in this species. Moreover, as previously argued (Vitturi et al., 1988), this result does not agree with 2n = 34 suggested by Thiriot-Quievreux & Ayraud (1982) for the males of L. (Melara- phe) neritoides from Villefranche-sur-Mer, France. Because of the hypothesis of a chro- mosomal polymorphism in this species has already been excluded (Vitturi et al., 1988), a karyotype revision of the French population would be desirable. Male XO sex determination mechanism has been described in some grasshoppers (Cabrero et al., 1985; Vitturi et al., 1993) and fishes (Chen, 1969; Le Grande, 1975). How- ever, according to our knowledge, it has never been reported in mesogastropods, be- Cause neritid species, possessing this sex- determination mechanism (Vitturi & Catalano, 1988, and authors quoted by them) belong to Archaeogastropoda (Franc, 1968; Boss, 1982). The same chromosome number (2n = 34) but with small differences in karyotypes have been found in three out of four L. saxatilis populations analyzed so far. Observed differ- ences are to be probably related to the fact that some chromosome pairs may be close to the limits between different chromosome categories. 128 VITTURI ЕТ AL. новь £a JA 3% 46 £x à aa ху $5 ta à Е Tan ne /.© ta4 CLS #16 (a wx PO Mey, ep FA" FIG. 5. Giemsa stained karyotype from male gonads of L. (Melaraphe) neritoides. FIG. 6. Karyogram obtained from five spermatogonial metaphases of L. (Melaraphe) neritoides. FIG. 7. NOR phenotypes (A and В) of L. (Melaraphe) neritoides (д = giemsa stained and п = silver stained). FIG. 8. Diakinetic bivalents from male gonads of L. (Melaraphe) neritoides (arrow indicates the unpaired chromosome). Scale for figures from 1 to 8: 10 um = 26 mm Considering NOR patterns determined for 3. at least three chromosome pairs are in- L. (Melaraphe) neritoides and L. saxatilis, volved in nucleolus organization in L. some considerations can be drawn: saxatilis, whereas only two operate in L. oi (Melaraphe) neritoides. This implies a 1. intra-specific polymorphism due to dif- smaller variation of NOR patterns in the ferent number of silver-positive chro- second species. mosomes per cell occurs in both spe- cies; On the whole, data from the literature doc- 2. chromosomal location of active NOR ument a wide distribution of NOR polymor- sites differs from one species to an- phism within animal kingdom (Foresti et al., other; 1981; Gold & Amemiya, 1986; King et al., KARYOTYPES OF PERIWINKLES 129 number of cells red fluorescence Littorina (Melaraphe) punctata 43 C Littorina (Melaraphe) neritoides 42 B Littorina saxatilis 50 D Mytilus edulis FIG. 9. A-D. Histograms of red fluorescence emission in cell suspensions obtained from three species of Littorinidae and Mytilus edulis (control). A: L. (Melaraphe) punctata; В: L. saxatilis; С: L. (Melaraphe) neri- toides; D: Mytilus edulis. Red fluorescence emission and number of scored cells are expressed in arbitrary units. 1990; Vitturi et al., 1991a). Particularly, in mollusks, intraspecific NOR variations have been described for two prosobranch gastro- pods (Vitturi & Catalano, 1989, 1990), two pulmonates (Vitturi et al., 1991b; Vitturi, 1992) and four oysters (Thiriot-Quievreux & Insua, 1992; Insua & Thiriot-Quievreux, 1993). Haploid genome sizes (C-value) vary from 0.8 to 1.3 pg in five littorinid species (Hine- gardner, 1974; present paper), whereas the diploid chromosome number is 34, except for males of L. (Melaraphe) neritoides. Ob- tained results allow us to speculate that large differences in genome size of periwinkles and loss or accumulation of DNA occur within chromosomes without changing their num- ber. As already reported for other organisms (Hutchinson et al., 1980; Gold & Prince, 1985; Olmo et al., 1989; Vitturi et al., 1993), ob- served C-value variability ought to reflect pri- marily gain or losses of repeated DNA se- quences. As in littorinid species, wide genome size variations with no substantial change in chro- mosome number have been already reported among species of the same genus in culicid mosquitoes (Diptera) (Nagesh Rao & Rai, 1990) and leaf beetles (Coleoptera, Chry- 130 VITTURI ET AL. TABLE 3. Genome size evaluation in three periwinkle species. Relative fluorescence is referred to Mytilus edulis Fluorescence in Arbitrary Units Percentage Estimate Number of Average Standard of Relative Haploid DNA Standard Species Samples modal Value Deviation Fluorescence Content (pgs) Deviation Littorina (Melaraphe) punctata 3 25.333 0.471 50.667 0.811 0.0301 Littorina (Melaraphe) neritoides 3 43 0.816 86 1.376 0.0522 Littorina saxatilis 4 42.250 0.829 84.500 1.352 0.0531 TABLE 4. Chromosome numbers in six species of the genus Littorina Species n 2n Origin L. brevicula 117 — Japan L. strigta 1% 34 Japan L. neritoides 17 34 Provence, France L. neritoides 11% 33 L. neritoides 17 33 L. punctata We 34 L. saxatilis 17 34 Cornwall, England L. saxatilis 17 34 L. saxatilis 17 34 Barents Sea L. saxatilis — 34 L. obtusata — 34 Lagoon of Venice, Italy Gulf of Palermo, Italy Gulf of Palermo, Italy Northern Sea, Sweden Lagoon of Venice, Italy Northern Sea, Sweden Reference Nishikawa, 1962 Nishikawa, 1962 Thiriot-Quievreux 4 Ayraud, 1982 Present paper Vitturi et al., 1988 Present paper Janson, 1983 Janson, 1983 Birstein & Mikhailova, 1990 Present paper Janson, 1983 somelidae) (Petitpierre & al., 1993). Further- more, no direct correlation between DNA content and chromosome number were ar- gued also for pleurocerid snail genus Semi- sulcospira (Mesogastropoda) (Nakamura 4 Ojima, 1990). Although karyological data for L. punctata are limited to the diploid chromosome num- ber (2n = 34) and spermatocyte bivalent mor- phology, they allow us to exclude a XO sex- determining mechanism for this species. Hence, the wide karyological differences (ge- nome size and presence of sex chromo- somes) between L. punctata and L. neri- toides make location of these species in the same subgenus (Nordsieck, 1968; Torelli, 1982) unjustified. Alternatively, in accordance with Rosewater (1970), the taxon Melaraphe should contain only L. neritoides, which dif- fers from all other littorinids by a pair of cusps on the basal part of the central tooth in the radula. The three species of Littorina here studied display a great variability when different cy- togenetical parameters (i.e., karyotype mor- phology, NOR patterns, sex chromosomes, genome size) are considered, and, therefore, karyology may be very useful in order to de- fine the frequently rearranged taxonomy of the genus (Bandel, 1974). ACKNOWLEDGEMENTS Financial support by the ministero per Г Università e la Ricerca Scientifica e Tecno- logica (60%, 1992-93), Roma, is gratefully acknowledged. LITERATURE CITED BANDEL, K., 1974, Studies on Littorinidae from the Atlantic. The Veliger, 17: 92-114. BIRSTEIN, V. J. & М. A. MIKHAILOVA, 1990, On the karyology of trematodes of the genus Microphal- lus and their intermediate gastropod host, Litto- rina saxatilis II. Karyological study of Littorina saxatilis (Gastropoda: Prosobranchia). Genetica, 80: 167-170. BOSS, K. J., 1982, Mollusca (ana) classification of Mollusca. In Synopsis and classification of living organisms,” ed. $. P. PARKER, 1: 945-1166, 2: 1092-1096. New York: McGraw Hill. KARYOTYPES OF PERIWINKLES 131 CABRERO, J., J. P. M. CAMACHO & Е. PASCUAL, 1985, Cytotaxonomic studies on pamphagid ge- nus Eumigus. Detection of two chromosomal races in E. monticola (Rambur) (Insecta, Or- thoptera). Caryologia, 38: 1-12. CHEN, T. R., 1969, Karyological heterogamety of deep sea fishes. Postilla, 130: 1-29. FORESTI, F., L. Е. ALMEIDA TOLEDO & 5. A. TO- LEDO F., 1981, Polymorphic nature of nucleolus organizer regions in fishes. Cytogenetics and Cell Genetics, 31: 137-144 FRANC, A., 1968, Sous-class des prosobranches In: Traité de Zoologie, Ed. P. P. Grassé, Tome V., Masson et Cie, Paris, 40: 324. GOLD, J. В. & С. T. AMEMIYA, 1986, Cytogenetic studies in North American minnows (Cyprinidae) XII. Patterns of chromosomal nucleolus orga- nizer regions variation in 14 species. Canadian Journal of Zoology, 65: 1869-1877. GOLD J. В. & J. PRINCE, 1985, Genome size vari- ation among North American minnows (Cyprin- idae) |. Distribution of the variation in five spe- cies. Heredity, 54: 297-305. HINEGARDNER, R., 1974, Cellular DNA content of the Mollusca. Comparative Biochemistry and Phisiology, 47 B: 447-460. HOWELL, W. М. & D. A. BLACK, 1980, Controlled silver-staining of nucleolus organizer regions with a protective colloidal developer a 1-step method. Experientia, 36: 1014-1015. HUTCHINSON, J., В. К. J. NARAYAN & H. REES, 1980, Constraints on the composition of supple- mentary DNA. Chromosoma, 78: 137-145. INSUA, A. & С. THIRIOT-QUIEVREUX, 1993, Kary- otype and nucleolus organizer regions in Ostrea puelcana (Bivalvia: Ostreidae) The Veliger, 36: 215-219. JANSON, K., 1983, Chromosome number in two phenotypically distinct populations of Littorina saxatilis Olivi, and in specimens of the Littorina obtusata (L.) species-complex. Journal of Mol- luscan Studies, 49: 224-227. KING, M., М. CONTRERAS & В. L. HONEYCUTT, 1990, Variation within and between nucleolar or- ganizer regions in Australian Hylid frogs (Anura) showed by 18$ + 28$ in situ hybridization. Ge- netica, 80: 17-29. LE GRANDE, W. Н., 1975, Karyology of six species of Louisiana flat-fishes (Pleuronectiformes: Os- teichthyes). Copeia, 3 (1975): 516-522. LEVAN, A. K. FREDGA, 8 A. A. SANDBERG, 1964, Nomenclature for centromeric position of chro- mosomes. Hereditas, 52: 201-220. NAGESH RAO, P. £ K. S. RAI, 1990, Genome ev- olution in the mosquitoes and other closely re- lated members of superfamily Culicoidea. Her- editas, 113: 139-144. NAKAMURA, H. K. 8 Y. OJIMA, 1990, Cellular DNA contents of the freshwater snail genus Semisul- cospira (Mesogastropoda: Pleuroceridae) and some cytotaxonomical remarks. American Mal- acological Bulletin, 7(2): 105-108. NISHIKAWA, S., 1962, A comparative study of chromosomes in marine gastropods with some remarks on cytotaxonomy and phylogeny. Jour- nal Shimonoseki Collected Fisheries, 11: 119- 186. NORDSIECK, F., 1968, Die Europaischen Meeres- gehouserhneken (Prosobranchia). Stuttgart, Fisher: 1-273. OLMO, E., T. CAPRIGLIONE 8 G. ODIERNA, 1989, Genome size evolution in vertebrates: trends and constraints. Comparative Biochemistry and Physiology, 92B: 447-453. PARENZAN, P., 1970, Carta d’ identita delle con- chiglie del Mediterraneo (Gasteropodi), Vol. |. Bios Taras Eds. PETITPIERRE, E., SEGARRA, С. & С. JUAN, 1993, Genome size and chromosomal evolution in leaf beetles (Coleoptera, Chrysomelidae). Heriditas, 119: 1-6. ROSEWATER, J., 1970, The family Littorinidae in the Indo-Pacific. Part. |. The super-family Lit- torinidae. Indo-Pacific. Mollusca, 2: 417-506. THIRIOT-QUIEVREUX, С. & М. AYRAUD, 1982, Les karyotypes de quelques especes de bivalves et gasteropodes marins. Marine Biology, 70: 165- 172: THIRIOT-QUIEVREUX, С. & А. INSUA, 1992, Nu- cleolar organiser region variation in the chromo- somes of three oyster species. Journal of Exper- imental Marine Biology and Ecology, 157: 33-40. TORELLI, A., 1982, Gasteropodi conchigliati. Guide per il riconoscimento delle specie animali delle acque lagunari e costiere Italiane 8. CNR. 238 pp. VITTURI, R., 1992, Conventionally stained chromo- somes, constitutive heterochromatin and nucle- olus organizer regions (NORs) in Milax nigricans (Gastropoda, Pulmonata). Chromatin, 1: 147- 155: VITTURI, В. & Е. CATALANO, 1988, A male ХО sex-determining mechanism in Theodoxus me- ridionalis (Neritidae) (Prosobranchia, Archaeo- gastropoda). Cytologia, 53: 131-138. VITTURI, R. 4 E. CATALANO, 1989, Spermatocyte chromosomes and nucleolus organizer regions (NORs) in Tricolia speciosa (Mühlfeld, 1824) (Prosobranchia, Archaeogastropoda). Malacolo- gia, 31(1): 211-216. VITTURI, R. 4 E. CATALANO, 1990, Spermatocyte chromosome banding studies in Buccinulum corneum(Prosobranchia: Neogastropoda): varia- tion in silver-NOR pattern. Marine Biology, 104: 259-263. VITTURI, В., С. MANSUETO & Е. FICARELLA DI PAOLA, 1993, Heterochromatin variation in four species of the genus Pamphagus (Orthoptera: Pamphagidae) analized by C-banding. Biologi- sches Zentralblatt, 112: 335-341. VITTURI, R., Е. CATALANO, М. В. LO CONTE, А. М. ALESSI, Е. P. AMICO 8 D. COLOMBERA, 1991a, Intra-populational and intra-individual mosaicism of Uranoscopus scaber L. (Perci- formes, Uranoscopidae). Heredity, 67: 325- 330. 132 VITTURI ET AL. VITTURI, R., E. CATALANO, M. MACALUSO, 4 analysis of Helicella virgata (Pulmonata: Heli- B. ZAVA, 1988, The karyology of Littorina neri- cidae): silver-stained and C-banded chromo- toides (Linnaeus, 1758) (Mollusca, Prosobran- somes. Journal of Heredity, 82: 339-343. chia). Malacologia, 29: 319-324. VITTURI, R., D. COLOMBERA, E. CATALANO & F. P. AMICO, 1991b, Spermatocyte chromosome Revised Ms. accepted 26 October 1994 MALACOLOGIA, 1995, 37(1): 133-156 POPULATION GENETICS AND SYSTEMATIC STATUS ОЕ ONCOMELANIA HUPENSIS (GASTROPODA: POMATIOPSIDAE) THROUGHOUT CHINA George М. Davis', Zhang УР, Guo Yuan Hua’, & Christina Spolsky' ABSTRACT The coevolved relationships of populations considered to be Oncomelania hupensis through- out China with Schistosoma japonicum are of interest to evolutionary biologists and consider- able importance to medical science relative to understanding the differentiation of the parasite throughout its range. As populations of Oncomelania dispersed and diversified in a direction from Burma-western Yunnan, China, throughout China to Japan and the Philippines, the par- asite has had to modify genetically with the genetically changing snail populations or become regionally extinct. This hypothesis predicts that measures of genetic distance among snail populations parallel genetic diversification among parasite populations. Our question here 15: to what extent have populations of Oncomelania diverged genetically throughout China, and what are the implications for systematic status of the populations? Measures of population divergence for Oncomelania can be derived from shell characters, anatomical characters, or molecular genetic distances. In this paper, we explore genetic di- vergence based on allozyme data involving 14 populations that are widespread throughout China, are in divergent drainage systems, and have different shell phenotypes. We find con- siderable genetic differentiation occurring throughout China. We also find coherent discernable patterns to the genetic differentiation. Careful examination of these patterns provides evidence for one case of cross-country transport of snails, and for the existence of three exceptional populations in which hybridizations between subspecies may have occurred. Excluding these four populations, patterns of genetic differentiation are in general correlated with geographic distribution and shell sculptural and shape characters. These patterns thus serve as the basis for subdivision of O. hupensis into three discrete subspecies. O. h. robertsoni, О. В. tangi, and O. h. hupensis. INTRODUCTION The rissoacean gastropod genus Oncomel- ania is restricted to Asia and has two species based on morphological criteria (reviewed in Davis, 1994); Oncomelania hupensis poly- typic species distributed from northern Burma (fossil), western China to Japan, Philippines, and Sulawesi, and O. minima in northwestern Honshu, Japan. The genus is one of eight genera of the Pomatiopsidae: Pomatiopsinae deriving from a Gondwanian origin (Davis, 1979). The other genera are now found in South Africa, South America, Australia, Ja- pan, and the U.S.A. The genus in the U.S.A., Pomatiopsis, derived from Oncomelania of Japan. Oncomelania is of considerable interest to very different groups of people. To the spe- cialist in tropical medicine and parasitology, O. hupensis is associated with the transmis- sion of the human blood worm Schistosoma japonicum. The two species of Oncomelania also transmit different species of the lung fluke Paragonimus (Davis et al., 1994). For the systematist, variation in size and shell sculpture throughout the range of the genus has been the basis for debates about the number of genera and species involved (Bartsch, 1936a; Abbott, 1948; Davis, 1971; Kang, 1981, 1985; Liu et al. 1981; Lou et al. 1982). The genus now plays a central role in the debate of defining what a species is. Does one today recognize polytypic species or subspecies; to what extent should one rely on measures of genetic distance to define species (Woodruff et al., 1988; Davis, 1994)? The coevolved relationships of populations considered to be Oncomelania hupensis throughout China with Schistosoma japoni- cum are of interest to evolutionary biologists and considerable importance to medical sci- ence relative to understanding the differenti- ation of the parasite throughout its range (re- ‘Academy of Natural Sciences of Philadelphia, 1900 Benjamin Franklin Parkway, Philadelphia PA 19103. 2Institute of Parasitic Diseases, Chinese Academy of Preventive Medicine, Shanghai, China. 134 DAVIS, ZHANG, GUO & SPOLSKY viewed by Davis, 1992). As populations of Oncomelania dispersed and diversified in a direction from Burma-western Yunnan, China, throughout China to Japan and the Philippines (Davis 1979), the parasite has had to modify genetically with the genetically changing snail populations or become re- gionally extinct. This hypothesis predicts that measures of genetic distance among snail populations parallel genetic diversification among parasite populations. Measures of population divergence for On- comelania can be derived from shell charac- ters, anatomical characters, or molecular ge- netic distances. In this paper, we explore genetic divergence based on allozyme data involving 14 populations that are widespread throughout China, are in divergent drainage systems, and have different shell pheno- types. We find considerable genetic differen- tiation occurring throughout China. We also find coherent discernable patterns to the ge- netic differentiation. Careful examination of these patterns provides evidence for one case of cross-country transport of snails, and for the existence of three exceptional popu- lations in which hybridizations between sub- species may have occurred. Excluding these four populations, patterns of genetic differ- entiation are in general correlated with geo- graphic distribution and shell sculptural and shape characters. These patterns thus serve as the basis for subdivision of О. hupensis into three discrete subspecies: О. h. robert- soni, О. В. tangi, and O. В. hupensis. MATERIALS AND METHODS Specimens Studied Two groups of populations were studied seven months apart: all snails were alive when brought to the United States by Dr. Guo. They were collected by members of lo- cal anti-epidemic stations throughout China and sent to Dr. Guo in Shanghai. The 14 lo- calities sampled are listed in Table 1 and are shown in Figure 1. Group | represents seven populations from the most geographically separated locations in China collected in No- vember 1984. Group II snails, collected in November 1985 were from central to eastern China, with populations concentrated in the Hubei-Hunan-Anhui triangle. Shell categories are based on the major phenotypic classes of variants (Figs. 2-5): (1) smooth and no varix = S; (2) smooth but with pronounced varix = S,V; (3) smooth, with varix but with sporadic low ribs on penulti- mate and body whorl; these may appear as exaggerated growth lines = S +; (4) ribbed and with varix, the ribs are numerous (15-18 on the lower whorls) and pronounced but low in profile = R; (5) pronounced high profile ribs, few in number on the lower whorls (12-13), with varix = R+. Each population consists of one phenotypic class only. Electrophoresis Horizontal starch gel electrophoresis of tissue proteins was followed by staining for the following 30 loci: AAT-1, AAT-2 (as- partate aminotransferase, 2.6.1.1); AK (adenosine kinase, 2.7.1.20); AO (aldehyde oxidase, 1.2.3.1); ACPH (acid phospha- tase, 3.1.3.2); APH (alkaline phosphatase, 3.1.3.1); CK (creatine kinase, 2.7.3.2); EST-1, EST-2, EST-3 (esterase, 3.1.1.1); GDH (glu- tamate dehydrogenase, 1.4.1.2); G6PD-1, G6PD-2 (glucose-6-phosphate dehydroge- nase, 1.1.1.49); GPI (glucose-6-phosphate isomerase); ISDH-1, ISDH-2, ISDH-3 (isoci- trate dehydrogenase, 1.1.1.42); LDH (L-lac- tate dehydrogenase, 1.1.1.42); MDH (malic dehydrogenase, 1.1.1.37); ME-1, ME-2 (malic enzyme, 1.1.1.40); MPI (mannose-6-phos- phate isomerase, 5.3.1.9); NADD-1 (NADH dehydrogenase, 1.6.99.3); 6PGD (phospho- gluconate dehydrogenase, 1.1.1.44); PGM-1, PGM-2 (phosphoglucomutase, 5.4.2.2); OCT (octopine dehydrogenase, 1.5.1.11); SDH-1; SDH-2 (sorbitol dehydrogenase, 1.15.1.1) ; XDH (xanthine dehydrogenase, 1.1.1.204). Procedures are those of Ayala et al. (1973) as modified by Dillon and Davis (1980), Davis (1983), Davis et al. (1981, 1988), and most recently for Oncomelania, by Davis et al. (1994). Because the number of snails was limited, and because the two groups were studied some months apart, it was not possible to obtain results for all loci for all populations. In comparing the two groups of populations, in- sufficient snails were available to fill in miss- ing data for some loci and to be absolutely certain of homology of scoring among some populations. Accordingly, the results are given for each group of populations sepa- rately before giving the combined data set. The combined data set for 14 populations involved 25 loci. The following loci were not included: AAT-2; ME-2; CK; GDH; ISDH-3. Genetic parameters were calculated using BIOSYS-1 (Swofford & Selander, 1981). POPULATION GENETICS ОЕ ONCOMELANIA HUPENSIS TABLE 1. Populations, localities, and catalog numbers for the fourteen populations of the two groups. Localities are listed as they occur west to east. Latitudes and longitudes are given. They are provided for the county town when the precise locality in the county is not known. Shell sculptural types are as follows: Smooth; Smooth+ = smooth with a few low riblets scattered on the last two whorls of some shells, some looking as strong growth lines; Ribs+ = ribs very strong, few and high, about 11-13 on the body whorl; Ribbed = ribs strong, low but pronounced, many, ¡.e. some 18-19 on the body whorl or penultimate whorl. Group | snails collected in October 1984: e Yunnan Province; Dali County 25744'43"N; 100°7’16”Е; ANSP 398317, A18327 shell: smooth; no varix . Sichuan Province; XiChang City 2657'49"N; 10277'58“E; ANSP 398322; A18333 shell: smooth; no varix . Hubei Province; JianLi County 29°49’N; 112°54’E; ANSP 398320; A18330 shell: ribbed; varix strong . Jiangxi Province; PengZe County 29°54’N; 116°32’E; ANSP 398321; A18331 shell: ribs+; varix strong . Anhui Province; GuiChi City 117°20.6’E; 30°30’N; CIPD 0356; ANSP 398319; A18329 shell: ribs+; varix strong . Fujian Province; FuQing County 25°43’38”М; 119°24’00”Е; ANSP 398317; A18332 shell: smooth; varix strong and wide . Fujian Province; XiaPu County 26°50/11”М; 120°E; ANSP 398323; A183333 shell: smooth; varix strong and wide Group Il collected in November 1985: 1(8). Guangxi Province; GuiPing County 23°23’27”’N; 110°04’42”E; ANSP 375733 shell: smooth+; varix strong 2(9). Hubei Province; JiangLing County 30°20’57”М; 112°11’E; ANSP 375528 shell: ribbed; varix strong 3(10). Hunan Province; YueYang City 29722'52"N; 113°06’00”Е; ANSP 375732 shell: ribs+; varix strong 4(11). Hubei Province; HanYang County 30734'46"N; 114°01”06”Е; ANSP 375731 shell: ribs+; varix strong 5(12). Anhui Province; TongLing County/City 30°12’35”М; 116°05’27”Е; ANSP 375730 shell: ribs+; varix strong 6(13). Anhui Province; NingGuo County 30°22’23”М; 118°58’21”Е; ANSP 37573 shell: smooth; varix strong 7(14). Zhejiang Province; Chang Xing County near Anji; 31°01’35’N; 119°54’29°E; ANSP 375729 shell: smooth; no varix = DA, Fig. 1 = XI, Fig. 1 = Jl, Fig. 1 = PZ, Fig. 1 = GC, Fig. 1 = РО, Fig. 1 = XP, Fig. 1 = GP, Fig. 1 = JL, Fig. 1 = YY, Fig. 1 = НУ, Fig. 1 = TL, Fig. 1 = NG, Fig. 1 = СХ, Fig: 1 135 Hardy-Weinberg equilibrium was analyzed for all polymorphic loci. Nei’s (1978) genetic distance and Cavalli-Sforza & Edwards’ (1967) arc distance were calculated and phenograms constructed using the UPGMA method. Homology of alleles at certain loci could not be rechecked because of lack of specimens, and therefore Nei’s (1978) unbi- ased minimum distance (mD) was used so as not to inflate D due to possible error. Un- rooted trees based on mD were also con- structed using the FITCH program of PHYLIP version 3.4 (Felsenstein, 1989). This phyloge- netic analysis program does not assume equal rates of evolution. Twenty repetitions of FITCH were run with randomized input order 136 DAVIS, ZHANG, GUO 8 SPOLSKY SHANXI | HENAN LES te | ANHUI | | | | TL | HY о / SICHUAN } an, / ÈS HU | 1 GC y o R | | 2 RUE 5 a Г. PR & e JL y) и PZ ZHEJIANG Zi | YY Jl Nanchang E | Changsha A | JIANGXI | ^_--ХР | GUIZHOU HUNAN р < eGuiyang QEr Hai Е | Fuzhou FUJIAN = 25 Dali >-DA | N AS Kunmin: ig | e | | y 1 < = GUANGXI | YUNNAN GP G ANGDONG N 4 Guangzhou Canton) . + Jiang в. \ Nanning | | o 90 180 270 360km | 20° 105° 110° FIG. 1. Мар of localities in southern China. West to east: DA = Dali; Х! = XiChang; GP = GuiPing; JL JiangLing; Jl = JianLi; YY = YueYang; HY = HanYang; PZ = PengZe; GC = GuiChi; TL = TongLing; NG NingGuo; FQ = FuQing; CX = ChangXing; XP = XiaPu. and optimization by global branch rearrange- ment. RESULTS Indices of genetic variability are given in Table 2. Mean heterozygosity 1$ low, with means ranging from 0.008 to 0.093 (1.1 to 1.6 alleles per locus). The percentage of poly- morphic loci ranged from 4.0 to 28.0. The lowest levels (4.0 to 8.0) involved smooth- shelled populations from Fujian and Sichuan provinces in the west, and from Zhejiang Province in the east. The mean value for smooth-shelled populations was 9.3 + 6.0 (4.0-20.0; N = 6). The highest levels (20.0 to 28.0) involved ribbed snails from Hubei, An- hui, and Jiangxi provinces. The percentage of polymorphic loci among ribbed snails was 20.6 + 4.7 (16-28; N = 7). In the analysis of the seven populations in group |, allele frequencies for 29 loci involving 67 alleles are given in Table 3. Nei’s D and arc distances are given in Table 4. Invariant loci and loci with fixed alternative alleles are 115° 120° given in Table 5. Phenograms based on these distances are given in Figure 6. Group | populations represent highly diver- gent locations in six different provinces from Yunnan to Jiangxi provinces in the interior, and from Fujian Province on the coast. Nei’s D ranges from 0.107 (close geography and shell type: Yunnan and Sichuan: smooth shells) to 0.346 (Fujian smooth shell type vs. Jiangxi ribbed shell type). The mean D was 0.241 + 0.066. The corresponding value for arc distance is 0.457 + 0.059 (range of 0.314—0.538). Results from group II are given in Tables 6 and 7 involving 60 alleles at 28 loci. Nei’s D averaged 0.234 + 0.095 (range of 0.089 to 0.382); arc D averaged 0.440 + 0.087 (range of 0.291 to 0.559). Corresponding pheno- grams are given in Figure 7. Nine loci were monomorphic in the first group, 12 in the second (Table 5). One locus in each group was invariant except for one population; thus there were 19 and 14 infor- mative loci, respectively. There were fixed differences at eight loci in each group (Table 5). There were minor deviations from Hardy- POPULATION GENETICS ОЕ ONCOMELANIA HUPENSIS 137 180 270 360 km | 105* FIG. 2. Localities as in Fig. 1, but showing the distribution of shell types. Relative shell sizes are as shown. Weinberg equilibrium in both groups of pop- ulations (Table 8). In the first group, these involved four populations, five of 25 polymor- phic loci (summed over all populations), and primarily reflected heterozygote deficiency. In the second group, deviations occurred in- five populations at three of 28 polymorphic loci; the esterase-2 locus was involved in three of the five populations. Grouping pop- ulations from west to east along the Yangtze River drainage, Nei's D averaged as follows: Yunnan-Sichuan, 0.107 (group 1; Hubei- Hunan, 0.089—0.102 (group II, М = 3); Anhui, 0.231 (group Il, N = 2). For the combined data-set, the matrix of Nei’s (1978) unbiased minimum distance is given in Table 9. The UPGMA phenogram based on these data is given in Figure 8. The greatest Nei’s minimum distance was 0.453 (between Gui Ping and Fu Qing). In the phe- nogram, the populations are grouped in two major clusters. The upper cluster 1$, with three exceptions, a smooth-shelled cluster with shells having no varix (Dali, Chang Xing, Xi Chang) at the top; and the Fujian Province populations with smooth shells and low wide varix (Fu Qing, Xia Pu) at the bottom. Three populations seem out of place because they have ribbed shells: Tong Ling, Jian Li, and Gui Chi. Tong Ling has relatively low genetic distances to the Yunnan (Dali) and Sichuan (Xi Chang) populations and the lowest ge- netic distance (mD = 0.092) to the Zhejiang population (Table 9). The presence of ribbed shells in these snails, which are genetically closest to smooth-shelled populations, sug- gests the possibility that this is a hybrid pop- ulation. The same might be the case for the Jian Li and Gui Chi populations, which also have closer genetic affinity with the upper section taxa of the phenogram. The lower cluster, with an average pairwise mD of 0.160 + 0.07, is one characterized by populations with ribbed shells, except for the Ning Guo (Anhui) and the Gui Ping (Guangxi) smooth-shelled populations. These excep- tions will be discussed later. In sum, popula- tions clustered in Figure 8 appear to repre- sent three genetically cohesive groups: the Yunnan, Sichuan, and Zhejiang smooth- shelled populations with no varix; the two smooth-shelled Fujian populations with low wide shells and a low wide varix; a cohesive group of ribbed-shelled populations, which DAVIS, ZHANG, GUO 8 SPOLSKY Dune FE a FIG. 3. Examples of shells from six localities: A. Dali, Yunnan Province = Oncomelania hupensis robertsoni; В. XiChang, Sichuan Province = O. В. robertsoni; С. ChangXing, Zhejiang Province = O. h. robertsoni; D. FuQing, Fujian Province = O. h. tangi; E. XiaPu, Fujian Province = O. В. tangi; Е. NingGuo, Anhui Province = O. В. hupensis (fausti form). The shell at the top left is 6.0 mm; others are printed to the same scale. POPULATION GENETICS OF ONCOMELANIA HUPENSIS 139 FIG. 4. Examples of shells from six localities: A. GuiPing, Guangxi Province = O. hupensis guangxiensis; B. JianLi, Hubei Province = O. h. hupensis; C. JianLing, Hubei Province = O. h. hupensis; D. PengZe, Jiangxi Province = O. h. hupensis; E. GuiChi, Anhui Province = O. h. hupensis; F. YueYang, Hunan Province = O. h. hupensis. The shell at the top left is 6.69 mm long; others are printed to the same scale. 140 DAVIS, ZHANG, GUO & SPOLSKY FIG. 5. Examples of shells from two localities: A. TongLing, Anhui Province = Oncomelania hupensis hupensis; B. HanYang, Hubei Province = O. h. hupensis. The top left shell is 7.75 mm long; others are printed to the same scale. includes two exceptions that will be dis- cussed later. In addition, there are three pos- sible hybrid populations in the upper cluster. The general topology of the UPGMA pheno- gram is confirmed by FITCH analyses. Figure 9 gives the best unrooted FITCH tree (sum of squares = 2.165); the 20 repetitions pro- duced only four unique trees, with three clear groupings in each case: Yunnan-Sichuan- Zhejiang (Dali group), Fu Qing—Xia Pu (Fujian group), and a mainly ribbed group of six pop- ulations (Han Yang, Gui Ping, Yue Yang, Peng Ze, Jiang Ling, and Ning Guo). The dif- ferences in the four FITCH trees occurred only in relative placement and branch lengths of the three possibly hybrid populations; in all cases, however, these three populations were positioned between the two smooth- shelled groups on the one hand and the ribbed group on the other. The former two groups are found in the upper cluster in the UPGMA phenogram; the latter group corre- sponds exactly to the lower cluster in the UP- GMA tree. Thus, the differences between the UPGMA and FITCH analyses lie mainly in the placement of the three putative hybrid pop- ulations. Genetic Distance, Shell Sculpture and Geographic Distance. Pairwise comparisons among populations on the basis of Nei’s minimum D from the smallest value to 0.156, along with the shell type, are given in Table 10. The value of 0.156 was arbitrarily chosen to represent the upper limit of divergence that one might the- oretically expect between populations based on a number of studies of diverse taxa (re- viewed in Ayala and Aquadro, 1982). It 1$ clear, even without doing a Mantel test (Rice, 1989), that there is no correlation between geographic distance and genetic distance when one considers all 14 populations. The two farthest separated populations have the lowest genetic distance (mD = 0.007 over >2,000km), whereas closest neighbors have an mD of 0.204 at 44 km, and of 0.339 at 72 km. This lack of correlation provided the pre- liminary indication that the set of 14 popula- POPULATION GENETICS OF ONCOMELANIA HUPENSIS 141 TABLE 2. Genetic variability at 25 loci for all 14 populations studied. Mean Heterozygosity Mean sample Mean no. size per alleles per % loci direct- Hdywbg Population locus locus polymorphic count expected 1. Dali (Yunnan) 30.0 : 12.0 0.028 0.035 (0.0) (0.1) (0.017) (0.021) 2. FuQing (Fujian) 25:0 р 8.0 0.008 0.013 (0.0) (0.1) (0.017) (0.010) 3. GuiChi (Anhui) 14.9 a 24.0 0.077 0.093 (0.1) (0.2) (0.036) (0.039) 4. JianLi (Hubei) oro : 20.0 0.045 0.082 (0.0) (0.2) (0.021) (0.039) 5. PengZhe (Jiangxi) 25.0 у 28.0 0.093 0.110 (0.0) (0.2) (0.036) (0.043) 6. XiChang (Sichuan) 16.2 . 4.0 0.008 0.022 (1.0) (0.0) (0.008) (0.022) 7. XiaPu (Fujian) 25.0 . 8.0 0.026 0.022 (0.0) (0.1) (0.020) (0.017) 8. ChangXing (Zhejiang) 25.0 4 4.0 0.010 0.018 (0.0) (0.0) (0.010) (0.018) 9. TongLing (Anhui) 2510 : 16.0 0.064 0.082 (0.0) (0.2) (0.032) (0.040) 10. JiangLing (Hubei) 25.0 р 20.0 0.064 0.082 (0.0) (0.2) (0.032) (0.040) 11. YueYang (Hunan) 25.0 ; 16.0 0.085 0.082 (0.0) (0.2) (0.041) (0.039) 12. GuiPing (GuangXi) 25.0 . 12.0 0.074 0.061 (0.0) .1) (0.041) (0.034) 13. NingGuo (Anhui) 25.0 ; 20.0 0.074 0.055 (0.0) 1) (0.044) (0.028) 14. HanYang (Hubei) 25.0 : 20.0 0.070 0.073 (0.0) .2) (0.035) (0.035) tions is not a valid grouping, and suggested the existence of discrete subgroups of O. hu- pensis. In Table 11 are given pairwise comparisons between populations listed by an increasing value of D from Tables 3 and 7. Crisscrossing through southern China the values gradually increase by increments of 0.020 + 0.01 (N = 15). There are no big gaps to suggest a change from population groupings to dis- crete species. However, when populations are separated into groups based on chonchological and geographic criteria (Table 12), significant dif- ferences are evident among the different groups of populations. The Dali group of three populations with smooth shells without varix have a small average distance among them (0.081). A relatively small distance is found between the two Fujian populations with smooth shells but with a low, wide varix (0.154). Distances jump to >0.200 for inter- group comparisons of smooth-shelled popu- lations. When all ribbed shells are compared, the average distance is 0.204 + 0.085; one stan- dard deviation ranges from 0.119 to 0.289. However, if one excludes the three putative hybrid populations, the remaining six popu- lations form a genetically cohesive group: this group includes two smooth-shelled pop- ulations where shells have а varix. This pre- dominantly ribbed-shelled set of populations has a much lower average pairwise distance, 0.160. This distance is similar to the interpop- ulational differences within the Fujian group (0.154) and to the range of the Dali group (0.007 to 0.127). Thus, there are no signifi- cant within-group differences in distances between the coherent ribbed-shelled group of six and the two major smooth-shelled groups, Dali and Fujian. There is a clear gap between interpopulation distances within groups (Dali, Fujian, and ribbed; mD < 0.160) and intergroup distances between smooth and ribbed populations (mD > 0.300). For the three ribbed, possibly hybrid populations (Tong Ling, Gui Chi, Jiang Li), comparisons of each to the Dali group, the Fujian group, the 142 DAVIS, ZHANG, GUO 8 SPOLSKY TABLE 3. Allele frequencies for seven populations of Oncomelania hupensis from throughout China (group I). 29 loci; 67 alleles. “Same number of individuals from each population at all loci. There were nine invariant loci; see Table 5. AAT-2 was not detected in XiChang, population 6. POPULATION Locus DALI FUQING GUICHI JIANLI PENGZE XICHANG XIAPU (Ny 30 25 19 15 25 18 25 ААТ-1 А 1.0 0.0 0.97 1.0 0.0 0.0 1.0 В 0.0 1.0 0.0 0.0 1.0 1.0 0.0 С 0.0 0.0 0.03 0.0 0.0 0.0 0.0 ААТ-2 А 1.0 1.0 0.87 1.0 1.0 —= 0.74 В 0.0 0.0 0.13 0.0 0.0 — 0.26 ACPH A 0.0 0.0 1.0 0.0 0.0 0.0 0.0 B 1.0 1.0 0.0 1.0 1.0 1.0 1.0 AK A 0.0 0.0 1.0 0.0 0.0 0.0 0.0 B 1.0 1.0 0.0 1.0 1.0 1.0 1.0 APH A 1.0 1.0 1.0 0.0 1.0 1.0 1.0 B 0.0 0.0 0.0 1.0 0.0 0.0 0.0 EST-1 A 1.0 0.86 1.0 1.0 0.96 0.0 0.0 B 0.0 0.0 0.0 0.0 0.02 0.50 0.0 C 0.0 0.0 0.0 0.0 0.0 0.50 0.0 D 0.0 0.14 0.0 0.0 0.02 0.0 1.0 EST-2 A 1.0 1.0 0.56 0.43 0.78 1.0 0.72 B 0.0 0.0 0.40 0.30 0.18 0.0 0.0 С 0.0 0.0 0.03 0.17 0.04 0.0 0.28 D 0.0 0.0 0.0 0.10 0.0 0.0 0.0 EST-3 A 1.0 0.0 0.92 120 1.0 1.0 1.0 B 0.0 1.0 0.0 0.0 0.0 0.0 0.0 C 0.0 0.0 0.08 0.0 0.0 0.0 0.0 G6PD-2 A 0.97 1.0 1.0 1.0 1.0 1.0 1.0 B 0.03 0.0 0.0 0.0 0.0 0.0 0.0 СР! А 0.0 0.0 0.57 0.90 0.70 1.0 0.0 В 0.0 1.0 0.40 0.07 0.28 0.0 1.0 С 0.0 0.0 0.03 0.0 0.0 0.0 0.0 D 0.93 0.0 0.0 0.0 0.0 0.0 0.0 E 0.07 0.0 0.0 0.0 0.0 0.0 0.0 | 0.0 0.0 0.0 0.03 0.02 0.0 0.0 LDH A 1.0 1.0 1.0 1.0 0.0 1.0 1.0 B 0.0 0.0 0.0 0.0 1.0 0.0 0.0 MDH A 0.0 1.0 1.0 1.0 1.0 0.0 1.0 B 1.0 0.0 0.0 0.0 0.0 1.0 0.0 NADD A 1.0 1.0 1.0 1.0 0.0 1.0 1.0 B 0.0 0.0 0.0 0.0 1.0 0.0 0.0 POPULATION GENETICS OF ONCOMELANIA HUPENSIS 143 TABLE 3. (Continued) POPULATION Locus DALI FUQING GUICHI JIANLI PENGZE XICHANG XIAPU OCT A 1.0 1.0 0.70 0.57 0.32 1.0 1.0 B 0.0 0.0 0.0 0.0 0.24 0.0 0.0 C 0.0 0.0 0.0 0.0 0.04 0.0 0.0 D 0.0 0.0 0.30 0.30 0.26 0.0 0.0 E 0.0 0.0 0.0 0.13 0.12 0.0 0.0 E 0.0 0.0 0.0 0.0 0.02 0.0 0.0 6PGD A 1.0 1.0 0.0 1.0 1.0 1.0 B 0.0 150 1.0 PGM-1 A 0.72 0.0 0.07 0.0 0.04 1.0 0.0 B 0.28 1.0 0.53 0:77 0.62 0.0 1.0 С 0.0 0.0 0.37 0.20 0.34 0.0 0.0 D 0.0 0.0 0.03 0.03 0.0 0.0 0.0 PGM-2 A 1.0 0.0 1.0 1.0 0.90 1.0 0.08 В 0.0 1.0 0.0 0.0 0.02 0.0 0.93 С 0.0 0.0 0.0 0.08 0.0 0.0 SDH-1 A 0.82 0.94 1.0 0.0 0.72 1.0 1.0 B 0.18 0.06 0.0 0.90 0.0 0.0 0.0 C 0.0 0.0 0.0 0.10 0.0 0.0 0.0 D 0.0 0.0 0.0 0.0 0.28 0.0 0.0 XDH A 1.0 1.0 1.0 1.0 0.0 1.0 1.0 В 0.0 0.0 0.0 1.0 0.0 0.0 TABLE 4. Pairwise genetic distances among the seven populations of group I; Nei’s (1978) D below the diagonal; arc D above the diagonal. The lower set includes the AAT-2 locus, which was not scorable for XiChang and therefore excluded from the upper set. POPULATION DA РО GC Jl PZ XI XP Dali — 0.443 0.456 0.402 0.538 0.339 0.392 FuQing 0.223 — 0.499 0.466 0.503 0.460 0.314 GuiChi 0.221 0.287 — 0.436 0.482 0.510 0.448 JianLi 0.171 0.255 0.217 — 0.513 0.483 0.417 PengZe 0.334 0.302 0.278 0.312 — 0.515 0.521 XiChang 0.107 0.227 0.292 0.244 0.301 — 0.455 XiaPu 0.176 0.108 0.233 0.200 0.346 0.230 — POPULATION DA FQ GC Jl PZ XI XP Dali — 0.395 0.411 0.364 0.481 — 0.358 FuQing 0.214 — 0.449 0.419 0.452 — 0.289 GuiChi 0.214 0.277 — 0.391 0.432 — 0.404 JianLi 0.164 0.245 0.210 — 0.456 — 0.381 PengZe 0.320 0.289 0.268 0.298 — — 0.473 XiChang — — — — = — — XiaPu 0.173 0.107 0.227 0997 0.337 — = ribbed group of five, and separately the Ning est to the Dali group (Dali group—Tong Ling, Guo population, show that in each case, the 0.116; Dali group—Jiang Li, 0.189; Dali putative hybrid population is genetically clos- group—Gui Chi, 0.227). These three excep- 144 DAVIS, ZHANG, GUO & SPOLSKY TABLE 5. Loci fixed for a single allele or having alternative alleles Invariant loci Group 1 AO X Group 2 > A | >< << >< | << < m Sr EEE ui № Invariant except population G6PD-2 EST-1 Fixed for alternative alleles | © > | хх | = © E De SS | За | CPR ee | Alternative allele except population 3 (GC) 3 (GC) = MDH = 8 (JL) tional populations are also fairly closely re- lated to each other (average mD 0.187), but not clearly to any other group. DISCUSSION Population Genetics Populations of Oncomelania studied here are not unusual in having few alleles per lo- cus (1.0 to 1.6), a low mean heterozygosity (0.008 to 0.036), and low percentage of poly- morphic loci (4 to 28). In a previous study of Oncomelania from China and the Philippines, Woodruff et al. (1988) found 16 polymorphic loci among 21 loci studied, with 1.8 to 2.1 alleles per locus, in two populations from Gui Chi in Anhui Province; mean heterozygosity was 0.19 to 0.20, and the percentage of poly- morphic loci ranged from 52 to 62. The major difference between these two studies of On- comelania 1$ that in this study we had results from 29 loci. Had we studied only 21 loci, Gui Chi of our study would have had 33% poly- morphic loci rather than 24%. This, however, is not the only source of discrepancy. Wood- ruff et al. (1988) reported polymorphism at three loci where we found none: LAP, ACP, ME; in addition, they had a polymorphism at PEP locus we did not study. Even with ex- tensive additional screening of our Gui Chi population (250 additional individuals; 38 loci), the number of polymorphic loci still re- mains relatively low, 44%. The results found in our study were similar to those found in sister taxa of rissoacean snails in the genus Hydrobia (Hydrobiidae): Davis et al. (1988, 1989), one species, six populations (30 loci); Haase (1993), three species (25 loci); and in Truncatella: Rosen- berg (1989), five species (19 loci), one spe- cies with four populations. п the Truncatella study, four populations had a mean heterozy- gosity above 0.036 (0.037 to 0.077). How- ever, the 0.077 value derived from a mean sample size of 5.3 snails analyzed per locus. The mean number of alleles per locus varied from 1.1 to 1.4; mean heterozygosity varied from 0.006 to 0.077; the percentage of poly- morphic loci varied from 5.3 to 31.6. In sum- mary, the Oncomelania populations studied here are normally outbreeding rissoacean- grade snails with an apparent usual pattern of low heterozygosity and low percentage of polymorphic loci. Genetic distance and taxonomic discrimination Where do populations stop and higher taxa begin? This topic was reviewed extensively by Davis (1994) with particular reference to Oncomelania hupensis. Large genetic dis- tances by themselves do not serve to define species. There is no magical cut off point be- low which are populations and above which are species. We have demonstrated the gradual rise of Nei’s D to the 0.400 level in pairwise comparisons of populations across China, but again, this does not hold if one excludes the three candidate hybrid popula- tions. We documented the very great inter- population variance in Nei's minimal D for the ribbed populations traditionally classified as POPULATION GENETICS OF ONCOMELANIA HUPENSIS 145 МЕ! '5 1978 D CHANG XING, ZHEJIANG $ TONG LING, ANHUI JIANG LING, HUBEI НАМ YANG, HUBEI GUI PING, $ А+ У R У YUE YANG, HUNAN В+ У R+ У GUANGXI 5+ У 5 V ARC D NING GUO, ANHUI А 1 1 == J 12 08 04 te) CHANG XING, ZHEJIANG S NV TONG LING, ANHUI R+ V JIANG LING, HUBEI в у YUE YANG, HUNAN А+ У HAN YANG, HUBEI А+ У GUI PING, GUANGXI S+ V NING GUO, ANHUI 5 У 1 _ —E =i 1 2 НЕЕ -60 .54 .48 .42 .36 .30 24 = 5. Е 18 12 .06 о FIG. 6. UPGMA derived phenograms based on Nei’s 1978 D and Cavalli-Sforza's Arc D for group | populations. Oncomelania hupensis hupensis (range of 0.048 to 0.324). Woodruff et al. (1988) showed that On- comelania hupensis quadrasi populations from the Philippines differed from Oncomel- ania hupensis hupensis from China by a D of 0.62 + 0.04 and on this basis, invoking the “evolutionary species concept,” stated that two species rather than subspecies were in- volved. On the other hand, Davis (1994) noted that the land snail Cepaea nemoralis introduced from Europe to the southern U.S.A. differed by a D of 0.631 between con- tinents (Johnson et al., 1984) and that a pop- ulation near Pavia, Italy, differed from another from Florence, Italy, by D = 0.391. This spe- cies, used in various paradigms in evolution, is well known for geographic variation in shell polymorphisms and allozymes. Quoting Johnson et al. (1984), “The decoupling of ge- netic distance from speciation emphasizes the limitations of viewing the process of spe- ciation solely in genetic terms” (see also Stine, 1989 and Murray et al., 1991). The large genetic distances in our and the Woodruff studies are caused by six or more loci with alternative allele(s). Alternative alle- les in allopatric populations are largely re- sponsible for the results seized upon by Woodruff et al. (1988) to defend an evolution- ary or phylogenetic species concept. But is this justified? Species concepts and Oncomelania In describing new species, and especially in applying molecular genetic data in the pro- cess, one must commit to a species concept and be prepared to defend that concept. Davis (1994) reviewed this topic in arguing for Templeton's (1989) “cohesion” model, which includes the biological species con- cept whenever it applies as well as Patter- son's (1985) recognition concept. The cohe- sion model integrates population genetics and ecology with standard studies of mor- phology. The concept can be applied to all organisms from outbreeders to syngameons or parthogenetic organisms. Among populations of Oncomelania hu- pensis one finds considerable cohesion! Given that each morphological character 1$ controlled by one to several genes, morpho- logical distance is a measure of relative ge- netic distance. Aside from size and a few shell characters, the allopatric populations of this species are qualitatively the same. Ac- cordingly, the relative genetic divergence ex- pressed in morphology 15 extremely low. Re- productive cohesion is also large. Allopatric 146 DAVIS, ZHANG, GUO 8 SPOLSKY TABLE 6. Allele frequencies for seven populations of Oncomelania hupensis from south-central China (group II). 28 loci (no CK; GDH). М = 25 for all populations, all loci. There were 12 invariant loci; see Table 5. Population Locus ChangXing TongLing JiangLing YueYang GuiPing NingGuo HanYang AAT-1 A 1.0 1.0 0.0 1.0 1.0 0.0 1.0 B 0.0 0.0 1.0 0.0 0.0 0.0 0.0 C 0.0 0.0 0.0 0.0 0.0 1.0 0.0 AAT-2 A 1.0 1.0 0.92 0.96 1.0 1.0 1.0 B 0.0 0.0 0.02 0.0 0.0 0.0 0.0 C 0.0 0.0 0.06 0.0 0.0 0.0 0.0 D 0.0 0.0 0.0 0.04 0.0 0.0 0.0 ACPH A 0.0 0.0 0.0 0.0 1.0 0.0 0.0 B 1.0 1.0 1.0 1.0 0.0 1.0 1.0 EST-1 A 1.0 1.0 0.44 1.0 1.0 1.0 1.0 B 0.0 0.0 0.44 0.0 0.0 0.0 0.0 С 0.0 0.0 0.0 0.0 0.0 0.0 0.0 D 0.0 0.0 0.12 0.0 0.0 0.0 0.0 EST-2 A 1.0 0.46 1.0 0.46 0.46 0.50 0.78 B 0.0 0.42 0.0 0.46 0.0 0.50 0.0 C 0.0 0.12 0.0 0.08 0.54 0.0 0.22 СР! А 0.0 0.80 0.70 0.72 0.48 0.38 0.80 В 0.0 0.0 0.14 0.16 0.0 0.62 0.20 С 0.0 0.02 0.0 0.0 0.0 0.0 0.0 D 1.0 0.02 0.04 0.04 0.0 0.0 0.0 Е 0.0 0.10 0.02 0.02 0.0 0.0 0.0 E 0.0 0.02 0.04 0.02 0.0 0.0 0.0 G 0.0 0.02 0.06 0.04 0.52 0.0 0.0 ISDH-2 A 1.0 1.0 0.0 0.0 0.0 0.0 0.0 B 0.0 0.0 1.0 1.0 1.0 1.0 1.0 LDH A 1.0 1.0 0.0 0.0 0.0 1.0 0.0 С 0.0 0.0 1.0 1.0 0.0 0.0 1.0 D 0.0 0.0 0.0 0.0 1.0 0.0 0.0 MDH A 0.0 0.0 0.98 1.0 0.0 0.0 0.0 B 1.0 1.0 0.0 0.0 1.0 1.0 1.0 C 0.0 0.0 0.02 0.0 0.0 0.0 0.0 ME-2 A 1.0 О 1.0 0.0 0.0 1.0 1.0 В 0.0 0.0 0.0 1.0 1.0 0.0 0.0 МАРО А 1.0 1.0 0.0 0.0 0.0 0.0 0.0 В 0.0 0.0 1.0 1.0 1.0 0.0 1.0 С 0.0 0.0 0.0 0.0 0.0 1.0 0.0 OCT A 1.0 0.54 0.40 0.34 120 0.88 0.56 B 0.0 0.14 0.0 0.06 0.0 0.08 0.04 C 0.0 0.0 0.0 0.0 0.0 0.04 0.0 D 0.0 0.26 0.56 0.60 0.0 0.0 0.38 E 0.0 0.06 0.04 0.0 0.0 0.0 0.02 POPULATION GENETICS ОЕ ONCOMELANIA HUPENSIS 147 TABLE 6. (Continued) Locus ChangXing TongLing JiangLing 6PGD A 1.0 1.0 0.0 B 0.0 0.0 1.0 PGM-1 A 0.32 0.0 0.0 B 0.68 0.64 0.52 C 0.0 0.34 0.44 D 0.0 0.02 0.04 Е 0.0 0.0 0.0 РСМ-2 А 140 1.0 1.0 B 0.0 0.0 0.0 XDH A 1.0 0.0 0.0 B 0.0 1.0 1.0 Population YueYang GuiPing NingGuo HanYang 0.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 0.0 0.0 0.06 0.02 0.62 0.46 0.94 0.44 0.38 0.54 0.0 0.50 0.0 0.0 0.0 0.02 0.0 0.0 0.0 0.02 1.0 1.0 0.02 0.98 0.0 0.0 0.98 0.02 0.0 0.0 0.0 0.0 1.0 1.0 1.0 1.0 TABLE 7. Matrices of genetic distances for group Il populations. Nei's (1978) D is given above the diagonal; arc D below the diagonal. Population CX TL JL ChangXing — 0.089 0.381 TongLing 0.306 — 0.293 JiangLing 0.559 0.492 — YueYang 0.558 0.470 0.310 GuiPing 0.556 0.490 0.468 NingGuo 0.506 0.448 0.469 HanYang 0.487 0.397 0.302 populations from the Philippines, Japan, China can readily interbreed and produce vi- able F,, F, РГ, generations (reviewed in Davis, 1980, 1981). Thus, there is no barrier to reproduction; clearly, allozymic differenti- ation does not impinge on overall genetic co- hesiveness. It appears that gene divergence, as seen in allozymes, can change relatively rapidly in Oncomelania hupensis, while over- all morphological and breeding system rec- ognition patterns do not. Because the fundamental niche parame- ters for allopatric populations of the species are the same (van der Schalie and Davis, 1968; Davis, 1971), the populations are de- mographically exchangeable. Behavior is the same for all populations, whether from Ja- pan, the Philippines or China; that 1$ these snails climb upward. In dry periods, they can be found cemented to stone walls bounding canals or, in the Yangtze flood plain, a meter or more up on tree trunks. All populations lay their eggs singly, each coated with fine mud. YY GP NG HY 0.382 0.364 0.301 0.264 0.267 0.286 0.231 0175 0.102 0.261 0.263 0.093 — 0.155 0.287 0.089 0.373 — 0.296 0.140 0.483 0.506 = 0.193 0.291 0.360 0.403 —= Species of Oncomelania In contrast to the biological and ecologi- cal cohesion among populations of On- comelania hupensis, there is considerable disruption between the two recognized spe- cies of Oncomelania: O. hupensis and О. minima (Bartsch, 1936a). Oncomelania min- ima comes from northwestern Honshu, Ja- pan. The type locality is the Noto Peninsula, Ishikawa Prefecture. The species are readily distinguished by differences in shell morphol- ogy, genital and gill anatomy, and ecology (Davis, 1971). Oncomelania minima can be amphibious, as is O. hupensis, but the former is abundant on sticks, leaves, rocks in a nar- row mountain stream. Oncomelania minima is also abundant on rock slabs over which there was a great amount of water seepage- flow. Thus, these snails live much as do many species of the sister genus Tricula (Pomati- opsidae: Triculinae); this mode of life is not that of O. hupensis. 148 DAVIS, ZHANG, GUO 8 SPOLSKY NEI'S 1978 D DA Ll, YUNNAN $ NV XI CHANG, SICHUAN $ NV JIAN Li, HUBEI в У FU QING, FUJIAN Ss У XIA PU, FUJIAN SV GUI CHI, ANHUI В+ У РЕМС 2НЕ, JIANGXI В+ У L 4 L 1 1 1 L 1 L L 1 1 4 1. L 1 4 J 40 36 32 .28 24 20 16 12 .08 04 o ARC D DA LI, YUNNAN $ NV XI CHANG SICHUAN $ NV FU QING, FUJIAN SAV, XIA PU, FUJIAN SAN GUI CHI, ANHUI R+ V JIAN Li, HUBEI R У РЕМС 2НЕ, JIANGXI В+ У L N 1 1 1 L 1 + 1 1 it 1 = 1 = L L 1 J -60 .54 .48 .42 .36 -30 .24 18 12 -06 o FIG. 7. UPGMA derived phenograms based on Nei’s 1978 D and Arc D for group И populations. TABLE 8. Deviations from Hardy-Weinberg (H. W.) equilibrium for all populations studied; Р = probability; P level accepted = 0.05. No. loci Locus not Probability Population polymorphic in H.W. x pooled exact GROUP | POPULATIONS (29 loci) Dali 4 SDH-1 0.01 — 0.03 FuQing 2 — — — — GuiChi 6 EST-3 0.0 — 0.04 Лапы 5 EST-2 0.01 0.02 0.03 OCT 0.0 0.02 0.03 PengZhe Uf SDH-1 0.002 — 0.01 XiChang 1 — — — — XiaPu 2 — — — — GROUP II POPULATIONS (28 loci) ChangXing 1 PGM-1 0.02 — 0.03 TongLing 4 EST-2 0.001 0.02 0.04 JiangLing 5 — — — = YueYang 5 AAT-2 0.0 — 0.02 GuiPing 3 EST-2 0.0 — 0.02 NingGuo 5 EST-2 0.0 — 0.0 HanYang 5 — — — — Тре type of morphological discontinuity and ecological divergence that allows recog- nition of these two species parallels that of the types of character-state changes seen among species of the sister subfamily Tricu- linae. Closely related triculine species differ by a combination of character states, such as given above and: penis with papilla in one species, without in another; seminal vesicle a coil dorsal to the gonad vs. a knot on the posterior stomach; penis with pronounced ejaculatory duct in the base of the penis in one species vs. absent in another; outer mar- ginal tooth with specialized outer cusp vs no POPULATION GENETICS OF ONCOMELANIA HUPENSIS 149 TABLE 9. Matrix of genetic distances (mD) for the combined set of fourteen populations. R = ribbed; S = smooth; V = marix; NV = no varix. Population Dali, Yunnan FuQing, Fujian $, V — GuiChi, Anhui В+, V — JianLi, Hubei В, V — PengZe, Jiangxi R+, V — XiChang, Sichuan S, NV XiaPu, Fujian S, V ChangXing, Zhejiang S, NV TongLing, Anhui В+, V JiangLing, Hubei R, V DA FQ GC Jl PZ S,NV — 0.218 0.209 0.166 0.297 0.110 0.215 0.007 0.101 0.344 0.308 0.296 0.294 0.255 0.266 0.241 0.276 0.225 0.154 0.205 0.254 0.334 0.376 0.453 0.296 0.404 0.202 0.247 0.270 0.262 0.203 0.204 0.299 0.245 0.260 0.325 0.286 0.274 0.232 0.234 0.169 0.156 0.339 0.280 0.378 0.372 0.324 0.275 0.349 0.291 0.208 0.097 0.129 0.231 0.256 0.167 XI XP CX TL JL YY GP NG НУ — 0.266 0.127 0.155 0.292 0.367 0.361 0.328 0.307 — 0.201 0.243 0.306 0.286 0.357 0.253 0.315 — 0.092 0.337 0.300 0.290 0.281 0.249 YueYang, Hunan В+, V — 0.150 0.222 0.048 GuiPing, Guangxi S+, V — 0.232 0.098 NingGuo, Anhui SV — 0.185 HanYang, Hubel В+, V — NEIL'S MINIMUM DISTANCE DA LI, YUNNAN $ NV CHANG XING, ZHEJIANG S NV TONG LING, ANHUI А+ У XI CHANG, SICHUAN $ NV JIAN LI, HUBEI в Vv GUI CHI, ANHUI R+ V FU QING, FUJIAN sv XIA PU, FUJIAN зу PENG РНЕ, JIANGXI R+ У JIANG LING, HUBEI в V YUE YANG, | HUNAN R+ У HAN YANG, HUBEI R+ V GUI PING, GUANGXI S+ V NING GUO, ANHUI sv .12 .08 .04 lo) FIG. 8. UPGMA derived phenogram based on Nei's minimum D (Table 9) for both population groups combined. specialized outer cusp; and so forth. In sum, species in the Pomatiopsinae and the sister clade, Triculinae, regularly demonstrate mor- phological discontinuities equatable to con- siderable relative genetic divergence. Such discontinuity is not found within the On- comelania hupensis species complex. The Oncomelania hupensis Polytypic Species Complex The data we have thus far for this species indicates that overall cohesiveness—mor- phological, genetic, and ecological—has not been disrupted to the extent that allopatric populations in this complex should be ac- corded species status. Nevertheless, evolu- tion towards species rank has progressed to varying degrees in many allopatric popula- tions both within and outside China. In China, Oncomelania is distributed in 12 provinces and 347 counties (Kang, 1985). It has been argued by Liu et al. (1981) that, based on shell differences, there are five sub- species in mainland China: O. hupensis hu- pensis; O. hupensis robertsoni (Bartsch, 1946); O. hupensis tangi (Bartsch, 1936b); O. hupensis fausti (Bartsch, 1925); and O. hu- pensis guangxiensis (Liu et al., 1981). On the basis of the data presented here, we are in- clined to accept three of them. Before discussing allozyme results and the 150 DAVIS, ZHANG, GUO & SPOLSKY O. hupensis hupensis Han Yang x JiangLing GuiChi JianLi O. h. robertsoni O. h. tangi HJ 0.10 FIG. 9. An unrooted FITCH tree based on Nei's mD for both population groups combined. Line lengths are proportional to branch lengths. basis for accepting three subspecies, it is im- portant to understand the genetic basis for shell characters. Historically, the plesiomor- phic relevant character states are: small size, smooth shell, no varix, colorless apex (Davis, 1979). Today, differences among populations are seen in the increased shell size, sculpture (smooth to heavily ribbed), varix or lack of same, width/length ratios, and apex color. Hybridization and schistosome susceptibility POPULATION GENETICS ОЕ ONCOMELANIA HUPENSIS 191 TABLE 10. Comparison of Nei's mD (юг mD < 0.157), shell morphology, and geographic distance; $ = smooth shells; R = ribbed shells (see Table 1 caption). Distances in km. 0.007 Dali, Yunnan x ChangXing, Zhejiang 0.048 HanYang, Hubei x YueYang, Hunan 0.059 JiangLing, Hubei x YueYang, Hunan 0.092 ChangXing, Zhejiang x TongLing, Anhui 0.092 JiangLing, Hubei x HanYang, Hubei 0.097 PengZe, Jiangxi x JiangLing, Hubei 0.098 GuiPing, Guangxi x HanYang, Hubei 0.101 Dali, Yunnan x TongLing, Anhui 0.110 Dali, Yunnan x XiChang, Sichuan 0.127 XiChang, Sichuan x ChangXing, Zhejiang 0.129 PengZe, Jiangxi x YueYang, Hunan 0.150 YueYang, Hunan x GuiPing, Guangxi 0.154 XiaPu, Fujian x FuQing, Fujian 0.155 XiChang, Sichuan х TongLing, Anhui 0.156 JianLi, Hubei x TongLing, Anhui 0.204 GuiChi x TongLing/ANHUI 0.339 JiangLing x JianLi/HUBEI SxS 2,052 (most distant) R+ x R+ 62.5 Rx R+ 141.1 SxR+ 199.0 Rx R+ 175.0 R+ xR 438.3 S+ x В+ 899.3 S x R+ 1,854 SxS 321.4 SXS 1,784 R+ x R+ 312.0 R+ x 5+ 660.0 эхо 142.3 S x В+ 1,567 Rx R+ 442.0 R+ x В+ 44.4 (closest neighbors) RxR 72.1 TABLE 11. Pairwise comparisons of Nei's (1978) D values in a gradation from small to large values along with the provinces involved. From Tables 4 and 7. то Populations compared 0.089 TongLing x ChangXing 0.102 HanYang x YueYang 0.140 HanYang x GuiPing 0.155 GuiPing x YueYang 0.176 XiaPu x Dali 0.193 HanYang x NingGuo 0.221 GuiChi x Dali 0.255 JianLi x FuQing 0.261 GuiPing x JiangLing 0.287 GuiChi x FuQing 0.292 GuiChi x XiChang 0.301 XiChang x PengZe 0.334 PengZe x Dali 0.346 XiaPu x PengZe 0.364 GuiPing x ChangXing 0.382 YueYang x ChangXing Geographic location Anhui—Zhejiang Provinces Hubei Province Hubei—Guangxi Provinces Guangxi—Hubei Provinces Fujian—Yunnan Provinces Hubei—Anhui Provinces Anhui—Yunnan Provinces Hubei—Fujian Provinces Guangxi—Hubei Provinces Anhui—Fujian Provinces Anhui—Sichuan Provinces Sichuan—Jiangxi Provinces Jiangxi—Yunnan Provinces Fujian—Jiangxi Provinces Guangxi—Zhejiang Provinces Hubei—Zhejiang Provinces studies underscore the importance of these sculptural aspects (Davis and Вий, 1973). Ribbing is dominant to smooth, involving a single gene with multiple alleles. Large size is dominant to small as is higher whorl number. Hybrid vigor was demonstrated, as F, snails had 7.5 to 8.5 whorls, whereas parental snails had 5.0 to 6.5 whorls. Susceptibility 1$ dominant to resistance. In addition to shell characters, pigmentation is dominant to albi- nism. Ribbing 1$ restricted to the mainland of China and especially to low-land flood plains and marshy areas adjacent to the Yangtze River or rivers flowing into the Yangtze. At higher elevation, above the effects of annual flooding, the shells of snails become smooth. All snail populations in the highland areas west of the three gorges barrier on the Yangtze River, in Yunnan and Sichuan, are smooth and without a varix. Ribbing seems to be selected for relative to dispersal and survival during rampaging floods of the Yangtze River drainage. The varix, the rib-like thickening at the lip of the shell is, it seems, the last rib. Yet, there are smooth shells that may or may not have a varix. All shells with ribs have a varix. Reten- tion of the varix in smooth shells seems to be the genetic loss of all ribs except the last one. It is also probable that the loss of ribbing with retention of the varix has occurred indepen- 152 DAVIS, ZHANG, GUO 8 SPOLSKY TABLE 12. Comparison of populations based on shell sculpture using Nei’s minimum D |. Smooth x Smooth Shells Dali x Sichuan x Zhejiang [no varix] Fujian [strong, wide varix] Dali groups x Fujian group NingGuo [smooth, varix] х Dali group NingGuo x Fujian group GuiPing [smooth, varix] х Dali group GuiPing x Fujian group |. Ribbed x Ribbed Shells All ribbed populations (incl. hybrids) Ribbed populations (excluding hybrids) 0.081 + 0.06 0.154 0.230 + 0.024 0.301 + 0.024 0.274 0.316 + 0.036 0.405 CNL, EN EN CN BL, вет ито ND © D © O — w 0.204 + 0.085 0.160 + 0.092 ZZ — N er a (includes НУ, GP, YY, PZ, JL, NG = Ribbed “Group of 6”) Smooth x Ribbed Dali group x All ribbed populations Dali group x Group of 6 Fujian group x All ribbed populations Fujian group x Group of 6 NingGuo x rest of group of 6 GuiPing x rest of group of 6 IV. TongLing [ribbed, possible hybrid] x Other Groups TongLing x Dali group TongLing x Fujian group TongLing x Group of 5 TongLing x NingGuo V. GuiChi [ribbed, possible hybrid] x Other Groups GuiChi x Dali group GuiChi x Fujian group GuiChi x Group of 5 GuiChi x NingGuo VI. JiangLi x Dali group JiangLi x Fujian group JiangLi x Group of 5 JiangLi x NingGuo dently in different geographic regions and at different times. Thus, one observes the fol- lowing characters and character-states. Varix: absent (0), present (1); Varix shape: high and pronounced (0), low and wide (1). The subspecies we accept and the basis for accepting them are given below: Oncomelania hupensis hupensis Gredler, 1881, is strongly ribbed and with a strong rib-like varix. Shells are tall, large. This sub- species occurs throughout the mid to lower Yangtze River basin, especially Hubei, Hunan, and Anhui provinces, and the Bei River in Guangdong Province (Liu et al., 1981). At first glance, our populations 3, 4, 5, 9, 10, 11, 12 (Table 1) appear to fit this de- scription. Nei’s mD among these seven pop- ulations is 0.204 + 0.085 (N = 21). As dis- JiangLi [ribbed, possible hybrid] x Other Groups 0.257 + 0.077 0.304 + 0.056 0.296 + 0.053 0.334 + 0.089 0.227 + 0.036 0.182 + 0.042 225.726 FL, FL, FL CE NL A UT NE Hall Sn On SS 105) MORO 0.116 + 0.032 0.249 0.212 + 0.048 0.217 CLE, PEPE ou out — O1 ND © 0.227 + 0.034 0.264 0.267 + 0.027 0.325 A de I mM — O1 D © 0.189 + 0.033 0.238 0.319 + 0.052 0.378 PATA PS PE, шит + On Ww cussed below, however, we also include two smooth-shelled populations, 8 and 13, within this grouping, but exclude ribbed popula- tions 3, 5, and 12. We included within O. hupensis hupensis the nominal subspecies O. h. fausti and O. h. guangxiensis. Oncomelania h. fausti has O. h. hupensis-sized shells that are smooth but with a strong varix. These smooth-shelled snails live in uplands beyond the reach of flooding of the Yangtze River and tributaries. The two nominal taxa live over the same geo- graphical regions. Lou et al. (1982) con- cluded that they are synonymous. In Hubei Province, one finds streams where the smooth form is at the headwaters and the ribbed form is along the flood plain. In inter- mediate zones, there appears to be intergra- POPULATION GENETICS ОЕ ONCOMELANIA HUPENSIS 153 dation of the two sculptural types. It is ap- parent that ribbing is associated with flood plains and the smooth shells associated with upland areas beyond the effects of flooding. There are no physical or reproductive barriers between the sculptural types. The upland habitats only reflect elevation and freedom from annual flooding, not differences in the microhabitat. The one population of the “fausti” type that we studied, no. 13 from Ning Guo, Anhui Province, differs from the other members of the coherent group of six ribbed populations (Table 12) by an mD of 0.227 + 0.062 (N = 13); from the two smooth- shelled groups, Dali and Fujian, by 0.301 and 0.274, respectively. Further, in both phyloge- netic analyses (Figs. 8, 9), this “fausti” type clusters with the group of predominantly ribbed-shelled taxa. We noted previously that Nei’s mD for the latter populations is 0.160 + 0.07. Accordingly, we do not con- sider O. h. fausti to be a distinct subspecies but a smooth form of O. h. hupensis. Liu et al. (1981) described a new subspe- cies, O. hupensis guangxiensis from the northwestern part of Guangxi along the Yu Jiang and Hongshui river systems. The shells were described as medium sized, rather thin, smooth and with a weak varix. The type and the description of the taxon puts emphasis on deep sutures and especially rounded body whorl. Our Guangxi population from Gui Ping is at the confluence of the Yu Jiang and Hongshui rivers. One of our specimens re- sembles the figured type, but the remaining somewhat resemble the “O. h. fausti’ type. The shells are smooth and the varix weak in 90% of the snails. However, there are weak scattered ribs or pronounced growth lines on the shells of some individuals. It differs from the two smooth-shelled groups by Nei’s тб of 0.316 and 0.405 respectively; it differs from the other members of the coherent group of six ribbed populations by 0.182. As with our “fausti” population, this population clusters even more tightly with the coherent group of O. h. hupensis. The faint ribs seen on some shells and its genetic affinity for taxa in the lower cluster indicates to us that it 1$ part of the ribbed-shell group, i.e. O. A. hu- pensis, and not a distinct subspecies. How- ever, topotypes must be studied to confirm our opinion. This then brings into question the affinity of the three ribbed populations in the upper cluster of Figure 8, Tong Ling, Jian Li, and Gui Chi. Because morphologically and geo- graphically, these populations appear to be O. h. hupensis, but genetically and phyloge- netically, they appear most closely allied to the smooth-shelled Dali group, we raise the possibility that these three populations are a result of hybridization between smooth- shelled taxa, such as O. h. robertsoni as dis- cussed below (Dali, Chang Xing, and Xi Chang), and a ribbed O. h. hupensis popula- tion. It is particularly compelling to make this hypothesis given: (1) that ribbing is dominant to smooth and the populations in question come from the flood plains; (2) an average mD to the three O. h. robertsoni populations of only 0.116 + 0.034; and (3) the close geo- graphic proximity ofthe Tong Ling population and the transported smooth-shelled popula- tion from Chang Xing, Zhejiang Province; the genetic distance between these two popula- tions is only 0.092. In further discussion, when we refer to ribbed populations, we therefore include only the coherent group of six populations that cluster together in phy- logenetic analyses (Figs. 8, 9). Oncomelania hupensis robertsoni (our populations Dali, Xi Chang, Chang Xing) 1$ smooth and without varix. It is located in Si- chuan and Yunnan provinces at altitudes from 200 m up to 2,000 m (Liu et al., 1981) in ditches along the slopes of hills, at the edges of fields and irrigation canals in basin areas. The geographic region is isolated from other provinces with respect to invasion down the Yangtze River by the long stretch of the treacherous Yangtze Gorges west of Hubei Province. One population assigned to this subspecies is enigmatic in that it comes from Zhejiang Province at the eastern end of China (Group Il, Chang Xing), a locality that is the farthest removed from Yunnan-Sichuan of all populations studied, yet the least divergent from the Yunnan population (Nei’s mD of 0.007). The presence of this population can- not be due to dispersal by flotation down the Yangtze, given the time involved to disperse over such a great distance and the genetic near-identity, but rather to a recent introduc- tion either by man or birds. These three pop- ulations cluster together tightly by both methods of analysis (Figs. 8, 9; mD = 0.081). They differ from the ribbed group (О. h. hu- pensis) by an average mD of 0.305 + 0.071 (N = 34). They also differ from the smooth- shelled Fujian group (O. hupensis tangi) by 0.220 + 0.024. Thus, there is a significant dif- ference in genetic distance between these two smooth-shelled groups. 154 DAVIS, ZHANG, GUO 8 SPOLSKY Oncomelania hupensis tangi (our Fu Qing, Xia Pu populations) has smooth shells with a low but very wide varix. The width of the shells is greater relative to height than in O. h. robertsoni. Oncomelania h. tangi lives along the coast of Fujian Province in hilly environ- ments (from 50 to 500 m altitude) as well as in small ditches of the seaside lowlands. This region is very isolated from other regions with Oncomelania (Fig. 1). Our two populations form a separate cluster (Figs. 8, 9); they differ from all others by an mD of 0.285 + 0.066 (N = 24); from the ribbed group by 0.334 + 0.089 (N = 10). Given an intragroup mD of 0.154, there is a decided gap between this taxon and О. h. hupensis or O. В. robertsoni. There are no fixed diagnostic alleles unique to any of the nominal subspecies. While ge- netic distance per se is not a measure of sub- species status, a subspecies should be co- herent genetically as well as morphologically. Phylogenetic clusters shown in Figs. 8 and 9 do provide evidence for genetic cohesive- ness supporting the three subspecies we ac- cept, if the arguments we provided concern- ing “fausti,”” “guangxiensis,” and possible hybrids hold up. Additional arguments to support the subspecies are: (1) regional iso- lation; (2) shell size/shape; (3) presence or absence of shell ribbing, but recognizing that hybridization in the flood plains will yield a ribbed shell; (4) presence or absence of varix; (5) varix, if present, low and wide (0), weak (1), strong and rib-like (2); (6) differ- ences in susceptibility to schistosomes (Liu et al., 1981); (7) low interpopulation genetic distances as well as low percentage of poly- morphic loci within the O. h. robertsoni and О. В. tangi smooth-shelled groups, as well as within the coherent O. h. hupensis group. Hope and McManus (1994) carried out PCR-based RFLP analyses of variation in the ITS-region of rDNA repeat unit among four populations of Oncomelania hupensis т China, three from the Philippines, and one from Japan. Their data are puzzling because the largest divergence they find is between the Yunnan and Sichuan populations in China, populations we classify as O. hupen- sis robertsoni because of cohesiveness in al- lozymes, shell sculpture, lack of varix, and biogeographic closeness; and additionally, because of lowest divergences among pop- ulations in cytochrome b gene sequences (Spolsky and Davis, unpubl.). The discrep- ancy may be the result of the paucity of data points (DNA fragment bands) in the RFLP LE] ‘6 analyses and of the inability to distinguish small differences in fragment size. We there- fore advise caution in choice of tool to dem- onstrate population divergence. All studies of populations should give illustrations of the shells to ascertain shell phenotypes. Voucher specimens should be deposited for refer- ence. SUMMARY The assortment of four shell characters to distinct geographic regions and regional т- fectivity patterns argue for subspecific sta- tus. Low allozymic heterozygosity, unique combinations of shell characters, coherent genetic clustering and geographic isolation indicate the usefulness of recognizing sub- species for two smooth-shelled taxa. The presence of a different type of varix on the shells of these two nominal subspecies indi- cates to us that they had independent ori- gins. The phylogenetic analyses confirm this. Major points to be considered are: 1. The three regional sets of populations should be treated as subspecies. Allozyme differences are insufficient by themselves, but together with geographic location and shell characters, enable recognition of each of these population sets as subspecies. 2. In hilly areas above the Yangtze River flood effects, ribbing becomes reduced and lost. This does not affect subspecific status. 3. The disjunction between morphology and genetic distance in three ribbed popula- tions indicates possible hybridization be- tween subspecies, in particular between O. h. robertsoni and O. h. hupensis. 4. The considerable interpopulation ge- netic distance between allopatric popula- tions of the same shell type, especially in the O. hupensis hupensis complex (that includes all populations with ribs), shows genetic change in the absence of anatomical change. We predict that there would be parallel ge- netic differences in the populations of Schis- tosoma japonicum transmitted by these allo- patric snail populations; such genetic differentiation should be visible in DNA se- quences of genes such as cytochrome b. Molecular differences in the snails and their parasites evolve at a different tempo and mode than does the anatomical ground plan. Would sufficient differences be found to de- fine a series of morphostatic (term defined in Davis, 1992) subspecies where differences POPULATION GENETICS OF ONCOMELANIA HUPENSIS 159 might be driven by localized parasitism in a tightly coevolved system as suggested by Davis (1992)? 5. A weakness of this paper is that there was not enough material to confirm the allele align- ments on which the integrated data set in Table 9 is based. Each data set by itself is confirmed. What is now needed are: (1) a de- tailed analysis of populations within a prov- ince grouped along river drainage systems from highlands to point of entry to the Yangtze River or other primary river; (2) a comparison between selected populations of the different subspecies using more than 50 snails per population in order to increase reliability in estimates of polymorphic loci; (3) a careful analysis of populations of O. hupensis hupen- sis from selected streams where there is a demonstrated change from ribbed shells on the flood plains to smooth shells at higher elevations. Clearly, this baseline study has set the stage for future, carefully targeted studies. ACKNOWLEDGMENTS We thank Caryl Hesterman for running the gels; all gels were scored by Davis. Graphics were prepared by Susan Trammell. We are indebted to Dr. Margaret Mulvey and to Dr. Hsiu-Ping Liu of the Savannah River Ecology Laboratory for their independent critical re- view of the manuscript; they made valuable suggestions for improvement. This work was supported by N.I.H. grant TMP Al 11373 to Davis. The support of the Institute of Parasitic Diseases, Chinese Academy of Preventive Medicine is gratefully acknowledged. LITERATURE CITED ABBOTT, В. T., 1981, Handbook of medically im- portant mollusks of the Orient and the western Pacific. Bulletin of the Museum of Comparative Zoology, Harvard University, 100: 246-328. AYALA, Е. & С.Е. AQUADRO, 1982, A comparative summary of genetic distances in the vertebrates. Evolutionary Biology, 15: 151-158. AYALA, F., D. HEDGECOCK, G. ZUMWALT & J. VALENTINE, 1973, Genetic variation in Tridacna maxima, an ecological analog of some unsuc- cessful evolutionary lineages. Evolution, 27: 177-191. BARTSCH, P., 1925, Some new intermediate hosts of the Asiatic human blood fluke. Journal of the Washington Academy of Sciences, 15: 71-73. BARTSCH, P., 1936a, Molluscan intermediate hosts of the Asiatic blood fluke, Schistosoma japonicum, and species confused with them. Smithsonian Miscellaneous Collections, 95: 1-60, 8 pls. BARTSCH, P., 1936b, A new intermediate host of the Asiatic blood fluke, Schistosoma japonicum. Proceedings of the Biological Society of Wash- ington, 49: 139-142. BARTSCH, P., 1946, Schistosomophora in China, with descriptions of two new species and a note on their Philippine relative. Smithsonian Miscel- laneous Collections, 104: 1-7. CAVALLI-SFORZA, L. L. & A. W. EDWARDS, 1967, Phylogenetic analysis: models and estimation procedures. Evolution, 21: 550-570. DAVIS, С. M., 1971, Mass cultivation of Oncomel- ania (Prosobranchia: Hydrobiidae) for studies of Schistosoma japonicum. Bio-Medical Reports of the 406th Medical Laboratory, 19: 85-161. DAVIS, С. M., 1979, The origin and evolution of the gastropod family Pomatiopsidae, with emphasis on the Mekong River Triculinae. Monograph of the Academy of Natural Sciences of Philadel- phia, 20: 1-120. DAVIS, G. M., 1980, Snail hosts of Asian Schisto- soma infecting man: origin and coevolution. In: J. BRUCE ET AL., eds. The Mekong schistosome. Ma- lacological Review, Supplement, 2: 195-238. DAVIS, G. M., 1981, Different modes of evolution and adaptive radiation in the Pomatiopsidae (Prosobranchia: Mesogastropoda). Malacologia, 21: 209-262. DAVIS, G. M., 1983, Relative roles of molecular genetics, anatomy, morphometrics, and ecology in assessing relationships among North Ameri- can Unionidae (Bivalvia). In: с. $. OXFORD & D. ROLL- INSON, eds. Protein polymorphism: adaptive and taxonomic significance. Systematics Association Special Volume, 24: 193-221. Academic Press. DAVIS, G. M., 1992, Evolution of prosobranch snails transmitting Asian Schistosoma; coevolu- tion with Schistosoma: A review. Progress in Clinical Parasitology 3: 145-204. DAVIS, G. M., 1994, Molecular genetics and taxo- nomic discrimination. The Nautilus, Supplement, 2: 3-23. DAVIS, С. M., W. H. HEARD, S. L. H. FULLER, & С. HESTERMAN, 1981, Molecular genetics and speciation in Elliptio and its relationships to other taxa of North American Unionidae (Bivalvia). Bi- ological Journal of the Linnean Society, 15: 131- 150: DAVIS, С. M., С.Е. CHEN, Z. В. KANG 4 Y. У. LIU, 1994а, Snail hosts of Paragonimus т Asia and the Americas. Biomedical and Environmental Sciences, 7: 369-382. DAVIS, G. M., С. E. CHEN, X. P. ZENG, $. H. YU & L. Ll, 1994b, Molecular genetic and anatomical relationships among pomatiopsid (Gastropoda: Prosobranchia) genera from southern China. Proceedings of The Academy of Natural Sci- ences of Philadelphia, 145: 191-207. 156 DAVIS, ZHANG, GUO 8 SPOLSKY DAVIS, G. M., V. FORBES 4 G. LOPEZ, 1988, Spe- cies status of northeastern American Hydrobia (Gastropoda: Prosobranchia): ecology, morphol- ogy and molecular genetics. Proceedings of the Academy of Natural Sciences of Philadelphia, 140: 191-246. DAVIS, G. M., M. MCKEE 8 G. LOPEZ, 1989, The Identity of Hydrobia truncata (Gastropoda: Hydrobiinae): comparative anatomy, molecular genetics, ecology. Proceedings of the Academy of Natural Sciences of Philadelphia, 141: 333- 359. DAVIS, С. М. & М. RUFF, 1973, Oncomelania hu- pensis (Gastropoda: Hydrobiidae) hybridization, genetics, and transmission of Schistosoma japonicum. Malacological Review, 6: 181-197. DILLON, В. & G. М. DAVIS, 1980, The Goniobasis of southern Virginia and northern North Carolina: genetic and shell morphometric relationships. Malacologia, 20: 83-98. FELSENSTEIN, J., 1989, PHYLIP—Phylogeny In- ference Package (Version 3.2). Cladistics, 5: 164-166. GREDLER, P. V., 1881, Zur Concylien-Fauna von China. Ill. Stück. Jahrbücher der Deutschen Malakozoologischen Gesellschaft, 8: 110-132. HAASE, M., 1993, The genetic differentiation in three species of the genus Hydrobia and sys- tematic implications (Caenogastropoda, Hydro- biidae). Malacologia, 35: 389-398. HOPE, M. & D. P. MCMANUS, 1994, Genetic vari- ation in geographically isolated populations and subspecies of Oncomelania hupensis deter- mined by a PCR-based RFLP method. Acta Tropica, 57: 75-82. JOHNSON, М. S., O. С. STINE & J. MURRAY, 1984, Reproductive compatibility despite large- scale genetic divergence in Cepaea nemoralis. Heredity, 53: 655-665. KANG, Z. B., 1981, A review of the taxonomical history of Oncomelania. Academic Journal of Hubei Medical College, 2: 54-55. KANG, Z. B., 1985, Comments on some problems of taxonomy of Oncomelania. Japanese Journal of Parasitology, 34: 93-100. LIU, У. Y., T. К. LOU, Y. Х. WANG & W. 2. ZHANG, 1981, Subspecific differentiation of oncomela- niid snails. Acta Zootaxonomica Sinica, 6: 253- 266, 1 pl. LOU, Т. Z., Y. Y. LIU, W. 2. ZHANG 8 У. X. WANG, 1982, A discussion on the classification of On- comelania (Mollusca). Sinozoologia, 2: 97-114, 3 pls. MURRAY, J., O. C. STINE & M. S. JOHNSON, 1991, The evolution of mitochondrial DNA in Partula. Heredity, 66: 93-104. NEI, M., 1978, Estimation of average heterozygos- ity and genetic distance from a small number of individuals. Genetics, 89: 583-590. PATTERSON, H. E. H., 1985, The recognition con- cept of species. In: E. $. VRBA, ed., Species and speciation. Transvaal Museum Monograph, 4: 21-29. Pretoria, South Africa. RICE, W. R., 1989, Analyzing tables of statistical tests. Evolution, 43: 223-225. ROSENBERG, G., 1989, Phylogeny and evolution of terrestriality of the Atlantic Truncatellidae (Prosobranchia: Gastropoda: Mollusca). Unpubl. thesis; Harvard University. STINE, O. C., 1989, Cepaea nemoralis from Lex- ington, Virginia: the isolation and characteriza- tion of their mitochondrial DNA, the implications for their origin and climatic selection. Malacolo- gia, 30: 305-315. SWOFFORD, D. L. & В. SELANDER, 1981, BIO- SYS-1. A computer program for the analysis of allelic variation in genetics. Univ. of Illinois, Ur- bana. 60 pp. TEMPLETON, A. R., 1989, The meaning of species and speciation: a genetic perspective. In: D. OTTE & J. ENDLER, eds. Speciation and its conse- quences. pp. 3-27. Sinauer Associates, Inc., Sunderland, Mass. VAN DER SCHALIE, H. & G. M. DAVIS, 1968, Cul- turing Oncomelania snails (Prosobranchia: Hy- drobiidae) for studies of oriental schistosomia- sis. Malacologia, 6: 321-367. WOODRUFF, D., К. С. STAUB, Е. $. UPATHAM, V. VIYANT & Н. С. YUAN, 1988, Genetic variation in Oncomelania hupensis: Schistosoma japonicum transmitting snails in China and the Philippines are distinct species. Malacologia, 29: 347-361. Revised Ms. accepted 1 June 1995 MALACOLOGIA, 1995, 37(1): 157 FENTER TO THe Е НОВ RESPONSE TO BOUCHET 8 ВОСВОГ; “THE LOTTERY OF BIBLIOGRAPHICAL DATABASES: A REPLY TO EDWARDS 8 THORNE” М. A. Edwards! & М. J. Thorne? While not wishing to protract this exchange unnecessarily, we would like to offer a final, brief, response to the above. Which Names Get Omitted? The analysis of names omitted 1$ based оп material published mostly between 1940 and 1975, that is between 20 and 55 years ago. During most of this period the Record was compiled by volunteer specialists in the dif- ferent animal groups. The work was done in their spare time so that, while it is unfortu- nate, it is not surprising that material, whether from obscure or main stream publications, was overlooked. Bouchet 8 Rocroi say that **. . . in an era of frequent and easy travel ... the staff of ZR has attended only once ... an International Malacological Congress . . .” and, atten- dance at meetings would “. . . greatly en- hance the efficiency of ZR...”. There is a simple explanation for our ab- sence at meetings and that is although travel may be easy, it is expensive. We are occa- sionally able to attend local meetings (hence our presence at the 1986 congress in Edin- burgh), but the cost of attending the annual and other meetings of all the main zoological disciplines in Europe, USA, Russia, or any- where else, is quite beyond the capacity of our budget. As referred to in our earlier re- sponse, Zoological Record production is heavily subsidized, and increased travel would add considerably to the already large overheads borne by BIOSIS. It must also be said that while the meetings we have attended have certainly been useful, they have not led to the discovery of signifi- cant numbers of new titles for indexing. The Risks of a “List of Available Generic Names in Zoology’? Based on Nomenclator Zoologicus and ZR” The policy of the Zoological Record and the Nomenclator Zoologicus is to provide in- formation and not to adjudicate on the status of the names listed. The Nomenclator seeks to give details of the first use of a name and it is not practical to check back on previous volumes to determine the availability of names under the Code. A note of any rele- vant information on errors, omissions, and other matters is always welcomed by the ed- itor for future action. As mentioned above, The Zoological Record is subsidized by BIOSIS and the No- menclator Zoologicus is compiled on a vol- untary basis. Each is produced as a service to zoology and not as a financial proposition. К is our understanding that proposals on how a registry of names might be compiled or made available have yet to be finalized, therefore further discussion seems prema- ture. The editor-in-chief of Malacologia welcomes let- ters that comment on vital issues of general im- portance to the field of Malacology, or that com- ment on the content of the journal. Publication is dependent on discretion, space available and, in some cases, review. Address letters to: Letter to the Editor, Malacologia, care of the Department of Malacology, Academy of Natural Sciences, 19th and the Parkway, Philadelphia, PA 19103. The Zoological Society of London, Regent's Park, London, NW1 4RY, United Kingdom. 2BIOSIS, U.K., Garforth House, 54 Micklegate, York, North Yorkshire YO1 1LF, United Kingdom. 157 ЕАВАТА Ibanez, M., E. Ponte-Lira & М. В. Alonso. 1995. EL GENERO CANARIELLA HESSE, 1918, Y SU POSICION EN LA FAMILIA HYGROMIIDAE (GASTROPODA, PULMONATA, HELICOIDEA). MALACOLOGIA 36(1-2): 111-137. Figures 12-20 are reprinted here that include the scale bar under Figs. 19 and 20. The scale bar, and the edges of the shells at the bottoms of figures 15 and 16 were cut out in the original printing. 158 ы , ' ' La 3 FIGS. 12-20. Concha y SEM detalles. (12) Canariella discobolus (Barranco de la Rajita, La Gomera). (13) Canariella gomerae. Lectotipo de Helix (Gonostoma) дотегае (МНМ; es un ejemplar pequeño dentro de la especie). (14-15) Canariella hispidula. (14) Lectotipo de Helix (Gonostoma) hispidula subhispidula (ZMZ). (15) Lectotipo de Helix (Ciliella) lanosa (ZMZ). (16) Canariella leprosa (El Draguillo, Tenerife). (17-18) Cana- riella eutropis. (17) Lectotipo de Helix eutropis (NMB). (18) Mandíbula de un ejemplar de Morro del Cava- dero, Fuerteventura). (19-20) Rádula de Canariella planaria (Benijo, Tenerife). (19) Diente central y primeros dientes laterales. (20) Dientes laterales próximos al margen radular. Escala: (12-17) 5 mm; (18) 200 um; (19-20) 20 um. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Vol. Publication dates 28, No. 29, No. 29, No. 30, No. ©: No: 31, No. 32, No. 32, No. 33, No. 34, No. 35, No. 35, No. 36, No. 19 Jan. 1988 28 June 1988 16 Dec. 1988 1 Aug. 1989 29 Dec. 1989 28 May 1990 30 Nov. 1990 7 June 1991 6 Sep. 1991 9 Sep. 1992 14 July 1993 2 Dec. 1993 8 Jan. 1995 WHY NOT SUBSCRIBE ТО MALACOLOGIA? ORDER FORM Your name and address IIOIOK|S5za Send U.S. $36.00 for a personal subscription (one volume) or U.S. $55.00 for an institutional subscription. Make checks payable to “MALACOLOGIA.” Address: Malacologia Department of Malacology Academy of Natural Sciences 1900 Benjamin Franklin Parkway Philadelphia, PA 19103-1195, U.S.A. MALACOLOGIA, 1995, 37, МО. 1 37 МО. 1 INSTRUCTIONS FOR AUTHORS 1. MALACOLOGIA publishes original re- search on the Mollusca that is of high quality and of broad international interest. Papers combining synthesis with innovation are par- ticularly desired. While publishing symposia from time to time, MALACOLOGIA encour- ages submission of single manuscripts on diverse topics. Papers of local geographical or systematic interest should be submitted elsewhere, as should papers whose primary thrust is physiology or biochemistry. Nearly all branches of malacology are represented on the pages of MALACOLOGIA. 2. Manuscripts submitted for publication are received with the tacit understanding that they have not been submitted or published elsewhere in whole or in part. 3. Manuscripts may be in English, French, German or Spanish. Papers in lan- guages other than English must include a translation of the Abstract in English. Authors desiring to have their abstracts published in other languages must provide the transla- tions (complete with main titles). Include all foreign accents. Both American and British spellings are allowed. 4. Unless indicated otherwise below, contributors should follow the recommenda- tions in the Council of Biology Editors (CBE) Style Manual (ed. 5, 1983) available for U.S. $24.00 from CBE, 9650 Rockville Pike, Bethesda, MD 20814, U.S.A. 5. Be brief. 6. Manuscripts must be typed on one side of good quality white paper, double- spaced throughout (including the references, tables and figure captions), and with ample margins. Tables and figure captions should be typed on separate pages and put at the end of the manuscript. Make the hierarchy of headings within the text simple and consis- tent. Avoid internal page references (which have to be added in page proof). 7. Choose a running title (a shortened version of the main title) of fewer than 50 letters and spaces. MALACOLOGIA 1993 8. Provide a concise and informative Ab- stract summarizing not only contents but re- sults. A separate summary generally 1$ super- fluous. 9. Supply between five and eight key (topic) words to go at the end of the Abstract. 10. Use the metric system throughout. Mi- cron should be abbreviated um. 11. Illustrations are printed either in one column or the full width of a page of the journal, so plan accordingly. The maximum size of a printed figure is 13.5 x 20.0 cm (preferably not as tall as this so that the cap- tion does not have to be on the opposite page). 12. Drawings and lettering must be dark black on white, blue tracing, or blue-lined paper. Lines, stippling, letters and numbers should be thick enough to allow reduction by 2 or Уз. Letters and numbers should be at least 3mm high after reduction. Several drawings or photographs may be grouped together to fit a page. Photographs are to be high contrast. High contrast is especially im- portant for histological photographs. 13. All illustrations are to Бе numbered sequentially as figures (not grouped as plates or as lettered subseries), and are to be ar- ranged as closely as possible to the order in which they are first cited in the text. Each figure must be cited in the text. 14. Scale lines are required for all nondi- agrammatic figures, and should be conve- nient lengths (e.g., “200 um,” not “163 шт”). Magnifications in captions are not accept- able. 15. All illustrations should be mounted, numbered, labeled or lettered, ¡.e. ready for the printer. 16. A caption should summarize what 1$ shown in an illustration, and should not du- plicate information given in the text. Each let- tered abbreviation labeling an individual fea- ture in a figure must either be explained in each caption (listed alphabetically), or be grouped in one alphabetic sequence after the Methods section. Use the latter method if many abbreviations are repeated on different figures. Ss 4 i ве. 17. Tables are to be used sparingly, and Ш — vertica! lines not at all. I 18. References cited in the text must be in x Y AE the Literature Cited section and vice versa. | (ES ‚ Follow a recent issue of MALACOLOGIA for | bibliographic style, noting especially that se- rials are cited unabbreviated. Supply pagina- a \ tion for books. Supply information on plates, my etc., only. if they are not included in the J pagination. La % 19,-In © systematic papers, synonymies u à: A should’ not give complete citations’ but ‘ 1 as should relate by author, date and page to the | > Literature Cited section. Е, _, 20: For systematic papers, all new type- A specimens,must be deposited in museums я where they. may be studied by other scien- be Cu tists. Likewise MALACOLOGIA requires that NI voucher specimens upon, which a paper is M N based be deposited in a тизеит where they | may eventually be reidentified, /. 21. Submit each manuscript in triplicate. | The second And third SERIES can be repro- a _ductions. 22. Authors; who want illustrations re- turned should request this at the time of or- dering reprints. Otherwise, illustrations will be _ maintained for six nenne only alter, publica= tion. \ р - 1 REPRINTS AND PAGE COSTS | 23. When 100 or more reprints,are or- С _dered, ап author гесемез 25 additional cop- TE vik es free. Reprints must be ordered at the time APE . ‚ / proof is returned to the Editorial Office. Later orders cannot be considered. For each au- thors' change in page proof, the cost is U. >. ~~ $3.00 or more. ; 24. When an article is 10/or more printed pages long, MALACOLOGIA requests that an я -author pay рай of the publication.costs if { г grant ог institutional support is available. 7 А ~ SUBSCRIPTION PURCHASE 25. fective Jan. 1995, personal subserip- tions are U.S. $36.00. Institutional, agency or dealership subscriptions are U.S. $55.00. Back issues and single volumes: $36,00 for : individual purchaser; $55.00 for institutional : or agency purchasers. There is a one dollar | handling charge per volume for all purchases ‚ of single volumes. Address inquiries to the. Subscription Office. / | we 1 ne H y "Г т LAON PAS И Ne G ie A (a N ( > | (A “| E N | m iy уж } Y nee | i y ER AM { N ve | и VOL. 37, NO. 1 MALAGQLOGIA Y "Nm Мой ae м R | ГА. | ми и. CONTENTS СИ | a a x М. LAZARIDOU-DIMITRIADOU ; IS e The Life Cycle, Demographic Analysis, том and Secondary Production of the г Snail Helicella (Xerothracia) Papp! (Schútt, 1962) т ose Pulmonata) in E. Macedonia (Greece) ...... dde LATE CARO OA q ER: | HILARY PIGGOTT & GEORGES DUSSART И À y 5 _ Egg-Laying . and Associated Behavioural : Responses of meta Peregra EBEN (Müller) and Lymnaea Stagnalis (L.) to Calcium in their Environment coves Я Kr - LUIZ RICARDO LOPES DE SIMONE N > y Ke Ne = Anatomical Study оп Топпа Galea Mine, 1758) and Tonna Maculosa (Dillwyn, a | 181 2 (Mesogastropoda, Tonnoidea, Tonnidae) from Brazilian Region. ASS DEP i 3 N Hoe D. ACUÑA & М. A. MUÑOZ e t, \ | a "A Taxonomic Application of Multivariate Mixture Sie in Patellidae . igual Mua aM __ № ELEUTHERIADIS & М. LAZARIDOU-DIMITRIADOU = и ee was The Life Cycle, Population Dynamics, Growth and secondary Production of they ‘Snail Viviparus Contectus (Millet) (Gastropoda: Prosobranchia) i in the Marshes of | < LU ‘the River Strymonas, Serres, Macedonia, Northern Greece Ll N Ba), 4 |. \ 4 KATERINE COSTIL & JACQUES DAGUZAN | У м > ANAL a E Comparative Life Cycle and Growth of two rear ter Gastropod Species, N | Hal m x ори Planorbarius Corneus: (L.) and Planorbis Planorbis (LT A ee Les au _ 53 Гм PE KENNETH C. EMBERTON Bu NS X Res ( 1 ei 13 IR WANN When Shells Do Not Tell: 145 Million Years of Evolution i in North M wit UP ar у Ar УГ Polygyrid Land Snails, with a Revision and Conservation Priorities . + 9 Y y 3 AXHONEK к] 5 A > yd N Sy 1 | GebatabhieDishtbytion and ‘Shell Colour and Banding Polymorphism, in Mar 7 $ ! Hs ginal Populations of Cepaea Nemoralis (Gastropoda, Helicidae) RR 15 CR ane he R. VITTURI, A. LIBERTINI, М. PANOZZO & G. MEZZAPELLE Ni a | bs IA MC Karyotype Analysis and Genome Size in Three нала Species of у TES % | winkles (Prosobranchia: Mesogastropoda) ia ere PR RT ` ño A “GEORGE M. DAVIS, ZHANG YI, GUO YUAN HUA & CHRISTINA. SPOLSKY в. N ln \ 4) 7 Population Genetics and Systematic Status of Oncomelania Hupensis y an, N _ tropoda: HEURES) ее China AA Bd HE a Pine м. И t | EEE о | | AUTRE | LETTER TO THE EDITOR A ем А. EDWARDS & M. “Y. THORNES : I о у | Response to Bouchet & Rocroi; “The Lottery of Bibiographical Dép RENE | ig ne) ly to to Edwards a AO rn a LANCE NOM Ad RE EN, ’ Pa, rl NA Ain AT IS AN eaten A a Si у nh rey vay } Ms on { à IN yo, Y уз \ on 2 ig y у We - у ae у’ Wy у \ ) \ A \ RS LE Gite RUN" Wr) 3 4 De у | > h f J ; ) i‘ ÿ : | 1 ne : | у | N ln N | | IN Mu l | А EN d \ PL di, par К JA de т te 2 NE Au hr у к EN ig | уу к НИ Er ch MA LAS LE MENT Pat BR, PANA , PEER L Boh: UA MRAP CANS A ae | ‘ te ty dd ES Bay Fees ar vy JN №. м и р | VAE Да NOL. 37, NO. 2 | A Е 1996 1 996 a М АК 1 D we i | | pl ARAN А RD À ie с aur JERSN X N 'MALACOLOGIA a ( Т \ № № Je À 3 y 73 $ » . - 4 ) L Е ? Journal International de Malacologie he я fi { y | | E ( E | + Une р oF у N р A a \ ‘ A fi \ < 7 Международный Журнал Малакологии ] 5 ‘ ] a \ AE) ) JA CAL PI DECO A | 4 A | ú SA A tr к Ú (4 = x Ex | À 4 Le \ Internationale Malakologische Zeitschrift _ EE AN hy re | М f ; { 1) LA A " i | } я iy A у 4 % { | | he ( MALACOLOGIA ми ( \c a Editor-in-Chief: Weg 3 GEORGE M. DAVIS DEP N Editorial and Subscnption Offices: — ES E 7 Department of Malacology a ARA a) The Academy of Natural Sciences of Philadelphia DAT / ' 1900 Benjamin Franklin Parkway An. Aa 8 Philadelphia, Pennsylvania 19103-1195, U.S.A. - — Co-Editors: | | N “< х e ру EUGENE СОАМ и | CAROL JONES | California Academy of Sciences .. © | Denver, СО _ yy a San Francisco, CA A Dips a CNRC > #4 4 И Assistant Managing Editor: ie | oh eee Eu т CARYL НЕЗТЕВМАМ N 1 D Associate Editors: 3.) N Е И | N ANNE GISMANN À de University of Michigan ( Masti 1 AS € wos E , À 8 y Na | Ann Arbor A Egypt Eu " в. À HA "( п à MALACOLOGIA is published Er the INSTITUTE OF MALACOLOGY, the Sponsor Members of Fe which (also serving as o are: > у Be RE OO RUDIGER BIELER ‚ МАМ KOHN) ие ae Field Museum, Chicago O: pra University of Washington, Seattle » и) 2 JOHN BURCH JAMES NYBAKKEN | Pag a У 7 MELBOURNE R. CARRIKER, Moss Landing Marine Laboratory | A QE A ¡NA ` President Elect y | ¡California AS FRE Е, of Delaware, Lewes CLYDE F. E. ROPER \\ Là à E GEORGE M. DAVIS Smithsonian Institution = в. = Secretary and Treasurer Washington, D.C. = r ON CAROLE $. HICKMAN, President | SHI-KUEI WU | Ч. e ‚University of California, Berkeley ni University of SOS Museum, Bould AR ¡IA | А # NER. y Ви LS Fer \ а Members an N р: zZ { 230 А: © EDMUND GITTENBERGER JACKIE L. VAN GOETHEM il tale Secretary, UNITAS MALACOLOGICA | y Treasurer, UNITAS MALACOLOGICA : JE Rijksmuseum van Natuurlijke Ad Koninklijk Belgisch Instituut _ Historie | | Ия lige voor Natuurwetenschappen _ Leiden, Nettfenands A | Brussel, Belgium / IA | A UE И Va ел. AAA: a PEUR NES Emeritus Members po! or y AOS р od FRANCIS ALLEN, Emerita | ROBERT ROBERTSON - ‘pi “Environmental Protection Agency Y | The Academy of Natural Scenes | J , | . Washington, D.C. a PS ) | dE KENNETH J. BOSS ) EN | Le Ko of Comparative Zoology _ > W. D. RUSSELL-HUNTER | M m Bei NS ambridge, Massachusetts KU | Easton, Maryland _ Of 0 AN 2% REN, re Ser | ae ) A N | Copyright © 1996 by the Institute of Malacology. SA TAN je ) Rh: > д i Ur à hol À 1 (9 dus | | | À a 5 |) ARE ds L | , | A ие wg aX у | \ 4 $ VEN 4 м J. А. ALLEN Marine Biological Station Millport, United Kingdom E. E. BINDER Muséum d'Histoire Naturelle Genève, Switzerland A. J. CAIN University of Liverpool United Kingdom P. CALOW University of Sheffield United Kingdom J. G. CARTER University of North Carolina Chapel Hill, U.S.A. R. COWIE Bishop Museum Honolulu, HI., U.S.A. А. Н. CLARKE, Jr. Portland, Texas, U.S.A. В. С. CLARKE University of Nottingham United Kingdom R. DILLON College of Charleston SC, U.S.A. С. J. DUNCAN University of Liverpool United Kingdom D. J. EERNISSE California State University Fullerton, U.S.A. E. GITTENBERGER Rijksmuseum van Natuurlijke Historie Leiden, Netherlands F. GIUSTI Universita di Siena, Italy A. N. GOLIKOV Zoological Institute St. Petersburg, Russia 1996 EDITORIAL BOARD $. J. GOULD Harvard University Cambridge, Mass., U.S.A. A. V. GROSSU Universitatea Bucuresti Romania T. HABE Tokai University Shimizu, Japan R. HANLON Marine Biomedical Institute Galveston, Texas, U.S.A. J. А. HENDRICKSON, уг. Academy of Natural Sciences Philadelphia, PA, U.S.A. D. M. HILLIS University of Texas Austin, U.S.A. K. E. HOAGLAND Association of Systematics Collections Washington, DC, U.S.A. B. HUBENDICK Naturhistoriska Museet Goteborg, Sweden S. HUNT Lancashire United Kingdom R. JANSSEN Forschungsinstitut Senckenberg, Frankfurt am Main, Germany R. N. KILBURN Natal Museum Pietermaritzburg, South Africa M. A. KLAPPENBACH Museo Nacional de Historia Natural Montevideo, Uruguay J. KNUDSEN Zoologisk Institut & Museum Kobenhavn, Denmark A. LUCAS Faculté des Sciences Brest, France C. MEIER-BROOK Tropenmedizinisches Institut Túbingen, Germany Н. К. MIENIS Hebrew University of Jerusalem Israel J. E. MORTON The University Auckland, New Zealand J. J. MURRAY, Jr. University of Virginia Charlottesville, U.S.A. R. NATARAJAN Marine Biological Station Porto Novo, India J. VKLAND University of Oslo Norway T. OKUTANI University of Fisheries Tokyo, Japan W. L. PARAENSE Instituto Oswaldo Cruz, Rio de Janeiro Brazil J. J. PARODIZ Carnegie Museum Pittsburgh, U.S.A. J. P. POINTER Ecole Pratique des Hautes Etudes Perpignan Cedex, France W. Е. PONDER Australian Museum Sydney ©] ZW Academia Sinica Qingdao, People’s Republic of China D. G. REID The Natural History Museum London, United Kingdom N. W. RUNHAM University College of North Wales Bangor, United Kingdom S. G. SEGERSTRLE Institute of Marine Research Helsinki, Finland A. STANCZYKOWSKA Siedlce, Poland F. STARMUHLNER Zoologisches Institut der Universitat Wien, Austria Y. |. STAROBOGATOV Zoological Institute St. Petersburg, Russia W. STREIFF Université de Caen France J. STUARDO Universidad de Chile Valparaiso S. TILLIER Muséum National d’Histoire Naturelle Paris, France R. D. TURNER Harvard University Cambridge, Mass., U.S.A. J.A.M. VAN DEN BIGGELAAR University of Utrecht The Netherlands J. A. VAN EEDEN Potchefstroom University South Africa N. H. VERDONK Rijksuniversiteit Utrecht, Netherlands B. R. WILSON Dept. Conservation and Land Management Kallaroo, Western Australia H. ZEISSLER Leipzig, Germany A. ZILCH Forschungsinstitut Senckenberg Frankfurt am Main, Germany MALACOLOGIA, 1996, 37(2): 163-332 ANATOMY AND SYSTEMATICS OF THE WESTERN ATLANTIC ELLOBIIDAE (GASTROPODA: PULMONATA) António M. de Frias Martins Departamento de Biologia, Universidade dos Açores, P-9502 Ponta Delgada Codex, Sao Miguel, Açores, Portugal ABSTRACT Various conchological, radular and anatomical characters of the 18 Western Atlantic species of the pulmonate family Ellobiidae are evaluated and used in a systematic review of the family. The conchological features, especially protoconch, resorption of inner whorls, apertural den- tition and radular morphology, are useful at the specific and generic levels. Features of the radula of the Melampinae change greatly with increasing age. The youngest individuals have strongly cusped crowns. Although the cusps usually disappear with age, some species retain various features of the juvenile radula. The reproductive and central nervous systems are most useful in defining subfamilial relationships. The monaulic, entirely glandular condition of the pallial gonoducts, and the greater width of the visceral nerve ring are hereby considered primitive. Morton’s (1955c) subfamilial division of the halophilic Ellobiidae is corroborated. The Pythiinae have a monaulic, entirely glandular pallial gonoduct and a wide visceral nerve ring. The Ellobiinae have a diaulic reproductive system with entirely glandular pallial gonoducts, and a long visceral nerve ring. The Pedipedinae have a monaulic/incipient semidiaulic, partly glan- dular pallial gonoduct, and a concentrated visceral nerve ring. The Melampinae are character- ized by an advanced semidiaulic reproductive system with nonglandular pallial gonoducts, and concentration of the ganglia of the visceral nerve ring. The present work documents that Microtralia Dall, 1894, belongs in the Pedipedinae, not in the Melampinae; that Laemodonta Philippi, 1846, belongs in the Pythiinae, not in the Pedipe- dinae; that Leuconia succinea Pfeiffer, 1854, belongs in the Pedipedinae and in the new genus Creedonia; that Apodosis Pilsbry £ McGinty, 1949, is synonymous with Leuconopsis Hutton, 1884; that Myosotella Monterosato, 1906, type species Myosotella payraudeaui “Shuttleworth” Pfeiffer, 1856 [= Auricula myosotis Draparnaud, 1801], is removed from Ovatella Bivona, 1832, and restored to generic rank; that Detracia Gray, 1840, as noted by Zilch (1959), is a subgenus of Melampus Montfort, 1810; that Melampus monile (Bruguière, 1789) belongs in the subgenus Detracia Gray, 1840; and that Detracia clarki Morrison, 1951, is a junior secondary homonym and is herein renamed Melampus (Detracia) morrisoni. Leuconopsis manningi new species, from Ascension Island, is described. The phylogenetic relationships within the Ellobiidae are discussed, a tentative cladogram of the family is presented, some distributional patterns are considered and reference is made to the fossil record. Key words: Archaeopulmonata, Ellobiidae, systematics, shell, radula, anatomy, genitalia, nervous system, Western Atlantic, mangroves, salt marshes. TABLE OF CONTENTS Introduction Materials and Methods Abbreviations Used in Figures Taxonomic Characters Classification Outline, Western Atlantic Ellobiidae Systematics Family Ellobiidae H. & A. Adams in Pfeiffer, 1854 Subfamily Ellobiinae H. & A. Adams in Pfeiffer, 1854 Genus Ellobium Róding, 1798 Subgenus Auriculodes Strand, 1928 163 Ellobium (Auriculodes) dominicense (Férussac, 1821) Genus Blauneria Shuttleworth, 1854 Blauneria heteroclita (Montagu, 1808) Subfamily Pythiinae Odhner, 1925 Genus Myosotella Monterosato, 1906 Myosotella myosotis (Draparnaud, 1801) Genus Laemodonta Philippi, 1846 Laemodonta cubensis (Pfeiffer, 1854) Subfamily Pedipedinae Fischer & Crosse, 1880 Genus Pedipes Scopoli, 1777 Pedipes mirabilis (Mühlfeld, 1816) 164 MARTINS Pedipes ovalis C. B. Adams, 1849 Genus Сгеедота new genus Creedonia succinea (Pfeiffer, 1854) Genus Microtralia Dall, 1894 Microtralia occidentalis (Pfeiffer, 1854) Genus Leuconopsis Hutton, 1884 Leuconopsis novimundi (Pilsbry & McGinty, 1949) Leuconopsis manningi new species Leuconopsis sp. Subfamily Melampinae Pfeiffer, 1853 Genus Melampus Montfort, 1810 Subgenus Melampus s.s. Melampus (Меатриз) coffeus (Linnaeus, 1758) Melampus (Melampus) bidentatus Say, 1822 Subgenus Detracia Gray, 1840 Melampus (Detracia) bullaoides (Montagu, 1808) Melampus (Detracia) floridanus Pfeiffer, 1856 Melampus (Detracia) paranus (Morrison, 1951) Melampus (Detracia) monile (Bruguiere, 1789) Melampus (Detracia) morrisoni new name Genus Tralia Gray, 1840 Subgenus Тга/а $.5. Tralia (Tralia) ovula (Bruguiere, 1789) Conclusions Phylogeny and Classification Zoogeography of the Ellobiidae Acknowledgments Literature Cited Appendix INTRODUCTION The Ellobiidae are primitive pulmonate gastropods that characterize the malaco- fauna of the upper and supra-littoral zones of the mangroves of the tropical regions and salt marshes of temperate regions. The Ello- biidae were first assigned familial rank by Lamarck (1809) when he included his Auric- ula [= Ellobium Róding], along with three other unrelated genera, within the “auricula- cées”. Since then, several comprehensive works have been published. The group was illustrated in Reeve's Conchologia Systemat- ica (1842) and Conchologia Iconica (1877). A pictorial presentation was given in Martini 8 Chemnitz’ Conchylien-Cabinet by Küster (1844) and Kobelt (1897-1901). Pfeiffer (1854b) outlined a monograph of the Auricu- lacea [= Ellobiidae] in his Synopsis and fully developed the work in his Monographia in 1856, which he revised and completed twenty years later. Odhner (1925) rearranged the classification of the family on the basis of radular morphology; Morton (1955c) in- cluded morphology of the stomach and re- productive organs in his review of the group. Only a few genera have received compre- hensive treatment. The genus Plecotrema [= Laemodonta Philippi] was first revised by H. 8 A. Adams (1853) and was studied by Sykes (1895) and, more recently, by Huben- dick (1956). The genera Ellobium Róding and Melampus Montfort were studied by H. & A. Adams (1854). Cox (1882) worked on the no- menclature and distributuion of Pythia Ród- ing, and Connolly (1915) did a similar study on the genus Marinula King. Noteworthy are the detailed anatomical and histological studies on Melampus boholensis H. & A. Ad- ams (Koslowsky, 1933), Myosotella myosotis (Draparnaud) (Meyer, 1955; Morton, 1955b) and Auriculinella (L.) bidentata (Montagu) (Morton, 1955b). Marcus & Marcus (1965a, b) discussed the anatomy of Melampus (M. coffeus (Linnaeus), Melampus (D.) paranus (Morrison), Ellobium (A.) dominicense (Férus- sac) and Blauneria heteroclita (Montagu). Giusti (1973) discussed the radula and anat- omy of Ovatella firminii (Payraudeau), and the shell, radula and anatomy of Myosotella my- osotis (Draparnaud) were dealt with by Giusti (1973, 1976) and Cesari (1973, 1976). The Western Atlantic ellobiids were т- cluded in the very earliest conchological re- ports of American scientists. Say (1822), the first New World malacologist, described the common Melampus (M.) bidentatus. Gould (1841) illustrated Say's species and Myoso- tella myosotis (Draparnaud), which is thought to have been introduced to North America from Europe. Study of American ellobiids was particularly influenced by Binney (1859, 1865) and Dall (1885). Binney (1859) figured most of the common species; his later figures (1865) were copied by subsequent workers (Tryon, 1866; Dall, 1885, 1889; M. Smith, 1937; Abbott, 1974), sometimes without crit- ical investigation. For example, Binney's in- accurate representation of Melampus (D.) flo- ridanus Pfeiffer in fact represents a dwarf Melampus (M.) bidentatus Say. Morrison (1946, 1951a, 1951b, 1954, 1958, 1959, 1964) addressed several aspects of American ellobiid systematics, life history and ecology, and Clench (1964) revised the Western Atlantic Pedipes and Laemodonta. The only detailed comprehensive anatom- ical research on Western Atlantic ellobiids WESTERN ATLANTIC ELLOBIIDAE 165 was that of Marcus 8 Marcus (1965a, b) on the four species mentioned above. Several aspects of the life history and anatomy of Melampus (M.) bidentatus Say have been investigated, almost exclusively in the New England area. Hausman (1932), Holle 8 Dineen (1957) and Grandy (1972) focused on various aspects of the ecology of this spe- cies, while Apley (1970) and Russell-Hunter et al. (1972) did extensive research on its early life history. Additional investigations have involved the morphology of the nervous system (Price, 1977; Kahan 8 Moffett, 1979), several aspects of physiology and behaviour (Price, 1979, 1980; Hilbish, 1981; Capaldo, 1983), locomotion (Moffett, 1979) and feed- ing (Thompson, 1984). In the present work particular attention has been paid to shell morphology, the radula and internal anatomy, especially the repro- ductive and nervous systems. This holistic approach helps to clarify the systematic po- sition and phylogenetic relationships of the Western Atlantic ellobiids. MATERIALS AND METHODS Materials Thousands of specimens from many local- ities were studied to understand inter- and intrapopulational variation in shell morphol- ogy. To accomplish this | studied the collec- tions at the Museum of Comparative Zool- ogy, Harvard University, Cambridge, at the American Museum of Natural History, New York, at the Academy of Natural Sciences of Philadelphia and at the United States Mu- seum of Natural History, Washington, D. C. Because museum collections were very poor in material with preserved soft tissue, the ma- jority of the internal anatomical work was done on specimens from my collections. Most of the Western Atlantic material was obtained during field trips along the Atlantic coast of the United States, to Bermuda, the Bahamas, Puerto Rico and Venezuela. Some specimens from R. C. Bullock's collection were also kindly made available to me. Field trips were very important in providing large series of most recorded species and in allow- ing examination of living animals in their hab- itats. Most of this material is now in my col- lection. Material not from the Western Atlantic, be- sides that in the museum collections men- tioned above, included Azorean ellobiids from my own collection, specimens from Ma- laysia sent by A. Sasekumar and another se- ries of specimens from Hong Kong sent by B. S. Morton; all are now part of my collection. The British Museum (Natural History) has kindly allowed me to work on preserved specimens of Marinula. Most of the Western Atlantic species of el- lobiids were first studied and described by European scientists and much of the type material is thought to be in European muse- ums. Only the type material studied in brief visits to the British Museum (Natural History) and to the Museum National d’Histoire Na- turelle de Paris, as well as that kindly sup- plied by the Muséum d'Histoire Naturelle de Genève, were incorporated in this work. Sherborn (1940) and Dance (1966) have been used to locate tentatively the collections that might contain required type material. Throughout the text, the museums and collections in which the studied material 1$ deposited are indicated by the following ab- breviations: AMNH American Museum of Natural His- tory, New York, NY, U.S.A. ANSP Academy of Natural Sciences of Philadelphia, PA, U.S.A. BMNH The Natural History Museum [for- merly British Museum (Natural His- tory)], London, U.K. R.B. Private collection of R. C. Bullock, University of Rhode Island, King- ston, RI, U.S.A. FMNH Field Museum of Natural History, Chicago, IL, U.S.A. LSL Linnaean Society of London, U.K. A.M. Private collection of A. M. F. Mar- tins, University ofthe Azores, Ponta Delgada, Sáo Miguel, Azores, POR- TUGAL. MCZ Museum of Comparative Zoology, Harvard University, Cambridge, MA, U.S.A. MHNG Museum d'Histoire Naturelle de Geneve, SWITZERLAND MNHNP Museum National d’Histoire Na- turelle de Paris, FRANCE NHMB Natural History Museum of Basel, SWITZERLAND RAMM Royal Albert Memorial Museum, London, U.K. USNM National Museum of Natural History [formerly United States National Museum], Washington, DC, U.S.A. 166 MARTINS Methods Observation and Collection of Live Animals. Observations of external morphology were made in the field and in the laboratory. The animals were photographed with Koda- chrome film. Notes on the habitat were taken during collecting. Besides extensive search and collecting in a variety of habitats, six transects were made in January, 1981, in the mangroves of the Florida Keys. Duplicate transects were made in May, 1982, at two of the 1981 sites, one on the previously disturbed site, another adja- cent to it. All ellobiids found in the transects were collected and preserved. Qualitative analysis of this data is included in notes on habitats of the different species. Preservation. Most animals were immersed directly in 70% ethanol. Some were relaxed overnight in isotonic MgCl, (75.2 g Мас of distilled water) and then preserved in 70% ethanol. Some of the contracted and relaxed animals were fixed in Bouin's solution after the shell was cracked to allow better pene- tration of the fixative; others were frozen in fresh water for later dissection. This latter method seemed quite useful, because the or- gans maintained their original colors and softness, allowing easier dissection many months later. Measurements. Various numbers of speci- mens from different localities were selected (Table 1, Appendix). Shells and dissected re- productive systems were drawn using a Wild M8 microscope with drawing tube. All mea- surements were taken from these drawings using a GTCO digitizer and IBM microcom- puter. Radular teeth were counted from SEM photographs. Shell and Radula Preparations for Scanning Electron Microscope. Juveniles of most spe- cies and adults of the smaller species were mounted for SEM observation of the entire shell or the protoconch, or both. The shells were cleaned in 95% ethanol in an ultrasonic cleaner for two to ten seconds, depending on the fragility of the specimen, and then were mounted on a stub with double-sided tape. The radulae were first cleaned in KOH (two pellets/10 ml distilled water), washed in dis- tilled water and in 70% ethanol. Ultrasonic cleaning was reduced to two seconds for each step. The radula was mounted on a piece of cover slip, to which it adhered when dry, and the cover slip was affixed to the stub with double-sided tape. The use of 70% eth- anol alone had the advantage of slower evap- oration, which was preferable when small pieces had to be manipulated at the exact moment they dried, to ensure proper posi- tioning and good adhesion. | found very help- ful the use of human eyelashes attached to dissecting needles with “superglue.” They are fine, flexible, but sufficiently rigid for clearing the membranes of the radula without tearing, and for facilitating positioning while mounting. All specimens were first coated with a sin- gle layer of carbon and then two layers of gold-palladium (60:40) in a Denton DV-502 vacuum evaporator, and examined in an ISI MSM-3 SEM. Histology. Serial sections were made of specimens of every species collected. Some specimens were relaxed overnight in isotonic MgCl, before fixation in Bouin’s solution. For most of the specimens the shell was cracked and the pieces removed to allow better fixa- tion. Whole animals, dissected reproductive or- gans and stomach were embedded sepa- rately. Embedding was done with an Auto- technicon Duo, Model 2A. The specimens were dehydrated in S-29 and embedded in Paraplast. The blocks were refrigerated until sectioned. Sectioning was done with a Spen- cer 820 microtome. The thickness of the sec- tions varied from 8 to 15 um. Best results were achieved by keeping the block and the blade refrigerated during sectioning. The preparations were stained with Heidenhain's aniline blue, following Luna (1968). ABBREVIATIONS USED IN FIGURES aa anterior aorta acpn anterior cutaneous pedal nerve ad anterior diverticulum adgl anterior lobe of digestive gland ад! albumen gland al aperture length aln anterior labial nerve alpn anterolateral pedal nerve ато! anterior mucous gland ampn anteromedial pedal nerve an aortic nerve angl anal gland ann anal nerve aoen anterior esophageal nerve au auricle WESTERN ATLANTIC ELLOBIIDAE 167 anterior vas deferens aperture width buccal bulb buccal commissure buccal ganglion bursa bursa duct body whorl body whorl length central tooth gastric caecum cardiac region of stomach cerebrobuccal connective cerebral commissure cerebral ganglion cutaneous-lateral pleural nerve columellar muscle columellar muscle nerve cerebropedal connective cerebropleural connective crop columellar tooth columellar tooth width penial diverticulum eye external pallial nerve elbow of vagina foot female genital opening fertilization pouch genital nerve hermaphroditic duct hypobranchial gland intestine inner lip internal pallial nerve kidney kidney pore lateral teeth lower pneumostomal gland marginal teeth mantle skirt mantle skirt artery nerve muscular band mucous gland mantle lappet medial lip nerve mantle organ medial pallial nerve muscular strand of stomach mantle skirt vein nuchal nerve esophagus osphradial ganglion outer lip ocular nerve open spermatic groove ovotestis propodium posterior artery pedal commissure posterior cutaneous pedal nerve posterior cutaneous visceral nerve posterior diverticulum posterior lobe of digestive gland penis pericardium penial nerve penial retractor muscle pedal ganglion pallial gland pharyngeal retractor muscle nerve pharyngeal nerve pleural ganglion posterior lip nerve pleuropedal connective posterior lateral pedal nerve pleuroparietal connective posterior mucous gland posteromedial pedal nerve pneumostome pneumostomal nerve posterior esophageal nerve first posterior pedal nerve second posterior pedal nerve prostate gland parietocutaneous nerve parietal ganglion anterior left parietal ganglion posterior left parietal ganglion parietovisceral connective anterior parietal tooth posterior parietal tooth peritentacular nerve width of posterior parietal tooth pulmonary vein posterior vas deferens pyloric region of stomach rectum riblets roof of mantle cavity stomach salivary gland salivary gland nerve shoulder of body whorl shell length subpedal commissure spermoviduct spire spire length statocyst statocyst nerve seminal vesicle shell width 168 MARTINS 18 transitional teeth t tentacle tem tentacular control muscle tn tentacular nerve upe unwrapped penis ира! upper pneumostomal gland V vagina ve ventricle vg visceral ganglion wpe wrapped penis TAXONOMIC CHARACTERS Мауг (1969: 121) stated, “А taxonomic character is any attribute of a member of a taxon by which it differs or may differ from a member of a different taxon.” Application of this definition cannot be uniform and gener- alized. Although there is consensus that a ho- listic approach is essential to sound classifi- cation (Mayr, 1969; Solem, 1978), one must be aware of the difference between charac- ters used at the species level or even for ge- neric grouping and those used for higher taxa. Characters that stress differences are used to define lower taxa, whereas those characters sensitive to convergence and seemingly less affected by environmental factors are used to define phylogenetic relationships among higher taxa. For example, the pattern of the spiral grooves on the shell is useful in sepa- rating Melampus (M.) coffeus from Melampus (M.) bidentatus, whereas the arrangements of the nervous and reproductive systems are the most consistent characters in defining the subfamilies of the Ellobiidae. In the Ellobiidae the shell and radular morphology are useful mostly at the generic or specific level. Harry (1951) and Hubendick (1978) pointed out the value of anatomical studies for clarifying tax- onomic relationships within the group. Morton (19550), followed by Marcus (1965) and Mar- cus & Marcus (1965a, b), adopted this holistic approach by including analyses of the stom- ach and reproductive organs; this approach led to somewhat surprising results, such as inclusion of Auriculinella and Blauneria within the Ellobiinae. In the present study conchological, radular and anatomical characters are used. Each of these different characters will now be con- sidered in more detail. FIG. 1. Conchological characters. A, Shell termi- nology; B, Morphometry. Conchological Characters (Fig. 1) The shell, more than other molluscan structures, has the obvious advantages of permanence and ease of study. Traditionally it has been the important basis for distinc- tion of most taxa (Zilch, 1959). Application of mathematical models and statistical analyses has provided tools for the interpretation of shell morphometry with accuracy and preci- sion (Sokal & Sneath, 1963). Mathematical analysis of geometry of shell coiling has also been used (Raup, 1961, 1962, 1966, 1967; Raup 4 Michelson 1965; Rex & Boss, 1976; Warburton, 1979; Harasewych, 1981). This method aims at providing an opportunity for interpretation of evolutionary changes in shell morphology in functional terms and as an in- dication of strategies of adaptation to differ- ent habitats (Vermeij, 1971). Such interpreta- tion has been challenged recently (Gould, 1984). In spite of modern refinements in anal- yses of conchological characters, it remains true that some, such as shell shape, have limited weight in assessing phylogenetic re- lationships because shell morphology is of- ten strongly influenced by diverse environ- mental parameters (Hubendick, 1978; Solem, 1978). Other shell characters, such as the proto- conch, resorption of the inner whorls and ap- ertural dentition, demonstrate more nearly constant patterns and are useful at the spe- cific and even generic level. The gastropod protoconch indicates the type of larval development that organisms in the different groups have undergone (Dall, 1924; Lutz et al., 1984). It might also have other very distinctive features that make it a WESTERN ATLANTIC ELLOBIIDAE 169 useful taxonomic character (Walter, 1962; Bouchet 8 Warén, 1980; Turner & Lutz, 1984). Study of the protoconch has been aided greatly by the use of the scanning elec- tron microscope, an increasingly important tool in malacology (Solem, 1970; Calloway 8 Turner, 1978). The SEM has been used to examine small and juvenile specimens, and to study the external morphology of the shell, radula and larvae. In this study the SEM was used to examine the protoconchs and radu- lae of most of the Western Atlantic ellobiids, and to provide photomicrographs of small and juvenile specimens. Ellobiid protoconch morphology proved a very useful taxonomic character in most cases. The Melampinae, for example, have only one type of heterostrophic protoconch, which shows one-half of each nuclear whorl. This feature might reflect the fact that, as far as is known, all have free-swimming larvae. Indeed, a similar type of protoconch occurs in Pyramidellidae having larvae with a long pelagic phase (Haszprunar, 1985). The mor- phology of the protoconch in the other sub- families of the Ellobiidae does not show an exclusive subfamilial pattern. For example, all the Pythiinae, Ellobiinae and the pedipedin- ian genera Pedipes and Creedonia have a bulbous protoconch with an umbilicus and a laterally facing aperture. The protoconch of the pedipedinian genera Microtralia and Pseudomelampus sits atop the teleoconch with the aperture facing the columellar axis, as in the Melampinae, rather than laterally as in all pythiinians and ellobiinians. Particular anatomical features indicate that these two genera belong to the Pedipedinae, however. The protoconch is very uniform within a spe- cies and, in the case of the West Indian Pe- dipes, it was the only consistent diagnostic conchological character that allowed clear separation of species. Shell resorption, as seen in the Ellobiidae, also occurs in the Neritacea, Helicinidae and Conidae. It was first noted by Montagu (1803: 235) in his Voluta denticulata [= Myosotella myosotis (Draparnaud)] and was reported for most members of the Ellobiidae by Gray (1840: 220-221). Crosse & Fischer (1879, 1882), however, studied the phenomenon in more detail and are usually credited with its discovery. Resorption of the inner whorls pro- vides a larger cavity in which the organs of the visceral mass can be rearranged with spatial economy. For example, in Melampus and Mi- crotralia, which show a high degree of resorp- tion, the conspicuous ovotestis has moved to an apical position and displaced the posterior lobe of the digestive gland; in Pedipes and Creedonia, which do not гезо the inner whorls, the ovotestis lies embedded in the apical, conspicuous posterior lobe of the di- gestive gland. This character varies within the different subfamilies of the Ellobiidae, but can be useful at lower taxonomic levels. In Melam- pus s.s., for example, the partition о the inner whorls occupies only one-fourth of the body whorl (Figs. 225, 267), whereas in the subge- nus Detracia it occupies at least three-fourths of the body whorl (Figs. 302, 316). The ex- treme case of variation within one subfamily occurs in the Pedipedinae, in which Pedipes and Creedonia completely retain the inner whorls (Figs. 106, 128, 153), whereas in Mi- crotralia resorption reaches the most ad- vanced stage in the Ellobiidae with less than a quarter of the interior partitions left (Fig. 178). In conjunction with other features, the lack of shell resorption justified the creation of the new genus Creedonia. The degree of re- sorption also supported the separation of De- tracia as a Subgenus of Melampus s.l. and was helpful in the interpretation of some anatom- ical differences observed between that sub- genus and Melampus s.s. Apertural dentition, an important character in gastropod classification, is а conspicuous feature of the Ellobiidae. One of the most commonly accepted functions of the aper- tural dentition is that of constituting a barrier against predators. From my observations on the disposition of the various branches of the columellar muscles along the conspicuous internal lamellae of Melampus (D.) bullaoides (Fig. 302), | think that this feature also helps in positioning the shell during locomotion. Although variable, there are some general patterns of apertural dentition. On this basis one can characterize broadly the different subfamilies as follows: Ellobiinae with bipli- cate inner lip, with columellar and parietal teeth very close together; Pythiinae with evenly spaced triplicate inner lip with first pa- rietal tooth strongest; Pedipedinae with two columellar teeth and strong parietal tooth; Melampinae with inner lip dentition restricted to anterior half, columellar and posterior pa- rietal teeth conspicuous, outer lip dentate. There are exceptions to these patterns, how- ever, forthe species in the melampinian genus Tralia have an inner lip structure very similar to that of the pythiinian genus Myosotella. The inner lip dentition of Microtralia deviates from 170 MARTINS FIG. 2. Terminology for radular teeth of Melampus (M.) coffeus. A, Top view of central tooth; B, Top view of first lateral tooth; C, Lateral view of first lateral tooth; D, Top view of tenth marginal tooth; E, Top view of 20th marginal tooth. the pattern of the Pedipedinae in having only one columellar tooth. The apertural structure of the pythiinian genera Cylindrotis and Au- riculastra resembles that of the Ellobiinae and Ellobium (E.) aurisjudae has a conspicuous posterior parietal tooth. Radular characters The molluscan radula is a valuable charac- ter in the classification of higher taxa and is the basis of phylogenies proposed for the Gastropoda (Gray, 1853; Troschel, 1856- 1893; Mörch, 1867). Some authors have stud- ied ellobiid radulae in an attempt to divide the Ellobiidae into subfamilies. Classifications of the ellobiids by Odhner (1925) and Thiele (1931) were based mainly on radular morphol- ogy, but these authors differed in their sub- familial division. Odhner adopted six subfam- Шез whereas Thiele recognized only three. Observation of the radula with a light micro- scope provides only limited information on the intricate articulation of the different teeth with one another (Figs. 250, 395). The SEM opened new vistas in the study of radular morphology and function (Solem, 1972b, 1974). The terminology used in this study follows that of Fretter & Graham (1962) and Ober- holzer et al. (1970) (Fig. 2). The radula of the ellobiids characteristically has many teeth in each row; the central tooth usually has a small crown. In most species the transition from the lateral teeth, which have a strong mesocone, to the pectinate marginal teeth is gradual. Morphology of the radula in the Melampinae and Ellobiinae 1$ rather uniform, but it varies in the Pedipedinae and Pythiinae. The radula of the Melampinae undergoes a series of morphological changes with age. The deeply indented crown of the lateral teeth of very young individuals becomes the unicuspid, triangular crown of adults. Some species, however, seem to have a radula with neotenic features, for example Melampus (D.) floridanus and Melampus (D.) paranus, which, as adults, have a conspicuous ectocone on the lateral teeth. This structure, present in the radula of the juveniles of some species of the Melampinae (Figs. 243-249, 370, 371), disap- pears with age. Marcus 8 Marcus (1963, fig. 8) observed the same morphological change in the radula of Ellobium (A.) dominicense. Their illustration of the radula of a very young specimen of that species shows a striking resemblance to the radula of an adult Blaune- па heteroclita. The Pedipedinae and Pythiinae display great radular diversity with as many as three radular types in each subfamily. Radulae of some Pedipedinae, such as Microtralia, re- semble that of the Melampinae, whereas in the Pythiinae the strong mesocone on the radular teeth of Cassidula and Pythia resem- bles that in adult Ellobium. The radula of the ellobiids is a much more useful character at the generic level than at the specific level. The minute differences in the radulae developed by analysis of closely related pairs such as Melampus (M.) cof- feus—Melampus (M.) bidentatus, Pedipes mirabilis—Pedipes ovalis and Leuconopsis novimundi—Leuconopsis manningi failed to provide morphological evidence useful in the separation of these species pairs. On the other hand, the different genera, mainly within the subfamilies Pythiinae and Pedipedinae, are readily distinguished on the basis of their radular morphology. Anatomical characters A series of anatomical characters com- monly used in devising classifications was listed by Solem (1978). Because all charac- ters do not have the same taxonomic value weighting always must be applied. Those characters having greater influence on the cohesion of the group should be used in phy- logenetic studies. Those same characters should be the least affected by nongenetic WESTERN ATLANTIC ELLOBIIDAE 171 factors, such as environmental and com- petitive pressures, exemplified by habitat and food. Thus the reproductive and the ner- vous systems ought to be considered prime taxonomic characters for the interpretation of phylogenetic relationships among higher taxa. Stomach: Graham's comprehensive studies (1939, 1949) of the functional morphology of the molluscan stomach showed an evolution- ary trend toward the disappearance of the crystalline style and simplification of the sort- ing areas in conjunction with the adoption of a macrophagous carnivorous diet. He ob- served the forward migration of the cardiac opening, with consequent reduction of the stomach to a blind sac into which the diges- tive gland discharges, and the increase in the muscularity of the mid-section to form a giz- zard. К is generally recognized that the basom- matophoran stomach originated from the prosobranch condition and it appears to me that it evolved along two different lines. The lower basommatophoran or archaeopulmo- nate stomach shows a tendency toward a forward migration of the cardiac opening. Otina otis (Turton), a primitive marine pulmo- nate, retains a vestigial style sac and has a rudimentary gizzard (Morton, 1955a). In the higher, limnic basommatophorans the stom- ach remains open-ended, with esophageal and intestinal openings at opposite ends, as shown in Carriker (1946) and Morton (19550). In this group the simplest stomach occurs in Acroloxus, which lacks musculature, has a well-developed caecum similar to a style sac and a structure similar to a crystalline style (Hubendick, 1978). Morton (1952, 1953) also investigated the functional morphology of the gastropod stomach and, on the basis of the disposition of the internal ciliary patches and of the tendency toward stronger muscularity of the mid-section, used it as a character in the classification of the ellobiids (Morton, 19550). In this study only the external appearance of the stomach was noted. Without an under- standing of the functional morphology of the internal parts, phylogenetic inferences and use in classification would be unwarranted and possibly misleading. Reproductive System: Traditionally the mol- luscan reproductive system has been ac- corded special value in the understanding of the phylogenetic relationships among higher taxa (Duncan, 1960a, b; Visser, 1977, 1981; Gosliner, 1981; Haszprunar, 1985, 1988; Sal- vini-Plawén 8 Haszprunar, 1987). The impor- tance of the reproductive system in gastro- pod evolutionary studies is corroborated by my studies. A basic plan of the gastropod reproduc- tive system consists of a posteriorly located gonad, a middle glandular section and an anterior duct associated with the pallial re- gion, primitively glandular owing to its prob- able origin from the hypobranchial gland (Fretter, 1984). This simple tube becomes in- creasingly complex with the appearance of specialized evaginations and of the herma- phroditic condition (Ghiselin, 1966). In proso- branchs gonochorism is the rule, a condition currently considered primitive (Haszprunar, 1988). Cases of protandry and of simulta- neous hermaphroditism exist in archaeo- gastropods and mesogastropods, however (Fretter, 1946). Simroth (1907) and Huben- dick (1945) both thought that hermaphrodit- ism was the original condition among the gas- tropods. Krull (1935, fide Fretter & Graham, 1962) also shared Simroth's view and, based on the fact that the pallial oviduct of the prosobranch hydrobiids is divided in a man- ner similar to that of the monaulic pulmo- nates (species with one bisexual duct), he suggested that the hydrobiids were the most primitive gastropods. This view has not been accepted by later authors. In the euthy- neurans (opisthobranchs and pulmonates), once commonly thought to have evolved from the archaeogastropods (Pelseneer, 1894a; Hubendick, 1945; Morton, 1955c), but more probably from the mesogastropods (Fretter, 1946; Boettger, 1954; Duncan, 1960a; Gos- liner, 1981) or Apogastropoda (Salvini-Plawén 8 Haszprunar, 1987; Haszprunar, 1988), her- maphroditism 1$ the universal condition (Ghis- elin, 1969). Opinions also differ as to which of the two hermaphroditic conditions appeared first, monaulic (one bisexual duct) or diaulic (two separate sexual ducts). Pelseneer's sug- gestion (1894b: 19) that hermaphroditism in mollusks arose by the secondary addition or grafting of a male system to the female indi- vidual has led to the view that топацу 1$ the primitive condition (Ghiselin, 1966; Marcus 4 Marcus, 1965b; Visser, 1977, 1981; Huben- dick, 1978). Solem (1972a, 1978) considered diauly the primitive condition, however, and stated that partial or total fusion of the male and female reproductive tracts has evolved independently in several groups. The choice 172 MARTINS of one or another hypothesis has obvious phylogenetic implications for the use of the reproductive system. Visser (1981) rejected Solem's opinion because there is no evidence of two separate gonads with two separate gonoducts in primitive gastropods. Visser, in contrast to Pelseneer, stated that hermaph- roditism in the Basommatophora, unlike that of the Stylommatophora, was derived from a male prosobranch. As evidence he cited the consistency ofthe penial structure throughout the basommatophorans (see also Hubendick, 1978). My work has led me to support the most commonly held view, namely that monauly and a glandular pallial gonoduct represent the primitive condition. The tendency toward reduction of the glandular elements of the re- productive system to the proximal, nonpallial portion is hereby taken as a derived trend. Supporting this view 1$ the presence of glan- dular pallial gonoducts among littorinids and the primitive opisthobranchs (Gosliner, 1981). Existence of such glandular ducts in groups otherwise clearly primitive (Pythiinae, Ellobii- nae) also is taken as supportive circumstan- cial evidence for the case. The terminology used in this study follows that of Duncan (1975), Visser (1977), Berry (1977) and Tompa (1984). Histological stud- ies were carried out to clarify some critical features of basic morphology, such as the extent of the mucous and prostate glands and to establish the aulic condition, the site of separation of male and female ducts. No distinction was made between the different components of the penial complex (penial sheath, preputium and penis) and this entire structure is herein called the penis. The de- gree of adhesion of the anterior vas deferens to the penis is also considered, the free con- dition being interpreted as derived. Central Nervous System. Use of the pulmo- nate central nervous system as a primary tax- onomic character has become increasingly accepted (Bargmann 1930; Van Mol, 1967; Bishop, 1978, Haszprunar, 1985, 1988; Sal- vini-Plawén 8 Haszprunar, 1987). Morton (1955c) and Regondeau et al. (1976) agreed that within the gastropods concentration of the ganglia is a derived character, but Morton shared Fretter 8 Graham's concern (1949) that sole reliance on characters of the ner- vous system to establish phylogenetic rela- tionships can be misleading. Haszprunar (1985, 1988) emphasized the possibility that concentration of ganglia and consequent eu- thyneury could be associated with small size in some cases. The degree of concentration of the ganglia of the central nervous system is considered important because the complexity of an en- tire system 15 generally unaffected by envi- ronmental pressures. Any major change in the arrangement of the ganglia probably would mean a greater rearrangement at most levels of anatomical organization. For this reason the morphology of the central nervous system is considered herein to be a taxo- nomic character useful at higher levels of classification. A detailed treatment of the ellobiid central nervous system 1$ provided for Ellobium (A.) dominicense (Fig. 21) and Melampus (М.) coffeus (Fig. 255). The terminology adopted here is from several sources (Simroth 1912, 1925-1928; Bargmann, 1930; Carriker, 1946; Brisson, 1963; Price, 1977). For most species only the relative concentration of the ganglia seemed important, but the nerves were found to approximate the pattern in Melam- pus (M.) coffeus. CLASSIFICATION OUTLINE, WESTERN ATLANTIC ELLOBIIDAE Family Ellobiidae H. & A. Adams in Pfeiffer, 1854 Subfamily Ellobiinae H. & A. Adams in Pfeiffer, 1854 Genus Ellobium Róding, 1798 Subgenus Auriculodes Strand, 1928 Ellobium (A.) dominicense (Férussac, 1821) Genus Blauneria Shuttleworth, 1854 Blauneria heteroclita (Montagu, 1808) Subfamily Pythiinae Odhner, 1925 Genus Myosotella Monterosato, 1906 Myosotella myosotis (Draparnaud, 1801) Genus Laemodonta Philippi, 1846 Laemodonta cubensis (Pfeiffer, 1854) Subfamily Pedipedinae Fischer & Crosse, 1880 Genus Pedipes Scopoli, 1777 Pedipes mirabilis (Mühlfeld, 1816) Pedipes ovalis C. B. Adams, 1849 Genus Creedonia new genus Creedonia succinea (Pfeiffer, 1854) Genus Microtralia Dall, 1894 Microtralia occidentalis (Pfeiffer, 1854) Genus Leuconopsis Hutton, 1884 Leuconopsis novimundi (Pilsbry & McGinty, 1949) Leuconopsis manningi new species Leuconopsis sp. Subfamily Melampinae Stimpson, 1851 WESTERN ATLANTIC ELLOBIIDAE 173 Genus Melampus Montfort, 1810 Subgenus Melampus $.5. Melampus (M.) coffeus (Linnaeus, 1758) Melampus (M.) bidentatus Say, 1822 Subgenus Detracia Gray т Turton, 1840 Melampus (D.) bullaoides (Montagu, 1808) Melampus (D.) floridanus (Pfeiffer, 1856) Melampus (D.) paranus (Morrison, 1951) Melampus (D.) monile (Bruguiere, 1789) Melampus (D.) morrisoni new name Genus Tralia Gray in Turton, 1840 Subgenus Тга/а s.s. Tralia (T.) ovula (Bruguière, 1789) SYSTEMATICS Family Ellobiidae H. & A. Adams in Pfeiffer, 1854 Auriculidae Lamarck, 1809: 321 [corrected from “Les Auriculacées” by Gray, 1840]. Auriculae Lamarck. Férussac, 1821: 32. Auriculadae Lamarck. Gray, 1824: 107. Auriculacea Lamarck. Blainville, 1824: 245. Auriculaceae Lamarck. Menke, 1828: 19. Auriculoidea Lamarck. Cristofori & Jan, 1832: 6. Auriculidea Lamarck. Beck, 1837: 101. Auriculata Lamarck. Sismonda, 1842: 26. Auriculiadae Lamarck. De Kay, 1843: 57. Auriculina Lamarck. Agassiz, 1847: 41 [cor- rection for Auriculacea]. Non Grateloup, 1838, nec Gray, 1847a. Carychiadae (Leach MS) Gray, 1847b: 269. Auriculae’inae Lamarck. Strobel, 1850: 32. Conovulidae Clark, 1850: 444. Melampidae Stimpson, 1851: 51. Ellobiidae H. & A. Adams in Pfeiffer, 1854b: 146 [in synonymy with Auriculacea La- marck]. Description: Shell spiral, dextral (except in Blauneria), oval-conic, sometimes umbili- cate, smooth or with spiral sculpture, cov- ered with brownish periostracum; aperture elongate, round at base, angulate posteriorly, with strong folds on inner lip, outer lip sharp or weakly reflected, often dentate. Inner whorls resorbed (except in Pedipes and Creedonia). Protoconch heterostrophic. Animal completely retractable into shell. Head separated from foot by transverse groove, into which a large mucous gland opens. Operculum present in embryos, ab- sent in adults. Mouth T-shaped; horny upper jaw sometimes with folded extremities lining lateral lips; one pair of subcylindric, contrac- tile or subretractile tentacles; eyes sessile, medial to base of tentacles; foot long, ante- riorly blunt, sometimes transversely divided, posteriorly tapered and entire or bifid; pneu- mostome on right side, medial to anal aper- ture. Radula broad, elongate; teeth numerous; central tooth equilateral; lateral teeth inequi- lateral, becoming shorter toward outer edges of radula, abruptly or gradually changing into marginal teeth. Digestive system moderately long; salivary glands usually elongate; esophagus long, thin walled, longitudinally grooved, opening posteriorly into wide crop; stomach generally tripartite with thin-walled cardiac region, muscular medial and pyloric regions and thin-walled, smaller posterior caecum; diges- tive gland usually bilobed, emptying anteri- orly at crop, posteriorly at gastric caecum. Reproductive system hermaphroditic; ovo- testis acinose and embedded in digestive gland or leaf-like and covering part of stom- ach; hermaphroditic duct with generally con- voluted seminal vesicle; glandular complex composed of whitish albumen gland, convo- luted posterior mucous gland, straight ante- rior mucous gland and prostate gland cover- ing pallial ducts (except in Melampinae); fertilization chamber follows posterior mu- cous gland and gives rise to oviduct and spermiduct, which might or might not be completely separate for their entire length; bursa duct and bursa arising from vagina at variable distances from aperture; female ap- erture medial to pneumostome, anterior to union of mantle with neck; male aperture on right corner of cephalic groove, under right tentacle; a fold of skin (sperm groove) runs from near female to male aperture, functional only in Pythia; in all others the vas deferens lies embedded in neck skin; it separates from skin inward near male aperture and enters penis at posterior end; penial complex (penis and penial sheath) lying over buccal bulb and cerebral ganglia. The hypoathroid, pentaganglionate central nervous system is of the basommatophoran type (Bargmann, 1930; Haszprunar, 1985), composed of 11 discrete ganglia, joined by connectives of various lengths: paired cere- bral, buccal, pleural, parietal and pedal gan- glia, and an unpaired visceral ganglion. Remarks: The Ellobiidae were first assigned familial rank by Lamarck (1809) under the vernacular Les Auriculacées. The group т- 174 MARTINS cluded Lamarck's Auricula and three other genera (Melanopsis, Melania and Limnaea) that were assigned subsequently by other authors to different families. Many incorrect Latinizations of Lamarck's vernacular name followed; Blainville's (1824) Auriculacea be- came well established and was used in major monographs on the family (Reeve, 1842; Küster, 1844; Pfeiffer, 1856a, 1876; Kobelt, 1898). The correct Latin designation, Auriculidae, was first used by Gray (1840) and was widely accepted until the 1920s, when the names Ellobiidae H. & A. Adams т Pfeiffer, 1854, and Melampidae Stimpson, 1851, replaced Lamarck's name. According to the International Code of Zoological Nomenclature Art. 11 (e) the name Auriculidae has priority because, when first published, it was based upon the name then valid for the contained genus Auricula La- marck. Odhner (1925), however, preferred the name Ellobiidae H. 4 A. Adams because the type genus, Auricula Lamarck, 1799, is a synonym of Ellobium Róding, 1798. The name Ellobiidae has been in general use since that time. Works dealing exclusively with the family, such as those of Odhner (1925), Morton (1955b, c), Hubendick (1956), Clench (1964), Marcus (1965), Marcus & Mar- cus (1965a, b), Cesari (1973, 1976), or gen- eral ones, such as those of Thiele (1931), Zilch (1959), Hyman (1967), Franc (1968), Fretter (1975), Jones (1975), Runham (1975), Berry (1977), Hubendick (1978), Solem (1978, 1985), Boss (1982) and Haszprunar (1985, 1988), and even popular books, such as those of Morris (1973), Humphrey (1975), Emerson & Jacobson (1976) and Rehder (1981), are the most obvious examples of the acceptance of the name Ellobiidae. Recently the name Melampidae Stimpson, 1851, has appeared in some influential mal- acological works, such as those of Keen (1971), Abbott (1974), Rios (1975) and Kay (1979). Morrison's reintroduction (1964) of the name Melampidae was unfortunate in several ways. It was an unnecessary distur- bance of taxonomic stability, because the name Ellobiidae had already been universally accepted. It also required a change to a dif- ferent type genus for the family. The appear- ance of the term Melampidae in influential malacological works obviously was leading to widespread use and consequent renewed taxonomic confusion. Strict application of the law of priority to family-group names would upset general use of the name Ellobiidae. In accordance with Art. 23 (d) of the ICZN, a petition should be submitted to the International Commission on Zoological Nomenclature to place the name Ellobiidae on the Official List of Family- Group Names in Zoology, and to place the names Auriculidae and Melampidae on the Official List of Rejected Names. Credit usually is given to H. & A. Adams (1855b) for the introduction of the name El- lobiidae. However, Pfeiffer (1854b), who had access to the Adams brothers’ manuscript, referred to the to-be-proposed family name, but continued to use the name Auriculacea. For this reason the name Ellobiidae, which should be credited to H. & A. Adams, must take the date 1854, when it was first pub- lished by Pfeiffer as a synonym. The family Ellobiidae varies greatly in mor- phology and anatomy, but it is nevertheless readily identifiable as a group at the familial level. Starobogatov (1976) exaggerated the differences in the reproductive system and raised the family name to ordinal status and considered existing subfamilies separate families. This view has not gained accep- tance and | have concluded that the differ- ences within the ellobiids are reconcilable within a single family. Odhner (1925), using radular characters, and Zilch (1959), using shell morphology, recognized six subfamilies, Carychiinae, Mel- ampinae, Pedipedinae, Pythiinae, Cassiduli- nae and Ellobiinae. Morton (1955c) and Hu- bendick (1978) merged the Cassidulinae with the Pythiinae on the basis of the similarities of their reproductive systems, and assigned the latter name to the group. My studies support Morton’s conclusions and | have followed his scheme of classification for the division of the Ellobiidae into subfamilies. Zilch (1959) recognized 20 genera of living halophilic ellobiids. Zilch’s classification is accepted here with certain modifications and 21 genera are recognized in this paper. Zilch considered Sarnia H. & A. Adams a subgenus of Tralia Gray, but Marincovich (1973), on the basis of radular morphology, placed it in the Ellobiinae. Sarnia (Fig. 181) shows strong conchological similarity to Pseudomelampus and Microtralia, and for that reason | include it in the Pedipedinae. Further information on the reproductive and nervous systems is needed to confirm the systematic position of this genus, however. | have synonymized herein Apodosis Pilsbry & McGinty with Leu- WESTERN ATLANTIC ELLOBIIDAE 175 ELLOBIINAE CARYCHIINAE PEDIPEDINAE 0002000 a A, Lan à À AMIA FIG. 3. Pictorial review of subfamilies of Ellobiidae, from most primitive to most advanced. A, Pythia (P.) scarabaeus (Linnaeus), radula; B, Pythia (P.) scarabaeus, reproductive system; C, Pythia (P.) plicata (Fér- ussac), central nervous system; D, Pythia (P.) scarabaeus; E, Рута (Trigonopythia) trigona (Troschel); Е, Ophicardelus australis (Quoy & Gaimard); G, Myosotella myosotis (Draparnaud); H, Ovatella firminii Pay- raudeau; |, Laemodonta octanfracta (Jonas); J, Allochroa bronni (Philippi); К, Cassidula (C.) aurisfelis (Bru- guière); L, Cassidula (Cassidulta) doliolum (Petit); M, Cylindrotis quadrasi Móllendorff; М, Auriculastra subula (Quoy & Gaimard); O, Ellobium (E.) aurismidae (Linnaeus), radula; P, Ellobium (E.) aurismidae, reproductive system; Q, Ellobium (E.) aurismidae, central nervous system; R, Ellobium (E.) aurismidae; S, Ellobium (Auriculodes) gangeticum (Pfeiffer); T, Auriculinella (Leucophytia) bidentata (Montagu); U, Blauneria hetero- clita (Montagu); V, Carychium tridentatum (Risso), radula; W, Carychium tridentatum, reproductive system; X, Carychium tridentatum, central nervous system; Y, Carychium minimum Müller; 2, Zospeum spelaeum (Rossmássler); AA, Pedipes mirabilis (Mühlfeld), radula; BB, Pedipes pedipes (Bruguiere), reproductive system; CC, Pedipes pedipes, central nervous system; DD, Pedipes pedipes; EE, Marinula pepita King; FF, Creedonia succinea (Pfeiffer); GG, Pseudomelampus exiguus (Lowe); HH, Sarnia frumentum (Petit); Il, Microtralia occidentalis (Pfeiffer); JJ, Leuconopsis obsoleta (Hutton); KK, Melampus (M.) coffeus (Linnaeus), radula; LL, Melampus (M.) coffeus, reproductive system; MM, Melampus (M.) coffeus, central nervous system; NN, Melampus (M.) coffeus; 00, Melampus (Micromelampus) nucleolus Martens; PP, Melampus (Detracia) bullaoides (Montagu); QQ, Melampus (Signia) granifer (Mousson); ВВ, Tralia (T.) ovula (Bruguière); SS, Tralia (Persa) costata (Quoy & Gaimara). conopsis Hutton, and have created the genus Creedonia. Myosotella Monterosato, treated by Zilch as a subgenus of Ovatella Bivona, is given herein generic status. A pictorial review of the subfamilies 1$ pre- sented in Figure 3. The radula, nervous sys- tem and reproductive system of the type or of a representative species of each subfamily are shown, as well as the shells of all the type species of the genera and subgenera herein recognized. The reproductive and nervous systems provided the most consistent basis for the separation of the subfamilies. Detailed descriptions of general anatomy and of the nervous system are provided un- der Melampus (M.) coffeus; these descrip- tions will be used as standards of compari- son in discussions of other species. Habitat: Ellobiids are mainly tropical. They commonly occur around the high-tide mark in mangrove areas, under rocks or pieces of de- caying wood. In extratropical regions they live in eurhyaline environments of salt marshes or in upper littoral rocky areas. Morton (1955c) divided the ellobiids into 176 MARTINS four groups according to habitat. The upper- tidal marine ellobiids, such as Melampus, Myosotella, Ophicardelus, Cassidula and El- lobium, prefer the high-tidal fringe of man- groves and marshes, never venturing far from the reach of the highest spring tides. The in- tertidal and crevice-dwelling species include the minute ellobiids of the genera Auricu- linella, Leuconopsis, Pedipes, Microtralia and Marinula, to which Creedonia, Laemodonta and Blauneria might be added, which live buried at different depths in the sediment or under partly buried rocks, roots and fallen branches in the upper intertidal zone. Pythia is the only coastal terrestrial ellobiid; it always frequents moist places near the shore, al- though out of reach of the highest tides. The inland terrestrial ellobiids are Carychium and Zospeum, which live in very humid environ- ments, frequently under forest leaf litter or in caves. The ellobiids are commonly thought to have evolved from an estuarine ancestor. Con- quest of the terrestrial habitat brought about modifications in the structure of the larval stages. Such modifications, however, are not exclusively related to distance from the sea and a single feature, such as suppression of a free-swimming veliger, can exist in intertidal and terrestrial species. A more or less mod- ified veliger stage 1$ present throughout the family. The Melampinae have a free-swim- ming veliger larva. Apley (1970) recorded two to six weeks of planktonic life for Melampus (M.) bidentatus, while Marcus & Marcus (1965a) suggested equally long periods for the veligers of Melampus (M.) coffeus and Melampus (D.) paranus. The veliger stage of the other subfamilies passes inside the egg and the embryo crawls immediately after hatching. Larvae of Ellobium (A.) dominicense have а ciliate velum and can swim for very short periods of time (Ewald, 1963). That same ciliated structure was found in Blauneria heteroclita by Marcus & Marcus (1963). Mor- ton (1955c) observed that the velum of the larvae of Myosotella myosotis and of Auricu- linella (L.) bidentata is reduced and lacks cilia. Another feature of larval ellobiids is the widespread presence of an operculum, which is lost at an early age. Blauneria heteroclita, which normally reaches 7 mm in length, sheds the operculum at a shell length of about 0.7 mm (Marcus & Marcus, 1963). The reduced operculum of Myosotella myosotis and of Au- riculinella (L.) bidentata helps to break the shell during hatching (Morton 1955b). Range: The family Ellobiidae has worldwide distribution, but appears to have three main centers, a large Indo-Pacific center, charac- terized by Ellobium, Cassidula and Pythia; a smaller West Indian center, characterized by Melampus; and a much poorer Mediterra- nean region, characterized by Myosotella and Ovatella. The fossil record of the Ellobiidae 1$ rela- tively poor and 1$ insufficient for the determi- nation of evolutionary lineages. The presence of the Indo-Pacific genera Ellobium and Cas- sidula in Europe during the Eocene and Miocene (Zilch, 1959) suggests that the Eu- ropean shores were connected with the Indo- Pacific region. This is consistent with the ex- istence of the Tethys Sea which, in various ways, extended longitudinally from Australia through Europe and northern Africa to the tropical West Indies and eastern Pacific. Existence of this seaway is indicated by the distribution of several groups of inverte- brates, and has been more extensively stud- ied for the Mediterranean region. Evidence of a Tethyan distribution in American faunas was found in Foraminifera (C.G. Adams, 1967), Ostracoda (McKenzie, 1967) and in bi- valves and gastropods (Palmer, 1967). The ellobiid fossil record does not provide any new information about Tethyan relationships between Europe and America. The ellobiid fossils of North America are represented by the melampinine genera Rhytophorus and Melampoides from the Cretaceous of Wyo- ming (White, 1895; Henderson, 1935) and by the more recent Melampus, Marinula, Tralia and Pedipes, from the Eocene, Miocene and Pleistocene (Conrad, 1862; Dall, 1912; Wood- ring, 1928; Gibson-Smith & Gibson-Smith, 1979, 1982, 1985). The Mesozoic genera seem not to have European counterparts, but Rhytophorus was recorded from the Lower Cretaceous of China (Zhu, 1980). Present records are too sparse to allow elab- oration of the meaning of such an occur- rence. The Cenozoic genera represent only the Recent ellobiid fauna of the West Indian region. Subfamily Ellobiinae H. & A. Adams т Pfeif- fer, 1854 Auriculea Pfeiffer, 1853a: 9. Ellobiinae “Н. & A. Adams” Pfeiffer, 1854b: 146. Auriculinae H. & A. Adams, 1855a: 30 [emen- dation of Auriculea Pfeiffer, 1853]. WESTERN ATLANTIC ELLOBIIDAE 177 Description: Shell very small and thin (Auri- culinella, Blauneria) to large and thick (Ello- bium), dextral except in Blauneria. Spire low to high, with very faint to marked and granu- lar spiral lines. Body whorl 60-80% of shell length, smooth or sculptured like spire. Ap- erture 70-80% of length of body whorl, oval- elongate; columellar tooth small, very oblique; anterior parietal tooth stronger, perpendicular to (Auriculinella) or weakly oblique to columel- lar axis; smaller posterior parietal teeth some- times present; outer lip thin and sharp to thick and weakly reflected, smooth internally. Pro- toconch smooth, prominent, with umbilicus- like slit in apex. Radula with central tooth small, triangular; lateral teeth bicuspid, with endocone smaller than mesocone; marginal teeth similar to lat- eral teeth but smaller. Animal whitish; eyes often concealed by thick skin; tentacles short, subcylindric or with dilated tips; foot entire (Ellobium) or transversely divided. Mandible corneous, semilunate. Stomach tripartite; mid-section very muscular. Spermiduct separates from oviduct before the latter leaves posterior glandular complex; anterior mucous gland covers entire length of vagina; spermiduct surrounded by prostate gland and might communicate with base of bursa duct near vaginal opening. Penis large and complex to small and simple (Blauneria); associated vas deferens adheres to penis. Visceral nerve ring long, with evidence of streptoneury in Ellobium; right parietovisceral connective very short. Remarks: There has been confusion in defin- ing the limits of the subfamily Ellobiinae. Pfeiffer (1853a) was the first to try to group the genera of the Ellobiidae into higher taxa. On the basis of the absence or presence of a reflected outer lip he recognized the subfam- ilies Melampea and Auriculea. He assigned Pythia, Auricula [= Ellobium] and Carychium to the latter group. The two subfamilial names were emended to Melampinae and Auriculinae by H. & A. Adams (1855a). Pfeiffer (1854b), after seeing the unpublished manu- script of the Genera of Recent Mollusca by H. & A. Adams (1855), mentioned some of the Adams’ conclusions, including the names Melampinae and Ellobiinae, and it is H. & A. Adams (in Pfeiffer) who should be credited with the introduction of the latter name (see the remarks under the Ellobiidae). Pfeiffer (1854b) continued to use his previous names and to the existing list of the Auriculea he added Plecotrema [= Laemodonta], Cassid- ula, Alexia [= Myosotella] and Blauneria. Two years later in his Monografia Pfeiffer (1856a) tentatively included the genus Leuconia [= Auriculinella] in this subfamily. Odhner (1925) noted the peculiar radula of Ellobium and he admitted only this genus to the subfamily. He wrongly stated that Ello- bium (E.) aurismidae lacks the central radular tooth (Fig. 17). Zilch (1959), who used con- chological characters, also considered the Ellobiinae monotypic. Thiele (1931), on the basis of radular morphology, reached quite different conclusions and he included in the Ellobiinae the subfamilies Melampinae, Pythi- inae and Cassidulinae. Studies of the comparative anatomy of the group are essential to an understanding of the taxonomic relationships within the Ellobi- inae and of the entire family as well. Morton (1955с) noticed the similarity of the reproduc- tive tracts of Auriculinella and Ellobium, and placed those two genera in the subfamily El- lobiinae. Likewise on the basis of reproduc- tive structures Marcus (1965) and Marcus 4 Marcus (1965b) added В/аипепа. In spite of the sinistrality of В/аипепа and the fact that Blauneria and Auriculinella are much smaller than Ellobium, the dentition of the inner lip shows a constant pattern in all three genera of the subfamily. This conchological similar- ity, corroborating the evidence shown by the reproductive system, makes these features useful phylogenetic characters. | therefore concur with the inclusion of Auriculinella and Blauneria in the subfamily Ellobiinae. Habitat: The various genera of the Ellobiinae live in somewhat different habitats. Ellobium is common on the muddy surface of Indo- Pacific mangroves, just below the high-tide mark, around roots and decaying wood (Berry et al. 1967). Blauneria lives buried in the black sediment, and under rocks and rot- ting vegetable material at the high-tide mark (Marcus & Marcus 19656; Martins, personal observation). Auriculinella lives closer to the low-tide mark than the other two genera; in the Azores it lives under rocks buried in gravel, sometimes into the intertidal zone (Martins, 1980). Range: The Ellobiinae have a worldwide dis- tribution, with only partial overlap of the dif- ferent genera. Ellobium, which is character- istic of the Indo-West Pacific mangroves, has a single representative in the tropical Eastern 178 MARTINS Pacific and another in the tropical Western Atlantic. Blauneria occurs only in the Western Indo-Pacific and in the Western Atlantic. Au- riculinella is restricted to the Mediterranean, the eastern North Atlantic and Macaronesian Islands. The subfamily seems to have had a Tethyan distribution, which is shown by the present distribution of Ellobium and by the presence of Ellobium and the Blauneria-like Stolidoma Deshayes in the Jurassic and Oli- gocene deposits of Europe (Degrange- Touzin, 1893; Zilch, 1959; Huckriede, 1967). Genus Ellobium Röding, 1798 Ellobium Röding, 1798: 105. Type species by subsequent designation of Gray (184 7a): Ellobium midae Röding, 1798 [= Bulla aurismidae Linnaeus, 1758]. Auricula Lamarck, 1799: 76. Type species by monotypy: Auricula midae (Röding, 1798) [= Bulla aurismidae Linnaeus, 1758]. Auriculus Montfort, 1810: 310. Type species by monotypy: Auriculus judae Montfort, 1810 [= Bulla aurisjudae Linnaeus, 1758]. Marsyas Oken, 1815: 305 [new name for El- lobium Röding]. Geovula Swainson, 1840: 344 [new name for Ellobium Röding]. Description: Shell moderately large and thin (25 mm) to large and thick (100 mm) and cov- ered with рае brown periostracum. Spire low to moderately high, sculptured with granular spiral lines crossed by more or less conspic- uous axial cords. Body whorl about 80% shell length, with same sculpture as spire, sometimes weakly depressed dorsoventrally. Aperture about 80% length of body whorl; small, very oblique, twisted columellar tooth; stronger anterior parietal tooth; posterior pa- rietal tooth sometimes present; outer lip thin to thick, sharp to weakly reflected. Radula with central tooth small, very nar- row, without mesocone, with ectocones curved inwards; lateral teeth with very wide, bicuspid crown; marginal teeth similar to lat- eral teeth, but smaller. Remarks: The name Ellobium Róding, 1798, was ignored for a long time in favor of its junior synonym Auricula Lamarck, 1799. Most probably the reason for maintaining the junior name was the acceptance of La- marck's work and ignorance of the Bolten Catalogue published by Róding in 1798 (Fis- cher, 1857; Dall, 1915). The vernacular name Auricule was first published by Lamarck in the Actes de la Société d'Histoire Naturelle de Paris in 1795 or 1796 (fide Férussac, 1821: 95), but the Latinized name Auricula first ap- peared in Lamarck's (1799) Prodrome, pub- lished in the Mémoires of the same society. Montfort (1810) pointed out that Lamarck (1799) had confused Auricula midae and Au- ricula judae by including in the references Ar- genville’s (1757: 226, pl. 10 [13], fig. G) “oreille de Midas,” which Montfort identified with Au- ricula judae. Montfort, then, renamed La- marck’s genus Auriculus and selected for its type species Auriculus judae [= Ellobium (E.) aurisjudae (Linnaeus)]. Pfeiffer (1876) pre- ferred Auriculus Montfort to Ellobium Róding and Auricula Lamarck, both of which he dis- missed as vague, owing to the heterogeneous assemblage of species that they included. The genus Ellobium is conchologically well Characterized by its auriform shape, by a finely reticulate sculpture and by the conspic- uous straw-colored to dark brown репо- stracum (Figs. 4-9). The central tooth of the radula is greatly reduced but not lost in Ello- bium (E.) aurismidae, as Odhner (1925) erro- neously reported (Fig. 17). The mesocone of the central tooth has been lost and the ecto- cones curl inwards and resemble a pair of pincers (Figs. 13, 14). Zilch (1959), on the basis of conchological characters, recognized the subgenera Ello- bium s.s. and Auriculodes Strand. Ellobium was Characterized as having a large, thick shell with a thick, reflected outer lip (Fig. 9), whereas Auriculodes had a smaller, thinner shell, with the outer lip sharp and weakly re- flected (Figs. 4-8). Some scattered informa- tion on the reproductive system of species belonging to both subgenera (see remarks under Auriculodes) indicates that the penis and vagina are usually more complex in Ello- bium s.s. More detailed research on a greater number of species of both subgenera is needed, however, to clarify the relative taxo- nomic positions of Ellobium s.s. and Auricu- lodes. Pending additional information, on the basis of shell thickness, | concur with Zilch (1959) in the recognition of these subgenera. Subgenus Auriculodes Strand, 1928 Autonoe Guppy, 1868: 244. Type species by monotypy: Autonoe riparia Guppy, 1868 [= Auricula dominicense Ferussac, 1821]. Non Leach, 1852. WESTERN ATLANTIC ELLOBIIDAE 179 Auriculina Kobelt, 1898: 77. Type species by original designation: Auricula (Auriculina) дапдейса Pfeiffer, 1855. Non Grateloup, 1838, nec Agassiz, 1847. Auricella Móllendorff, 1898: 160. Type spe- cies by original designation: Auricula (Auricella) auricella Férussac, 1821 [= Bulimus auricula Bruguière, 1789]. Non Jurine, 1817. Auriculodes Strand, 1928: 64 [new name for Auriculina Kobelt, 1898]. Autonoella Wenz, 1947: 36 [new name for Autonoe Guppy, 1868]. Description: Shell to 25 mm long, thin to somewhat solid. Spire with fine spiral lines, sometimes granular and crossed by axial granular cords. Body whorl not flattened dor- soventrally, smooth and shiny or with granular appearance as in spire. Inner lip of aperture with very oblique, twisted columellar tooth and somewhat stronger, weakly oblique pa- rietal tooth; outer lip sharp, sometimes slightly thick and somewhat sinuous at mid-length, slightly reflected in gerontic specimens. Animal with portion of vagina anterior to confluence with bursa duct straight, very short; associated vas deferens adhering to anterior vagina; penis moderately long, straight; associated vas deferens adhering to penis. Remarks: Guppy (1868), on the basis of a single beach specimen, introduced Autonoe [= Autonoella Wenz], which he considered al- lied to Melampus and Laimodonta [= Laem- odonta]. Later, in the revised list of the spe- cies of Trinidad, Guppy (1872: 7) observed under Synonyms, etc., “Comp. Auricula pel- lucens [= Ellobium (A.) dominicense (Férus- sac)].” Thiele (1931) considered Autonoe a subgenus of Melampus, as did Zilch (1959) for Autonoella Wenz, a replacement name for the preoccupied Autonoe Guppy. From the original description of Autonoella riparia (Guppy), and from Guppy’s illustration (1871: pl. 17, fig. 1), it seems that the specimen con- sidered was a juvenile of Ellobium (A.) domin- icense (Férussac). In view of this, | consider Autonoella Wenz a junior synonym of Auric- ulodes Strand. Kobelt (1898) proposed Auriculina at the same time as Móllendorff (1898) introduced Auricella for the smaller and thinner-shelled forms of Ellobium s.l. Because both names were preoccupied, Strand (1928) introduced the substitute name Auriculodes for Kobelt's Auriculina. Only two species of Auriculodes have been investigated anatomically, and they appar- ently differ greatly from each other in their pallial gonoducts. According to Marcus 4 Marcus (1965b) and Martins (this study) Ello- bium (A.) dominicense has a very short, straight vaginal section anterior to the con- fluence with the bursa duct, and a moder- ately long, straight penis. Knipper & Meyer (1956) briefly described the reproductive sys- tem of Ellobium (A.) gaziense (Preston, 1913) and they mentioned the lack of separation between male and female ducts. This feature is not typical of the subfamily and could lead to removal of Ellobium (А.) gaziense, a spe- cies with typical Auriculodes shell (Fig. 8), from the Ellobiinae. Knipper 8 Meyer's rep- resentation of the nervous system 1$ so sim- ilar to that of Ellobium (A.) dominicense (Mar- tins, this paper), however, that the accuracy of their report on the reproductive system should be questioned instead. Apparently there are variations in the penial structure of Ellobium s.s. as well. The highly coiled penis of Ellobium (E.) aurisjudae (Linnaeus) is typi- cal ofthe nominate subgenus (Morton, 1955b; Berry et al., 1967). Sumikawa 8 Miura (1978) observed a thick, straight penis in Ellobium (E.) chinense (Pfeiffer) although this species retains the characteristic long, coiled anterior vagina. Odhner (1925: pl. 1, fig. 10), on the other hand, represented a small, somewhat thickened, straight penis, and an equally straight vagina for Ellobium (E.) subnodosum (Metcalfe, 1851). All of these scattered ana- tomical observations on the genus hardly al- low conclusions to be drawn concerning the correlation between conchological and ana- tomical characters of these two subgenera, but | find the conchological characters suffi- cient to justify the separation of Auriculodes from Ellobium s.s. Habitat: Species of the subgenus Auricu- lodes prefer to live above the high-tide mark of mangrove swamps, gathering wherever there is rotten wood (Morrison, 1946; Marcus & Marcus, 1965b; Keen, 1971; Martins, per- sonal observation). Range: The subgenus Auriculodes is known from the eastern coast of Africa (Knipper & Meyer, 1956) and throughout the Indo-Pa- cific region. It is represented along the west- ern coast of Central America by Ellobium (A.) stagnale (Orbigny, 1835) and in the West In- dian region to Brazil by the closely related Ellobium (A.) dominicense (Férussac). 180 MARTINS Ellobium (Auriculodes) dominicense (Férussac, 1821) Figs. 4-7, 10-16, 18-22 Auricula dominicensis Férussac, 1821: 103 [Santo Domingo Island (Hispaniola); lec- totype herein selected MNHNP (Fig. 4)); Beck, 1837: 103; Beau, 1858: 15. Auricula pellucens Menke, 1828: 78 [Demer- ara (Guyana), South America; location of type unknown]; Menke, 1830: 36, 131; Kúster, 1844: 17, pl. 2, figs. 16, 17; Pfeif- fer, 1854b: 151; Pfeiffer, 1856a: 137; Bin- ney 8 Bland, 1870: 87; Simpson, 1889: 68. Conovulus pellucens (Menke). Voigt, 1834: hl Ellobium pellucens (Menke). H. & A. Adams, 1855b: 237; Morrison, 1951b: 10; Perry & Schwengel, 1955: 197, pl. 39, fig. 185; Morrison, 1958: 123; Marcus, 1965: 124-128 [taxonomy]; Marcus & Marcus, 1965b: 426-438, pl. 1, figs. 1-7, pl. 2, figs. 8-11, pl. 3, figs. 12-16 [anatomy, ecology, taxonomy]; Rios, 1970: 139; Abbott, 1974: 334, fig. 4106 [illustration from Dall (1885)]; Rios, 1975: 159, pl. 48, fig. 769; Altena, 1975: 88; Vokes & Vokes, 1983: 60, pl. 22, fig. 18. Autonoe riparia Guppy, 1868: 244 [Mayaro Point, Trinidad; type presumed to be in Victoria Institute, Trinidad, destroyed by fire in 1920 (Sherborn, 1940); Guppy, 1871: 306, pl. 17, fig. 1 [type figured]; Guppy, 1872: 7. Melampus riparius (Guppy). Pfeiffer, 1876: STE Auriculus pellucens (Menke). Pfeiffer, 1876: 359. Auricula (Auriculastrum) pellucens Menke. Dall, 1885: 275, pl. 18, fig. 8; Dall, 1889: 90, pl. 47, fig. 8; Maury, 1922: 54. Auriculastra pellucens (Menke). Kobelt, 1898: 101, pl. 15, figs. 5, 6; Haas, 1950: 197; Ewald, 1963: 11-14 [larval history]. Melampus (Autonoe) riparius (Guppy). Ko- belt, 1898: 213, pl. 25, figs. 5, 6; Thiele, 1931: 467. Auriculastrum pellucens (Menke). С.М. Johnson, 1934: 158; M. Smith, 1937, pl. 67, fig. 8 [plate from Dall (1885)]; Webb, 1942, pl. 11, fig. 21; M. Smith, 1951: 145, pl. 55, fig. 2, pl. 67, fig. 8; Coomans, 1958: 103. Melampus (Autonoella) riparius Zilch, 1959: 66, fig. 210. (Guppy). Description: Shell (Figs. 4-7, 10, 11) to 27 mm long, oval-elongate, somewhat solid, whitish-yellowish, covered with brownish pe- riostracum. Spire with as many as eight weakly convex whorls; sculpture as in sub- genus. Body whorl about 85% of shell length, subcylindric, smooth or with same sculpture as spire. Aperture about 80% length of body whorl. Inner partition of whorls occupying one-third of body whorl (Fig. 6). Protoconch smooth, prominent, with about one whorl vis- ible; lip weakly reflected at base, forming um- bilicus-like perforation in apex (Figs. 10, 11). Animal white; tentacles partly retractable, moderately long, subcylindrical, with swollen tip; eyes inside base of tentacles, deep in integumentum, barely visible; foot entire; mantle skirt white; anal opening continued by fold of mantle skirt forming longitudinally split tubular extension. Kidney long, narrow, whit- ish. Radula (Figs. 13-16, 18) with formula (26+1+26)x70. Central tooth small; base roughly rhombic; posterior portion elongate, emarginate at anterior quarter, where crown of next tooth seems to articulate; crown very small; mesocone lacking; endocones thin, sometimes curled inwards. Lateral and mar- ginal teeth not sharply distinct, here de- scribed always as lateral teeth; first seven to 12 with base short and wide, weakly pro- jected lateroanteriorly, with median anterior notch with which posterior process of crown of next tooth articulates; crown wide, roughly triangular, bicuspid, with conspicuous poste- rior process; mesocone wide, somewhat rounded anteriorly; endocone sometimes barely defined, mainly in adult specimens; gradual narrowing of crown and somewhat sharper definition of endocone marks teeth 12 to 21; base shorter and narrower than that of first group of lateral teeth, with lateral pro- jection resembling basal ectocone; crown somewhat narrow, elongate; endocone first very rudimentary, then absent; no clearly de- fined ectocone. Digestive system with mandible solid, crescentic, with concave, sharp anterior edge and tapered ends (Fig. 12). Salivary glands fusiform, separated from each other, attaching to whitish esophagus by thin liga- ments. Stomach tripartite (Fig. 19); anterior portion membranous, comprising cardiac and pyloric regions; mid-portion (gizzard) very muscular, subcylindric; gastric caecum thin, receiving posterior diverticulum anteri- WESTERN ATLANTIC ELLOBIIDAE 181 FIGS. 4-11. Ellobium. (4) E. (A.) dominicense (Férussac), lectotype (MNHNP), Santo Domingo [Hispaniola], sl 16.2 mm. (5) E. (A.) dominicense, Demerara, Guyana (ANSP 22251), sl 22.3 mm. (6) E. (А.) dominicense, Big Torch Key, Florida, sl 20.6 mm. (7) E. (А.) dominicense, Big Torch Key, Florida, sl 23.0 mm. (8) E. (А.) gaziense (Preston), syntype (BMNH 1969103), Gazi, British East Africa [Kenya], sl 18.2 mm. (9) E. (E.) aurismidae (Linnaeus), Malaysia, sl 90.4 mm. (10) Е. (A.) dominicense, lateral view of spire and protoconch, Big Torch Key, Florida. (11) E. (А.) dominicense, top view of spire and protoconch, Big Torch Key, Florida. Scale 1 mm. orly. Digestive gland bilobed, brownish; pos- terior lobe conic, partly covering ovotestis. Reproductive system (Fig. 20) with ovotes- tis follicular, covering posterior portion of stomach, beneath posterior lobe of digestive gland; hermaphroditic duct thin, straight; separation of male and female ducts just an- terior to fertilization chamber; secondary connection of posterior vas deferens with an- terior end of bursa duct; bursa duct as long as pallial gonoducts, emptying into oviduct a short distance from female aperture; anterior mucous gland covers oviduct as far as con- fluence with bursa duct. Penis moderately 182 MARTINS FIGS. 12-17. Ellobium, mandible and radular teeth. (12) E. (А.) dominicense, mandible, Big Torch Key, Florida; scale 1 mm. (13) Е. (А.) dominicense, radula of young specimen, Big Torch Key, Florida, sl 5.4 mm; scale 100 um. (14) E. (А.) dominicense, radula of young specimen, Big Torch Key, Florida, sl 5.4 mm; scale 100 um. (15) E. (А.) dominicense, radula of young specimen, Big Torch Key, Florida, sl 5.4 mm; scale 200 um. (16) E. (4.) dominicense, radula of young specimen, Big Torch Key, Florida, sl 5.4 mm; scale 200 um. (17) E. (E.) aurismidae, radula, Malaysia, sl 90.4 mm; scale 200 um. long, somewhat thin, simple; ramifications of right tentacular retractor muscle attach to pe- nis near male aperture; associated vas defe- rens adhering to penis; penial retractor short, attaching to nuchal region. Nervous system (Fig. 21) with ganglia wrapped in thick connective tissue; cerebral commissure twice length of cerebral gan- glion; pedal commissure very short; left ce- rebropedal and cerebropleural connectives somewhat longer than right ones, about as long as cerebral commissure; pleuroparietal WESTERN ATLANTIC ELLOBIIDAE 183 Cc 1L 2L 7L 8L 17L 20L 24L (AP 9L 10L yi 12L я | ON ( = FIG. 18. Ellobium (A.) dominicense, radula, Big Torch Key, Florida. Scale 10 um. FIG. 19. Ellobium (A.) dominicense, stomach, Big Torch Key, Florida. Scale 1 mm. connectives very long; left parietovisceral connective shorter than right one, beneath branch of columellar muscle coming from be- hind right tentacle; right parietovisceral con- nective crossing over left one before insertion in visceral ganglion (rudiment of chiasto- neury); cerebral ganglia as large as pedal ganglia; left parietal ganglion double; anterior portion of left parietal ganglion giving off nerve to artery in mantle skirt, posterior por- Е hd в У = RE, per $ su A EN FIG. 20. Ellobium (A.) dominicense, reproductive system, Big Torch Key, Florida. Scale 1 mm. tion giving off pallial internal and parietal cu- taneous nerves; osphradial ganglion present on pneumostomal nerve; tentacular nerves split at their origin; penial nerve coming off median lip nerve. Remarks: Ellobium (A.) dominicense (Férus- sac, 1821) has been considered a synonym of Ellobium (А.) pellucens (Мепке, 1828) al- though Dall (1885: 276) stated that Férus- sac's species was described in such a way as to be unidentifiable. Férussac's reference (1821: 103) to the sculpture and size of the shell and comparison with Auricula auricella, which he had just introduced, constitute enough indication to recognize the species. Contention might arise owing to the fact that Férussac's Auricula auricella was not de- scribed (Pfeiffer, 1856a: 134, footnote), but the author referred to Bulimus auricula Bru- guière, 1789, Lister (1770: pl. 577, fig. 326) [error for 32b] and Gualtieri (1742: pl. 55, fig. F). Férussac's description contains enough information to allow identification of the spe- cies and his name has priority over that of Menke (1828). Emerson & Jacobson (1976) considered Ellobium auricula (Bruguière) to be the earli- est name for the West Indian Ellobium. Bru- guiere (1789: 342) provided a description of a Bulimus aurícula that indeed could apply to 184 MARTINS gn ст ргсп тап a prg1 plprc pmpn * ‘ x \ ` E A и ‘ и 4 ppni / / plg и LA 0 ‘ Ц р!рп iu D---- + 2 st ampn pen FIG. 21. Ellobium (A.) dominicense, central nervous system, Big Torch Key, Florida. Scale 1 mm. the West Indian shells. No locality was given in the original description, and reference was made to the Gualtieri and Lister illustrations just cited. Férussac, as stated above, intro- duced without description Auricula auricella from Baie des Chiens Marins, New Cale- donia. He mentioned in his synonymy of Bru- guière’s name the same synonymic refer- ences given by that author. The fact that some Indo-Pacific species are conchologi- cally very similar to the West Indian species contributed to this confusion. Ellobium (A.) dominicense has been stated erroneously to live in Natal, East Africa (Krauss, 1848), prob- ably the result of a misidentification of Ello- bium (A.) gaziense (Preston, 1913), and Pfeiffer (1856a) considered the Indian Ello- bium ceylanicum H. 8 A. Adams, 1854, a jun- ior synonym of Auricula pellucens Menke, 1828 [= Ellobium (A.) dominicense (Ferussac, 1821)]. All this circumstantial evidence indi- cates that Bruguiere (1789) had described an Indo-Pacific shell, which was deposited at the Museum d’Histoire Naturelle de Geneve (Mermod & Binder, 1963). I therefore dis- agree with Emerson & Jacobson (1976), who misidentified Bruguiere’s name for the West Indian species. Ellobium (A.) dominicense has been placed wrongly in Auriculastra [Auriculastrum is an unjustified emendation (Marcus & Marcus, 1965b)] by Dall (1885) and others. Martens (1880) created Auriculastra as a subgenus of Marinula for those species similar to Ellobium s.s., but with visible eyes and knobbed ten- tacle tips. Ellobium (A.) dominicense has these characteristics, a fact which might ex- plain Dall’s decision. However, Auriculastra elongata (Parreyss, 1845), also originally listed by Martens and very similar to the type species, Auriculastra subula (Quoy & Gaim- ard, 1832), has a very different radula (Odh- ner, 1925) and appears to belong in the Pythiinae. The nervous system and the radula of the specimens of Ellobium (A.) dominicense here examined, collected on Big Torch Key, Flor- ida, differ from those of animals from Brazil studied by Marcus & Marcus (1965b). The central nervous system of the Floridian spec- imens is very similar to that of Ellobium (A.) gaziense (Preston), illustrated by Knipper & Meyer (1956: 106, fig. 6), differing only in that the left parietal ganglion is double in Ellobium (A.) dominicense. In the Brazilian specimens (Marcus 8 Marcus, 1965b: 431, pl. 3, fig. 13) there is only one left parietal ganglion. The latter authors did not refer to the osphradial WESTERN ATLANTIC ELLOBIIDAE 185 ganglion and their figures indicate that the pleuroparietal connectives are shorter than the cerebropleural connectives. In the Florid- ian specimens the pleuroparietal connectives are three times longer than the cerebropleu- ral connectives. The radula of the specimens from Florida is very similar to those of Ello- bium (E.) aurismidae (Fig. 17) and Ellobium (E.) aurisjudae, both from Malaysia. Marcus & Marcus (1965b: 433, pl. 2, fig. 8) described and figured a tricuspid central tooth with a small, triangular mesocone and rudimentary ectocones. In the specimens from Florida the mesocone is lacking and the slender ecto- cones are sometimes curved inwards, re- sembling small fangs. Preserved material from northern South America was not avail- able for comparative anatomical study; how- ever, intrapopulational variability in shell shape and intensity of sculpture is seen throughout the range of the species, al- though the sculpture seems to be more marked in northern South American speci- mens (Fig. 5). | am unsure about the phylo- genetic significance of the anatomical differ- ences observed in the Brazilian specimens. Should further comparative anatomical re- search establish that the South American specimens are a separate taxon, Menke’s name pellucens is available. Deposited in the Muséum National d’His- toire Naturelle de Paris are two syntypes of Auricula dominicensis Férussac, from which a lectotype is herein selected and figured (Fig. 4). Habitat: Ellobium (A.) dominicense lives in protected embayments in which mangrove growth is thin, buried in the soft black humic sediment or under rotting logs seldom cov- ered by high tide. It seems to be an oppor- tunistic species, usually found in colonies and apparently with very limited movement once established. Great numbers of shells clustered in a small area, indicative of former colonies, are often found without living ani- mals in the immediate vicinity. lt seems that the colonies are destroyed by lack of food or by some environmental change, even though apparently suitable habitats exist a few meters away (Ewald, 1963; Marcus & Mar- cus, 1965b; Martins, personal observation). Range: Florida, from Miami to Cedar Key (Dall, 1885); Dominican Republic (Férussac, 1821); Haiti; Guadeloupe (Beau, 1858); Trin- idad (Guppy, 1868); Yucatan, Mexico, to 90 75 60 45 30 FIG. 22. Ellobium (A.) dominicense, geographic distribution. Open circle, locality from literature. Cananeia, Brazil (Marcus & Marcus, 1965b). (Fig. 22). Specimens Examined: FLORIDA: Golden Beach (MCZ 157854); Miami (ANSP 77056; MCZ 104943); Virginia Key (USNM 338303); Key Biscayne (ANSP 345210; USNM 700836, 700911); Coconut Grove (MCZ 201646); El- liot Key (MCZ 110206); Key Largo (ANSP 192837; MCZ 243979; USNM 590644, 701421); Card Sound, Key Largo (A.M.); Rab- bit Key (ANSP 88136); Big Pine Key (ANSP 106384); Big Torch Key (USNM 61046, 492482, 492484; A.M.); Middle Torch Key (USNM 663960); Oyster Bay (USNM 37596); Lossman Key (MCZ 291093); Cape Sable (MCZ 291085, 292564; USNM 525156); Rog- ers River (MCZ 3981); 2.5 km E of Chokolos- kee Key (MCZ 58955); Harris Island, Ten Thousand Islands (USNM 381326); Blue Hill Island, near Goodland Point (ANSP 82742); $ of Cape Romano (ANSP 62833); Marco (MCZ 292565); Bonita Springs (MCZ 291088); Carl E. Johnson Park, near Little Carlos Pass (A.M.); Fort Myers (ANSP 66963, 140799; USNM 87733, 492483); Punta Rassa (MCZ 291091, 291094, 292566; USNM 39804); Punta Gorda (USNM 592297); Sanibel Island (ANSP 170650; MCZ 13721, 291089, 186 MARTINS 291090, 292563); Bokeelia (MCZ 291087); Wulfert (ANSP 219866). HAITI: lle-a-Vache (USNM 403877, 404948). MEXICO: Silam, Yucatán (ANSP 62656). VENEZUELA: М of Sinamaica, Zulia (USNM 536129). GUYANA: Demerara (ANSP 2225, 22241; MCZ 146522; USNM 31572, 58857, 119552). FRENCH GUIANA: Cayenne (MCZ 102934; USNM 126413). SURINAME: Saramacca (USNM 635276). Genus Blauneria Shuttleworth, 1854 Blauneria Shuttleworth, 1854a: 148. Type species by monotypy: Blauneria cuben- sis (Pfeiffer, 1841) [= Voluta heteroclita Montagu, 1808]. Blanneria Shuttleworth. Dall, 1885: 287 [in synonymy; error for Blauneria]. Blaumeria Shuttleworth. Verrill, 1901: 35 [er- ror for Blauneria]. Description: Shell to 8 mm long, elongate, fragile, translucent, whitish, sinistral. Spire with as many as nine flattened whorls. Body whorl about 60% of shell length. Umbilicus absent. Aperture about 70% of length of body whorl, oval-elongate; inner lip with very small columellar tooth; outer lip sharp, smooth. Pro- toconch prominent, smooth, with about one and one-half whorls visible. Radula having central tooth with wide, tri- angular, emarginate base; crown small, uni- cuspid. Lateral teeth gradually becoming smaller toward margin of radula, bicuspid, with strong mesocone and much smaller en- docone; no morphological distinction be- tween lateral teeth and marginal teeth. Animal whitish, translucent, with short, cylindrical tentacles and very conspicuous black eyes. Foot transversely divided. Ar- rangement of organs sinistral. Separation of male and female ducts just anterior to fertili- zation chamber, before oviduct enters pos- terior glandular complex; posterior vas defe- rens secondarily communicates with anterior end of bursa duct. Penis small, simple; asso- ciated vas deferens adhering to penis. Con- nectives of visceral nerve ring long. Remarks: The genus Blauneria is readily identifiable because it is the only sinistral el- lobiid taxon. The history of this once enig- matic small group, before it was placed tim- idly in a separate genus by Shuttleworth (1854a), is connected with that of the type species Blauneria heteroclita (Montagu), and will be discussed in the remarks under that species. Once it was discovered to be a member of the Ellobiidae, the genus Blauneria was placed in different subfamilies, depending upon which character assumed greater im- portance in the classification scheme of the particular malacologist. Fischer & Crosse (1880) included Blauneria and other “marine” ellobiids with an elongated spire in the Auri- culinae [= Ellobiinae]. Odhner (1925), on the basis of radular characters, considered the genus to belong to the Cassidulinae. Thiele (1931), who based his classification largely upon Odhner's radular studies, did not rec- ognize the subfamilies Pythiinae and Cassid- ulinae, and placed Blauneria, together with many other genera, in the heterogeneous subfamily Ellobiinae. Zilch (1959), probably on the basis of conchological similarities with the dextral Cylindrotis Móllendorff, 1895, re- moved Blauneria to the Pythiinae. Finally, Marcus (1965) and Marcus & Marcus (1965b), followed by Hubendick (1978), included Blauneria in the Ellobiinae owing to similari- ties of the reproductive system with those of Ellobium and Auriculinella. My anatomical studies confirm the taxonomic conclusions of these latter authors. Habitat: Blauneria commonly lives in man- groves at the high-tide mark in the sediment under rocks and decaying branches. Pease (1869: 60) reported that his Blauneria gracilis from Hawaii lives in the same habitat as Pe- dipes, in the crevices of stones covered at high tide. He observed that Blauneria never crawls on the sides or tops of the rocks dur- ing low tide, but only around the base, which was always wet. Range: The genus is known from the warm regions of the Indo-Pacific and from the trop- ical Western Atlantic. There is no known fossil record of this sinistral genus, but the concho- logically closely related, dextral Stolidoma Deshayes, 1863, has been recorded from strata as old as the Paleocene of Europe (De- grange-Touzin, 1893; Zilch, 1959). Zhu (1980) described a Blauneria ? elliptiformis from the Cretaceous of northeastern China. Blauneria heteroclita (Montagu, 1808) Figs. 23-40 Voluta heteroclita Montagu, 1808: 169 [Dun- bar, Scotland (error), herein corrected to Matanzas, Cuba; location of type un- WESTERN ATLANTIC ELLOBIIDAE 187 known]; Laskey, 1811: 398, pl. 81, figs. 1, 2; Turton, 1819: 254. Acteon heteoclita (Montagu). Fleming, 1828: >37 Achatina (?) pellucida Pfeiffer, 1840: 252 [Cuba; location of type unknown]. Tornatellina cubensis Pfeiffer, 1841: 130 [Cuba; location of type unknown]. Auricula heteroclita (Montagu). Thorpe, 1844: 146. Tornatella heteroclita (Montagu). Forbes & Hanley, 1852: 526. Blauneria cubensis (Pfeiffer). Shuttleworth, 1854a: 148; Franc, 1968: 525. Blauneria pellucida (Pfeiffer). Pfeiffer, 1854b: 152; Pfeiffer, 1856a: 153; H. 8 A. Adams, 1858: 643, pl. 138, fig. 8; Binney, 1859: 175, pl. 53, fig. 2; Binney, 1860: 4; Binney, 1865: 21, text fig. 22; Mórch, 1878: 5. Oleacina (Stobilus) cubensis (Pfeiffer). H. 8 A. Adams, 1855a: 136. Odostomia (Tornatellina) cubensis (Pfeiffer). Shuttleworth, 1858: 73. ? Odostomia cubensis (Pfeiffer). Poey, 1866: 394. Blauneria heteroclita (Montagu). Pfeiffer, 1876: 368; Arango y Molina, 1880: 60; Fischer 8 Crosse, 1880: 9, pl. 34, figs. 14, 14a, 14b [anatomy, radula, taxonomy]; Dall, 1885: 287, pl. 17, fig. 6; Dall, 1889: 92, pl. 47, fig. 14; Simpson, 1889: 60; Crosse, 1890: 259; Kobelt, 1900: 260, pl. 31, figs. 19, 20; Dall & Simpson, 1901: 369; Davis, 1904: 126; Peile, 1926: 88; Thiele, 1931: 466; Bequaert & Clench, 1933: 538; C.W. Johnson, 1934: 160; M. Smith, 1937: 147, pl. 67, fig. 14 [plate from Dall (1885)]; Morrison, 1951b: 10; Coomans, 1958: 104; Nowell-Usticke, 1959: 88; Zilch, 1959: 74, fig. 241; Warmke & Abbott, 1961: 152; Marcus, 1965: 124-128 [taxonomy]; Marcus & Marcus, 1965b: 438-446, pl. 4, figs. 25-29 [anatomy, taxonomy, habitat]; Rios, 1970: 139; Abbott, 1974: 334, fig. 4104 [illustration from Binney (1859)]; Altena, 1975: 87, fig. 42; Rios, 1975: 159, pl. 48, fig. 768; Hubendick, 1978; 20, fig. 164, 24, fig. 176 [nervous and reproduc- tive systems redrawn from Marcus & Marcus (1965b)]; Vokes & Vokes, 1983: 60, pl. 31, fig. 19; Jensen & Clark, 1986: 458, pl. 153. Blanneria pellucida (Pfeiffer). Dall, 1885: 287 [error for Blauneria; in synonymy]. Blaumeria heteroclita (Montagu). 1901: 35 [error for Blauneria]. Verrill, Description: Shell (Figs. 23-32) with length to 7 mm, elongate, fragile, transparent to translucent, shiny, whitish. Spire with as many as nine flat or weakly convex whorls; very faint spiral lines on teleoconch, crossed by irregular growth lines. Body whorl about 60% shell length in gerontic specimens, 70- 75% in young individuals. Aperture about 70% body whorl length, oval-elongate; inner lip weakly canaliculate at base, with small, very oblique columellar tooth, stronger, ob- lique parietal tooth at mid-length of aperture; outer lip sharp, smooth inside. Partition of inner whorls occupying about three-quarters of the body whorl (Fig. 26). Protoconch smooth, well developed, with one and one- half whorls visible, leaving umbilicus-like per- foration on apex (Figs. 30-32). Radula (Figs. 33-36) having formula (17 + 1 + 17) x 70. Base of central tooth wide, tri- angular, deeply emarginate anteriorly; crown very small, narrow, unicuspid. Lateral teeth 15 to 18; base quadrangular, anteriorly oblique away from central tooth, with small notch on anterior edge; crown wider and longer than base, bicuspid; mesocone strong, long; en- docone less than half the length of mesocone; from about sixth lateral tooth outward a pro- cess develops on posterolateral edge of crown, which articulates with notch in base of next tooth. Marginal teeth not morphologi- cally distinct from lateral teeth except in grad- ual decrease in size. Animal has external anatomy as in genus. Stomach (Fig. 37) with thin, somewhat di- lated cardiac region, and smaller, slightly thicker pyloric region; gizzard very muscular, barrel-shaped; gastric caecum invaginable, without posterior diverticulum. Reproductive system (Fig. 38) with ovotes- tis apical, granular, orange; hermaphroditic duct simple, with some pouch-like dilations (seminal vesicle) as it approaches albumen gland; male and female ducts separating just anterior to fertilization chamber; spermiduct thick, covered with prostatic tissue, commu- nicating with bursa duct where the latter opens into vagina; anterior mucous gland covers oviduct until confluence with bursa duct. Penis small, simple; associated vas deferens adhering to penis; penial retractor very short, attaching to nuchal region. Nervous system (Fig. 39) with cerebral ganglia largest; cerebral commissure as long as width of cerebral ganglion; pedal commis- sure very short; right cerebropedal and cere- bropleural connectives longer than left coun- 188 MARTINS FIGS. 23-35. Blauneria heteroclita (Montagu). (23) Hungry Bay, Bermuda, sl 6.7 mm. (24) Hungry Bay, Bermuda, sl 5.2 mm. (25) Hungry Bay, Bermuda, sl 4.3 mm. (26) Hungry Bay, Bermuda, sl 6.3 mm. (27) Plantation Key, Florida, sl 3.5 mm. (28) Matanzas, Cuba (MCZ 131769), sl 3.7 mm. (29) Isla Mujeres, Yucatán, Mexico (R.B.), sl 3.5 mm. (30) Lateral view of spire and protoconch, Big Pine Key, Florida. (31) Top view of spire and protoconch, Crawl Key, Florida. (32) Top view of spire and protoconch, West Summerland Key, Florida. (33) Lateral and central teeth of radula, Hungry Bay, Bermuda, sl 4.5 mm. (34) Lateral and central teeth of radula, Hungry Bay, Bermuda, sl 4.5 mm. (35) Lateral teeth of radula, Hungry Bay, Bermuda, si 4.5 mm. Scale, Figs. 30-32, 1 mm; Figs. 33-35, 100 um. WESTERN ATLANTIC ELLOBIIDAE 189 C IL 2b 3L A\ } 7 № À | } A A = ES 8L 9L 13L 14L 15L ES == AS der uy | | YA | и \ V FIG. 36. Blauneria heteroclita, radula, Hungry Bay, Bermuda. Scale 10 um. FIG. 37. Blauneria heteroclita, stomach, Hungry Bay, Bermuda. Scale 1 mm. terparts; left pleuroparietal and right parie- tovisceral connectives very long, the latter somewhat shorter than the former; right pleu- roparietal and left parietovisceral connectives about same size, about half length of cerebral commissure. Remarks: Ваипепа heteroclita (Montagu) was originally thought to belong to the En- glish malacofauna. The appearance of this Western Atlantic shell on the shores of Dun- bar, Scotland, can be attributed to the dump- ing of ballast of ships from the West Indies. FIG. 38. Blauneria heteroclita, reproductive sys- tem, Hungry Bay, Bermuda. A-C, transverse sec- tions and their locations. Scale 1 mm. FIG. 39. Blauneria heteroclita, central nervous sys- tem, Hungry Bay, Bermuda. Scale 1 mm. This little, fragile and elegant shell puzzled the European naturalists for some time. Pfeiffer, within 12 months, introduced the names Achatina ? pellucida (1840) and Tor- natellina cubensis (1841) for specimens from Cuba. H. 8 A. Adams (1855b, 1858) treated those two names as referring to species in very different groups. They assigned Torna- tellina cubensis to the terrestrial Oleacina, and they followed Pfeiffer (1854b) in allocat- ing Achatina ? pellucida to Blauneria. The species in question was placed in seven dif- ferent genera before Shuttleworth (1854a) 190 MARTINS hesitantly proposed that “Odostomia cuben- sis’’ probably should belong to a separate genus. Shuttleworth, in a presentation made at the Lyceum of New York a month before the appearance of that paper but published four years later, had considered the species to be marine on the word of the naturalist Blauner. Pfeiffer (1854b), upon receiving a communication from Gundlach that the ani- mal in question had conspicuous eyes at the base of the tentacles (Pfeiffer 1856a: 153), immediately adopted Shuttleworth's name Blauneria and placed the genus within the Auriculidae [= Ellobiidae]. | have found some discrepancies between the specimens | studied and those from Bra- zil examined by Marcus 8 Marcus (1965b). The Marcuses stated (p. 443) that the en- docone of the radular teeth is basal. The SEM photographs of my Bermudian specimens clearly show the endocone as part of the crown, not of the base (Figs. 33-35). Another discrepancy is found in the lengths of the pleuroparietal connectives of the visceral nerve ring (pl. 5, fig. 28). Based upon my ob- servations in the current study | suspect that the Marcuses reversed the right and left con- nectives. Blauneria differs from all other ellobiids in its sinistrality. Gerontic specimens have a very elongate and slender shell (Fig. 23). Most commonly, however, the body whorl of the shell is longer and wider than the spire (Figs. 27-29). This form has been the one commonly illustrated, represented by Binney (1865) and copied by Dall (1885), M. Smith (1937) and Abbott (1974). Habitat: Blauneria heteroclita lives in man- groves above the high-tide mark, where it is usually deeply buried in the soft sediment un- der rocks, rotten wood or on the roots of the propagules, where it occurs with Laemo- donta, Creedonia and Microtralia. Marcus 4 Marcus (1965b) stated that these animals are common in decaying banana trees washed ashore in Cananeia, Brazil. Range: Bermuda; Florida to Texas and Yu- catán, Mexico; West Indies; Panama (Olsson 8 McGinty, 1958); Suriname (Altena, 1975); Brazil (Fig. 40). Binney (1859: 176) stated, “Dr. Foreman collected a few specimens in a garden of Washington city. He believes them to have been brought on plants from Charleston, S.C.” Both places are distant from the range 90 75 60° 45 30 FIG. 40. Blauneria heteroclita, geographic distribu- tion. Open circle, locality from literature. of the species and, because there has been no confirmation of either record, | do not in- clude them in the range of the species. Specimens Examined. BERMUDA: Fairyland (ANSP 99076); Old Road, Shelly Bay (A.M.); Cooper's Island (ANSP 131647); Hungry Bay, S of Ely's Harbour (both A.M.). FLORIDA (USNM 39843, 67953): St. Augustine (USNM 663064); Rose Bay, N of New Smyrna Beach (A.M.); Miami (MCZ uncatalogued); Barnes Sound (ANSP 196748; MCZ 291100); Key Largo (MCZ uncatalogued; USNM 597460); Tavernier Key (USNM 492513); S of Ocean Dr., Plantation Key (A.M.); Lignumvitae Key (ANSP 156648; MCZ 294648); Lower Mate- cumbe Key (USNM 492521); Long Key (A.M.); Grassy Key (ANSP 397277; MCZ 291102; A.M.); Crawl Key (A.M.); Big Pine Key (ANSP 104106); end of Long Beach Drive and W of Kohen Avenue, both Big Pine Key (both A.M.); Sugarloaf Key (ANSP 88804, 104107); Boca Chica Key (USNM 270352); Cape Sable (MCZ 291099, 291101); Marco (ANSP 22470; USNM 37615, 37616); Semi- nole Point (ANSP 105422); Starvation Key (ANSP 130061); Fort Myers (USNM 492512); E of St. James, Pine Island (ANSP 93432); Captiva Island (ANSP 149907); Sarasota Bay (USNM 30626); Mullet Key (USNM 652410, WESTERN ATLANTIC ELLOBIIDAE 191 653108; A.M.); Tampa Bay (USNM 37614); Boca Ciega Bay (ANSP 9570); Shell Key (USNM 466212); Clearwater Island (ANSP 9350). ALABAMA: Coden Beach (USNM 422371). TEXAS: Galveston (MCZ 227843); Port la Vaca (MCZ 223050); N end of Padre Island, 45 km S of Рой Aransas (MCZ 228745). MEXICO: Isla Mujeres, Quintana Roo, Yucatán (В.В.). BAHAMA ISLANDS: GRAND BAHAMA ISLAND: North Hawksbill Creek (ANSP 370564); South Hawksbill Creek (ANSP 371810); GREAT ABACO IS- LAND (ANSP 299496); ANDROS ISLAND: South Mastic Point (A.M.); Stafford Lake (ANSP 151931); Mangrove Key (USNM 180672, 269947, 270198); Smith's Place, South Bight (USNM 257569, 269649); Linder Key (USNM 270224); NEW PROVIDENCE IS- LAND: Nassau (MCZ uncatalogued); SE shore of Lake Cunningham (ANSP 299720); Bonefish Pond (A.M.); ROYAL ISLAND (USNM 468124); AKLINS ISLAND: between Pleasant Point and Claret Cove (MCZ 225524). CUBA (ANSP 22471; MCZ uncata- logued; USNM 39842, 57726, 492511): Ha- bana (ANSP 130745, 326340; MCZ 233993); Salt Works, Hicacos Peninsula (ANSP 157338); La Chorrera (MCZ 128256, 167956); Сало (MCZ 167955); Matanzas (MCZ 131769; USNM 492510); Batabanó (ANSP 93730; MCZ 167957). JAMAICA (ANSP 16705, 22472; USNM 94765): Green Island Harbor (USNM 440791); Montego Bay (ANSP 329122); Port Morant (USNM 423688); King- ston (USNM 427130, 467555); Hunt's Bay (USNM 427117). HAITI: lle-a-Vache (USNM 403701, 403859, 403872, 404947); Landep- rie Bay (USNM 383264); between Vieux Bourg and Baie des Flamands (USNM 402467); Aquin (USNM 403149); Bizoton (USNM 403324). PUERTO RICO: Punta Are- nas, N of Joyuda (A.M.); Puerto Real (A.M.). VIRGIN ISLANDS: ST. THOMAS (ANSP 22473). CARIBBEAN ISLANDS: GRAND CAYMAN ISLAND (ANSP 209768). BRAZIL: Cananeia (ANSP 305213; USNM 699448). Subfamily Pythiinae Odhner, 1925 Scarabinae Fischer & Crosse, 1880: 5. Pythiinae Odhner, 1925: 14. Description: Shell variable in size. Aperture usually heavily dentate; one columellar tooth; one to three, commonly two parietal teeth, anterior one strongest; outer lip generally in- ternally dentate. Radula with mesocone of lateral teeth tri- angular, usually pointed; marginal teeth be- coming smaller toward margin, with as many as three subequal cusps. Animal with rudimentary anterior tentacles sometimes present; foot entire. Pallial gono- duct entirely hermaphroditic; anterior mu- cous gland and prostate gland covering spermoviduct along entire length; bursa duct emptying near vaginal opening; spermatic groove open in Pythia; penis simple; vas def- erens adhering to penis externally or free in haemocoel. Ganglionic connectives of vis- ceral nerve ring long, leaving pedal ganglia mid-way between cerebral ganglia and vis- ceral ganglion; right parietovisceral connec- tive longer than left one. Remarks: Fischer & Crosse (1880) created the subfamily Scarabinae for Scarabus Mont- fort, 1810 [= Pythia Róding, 1798] on account of its oddly shaped, dorsoventrally flattened shell. Odhner (1925) used the name Pythiinae because by that time Scarabus Montfort was recognized as a junior synonym of Pythia Roding; he included Alexia [= Myosotella] and Blauneria on the basis of radular characters. Cassidula and Ophicardelus were added by Morton (1955c), who merged Odhner's Cas- sidulinae with the Pythiinae. Zilch (1959) re- verted to Odhner’s division and included in the Pythiinae the Recent genera Pythia, Ova- tella, Cylindrotis and Blauneria and removed Ophicardelus and Cassidula to the Cassiduli- nae. Marcus (1965) and Marcus 8 Marcus (1965b) noted that in Blauneria the spermi- duct and oviduct separate before the her- maphroditic duct enters the glandular com- plex, and so removed this genus to the Ellobiinae. In consideration of shell and ana- tomical features, | have concluded that Lae- modonta must be included in the Pythiinae. Dall (1885) included Sayella within the El- lobiidae and Zilch (1959) listed it, with a ques- tion mark, within the Pythiinae. Morrison (1939), however, showed that Sayella Dall is not an ellobiid, but a pyramidellid opistho- branch. Separation of the Cassidulinae from the Pythiinae, as Odhner (1925) proposed and Zilch (1959) supported, 15 not justifiable. The two groups are similar in the basic pattern of the inner lip teeth of the shell aperture but their radular morphology shows too much di- versity and overlap to constitute a useful tax- onomic character at the subfamilial level. Both groups have a similar plan of the ner- 192 MARTINS vous system and, for this reason, Morton (1955c) regarded Odhner's Cassidulinae as superfluous. The nervous system of the Cas- sidulinae indeed shows the elongate right pa- rietovisceral connective, characteristic of the Pythiinae. Morton erroneously stated that the pallial gonoduct of Cassidula is very similar to that of Myosotella in that it remains hermaph- roditic until the vaginal aperture. According to Berry et al. (1967), Berry (1977) and Mar- tins (personal observation) the vas deferens of Cassidula aurisfelis (Bruguière) separates from the oviduct some distance before the vaginal opening, and runs free until entering the neck skin to follow the spermatic groove. This feature can be considered secondary to the general pattern of the reproductive sys- tem, however, for the bursa duct opens at the same position relative to the separation of the vas deferens in Cassidula as it does in the other Pythiinae. The same arrangement oc- curs in the Ellobiinae. Ellobium (E.) aurisjudae also has a long, nonglandular vagina, which is in accordance with the highly specialized penial complex of the species. Ellobium (E.) aurismidae, on the other hand, has a less specialized penis and lacks the long, non- glandular vagina (Morton, 1955c; Berry et al., 1967; Martins, personal observation). In both species, however, the bursa duct opens at the anterior end of the glandular portion of the oviduct. In view of the similarities of the repro- ductive and nervous systems of the two groups, as well as their similar patterns of apertural dentition, Morton's decision (1955c) to merge the Cassidulinae with the Pythiinae is hereby followed. Habitat: The Pythiinae contain very primi- tive ellobiids such as Pythia, Myosotella, Ophicardelus and Cassidula. These groups have left the proximity of the sea and are less dependent upon that element than all other halophilic ellobiids. Руа has acquired a semiterrestrial habitat, and Myosotella, Oph- icardelus and Cassidula were placed by Morton (1955c) among the “supratidal and estuarine ellobiids.” Laemodonta lives in rocky areas at the high-tide mark, with Pe- dipes, and in the mangroves at or just below the high-tide mark, under rocks and fallen branches. Range: The Pythiinae have a worldwide distribution. Pythia, Cassidula and Ophicar- delus are characteristic of the tropical Indo- Pacific; Laemodonta, also common in the Indo-Pacific, is represented in the West Indies by one species. Ovatella and Myosotella are represented in the Mediterranean, but the latter has been introduced to eastern North America (Binney, 1859; Verrill, 1880), Califor- nia (Hanna, 1939), western South America, South Africa and Australia (Climo, 1982). Genus Mysotella Monterosato, 1906 Phytia Róding, 1798. Gray, 1821: 231 [mis- spelling of Pythia]. Phitia Gray. Blainville, 1824: 246 [misspelling of Gray’s misspelling of Ру]. Phythya Gray. Deshayes, 1832: 762 [mis- spelling of Gray’s misspelling of Руа]. Jaminia Brown, 1827, pl. 51. Type species by subsequent designation of Gray (1847a): Jaminia denticulata (Montagu, 1803) [ = Auricula myosotis Draparnaud, 1801]. Non Risso, 1826. Alexia “Leach” Gray, 1847a: 179. Type spe- cies by monotypy: Alexia denticulata (Montagu, 1803) [= Auricula myosotis Draparnaud, 1801]. Non Stephens, 1835. Kochia Pallary, 1900: 239. Type species by subsequent designation of Monterosato (1906): Alexia (Kochia) oranica Pallary, 1900 [= Auricula myosotis Draparnaud, 1801]. Non Frech, 1891. Myosotella Monterosato, 1906: 126. Type species by original designation: Myoso- tella payraudeaui “Shuttleworth” Pfeif- fer, 1856a [= Auricula myosotis Drapar- naud, 1801]. Nealexia Wenz, 1920: 190 [new name for A/- exia Gray, 1847, non Stephens, 1835]. Description: Shell to 10 mm long, fragile to somewhat solid, pale yellow to purplish red. Spire high, with as many as eight weakly con- vex, spirally striated whorls; only one spiral row of hairs in juveniles. Aperture oval-elon- gate; inner lip with small, very oblique col- umellar tooth, strong anterior parietal tooth and usually one, sometimes more, parietal teeth becoming smaller posteriorly; outer lip sharp, weakly reflected, commonly with one or more inner tubercles. Protoconch smooth, large, with one and one-half protruding whorls, leaving umbilicus-like slit in apex of shell (Figs. 76, 77). Radula with base of central tooth wide, emarginate half of its length; crown of mar- ginal teeth pointing medially, mesocone stronger than endocone. WESTERN ATLANTIC ELLOBIIDAE 193 Animal grayish-white; neck and tentacles sometimes darkly pigmented. Hermaphro- ditic duct convolute; pallial gonoduct her- maphroditic as far as the vaginal арейиге; anterior mucous gland and prostate gland cover entire length of spermoviduct; bursa duct emptying near vaginal aperture; penis short, thick; vas deferens adhering to penis. Ganglia of visceral nerve ring widely spaced; osphradial ganglion present. Remarks: The majority of modern literature has treated Myosotella Monterosato, 1906, as a subgenus of Ovatella Bivona, 1832. The anatomy of the type species of Myosotella, Myosotella myosotis (Draparnaud), has been studied extensively (Meyer, 1955; Morton, 1955b) and Giusti (1973) looked briefly into the anatomy of the type species of Ovatella, Ovatella firminii (Payraudeau, 1826). A study of the anatomy of Ovatella aequalis (Lowe, 1832) from the Azores (Martins, personal ob- servation) revealed the presence of a pallial gland, not noted by Giusti (1973) for Ovatella firminii, similar to that in Carychium tridenta- tum (Müller) (Morton, 1955b), Руа scara- beus (Gmelin, 1791) (Plate, 1897), Cassidula labrella (Deshayes, 1830) (Renault, 1966) and Laemodonta cubensis (Pfeiffer, 1854) (Mar- tins, this study). In another work (Martins, 1980) Ovatella aequalis was shown to have a tripartite mandible with tapering ends, whereas that of Myosotella myosotis 1$ entire and quadrangular. These two characteristics, corroborated by differences in the proto- conch, justify the attribution of generic rank to Myosotella. Some modern authors, following Kennard 8 Woodward (1919), treat Myosotella as a junior synonym of Gray’s misspelling “Phy- па’ (Morrison, 1951а; M. Smith, 1951; Мс- Millan, 1968; Keen, 1971; Climo, 1982). The word “Phytia” appeared in Gray (1821) and is obviously a misspelling of Pythia Róding, 1798, for two reasons. First, as Watson (1943) pointed out, the family Ellobiidae was, at the time of Gray’s publication, divided into very few genera, and Carychium Müller, 1774, Рута Roding, 1798, and Auricula La- marck, 1799, all had been established many years earlier. Pythia had been introduced for Pythia helicina Róding [= Helix scarabeus Gmelin], a species which has a row of tuber- cles inside the outer lip. Group b of Gray's “Order 1. Adelopneumona” included the am- phibious Auricula, Carychium and “Phytia.” Gray's only example of “Ррува’” was Voluta denticulata Montagu, a form of Myosotella myosotis that also has two or more tubercles inside the ощег lip. It can be assumed, there- fore, that Gray was including Voluta denticu- lata Montagu within the already known genus Pythia Róding on the basis of the dentition of the outer lip. Second, Gray's publication is notorious for the number of misspellings it contains. For example, in the first nine lines of page 231, on which “Phytia” appears in the fifth line, one can read: Clauselia [= Clau- silia], Ancillus [= Ancilus] and Phaneropneu- mana [= Phaneropneumona] and, near the bottom of the page, Neritino [= Neritina]. Fur- thermore, Gray (1847a) corrected “Phytia” to Pythia. In view of the above, “Phytia” of Gray must be treated as a misspelling of Pythia Rôding, in accordance with Articles 19 and 32 ii of the ICZN, and as such it lacks taxo- nomic standing. Сгау’$ misspelling was later misspelled by Blainville (1824) and Deshayes (1832). Gray (1847a) also introduced Alexia for Vo- luta denticulata Montagu [= Аипсша myoso- tis Draparnaud]. Stephens (1835) had used the same name for a genus of Coleoptera, however, rendering Gray's name preoccu- pied. This fact prompted Wenz (1920) to pro- pose Nealexia as a new name for Alexia Gray, but Myosotella Monterosato, 1906, has pre- cedence over Wenz' name. In two more instances Gray made mistakes concerning Ovatella [= sensu Myosotella]. In 1840 he used Ovatella Bivona as a subgenus of Conovulus Lamarck for Voluta denticulata Montagu; later (1847a) he included ‘‘Ovatella Gray non Bivona” in the synonymy of his A/- exia. Because Voluta denticulata was not in- cluded in Bivona's (1832) original species, “Ovatella 'Bivona' Gray” must be treated as a misuse of Ovatella Bivona. Gray (1847a) also included in the synonymy of his Alexia the name Jaminia Brown, 1827, but the latter name was preoccupied by Jaminia Risso (1826). Pallary (1900) proposed Kochia as a sub- genus of A/exia Gray and he included, among other species, Alexia (K.) denticulata (Mon- tagu) and Alexia (K.) oranica Pallary [both junior synonyms of Myosotella myosotis (Draparnaud)]. The latter species was se- lected as type species by Monterosato (1906). Pallary (1921), unaware of Montero- sato’s selection, proposed Alexia (K.) dentic- ulata as the type species of Kochia, noting at the same time that this name was preoccu- pied by Kochia Frech (1891). 194 MARTINS Monterosato (1906) considered Montagu's Voluta denticulata and Draparnaud's Auricula myosotis not only as being different species, but as belonging to different genera. Leaving the former within Gray's Alexia, he included the latter within his genus Myosotella, which he created for a group of species under Pfeiffer’s Alexia #2 (1856a: 147); he desig- nated Myosotella payraudeaui ('“Shuttle- worth” Pfeiffer, 1856) as the type species. On the basis of Pfeiffer's description, | consider Myosotella раугаиаеаш conspecific with My- osotella myosotis (Draparnaud). Monterosa- to's name, then, 1$ the earliest available name for the subgenus that includes Myosotella myosotis (Draparnaud). Habitat: Myosotella lives mainly above the high-tide mark, sometimes even away from the influence of spring tides (Morton, 1955b). Range: Although it has a worldwide distribu- tion Myosotella is generally absent from the tropics. Myosotella myosotis (Draparnaud, 1801) Figs. 41-84 Auricula myosotis Draparnaud, 1801: 53 [Mediterranean coast; type probably in Vienna (Locard, 1895)]; Draparnaud, 1805: 56, pl. 3, figs. 16, 17; Férussac, 1821: 103; Lamarck, 1822, 6: 140; Blain- ville, 1824: 246; Blainville, 1825: 453, pl. 37 bis, fig. 6; Gould, 1833: 67; Griffith & Pidgeon, 1834: 36; Küster, 1844: 19, pl. 1, figs. 15-17; Moquin-Tandon, 1851: 348-351 [апаюту]. Voluta denticulata Montagu, 1803: 234, pl. 20, fig. 5 [Devon, England; lectotype herein selected RAMM 4100 (Fig. 41); paralectotypes RAMM 4100]; Dillwyn, 1817: 506; Wood, 1825: 90, pl. 19, fig. 18. Voluta ringens Turton, 1819: 250 [England; lectotype herein selected USNM 85901 1 (Fig. 42); paralectotype USNM 55351]. Voluta reflexa Turton, 1819: 251 [Exmouth, England; holotype USNM 55370 (Fig. 44)]. Phytia denticulata (Montagu). Gray, 1821: 132; Gardiner, 1923: 64; Germain, 1931: 561, text fig. 597. Auricula veneta Von Martens, 1824: 433 [Venice: location of type unknown (fide Cesari, 1976)]. Jaminia denticulata (Montagu). Brown, 1827, ВТ. tig. 6. Jaminia quinquedens Brown 1827, pl. 51, fig. 11 [Prestonpans, England; type probably at Manchester (Sherborn, 1940)]. Acteon denticulatus (Montagu). Fleming, 1828: 337. Auricula tenella Menke, 1828: 36 [Type local- ity herein designated to be Norderney Is- land; location of type unknown]; Menke, 1830: 131; Küster, 1844: 57. Carychium personatum Michaud, 1831: 73, pl. 15, figs. 42, 43 [Bretagne, France; lectotype herein selected MNHNP (Fig. 45)]. Melampus borealis Conrad, 1832: 345 [New- port, Rhode Island; type material pre- sumed lost (Baker, 1964)]; Jay, 1839: 59; Н. & A. Adams, 1854: 10. Melampus gracilis Lowe, 1832: 288 [Madeira; location of type unknown]. Auricula myosotis Lamarck. Orbigny, 1835: 28: Руша denticulata (Montagu) Gray. Beck, 1837: 103: Руа myosotis (Draparnaud). Beck, 1837: 104. Auricula reflexilabris Orbigny, 1837: 326, pl. 42, figs. 1-3 [Lima, Peru; lectotype herein selected ВММН 1854.12.4.242 (Fig. 46)]. Auricula (Auricula) myosotis (Draparnaud). Anton, 1839: 48. Auricula denticulata (Montagu). Gould, 1841: 199, fig. 129; De Kay, 1843: 58, pl. 5, fig. 93; Küster, 1844: 54, pl. 8, figs. 1-5; Reeve, 1877, pl. 7, fig. 61. Auricula mysotis Draparnaud. Sowerby, 1842: 99 [misspelling of myosotis]. Auricula denticulata var. borealis (Conrad). De Kay, 1843: 58, pl. 5. fig. 91. ?Auricula sayi Küster, 1844: 42, pl. 6, figs. 14, 15 [United States of America; location of type unknown (nomen dubium)]. Auricula microstoma Küster, 1844: 52, pl. 1, figs. 18, 19 [Budua, Dalmatia; location of type unknown]. Auricula kutschigiana Küster, 1844: 54, pl. 8, figs. 11-14 [Servola near Trieste; Lissa Island; location of type unknown]. Auricula biasolettiana Kúster, 1844: 56, pl. 8, figs. 18-20 [Niza; Trieste; coast of Dal- пана; location of type unknown]. Auricula myosotis var. elongata Küster, 1844: 69, pl. 8, figs. 21, 22 [Zara; location of type unknown]. Auricula myosotis var. adriatica Kúster, 1844: 69, pl. 8, figs. 23, 24 [Trieste; Istria; Dal- matia; Zara; location of type unknown]. WESTERN ATLANTIC ELLOBIIDAE 195 Auricula ciliata Morelet, 1845: 77, pl. 7, fig. 4 [Alcácer do Sal, Alentejo, Portugal; lec- totype herein selected BMNH 1893.2.4.831 (Fig. 47)]. Аипсша botteriana Philippi, 1846: 97 [Lesina Island, Dalmatia; location of type un- known]. Melampus denticulatus auct. 185152: Alexia denticulata (Montagu). Leach, 1852: 97; Locard, 1882: 182; Adam, 1947: 39; Sevo, 1974: 5, fig. 5. Alexia obsoleta Pfeiffer, 1854a: 111 [Tergesti, Adriatic Sea; location of type unknown]; Kobelt, 1898: 131, pl. 19, figs. 5, 6. Alexia myosotis (Draparnaud). Pfeiffer, 1854b: 151; Pfeiffer, 1856a: 148; Binney, 1859; 172, pl..15, 119.88; plu 79, Tig: 16; Binney, 1860: 4; Binney, 1865: 4, figs. 2-4; Tryon, 1866: 6, pl. 18, figs. 1, 2; Pfeiffer, 1876: 365; Nevill, 1879: 227; Verrill, 1880: 250; Locard, 1882: 183; Apgar, 1891: 130; Schneider, 1892: 116; Whiteaves, 1901: 208; C.W. Johnson, 1915: 178; Morse, 1921: 21, pl. 7, fig. 44; Nobre, 1930: 165, pl. 7, fig. 70; Nobre, 1940: 36; Adam, 1947: 38; La Rocque, 1953: 262; Porter, 1974: 300; Sevo, 1974: 6, fig. 6. Conovulus denticulatus (Montagu). Clark, 1855: 297. Alexia bermudensis H. 8 A. Adams, 1855a: 33 [Bermuda; lectotype herein selected BMNH 1969105 (Fig. 48)]: H. & A. Ad- ams, 1855b: 241; Pfeiffer, 1856a: 152; Pfeiffer, 1876: 367; Kobelt, 1901: 282, pl. 33, fig. 3; Fénaux, 1939: 43, pl. 1, fig. 6. Conovulus (Alexia) denticulata (Montagu). Woodward, 1856: 174. Alexia payraudeaui “Shuttleworth” Pfeiffer, 1856a: 147 [Corsica; Nizza; Tergesti; lo- cation of type unkown]; Pfeiffer, 1876: 365; Kobelt, 1898; 130, pl. 17, figs. 21, 22: Melampus turritus (Say MS) Binney, 1859: _ 174 [Rhode Island; type presumably de- posited at ANSP, probably lost]. Auricula bicolor Morelet, 1860: 206, pl. 5, fig. 7 [Pico, Azores; lectotype herein se- lected BMNH 1893.2.4.822 (Fig. 49)]. Auricula vespertina Morelet, 1860: 210, pl. 5, fig. 9 [Pico, Azores; lectotype herein se- lected ВММН 1893.2.4.825 (Fig. 50)]. Alexia micheli Bourguignat, 1864: 140, pl. 8, figs. 34-36 [La Calle and Cherchell, Al- geria; lectotype herein selected MHNG (Fig. 51)]. Non Mittré, 1841. Stimpson, Alexia micheli var. triplicata Bourguignat, 1864; 141, pl. 8, figs. 37, 38 [La Calle, Algeria; lectotype herein selected MHNG (Fig. 52)]. Alexia algerica Bourguignat, 1864: 141, pl. 8, figs. 23-26 [Algeria; lectotype herein se- lected MHNG (Fig. 53)]; Kobelt, 1898: 128, pl. 17, figs. 18, 19. Alexia algerica var. quadriplicata Bourguig- nat, 1864: 142, pl. 8, figs. 27-30 [Algeria; lectotype herein selected MHNG (Fig. 54)]. Alexia loweana Pfeiffer, 1866: 145 [Madeira Island; location of type unknown]. Melampus myosotis (Draparnaud). Jeffreys, 1869: 106, pl. 4, fig. 2 [Voluta ringens Turton illustrated (Fig. 43), probably type material]. Alexia setifer Cooper, 1872: 153, pl. 3, figs. A1-A3, А5-Аб [San Francisco Bay, California; holotype ANSP 22513a (Fig. 55). Alexia setifer var. tenuis Cooper, 1872: 154, pl. 3, fig. A4 [San Francisco Bay, Califor- nia; holotype ANSP 22513b (Fig. 56)]. Alexia (Auricula) myosotis var. hiriarti Follin & Bérillon, 1874: 88 [Biarritz lighthouse; lectotype herein selected MNHNP (Fig. 57). Alexia setigera Cooper. Pfeiffer, 1876: 368; Fénaux, 1939: 43 [error for setifer]. Auricula (Alexia) meridionalis Brazier, 1877: 26 [Port Adelaide, South Australia; holo- type ANSP 22506a (Fig. 58)]. Auricula watsoni Wollaston, 1878: 269 [Ma- deira; lectotype herein selected BMNH 1895.2.2.411 (Fig. 59)]. Auricula watsoni scrobiculata Wollaston, 1878: 269 [Salvages Islands (Madeira); lectotype herein selected BMNH 1895. 2.2.417 (Fig. 60)]. Auricula bicolor var. subarmata Wollaston, 1878: 466 [Lanzarote (Canary Islands); location of type unknown]. Auricula (Alexia) denticulata (Montagu). Fisch- er, 1878: 309-312. Alexia setifera Cooper. Nevill, 1879: 226 [un- justified emendation of setifer]. Alexia borealis Say Cooper. Nevill, 1879: 227. Alexia hiriarti Follin 8 Bérillon. Locard, 1882: 183. Alexia biasoletina (Küster). Locard, 1882: 183 [misspelling of biasolettiana]. Alexia ciliata (Morelet). Locard, 1882: 184; Kobelt, 1898: 129, pl. 17, fig. 20. Tralia (Alexia) myosotis (Draparnaud). Dall, 1885: 277; Dall, 1889: 92, pl. 52, fig. 9. 196 MARTINS Tralia (Alexia) myosotis var. ringens (Turton). Dall, 1885: 278. Tralia (Alexia) myosotis forma junior Dall, 1885: 278 [new name for Auricula ciliata Morelet and Alexia setifer Cooper]. Alexia cossoni Letour neux 8 Bourguignat, 1887: 130 [Gabès and Cheiba, Cape Bon, Tunisia; lectotype herein selected MHNG (Fig. 61). Alexia terrestris Letourneux & Bourguignat, 1887: 130 [El-Hamma, $ of Gabes, Tu- nisia; holotype MHNG (Fig. 62)]. Alexia globulus Bourguignat, in Letourneux & Bourguignat, 1887: 131 [Gabes, Tunisia; holotype MHNG (Fig. 63); on museum la- bel as Alexia ovum Bourguignat]. Alexia letourneuxi Bourguignat, т Letourneux 8 Bourguignat, 1887: 131 [Mandara, near Alexandria, Egypt, and Djerba Is- land, Tunisia; lectotype herein selected MHNG (Fig. 64)]. Alexia pechaudi Bourguignat, ín Letourneux 8 Bourguignat, 1887: 132 [Macta near Oran and Majerda, Tunisia; holotype MHNG (Fig. 65)]. Alexia acuminata Morelet, 1889: 15, pl. 1, fig. 11 [Port Elizabeth, Cape Colony, South Africa; specimen marked “type” broken, lectotype herein selected BMNH 1893.2.4.838 (Fig. 66)]. Alexia pulchella Morelet, 1889: 15, pl. 1, fig. 10 [Port Elizabeth, Cape Colony, South Africa; lectotype herein selected ВММН 1911.8.8.39 (Fig. 67)]. Alexia armoricana Locard, 1891: 132 [west coast of France; lectotype herein se- lected MNHNP (Fig. 68)]. Alexia exilis Locard, 1893: 62 [Le Croisic, Loire-Inférieure; Porquerolles (France); herein restricted to Porquerolles; lecto- type herein selected MNHNP (Fig. 69)]. Alexia parva Locard, 1893: 62 [Le Croisic, Loire-Inférieure (France); lectotype here- in selected MNHNP (Fig. 70)]. Alexia ringicula Locard, 1893: 62 [Arrdudon, Morbihan (France); lectotype herein se- lected MNHNP (Fig. 71)]. Auricula (Alexia) myosotis Pelseneer, 1894a: 73, figs. [anatomy]. Alexia bicolor (Morelet). Kobelt, 1898: 134, pl. 24, fig. 3. Alexia vespertina (Morelet). Kobelt, 1898: 135, pl. 24, fig. 4. Alexia (Kochia) oranica Pallary, 1900: 240, pl. 6, figs, 2, 2a [Oran, Tunisia; lectotype herein selected MNHNP (Fig. 72)]. Draparnaud. 195-208 Alexia myosotis marylandica Pilsbry, 1900a: 40 [Mouth of St. Leonards Creek, Patux- ent River, Maryland; lectotype by Baker (1964) ANSP 22483a (Fig. 73)]; C.W. Johnson, 1934: 159. Alexia myosotis bermudensis Pfeiffer. Pilsbry, 1900b: 504. Alexia oranica Pallary. Kobelt, 1901: 280, pl. 31, figs. 8, 9. Alexia bidentata Montagu forma americana Kobelt, 1901: 312, pl. 33, figs. 1, 2 [Ber- muda; type Senckenberg Museum, Frankfurt-am-Main (not seen)]. Myosotella myosotis (Draparnaud). Montero- sato, 1906: 126. Phytia myosotis var. bermudensis (H. & A. Adams). Peile, 1926: 88. Phytia myosotis (Draparnaud). Ellis, 1926: 96, pl. 2, fig. 3, pl. 5, fig. 49; Germain, 1931: 560, text figs. 295, 296, pl. 18, figs. 535, 536; McMillan, 1947: 264; McMillan, 1949: 67; M. Smith, 1951: 145, pl. 55, fig. 3, pl. 71, fig. 9; McMillan, 1968: 165; Climo, 1982: 43-48, fig. 1, A-L. Alexia (Myosotella) myosotis (Draparnaud). Thiele, 1931: 466. Phytia myosotis myosotis (Draparnaud). Winckworth, 1932: 238. Phytia myosotis denticulata (Montagu). Winckworth, 1932: 238. Alexia myosotis myosotis (Draparnaud). C.W. Johnson, 1934: 159. Alexia myosotis var. varicosa Fénaux, 1939: 44, pl. 1, fig. 3 [Provence, France; type probably in Fenaux’s collection, Ecole des Mines, Paris]. Alexia subflava Fénaux, 1939: 45, pl. 1, fig. 9 [Bermuda; type in Fénaux's collection, Ecole des Mines, Paris (not seen)]. Phytia bermudensis (H. & A. Adams). Morri- son, 1951b: 10. Phytia myosotis marylandica (Pilsbry). Morri- son, 1951b: 10; Burch, 1960a: 182 [chromosomes]. Phytia myosotis borealis (Conrad). Morrison, 1951b: 10. Ovatella myosotis (Draparnaud). Meyer, 1955: 1-43, pls. 1, 2 [anatomy, taxon- omy, life history]; Morton, 1955b: 119- 131, figs. [anatomy, life history]; Morton, 1955c: 127-168 [anatomy, taxonomy, evolutionary relationships]; Bousfield, 1960: 14, pl. 1, fig. 10; Coomans, 1962: 90; Kensler, 1967: 391-406 [ecology]; Jacobson & Emerson, 1971: 65, text fig.; Baranowski, 1971: 143; Abbott, 1974: 334, fig. 4103; Emerson 8 Jacobson, WESTERN ATLANTIC ELLOBIIDAE 197 1976: 192, pl. 26, fig. 28; Hubendick, 1978: 1-45 [taxonomic relationships); Morrell, 1980: 208-209; Rehder, 1981: 650, fig. 232; Jensen 4 Clark, 1986: 458, figured. Ovatella (Myosotella) myosotis (Draparnaud). Zilch, 1959: 73, fig. 236; Cesari, 1973: 181-210 [taxonomy, distribution, ecol- ogy]; Giusti, 1973: 124, figs. 4 A-N, pl. 2, figs. 1-4, pl. 3, figs. 1-3; Giusti, 1976; Cesari, 1976: 3-19, 5 pls. [taxonomy, anatomy, polymorphism]; Martins, 1978: 24, pl. 3, figs. 4, 4a, 4b, pl. 4, figs. 4, 4a, 4b, pl. 5, figs. 5, 6, D; Martins, 1980: 1-24, pl. 2, figs. f-o [habitat]. Ovatella (Alexia) myosotis (Draparnaud). Rus- sell-Hunter & Brown, 1964: 134. Ovatella myosotis bermudensis (H. 8 A. Ad- ams). Abbott, 1974: 334 [fig. 4105, erro- neously stated to be Microtralia occiden- talis, appears to be Myosotella myosotis from Bermuda]. Description: Shell (Figs. 41-77) to 12 mm long, oval-elongate, fragile to somewhat solid, commonly pale yellow to purplish red, rarely whitish. Spire high, with as many as eight somewhat convex whorls; first three whorls of teleoconch with spiral rows of pits, becoming fewer as spire progresses (Figs. 76, 77); row of hairs, in juveniles, anterior to spiral rows of pits (Fig. 77). Body whorl about 70% of shell length, smooth except for faint, irregularly spaced growth lines. Aperture about 80% body whorl length, oval-elongate, anteriorly rounded; inner lip with small, very oblique and somewhat twisted, white col- umellar tooth; anterior parietal tooth stron- gest, white, of variable thickness and perpen- dicular to columellar axis; usually one, rarely none, sometimes as many as four posterior parietal teeth that gradually become smaller posteriorly; outer lip sharp, often weakly re- flected in gerontic specimens, commonly with one, sometimes with as many as six whitish tubercles. Partition of inner whorls only in body whorl (Fig. 75). Protoconch as in genus. Animal grayish white to yellowish brown; neck usually darkly pigmented; tentacles subcylindric, darker than neck; rudimentary anterior tentacles present; foot not trans- versely divided, yellowish; mantle skirt gray- ish with dark spots. Radula (Figs. 78-80) having formula (20 + 11 +1+ 11 +20) x 80. Width of central tooth base twice that of lateral teeth, with central emargination, anterior portion of arms some- what sinuous; crown small, posteriorly de- pressed, unicuspid; mesocone triangular, somewhat rounded. Lateral teeth eight to 13; base quadrangular, elongate, oblique, with rounded lateral prominence over anterior third; crown cuneiform, about half length of base, posteriorly rounded. Marginal teeth 17 to 25; base becoming reduced anteriorly, projecting and square posteriorly; crown pointing medially, bicuspid; endocone some- what smaller than mesocone. Stomach with anterior membranous cham- ber, median muscular gizzard and posterior membranous gastric caecum (Fig. 81). Reproductive system (Fig. 82) with ovotes- FIGS. 41-60. Myosotella myosotis (Draparnaud). (41) Voluta denticulata Montagu, lectotype (RAMM 4100), Devon, England, sl 8.5 mm. (42) Voluta ringens Turton, lectotype (USNM 859011), British Isles, sl 8.4 mm. (43) Voluta ringens Turton, figured in Jeffreys’ British Conchology, pl. 98, fig. 29 (USNM 67947), sl 8.5 mm. (44) Voluta reflexa Turton, holotype (USNM 55370), British Isles, sl 9.2 mm. (45) Carychium personatum Michaud, lectotype (MNHNP), Boulogne, France, sl 6.5 mm. (46) Auricula reflexilabris Orbigny, lectotype (BMNH 1854.12.4.242), Lima, Peru, sl 9.0 mm. (47) Auricula ciliata Morelet, lectotype (BMNH 1893.2.4.831), Portugal, sl 7.8 mm. (48) Alexia bermudensis H. & A. Adams, lectotype (ВММН 1969105), locality not given [Bermuda], sl 7.6 mm. (49) Auricula bicolor Morelet, lectotype (BMNH 1893.2.4.822), Pico, Azores, sl 9.7 mm. (50) Auricula vespertina Morelet, lectotype (BMNH 1893.2.4.825), Area [Areia] Larga, Pico, Azores, sl 7.8 mm. (51) Alexia micheli Bourguignat, lectotype (MHNG), La Calle, Algeria, sl 9.2 mm. (52) Alexia micheli var. triplicata Bourguignat, lectotype (MHNG), La Calle, Algeria, sl 8.0 mm. (53) Alexia algerica Bourguignat, lectotype (MNHG), Mostaghanem, Algeria, sl 9.4 mm. (54) Alexia algerica var. quadriplicata Bourguignat, lectotype (MHNG), Cape Caxine near Alger, Algeria, sl 6.8 mm. (55) Alexia setifer Cooper, holotype (ANSP 22513a), San Francisco, California, sl 7.1 mm. (56) Alexia setifer var. tenuis Cooper, holotype (ANSP 22513b), San Francisco, California, sl 6.4 mm; Baker (1964) gave the length as 7.7 mm, which does not match the length of the shell marked as type. (57) Alexia (Auricula) myosotis var. hiriarti Follin & Bérillon, lectotype (MNHNP), Biarritz lighthouse, France, sl 10.1 mm. (58) Auricula (Alexia) meridionalis Brazier, holotype (ANSP 22506a), Port Adelaide, South Australia, sl 8.3 mm. (59) Auricula watsoni Wollaston, lectotype (BMNH 1895.2.2.411), Madeira, sl 8.1 mm. (60) Auricula watsoni scrobiculata Wollaston, lecto- type (BMNH 1895.2.2.417), Salvages Islands [Madeira], sl 7.5 mm. MARTINS FIGS. 41-60. 199 WESTERN ATLANTIC ELLOBIIDAE FIGS. 61-77. 200 MARTINS FIGS. 78, 79. Myosotella myosotis, radular teeth, Newport River, North Carolina, sl 5.1 mm. Scale 100 um. tis light colored, between lobes of digestive gland; hermaphroditic duct long, dilated, convoluted; pallial gonoduct hermaphroditic along its entire length; anterior mucous gland and prostate gland cover entire length of spermoviduct; bursa duct as long as sper- moviduct, emptying near vaginal opening; spermatheca spherical. Penis short, thick; length; connectives of visceral nerve ring long; right pleuroparietal connective twice as long as left one; left parietovisceral connec- tive longer than right one, sometimes with ganglionic swelling on anterior third, from which internal pallial nerve originates; rudi- mentary osphradial ganglion arising from pneumostomal nerve. associated vas deferens adhering to penis. Nervous system (Fig. 83) with cerebral commissure one and one-half times width of cerebral ganglion; left and right cerebropleu- ral and cerebropedal connectives of same Remarks: Myosotella myosotis is an ex- tremely variable species especially known from European coasts. Within one population the shape of the shell can vary from slim and FIGS. 61-77. Myosotella myosotis (Draparnaud). (61) Alexia cossoni Letourneux & Bourguignat, lectotype (MHNG), Lagune de l'oued Cheiba, (Cap Bon), Tunisia, sl 7.3 mm. (62) Alexia terrestris Letourneux & Bourguignat, holotype (MHNG), El Hamma, $ of Gabes, Tunisia, sl 5.4 mm. (63) Alexia globulus Bourguig- nat, holotype (MHNG), Gabés, Tunisia, sl 5.7 mm. (64) Alexia letourneuxi Bourguignat, lectotype (MHNG), Mandara, near Alexandria, Egypt, sl 5.7 mm. (65) Alexia pechaudi Bourguignat, holotype (MHNG), La Mactra, near Oran, Tunisia, sl 5.2 mm. (66) Alexia acuminata Morelet, lectotype (BMNH 1893.2.4.838), Natal, sl 5.0 mm. (67) Alexia pulchella Morelet, lectotype (BMNH 1911.8.8.39), Port Elizabeth, South Africa, sl 5.0 mm. (68) Alexia armoricana Locard, lectotype (MNHNP) Brest, Finisterre, France, sl 5.1 mm. (69) Alexia exilis Locard, lectotype (MNHNP), Porquerolles, France, sl 6.1 mm. (70) Alexia parva Locard, lecto- type (MNHNP), Le Croisic, Loire-Inférieure, France, sl 5.0 mm. (71) Alexia ringicula Locard, lectotype (MNHNP), Arrdudon, Morbihan, France, $1 5.0 mm. (72) Alexia (Kochia) oranica Pallary, lectotype (MNHNP), Oran, Tunisia, $1 6.0 mm. (73) Alexia myosotis marylandica Pilsbry, lectotype (ANSP 22483a) Patuxent River, Maryland, sl 8.0 mm. (74) Jamestown, Rhode Island, sl 6.7 mm. (75) Old Road, Shelly Bay, Bermuda, sl 6.3 mm. (76) Lateral view of spire and protoconch, Jamestown, Rhode Island. (77) Top view of spire and protoconch, Sao Miguel, Azores. Scale 1 mm. WESTERN ATLANTIC ELLOBIIDAE 201 il (E 10L 111 1M 2M 14M 15M Ола | J FIG. 80. Myosotella myosotis, radula, Beaufort, North Carolina. Scale 10 um. FIG. 81. Myosotella myosotis, stomach, Bermuda. Scale 1 mm. high spired to globose, and the color ranges from pale yellow to purplish red (Martins, per- sonal observations in Bermuda and Azores). Similar variability occurs in the apertural mor- phology, in which the number of parietal and outer Пр teeth can vary considerably. It was the variability of these characters that evoked most of the many names given to this spe- cies. According to Locard (1895), Drapar- naud (1801) was aware of this variability when he described Auricula myosotis, be- cause the 113 syntypes included examples of the dentate form later described by Mon- tagu (1803) as Voluta denticulata. Michaud (1831), who completed Draparnaud's work, described Montagu's form as Carychium personatum (Fig. 45). Even quite recently the question of the con- specificity of the European forms included in the genus Myosotella has been extensively FIG. 82. Myosotella туозой$, reproductive sys- tem, Bermuda. A-C, transverse sections and their locations. Scale 1 mm. FIG. 83. Myosotella myosotis, central nervous sys- tem, Bermuda. Scale 1 mm. debated. Germain (1931) accepted two Eu- ropean species, Phytia myosotis (Drapar- naud), with only one posterior parietal tooth, and Phytia denticulata (Montagu), with a heavily dentate aperture. Winckworth (1932) treated both as subspecies of Phytia. Watson (1943) noted the differences between the two forms but added that there are intermediates. The slight differences he found in radular fea- tures could be explained by the different sizes of the specimens studied, and the differences in shell morphology could be attributed to the 202 MARTINS more saline habitat of Phytia denticulata. Fénaux (1939), after examining hundreds of specimens from a stretch of coast between Toulon and Agde, southern France, found al- most all the “species” described from Eu- rope. Cesari (1973) was inclined to treat Ova- tella denticulata as a synonym of Ovatella myosotis but later (1976), as did Watson (1943), he considered the case of Ovatella denticulata unclear pending a definite ana- tomical comparison. Considering the high de- gree of shell variability of Myosotella myosotis (sensu lato), a wide range of anatomical vari- ability is to be expected. The same condition is found in the Western Atlantic Melampus (M.) bidentatus, which exhibits high variability in shell morphology as well as in anatomical characters (see the remarks under that spe- cies). On the basis of the great range of vari- ability in shell morphology, | think it justifiable to consider Myosotella myosotis as the only species living in Europe and North Africa. The names Voluta ringens Turton, Voluta reflexa Turton, Auricula tenella Menke, Carychium personatum Michaud, Auricula botteriana Philippi, Alexia letourneuxi Bourguignat, Al- exia armoricana Locard, Alexia ringicula Lo- card and Alexia oranica Pallary all pertain to the dentate morph of Myosotella myosotis. | have concluded previously (Martins, 1978, 1980) that Auricula vespertina Morelet and Auricula bicolor Morelet from the Azores are conspecific with Myosotella myosotis. Upon inspection of the type material of Wol- laston's Auricula watsoni and Auricula wat- soni scrobiculata from Madeira (Figs. 59, 60) | also include them in the synonymy of Муо- sotella myosotis. Shell morphology can be affected by envi- ronmental factors. The Bermudian speci- mens (Figs. 48, 75) are larger and thicker than the specimens from New England (Fig. 74), but similar to those | found in North Carolina [Alexia myosotis marylandica Pilsbry (Fig. 73)] and in the Azores. The thickening and en- hanced color of the shell seen nearer the warm regions is also observed in Melampus (M.) bidentatus (see the remarks under the species), and should be considered an envi- ronmentally determined character of little taxonomic value. The names Alexia myosotis marylandica Pilsbry and Alexia bermudensis H. 8 A. Adams, the latter considered a sub- species by Abbott (1974), are obviously only morphological variations of Myosotella myo- $015. Alexia subflava Fénaux, also from Ber- muda, was based upon a form with unusual apertural features, but it is clearly within the range of variation of Myosotella myosotis, and it too must be considered synonymous. Myosotella myosotis (Draparnaud) and Au- riculinella (L.) bidentata (Montagu, 1801) are often confused. The latter was erroneously reported from America and Bermuda. Dall's (1885) statement that Melampus (Leuconia) bidentatus (Montagu) [= Auriculinella (Leuco- phytia) bidentata (Montagu)] lived in America was based on Binney's remarks about Myo- sotella myosotis. Binney (1859: 174), after de- scribing the animal, noted that it differed from Н. 8 A. Adams’ illustration of the animal of Alexia denticulata (1855b: pl. 82, fig. 5). He mentioned that, “from the exterior of the an- imal there appears no difference between it and Melampus bidentatus.” Apparently Dall (1885) wrongly concluded that the species in question should also have the foot trans- versely divided, a characteristic shared by Melampus (M.) bidentatus and Auriculinella (L.) bidentata, but not by Myosotella myoso- tis. From Dall’s description of Melampus (Leuconia) bidentatus (Montagu) it is clear that he was confused about differences be- tween the shell of Myosotella myosotis and that of Auriculinella (L.) bidentata. Kobelt (1901: 283) briefly described a sup- posedly biplicate variation of Alexia bermu- densis H. & A. Adams, to which he later (p. 312, caption of pl. 33, figs. 1, 2) gave the name Alexia bidentata Montagu forma amer- icana. The illustration hardly differs from that of Alexia bermudensis (pl. 33, fig. 3), which H. & A. Adams (1855a: 33) described as having “columella biplicata” (Figs. 48, 75). As noted above, Myosotella myosotis varies greatly in apertural morphology, especially in the con- spicuousness of the posterior parietal teeth. Alexia bidentata Montagu forma americana Kobelt is just a phenotypic variant of Myoso- tella myosotis. Three other names were applied to North American specimens. Kuster (1844) de- scribed and figured an Auricula sayi. п the words of Binney (1859: 178), “Kúster's figure represents no known American shell. There exists, however, a strong resemblance be- tween it and his figure of Alexia myosotis.” Pfeiffer (1856a) tentatively assigned Kúster's name to Marinula, and compared it with Au- ricula infrequens C. B. Adams, 1852, from Panama. After examining C. B. Adam’s type material | disagree with Pfeiffer’s сотрап- son. The dentition shown in Kúster's illustra- tion resembles that of Creedonia succinea, WESTERN ATLANTIC ELLOBIIDAE 203 although the shell is too globose and acumi- nate to be referred with certainty to that spe- cies. In view of the conflicting diagnostic characters derived from the illustration and from the description given by Kúster, | con- sider Auricula sayi Kúster a nomen dubium. The other two problematic names are Melampus borealis Conrad and Melampus turritus (Say MS) Binney, both from Rhode Island, and both undoubtedly conspecific with Myosotella myosotis. The former was misidentified by Pfeiffer (1856a) who, based upon misidentified specimens from Georgia in the Cuming collection, wrongly assigned them to Melampus bidentatus Say, var. y bo- realis Conrad. Pfeiffer’s description of this variety (1856a: 46) mentioned an “outer lip with a white callus, regularly with 6-10 pli- сае.” Меатриз (D.) floridanus, another Georgian species, has as many as ten riblets inside the outer lip, but it is doubtful that Pfeiffer, who had introduced the latter spe- cies 11 pages before, would have confused it with Myosotella myosotis. Melampus (M.) bi- dentatus normally exhibits the sort of dentic- ulation on the outer lip mentioned by Pfeiffer, but this feature never has been found in Му- osotella myosotis. One must conclude that Pfeiffer relied on misidentified specimens when he identified his variety with Ме/атри$ borealis Conrad. The description of Меат- pus turritus, found by Binney (1859) among Say's unpublished manuscripts, was pub- lished by that author only to provide addi- tional information about Myosotella myosotis. Myosotella myosotis can be differentiated conchologically from Melampus (M.) bidenta- tus and Melampus (D.) floridanus, with which it associates, by its less globose shape, by its shorter, anteriorly rounded and wider aper- ture, and by its lack of riblets within the outer lip. № differs from Creedonia succinea by its pointed spire and by the dentition of its inner lip, which in Creedonia has a very strong, posteriorly located parietal tooth. Some dwarf, thin-shelled forms of Tralia (T.) ovula can be confused with the solid, deeply col- ored forms of Myosotella myosotis. This fact probably accounts for Dall's (1885) report of the latter species from Jamaica. The nonmu- cronate apex, the sinuous outer lip and the white, equidistant teeth of the inner lip of Tra- lia constitute sufficient diagnostic characters, however. Gerontic individuals of Myosotella myosotis have a weakly reflected outer lip, a feature that led early authors to insist on in- cluding this species in the genus Auricula. Detailed studies of life history and anatomy were published simultaneously by Morton (1955b) and Meyer (1955). Habitat: Myosotella myosotis lives in salt marshes and adjacent areas, preferring piles of rocks and detritus above the high-tide mark. In Bermuda this species commonly lives under piles of rocks, farther onto land than any other halophilic ellobiid, a situation also observed in the Azores (Martins, 1980). Range: Myosotella myosotis is well known as a Mediterranean and Eastern Atlantic spe- cies. “Species” very similar to Myosotella myosotis have been described from extra- European shores, such as Orbigny’s (1837) Auricula reflexilabris (Fig. 46) from the Pacific coast of South America, Cooper's (1872) Al- exia setifer (Figs. 55, 56) from California, Bra- zier's (1877) Auricula (Alexia) meridionalis (Fig. 58) from southern Australia and More- let's (1889) Alexia acuminata (Fig. 66) and Al- еха pulchella (Fig. 67) from South Africa. Hanna (1939) included Cooper's species among the “Exotic Mollusca in California” and | concur with Paulson (1957) in conclud- ing that the Californian Alexia setifer Cooper does not differ from eastern American or Eu- ropean specimens. Alexia setifer Cooper, as well as Auricula ciliata Morelet, were named on the basis of the presence of hairs on the spire of juveniles. Clark (1855) first noted that this condition occurs in Myosotella myosotis. Taking into consideration the well-docu- mented morphological plasticity shown by Myosotella myosotis, | concur with Climo (1982) in synonymizing Brazier's species, as well as the others just mentioned. The wide range of Myosotella myosotis is attributed to its estuarine and supralittoral habits; most probably the animals were car- ried about in ballast or as egg masses laid on deck equipment and cargo that came in di- rect contact with marsh communities (Climo, 1982). In the Western Atlantic this species occurs from Halifax, Nova Scotia (Bousfield, 1960), to Georgia, Bermuda and Cuba (Fig. 84). The Cuban specimen at the USNM (383711) should be classed as a spurious report until further confirmation because it is far from the range of the species in the Western Atlantic. Dall's герой of this species from Jamaica 1$ doubtful and it has not been confirmed by recent collections. Stimpson (1851), followed by Binney (1859, 1865), Verrill (1880) and Dall (1885), remarked that this species probably 204 MARTINS 45 30 FIG. 84. Myosotella myosotis, geographic distribu- tion, Western Atlantic. Open circle, locality from literature. was introduced to the eastern coast of North America. Specimens Examined: MAINE: Castine (MCZ 4180; USNM 492501); Portland (MCZ uncatalogued; USNM 24865, 73394); New- castle (MCZ 34005). NEW HAMPSHIRE: Fabian Point, Great Bay, Newington (R.B.). MASSACHUSETTS (ANSP 22508; USNM 27740, 27913): Manchester (USNM 39800); Marblehead (MCZ 199478; USNM 492503); Boston (MCZ uncatalogued; USNM 41240); Woods Hole (ANSP 357609; MCZ 34004; USNM 158953, 525155); New Bedford (ANSP 22494; MCZ uncatalogued; USNM 139801). RHODE ISLAND (MCZ 34003; USNM 539238): Newport (MCZ 68946, 163167; USNM 39799, 67730); Warren (ANSP 60355); Maple Creek, Jamestown (A.M.); Wickford (MCZ 294645). CONNECTI- CUT: Branford (MCZ 34847; USNM 492502); New Haven (USNM 83471). NEW YORK: New York Harbor (USNM 492485); Cold Spring Harbor (MCZ 294167); Staten Island (MCZ 56738, 61847, 119477; USNM 59729, 407787, 492500). NEW JERSEY: Cape May (MCZ uncatalogued). MARYLAND: Patuxent River (ANSP 22483, 359154; USNM 492486); St. Leonards Creek, Patuxent River (ANSP 60971; USNM 465806); Crisfield (USNM 618924). VIRGINIA: Mollusk (USNM 791448); Watts Bay (USNM 701628); Chincoteague Is- land (MCZ uncatalogued); Fisherman’s Island (USNM 422292); Norfolk (USNM 637142). NORTH CAROLINA: mouth of Newport River, S of Beaufort (A.M.) SOUTH CAROLINA: Charleston (MCZ uncatalogued); McClellan- ville (USNM 663059). GEORGIA: Isle of Hope (USNM 663053); Thunderbolt (USNM 663055). BERMUDA (ANSP 48594, 48595, 48596, 62743, 78217, 79937, 85588; MCZ 8972, 9971, 24407, 24408, 24409, 294163, 294166; USNM 6529a, 6537, 94436, 94437, 101401, 151271, 250298a, 492487, 492488, 482490): Hamilton (USNM 152134, 171941); Hamilton Beach (MCZ uncatalogued); Fairy- land (ANSP 99074; USNM 208070); Gibbet Island (MCZ 294162); Flatts (ANSP 88572; MCZ 294164; USNM 1719340); Shelly Bay (MCZ 294165; USNM 492489); Old Road, Shelly Bay (A.M.); N of Shelly Bay Beach (A.M.); S of Coney Island (R.B.); Ferry Reach Park (R.B.); N of Long Bird Bridge (A.M.); Castle Harbour (ANSP 143320); Cooper’s Is- land (ANSP 131644); Spitall Pond (A.M.); Hungry Bay (ANSP 88580; USNM 171947; A.M.); Paget (USNM 714209); W of Somerset Bridge (A.M.); S of Ely’s Harbour (A.M.); Man- grove Bay (A.M.). CUBA: Los Canos, Guan- tanamo (USNM 383711). Genus Laemodonta Philippi, 1846 Laemodonta Philippi, 1846: 98. Type species by monotypy: Laemodonta [Auricula] striata (Philippi, 1846) [= Pedipes octan- fracta Jonas, 1845]. Non Martens, 1824, nec Anton, 1839. Laimodonta Bronn, 1847: 4 [nomen nudum, fide Herrmannsen, 1852]. Non “Nuttall” H. & A. Adams, 1855a. Plecotrema H. & A. Adams, 1853: 120. Type species by original designation: Ple- cotrema typica H. & A. Adams, 1853. Bullapex Haas, 1950: 199. Type species by monotypy: Laemodonta (Bullapex) cubensis (Pfeiffer, 1854). Description: Shell to 9 mm long, oval-conic, solid, sometimes hirsute. Umbilicus present. Spire moderately high, sculptured with more or less marked spiral cords. Body whorl 70- 75% shell length, sometimes truncate at base, with same sculpture as spire. Aperture about 70% length of body whorl, oval-elon- WESTERN ATLANTIC ELLOBIIDAE 205 gate, narrow; inner lip with three subequal teeth, one oblique columellar tooth and two parietal teeth; outer lip thickened, with one to three teeth about same size as parietal teeth. Protoconch smooth, globose, prominent. Remarks: Philippi (1846: 98), following the description of his Auricula striata, noted, “Laemodonta striata Adams (ubi?). Bronn placed under this name this species from Sandwich Islands [Hawaii]. The name Lae- modonta appeared as a nomen nudum in a sales catalogue of shells prepared by Bronn (1847, fide Sykes, 1894), who had sent the shells to Philippi. Although Laemodonta Phil- ippi, 1846, was introduced in synonymy, this name must be accepted as valid according to the ICZN, Article 11, (а). Thiele (1931) and Zilch (1959) used Laemodonta and since 1961 the name has been universally ac- cepted (Clench, 1964; Franc, 1968; Abbott, 1974; Hubendick, 1978; Kay, 1979). The names Laimodonta and Laemodonta have been confused in many instances, the second being taken wrongly for a misspelling of the first. Laimodonta (Nuttall MS) H. 4 A. Adams, 1853, was introduced for a group of shells different from those assigned to Lae- modonta Philippi. Often credit was given er- roneously to Nuttall for the introduction of Laimodonta. It appears, however, that Nuttall never published the name (Sykes, 1894). Nevill (1879) considered “Laimodonta Nut- tall” [= emendation of Laemodonta Philippi] and Laimodonta H. 8 A. Adams to be differ- ent taxa and Ancey (1887) introduced A/- lochroa to replace H. 8 A. Adams” suppos- edly preoccupied name. Sykes (1894), apparently unaware of Апсеу’$ introduction, also stated that Laimodonta H. 8 A. Adams was preoccupied, not by Philippi’s (1846) or by Bronn's (1847) names, which he consid- ered undescribed, but by Laimodon Gray, 1841, a genus of birds. Sykes proposed the new name Enterodonta. Laimodon Gray, 1841, cannot be considered a homonym of Laimodonta or of Laemodonta. According to the ICZN, Art. 32 (a), in spite of the fact that Philippi (1846) misspelled Bronn's name, his spelling is to be considered the correct orig- па! spelling. Laimodonta (Nuttall MS) H. & A. Adams, although not a homonym of Philippi’s name, has been abandoned in favor of Al- lochroa in important malacological works (Thiele, 1931; Zilch, 1959; Franc, 1968; Kay, 1979). A permanent solution to this problem would be the placement of Laimodonta ‘‘Nut- tall” H. & A. Adams on the Official List of Rejected Names in Zoology. Sykes (1894) and Hubendick (1956) in their monographs on Laemodonta preferred the name Plecotrema H. 8 A. Adams to Philippi’s name. Sykes recorded in his synonymy Lira- tor Beck, which he felt had not been properly introduced. The name Lirator, indeed, was used by Beck (1837) for an undescribed Melampus (Lirator) multisulcatus from Opara Island. Pfeiffer (1856a, 1876) tentatively iden- tified Beck's species with Laemodonta striata Philippi [= Pedipes octanfracta Jonas] and Hubendick (1956) accepted Pfeiffer's opinion without query. Because only circumstantial evidence connects Beck's names to a recog- nized species, however, one must conclude that both Lirator and Melampus (Lirator) mul- tisulcatus are nomina nuda. Hubendick (1956: 111) stated that Aguayo 8 Jaume (1947) had given “strong reasons for maintaining Plecotrema as the valid name of the genus.” In fact, what Aguayo & Jaume (1947: No. 132) had stated was, “in the im- possibility of deciding about the priority of Laemodonta [Philippi] 1847 [sic] over Laimo- donta [Bronn] 1847, and about the validity of Lirator [Beck], we have decided to use the genus Plecotrema [H. 8 A. Adams] as many modern authors do.” Later Hubendick (1978) used Laemodonta Philippi and relegated Ple- cotrema to synonymy. As stated above, Lae- modonta Philippi is now universally ac- cepted. The genus Laemodonta has been assigned erroneously to the subfamily Pedipedinae on the basis of the shell. Although described as “fairly uniform” (Hubendick, 1956: 110) or as a “convenient group” (Sykes, 1894: 241), the name Laemodonta is used now for a mixture of genera. Examination of the radula of “Lae- modonta” punctigera H. & A. Adams from Malaysia has a very wide, rounded meso- cone, typical of the Cassidula group, and that species will be assigned to another genus pending more research. H. & A. Adams (1853: 120) had noted that Plecotrema [= Laemodonta] was “а genus of small shells allied to Cassidula.” The present study of Laemodonta cuben- sis leads me to include this West Indian spe- cies within the Pythiinae on the basis of the radula and the reproductive and the nervous systems. Although preserved material of the type species was not available for anatomical comparisons, the remarkable resemblance of shell morphology, especially apertural denti- 206 MARTINS tion, of the West Indian species to Laemo- donta octanfracta (Fig. 31) has led me to con- clude that they are congeneric. № further anatomical studies indicate the necessity of taxonomic separation, the name Bullapex Haas 15 available for Laemodonta cubensis. | concur with Hubendick (1956) that the subgenus Bullapex Haas cannot be justified on the basis of shell characters alone. The true umbilicus of the West Indian species is often reduced to an umbilical depression, similar to the pseudoumbilicus mentioned by Haas (1950: 199) for Laemodonta clausa H. 8 A. Adams, 1853 [= Laemodonta octanfracta (Jonas)]. According to Hubendick (1956: 114) the inflated apex can be explained as an eco- logically influenced character and as such 1$ unreliable. The apex (protoconch) of Laemo- donta octanfracta, although not so prominent as that of Laemodonta cubensis, appears to be somewhat inflated (Fig. 92). According to Sykes (1894) and Hubendick (1956) the genus Laemodonta appeared in the Eocene. These earlier species, unlike those recorded from the Miocene, are smoother and more similar to the West Indian Laemodonta cubensis. The Miocene species have the heavy sculpture of the Indo-Pacific group. Hubendick concluded that the West Indian species and the Indo-Pacific group probably had common ancestors in the Tethys Sea. Habitat: Because most of the data available to me pertain to Laemodonta cubensis, de- scription of the soft parts and comments on the habitat are presented under that species. Range: Hubendick (1956) noted the discon- tinuous distribution of Laemodonta. Most of the representatives are from the western Indo-Pacific, with one species in the West Indies and Bermuda. The genus is not repre- sented in the Recent fauna of the Mediterra- nean or Eastern Atlantic. Laemodonta cubensis (Pfeiffer, 1854) Figs. 85-87, 89, 90, 93-101 Plecotrema cubensis Pfeiffer, 1854b: 153 [Cárdenas, Cuba; location of type un- known]; Pfeiffer, 1856a: 107; Pfeiffer, 1876: 348; Arango y Molina, 1880: 60; Crosse, 1890: 259; Kobelt, 1900: 236; Peile, 1926: 88; Aguayo & Jaume, 1947: 132; Hubendick, 1956: 111, text. fig. 1A, pl. 23, fig. 7 [distribution]. Plecotrema cubense Pfeiffer. Sykes, 1895: 245; Pilsbry, 1900b: 504, pl. 62, fig. 11. Laemodonta cubensis (Pfeiffer). Thiele, 1931: 464; Morrison, 1951b: 9; Morrison, 1958: 118-124 [habitat]; Abbott, 1974: 333, fig. 4101; Emerson & Jacobson, 1976: 190, pl. 26, fig. 20; Rehder, 1981: 650, fig. 222; Jensen 8 Clark, 1986: 458, figured. Laemodonta (Bullapex) cubensis (Pfeiffer). Haas, 1950: 199, pl. 22, figs. 6-8; Zilch, 1959: 69, fig. 225; Clench, 1964: 123, pl. 79 [taxonomy, distribution]; Vokes 4 Vokes, 1983: 60, pl. 31, fig. 18. Description: Shell (Figs. 85-87, 89, 90) to 3.5 mm long, oval, somewhat solid, pale yellow to light brown, hirsute. Narrow umbilicus or umbilical depression present. Spire moder- ately high, whorls as many as six and one- fourth, weakly convex, with two incised spiral lines near suture; first whorls of teleoconch with fine, compact spiral striae, crossed by very fine, somewhat irregular growth lines. Body whorl about 70% shell length, with in- cised spiral lines. Aperture oval; inner lip with three evenly spaced teeth; columellar tooth oblique toward base, moderately strong; an- terior parietal tooth smallest, oblique poste- riorly; outer lip sharp, with two conspicuous teeth, sometimes with one or two much smaller tubercles posteriorly. Partition of in- ner whorls occupying about three-quarters of body whorl (Fig. 86). Protoconch whitish, smooth, inflated, oblique or perpendicular to columellar axis of teleoconch (Figs. 89, 90). Animal whitish, translucent; tentacles long, thin, subcylindric, translucent; foot entire, rounded posteriorly. Pallial cavity long; kid- ney long and thin; mantle gland curved, tu- bular, empties near vaginal opening. Radula (Figs. 93-97) having formula [24 + (7 +7) + 1 +(7 + 7) + 24] x 100. Central tooth at about same plane as lateral teeth; base triangular, weakly emarginate anteriorly, with lateral prominences at mid-length; length of crown about half that of crown of lateral teeth, unicuspid; mesocone some- what sharp. Lateral teeth seven to ten; base weakly bent medially at posterior third, with lateral and medial prominences, the latter anteriormost; crown about half the length of the base, unicuspid, cuneiform. Transitional teeth five to nine; base similar to that of lat- eral teeth; crown bicuspid; endocone some- what shorter and weaker than mesocone. Marginal teeth 21 to 24; base becoming WESTERN ATLANTIC ELLOBIIDAE 207 FIGS. 85-92. Laemodonta, Ovatella. (85) L. cubensis (Pfeiffer), West Summerland Key, Florida, sl 2.9 mm. (86) L. cubensis, Grassy Key, Florida, sl 3.2 mm. (87) L. cubensis, Crawl Key, Florida, sl 0.84 mm. (88) O. aequalis (Lowe), Sáo Miguel, Azores, sl 9.3 mm. (89) L. cubensis, lateral view of spire and protoconch, Grassy Key, Florida. (90) L. cubensis, top view of spire and protoconch, Grassy Key, Florida. (91) O. aequalis, top view of spire and protoconch, Säo Miguel, Azores. (92) L. octanfracta (Jonas), top view of spire and protoconch, Hawaii. Scale 1 mm. shorter and wider than that of lateral teeth, developing lateral basal cusp covered by next tooth; crown tricuspid; endocone, me- socone and ectocone sharp, becoming subequal, with mesocone somewhat longer and stronger. Digestive system with salivary glands small, attaching posteriorly to esophagus through large area. Stomach (Fig. 98) tripartite; ante- rior portion corresponding to cardiac region, thin, dilated; mid-portion very muscular, with muscle also covering pyloric region; gastric caecum somewhat thin, not muscular, receiv- ing dilated, pouch-like posterior diverticulum 208 MARTINS FIGS. 93-96. Laemodonta cubensis, radular teeth, Grassy Key, Florida, sl 3.5 mm. Scale 50 um. C 1L2L 11L 1T 8T 1M 2M 3M 4M 11M 12M N | J KL) FIG. 97. Laemodonta cubensis, radula, Grassy Key, Florida. Scale 10 um. anteriorly, at boundary with mid-region. Di- gestive gland with two subequal lobes. Reproductive system (Fig. 99) with ovotes- tis between lobes of digestive gland; seminal vesicle of hermaphroditic duct convoluted at mid-length; pallial gonoduct hermaphroditic to the vaginal aperture; anterior mucous gland and prostate gland cover entire length of spermoviduct; bursa duct about same length as spermoviduct and empties just posterior to vaginal opening; bursa spherical. Penis short, thin; vas deferens adhering to penis; penial retractor about as long as penis, inserting on floor of pallial cavity. Nervous system (Fig. 100) with cerebral commissure short, about half width of cere- FIG. 98. Laemodonta cubensis, stomach, Ber- muda. Scale 1 mm. bral ganglion; right cerebropedal and cere- bropleural connectives two-thirds length of left ones; left and right connectives of vis- WESTERN ATLANTIC ELLOBIIDAE 209 FIG. 99. Laemodonta cubensis, reproductive sys- tem, Hungry Bay, Bermuda. A-C, transverse sec- tions and their locations. Scale 1 mm. FIG. 100. Laemodonta cubensis, central nervous system, Hungry Bay, Bermuda. Scale 1 mm. ceral nerve ring equal; pleuroparietal connec- tives very short; parietovisceral connectives very long, the right one longer; visceral gan- glion beneath tentacle retractor muscle. Remarks: Laemodonta cubensis is the only representative of the genus in the Atlantic. All other species live in the Indo-Pacific region. The West Indian species is somewhat iso- lated conchologically, owing to its thinner, much less sculptured shell. Kobelt (1900) and Thiele (1931) were not sure whether this spe- cies should even belong to this genus. As stated in the remarks under the genus, this species is included in Laemodonta because of the great similarity of its apertural morphol- ogy to that of the type species, Laemodonta striata (Philippi, 1846) [= Laemodonta octan- fracta (Jonas, 1845)]. The description of the radula of Plecotrema clausa H. 8 A. Adams, 1853, a junior synonym of the type species, given by Odhner (1925) is very similar to that of Laemodonta cubensis except for the lesser number of teeth in a row in the latter. In the original description of Laemodonta cubensis, Pfeiffer (1854b: 153) characterized the shell as “hispidula” [slightly hairy], a fea- ture also noticed by Haas (1950). A pilose shell appears also in some Pacific species; Garrett (1872) noted that his Plecotrema hir- suta [= Laemodonta molinifera (H. £ A. Ad- ams)] had short, curved hairs. The presence of a pallial gland was some- what unexpected in Laemodonta. This organ of unknown function was first noticed by Plate (1897) in Phytia scarabeus (Linnaeus) and observed later in Carychium tridentatum (Risso) (Morton, 1955b), in Cassidula labrella (Deshayes) (Renault, 1966) and in Ovatella aequalis (Lowe) (Martins, personal observa- tion). Carychium lives inland, frequently in the mountains and, although preferring humid environments such as forest leaf litter, it is obviously a terrestrial species. Pythia is also considered a terrestrial ellobiid because it lives in the upper fringe of mangroves. Ova- tella aequalis lives just above the high-tide limit and Laemodonta at or just below the high-tide mark. Information is not available concerning the precise habitat of Cassidula labrella. Morton (1955b, c) advanced the hy- pothesis that this was probably a case of par- allel evolution in response to some environ- mental parameter associated with terrestrial life. According to that same author, the pos- sible functions of the pallial gland range from help in forming egg cases to aid in keeping the body moist or secretion of bacteria-killing substances as a protective device while the animal is crawling. The presence of the pallial gland in two supposedly marine ellobiids de- папа$ a review of the hypotheses about the evolution and function of this organ. Laemodonta cubensis 1$ very distinct from all other West Indian mollusks because of its hirsute, oval shell and its apertural dentition. Its protoconch and juveniles are very similar to the protoconch and hirsute juveniles of the Macaronesian and western European Ova- 210 MARTINS 90 75 16 OI 45 30 FIG. 101. Laemodonta cubensis, geographic dis- tribution. Open circle, locality from literature. tella aequalis (Lowe) (Figs. 89, 90). The strength of the palatal tooth of the aperture (Figs. 85, 88), the similarity of the radular teeth and the presence о the pallial gland (personal observations) also suggest a generic relation- ship between Laemodonta cubensis (Pfeiffer) and Ovatella aequalis (Lowe). Habitat: Laemodonta cubensis lives at or just below the high-tide mark, aggregating under half-buried porous rocks, rotting wood and leaves, and among the roots of propagules, together with Pedipes, Blauneria, Microtralia and Creedonia. lt is common, along with Pe- dipes, in rocky areas, either under loose stones near the sediment, or in crevices in rock beds at about the high-tide mark. They prefer the part of loose stones that touches the sediment. Range: Bermuda; Captiva Island, on the western coast of Florida, south to the Florida Keys; Bahamas and Cuba, Jamaica south to Barbados; Mexico (Vokes & Vokes, 1983) (Fig. 101). Specimens Examined: FLORIDA: Third Ragged Key above Sand Key (USNM 462737, 614608); Key Largo (MCZ 235475); N of Tavernier Key, Key Largo (A.M.); S of Ocean Drive, Plantation Key (A.M.); Indian Key Fill, N of Indian Key Channel (A.M.); In- dian Key (USNM 462893, 492557); Long Key (A.M.); Grassy Key (A.M.); Crawl Key (MCZ 235469; A.M. Bonefish Key (ANSP 174978, 219861; МСУ 110178; USNM 599365); Knight Key (A.M.); Bahia Honda Key (ANSP 104109; MCZ 235472; USNM 492464); West Summerland Key (A.M.); Big Pine Key (ANSP 104110); Long Beach Drive and W end of Ko- hen Avenue, both Big Pine Key (A.M.); Little Torch Key (MCZ 235470); Big Torch Key (A.M.); Ramrod Key (MCZ 235471); Sugarloaf Key (ANSP 89559); Boca Chica Key (ANSP 104111; USNM 270348); Garden Key, Dry Tortugas (USNM 492509, 492522); Seminole Point (ANSP 105438); Captiva Island (ANSP 149410). BERMUDA (ANSP 62842, 293543; MCZ 24229; USNM 250298): Fairyland (ANSP 99078); N of Shelly Bay Beach (A.M.); North of Long Bird Bridge (A.M.); near St. Georges (ANSP 100821); Castle Harbour (ANSP 143321); Hungry Bay (A.M.); W of Somerset Bridge (A.M.); Ely's Harbour (A.M.); Mangrove Bay (A.M.). BAHAMAS ISLANDS: GRAND BAHAMA ISLAND: Hepburn Town, Eight Mile Rock (ANSP 375427, 375455); Bell Channel, Lucaya (ANSP 370709); Dead Mans Reef [Sandy Bevan's Cay] (ANSP 371224); Bahama Beach Canal (ANSP 371802); Silver Cone Canal (ANSP 372886); North Riding Point (ANSP 375563); West End (ANSP 368764); GREAT ABACO ISLAND: Mores Is- land (MCZ 116720); Sand Bank, Crossing Bay (MCZ 235474); S of Witch Point (MCZ 235478); Wilson City (MCZ 235479); North Hawksbill Creek (ANSP 370566); ANDROS ISLAND (ANSP 151873): Morgan's Bluff (А.М.); South Mastic Point (A.M.); Mangrove Cay (USNM 590609, 614605); Solomon Pond, Mangrove Cay (USNM 614606); First island off Mintie Bar, SE of South Bight (USNM 590610, 614607); NEW PROVI- DENCE ISLAND: Bar Point (A.M.); W of Rock Point (A.M.); W of Clifton Point (A.M.); E of Clifton Pier (A.M.); shore of Millars Road (A.M.); Malcolm Creek (A.M.); ELEUTHERA ISLAND: S of Rock Sound (MCZ 235473); EXUMA CAYS: NE coast, Hog Cay (MCZ 235476), Western End, Hog Cay (MCZ 235477). CUBA (ANSP 22544): near Habana (ANSP 130743); El Vedado, Habana (MCZ uncatalogued). JAMAICA: Falmouth (ANSP 397272); Robin's Bay (USNM 442000). PU- ERTO RICO: San Juan (R.B.). VIRGIN 1$- LANDS: ST. THOMAS (USNM 6427). LESSER ANTILLES: BARBADOS: off Laza- reto (USNM 502107). WESTERN ATLANTIC ELLOBIIDAE 211 Subfamily Pedipedinae Fischer & Crosse, 1880 Pedipedinae Fischer 4 Crosse, 1880: 5. Description: Shell to 11 mm long, globose to elongate. Spire low to high, with as many as six and one-half whorls. Body whorl 80- 90% of shell length. Aperture broad to nar- row; columellar teeth one or two; parietal teeth one or two; outer lip smooth, with one strong tooth or with internal axial, ribbed cal- losity (Pseudomelampus). Inner whorls re- sorbed except in Pedipes and Creedonia. Animal whitish; foot transversely divided (except in Microtralia), posteriorly tapered, tip rounded. Mantle skirt broad, fused posteri- orly. Pallial cavity not occupying entire body whorl; kidney white, long to broadly triangu- lar; pneumostomal glands white and anterior to kidney; anal gill well developed; mantle or- gan lacking. Radula having very variable formula. Cen- tral tooth slightly posterior to lateral teeth, unicuspid or tricuspid. Lateral teeth bicuspid. Transition to marginal teeth gradual. Marginal teeth with as many as five cusps. Digestive system with mandible broadly rectangular, composed of numerous longitu- dinal fibers. Salivary glands white, small, fusi- form. Digestive gland of two roughly equal lobes; anterior lobe empties into crop through wide anterior diverticulum, just be- fore crop enters stomach; posterior lobe empties into gastric caecum through poste- rior diverticulum. Stomach tripartite, middle section very muscular and with a caecum. Reproductive system with hermaphroditic duct not convoluted (except in Marinula s.s.), posteriorly dilated; anterior mucous gland and prostate gland extending over proximal half of spermoviduct; bursa duct emptying just posterior to female opening, at which vas deferens separates from vagina (except in Pseudomelampus and Leuconopsis). Penis thick, usually with more or less developed diverticulum, simple in Microtralia; vas defe- rens free, enters penis apically; penial retrac- tor short, attached to columellar muscle or to floor of pallial cavity. Nervous system with cerebral ganglia well developed; cerebropedal connectives about as long as cerebropleural connectives; pleu- roparietal connectives and parietovisceral connectives very short, somewhat longer in Leuconopsis. Remarks: Fischer & Crosse (1880) created the subfamily Pedipedinae for Pedipes, the only ellobiid genus then known to retain its inner whorls. On the basis of radular charac- ters Odhner (1925) added to the subfamily the genera Marinula and Plecotrema |= Laemo- donta], Thiele (1931) included Pseudomelam- pus and Leuconopsis and Morton (1955c) added Rangitotoa [= Microtralia]. In 1959 Zilch transferred Rangitotoa to the Melampinae and added Apodosis, which | consider a junior synonym of Leuconopsis. Abbott (1974) listed Microtralia within the Cassidulinae and erro- neously considered Ovatella [sensu Myoso- tella] to belong to the Pedipedinae. My anal- ysis Of nervous and reproductive systems leads to the inclusion of Microtralia in the Pe- dipedinae and to the removal of Laemodonta, Ovatella and Myosotella to the Pythiinae. A new genus, Creedonia, is here created upon the basis of the conchological and radular characters of Creedonia succinea (Pfeiffer, 1854), formerly placed in the genus Marinula. Creedonia does not resorb the inner whorls, its central and lateral radular teeth are broad with only a few in arow and the marginal teeth have several endocones but lack ectocones. The radulae in those species of Marinula stud- ied have numerous, very long, narrow lateral teeth and marginal teeth with one endocone and several ectocones (Figs. 163-168). The subfamily is best characterized ana- tomically. The nervous and reproductive sys- tems have very consistent patterns, whereas the shell and radula vary somewhat. The short connectives of the visceral ring cause the concentration of those ganglia. This fea- ture also exists in the Melampinae and sets these two groups apart from the remaining subfamilies, which have long visceral ring connectives. The cerebral ganglia in the Pe- dipedinae are proportionally much larger than the other ganglia. The reproductive sys- tem differs from that of the Melampinae by its acinose ovotestis, its unconvoluted seminal vesicle of the hermaphroditic duct, its longer spermoviduct, its junction of the bursa duct near the female opening, and its elaborate penial complex, sometimes with a long diver- ticulum. From the other three subfamilies it differs by its elaborate penial complex with free anterior vas deferens, and by its prostate and anterior mucous glands that cover only the posterior half of the spermoviduct. The genus Leuconopsis deviates in some anatomical features from the typical Pedipe- dinae pattern in that its visceral nerve ring is 212 MARTINS longer and its reproductive system is semi- diaulic, the vas deferens and vagina sepa- rating from the common spermoviduct half- way along the pallial gonoduct. The visceral nerve ring is not so long as that of any of the species belonging to the Pythiinae and Ello- biinae here studied and does not justify per se the exclusion of Leuconopsis from the Pedipedinae. The organization of the pallial gonoducts of Leuconopsis resembles that of Pseudomelampus (Martins, personal obser- vation) in that the vas deferens separates from the spermoviduct some distance before reaching the female opening, giving rise to a long, nonglandular vagina. The arrangement of the reproductive organs in Leuconopsis is, then, within the range of variation seen within the Pedipedinae. Shell and radular charac- ters also justify the inclusion of Leuconopsis within the Pedipedinae. The Pedipedinae are represented in the West Indies by Pedipes, Creedonia, Leu- conopsis and Microtralia, all of which are readily distinguishable on conchological characters. Pedipes 1$ globose, generally heavily sculptured, with two strong columel- lar teeth, a large oblique, posteriorly placed parietal tooth and a readily visible callous tooth on the outer lip, opposite the parietal tooth. Creedonia has an apertural configura- tion similar to that of Pedipes, but lacks the tooth on the outer lip and has an elongate, smooth and very thin shell. The minute Leu- conopsis has an elongate, thick shell that lacks the parietal and teeth on the outer lip. The shell of Microtralia is very thin and trans- parent and has a long, narrow aperture with the outer lip smooth inside, one columellar tooth and two anteriorly placed parietal teeth, of which the anterior is the strongest. Habitat: The Pedipedinae are a group of small species that live at or below the high- tide zone (Morton, 1955c; Martins, 1980), sometimes reaching the low intertidal area (Spencer, 1979). They live mainly in crevices and under half-buried rocks. Some (Cree- donia, Microtralia) live in the soft, black sed- iment of the high-tide region of mangroves, or in the spaces around the underground root system of propagules and adult plants. Range: Worldwide, warm and temperate ar- eas. In the Western Atlantic the Pedipedinae occur in Bermuda and from south Florida to the West Indies, Central America south to Brazil, Ascension Island and Tristan da Cunha (Connolly, 1915). Genus Pedipes Scopoli, 1777 Pedipes Scopoli, 1777: 392. Type species by subsequent designation of Gray (1847a): Pedipes afra (Gmelin, 1791) [= Pedipes pedipes (Bruguière, 1789)]. Carassa Gistel, 1847 [1850]: 555 [substitute name for Pedipes Scopoli]. Description: Shell to 6 mm long, globose, solid to fragile, white to dark brown. Spire low, with as many as five rapidly expanding convex whorls, with incised spiral grooves. Body whorl averaging 87% of shell length. Aperture about 70% length of body whorl, ovate, widely rounded at base; exposed por- tion of columella flat and depressed, with two strong columellar teeth, posterior one stron- ger; parietal tooth strongest, oblique; outer lip sharp, smooth or with thick, elevated tooth opposite parietal tooth. Inner whorls not resorbed. Protoconch with apex involute in first teleoconch whorls, smooth, translu- cent, yellowish to brown. Animal grayish; tentacles long, pointed, with transparent base, blackish toward tip. Radula with 150 to 450 teeth in a row; first 30 to 40 rows without marginal teeth. Central tooth at same level as lateral teeth; base long, slightly indentate anteriorly, wide and angular in first quarter, longitudinally de- pressed in the middle, becoming narrower posteriorly; crown as wide as base, falciform, with long mesocone. Lateral teeth as narrow as but longer than central tooth, laterally compressed; base irregularly thickened lon- gitudinally; crown falciform, laterally com- pressed; mesocone long; endocone half length of mesocone, pointed. Transitional teeth four to eight, with base becoming shorter, crown becoming wider; two or three endocones. Marginal teeth cteniform, with very thin, short base; wide rounded crown, connected to base by long neck bent up- wards; as many as six short endocones; me- socone somewhat stronger than endocones. Visceral mass coiled. Digestive gland yel- lowish; posterior diverticulum becoming pouch-like before entering anterior portion of gastric caecum. Stomach tripartite; anterior thin-walled, pouch-like section that receives crop on left and empties into intestine on right; middle thick-walled, very muscular giz- zard; posterior membranous, extensible cae- WESTERN ATLANTIC ELLOBIIDAE 213 cum that receives posterior diverticulum at junction with gizzard. Ovotestis acinose, em- bedded in posterior lobe of digestive gland; hermaphroditic duct dilated along most of its length; bursa spherical. Penis unevenly thick- ened, convoluted in middle. Cerebral com- missure long; connectives of visceral ring very short; penial nerve originating from right middle labial nerve. Remarks: The name Pedipes was first used by Adanson (1757), who gave a very detailed and accurate description of a species from Senegal later described as Pedipes pedipes and accepted as the type species of the ge- nus. Adanson, a contemporary of Linnaeus, derived the generic group name from the French word, pietin, meaning pedestrian, which refers to the way in which the animal progresses, first advancing the anterior half of the transversely divided foot, then moving the posterior half, seemingly advancing by steps. Adanson’s work (1757) antedates the starting point of zoological nomenclature and therefore he is not credited with introducing Pedipes. The first post-Linnaean use of Pe- dipes was that of Scopoli (1777), who briefly characterized Pedipes “Adanson” as having the shell aperture diversely dentate. The only known species, not mentioned by Scopoli, was Adanson’s African “pietin,” introduced by Bruguiere (1789) as Bulimus pedipes. He- Их afra Gmelin, 1791 [= Pedipes pedipes (Bruguiere, 1789)] was designated type spe- cies of Pedipes by Gray (1847a). Férussac (1821) used Pedipes in a re- stricted sense but Blainville (1824), using the sharp outer lip as the major diagnostic character, included a heterogeneous as- semblage, i.e., Conovulus [= Melampus] and the opisthobranch genus Tornatella. Lowe (1832), following Lamarck (1822) and Menke (1828), treated Pedipes as a genus of the family “Plicacea” and tried to prove that the animal was a pectinibranch. After Crosse & Fischer (1879) noticed that Pedipes was the only Known ellobiid that did not resorb the inner whorls of the shell, the genus could be separated easily from otherwise conchologi- cally similar groups, e.g., Marinula. In this study the genus Creedonia too was found not to resorb the inner whorls. The radula of Pe- dipes too is unique because the posterior- most 30 to 40 rows lack marginal teeth (Fig. 140). The genus Pedipes is represented in the Western Atlantic by two species. The larger (6 mm) Pedipes mirabilis has a thick, globose, generally heavily sculptured shell; the visible part of the protoconch consists of hardly more than one whorl and has a sinuous, elongate apertural lip. The smaller (3.5 mm), more elongate Pedipes ovalis has a thicker- shelled, rock-dwelling form and a thinner- shelled, mangrove-inhabiting form. The pro- toconchs of both forms are identical, with more than one and one-third whorls visible and with round, not sinuous, apertural lips. Habitat: The genus Pedipes lives in man- groves near the sea, in which animals are abundant under fallen leaves and branches below the high-tide mark. They also live along open rocky shores in crevices and un- der stones frequently covered by waves at high tide. Range: Worldwide, warm temperate to trop- ical regions. In the Western Atlantic they oc- cur in Bermuda, southern Florida to Texas, the West Indies and Central America south to Brazil and Ascension Island. Pedipes mirabilis (Mühlfeld, 1816) Figs. 102-106, 108-110, 112-120 Turbo mirabilis Mühlfeld, 1816: 8, pl. 2, figs. 13a, b [Locality unknown, herein desig- nated Cabo Rojo lighthouse, Puerto Rico; type specimens presumed lost, fide Clench (1964); neotype herein des- ignated MCZ 188476a (Fig. 102)). Pedipes mirabilis (Mühlfeld). Beck, 1837: 105; Pfeiffer, 1856a: 70; Pfeiffer 1876: 333; Mörch, 1878: 5; Nevill, 1879: 221; Pilsbry, 1900b: 503; Dall & Simpson, 1901: 369, pl. 53, fig. 8; Aguayo & Jaume, 1947: 218; Morrison, 1951b: 9; Morrison, 1958: 121 [ecology]; Nowell- Usticke, 1959; 88; Warmke & Abbott, 1961; 152, pl. 28, fig. j; Rios, 1970: 138; Morris, 1973: 274, pl. 74, fig. 12; Rios, 1975: 159, pl. 48, fig. 767; Rosewater, 1975: 23; Emerson & Jacobson, 1976: 189, pl. 21, fig. 20; Rehder, 1981: 648, fig. 234; Vokes & Vokes, 1983: 60, pl. 31, fig. 16 [juvenile; not positively this spe- cies]; Mahieu, 1984: 314 pp; Jensen & Clark, 1986: 458, figured. Pedipes quadridens Pfeiffer, 1840: 251 [Cuba; location of type unknown]; C. B. Adams, 1849: 41, 42; C. B. Adams, 1851: 186; Pfeiffer, 1854b: 148; Shuttle- worth, 1854b: 102; H. & A. Adams, 214 MARTINS 1855b: 149; Shuttleworth, 1858: 73; Poey, 1866: 394. Pedipes globulosus C. B. Adams, 1845: 12 [Jamaica; lectotype by Clench 4 Turner (1950), MCZ 177347 (Fig. 103)]; Clench & Turner, 1950: 288, pl. 49, fig. 9. Pedipes globulsus “Petit” Pfeiffer, 1856a: 70 [Haiti; type from Cuming's collection, not seen at ВММН]; Pfeiffer, 1876: 333. Non “Férussac” H. & A. Adams, 1854 (nomen nudum). Pedipes mirabilis (Mühlfeld) [in part] Arango у Molina, 1880: 60; Dall, 1889: 92, pl. 47, fig. 17; Crosse, 1890: 259; Kobelt, 1900: 255, pl. 24, figs. 19, 20; Maury, 1922: 54; C. W. Johnson, 1934: 159; M. Smith, 1937: 145, pl. 55, fig. 8 [probably Pe- dipes ovalis; pl. 67, fig. 17 is Pedipes ovalis]; M. Smith, 1951: [same illustra- tions as in first edition, 1937]; Clench, 1964: 119, pl. 76, figs. 1, 3, pl. 77 [fig. 2 is lectotype of Pedipes ovalis C. B. Ad- ams; systematics, distribution]; Ап- drews, 1971: 144, text fig. [figure prob- ably is of Pedipes ovalis]; Abbott, 1974: 333, fig. 4096 [in part]; Andrews, 1977: 181, text fig. [figure probably is of Pe- dipes ovalis]. Pedipes mirabilis Megerle. Peile, 1926: 88. Description: Shell (Figs. 102-106, 108-110) to 6 mm long, globose, very solid, white to brown. Spire low, as many as five con- vex whorls, sculptured with incised spiral grooves and fine axial striae. Body whorl av- eraging 88% of shell length, with average of 22 deeply incised spiral grooves. Sculpture as on spire; spiral grooves sometimes subdi- vided by fine spiral cords. Aperture about 70% of length of body whorl, widely ovate, round to angular at base, sometimes with weak angle at shoulder; columella flat and weakly concave, with two strong, rounded, subequal teeth perpendicular to columellar axis; parietal tooth strongest, oblique and slightly curved anteriorly; outer lip wide and smooth in juveniles, thick and crenulated in adults owing to grooves of body whorl; op- posite parietal tooth one large tooth very weakly extends inside aperture. Protoconch with barely more than one whorl visible, ap- ertural lip sinuous (Figs. 108-110). Radula (Figs. 112-116) as in genus; formula [120 + (6 + 70) + 1 + (70 + 6) + 120] x 120. Stomach (Fig. 117) as in genus. Reproductive system (Fig. 118) with her- maphroditic duct with longitudinally dilated seminal vesicle; bursa duct longer than sper- moviduct and albumen gland conbined. Pos- terior half of penis thicker than anterior por- tion. Nervous system (Fig. 119): left cerebrope- dal and cerebropleural connectives longer than right ones; left parietovisceral connec- tive twice length of right one; visceral gan- glion largest of five in visceral ring; left pleural and left parietal ganglia smaller than right counterparts. Remarks: In spite of the great variability shown by West Indian Pedipes, recent au- thors consider all of the named forms con- specific. According to Clench (1964) variabil- ity in Pedipes is a result of colonization strategy. Most colonies might have begun from one individual or from one cluster of eggs. Clench based this observation upon the meager representation of Pedipes in mu- seum collections, because he clearly stated (р. 118), “there is nothing in the literature... concerning their life history.” The colonies are not so rare as Clench implied. Pedipes species are among the most common West Indian ellobiids just below the high-tide mark, at least in mangroves (Martins, personal ob- servation). Pedipes mirabilis prefers piles of loose rocks around the high-tide mark. The shell is always thick, deeply grooved, with the aper- ture constricted in adults by a thick outer lip tooth. The body whorl of gerontic animals shows asymmetric growth. The name “quadridens” of Pfeiffer (1840) reflects the change in apertural aspect with age, and the names “globulosus” of С. В. Adams (1845) and “globulus” of Pfeiffer (1856a) refer to the allometric growth of this species. In 1849 C. B. Adams cautiously introduced Pedipes ovalis, calling attention to its strong affinity with Pfeiffer's Pedipes quadridens. Al- though the latter species is here considered a junior synonym of Pedipes mirabilis, Pedipes ovalis is recognized here as a separate spe- cies on the basis of size, sculpture and pro- toconch. The thick-shelled, heavily toothed Pedipes ovalis that C. B. Adams described from Jamaica rarely occurs in mangroves (Martins, personal observation), where the much thinner-shelled, smoother Pedipes tri- dens Pfeiffer [= Pedipes ovalis C. B. Adams] abounds. Their similar protoconchs and the sizes suggest, however, that both rock- dwelling and mangrove-dwelling forms are expressions of the same species. The simi- WESTERN ATLANTIC ELLOBIIDAE 215 FIGS. 102-111. Pedipes. (102) P. mirabilis (Mühlfeld), neotype (MCZ 188476a), Cabo Rojo lighthouse, Puerto Rico, sl 4.7 mm. (103) P. globulosus C. B. Adams, lectotype (MCZ 177347), Jamaica, sl 4.6 mm. (104) P. mirabilis, Puerto Cabello, Venezuela, sl 6.0 mm. (105) P. mirabilis, Rio Grande do Norte, Brazil (ANSP 300179), sl 3.8 mm. (106) P. mirabilis, Morgan's Bluff, Andros Island, Bahamas, sl 5.0 mm. (107) P. pedipes (Bruguiere), Senegal (AMNH 22590), sl 7.7 mm. (108) P. mirabilis, lateral view of spire and proto- conch, Maravén, Venezuela. (110) P. mirabilis, top view of spire and protoconch, Shelly Bay, Hamilton, Bermuda. (111) P. pedipes, top view of spire and protoconch, Sáo Miguel, Azores. Scale 1 mm. larities of Pedipes ovalis with Pedipes mira- bilis should, then, be interpreted as adapta- tions for life in rocky environments. A more detailed comparison between these two spe- cies is presented under the remarks on Pe- dipes ovalis. Pedipes mirabilis is similar to the Eastern Atlantic Pedipes pedipes (Bruguière), mostly in the shape of the protoconch (Fig. 111). The Eastern Atlantic species, however, has a double outer lip tooth and a bifid, downward- curved parietal tooth (Fig. 107). 216 MARTINS FIGS. 112-115. Pedipes mirabilis, radular teeth. (112) Shelly Bay, Bermuda, sl 2.3 mm. (113-115) El Palito, Venezuela, sl 3.2 mm. Scale 50 um. CAL 9L 65L 75L80L1T 2T ST 1M I —~ 2M 3M 4M 38M 40M 81M 124M / FIG. 116. Pedipes mirabilis, radula, El Palito, Ven- ezuela. Scale 10 um. H. & A. Adams (1854) listed a Pedipes globulus Férussac, which might be confused with the homonym introduced by Pfeiffer AA FIG. 117. Pedipes mirabilis, stomach, Bahamas. Scale 1 mm. WESTERN ATLANTIC ELLOBIIDAE ZY. ot FIG. 118. Pedipes mirabilis, reproductive system, Clifton Pt., New Providence, Bahamas. A-C, trans- verse sections and their locations. Scale 1 mm. FIG. 119. Pedipes mirabilis, central nervous sys- tem, Clifton Pt., New Providence, Bahamas. Scale 1 mm. (1856a) for a West Indian specimen. Such a name does not appear in Férussac (1821). The Adams brothers might have intended to refer to Pedipes ovulus, which Férussac (1821: 109) described as “longer than afra [= FIG. 120. Geographic distributions, Pedipes mira- bilis (circles) and Leuconopsis manningi (star). Open circle, locality from literature. Pedipes pedipes], smooth and shiny, without tooth on the outer lip.” Perhaps Férussac was dealing with a specimen of Marinula, which Connolly (1915) with doubt referred to Marinula xanthostoma H. & A. Adams. Pe- dipes globulus was described by Pfeiffer (1856a) using Petit’s manuscript name in the Cuming collection, and it is considered syn- onymous with Pedipes mirabilis. The name Pedipes globulus “Еегиззас” H. & A. Adams should be considered a nomen nudum. A Pedipes, tentatively assigned to Pedipes mirabilis, was found in the Early Miocene Cantaure Formation in Venezuela (Gibson- Smith & Gibson-Smith, 1979). Recently (1985) the Gibson-Smiths described those fossils as Pedipes mirandus, which | consider а junior synonym of Pedipes ovalis (see the remarks for this species). Habitat: Pedipes mirabilis usually lives on rocky shores, often where wave action is strong. The animals aggregate in fairly large numbers under rocks at or just below the high-tide mark. Range: Bermuda; Florida, Texas; West In- dies, Central America, northern South Amer- ica to Sao Paulo, Brazil (Rios, 1975); Ascen- sion Island (Fig. 120). Specimens Examined: FLORIDA: Daytona (USNM 162346, 253173); Indian River (USNM 758222); Lake Worth (MCZ 205366; 218 MARTINS USNM 599349); Palm Beach Inlet (MCZ 110215; USNM 543392); Boca Raton (ANSP 219865); S Bayshore Dr., Miami (USNM 701950); Biscayne Bay (MCZ 291105); Hau- lover Beach Park, Biscayne Bay (USNM 809777). TEXAS: Port Aransas (MCZ 225522, 229626); Mustang Island, Port Aransas (MCZ 235614); South Padre Island (ANSP 319092; USNM 758649); Port O'Connor (USNM 711183). BERMUDA (ANSP 48599, 62741; MCZ 9952, 24251, 167937): Flatts (USNM 171948); Shelly Bay (MCZ 25523; A.M.); М of Long Bird Bridge (A.M.); W of Somerset Bridge (A.M.); Ireland Island (USNM 712378). BAHAMA ISLANDS: GRAND ВАНАМА IS- LAND: W of Eight Mile Rock (R.B.); Hepburn Town, Eight Mile Rock (ANSP 370410); Car- avel Beach, Freeport (ANSP 370228); Tama- rind Shipway, Lucaya (ANSP 370708); GREAT ABACO ISLAND: Wilson City (ANSP 299513; USNM uncatalogued); Sweeting's Village (MCZ 24142); Sand Bank, Crossing Bay (MCZ 116721); Mores Island (MCZ 116719); ANDROS ISLAND: Morgan's Bluff (A.M.); South Mastic Point (A.M.); Mangrove Cay (USNM 180462a); PARADISE ISLAND (A.M.); NEW PROVIDENCE ISLAND: Bar Point (A.M.); Delaport Point (A.M.); Rock Point (A.M.); Clifton Point (A.M.); E of Clifton Pier (A.M.); ROYAL ISLAND (MCZ 78360). CUBA: El Vedado (MCZ 167983); Matanzas Bay (ANSP 167481; MCZ 83308, 109334, 167984); Peñas Altas (MCZ 127866); Playa de Bellamar (ANSP 222590, 345332); Ver- salles (MCZ 92075); Muelle de la Aduana, Matanzas (MCZ 188903); Chivera, Bahia de Santiago (MCZ 167985); Cayo Francés (MCZ 167982); Guantánamo Bay (ANSP 313059). JAMAICA (ANSP 22565, 22570, 22572; MCZ 117347, 117348, 185170; USNM 90459, 94747): Montego Bay (USNM 441609); Rob- ins Bay (MCZ 167896; USNM 441978); Jack's Bay (MCZ 167895; USNM 441836); Manchioneal (USNM 492493); Port Morant (USNM 423674); Rock Fort (MCZ 167894; USNM 423792); Kingston (USNM 442594); Kingston Harbor (MCZ 314005); Palisadoes (USNM 442540); Mouth of Rio Cobre, Port Royal (USNM 426870); Hunt's Bay (USNM 441675); Little River (USNM 492506). HAITI: St. Louis (MCZ 167899; USNM 439397); Port Salut (MCZ 167891; USNM 440000); Les Cayes (USNM 439780); Ааит (USNM 367339, 440107); Baïe Anglaise, near Aquin (USNM 439605); Saltrou (MCZ 167897, 167898, 223892; USNM 439341); W of Me- tesignix (USNM 404730); Bizoton (USNM 439843). DOMINICAN REPUBLIC: Santo Do- mingo (ANSP 60920; USNM 492507); Santa Bárbara de Samaná (ANSP 173412; MCZ 57783); Cayo Chico, 4 km E of Santa Bárbara de Samaná (MCZ 57784). PUERTO RICO: Pi- ñones, W of Boca de Cangrejos (A.M.); Pu- erta de Tierra, San Juan (A.M.); Punta Arenas, N of Joyuda (A.M.); Cabo Rojo lighthouse (MCZ 188476, 188476a); Humacao (MCZ 166297); Ensenada Honda, Culebra Island (USNM 159675). VIRGIN ISLANDS: ST. CROIX (USNM 621393, 706774): Christian- sted (MCZ 188477); ST. THOMAS (ANSP 22569; USNM 119543); GUANA ISLAND: North Beach (MCZ 89245); ST. JOHN (ANSP 22568). LESSER ANTILLES: ST. THOMAS (MCZ 294220); ST. KITTS (MCZ 167935); BARBUDA: Spanish Point (ANSP 353819); GUADELOUPE (ANSP 22566; MCZ 181419); MARTINIQUE (MCZ 167936, 294221; USNM 612694); Pointe Pie, 2.5 km S of Ste. Anne (MCZ 248315); GRENADINES: Union, Admi- ralty Bay, Bequia Island (MCZ 216484); BAR- BADOS (MCZ 167900, 167939; USNM 502106); TOBAGO: Buccoo Bay (ANSP 188276); TRINIDAD (MCZ 90508): Toco (MCZ 62326). CARIBBEAN ISLANDS: CAY- MAN ISLANDS: Cayman Brac (MCZ 294222); ARUBA (USNM 663655). CURAÇAO: Port Marie & Daaibooi Baai (R.B.). COSTA RICA: Portete (USNM 702836, 706405). PANAMA: Toro Point, Fort Sherman (USNM 734066); Limon Bay, inside Toro Point (USNM 732870; R.B.); Fort Randolph (USNM 759237). CO- LOMBIA: Sabanilla (MCZ 167890; USNM 103468, 193615). VENEZUELA: Cayo Punta Brava (A.M.), Parque Nacional de Morrocoy, Tucacas (A.M.); El Palito (A.M.); Puerto Ca- bello (A.M.); Maravén, Borborata (A.M.). BRAZIL: Praia do Forte, Natal, Rio Grande do Norte (ANSP 300179). ATLANTIC ISLANDS: ASCENSION ISLAND (USNM 735717). Pedipes ovalis C. B. Adams, 1849 Figs. 121-148 Pedipes ovalis C. B. Adams, 1849: 41 [Ja- паса; lectotype by Clench & Turner (1950) MCZ 177349 (Fig. 121)]; C. B. Ad- ams, 1851: 186; Pfeiffer, 1854b: 148; H. 8 A. Adams 1855b: 249; Pfeiffer, 1856a: 70; Pfeiffer, 1876: 333; Clench 4 Turner, 1950: 321, pl. 141, fig. 14 [lectotype fig- ured]; Morrison, 1951b: 9; Morrison, 1958: 121 [ecology]; Morton, 1955: 127-168 [evolution]. Pedipes tridens Pfeiffer, 1854b: 148 [nomen nudum]. WESTERN ATLANTIC ELLOBIIDAE 219 Pedipes tridens Pfeiffer, 1855: 122 [Bermuda and Cärdenas, Cuba, herein restricted to Bermuda; lectotype herein selected BMNH 1967590 (Fig. 122)]; H. & A. Ad- ams, 1855b: 249; Pfeiffer, 1856a: 72; Pfeiffer, 1876: 333; Pilsbry, 1900b: 503, pl. 62, fig. 10; Peile, 1926: 88; Haas, 1950: 198, pl. 22, fig. 4. Pedipes naticoides Stearns, 1869: 108, text fig. [Rocky Pt., Tampa Bay, Florida; ho- lotype USNM 37598 (Fig. 123)]; Pfeiffer, 1876: 334; Dall, 1883: 323; Dall, 1885: 279, pl. 18, fig. 17; Simpson, 1889: 69. Pedipes mirabilis (Mühlfeld) [in part]. Arango y Molina, 1880: 60; Dall, 1889: 92, pl. 47, fig. 17; Crosse, 1890: 259; Kobelt, 1900: 255, pl. 24, figs. 19, 20; Maury, 1922: 54; C.W. Johnson, 1934: 159; M. Smith, 1937: 145, pl. 55, fig. 8 [probably Pe- dipes ovalis; pl. 67, fig. 17 is Pedipes ovalis]; М. Smith, 1951: [same illustra- tions as in first edition, 1937]; Clench, 1964: 119, pl. 76, figs. 1, 3, pl. 77 [fig. 2 is lectotype of Pedipes ovalis C. B. Adams; systematics, distribution]; An- drews, 1971: 144, text fig. [figure prob- ably is of Pedipes ovalis]; Abbott, 1974: 333, fig. 4096 [in part]; Andrews, 1977: 181, text fig. [figure probably is of Pe- dipes ovalis]. Non Mühlfeld, 1816. Pedipes insularis Haas, 1950: 197, pl. 22, fig. 3 [Lover’s Lake, St. George’s, Bermuda; holotype FMNH 30171 (not seen); para- type ANSP 212176 (Fig. 124)]. Pedipes mirabilis, forma ovalis C. B. Adams. Robertson, 1960: 22. Pedipes mirandus Gibson-Smith & Gibson- Smith, 1985: 88, fig. 1 [Early Miocene Cantaure Formation, Paraguana Penin- sula, Venezuela; holotype NHMB No. H 17113 (not seen)]. Description: Shell (Figs. 121-139) to 3.5 mm long, oval, solid to thin, yellow to dark brown; spire low, whorls four and one-half, convex, sculptured with incised spiral grooves; mi- crosculpture of grooves composed of very fine, irregular, compressed axial lamellae, sometimes crossed by spiral lines; ribs smoothish, sometimes with incised spiral lines; slightly matte appearance caused by fine growth lines crossing spiral ribs. Body whorl averaging 85% of shell length, with 15 to 34 deeply incised spiral grooves. Aperture about 70% of length of body whorl, widely ovate to squarish, round to somewhat angu- late at base; columella flat and weakly con- cave, with two rounded teeth, the posterior one stronger, anterior one sometimes very weak; parietal tooth oblique, longest; outer lip somewhat angular posteriorly, frequently smooth in thin-shelled forms (Figs. 122-124, 126-128); thick-shelled forms usually with strong, ridge-like tooth opposite parietal tooth, extending inside aperture (Figs. 121, 125, 129, 132). Protoconch with more than one and one-third whorls visible, apertural lip round, not sinuous (Figs. 133-139). Radula (Figs. 140-144) as in genus; for- mula [75 + (5 + 50) + 1 + (50 + 5) + 75] x 120. Stomach (Fig. 145) as in genus. Reproductive system (Fig. 146) with her- maphroditic duct anteriorly dilated to form seminal vesicle; bursa duct shorter than sper- moviduct and albumen gland combined. An- terior half of penis thicker than posterior half. Nervous system (Fig. 147) with left cere- bropedal and cerebropleural connectives about twice length of right ones; left parieto- visceral connective about as long as right one; visceral ganglion largest of five in vis- ceral nerve ring; left pleural ganglion and left parietal ganglion three times larger than right counterparts. Remarks: Pedipes ovalis is very variable (Figs. 121-132). A stout, highly sculptured form could be confused with Pedipes mirabi- lis. In fact, C. B. Adams (1849: 41) introduced his description of Pedipes ovalis with the words, “Pedipes ovalis may be a variety of Pedipes quadridens Pfeiffer [= Pedipes mira- bilis (Múhltfeld)].” Аз С. В. Adams pointed out, it differs from Pedipes mirabilis by the smoothness of its body whorl and the less conspicuous tooth on the outer lip. The outer lip tooth in Pedipes ovalis is often ridge- shaped and it gradually diminishes into the aperture, whereas in Pedipes mirabilis this tooth is more tubercle-shaped. In Pedipes ovalis the anterior columellar tooth usually is weaker than the posterior one. The most consistent character differentiating these species, however, is the protoconch, which in Pedipes ovalis is larger and has a rounded, not sinuous, lip. Smoother, thin-shelled examples were named Pedipes tridens by Pfeiffer (1855), Pe- dipes naticoides by Stearns (1869) and Pe- Чрез insularis by Haas (1950) (Figs. 122-124). This form differs from the typical thick-shelled form in the greater number of grooves on the body whorl and in the wider, somewhat quadrangular aperture that some- 220 MARTINS FIGS. 121-132. Pedipes ovalis C. B. Adams. (121) Lectotype (MCZ 177349), Jamaica, sl 3.1 mm. (122) P. tridens Pfeiffer, lectotype (BMNH 1967590), Bermuda, sl 3.4 mm. (123) P. naticoides Stearns, holotype (USNM 37598), Tampa Bay, Florida, sl 2.4 mm. (124) P. insularis Haas, paratype (ANSP 212176), Lover's Lake, Bermuda, sl 2.4 mm. (125) Clifton Pt., New Providence, Bahamas, sl 2.3 mm. (126) Shore of Millars Road, New Providence, Bahamas, sl 2.3 mm. (127) Crawl Key, Florida, sl 2.3 mm. (128) Plantation Key, Florida, sl 3.5 mm. (129) Punta Arenas, Puerto Rico, sl 2.8 mm. (130) Isla Mujeres, Yucatán, Mexico (R.B.), sl 2.6 mm. (131) Fort Sherman, Panama (USNM 620532), sl 3.3 mm. (132) Puerto Cabello, Venezuela, sl 3.0 mm. times has a weak tooth inside the outer lip. The smoothness of the body whorl is very evident in the thin-shelled form although there is much variability and overlap with the thick-shelled form. Owing to unifying fea- tures, such as the identical protoconch, and the continuation and gradual disappearance of the outer lip tooth into the aperture, how- ever, the thin-shelled form should be consid- ered conspecific with Pedipes ovalis. The thick-shelled forms of Pedipes ovalis live mostly in rocky habitats, whereas the thin-shelled forms are predominantly man- grove-dwellers. In Punta Arenas, Puerto Rico, both species of Pedipes live in an area in which mangrove trees cover the rocky shore. At this site Pedipes ovalis showed a wide range of thickness and corresponding variability in the conspicuousness of the tooth on the outer lip (Fig. 129). In the Florida WESTERN ATLANTIC ELLOBIIDAE 221 FIGS. 133-139. Pedipes ovalis. (133) Juvenile, Crawl Key, Florida, sl 0.45 mm. (134) Juvenile, Crawl Key, Florida, sl 0.55 mm. (135) Juvenile, Crawl Key, Florida, sl 0.55 mm. (136) Lateral view of spire and proto- conch, Clifton Pt., New Providence, Bahamas. (137) Top view of spire and protoconch, Clifton Pt., New Providence, Bahamas. (138) Top view of spire and protoconch, Punta Arenas, N of Joyuda, Puerto Rico. (139) Top view of spire and protoconch, Isla Mujeres, Yucatán, Mexico. Scale 1 mm. Keys, in which | failed to collect Pedipes mi- rabilis and from which | could not confirm any museum records referring to that species, Pedipes ovalis in most mangroves appears as Pfeiffer’s Pedipes tridens or Stearns’ Pe- dipes naticoides. In rocky areas, however, the sculpture and shape approach those of Pedipes mirabilis. Anatomical research yielded some small differences in the reproductive and nervous 222 MARTINS FIGS. 140-143. Pedipes ovalis, radular teeth. (140) Whole radula, Ely's Harbour, Bermuda, si 3.1 mm. (141) Morgan's Bluff, Andros Island, Bahamas, sl 2.7 mm. (142, 143) Ely’s Harbour, Bermuda, sl 2.7 mm. Scale, Fig. 140, 1 тт; all others, 50 um. WESTERN ATLANTIC ELLOBIIDAE 223 С 1L2L3L 34L т 3T AT MUA ASS 1M 2M 14M 60M 62M DER ACTAS FIG. 144. Pedipes ovalis, radula, Ely's Harbour, Bermuda. Scale 10 um. FIG. 145. Pedipes ovalis, stomach, Florida. Scale 1 mm. systems, and counts of radular teeth are lower in Pedipes ovalis. On that basis, but mostly on the bases of the protoconch, the generally more disparate sizes of the col- umellar teeth, the shape of the outer lip tooth and the maximal size, Pedipes ovalis is con- sidered distinct from Pedipes mirabilis. The resemblance of the two species can be inter- preted as convergence due to adaptation to the same environmental pressures of the rocky shore. The gradation from the thick- shelled, rock-dwelling forms to the thin- shelled, mangrove-dwelling populations, to- gether with the retention of the same pattern of protoconch and shape of the tooth on the ot...-- FIG. 146. Pedipes ovalis, reproductive system, Florida. Scale 1 mm. FIG. 147. Pedipes ovalis, central nervous system, Florida. Scale 1 mm. outer lip, amply justify the inclusion of Pe- dipes tridens, Pedipes naticoides and Pe- dipes insularis as junior synonyms of Pedipes ovalis. As stated under the remarks on the previ- ous species, Gibson-Smith 8 Gibson-Smith (1985) described a Pedipes mirandus from the Early Miocene Cantaure Formation of Venezuela. The authors did not mention the shape of the protoconch, the decisive char- acter for the separation of the Western Atlan- tic species. Judging from the accentuated difference in size of the columellar teeth, however, | consider Pedipes mirandus a jun- 224 MARTINS FIG. 148. Pedipes ovalis, geographic distribution. ior synonym of Pedipes ovalis. The specimen of the latter species that | collected in Vene- zuela (Fig. 132) closely resembles the illustra- tion of the holotype of Pedipes mirandus (Gibson-Smith 8 Gibson-Smith, 1985: 88, fig. 1). Habitat: Pedipes ovalis often occurs with Pe- dipes mirabilis under rocks and in crevices at or just below the high-tide mark. The thinner- shelled forms are very common in man- groves under leaves, twigs and rocks at or just below high-tide mark. The juveniles ven- ture farther into the intertidal zone than do any other West Indian ellobiid. Range: Bermuda; Florida; West Indies; Mex- ico south to Panama and Venezuela (Fig. 148). Specimens Examined: FLORIDA: Waveland (USNM 123531); Miami (ANSP 320358; USNM 159439, 330934); Ocean Beach (USNM 270714); Third Ragged Key above Sand Key (USNM 462738); Key Largo (USNM 597459); Tavernier Key (USNM 492504); Plantation Key (MCZ 188973, 291000, 291003); Ocean Dr., Plantation Key (A.M.); Upper Matecumbe Key (USNM 492492); In- dian Key (MCZ 167889; USNM 492520); In- dian Key Fill, N of Indian Key Channel (A.M.); Lignumvitae Key (ANSP 156683); Lower Matecumbe Key (MCZ 167893; USNM 492495); Long Key (ANSP 219860; A.M.): Grassy Key (ANSP 89560, 397279; MCZ 188970; A.M.); Crawl Key (MCZ 188972, 289998, 289999; A.M.); Bonefish Key (ANSP 227991); Knight Key (MCZ 188971); Bahia Honda (ANSP 104115; MCZ 188969); West Summerland Key (A.M.); Big Pine Key (ANSP 104114, 227999; MCZ 291104); W end of Ko- hen Avenue and Long Beach Drive, both on Big Pine Key (A.M.); Little Torch Key (MCZ 188974, 291108); Big Torch Key (ANSP 104112); Ramrod Key (MCZ 188975; USNM 599368); Sugarloaf Key (ANSP 89561, 104113; MCZ 188478); Boca Chica Key (MCZ 167892; USNM 270349); Key West (ANSP 22563; USNM 36017, 492494); SW channel, Dry Tortugas (USNM 492505); Gar- den Key, Dry Tortugas (USNM 590210); Fla- mingo Key (ANSP 294313); Cape Sable (MCZ 291103); Seminole Point (ANSP 105432); Sanibel Island (MCZ 84103); Tarpon Bay, Sanibel Island (MCZ 84339); Captiva Island (ANSP 149408); Starvation Key (ANSP 130059); Palmetto (A.M.); Mullet Key (USNM 652408, 653109; A.M.); Shell Key (USNM 466287); Tampa Bay (MCZ 239222; USNM 37598a); Anclote Key (ANSP 22564). MEX- ICO: Isla Cancun, Quintana Roo (ANSP 285534). BERMUDA: (ANSP 48597, 48600, 48601, 48602; MCZ 9952a, 74809, 314027; USNM 6523, 94438, 492496): Fairyland (ANSP 99077, 111096; USNM 208071); Flatts (USNM 171963); Shelly Bay (MCZ 225523); Old Road, Shelly Bay (A.M.); Coney Island (А.М.); N of Long Bird Bridge (A.M.); Nonsuch Island (MCZ 248274); Lover's Lake (ANSP 212176); Cooper's Island (ANSP 131648); Hungry Bay (A.M.); W of Somerset Bridge (A.M.); Ely's Harbour (A.M.); Mangrove Bay (A.M.). BAHAMA ISLANDS: BIMINI (ANSP 325624): East Well, East Bimini (ANSP 326449); N. end of Pigeon Cay, Bimini La- goon (ANSP 326022; USNM 656173); S end of Pigeon Cay (ANSP 326017); Cavelle Pond, South Bimini (ANSP 325548); Tokas Cay (ANSP 325831); GRAND BAHAMA ISLAND: W of Eight Mile Rock (R.B.); Running Mon Canal (ANSP 369780); North Hawksbill Creek (ANSP 370569); Dead Mans Reef [Sandy Be- van’s Cay] (ANSP 371226, 371285); Sweet- ings Cay (ANSP 374312); Riding Point (ANSP 371521); West End (ANSP 368763, 371933); GREAT ABACO ISLAND: West Point (ANSP 299478); Gorling Cay (ANSP 299549); AN- DROS ISLAND: Morgan's Bluff (A.M.); South WESTERN ATLANTIC ELLOBIIDAE 220 Mastic Point (A.M.); Danlin Bay (USNM 180671); Mangrove Cay (ANSP 325639; USNM 180462); First island off Mintie Bar, SE end of South Bight (USNM 271784); NEW PROVIDENCE ISLAND: Delaporte Point (A.M.); E of Clifton Pier (A.M.); Clifton Bluff (MCZ 205367); Clifton Point (A.M.); Millars Road (A.M.); Malcolm Creek (A.M.); ROYAL ISLAND (MCZ 78360, 167901; USNM 468120); ELEUTHERA ISLAND: Governor's Harbor (MCZ 167995); EXUMA CAYS: Hog Cay (MCZ 225560, 225561); CAY SAL BANK: Salt Lagoon, Cay Sal (USNM 513429). CUBA (USNM 492498): Dimas (USNM 614603); Ha- bana (ANSP 130744); Las Villas, Caibarién (USNM 608763). JAMAICA (MCZ 177348a, 177349, 177350, 185170a; USNM 90460, 94748): Falmouth (ANSP 397266); Robin's Bay (MCZ 167896a; USNM 441978a); Jack's Bay (MCZ 167895а; USNM 441836a); Port Morant (USNM 423674a); Palisadoes (USNM 442540a). HAITI: St. Louis (USNM 439397a); Port Salut (USNM 440000a); Bizoton (USNM 439843a). PUERTO RICO: San Juan (R.B.); Punta Arenas, N of Joyuda (A.M.); Cabo Rojo lighthouse (МСУ 1884765). VIRGIN IS- LANDS: ST. CROIX (USNM 706775); ST. THOMAS (ANSP 22562). LESSER ANTILLES: ST. KITTS (MCZ 167935a; USNM 492491); GRENADA: Caliveny Harbor (ANSP 296716); ST. MARTIN (MCZ 250474). MEXICO: Isla Mujeres, Quintana Roo (R.B.). BELIZE: Twin Cays (USNM 841329); Drowned Cays (ANSP 284811). PANAMA: Devil's Beach, Fort Sher- man (USNM 620532). CARIBBEAN 15- LANDS: ST. ANDREWS ISLAND (ANSP 155415). VENEZUELA: Puerto Cabello (A.M.). Genus Creedonia new genus Type species: Сгеедота succinea (Pfeif- fer, 1854). Description: Shell to 3.8 mm long, oval-elon- gate, fragile. Spire moderately high, trun- cated, with rounded apex; as many as four smooth, weakly convex whorls. Body whorl about 80% of shell length. Aperture oval- elongate, about 70% of body whorl length, posteriorly acuminate, rounded at base; col- umella somewhat oblique and twisted; col- umellar teeth two, posterior one stronger; pa- rietal tooth a little stronger than posterior columellar tooth; outer lip sharp, smooth. In- ner whorls not resorbed. Protoconch large, smooth, with nuclear whorls covered by first whorls of teleoconch. Radula with about 45 teeth in a row; central tooth wide, with triangular base, small, uni- cuspid crown; lateral teeth with strong en- docone; transitional teeth with two еп- docones; marginal teeth with as many as five endocones. Animal whitish; tentacles long, pointed. Visceral mass coiled. Pallial cavity elongate; kidney long, thin. Hermaphroditic duct some- what dilated in the middle; penis with long diverticulum. Nervous system with long cere- bral commissure. Remarks: The genus Creedonia is created for Creedonia succinea (Pfeiffer) upon the ba- sis of shell, radular and anatomical charac- ters. This new genus is closely related to Pe- dipes and Marinula, and the type species was formerly included in one or the other genus. Creedonia, like Pedipes, does not resorb its inner whorls and, like Marinula, has a smooth shell and a smooth outer lip. The three gen- era characteristically have two columellar teeth and one strong parietal tooth. As stated above, Creedonia succinea for- merly was considered to belong to the genus Marinula. Only twice have some species of Marinula been assigned tentatively to new genera. Swainson (1855) introduced the ge- nus Cremnobates in which he included his three species Cremnobates cornea, Cremno- bates parva and Cremnobates solida, all from Tasmania. Hedley & Suter (1910) noted that Cremnobates cornea is a junior synonym of Ophicardelus australis (Quoy 8 Gaimard, 1832) and that Cremnobates solida is con- specific with Marinula patula (Lowe, 1832). They therefore selected Cremnobates parva (Fig. 156) as type of the genus. Connolly (1915) considered Cremnobates parva allied to Marinula xanthostoma H. 8 A. Adams, 1855. Iredale (1936: 328) proposed Mar- ipythia for Marinula xanthostoma H. & A. Ad- ams on the basis of Connolly's opinion that that species “could not be classed under Marinula.” This is a misinterpretation of the statements of Connolly (1915: 118) who, after tracing the tortuous history of Marinula xan- thostoma, concluded, “the typical form of xanthostoma is on the extreme borderland of Marinula,” but added that intermediate forms occurred in different localities, a fact making the connection with Marinula less doubtful. Research on the anatomy of a Marinula cf. xanthostoma H. 8 A. Adams, conchologically related to Cremnobates parva, revealed a re- productive system similar to that of Pedipes. 226 MARTINS The reproductive system of Marinula pepita King, 1832, the type species of the genus, differs considerably from that of the Adams’ species, leading to the conclusion that they are at least subgenerically separated. The similarity of the radular teeth of Marinula pepita to those of Marinula filholi (Hutton, 1878) (Figs. 163-168), conchologically allied to Marinula xanthostoma, casts doubt upon their generic separation. Because | lacked an opportunity to examine the anatomy of Cremnobates parva to assess its relationship to Marinula xanthostoma, | think a decision about the synonymy of the names proposed by Swainson and Iredale is unwarranted. The genus Marinula has been confused with Ovatella [Pythiinae] on the basis of the appar- ent similarity of the dentition of the inner lip. H. & A. Adams (1855b) created the subgenus Monica to include the Mediterranean Monica firminii (Payraudeau, 1826) [= Ovatella firminii), and the Madeiran Monica aequalis (Lowe, 1832) [= Ovatella aequalis] and Monica gracilis (Lowe) [= Ovatella aequalis]. The shells of Marinula are easily separated from those of Ovatella on the basis of their apertural teeth. Marinula, Pedipes and Creedonia all have two conspicuous columellar teeth, whereas Ova- tella has only one columellar tooth. The pari- etal tooth of Marinula is the strongest of the three inner lip teeth, whereas in Ovatella the anterior parietal tooth is the strongest (Fig. 88). Connolly (1915) added as a diagnostic character of the genus the absence of teeth on the outer lip, but the Eastern Pacific Marinula concinna (C. B. Adams, 1852) and Marinula brevispira (Pilsbry, 1920) have a thick, ridge-like tooth opposite the parietal tooth. Anatomical research on these species is needed to ascertain their phylogenetic re- lationships, however. Marinula is known from the Indo-Pacific and it is well represented along the Pacific coasts of Central and South America; it has been reported from the South Atlantic Islands and from South Africa as well (Connolly, 1915). The new genus Creedonia differs from Marinula by having a thinner, smaller shell that is less than half the size of that of any species included in Marinula, with the possi- ble exception of Marinula mandroni Velain, 1877, which Connolly (1915) suspected to have been named after a young specimen of Marinula velaini Connolly, 1915. In Creedonia the columella is twisted and oblique, instead of flat and straight, and the anterior columel- lar tooth is always conspicuous, whereas in Marinula it is very small (Figs. 155-157). The spire in Creedonia is more elevated, the apex is truncate and perforated (Fig. 158) instead of acuminate and obliterated as in Marinula (Fig. 159). As stated above, Creedonia ani- mals do not resorb the inner whorls of the shell (Fig. 153), whereas those of Marinula species do. The radula of Creedonia succinea differs from that of Marinula in its broad central and lateral teeth and in the very small number of teeth in a row (Table 3, Appendix). The mar- ginal teeth have several endocones but no ectocones, whereas in the Neozealandic Marinula Вой (Hutton) and in Marinula pe- pita King there are one or two endocones and several ectocones (Figs. 163-168). The genus Creedonia is named in honor of the Rev. Joseph Dennis Creedon, Pastor of Christ the King Church, Kingston, Rhode Is- land, as an expression of my gratitude for his support in this research and for his invaluable friendship. Creedonia succinea (Pfeiffer, 1854) Figs. 149-154, 158, 160-162, 169-173 Leuconia succinea Pfeiffer, 1854b: 156 [Cár- denas, Cuba; location of type unknown]; Pfeiffer, 1856a: 157; Pfeiffer, 1876: 370; Arango y Molina, 1880: 61; Crosse, 1890: 260; H. & A. Adams, 1855b: 248. Pedipes elongatus Dall, 1885: 279, pl. 18, fig. 4 [Marco, Florida; lectotype herein se- lected USNM 859012 (Fig. 149); five paralectotypes USNM 37599]; Dall, 1889: 92, pl. 47, fig. 4; Simpson, 1889: 60; Kobelt, 1900: 258, pl. 24, figs. 17, 18; Maury, 1922: 54; C. W. Johnson, 1934: 159; M. Smith, 1937, pl. 67, fig. 4 [pl. from Dall (1885)]; Emerson 4 Jacobson, 1976: 190; pl. 26, tig 2% Marinula succinea (Pfeiffer). Morrison, 1951b: 9; Morrison, 1958: 118-124 [habitat]; Abbott, 1974: 333, fig. 4100 [not fig. 4108]; Vokes & Vokes, 1983: 60, pl. 31, A Description: Shell (Figs. 149-154, 158) to 3.8 mm long, oval-elongate, fragile, shiny, translucent, pale yellow to golden brown. Spire truncate, with as many as four and one-half weakly convex, apparently smooth whorls; very fine spiral lines visible under high magnification, crossed by weak, irregu- larly spaced growth lines; spiral depression just below suture. Body whorl about 80% of shell length, smooth. Aperture oval-elongate, WESTERN ATLANTIC ELLOBIIDAE 227 FIGS. 149-159. Creedonia, Marinula. (149) Pedipes? elongatus Dall, lectotype (USNM 859012), Marco, Florida, sl 3.9 mm. (150) C. succinea (Pfeiffer), Crawl Key, Florida, sl 2.3 mm. (151) C. succinea, Big Pine Key, Florida, sl 3.3 mm. (152) C. succinea, Isla Mujeres, Yucatán, Mexico (R.B.), sl 4.3 mm. (153) С. succinea, Isla Mujeres, Yucatán, Mexico (R.B.), sl 3.3 mm. (154) С. succinea, lateral view of spire and protoconch, Crawl Key, Florida. (155) М. pepita King, syntype (BMNH 1968882), Chiloe Island, Chile, sl 10.1 mm. (156) М. parva (Swainson), New Zealand (USNM 98181), sl 6.4 mm. (157) M. filholi (Hutton), New Zealand, (USNM 681303), $1 5.4 mm. (158) С. succinea, top view of spire and protoconch, Crawl Key, Florida. (159) M. filholi, top view of spire and protoconch, New Zealand (USNM 681303). Scale 1 mm. about 70% of length of body whorl, round at anterior columellar tooth conspicuous, pos- base; columella somewhat oblique, twisted; terior columellar tooth twice the size of ante- columellar teeth two, oblique toward base; rior; parietal tooth lamelliform, as large as or 228 MARTINS somewhat larger than posterior columellar tooth; outer lip sharp, smooth. Inner whorls not resorbed (Fig. 153). Protoconch large, smooth, whitish, translucent; nuclear whorls enveloped by first whorl of teleoconch, leav- ing pit in apex of shell (Figs. 154, 158). Radula (Figs. 160-162, 169) with formula [12 + (2 + 12) + 1 + (12 + 2) + 12] x 80. Base of central tooth as wide as that of lateral teeth, rhomboidal, with anterior end much shorter than posterior, rounded; crown as wide as posterior end of base; mesocone small, triangular, with rounded tip; no ecto- cones. Lateral teeth eight to 12; base qua- drangular, medially bent at half-length; crown as wide as posterior end of base, triangular, with rounded tip; endocone about half the length of mesocone, strong, weakly pointed. Transitional teeth two, with base wider than that of lateral teeth, with two subequal en- docones. Marginal teeth 12 to 14; base be- comes shorter and wider; mesocone be- comes smaller as teeth approach lateral edge of radula; first marginal tooth with three subequal endocones; fourth endocone ap- pears on fourth marginal tooth, fifth en- docone on tenth marginal tooth. Animal whitish, translucent; tentacles mod- erately long, somewhat pointed, translucent, with bulbous base. Foot transversely divided. Pallial cavity elongate; kidney broad, triangu- lar, white. Digestive system with salivary glands small, fusiform. Stomach globose, very mus- cular; gastric caecum conspicuous, membra- nous (Fig. 170). Digestive gland bilobed; anterior lobe covers most of stomach and empties into pouch-like posterior crop through dilated anterior diverticulum; intes- tine very dilated as it comes off the stomach. Reproductive system (Fig. 171) with ovo- testis acinose, embedded in posterior lobe of digestive gland; hermaphroditic duct with irregularly dilated seminal vesicle; fertilization pouch bilobed, very conspicuous; albumen gland large, triangular; posterior mucous gland weakly convoluted; anterior mucous gland and prostate gland cover posterior half of spermoviduct. Bursa duct thick, shorter than spermoviduct; bursa elongate. Penis with several pouch-like dilations, with very long diverticulum wrapped around esopha- gus and salivary glands; short penial retractor attaches to columellar muscle; vas deferens short, free. Nervous system (Fig. 172) with cerebral commissure just shorter than width of cere- bral ganglion; left cerebropedal and cere- bropleural connectives shorter than right ones; pedal commissure very conspicuous; cerebral ganglia large, elongate laterally; left pleural ganglion about one-fourth size of right one; left parietal ganglion about one-tenth size of right one; visceral ganglion largest of visceral ring, somewhat smaller than pedal ganglia. Remarks: Creedonia succinea was originally assigned by Pfeiffer (1854b) to the genus Leuconia Gray [= Auriculinella Tausch, 1886]. The species appeared in the literature under this name until placed by Morrison (1951b) in the genus Marinula King, 1832, in which it has remained until now. Dall (1885) apparently was not aware of Pfeiffer's species when he introduced Pe- dipes elongatus for specimens from Marco, Florida. Creedonia succinea is one of the few species of ellobiids that shows little morpho- logical variation. It cannot be confused with any other West Indian species. The superfi- cial similarity to the Mediterranean Ovatella was already pointed out in the remarks under the genus Creedonia. In Creedonia the col- umellar tooth 1$ double and the parietal tooth is the strongest or at least as strong as the posterior columellar tooth. In Ovatella there is only one columellar tooth and the first pari- etal tooth 1$ the strongest. The same applies to the introduced Myosotella myosotis, with the difference that in this species the poste- rior parietal tooth is either absent or weaker than the anterior parietal tooth. The spire of Creedonia succinea is truncate and the pro- toconch gives it a mucronate appearance. The elongate, smooth, translucent shell, with flat whorls, separates Creedonia succinea from the thin-shelled form of Pedipes ovalis with which it occurs. Microtralia and Blaune- ría also occur with Creedonia; the former differs from Creedonia in having a narrow aperture with much smaller inner lip teeth and a very short spire. В/аипепа 1$ sinistral, has a high spire and is white and transpar- ent, whereas Creedonia is straw-colored to brown. Connolly (1915: 105), in his monograph on the genus Marinula, apparently was not ac- quainted with Pfeiffer's species. He men- tioned “Pythia abbreviatus Beck,” criticizing Pfeiffer's (1856a) questionable attribution of it to Marinula in these terms: “whatever may be its true genus, as the shell is said to come from the Antilles it is quite unlikely to be a WESTERN ATLANTIC ELLOBIIDAE 229 FIGS. 160-168. Creedonia, Marinula, radular teeth. (160-162) C. succinea, Long Key, Florida, sl 3.0 mm. (163) M. filholi, New Zealand, sl 5.4 mm. (164) M. filholi, New Zealand, sl 5.5 mm. (165) M. filholi, New Zealand, sl 5.4 mm. (166-168) M. tristanensis Connally [= M. pepita King], Gough Island (BMNH), $1 10.8 mm. Scale 50 um. 230 MARTINS С iL 2L 11L 12L 1M 2M 3M 15M 16M IAS ANA A FIG. 169. Creedonia succinea, radula, Long Key, Florida. Scale 10 um. lee an ‚I ] dj E SA el FIG. 170. Creedonia succinea, stomach, Crawl Key, Florida. Scale 1 mm. FIG. 171. Creedonia succinea, reproductive sys- tem, Crawl Key, Florida, sl 3.3 mm. A-C, trans- verse sections and their locations. Scale 1 mm. Marinula.” Beck (1837: 105) had listed, with- out description, a “Pythia abbreviatus” from the West Indies, placing the name after Pythia aequalis (Lowe, 1832) [= Ovatella ae- ppre. pipe cpe = IM a cc bg / ` ИВ Аа и a N \ = / Pg. x N SS == ` S ` > ` ` ` FIG. 172. Creedonia succinea, central nervous sys- tem, Crawl Key, Florida, sl 3.0 mm. Scale 1 mm. qualis] and Pythia patulus, which is question- ably referred by Connolly (1915) to Marinula xanthostoma H. & A. Adams. Pfeiffer (1856a) did not see Beck's specimens but tentative- ly assigned Pythia abbreviatus Beck to Marinula, no doubt on the basis that Beck listed it between two species that Pfeiffer considered to be Marinula. The only other species in the Western Atlantic that at first glance could be confused with Creedonia succinea is Myosotella myosotis, which does not live in the West Indies. In spite of the fact that some circumstancial evidence seems to indicate that Beck's name refers to Creedonia succinea, Рута abbreviatus Beck must remain a nomen nudum. Habitat: Individuals of Creedonia succinea live about the high-tide mark, the juveniles venturing a short distance into the intertidal zone. They live within the sediment, some- times 10 to 15 cm deep, and they occur fre- quently under half-buried rotting wood or rocks and on the roots of mangrove propa- gules, together with Pedipes, Microtralia and Blauneria. Range: Georgia ?, Florida Keys and the Ba- hama Islands south to Cuba and Jamaica; Mexico (Fig. 173). The USNM record from Isle of Hope, Georgia, collected by Hubricht, is so distant from the normal range that it could be explained better as the result of ac- cidental transportation by currents. Specimens Examined: GEORGIA: Isle of Hope (USNM 663054). FLORIDA: S of Ocean Drive, Plantation Key (A.M.); Lignumvitae Key WESTERN ATLANTIC ELLOBIIDAE 231 (ANSP 156694); Long Key (A.M.); Grassy Key (А.М.); Crawl Key (A.M.); Big Pine Key (ANSP 293553); Long Beach Drive and W of Kohen Avenue, both Big Pine Key (A.M.); Newfound Harbor (USNM 272639); Big Torch Key (ANSP 104105); Sugarloaf Key (ANSP 89566, 104104); Ramrod Key (MCZ 235471a); Boca Chica Key (USNM 590597); Key West (USNM 450693); Seminole Point (ANSP 105410); Marco (ANSP 22578; USNM 37599, 859012); Captiva Island (ANSP 149409); Mullet Key (USNM 652409; A.M.); Mullet Key Bayway (USNM 653110). BAHAMA ISLANDS: GRAND BAHAMA ISLAND: South Hawksbill Creek (ANSP 371809); ANDROS ISLAND: South Mastic Point (А.М.). CUBA: Matanzas (MCZ 131760). JAMAICA: Kingston (USNM 442584). MEXICO: N end of Ascension Bay, Quintana Roo (USNM 736105); Isla Mujeres, Quintana Roo (R.B.). Genus Microtralia Dall, 1894 Microtralia Dall, 1894: 117. Type species by monotypy: Auricula ? (Microtralia) mi- nuscula (Dall, 1889) [= Leuconia occi- dentalis Pfeiffer, 1854]. Rangitotoa Powell, 1933: 148. Type species by monotypy: Rangitotoa insularis Pow- ell, 1933. Description: Shell to 3.8 mm long, subcylin- dric, fragile, translucent white. Spire low to moderately high, with as many as seven weakly convex whorls. Body whorl 80% of shell length. Aperture narrow, about 90% of body whorl length; inner lip with small, ob- lique columellar tooth; anterior parietal tooth very near columellar tooth, strong; posterior parietal tooth very small, about mid-length of aperture; outer lip thin, sharp. Protoconch smooth, globose; nuclear whorls deeply em- bedded in first whorl of teleoconch. Radula with 55 to 79 teeth in a row. Central tooth at same level as lateral teeth; base broad, triangular, anteriorly emarginate; crown small, tricuspid. Base of lateral teeth quadrangular, weakly bent medially; crown less than half length of base, with large me- socone, small ectocone. Transitional teeth with one endocone. Marginal teeth wide, pectinate, with as many as six ectocones. Animal whitish to rusty brown, translucent. Foot not divided transversely, posteriorly en- tire, round. Eyes lacking. Tentacles short, subcylindric. Hermaphroditic duct dilated an- teriorly into a pouch-like seminal vesicle; an- "90 75. 60 45 30 FIG. 173. Creedonia succinea, geographic distribu- tion. terior mucous gland covering posterior half of spermoviduct; vas deferens free from penis. Connectives of visceral ring very short. Remarks: Since its introduction by Dall (1894) the genus Microtralia has been considered to belong to very different taxonomic groups. Its uncertain taxonomic position is the result of the different weights given by different au- thors to the various taxonomic characters. The etymology ofthe word implies similarity to Tralia, a member of the Melampinae. Dall (1894) tentatively placed Microtralia in the ge- nus Auricula [= Ellobium], a member of the Ellobiinae. Thiele (1931) considered Microtra- lía a subgenus of Melampus. Powell (1933), although recognizing the uniqueness of the genus, followed Odhner's (1925) radula- based classification and placed his Rangito- toa, here considered а junior synonym of М/- crotralia, in the Melampinae. Powell stressed the radular affinities of his genus with the Carychiinae. Morton (1955b), on the basis of anatomy and habitat preferences, placed Rangitotoa [= Microtralia] within the Pedipe- dinae. Zilch (1959) treated Microtralia as a subgenus of Melampus, and he considered Rangitotoa as a separate genus of the Melampinae. Abbott (1974) considered Mi- 232 MARTINS crotralia a genus of the subfamily Cassiduli- nae. Although the shell is not typical of the Pedipedinae, the dentition of the inner lip and the protoconch of Microtralia are similar to those of the more solid Pseudomelampus and Sarnia (Figs. 180, 181). The central and lateral teeth of the radula of this Eastern At- lantic genus closely resemble those of Pe- dipes, but the pectinate marginal teeth with as many as six ectocones are very similar to those of Pseudomelampus (Martins, per- sonal observation). Analysis of the reproduc- tive and nervous systems indicate the sys- tematic position of Microtralia within the Pedipedinae. The Neozealandic Rangitotoa insularis Powell, 1933, is quite similar to the West In- dian Microtralia occidentalis (Pfeiffer, 1854), especially in shell and radular characters, and Climo (1982) considered them conspe- cific (Fig. 179). Habitat: These animals live near the high- tide mark, under rocks partly buried in mud (Powell, 1933). In West Indian mangroves Mi- crotralia lives in the black sediment at the high-tide mark, preferably under rotting, half- buried branches (Martins, personal observa- tion). Range: Sporadic records from Easter Island (Rehder, 1980), Hawaii (Pease, 1869), New Zealand (Powell, 1933), Japan (Habe, 1961) and South Africa (Turton, 1932) indicate an Indo-Pacific distribution. In the West Indian region the genus 1$ represented by Microtra- lia occidentalis (Pfeiffer). Microtralia occidentalis (Pfeiffer, 1854) Figs. 174-178, 182-193 Leuconia occidentalis Pfeiffer, 1854b: 155 [Cárdenas, Cuba; location of type un- known]; H. 8 A. Adams, 18556: 248; Pfeiffer, 1856a: 157; Pfeiffer, 1876: 370; Arango y Molina, 1880: 61; Crosse, 1890: 260. Tralia (Alexia?) minuscula Dall т Simpson, 1889: 69 [Magill's Bay, Tampa, Florida, and Exuma Island, Bahamas, herein re- stricted to Magill’s Bay, Tampa, Florida; lectotype herein selected USNM 61211 (Fig. 174); two paralectotypes USNM 859503]. Tralia minuscula Dall. Dall, 1889: 92. Auricula ? (Microtralia) minuscula (Dall). Dall, 1894: 117, fig. 7 [Fig. 175]. Leucopepla occidentalis (Pfeiffer). Peile, 1926: 88. Microtralia occidentalis (Pfeiffer). Pilsbry, 1927: 125; Morrison, 1951b: 10; Abbott, 1974: 334 [not figured; fig. 4105, errone- ously referred to this species, represents Myosotella myosotis]; Jensen & Clark, 1986: 456 ffig. on раде 456, wrongly stated to represent this species, is of Myosotella myosotis]. Auriculastrum (Microtralia) minusculum (Dall). C.W. Johnson, 1934: 159. Auriculastra nana Haas, 1950: 197, pl. 22, figs. 1, 2 [Lover's Lake, St. George’s Is- land, Bermuda; holotype FMNH 30169 (not seen); paratype ANSP 212177 (Fig. 176)]. Melampus (Microtralia) minusculus Zilch, 1959: 65, fig. 208. (Dall). Description: Shell (Figs. 174-178, 182-184) to 3.8 mm long, subcylindric, fragile, translu- cent, white to yellowish. Spire low to moder- ately high; whorls to five and three-fourths, weakly convex, sculptured with very fine, un- dulating spiral lines that extend over body whorl. Body whorl about 80% of shell length, crossed by faint, compact growth lines. Ap- erture about 90% of body whorl length, nar- row; inner lip with three teeth on anterior half; columellar tooth small, oblique, twisted; an- terior parietal tooth strong; posterior parietal tooth very small, sometimes reduced to a FIGS. 174-184. Microtralia, Rangitotoa, Pseudomelampus, Sarnia. (174) Tralia (Alexia?) minuscula Dall, lectotype (USNM 61211), Magill's Bay, Tampa, Florida, sl 3.5 mm. (175) Auricula? (Microtralia) minuscula Dall, Atkins Island, Bahamas (USNM 127487), si 2.3 mm; figured by Dall (1894, fig. 7). (176) Auriculastra nana Haas, paratype (ANSP 212177), Lover's Lake, Bermuda, sl 3.2 mm. (177) M. occidentalis (Pfeiffer), Hungry Bay, Bermuda, sl 3.5 mm. (178) М. occidentalis, Hungry Bay, Bermuda, $1 3.6 mm. (179) A. insularis Powell, paratype (ANSP 242319), Rangitoto Island, Auckland, New Zealand, sl 3.2 mm. (180) P. exiguus (Lowe), lectotype (BMNH 1875.12.31.109), Madeira, sl 5.8 mm. (181) S. frumentum (Petit), syntype? (BMNH 1843.11.24.58), Lima, Peru, sl 7.0 mm. (182) M. occidentalis, top view of spire and protoconch, Plantation Key, Florida. (183) M. occidentalis, Hungry Bay, Bermuda, sl 3.1 mm. (184) M. occidentalis, lateral view of spire and protoconch, Hungry Bay, Bermuda. Scale 1 mm. 233 WESTERN ATLANTIC ELLOBIIDAE FIGS. 174-184. 234 MARTINS FIGS. 185-188. Microtralia occidentalis, radular teeth. (185) Hungry Bay, Bermuda, sl 3.9 mm. (186) Grassy Key, Florida. (187, 188) Hungry Bay, Bermuda, sl 3.9 mm. Scale 20 um. barely visible callus at mid-length of aperture; outer lip sharp, parallel to body whorl, sinu- ous. Inner wall of whorls occupying less than one-quarter of body whorl (Fig. 178). Proto- conch globose; nuclear whorls deeply invo- luted in first whorl of teleoconch; only small portion of lip showing (Figs. 182, 184). Animal whitish to rusty brown; tentacles short, subcylindric, with tip weakly pointed or somewhat flat and expanded. Eyes lacking. Mantle skirt whitish with brownish tinge along border. Pallial cavity somewhat elongate; kidney broadly triangular, anteriorly rounded, covering most of pallial cavity; pneumo- stomal and anal openings prolonged by a tube-like flap of mantle skirt; anal gill well de- veloped. Radula (Figs. 185-189) having formula [15 + (3 + 16) + 1 + (16 + 3) = 15] x 95. Crown of central tooth small, as wide as posterior end of base, tricuspid; mesocone small, blunt to weakly pointed; ectocones very small but well defined. Lateral teeth 12 to 18; crown ST FIG. 189. Microtralia occidentalis, radula, Grassy Key, Florida. Scale 10 um. wider than base, bicuspid; mesocone broadly rounded anteriorly, becoming more pointed and longer toward marginal teeth. Transitional teeth two to three, with small en- docone, thinner and longer ectocone. Mar- WESTERN ATLANTIC ELLOBIIDAE 235 FIG. 190. Microtralia occidentalis, stomach, Ber- muda. Scale 1 mm. ginal teeth 13 to 19; base short and wide, with lateral flare, on which endocone of next tooth articulates; Crown gradually widening and mesocone gradually becoming shorter and thinner toward margin; first marginal tooth with two ectocones; additional ecto- cones appearing on fourth, eighth and twelfth marginal teeth; sometimes a sixth ectocone appears on twelfth marginal tooth in some rows. Digestive system (Fig. 190) having diges- tive gland with two subequal lobes. Posterior crop dilated, receiving anterior diverticulum just before joining stomach. Anterior portion of stomach thin, with inner thickening be- tween entrance of esophagus and exit of intestine; mid-stomach gizzard-like, thickly muscular; gastric caecum thin, dilated, re- ceiving posterior diverticulum at junction with gizzard. Reproductive system (Fig. 191) with ovotestis acinose, trilobed, conical, at poste- rior tip of visceral mass, covering stomach; hermaphroditic duct straight, with anterior, pouch-like seminal vesicle connecting with convoluted fertilization chamber; albumen gland and posterior mucous gland large; an- terior mucous gland and prostate gland cov- ering posterior half of spermoviduct; bursa duct as long as spermoviduct, thick, empty- FIG. 191. Microtralia occidentalis, reproductive system, Hungry Bay, Bermuda. A, B, transverse sections and their locations. Scale 1 mm. ing near opening of vagina; bursa oval-elon- gate; vas deferens separates from oviduct near opening of vagina. Penis short, thick; associated vas deferens free, somewhat longer than penis; penial retractor about as long as penis, inserting on penis subapically, attaching to anterior portion of floor of pallial cavity. Nervous system (Fig. 192) with cerebral commissure somewhat shorter than width of cerebral ganglion; left cerebropleural and cerebropedal connectives longer than right ones; connectives of visceral ring very short, causing agglomeration of ganglia; pedal commissure short but conspicuous. Cerebral ganglia largest; pleural ganglia well devel- oped; left parietal ganglion very small; right parietal ganglion and visceral ganglion about same size. Penial nerve branching from me- dial lip nerve. Remarks: Originally Pfeiffer (1854b) assigned Microtralia occidentalis to the genus Leuco- nia Gray, 1840, which, because it was preoc- cupied, was renamed Leucopepla by Peile (1926). Pilsbry (1927) showed that on the ba- sis of shell characters Microtralia occidentalis could not be placed in Leucopepla [= Auricu- linella]. The latter genus belongs in the Ello- biinae on the basis of its nervous and repro- 236 MARTINS FIG. 192. Microtralia occidentalis, central nervous system, Hungry Bay, Bermuda. Scale 1 mm. ductive systems, and Microtralia (see re- marks under the genus) rightly belongs in the Pedipedinae. Dall (1889), apparently unaware of Pfeif- fer's name, described Tralia (Alexia?) minus- cula (Fig. 174) for which he created, in 1894, the subgenus Microtralia, tentatively remov- ing it to the genus Auricula. The odd combi- nations of names representing such different groups indicate the extent to which Dall was confused about the relationships of this small species. Microtralia occidentalis shows some mor- phological variation within populations and across its geographical range. Bermudian specimens are brownish and have the tips of the tentacles somewhat flat and broad. Flo- ridian specimens are usually whitish, some- times yellowish brown, and the tentacles are subcylindrical with blunt or weakly pointed tips. Bahamian examples are rusty brown and the tips of the tentacles are intermediate in shape between Bermudian and Floridian specimens. The radulae of specimens from Bermuda have rounder and somewhat shorter cusps than do those from Florida, but other- wise show no other morphological differ- ences. Intrapopulational variations in shell morphology, especially the height of the spire and the strength of the apertural teeth, occur throughout the range of the species. Haas (1950) apparently was unaware of Pfeiffer’s or Dall’s names when he introduced Auriculastra nana from Bermuda, for he did not refer to either author in the original description. Haas’ species (Fig. 176) does not differ from Microtralia occidentalis and it must therefore be considered a junior syn- onym of the latter. Climo (1982), as noted under the remarks for the genus, synonymized Rangitotoa insu- laris Powell with Microtralia occidentalis, con- sidering the former to have been introduced in New Zealand and possibly also in Rapa Iti Island and Easter Island. | concur with Cli- mo’s taxonomic decision about the genera, on the basis of the conchological and radular similarities; however, the widespread distri- bution of the genus (See the remarks for the genus) and the anatomical differences ob- served in Microtralia alba (Gassies, 1865) from Hong Kong (Martins, 1992), preclude an immediate synonymization of both species. An anatomical peculiarity of Microtralia oc- cidentalis is the absence of eyes, confirmed by histological examination. Concealment of the eyes under the skin has been reported for several species of the genus Ellobium (Pelse- neer, 1894a: 75, note 1). In the West Indian species Ellobium (A.) dominicense the eyes, although covered by thick skin, are readily visible. The shell of Microtralia occidentalis is not confused easily with that of any other West Indian ellobiid. It can resemble the very thin- shelled juveniles of some populations of Tra- lia ovula, however. Microtralia has a large, rounded protoconch, faintly incised, undulat- ing lines on the spire and a posterior parietal tooth that is anterior to the mid-length of the aperture. In Tralia ovula the apex is mu- cronate, the spire has marked, pitted lines and the posterior parietal tooth 15 in the pos- terior half of the aperture. Microtralia commonly occurs with Pedipes ovalis, Laemodonta cubensis, Blauneria het- eroclita and Creedonia succinea and 1$ readily distinguished from them. Pedipes is globose and has a rounded, strongly dentate aperture, Laemodonta is oval-elongate and hirsute and has a heavily dentate aperture, Blauneria has a sinistral shell and Creedonia has a truncated spire and very different inner lip dentition. Habitat: Microtralia occidentalis lives at or above the high-tide mark, buried in the sediment sometimes 10 to 15 cm deep, in the company of В/аипепа heteroclita and Creedonia succinea. The animals are quite common under partly buried, rotting wood or porous rocks, and on the roots of mangrove propagules, on which Laemodonta cubensis and Pedipes ovalis also abound. WESTERN ATLANTIC ELLOBIIDAE 237 90 75. 60 45 30 FIG. 193. Microtralia occidentalis, geographic dis- tribution. Range: Bermuda; Clearwater, Florida, south to the Florida Keys and the Greater Antilles (Fig. 193). Specimens Examined: BERMUDA (USNM 250297); Fairyland (ANSP 99075; USNM 208069); Old Road, Shelly Bay (A.M.); Lover's Lake, St. George's (ANSP 212172); near St. George's (ANSP 1008220); Castle Harbour, near Harrington House (ANSP 143322); Coo- pers Island (ANSP 131645); Hungry Вау (A.M.); S End of Ely's Harbour (A.M.); Man- grove Bay (A.M.). FLORIDA: N of Tavernier Creek, Key Largo (A.M.); S of Ocean Drive, Plantation Key (A.M.); Lignumvitae Key (ANSP 156682); Long Key (A.M.); Grassy Key (A.M.); Crawl Key (A.M.); Bahia Honda Key (ANSP 104108); Big Pine Key (ANSP 104102); W of Kohen Avenue, Big Pine Key (A.M.); Big Torch Key (ANSP 104001; A.M.); Sugarloaf Key (ANSP 89558); Boca Chica Key (ANSP 152503; USNM 270350); Semi- nole Point (ANSP 105409); Blue Hill, Hors Is- land (ANSP 99199); Captiva Island (ANSP 131836); McGill's Bay, near Tampa (USNM 61211, 859503); Boca Ciega Bay (ANSP 9571); Pinellas Point (USNM 83255); Clear- water Island (ANSP 9351). ВАНАМА IS- LANDS: GREAT ABACO ISLAND: Mores Is- land (MCZ 294207); ANDROS ISLAND: South Mastic Point (A.M.); Stafford Lake (ANSP 294338); Mangrove Cay (USNM 270214b); NEW PROVIDENCE ISLAND: W of Clifton Point (A.M.); E of Clifton Pier (A.M.); shore of Millars Road (A.M.); Bonefish Pond (A.M.); AKLINS ISLAND: Pinnacle Point (USNM 390857a); SAN SALVADOR (USNM 127487): Bob's Key, S. Ferdinand (USNM 360499). CUBA (ANSP 22482): near Habana (ANSP 130794). JAMAICA: Falmouth (ANSP 397269); Robin's Bay (USNM 441980, 442113); Kingston (USNM 395452b); Rio Co- bre, Port Royal (USNM 426889); Hunt's Bay (USNM 441642); Rock Fort (USNM 467164). HAITI: Gonave Island (USNM 380184). DO- MINICAN REPUBLIC: Rio Guayabin by Sa- baneta Road (ANSP 160398). PUERTO RICO: Puerta de Tierra, San Juan (A.M.); Pu- erto Real (A.M.). Genus Leuconopsis Hutton, 1884 Leuconopsis Hutton, 1884: 213. Type spe- cies by monotypy: Leuconopsis obsoleta (Hutton, 1878). Apodosis Pilsbry & McGinty, 1949: 9. Type species by monotypy: Apodosis novi- типа! Pilsbry & McGinty, 1949. Description: Shell to 4 mm long, oval-conic to oblong-conic, somewhat thin to solid. Spire low to moderately high, with as many as six and one-half flat, striated whorls. Body whorl about 80% of shell length. Aperture about 75% length of body whorl, approximately oval, posteriorly angulate; inner lip with strong submedian columellar tooth, usually with weak anterior secondary tooth; outer lip sharp. Protoconch smooth; nuclear whorls embedded in first whorl of teleoconch. Radula with 87 to 111 teeth in a row. Base of central tooth widened anteriorly, sharply constricted posteriorly; crown thin, falciform; mesocone long, sharp. Base of lateral teeth abruptly bent medially at half length; crown as wide as base, with strong endocone. Transi- tional teeth lacking. Marginal teeth with strong mesocone, weaker endocone and ectocone. Remarks: The genus Apodosis was created by Pilsbry & McGinty (1949) for the smallest and rarest West Indian ellobiid. In the original description the authors stated (p. 10) that they were “strongly inclined to treat Apo- dosis as a subgenus of the antipodal genus Leuconopsis Hutton.” The shape of the shell, oblong-conic in the Atlantic species vs oval- 238 MARTINS conic in the type species (Fig. 217) the unim- pressed suture and the inner thickening of the outer lip led them to establish a new ge- nus. Examination of additional specimens of Leuconopsis novimundi revealed that the thickness of the outer lip varies with the thickness of the shell, and that the distinct thickened outer lip actually does not appear in some thin-shelled specimens. This varia- tion was observed in Leuconopsis manningi n. sp. from Ascension Island and in Leu- conopsis rapanuiensis Rehder, 1980, from Easter Island. The other characters men- tioned by Pilsbry 4 McGinty are significant only at the specific level. The most obvious generic shell characters are the absence of a parietal tooth and the presence of a weak secondary columellar tooth just anterior to the primary submedian columellar tooth. This columellar structure is reminiscent of Pe- dipes, Marinula and Сгеедота. Powell (1933) illustrated the radula of Leu- conopsis obsoleta (Hutton, 1878). The mor- phology of the teeth is very similar to that of the West Indian Apodosis novimundi. On the basis of shell and radular characters, Apo- dosis Pilsbry 8 McGinty must be considered a junior synonym of Leuconopsis Hutton. Habitat: The genus lives intertidally under rocks (Hutton, 1884). Powell (1933: 150) found Leuconopsis obsoleta “in sheltered harbour bays towards high-tide, .. . and on cliffs, just above high-tide mark, in situations where fresh-water seepage occurs.” | col- lected Leuconopsis novimundi on New Prov- idence Island, Bahama Islands, in crevices of cliffs, just above high tide, but | did not see any indication of freshwater seepage. Range: The genus Leuconopsis occurs in the Pacific in Australia, New Zealand and Easter Island. In the Atlantic it is represented by Leuconopsis novimundi (Pilsbry & Мс- Ginty) from the Florida Keys, Bahamas and Jamaica, and by Leuconopsis manningi, herein described, known only from Ascen- sion Island. A possible third species from St. Thomas is left unnamed owing to lack of suit- able material. Leuconopsis novimundi (Pilsbry & McGinty, 1949) Figs. 194-204 Apodosis novimundi Pilsbry & McGinty, 1949: 10, pl. 1, fig. 1 [Clifton Bluff, New Providence, Bahamas; holotype ANSP 185474a (Fig. 194)]; Morrison, 1951b: 9; Zilch, 1959: 70, fig. 227; Franc, 1968: 525; Abbott, 1974: 334, fig. 4102. Non “Pilsbry & McGinty” Rosewater, 1975 [misindentification of Leuconopsis man- ningi Martins n. sp.]. Description: Shell (Figs. 194-197) to 3.4 mm long, oblong-conic, solid, uniformly light yellow to pale brown. Umbilical area marked by shallow excavation. Spire moderately high, whorls аз many as six and one-half, flat and heavily sculptured with numerous spiral cords, intersected by compact axial, somewhat regularly spaced, fine growth lines, giving shell a matte appearance. Body whorl convex, about 75% of shell length and with same sculpture as spire. Aperture about 70% body whorl length, subaxial, narrowly ovate; inner lip with partly hidden tooth at point of juncture of columella and parietal wall; occasionally secondary columellar tooth present as faint callosity just anterior to columellar tooth; outer lip sharp. Protoconch oblong, smooth, transparent, with sinuous lip (Fig. 197). Animal whitish gray; foot dirty white, trans- versely divided; tentacles short, transparent, subcylindric, with rounded tip; mantle skirt slightly lighter than rest of animal. Radula (Figs. 198-201) having formula (33 + 10 + 1 + 10 + 33) x 75+. Base of central tooth with quadrangular anterior half, round at tip; width of posterior half abruptly re- duced to half; crown as wide as posterior half of base, falciform; mesocone just over half length of base, pointed. Lateral teeth bicus- pid; mesocone sharp, as long as, but stron- ger than, that of central tooth; endocone sharp, almost as long and strong as meso- cone. Marginal teeth tricuspid; ectocone be- coming as long as endocone; mesocone be- coming smaller, but remaining the strongest cusp. Digestive system with posterior crop wide, with strong internal folds; stomach very mus- cular, with gastric caecum where posterior diverticulum empties. Reproductive system (Fig. 202) semidiau- lic, with vas deferens separating from oviduct almost at half-length of pallial gonoduct; an- terior mucous gland covers spermoviduct nearly to separation of vas deferens; bursa duct empties near female opening. Penial complex dilated, pouch-like in mid-section; posterior section thinner, coming out of pouch as blunt diverticulum; penial retractor ee WESTERN ATLANTIC ELLOBIIDAE 239 FIGS. 194-200. Leuconopsis novimundi (Pilsbry & McGinty), Clifton Bluff, New Providence, Bahamas. (194) Holotype (ANSP 185474a), sl 3.36 mm. (195) Shell length 2.73 mm. (196) Lateral view of spire and proto- conch. (197) Top view of spire and protoconch. (198) Central, lateral and marginal teeth of radula. (199) Marginal teeth of radula. (200) Marginal teeth of radula. Scale, Figs. 196, 197, 1 mm; Figs. 198-200, 20 um. multifid, short, attaching to end of posterior section of penis subapically; vas deferens free, entering penis at base of thinner poste- rior section. Nervous system (Fig. 203) with cerebral ganglia largest; left pleural ganglion larger than right one; left parietal ganglion much FIG. 201. Leuconopsis novimundi, radula, Clifton smaller than right one; visceral ganglion Bluff, New Providence, Bahamas. Scale 10 um. 240 MARTINS FIG. 202. Leuconopsis novimundi, reproductive system, Clifton Bluff, New Providence, Bahamas. Penis drawn under camera lucida, pallial gonoducts reconstructed from histological sections. A-L, transverse sections and their locations. Scale 1 mm. about as large as right parietal ganglion. Ce- rebral commissure somewhat longer than width of cerebral ganglion; right and left cere- bropedal and cerebropleural connectives roughly equal; left pleuroparietal connective very small; all other visceral ring connectives equal, about half length of cerebropleural connectives. Remarks: Leuconopsis novimundi shows some anatomical deviations from typical members of the Pedipedinae such as Pe- dipes and Creedonia (see the remarks under the subfamily). The origin of the vas deferens from the mid-section of the spermoviduct- vaginal tract was unexpected in this species because Morton (1955b) stated that the sep- aration of the vas deferens from the sper- moviduct in Leuconopsis obsoleta occurs at the vaginal opening, as in Ovatella [sensu Myosotella]. Nevertheless, Leuconopsis novi- mundi is placed in the Pedipedinae on the basis of the presence of the double columel- lar tooth in the aperture, the shape of the crown of the central and lateral teeth, the prostate and anterior mucous gland not FIG. 203. Leuconopsis novimundi, central nervous system, Clifton Bluff, New Providence, Bahamas. Scale 1 mm. reaching the female opening, and the rela- tively short visceral nerve ring, as compared to that of the Pythiinae or Ellobiinae. WESTERN ATLANTIC ELLOBIIDAE 241 Rosewater (1975) erroneously identified a small Leuconopsis from Ascension Island as Apodosis novimundi. These specimens (USNM 735149, 859015) differ from Leu- conopsis novimundi in the differently sculp- tured shell that lacks the matte appearance and in the readily visible secondary columel- lar tooth. They represent a previously unde- scribed species that is introduced as Leu- conopsis manningi п. sp. in this report (Figs. 205, 206). The protoconch of Leuconopsis novimundi is very similar to that of Leuconop- sis manningi in having a sinuous lip (Figs. 207, 208), a characteristic lacking in Leu- conopsis rapanuiensis (Fig. 216). Leuconopsis novimundi is readily distin- guished from all other Western Atlantic ello- biids by its oval-elongate shell that has a matte appearance and by its lack of parietal teeth in the aperture. Habitat: The two specimens collected by R. Robertson in 1964 on Pigeon Cay, Bimini, in algae on mangrove roots were probably the first ones to be collected alive. All other spec- imens in museum collections, including those of Pilsbry and McGinty, seem to have been obtained from beach drift. Despite thorough field work and patient rock-combing, all but one of the specimens | found alive came from a cave at the western tip of Clifton Bluff (Clif- ton Pt.), New Providence Island, Bahamas, kindly indicated to me by T. L. McGinty. The cave formed from a double crack in the coral bed, running from sea to shore. lt is open above, so direct sunlight illuminates it a few hours a day. At high tide water enters the main opening from the ocean, as well as the bottom of the double crack. One specimen was found among stones that had collected in one such crack, just above high-tide mark. Eight others were found in the crevices as deep as 20 cm in the wall and were obtained by chipping away the wet layers of coral in a band about 15 cm wide just above the black zone. The animals apparently feed on the de- tritus that collects in these crevices. The eight specimens were found within a radius of about 20 cm, together with Pedipes ovalis, Laemodonta cubensis and young Melampus (D.) monile. The other live specimen was found at Morgan's Bluff, Andros Island, Ba- hamas, under stones at the high-tide mark of a tidal pool. Jamaica Range: Florida Keys, Bahamas, (Fig. 204). 60 “45 30 FIG. 204. Leuconopsis novimundi, geographic dis- tribution. Specimens Examined: FLORIDA: Indian Key (USNM 492557a). BAHAMA ISLANDS: GRAND BAHAMA ISLAND: Gold Creek (ANSP 369338); Hepburn Town, Eight Mile Rock (ANSP 370409); Caravel Beach [John Jack Point], Freeport (ANSP 370225); BIMINI ISLANDS: N end of Pigeon Cay (ANSP 329623); ANDROS ISLAND: Morgan’s Bluff (A.M.); Mangrove Cay (USNM 1804625); First island off Mintie Bar, SE of South Bight (USNM 271888); NEW PROVIDENCE IS- LAND: Clifton Bluff (ANSP 185474; А.М.). JA- MAICA: Jack’s Bay (USNM 441915). Leuconopsis manningi n. sp. Figs. 120, 205-212 Apodosis novimundi Pilsbry & McGinty. [Type locality: English Bay, Ascension Is- land; holotype USNM 859015 (Fig. 205); 11 paratypes USNM 859015 (Fig. 206)].Rosewater, 1975: 23. Non Pilsbry & McGinty, 1949. Description: Shell (Figs. 205-208) to 2.1 mm long, ovate, solid, uniformly pale to dark brown, smooth, shiny. Umbilical excavation very weak. Spire short, whorls as many as five and one-fourth, flat, sculptured with two 242 MARTINS Mi? pd à | FIGS. 205-211. Leuconopsis manningi n. sp., English Bay, Ascension Island. (205) Holotype (USNM 859015), si 2.02 mm. (206) Paratype (USNM 859015), sl 1.58 mm. (207) Lateral view of spire and proto- conch. (208) Top view of spire and protoconch. (209) Central and lateral teeth of radula. (210) Marginal teeth of radula. (211) Marginal teeth of radula. Scale, Figs. 207, 208, 1 тт; Figs. 209-211, 20 um. to seven incised spiral lines on shoulder in adults, juveniles often spirally striated on en- tire length; growth lines very faint. Aperture 75% of body whorl length, ovate; inner lip with strong posterior columellar tooth about mid-length of aperture, with conspicuous but much smaller anterior secondary tooth; outer lip sharp. Protoconch oblong, smooth, trans- parent, dark brown, with sinuous lip. Radula (Figs. 209-212) having formula (32+11+1+11+32)x70+. Radular morphol- ogy as in Leuconopsis novimundi. Animal unknown. Remarks: Leuconopsis manningi was first mentioned in the literature by Rosewater (1975), who misidentified it as Apodosis novimundi. This new species differs from WESTERN ATLANTIC ELLOBIIDAE 243 C iL 2L 11L 1M 2M 22M 23M 31M 32M | | IM LA a Аа» y a à en LU ALL MA MA à N y ) 7 \ ] PRA L a | у L