a5 pea a he CERN En AURAS AR A A TA DCE HO у DE TIR À У 2 № И __ AA AE RRA net un BED lol N O ET TEE ATEN SH EXT "e EN PSA esas: в» вое 46 sin му Y {+ Mike, Нин ми een HG AAA LENS PTT CAR Em une a STO ned A ra СЕ учим LA note. р # [ > $ r ee ‘ , 7 Auch AA DP IE BL ме De Rr DOM MERCIA A NS A e Bait? ine Baste tha и PH HET] nd ee RATER dd С Ce ohh A UN. “nee Y hae 3 (2524 Ty Rs Mx grees CL 0:15 Os т 4 + $ CHINESE LES etter AR Reese es SS Ls 2 Leyte the Dotter tide ha PE OPEN AD 50 Ags Oot o Я x + лена D ee ae и nu. IT PE NEA La A, x SEHEN: UNE UE СТ A IA о мае > ao 1s Zaren Nennen p E MALA AI AA, tu ER N , he né: » Яя soi у Ая É IL} Е ао ` Ate AG). ма HUILE ; DD PRADO Nise ai os 0 AAN и EA 1 м: ae № А СОН СЯ , Jarre en the te me “ir » q mi DARAN AN En А Par ate at ‘ LA tee Чита, » A Duo OR ; 5 mr Nach, e MAA sd ma: С RE nf mate a LE CRE » A ps И, : ) se ОА sr CR » » . NS Sana RR y wet ar} x ven x ESTA И : eats etary ahha ee H NIE À OA riada ty O O Há Des Menton ER se MERE el № nena cers US НЫ PMNS Eva poms gf манны OUR BOUT CON ET het, En Mens qd Ra bie À 9 oF ae tees os 094 EEE RER CES ass yee Roue DT EUX МС im: TE CPE ках Le Ex re re ER m ala Et re гы y ie ee apa te + PE 2 er À i 5 СРР Е } oy Fd at id 4 ; D В 3 7 Rete PEGA i wag ayn D 2 CI a y al we ‘ sat IA As и С Y de y HO jur ; i is О 7} i PAY weg й yo 4 tré 5 В ай, И} A 3 y Pater) 19.4 APR pas rs À 1 ANT À # $ y $4 et EUA NE Hip ; ety ELITE ee } : В 0 i y ba ada re р 7 в werke Г te 4 SP e TS } : ‘ ATX 5 n % АЕ > i Way) Y . { 4 a . 4 ’ PA } \ ния Re Sy ee pa pes À En À ‘ iS hir AS i Г CHEN у x $ y 4 7 у терр ня Pra: te м , у MR АОИ Нан : NL НЙ. НАК К nee wa SFMT TS ARG } % 49 3 ess a 3 Ar Burn ER у ; y nai ; : ори му Se AS ‘2 58 i 1 Fat RENE SU р my i ‘ eS Bag OS ee Hs q AO Are) SU UE Fits eaters Fr = HARVARD UNIVERSITY К Library of the Museum of Comparative Zoology VOE. 27 1986 MALACOLOGIA nternational Journal of Malacology Revista Internacional de Malacologia Journal International de Malacologie Международный Журнал Малакологии Internationale Malakologische Zeitschrift Z. A. Filatova К. Нам! А. Муга Кееп 1905-1986 see pp. 375—404 C. M. Yonge Publication dates Vol. 26, No. 1-2—9 July, 1985 Vol. 27, No. 1-7 March, 1986 MALACOLOGIA, VOL. 27 CONTENTS R. A. D. CAMERON Environment and diversities of forest snail faunas from coastal British (Grol ign TEE O O A! M. A. CHAUDHRY 8 E. MORGAN Factors regulating oviposition in Bulinus tropicus in snail-conditioned E o a IE А A ae Oe ee TRE Fa T. C. CHENG 4 E. J. PEARSON Modification and evaluation of Burch and Cuadros's medium for the main- tenance of the testes of a marine gastropod ............................. 1. DEYRUP-OLSEN, A. W. MARTIN & В. T. PAINE The autotomy escape response of the terrestrial slug Prophysaon foliolatum (RulmonatarArionidae) PR RE Meese etic o Ao H. L. FAIRBANKS The taxonomic status of Philomycus togatus (Pulmonata: Philomycidae): a morphological and electrophoretic comparison with Philomycus ОЕ A ee EEE Е. J. GARCIA, J. С. GARCIA & J. L. CERVERA Estudio morfológico de las espículas de Doriopsilla areolata (Gastropoda: NUGIDEANCNIA)NE RE AN ates tes ARTE Te Maia hig yn VS N M. G. HADFIELD ExiüncionuniHawWaiian achatinelline Snails 2... 00.0 nee eee ee S. A. HARRIS, F. M. da SILVA, J. J. BOLTON & A. C. BROWN Algal gardens and herbivory in a scavenging sandy-beach nassariid whelk . R. HERSHLER & G. LONGLEY Phreatic hydrobiids (Gastropoda: Prosobranchia) from the Edwards (Balcones Fault Zone) Aquifer region, south-central Texas ................ . HERSHLER & W. L. MINCKLEY Microgeographic variation in the banded spring snail (Hydrobiidae: Mexipyrgus) from the Cuatro Ciénegas basin, Coahuila, México .......... K. E. HOAGLAND Genetic variation in seven wood-boring teredinid and pholadid bivalves with different patterns of life history and dispersal ............................. . JABLONSKI & К. W. FLESSA The taxonomic structure of shallow-water marine faunas: implications for PRANCTOZOIC EXINCUONS as ae oa EN A AAN: M. S. JOHNSON, J. MURRAY 8 B. CLARKE High genetic similarities and low heterozygosities in land snails of the genus: Samoana от the Society Islands. nee H. A. JONES, R. D. SIMPSON & C. L. HUMPHREY The reproductive cycles and glochidia of fresh-water mussels (Bivalvia: Hyriidae) of the Macleay River, northern New South Wales, Australia ..... P. W. KAT Hybridization in a unionid faunal suture zone ............................. D о 341 249 173 307 271 83 67 299 127 357 323 43 97 185 107 A. M. KEEN (posthumous) Some important souces of molluscan generic type designations .......... 403 - A. KITCHELL, С. H. BOGGS, J. A. RICE, J. Е. KITCHELL, A. HOFFMAN & J. MARTINELL Anomalies in naticid predatory behavior: a critique and experimental observations +... Pee wa a bate ое по В 291 Р. MORDAN 4 $. TILLIER New Caledonian charopid land snails. I. Revision of the genus Pararhytida (Gastropoda: Charopidae) ..............2. 222.000 one ae oe sel eer 203 . D. PARASHAR & K. M. RAO Effects of long-term exposure to low concentrations of molluscicides on a fresh-water snail, Indoplanorbis exustus, a vector of schistosomiasis ...... 265 = 09) O. $. РЕН! & J. D. THOMAS Polymorphism т a laboratory population of Biomphalaria glabrata trom a seasonally drying habitat in north-east Brazil ................... 313 C. S. RICHARDS 4 D. J. MINCHELLA Genetic studies of biphallic Biomphalaria glabrata ........................ 243 D. RITTSCHOF 4 A. B. BROWN Modification of predatory snail chemotaxis by substances in bivalve prey OAOFS: 2:1. a В I CEE ETES 281 В. ROBERTSON 4 Е. V. СОАМ А. Муга Keen (1905-1986), :.....2.... sas cas 02e DEEE 375 С. J. VERMEIJ Molluscan extinction: introduction to a symposium ........................ 1 С. J. VERMEIJ 4 Е. J. PETUCH Differential extinction in tropical American molluscs: endemism, architecture, andthe’ Panama land ibridge . 2s... 0346. ASRS nas au en EEE 29 P. WARD Cretaceous ammonite shell Shapes. ....... 2... 2... 20.0 eee 3 Muy? MCZ VOL. 27, NO. 1 LIBRARY 1986 MAR 11 1986 ВУАКО LIV ERSITY MALACOLOGIA International Journal of Malacolog y Revista Internacional de Malacologia Journal International de Malacologie | Международный Журнал Малакологии Internationale Malakologische Zeitschrift MALACOLOGIA Editors-in-Chief: GEORGE M. DAVIS ROBERT ROBERTSON Editorial and Subscription Offices: Department of Malacology The Academy of Natural Sciences of Philadelphia Nineteenth Street and the Parkway Philadelphia, Pennsylvania 19103, U.S.A. Associate Editors: JOHN B. BURCH University of Michigan, Ann Arbor ANNE GISMANN Maadi, А. В. Egypt Editorial Assistant: MARY DUNN Assistant Managing Editor: CARYL HESTERMAN MALACOLOGIA is published by the INSTITUTE OF MALACOLOGY, the Sponsor Members of which (also serving as editors) are: KENNETH J. BOSS, President-Elect Museum of Comparative Zoology Cambridge, Massachusetts JOHN B. BURCH MELBOURNE R. CARRIKER, President University of Delaware, Lewes GEORGE M. DAVIS Secretary and Treasurer PETER JUNG, Participating Member Naturhistorisches Museum, Basel, Switzerland JAMES NYBAKKEN, Vice-President Moss Landing Marine Laboratories California OLIVER E. PAGET, Participating Member Naturhistorisches Museum, Wien, Austria ROBERT ROBERTSON CUYDE.F.. E: ROPER Smithsonian Institution Washington, D.C. W. D. RUSSELL-HUNTER Syracuse University, New York NORMAN F. SOHL United States Geological Survey Washington, D.C. SHI-KUEI WU University of Colorado Museum, Boulder J FRANCIS ALLEN, Emerita Environmental Protection Agency Washington, D.C. ELMER G. BERRY, Emeritus Germantown, Maryland Copyright O 1986 by the Institute of Malacology EDITORIAL BOARD J. A. ALLEN Marine Biological Station Millport, United Kingdom E. E. BINDER Museum d'Histoire Naturelle Genéve, Switzerland A. J. CAIN University of Liverpool United Kingdom P. CALOW University of Sheffield United Kingdom A. H. CLARKE, Jr. Portland, TX, U.S.A. B. C. CLARKE University of Nottingham United Kingdom R. DILLON College of Charleston SC, U.S.A. E. FISCHER-PIETTE Muséum National d'Histoire Naturelle Paris, France У: ЕВЕТТЕВ University of Reading United Kingdom E. GITTENBERGER Rijksmuseum van Natuurlijke Historie Leiden, Netherlands Е. GIUSTI Universita di Siena Italy A. N. GOLIKOV Zoological Institute Leningrad, U.S.S.R. S. J. GOULD Harvard University Cambridge, MA, U.S.A. A. V. GROSSU Universitatea Bucuresti Romania T. HABE Tokai University Shimizu, Japan A. D. HARRISON University of Waterloo Ontario, Canada K. HATAI Tohoku University Sendai, Japan J. A. HENDRICKSON, Jr. Academy of Natural Sciences Philadelphia, PA, U.S.A. K. E. HOAGLAND Lehigh University Bethlehem, PA, U.S.A. B. HUBENDICK Naturhistoriska Museet Göteborg, Sweden S. HUNT University of Lancaster United Kingdom R. JANSSEN Forschungsinstitut Senckenberg Frankfurt am Main, Germany (Federal Republic) A. M. KEEN Santa Rosa CAUSA" АВ. М. KILBURN Natal Museum Pietermaritzburg, South Afica M. A. KLAPPENBACH Museo Nacional de Historia Natural Montevideo, Uruguay J. KNUDSEN Zoologisk Institut & Museum Kobenhavn, Denmark А. J. KOHN University of Washington Seattle, U.S.A. Y. KONDO Bernice P. Bishop Museum Honolulu, HI, U.S.A. J. LEVER Amsterdam, Netherlands A. LUCAS Faculté des Sciences Brest, France С. MEIER-BROOK Tropenmedizinisches Institut Túbingen, Germany (Federal Republic) H. K. MIENIS Hebrew University of Jerusalem Israel J. E. MORTON The University Auckland, New Zealand J. J. MURRAY, Jr. University of Virginia Charlottesville, U.S.A. R. NATARAJAN Marine Biological Station Porto Novo, India J. 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, PA, U.S.A. W. F. PONDER Australian Museum Sydney A. W. B. POWELL Auckland Institute & Museum New Zealand R. D. PURCHON Chelsea College of Science £ Technology London, United Kingdom QI Z.-Y. Academia Sinica Qingdao, People's Republic of China N. W. RUNHAM University College of North Wales Bangor, United Kingdom S. G. SEGERSTRÁLE Institute of Marine Research Helsinki, Finland G. A. SOLEM Field Museum of Natural History Chicago, U.S.A. F. STARMUHLNER Zoologisches Institut der Universitát Wien, Austria У. |. STAROBOGATOV Zoological Institute Leningrad, U.S.S.R. W. STREIFF Université de Caen France J. STUARDO Universidad de Chile Valparaiso T. E. THOMPSON University of Bristol United Kingdom S. TILLIER Muséum National d'Histoire Naturelle Paris, France F-TOFEOEETTO Societa Italiana di Malacologia Milano R. D. TURNER Harvard University Cambridge, MA, U.S.A. W. $. $. VAN BENTHEM JUTTING Domburg, Netherlands J. A. VAN EEDEN Potchefstroom University South Africa N. H. VERDONK Rijksuniversiteit Utrecht, Netherlands B. R. WILSON Nedlands Western Australia C. M. YONGE Edinburgh, United Kingdom EISZEISSEER Leipzig, Germany (Democratic Republic) A. ZILCH Forschungsinstitut Senckenberg Frankfurt am Main, Germany (Federal Republic) y à 4 Ay Mare = JN MALACOLOGIA, 1986, 27(1): 1-2 MOLLUSCAN EXTINCTION: INTRODUCTION TO A SYMPOSIUM Geerat J. Vermeij Department of Zoology, University of Maryland, College Park, MD 20742, U.S.A. Most of the species that have ever lived are extinct. This simple fact may seem a tired cliché yet it has profound implications for the study of evolution and, perhaps even more importantly, for study of the effects that humans are having on the world's biota. Few topics are as theoretically interesting and as important to human welfare as ex- tinction. Molluscs are an exceptionally attractive group for the study of extinction. Their fossil record is better than that of most other groups, and shelled molluscs are ecologically and biogeographically perhaps the best under- stood animals among invertebrates. A host of important questions awaits the willing investigator. Which factors bring about the extinction of species? How do extinct spe- cies differ morphologically, ecologically, and biogeographically from survivors? Which fea- tures enable species to persist through crises, and how do these features affect post-crisis evolution? What is the tempo of extinction; does extinction occur continuously, or is it concentrated during certain brief intervals of time? Are there qualitative differences be- tween the famous mass extinctions, such as those at the end of the Permian and Creta- ceous Periods, and other less devastating events such as those in the Pliocene? What effects has extinction had on surviving spe- cies? Can the extinctions of the geological past be held up as useful models for the extinctions taking place in the present day as humans are eliminating species and habitats on an ever increasing scale? During a symposium that was held on the morning of August 7, 1983, at the annual meeting of the American Malacological Union in Seattle, Washington, five papers dealing with various aspects of extinction were pre- sented. Four of these papers are published in this issue. P. Ward examines the pattern of selectivity and timing of extinction in ammonoids during the Cretaceous Period. His analysis goes a long way toward understanding the eventual global failure of these externally shelled ani- mals. G. J. Vermeijand E. J. Petuch also examine the pattern of selectivity of extinction. Their analysis highlights the extinction of gastro- pods and pelecypods during the Pliocene in tropical America. Perhaps the most important conclusion is that the striking architectural differences currently existing between the gastropod faunas of the tropical Western At- lantic and Eastern Pacific are, at least in part, attributable to differential extinction after the uplift of the Panama Isthmus in the Pliocene. D. Jablonski and K. Flessa present an anal- ysis of the distribution of molluscan and echinoderm families on oceanic islands and continental shelves. They show that a reduc- tion in shallow-shelf area owing to a cata- strophic lowering of sea level, such as the one that is believed by some to have contributed to the mass extinctions near the close of the Permian and Cretaceous Periods (or the Pal- eozoic and Mesozoic eras), would not cause a major extinction at the family level among living molluscs. They suggest that changes in sea level have in general been of secondary importance as direct agencies of extinction. The close association between sea level and extinction in the fossil record may be fortu- itous, or an indirect consequence of biogeo- graphic or climatic perturbations. The final paper, by М. Hadfield, is an ele- gant analysis of the way in which populational characteristics have rendered several Hawai- ian endemic land snails exceptionally suscep- tible to decimation and extinction as a result of chronic harvesting by overzealous human col- lectors. His work is a beautiful example of how theoretical ecology can be applied to the prac- tical problems arising when humans tamper with the species and environments around them. Hadfield's work is also important in that it highlights the value of the scholarly gather- ing together of anecdotal accounts from un- published notebooks and other obscure sources more familiar to the historian of hu- man affairs than to a biologist. 2 VERMEIJ These contributions illustrate the important role that the study of molluscs can play in enhancing our understanding of extinction. They also forcefully remind us that results and implications of great practical significance of- ten flow from studies of seemingly abstruse and esoteric topics. The man in the street might be disinclined to fund research on am- monoids or theoretical ecology, yet such re- search provides the essential groundwork for the resolution of problems of great practical significance, such as the human impact on habitats and species. MALACOLOGIA, 1986, 27(1): 3-28 CRETACEOUS AMMONITE SHELL SHAPES . Peter Ward! University of California, Davis ABSTRACT Cretaceous ammonite genera, and Cretaceous ammonite species of the Great Valley Se- quence, California, are analyzed for shell morphology on a stage by stage basis. Each taxon was measured for shell ornamental rugosity (shell rib width divided by whorl height), and overall shell shape (either planispiral or heteromorphic, with planispiral taxa being measured for the S statistic of Raup, 1967). For both generic and species level data, shell ornamental rugosity for the Cretaceous stages was compared by placing all taxa into one of four ornamental categories, based on the rib width/whorl height measurement. It was found that at both the species and generic level, shell ornamental rugosity increased progressively during the lower Cretaceous, and reached a peak in the Cenomanian-Turonian. During the remainder of the Cretaceous, rugosity of the generic and species level populations diminished. By placing all members of the two data sets into one shell shape category (either ornamentea, with rib width/whorl height ratios of .15 or greater), streamlined planispiral (S values of .5 or less), heteromorphic (all non- planispiral shells) or non-streamlined, non-coarsely ornamented planispiral, stage by stage examination of the relative proportions of these shell shape categories was carried out. In addition to the mid-Cretaceous peak of coarsely ornamented forms, these analyses revealed two peaks of heteromorphic diversity, during the Lower and Upper Cretaceous. Streamlined forms also showed their highest relative abundances during the latest Cretaceous. INTRODUCTION Perhaps the greatest difference between marine ecosystems of the Mesozoic and Ce- nozoic Eras was the presence of thousands of species of chambered cephalopods, mainly ammonites, in the former. During the Me- sozoic Era the ammonoids were numerically abundant in all of the world's seas, and judg- ing from their commonality in a variety of sedimentary facies, must have been impor- tant constituents in a wide variety of marine ecosystems. They cannot be excluded from any discussion of evolution within the pelagic realm, for they played a vital part in the marine Mesozoic history of the earth. The mode of life of Mesozoic ammonoids has been the topic of scientific inquiry for over a century, but remains controversial because of the complete lack of living ammonoid spe- cies. Our nearest living analogue is the Indo- Pacific cephalopod Nautilus, the last vestige of the Nautiloidea, and last surviving exter- nally shelled cephalopod. As in ammonoids, the external shell in Nautilus provides not only protection of the soft parts, but perhaps more importantly serves as a buoyancy producing organ, equivalent in function to the swim- bladder of a fish. Nautilus and ammonoid shells are separated into two parts: a large cavity that contains the living soft parts, and a closed portion that is gas and liquid filled, and is partitioned by numerous calcareous septa that serve as strengthening structures. The chambered shell region is of sufficiently low density to buoy up the heavy, calcareous shell and soft parts so as to provide neutral buoy- ancy for the active shell and animal. With this buoyancy system, and presumed squid-like water jet propulsion system, Me- sozoic ammonoids must have been functional equivalents of fish. The recent, exhaustive survey of paleobiological research on the functional morphology and paleoecology of ammonites (Jurassic and Cretaceous am- monoids) recently published by Kennedy 8 Cobban (1976) emphasized many aspects of the convergence between ammonites and fish, and also listed several aspects of am- monoid paleobiology that are relevant to un- derstanding Mesozoic pelagic ecoystems. These can be summarized as follows: ‘Present address: Department of Geological Sciences, University of Washington, Seattle, WA 91895, U.S.A. 4 WARD 1. All ammonites passed through a plank- tonic post-hatching stage, spending some un- known period in the surface regions of the seas, thus being affected by changing evolu- tionary conditions within the plankton. While in the plankton, hatchling ammonites may have fed on other zooplankton species, and in turn been prey of still others. In contrast to the hatchlings, juvenile and adult ammonites in- habited a variety of environments (based on interpretations of shell morphologies). Plank- tonic, nektonic, and nektobenthonic species have all been identified. 2. All living cephalopods are capable of mantle-powered swimming. Ammonites were probably no exception, and may have swum in a fashion very similar to that of Recent Nautilus. Studies on musculature and hydro- dynamic properties of the shell, however, suggest that the majority of species were relatively weak or slow swimmers. Some ammonite species (heteromorphic ammo- nites) had shell shapes that seem adapted to anything but swimming. The current opin- ion among ammonite specialists is that many species were very slow swimmers, and may have been better adapted for slow vertical excursions, rather than lifestyles necessitat- ing rapid horizontal swimming. 3. Most ammonites appear to have ex- ploited low levels in marine food chains, based on functional morphology of jaw parts, and exceptionally preserved fossils with crop contents still preserved. While all ammo- oids are thought to have been carnivores (like all extant cephalopods), many species have been interpreted as zooplankton feed- ers. The ammonites disappeared at the end of the Cretaceous Period. Their final extinction coincided with the extinction of other promi- nent groups of Mesozoic organisms, such as dinosaurs, other molluscs (most notably the majority of belemnites and all inoceramid and rudistid bivalves) and many groups of marine plankton. Extinction of the last remaining am- monite species came at the end of a long-term (10 to 15 million years) diversity decline. Dur- ing this last phase in the history of the ammo- nites, shell morphologies as well as charac- teristic evolutionary tempos of the constituent ammonite species were also changing (Ward, 1983; Ward 8 Signor, 1983). The purpose of this paper 1$ to better document the nature of the Cretaceous ammonite record in terms of shell morphologie. MATERIALS AND METHODS Ammonite shell morphology has been ex- amined and described in the following way. To describe shell shape, Cretaceous planispiral ammonite genera illustrated in the Treatise on Invertebrate Paleontology (Pt. 4) have been measured, and the W, D, and S parameters of Raup computed (Appendix 1). From these same specimens and from Cretaceous he- teromorphic genera from the same source, ornamental rugosity has been computed by dividing rib width by whorl height (Appendix 1). Based on these measurements, ammonite taxa have been placed in shell shape catego- ries, grouped according to one descriptor (or- nament rugosity or streamlining), or together (by assigning taxa to one shell shape cate- gory for each time period examined and then graphing the percentages of these distribu- tions). These graphs and table only provide a model for ammonoid shell shape distributions, and the limitations of the data cannot be un- derestimated. The generic даа base 1$ old, and the methodology of assigning a genus to an entire stage if it appears in any part of that stage greatly inflates the ranges of the taxa. Most ammonite genera ranged far less than a single stage. Secondly, the use of genera to construct shell shape distributions tends to give equal weight to all taxa. Obviously, some genera were far more speciose than others, so that diverse taxa are penalized. Thirdly, the data measured for a given genus are based on measurement of a single species of that genus; values for other species of the same genus could be different. In defense of the data base, however, it can be demonstrated that similar trends in shell shape distributions appear to be present at the species level of ammonite genera occurring in a single depo- sitional system (the Great Valley Sequence of California). These species have been tabu- lated and categorized as above. RESULTS 1. Generic level analyses A. Cretaceous ammonite ornamental rugosity—generic level Shell ornament rugosity has previously been discussed by Ward (1981), who quantitatively demon- CRETACEOUS AMMONITE SHELL SHAPES 5 strated that shell ornamental rugosity т- creased among ammonoid genera throughout the history of the group. That study was made at a system or series level of resolution. In this study | have examined the Cretaceous am- monite genera at the stage level. The meth- odology used here 1$ also somewhat different than in my previous study, in that here the rugosity measure is computed by dividing av- erage rib width (for non-major ribs, taken at mid flank) by whorl height, rather than by whorl diameter at the point of measure. Using this new methodology, heteromorphic as well as planispiral ammonites can be directly com- pared. The rugosity of combined heteromorphic and planispirally coiled genera for Cretaceous stages 15 listed in Table 1, and figured in Fig. 1. The four categories of rugosities are 0-.049, .05-.099, .10-.149, and greater or equal to .15. These are approximately equiv- alent to the four ornamental categories of Ward (1981). The combined groups as a per- centage of the entire fauna for each stage illustrated in Fig. 1 clearly show that there was a maximum of shell ornamental rugosity dur- ing the middle part of the Cretaceous, with minima at the start and end of the Cretaceous, a trend described earlier by Ward (1983). The major expansion of more coarsely orna- mented ammonoid genera began during the Hauterivian, and increased in numbers pro- gressively through the Turonian. The Albian, Cenomanian, and Тиготап all had ammonite faunas in which the majority of taxa belonged to the two highest ornamental categories. Fol- lowing the Turonian, coarsely ornamented forms gradually lessened in number (as a percentage of the entire fauna). By Campan- ian and Maastrichtian time, the two highest ornamental categories comprised less than 20% of the entire fauna. Ward (1983) suggested that Lower and Up- per Cretaceous heteromorphic ammonite fau- nas would be distinguished from one another on the basis of shell ornament. Shell orna- mental values for heteromorphic ammonite genera are listed in Table 2. From these data, it is clear that shell ornament rugosity in the Lower and Upper Cretaceous heteromorphic faunas 15 differentiable. B. Streamlining Streamlining in ammonites 15 related to both whorl profile and ornamentation. Raup (1967) showed there to be an inverse corre- lation between highly streamlined forms and degree of shell ornamental rugosity. Such a correlation is also observable in my data. The shell shapes best designed for high stream- lining efficiency are those with involute shells (low D) and especially low whorl breadth to height ratios (S), and nonexistent or only weak ribbing and tuberculation. To search for trends in streamlining during the Cretaceous, S values for the planispiral ammonite sample from Appendix 1 have been tabulated, and placed into categories of either 0-.49, .5-.99, 1.0-1.49, and greater than, or equal to 1.5. These data are listed in Table 3. The most highly streamlined forms, with very сот- pressed cross sections having S values less than .5, show maxima during the Hauterivian, Campanian, and Maastrichtian. The Maastric- tian shows the highest individual level of highly streamlined forms, with 29%. The Late Cretaceous trend of increasing proportions of highly streamlined forms appears to be real at the species level as well. | have recently been able to study types of many Maastrichtian ammonite species in European and North American museums; within these collections, the great abundances of highly streamlined species belonging to Sphenodiscus, Libyco- ceras, Hauericeras, and Pachydiscus are quickly evident, and in my mind indicate a significant trend in ammonite shell shapes. Other than the Hauterivian maximum, highly compressed ammonites appear to remain at approximately constant levels throughout the Lower Cretaceous, and then increase in num- bers during the Upper Cretaceous. Highly depressed ammonite shells are those with S values of greater than 1.5. These types of shells show maxima during the mid- dle part of the Cretaceous, and perhaps not surprisingly, many of the highly depressed species are also those with the coarsest shell ornament. The Maastrichtian also shows a high percentage of depressed forms, but this appears to be an artifact of the very small Maastrichtian sample, rather than a real trend. To compare Cretaceous streamlining effi- ciency with the set of Jurassic ammonites (Ward, 1980), percentages of shells having low О (О-.33), and either S values of less than .5, or between .5 and 1.0 have been computed for Lower, Middle and Upper Jur- assic and Lower and Upper Cretaceous (Ta- ble 3). From this table, it is apparent that the 6 WARD TABLE 1. Rugosity of shell ornament, Cretaceous planispiral and heteromorphic genera. Values in absolute numbers and percentages of total for each stage (parentheses). Values derived from division of rib width by whorl height. Stage n 0 -.049 .05 — .099 .10-.149 = 15 Maastrichtian 23 13(56) 6(26) 2(09) 2(09) Campanian 48 17(35) 21(44) 4(08) 6(12.5) Santonian 44 14(32) 18(41) 4(09) 8(18) Coniacian 45 13(29) 11(24) 8(18) 13(29) Turonian 43 12(30) 8(19) 8(19) 15(35) Cenomanian 55 18(33) 6(11) 15(27) 16(29) Albian 88 23(26) 13(15) 25(28) 27(31) Aptian 43 13(30) 12(28) 9(21) 9(21) Barremian 42 14(33) 10(24) 10(24) 8(19) Hauterivian 35 12(34) 12(34) 6(17) 5(14) Valanginian 37. 14(38) 14(38) 5(14) 4(11) Berriasian 24 7(29) 12(50) 2(08) 3(12.5) 5 я я я S $ в < я 5 Se fb eS zum. Soe oS eee eS 0: eS rn = 100 80 60 40 20 O 2 2 : : n=24 37 35 42 43 88 55 43 45 44 48 23 FIG. 1. Shell ornamental rugosity of Cretaceous ammonite genera listed in Appendix 1. For each stage, taxa have been placed in one of four categories. Shell rugosity was determined by dividing rib width by whorl height at the point of measure. These categories are then given a percent frequency value. N values at bottom of graph refer to number of taxa used in each stage. CRETACEOUS AMMONITE SHELL SHAPES 7 TABLE 2. Rugosity of shell ornament, Cretaceous heteromorphic ammonite genera. Because of the low number of taxa, values are in absolute numbers, rather than percentages. Stage n 0 —.049 Maastrichtian 9 4 Campanian 14 4 Santonian 10 2 Coniacian 7 1 Turonian 8 2 Cenomanian 9 1 Albian 12 1 Aptian 13 1 Barremian 18 2 Hauterivian 12 1 Valanginian 3 0 Berriasian 2 0 05.099 .10—.149 = 115 4 1 0 8 2 0 8 0 0 5 1 0 J 0 2 2 2 4 2 6 2 3 4 D 5 7 5 7 2 2 0 2 1 0 0 2 TABLE 3. The temporal distribution of streamlined ammonites. Time D=0=:335/= 105 Lower Jurassic 4.5% Middle Jurassic 10.4% Upper Jurassic 8.3% Lower Cretaceous 12.3% Upper Cretaceous 18.4% percentage of highly streamlined taxa was higher in the Cretaceous than any time in the Jurassic. C. Cretaceous shell shapes The nature of the Cretaceous ammonite shell categories can be better understood by combining the ornamental, streamlining, and heteromorphic categories into a single stage by stage compilation. To do this, four shell shape categories have been tabulated as a percentage of the total sample of each stage. Shells were placed in the category of orna- mental planispiral if shell rugosity was .15 or more. Shells were placed in the category of streamlined planispiral if they had S values of less than .5. The remainder were planispirals, or heteromorphic. These shape categories were then plotted as a percent-fraction of the entire sample for each stage. If a taxon ap- peared in any part of a stage, it was consid- ered to range throughout that stage. Shell shape categories for the Cretaceous stages are shown in Fig. 2. Frequencies of the different shapes are shown in Fig. 3. Per- DI=S0O=33 SD 210 1 and 2 combined 22.5% 27.0% 33.1% 43.5% 29.0% 37.0% 38.3% 54.2% 40.8% 59.2% haps the most significant aspect of this dia- gram is found within the category of orna- mented ammonites. As described above, coarsely ornamented ammonites comprised a minority of the ammonite faunas at the be- ginning and end of the Cretaceous. The Ber- riasian Stage, and to a lesser extent the Valanginian Stage as well, are composed of taxa that are continuations of Jurassic fami- lies. Almost all showed the Jurassic pattern of moderately ornamented, non-streamlined shell shapes. This overall shell shape pattern suggests that neither speed, nor shell defen- sive measures were necessary or advanta- geous in the pre-Cretaceous ammonite record. By the middle of the Cretaceous, however, and into the lower part of the upper Cretaceous, these coarsely ornamented forms became among the most diverse am- monite taxa. Beginning in the Coniacian Stage, and increasing in tempo in the San- tonian and Campanian Stages, the coarsely ornamented forms showed marked reduc- tions in diversity through extinction. By late Campanian and Maastrichtian times, this shell shape category was virtually non-exis- CRETACEOUS AMMONITE SHELL SHAPES (o) Berriasian Valanginian Hauterivian Barremian Albian Cenomanian Coniacian Santonian Aptian Maastrichtian Campanian я = я о я = E : heteromorphic Coarsely ornamented "N .‘planispiral O > 5 n-streamline AN! поп-огпате ИНИННИИИИ planispiral (OS RTS 60 99 92 63 29 FIG. 3. Shell shape categories for generic level data. All taxa were placed in one of four shell shape categories: streamlined, for planispiral shells with S values (whorl breadth divided by whorl height) less than 0.5; non-streamlined, non-ornamented planispirals, for planispiral shells with S values greater than 0.5, and shell ornament rugosities of less than 0.15; coarsely ornamented planispirals, with shell rugosity of greater than or equal to 0.15; and heteromorphic, for non-planispirally coiled shells. For each stage, the number of taxa (in value at bottom of graph) are placed in one of the shell shape categories, and the percent frequency for the entire stage sample calculated and graphed. tent. The reduction in the ornamented shell category can also be observed in other ways. Ward 8 Signor (1983) have recently exam- ined the nature of evolutionary tempo in Jur- assic and Cretaceous ammonites. In a com- panion piece (Ward 8 Signor, in press), we looked at the nature of the clade shapes. At the same time as overall ammonite diversity declined during the latest part of the Creta- ceous, the average shapes of the majority of family clades as replicated by spindle dia- grams also began to change. Characteristic clade shapes for the highly ornamented am- monites were short but wide, indicating high origination and high extinction rates. The ex- tinctions of the ornamented ammonites near the end of the Cretaceous left mainly long- ranging, low diversity ammonite families in the final stages of the Cretaceous. These groups included the Upper Cretaceous he- у FIG 2. Upper Cretaceous ammonites from the North Pacific Province, illustrating shell shape categories used in this paper. 1. Pseudoxybeloceras nanaimoense, a large Campanian heteromorphic ammonite. Shell ornamental rugosity between .06 and .08. 2. Three ammonites from the same concretion. Two specimens of the heteromorphic ammonite Hyphantoceras venustum, and a single planispiral Tetragonites popetensis. Note the short spines on the heteromorphs. Santonian of Mill Creek, California. 3. Bostrychoceras elonga- tum, a Santonian heteromorphic ammonite. This species has fairly coarse ornament for a heteromorph. О = .09 to .10. 4. Gaudryceras denseplicatum, a Santonian planispiral. This ammonite is neither stream- lined, nor ornamented. Ornamentation is very fine ribs. 5. Glyptoxoceras subcompressum, a Santonian through Maastrichtian heteromorph. 6. Desmophyllites diphyloides, a Campanian desmoceratid. This am- monite is neither highly streamlined, nor ornamented. 7a, b. Hauericeras gardeni, a highly streamlined ($ = .49) desmoceratid planispiral of the Santonian. 8. Epigoniceras epigonum, a Santonian and Campanian tetragonitid ammonite. 9. Submortoniceras chicoense, a coarsely ornamented planispiral from the Campan- ian. 10. Eupachydiscus haradai, a moderately ornamented planispiral from the Santonian. 10 WARD teromorph families, Nostoceratidae and Diplomoceratidae, and non-ornamented planispirals such as the Phylloceratidae, De- smoceratidae and Tetragonitidae. Heteromorphic ammonites are the other shell shape category that distinguishes the Cretaceous from all previous periods. Al- though hetermorphics are known for the Triassic and Jurassic as well, they occur at these times at very low diversity. Their diver- sifications during the Cretaceous set this sys- tem apart from the rest of ammonoid history. Although heteromorphic ammonites are here categorized as one shell grouping, the enormous range of morphology evolved by the disparate heteromorphic families sug- gests a wide range of adaptation. In this re- gard this category is probably much more artificial than either the streamlined or orna- mented planispiral categories. Within the Lower Cretaceous the more massive, heavily sculpted forms such as members of the An- cyloceratidae and Crioceratidae are ammo- nites that appear closely allied to the coarsely ornamented planispirals of the time, and in some cases are phylogenetically related (Wiedmann, 1969). Wiedmann's finding that planispiral and heteromorphic shell shapes freely transformed from one category to the other is further evidence that these two cate- gories, the coarsely ornamented heteromo- rophs and planispirals of the Aptian, Albian, and Cenomanian stages were ecologically al- lied. In contrast, the more finely sculpted, del- icate heteromorphic forms of the Upper Cre- taceous, initiating with the Hamitidae and con- tinuing with the Nostoceratidae and Diplomo- ceratidae, suggest different adaptations. These will be discussed further below. 2. The Cretaceous ammonite record of the Great Valley Sequence, California During the Late Jurassic and throughout the Cretaceous Period, subduction of ocean crust along the present day California coast- line produced the Sierran Arc. Thick wedges of clastic sediments were deposited along the continental shelf and slope west of this north- south trending arc. The stratal record of this event, the Great Valley Sequence, contains a rich ammonite record. The fossils are often preserved unaltered, and in many areas, field locales are locally abundant. All of the stages of the Cretaceous are represented, and am- monite zonation for the entire Cretaceous has been proposed (Ward 8 Signor, 1983). A. Stage level ornamentation rugosities Using the same method as for the generic level data, the Sacramento Valley ammonite species of the Valanginian to Maastrichtian Stages have been measured for S (whorl width divided by whorl height), and for orna- mental rugosity (rib width divided by whorl height at the point of measure). For the latter measure, the heteromorphic ammonite spe- cies assignable to Baculites have been omit- ted. Species of this genus are either non- ornamented, or can have crescent-shaped swellings on the flanks that are not really true ribs, even though this shell ornament, as in other ammonites, probably served to strengthen the shell. The results of the ornamental analyses, tabulated as for the generic data, are shown in Fig. 4 (from data in Appendix 3). Hetero- morphic as well as planispiral species have been included. As for the generic level data, the lower and upper Cretaceous stages show a lower degree of average shell ornamenta- tion than do the middle stages. The ammo- nites of the Aptian through Turonian of the Great Valley Sequence can be characterized as having a higher number of more coarsely ornamented species than did the ammonite faunas of earlier and later stages. The four shell shape categories used for the generic level data are also used to analyze the shell shape distributions of the species level data (Fig. 5). The only major difference between this figure for species of the Great Valley Sequence, and the figure for the ge- neric level data (Fig. 3) is in the category of streamlined planispiral ($ = .5). In the Great Valley Sequence, there appeared to be a rel- atively lower percentage of highly streamlined species than for the Cretaceous ammonite record as a whole. This may be due to the differing taxonomic levels of analysis (generic vs. species level), or may be real. It is my feeling that the Great Valley Sequence did have fewer, highly streamlined species rela- tive to other ammonite shell shapes at any given time, due to the deepwater facies and environments that characterized most Great Valley Sequence paleoenvironments. Most highly streamlined genera of the Cretaceous, such as Placenticeras and the numerous neo- ceratites, appear to have been restricted to CRETACEOUS AMMONITE SHELL SHAPES 11 Valanginian Hauterivian я = Я d я я is) a Aptian Albian 100 80 60 40 20 о п =23 2425, 14. 28 Cenomanian Turonian Coniacian Santonian Campanian Maastrichtian 2322222072020 FIG. 4. Shell ornamental rugosity for Great Valley Sequence species data. Ornamental categories as in Fig. 1. extremely shallow water (Kennedy & Cobban, 1976). These genera are completely absent from the Great Valley Sequence. B. Shell shape categories The stage-level percentages of major shell categories from the great Valley Sequence of California are shown in Fig. 5 (from data in Appendix 3). The three earliest Cretaceous stages in California have ammonite faunas that are more closely allied to Jurassic faunas than to those of the rest of the Cretaceous. The majority of taxa are from the ammonitidid families Olcostephanidae, Craspeditidae, and Berriasellidae. The dominant shell forms are non-streamlined, with prominent (but not coarse) ornament of finely incised ribs and small tubercles. In California the Berriasian, Valanginian, and Hauterivian Stages are best represented in the Paskenta region of the northwestern Great Valley, where strata are dominated by species of Neocosmoceras and Kilianiceras in the Berriasian, and large num- bers of Thurmanniceras in the Valanginian. This trend of large numbers of moderately sculpted, non-streamlined planispiral species continued into the early Hauterivian age. In the later Hauterivian, however, the seeds of the coming revolution in shape were already sown; early heteromorphic genera such as Crioceratites, Acrioceras, and the first Shas- ticrioceras signalled the end of the dominance of ammonitinid planispiral ammonites, which had made up nearly the entirety of ammonite faunas throughout the Jurassic and into the lowermost Cretaceous of California. During the Barremian and Aptian ages, he- teromorphic ammonite shapes were clearly increasing numerically in terms of the total spectrum of shell shapes present. Barremian heteromorphic ammonites included common gyroconic species of very large size, with planispiral shells of whorl expansion rates so 12 Hauterivian Barremian Albian Valanginian Aptian 100 80 60 40 Cenomanian non-streamlined, non-ornamented planispiral WARD Turonian Coniacian Santonian Campanian Maastrichtian hetero =z DE Streamlinec FIG. 5. Shell shape categories for Great Valley Sequence species. Shell shape categories as in Fig. 3. large that the outer shell wall was no longer in contact with previously secreted shell. Most important of these were specimens of Shas- ticrioceras, which were endemic to, but ex- tremely diverse in the North Pacific Province, and important in biostratigraphic zonation be- cause of their ubiquity and short stratigraphic ranges. Many of these species attained more than a meter in diameter. The second most abundant group of heteromorphs was the an- cylocerids, some of which also attained very large size. Aptian heteromorphic shell faunas were largely similar to those of the Barremian, but with the addition of large, hook-like shell species assigned to Hamulina and Anahamu- lina. Another important shape trend initiated in the Barremian-Aptian was the emergence of coarsely ornamental planispiral species, such as Pulchellia and Cheloniceras. These mas- sively armored species continued to increase in diversity in the Great Valley Sequence until the end of the Turonian. The Albian and Cenomanian ages, in con- trast to the Aptian, are widespread throughout the North Pacific Province; Albian and Ce- nomanian strata with ammonite faunas are known from Alaska, several areas in British Columbia, and in many areas of California, although the Ono region of northern California is by far the most complete and fossiliferous. No single ammonite shell group dominates Californian Albian-Cenomanian shell shape distributions. New additions to these faunas were by the successive radiations of desmo- ceratid and tetragonitid ammonites, to be di- verse components from the Albian until the very end of the Cretaceous. These species had smooth, extremely well-streamlined shells. Ornamented species also continued to diversify in the North Pacific Province. Among heteromorphic ammonites the crioceratid and ancylocerid species of the Barremian-Aptian were replaced by torticonic forms, such as Turrilites, and, for the first time, large numbers of delicately ornamented hamitids, assignable to Hamites and Stomohamites. The remainder of the Upper Cretaceous Stages in the Great Valley Sequence, the Coniacian, Turonian, Santonian, Campanian, and Maastrichtian, show three main trends. First, the large numbers of coarsely orna- mented species diminished in numbers fol- CRETACEOUS AMMONITE SHELL SHAPES 13 lowing the Turonian. Secondly, large numbers of small to medium sized heteromorphic am- monites, especially the baculitids, became in- creasingly important. Finally, streamlined planispirals increased in abundance until the end of the Maastrichtian. The Maastrichtian of the North Pacific realm is nowhere near as widely exposed as is the Campanian. Only Lower Maastrichtian strata have been definitely identified on the basis of molluscs, and much confusion still exists about the placement of the Campanian- Maastrichtian boundary in the North Pacific Province (see Jones, 1963, for a discussion), let alone the Lower-Upper Maastrichtian boundary. By Maastrichtian time ammonite diversity in the North Pacific Realm had dropped markedly. Only 14 Maastrichtian species are known from California as com- pared to 37 during the Campanian. The Maas- trichtian species are almost all holdovers from the Campanian. They make up an interesting assemblage of shape categories, for the mor- phologic make-up of the Maastrichtian ammo- nites sheds light on selective conditions ap- parently operating on ammonites immediately prior to their complete, world-wide, extinction at the end of the Maastrichtian. Of the 14 California Maastrichtian species, 10 are he- teromorphic (5 baculitids and 5 nosto- and diplomoceratids). 5. Numerical abundance of North Pacific Cretaceous shell shapes The diversity of specific shell shapes dis- cussed above for each stage of the California Cretaceous does give a picture of evolution- ary trends occurring within ammonite commu- nities of the North Pacific Province. Of equal interest is information about the relative abun- dance of each taxon. Biofacies distributions of Cretaceous am- monites from various times and places around the globe have been studied by Scott (1940), Kauffman (1967) and Kennedy & Cobban (1976), and most recently by Tanabe (1979). Tanabe's paper is significant in its thoroughness and detail, as well as being the only study of Cretaceous faunas similar to those discussed in this paper. Tanabe fol- lowed the works of Ziegler (1967) and Wenat (1971) in sampling individual outcrops, and tabulating the percentage of each species or morphotype within the sample. | have made similar studies on selected outcrops of North Pacific Cretaceous strata in California, Washington, and British Columbia (Fig. 6). Although these studies have been very few, they do illustrate trends in abun- dances of various Cretaceous shell shapes through time. Abundance trends from absolute numbers of individual species show the same trend illustrated by diversity trends: heteromorphic ammonites became increasingly abundant (as a percentage of the entire ammonite fauna at any locality) as well as increasingly diverse through time. In the late Cretaceous the most abundant of all ammonites were species of Baculites. The dominance of baculitid species in vir- tually all marine, Late Cretaceous facies of California, Washington, and southeastern British Columbia is in marked contrast to the facies patterns of similarly aged strata in Ja- pan. Matsumoto (1960) first noted that abun- dance patterns in California and Japan showed significant differences: “The apparent difference may be due to ecological or sedi- mentary environment. Members of the Tet- ragonitidae, such as Tetragonites, An- agaudryceras, Gaudryceras, and some of the Desmoceratidae, such as Desmoceras (Pseudouhligella), Damesites, and Me- sopuzosia are persistent and occur abun- dantly in various stages of the Japanese Up- per Cretaceous. They do occur in California but the occurrence is sporadic and restricted to particular beds. The Baculitidae are fairly common and occur at various levels of the Upper Cretaceous of California, being often predominant over other ammonites. In the Japanese Cretaceous they are not rare but never so abundant as the desmoceratids and tetragonitids.” Matsumoto attributed these distribution dif- ferences to facies differences between Cali- fornia and Japan. Baculites 1$ the dominant ammonite in virtually all ammonitiferous, post- Turonian strata in California and British Co- lumbia, irrespective of facies. Only strata of earliest and latest Santonian age (Venustum and Elongatum Zones, and Schmidti Zone) have facies where baculitids are not the nu- merically dominant species. If the Upper Cre- taceous of Japan can be characterized as being composed of high diversity, high equi- tability ammonite faunas, with tetragonitids and desmoceratids as the most abundant species, then this is a marked contrast with the Californian Upper Cretaceous. The Cre- 14 AGE Middle Campanian Early Campanian Late Conlaclan or Early Santonian Late Albian LOCALITY Sucia Island, Washington Trent River, Vancouver Island Cache Creek, Colusa County California McCarty Creek, Tehama County, WARD FACIES DOMINANT GENERA Baculites Canadoceras Hoplitoplacenticeras concretionary siltstone mudstone Baculites Canadoceras Gaudryceras Baculites Protexanites Puzosia Desmoceras California McCarty Creek, Tehama County, Callfornla Valanginlan turbidites Hamites Lytoceras Thurmanniceras FIG. 6. Shell shapes and dominant genera from selected Cretaceous outcrops, Great Valley Sequence and Nanaimo Group, Vancouver Island, Canada. P = planispiral, H = heteromorphic. taceous record in the Great Valley Sequence shows a steady increase in the numbers of heteromorphic as compared to planispiral am- monites. During the Late Cretaceous the he- teromorphic ammonites are by far the most abundant species in most Cretaceous out- crops of California, Washington, and British Columbia. DISCUSSION The shell shapes themselves of the North Pacific ammonites can be used as a clue to mode of life. Raup (1967) and Chamberlain (1976, 1980) examined the spectrum of evolved ammonoid shell shapes, and identi- fied streamlining and stability as major adap- tive influences on shell shape. Perhaps one of the most surprising findings of these studies was the presence of so many ammonite spe- cies which were so poorly streamlined. For many species, evidently, rapid swimming speeds or great maneuverability, which would be possible with weakly ornamented, com- pressed shells of high stability, were not of overriding selective value in their particular habitats. This generalization is especially ev- ident in reviewing the evolutionary history of heteromorphic, or non-planispirally coiled am- monites. These forms were anything but streamlined (although most had very high shell stability). Many workers have consid- ered heteromorphic species as secondary ben- thonic adaptations, ammonites which became benthonic in ways analogous to living octo- pus. Certainly this argument holds for the massive heteromorphs of the Lower Creta- ceous. The greatest single argument against this hypothesis, however, lies in the morphol- ogy of the heteromorphic shells themselves, which have lost none of the components of the buoyancy maintenance system of all other CRETACEOUS AMMONITE SHELL SHAPES 15 ammonites, and indeed have undergone ad- aptations of the shell which appear to have increased efficiency of the shell as a buoy- ancy organ (Ward, 1979; 1983; Klinger, 1980). Packard (1972) has argued that many heteromorphic ammonites were passive, nearly planktonic organisms that lived in the mesopelagic regions in ways analogous to the present day cranchiid squids. This view was reiterated by Ward (1976), Ward 8 We- stermann (1977), Ward (1979), and Klinger (1980). If this generalization is true for a ma- jority of heteromorphic species, it would sug- gest that a major evolutionary trend in ammo- nites of the North Pacific Cretaceous was towards increasing diversities and abun- dances of mesopelagic species. An important breakthrough in interpreting the mode of life and living environment of ammonites was the discovery by Wester- mann (1971) that relative siphuncle strength in ammonites could be used as a measure of overall shell strength, and hence be useful as a key to bathmetry. This view has been chal- lenged (Saunders & Wehman, 1977; Cham- berlain & Moore, 1983); however, it is my belief that this measure may indeed yield in- formation on maximum, potential depth limits (Westermann 8 Ward, 1980), if not in terms of strength, than perhaps in terms of siphuncular water movement (Ward, 1983). Westermann showed that ammonitinid am- monites, which made up the bulk of Jurassic ammonite faunas, had large diameter, thin- walled siphuncles, and they may be restricted to shelf environment of one to two hundred meters maximum depth because of the potential weakness against ambient pres- sure explosion of these structures. These siphuncles contrasted markedly to supposed siphuncular strengths of phylloceratids and lytoceratids, and to that of Nautilus, all of which show a siphuncular strength potential of at least 700 m. Since Westermann's stud- ies, other siphuncular diameters have been measured, including those for a number of species of the North Pacific Province Creta- ceous ammonites (Ward & Westermann, 1977; Tanabe, 1979; Ward 8 Signor, 1983). Based on generalizations from published ammonitonid siphuncle strength figures, all ammonitid species of the California Berrias- ian, Valanginian, and Hauterivian should be assigned to shallow water habitats, with max- imum submersion limits of perhaps 100 to 200 m (Westermann, 1971). In these stages, only the lytoceratids and phylloceratids may have been deepwater forms. Interestingly, Tanabe (1979) showed that all coarsely ornamented ammonites he examined also showed simi- larly weak siphuncles, and thus, perhaps, re- striction to shallower water. Certainly the fa- cies distributions of these ammonites support such a generalization (Tanabe, 1979). Other shallow water species (based on siphuncular strength) include the Upper Cretaceous Placenticeratidae, which in the North Pacific are represented by the Campanian genera Metaplacenticeras and Hoplitoplacenticeras. Lytoceratids, phylloceratids, desmoceratids, and tetragonitids have had much stronger siphuncles (Ward 8 Signor, 1983) and thus the capability to inhabit much deeper habitats, perhaps to depths as deep as 1000 m. Many of these species may have been shallow wa- ter species by choice; nevertheless, the ca- pability of deeper habitats was present. Heteromorphic shells provide additional problems of interpretation, and should be ex- amined on a case by case basis, rather than be categorized as one large group. Clearly, large, ornamented heteromorphic species of the lower Cretaceous, such as Ancyloceras, Heteroceras, and Toxoceras seem much more analogous to the large, coarsely orna- mented planispiral ammonites of the lower and middle Cretaceous than they do to their less robust, finely sculpted descendents of the Upper Cretaceous. Siphuncle strength measurements of these Lower Cretaceous heteromorphic ammonites (Westermann, 1982; Ward & Signor, 1983) also show them to be similar to the ornamented ammonites in having large but thin-walled, and thus weak siphuncles. The Upper Cretaceous heteromor- phs, such as Baculites, Hyphantoceras, Bos- trychoceras, and Glyptoxoceras had stronger siphuncles (Ward & Signor, 1983) and are here interpreted as mesopelagic forms, while the more massive Lower Cretaceous forms may have been nektobenthonic, or perhaps entirely benthonic. The major conclusion to be drawn from the published strength figures is that shallow wa- ter species, such as the early Cretaceous ammonitinids, and mid to upper Cretaceous coarsely ornamented species became т- creasingly restricted in number during the Cretaceous, and were completely replaced by deeper water species (such as nostoceratids, phylloceratids, desmoceratids, tetragonitids) by the latest Cretaceous. It may be that the 16 WARD absolute number of ammonites living in shal- low water habitats never changed during the Cretaceous, or perhaps even increased. lt 1$ apparent, however, that even if ammonites inhabited the same shallow water habitats throughout the Cretaceous, by Late Creta- ceous time many were capable of inhabiting deeper habitats as well. It is my feeling that shallow water, nektonic ammonites gradually decreased in diversity and abundance throughout the Cretaceous. The ammonites may have moved off the shallow shelves in the late Cretaceous, either to deeper water benthic habitats or into the water column. Those that remained in shallow waters, such as the Placenticeratidae and Sphenodiscidae, were all highly streamlined. CONCLUSION Evolution of Cretaceous ammonites ap- pears to show several trends. Early Creta- ceous faunas are dominated by shallow water species, which were gradually replaced by deeper water species during the Cretaceous. Those species which continued to live in shal- low water either became coarsely orna- mented, with protective ribbing and spines, or evolved shells of high streamlining efficiency. By latest Cretaceous time shallow water spe- cies made up only a small fraction of the total ammonite fauna. The increase in diversity and abundance of deep water planispiral forms (Phylloceratidae, Desmoceratidae, Tetragonitidae) and he- teromorphic forms (Nostoceratidae, Baculit- idae, Diplomoceratidae) suggests that ammo- nites migrated from their Jurassic-Lower Cre- taceous shelf habitats into deeper benthic or mesopelagic habitats. This trend could have been due to increased competition in shallow shelf areas by newly evolving teleost fishes and/or dibranchiate cephalopods, or increas- ing predation in shallow water by teleosts, sharks, shell-crushing crustaceans, or marine reptiles. The Cretaceous evolution of heteromorphic ammonites distinguishes the Cretaceous am- monite faunas from those of all preceding systems. | believe that the high diversity and abundance of heteromorphic ammonites is a signal that fundamental reorganization was occurring within the Cretaceous pelagic realm. | would interpret many of the Late Cretaceous heteromorphic species as me- sopelagic zooplankton feeders (Ward & We- stermann, 1977; Ward, 1979). Within this context, the final extinction of the ammonites at the end of the Cretaceous may be due in part to environmental changes within the plankton. By Maastrichtian time all North Pacific shallow water ammonites were already extinct. All remaining species may have been deep-water forms, and most were heteromorphic. If those heteromorphic forms were indeed vertically migrating, mesopelagic zooplankton feeders, then they would have been very susceptible to the catastrophic changes within the zooplankton communities that occurred at the end of the Cretaceous either as adults, or in their planktonic hatch- ling stages. It is perhaps no coincidence that the only externally shelled cephalopods that survived the terminal Cretaceous extinction were nautilids, which are high in the food chain (large crustacean carnivores) without planktonic larval stages. REFERENCES CITED CHAMBERLAIN, J., 1976, Flow patterns and drag coefficients of cephalopod shells. Palaeontology, 19: 539-563. CHAMBERLAIN, J., 1980, Hydromechanical de- sign of fossil cephalopods. In: HOUSE, M. & SENIOR, J., eds., The Ammonoidea, Academic Press, p. 289-336. CHAMBERLAIN, J. A. Jr., & MOORE, W. A. Jr., 1982, Rupture strength and flow rates of Nautilus siphuncular tubes. Paleobiology, 8: 408—425. JONES, D., 1963, Upper Cretaceous ammonites from southern Alaska. [United States] Geological Survey Professional Paper, 432: 53 p. KAUFFMAN, E. G., 1967, Coloradoan macroinver- tebrate assemblages, central Western Interior United States. In KAUFFMAN, E. G. & KENT, H. G., eds., Paleoenvironments of the Cretaceous Seaway—a symposium. Colorado School of Mines, р. 67-143. KENNEDY, W. J. 8 COBBAN, W. A., 1976, Aspects of ammonite biology, biogeography, and biostrat- igraphy. Special Papers in Paleontology, 17:94 p. KLINGER, H., 1980, Speculations on buoyancy control and ecology in some heteromorph am- monites. In HOUSE, M. & SENIOR, S., eds., The Ammonoidea, Academic Press, р. 337-355. MATSUMOTO, T., 1960, Upper Cretaceous ammo- nites of California, Pt. Ш. Memoirs of the Faculty of Science Kyushu University, ser. D: 91-171. PACKARD, A., 1972, Cephalopods and fish, the CRETACEOUS AMMONITE SHELL SHAPES A, limits of convergence. Biological Reviews, 47: 241-307. RAUP, О. M., 1967, Geometric analysis of shell coiling: coiling in ammonoids. Journal of Paleon- tology, 40: 1178-1190. SAUNDERS, B. & WEHMAN, D., 1977, Shell strength of Nautilus as a depth limiting factor. Paleobiology, 3: 83-89. SCOTT, G., 1940, Paleoecological factors control- ling the distribution and mode of life of Creta- ceous ammonoids in the Texas area. Journal of Paleontology, 14: 299-323. TANABE, K., 1979, Paleoecological analysis of am- monoid assemblages in the Turonian Scaphites facies of Hokkaido, Japan. Palaeontology, 22: 609-630. WARD, P., 1976, Upper Cretaceous ammonites (Santonian-Campanian) from Orcas Island, Washington. Journal of Paleontology, 50: 454—461. WARD, Р., 1979, Functional morphology of Creta- ceous helically-coiled ammonite shells. Pal- eobiology, 5: 415-422. WARD, P., 1981, Shell sculpture as a defensive adaptation in ammonoids. Paleobiology, 7: 96-100. WARD, P., 1983, The extinction of the ammonites. Scientific American, 249: 136-147. WARD, P. & SIGNOR, P., 1983, Evolutionary tempo in Jurassic and Cretaceous ammonites. Paleobiology, 9: 183-198. WARD, P. & WESTERMANN, G., 1977, First oc- currence, systematics, and functional morphol- ogy of Nipponites from the Americas. Journal of Paleontology, 51: 367-372. WENDT, J., 1971, Genese und Fauna submariner sedimentárer Spallenfúllungen im mediterranen Jura. Palaeontographica, A136: 121-192. WESTERMANN, С. Е. G., 1971, Form, structure and function of shell and siphuncle in coiled Mesozoic ammonoids. Life Sciences Con- tributions, Royal Ontario Museum, no. 78: 39 p. WESTERMANN, G. E. G., 1982, The connecting rings of Nautilus and Mesozoic ammonoids: im- plications for ammonoid bathymetry. Lethaia, 15: 373-384. WESTERMANN, С. Е. С. 8 WARD, P., 1980, Sep- tum morphology and bathymetry in Cepha- lopoda. Paleobiology, 6: 48-50. WIEDMANN, J., 1969, The heteromorphs and ат- monite extinction. Biological Reviews, 44: 563—602. , ZIEGLER, B., 1967, Ammoniten-Okologie am Beispiel des Ober-Jura. Geologische Runds- chau, 56: 439-464. APPENDIX 1 Geometry and ornament of Cretaceous ammonite Taxon LYTOCERATIDAE Lytoceras Suess Pterolytoceras Spath Eulytoceras Spath Metalytoceras Spath Ammonoceratites Rafinesque Argonauticeras Anderson Pictetia Uhlig Megalytoceras Buckman Metrolytoceras Buckman PROTETRAGONITIDAE Protetragonites Hyatt Leptotetragonites Spath Hemitetragonites Spath TETRAGONITIDAE Eogaudryceras Spath Eotetragonites Breistroffer Anagaudryceras Shimizu Mesogaudryceras Spath Zelandites Marshall Parajaubertella Matsumoto Vertebrites Marshall Gaudryceras De Grossouvre Kossmatella Jacob Gabbioceras Hyatt Tetragonites Kossmat genera. Age W D S Or Sin-UpCr 1.67 .42 .9 Tith-Val 2.25 .44 1.2 .0 Haut-Bar 1.93 4 .67 .04 Val Apt-Cen 2.25] 42 USA .03 Apt-Alb 242 .29 1.18 .03 Alb 10.94 .35 1.14 .02 Ba) 7.84 .36 .64 .02 2.35 39 .02 Tith-Val 1.87 .54 1.08 .02 Ber-Val 2.07 ES .68 .02 Haut-Alb 2233 .41 1.06 .02 Bar-Alb 2.14 32 1515 .02 Bar-Alb 2.01 .41 1.30 .02 Alb-Maas 3.51 .41 1.10 .03 Ceno Swe 137 1.59 .03 Alb-Maas 2.71 .21 .59 .02 Сепо 1.49830 2.04 Sant-Maas 1.89 .36 .79 Tur-Maas 1253 57 2.78 .03 Alb-Ceno 1.89 .36 .79 Apt-Ceno 2.51 25 2.27 Alb-Ceno РЗ 35 1.38 18 APPENDIX 1 continued WARD Taxon Age Epigoniceras Spath Tur-Maas Pseudophyllites Kossmat Camp-Maas PHYLLOCERATIDAE Phylloceras Suess Sin-Val Partschiceras Fucini Sin-Bar Phyllopachyceras Spath Bar-Maas Hypophylloceras Salfield Haut-Maas Calliphylloceras Spath Hett-Alb Holcophylloceras Spath Baj-Apt CRASPEDITIDAE Subcraspedites Spath Infravalang Paracraspedites Swinnerton Infravalang Platylenticeras Hyatt Valang Tolypeceras Hyatt Valang Proleopoldia Spath Neocom Paquiericeras Sayn Valang Temnoptychites Pavlow L. Neocom Tollia Pavlow Infravalang Praetollia Spath Infravalang Hectoroceras Spath Infravalang OLCOSTEPHANIDAE Spiticeras Uhlig S. (Negreliceras) Djanlidze Groebericeras Leanza Aspidostephanus Spath Olcostephanus Neumayr O. (Rogersites) Spath Subastieria Spath Parastieria Spath Saynoceras Munier-Chalmas Polyptychites Pavlow P. (Euryptychites) Pavlow Valanginites Kilian Dichotomites Koenen Neocraspedites Spath Speetoniceras Spath Simberskites Pavlow Craspedodiscus Spath Subthurmannia Spath Subalpinites Mazenet BERRIASELLIDAE Berriasella Uhlig Pseudargentiniceras Spath Argentiniceras Spath Thurmanniceras Cossmann Neocomites Uhlig Odontodiscoceras Spath Calliptychoceras Spath Lissonia Gerth Cuyaniceras Leanza Limaites Lisson Parandiceras Spath Frenguelliceras Leanza Lyticoceras Hyatt Favrella Douvillé Neocosmoceras Blanchet Kilianella Uhlig Sarasinella Uhlig U. Jur (U. Tithon) L. Cret (Berrias) L. Cret (Berrias) L. Cret (Berrias) L. Valang U. Valang L. Haut L. Haut U. Valang L. Valang U. Valang U. Valang U. Valang-L. Haut U. Valang-Haut L. Haut L. Haut U. Haut L. Neocom Berrias Tithon-Ber U. Tithon-Ber Berrias Berrias-Val Berrias-Val Valang Valang Berrias Berrias Berrias-Val Valang Valang Valang-Haut L. Haut Berrias Berrias-Val L. Valang CRETACEOUS AMMONITE SHELL SHAPES 19 APPENDIX 1 continued Taxon Age W D S Or Neohoploceras Spath Valang 1.56 .25 1.29 .08 Wichmanniceras Leanza Valang 1.67 .59 1.33 al Distoloceras Hyatt U. Valang-Haut 1.89 .36 .89 .14 Acanthodiscus Uhlig L. Haut 2.25 .29 1.2 112 Leopoldia Mayer-Eymer U. Valang-Haut 3.11 23 oi al Saynella Kilian L. Haut 2.64 15 .41 Hatchericeras Stanton L. Haut 2.60 ay, .58 Suboosterella Spath L. Haut 2.25 .38 .47 113 Delphinites Sayn Valang 2.42 2j 73 0 OOSTERELLIDAE Oosterella Kilian U. Valang 2.91 .28 .48 Pseudoosterella Spath U. Valang 2.15 .41 285 .08 DESMOCERATIDAE Eodesmoceras Spath Valang-Bar 2.12 .38 1 Е. (Miodesmoceras) Wright Bar 1.62 25 .71 Barremites Kilian L. Haut-Bar 2.19 ale) .38 В. (Raspailiceras) Wright Haut-Bar 2.10 x А В. (Barremites) Ваг 2.19 15 .38 Subsaynella Spath U. Haut-Bar 3.06 .14 15 .04 Valdedorsella Breistroffer L. Haut-Apt 1.96 .29 1227 .07 Pseudohaploceras Hyatt Bar-Apt 2.16 .36 .05 Callizoniceras Spath U. Bar 1.78 .19 .54 112 Melchiorites Spath L. Apt-Alb 2.37 .25 .8 Puzosia Bayle Е. Alb-U. Tur 2.25 -39 .8 Ki] P. (Anapuzosia) Matsumoto L. Alb-Cen 1.44 .47 Bhimaites Matsumoto U. Alb-Cen Pip 32 .84 Lytodiscoides Spath U. Alb 2.25 Si Hedi .07 Pachydesmoceras Spath U. Alb-Tur 2.40 .29 .04 Silesitoides Spath L. Alb-M. Alb 1.47 .53 .04 Jimboiceras Matsumoto Turon-L. Sant 1.78 .38 1 .06 Parapuzosia Nowak U. Cenom-Camp 1.99 35 .06 Mesopuzosia Matsumoto Tur-Coni 1 .26 .86 .08 Kitchinites Spath Santon-Camp 2.40 .32 .65 К. (Neopuzosia) Matsumoto Santon-Camp .32 .07 Zurcherella Casey U. Bar-U. Apt 3.33 .26 .70 .07 Uhligella Jacob U. Apt-M. Alb 1.25 .26 1 .09 Pseudosaynella Spath L. Apt-L. Alb 2.39 12 37 Beudanticeras Hitzel L. Alb-U. Alb 2.19 .18 .5 Desmoceras Zittel U. Apt-Cenom 2.42 21 1.18 D. (Pseudouhligella) Matsumoto U. Alb-Tur 2.83 .19 all Tragodesmoceroides Matsumoto Tur 2.04 13 1.07 Damesites Matsumoto Cenom-Camp 2.25 VE .69 Desmophyllites Spath Santon-Camp 1.92 13 .49 Hauericeras De Grossouvre Coni-Maas 1.59 .29 .46 HOLCODISCIDAE Spitidiscus Kilian Haut 2.89 .29 125 Plesiospitidiscus Breistroffer U. Haut 2.10 .21 .65 Holcodiscus Uhlig Bar EN .34 .89 .05 Metahoplites Spath Bar 3.16 .19 UT Astieridiscus Kilian Bar 2.85 .30 1 .07 SILESITIDAE Silesites Uhlig Bar 1.92 .44 15 .08 Neosilesites Breistroffer U. Apt-L. Alb 1.89 155 KOSSMATICERATIDAE Hulenites Matsumoto U. Alb 2.25 .29 .06 Holcodiscoides Spath Tur 2.32 .38 1.05 .06 Yokoyamaoceras Wright/Matsumoto Tur-Coni 2.37, 3 .86 20 APPENDIX 1 continued Taxon WARD Kossmaticeras De Grossouvre K. (Natalites) Collignon Grossouvreites Kilian & Reboul Gunnarites Kilian 8 Reboul Maorites Marshall Pseudokossmaticeras Spath Neograhamites Spath Jacobites Kilian & Reboul Brahamites Kossmat PACHYDISCIDAE Eopachydiscus Wright Lewesiceras Spath Pseudojacobites Spath Pachydiscoides Spath Nowakites Spath Canadoceras Spath Patagiosites Spath Anapachydiscus Yabe & Shimuzu Pachydiscus Zittel Menuites Spath Pseudomenuites Matsumoto Eupachydiscus Spath Bayleites Collignon Tragodesmoceras Spath MUNIERICERATIDAE Muniericeras De Grossouvre PULCHELLIIDAE Nicklesia Hyatt Pulchellia Uhlig Coronites Hyatt Subpulchellia Hyatt Psilotissotia Hyatt Lopholobites Hyatt TROCHLEICERATIDAE Trochleiceras Fallot & Termier DOUVILLEICERATIDAE Paraspiticeras Kilian Procheloniceras Spath Roloboceras Casey Cheloniceras Hyatt C. (Epicheloniceras) Casey Diadochoceras Hyatt Parahoplites Anthula Acanthohoplites Sinzow Paracanthoplites Stoyanow Hypacanthoplites Spath Gargasiceras Casey Colombiceras Spath Douvilleiceras De Grossouvre Astiericeras Parona & Bonarelli DESHAYESITIDAE Deshayesites Kazansky Dufrenoyia Burckhardt ENGONOCERATIDAE Knemiceras Böhm Parengonoceras Spath Engonoceras Neumayr & Uhlig Protengonoceras Hyatt U. Tur-Camp U. Camp Camp Camp Camp U. Camp-Maas Camp Camp Maas U. Alb U. Cenom U. Turon Coni-Santon Coni-Santon U. Santon-Camp U. Camp-Maas Coni-Maas Camp-Maas U. Santon-Camp Camp Coni-Camp Santon-L. Camp L. Turon-Coni Coni Bar U. Haut-U. Bar Bar Bar-Apt U. Haut-Bar Bar U. Apt-L. Alb Bar L. Apt-U. Apt L. Apt U. Apt U. Apt U. Apt U. Apt U. Apt-L. Alb L. Alb-M. Alb Low M. Alb L. Apt-U. Apt U. Apt M. Alb-U. Alb L. Alb-M. Alb M. Alb-Cenom M. Alb CRETACEOUS AMMONITE SHELL SHAPES 21 APPENDIX 1 continued Taxon Age W D 5 Ог Metengonoceras Hyatt U. Alb 2.07 .07 Epengonoceras Spath Cenom-L. Turon 2:31 alo .58 Neolobites Fischer Cenom 2:25 .12 .43 .05 PLACENTICERATIDAE Proplacenticeras Spath Cenom-Con 232 .09 53 Metaplacenticeras Spath Santon-Camp .06 Placenticeras Meek U. Santon-L. Camp 1.64 .16 .59 Stantonoceras Johnston U. Santon-L. Camp 1275 22 1403 105 Diplacmoceras Hyatt L. Camp 2.92 ali 45 Haresiceras Reeside U. Santon 2.94 .08 .64 Hoplitoplacenticeras Spath L. Camp-Maas 2.10 Al 1.04 1174 LYMERIELLIDAE Proleymeriella Breistroffer L. Alb 2.16 .28 .78 .13 Leymeriella Jacob L. Alb-M. Alb 2.16 .36 .94 .25 Epileymeriella Breistroffer L. Alb 3.06 29 8 ali Aioloceras Whitehouse U. Apt-L. Alb 2.51 18 .65 HOPLITIDAE Cleoniceras Parona 8 Bonarelli L. Alb-M. Alb 2.54 12 .42 С. (Neosaynella) Casey L. Alb 2.42 sn 52 Puzosigella Casey Up. Е. Alb 2.10 .21 uke .075 Farnhamia Casey L. Alb 1.89 en Tetrahoplites Casey L. Alb 1.86 37 1.32 29 Pseudosonneratia Spath L. Alb-M. Alb 1.64 .38 .95 At Hoplites Neumayr M. Alb 1.91 2 .69 .14 Anahoplites Hyatt М. Alb-U. Alb 2.64 alo 05 ss) Epihoplites Spath M. Alb-U. Alb 2.66 .29 1105 A Discohoplites Spath U. Alb 2.07 .30 .69 Hyphoplites Spath U. Alb-L. Cenom 2135 122 .56 .21 Sonneratia Bayle Up. L. Alb 2.32 ‚Si .45 li Tetrahoplitoides Casey L. Alb JO ‚25 Protohoplites Spath Up. L. Alb-Low M. Alb 2.09 :39 1.65 P. (Hemisonneratia) Casey M. Alb 251 .26 1.14 2 Otohoplites Steinmann Up. L. Alb-L. M. Alb 2.25 El .96 .19 Dimorphoplites Spath М. Alb-U. Alb 2.12 .31 .86 .16 Lepthoplites Spath U. Alb 1.96 125 .69 .08 Pleurohoplites Spath U. Alb 2.01 32 9 .09 P. (Arrhaphoceras) Whitehouse U. Alb 2.35 .30 ES Al Cyamhoplites Spath L. Alb 2.32 Sil .41 .14 Lemurcoceras Spath L. Alb 1.99 .38 1:3 Arcthoplites Spath L. Alb 1.99 .26 .91 .2 Gastropolites McLearn M. Alb 1.96 .19 .76 .14 SCHLOENBACHIIDAE Schloenbachia Neumayr Up. U. Alb-U. Cenom 1.96 .26 .96 112 Euhystrichoceras Spath L. Cenom 1.65 33 Toute Prionocycloides Spath Cenom 2.25 .2 .67 Tropitoides Spath Cenom 2.32 .06 153 .07 Prohauericeras Nowak Turon 2.19 .18 .46 FORBESICERATIDAE Forbesiceras Kossmat Cenom 2.91 .48 .03 BRANOCERATIDAE Eubrancoceras Breistroffer Up. L. Alb-L. M. Alb 2.33 .38 .83 .25 E. (Parabrancoceras) Breistroffer L. Alb 1.62 .43 1 Brancoceras Steinmann Up. L. Alb-M. Alb 2.15 .36 .86 «16 Hysteroceras Hyatt Up. M. Alb-L. U. Alb 1.89 .41 .85 .25 Mojsisoviczia Steinmann M. Alb 1.65 .25 .46 ce. Venezoliceras Spath Up. M. Alb 1.6? 25 .65 Oxytropidoceras Sticler Up. L. Alb-M. Alb 3.06 Pi .64 .06 Dipoloceras Hyatt Up. M. Alb-L. U. Alb 2.06 .33 1-55 .08 APPENDIX 1 continued Taxon Mortoniceras Meek M. (Dieradoceras) Van Hoepen M. (Durnovarites) Spath M. (Cantabrigites) Spath Neokentroceras Spath Aresoceras Van Hoepen Cainoceras Van Hoepen Prohysteroceras Spath Neoharpoceras Spath Spathiceras Whitehouse FLICKIIDAE Flickia Pervinquiere Ficheuria Pervinquiere Adkinsia Böse Prolyelliceras Spath LYELLICERATIDAE Lyelliceras Spath Tegoceras Hyatt Neophlycticeras Spath Stoliczkaia Neumayr Budaiceras Böse Salaziceras Breistroffer Mantelliceras Hyatt M. (Cottreauties) Collignon Sharpeiceras Hyatt Acompsoceras Hyatt Calycoceras Hyatt Paracalycoceras Spath Eucalycoceras Spath Acanthoceras Neumayr Neosaynoceras Breistroffer Euomphaloceras Spath Kanabiceras Reeside & Weymouth Romaniceras Spath Protacanthoceras Spath Dunveganoceras Warren & Stelck Utaturiceras Wright Metoicoceras Hyatt Watinoceras Waren Benueites Reyment Mammites Laube & Bruder Kamerunoceras Reyment Pseudaspidoceras Hyatt Metasigaloceras Hyatt Borrissjakoceras Arkangelsky BINNEYITIDAE Binneyites Reeside VASCOCERATIDAE Spathites Kummel & Decker Gombeoceras Reyment Paravascoceras Furon Vascoceras Choffat Paramammites Furon Fagesia Pervinquiere Thomasites Pervinquiere Neoptychites Kossmat WARD Age Up. M. Alb-Up. Alb Low. U. Alb Up. U. Alb Up. U. Alb Low. U. Alb Low. U. Alb Low. U. Alb Low. U. Alb Up. U. Alb Top. U. Alb-L. Cenom U. Alb-L. Cenom U. Alb-L. Cenom L. Cenom L. Alb L. Alb-M. Alb L. Alb-M. Alb M. Alb-C. Alb U. Alb-L. Cenom L. Cenom U. Alb L. Cenom L. Cenom Cenom L. Cenom Cenom Cenom U. Cenom-basal Tur Up. L. Cenom-U. Cenom L. Cenom U. Cenom U. Cenom-L. Tur U. Cenom U. Cenom U. Cenom U. Cenom L. Turon L. Turon L. Turon Turon L. Turon L. Turon L. Turon U. Cenom-Tur Coni . Turon . Turon . Turon Turon Turon Turon . Turon . Turon — ine) a — — aren ро —b 1 4 md À DELL wo — — © © © .22 CRETACEOUS AMMONITE SHELL SHAPES APPENDIX 1 continued Taxon Age W TISSOTIIDAE Pseudotissotia Peron. L. Turon 1.96 P. (Bauchioceras) Reyment L. Turon 2.14 P. (Wrightoceras) Reyment L. Turon 27 Choffaticeras Hyatt Turon 1.28 C. (Leoniceras) H. Douvillé Turon 12 Plesiotissotia Peron. Coni 2.20 Heterotissotia Peron. U. Turon-Coni 1.66 Tissotia H. Douville Coni-L. Santon 1.47 T. (Metatissotia) Hyatt Coni-L. Santon 2.42 Hemitissotia Peron. Coni Buchiceras Hyatt Coni 1.70 Hoplitoides von Koenen L. Turon-Coni 1.87 COILOPOCERATIDAE Glebosoceras Reyment L. Turon 1.81 Coilopoceras Hyatt Е. Turon-Coni 10 COLLIGNONICERATIDAE Collignoniceras Breistroffer Turon 1.84 C. (Selwynoceras) Warren & Stelck L. Turon 2.34 Prionocyclus Meek Turon 2.47 Subprionocyclus Shimizu U. Turon 2:37 Germaniceras Breistroffer U. Turon 2.25 Niceforoceras Basse Coni 3.06 Subprionotropis Basse Coni 3.06 Gauthiericeras De Grossouvre U. Turon-Coni 2.25 Peroniceras De Grossouvre Coni 2.20 Yabieceras Tokunaga & Shimizu Coni Protexanites Matsumoto L. Coni-L. Santon 2.04 Texanites Spath U. Coni-L. Camp 1.84 Paratexanites Collignon L. Coni-L. Santon P. (Parabevahites) Collignon U. Coni-L. Santon 1.99 Bevahites Collignon U. Santon-M. Camp 2.36 Submortoniceras Spath Camp Menabites Collignon U. Santon-M. Camp 173 Barroisiceras De Grossouvre Coni 2.09 Solgerites Reeside Coni 1.60 Forresteria Reeside Coni 2.64 F. (Reesideoceras) Basse Coni 2.42 Е. (Harleites) Reeside Coni 1.83 Lenticeras Gerhardt Coni-L. Santon 1.14 Paralenticeras Hyatt U. Coni-L. Santon Eulophoceras Hyatt U. Coni-L. Camp 3.06 Pseudoschloenbachia Spath U. Santon 2.64 Diaziceras Spath U. Santon 1.70 Manambolites Hourca U. Camp-Maas 1.50 SPHENODISCIDAE Daradiceras Sornay & Tessiuer Maas 2.33 Sphenodiscus Meek Maas 2.25 Libycoceras Hyatt Maas 2.04 Indoceras Noetling Maas 1299 33 225 24 WARD APPENDIX 2 Ornamental rugosities of heteromorphic ammoni- Ptychoceras tes. HAMITIDAE Hamites Taxon Age Or Hemiptychoceras Stomohamites BOCHIANITIDAE ANISOCERATIDAE Protancyloceras Ti-Be 18 Protanisoceras Bochianites Ti-Haut 25 Anisoceras Juddiceras Val 1125 Idiohamites ANCYLOCERATIDAE Allocrioceras Aegocrioceras Haut .09 Phlycticrioceras Crioceratites Ha-Ba .045 TURRILITIDAE Balearites Ha .05 Proturrilitoides Paracrioceras Ha-Ba .09 Turrilitoides Menuthiocrioceras Ha .08 Ostlingoceras Hoplocrioceras Ha-Ba .14 Pseudohelicoceras Shasticrioceras Ba als Hypoturrilites Pedioceras Ba-Apt .05 Turrilites Parancyloceras Ba 5 NOSTOCERATIDAE Acrioceras Ha-Ba .07 Bostrychoceras Aspinoceras Ba 118 Nipponites Uhligia Ba-Apt .07 Nostoceras Ancyloceras Ba in Exiteloceras Toxoceras Apt .16 Solenoceras Australiceras Apt UN Neocrioceras Tropaeum Ba-Apt 2 Didymoceras Hamiticeras .21 DIPLOMOCERATIDAE HETEROCERATIDAE Glyptoxoceras Heteroceras Ba-Apt AS Diplomoceras Colchidites Ba-Apt SW Polyptychoceras Dissimilites Ba sl Pseudoxybeloceras HEMIHOPLITIDAE SCAPHITIDAE Pseudothurmannia Ha-Ba dí Clioscaphites Hemihoplites Ha-Ba 7 Hoploscaphites PTYCHOCERATIDAE Discoscaphites Anahamulina Ha-Ba .05 Acanthoscaphites Hamulina Ba A Apt-Alb Apt-Alb Alb Alb-Tur Alb Alb-Tur Alb-Ceno Tur Coni Alb Alb Alb-Ceno Alb Alb Ceno Tur-Sant Camp Camp Ca-Ma Camp Ca-Ma S-Ma Ca-Ma Coni-Camp Coni-Ca 5-С Са-Ма Са-Ма Сатр CRETACEOUS AMMONITE SHELL SHAPES 25 APPENDIX 3 S and Or values for Great Valley Sequence ammonites, California. Taxon Age 5 Ог PACHYDISCIDAE Anapachydiscus californicus Camp 1.0 Eupachydiscus teshioensis Sant 123 .15 Eupachydiscus hardai Sant 1.0 .14 Pachydiscus egertoni Camp .8 .08 Pachydiscus subcompressus Maas SÍ :05 Pachydiscus buckhami Sant-Camp 1.0 .06 Canadoceras yokoyami Sant-Camp .9 .09 Canadoceras newberryanum Camp .8 .10 Patagiosites arbucklensis Camp .8 .07 Menuites sp. Camp 1a ACANTHOCERATIDAE Graysonites wooldridgei Ceno .65 .17 Calycoceras spinosum Ceno 1.02 И Calycoceras orientale Ceno .9 110 Calycoceras boulei Ceno 1.2 aS Calycoceras stoliczkai Ceno 1.48 a Acanthoceras whitei Ceno 182 .18 Acanthoceras sp. Ceno Si Romaniceras deveroide Tur 1625 .21 Romaniceras pseudodeverianum Tur .23 Eucalycoceras shastense Tur 1.0 117 Kanabiceras septemseriatum Tur 1.4 19 VASCOCERATIDAE Plesiovascoceras californicum Tur 1.8 3 COLLIGNONICERATIDAE Collignoniceras woollgari Tur 1.0 15 Subprionocyclus branneri Coni .8 .20 Subprionocyclus neptuni Coni 7 25 Subprionocyclus normalis Coni 57, 15 Peroniceras tehamense Coni? 1.0 :23 Texanites kawasakii Coni? 3115 Protexanites thompsoni Coni .16 Submortoniceras chicoense Camp 7 .23 Pseudoschloenbachia boulei Sant-Camp 5 .02 Pseudoschloenbachia sp. Sant 0 Metaplacenticeras pacificum Camp .45 .08 Hoplitoplacaticeras vancouverense Camp .48 2 PULCHELLIIDAE Pulchellia popenoi .6 .14 BRANCOCERATIDAE Mortoniceras sp. Alb 1.14 .30 Oxytropidiceras packardi Alb :55 12 DOUVILLEICERATIDAE Douvilleiceras spiniferum Alb 15338 .19 Acanthoplites gardneri Apt .66 .16 Paraphoplites dallasi Apt .94 .09 Cheloniceras sp. Bar 116 DIPLOMOCERATIDAE Scalarites mihoensis Tur-Coni .075 Ryugusella ryugasensis Sant-Camp .09 Glyptoxoceras subcompressum Sant-Maas .08 Glyptoxoceras indicum Camp .06 Polyptychoceras vancouverense Sant .25 NOSTOCERATIDAE Bostrychoceras elongatum Sant .13 Bostrychoceras otsukai Sant .07 26 APPENDIX 3 continued Taxon Pseudoxybeloceras lineatum Didymoceras vancouverensis Hyphantoceras venustum Neocrioceras sp. Nipponites occidentalis HAMULITIDAE Euptychoceras jordanense Anahamulina aldersona Anahamulina wilcoxensis CRIOCERATIDAE Acrioceras hamlini Acrioceras voyanum Acrioceras vespertinum Shasticrioceras pontiente Shasticrioceras whiteneyi Crioceratites sp. Crioceratites latus Hemibaculites sp. Shasticrioceras patricki Hoplocrioceras wintunium Hoplocrioceras redmondi Hoplocrioceras duncanense HAMITIDAE Hamites sp. Stomohamites sp. TURRILITIDAE Pseudohelicoceras ANCRYLOCERATIDAE Ancyloceras elephans Ancyloceras thomeli Dissimilites sp. Toxoceras blandi Toxoceratoides royerianus Toxoceratoides saulae Toxoceratoides starkingi Toxoceratoides corae Toxoceratoides greeni Heteroceras jeletzkyi Hemihoplites popenoi Arquethites sp. TETRAGONITIDAE Eogaudryceras hertleini Eogaudryceras aurarium Anagaudryceras whitneyi Eotetragonites wintunius Eotetragonites shoupi Eotetragonites gainesi Gabbioceras angulatum Protegragnoites crebrisulcatus Anagaudryceras mikobokense Gaudryceras denseplicatum Gaudryceras denmanense Vertebrites kayei Zelandites inflatus Tetragonites glabrus Tetragonites popetensis Epigoniceras epigonium Pseudophyllites indra Age Sant-Camp Maas Coni-Sant Coni-Sant Tur Bar Apt Apt Haut Haut Haut Bar Bar Bar Haut-Bar Bar Bar Haut Haut Haut Alb Alb Ceno Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Bar Apt Alb Alb Apt Apt Alb Apt Alb Maas Tur-Camp Sant-Camp Maas Coni-Sant? Sant Sant-Camp Camp Camp-Maas — hd do sd (6%) ooronsooo — — CRETACEOUS AMMONITE SHELL SHAPES 27 APPENDIX 3 continued Taxon Age $ Ог PHYLLOCERATIDAE Phylloceras aff. P. thetys Bar 6 02 Partschiceras occidentale Bar .9 .07 Phyllopachyceras trinitense Val-Haut .07 Phyllopachyceras итрдиапит Val-Haut .94 .03 Hypophylloceras onoense Haut .54 .02 Neophylloceras ramosum Coni-Maas .48 .02 Neophylloceras hetonaiense Maas .56 .02 Calliphylloceras sp. Alb .6 0 Phylloceras aldersona Apt .02 Hypophylloceras californica Alb .03 Neophylloceras serititense Alb .02 LYTOCERATIDAE Lytoceras batesi Bar-Alb .04 Lytoceras saturnale Val-Bar .02 Eulytoceras phestum Bar 1.0 KOSSMATICERATIDAE Marshallites sp. Camp .60 .06 BOCHIANITIDAE Bochianites sp. Val-Haut 2125 DESMOCERATIDAE Desmoceras kossmati Ceno 1.0 0 Desmoceras japonicum Ceno 0 Desmoceras merriami Alb .92 0 Barremites sp. Bar .05 Desmoceras dawsoni Alb 6 0 Desmophyllites diphylloides Sant-Camp 75 0 Damesites damesi Sant 7 0 Damesites hetonaiensis Sant .65 0 Hauericeras angustatum Sant-Camp 5 0 Hauericeras rembda Maas 15 0 Puzosia intermedia Alb 73 05 Puzosia subquadrata Alb 1.0 05 Puzosia hofmanni Alb 1.06 .05 Melchiorites indigenes Bar .93 .09 Mesopuzosia pacifica Alb 75 .07 Mesopuzosia indopacifica Sant 15 .08 Pachydesmoceras pachydiscoide Alb .85 .05 Leconteites lecontei Alb .74 .07 Beaudanticeras breweri Alb .56 .07 Brewericeras hulenensis Alb .48 .06 Cleoniceras susukii Alb 57 .06 BERRIASELLIDAE Thurmanniceras stippi Val 255 .07 Thurmanniceras californicum Val .64 .07 Thurmanniceras wilcoxi Val .74 .04 Hannaites riddlensis Haut .55 .04 Hannaites truncatus Haut .64 .10 Neocomites indicus Val .07 Kilianella crassiplicata Val .82 .08 Kilianella besairiei Val 5161 Sarasinella angulata Val .07 Sarasinella hyatti Val .08 Sarasinella densicostata Val .06 HOLCODISCIDAE Spitidiscus oregonensis Val .64 .04 OLCOSTEPHANITIDAE Homolsomites stantoni Val VE .05 28 APPENDIX 3 continued WARD Taxon Age S Or Homolsomites mutabilis Val .68 .09 Olcostephanus pecki Val 1.0 .04 Speetoniceras agnessense Haut AE Spitidiscus oregonensis Haut .64 .04 Polyptychites trichotonus Val .07 Wellsia packardi Haut .08 Wellsia oregonensis Haut .63 .04 Wellsia vigorosa Haut .06 Neocraspedites giganteus Haut .67 .04 Simbirskites lecontei Haut 1.0 .07 Simbirskites broadi Haut 1.27 .12 Hertleinites aguila Haut .65 .06 Н. rectoris Haut 1.04 .06 Hollisites lucasi Haut tel .07 Hollisites inflatus Haut 1.07 .04 MALACOLOGIA, 1986, 27(1): 29-41 DIFFERENTIAL EXTINCTION IN TROPICAL AMERICAN MOLLUSCS: ENDEMISM, ARCHITECTURE, AND THE PANAMA LAND BRIDGE Geerat J. Vermeij 8 Edward J. Petuch Department of Zoology, University of Maryland, College Park, MD 20742, U.S.A. ABSTRACT The uplift of the Central American isthmus during the Pliocene triggered a substantial impoverishment in the marine biota of tropical America. We tabulated all Pliocene subgeneric taxa and their living descendants in 18 families of gastropods and three families of pelecypods for each of three regions: (1) the Caloosahatchian Province, centered in Florida; (2) the Atlantic Gatunian region, comprising the rest of the tropical Western Atlantic, and (3) the Pacific Gatunian, corresponding to the modern tropical Eastern Pacific. Extinction affected molluscs more in the Caloosahatchian (32%) and Atlantic Gatunian regions (32%) than in the Pacific Gatunian (15%). In all regions, endemic taxa зиНегед more than 50% extinction. Because the Atlantic faunas were richer in endemics than was the Pacific Gatunian, part of the interoceanic difference in the impact of extinction 15 attributable to the high susceptibility to extinction of narrowly distributed taxa. The tendency for hard-bottom gastropods to be somewhat more resistant to extinction than were soft-bottom taxa is shown to be partly the result of an artifact of geographical range, there being relatively few endemic taxa among hard-bottom gastropods. Hard-bottom taxa with a narrow or thick-lipped aperture were more susceptible to extinction in the Atlantic than were their wide-apertured counterparts. This pattern, which is not an artifact of geographical range, resulted in a post-Pliocene decline in the incidence of apertural protective devices among Atlantic hard-bottom gastropods, while the incidence of these features in the Eastern Pacific remained at the high Pliocene level. Among soft-bottom gastropods, the incidence of narrow and thick- lipped apertures has increased from the Pliocene to the Recent in each of the three regions of tropical America. An examination of refuges to which previously more widely distributed taxa have become restricted shows that high productivity could have played a role in the persistence of many populations. The Eastern Pacific is the most important of these refuges, but the north and east coasts of South America have also been important. Key words: extinction; Panama land bridge; productivity; predation; molluscs; endemism. INTRODUCTION The uplift of the Central American isthmus, which created a continuous land bridge be- tween North and South America between 3.5 and 3.1 million years ago (Saito, 1976, Keigwin, 1978), was one of the most impor- tant events in recent earth history. It permitted a large-scale two-way migration of land mam- mals (Marshall et al., 1982), intensified north- south circulation in the oceans (Emiliani et al., 1972; Holcombe 8 Moore, 1977; Kaneps, 1979), altered the chemistry of the Atlantic Ocean (Keigwin, 1982), and perhaps created conditions favorable for northern-hemisphere glaciation (Weyl, 1968). Together with peri- odic glaciations which raised and lowered sea levels and sea surface temperatures, the up- (29) lift triggered a substantial biotic impoverish- ment by the extinction of numerous shallow- water lineages of molluscs (Woodring, 1966; Vermeij, 1978; Petuch, 1982), barnacles (Spivey, 1981), corals (Porter, 1972; Dana, 1975; Frost, 1977; Heck 8 McCoy, 1978), and other groups. It is evident from the distribution of Recent molluscs that the extinctions in marine tropical America did not affect all regions equally. Many molluscan taxa which in the Pliocene occurred in both the Atlantic and Pacific areas of the Americas are today confined to the Eastern Pacific. The number of taxa that have become confined to the Atlantic is only 1/8 as large as that of taxa confined to the Pacific (Woodring, 1966; Vermeij, 1978). Extinction therefore seems to have affected the Atlantic biota far more profoundly than it affected the 30 VERMEN 8 PETUCH Pacific. For scleractinian corals, however, the trend may have been the reverse (Porter, 1972; Heck & McCoy, 1978). The study of extinction in tropical America might shed light on several interesting prob- lems. The Recent faunas on the Atlantic and Pacific coasts of tropical America differ т many important respects. A study of the inci- dence of cognate species (closely similar At- lantic and Pacific congeners) showed that hard-bottom gastropods are taxonomically more divergent than are gastropods from un- consolidated (soft) bottoms (Vermeij, 1978). Does this pattern mean that extinction af- fected hard-bottom gastropods more than it affected species on soft bottoms, or is the pattern the result of differential diversification on the two coasts? Moreover, there is a con- siderable difference between Western Atlan- tic and Eastern Pacific hard-bottom gastro- pods in the incidence of certain antipredatory features. Caribbean faunas show a lower in- cidence of narrow apertures (aperture length/ aperture width greater than 2.5), strong sculp- ture, low spires (apical half-angle greater than 45°), and thick outer lips than do assemblages in the Eastern Pacific (Vermeij, 1978). Eco- logical studies have shown that the architec- tural contrast reflects a higher intensity of predation by shell-breakers in the Eastern Pacific (Bertness, 1982). When did these dif- ferences arise, and did differential extinction contribute to them? What was the architec- tural history of soft-bottom gastropods, among which regional differences in preda- tion-related architecture are not evident today in tropical America (Vermeij, 1978; Vermeij et al., 1980)? New reconstructions of pre-isthmian bioge- ography (Petuch, 1982) and the discovery of a relict fauna in northern South America (Petuch, 1976, 1981a) now permit a refine- ment of our understanding of marine extinc- tions in tropical America. During the Miocene and Pliocene, tropical America was divided into two biogeographical provinces: (1) the Caloosahatchian Province, centered in Flor- ida and extending to the Carolinas, Yucatan, and northern Cuba; and (2) the Gatunian Province, extending in the Atlantic from Nica- ragua and the West Indies to central Brazil and in the Pacific from the Gulf of California to northern Peru (Petuch, 1982). After the formation of the Central American isthmus, the Pacific and Atlantic parts of the Gatunian Province became mutually isolated and acquired distinctive though closely allied biotas. We ask three questions in this paper. First, what proportion of the Pliocene fauna is still extant in Florida, the Atlantic Gatunian Re- gion, and the Eastern Pacific (Pacific Gatun- ian Region)? Second, what factors have influ- enced differential extinction on the two coasts of tropical America? Does a narrow range during the Pliocene imply a higher than aver- age susceptibility to extinction and, if so, does the small number of endemic Pacific taxa during the Pliocene explain the relatively small impact of extinction in that region? Fi- nally, were the extinctions selective with re- spect to habitat type and predation-related architecture? METHODS Petuch (1982) compiled a list of all subge- nera and species groups of larger mesogas- tropods and neogastropods that have been described from Pliocene strata of tropical America. Because many of these families have not been studied comprehensively in recent years, we have chosen to restrict our analysis to 18 of the families treated by Pe- tuch. Our revised and emended compilation (see the Appendix) is based on a re- evaluation of several general systematic treatments (Olsson, 1964; Keen, 1971) and on recent papers on Strombidae (Abbott, 1960), Tonnacea (Abbott, 1968; Beu, 1980), Muricidae (Vokes & D'Attilio, 1982), Colum- bellidae (Radwin, 1977a, b), Buccinidae (Vokes, 1970; Cernohorsky, 1981), Mitridae (Cernohorsky, 1976), and Cancellariidae (Petit, 1967, 1970, 1976). In addition, we have compiled a list for the pelecypod families Ar- cidae (Reinhart, 1935; Olsson, 1961, 1964; Woodring, 1973), Cardiidae (Keen, 1980), and Lucinidae (Bretsky, 1976). We have fol- lowed Woodring’s (1973) interpretation of the arcid groups established by Olsson (1961). We accept Olsson’s (1964) lucinid taxa to- gether with Woodring’s (1982) range exten- sions despite Bretsky’s (1976) uncertainty about the status of some of the fossil forms. Many of the subgenera that we have tabu- lated may be subject to reinterpretation and revision in the future. Some readers will object that we have taken only a subset (indeed, a minority of EXTINCTION IN TROPICAL AMERICAN MOLLUSCS 31 families) of gastropods and pelecypods. Un- fortunately, many groups have not received serious attention from investigators on one or both coasts of tropical America, so that the stratigraphical and geographical distribution of many genera and subgenera remains in doubt. Some families, like the Naticidae, are well monographed in the Eastern Pacific (Marincovich, 1977), but remain little under- stood in the Western Atlantic. Small-shelled families and most opisthobranchs are poorly known. In order to have more control over the quality of our data, we elected not to include these families. We must assume that the ev- olutionary behavior of the families which we have studied 1$ similar to that of families that we omitted. Six biogeographical categories of tropical American Pliocene molluscs may be recog- nized: (1) Caloosahatchian endemics; (2) At- lantic Gatunian endemics; (3) Pacific Gatun- ian endemics; (4) Gatunian taxa, found in both the Atlantic and Pacific parts of the Gatunian Province; (5) Atlantic taxa, those found in both the Caloosahatchian and Atlantic Gatunian regions; and (6) pan-American taxa, those found in all three regions. Many Pliocene taxa that during the Pliocene were found in both the Atlantic and the Pacific have in the Recent fauna become restricted to the Pacific; they have been referred to as Paciphilic (Wood- ring, 1966). Caribphilic taxa are those that have become restricted in the Recent fauna to the Western Atlantic. Although a more quantitative treatment of geographical range would be desirable, we believe that artifacts of preservation mitigate against greater precision. Although the Pliocene fauna is known from many localities throughout tropical America, some environ- ments may be represented by only a few sites. Taxa confined to these environments would then seem to be endemic when in fact their limited distribution represents the rarity or poor preservability of shells in some envi- ronments. We calculated the impact of extinction as follows. For each of the three regions of trop- ical America, the number of taxa that became locally extinct was divided by the total number of taxa that were present in that region at or before the time of the isthmian uplift. Throughout this paper, we treat the Recent faunas of America as impoverished versions of the Pliocene fauna. At the supraspecific level this is a valid procedure, because very few subgenera appear in the Pleistocene that were not already present during the Pliocene. The few new Pleistocene genera, such as Charonia (Cymatiidae) and Mammilla (Nati- cidae), appear to be immigrants from the Indo-West-Pacific or Eastern Atlantic (Marin- covich, 1977; Emerson, 1978; Vermeij, 1978; Petuch, 1982). Some groups became extinct in part or all of tropical America but were later re-introduced from the Western Pacific. We believe that the number of such groups is small, and that we have only slightly under- estimated the levels of extinction. The habitats of gastropods were estab- lished by analogy with the known habits of living forms. Taxa considered to have narrow apertures or a thick Пр include all Cypraeacea (Cypraeidae, Eratoidae, Ovulidae), Tonnacea (except Ficidae and Tonna), Columbellidae (except Mitrella), Olividae (except Ancilla, Eburna, and Agaronia), Mitridae, Marginell- idae, Conidae, Strombidae, and some Volu- tidae (Plicoliva), Muricidae (Eupleura, Vitu- laria), and Buccinidae (Bailya, Engina, No- rthia). We have chosen not to employ statistical tests to our data. Our intent is to discover trends and effects whose significance de- pends less on statistical cut-off points than on biological impact. RESULTS Before the establishment of the Central American isthmus, the Atlantic part of the Gatunian Province was slightly richer in taxa than was the Pacific portion. Of the groups tabulated in the Appendix, only the Tonnacea and Cancellariidae had a slightly higher Pa- cific than Atlantic diversity in the Pliocene Gatunian Province (Table 1). The Caloosaha- tchian Province was generally less rich than the Atlantic Gatunian region, although the Muricidae, Fasciolariidae, and Volutidae had a larger number of taxa in the Caloosahatch- ian. Diversities in the Pacific Gatunian and Caloosahatchian regions were roughly com- parable (Table 1). The patterns of diversity are somewhat dif- ferent today. The Atlantic and Pacific portions of the former Gatunian Province now support roughly the same number of taxa (Table 1). The region corresponding to the Neogene Caloosahatchian Province still has a gener- 32 VERMEN 8 PETUCH TABLE 1. Diversity and extinction of tropical American molluscs. Мь = number of subgenera present in the Pliocene. Мн = number of subgenera surviving from Pliocene to Recent. E = percentage of extinction of Pliocene taxa. Caloos. Atl. Gatun. Pac. Gatun. Taxon Np Na E Np Na E Np Na E Gastropoda Turritellidae 7; 4 43% 8 4 50% 7 4 43% Strombidae 3 3 0 4 2 50% 2 2 0 Cypraeacea 8 3 63% 17 11 35% 13 11 15% Tonnacea Uf 6 14% 18 11 15% 14 9 36% Muricidae 30 19 27% 24 19 21% 18 16 11% Columbellidae 1e 8 38% 24 19 21% 16 14 13% Buccinidae 14 U 50% 18 8 56% ШИ 14 18% Fasciolariidae 8 6 25% TL 6 14% 4 4 0 Mitridae 4 3 25% 10 5 50% 8 8 0 Olividae 9 5 44% 117 in 35% 9 9 0 Volutidae 6 5 17% 4 4 0 3 2 33% Marginellidae 12 11 8% 12 WZ 0 5 5 0 Cancellariidae 10 4 60% 18 6 67% 19 16 16% Conidae 6 4 33% 7 6 14% 7 6 14% Total 137 88 36% 183 124 32% 142 121 15% Pelecypoda Arcidae 115 10 33% 20 12 40% 20 18 10% Lucinidae 16 14 13% 18 16 11% 16 12 25% Cardiidae 8 7 13% 13 6 54% 13 11 15% Total 39 31 21% 51 34 33% 49 41 17% Total 176 119 32% 234 158 32% 191 162 15% ally lower diversity than does either portion of the former Gatunian Province, but the Strom- bidae and Volutidae reach their highest subgeneric number in Florida and surround- ing waters. Differential extinction was responsible for the equalization of diversity in the two portions of the former Gatunian Province. Of the 234 taxa tabulated in the Appendix from the Pliocene Atlantic Gatunian region, 32% have become regionally extinct. For the 191 Pliocene Pacific Gatunian gastropods and pe- lecypods, the corresponding percentage is 15%. These estimates suggest that the im- pact of regional extinction was roughly two times greater in the Atlantic part ofthe Gatun- ian Province than in the Pacific part. The only families that suffered greater extinction in the Atlantic portion were the Volutidae, Luci- nidae, and those in the Tonnacea (Table 1). Even more taxa would have become re- gionally extinct in the Atlantic Gatunian region were it not for two refuges where taxa whose distribution was much broader during the Pliocene survive today as relicts. Atlantic rep- resentatives of six taxa (Broderiptella, Mu- racypraea, Panamurex, Sincola, Aphera, and Subcancilla) persist in the waters off eastern Colombia and Venezuela (Petuch, 1976, 1981a). Five taxa survive in the Western At- lantic only in Brazil (Pusula, Northia, Plicoliva, Bullata, and Miltha). Ancilla and Eburna are found today only in northern and eastern South America, but like the other taxa men- tioned above they had much wider distribu- tions in tropical America during the Pliocene. Vermeij (1978) and Petuch (1979, 1981a) have listed additional taxa that have become restricted in the Recent fauna to the Brazilian and Colombo-Venezuelan refuges. The Caloosahatchian Province suffered substantial impoverishment of subgenera, es- pecially among gastropods (Table 1). Of the 176 Pliocene taxa tabulated in the Appendix, 32% have become regionally extinct. Impov- erishment was therefore similar in magnitude to that in the Atlantic Gatunian region and greater than that in the Eastern Pacific. The only families suffering less extinction in the Caloosahatchian Province than in the Eastern Pacific are the Volutidae, Lucinidae, and groups in the Tonnacea, the same groups which also did relatively well in the Atlantic Gatunian (Table 1). EXTINCTION IN TROPICAL AMERICAN MOLLUSCS 33 Vermeij's (1978) suggestion that the Florida region serves as a refuge 1$ not substantiated by our data. Of the taxa we studied, only three (Metulella, Stewartia, and Dinocardium) have survived in Florida after disappearing from the Gatunian region. Petuch (1981b) regarded the area around Roatan Island, Honduras, as a refuge for several Caloosahatchian relicts, but only one taxon (Pleioptygma) among the 176 Pliocene taxa we studied seems to have become restricted to this area. As Woodring (1966) and Vermeij (1978) recognized, the Eastern Pacific serves as a refuge for many tropical American taxa which during the Pliocene lived in both the Atlantic and the Pacific. Of the 57 taxa that became regionally extinct in the Caloosahatchian Province, 16 (28%) are Paciphiles, whereas 11 (Broderiptella, Pusula, Panamurex, Sin- cola, Strombina, Ancilla, Eburna, Subcan- cilla, Bullata, Aphera, and Miltha) (19%) have become restricted to refuges in the former Atlantic Gatunian region. Paciphilic taxa com- prise 39 of 76 regionally extinct subgenera (51%) in the Atlantic Gatunian region; these taxa constitute 15% of the Pliocene fauna in that region. Caribphilic taxa comprise 9 (17%) of 30 taxa which became regionally extinct in the Eastern Pacific. These data support the earlier finding that the Atlantic refuges are less important as sanctuaries than is the East- ern Pacific for Pliocene relicts. Several taxa listed in the Appendix as being extinct in the Americas still persist in the Indo- West-Pacific, Australia, or southern Africa. They include Dolomena, Labiostrombus, Cypraeovula, Pustularia, Subpterynotus, Harpeola, Dibaphus, Omogymna, Strephona, Neocylindrus, and Hawaiarca. Of the 62 taxa that became globally extinct in the Americas, 11 (18%) belong to this relict category. That the Indo-West-Pacific has acted as a refuge for corals and other animals throughout the Cenozoic is well known (Vermeij, 1978; Heck 8 McCoy, 1978). Our data show clearly that endemism (de- fined as occurrence during the Pliocene in only one of the three regions in tropical Amer- ica) is associated with a high probability of post-Pliocene extinction. Regional extinction affected 25 of 50 Caloosahatchian taxa (50%), 19 of 34 Atlantic Gatunian taxa (56%), and 6 of 10 Pacific Gatunian taxa (60%). By contrast, the 102 pan-American taxa suffered only 24% local extinction in the Caloosaha- tchian, 20% extinction in the Atlantic Gatun- ian, and 14% regional extinction in the Pacific Gatunian region. Habitat and apertural form clearly influenced the likelihood of extinction in tropical American gastropods. Soft-bottom gastropods were somewhat more affected by extinction than were hard-bottom forms in each of the three regions. This results chiefly from the large number of extinction-vulnerable wide-aper- tured taxa among soft-bottom gastropods. As Vermeij (1978) already suspected, At- lantic Gatunian and Caloosahatchian hard- bottom gastropods with a narrow or thick- lipped aperture were more prone to extinction than were co-occurring hard-bottom forms with a wide or thin-lipped aperture (Table 2). Inthe Pacific Gatunian region, the situation for hard-bottom gastropods was the reverse; wide-apertured taxa were more prone to ex- tinction than were narrow-apertured taxa (Ta- ble 2). Among gastropods from soft (uncon- solidated) bottoms, taxa with a narrow or thick-lipped aperture were less susceptible to extinction than were broad-apertured taxa. This difference was evident in all three re- gions (Table 2). The architectural difference between hard- bottom gastropod assemblages of the mod- ern-day Western Atlantic and Eastern Pacific is at least in part attributable to the selective TABLE 2. Effect of habitat and apertural shape on susceptibility of gastropods to extinction. Caloos. Atl. Gatun. Pac. Gatun. Category Np E Np E Np E Soft-bottom, broad aperture 47 53% 53 49% 46 22% Soft-bottom, narrow aperture 43 23% 67 28% 44 16% Soft-bottom, total 90 39% 120 38% 90 19% Hard-bottom, broad aperture 27 26% 24 8% 20 15% Hard-bottom, narrow aperture 20 35% Эй 32% 31 3% Hard-bottom, total 47 30% 61 23% 51 8% 34 VERMEIJ 8 PETUCH extinction of narrow-apertured and thick- lipped gastropod taxa in the Atlantic. During the Pliocene, the Atlantic and Pacific parts of the Gatunian Province had about the same incidence of modified apertures (61% and 60% respectively). Among the Pliocene sur- vivors in the modern fauna, the incidence of narrow-apertured and thick-lipped forms has remained at the Pliocene level in the Eastern Pacific (64%), but has fallen in the Caribbean to 53% (Table 3). In the Caloosahatchian Province, the incidence of modified apertures has remained rather low (40% and 37% re- spectively for the Pliocene and Recent) (Ta- ble 3). TABLE 3. Incidence (1!) of narrow-apertured taxa among Pliocene and Recent gastropods in tropical America. Category Np | Na | Caloosahatchian Hard-bottom 48 40% 32 37% Soft-bottom 90 48% 55 60% Atlantic Gatunian Hard-bottom 61 61% 47 53% Soft-bottom 122 55% 75 64% Pacific Gatunian Hard-bottom 51 60% 47 64% Soft-bottom 90 49% 73 51% For the soft-bottom component of the gas- tropod fauna, there was a modest increase in the incidence of narrow apertures and thick lips in all three regions, though in the Eastern Pacific this increase was slight (Table 3). The higher incidence of modified apertures that characterized the Atlantic Gatunian soft- bottom gastropods relative to those of the Pacific Gatunian has been maintained in the Recent fauna. Although Vermeij (1978) was unable to detect this present-day difference, he had very few Recent assemblages, partic- ularly from the Eastern Pacific. DISCUSSION In general, the two Atlantic provinces of tropical America were more affected by ex- tinction than was the Eastern Pacific. This conclusion applies to gastropods and pe- lecypods (Table 1), to soft-bottom and to hard-bottom gastropods (Table 2), and to nar- row-apertured and wide-apertured gastro- pods (Table 2). Our findings are in excellent agreement with earlier inferences drawn from distributional patterns of extant species (Woodring, 1966; Vermeij, 1978). Conclusions about the impact of extinction inevitably depend on the quality of information about the stratigraphical and geographical distribution of the taxa in question. One pos- sible source of error in our data is the distri- bution of sites from which Pliocene fossils have been collected in tropical America. As Woodring (1966) has pointed out, a majority of localities is located in the Caribbean part of the Gatunian region. Although Eastern Pacific localities in Nicaragua, Costa Rica, Panama, Colombia, Ecuador, and Peru are rich in spe- cies, they are outnumbered by Caribbean lo- calities in Central America, northern South America, and the West Indies. If sampling of the Pliocene Pacific Gatunian region was less complete than that in the Caribbean, some extinction-prone endemic taxa might have been missed, and some taxa now known only from the Atlantic Gatunian might be found in the Pacific portion of that province as well. Estimates of regional extinction in the Eastern Pacific would therefore probably be some- what higher if more fossil-bearing localities were available. Given the large differences between present estimates of Pacific and At- lantic Gatunian extinctions, however, we do not believe that new fossil discoveries will alter our findings significantly. Another source of error, or more precisely of variation, is the choice of taxonomic group. Families differed widely in their susceptibility to extinction (Table 1). Nevertheless, the geo- graphical, habitat, and architectural patterns of selectivity in extinction are evident in many families. We therefore take these patterns to reflect selectivity that transcends taxonomic considerations. The difference in the impact of extinction between the Atlantic and the Pacific may be due in part to the high incidence of extinction- vulnerable endemic taxa in the Pliocene At- lantic. Our data show clearly that Pliocene endemic taxa (those confined to a single re- gion during the Pliocene) had a much higher probability of extinction than did pan- American taxa (those occurring in all three regions during the Pliocene). Because the Pacific Gatunian faunas contained fewer en- demics (5%) than did either the Atlantic Ga- tunian (14%) or the Caloosahatchian fauna (28%), the greater collective resistance of EXTINCTION IN TROPICAL AMERICAN MOLLUSCS 35 eastern Pacific taxa to extinction can be inter- preted partly as an artifact of geographical range. The correspondingly large contribution of pan-American taxa to the Pliocene Eastern Pacific fauna (53%) adds further support to this interpretation. Geographical artifacts also explain some other features of post-Pliocene extinction in tropical America. Hard-bottom gastropods were less affected by extinction than were gastropods in unconsolidated sediments (Ta- ble 2). The percentage of endemic taxa among soft-bottom gastropods is higher both in the Caloosahatchian Province (31%) and Atlantic Gatunian region (21%) than it is among hard-bottom gastropods from these two regions (27% and 15% respectively). More importantly, the percentage of pan- American taxa is substantially lower in soft- bottom gastropods both in the Caloosahatch- ian Province (44%) and Atlantic Gatunian re- gion (33%) than among hard-bottom forms (66% and 51% respectively). The collectively greater resistance of hard-bottom gastropods to extinction therefore seems to be associated with relatively broad geographical ranges. We do not know why hard-bottom snails should tend to have broader geographical dis- tributions than soft-bottom gastropods in trop- ical America. It is interesting that many of the hard-bottom gastropods belong to families that have planktonically dispersing larvae. Any statistical association between distribu- tion, habitat, and dispersibility may be quite fortuitous, but our present understanding of these relationships is still rudimentary. Several aspects of selective extinction can- not be explained as artifacts of geographical range. п each of the three regions of Pliocene tropical America, soft-bottom gastropods with a narrow or thick-lipped aperture were more resistant to extinction than were broad- apertured forms from the same habitats (Ta- ble 2). If this pattern were the consequence of a geographical artifact, narrow-apertured forms should show a lower incidence of en- demism than should broad-apertured taxa. This is indeed so in the Caloosahatchian Province (26% and 36% endemism, for nar- row-apertured and broad-apertured subge- nera respectively), but not in the Atlantic Ga- tunian region (19% and 11% respectively). Moreover, the incidence of pan-American taxa, which should be higher in the more extinction-resistant narrow-apertured forms, is either the same for the two groups (44% in the Caloosahatchian Province), or lower in the narrow-apertured forms (28% versus 42% in the Atlantic Gatunian region). Among hard-bottom gastropods, extinction affected narrow-apertured taxa more pro- foundly than wide-apertured forms in both At- lantic regions of tropical America, whereas in the Eastern Pacific the narrow-apertured forms were less affected by extinction than were gastropods with a broad aperture. Again, this complex pattern would not have been predicted from the incidences of en- demic and pan-American taxa. In the Caloosahatchian Province, for example, en- demics comprise 56% of broad-apertured taxa and only 10% of the more extinction- prone narrow-apertured forms. The relative reduction of narrow-apertured and thick-lipped gastropods from the Pliocene to the Recent in hard-bottom Atlantic environ- ments suggests that selection in favor of shell armor decreased. This hypothesis is consis- tent with the observation that the ecological impact of shell-breaking predators is less on the Atlantic coast of Panama than on the Pacific side (Bertness, 1982). The rise in in- cidence of narrow-apertured taxa among gastropods from unconsolidated bottoms in all three regions suggests an increase in se- lection for armor in this habitat. No ecological or other data have yet come to light which support this interpretation. Vermeij et al. (1980) found no temporal change in the fre- quency of repaired injuries in terebrid gastro- pods in tropical America, as would have been expected if the hypothesis were correct. Predators probably did not play a direct role in bringing about the extinction of broad- apertured taxa. Plausible instances of extinc- tion due to biotic agents are rare, and invari- ably involve the bringing together of two biotas with very different evolutionary histo- ries (Vermeij, 1978). Post-Pliocene immigra- tion into the Atlantic seems to have been of negligible magnitude, and all the available evidence suggests that the chief predators of modern Atlantic gastropods were already present in tropical America during the Pliocene. We believe instead that narrow-apertured forms may, on average, have lower individual and population growth rates than do broad- apertured taxa, at least on hard bottoms, and that their populations rebound less rapidly after being decimated by a catastrophe. Di- rect estimates of individual growth rates, egg 36 VERMEN 8 PETUCH production, and other life-history characteris- tics are required for the evaluation of this hypothesis. No data of this kind currently ex- ist. The hypothesis is, however, consistent with the properties of the tropical American refuges that we have documented in this study. Both the Pacific and the Atlantic ref- uges are characterized by high productivity (Vermeij, 1978; Petuch, 1981a, 1982; Anton- ius, 1980). Birkeland (1982) has marshalled an impressive body of evidence to support his view that massive starvation is typical for many marine invertebrate larvae under con- ditions of low productivity, whereas mass survival of larvae is possible when nutrient levels are increased either through upwelling or by rain-induced terrestrial runoff. High- productivity environments might therefore protect many species from repeated decima- tion to dangerously low population levels by providing conditions for rapid population ex- pansion. Species with high individual growth rates might be less affected by decimation and routine starvation than those with slower growth, smaller internal volume, and other features associated with the production of ar- mor. There is some evidence for the hypothesis that the Caribbean Sea suffered a reduction in productivity after the uplift of the Panama land bridge. Keigwin (1982) has documented a Late Neogene decline in nutrient levels in deep Caribbean waters, whereas deep wa- ters in the Eastern Pacific have remained consistently rich in nutrients. This pattern is also consistent with the history of scleractin- ian corals, which, unlike most molluscan groups, suffered more extinction in the East- ern Pacific than in the Atlantic (Heck & Mc- Coy, 1978). Birkeland (1977) has shown that corals are outcompeted by suspension- feeding animals lacking algal symbionts when they co-occur as newly settled juveniles on panels under conditions of upwelling. A ca- tastrophe such as a sudden drop in temper- ature could therefore have had much more profound effects on corals in the Eastern Pa- cific than in the Caribbean, where recruits would stand a better chance of success in the generally less productive waters. The reasons for selective extinction with respect to gastropod aperture shape remain shrouded in mystery, but the consequences of selectivity are clear. Selective extinction helped to bring about a change from an ar- chitecturally homogeneous Pliocene fauna of hard-bottom gastropods in Gatunian America to architecturally divergent faunas on the Pa- cific and Atlantic coasts. Selective diversifica- tion has perhaps exaggerated this divergence between the faunas, as suggested by Vermeij (1978), but we have no new evidence on this point. The patterns of extinction that we have un- covered in the Late Neogene molluscan fau- nas of tropical America may have properties that apply to other extinction events. The greater susceptibility of endemics to extinc- tion may be generally true and is well known for human-caused extinctions (Vermeij, 1985). Valentine (1973), Boucot (1975), and many others have pointed out that residual faunas after major episodes of extinction show low provinciality and cosmopolitan dis- tributions of taxa. Our study suggests that extinction and its consequences depend on geography. Not only was extinction less profound in the East- ern Pacific than in the Atlantic, but the pat- terns of selectivity were different. These re- sults provide good reasons for caution. It is unsafe to generalize from single studies of extinction. Not only do we need to understand why certain morphological traits are associ- ated with stratigraphical persistence, but we need to know how these possibly fortuitous associations are influenced by geographical and historical peculiarities. ACKNOWLEDGMENTS We are grateful to the (U.S.) National Sci- ence Foundation for supporting our research. The senior author is indebted to Egbert G. Leigh for instilling an early interest in the his- tory of tropical America. We have greatly prof- ited from critical appraisals of our manuscript by M. L. Reaka, D. Jablonski, J. W. Valentine, and two anonymous reviewers. REFERENCES CITED ABBOTT, R. T., 1960, The genus Strombus in the Indo-Pacific. Indo-Pacific Mollusca, 1: 33-146. ABBOTT, R. T., 1968, The helmet shells of the world (Cassidae). Indo-Pacific Mollusca, 2: 2-202. ANTONIUS, A., 1980, Occurrence and distribution of stony corals in the Gulf of Cariaco, Venezuela. EXTINCTION IN TROPICAL AMERICAN MOLLUSCS 37 Internationale Revue der Gesamten Hydrobiolo- gie, 65: 321-338. BERTNESS, M. D., 1982, Shell utilization, preda- tion pressure, and thermal stress in Panamanian hermit crabs: an interoceanic comparison. Jour- nal of Experimental Marine Biology and Ecology, 64: 159-187. BEU, A. G., 1980, Australian gastropods of the family Bursidae. 1. The families of Tonnacea, the genera of Bursidae, and revision of species pre- viously assigned to Tutufa Jousseaume, 1881. Records of the Australian Museum, 33: 248-324. BIRKELAND, C., 1977, The importance of rate of biomass accumulation in early successional stages of benthic communities to the survival of coral recruits. Proceedings of the Third Interna- tional Coral Reef Symposium, 1: 15-21. BIRKELAND, C., 1982, Terrestrial runoff as a cause of outbreaks of Acanthaster planci (Echinoder- mata: Asteroidea). Marine Biology, 69: 175-185. BOUCOT, A. J., 1975, Evolution and extinction rate controls. Elsevier, Amsterdam, 427 p. BRETSKY, $. S., 1976, Evolution and classification of the Lucinidae (Mollusca; Bivalvia). Palaeonto- graphica Americana, 8: 219-337. CERNOHORSKY, W. O., 1976, The Mitridae of the world. I. The subfamily Mitrinae. Indo-Pacific Mo- llusca, 3: 273-528. CERNOHORSKY, W. O., 1981, The family Bucci- nidae. Part |: the genera Nassaria, Trajana and Neoteron. Monographs of Marine Mollusca, 2: 201-284. DANA, T. F., 1975, Development of contemporary Eastern Pacific coral reefs. Marine Biology, 33: 355-374. EMERSON, W.K., 1978, Mollusks with Indo-Pacific faunal affinities in the eastern Pacific Ocean. Nautilus, 92: 91-96. EMILIANI, C., GARTNER, S. & LIDZ, B., 1972, Neogene sedimentation on the Blake Plateau and the emergence of the Central American isth- mus. Palaeogeography, Palaeoclimatology, Palaeoecology, 11: 1-10. FROST, S. H., 1977, Miocene to Holocene evolu- tion of Caribbean Province reef-building corals. Proceedings of the Third International Coral Reef Symposium, 2: 353-359. НЕСК, К. L., Jr. & MCCOY, E. D., 1978, Long- distance dispersal and the reef-building corals of the Eastern Pacific. Marine Biology, 48: 349-356. HOLCOMBE, T. L. & MOORE, W. S., 1977, Paleo- currents in the eastern Caribbean: geological evidence and implications. Marine Geology, 23: 35-56. KANEPS, A. G., 1979, Gulf Stream: velocity fluctu- ation during the Late Cenozoic. Science, 204: 297-301. KEEN, A. M., 1971, Seashells of tropical West America. Stanford University, Palo Alto, Calif., 1064 p. KEEN, A. M., 1980, The pelecypod family Card- iidae: a taxonomic summary. Tulane Studies т Geology and Paleontology, 16: 1-40. KEIGWIN, L. D., Jr., 1978, Pliocene closing of the Isthmus of Panama, based on biostratigraphic evidence from nearby Pacific Ocean and Carib- bean Sea cores. Geology, 6: 630-634. KEIGWIN, L. D., Jr., 1982, Isotopic paleoceanogr- aphy of the Caribbean and East Pacific: role of Panama uplift in Late Neogene time. Science, 217: 350-353. MARINCOVICH, L., Jr., 1977, Cenozoic Naticidae (Mollusca: Gastropoda) of the northeastern Pa- cific. Bulletins of American Paleontology, 70: 160-494. MARSHALL, |. G., WEBB, $. D., SEPKOSKI, J. J., Jr. & RAUP, D. M., 1982, Mammalian evolution and the great American interchange. Science, 215: 1351-1357. OLSSON, A. A., 1961, Mollusks of the tropical eastern Pacific, particularly from the southern half ofthe Panamic-Pacific faunal province (Pan- ama to Peru). Panamic-Pacific Pelecypoda. Pa- leontological Research Institute, Ithaca, New York, 574. OLSSON, A. A., 1964, Neogene mollusks from northwestern Ecuador. Paleontological Re- search Institute, Ithaca, New York, 256 р. PETIT, R. E., 1967, Notes on Cancellariidae (Mol- lusca: Gastropoda). Tulane Studies in Geology, 5: 217-219. PETIT, R. E., 1970, Notes on Cancellariidae (Mol- lusca: Gastropoda)—I!. Tulane Studies in Geol- ogy and Paleontology, 8: 83-88. PETIT, R. E., 1976, Notes on Cancellariidae (Mol- lusca: Gastropoda)—Ill. Tulane Studies in Geol- ogy and Paleontology, 12: 33-43. PETUCH, E. J., 1976, An unusual molluscan as- semblage from Venezuela. Veliger, 18: 322-325. PETUCH, E. J., 1979, New gastropods from the Abrolhos archipelago and reef complex, Brazil. Proceedings of the Biological Society of Wash- ington, 92: 510-526. PETUCH, Е. J., 1981a, А relict Neogene caenogas- tropod fauna from northern South America. Malacologia, 20: 307-347. PETUCH, E. J., 1981b, A volutid species radiation from northern Honduras, with notes on the Hon- duran Caloosahatchian secondary relict pocket. Proceedings of the Biological Society of Wash- ington, 94: 1110-1130. PETUCH, E. J., 1982, Geographical heterochrony: contemporaneous coexistence of Neogene and Recent molluscan faunas in the Americas. Palaeogeography, Palaeoclimatology, Palaeoe- cology, 37: 277-312. PORTER, J. W., 1972, Ecology and species diver- sity of coral reefs on opposite sides of the isth- mus of Panama. Bulletin of the Biological Society of Washington, 2: 89-116. RADWIN, G. E., 1977a, The family Columbellidae in the Western Atlantic. Veliger, 19: 403-417. RADWIN, G. E., 1977b, The family Columbellidae 38 VERMEN 8 PETUCH in the Western Atlantic. Part Il A. Pyreninae. Veliger, 20: 119-133. REINHART, P. W., 1935, Classification of the pe- lecypod family Arcidae. Bulletin du Musée royal а’ Histoire naturelle de Belgigue, 11(3): 1-68. SAITO, T., 1976, Geologic significance of coiling direction in the planktonic Foraminifera Pullena- tina. Geology, 4: 305-309. SPIVEY, Н. R., 1981, Origins, distribution, and zoo- geographic affinities of the Cirripedia (Crustacea) of the Gulf of Mexico. Journal of Biogeography, 8: 153-176. VALENTINE, J. W., 1973, Evolutionary paleoeco- logy of the marine biosphere. Prentice Hall, En- glewood Cliffs, New Jersey, 511 p. VERMEIJ, С. J., 1978, Biogeography and adapta- tion: patterns of marine life. Harvard University, Cambridge, Mass., 332 p. VERMEIJ, С. J., 1985, The biology of human- caused extinction of species. In NORTON, B. G. 8 SHUE, H., ed., The preservation of species. п press. VERMEIJ, С. J., ZIPSER, E. & DUDLEY, E. C., 1980, Predation in time and space: peeling and drilling in terebrid gastropods. Paleobiology, 6: 352-364. APPENDIX Pliocene molluscs and their living descendants in tropical America. 1 = Caloosahatchian endemic. 2 = Atlantic Gatunian endemic. 3 = Pacific Gatunian endemic. 4 = Pacific and Atlantic Gatunian. 5 = Caloosahatchian and Atlantic Gatunian. 6 = Pan- American. A = Western Atlantic. Р = Eastern Pacific. ex = extinct. s = soft-bottom. h = hard- bottom. Taxon Pli Rec Habitat Turritellidae Bactrospira CAeX 5 Broderiptella CAE $ Eichwaldiella 6 ex $ Lemintina 4 ex 5 Springvaleia COX $ Torcula 0 АР S Torculoidella 1 ex Ss Turritella 6 AP 5 Vermicularia 6 AP h Strombidae Dolomena 27 пех $ Labiostrombus 2 ex $ Lentigo SER $ Strombus CAPTA? $ Tricornis gigas group 1 AP $ Т. gallus group AO $ Сургаеасеа Cymbula 4 АР h Cyphoma ATAR h Cypraeovula 1 ex h VOKES, E. H., 1970, The genus Trajana (Mollusca: Gastropoda) in the New World. Tulane Studies in Geology and Paleontology, 7: 75-83. VOKES, E. H. 8 D'ATTILIO, A., 1982, Review of the muricid genus Attiliosa (Mollusca: Gastropoda). Veliger, 25: 67-71. WEYL, P. K., 1968, The role of the ocean in climatic change: a theory of the ice ages. Meteorology Monographs, 8: 37—62. WOODRING, W. P., 1966, The Panama land bridge as a sea barrier. Proceedings of the American Philosophical Society, 110: 425—433. WOODRING, W. P., 1973, Geology and paleontol- ogy of Canal Zone and adjoining parts of Pan- ama: description of Tertiary mollusks (additions to gastropods, scaphopods, pelecypods: Nucu- lidae to Malleidae). [United States] Geological Survey Professional Paper, 306-E: 453-539. WOODRING, W. P., 1982, Geology and paleontol- ogy of Canal Zone and adjoining parts of Pan- ama: description of Tertiary mollusks (Pe- lecypods: Propeamussiidae to Cuspidariidae; additions to families covered in P 306-E; addi- tions to gastropods; cephalopods). [United States] Geological Survey Professional Paper, 306-F: 541-759. Eocypraea 2. ex h Erato 6 АР h Erosaria 47 -Р h Jenneria GR h Luria 4 AP h Macrocypraea 4 АР h Marginocypraea 2 ex h Muracypraea 4 А 5 Propustularia 2 А h Pseudocyphoma 2 А h Pseudozonaria дер h Pustularia CAEN r Pusula 6 AP h Simnia 6 AP h Siphocypraea 1 ех 5 Тима 6 АР h Tonnacea Casmaria 4 АР 5 Cassis 6 А $ Dalium 4 A $ Cypraecassis AA $ Echinophoria 4 АР $ Ficus communis group фо $ Е. lanza group SEX $ Е. ventricosa group 4 АР $ Мага о р $ Miogalea LAIEX $ Morum 4 АР h Neosconsia 3 ex 5 Oniscoidea 6 АР h Sconsia BEER 5 Semicassis CA 5 Топпа 6 АР $ EXTINCTION IN TROPICAL AMERICAN MOLLUSCS APPENDIX CONTINUED Pliocene molluscs and their living descendants in tropical America. 1 = Caloosahatchian endemic. 2 = Atlantic Gatunian endemic. 3 = Pacific Gatunian endemic. 4 = Pacific and Atlantic Gatunian. 5 = Caloosahatchian and Atlantic Gatunian. 6 = Pan- American. А = Western Atlantic. Р = Eastern Pacific. ex = extinct. s = soft-bottom. h = hard- bottom. Taxon Pli Rec Habitat Muricidae Acantholabia 1 ex h Acanthotrophon 1 AP h Aspella 6 HAIR h Attiliosa 4. АР h Calotrophon 6 AP h Chicoreus florifer group 1 А h C. brevifrons group CEA h C. shirleyae group ES h Dermomurex 1 AP h Eupleura caudata group 1 А h E. thompsoni group 4 UAB h Favartia 6 AP h Laevityphis A h Microrhytis чех h Miocenebra 1 ex h Muricanthus 6 AP h Muricopsis 1 AP h Murex CAP Ss Murexiella 6 AP h Murexsul 1 A h Neurarhytis 1 ex h Panamurex 6 А h Phyllonotus 6 ПАР $ Pilsbrytyphis 2 ex $ Pteropurpura 4 АР h Pterorhytis 1 ex h Pterotyphis 4 АР $ Pterynotus 4 АР h Purpurellus AE h Rugotyphis 1 ex $ Siphonochelus 2 А h Siratus 2 А h Subpterynotus 6 ex h Talityphis CANAS $ Trachypollia 6 AP h Tripterotyphis 6 AP $ Typhinellus SA $ Urosalpinx c A h Vitularia GIP h Genus unnamed 1 ех h Columbellidae Alcira LEX $ Anachis GAP h Astyris CANTAR h Cigclirina 4 Р $ Columbella 6 AP h Columbellopsis 2 А h Conella CIA h Costoanachis 6 AP h Cotonopsis Eurypyrene Litotrema Macgintopsis Mazatlania Metulella Mitrella Nassarina Nitidella Parametaria Parvanachis Sincola Streptorygma Strombina Strombinella Zafrona Zanassarina Buccinidae Agassitula Antillophos Bailya Calophos Celatoconus Cymatophos Engina Floritula Fusinosteira Gemophos Gordanops Metaphos Metula Minitula Monostiolum Nerva Nicema Northia Pisania Ptychosalpinx Rhipophos Solenosteira Strombinophos Thiarinella Trajana Fasciolariidae Cinctura Dolicholatirus Fasciolaria tulipa group F. gorgasiana group Fusinus Heilprinia Luecozonia Liochlamys Pleuroploca Polygona Terebraspira Triplofusus Olividae Agaronia Ancilla Callianax Dactylidella Dactylidia ONA=B=-NAAD=DRAÁ0ODAÁA=DDDODANm OPrPODOOOD-OUIR—-DO> OP PB ON UN a nRRORB CO (o Е ФИ E SAA (ЕЕ DINA IIA HP) TEN ФО O NN OOS OO SO Sow sn 00000 40 VERMEN 8 PETUCH APPENDIX CONTINUED Pliocene molluscs and their living descendants in tropical America. 1 = Caloosahatchian endemic. 2 = Atlantic Gatunian endemic. 3 = Pacific Gatunian endemic. 4 = Pacific and Atlantic Gatunian. 5 = Caloosahatchian and Atlantic Gatunian. 6 = Pan- American. A = Western Atlantic. Р = Eastern Pacific. ex = extinct. s = soft-bottom. h = hard- bottom. Taxon Pli Rec Habitat Eburna SA 5 Jaspidella A S Macgintiella IA 5 Mansfieldella 1 ex $ Minioliva 4 АР $ Neocylindrus ¡MEX $ Niteoliva 4 AP $ Oliva 6 AP $ Olivella 6 АР $ Отодутпа 2 ex 5 Pachyoliva 4 P 5 Strephona 2 ex 5 Strephonella a > 5 Toroliva 5 ex $ Mitridae Atrimitra 4, (à h Dibaphimitra 1 A $ Dibaphus 2 1eXx $ Fusimitra 6 AP $ Isara 4 АР $ Nebularia AA h Pleioptygma 1 A $ Prochelaea EX $ Scabricola 4 АР h Strigatella A h Subcancilla 6 AP 5 Tiara 4 P Ss Volutidae Aurinia 1 A $ Calliotectum 8 7 5 Clenchina 1 A 5 Enaeta 6 AVP 5 Harpeola 1 ex $ Lyria БА $ Mysterostropha 3 ex $ Plicoliva* 2 А S Scaphella 1 A $ Voluta 2 А $ Marginellidae Bullata 5 А $ Cypraeolina 1 A Ss Eburnospira 5 A S Egouena 5: A $ Eratoidea 1 A $ Gibberula SNA 5 Leptegouana DA $ Marginella 2 А $ Microspira 6 АР $ Persicula CA s Prunum 6 AP $ Radicea Serrata Volvarina Cancellariidae Admetula Agatrix Aphera Bivetiella Bivetopsia Bonnelitia Calcarata Cancellaria Charcolleria Euclia Extractrix Hertleinia Marksella Massyla Narona Olssonella Perplicaria Pyruclea Sveltia Trigonostoma Ventrilia Conidae Asprella Conasprella Contraconus Leptoconus Lithoconus Pyruconus Stephanoconus Ximeniconus Arcidae Acar Arca Arcopsis Arcoptera Barbatia Caloosarca Cucullaearca Cunearca Eontia Fugleria Grandiarca Granoarca Hawaiarca Larkinia Lunarca Noetia Obliquarca Potiarca Rasia Sheldonella Taeniarca Tosarca Lucinidae Anodontia Armimiltha Bellucina Callucina о u» POPOODFPOODD-TRODODDODD—- POP oO OO + AMDAARHAM--WAMDFAHAADHPHLAOHA D0- 0 nnn фроодофоооо,фо DVD DD Oo TN N nn 0 EXTINCTION IN TROPICAL AMERICAN MOLLUSCS 41 APPENDIX CONTINUED Pliocene molluscs and their living descendants in tropical America. 1 = Caloosahatchian endemic. 2 = Atlantic Gatunian endemic. 3 = Pacific Gatunian endemic. 4 = Pacific and Atlantic Gatunian. 5 = Caloosahatchian and Atlantic Gatunian. 6 = Pan- American. A = Western Atlantic. P = Eastern Pacific. ex = extinct. s = soft-bottom. h = hard- bottom. Taxon Pli Rec Habitat Cavilinga 6 AP Codakia 6 AP Ctena 6 AP Divalinga АР Eulopia 2 А Неге 3 P Lepilucina 4 ex Levimyrtea SINEX Lucina 5 A Lucinisca 67 АР Lucinoma 6 AP Miltha 6 AP Myrtea 2 А Parvilucina OEA. Phacoides 6 A Pleurolucina 6 AP Stewartia 5 А Cardiidae Acrosterigma 1 Р Americardia 6 AP Apiocardia 4 Р Dallocardia 6 AP Dinocardium 6 A Laevicardium 6: "АР Lophocardium 4 Р Mexicardia 4 P Microcardium 4 AP Nemocardium 4 ex Papyridea 6 AP Phlogocardia AP Trachycardium 6 АР Trigoniocardia Gr AIP “Although originally placed in the Olividae by Petuch (1979), the genus Plicoliva Petuch 1979 now appears to belong to the Volutidae, subfamily Lyriinae. This change in familial placement was made in order to accommodate a number of volutid characteristics that are found in Plicoliva, such as an unchanneled suture, few but evenly sized col- umellar plications, and a lyriine papillate pro- toconch. Furthermore, this volutid genus has been found in the fossil record of the Caribbean. An examination of the type of Prochelaea gabbi Pilsbry & Johnson 1917 showed that this species is not referable to that mitrid genus, but belongs to Plicoli- va. P. zelindae Petuch 1979 is a Brazilian relict. \: ES © 6 >. 1 i) tut Ian р 5 | | i y | | Fu | | | ау По | | schen 01 Sie Er remet 15 | п Ала 2 EL \ 254 DEN DEL © M : | ' à e MA Ka MALACOLOGIA, 1986, 27(1): 43-66 THE TAXONOMIC STRUCTURE OF SHALLOW-WATER MARINE FAUNAS: IMPLICATIONS FOR PHANEROZOIC EXTINCTIONS David Jablonski? & Karl W. Flessa? ABSTRACT The taxonomic and biogeographic structure of Recent shallow marine faunas provides a means of evaluating the causes and magnitudes of extinctions in the fossil record. We assem- bled data on the distribution of families of marine gastropods, bivalves, echinoderms and scleractinians and on the number of species within families in gastropod, bivalve and echinoid faunas. The 22 oceanic islands for which we collected data harbor a very large proportion (87%) of the global, shallow water marine fauna, and 78% of the families are at two or more of the 22 islands. This suggests that even if eustatic lowering of sea level ravaged the continental shelf faunas, oceanic islands would provide a safe haven for representatives of the great majority of the shallow marine benthic families. Continental shelf bivalve and echinoid faunas have significantly more species per family than island bivalve and echinoid faunas (a proportion of 1.5:1 and 1.3:1, respectively), though gastropod faunas show no such difference. Gastropod faunas display persistently higher species-family ratios than bivalve faunas, and echinoid faunas have the lowest ratios of all three classes. Species-family ratios are diversity-dependent, so that island-continental shelf and class-to-class differences in species-family ratios appear to be a direct consequence of differing species richness among the faunas and classes. The fossil record suggests that species richness within clades may not be an adequate measure of resistance to mass extinction. Tropical clades appear to suffer disproportionately during times of mass extinction, and in general species-rich clades are no better represented among survivors than are species-poor clades. The linkage between speciation and extinction rates generates species-rich, but evolutionarily volatile clades. Species richness within clades may, however, contribute to a clade's resistance to background extinction. That different factors contribute to extinction-resistance during times of mass vs. background extinction suggests that macroevolutionary processes during those times are qualitatively as well as quantitatively different. Key words: biogeography; extinction; species-family ratios; mollusks; echinoderms; corals. INTRODUCTION Extinction is the fate of all species. п addi- tion to persistent levels of background ex- tinctions, at least five mass extinction events of various magnitudes have plagued the ma- rine biota since the beginning of the Phan- erozoic (Newell, 1967; Raup & Sepkoski, 1982). Most recent analyses of extinctions have emphasized global tallies of higher taxa, but this approach yields few insights into the biogeographic or ecologic controls of such events. The biodistributional patterns of victims and survivors hold considerable promise as indicators of the processes un- derlying mass extinctions, but reliable data are difficult to accumulate on the appropri- ately large scales. Examination of present- day taxonomic, ecologic, and biogeographic patterns provides one complementary ap- proach to the extinction problem. For exam- ple, by using the modern biota to estimate probable faunal response to a given pertur- bation, we can evaluate hypotheses avail- able to explain the magnitudes and patterns of extinction observed in the fossil record. Here we analyze the taxonomic and biogeo- graphic structure of recent marine organ- isms, with emphasis on bivalves, gastropods and echinoids, in order to test hypotheses ‘Department of Ecology and Evolutionary Biology, University of Arizona, Tucson, Arizona 85721, U.S.A. Present address: Department of Geophysical Sciences, University of Chicago, 5734 South Ellis Avenue, Chicago, IL 60637, U.S.A. 2Department of Geosciences, University of Arizona, Tucson, Arizona 85721, U.S.A. JABLONSKI & FLESSA 4 "8151099 YINOS ‘WOS !sausJeosey 'SYW 'SESANDIEN ‘YYW ‘2112PEN ‘GVW ‘euenboen ‘OVW :usjenbiey ‘Yay :uaAeyy чег ‘NYP :puejaa] ‘351 ‘немен ‘МУН (wend ‘NO “lia ‘rid :Buluue NV y :$90.эе- ‘31 ‘191523 ‘SWF !4009 ‘OOD ‘BuI88H-S0909 MOD :S0909 ‘909 :uonaddıly ‘119 ‘weuyjeuN ‘ино !epnwueg 'Yy39 ‘S210Zy ‘OZV ‘UOISU99SY ‘OSV ‘елаерм ‘CIV :зиоцелелаам ‘S|2109 UEIUNDEIEIOS pue ‘зрюшцое ‘Splouniydo ‘spioyayse “spodosseb youeiqosoid ‘зэлема 10} pajoa¡109 элэм |9A8| ¡e1ue, ay) Je вер eouasqe-aouasald yolym 10} Spuejsi 9IUBa9O ‘| "DIA TAXONOMIC STRUCTURE AND EXTINCTIONS 45 and make inferences regarding mass extinc- tions in the geologic past. Because the present-day fauna is by far the best-known biota of the Phanerozoic, it can provide a relatively complete—if still imperfect—assessment of patterns of spe- cies-level diversity and distribution, and how those patterns relate to the higher taxonomic levels that usually serve as bases for paleon- tological analysis (see Raup, 1979a, for a discussion of the biases hindering species- level studies of Phanerozoic diversity). In many respects the Recent biota represents a biogeographic extreme for Phanerozoic marine invertebrates: this is apparently a period of near-maximum latitudinal thermal gradients and maximum development of longitudinal land barriers, yielding very high provinciality and presumably minimum mean geographic ranges for species and higher taxa (see Valentine, 1969, 1973; Valentine et al., 1978; Raup, 1982). Conse- quently, analysis of present-day geographic distributions can give us an end-mem- ber against which to test extinction hypothe- ses. One extinction hypothesis for marine organisms is the proposed link between marine regression and mass extinction through species-area relationships. Some authors have emphasized climatic and other indirect effects of regression as a causal factor (e.g., Haq, 1973; Fischer 8 Arthur, 1977; Cavelier et al., 1981), but Schopf (1974) and Simberloff (1974) have main- tained that the Permo-Triassic extinction, the largest of the Phanerozoic, can be explained directly through loss of habitable shelf area (but see Schopf, 1979; Wise & Schopf, 1981, for a very different view- point). The species-area relationship is often best described by the power function S = kA’, where $ is the number of species, A is the area of the geographic isolate, and k and z are fitted constants, with z values usu- ally clustering near 0.3 (Connor & McCoy, 1979; Flessa 8 Sepkoski, 1978). Though grounded in studies of island biogeography, its extrapolation to larger-scale questions of global extinction and evolution has met con- siderable acceptance among paleontologists (e.g. Flessa, 1975; Gould, 1976; McLaren, 1983). This is despite misgivings among some biologists regarding the general ex- planatory power of island biogeographic the- ory (e.g., Simberloff, 1976, 1981; Connor 8 McCoy, 1979; Gilbert, 1980) and the actual applicability of the species-area relationship to the fossil record (Flessa 8 Sepkoski, 1978). In this paper we assess the probability that a marine regression could produce a mass extinction in the Recent biota through reduction in habitable shelf area. We test this hypothesis by comparing shallow marine faunas of continental shelves to those of oceanic islands. Islands would ap- pear to be immune from the effects of lower sea level. While drops in relative sea level reduce habitable epicontinental and conti- nental margin sea, conical oceanic islands will actually gain slightly in perimeter and therefore in shallow-water area (see Stanley, 1979; Jablonski, in press a). At the same time, previously-drowned seamounts emerge into the photic zone, tending to replace lagoonal habitats that may be exposed dur- ing regression, so that net destruction of habitat types also will be limited in oceanic settings. In order to estimate the proportion of the marine biota inhabiting oceanic islands, and thus exempt from extinction by area effects during regression, we surveyed the literature of island faunas for representa- tives of the benthic families of six skeleton- ized invertebrate classes. There is, of course, no guarantee that those families would persist in their island refugia for geo- logically significant periods of time. For ex- ample, if all the families are monospecific on each island, normal attritional extinction could rapidly remove a significant number of families. To test for this possibility, we com- pare the frequency distribution of species within families for mainland and island fau- nas. These frequency distributions are sim- ply the “hollow curves” of Willis (1922; see also Williams, 1964; Anderson, 1974). The shape of the species-family frequency distribution, or simple species-family ratios, may at least partly determine a clade's resistance to extinction; all other factors being equal, the more species in the family, the lower its probability of extinction. This probabilistic model of extinction has received few empirical tests, and we provide a prelim- inary comparison among bivalves, gastro- pods and echinoids to assess the role of tax- onomic structure in shaping rates and pat- terns of extinction of major groups of marine organisms. 46 JABLONSKI & FLESSA AAA >> FIG. 2. Oceanic islands and mainland areas for which frequency distributions of species within families were collected for bivalves, shelled gastropods, and echinoids. Abbreviations: ALD, Aldabra; ARA, southeast Arabia; ARC, Arctic Canada; ASC, Ascension; BCO, British Columbia; BER, Bermuda; BON, Bonin; BRA, Brazil; BRI, Britain; CAY, Grand Cayman; CHA, Chatham; COK, Cocos-Keeling; EAS, Easter; FAE, Faeroes; FAL, False Bay; FAN, Fanning; FUN, Funafuti; САО, Gulf of Aqaba; GHA, Ghana; GOA, Arabian Gulf; СОМ, Gulf of Mexico; GRE, Grenada; HAW, Hawaii; ICE, Iceland; ITA, Italy; JAM, Jamaica; JAN, Jan Mayen; JAP, Japan; KRG, Kerguelen; KRM, Kermadecs; KUT, Gulf of Kutch; MAL, Maldives; MAR, Marquesas; MAU, Mauritius; MED, Mediterranean; MEX, Pacific Mexico; MON, Monterey Bay; NIU, Niue; NZA, New Zealand, Aupourian Province; NZC, New Zealand, Cookian Province; NZF, New Zealand, Forsterian Province; OMA, Oman; PAG, Pagan; PAN, Atlantic Panama; PTB, Point Barrow, Alaska; RED, Red Sea; SAF, South Africa; SAG, Sagami Bay; SPM, Spanish Mediterranean; SUE, Gulf of Suez; SUR, Surinam; TEX, Texas; WFL, West Florida. MATERIALS AND METHODS Families. Distribution of families on islands was assessed using a revised and updated version of Jablonski’s (in press a) data. The distributions are based on published records for 22 oceanic islands of three major phyla of benthic marine invertebrates (see Faunal Ref- erences): Mollusca (bivalves, prosobranch gastropods), Echinodermata (asteroids, oph- iuroids, echinoids), and Coelenterata (scler- actinian corals). Only volcanic islands sur- rounded by oceanic crust were used. Classi- fication follows the Treatise on Invertebrate Paleontology and, for gastropods, Taylor & Sohl (1962). To compensate for differences in sampling and preparation procedures among islands, families were included only if living representatives occur in depths of 100 m or less, and if maximum adult size exceeds 5 mm. A total of 276 families met these criteria. Sites ranged from one of the world’s most northerly islands, Jan Mayen (71° N), to South Georgia and Macquarie Islands (both approx- imately 54° S), and include localities from most of the major marine biogeographic re- gions (Fig. 1). Species. Frequency distributions of species within families were compiled from the litera- ture for oceanic islands (15 molluscan faunas, 13 echinoid faunas) and continental shelves (22 molluscan faunas, 15 echinoid faunas) (see Faunal References). Owing to their prox- imity to continental shelves, Caribbean is- lands were recorded as continental rather than oceanic. In data compilations such as this, presence/absence of families is a far more reliable observation than the detailed apportionment of species within each family, and critical evaluation of the island faunas led to deletion of some that seemed complete at the family level but considerably less so for species; other islands were added that lacked the broad multi-phylum coverage needed for the familial analyses but included good ac- counts of one or more of the three classes of most interest. Consequently, the islands char- acterized at the species level do not constitute an exact subsample of those characterized at the familial level. Also, shelled opisthobran- chs as well as prosobranchs were included in the gastropod species counts. So that the data on the shelf biotas would be comparable to the island ones, we attempted to use continental shelf data from individual sites or small, well-defined areas rather than large-scale regional compi- lations. Continental shelf coverage ranges from the Canadian Arctic (up to about 80° N) to the Forsterian Province at the southern end of New Zealand (ca. 47° S). As with the oceanic islands, we were able to include most of the major marine biogeographic re- gions (Fig. 2). Despite our critical approach, we recognize that the species-level data must still be considerably more heterogeneous and less reliable than the family data. Nevertheless, because families can go extinct only if their member species do so, we feel that our species-level data, however flawed, pro- vide some insight into the processes of ex- tinction. RESULTS Oceanic islands harbor a remarkably large proportion of today’s shallow-water families (Fig. 3). Of the 276 families considered here, 239 (87%) have species recorded from one or more of the 22 oceanic islands in Fig. 1, and at least 200 (78%) are present on two or more TAXONOMIC STRUCTURE AND EXTINCTIONS 47 BIVALVIA GASTROPODA ECHINOIDEA 80% ON ISLANDS 97% ON ISLANDS 76% ON ISLANDS dl 90 40 © о "| — FAMILIES wo о —l o 90 82 38 ASTEROIDEA OPHIUROIDEA SCLERACTINIA 82% ON ISLANDS 76% ON ISLANDS 100% ON ISLANDS 30 30 30 Ф20 20 20 WwW - = ео 10 10 0 0 0 28 23 17 13 D 21 FIG. 3. Number of shallow-water families in the world biota (white bars) represented on 22 oceanic islands (black bars). At the family level, a high proportion of the living biota is represented on oceanic islands. 48 JABLONSKI 8 FLESSA of the islands. These results are essentially identical to Jablonski's (in press a) earlier analysis of the same groups. Classes are not equally represented, but even the more poorly represented classes, the ophiuroids and echinoids, have about 76% of their families recorded on the 22 islands (70% and 66%, respectively, on two or more islands). Next come the bivalves and asteroids, each with about 80% of their families on the islands (70% and 64% on two or more islands). Es- pecially well-represented are the prosobranch gastropods (99% on at least one island, 85% on two or more islands) and the scleractinian corals (100% on at least one island, 90% on two or more of the islands). However, at the species level the mainland and oceanic island faunas exhibit different taxonomic structures for two of the three classes we examined in detail. Both the echinoids (Figs. 4-9) and the bivalves (Figs. 10-18) have fewer species per family in the oceanic island faunas than in the continental shelf faunas, and also have a higher propor- tion of monospecific families (in effect, а mea- sure of the steepness of the hollow curve) on islands relative to mainlands. These differ- ences are significant at p = 0.05 by the Wilcoxon two-sample test (Sokal 8 Rohlf, 1969: 391-395). Although means of such non-normally distributed ratios are difficult to interpret, they are a rough measure of the magnitude of the differences. Mainland bivalves have an average species-family ratio of 3.76 and an average proportion of monospecific families of 34%; for island bi- valve faunas the average species-family ratio is 2.32 and the average proportion of monospecific families is 48%. Mainland fam- ilies, then, exceed island bivalves in average number of species per family by a factor of approximately 1:1.5. Mainland echinoid fau- nas have an average species-family ratio of 2.29, and the average proportion of mono- specific families is 42%; for island faunas the average species-family ratio is 1.74 and the average proportion on monospecific families is 58%. Mainland echinoid families, then, ex- ceed island families in average number of species per family by a factor of about 1:1.3. In contrast to the echinoids and bivalves, gastropods (Figs. 19—31) show no significant difference between island and mainland fau- nas in species-family ratios or percentage of monospecific families (Wilcoxon two-sample test). Not only 1$ there variation among areas for two of the three classes we analyzed, but there is significant variation among the three taxa, with species-family ratios significantly higher for gastropods than for bivalves, which in turn are significantly higher than echinoids. Accordingly, the percentage of monospecific families 15 significantly lower for gastropods than for bivalves, and significantly lower for bivalves than for echinoids (Wilcoxon signed ranks tests, p = 0.01). DISCUSSION The family data suggest that reduction in continental shelf area during marine regres- sion is not sufficient in itself to produce a mass extinction at the familial level in marine invertebrates (see also Jablonski, in press a). Even if the entire continental shelf biota was eradicated, this would remove at most 13% of the Recent families; the rest are estab- lished on islands not subject to reduction in habitable shallow-water area during a drop in relative sea level. As Jablonski (ibid.) pointed out, this is a conservative estimate, because: (1) few scenarios include complete annihila- tion of the shelf fauna—some families should persist there as well; (2) only 22 islands out of the thousands actually present in the world ocean were surveyed, and even for these the faunas are doubtless incompletely mono- graphed; and (3) provinciality is unusually high in today’s oceans, even at the family level (e.g. Valentine, 1973; Campbell & Val- entine, 1977), so that on average families in the geologic past can be expected to have been even more widespread and better-repre- sented on oceanic islands than in the Recent. We do not claim, however, that oceanic is- lands have constituted the sole haven for shallow-water benthos during marine regres- sions. Because less radical environmental re- ductions than postulated here are surely the rule, other refugia will also be available. Therefore survivorship of taxa within prov- inces need not be closely related to the num- ber of islands within the region [see, for ex- ample, Vermeij & Petuch’s (1985) and Ver- meij's (in press)] observation of greater sur- vivorship of Neogene molluscan taxa in the island-poor tropical Eastern Pacific relative to the island-rich Caribbean. Habitats will not have equal probabilities of 49 TAXONOMIC STRUCTURE AND EXTINCTIONS ‘зпцийпеи| pue 'uajanblay ‘e1qep|Y 'DIPeUIIA ‘SOAIPIEN ‘UIUOG :SEUNEJ BOW YINOS pue ‘шея ‘Aje} ‘O9IXON 911984 :Seuney PIOUIU98 pues! эшеээо ш Suonnqinsip Aouanbay Анше}-эюэ4$ ‘2 ‘914 plouIy9e yays ¡ejueupyuos и! Suonqusip Ácuanbas] Апше!-заюэ4$ ‘9 ‘914 ot WV4NI dS OL OL Wv4NI dS eS < ol OL OL ol OL OL Ot OL эзачинзя | AAGIWW | NINO8 ATW LI O9IX3W 91419Vd OL 0 OL ol OL V3QIONIHD3 V3QIONIHD3 ‘INH UeIqe1y pue 'zans jo JIN9 ‘заэиалаэН ‘еаелу илэ}5еэцпос ‘eqeby jo no ‘(2861 ‘29114 чо paseq) eas pay :$еипв} гипе- ay] pue 'sONSIYEIS BAIND 10} | э4е1 ses ‘saloeds Z| surejuoo Аше; PIOUIYOS дэц$ |езиэициоэ и! suonnquisip Aouanbau Ашие]-5э19э4$ “Gg ‘914 auo Ajuo alum ‘adaide зэюэ4$ эио aney salle, Z| ‘uedeg 10} “ajduexa 104 ‘salads “ja ‘€ ‘с ‘| Buiney випе} yoes ul Saillie} jo 1aquinu eu} MOUS swe160}SIH ‘(еэц\у YINOS) {eg asje4 pue ‘рчееэу мэм ‘ueder 'ueauella) Wv4 Be ‘dS рей ysiueds ‘(0861 ‘UeWOY UO paseq) eas pay ‘oaıxayy JO ¡no :seuney 3 рюшцоэ yeys ¡ejuaunuos и! зиоцпаер Aouenbey Апше}-зэаюэ4$ “y ‘914 о 319 NVISVYV om ATIWW4 NI $3036 vIgvuv 3S 5 ot Ot o! 3 m U 230$ 30 319 AVS 3514 ONVIVIZ MIN Nvdvr AON3NO SHS g S VAGIONIHD3 = vas аэн [1$ ot V8avov 30 41n9 x LIGSW HSINVdS VAGIONIH93 и 50 JABLONSKI 8 FLESSA TABLE 1. Species-family frequency data for continental shelf and oceanic island echinoid faunas. See Fig. 2 for locations, and Figs. 4-9 for frequency distributions. Number of Location species |. Continental shelf faunas Southeast Arabia 34 Britain 19 False Bay (South Africa) i Arabian Gulf 19 Gulf of Mexico 45 Gulf of Aqaba 29 Italy 25 Japan 95 Pacific Mexico 67 New Zealand 28 Red Sea 48 South Africa 35 Spanish Mediterranean 46 Gulf of Suez 25 |. Oceanic island faunas Aldabra 30 Ascension 10 Bermuda 13 Bonin 22 Easter Y Faeroes 9 Hawaii 27 Iceland 11 Jan Mayen 3 Kerguelen 4 Kermadecs 11 Maldives 22 Mauritius 41 Pagan 17 elimination during regression—there will al- ways be terrigenous intertidal and innershelf environments, for example. Therefore, taxa that penetrate shallow-water or intertidal en- vironments would be less likely to suffer hab- itat loss than those confined to deeper-water environments on the shelf, and a supplemen- tary approach to our island-mainland compar- isons might be a tabulation of families having representatives in such nearshore habitats. However, there is a confounding factor that must be taken into consideration before this can be tested in the fossil record: taxa in these nearshore environments have biological at- tributes (e.g. broader environmental toler- ances, dispersal capabilities, and geographi- cal ranges than offshore taxa) that further enhance a clade's extinction-resistance (see Jackson, 1974; Jablonski, 1980, 1982; Ja- blonski & Valentine, 1981). Consequently, preferential survival of nearshore taxa across Number of | % monospecific Species-family families families ratio 14 36 2.4 8 42 2.4 5 43 1.4 lal 54 17 20 30 2.2 13 38 2.2 12 50 2.1 28 43 3.4 17 24 3.9 14 5% 2.0 18 39 277, 17 47 2.0 24 54 1.9 14 36 1.8 13 31 2.3 8 75 ES 10 70 ES 11 73 2.0 6 83 1.2 6 50 eS 9 11 3.0 9 78 ez 2 50 18 3 67 19 9 67 ES 14 69 1.6 15 33 DA, 8 50 eal extinction boundaries might be for reasons other than the inevitable persistence of their habitat types during regression. The species-level data present a somewhat different picture from the family level data. There are signficantly fewer species per fam- ily on islands relative to continental shelves for echinoids and bivalves, though not for gastropods. This raises the possiblity that, at least for some clades, families on islands may be more extinction-prone than they are on the mainland. However, it is not clear that differ- ences in island-mainland species-family ra- tios ranging from 1:1.3 to 1:1.5 are sufficient to offset the advantages of widespread distri- bution at the family level among oceanic is- lands. The fact that some, but not all, clades have different taxonomic structures on islands vs. mainlands adds another level of complex- ity to the problem, and deserves further inves- tigation. TAXONOMIC STRUCTURE AND EXTINCTIONS 51 < LL] = O E a О E Lu M Z : < = > © < = a ene of = а. (dp) м < LL] = O = 2 = = O Lu BERMUDA 5 SP. IN FAM. 5 5 ICELAND 5 5 5 5 ASCENSION FAEROES JAN MAYEN 5 A9NINOJ84W3 FIG. 9. Species-family frequency distributions in oceanic island echinoid faunas: Pagan, Hawaii, and Easter. FIG. 8. Species-family frequency distributions in oceanic island echinoid faunas: Iceland, Bermuda, Ascension, Jan Mayen, and Faeroes. The basic factor governing species-family relationships appears simply to be local spe- cies richness. Species-family ratios are diver- sity- dependent, so that the species-family ra- tio increases monotonically with number of species. This 15 in itself a consequence of the curvilinear shape of the relationship between families and species in a fauna or clade (Figs. 32, 33). Though always a positive function, the slope of the species vs. family curve de- creases, so that larger faunas or clades have a higher species-family ratio than small fau- nas or small clades. Simberloff (1970, 1978) and Járvinen (1982) discuss this aspect of taxonomic structure within clades as it affects species-to-genus ratios. There are no significant differences among the three groups we examined in the relation- ship between species richness and the appor- tionment of these species among families. Confidence limits for the slopes of the regres- sion lines (log-transformed data) all overlap. Thus, echinoids do not have exceptionally low species-family ratios, given the number of species in the class; at each locality their taxonomic structure simply conforms to that of the bivalves and gastropods. Similarly, the gastropods do not have inordinately high spe- cies-family ratios (despite, for example, the apparent outlier of 11.53 for the Hawaiian fauna), but fall on the line projected for bivalves and echinoids. Because the clades accumulate families at a similar proportion as species richness increases, no major differ- ences in macroevolutionary processes among the three groups are indicated. A given speciation event is as likely to lead to a new family in the echinoids as in the gastropods, and no group is more adept than the others in producing new families in the course of spe- ciation. The pattern suggests, therefore, that the differences in taxonomic structure among these clades is a consequence of local spe- cies diversities. As a group, the echinoids are expected to have a low species-family ratio, because they have few species per locality; bivalves are intermediate and gastropods have highest species richness and thus high- est species-family ratios. This pattern is also the basis of the island-mainland differences. Island bivalve and echinoid faunas are rela- tively paucispecific, and thus their species- family ratios are lower than on the mainland. Gastropod faunas have comparable numbers of species on islands and continental shelves, JABLONSKI & FLESSA 52 ‘эоум е SE иваиеллэураи\ AU} PUE ‘иешо ‘eueYy ‘шеи :SEUNE} эмела иэцз ¡ejuaunuos ui suonnquisip AoUanba.y Ápiuej-sandads “El ‘914 WV3N dS o! oe 02 ot NYWO 5 < ol OL NV3NVHH3LIO3WN ot oz ot YNVH9S му Lidge ol ot VIATVAIS "(ESB y) моллея JUIOd pue “eiquinjo) у$циа '9191y ueipeueo ‘(елое)) {eg Áalajuoy :seune} элела Jays ¡ejuaunuos ш зиоцпащер Aouanbau Ápuuej-sendads ‘|| ‘914 АЛМУ 3 М 590345 02 o! 02 о U т po] = MOYHVE Id 4 VISNN109 HSILIHS 2 ol ol 02 OL 02 ol ¡ENE DILOYY NVIOVNVO АУЯ A3H31INOW ol ol VIATVAIg ешвие« 91 ue y pue ‘шеципб ‘ереиал9 'puejsj ueWÄeN puelo) :seunej эмела Hays ¡ejuaunyuos и! suonnquisip Aouanbayy Апше}-зэюэ4$ “ZI ‘914 AIWVY1 NI S31934S 02 ol 02 OL VWVNVd OLLNVILY WVNIUNS ™ AONANOSY4 ol 0 VOVNIYO NVWAVO ANVH9 VIATVAI où ol 'sa9uaJaj9oy Jeuney au) pue ‘$94$4е]$ эл/пэ JO} г э!ае | ээс ‘(eouy yinos) {eg asje 4 pue “¡ze1g ‘(ueder) {eg iwebes :seune} эмела JUS ¡jejuaunuos ш зиоцпаиер AoUanbay Апше]-5эю0э4$ ‘OL ‘914 АЛМУЗ М $90345 ol 0€ 02 o! AON3N0 344 Ava 357193 11ZVH8 o ol МУЗУГ ‘Ava INVOVS VIATVAIg TAXONOMIC STRUCTURE AND EXTINCTIONS 53 TABLE 2. Species-family frequency data for continental shelf and oceanic island bivalve faunas. See Fig. 2 for locations, and Figs. 10-18 for frequency distributions. Number of Location species |. Continental shelf faunas Arctic Canada 68 British Columbia 121 Brazil 291 Britain 183 Grand Cayman Is. 73 False Bay (South Africa) 90 Ghana 61 Grenada 40 Jamaica 235 Gulf of Kutch 91 Mediterranean 149 Monterey Bay (California) 172 Aupourian Province (New Zealand) 264 Cookian Province (New Zealand) 152 Forsterian Province (New Zealand) 169 Oman 92 Atlantic Panama 138 Point Barrow (Alaska) 37 Sagami Bay (Japan) 393 Surinam 135 Texas 145 West Florida 155 |. Oceanic island faunas Aldabra 93 Ascension 22 Bermuda 122 Chatham 114 Cocos-Keeling 68 Easter 15 Faeroes 83 Fanning 52 Funafuti 79 Hawaii 139 Iceland 93 Jan Mayen 28 Kerguelen 29 Marquesas 11 Niue 6 Pagan 26 and as expected they lack significant differ- ences in species-family ratios between the two habitats. lt appears that what needs to be explained is not a source of evolutionary nov- elties that might maintain a high proportion of families to species in island faunas, but the reasons for such prolific within-habitat spe- ciation in gastropods relative to bivalves and echinoids, or, from a different perspective, why bivalves are impoverished at the species Number of % monospecific Species-family families families ratio 21 33 3.2 35 37 3.5 51 20 5.7 45 29 4.1 20 30 3.6 37 41 2.4 27 44 2.3 18 90 2.2 43 30 9.9 27 37 3.4 49 37 3.0 38 26 4.5 48 27 5.5 40 20 3.8 41 22 4.1 31 92 3.0 42 36 3.3 14 36 2.6 66 17 6.0 38 45 3.6 44 43 3.3 38 32 4.1 29 32 3.2 15 67 1.5 39 41 3.1 37 27 3.1 28 39 2.4 13 85 1.2 34 47 2.4 23 43 2.3 32 62 2.5 39 26 3.6 30 37 3.1 12 25 2.3 16 50 1.8 8 75 1.4 5 80 12 13 38 2.0 level on tropical oceanic islands (possibly ow- ing to lack of suitable habitats on atolls). Thus, the pattern of distribution of species-family ratios among clades could result from (a) dif- ferential extinction rates near an equilibrium (for example, differences in their ability to accommodate new species as they arrive), or (b) differential speciation rates in a situation far from equilibrium. What, then, 1$ the relationship between tax- JABLONSKI & FLESSA 54 ‘UOISU99SY pue 'usÁeyy чег 'puejao] ‘ерпшаэ8 'saoJaey :seuney 'ueBeg pue 'Buluuey 'nnjeuny ‘sesanbieyy ‘eniN ‘иемен :seunej эмела Puejsı omesoo ш suoynquisip Аэцепбэц Апше}-5э10949$ “ZL ‘914 эмела pue¡si эшеээо ul зиоцпаиер Aouenbay Alıwej-seweds ‘91 ‘914 OL ol OL ИУ М dS res ol S NOISN39Sv N3AVW NYP GNV1390! NVOYd ILNJVNN4 ONINNY 4 ol OL ot 0+ o! ol 02 OL OL 01 3NIN ol S3083VWW3 ИУМУН ol SVS3NOYVIN ol o! OL AIN3INO38YW3 2 o АЭМЭПОЗНЗ ol vanny38 a VIATVAIA "yoıny JO ипэ pue '(puejeaz мэм) ээшлола чеиподпу “(pueje ‘воешег pue ‘ерис|+ 1sam 'sexa] :seung, -8Z MON) ээшлол4 UBINOO) ‘(ришеэ? MON) ээшлол4 иеиз}$104 ‘5еипе} эмела Jays ¡ejuaunuos ul зиоцпацер Аэцэпбэц Апше}-зэю09э4$ “pl ‘914 элела Jjeys ¡ejuaunuos Ul зиоцпащер Aouanbau Ápiuej-seidads “Sl ‘914 АТИУЗ М S31934S oe ИЗМ dS 02 0+ 02 OL 02 ol VOIVNVIP HOLNA 30 31N9 (ZN) NvIHNOdNY OL 02 ol ol OL SvX31 vVOIHOI3 1S3M oL A CZN) NVINOOO (ZN) NVIH3LSHOJ VIATVAIg or 2 VIAIVAIS ™ AON3NO343 ? A9N3NO383 O o o N TAXONOMIC STRUCTURE AND EXTINCTIONS 55 BIVALVIA 10 1 10 10 COCOS-KEELING KERGUELEN EASTER ALDABRA 10 CHATHAM SP. № РАМ. 10 (o) 10 AON3N0384 FIG. 18. Species-family frequency distributions in oceanic island bivalve faunas: Chatham, Aldabra, Cocos-Keeling, Kerguelen, and Easter. onomic structure and extinction? It might be predicted that variations in species-family ra- tios determine susceptibility to mass extinc- tion among clades: given the same percent- age of species extinction, paucispecific clades are more likely than speciose clades to be reduced to such low species richnesses that random events will remove the last few surviving species. However, although this generalization may hold for extreme cases, it is not sufficient to explain patterns of extinc- tion among clades during mass extinctions (Jablonski, in press). Many low-diversity clades exhibit impressive longevities, and high species richness does not always confer great extinction-resistance on a clade. For example, Ward & Signor (1983) found that speciose ammonite clades were among the shortest-lived. With respect to mass extinc- tions, Jablonski (in press b) found that paucispecific bivalve and gastropod genera had an equal, or even higher, probability of surviving the Cretaceous-Tertiary boundary than speciose genera (Fig. 34) and the liter- ature shows that mass extinction events ter- minate both paucispecific and species-rich clades (e.g. both trilobites and productid bra- chiopods at the end of the Permian). Taxonomic structure alone does not appear to be a major determinant of extinction or survival among clades during mass extinc- tions, probably because biological attributes that affect extinction and speciation rates are not randomly distributed among taxa but tend to covary. Consequently, clades that tend to exhibit high speciation rates tend to be extinc- tion-prone as well, and cladesthattendtoresist speciation tend to have extinction-resistance species. For example, in bivalves and gastro- pods broad larval dispersal capability tends to reduce speciation rates and is generally ac- companied by broad geographic ranges and high degree of environmental tolerances, which also impart extinction-resistance to those species. Conversely, species with low dispersal capabilities, and thus high speciation rates, also tend to have restricted geographic ranges and narrow environmental tolerances, and as predicted exhibit high extinction rates as well (Jackson, 1974; Jablonski, 1980, 1982; Hansen, 1980; Jablonski 4 Lutz, 1983). There- fore, the attributes that cause a clade to exhibit high speciation rates also impart upon it a volatility that may make it particularly vulner- able to environmental changes during mass extinction events (Stanley, 1979; Jablonski, in press b). Biogeographic distributions may be far more important than species richness in de- termining a clade's survival or disapperance during mass extinctions. The few data avail- able indicate that clades with representatives in more than one province have a greater probability of surviving mass extinction events than clades, however speciose, that are confined to a single province (e.g. Br- etsky, 1973; Boucot, 1975; Jablonski, in press). Many authors have suggested that the tropical marine biota appear to suffer dis- proportionately during mass extinctions (e.g. Kauffman, 1979, Cavelier et al., 1981; Boucot, 1983; Jablonski, 1984, in press b). If the tropics are subject to major perturbations during mass extinctions, this would further contribute to the lack of correlation between species richness and clade survival at those times. Widespread clades whose peak diver- sities are in the tropics would be more se- verely affected than clades in which diversity is evenly distributed with latitude or is highest outside the tropics. Clades restricted entirely to the tropics, which are often among the JABLONSKI 8 FLESSA 56 | ‘ри uewAeD риелэ pue ‘epeuald “Weuuns ‘ешеиез эциецу :seuney “eIquIn¡o)) узциа pue ‘эцолу UBIPeueo ay) ‘(еузем) MOG JUIO”d :seuney podousef Jays |езиэицио ul suonnquisip Ácuanbaay Ajimej-Saioeds ‘ее ‘914 podonseß Jays |езиэициоэ и! suonnqusip Aduanbayj Анцшие}-5э0э4$ “|Z ‘914 al RARE NISTORSS ATV NI $30346 n Oz ol à 5 5 < VISNN109 HSILIH8 D NvWAvO амуно | FO! VavN349 ot : 019 о < ol oe 02 ol ae Se 02 ol ol WvNIHns VWVNVd DOILNVILV los Gi DILOHV NVIOYNVO MOËUUVE “ld 9904041595 vaOdOHLSVO E 2 “eoujy UNOS ‘Aeg asje4 pue |IZP1g :seuney ‘S99U919J9H ¡euney4 ay) pue '594$4е1$ aruno рододзеб jjeys ¡ejuauyuos ui зиоцпащер Aduanbayy Апше}-5э09э95 ‘Oz `914 40) € 91921 995 “(euwoye9) Aeg Аэлэзиои\ pue (ueder) {eg iweBes :seune} podousef пацз ¡ejuaunuoo ui зиоцпащер Аэцэпбэд Ашие}-5э10э4$ ‘61 ‘914 АТИ М $903«$ ot 02 ATINV14 NI $90345 02 ol 2 Fo TE = V91H14v HINOS'AV8 AS 1Vv4 ADN3N0 344 AVS A3H31INOW ANNO 343 © o ¡NANA a = a 02 05 0€ 02 ol 11ZVH8 Nvdvr'AVS INVOYS ol ol vaodOH1Sv9 vaodOH1sv9 TAXONOMIC STRUCTURE AND EXTINCTIONS 57 TABLE 3. Species-family frequency data for continental shelf and oceanic island gastropod faunas. See Fig. 2 for locations, and Figs. 19-31 for frequency distributions. Number of Location species |. Continental shelf faunas Arctic Canada 103 British Columbia 209 Brazil 528 Britain 245 Grand Cayman 15. 204 False Bay (South Africa) 217 Ghana 99 Grenada 119 Jamaica 459 Mediterranean 214 Monterey Bay (California) 371 Aupourian Province (New Zealand) 843 Cookian Province (New Zealand) 353 Forsterian Province (New Zealand) 418 Oman 265 Atlantic Panama 389 Point Barrow (Alaska) 69 Sagami Bay (Japan) 696 Surinam 115 Texas 181 West Florida 166 |. Oceanic island faunas Aldabra 363 Ascension 50 Bermuda 192 Chatham 212 Cocos-Keeling 380 Easter 96 Faeroes 79 Fanning 380 Funafuti 245 Hawaii 693 Iceland 152 Jan Mayen 62 Kerguelen 84 Marquesas 125 Niue 199 Pagan 166 most speciose of all, would paradoxically be most extinction-prone. In simulating this lati- tudinal pattern of extinction in the modern bi- ota, Jablonski (in press a) found that elimina- tion of families restricted to the tropical is- lands in our survey resulted in an extinction of the magnitude of the Permo-Triassic, with clades ranked appropriately (that 1$, reef- building corals most severely reduced, echi- noderm classes intermediate, and bivalves and gastopods least severely affected). Number of | % monospecific Species-family families families ratio 22 45 4.7 50 34 4.1 76 24 7.0 52 37 4.7 51 33 4.0 40 177 5.4 39 dd 2:5 43 40 2.8 55 16 8.4 62 27 3.4 50 14 7.4 112 25 Teo 89 36 4.0 71 35 5:9 53 30 5.0 65 29 6.0 19 47 3.6 89 20 7.8 39 38 3.0 53 36 3.4 52 31 3.2 58 28 6.3 30 70 1.6 51 25 3.8 58 36 3.7 56 27 6.8 37 43 2.6 31 45 215 59 25 6.4 64 31 4.6 61 16 AES 39 36 3.9 20 55 3.1 28 46 3.0 27 33 4.6 33 36 6.0 37 24 4.5 SUMMARY AND CONCLUSIONS The taxonomic structure of the modern ma- rine biota indicates that many families would be represented on oceanic islands following marine regression. Even if withdrawal of the seas from continental shelves completely eradicated the shelf benthos, 87% of the fam- ilies would be represented on oceanic islands, which would not be subject to areal reduction during regression. However, for some taxa JABLONSKI 8 FLESSA 58 (puejeaz MON) S9UIAOI4 UEIHOOD pue (риуееэ7 MAN) ээшлол4 иеиа}$104 :seunej podonseß jjays ¡ejuaunuos ul зиоцпащер Aduanbayy Апше}-зэюэ4< ‘92 ‘914 AW М $30395 or 05 (ZN) МУ! 315303 vaodOH1Sv9 ‘воешег pue ‘EPIIO|4 JSOM ‘SEX9] :seuney podoujseb ‚jays ¡ejueunuos ul suonnquisip Aouenbay Анше}-заюэа< ‘pz ‘914 A ЧИУЗ М $39345 os 05 02 ol pa] Lp Y 3 volver 5 SZ OL 02 ol 02 01 вл ЧЕ =>: u vaIHOIJ 153M Svx31 Lou OL vVAOdOULSVO ‘(puejesZ мэм) ээшлол4 ueunodny au} pue uleyug :seune, podo.]se6 Jays ¡ejuaunuoo и! зиоцпащер АэцепЬэ4 Анше}-заюэ4< “gz ‘914 АТИУЗ М $39345 05 OC ol ool os Ov (ZN) NVIHNOdNVY o АЭМЭПОЗНЗ МУ 15а о a OL VIOJOYLSVO ог '8loym e se UPSUPJISJIPAN ay) pue ‘иешо 'eueyo :звипе} podousef jays ¡ejuaunuos и! зиоцпае!р Aduanbayy Allwe)-selwadg “ez ‘914 си, og ИУУМ dS oz OL Ce “al АВГ U ^ z NV3NVyY83LI03N 5 < OL 02 ol ol О D NVWO VNVHD + OL OL VaodOH1Sv9 o 19583 pue 'Bulj9ay-s0907) ‘eiqeply :seuney :'s90J9e y ay) pue 'epnuwiag 'uoisuaosy 'usÁey URL 'puejao] :seune) podonseß pues! o1uesoo ui suonnqusip Aouanbay Апше}-зэюэ4$ “OE ‘914 роаодзеб pueisı oiuesoo ul suonnquisip Aouanbay Ájiuej-seldads ‘62 ‘914 ANY М $3035 ol ol où VAN dS o! 05 02 OL Su — En À EA no! à À 2 5 т S30H3v4 5 331593 ONI133H-S0909 2 5 vanııy38 Io, NOISNZOSV A ol ol (ap) - PA O oz = 05 03 ol S 02 ; ol = 02 ot = x vHgva1v LU N3AVN МУГ ОМУ 1391 o OL 2 VIAOJOYLSVID VAOdOYLSVO A ‘à Ww = 'sesanbueyy pue 'ynyeuny 'Buluuey :seune} ‘иемен pue ‘uebed 'anin :seune} = podoseb pues: эшеээо и! suonnquisip Аэцэпбэд Ápuej-salads ‘82 ‘914 рододзеб pues! oiuesoo ui suonnquisip Aouanbay Айше]}-зэюэ4а$ ‘/е ‘914 =) es + ATINV1 NI S3193dS So ide on EN NI a 5 (ep) 05 02 o! | ый O ew as A PA ll la DOS 3 = : : O 02 on ILNJYVNNA | 5 ИУМУН Z 5 > 2 2 O Loi pl x < SVSINOYVN | de E a > DEA QE NV9OVd E 028 el + “y E] ry 0€ 02 ol “tls | m Sl E ONINNV y 3NIN ‘|; Lo 01 VIOAOYLSVD vaodO4H1Sv9 60 JABLONSKI 8 FLESSA GASTROPODA 10 CHATHAM 10 > KERGUELEN O - Lu cr Te 10 20 SP. IN FAM. FIG. 31. Species-family frequency distributions in oceanic island gastropod faunas: Chatham and Kerguelen. the security offered by these island refuges may be offset by differences in species-family ratios between mainlands and islands. For bivalves and echinoids—though not for gas- tropods—families on islands tend to have fewer species and thus may be less extinction- resistant than families on mainlands. Just as Simberloff (1970, 1978) found that species- genus and species-family ratios on islands commonly could be explained in terms of local species richness without recourse to compet- itive interactions, we found that species-family ratios within clades could be explained in terms of local species richness without recourse to different macroevolutionary processes among the three groups we examined. It is not clear whether the differences in species-family ratios between mainland and island faunas are of sufficient magnitude to undermine the refugium effect. Taxonomic structure, e.g. species-family ratios, appears to be of secondary importance in determining survival during mass extinctions because the biological attributes that give rise to speciose clades tend to make those species extinction- prone. We suggest that taxonomic structure is more important during intervals of back- ground extinction, though here, too, other bi- ological properties obviously are also signifi- cant. In contrast, biogeographic patterns ap- pear to have greater significance for survivor- ship during mass extinctions. Like Raup (1979b, 1982) we have used the taxonomic structure of the present-day biota to infer ex- tinction patterns and probabilities in the fossil record. These results must be approached with caution, however, because species-fam- ily ratios have apparently changed since the Cambrian, and Recent taxa are on average the most speciose of the Phanerozoic (Valen- tine, 1969, 1970, 1973; Raup, 1975; Se- pkoski ef al., 1982). Relative representation of clades on oceanic islands vs. mainlands thus may well have changed through time. Differences we observed in species-family ratios between island and mainland echinoids and bivalves could be a relatively recent phenomenon, or conversely, the is- land-mainland similarities between gastropod faunas may be a consequence of the explo- sive post-Paleozoic speciation in the tropical gastropods. The trend of increasing species- family ratios through the Phanerozoic has ad- ditional implications for both mass and back- ground extinctions. For example, as Valen- tine (1974) points out, a given magnitude of familial extinction could result from an in- creasingly smaller quantity of species-level extinctions going back through the Phan- erozoic. Finally, the observed decline in background extinction rates of Phanerozoic families (Raup & Sepkoski, 1981) need not reflect improvement in biological adaptation, but rather may simply be a function of in- creasing species-family ratios through time (Flessa & Jablonski, 1985). м FIG. 32. Number of families plotted against number of species in continental shelf and oceanic island faunas of echinoids and bivalves. See Table 1 for data. TS FIG. 33. Number of families plotted against number of species in continental shelf and oceanic island faunas of bivalves and gastropods. See Table 1 for data. FAMILIES FAMILIES TAXONOMIC STRUCTURE AND EXTINCTIONS 61 60 O e) о oO 40 o р : о = Ф оо о О O 8 o° om O 20 = бо e в Г] ECHINOIDEA = BIVALVIA o 100 200 300 400 SPECIES A 100 A A A A £ a à A 60 A Pr adm SO 4 49, 0 a APS à GASTROPODA 4 20 %& A BIVALVIA o 200 400 600 800 SPECIES 62 JABLONSKI & FLESSA A. BIVALVE GENERA EXTINCT SURVIVE n=63 Nn=53 40 40 40% < 42% m 20 210 WwW oO 0 9 1-2 3+ VEZ SET spp SPP spp spp B. GASTROPOD GENERA EXTINCT SURVIVE n=95 n=6 1 50 50 50% < m 33% = 25 2:5 o Ww © 0 0 12. 3 152 3+ spp spp spp spp FIG. 34. Survivorship of species-rich and species- poor molluscan genera across the Cretaceous- Tertiary boundary. A, Gulf and Atlantic Coastal Plain bivalves; B, Gulf and Atlantic Coastal Plain gastropods. Note that species-rich genera are at least as well-represented among the victims of the extinction as among the survivors, suggesting that taxonomic structure does not play a major role in determining clade survival during mass extinction events. After Jablonski (in press b). ACKNOWLEDGMENTS We thank Susan M. Kidwell, Earl D. McCoy, and Geerat J. Vermeij for valuable comments and criticism. John D. Taylor and Geerat J. Vermeij provided unpublished faunal lists for the molluscan faunas of Aldabra and the Marianas, respectively, and advice and assistance with published sources came from David J. Bottjer (and the Hancock Library, University of Southern California), Philip W. Signor, and J. Wyatt Durham. Supported in part by NSF Grant EAR 81-21212. REFERENCES CITED ANDERSON, S., 1974, Patterns of faunal evolution. Quarterly Review of Biology, 49: 311-332. BOUCOT, A. J., 1975, Evolution and extinction rate controls. Amsterdam: Elsevier, 427 p. BOUCOT, A. J., 1983, Does evolution take place in an ecological vacuum? |. Journal of Paleontol- ogy, 57: 130: BRETSKY, P. W., 1973, Evolutionary patterns in the Paleozoic Bivalvia: documentation and some theoretical considerations. Geological Society of America, Bulletin, 84: 2079-2096. CAMPBELL, C. A. & VALENTINE, J. W., 1977, Comparability of modern and ancient faunal provinces. Paleobiology, 3: 49-57. CAVELIER, C., CHATEAUNEUF, J. J., POMEROL, C., RABUSSIER, D., RENARD, M. & VERGN- AUD-GRAZZINI, C., 1981, The geological events at the Eocene/Oligocene boundary. Palaeogeo- graphy, Palaeoclimatology, Palaeoecology, 36: 223-248. CONNOR, E. F. & McCOY, E. D., 1979, The sta- tistics and biology of the species-area re- lationship. American Naturalist, 113: 791- 833. FISCHER, A. G. & ARTHUR, M. A., 1977, Secular variations in the pelagic realm. Society of Eco- nomic Paleontologists and Mineralogists, Special Publication, 25: 19-50. FLESSA, K. W., 1975, Area, continental drift and mammalian diversity. Paleobiology, 1: 189-194. FLESSA, К. W. & JABLONSKI, D., 1985, Declining Phanerozoic background extinction rates: effect of taxonomic structure? Nature, 313: 216-218. FLESSA, К. W. 8 SEPKOSKI, J. J., Jr., 1978, On the relationship between Phanerozoic diversity and changes in habitable area. Paleobiology, 4: 359-366. GILBERT, F. S., 1980, The equilibrium theory of island biogeography. Journal of Biogeography, 7: 209-235. GOULD, 5. J., 1976, Palaeontology plus ecology as palaeobiology. In: MAY, R. M., ed., Theoretical ecology: principles and applications. Saunders, Philadelphia, p. 218-236. HANSEN, T. A., 1980, Influence of larval dispersal and geographic distribution on species longevity in neogastropods. Paleobiology, 6: 193-207. HAQ, B. U., 1973, Transgression, climatic change and the diversity of calcareous nannoplankton, Marine Geology, 15: M25-M30. JABLONSKI, D., 1980, Apparent versus real biotic effects of transgressions and regressions. Pal- eobiology, 6: 397—407. JABLONSKI, D., 1982, Evolutionary rates and modes in Late Cretaceous gastropods: role of larval ecology. Proceedings of the 3rd North American Paleontological Convention, 1: 257- 262. JABLONSKI, D., in press a, Marine regressions and mass extinctions: a test using the modern biota. TAXONOMIC STRUCTURE AND EXTINCTIONS 63 In VALENTINE, J. W., ed., Phanerozoic diversity patterns: profiles т macroevolution. Princeton: Princeton, University Press. JABLONSKI, D., in press b, Causes and conse- quences of mass extinctions: a comparative ap- proach. In: ELLIOTT, D. K., ed., Dynamics of extinction. New York: Wiley. JABLONSKI, D. & LUTZ, В. A., 1983, Larval ecology of marine benthic invertebrates: Pal- eobiological implications. Biological Reviews, 58: 21-89. JABLONSKI, D. & VALENTINE, J. W., 1981, On- shore- offshore gradients in Recent eastern Pa- cific shelf faunas and their paleobiogeographic significance. In: SCUDDER, С. ©. E. 4 REVEAL, J. L., ed., Evolution today, Proceedings of the Second International Congress of Systematic and Evolutionary Biology. Pittsburgh: Carnegie- Mellon University, р. 441—453. JACKSON, J. В. C., 1974, Biogeographic conse- quences of eurytopy and stenotopy among ma- rine bivalves and their evolutionary significance. _American Naturalist, 108: 541-560. JARVINEN, O., 1982, Species-to-genus ratios in biogeography: a historical note. Journal of Bio- geography 9: 363-370. KAUFFMAN, E. G., 1979, The ecology and bioge- ography of the Cretaceous-Tertiary extinction event. In: CHRISTENSEN, W. K. & BIRKELUND, T., ed., Cretaceous-Tertiary Boundary Events Symposium. University of Copenhagen, 2: 29-37. MCLAREN, О. J., 1983, Bolides and biostratigraphy. Geological Society of America Bulletin, 94: 313-324. NEWELL, N. D., 1967, Revolutions in the history of life. Geological Society of America Special Paper 89: 63-91. ВАЧР, D. M., 1975, Taxonomic diversity estimation using rarefaction. Paleobiology, 1: 333-342. RAUP, D. M., 1979a, Biases in the fossil record of species and genera. Bulletin of the Carnegie Museum of Natural History, 13: 85-91. RAUP, О. M., 1979b, Size of the Permo-Triassic bottleneck and its evolutionary implications. Sci- ence, 206: 217-218. RAUP, D. M., 1982, Biogeographic extinction: a feasibility test. Geological Society of America Special Paper, 190: 277-281. RAUP, D. M. & SEPKOSKI, J. J., Jr., 1982, Mass extinctions in the marine fossil record. Science, 215: 1501-1503. SCHOPF, Т. J. M., 1974, Permo-Triassic extinction: relation to seafloor spreading. Journal of Geol- ogy, 82: 129-143. SCHOPF, T. J. M., 1979, The role of biogeo- graphic provinces in regulating marine faunal diversity through geologic time. In: GRAY, J. & BOUCOT, A. J., ed., Historical Biogeography, plate tectonics, and the changing environment. Corvallis: Oregon State University Press, p. 449-457. SEPKOSKI, J. J., Jr., ВАМВАСН, В. K., RAUP, D. M. & VALENTINE, J. W., 1981, Phanerozoic marine diversity and the fossil record. Nature, 293: 435—437. SIMBERLOFF, D. S., 1970, Taxonomic diversity of island biotas. Evolution, 24: 23-47. SIMBERLOFF, D. S., 1974, Permo-Triassic extinc- tions: effects of area on biotic equilibrium. Jour- nal of Geology, 82: 267-274. SIMBERLOFF, D. S., 1976, Species turnover and equilibrium island biogeography. Science, 194: 572-578. SIMBERLOFF, D. S., 1978, Use of rarefaction and related methods in ecology. In: DICKSON, K. L., CAIRNS, J. Jr. & LIVINGSTON, R. J., ed., Bio- logical data in water pollution assessment: quan- titative and statistical analyses. American Soci- ety of Testing and Materials, Special Technical Publication, 652: 150-165. SIMBERLOFF, D. S., 1981, Community effects of introduced species. In: NITECKI, M. N., ed., Bi- otic crises in ecologic and evolutionary time. New York: Plenum, p. 53-81. SOKAL, В. В. & ROHLF, Е. J., 1969, Biometry. San Francisco: Freeman, 776 p. STANLEY, S. M., 1979, Macroevolution. San Fran- cisco: Freeman, 332 p. TAYLOR, D. W. & SOHL, N. F., 1962, An outline of gastropod classification. Malacologia, 1: 7-32. VALENTINE, J. W., 1969, Patterns of taxonomic and ecological structure of the shelf benthos during Phanerozoic time. Palaeontology, 12: 684-709. VALENTINE, J. W., 1970, How many invertebrate species? A new approximation. Journal of Pale- ontology, 44: 410-415. VALENTINE, J. W., 1973, Evolutionary paleoeco- logy of the marine biosphere. Englewood Cliffs, М. J.: Prentice-Hall, 511 р. VALENTINE, J. W., 1974, Temporal bias in extinc- tions among taxonomic categories. Journal of Paleontology, 48: 549-552. VALENTINE, J. W., РОМ, Т. С. & PEART, D. 1978, А provincial model of Phanerozoic marine diversity. Paleobiology, 4: 55-66. VERMEI)J, С. J., in press, An examination of envi- ronments and geographical areas containing modern survivors of extinction-ravaged taxa. In: ELLIOTT, D. K., ed., Dynamics of extinction. New York: Wiley. WARD, P. D. & SIGNOR, P. W., Ш, 1983, Evolu- tionary tempo in Jurassic and Cretaceous am- monites. Paleobiology, 9: 183-198. WILLIAMS, C. B., 1964, Patterns in the balance of nature.... London: Academic Press, vii + 324 р. WILLIS, J. С., 1922, Age and area. Cambridge: Cambridge University Press, 259 p. WISE, К. P. & SCHOPF, T. J. M., 1981, Was marine faunal diversity in the Pleistocene af- fected by changes in sea level? Paleobiology, 7: 394-399. 64 JABLONSKI & FLESSA FAUNAL REFERENCES ABBOTT, R. T., 1958, The marine mollusks of Grand Cayman Island, British West Indies. Acad- emy of Natural Sciences of Philadelphia Mono- graph 11: 138 p. ABBOTT, R. T., 1974, American seashells. Ed. 2. New York: Van Nostrand Reinhold, 663 p. ALTENA, C. O. van REGTEREN, 1971, The marine Mollusca of Suriname (Dutch Guiana) Holocene and Recent. Part Il. Bivalvia and Scaphopoda. Zoologische Verhandelingen (Leiden), 119: 100 p. ALTENA, C. O. van REGTEREN, 1975, The marine Mollusca of Suriname (Dutch Guiana) Holocene and Recent. Part lll. Gastropoda and Ce- phalopoda. Zoologische Verhandelingen (Leiden), 139: 104 p. ANDREWS, J., 1977, Shells and shores of Texas. Austin: University of Texas Press, 365 p. BASSINDALE, R., 1961, On the marine fauna of Ghana. Proceedings of the Zoological Society of London, 137: 481-510. BERNARD, Е. R., 1970, A distributional checklist of the marine molluscs of British Columbia, based on faunistic surveys since 1950. Syesis, 3: 75-94. BOSCH, D. & BOSCH, E., 1982, Seashells of Oman. London: Longman, 206 p. CASO, M. E., 1961, Estado actual de los cono- cimientos acerca de los equinodermos de Mexico. Ph.D. dissertation, Departamento de Biologia, Universidad Nacional Autoñoma, Mex- ico, 388 p. CERNOHORSKY, W. O., 1970, The littoral marine molluscs of Niue Island. Records of the Auckland Institute and Museum, 7: 175-186. CERNOHORSKY, W. O., 1971, Marine shells of the Pacific. Rev. ed. Sydney: Pacific Publications, 248 p. CERNOHORSKY, W. O., 1972, Marine shells of the Pacific. Sydney: Pacific Publications, 2: 411 p. CERNOHORSKY, W. O., 1978, Tropical Pacific marine shells. Sydney: Pacific Publications, 352 p. CERNOHORSKY, W. O., 1978, Report on the mol- luscan fauna of the Lau Group, Fiji Islands. Bul- letin of the Royal Society of New Zealand, 17: 39-52. CHERBONNIER, G. 8 GUILLE, A., 1975, Echinod- ermes récoltés aux îles Kerguelen. Bulletin du Museum National d'Histoire Naturelle (Paris), Zoologie, 210: 603-630. CLARK, A. M. & COURTMAN-STOCK, J., 1976, The echinoderms of southern Africa. London: British Museum (Natural History), 238 p. CLARK, A. M. 8 ROWE, F. W. E., 1971, Monograph of shallow-water Indo-West Pacific echinoderms. London: British Museum (Natural History), 238 p. CLARK, А. М. & SPENCER-DAVIES, P., 1965 (1966), Echinoderms of the Maldive Islands. An- nals and Magazine of Natural History, ser. 13, 8: 597-612. CLARK, H. L., 1942, The echinoderm fauna of Bermuda. Bulletin of the Museum of Compara- tive Zoology, Harvard University, 89: 367- 391. COLES, J., 1981, A check-list of the molluscs of the Cook Islands (Mollusca). Poirieria, 11(1): 7-11; 11(2): 17-21. D'ANGELO, G. & GARGIULLO, S., 1978, Guida alle Conchiglie Mediterranee. Milano: Fabbri Editori, 223 p. DAUTZENBERG, Ph. 8 BOUGE, J. L., 1933, Les mollusques testacés marins des etablissements français de l'Océanie. Journal de Conchyliologie, 77: 41-108, 145-326, 351-469. DAY, J. H., FIELD, J. G. & PENRITH, М. J., 1970, The benthic fauna and fishes of False Bay, South Africa. Transactions of the Royal Society of South Africa, 39: 1-108. DELL, R. K., 1960, Chatham Island marine Mol- lusca based upon the collections of the Chatham Islands Expedition, 1954. Bulletin of the New Zealand Department of Scientific and Industrial Research, 139: 141-157. DELL, R. K., 1964, Marine Mollusca from Macquarie and Heard Islands. Records of the Dominion Museum, Wellington, New Zealand, 4: 267-301. FELL, F. J., 1974, The echinoids of Easter Island (Rapa Nui). Pacific Science, 28: 147-158. FELL, H. B., 1960, Archibenthal and littoral echinoderms of the Chatham Islands. Bulletin of the New Zealand Department of Scientific and Industrial Research, 139: 55—75. FRETTER, V. & GRAHAM, A., 1962, British prosobranch molluscs. London: Ray Society, 755 р. HEDLEY, C., 1899, The Mollusca of Funafuti. Mem- oirs of the Australian Museum, 3: 397-510, 547-565. НЕММЕМ, J. 4 HEMMEN, C., 1979, Beitrag zur Kenntnis der Meeres Mollusken-Fauna der Kar- ibischen See. Grenada. Jahrbuch der Nassauis- chen vereins fur Naturkunde, Wiesbaden, 104: 137-172. HUMFREY, M., 1975, Sea shells of the West In- dies. New York: Taplinger, 351 p. KAY, E. A., 1971, The littoral marine molluscs of Fanning Island. Pacific Science, 25: 260-281. KAY, E. A., 1979, Hawaiian marine shells. Reef and shore fauna of Hawaii, section 4: Mollusca. Bern- ice Pauahi Bishop Museum Special Publication 64(4): 653 p. KAY, E. A. & SWITZER, M. F., 1974, Molluscan distribution patterns in Fanning Island lagoon and a comparison of the mollusks of the lagoon and the seaward reefs. Pacific Science, 28: 275-295. KENSLEY, B., 1973, Sea-shells of southern Africa- gastropods. Capetown: Maskew Miller. KUNDU, H. L., 1965, On the marine fauna of the TAXONOMIC STRUCTURE AND EXTINCTIONS 65 Gulf of Kutch. Part Ш. Pelecypoda. Journal of the Bombay Natural History Society, 62: 209-236. KURODA, T., НАВЕ, T. & OYAMA, K., 1971, The sea shells of Sagami Bay. Tokyo: Maruzen. xix + 739 + 489 + 51 p., 121 pl., 1 map. LEMCHE, H., 1929, Gastropoda Opisthobranchia. The zoology of the Faeroes, 3(1, 53): 20 p. LEMCHE, H., 1939, Gastropoda Opisthobranchia. The zoology of Iceland, 4(61): 54 p. LIEBERKIND, I., 1929, Echinodermata. The zool- ogy of the Faeroes, part 53: 20 p. LUBINSKY, I., 1980, Marine bivalve molluscs of the Canadian central and eastern Arctic: faunal com- position and zoogeography. Canadian Bulletin of Fisheries and Aquatic Sciences, 207: 111 p. MacGINITIE, N., 1959, Marine Mollusca of Point Barrow, Alaska. United States National Museum Proceedings, 109: 59-208. MACPHERSON, E., 1971, The marine molluscs of Canada: prosobranch gastropods, chitons and scaphopods. National Museum of Canada Pub- lications in Biological Oceanography, 3: 149 p. MADSEN, Е. J., 1949, Marine Bivalvia. The zoology of Iceland, 4(63): 116 p. MAES, V. O., 1967, The littoral marine mollusks of Cocos- Keeling Islands (Indian Ocean). Proceed- ings of the Academy of Natural Sciences of Phil- adelphia, 119: 93-217. MARCUS, E., 1977, An annotated checklist of west- ern Atlantic warm water opisthobranchs. Journal of Molluscan Studies Supplement, 4: 22 p. McKNIGHT, D. G., 1968, Some echinoderms from the Kermadec Islands. New Zealand Journal of Marine and Freshwater Research, 2: 505-526. McKNIGHT, D. G., 1969, An outline of the New Zealand shelf fauna. Benthos survey, station list, and distribution of the Echinoidea. New Zealand Department of Scientific and Industrial Research Bulletin 195(= New Zealand Oceanographic In- stitute Memoir, 47): 91 p. McKNIGHT, D. G., 1972, Echinoderms collected by the Cook Islands eclipse expedition, 1965. New Zealand Oceanographic Institute Research Bul- letin 1(3): 37—45. MICHEL, C., 1974, Notes on marine biology studies made in Mauritius. Bulletin of the Mauritius Insti- tute, 7(2): 287 p. MORTENSEN, Th., 1927, Handbook of the echinoderms of the British Isles. London: Oxford University Press, 471 p. MORTENSEN, Th., 1950, Echinoidea. Report of the British, Australian and New Zealand Antarctic Research Expedition, ser. B, 4: 287-310. NISIYAMA, S., 1968, The echinoid fauna from Ja- pan and adjacent regions. Part Il. Palaeontologi- cal Society of Japan Special Paper 13: 491 p. OLSSON, A. A. & McGINTY, Т. L., 1958, Recent marine molluscs from the Caribbean coast of Panama with the description of some new gen- era and species. Bulletins of American Paleon- tology 39(177): 1-58. PAWSON, D. L., 1978, The echinoderm fauna of Ascension Island, South Atlantic Ocean. Smithsonian Contributions in Marine Science 2: 31 p. PERRY, L. М. 8 SCHWENGEL, J. S., 1955, Marine shells of the western coast of Florida. Ithaca: Paleontological Research Institution, 262 p. PETERSEN, G. H., 1968, Marine Lamellibranchi- ata. The zoology of the Faeroes, 3(1, 55): 80 p. POWELL, A. W. B., 1957, Mollusca of Kerguelen and Macquarie Islands. Report of the British, Australian and New Zealand Antarctic Research Expedition, ser. B, 6: 107-149. POWELL, A. W. B., 1976, Shells of New Zealand. Ed. 5 revised. Christchurch: Whitcoulls, 154 p. POWELL, A. W. B., 1979, New Zealand Mollusca. Auckland: Collins, 500 p. PRICE, A. R. G., 1982, Echinoderms of Saudi Arabia. Comparison between echinoderm faunas of Arabian Gulf, SE Arabia, Red Sea and Gulf of Aqaba and Suez. Fauna of Saudi Arabia, 4: 3-21. REHDER, H. A., 1968, The marine molluscan fauna of the Marquesas Islands. Bulletin of the Ameri- can Malacological Union, 1968: 29-32. REHDER, H. A., 1980, The marine mollusks of Easter Island (Isla de Pascua) and Sala y Gomez. Smithsonian Contributions to Zoology, 289: 167 p. REHDER, H. A. £ WILSON, B. R., 1975, New species of marine mollusks from Pitcairn Island and the Marquesas. Smithsonian Contributions to Zoology, 203: 16 p. RIOS, E. C., 1970, Coastal Brazilian seashells. Fundacáo Cidade do Rio Grande, Museu Oce- anografico de Rio Grande, 225 p. RODRIGUEZ, J., 1980, Echinoderms (except Ho- lothuroidea) of the southern Mediterranean coast of Spain. In: JANGOUX, M., ed., Echinoderms: present and past. Rotterdam: A. A. Balkema, p. 127-131. ROMAN, J., 1980, Une monographie des echinides de la Mer Rouge. Principaux résultats. In: JAN- GOUX, M., ed., Echinoderms: present and past. Rotterdam: A. A. Balkema, p. 133-136. ROSEWATER, J., 1975, An annotated list of the marine mollusks of Ascension Island, South At- lantic Ocean. Smithsonian Contributions to Zool- ogy, 189: 41 p. SERAFY, D. K., 1979, Echinoids (Echinodermata: Echinoidea). Memoirs of the Hourglass Cruises, 5 (3): 120 p. SHIGEI, M., 1970, Echinoids of the Bonin Islands, Journal of the Faculty of Sciences of Tokyo Uni- versity, 12: 1-22. SKJAEVELAND, 5. H., 1973, Echinoderms of Jan Mayen Island. Astarte, 6: 69-74. SLOAN, N. A., CLARK, A. M. & Taylor, J. D., 1979, The echinoderms of Aldabra and their habitats. Bulletin of the British Museum (Natural History), Zoology 37: 81-128. SMITH, А. G. & GORDON, M., Jr., 1948, The ma- rine mollusks and brachiopods of Monterey Bay, 66 JABLONSKI 8 FLESSA California, and vicinity. Proceedings of the Cali- fornia Academy of Sciences, Ser: 4, 26: 147-241. SNELI, J. A. & STEINNES, A., 1975, Marine Mo- Низса of Jan Mayen Island. Astarte, 8: 7-16. SPARCK, В. 8 THORSON, G., 1931, Marine Gas- tropoda Prosobranchiata. The Zoology of the Faeroes, no. 52: 56 p. TEBBLE, N., 1976, British bivalve seashells. Ed. 2. Edinburgh: Her Majesty's Stationery Office, 212 p. THOMPSON, T. E., 1976, Biology of opisthobranch molluscs. London: Ray Society, 1: 207 p. THORSON, G., 1941, Marine Gastropoda Prosobranchia. The zoology of Iceland, 4(60): 150 p. TORTONESE, E., 1965, Fauna d'Italia. Vol. VI. Echinodermata. Bologna: Edizioni Calderini, 422 p. VERMEIJ, С. J., KAY, Е. A. & ELDREDGE, L. G., 1983, Molluscs of the northern Mariana Islands, with special reference to the selectivity of oce- anic dispersal barriers. Micronesica, 19: 27- 55. WALLER, T. R., 1973, The habits and habitats of some Bermudian marine mollusks. Nautilus, 87: 31-52. MALACOLOGIA, 1986, 27(1): 67-81 EXTINCTION IN HAWAIIAN ACHATINELLINE SNAILS Michael G. Hadfield Pacific Biomedical Research Center, University of Hawaii, Honolulu, HI 96822, U.S.A. ABSTRACT Growth data point strongly to late maturity in the achatinellines, and information from both dissection and examination of age-frequency histograms supports the idea that annual fecundity is near one. The limited lifespan extrapolated suggests a reproductive life of perhaps six years; lifetime fecundity could thus be as low as 6, and seems certain not to exceed 24. This assemblage of traits predisposes populations of achatinellines to extinction in the face of predation, particularly of the selective, shell-collecting sort. Key words: Achatinella; extinction; Hawaiian tree snails; life histories; mark-recapture studies; demography. INTRODUCTION That populations, varieties and species of the Hawaiian arboreal snails (subfamily Acha- tinellinae) are going extinct has been known for more than 120 years (Frick, 1856). In the 1880's the alarm was sounded by at least two important students of Hawaiian shells, D. D. Baldwin (1887) and J. T. Gulick (quoted in A. Gulick, 1932). Baldwin (1887: 56), speaking of the habitats of Achatinella, noted, “It is also generally supposed that these shells are be- coming extinct by the ravages of cattle through our forests”; “Some of these hills have been denuded of woods, not only by cattle, but by the woodman's ax, and certain species are becoming rare”; and “The agen- cies now threatening the wholesale destruc- tion of these little gems of the forest are the rats and mice, which have become very abun- dant in mountain forests, particularly where there are no cattle.” In later years, awareness of the loss of the unique tree snail fauna increased. Henshaw, writing on the Partulina of the island of Hawaii (in Pilsbry & Cooke, 1912-1914) noted the great declines of numbers of individuals in some areas and their disappearance in oth- ers. Bryan (1935) put it mildly, “Land snails were formerly more plentiful in certain parts of the islands than they are today.” And Cooke (1941: 20), writing of Achatinella apexfulva, (67) said, “It must have been a rather abundant lowland species then but, to my knowledge, no typical live specimens have been found in the last 40 or 50 years.” More recently Kondo (Ms., 1970) presented quantitative estimates of extinction in Hawai- ian terrestrial gastropods (50% extinct) and succinctly outlined major causes for their dis- appearance. Hart (1975, 1978) repeated these figures in popular articles which had an important impact on public recognition of the problem. The culmination of this growing record of the wholesale disappearance of species of a unique subfamily of snails, endemic to the Hawaiian Islands, has been preparation of evidence for and, finally, a federal declaration of “endangered status” for the remaining spe- cies of the genus Achatinella (Federal Regis- ter, 1981). The documents relative to the dec- laration cite 22 species (20, after Welch's revisions, 1942, 1954, 1958) of Achatinella as extinct, with the remaining 19 (or 17) species endangered. Species in the achatinelline gen- era Partulina and Newcombia are probably in no better condition. The partulinas of Hawaii island are probably gone, and no good, recent census exists for the islands of Molokai, Маш and Lanai. On all of these islands, habitat destruction alone has had a major impact. Thus, extinction in the Hawaiian tree snails is an historical problem and continues today (Hadfield & Mountain, 1981). The goals of the 68 HADFIELD present account are not to further belabor this point, but rather to attempt first to clarify the magnitude and rate of the disappearance by asking, “How many snails were there?”, and second to examine the causes of mass mor- tality in these snails. Both of these goals must be met by examining the literature, little of which was written to answer these particular questions. Subsequently, | present results of recent field studies on the achatinellines and analyze them with a goal of ascertaining what features of the life history of these gastropods have made them so vulnerable to the agents of extinction. The systematics of the endemic Hawaiian subfamily Achatinellinae (Stylommatophora; Achatinellidae) are thoroughly reviewed in Pilsbry & Cooke (1912-1914) and elsewhere (Cooke & Kondo, 1960; Welch, 1938, 1942, 1954, 1958). METHODS Determination of primitive densities of achatinelline snails has depended almost en- tirely on extrapolating from collections. Only a single published attempt to estimate a popu- lation size for a Hawaiian tree snail species has been found; it is that of Henshaw for Partulina confusa (in Pilsbry and Cooke, 1912-1914). For other data | have used col- lecting notes of Baldwin, Gulick, Spalding and many others. As will be seen, comparing old collecting notes with recent experiences al- lows a very rough estimation of primitive snail densities. Causes of extinction of Hawaiian tree snails have been cited since the time of the original notes on shell collecting in Hawaii and added to by nearly every student of these snails. These are summarized from the literature and reinterpreted in the light of the collecting notes and modern field studies. Field methods used to study population sizes, growth rates, ages at maturity and other population characteristics were provided by Hadfield & Mountain (1981). These are mark- recapture studies which necessitate no sacri- fice of living snails. Added to these are some general results of field surveys where 2 to 4 experienced workers hiked transects and counted all visible snails for noted time dura- tions. To a lesser extent, information has been gleaned from assemblages of dead shells which accumulate on the ground beneath trees occupied by the achatinellines. Data utilized in an earlier publication (Had- field 8 Mountain, 1981) have been reanalyzed for some additional conclusions on life span and mortality. Study sites include two areas in the Wai- anae Mountains of Oahu: Kanehoa Trail (see Hadfield 8 Mountain, 1981) and another ridge line in the northern third of the Waianae Mountains. Elevations lie between 600 and 800 m. Here the forests are a mixture of native trees and shrubs such as Dodonaea spp., Metrosideros polymorpha, Dubautia sp., Scaevola gaudichaudiana, Osmanthus san- dwichensis, and Alyxia olivaeformis, and introduced forms including guava (Psi- dium guajava), silk oak (Grevillea robusta), Stachytarpheta jamaicensis, lantana (Lan- tana camara), etc. Snails seen inthe Waianae sites belong to the species Achatinella mus- telina (see Welch, 1938). A second major study site has been estab- lished on the Nature Conservancy's Kamakou Preserve on Molokai. At this site, elevation about 1200 т, Partulina redfieldii occupies scattered, small ohia trees (Metrosideros po- lymorpha) which are semi-isolated from one another by broad expanses of grassland. This formerly dense mesic forest was opened up by the grazing activities of feral livestock be- ginning in the mid- to late 1800's. Occasion- ally, snails are seen here on the shrubs Coprosma sp. and Wikstroemia sp. Data from the field studies have been ana- lyzed as described in Hadfield 8 Mountain (1981) for growth rate, age at first reproduc- tion and population density. Because growth rates for the Molokai snails were not the same for all sizes, as was seen in Achatinella mus- telina, the von Bertalanffy growth equation (Fabens, 1965) was calculated for Partulina redfieldii. lt provided a good fit for the data and allowed the size-frequency pattern to be transformed to reflect age frequencies. Longevity of snails from both areas, Wai- anae and Molokai, was extrapolated from age-frequency distributions by determining the average sizes of year classes in the age frequency histograms and dividing the number of lipped (= maximum sized) shells by the average size/year class to estimate the ex- pected number of year classes. The resulting number was added to the average age at lip development. Fecundity was estimated from EXTINCTION OF ACHATINELLINES 69 the literature and extrapolated from the histo- grams. A Leslie Matrix (Searle, 1966) was used to assess the effects of fecundity and annual survivorship on the malthusian parameter, r,,,, the intrinsic rate of increase of a population with a stable age distribution. Fecundity for each age (12 age classes) was entered as the top row of the matrix and annual survivorship as a diagonal across the next 11 rows (in a 12 x 12 matrix). A computer program evaluated the matrix to the 50th iteration and the domi- nant eigenvalue (A) was determined as the arithmetic mean of the 12 values in the 50th iteration. The value of r,, is obtained as the natural log of A (see Mertz, 1970, for a thor- ough discussion of these terms). The malthusian parameter usually proves to be the more useful estimator for comparing spe- cies with different life-history patterns. Briefly, г. varies around zero, with positive values reflecting growing populations and negative values shrinking ones. RESULTS AND DISCUSSION 1. How many snails were there? Henshaw (in Pilsbry 4 Cooke, 1912-1914) estimated densities of Partulina confusa on the Waimea Plains (island of Hawaii) from his observations there in 1903. He referred to this population as an “isolated colony” living in about 150 pua trees (Osmanthus sandwich- ensis) growing within an area of about a “half a mile square.” The trees were scattered, mostly separated, and 15 to 20 feet high. Henshaw stated (p. 97), “A rough estimate of the number of adult shells inhabiting this area when first visited is more than 75,000 shells, and it was possible to ride under the trees and from their trunks, leaves, and branches to pick shells literally by the handfuls. Cavities in the trunks and branches were usually packed with shells, mostly immature, from 50 to 75 being often found together.” Henshaw's estimate predicts 500 adult snails per tree! If the age frequency distribu- tion of P. confusa was similar to those we have seen for Р. redfieldii on Molokai in a somewhat similar setting, adult snails make up about 20-25% of the population. The total number of snails would thus have approached 300,000, equivalent to 2,000 per tree! Such figures seem impossible, but we shall never know. Henshaw reported that he gathered 1,100 adult shells from this spot in 6 hours, and his colleague, Mr. William Horner, did the same. Any place allowing the picking of 3 snails per minute was assuredly densely т- habited. Kondo (personal communication, July 1983) told this author that he visited the Waimea plain in 1946 and saw only dead shells of P. confusa covering the ground. Other estimates of snail densities can only be roughly extrapolated from collecting records. Such values must be very coarse because it is usually impossible to determine an important series of parameters. Was the collector collecting only large, mature shells or all sizes seen? If only large shells were taken, was the collector choosing the more attractive shells only, or all seen? Was the collector alone, or did he have companions taking an equal number of shells? How hard did the collector search? Was he interested in cleaning out the spot or only obtaining what was easy to find? Did the collector climb the trees or in other ways obtain shells higher than his reach (and was he on horseback when reaching?)? Nonetheless, the following quotations give some idea of the enormous densities of achatinellines encountered by the early shell collectors. J. T. Gulick, whose recorded collection was about 44,500 shells in about 3 years collecting (1851-53; see Clench, 1959), usually spoke of hundreds or thousands in each collecting lot (A. Gulick, 1932). He reported on riding into the northern Koolaus on July 28, 1853 with 10 others to a little valley and returning by four in the afternoon with over fourteen hun- dred shells (p. 120). On September 3, 1853, “Brother Thomas and | spent several hours in Punaluu Valley, where we procured over a hundred specimens of what | consider a new shell” (p. 124). On September 15, Gulick and two native men went into Waiawa Valley for most of a day and returned with 200 to 400 shells apiece (p. 125). Gulick also encour- aged the Hawaiian residents of rural villages to collect for him, and he would periodically ride about the islands buying the shells, indi- vidual lots often “amounting to several thou- sands,” and frequently filling his saddle bags with his purchases. In 1852, Samuel and Henry Alexander trav- eled from Oahu to Lanai with a Mr. Bailey and collected “several thousand” on that small 70 HADFIELD island (M. C. Alexander, 1934). A note in the Weekly Star, a script journal of the Punahou School in Honolulu, tells of a picnic on March 7, 1853 when, “... into the woods back of sugarloaf ... After dinner, we all dispersed about in the woods, for the purpose of procur- ing shells . . . the number procured that day was over four thousand.” The same journal for March 16, 1853: “Last Saturday was ren- dered famous by a pretty general expedition to the mountains in search of shells. Over (2,000) two-thousand specimens were brought back alive by the hardy adventurers comprising about fourteen species of the genus Achatinella.” Baldwin (1887, 1900) gives a number of incredible collecting tallies. In the valleys on the eastern side of the Waianae Range of Oahu he could secure “... the shells very rapidly, often getting a hundred or more from a single tree,” with the aid of a hook on a stick seven or eight feet long for pulling the branches of the trees downward. Four days collecting in the Waianae Range yielded over 2,000 shells. In one nine-day trip, Baldwin collected more than 3,500 shells; his daily yield varied from 152 to 864. Baldwin (1887) also reported a trip around Molokai when he collected 5,000 shells. And in Kohala, on the island of Hawaii, Baldwin reported (1887: 62), “during a recent visit to the locality, in a few minutes | collected several hundred speci- mens, picking them from trees and low bushes as rapidly as one would gather huck- leberries from a prolific field.” Cooke (1903), reporting on Achatinella mul- tizonata (= A. bellula) in Nuuanu Valley, col- lected 3,000 shells Нот a zone 100-400 yards wide and extending for about one mile on the northwestern side of the valley at ele- vations between 1,000 and 1,400 feet. These snails came from a number of semi-isolated populations within the described zone, sug- gesting fairly high local densities. From Emerson (Ms., undated, post-1900), we learn of a single ahakea (Bobea sp.) tree yielding 3 score of a rare shell and a banana tree which yielded a score of Achatinella bul- imoides. He also wrote of A. viridans so abun- dant on the midridge of Palolo Valley that they “...hung in clusters on the hoe vines” (probably Dioscorea sp.). 2. Modern notes Using mark-recapture techniques, we esti- mated a resident population of Achatinella mustelina in four small trees in a 25 т? quad- rate in the Waianae range to be about 210 animals (Hadfield 8 Mountain, 1981). The av- erage number of snails, all sizes included, observed on any visit to this site was 44, or about 21% of the population. | will use this estimate of observed-to-actual snails to com- pare current densities with the older collecting records for the Waianaes cited above. In the winter of 1982-1983, while conduct- ing surveys under contract to the Army Corps of Engineers in the region of Makua Valley, three other experienced snail observers and | made the observations listed in Table 1. The figures provide some very rough estimates of density to compare with Baldwin's (1887) col- lecting data from the 1880's. Baldwin had a companion when he gathered 2,000 plus shells in four days; together they took about 500 shells per day. № we assume they took only larger shells (Baldwin, 1887, cautioned against collecting young or immature shells) and that they were not minutely searching out the snails, then they were collecting snails, steadily, at a rate of about 31 per person hour. Our rate, seen in Table 1, assuming 50% “large shells”, varied from 1—7 shells per per- son hour of searching, an average of less than 4. If there is any validity to these calculations, then the best residual populations of Achati- nella mustelina amount to about 12% of those seen 100 years ago in the Waianae range. Another comparison may be made between our current study Site for Partulina redfieldii on TABLE 1. Transect sightings: Achatinella mustelina in the northern Waianae Range. No. large Date No. searchers Time No. snails Est. of total shells/person/hr* Nov. 13 1982 4 2 hr 29 138 2 Dec. 18 1982 4 3 hr 161 770 7 Jan. 15 1983 2 2% hr 48 229 5 Jan. 29 1983 4 1V2 hr 15 71 1 *Estimated to be 50% of all snails seen, divided by number of observers and number of hours. EXTINCTION OF ACHATINELLINES 71 Molokai and the populations of P. confusa described by Henshaw as existing on the Waimea Plain of Hawaii in 1903. These lo- cales share a common elevation, both are relatively wet, and both consist of isolated, low trees in a larger meadow at the time of obser- vation. As noted above, Henshaw's estimates ran to 500 adult snails per tree. In our study area, we have counted the numbers of snails in seven small ohia trees. These densely fo- liated trees are about 1.5-4.5 m high and have crown diameters ranging from 2-3 m. We have actually counted 84 snails in these bushes, searching each extensively on each of three visits to the study site. Counts of visible snails, expressed per tree, in the Kamakou meadow (Molokai) varied from 4 to 17; to some extent these numbers reflect the sizes of the trees. Data from three trees which have been sampled on three sep- arate occasions are tabulated together with Weighted-Mean estimates of populations (see Begon, 1979, pp. 13-16) in Table 2. The estimates vary between 8 and 29 snails per tree; they are in close agreement with esti- mates achieved by a second method, the Peterson Estimate (Begon, 1979), which uti- lizes data from only the first recapture. To determine an estimator of the number of snails present from those seen, the weighted- mean determined population size for each tree was divided into the average number of snails seen per tree on each of three visits; the mean of these three values, 55.7%, was the estimator so derived. Thus for another tree, counted only once, with a visible snail population of 17, the total population 1$ esti- mated to be 31. This is the upper limit to estimated population densities per tree in the Molokai study area. The estimates of numbers of Partulina con- fusa in the trees of the Waimea Plain given by Henshaw are one to two orders of magnitude greater than those arrived at for P. redfieldii, above, in a similar habitat. Of course, we cannot guess if the populations on Molokai were ever as dense as those of Waimea. From the foregoing, it is clear that the den- sities of achatinelline snails were once far greater than any seen today. While it would be possible to determine when whole habitats were eliminated for farming and logging, it is nearly impossible to learn just when the den- sities of still extant populations declined to their current levels. With such information, TABLE 2. Partulina redfieldii. Capture-recapture data and population estimates for three trees in the Kamakou study area, Molokai. Date Aug. 5 Feb. 13 June 16 Tree A: No. captured (n;) 15 12 11 No. previously marked (mi) — 6 7 No. released (г) iS 12 — Total no. marked (M) — 15 21 Min, — 180 231 Tree B: п, 4 8 6 m; — 4 5 r; 4 8 — M; — 4 8 Min; — 32 48 Tree C: п, 3 9 9 т; — 2 5 r; 3 9 — M; — 3 10 Min, = 27 90 Population size, N, = Mn, (Emi) + 1 N-Tree A = 29; av. no. snails seen = 44.3% of estimated total N-Tree B = 8; av. no. snails seen = 75% of estimated total N-Tree C = 15; av. no. snails seen = 47.9% of estimated total Average percent seen = 55.7% long-term rates of extinction could be calcu- lated. 3. Causes of mortality in achatinelline snails For more than one hundred years, different authors have speculated on the causes of extinction in the Hawaiian tree snails (see Introduction above). The list of causes pre- sented here includes factors cited by many authors in the past and also raises questions about causes of mortality perhaps not previ- ously considered. Habitat destruction is, of course, the single most important cause of population disap- pearance for any species. For the achatinel- lines this process may well have begun with the pre-historic landings of the Hawaiians. 72 HADFIELD Their clearing of lowland vegetation was prob- ably responsible for unrecorded extinctions of low altitude populations or even species (Kirch, 1982). By the mid-1800's the achatin- ellines were snails of the mountain slopes and ridges, but Pilsbry wrote about 1912 (in Pilsbry & Cooke, 1912-1914: xxvi), “Once forests with Achatinellidae and Endodontidae shaded the plains far seaward from the lovely peak of Kaneohe, where now dead shells may be picked up in plowed fields, or gathered out of “pockets” in the rocks. It has been the same in the northwest. Forest-snails are found in the sand-dunes of Kahuku, now far from where living tree-snails exist.” The shells of Achatinella caesia littoralis, an extinct subspecies, were found in the troughs of sand dunes near Kahuku, less than 100 feet from the sea. Pilsbry further noted, “Sixty years ago the Achatinellas were found in abun- dance at half the elevations now inhabited by them.” (ibid.: xlix). Pilsbry attributed these losses to both disappearance of trees and to declining humidities brought on by the grazing out of underbrush by cattle and goats. Much habitat went to cane and pineapple fields, some to reservoirs, and still more to forest cutting for logs. Reforestation, which might have provided a new home for relict achatinelline populations rarely included na- tive species, and trees such as eucalyptus, ironwood and Norfolk Island pine, planted widely on Oahu, have never provided suitable habitats for the achatinellines. Emerson (Mss., undated) noted that a prime habitat for a distinct variant of Achatinella rosea was deeply submerged by a sugar plantation res- ervoir, probably in the late 1890's. Accounts of total habitat destruction can be repeated for each of the islands. Farming occupies most of the suitable land on all the islands, and a drive from the coast to the upper elevations of Molokai takes one through extensive stands of the above-named exotic trees, supplemented with loblolly pines, Monterey cypress, and California coastal red- woods. This is not to say that achatinellines have not been seen on non-native vegetation. They have been reported on guava, lantana and the aboriginally introduced ti, banana and kukui. К is probable that lantana, reported in late 1800's publications to be locally infested with Achatinellas, proved unsuitable as a per- manent substratum; at no location where lan- tana 1$ currently abundant have | seen acha- tinellines on it except rarely. A slightly more subtle form of habitat de- struction was brought about by feral livestock, particularly goats and cattle. These browsers pushed forest destruction to much higher el- evations than the farmers did, eating the un- dergrowth and lower tree branches, destroy- ing seedlings, and probably transporting ex- otic plants ever higher into the native forests. Gulick, Frick, Baldwin, Emerson, Pilsbry and others spoke of the threat presented by cattle to native forests and their snail faunas. Although habitat destruction accounts for a large percentage of the loss of achatinelline snails, there were still many square miles of forest left mostly intact, particularly at upper elevations. While it may be that these upper elevations never harbored the greatest of achatinelline densities, it is certain that they were rich country for the snails. These snails too have mostly disappeared, and their dis- appearance must be related to more selec- tive agents of destruction. Perkins (1899) found only rats to be certain predators of tree snails, and he affirmed that native birds did not feed on them. The Polynesian rat had ob- viously been present a long time before Eu- ropeans were marveling at the great densi- ties of tree snails in the Hawaiian Islands. However, the introduction of the European rats appears to have presented a major threat to Hawaii's endemic snails, a fact ear- lier recorded by Gulick (A. Gulick, 1932: 411), Baldwin (1887), and Perkins (1899). To this day, one can find ample evidence of their rav- ages wherever tree snails still persist. To my knowledge, no other predator crushes the shells in the manner of predating rats; in a recent survey of dead shells collected from the ground in a small area in the northern Waianae Range (see Fig. 1), 10% of the ground shells appeared to have been eaten by rats. № the conclusions of Atkinson (1977) are correct, the predator here is Rattus rat- tus, which became widespread on Oahu only in the 1870's. There may have been other native snail predators. Frick (1856) wrote of a “centipede worm” that ate any tree snail unfortunate enough to land upon the ground; apparently the centipedes did not climb trees in search of prey. A terrestrial flatworm, Geoplana sept- emlineata (perhaps not native), is known to attack ground-living snails such as Achatina fulica (Mead, 1979), but | know of no evidence that the worms ascend trees to attack arbo- real snails. Geoplana would, however, appear EXTINCTION OF ACHATINELLINES 73 20 LL rs = (ep) E oF, 10 R NOBIS 507 as > 95 1513.5 155 Ager O7 а, 05 lipped shells 17.5 195 21.5 6 7. 8 Length (mm) and Year Class FIG. 1. Achatinella mustelina: size-, and extrapolated age-frequency distribution of dead shells collected at one time from a5 x 5 m quadrate in the northern Waianae Range. “Lipped shells” are those with terminal growth; they include, at least in part, shells of older ages than others in the size/age class. Age-frequencies were extrapolated from data provided in Hadfield & Mountain (1981).n = 138. to be a real danger to achatinellines blown or knocked to the ground. We have observed living Geoplana septemlineata in the forest litter at a place where Hurricane Iwa swept many living Achatinella mustelina to the ground in November 1982. We did not find evidence of unusually heavy mortality in these locales, however, and it appeared that the snails had rapidly ascended all nearby vege- tation (personal observations made in De- cember, 1982). Introduced ants have been cited as a seri- ous threat to Hawaii's native snails (Solem, 1976), but | know of no evidence to implicate them in predation on achatinellines. While the Hawaiians were known to make occasional leis from the shells of tree snails, there can be little doubt that a predator of major importance, arriving in abundance in the 1800’s and immediately commencing to depredate the achatinelline populations, was the European stock of Homo sapiens. How many shells were gathered as curiosities by itinerant voyagers we shall never know, but the recorded or semi-recorded collections are amazing in themselves. While many of the “missionary sons” were simply shell collect- ing, others such as J. T. Gulick seem to have had the motivations of a naturalist. These collectors scoured the islands from 1850 to 1900, and extensive collections were made even in the first decades of the twentieth century. How many did they collect? It is nearly impossible to determine, but the col- lecting notes cited above lend visions of tens of thousands of gastropods being frequently reduced to “shells.” J. T. Gulick himself wrote to G. T. Romanes in 1888, “The collection was made during the years 1851-1853, when | visited all the districts of the Island of Oahu 74 HADFIELD in person and accompanied by troops of na- tive assistants ransacked each valley.” And after speaking of forest and snail destruction, Gulick concluded, “The collection is therefore not only unique but will always remain unique...” (A. Gulick, 1932: 411). Emerson (Mss., undated) was of the certain opinion that the shell hunters were seriously reducing the densities of achatinelline popu- lations. He tells of finding a dense population of an Achatinella species in Waiomau (Palolo) Valley and reporting its location to a friend. This was a grievous mistake, he noted, be- cause the friend took a number of boys to the area, and, “The result was that the choice spot teaming with shells was ruthlessly plun- dered.” Kondo (1980) attributed extinctions to over- collecting; he reports for one locale, Kawaihalona Gulch, once rich in Achatinella bulimoides, ‘‘Collectors swarmed to Kawaihalona; rosea was soon gone forever.” Massive collections were made by Gulick (Emerson, Ms., undated, recalls visiting Gul- ick's house as a child and seeing there, “... boxes and kegs of evidently unsorted shells.”), Baldwin, Frick, Pease, Spalding, Meinecke (who collected more than 116,000; Kondo, Ms., 1980), and many others. How many populations might they have accounted for? Bryan (1935) wrote of “several collec- tions made of upwards of 10,000 specimens each.” К must be remembered that the achatinel- line snails are very slow-moving. They usually remain in the same tree for life, and named varieties were known from one small gulch, grove, or even a few trees. Predators as se- lective as human shell collectors could thus drastically affect such isolated populations with ease. Probably the most serious modern predator of native Hawaiian snails is the introduced, North American gastropod Euglandina rosea. While it was introduced to Hawaii between 1955 and 1956 to prey on the African snail, Achatina fulica, it has become abundant in many areas far beyond the range of A. fulica (summarized by Mead, 1979). Correlated with the spread of E. rosea into the remaining habitats of Achatinella has been disappear- ance of the endemics (Mead, 1979; Van der Schalie, 1969). This problem 1$ not unique to Hawaii. Clarke, Murray 8 Johnson (1984) have documented the disappearance of Par- tula on Moorea as populations of the intro- duced predator grow and expand there. Euglandina rosea was introduced to Moorea in 1977 where it spread rapidly to occupy nearly a third of the island by 1982. In that time, native populations of one Partula spe- cies had become extinct. The authors predict that by 1986 E. rosea will have spread throughout Moorea and eaten the remaining 7—9 partulid species endemic to that island. The population of Achatinella mustelina that we had observed in the Waianae Range of Oahu for three years was found to have disappeared coincidently with the invasion of the area by large numbers of E. rosea (Had- field 8 Mountain, 1981). Only dead shells of A. vulpina occurred in an area abundant in E. rosea in Halawa Valley, Oahu, in 1981 (un- published, personal observations). Itis, unfor- tunately, too late to learn if the dense popu- lations of achatinellines originally present in Hawaii could have withstood the ravages of this snail; the marginal ones remaining obvi- ously cannot. Still, there seems to be another, yet uniden- tified, source of mortality in the Achatinellinae and possibly other terrestrial snails of Hawaii. In the Waianae Range of Oahu, wherever living populations of Achatinella mustelina persist, the ground is littered with dead shells. In May 1983, we scoured the ground in a 5 by 5 m quadrate containing about 50 small trees, bushes and shrubs with living A. mustelina on them. The result of this survey was the dis- covery of 161 dead shells, spanning the entire size range (Fig. 1) found elsewhere for this species (Hadfield 8 Mountain, 1981). Only 23 of the shells found were broken and thus, at least potentially, the prey of rats; the remain- der were intact. Many of the large, lipped shells contained a small shell, that of the embryo they were brooding when they died. Euglandina rosea has not been seen in this area. The number of living A. mustelina seen on the trees in the quadrate was 44. № our sight- ing efficiency was similar to that experienced elsewhere in these mountains (Hadfield & Mountain, 1981), this would represent a living resident population of 150 to 200 snails. The dead shells had obviously perished over some lengthy but unknown time period as their condition varied from fresh and glossy to thin, eroded and colorless. There is no clear indication of what kills so many snails, but it is significant that the mor- tality is not size-dependent. In fact, the size- EXTINCTION OF ACHATINELLINES 75 40 w o Mean % of Samples о lipped shells hypothetical Mean «Glass FIG. 2. Achatinella mustelina: age-frequency distribution. Size-frequency data from Hadfield & Mountain (1981), adjusted to reflect age-frequency. Solid black bars represent immature shells for which age estimates are considered good. Stippled bars with solid outlines represent lipped shells (i.e., those with maximum growth); the year classes at which they are located represent only the age at which maximum size was attained. “Hypothetical” bars are projected year classes. See text for explanation. frequency distribution of the dead shells was similar to that of the live shells studied about 8 miles south of this location (see Hadfield € Mountain, 1981, and Fig. 2). Because it is not a characteristic of the study site for Partulina redfieldii on Molokai that dead shells are abundant on the ground, the observation may be taken as an indication that there is an additional mortality factor on Oahu. Still, some dead, but intact shells are present in the Kamakou meadow. In an effort to determine if the abundance of dead shells underlying the Achatinella-inhabited trees of Oahu was a recent occurrence, | talked with Dr. Yoshio Kondo, Malacologist Emeritus of the B. P. Bishop Museum, whose observations of Ha- waiian terrestrial snails began in the 1930's. He informed me that this had, in his experi- ence, always been so. What kills so many snails so non- selectively? It could be Geoplana septem- lineata, but this would require that rela- tively large numbers of snails are dislodged from the trees to the ground where the pred- ators lurk. Furthermore, Geoplana has been found in the Molokai area. Predatory ants could be present, but | have not seen them in these areas. Introduced terrestrial snails, such as Achatina fulica and Euglandina ro- sea harbor dense populations of the nema- tode, Angiostrongylus cantonensis (see Mead, 1979), and this worm may have т- vaded the mountain trees. Yet A. cantonen- sis is generally dispersed through the food chain or infects snails through contact with rat feces, and it is difficult to envision its entry into the achatinellines (Alicata, 1965). Addi- tionally, it has not been implicated in snail mortality. Another possible explanation for the death of so many Achatinellas on Oahu is a micro- bial disease. Mead (1979: 85ff) discusses 76 HADFIELD 2.0 Growth (mm/day x 0 10 Mid-length (mm) -2 y = (3.24 - 1.33X)10 20 30 FIG. 3. Partulina redfieldii: growth rate vs. shell length. Shell length 1$ plotted at the estimated mid-point of the mark-recapture interval for which the respective growth was recorded. such a disease in Hawaiian populations of Achatina fulica and its spread to other exotic snails. These others (Bradybaena similaris and Subulina octona) have invaded the high- est reaches of Oahu's mountains and could have transported such a disease. The current status of the remaining Achatinellas 15 too fragile to allow intense micropathological in- vestigation of them. 4. What makes the achatinellines so vulnera- ble? In order to understand the high rates of extinction in the achatinelline snails, as well as to determine means for circumventing fur- ther extinctions, life history characteristics of two species are being investigated through non-sacrificial field studies. Using mark- recapture techniques, we have estimated population densities, size-frequency distribu- tions and size-related growth rates. From these data, age-frequency distributions, age at first reproduction, and lifespan have been extrapolated. Fecundity estimates are still very rough (Hadfield & Mountain, 1981) and may remain so. From published observations on numbers of embryos in dissected snails and our age-frequency distribution histo- grams, a very low fecundity—probably only one offspring per year—seems likely. Analy- sis of life-history traits for each of the species is detailed below. Achatinella mustelina. Data for all life-his- tory parameters utilized, except lifespan, have been published (Hadfield & Mountain, 1981). Lifespan has been estimated from size (age)- frequency distribution as follows. In Fig. 2, catch data for A. mustelina are plotted by year class. Animals that have not reached full shell growth, and thus whose size predicts age fairly accurately, are represented by black EXTINCTION OF ACHATINELLINES TT bars. Because the snails form a thickened lip around the shell aperture and cease growing at or about the time of sexual maturity, those represented by solidly outlined, stippled bars are placed in the year class in which they stopped growing. They represent about 35% of all snails seen. The average size of the immature year classes is 7.3% of the total. If we assume very low mortality, the 35% of mature shells, divided by 7.3% per year class, would include nearly 5-year groups. Adding this on to the typical age at maturity, 6 years, we estimate a total number of year classes present at 11; the additional year classes are represented in Figure 2 by the “hypothetical” projection. This is admittedly a very coarse approximation, but is the only one currently available to us. For A. mustelina we have estimated the following life table parameters: age at first reproduction, 6 years; longevity, 11 years; fecundity 1 (or up to 4) per year. Partulina redfieldii: Growth rate was deter- mined by regressing all mark-recapture ob- tained growth information against estimated lengths of shells at the mid-points of mark- recapture intervals (Fig. 3). The regression equation thus provides a measure of the change in growth rate with size. The regres- TABLE 3. Partulina redfieldii. The distribution of sizes (= shell length) with age. Age Year class Size range 0—1 yr 0 0-12.53 mm 1-2 yr 1 12.54-17.09 mm 2-3 yr 2 17.10-19.91 mm 34 yr 3 19.92-21.64 mm 4—5 уг 4 21.65-22.71 тт 5—6 уг 5 22.72-23.37 тт > буг 6+ > 23.38 тт sion data and size at birth (5.12 mm) were placed into the von Bertalanffy growth equa- tion and solved for length at the end of each year to provide a year class-to-size-range ta- ble (Table 3). The size-frequency distribution for Partulina redfieldii in the Kamakou Pre- serve study area is shown in Figure 4, and the extrapolated age-frequency distribution in Figure 5. Data on the frequency of animals in imma- ture year classes, where a steady mortality 15 indicated, were extrapolated as described above for A. mustelina to provide an estimate of the number of year classes in the mature snail population (dotted outlines in Fig. 5). The summary of estimated life-history pa- rameters for P. redfieldii thus includes: age at first reproduction, 4 years; longevity, 11 years; fecundity, 1 (estimated up to 4) per year. The two achatinelline species studied both show the characteristics of late maturity and low fecundity, traits seemingly associated with evolution in a predator-free environment. Both traits may be related to the particularly large birth size in these snails. To understand the degree to which these life-history traits predispose such species to the threat of ex- tinction with various sorts of new predators, А. mustelina and P. redfieldii were examined with reference to r,,, the malthusian parame- ter or intrinsic rate of population increase. The basic question is, to what level can annual survivorship fall—given the typical life-history characters described above—and the r,, value for the population remain positive? For A. mustelina, with sexual maturity in the sixth year and annual fecundity of one until death after 11 years, survivorship must be greater than 0.825 annually for the population to remain stable or increasing (i.e., r,, 0). With fecundity estimated at 2 and 3 offspring per year, survivorship must still exceed 0.736 and 0.668, respectively, if the populations are not to decline. Similar figures for P. redfieldii, with an ear- lier estimated age at reproductive maturity and thus, with a similar lifespan, a higher lifetime fecundity, are: for г. to be greater than zero, with fecundities of 1 and 4, annual survivorships must exceed .734, and .573, respectively. These requisite survivorship numbers are high. For small populations of A. mustelina such as those isolated in single trees, a mor- tality factor (predator, collector, disease, etc.) that removed 20%, that is just 6 of 30 snails present, for example, on a regular basis would cause the population to decline and presum- ably go extinct. Other manipulations of these data may approach a realistic situation. For instance, picture some relatively isolated and unique variety of Achatinella on Oahu about 1850 or 1860. Add a series of collectors who go back year after year and collect just half of all the mature shells present. For the popula- tion growth characteristic to remain positive, annual survivorship of all immature year classes must be at least 90%. If a size- 78 HADFIELD 20 Mean % of Samples о SAS AZ SLT 5 535 12:5: A9 Se 22S 2352255 Class Mark (mm) FIG. 4. Partulina redfieldii: size-frequency distribution. All animals measured on each of three occasions were sorted into 2 mm size classes, and, for each occasion, the percentage of snails represented by each class was determined. Presented here are the means of the percentage determinations for the three visits. Total snails seen was 100. independent mortality factor is secondarily added, population extinction is almost guar- anteed. Did such repeated and selective collecting occur? The evidence is strong that it did. In addition to the references cited above relative to collections amassed by 19th century col- lectors, | sought indications of localized col- lecting pressures in the records of the B. P. Bishop Museum. Among many collections de- posited there, | found one to be particularly useful due to the precision of its site records (that of I. Spalding). In the catalogue for this collection, | found that the collector returned each year from 1908 to 1916 to Wailupe Val- ley on southeastern Oahu. In each of these years, he collected from 30 to 413 specimens of Achatinella fulgens in that locale. Examina- tion of the collection shows that these were mostly large, mature shells. The total taken in this period was 930. Additionally, at each visit to the valley the collector was accompanied by a shell-collecting friend whose take 1$ not recorded. Similar records are frequent and seem significant in terms of the models pre- sented above. The calculations made here are based only on varying mortalities. They tend to show that P. redfieldii, due to its earlier reproductive onset is more robust than A. mustelina. This may be the reason for the survival of the small, isolated, one-tree populations of P. red- fieldii which must have been in this state for a long time. There are, however, the indica- EXTINCTION OF ACHATINELLINES 79 m о Mean % of Samples O I; lipped shells hypothetical Year Class FIG. 5. Partulina redfieldii: age-frequency distribution of snails at the Molokai study site. The data in Fig. 4 were replotted in age classes predicted by the von Bertalanffy growth equation. Lipped shells are shown at the age at which terminal growth was reached. “Hypothetical” classes were roughly extrapolated from the shape of the distribution of the immature classes. The projected decline in the sizes of succeeding classes is shown by the horizontal dashed lines in year classes 3-5. There are no data to support a contention of zero mortality in adult snails, but the growth is so drawn to indicate maximum potential lifespan. tions of lower unexplained mortality (i.e., whole, dead shells beneath the trees) in the P. redfieldii of Kamakou Preserve, Molokai. None of these speculations include aspects of population density. To what low frequency can an achatinelline population fall before the probability of mating encounter falls too low for successful reproduction to occur? This is obviously an important question. It is Known that the destruction wrought on Hawaii's endemic achatinellines has also oc- curred in many other groups of native gastro- pods. Kondo (Ms., 1980) estimated at least 50% extinction for all of Hawaii's terrestrial snails. Most of Kauai’s unique Carelia disap- peared many generations ago; of about 200 species of endodontids formerly inhabiting the main islands, fewer than a dozen individ- uals have been seen in the last forty years. However, there are some groups of native and many exotic gastropods thriving in Ha- waii. Minute tornatellinids were found cover- ing all manner of weeds and crawling on the needles of Casuarina in the desolated hills above Kahuku, Oahu (personal observations, 1981). Succinea species are still abundant in 80 HADFIELD many Hawaiian mountain locales, and a few hardy Auriculella species persist in areas from which the achatinellines have disap- peared. We know far too little about the life histories of these snails to single out the par- ticular traits that favor survivorship. Certainly, any factor that increased fecundity would en- hance the probability of population persist- ance. This may be the case with oviparous Auriculellas and other forms (Cooke 4 Kondo, 1960). ACKNOWLEDGEMENTS For encouraging and providing access and assistance to our field studies in the Kamakou Preserve of Molokai, | express my deepest thanks to Mr. Alan Holt and Mr. Ed Misaki of the Nature Conservancy. To the Nature Con- servancy of Hawaii, | am especially indebted for being allowed to carry out the studies on Molokai. Data collected in the Northern Waianae Range in 1982-1983 were obtained during survey work for the U.S. Army Corps of En- gineers. | thank them for permission to utilize these data here. Barbara Shank, Peter Galloway, Marilyn Dunlap, Douglas Stoner, Stephen Miller, Carl Christensen, Cynthia Hunter and Carol Hop- per have all provided valuable field assis- tance. Writing this paper would have been impos- sible without the support of: E. Alison Kay, who dug deeply into her historical files to aid me; Stephen Ralston, who advised me on data analysis and wrote the computer pro- gram for multiplying the Leslie Matrix; Yoshio Kondo, who gave me his time and with it his incredible store of knowledge of Hawaiian land snails; Carl Christensen, who provided access to the collections and records of the B. P. Bishop Museum and aimed me toward a number of references | might otherwise have missed; Stephen Miller, who helped di- gest the data and who ran the computer pro- gram over and over; and Susan Grau, who prepared the graphs for this paper. Com- ments and suggestions on the manuscript made by E. A. Kay, C. Christensen and A. Holt improved and clarified it. To all of the above, my utmost thanks. REFERENCES CITED ALEXANDER, M. C., 1934, William Patterson Al- exander in Kentucky, the Marquesas, and Ha- иай. Honolulu, privately printed, xvii + 516 р. ALICATA, J. E., 1965, Biology and distribution of the rat lungworm, Angiostrongylus cantonensis, and its relation to eosinophilic meningitis and other neurological disorders of man and animals. In: DAWES, В., ed., Advances in Parasitology, Academic Press, New York, 3: 223-248. ATKINSON, I. A. E., 1977, A reassessment of fac- tors, particularly Rattus rattus L., that influenced the decline of endemic forest birds in the Hawai- ian Islands. Pacific Science, 31: 109-133. BALDWIN, D. D., 1887, The land shells of the Hawaiian Islands. Hawaiian Annual, 1887: 55—63. BALDWIN, О. D., 1900, Land shell collecting оп Oahu. Hawaii's Young People, 4(8): 239-243. ВЕСОМ, M., 1979, Investigating animal abundance: capture-recapture for biologists. University Park Press, Baltimore, 97 p. BRYAN, E. H., Jr., 1935, Hawaiian land shells. Hawaiian Nature Notes (Honolulu Star Bulletin, Ltd., Honolulu): 208-213. CLARKE, B., MURRAY, J. & JOHNSON, М. S., 1984, The extinction of endemic species by a program of biological control. Pacific Science, 38: 97-104. CLENCH, W. J., 1959, John T. Gulick's Hawaiian land shells. Nautilus, 72: 95-98. COOKE, С. M,, Jr., 1903. Distribution and variation of Achatinella multizonata from Nuuanu Valley. Occasional Papers, Bernice Pauahi Bishop Mu- seum, 11: 65—76. СООКЕ, С. M., Jr., 1941, Hawaiian land shells. Paradise of the Pacific, 53(12): 20-25. COOKE, С. M., Jr. 8 KONDO, Y., 1960, Revision of Tornatellinidae and Achatinellidae (Gastropoda, Pulmonata). Bernice Pauahi Bishop Museum Bulletin 221: 303 p. EMERSON, O. P. (Ms., undated, post-1900), The gay Achatinellidae and their habitats. (Bernice Pauahi Bishop Museum Library), 24 p. FABENS, A. J., 1965, Properties and fitting of the von Bertalanffy growth curve. Growth, 28: 265-289. FEDERAL REGISTER, 1981, Endangered and threatened wildlife and plants; listing the Hawaiian (Oahu) tree snails of the genus Achatinella, as endangered species. F.R., 46(8): 3178-3181. FRICK, D., 1856, Notes on Hawaiian terrestrial conchology. Sandwich Islands Monthly Maga- zine, 1: 137-140. GULICK, A., 1932, John T. Gulick evolutionist and missionary. University of Chicago Press, Chi- cago, 556 p. HADFIELD, М. G. & MOUNTAIN, В. S., 1981, A field study of a vanishing species, Achatinella mustelina (Gastropoda, Pulmonata), in the Wai- EXTINCTION OF ACHATINELLINES 81 anae Mountains of Oahu. Pacific Science, 34: 345-358. HART, A., 1975, Living jewels imperiled. Defend- ers, 50(6): 482-486. HART, A., 1978, The onslaught against Hawaii's tree snails. Natural History, 87(10): 46-57. KIRCH, P. V., 1982, The impact of prehistoric Po- lynesians on the Hawaiian ecosystem. Pacific Science, 36: 1-14. KONDO, Y. (Ms., 1970), Colloquium on endan- gered species of Hawaii; extinct land molluscan species. А report prepared for a meeting at the U.S. National Museum, 8 p. KONDO, Y. (Ms., 1980), Endangered land snails, Pacific. A report prepared for the International Union for the Conservation of Nature and Natural Resources, 15 p. MEAD, A., 1979, Economic malacology with partic- ular reference to Achatina fulica. FRETTER, V. 8 PEAKE, J., eds., Pulmonates, 2B. Academic Press, London, 150 p. MERTZ, D. B., 1970, Notes on methods used in life-history studies. In: CONNELL, J.H., MERTZ, О.В. 4 MURDOCK, W.W., eds., Readings т ecology and ecological genetics, Harper & Row, New York, p. 4-17. PERKINS, R. C. L., 1899, Introduction In: SHARP, D., ed., Fauna Hawaliensis, Cambridge Univer- sity Press, Cambridge, |: xv-ccxxviii. PILSBRY, H. A. & COOKE, C. M., 1912-1914. Manual of conchology, structural and systematic, ser. 2, vol. 22, Achatinellidae. Academy of Nat- ural Sciences of Philadelphia, 428 p. SCHALIE, H. VAN DER, 1969, Man meddles with na- ture—Hawaiian style. The Biologist, 51(4): 136-146. ЗЕАРЕЕ, $. R., 1966, Matrix algebra for the bio- logical sciences. Wiley, New York, 296 p. SOLEM, A., 1976, Endodontoid land snails from the Pacific Islands. Part |. Family Endodontidae. Field Museum Press, Chicago, 501 p. WELCH, D’A., 1938, Distribution and variation of Achatinella mustelina Mighels in the Waianae Mountains, Oahu. Bernice Pauahi Bishop Mu- seum Bulletin, 152: 164 p. WELCH, D’A., 1942, Distribution and variation of the Hawaiian tree snail Achatinella apexfulva Dixon in the Koolau Range, Oahu. Smith- sonian Miscellaneous Collections, 103(1): 1- 236. WELCH, D'A., 1954, Distribution and variation of the Hawaiian tree snail Achatinella bulimoides Swainson on the leeward and northern slopes of the Koolau Range, Oahu. Proceedings of the Academy of Natural Sciences of Philadelphia, 106: 63-107. WELCH, D'A., 1958, Distribution and variation of the Hawaiian tree snail Achatinella bulimoides Swainson on the windward slope of the Koolau Range, Oahu. Proceedings of the Academy of Natural Sciences of Philadelphia, 110: 123-212. MALACOLOGIA, 1986, 27(1): 83-96 ESTUDIO MORFOLOGICO DE LAS ESPICULAS DE DORIOPSILLA AREOLATA (GASTROPODA: NUDIBRANCHIA) F.J. García, J.C. García 4 J.L. Cervera Departamento de Zoología, Facultad de Biología, Universidad de Sevilla, Avenida Reina Mercedes s/n., Apartado 1.095, 41080 Sevilla, España RESUMEN La presencia de espículas en los Doridáceos constituye un importante mecanismo defensivo reforzado, frecuentemente, por la existencia de glándulas ácidas dispuestas por la superficie corporal. Se realiza un estudio detallado de la morfología de las espículas calcáreas de Doriopsilla areolata Bergh, así como de su disposición macroscópica en haces. Para ello, se ha considerado aisladamente varias regiones del cuerpo del animal, de las que se describe la morfología de los distintos tipos de espículas observadas en las mismas. Tales regiones son: región dorsal (de la que se describen también las espículas de las branquias y rinóforos), regiones laterales del cuerpo del animal y pie. Palabras llaves: espículas; taxonomía; Gastropoda; Nudibranchia; Doriopsilla. INTRODUCCION La presencia de espículas calcáreas es frecuente en algunos grupos de opisto- branquios (Acochlidiacea, Pleurobranchacea, Doridacea), una de cuyas funciones prin- cipales es probablemente la defensiva frente a posibles depredadores (Ros, 1976). Numerosos son los trabajos que citan la presencia de espículas en las especies de opistobranquios descritas, pero pocos son los que lo hacen con cierta profundidad. Así, Al- der 8 Hancock (1845-1855) describen la mor- fología general de las espículas de las espe- cies por ellos tratadas y su localización en el cuerpo; además, realizan una discusión sobre la morfología de las espículas en Do- ridáceos, así como de su disposición corporal y de su formación y desarrollo. Vayssiere (1901) describe también con detalle la morfo- logía de las espículas y su disposición corpo- ral. Frecuentemente, la descripción de las espículas se hace refiriéndose al aspecto general que presentan, sin concretar la zona corporal a la que pertenecen, o si lo hacen es someramente (entre otros autores, Pruvot- Fol, 1953; Marcus 8 Marcus, 1967; Edmunds, 1968; Bouchet, 1977; Ballesteros & Ortea, 1980). En ocasiones se describe la disposición de las espículas al actuar como mecanismo protector de órganos sensoriales (Kress, 1981). La abundancia de espículas que presenta (83) la especie Doriopsilla areolata Bergh, así como su diversidad morfológica, comprobada en observaciones anteriores, nos hizo abor- dar el estudio morfológico de las mismas con el fin de contribuir, en primer lugar, a un mejor conocimiento de dicha especie, ya que sobre el aspecto tratado en este trabajo pocos son los datos que se han aportado hasta la fecha. En segundo lugar, el permitir realizar estudios : encaminados a conocer con mayor precisión el papel que desempeñan en la defensa del animal. Y por último el presentar unos datos que permitan, al compararlos con los con- seguidos en futuros trabajos dedicados tam- bién a las espículas de otras especies próximas desde el punto de vista sistemático a D. areolata, ver si puede considerarse la morfología y disposición de las espículas como criterio taxonómico. MATERIAL Y METODO Para la elaboración de este trabajo se han utilizado 3 ejemplares de D. areolata captu- rados en aguas atlánticas del Sur de España (El Portil, Huelva: 37° 12’ 40” N, 7° 7’ 50" W) en la zona mediolitoral y primeros niveles infralitorales. El tamaño de los ejemplares conservados era de 1.7-2.5 cm de longitud. La disposición de los haces de espículas se observó al microscopio esteroscópico, para lo 84 GARCÍA, GARCÍA & CERVERA FIG. 1. Vista dorsal (A) y ventral (B) de un ejemplar de Doriopsilla areolata. cual se retiró la capa externa del tegumento del animal para observar así, las espículas y su disposición con mayor claridad y detalle. Las espículas esferoidales se encontraban a “flor” de la superficie debido posiblemente al tiempo (más de dos años) que llevaban los ejemplares incluidos en los líquidos con- servantes (glutaraldehido y tampón caco- dilato), mientras que en ejemplares de cap- ture e inclusión más reciente estas espículas aparecían cubiertas por el tegumento. DISPOSICION GENERAL DE LAS ESPICULAS El animal presenta por todo el cuerpo es- tructuras calcáreas dispuestas a modo de esqueleto, como se muestra en el corte trans- versal de un ejemplar, representado en la figura 2A. Mediante una observación más de- tallada se puede comprobar que las espículas se encuentran agrupadas de manera carac- terística según las distintas regiones corpo- rales. En la pared dorsal del cuerpo aparecen reunidas en haces dispuestos paralelamente a la superficie (Fig. 2D). El conjunto de haces presenta un aspecto de red de luz romboidal, que deja entrever en dicha luz, algunas espi- culas que la atraviesan de manera aislada. Desde los nudos de dicha red y dispuestos perpendicularmente a la superficie corporal, se alzan haces de espículas que van a formar el esqueleto de los tubérculos dorsales del animal, repartidos por toda la superficie dor- sal (Fig. 1A; Fig. 2B-D). En sus extremos se disponen espículas a modo de penacho, que pueden observarse por transparencia en el dorso del animal (Fig. 2B, C). Dispuestas por todo el tegumento y más superficialmente ESPICULAS DE DORIOPSILLA AREOLATA 85 FIG. 2. A, corte transversal del cuerpo de un animal. B, detalle de los tubérculos dorsales del manto. C, corte longitudinal de un tubérculo dorsal del manto. D, detalle de la disposición de los haces de espículas en el dorso del animal (no se representan los penachos apicales de espículas). E, detalle de la disposición de las espículas en el borde del manto y pared lateral del cuerpo. F, disposición de los haces de espículas del pie; la disposición de las espículas en el lado izquierdo de la figura sólo se representa parcialmente. Las escalas indican 1 mm. bm, borde del manto; ee, espículas esferoidales; hv, haces verticales; pa, penacho apical; plc, pared lateral del cuerpo; rb, retículo blanquecino externo; zmp, zona media del pie. 86 GARCÍA, GARCÍA 8 CERVERA FIG. 3. A, disposición de las espículas del tegumento que bordea el orificio branquial; pueden observarse acúmulos de espículas dispuestas horizontalmente (para ello, las branquias no se ilustran). В, disposición de las espiculas en las branquias. С, vaina rinofórica (el rinóforo se encuentra retraido). D, rinóforo. Las escalas indican 1 mm. ae, acúmulo basal de espículas; br, branquias; e, espículas; ee, espículas esferoidales; r, rinóforo; rb, retículo blanquecino externo. ESPICULAS DE DORIOPSILLA AREOLATA 87 que las anteriores, hay espículas esfe- roidales, que pueden estar de forma aislada o bien fusionadas entre sí formando grupos de número variable de ellas. El tegumento que bordea el orificio por donde emergen las branquias, presenta es- pículas esferoidales iguales a las del resto del dorso. Más internamente, se distinguen es- pículas agrupadas en haces laxos dispuestos principalmente en posición vertical, junto con otros intercalados oblicuamente. La zona más inferior de esta porción tegumentaria está rodeada por espículas dispuestas ho- rizontalmente a modo de anillo (Fig. 3A). En la base de los penachos branquiales, las espículas de morfología alargada se en- cuentran dispuestas horizontalmente o lige- ramente oblicuas. Enlospenachos propiamen- te dichos, estas espículas se hacen menos numerosas a la vez que se observan con mayor frecuencia las espículas esferoidales, las cuales son más numerosas en los extre- mos de aquéllos (Fig. 3B). En la vaina rinofórica (Fig. 3C), las espicu- las están en haces dispuestos verticalmente, los cuales se ramifican al llegar a la zona más externa de dicha vaina. Hay espículas esferoidales aisladas, dispuestas por toda la superficie. En los rinóforos (Fig. 3D), las espículas se encuentran en el tercio basal tanto en el tronco central como en las laminillas, en las cuales se disponen perpendicularmente al eje del rinóforo. En el borde del manto, los haces aban- donan la disposición que presentan en la zona central del dorso, para formar una red de organización menos geométrica (Fig. 1В; Fig. 2E). Los haces devienen más estrechos a medida que se aproximan a los bordes exteriores del manto a la vez que se ramifican más. Los haces verticales desaparecen tam- bién al aproximarse a los extremos del manto. Las espículas esféricas se hacen menos numerosas en la superficie inferior del borde del manto, así como en los lados del cuerpo del animal. Los haces de espículas del borde del manto, al pasar a la zona lateral del cuerpo, se hacen menos densos, quedan las espícu- las dispuestas de una manera menos apre- tada y se observa mayor cantidad de espícu- las aisladas o agrupadas en haces de escaso número (Fig. 2E). El pie, ventralmente, tiene una capa de espículas esferoidales agrupadas entre sí muy densamente (Fig. 2A). Más internamen- te, respecto a la capa anterior, están los haces de espículas alargadas, dispuestos transversalmente al eje longitudinal del ani- mal. Estos haces son más estrechos que los indicados para las regiones dorsal y lateral del cuerpo. Se aprecia también gran cantidad de espículas aisladas, sobre todo en la zona media del pie (Fig. 2F). En la zona próxima a los bordes del pie, los haces se hacen más finos a la vez que aumentan sus ramificacio- nes. DESCRIPCION MORFOLOGICA DE LAS ESPICULAS 1. Región dorsal Repartidas por toda la superficie dorsal se aprecian espículas esferoidales (Fig. 4А-С), de superficie rugosa o espinosa, que pueden encontrarse aisladas o bien fusionadas entre sí. Estas estructuras presentan tamaños que oscilan, en los ejemplares observados, entre 75-90 рт en las encontradas por las zonas centrales del dorso del animal y entre 40-115 um en las del borde del manto. Por debajo de estas espículas se disponen los haces de espículas descritos en el apartado anterior. Al observarlos al microscopio óptico se aprecia que la morfología de las espículas presenta algunas características según la región a la que pertenezcan. En la zona central, las espículas son li- neales, curvadas ligeramente por su centro, en cuyo caso pueden presentar una protuber- ancia en el área de curvatura (Fig. 4D-G); o bien estar curvadas de manera irregular (Fig. 4H-K). Por transparencia se observan líneas curvas concéntricas en los extremos, que se continúan siguiendo los ejes de las espículas, semejantes a las líneas de crecimiento de las escamas de los peces. Estas líneas curvas son menos perceptibles en las espículas con curvatura irregular y sobre todo en las de mayor tamaño. En los casos que presentan la protuberancia en la zona de curvatura, las prolongaciones longitudinales se dirigen a di- cha protuberancia (Fig. 4G). Ocasionalmente, se ven espículas lige- ramente curvadas y con varios abultamientos repartidos por su superficie (Fig. 4L). Los tamaños observados son muy varia- bles. Así, entre las del primer tipo, los tamaños oscilan entre 200-640 um y en las de curvatura irregular entre 115-385 um. 88 GARCÍA, GARCÍA & CERVERA Jesh eS O Вары RE DAS 2 д S a | EI de [= 1 A I 2, | | E Al El м DAN FIG. 4. Espículas de la zona central del dorso. Las escalas indican 30 um, salvo en Му Ñ que indican 10 um Las espiculas dispuestas verticalmente y dirigidas al interior de los tuberculos dorsales del animal son fusiformes, con uno de los extremos truncados y el otro redondeado (Fig. 4M). Las lineas concentricas son tam- bien visibles por transparencia en el extremo redondeado, y continúan más o menos hacia el otro extremo. El tamaño de estas espículas varía considerablemente según el tamaño del tubérculo dorsal, e incluso dentro de un mismo tubérculo. Miden desde 57 um hasta 580 um. En el borde del manto, las espículas alcanzan los tamaños mayores de todo el cuerpo (alcanzan 730 вт de longitud). La morfología es muy variable. Se distinguen: espículas fusiformes o ligeramente curvadas en el centro, con esta zona algo más ensan- chada que el resto de la espícula (Fig. 5A); otras tienen forma de L con rugosidades por ESPICULAS DE DORIOPSILLA AREOLATA 89 FIG. 5. A-G, espículas del borde del manto; H-L, espículas del tegumento que bordea el orificio branquial. Las escalas indican 50 um. 90 GARCÍA, GARCÍA 8 CERVERA su superficie (Fig. 5B); fusiformes con una protuberancia semejante a la descrita para la región central del dorso o a modo de espolón (Fig. 5C), que puede prolongarse para dar a la espícula un aspecto trirradiado (Fig. 5D, E); espículas tetrarradiadas, con un brazo más largo que los otros tres (Fig. 5F, G). Se observan también espículas por toda la región dorsal, pero más frecuentemente en los bordes del manto, con aspecto de prisma rectangular, que pueden estar aisladas o bien agrupadas en forma de cruz o de hexactina (por analogía con las espículas de dicho nom- bre de las esponjas), con tamaños entre 10-35 um (Fig. 4N, N). En el tegumento que bordea el orificio por donde emergen las branquias, las espículas, con frecuencia, presentan en el centro aproximadamente una protuberancia, uno de los extremos de la espícula suele estar curvado mientras que el otro se mantiene recto. Las líneas concéntricas se observan en los extremos y pueden llegar más o menos cerca de la zona central de la espícula. El tamaño de estas espículas es de 115-175 um; se ha visto una espícula perteneciente a este tipo cuyo tamaño era de 307 um (Fig. 5H-J). Espículas fusiformes con los extremos redondeados. Dentro de este modelo se dis- tinguen tres tipos según el tamaño, oscilando éstos para cada grupo entre 20-30 рт, 75-95 um y de 305-345 шт. Las líneas соп- céntricas no se observan en las mayores, mientras que en las pequeñas sólo se ven en los extremos (Fig. 5K). Espículas fusiformes, o algo curvadas, que presentan en la zona central una depresión. El tamaño observado para este tipo alcanza aproximadamente 230 um (Fig. 51). En las branquias el tamaño de las espícu- las esferoidales oscila aproximadamente en- tre 45-57 um. Entre las espículas de morfo- logía alargada se pueden distinguir varios tipos: - Espiculas con tamaño pequeño (10-15 um), fusiformes. - Espículas considerablemente mayores, entre las que se puede distinguir varios tipos: fusiformes con la zona central algo más en- sanchada que el resto; esta zona queda ade- más diferenciada de los extremos al pre- sentar un estrechamiento a cada lado; las líneas curvas concéntricas se observan por las dos áreas laterales de la espícula y en el área central sólo se ven las líneas longitudi- nales que continúan desde las líneas curvas anteriores. El tamaño de la espícula alcanza aproximadamente 285 um (Fig. 6A). Espiculas curvadas aproximadamente por el centro, con la superficie rugosa y las lineas curvas concentricas visibles solamente en los extremos. La zona central, en algunos casos, esta mas ensanchada. Las espiculas ob- servadas que encajan en este tipo presentan tamaños entre 260-392 um de longitud (Fig. 6B, C). Espículas curvadas aproximadamente a dos tercios de distancia de uno de los extre- mos. De los dos brazos así formados, el más largo se mantiene recto mientras que el otro suele estar curvado. En la zona de curvatura hay una prominencia que sobresale más o menos. Las líneas curvas concéntricas sólo se observan en los extremos, el resto queda más o menos opaco según el tamaño de la espícula; en las de menor tamaño se ven mejor dichas líneas concéntricas y sus pro- longaciones longitudinales. El tamaño de este tipo de espículas varía entre 150-270 um de longitud (Fig. 6D-F). Espículas fusiformes o curvadas por el cen- tro; la zona central está dilatada con respecto al resto de la espícula y a veces sobresale de dicha zona una prominencia a cada lado. La superficie de la espicula es rugosa y las li- ‘neas curvas no son visibles. Los tamaños observados oscilan entre 120 y 190 um (Fig. 6G, kl): En los rinóforos, en función del tamaño de las espículas encontradas, éstas se pueden reunir en dos grupos, espículas de tamaño menor de 25 um y espículas mayores de 45 шт. Entre las primeras se distinguen varios tipos: pequeñas estructuras calcáreas esferoidales aisladas y a veces agrupadas de dos en dos, con tamaños aproximados entre 3 y 13 um; espículas fusiformes de 7-11 um de longitud, dispuestas de manera aislada o agrupadas en forma de estrella de seis pun- tas que llegan a alcanzar 23 um (Fig. 61), o en forma de cruz con tamaños entre 5 y 10 um; espículas fusiformes (5-10 um de longitud) con tubérculos dispuestos en su superficie (Fig. 6J). Entre las espículas pertenecientes al segundo grupo se distinguen los modelos siguientes: espículas esferoidales, sobre todo en la zona próxima al vértice, de superficie rugosa, dispuestas aisladamente. Los tamaños observados en este tipo de espicu- las oscilan entre 45 y 75 um (Fig. 6K); es- pículas con una curvatura aproximadamente ESPICULAS DE DORIOPSILLA AREOLATA 9A FIG. 6. A-H, espiculas de las branquias; I-Q, espiculas de los rinöforos. Las escalas indican 30 pm. en la zona central, forma entre las dos semi- (Fig. 6L, M); en ocasiones no aparece la espículas ángulos muy variables, en la zona prominencia y la zona de curvatura está des- de curvatura suele aparecer una prominencia plazada hacia uno de los extremos, la 92 GARCÍA, GARCÍA & CERVERA superficie es rugosa y las líneas concéntricas se observan sólo en los extremos de la es- picula (Fig. 6N, N). El tamaño de las espicu- las de estos ultimos tipos oscila entre 75 y 305 um. Espiculas con una prominencia central. A cada lado de dicha prominencia la espicula presenta una curvatura. El tamaño que alcanzan es de aproximadamente 75 um (Fig. 60). Presentan a veces dos prominen- cias dispuestas en el mismo lado de la es- pícula (Fig. 6P), u opuestamente (Fig. 60). 2. Región lateral Al igual que en otras partes anteriormente citadas, se distinguen dos grupos de espícu- las atendiendo al tamaño. Las de tamaño reducido son ovaladas, a veces con una depresión en cada extremo (Fig. 7A, B); o bien tienen los extremos agudos en cuyo caso la espícula es más estrecha que las anteriores. Los tamaños que alcanzan oscilan entre 7-19 um y 7-11 pm, respectivamente. A veces se observan estas espículas agrupadas en forma de estrella de hasta doce puntas. Estas agrupaciones alcanzan tamaños entre 20 y 40 рт. Entre las del segundo grupo (mayores de 20 um) se distinguen los siguientes tipos: Espículas esferoidales, en menor número que en otras zonas del cuerpo del animal, tienen la superficie rugosa. Los tamaños ob- servados oscilan entre 20 y 35 ¡um (Fig. 7G). Espículas curvadas desde el centro, el cual suele ser algo más ancho que el resto de la espícula, mostrando además una prominen- cia por el lado convexo; próximos a los extre- mos suele haber otras protuberancias de me- nor tamaño que la central. Las líneas curvas concéntricas son visibles en los extremos. Los tamaños medidos de este tipo de espi- culas oscilan entre 92 y 153 um (Fig. 7C, D). Espículas semejantes al tipo anterior pero con los extremos que, bien vuelven a curvarse en sentido opuesto al seguido en el inicio de la curvatura, o bien no llegan a curvarse en sentido opuesto, pero sí lo hacen lo suficiente como para dar a la espícula un aspecto fusiforme con una curvatura central (Fig. 7E, F). En ninguna espícula de este tipo se han visto marcas concéntricas. El tamaño observado se encuentra entre 88 y 115 um para las primeras, y entre 134 y 173 um para las segundas. Espículas algo curvadas, con los extremos redondeados y de superficie bastante lisa, aunque a veces se observan pequeñas prominencias; las líneas concéntricas y sus prolongaciones longitudinales son visibles. El tamaño oscila entre 175 y 253 um (Fig. 7H). Espículas semejantes al tipo anterior pero de superficie rugosa y curvaturas más irregu- lares. Según el tamaño se pueden reunir en dos grupos, 75-115 um y 270-345 um (Fig. ZU). Espiculas fusiformes con una prominencia central. No se ven líneas concéntricas a pesar de su transparencia, el tamaño aproximado es de 90 um (Fig. 7J). 3. Pie Por toda la superficie ventral del pie se encuentran dispuestas, muy densamente, espículas esferoidales de superficie espi- nosa, que pueden estar independientes o fusionadas unas con otras (Fig. 7K, L). Los tamaños observados varían entre 20 y 45 um, aunque algunas llegan a las 75 шт. Espículas de aspecto fusiforme con los extremos agudos (entre 15 y 20 um), suelen estar aisladas o bien agrupadas a modo de estrella (Fig. 7M); en ocasiones estas espicu- las se encuentran atravesadas por pequeñas estructuras alargadas (Fig. 7N). Espiculas grandes (entre 345-460 рт), fusiformes, con los extremos redondeados; superficie rugosa y opaca (Fig. 7N). A veces se observa en uno de los lados una depresión de la espícula; por los bordes de dicha depre- sión se observan las líneas curvas concén- tricas (Fig. 70). El tamaño de las espículas de este tipo varía entre 155—460 um de longitud. Las mayores se encuentran en la zona central del pie; en los bordes de éste, las mayores observadas solo alcanzan 270 um y tienen los dos extremos de diferente grosor (Fig. ИР). Espículas con una curvatura еп la zona central, se observan a veces, por dicha zona, líneas longitudinales paralelas a los bordes de la espícula. Los tamaños observados os- cilan entre 125-385 ¡um de longitud (Fig. 70). En ocasiones se ven espículas semejantes a las anteriores pero de menor diámetro; estas espículas a veces están curvadas hasta for- mar aproximadamente un ángulo recto (Fig. 7R), e incluso se observan casos en los que se producen dos curvaturas a modo de zig- zag (Fig. 7S), que da a la espícula un aspecto ESPICULAS DE DORIOPSILLA AREOLATA 93 FIG. 7. A-J, espiculas de la pared lateral del cuerpo; K-Y, espiculas del pie. Las escalas indican 30 um, salvo en A, В, М, N que indican 10 um. 94 GARCÍA, GARCÍA 8 CERVERA de S. Estos tipos de espículas alcanzan tamaños comprendidos entre 115-350 рт. Espículas con una ligera curvatura central; en dicha zona presentan una prominencia más o menos manifiesta (Fig. 7T, U). Las espículas medidas alcanzan tamaños que os- cilan entre 85-210 um de longitud. A veces se observan dos e incluso tres prominencias. En estos casos la espícula está menos curvada (Fig. 7V). El tamaño varía entre 90-200 um, excepcionalmente lle- gan а 345 um. Esporádicamente se observan espículas con morfología que no encajan en los tipos anteriormente descritos, como son los repre- sentados en la figura 7Y-W. DISCUSION La existencia de espículas muy aparentes que se disponen irregularmente por todo el manto es señalada por Ros (1975) en Doriop- silla pusilla, lo cual hacen extensivo Ballest- FOT. 1. Detalle de las ramificaciónes de los haces de espículas del borde del manto. La escala corres- ponde a 200 pm. FOT. 2. Detalle de un haz de espículas del borde del manto. Se puede apreciar la disposición apre- tada de las espículas que lo integran. La escala corresponde a 100 um. eros 8 Ortea (1980), a todas las Doriopsilla. Sin embargo y en base a nuestras observacio- nes, еп D. areolata la disposición de las espı- “culas varía, según la región corporal consid- erada y dicha disposición se mantiene de un ejemplar a otro. Estudios futuros sobre la morfología de las espículas en otras especies de Doriopsilla podrán aclarar si existen o no importantes diferencias que permitan separar tales especies en base a la morfología de las espículas y por tanto, conocer si este carácter tiene una utilidad taxonómica. Aunque en cada región corporal las espí- culas presentan peculiaridades, todos los tipos encontrados en el cuerpo del animal pueden incluirse en varios grupos en función del tamaño y de la morfología de las espícu- las. Se distinguen así: espículas de aspecto fusiforme y tamaños no superiores a 25 um de longitud; espículas esferoidales, cuyos tamaños oscilan entre 7-90 um; y espiculas de morfología diversa, no esferoidal y ESPICULAS DE DORIOPSILLA AREOLATA 95 tamaños mayores de 50 um de longitud. Den- tro de este tercer grupo, las espículas ob- servadas se pueden considerar, de acuerdo a su morfología, en los cuatro tipos siguientes: 1) espículas fusiformes, sin prominencias muy sobresalientes; pueden ser algo curvadas y de superficie más o menos granu- losa, pero se manifiesta una tendencia a adoptar un aspecto fusiforme; 2) espículas curvadas en mayor o menor grado y con una prominencia sobresaliente, la cual deviene como un simple mamelón o bien como una rama perfectamente diferenciada; 3) espícu- las con una o varias curvaturas muy pronun- ciadas con aspecto de L o S, sin presentar prominencias muy sobresalientes; 4) espícu- las tetrarradiadas; suelen tener uno de los radios más largo que el resto, el cual a veces constituye la prolongación de otro de ellos, al configurarse ambos en un mismo eje. Las espículas fusiformes dispuestas en los tubérculos dorsales, debido al aspecto que presentan sus dos extremos (uno redon- deado y el otro truncado y de superficie rugosa), así como el que las líneas curvas concéntricas sólo se encuentren dirigidas hacia el extremo redondeado, hace pensar que estas espículas no sean sino espículas rotas. No obstante no hemos observado nin- guna espícula, en los tubérculos dorsales, que tengan los dos extremos semejantes y las líneas curvas concéntricas dirigidas a cada uno de ellos. Se podría considerar que mientras las demás espículas del cuerpo del animal presentan, en su formación, más de un sentido de crecimiento (lo que quedaría manifestado mediante las líneas concéntri- cas), en el caso de las espículas fusiformes de los tubérculos dorsales, sólo tuviese un sentido de crecimiento que sería el que indi- can las líneas concéntricas. Pruvot-Fol (1952) hace referencia a las espículas de D. areolata ilustradas por Vays- siere (1901, pl. 3, fig. 20), y al respecto se- ñala: “lls ont été représentés par Vayssiere avec leur surface rendue rugueuse par de nombreux nodules ou tubercules, et partielle- ment avec des branches latérales. J'ai trouvé ces spicules de taille extrémment dif- férentes, mais toujours simples avec de rares nodules. . . .” Las espiculas observadas рог nosotros generalmente tienen la superficie menos rugosa que las representadas por Vayssiere aunque la presencia de ramas la- terales que indica este autor sí aparecen con frecuencia en nuestras observaciones. El he- cho de que Pruvot-Fol encontrase las espicu- las simples y raramente con nódulos puede ser explicado si las observaciones que hubie- se realizado se centrasen a la región dorsal del cuerpo del animal, ya que como se ilustra en la figura 4, es ésta la zona del cuerpo en la que gran parte de las espículas son simples o con nódulos, pero carecen de ramas late- rales. Pruvot-Fol (1952) ilustra el extremo de una espícula (p. 410, fig. 9), en el que aparece una serie de líneas curvas con el lado con- vexo dirigido hacia el extremo de la espícula y el cóncavo hacia el centro de la misma; y entre estas líneas curvas y la silueta de la espícula algunas líneas longitudinales. Tam- bién señala para las espículas: “ ... surface striée longitudinalement, comme s'ils étaient entourés d'une sorte de gaine de fibrilles parallèles . .. Ala cassure, on voit qu'ils sont formés de couches concentriques in- égales. . Estas dos precisiones reali- zadas por Pruvot-Fol podemos compararlas por una parte con las líneas curvas que indi- camos en los extremos de muchas espículas observadas por nosotros, y por otra parte con las prolongaciones longitudinales de dichas líneas curvas. Thompson (1975) considera la separación de los géneros Dendrodoris y Doriopsilla como inadecuada y por tanto que el segundo es sinónimo del primero. Considera además a D. areolata como especie posiblemente si- nónima de Dendrodoris miniata (Alder & Han- cock, 1864), aunque no explica las razones que le llevan a adoptar tal paralelismo. Para D. miniata indica * . . . The mantle was rather smooth and the general aspect of the body convex but with a lens it is possible to see numerous low soft papillae on the upper pal- lial surface.” En D. areolata, por el contrario, aparecen numerosos tubérculos de tamaño considerable (sostenidos por acúmulos de espículas visibles por transparencia). Es- timamos por tanto que en función del carácter indicado, D. areolata no debe considerarse como sinónima de D. miniata. Ballesteros y Ortea (1980), describen en los ejemplares jóvenes de D. areolata un movimiento ondulatorio del borde del manto que no aprecian en los adultos, y consideran dicho comportamiento en los jóvenes como un mecanismo de advertir de su presencia a posibles enemigos, ya que su manto es de secreción ácida. La observación de los bor- des del manto de un ejemplar joven (de 9 mm 96 GARCÍA, GARCÍA & CERVERA de longitud) revela la inexistencia de espícu- las en el mismo, lo cual puede facilitar el movimiento ondulatorio a que se ha hecho alusión. Con la progresiva aparición y desar- rollo de las espículas, los bordes del manto se hacen más rígidos y dificultan con ello su movimiento, como ocurre en los ejemplares adultos en los que los movimientos quedan muy restringidos. AGRADECIMIENTOS Agradecemos a J.l. Navas, la ayuda pres- tada en la elaboración de este trabajo. REFERENCIAS CITADAS ALDER, J. 4 HANCOCK, A., 1845-1855, A mono- graph of the British nudibranchiate Mollusca. Ray Society, London, parts 1—7. BALLESTEROS, M. 8 ORTEA, J. A., 1980, Contri- bución al conocimiento de los Dendrodorididae (Moluscos: Opistobranquios: Doridáceos) del li- toral Ibérico. I. Publicaciones del Departamento de Zoología, Universidad de Barcelona, 5: 25-37. BOUCHET, P., 1977, Opisthobranches de profon- deur de l'océan Atlantique: Il. Notaspidea et Nudi- branchiata. Journal of Molluscan Studies, 43: 28—66. EDMUNDS, M., 1968, Opisthobranchiate Mollusca from Ghana. Proceedings of the Malacological Society of London, 38: 83-100. KRESS, A., 1981, A scanning electron microscope study of notum structure in some dorid nudibra- nchs (Gastropoda: Opisthobranchia). Journal of the Marine Biological Association of the United Kingdom, 61: 177-191. MARCUS, E. 8 MARCUS, E., 1967, American opis- thobranch mollusks. Studies in Tropical Oceano- graphy, Miami, 6: 1-256. PRUVOT-FOL, A., 1952, Compléments а la con- naissance anatomique de Doriopsilla areolata Bergh. Bulletin de la Société Zoologique de France, 77: 411—414. — PRUVOT-FOL, A., 1953, Etude de quelques opis- thobranches de la cóte Atlantique du Maroc et du Sénégal. Travaux de I’ Institut Scientifique Ché- rifien, Zoologie, 5: 1-105, 3 pl. ROS, J., 1975, Opistobranquios (Gastropoda: Eu- thyneura) del litoral ibérico. Investigaciones Pes- queras, 39: 269-372. ROS, J., 1976, Sistemas de defensa en los Opis- tobranquios. Oecologia Aquatica, 2: 41-77. THOMPSON, T. E., 1975, Dorid nudibranchs from eastern Australia (Gastropoda, Opisthobran- chia). Journal of Zoology, London, 176: 477-517. VAYSSIERE, A., 1901, Recherches zoologiques et anatomiques sur les Mollusques Opisthobran- ches du Golfe de Marseille (suite et fin). 3. An- nales du Museum d'Histoire Naturelle Marseille, 6: 1-130, 7 pl. A MORPHOLOGICAL STUDY OF THE SPICULES OF DORIOPSILLA AREOLATA (GASTROPODA: NUDIBRANCHIA) F.J. Garcia, J.C. García & J.L. Cervera Doridacean spicules are important for de- fense; they accompany acid glands on the body surface. The morphology of the calcar- eous spicules of Doriopsilla areolata Bergh 15 studied and illustrated in detail. The spicules are shown to be aggregated in bundles of fibers. Different body surfaces are considered separately because the spicules differ from place to place. The dorsal region (including gills and rhinophores) and lateral regions of the body and foot each have several kinds of spicules. MALACOLOGIA, 1986, 27(1): 97-106 HIGH GENETIC SIMILARITIES AND LOW HETEROZYGOSITIES IN LAND SNAILS OF THE GENUS SAMOANA FROM THE SOCIETY ISLANDS Michael $. Johnson’, James Murray? & Bryan Clarke? ABSTRACT Land snails of the genus Samoana from the Society Islands were examined for allozymic variation at 20 loci. The species S. diaphana, S. attenuata, S. burchi, and S. annectens form a tight group, with average genetic identities of 0.95. Combined with the results of earlier studies on the confamilial genus Partula, these findings indicate that speciation entailing little genic change at allozyme loci is common in the Partulidae. Unlike most species of Partula, however, these four species of Samoana have very low heterozygosities (H = 0.002). In the single case in which two alleles at a locus were common in the same population, there was a marked deficit of heterozygotes, suggesting that self-fertilization may be common. Comparison of allozymes of Samoana and Partula confirmed the distinctness of the genera, with genetic identities between them averaging approximately 0.25. Despite its current taxo- nomic placement, $. jackieburchi closely resembles Partula, and differs greatly from Samoana. The allozymes support conchological and other evidence that S. jackieburchi belongs in the genus Partula. It appears that $. jackieburchi is anatomically convergent with Samoana, and that genital anatomy 1$ not infallible as a taxonomic character in pulmonate snails. INTRODUCTION Five species of the land snail genus Samoana have been described from the So- ciety Islands, French Polynesia (Kondo, 1973, 1980). Unlike the closely related and co- occurring genus Partula, which has been the subject of detailed genetic and evolutionary studies (Crampton, 1916, 1925, 1932; Murray 8 Clarke, 1980), Samoana has received rel- atively little attention, due in part to the scar- city and inaccessibility of its populations. The work on Partula increases the interest in Samoana with respect to two questions. First, is Samoana similar to Partula in the genetic structure of its populations and in its mode of speciation? Second, do the two genera really represent distinct lineages? The close similarity of the two genera 15 suggested by both Samoana attenuata and S. diaphana having been originally described as species of Partula. Indeed, these two species had been placed in separate subgenera within Partula: S. attenuata in Partula sensu stricto (Pilsbry, 1909-1910) and $. diaphana in Leptopartula, along with the conchologic- ally similar P. arguta and P. turgida (Crampton 8 Cooke, 1953). The most obvious distinction between the two genera is in the structure of the male genitalia (Kondo, 1973): in Samoana the phallus is shorter and stouter, and the epiphallus is more distinct than in Partula. This anatomical difference was the basis for recognizing the recently described S. jackieburchi as a member of the genus Samoana (Kondo, 1980). The shells of $. jackieburchi are indistinguishable from those of P. otaheitana rubescens, raising the ques- tion whether genital anatomy provides an adequate basis for distinguishing the two genera. In an attempt to answer these questions about Samoana, we have examined elec- trophoretic patterns in the enzymes of all the species of Samoana from the Society Islands. We report here the results of these examina- tions, a comparison of Samoana and Partula, and a reappraisal of the taxonomic position of S. jackieburchi. MATERIALS AND METHODS Samples. Collections of adults and juve- niles were taken from the islands of Tahiti, Moorea, Raiatea, and Huahine. More than ‘Department of Zoology, University of Western Australia, Nedlands, Western Australia 6009, Australia. “Department of Biology, University of Virginia, Charlottesville, Virginia 22901, U.S.A. Department of Genetics, University of Nottingham, Nottingham NG7 2RD, United Kingdom. 98 JOHNSON, MURRAY 8 CLARKE MOOREA 172 5015 1499 40' W TAHITI 10 km FIG. 1. Map of Tahiti and Moorea, showing collection sites. Lines indicate major mountain ridges. Locality codes: M1 = Faatoai; M2 = Uufau; МЗ = Maatea; M4 = Hotutea; T1 = Fare Hamuta; T2 = Pirai; ТЗ = Tiarei. one site was sampled in Tahiti and Moorea, as shown in Fig. 1. The collections provided samples of all the species of Samoana that have been recorded from the Society Islands: S. attenuata (Pease), 48 individuals col- lected on three islands: Tahiti, from Pirai (N = 3) and Tiarei (N = 5) valleys; Moorea, from high in Faatoai Valley (N = 13) and from the pass at the head of Uufau Valley (N = 10); Raiatea, from Miti Miti Aute Rahi (N = 17). The species also occurs on the islands of Bora Bora and Tahaa. S. annectens (Pease), 1 individual col- lected on Huahine. S. burchi Kondo, 6 individuals collected on Tahiti, taken in a large area centering on Fare Hamuta on the Aorai Trail, at an elevation of approximately 800 m. S. diaphana (Crampton & Cooke), 53 indi- viduals collected on two islands: Moorea, from Faatoai (N = 20) and Uufau (N = 5) valleys, where it occurs sympatrically with S. attenuata, and from high in Maatea (N = 1) and Hotutea (N = 21) valleys, where $. at- tenuata is absent; Tahiti, from Fare Hamuta (N = 6), where it occurs sympatrically with S. burchi. This sample is notable, because S. diaphana has previously been reported only from Moorea (Kondo, 1973). S. jackieburchi Kondo, 33 individuals from Tahiti, from Tiarei Valley. These specimens were identified morpho- logically (see Kondo, 1973), and the identifi- cation of the Tahitian species was confirmed by Dr. Kondo. In the case of $. jackieburchi, identification required dissection to distin- guish it from P. otaheitana rubescens. Our sample of 34 adults from Tiarei included a single P. otaheitana. To increase our coverage of Samoana, a sample of 19 S. conica from Tutuila in Samoa (approx. 2000 km from Tahiti) was included. To allow comparisons between genera, data from 3 species of Partula were included: P. otaheitana, the species from which S. jackieburchi was recently removed (12 ani- mals from Pirai Valley and 19 from Papehue Valley, Tahiti); P. affinis, until recently consid- ered a subspecies of P. otaheitana (Kondo & Burch, 1979) (3 from Tiarei Valley, Tahiti); P. gibba, a more distant species of Partula (15 from Saipan). As a precaution, several spec- imens were dissected to confirm the presence of Partula-like genitalia in the Tahitian species of Partula. Allozymes in P. otaheitana and P. gibba have previously been compared with those in species of Partula from Moorea (Johnson et al., 1977). ALLOZYMES OF SAMOANA 99 Morphology. The inclusion of the speci- mens from Fare Hamuta т $. diaphana was tested by examining both shells and genitalia from adults (S. diaphana is distinguished by an almost transparent, semi-globose shell; Crampton 8 Cooke, 1953; Kondo, 1973). In comparing the Moorean and Tahitian speci- mens, we recorded the weight (Wt), length (L), and width (W) of the shell, as described before (Murray 8 Clarke, 1980). Taking the shell as an approximate cone, an estimate of shell volume was obtained as V = 1/12 W*L. Only the adults from Moorea and Tahiti were included in these analyses, although juveniles of S. diaphana were also easily distinguished from the other spe- cies. Comparisons of genital anatomy were based on the work of Kondo (1973), whose fig. 4 shows a distinctly inflated epiphallus for S. diaphana but not for the other species of Samoana. All of the adult Samoana from Moorea and Tahiti were examined for this characteristic. Electrophoresis. Allozymic variation was examined by conventional starch-gel elec- trophoresis. The enzymes and procedures were those used by Johnson et al. (1977), except that a tris-maleate buffer (Selander et al., 1971) was used for glutamate oxaloace- tate transaminase (Got-1 and Got-2 loci), iso- citrate dehydrogenase (/dh-1 and /dh-2), and phosphoglucomutase (Pgm-1 and Pgm-2), and a tris-EDTA-borate buffer (Selander et 9 S. diaphana 00 Width, mm al., 1971) was used for nucleoside phospho- rylase (Np). The enzymes represent 20 structural loci, each designated by an abbreviation of its en- coded enzyme. Allelic designations indicate the electrophoretic mobilities of the corre- sponding allozymes relative to those of P. gibba. А homozygous stock of P. gibba is maintained as a standard in the laboratory. Negative mobilities refer to cathodally mi- grating allozymes. Although we do not have data from genetic crosses in Samoana, seg- regations of the allozymes in laboratory crosses of Partula are consistent with Men- delian inheritance (Johnson et al., 1977, and unpublished). Based on the 20 loci, unbiased estimates of genetic identity (Nei, 1978) were calcu- lated between all pairs of species. The matrix of genetic identities was summarized by UPGMA clustering (Sneath & Sokal, 1973). RESULTS Morphology The differences in size and shape among $. diaphana, S. attenuata, and S. burchi are clear from Fig. 2, in which the widths and lengths of the shells from adults are plotted. Although there is considerable variation in each species, S. diaphana is consistently smaller and, relatively, wider than its sympa- S. burchí Length, mm FIG. 2. Comparison of width and length of shells among three species of Samoana. Lines encompass conspecific specimens. Open symbols = specimens from Moorea; solid symbols = from Tahiti. 100 JOHNSON, MURRAY 8 CLARKE Weight, mg Volume, mm3 FIG. 3. Comparison of weight and “volume” of shells among three species of Samoana. Open symbols = from Моогеа: solid = from Tahiti. tric congeners. In addition, S. diaphana has a lighter, less robust, shell than $5. attenuata and the even more heavily shelled S. burchi (Fig. 3). Although the shells of S. diaphana from Fare Hamuta are larger than most of those from Moorea, they clearly belong to the Moorean group. The dissections confirm this placement. The two adult S. diaphana from Tahiti had an inflated epiphallus, as did the dissected spec- imens from each of the four Moorean popu- lations. In contrast, none of the $. burchi or $. attenuata had an inflated epiphallus. Thus, the shells and genitalia were consistent in confirming the presence of $. diaphana on Tahiti. Electrophoresis With the exception of S. jackieburchi, the species of Samoana from the Society Islands have very similar allozymes. The four species that resemble each other ($. diaphana, $. attenuata, S. burchi, and S. annectens) will be considered before comparing them with the other species of Samoana and Partula. А striking feature of the four species 1$ their low variability. In most samples, no heterozygotes were found at any of the 20 loci, and the highest average observed heterozygosity was 0.012 (Table 1). The only locus with two com- mon alleles segregating in a population was Pgi, and the polymorphism occurred only in the sample of $. diaphana from Hotutea. Even in this case, however, there was a deficit of heterozygotes: the expectation according to the Hardy-Weinberg equilibrium was 9.0 he- terozygotes, but only 3 were observed (x*, = 7.00. Р.= 0:01). In addition to the paucity of variation within populations, little allozymic divergence was detected between species. Of the 20 loci ex- amined, 15 were identically monomorphic in all 4 species. Even among the 5 variable loci, the similarity between species was great (Ta- ble 1). Three of the variable loci (/dh-2, Pgm-1, and Pgm-2) had different alleles fixed in different populations. But these differences occurred within species. The greatest resem- blance was found between S. diaphana and ALLOZYMES OF SAMOANA 101 TABLE 1. Allelic frequencies at loci which are variable in Samoana from the Society Islands (excluding S. jackieburchi). Loci for which the same allele occurred at a frequency of >0.95 in all populations are not shown, but are included in the calculation of average observed heterozygosity (H). Locality codes: as in Fig. 1, except that В = Ralatea and Н = Huahine. diaphana attenuata burchi annectens Locus Allele M1 M2 M3 M4 Til M1 M2 T2 T3 R T1 H Sample size 20 5 1 21 6 13 10 3 5 17 6 1 Got-2 — 1.00 = — = = .10 .03 — — — 1.40 1.00; 1:00 1.00 1.00° 1.00’ 100 1:00 1.00 .90 197: 1.00 1.00 Idh-2 1.00 1.00 1.00 1.00 1.00 1.00 = — 1.00 1.00 — 1.00 1.00 .50 = = = — — 1.007 1.00 = — 1.00 — — Pgi .95 1.00 1.00 1.00 of 1:00 100 “1:00 1.00 1:00 100 1.00 1.00 50 = = = .69 = = = — = — — = Pgm-1 1.13 1.00 — 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.00 1.04 — 1.00 — = = = = — = = = = Pgm-2 1.12 1.00 1.00 1.00 1.00 — = — = — — — — 1.09 — == = = 1.00 = = 1.00 = = 1.00 = 1.04 = — — — = = = — 1.00 1.00 — — .87 = — = = = 1.00 1.00 = = — = 1.00 H (20 loci) .002 .000 000 .012 .000 000 000 000 .010 006 000 000 $. Бигс from Fare Hamuta, which were iden- tically homozygous for all 20 loci, giving a genetic identity of 1.00. The least resem- blance between populations of different spe- cies was 0.86. The identities between species were not appreciably different from those be- tween conspecific populations, which ranged from 0.89 to 1.00. To simplify comparisons with the other spe- cies, allelic frequencies in S. diaphana and S. attenuata were taken as the averages among conspecific populations, weighted by sample size. The great genetic similarities between S. diaphana, S. attenuata, S. burchi, and S. an- nectens, and the lack of variation within them, set the group apart from other species (Table 2). With the exception of the homozygous P. gibba, all the other species examined had higher heterozygosities, ranging from 0.027 in 5. conica to 0.135 in $. jackieburchi (Table 2). Furthermore, none showed a significant de- рациге from Hardy-Weinberg equilibrium. The pattern of genetic identities (Table 3) is well summarized by the phenogram in Fig. 4. S. diaphana, S. attenuata, S. burchi, and S. annectens form a very tight cluster, but this group has little similarity with either S. conica or $. jackieburchi. $. conica diverges strongly from all the other species. S. jackieburchi, however, falls clearly within the Partula clus- ter. Indeed, $. jackieburchi has an identity of 0.95 with P. otaheitana and 0.93 with P. affi- п/5. The close similarity of $. jackieburchito P. otaheitana also occurs in sympatry; the single specimen of P. otaheitana rubescens col- lected with the $. jackieburchi in the Tiarei sample is electrophoretically indistinguish- able from it. DISCUSSION The close genetic identities among species of Samoana from the Society Islands are con- sistent with those previously reported for spe- cies of Partula on Moorea (Johnson et al., 1977) and for P. affinis and P. otaheitana in the present study. Although they are by no means unique, such strong resemblances be- tween reproductively isolated species are un- usual (Thorpe, 1982). In the case of Partula, reproductive relationships have been estab- lished through extensive field and laboratory studies (Murray & Clarke, 1980), providing information that is not available for Samoana. Nevertheless, evidence for reproductive iso- lation in Samoana 1$ provided by the distinct- ness of species in sympatry. Our samples of S. diaphana include two sites of sympatry with S. attenuata and one with S. burchi. In each case, S. diaphana has a distinctly less robust, smaller, and more globose shell than does S. attenuata or S. burchi. Similarly, although 102 JOHNSON, MURRAY 8 CLARKE TABLE 2. Allelic frequencies and average observed heterozygosity (H) for 20 loci in species of Samoana and Partula. For loci not included in Table 1, $. diaphana, $. burchi, and $. annectens are identical to $. attenuata. Р. gibba is homozygous for the 1.00 allele at each locus. G6pd and Mdh-3 were monomorphic. Samoana Partula Locus Allele conica attenuta jackieburchi otaheitana affinis Sample size 19 48 33 31 3 Alph 1.02 .04 1.00 — .02 — 95 — — 1.00 96 1.00 87 = = — 02 = 55 .96 — — — — Got-1 1.70 .03 — — 1.50 97 1.00 21 18 17 1525 = = 519 1.00 — = .31 52 83 .80 — = = = — 71 — — .35 28 — 55 — = == 02 — Got-2 —.60 — .02 ‚25 — — — 1.00 — — 75 1.00 1.00 — 1.40 1.00 98 — Idh-1 1.06 1.00 — — — — 1.00 — — .60 .98 50 89 — 1.00 .40 02 50 ldh-2 1.30 .08 — — — — 1.00 .92 117 1.00 1.00 1.00 50 — 83 — Mah-1 т — — — 7 .33 1.00 1.00 — .86 61 67 95 1.00 — — — 81 — — .14 02 — Mah-2 1.50 1.00 — — 02 — 1.00 — — 1.00 98 1.00 80 — .99 — — — 70 — .01 — — — Mdh-4 10 1.00 — — — — 1.00 — 1.00 1.00 1.00 1.00 МР! 1.07 1e — — — — 1.00 87 — .94 1.00 .92 — — .06 18 — 86 — 1.00 — — — Np 1.15 — 1.00 — — — 1.00 1.00 — — — 84 — 1.00 94 1.00 73 — — = 06 Pep-2 1.13 1.00 — — — — 1.05 — 1.00 — — — 1.00 — — 1.00 1.00 1.00 Рер-4 1.03 — 1.00 — — — 1.02 1.00 — — — — 1.00 — — 1.00 .98 — 97 — — — .02 1.00 Рер-6 1.29 1.00 -- =- — — 1.23 — — 225 ALLOZYMES OF SAMOANA 103 Samoana Partula Locus Allele conica attenuta jackieburchi otaheitana affinis .85 — 1.00 — .09 — 6Pgd 1.00 1.00 — 1.00 1.00 1.00 .91 — 1.00 -- — — Ра! 1.42 .05 — — — — 1.00 — — 21 .74 —- .95 .95 1.00 — —- — .50 — — 79 .26 1.00 Pgm-1 Wels} 1.00 1.00 — — — 1.08 — — 03 .09 — 1.04 — — — -- — 1.02 — — .97 91 1.00 Pgm-2 1.09 = 06 = — — 1.04 "37 46 1.00 1.00 1.00 95 .63 — — — — 87 — 48 — — — Sod 1.60 1.00 — — —- — 1.00 — 1.00 1.00 1.00 1.00 H (20 loci) .027 .001 135 130 .067 Kondo (1973) considered the inflated epiphal- lus of his figured specimen of S. diaphana to be “a temporary non-reliable characteristic”, our observations indicate that it does indeed consistently discriminate S. diaphana from the other species. This consistent association of characteristics which are presumably inde- pendent genetically and developmentally т- dicates that S. diaphana is a distinct entity, reproductively isolated from the other spe- cies. As the differences between $. attenuata, S. burchi, and S. annectens are for their shells only, and as these forms have not been found together, inferences on their reproductive iso- lation are much weaker. Even excluding com- parisons between these three species, how- ever, the very high allozymic similarities, de- spite reproductive isolation, are confirmed by the comparisons with S. diaphana. Despite this similarity, the allozymes also provide further evidence for reproductive iso- lation between $. diaphana and $. attenuata. Where these two species occur together in Faatoai and Uufau valleys, they are fixed for alternate alleles at the /dh-2 and Pgm-2 loci (and at the Pgm-1 locus in Uufau) (Table 1). TABLE 3. Genetic identities among species of Samoana and Partula, based on 20 loci. For species represented by more than one population, allelic frequencies in the combined sample were used for interspecific comparisons. Samoana Partula con. dia. att. bur. ann. jac. ota. aff. S. conica — S. diaphana ES) .95 S. attenuata .34 .95 .90 $. burchi .36 .97 .96 — S. annectens Sn .98 .99 .98 — S. jackieburchi .28 .26 .25 .25 .26 — P. otaheitana 2, .26 .24 .24 25 .95 .90 Р. affinis 27 .26 .25 .25 .26 .93 .90 Р. gibba .31 .22 .19 .21 .22 .72 .74 .64 104 JOHNSON, MURRAY 8 CLARKE Genetic Identity 02 04 06 08 LO S conica $ diaphana $ burchi S.annectens S.attenuata $. jackieburchi P otahe/tana Paffinis P g'bba FIG. 4. Phenogram of genetic identities among species of Samoana and Partula, based on 20 loci. The complete absence of heterozygotes at these loci demonstrates a lack of genetic ex- change between the two species. Although such distinctness could also result from the coexistence of self-fertilizing strains (see later discussion), the finding of heterozygotes at some loci indicates that these species repro- duce at least partly through outcrossing. Thus, the pattern of allozyme variation sup- ports the existence of reproductive isolation, which was implied by the study of shells and genitalia. Nevertheless, $. diaphana and $. burchi, despite the reproductive isolation in- ferred from their conchological and anatomi- cal differences are allozymically identical at all 20 loci where they occur together. Thus, al- though allozymes are locally useful in dem- onstrating reproductive isolation, they are not invariably so. Similarly, while genetic identi- ties between species in many groups are usu- ally below 0.85 (Thorpe, 1982), this is not true of either Samoana or Partula, suggesting that a genetic “yardstick” does not exist. Although both genera have groups of spe- cies that closely resemble each other, the low heterozygosity in Samoana contrasts with the high values typical of most species of Partula (Table 2; Johnson et al., 1977). Another dif- ference between the genera is in terms of whether the genetic diversity of a species 1$ concentrated within populations, or repre- sented by differences between populations. Partitioning allozymic diversity into “within- population” and “between-population” com- ponents (Nei, 1973) shows different genetic patterns: in the Society Islands, only 2%, on average, of the intraspecific genic diversity in Samoana (excluding $. jackieburchi) occurs within populations, compared with 79% in Partula (unpublished data). Thus, although a single sample gives a good indication of al- lozymic diversity in the average species of Partula, several samples are required for spe- cies of Samoana. As emphasized by Kondo (1973), species of Samoana are remarkable for their scarcity. In part, this may be an artifact of their relative inaccessibility, since they generally occur at higher altitudes than Partula. Our samples of S. diaphana, for example, increase the re- corded sites for the species from 2 to 6 and suggest that it may be fairly continuously dis- tributed in Moorea at high altitudes. Neverthe- less, the area of high-altitude habitats 1$ rela- tively small, and reduced population sizes may contribute to the low genetic variation in Samoana compared with the more widely dis- tributed species of Partula. Despite their wide local distributions, the species of Partula are, almost without exception, endemic to single islands, whereas both S. diaphana and $. attenuata are found on more than one island. It seems likely that the species of Partula on Moorea represent a radiation within the island (Murray 8 Clarke, 1980; Johnson, Murray 8 Clarke, unpublished). In contrast, the occur- ALLOZYMES OF SAMOANA 105 гепсе of $. diaphana and $. attenuata on both Moorea and Tahiti implies interchanges be- tween islands. Despite their apparent ability to disperse, the fact that alternate alleles are fixed in different populations of both 5. di- aphana and $. attenuata within islands indi- cates that local populations are relatively iso- lated. Another possible reason for the low varia- tion in Samoana is self-fertilization. A compar- ison of heterozygosities in outcrossing and self-fertilizing species of terrestrial pulmon- ates (Selander & Ochman, 1983) shows that heterozygosities as low as those found in Samoana are common in species that self- fertilize. No heterozygotes have been de- tected in Partula gibba, which reproduces largely, if not completely, by self-fertilization, whereas all the Partula species that are known to outcross are highly heterozygous (Johnson et al., 1977; unpublished data). Al- though the low heterozygosities in Samoana may indicate selfing, the paucity of variation makes direct tests difficult. The large deficit of heterozygotes for Pgiin $. diaphana suggests that self-fertilization may be common, but ex- treme subdivision in an outcrossing popula- tion cannot be excluded. If the circumstantial evidence for selfing in Society Island Samoana proves to be correct, it clearly does not apply to the genus as a whole. Although not highly heterozygous, the sample of $. conica from Tutuila is in Hardy-Weinberg equilibrium for the Mpi and Pgm-2 loci. The contrasts with S. conica, in terms of both heterozygosity and genetic differences, emphasize the coherence of the Samoana species from the Society Islands. The close relationships among these species were also stressed by Kondo (1973) in his study of their reproductive anatomy. It was on anatomical grounds that S. jackieburchi was removed from P. otaheitana (Kondo, 1980). The elec- trophoretic data, however, do not support the placement of $. jackieburchiin Samoana. The Partula and Samoana from the Society Is- lands have very different sets of allozymes at 9 of 20 independent loci, and large differences at several others. They form two very distinct groups of species. The close clustering of species of Partula (Fig. 4) applies to all the 24 species that we have examined (unpublished data). The fact that S. jackieburchi most closely resembles the group of Partula spe- cies indicates clearly that it belongs in Partula. This conclusion is supported by other evi- dence. The shells of S. jackieburchi and P. otaheitana rubescens are indistinguishable, and differ from Society Island species of Samoana in their size, shape, robustness, color, and direction of coiling (photographs in Kondo, 1980). S. jackieburchi also lacks the maculation of the mantle that occurs in Samoana, but is uncommon in Partula. Fi- nally, a striking characteristic of S. diaphana, S. attenuata, S. burchi, and S. annectens is their very sticky mucus, a feature missing in Partula, and S. jackieburchi. Faced with the alternative of massive convergence of inde- pendent biochemical and morphological char- acters, Occam's razor requires the interpreta- tion that it is the genital anatomy of S. jackieburchi which is convergent. This neces- sary interpretation has disturbing implications for pulmonate taxonomy, which often relies on genital anatomy. Our data indicate that geni- talia are not infallible taxonomic characters. As in many cases of convergence, finer ana- tomical studies may show that the resem- blance of S. jackieburchi to Samoana is superficial. In any event, that resemblance emphasizes the danger of reliance on any particular taxonomic character. К is possible that the Samoana-like genita- lia of S. jackieburchi do not delineate a sep- arate species. The indistinguishable shells and allozymes of S. jackieburchi and P. otah- eitana rubescens are consistent with an in- terpreation of these entities as a single spe- cies with polymorphic genitalia. Only detailed field and laboratory studies will settle the is- sue. ACKNOWLEDGEMENTS We thank Peter Kendrick and Jane Prince for assistance, and Dr. Yoshio Kondo for con- firming our identifications of species of Samoana. Financial support was provided by the Australian Research Grants Scheme, the U.S.-Australian Cooperative Science Pro- gram (NSF: 2AS = 30), and the Science and Engineering Research Council. LITERATURE CITED CRAMPTON, H. E., 1916, Studies on the variation, distribution, and evolution of the genus Partula. The species inhabiting Tahiti. Carnegie Institu- tion of Washington Publications, 228: 1-311. 106 JOHNSON, MURRAY 8 CLARKE СВАМРТОМ, H. E., 1925, Studies on the variation, distribution, and evolution of the genus Partula. The species of the Mariana Islands, Guam and Saipan. Carnegie Institution of Washington Pub- lications, 228A: 1-116. CRAMPTON, H. E., 1932, Studies of the variation, distribution, and evolution of the genus Partula. The species inhabiting Moorea. Carnegie In- stitution of Washington Publications, 310: 1- 335. CRAMPTON, H. E. & COOKE, C. M., 1953, New species of Partula from southeastern Polynesia. Occasional Papers, Bernice Pauahi Bishop Mu- seum, 21: 135-159. JOHNSON, M. S., CLARKE, B. & MURRAY, J., 1977, Genetic variation and reproductive isola- tion in Partula. Evolution, 31: 116-126. KONDO, Y., 1973, Samoana of the Society Islands (Pulmonata: Partulidae). Malacological Review, 6: 19-33. КОМОО, Y., 1980, Samoana jackieburchi, new species (Gastropoda: Pulmonata: Partulidae). Malacological Review, 13: 25-32. KONDO, Y. 8 BURCH, J. B., 1979, Extrusive gen- ital anatomies and their internal postures in Par- tula affinis of Tahiti. Malacological Review, 12: 79-84. MURRAY, J. & CLARKE, B., 1980, The genus Partula on Moorea: speciation in progress. Pro- ceedings of the Royal Society of London, ser, B, 211: 83-117. NEI, M., 1973, Analysis of gene diversity in subdi- vided populations. Proceedings of the National Academy of Sciences, U.S.A., 70: 3321-3323. NEI, M., 1978, Estimation of average heterozygos- ity and genetic distance from a small number of individuals. Genetics, 89: 583-590. PILSBRY, H. A., 1909-1910, Family Partulidae. In: Manual of Conchology, ser. 2, 20: 155-336. SELANDER, R. K. & OCHMAN, H., 1983, The genetic structure of populations as illustrated by molluscs. Isozymes: Current Topics in Biological and Medical Research, 10: 93-123. SELANDER, В. K., SMITH, М. H., YANG, $5. Y., JOHNSON, W. E. & GENTRY, J. B., 1971, Bio- chemical polymorphism and systematics in the genus Peromyscus. |. Variation in the old-field mouse Peromyscus polionotus. Studies in Ge- netics VI, University of Texas Publication 7103: 49-90. SNEATH, P. H. A. & SOKAL, R. R., 1973, Numer- ical taxonomy; the principles and practise of nu- merical classification. Freeman, San Francisco, ху + 573 p. THORPE, J. P., The molecular clock hypothesis: biochemical evolution, genetic differentiation and systematics. Annual Review of Ecology and Sys- tematics, 13: 139-168. MALACOLOGIA, 1986, 27(1): 107-125 HYBRIDIZATION IN A UNIONID FAUNAL SUTURE ZONE Pieter W. Kat! Department of Earth & Planetary Sciences, The Johns Hopkins University, Baltimore, Maryland 21218, U.S.A. ABSTRACT Retreat of the Wisconsinan glaciers and expansion of unionid geographic ranges has resulted in re-establishment of contact between Interior Basin and northern Atlantic Slope species isolated by the Appalachian Mountains at the height of glaciation. One suture zone between these faunas occurs in the area around Lake Champlain, and molecular genetic, anatomical, and shell microstructural data indicate hybridization between species of Anodonta and Lampsilis. Additionally, introgression appears to occur over a wide geographic area. Elliptio populations around Lake Champlain exhibit no evidence of hybridization, but form a locally differentiated group when compared to northern Atlantic Slope E. complanata. Hybrid Anodonta and Lampsilis populations contain variant alleles not found among parental species. Probability of hybridization is proposed to be best predicted by similarity of glochidial hosts between unionid species, not necessarily by levels of electrophoretically determined genetic differentiation. Taxonomic impli- cations of the data are discussed. Key words: hybridization; zoogeography; genetics; Unionidae; Bivalvia. INTRODUCTION Hybrid zones have long been of special interest to evolutionary biologists. Introgress- ive hybridization (Anderson, 1949) can poten- tially enhance the level of genetic variation and thus the evolutionary flexibility of popula- tions. For instance, Sage 8 Selander (1979) and Hunt 8 Selander (1973) observed in- creased levels of heterozygosity as well as unique alleles in hybrid populations of frogs and mice. Such unique alleles have been proposed to arise through increased mutation rates among hybrid populations (Thompson & Woodruff, 1978), or through intragenic recom- bination between the different parental alleles (Watt, 1972). Also, Anderson & Stebbins (1954) have proposed that hybridization can trigger episodes of innovative diversification, and certain species of plants are known to have had a hybrid origin (Grant, 1966; Lewis, 1966; Gallez & Gottlieb, 1982). Hybrid zones are often established by changes in distribution of one or both of the taxa involved. Geographic ranges of all North American unionids have probably fluctuated to some extent during the repeated Quater- nary glacial episodes, but such fluctuations are best documented for species of the north- ern Interior Basin and Atlantic Slope faunas (Simpson, 1896; Ortmann, 1913; Baker, 1920; Johnson, 1970, 1980; Clarke, 1973; Kat, 1982, 1983a, b, с; Smith, 1982; Kat & Davis, 1984). For example, fossil evidence has indicated that populations of at least three Interior Basin species were present on the Atlantic Slope about 200,000 years ago, but were subsequently eliminated by glaciers (Kat, 1983b). Contact between these faunas has recently been re-established by migration out of Wisconsinan refugia. One such suture zone (Remington, 1968) is located in the area around Lake Champlain. This lake has had a varied postglacial history including a saltwater phase and connections to both Interior Basin and Atlantic Slope drainages (Simpson, 1896; Clarke 8 Berg, 1959; Elson, 1969; Johnson, 1980; Smith, 1982). As a consequence of these historic connections and/or a more re- cent immigration route (via the Erie Canal system, which links Lake Erie, Lake Cham- plain, and the Hudson River), Lake Cham- plain and surrounding drainages contain spe- cies of both faunal regions (Smith, 1983). This area thus provides a natural experiment to determine the degree of genetic interaction between various members of these previously isolated faunas. ‘Present address: Department of Malacology, National Museums of Kenya, Р.О. Box 40658, Nairobi, Kenya. (107) 108 Study of hybridization between Interior Ba- sin and Atlantic Slope unionid species is of special interest fortwo reasons. First, because of patterns of Wisconsinan glaciation and the location of the hybrid zone, there is little uncertainity about its origin. This hybrid zone represents a postglacial secondary contact between taxa that existed in allopatry at the height of glaciation, and therefore can not be explained as differentiation within a continuous series of populations (Endler, 1977; White, 1978). Second, the Lake Champlain fauna allows study of the degree of interaction between species belonging to lineages that are diversifying at different rates, between species that differ in observed levels of heterozygosity and polymorphism, and between species that exhibit various levels of genetic differentiation. Also included in this study is a genetic and morphological analysis of Anodonta “catar- acta” fragilis. This taxon was thought by Clarke & Rick (1963) to represent an intergrade between A. fragilis from Newfoundland and northern Atlantic Slope A. cataracta. Previous studies, however, have indicated that Nova Scotian A. “с.” fragilis are genetically distinct from A. cataracta, and that the taxon is more closely related to European than North Amer- ican anodontines (Kat, 1983d, e; Kat & Davis, 1984). The taxonomic status of anodontines that resemble A. fragilis but occur outside Newfoundland is therefore still uncertain. METHODS Populations analysed in this study were collected from Lake Champlain and adjacent areas, as well as the Delmarva Peninsula, Nova Scotia, Michigan, and Wisconsin. The bivalves were maintained in aquaria for at least two weeks. Four individuals from each population were then relaxed (Sodium nembu- tal) and fixed (10% formalin) in preparation for dissection. The remaining 20-25 individuals had wedges of tissue removed from the foot and viscera and these tissue samples were either homogenized and electrophoresed im- mediately or stored at -20° С for later analysis. Starch-gel electrophoresis has been used with good success in a series of taxonomic analyses of the Unionidae (Davis et al., 1981; Davis, 1983, 1984; Kat, 1983a, с, а; Kat & Davis, 1984). Methods for electrophoresis and enzyme staining were generally similar to those of Davis et al. (1981), and 15 loci of which at least eight were polymorphic among species were scored using the methods of Ayala et al. (1973). Nei’s (1972) genetic dis- tances were computer generated using a pro- gram written by Green (1979). This genetic distance matrix was then used in the mult- ivariate analysis program NT-SYS (Rohlf et al., 1972). Multidimensional scaling maxi- mized goodness-of-fit of the regression of ge- netic distance and distance in three-dimen- sional space, and a minimum spanning tree was derived from these adjusted distances. The minimum spanning tree summarizes tax- onomic relationships since distances between closely related taxonomic units are small, whereas those between distantly related tax- onomic units are large. Distance between tax- onomic units is here defined as a function of the observed Nei distances. Such multivariate analyses are especially useful in elucidation of relationships among taxa such as Elliptio that exhibit considerable polymorphism at a number of loci. Fine detail of unionid stomach anatomy can be used as a taxonomic tool at a variety of taxonomic levels (Kat, 1983a, c, d). Tech- niques for dissection and illustration are dis- cussed in Kat (1983d). Four individuals from each population were dissected and photo- graphed to determine levels of intrapopulation variability. Microstructure of conchiolin layers within the shell was examined with a scanning elec- tron microscope. Previous studies (Kat, 1983a, e) reveal that the conchiolin layer is divisible into three distinct regions, of which the central, reticulate region in particular con- tains species-specific characters. In the past, patterns of resemblance among unionid taxa based on conchiolin layer microstructure have been highly compatible with patterns of re- semblance suggested by electrophoretic data, and conchiolin layer microstructure was successfully used to discriminate among two races of Elliptio complanata and their hybrids (Kat, 1983a). Techniques for conchiolin layer preparation and microscopy are detailed in Kat (1983е). RESULTS A. Molecular genetics Distributions of alleles among loci that best discriminate species of Anodonta and Lamps- UNIONID HYBRIDIZATION 109 TABLE 1. Distribution of alleles among loci of Anodonta examined in this study. Loci that do not discriminate among species are not included. Enzyme Allele A. cataracta PGM | 24 1.00 РСМ Il 32 30 1.00 LAP 34 30 1.00 MDH | 18 15 1.00 MDH II = 8 АМ 1.00 НЕХ 34 31 1.00 МР! 26 20 .33 to .48 18 .52 to .67 ODH 15 6 1.00 ilis are presented in Tables 1 and 2. Similar data are not presented for populations of El- liptio because Elliptio dilatata from Wisconsin appears to possess only two relatively rare alleles not present among Atlantic Slope E. complanata examined to date: MDH | 14 and HEX 34. Also, the number of populations ex- amined and the high levels of polymorphism characteristic of Elliptio would require a table of excessive proportions. It is clear from Table 1 that the Anodonta population in Lake Champlain shares alleles characteristic of both Atlantic Slope A. catar- acta and Interior Basin A. grandis. However, the Lake Champlain population also pos- sesses alleles not present in either parental species (НЕХ 28 and PGM II 32), and is fixed Species A. grandis x cataracta A. grandis A. fragilis .26 .05 .74 .95 1.00 .29 1.00 fal 1.00 35 52 65 48 1.00 .50 1.00 1.00 .50 .35 :55 .65 .45 1.00 .20 .20 .72 .80 1.00 .08 1.00 65 O to .15 .35 .85 to 1.00 .61 T2 1.00 .39 .28 for MPI 26. Over 15 loci examined, A. catar- acta possesses 17 alleles, A. grandis 22, and A. cataracta x grandis 24: A. grandis shares 13 alleles with A. cataracta and 18 alleles with A. cataracta x grandis, while A. cataracta shares 14 alleles with A. cataracta x grandis. A. “cataracta” fragilis, however, is quite dif- ferent from both A. cataracta and A. grandis. А. “с.” fragilis has 18 alleles at the loci exam- ined of which it shares 10 with A. cataracta and A. grandis. There is no evidence from electrophoresis to suggest any genetic ex- change between A. cataracta and A. fragilis. Table 2 presents allele frequencies for nine loci at which Atlantic Slope Lampsilis radiata differ from Interior Basin L. siliquoidea. Again, there is good evidence to support a hybrid 110 KAT TABLE 2. Distribution of alleles among loci of Lampsilis examined in this study. Loci that do not discriminate among the parental species are not included. Species L. radiata x Enzyme Allele L. radiata siliquoidea L. siliquoidea СР! 16 1.00 .90 45 10 10 155 PGM | 18 0 to .34 .20 16 .62 to 1.00 ‚92 .80 14 .08 12 .04 to .20 PGM Il 30 .05 .45 28 .85 to 1.00 295 95 26 .05 to .15 LAP 34 .03 to .47 ‚83 .35 32 .53 to 1.00 .50 .45 30 .04 to .30 28 10 .20 МР! 24 1.00 1.00 .70 22 .30 6PGD 6 0 to .10 4 .70 to 1.00 .40 .45 2 .04 to .30 .60 150 G3PDH 11 O to .13 9 0 to .60 .40 7. .40 to 1.00 .60 1.00 GPDH 32 1.00 :95 .70 30 .05 .30 SOD Il =" ff 1.00 70 .08 = 2 .30 ¿92 origin of the Lampsilis population in Lake Champlain, and this population also pos- sesses an allele not present in either parental species (РСМ | 14). L. radiata possesses а total of 26 alleles over the 15 loci examined, L. siliquoidea 25, and L. radiata x siliquoidea 26. L. radiata shares 19 alleles with L. sil- iquoidea, and 20 alleles with L. radiata x sil- iquoidea, while L. siliquoidea shares 24 al- leles with the hybrid population. In contrast to these examples of hybridiza- tion between species of Anodonta and Lam- psilis, the population of Elliptio in Lake Cham- plain presents no evidence that it is of hybrid origin between Atlantic Slope E. complanata and Interior Basin E. dilatata. Rather, this population exhibits affinities to regional popu- lations of E. complanatain Vermontand Maine. Nei’s (1972) genetic distances and similar- ities between all pairs of Anodonta, Lampsilis, and Elliptio populations examined are pre- sented in Tables 3, 4, and 5, respectively. Table 3 indicates that A. “cataracta” fragilis 1$ genetically almost invariant from Nova Scotia through Maine and Vermont, and 15 distantly related to A. cataracta and A. grandis. A. cataracta and A. grandis are genetically sim- ilar at a level of 0.649 + .012, which is com- parable to levels of similarity among other species in the cataracta clade such as A. gibbosa from Georgia (Kat, 1983d). A. catar- acta x grandis exhibits intermediate levels of UNIONID HYBRIDIZATION 111 TABLE 3. Genetic distances (above the diagonal) and similarities (below the diagonal between all pairs of Anodonta populations examined in this study. See Appendix for locations of the collection sites. Population Species ME1 VT3 №54 NS6 NS2 NJ2 DE2 NJ1 vr MI ME1 — .001 .001 .001 .003 .544 .561 .568 .622 585 A. fragilis VT3 999 — .001 .001 .002 .542 559 566 72620) 7 583 A. fragilis NS4 9999) 7.999 — .001 .003 .541 .558 .564 .616 .538 A. fragilis NS6 ¿999 .999 .999 — .003 .541 ‚558 .564 .616 .538 А. fragilis №2 998’ 1998 999 5999 542 7.5570 56371609 .581 А. fragilis №2 .581 .582 .582 .582 .582 — .003 .003 .222 .453 A. cataracta DE2 SA .572 .572 .572 15713 .999 — .001 .224 .420 А. cataracta NJ1 .567 .568 .569 .569 .570 .998 .999 — .225 .423 А. cataracta VT1 2537. .538 .540 .540 .544 .801 .800 .799 — .124 hybrid MI 155% 558 584 584 560 .636 .657 .655 .883 — A. grandis TABLE 4. Genetic distances (above the diagonal) and similarities (below the diagonal) between all pairs of Lampsilis populations examined. See Appendix for locations of collection sites. Population Species №57 NS4 №1 МВ №53 VT3 DES МОТ ME4 DEI VT1 М №57 — .014 .004 .009 .021 .022 .020 .023 .029 .051 .056 .209 L. radiata №54 .986 — .011 .012 .033 .025 .024 .019 .031 .047 .052 .199 L. radiata №51 999’ .984 — 011 .025 .026 .022 .027 .033 .059 .064 .212 E. radiata МВ 991m 988) 2989) №1014. 01s Onl ОДО 029 047218 Era aleta NSS 979) 968 976. OC 0003: 20165 014 208i) 052229) L. radiata МТЗ 978 975) 2974.) 7982") 2983; —— 1 1019 016) 201167 027 027-2040 radiata РЕЗ +980’ 977 978 989.996. 2981. — 01 20110) 1026 2043 218 и. таагаа MD1 .977 .981 .974 .989 .984 .984 .989 — .003 .012 .038 .208 L. radiata MD4 .971 .970 .968 .983 .986 .984 .990 .995 — .013 .037 .203 L. radiata DE1 LEO ER AS AA AO UE O14 988 IE — 029 .220 L. radiata VT1 LEO 4.938 7.9545 NAS 2.9587 963, OA A hybrid MI .811 .820 .809 .804 .794 .804 .812 .816 .802 .815 .881 — LL. siliquoidea TABLE 5. Genetic distances (above the diagonal) and similarities (below the diagonal) between all pairs of Elliptio populations sampled in this study. See Appendix for locations of the collection sites. Population Species VT1 VT3 МЕТ VT2 MD4 MD2 MD3 РА МЕ? DE3 MD1 NJ2 NS8 №7 NS5 WI VT1 — .026 .046 .066 .041 .076 .059 .072 .052 .068 .041 .045 .052 .035 .033 .087 Е. complanata VT3 .976 — .015 .072 .016 .059 .041 .089 .016 .057 .017 .012 .016 .015 .033 .044 Е. complanata ME1 .955 .985 — .069 .016 .060 .048 .099 .025 .059 .023 .012 .024 .016 .040 .044 E. complanata VT2 936 .931 .932 — .063 .088 .093 .078 .098 .101 .076 .076 .087 .063 .097 .124 Е. complanata MD4 .960 .984 .984 .939 — .027 .023 .107 .035 .045 .022 .018 .028 .023 .053 .054 E. complanata MD2 .927 .943 .942 .916 .973 — .013 .144 .085 .035 .055 .057 .064 .064 .099 .072 Е. complanata MD3 .943 .960 .954 .911 .977 .987 — .129 .064 .027 .040 .042 .044 .049 .073 .070 E. complanata РА 931 .915 .905 .925 .898 866 .879 — .089 .134 .089 .120 .126 .089 .094 .157 E. complanata МЕ? .949 .984 .975 .907 .965 .919 .938 .915 — .067 .018 .019 .017 .020 .035 .053 Е. complanata DE3 .934 .944 .943 .904 .956 .966 .974 .874 935 — .043 .052 .052 .061 .086 .070 Е. complanata MD1 .960 .983 .977 .927 .978 .946 .960 .915 .982 .958 — .017 .019 .014 .035 .059 E. complanata №2 .956 .989 .988 .927 .982 .944 .959 .887 .981 .949 .983 — .007 .015 .037 .051 E. complanata №58 .949 .984 .976 .917 .972 .938 .957 .882 .983 .949 .982 .993 — .022 .037 .051 E. complanata №57 .966 .985 .984 .939 .977 .938 .953 .915 .980 .941 .987 .985 .979 — 014 046 E. complanata NS5 .968 .967 .961 .907 .948 .906 .930 .910 .966 .917 .966 .964 .963 .986 — .048 Е. complanata WI .91722:9572295727.,8837:9487.930779327.8557.9487793277.9437.9512:9507 :959) 7953) —= E. dilatata 112 similarity to both A. grandis (0.883) and A. cataracta (0.800). Populations of Lampsilis radiata from the Delmarva Peninsula to Nova Scotia exhibit an average level of interpopulation similarity of 0.979 + .012, and L. radiata and L. sil- iquoidea resemble each other at a level of 0.808 + .007 (Table 4). This degree of re- semblance is comparable to that observed among other species ofthe radiata clade such as southern Atlantic Slope L. splendida (Kat, 1983c). L. radiata x siliquoidea from Lake Champlain resembles L. radiata at a level of 0.954 + .010 and L. siliquoidea at a level of 0.881: the higher degree of resemblance to L. radiata reflects greater similarity in the fre- quencies of shared alleles. Interestingly, Clarke 8 Berg (1959) also classified the Lake Champlain Lampsilis population as more radiata-like than siliquoidea-like based on conchological characters. Populations of Elliptio complanata from the Delmarva Peninsula to Nova Scotia, Maine, and Vermont exhibit characteristically high levels of variability in genetic resemblance among populations, ranging from 0.993 to 0.866, with an average degree of resem- blance of 0.950 + .030 (Table 5). E. com- planata resembles E. dilatata from Wisconsin at a level of 0.934 + .028. This high level of resemblance among species within diversify- ing Elliptio clades is common (see Davis et al. 1981; Davis, 1984). Table 5 indicates, how- ever, that the Lake Champlain Elliptio popu- lation is most closely related to populations of E. complanata from Vermont, Maine, and Nova Scotia, and in fact exhibits less affinity with E. dilatata than other northeastern pop- ulations of E. complanata. Minimum spanning trees based on genetic distances and connecting all populations of Elliptio, Lampsilis, and Anodonta are illus- trated in Fig. 1. The distance measure be- tween populations is a function (variable over each analysis) of the Nei genetic distances, and thus corresponds to taxonomic related- ness. Such distances are small, for example, among populations of A. cataracta and A. “cataracta” fragilis, but considerable between these species. A. cataracta + grandis 1$ shown to be almost equidistant between A. cataracta and A. grandis, while L. radiata x siliquoidea clusters considerably closer to L. radiata than L. siliquoidea. The minimum spanning tree between Elliptio populations generally connects geographically neighbor- KAT ing populations. Divergent populations within this group are those from Joes Pond, Ver- mont, and the Susquehanna River, Pennsyl- vania, both due to high frequencies of other- wise rare alleles at loci such as ГАР and MPI. Table 6 contains levels of observed he- terozygosity and polymorphism for all popu- lations of Elliptio, Lampsilis, and Anodonta examined. Populations of Elliptio exhibit char- acteristically high levels of H and P (average Н = 0.139 + .014; average P = 0.517 + .054) except among peripheral populations in Nova Scotia (see Kat 8 Davis, 1984). He- terozygosity and polymorphism among popu- lations of Lampsilis are characteristically lower than those observed among Elliptio, except in the case of L. siliquoidea from Mich- igan, which possesses the highest level of H and P thus far observed for any lampsiline population (see Kat, 1983c). L. radiata x siliquoidea from Lake Champlain is not more heterozygous than either parent (average H for central range populations of L. radiata = 0.058 + .004), but exhibits a level of polymor- phism equal to that of L. siliquoidea. Levels of heterozygosity for anodontine populations presented here are higher than those pub- lished earlier (Kat, 1983d) due to inclusion of loci with fixed heterozygosities (GPI for all species and MDH | for A. “с.” fragilis). A. cataracta x grandis from Lake Champlain 1$ considerably more heterozygous than either parent species. B. Stomach anatomy Fine detail of stomach anatomy can be used to discriminate between hybrids and pa- rental species of the lampsilines and anodon- tines examined in this study. The Elliptio pop- ulation in Lake Champlain (Fig. 2) is very similar in stomach anatomy to E. complanata from eastern Canada (see Kat, 1983a), as well as E. dilatata from Wisconsin (not fig- ured). Stomach anatomy of L. radiata x sil- iquoidea from Lake Champlain (Fig. 3) is quite similar to that of L. radiata from eastern Can- ada (see Kat, 1983c) and the Delmarva Pen- insula (Fig. 4), and also that of L. siliquoidea from Michigan (Fig. 3A). Differences are ap- parent, however, in the curvature of the minor typhlosole fold. A. cataracta x grandis in Lake Champlain (Fig. 5) also differs from A. cataracta from Virginia and A. grandis from Tennessee and Michigan in details of the mi- nor typhlosole fold. This fold is gently rounded UNIONID HYBRIDIZATION 113 .041 1.977 VT3 Nm em М! < .023 .009 .019 .016 .016 ‚017 DEI MD4 NB NS3 №57 №51 №54 о .010 A. MD? DES Lampsilis 102 124 VT2 VT1 PA a .028 .032 .013 .030 .028 .028 1073 NS5 NS7 NS8 NJ2 МЕ? ——\ТЗ MEI М o = .032 MD4 MD1 o ES B. 033 .022 + DE3 MD3 MD2 ЕО NS4 NJ2 o o Ww o NS = 059 1.092 268 261 NS6 NS? ——— == Nu 1 УТ1 MI o o o Y VT3 DE2 o a С. MEI Anodonta FIG. 1. Minimum spanning trees for all populations of Lampsilis, Elliptio, and Anodonta. Among the lampsilines, all populations except MI (L. siliquoidea) and VT1 (hybrid) are L. radiata. Among Elliptio, all populations except WI (E. dilatata) are E. complanata, although populations from Joes Pond, Vermont (VT2) and the Susquehanna River, Pennsylvania (PA) are relatively distant from other E. complanata populations. Among the anodontines, NS4, NS6, NS2, VT3, and ME1 are populations of A. fragilis, NJ2, NJ1, and DE2 are populations of A. cataracta, VT1 is a hybrid, and MI represents A. grandis. The distance measure between populations is a function of the Nei genetic distance 114 TABLE 6. Levels of heterozygosity (H) and polymorphism (P) for all species included in this study. H E Elliptio complanata 022: 1037, 0.499 + .066 Elliptio dilatata 0.104 0.428 Lampsilis radiata 0.038 = :023 0.305 = 182 Lampsilis siliquoidea OAS 0.600 L. radiata x siliquoidea 0.053 0.600 Anodonta cataracta 0.098 + .003 0.142 Anodonta grandis 0.192 0.500 A. grandis x cataracta 0.256 0.570 Anodonta fragilis 0.148 + .006 0.185 = .039 in A. cataracta (see Kat, 1983d) but becomes more angular in A. cataracta x grandis (Fig. 5, 5A), and is V- or U-shaped in A. grandis (Fig. 5B, 5C). Overall stomach anatomy among anodontines of the cataracta group (А. cataracta, A. grandis, A. gibbosa) is quite similar (Kat, 1983d). A. “cataracta” fragilis, however, differs strongly from A. cataracta т stomach anatomy (Fig. 6, 6A) and there is no evidence from this character to suggest that A. cataracta and A. fragilis hybridize either in Nova Scotia or in New England. C. Conchiolin layer microstructure Conchiolin layers among unionids are com- posed of three parts: an upper, homogeneous region; a central, reticulate region that con- sists of a number of thin, usually vertically arranged lamellae that form chambers of var- ious shapes and dimensions; and a lower, very thin homogeneous region (Kat, 1983e). Microstructure of conchiolin layers can dis- criminate among parental species and Lake Champlain hybrids of Anodonta and Lamps- FIG. 2. Stomach floor of Elliptio complanata from Lake Champlain. Abbreviations: cm - conical mound, oes - oesophagus, р - sorting pouch, sai - sorting area 1, sa2 - sorting area 2, sp - sorting platform, ss & mg - style sac and midgut, t - major typhlosole, tm - minor typhlosole fold. Scale bar = 2 mm. UNIONID HYBRIDIZATION 115 FIG. 3. Stomach floor of Lampsilis radiata x siliquoidea from Lake Champlain; inset А depicts the minor typhlosole of L. siliquoidea from Michigan. Scale bar ilis, and suggests that Lake Champlain and surrounding areas are inhabited by a distinct subgroup of Elliptio complanata. E. complan- ata from Virginia and the southern Delmarva Peninsula are characterized by reticulate re- gions with short, widely-spaced lamellae that enclose triangular or rectangular chambers of various sizes (Kat, 1983a; Pl. 1:1). E. com- planata on the northern Atlantic Slope pos- sesses longer, straighter lamellae that en- close rather elongate, narrow chambers (Kat, 1983a; Pl. 1:2). E. dilatata from western North Carolina (Pl. 1:3), Wisconsin (Pl. 2:2), and western Ontario (Pl. 2:1) are characterized by a thin upper homogeneous region and long, vertical lamellae. E. complanata from Ver- 2 тт; structures as in Fig. 2. mont (Pl. 1:4 and 1:5) and Lake Champlain (Pl. 1:6) all possess highly characteristic curved and striated lamellae that enclose vari- ably shaped chambers. This particular con- chiolin layer microstructure has not been ob- served in any other region of the geographic range of E. complanata, although similarly striated lamellae occur in a hybrid zone be- tween races of E. complanata on the Del- marva Peninsula (Kat, 1983a). Conchiolin layer microstructure of Anodonta cataracta (Pl. 2:4) consists of a thin upper homogeneous layer underlain by a poorly defined reticulate region composed of small, irregular chambers. A. grandis pos- sesses a better defined reticulate region 116 KAT FIG. 4. Stomach floor of Lampsilis radiata from the Delmarva Peninsula (Andover Branch). Scale bar = 2 mm. FIG. 5. Stomach floor of Anodonta cataracta x grandis from lake Champlain; inset A represents an extreme variant in the same population; inset В depicts the minor typhlosole fold of A. grandis тот Tennessee; inset C shows the minor typhlosole fold of A. grandis from Michigan. Scale bar = 2 mm. UNIONID HYBRIDIZATION 117. FIG. 6. Stomach floor of Anodonta fragilis from Maine (Lake St. George). Inset shows the minor typhlosole fold of A. fragilis from Nova Scotia (Placide Lake). Scale bar = 2 mm. (Pl. 2:5), and A. cataracta x grandis pos- sesses a reticulate region characterized by thick, roughly vertical lamellae that enclose variably sized and shaped chambers (PI. 2:6). A. “cataracta” fragilis (Pl. 2:3) has a con- chiolin layer microstructure very different from that of A. cataracta. The conchiolin layer of Lampsilis radiata (PI. 3:1, 3:2, 3:3, and 3:4) is characterized by a thick upper homogeneous region and a re- ticulate region composed of poorly defined, digitiform to blocky lamellae. L. siliquoidea (PI. 3:6) also possesses a thick homogeneous region but has a reticulate region composed of densely packed, jagged lamellae. L. radiata x siliquoidea (Pl. 3:5) has a rather disorga- nized reticulate region composed of irregular, blocky lamellae. DISCUSSION Studies dealing with genetics of hybrid zones and dynamics of hybridization are nu- merous. Some of these studies indicate con- siderable genetic exchange within the hybrid zone and some, possibly asymmetrical, intro- gression elsewhere (e.g. Hunt & Selander, 1973; Avise & Smith, 1974; Patton et al. 1979; Moran et al., 1980; Hafner, 1982). Other studies report hybridization without any or much introgression beyond the often nar- row hybrid zone (e.g. Nevo & Bar-El, 1976; McDonnell et al., 1978; Sage 8 Selander, 1979; Barton et al., 1983). While hybridization between Anodonta grandis and A. cataracta and Lampsilis radiata and L. siliquoidea 1$ documented here, too few Interior Basin lo- calities in particular were examined to be able to determine the extent of introgression. There is some evidence, however, that intro- gression takes place over a wide geographic area: Atlantic Slope L. radiata and A. catar- acta exhibit a much higher frequency of fixed alleles among the 15 loci examined than do A. grandis and L. siliquoidea from Michigan, which both possess many “Atlantic Slope” alleles in low frequencies. Also, two of the loci KAT 118 "ur ор = sieq ajeos ‘меашецо aye7 Wo. ejeue/dwoo ‘3 “9 чиоциел шоц веивашоо y 'S y :BJ8J8/1p 3 “€ ‘вртив/аАшоо ‘3 ado|s suey изэциом ‘с ‘веивашоэ ‘5 ado|s 9nue|py UJSYINOS *| ‘ода! JO 9IN]INAISOLD1A J9ÁR| UONYDUOD “| 3 1vld = cr + +. 19 ‘WHOL, = sieq ajeog ‘иедшецо aye7 шоц sipuesb x EJOBJeJe9 у ‘9 :ueßlysiyy Woy SIPDUBIÓ “y ‘с -BIlUB A ÁSUUO Y LOI} воелеео “y ‘+ -21J095 влом WON 51/1624 “y “€ ¡OLIEJUO WOY вер ‘3 ‘г ‘UISUOOSIM шо вер "3 *| ‘вшороиу pue оца!/5 JO ainjonsysosdiwW Jake] UIJOIYOUOO “2 31vld UNIONID HYBRIDIZATION KAT 120 ‘uno = sieq ajeos “uebiyoy шод Bapionbijis ‘7 `9 ‘иеашецо 2427 шоц варюпЫи$ x вере! ‘7 ‘6 чиоциэл шод вере 7 y 2409$ enoN шо} вреире/ 7 “€ ‘PUEJUEW шо вере) ‘7 ‘с ‘2409$ BAON WON} BJeIpe4 ‘7 ‘| “SIISALUET JO 21NJON1SO19ILU 1эАе| UI|OIUDUDD “€ 31 VW ld = | TAO x ie. ›— UNIONID HYBRIDIZATION 121 surveyed possess soluble as well as mito- chondrial forms: MDH and SOD (Davidson & Cortner, 1967; Beckman, 1973; Harris & Hopkinson, 1978). Since mitochondrial genes are exclusively inherited through maternal lines, alleles at these loci can be used to estimate relative parental contributions to hy- brids and thus levels of introgression. Lake Champlain populations of Anodonta and Lam- psilis contain 65% cataracta and 70% radiata alleles at these loci, for example, while Mich- igan populations contain 45% cataracta and 8% radiata alleles. If incidence of introgres- sion is confirmed by genetic examination of “pure” populations of A. grandis and Lamps- ilis siliquoidea south of the maximal glacial advance, it would suggest that hybrids be- tween these species are not inferior to their parents in terms of maladaptation to external environments and/or disruption of balanced gene complexes (Sage & Selander, 1979: Moore, 1977; see below). The population of Elliptio in Lake Champlain exhibits no evi- dence of a hybrid origin, and is genetically and morphologically related to populations of At- lantic Slope E. complanata. Conchiolin layer microstructure of populations of E. complan- ata from Lake Champlain and Vermont is dif- ferent from that of other northern Atlantic Slope populations, but such divisions of E. complanata into locally differentiated popula- tions have been observed before and are apparently characteristic of this species (Kat, 1983a; Kat & Davis, 1984). If E. complanata and E. dilatata hybridize, it probably occurs farther west than Lake Champlain, and the location of the hybrid zone could well reflect differences in routes of recolonization taken by Elliptio when compared with Lampsilis and Anodonta. For example, Interior Basin Anodonta and Lampsilis apparently followed retreating glaciers closely: subfossil A. gran- dis and L. siliquoidea occur in Lake Algonquin (12,000 to 10,000 B.P.) and Transitional (10,000 to 6,000 B.P.) sediments, respec- tively, in southwestern Ontario (Miller et al., 1979), and subfossil Lampsilis (species unde- termined) occur in 7,000 year old sediments around Lake Champlain (Elson, 1969). E. complanata, according to Clarke & Berg (1959), occurs westward to the upper great lakes (Ontario, Huron, Superior), and proba- bly colonized this area via Lake Newbery, which inundated the present Finger Lakes basin in New York and drained into the Susquehanna River of the Atlantic Slope. Clarke & Berg (1959) also state that E. dila- tata co-occurs with E. complanata in the St. Lawrence River drainage, but geographic ranges of these phenotypically variable spe- cies need to be confirmed with more reliable taxonomic methods than the conchological characters of previous authors. Evolutionary biologists disagree on the sta- bility of hybrid zones through time. Remington (1968) has argued, for example, that suture zones are ephemeral, leading either to fusion of parental gene pools or separation with per- fection of reproductive isolating mechanisms. In contrast, others have indicated that zones of hybridization can be temporally stable and of ancient age (Short, 1972; Hunt & Selander, 1973; Sage & Selander, 1979). There can be little doubt that the hybrid zone between Inte- rior Basin and Atlantic Slope Anodonta and Lampsilis is of postglacial origin, and the po- sition of the hybrid zone in a repeatedly gla- ciated area implies that it will last only as long as the present interglacial stage. Evidence that a previous suture zone between these faunas was disrupted by glaciers is provided by the Fish House fossil assemblage near Camden, New Jersey (Kat, 1983b). Interest- ingly, this fauna contains no morphologic in- termediates between sympatric A. cataracta and A. grandis to suggest that those species were then hybridizing. This lack of genetic interaction was proposed to have resulted from perfection of isolating mechanisms in the zone of sympatry. These adaptations were subsequently lost as sympatric populations were eliminated by glaciers and the geo- graphic range of A. grandis restricted to ref- uges west of the Appalachian mountains. Re- establishment of contact between these spe- cies would thus again, at least initially, involve hybridization (Kat, 1983b). Observations that alleles that are either rare or absent in parental populations occur in appreciable frequencies in hybrid zones are not uncommon. This phenomenon was first described from a land snail hybrid zone (Clarke, 1968) and since then, Hunt & Selan- der (1973) have found variant esterase alleles in a hybrid zone between semispecies of house mice, and Sage & Selander (1979) described unusual alleles at five of ten loci among hybrid frog populations. Other exam- ples of rare alleles occurring within hybrid zones have been documented by Woodruff (1981) and Barton et al. (1983). In this study, the hybrid Anodonta population possesses 122 three variant alleles, while that of Lampsilis contains one. Hunt 8 Selander (1973) pro- posed that introgression modifies parental gene pools to relax incorporation against new alleles, and Stebbins (1971) suggested that minor alleles could be favored in low frequen- cies in the new genetic environment created by hybridization. More recently, Thompson 4 Woodruff (1978) and Woodruff et al. (1982) suggested that these new alleles might result from increased mutation rates among hybrids (as a consequence of heterozygosities involv- ing dissimilar alleles), and Watt (1972) pro- posed that such new alleles could result from intragenic recombination between parental al- leles. Such intragenic recombination has been proposed to explain patterns of allelic variability by various authors (Ohno et al., 1969; Koehn & Eanes, 1976; Morgan & Strobeck, 1979; Tsuno, 1981), and Golding & Strobeck (1983) showed mathematically that sympatry of two previously isolated popula- tions can increase the effective number of alleles maintained in the hybrid as well as parental populations. Whether these novel al- leles do actually spread from hybrid zones to contribute to allelic diversity of parental pop- ulations has not been documented. In this study, increased levels of heterozygosity could be proposed to account for the appear- ance of variant alleles in the Lake Champlain population of Anodonta but not for that of Lampsilis, since levels of heterozygosity of the hybrid lampsiline population are compa- rable to, or lower than, those of its parent species. Anodonta cataracta and A. grandis, and to a lesser extent Lampsilis radiata and Lamps- ilis siliquoidea, are genetically divergent spe- cies that apparently hybridize readily where their geographic ranges come into contact. The large and fertile (demibranchs of both hybrids were filled with glochidia at the time of collection) populations in Lake Champlain and elsewhere in the hybrid zone would ap- pear to indicate that there is little selection against hybridization, although relative levels of fitness of hybrids and parents are not known. Unionids seem susceptible to acci- dental hybridization because of their external mode of fertilization, but could theoretically experience penalties for hybridization be- cause of their complicated life cycle that in- cludes an obligate parasitic stage. Parasitism involves genes controlling host recognition as well as genes involved with glochidial survival while on the host, and such gene complexes would appear to be highly species-specific (Bush, 1975a, b; Kat, 1984). Hybridization will likely disrupt adapted gene complexes, ex- cept possibly among unionid species pos- sessing very generalized survival and host recognition genes. Such species are perhaps epitomized by A. grandis, which parasitizes over 30 fish hosts (Trdan & Hoeh, 1982), and to a lesser extent by L. siliquoidea, which parasitizes about 12 hosts (Trdan, 1981). Species like E. complanata and A. implicata, however, are only known to parasitize two and one hosts, respectively. Hybridization should be most strongly selected against among these more specialized unionids, unless both parental species parasitize the same host, an occurrence likely in the case of two E. com- planata races that hybridize on the Delmarva Peninsula (Kat, 1983a). Similarity of hosts should therefore be a more reliable predictor of hybridization between related unionid spe- cies than levels of electrophoretically de- tected differentiation. The data gathered in this study also have bearing on two taxonomic questions. First, Clarke & Berg (1959) suggested that in view of the width of the hybrid zone, Lampsilis siliquoidea should be reduced to the subspe- cies L. radiata siliquoidea. | disagree with this interpretation: L. siliquoidea from Michigan (which might constitute a periphery of the zone of introgression) already exhibits the same level of genetic divergence from L. radiata as do other recognized lampsiline species such as L. splendida from Georgia and Lampsilis sp. from Lake Waccamaw, North Carolina (Kat, 1983c). Also, stomach anatomy and conchiolin layer microstructure of these taxa are quite different. | suggest, therefore, that the taxon L. radiata siliquoidea be reserved to describe hybrid populations such as that in Lake Champlain, and that L. siliquoidea be used to describe Interior Basin populations outside the hybrid zone. Second, Clarke & Rick (1963) named Anodonta catar- acta fragilis to describe phenotypic (umbonal sculpture) intergrades between A. fragilis from Newfoundland and A. cataracta. In this study and others (Kat, 1983d, e; Kat & Davis, 1984) genetic, anatomical, and conchiolin layer microstructural data suggest that all populations of A. “cataracta” fragilis are very distinct from A. cataracta, and that there is no evidence to suggest any hybridization be- tween these taxa. | propose that, unless an UNIONID HYBRIDIZATION 123 analysis of A. fragilis (based on a diversity of data) from Newfoundland (the type locality) reveals substantial differences from A. “catar- acta” fragilis from Nova Scotia and northern New England, these taxa be considered syn- onymous and distinct from A. cataracta. ACKNOWLEDGMENTS Discussions with George Davis, Karl Kauf- man, Gene Meyer, Blaire Van Valkenburgh, and Bob Wayne improved versions of the manuscript. | am indebted to Richard Trdan for collecting specimens from Wiggins Lake in Michigan, without which this paper could not have been written. Bill Manning greatly facil- itated collecting in Lakes Memphremagog and Champlain. This study was supported, in part, by the National Science Foundation (DEB 78-01550 and DEB 81-18963), The Jes- sup Fund of the Academy of Natural Sci- ences, and a National Capital Shell Club scholarship. REFERENCES CITED ANDERSON, E., 1949, Introgressive hybridization. Wiley, New York, ix + 109 p. ANDERSON, E. 8 STEBBINS, G. L., 1954, Hybrid- ization as an evolutionary stimulus. Evolution, 8: 378-388. AVISE, J. С. & SMITH, М. H., 1974, Biochemical genetics of sunfish. |. Geographic variation and subspecific intergradation in the bluegill, Le- pomis macrochirus. Evolution, 28: 42-56. AYALA, F. J., HEDGECOCK, D., ZUMWALT, G. S. 8 VALENTINE, J. W., 1973, Genetic variation in Tridacna maxima, an ecological analog of some unsuccessful evolutionary lineages. Evolution, 27: 177-191. BAKER, F. C., 1920, The life of the Pleistocene or glacial period. . . . University of Illinois Bulletin 17(41): xiv + 476 p. BARTON, N. H., HALLIDAY, R. B. 8 HEWITT, G. M., 1983, Rare electrophoretic variants in a hy- brid zone. Heredity, 50: 139-146. BECKMAN, G., 1973, Population studies in north- ern Sweden. IV. Polymorphism of superoxide dismutase. Hereditas, 73: 305-313. BUSH, G. L., 1975a, Modes of animal speciation. Annual Review of Ecology and Systematics, 6: 339-364. BUSH, G. L., 1975b, Sympatric speciation in phytophagous parasitic insects. In PRICE, P.M., ed., Evolutionary strategies of parasitic insects. Plenum Press, London. CLARKE, A. H., Jr., 1973, The freshwater molluscs of the Canadian Interior Basin. Malacologia, 13: 1-509. CLARKE, A. H., Jr. 8 BERG, C. O., 1959, The freshwater mussels of central New York. Cornell University Agricultural Experiment Station Mem- oir 367: 1-79. CLARKE, А. Н., Jr. & RICK, A. M., 1963, Supple- mentary records of Unionacea from Nova Scotia with a discussion of the identity of Anodonta fragilis Lamarck. National Museums of Canada Bulletin, 199: 15-27. CLARKE, B. C., 1968, Balanced polymorphism and regional differentiation in land snails. In DRAKE, E. T., ed., Evolution and environment. Yale Uni- versity Press, New Haven and London, p. 351-368. DAVIDSON, В. G. 8 CORTNER, J. A., 1967, Mito- chondrial malate dehydrogenase: a new genetic polymorphism in man. Science, 157: 1569-1570. DAVIS, С. М., 1983, Relative roles of molecular genetics, anatomy, morphometrics and ecology in assessing relationships among North Ameri- can Unionidae (Bivalvia). In OXFORD, С. $5. 4 ROLLINSON, D., ed., Protein polymorphism: adaptive and taxonomic significance. Academic Press, London, p. 193-222. DAVIS, G. M., 1984, Genetic relationships among some North American Unionidae (Bivalvia): sib- ling species, convergence, cladistic relation- ships. Malacologia, 25: 629-648. DAVIS, G. M., HEARD, W. H., FULLER, $. L. Н. 4 HESTERMAN, C., 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. ELSON, J. A., 1969, Late Quaternary marine sub- mergence of Quebec. Revue Géographique de Montréal, 23: 247-258. ENDLER, J. A., 1977, Geographic variation, spe- ciation, and clines. Monographs in Population Biology (Princeton University Press), 10: ix + 246 p. GALLEZ, G. P. & GOTTLIEB, L. D., 1982, Genetic evidence for the hybrid origin of the diploid plant Stephanomeria diegensis. Evolution, 36: 1158-1167. GOLDING, С. В. & STROBECK, C., 1983, In- creased number of alleles found in hybrid popu- lations due to intragenic recombination. Evolu- tion, 37: 17-29. GRANT, V., 1966, The origin of a new species of Gilia in a hybridization experiment. Genetics, 54: 1189-1199. GREEN, D. M., 1979, A BASIC computer program for calculating indices of genetic distance and similarity. Journal of Heredity, 70: 429-430. HAFNER, J. C., 1982, Genetic interaction at a con- tact zone of Urodema bilobatum (Chiroptera: Phyllostomidae). Evolution, 36: 852-862. HARRIS, H. 8 HOPKINSON, D. A., 1978, Hand- 124 KAT book of enzyme electrophoresis in human genet- ics. North Holland Publishing Co., Amsterdam. HUNT, W. G. & SELANDER, В. K., 1973, Biochem- ical genetics of hybridization in European house mice. Heredity, 31: 11-33. JOHNSON, В. 1., 1970, The systematics and zoo- geography of the Unionidae (Mollusca: Bivalvia) of the southern Atlantic Slope region. Bulletin of the Museum of Comparative Zoology, 140: 263—450. JOHNSON, В. 1., 1980, Zoogeography of North American Unionidae (Mollusca: Bivalvia) north of the maximum Pleistocene glaciation. Bulletin of the Museum of Comparative Zoology, 149: 77-189. KAT, P. W., 1982, The relationship between he- terozygosity for enzyme loci and developmental homeostasis in peripheral populations of fresh- water bivalves (Unionidae). American Naturalist, 119: 824-832. KAT, P. W., 1983a, Patterns of electrophoretic and morphologic variability in a widely distributed unionid: an initial survey. Netherlands Journal of Zoology, 33: 21—40. KAT, P. W., 1983b, Fossil evidence from Fish House clays for the origin and changes in spe- cies composition through time of the northern Atlantic Slope unionid fauna (Mollusca: Bivalvia). Proceedings of the Academy of Natural Sciences of Philadelphia, 135: 85-101. KAT, P. W., 1983c, Morphologic divergence, genet- ics, and speciation among Lampsilis (Bivalvia: Unionidae). Journal of Molluscan Studies, 49: 133-145. KAT, P. W., 1983d, Genetic and morphological divergence among nominal species of North American Anodonta (Bivalvia: Unionidae). Malacologia, 23: 361-374. KAT, P. W., 1983e, Conchiolin layers among the Unionidae and Margaritiferidae (Bivalvia): mi- crostructural characteristics and taxonomic impli- cations. Malacologia, 24: 298-311. KAT, P. W., 1984, Parasitism and the Unionacea (Bivalvia). Biological Reviews, 59: 189-207. KAT, P. W. 8 DAVIS, G. M., 1984, Molecular ge- netics of peripheral populations of Nova Scotian Unionidae (Mollusca: Bivalvia). Biological Jour- nal of the Linnean Society, 22: 157-185. KOEHN, R. K. 8 EANES, W. F., 1976, An analysis of allelic diversity in natural populations of Drosophila: the correlation of rare alleles with heterozygosity. In KORBIN, S. 8 NEVO, E., ed., Population genetics and ecology. Academic Press, New York. LEWIS, H., 1966, Speciation in flowering plants. Science, 152: 167-172. McDONNELL, L. J., GARTSIDE, D. F. 8 LITTLE- JOHN, М. J., 1978, Analysis of a narrow hybrid zone between two species of Pseudophryne (Anura: Leptodactylidae) in southeastern Austra- lia. Evolution, 32: 602—612. MILLER, В. B., KARROW, P. Е. 8 KALAS, |. L., 1979, Late Quaternary mollusks from glacial Lake Algonquin, Nipissing, and Transitional sed- iments from southwestern Canada. Quaternary Research, 11: 93-112. MOORE, W. S., 1977, An evaluation of narrow hybrid zones in vertebrates. Quarterly Review of Biology, 52: 263-277. MORAN, С., WILKINSON, P. 8 SHAW, О. D., 1980, Allozyme variation across a narrow hybrid zone in the grasshopper, Caledia captiva. Heredity, 44: 69-81. MORGAN, К. 8 STROBECK, C., 1979, Is intragenic recombination a factor in the maintenance of genetic variation in natural populations? Nature, 277: 383-384. NEI, M., 1972, Genetic distance between popula- tions. American Naturalist, 106: 283-291. NEVO, E. 8 BAR-EL, H., 1976, Hybridization and speciation in fossorial mole rats. Evolution, 30: 831-840. ОНМО, S., STENIUS, C., CHRISTIAN, L. 4 SCH- IPMANN, G., 1969, De novo mutation-like events observed at the 6PGD locus of the Japanese quail, and the principle of polymorphism breeding more polymorphism. Biochemical Genetics, 3: 417-428. ORTMANN, A. E., 1913, The Alleghenian divide and its influence upon the freshwater fauna. Pro- ceedings of the American Philosophical Society, 52: 287-390. PATTON, J. L., HAFFNER, J. C., HAFFNER, М. $. & SMITH, M. F., 1979, Hybrid zones in Thomomys bottae pocket gophers: genetic, ph- enetic, and ecological concordance patterns. Ev- olution, 33: 860-867. REMINGTON, C. L., 1968, Suture-zones of hybrid interaction between recently joined biotas. Evo- lutionary Biology, 2: 321—428. ROHLF, E., KISHPAUGH, J. & KIRK, D., 1972, NT-SYS: numerical taxonomy system of multi- variate statistical programs. State University of New York, Stony Brook, New York. SAGE, D. E. & SELANDER, В. K., 1979, Hybrid- ization between species of the Rana pipiens complex in central Texas. Evolution, 33: 1069-1088. SHORT, L. L., 1972, Hybridization, taxonomy, and avian evolution. Annals of the Missouri Botanical Garden, 59: 447—453. SIMPSON, С. T., 1896, On the Mississippi Valley Unionidae found in the St. Lawrence and Atlantic drainage areas. American Naturalist, 30: 379-384. SMITH, D. G., 1982, The zoogeography of the freshwater mussels of the Taconic and southern Green Mountain region of northeastern North America (Mollusca: Pelecypoda: Unionacea). Canadian Journal of Zoology, 60: 261-267. SMITH, D. G., 1983, Notes on Mississippi River basin Mollusca presently occurring in the Hudson River system. Nautilus, 97: 128—131. THOMPSON, J. М. & WOODRUFF, В. C., 1978, UNIONID HYBRIDIZATION 125 Mutator genes—pacemakers of evolution? Na- ture, 274: 317-321. TRDAN, В. J., 1981, Reproductive biology of L. radiata siliquoidea (Pelecypoda: Unionidae). American Midland Naturalist, 106: 243- 248. TRDAN, R. J. & HOEH, W. R., 1982, Eurytopic host use by two congeneric species of freshwater mussel (Pelecypoda: Unionidae: Anodonta). American Midland Naturalist, 108: 381- 388. TSUNO, K., 1981, Studies on mutation at esterase loci in Drosophila viridis. |. Spontaneous mut- ation rates and newly arisen variants. Japanese Journal of Genetics, 56: 155-174. WATT, W. B., 1972, Intragenic recombination as a source of population genetic variability. American Naturalist, 106: 737—753. WHITE, M. J. D., 1978, Modes of speciation. Free- man, San Francisco, 455 p. WOODRUFF, О. S., 1981, Toward а genodynamics of hybrid zones: studies of Australian frogs and West Indian land snails. In ATCHLEY, W. R. 4 WOODRUFF, D. S., ed., Evolution and spe- ciation. Cambridge University Press, p. 171- 197. WOODRUFF, R. C., SLATKO, B. E. & THOM- PSON, J. N., 1982, Factors affecting mutation rate in natural populations. In ASHBURNER, M., CARSON, Н. L. 4 THOMPSON, J. М. eds, Ge- netics and biology of Drosophila, Vol. 3C, Aca- demic Press, New York. APPENDIX Classification of species mentioned in the text, and location of the collection sites. Unionidae Anodontinae Anodonta cataracta Say Anodonta fragilis Lamarck Anodonta gibbosa Say Anodonta grandis Say Ambleminae Lampsilini Lampsilis radiata (Gmelin) Lampsilis siliquoidea (Barnes) Lampsilis splendida (Lea) Pleurobemini Elliptio complanata (Lightfoot) Elliptio dilatata (Rafinesque) DE1 Andover Branch, Millington, Kent Co., Delaware DE2 Concord Pond, Sussex Co., Delaware DES Deep Creek, Nanticoke Acres, Sussex Co., Delaware MD1 Chester River, Millington, Kent Co., Maryland MD2 Mason Branch, Queen Anne, Queen Annes Co., Maryland MD3 Norwich Creek, Queen Anne, Talbot Co., Maryland MD4 Sassafras River, Sassafras, Cecil Co., Maryland ME1 Lake St. George, Waldo Co., Maine ME2 Kennebec River, Somerset Co., Maine MI Wiggins Lake, Gladwin Co., Michigan NB French Lake, Oromocto, Sunbury Co., New Brunswick NJ1 Delaware River, Burlington Co., New Jersey №2 Swartswood Lake, Sussex Co., New Jersey NS1 Lake Egmont, Cooks Brook, Halifax Co., Nova Scotia NS2 First Lake O' Law, Baddeck, Victoria Co., Nova Scotia NS3 Mattatall Lake, Wentworth Centre, Cumberland Co., Nova Scotia NS4 Newville Lake, Halfway River East, Cumberland Co., Nova Scotia NS5 Placide Lake, Havelock, Digby Co., Nova Scotia NS6 Shaw Lake, Arichat, Isle Madame, Richmond Co., Nova Scotia NS7 Grand Lake Shubenacadie, Grand Lake, Halifax Co., Nova Scotia NS8 Sydney River, Sydney, Cape Breton Co., Nova Scotia PA Susquehanna River, Cumberland Co., Pennsylvania VT1 Lake Champlain, South Hero, Grand Isle Co., Vermont VT2 Joes Pond, Danville, Caledonia Co., Vermont VT3 Lake Memphremagog, Newport, Orleans Co., Vermont WI St. Croix River, Hudson, St. Croix Co., Wisconsin MALACOLOGIA, 1986, 27(1): 127-172 PHREATIC HYDROBIIDS (GASTROPODA: PROSOBRANCHIA) FROM THE EDWARDS (BALCONES FAULT ZONE) AQUIFER REGION, SOUTH-CENTRAL TEXAS В. Hershler? 8 G. Longley Edwards Aquifer Research and Data Center, Southwest Texas State University, San Marcos, Texas 78666-4615, U.S.A. ABSTRACT This paper provides a systematic analysis of phreatic hydrobiids from 23 localities in south- central Texas, including 14 artesian wells and four springs in the Edwards (Balcones Fault Zone) Aquifer. Hauffenia micra (Pilsbry & Ferriss) and Ногайа nugax (Pilsbry & Ferriss) are redescribed as members of a new genus, and two additional new genera, seven new species and one new subspecies are also described (Table 2). Detailed morphological descriptions, provided for all taxa, emphasize characters of the shell, operculum, pallial cavity, digestive system, and reproductive system of both sexes. Two of the new genera are monotypic littoridinines having affinities with phreatic or epigean littoridinines from Mexico. The affinities of the third new genus, a hydrobiine which includes seven well-differentiated species, remain unclear. While all of the species are found in the Edwards (Balcones Fault Zone) Aquifer, at least four of the species are probably found in other aquifers of south-central Texas as well. With the description of seven new hydrobiid species, the rich and still poorly sampled troglobitic biota of the Edwards (Balcones Fault Zone) Aquifer now totals 39 troglobitic animal species, including four vertebrates. Key words: Edwards (Balcones Fault Zone) Aquifer; south-central Texas; troglobitic fauna; Hydrobiidae; morphology; systematics. INTRODUCTION The Hydrobiidae (Gastropoda: Rissoacea) are a large family (over 100 genera and 1000 species; С. M. Davis, 1979) of dioecious, gill- breathing snails that have radiated into di- verse fresh- and brackish-water habitats worldwide. Minute, unpigmented hydrobiids occupy groundwater habitats in numerous areas, with a large fauna occurring in karst regions of Europe (Vandel, 1965; Radoman, 1973), and lesser deployments occurring in North America (Morrison, 1949; Taylor, 1966), Mexico (Taylor, 1966; Hershler, 1984; 1985), Japan (Kuroda & Habe, 1958), and New Zealand (Ponder, 1966; Climo, 1974, 1977). Apart from the European deployment, little is known regarding the systematics and zoogeography of these taxa, in large part due the extremely small size (maximum shell di- mension often > 2 mm) of the snails and difficulties in sampling their habitat. One of the poorer known phreatic hydrobiid faunas is that of Texas. Ногайа nugax (Pilsbry 8 Ferriss, 1906) and Hauffenia micra (Pilsbry 8 Ferriss, 1906), described from river drift shells, have long been the sole described hydrobiids considered phreatic in Texas. As the anatomy of these taxa has not been stud- ied, it is not known whether they are conge- neric with any of the European phreatic hydrobiids (i.e., Horatia s.s. and Hauffenia 5.5.) that they resemble in shell features. Later collections from a cave (Reddell, 1965), spring (Taylor, 1974), stream drift (Strecker, 1935; Hubricht, 1940; Fullington, 1978; J. R. Davis, 1983), and artesian wells (Karnei, 1978; Longley, 1975, 1978, 1981) provided possible additional records for these species as well as probable new taxa, pointing to the presence of a diverse phreatic snail fauna in south-central Texas. Most of these collections have been from areas underlain by the Edwards (Balcones Fault Zone) Aquifer (Fig. 1). While phreatic faunas are known from several aquifers in Texas (Reddell, 1965, 1967, 1970; Reddell 8 ‘Present address: Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, DC 20560, U.S.A. (127) 128 HERSHLER 8 LONGLEY EDWARDS (Balcones Fault Zone) AQUIFER REGION AUSTIN SUBREGION SAN ANTONIO SUBREGION DRAINAGE FIG. 1. Map of the Edwards (Balcones Fault Zone) Aquifer Region showing the 31 localities considered in this paper. Squares refer to springs or caves, while circles refer to artesian wells that have (filled) or have not (open) yielded hydrobiids. Locality numbers are as in Appendix 1. Mitchell, 1969; Mitchell & Reddell, 1971), that of the Edwards is particularly rich, totalling 32 described animal species, including four ver- tebrates (Table 1). Twenty of these species, including nine amphipods, have been ob- tained from (and several are endemic to) the single well in the aquifer that has received continuous long-term sampling (Table 1), re- flecting not only the high local diversity of the fauna, but also the probability that many more taxa await discovery elsewhere in the aquifer. This diverse fauna poses a series of ques- tions to the biologist. What are the origins of the various faunal elements? What roles have hydrologic factors such as the presence of groundwater divides played in effecting spe- ciation in the aquifer? What is the food source for primary consumers in the deep artesian community, which has a surprisingly high trophic complexity (Longley, 1981)? Study of the Edwards Aquifer hydrobiids is clearly necessary not only to help provide a clearer picture of this unique groundwater ecosystem, but also to further our under- standing of the systematics, evolution, and adaptive radiation of the Hydrobiidae. This paper provides a systematic analysis of phreatic hydrobiids from 23 localities in south- central Texas, including 14 artesian wells and four springs in the Edwards Aquifer. We rede- scribe Ногайа nugax and Hauffenia micra as members of a new genus and also describe two additional new genera, seven new spe- cies and one new subspecies. A classification of these taxa is given in Table 2. Detailed morphological descriptions are given for all taxa. The systematic relationships of the var- ious taxa are assessed, with emphasis on comparisons with other phreatic hydrobiids that have been similarly studied. The Edwards (Balcones Fault Zone) Aqui- fer. A brief description of the aquifer is nec- essary as a prelude to discussion below. For further details the reader 1$ referred to Klemt et al. (1979). The Edwards (Balcones Fault Zone) Aquifer extends for 282 km (from TEXAS PHREATIC HYDROBIIDS 129 TABLE 1. Described biota of the Edwards Aquifer (from Reddell, 1965, 1967, 1970; Reddell & Mitchell, 1969; Longley, 1975, 1978; Bowman 8 Longley, 1976; Strenth, 1976; Young & Longley, 1976; Нап, 1978; Holsinger 8 Longley, 1980). Only troglobitic species found in the artesian zone of the aquifer are listed. Taxa marked with an asterisk have been collected from the artesian well at Southwest Texas State University. Platyhelminthes Kenkiidae *Sphalloplana mohri Hyman Mollusca Hydrobiidae *Horatia nugax (Pilsbry & Ferriss) Hauffenia micra (Pilsbry & Ferriss) Arthropoda Cypridae *Cypridopsis vidua obesa Brady & Robertson Cyclopidae *Cyclops cavernarum Ulrich *Cyclops learii Ulrich Cyclops varicans rebellus Lilljeborg Entocytheridae Sphaeromicola moria Hart Asellidae *Lirceolus smithi (Ulrich) Asellus pilus Steeves Asellus redelli Steeves Cirolanidae *Cirolanides texensis Benedict Monodellidae *Monodella texana Maguire Crangonyctidae *Stygobromus flagellatus (Benedict) *Stygobromus russelli (Holsinger) Stygobromus pecki (Holsinger) Stygobromus balconis (Hubricht) Stygobromus bifurcatus (Holsinger) Hadziidae *Texiweckelia texensis (Holsinger) *Texiweckelia insolita Holsinger *Texiweckelia samacos Holsinger “Allotexiweckelia hirsuta Holsinger Bogidiellidae *Parabogidiella americana Holsinger Artesiidae “Artesia subterranea Holsinger Sebidae *Seborgia relicta Holsinger Palaemonidae *Palaemonetes antrorum Benedict Palaemonetes holthuisi Strenth Dytiscidae *Hadeoporus texanus Young & Longley Chordata Ambystomidae *Typhlomolge rathbuni Stejneger Typhlomolge robusta Longley Ictaluridae Satan eurystomus Hubbs & Bailey Trogloglanis pattersoni Eigenmann Brackettville to north of Georgetown), paral- leling the Balcones Escarpment and Fauit Zone in south-central Texas, and varies in width from eight to 48 km (Fig. 1). The aquifer consists of a recharge and artesian (reservoir) zone. Recharge to the aquifer occurs largely by downward percolation from streams cross- ing areas where the Edwards outcrops (Fig. 1). To the south and east of the recharge zone the Edwards Formation dips downward (as- suming artesian conditions), and the top of the formation is as much as 600 m beneath ground level in Bexar County (Klemt et al., 1979). The Cretaceous Edwards limestone that comprises the aquifer is highly porous, due to the effects of solution and faulting. The aquifer is thought to have extensive water- filled caves and caverns (Pettit & George, 1956; Klemt et al., 1979). Note that the Balcones Fault Zone has the highest density of caves of any physiographic region in Texas (Smith, 1971). As a result of this high second- ary porosity, transmissivity is high in the aqui- fer, as seen in the occurrence of a large number of high capacity wells, some of which flow at ground level and discharge several thousand liters/second (Maclay 8 Small, 1976; Klemt et al., 1979). Several groundwa- ter divides are present in the aquifer, and smaller phreatic pools are also thought to have resulted from the intensive folding and fracturing in the bedrock (Holsinger & Long- ley, 1980). South and south-east of the “bad- water” line (reservoir boundary, Fig. 1), the Edwards water has sluggish circulation and 1$ no longer of good quality, having > 1000 mg/1 total dissolved solids (Klemt et al., 1979). Nat- ural discharge of the aquifer occurs (or has occurred) at major springs near Uvalde (Leona Springs), San Antonio (San Pedro, TABLE 2. Classification of phreatic hydrobiids from the Edwards (Balcones Fault Zone) Aquifer Region. Family Hydrobiidae Subfamily Hydrobiinae Phreatodrobia micra (Pilsbry & Ferriss) п. деп. Phreatodrobia nugax nugax (Pilsbry & Ferriss) Phreatodrobia nugax inclinata п. subsp. Phreatodrobia rotunda п. sp. Phreatodrobia conica n. sp. Phreatodrobia plana n. sp. Phreatodrobia imitata n. sp. Phreatodrobia punctata n. sp. Subfamily Littoridininae Balconorbis uvaldensis n. gen., n. sp. Stygopyrgus bartonesus п. gen., п. sp. 130 HERSHLER 8 LONGLEY San Antonio Springs), New Braunfels (Comal Springs), San Marcos (San Marcos Springs) and Austin (Barton Springs). Artificial dis- charge has also occurred in recent years through the hundreds of high capacity wells in the artesian zone. The Edwards (Balcones Fault Zone) Aquifer is separated from the Edwards (Plateau) Aquifer by a region where outcropped Edwards limestone has largely been eroded. MATERIALS AND METHODS Sampling and localities. The bulk of the material examined during this study was ob- tained by the well sampling program con- ducted by staff of the Edwards Aquifer Re- search and Data Center (EARDC) during 1976-1981. A total of 22 wells in the Edwards Aquifer were sampled often enough to either obtain troglobitic organisms or provide confi- dence that troglobites were absent in that area. Details for these wells regarding United States Geological Survey or Texas Board of Water Engineers well number, well owner, well depth, number of samples taken, and presence-absence of troglobitic fauna are given in Table 3. Fine-mesh funnel (constricted or open) nets were attached to pipes from artesian wells using hose clamps (Fig. 2A). The collecting vessel at the end of the net was usually a 3.8 liter plastic jar, although occasionally a small section of 64 um mesh netting material clamped to a section of polyvinylchloride (PVC) pipe (with screw-on cap) was used instead. All material collected was washed through netting attached to a plastic funnel and then transferred to 70% EtOH. Groundwater outlets of four springs were sampled as follows. The “Pipe” (or “Diver- sion”) orifice was sampled at San Marcos Springs. Developers of the spring capped this orifice (one of many feeding this spring) by cementing an old diving bell to the spring floor. A 29” culvert pipe was then attached to the opening of the bell to divert the spring flow elsewhere. A 29” diameter sampling net was placed at the end of the pipe to filter the water stream. At Barton Springs, sampling was done at the “Concession” Spring (Fig. 2B), which has been cemented over with several holes serving to release the spring flow. TABLE 3. Data regarding 22 Edwards Aquifer wells that have been sampled. Depth Number of Troglobites Well no. Owner (m) samples (Gay) = Southwest Texas State University 59 500 + + AY-68-29-923 Longhorn Portland Cement Co. (#2) 143 36 + AY-68-37-127 Brackenridge Zoo 124 45 4 AY-68-37-508 City Water Board (San Antonio) 402 48 + Artesian Station, Well 4 AY-68-36-918 Union Stockyards (#3) 412 8 + AY-68-37-710 City Water Board (San Antonio) 460 9 + Mission Station AY-68-43-115 J. H. Uptmore (#5) 227 10 + AY-68-44-107 Lakeland City Water Co. 555 5 = AY-68-44-215 City (Antonio) Public 358 22 + Service Board (#1) AY-68-43-107 Rio Vista Farms — 9 = AY-68-43-608 Verstraeten Brothers 513 10+ + AY-68-43-601 O. R. Mitchell 582 10+ + AY-68-43-505 J. W. Watts 610 2 = YP-69-43-103 King Farms = 3 + YP-69-43-801 D. C. Carnes — 5 = YP-69-50-? (=H-6-24) C. Reagan 386 4 + Н-6-43 W. С. Reagan 373 SEE + YP-69-50-109 R. Carnes 320 2 > YP-69-50-105 R. K. Dunbar 288 2 + Н-5-135 5. Moerbe 98 2 + Н-5-158 G. Ligocky 286 10+ + YP-69-50-501 Uvalde National Fish Hatchery — 2 + TEXAS PHREATIC HYDROBIIDS 131 AS FIG. 2. Sampling nets placed on outlets of the G. Ligocky Well, Uvalde County (A) and Barton (“Con- cession”) Springs, Travis County (B). Three- and six-inch PVC pipes with attached irrigation “socks” or funnel netting (and “cap” type collection vessel) were wedged into these holes to sample the water stream. Sim- ilar techniques were used to sample the nat- ural orifices at Comal (main spring) and Hueco Springs (“A”, smaller; and “В”, larger spring). Samples from springs were pro- cessed as above. Most of the wells and springs were sampled every 2-7 days during the sampling period. Samples from caves were obtained by vi- sual search for shells near permanent water and by sieving sediment with a fine hand sieve. Living snails were not obtained from any of these caves. The location of the 31 localities considered in this paper are given in Fig. 1. Note that three of the localities (all caves) sampled are from areas not underlain by the Edwards (Balcones Fault Zone) Aquifer (see below). Locality data are given in Appendix 1. Morphological study. Methods of morpho- logical study largely follow those of G. M. Davis (1979) and Hershler (1985). Live mate- rial was available only for two species. Ana- tomical study centered on the pallial cavity (and contained structures), digestive system, and reproductive systems of both sexes. Us- age of body surface references (left, right, dorsal, ventral) follows that of Fretter 8 Graham (1962). The nervous system was studied only in one species. Measurements of the length and width of the intestine coil in the pallial cavity refer to the maximum dimen- sions of the coil parallel and perpendicular to the length of the pallial cavity. To prepare snails for dissection, the calcified portion of the shell was first removed by placing individ- uals into concentrated Bouin's solution for 24 hours. Dissections were done using a Zeiss dissecting microscope (50X) fitted with an oc- ular micrometer, with specimens immersed in dilute Bouin's solution. Shells were measured using the same microscope (32X, 50X). Radulae were photographed using a scan- ning electron microscope. All radular data were obtained from these photographs. Shells were photographed using either a 35 mm camera attached to a Zeiss microscope, or the scanning electron microscope. Deposition of type material. Holotypes and paratypes are deposited in the Academy of Natural Sciences of Philadelphia. Reference is made to catalog numbers (ANSP) assigned to this material. All other material examined 15 housed in the EARDC collection. Taxonomic procedure. The higher level classification of the Hydrobiidae is currently confused, due to the high incidence of con- vergence and mosaic evolution within the family, and several classifications have re- cently been proposed (Taylor, 1966; Rado- man, 1973; С. M. Davis et al., 1982). Conver- gence in shell features is well documented in the Hydrobiidae (G. M. Davis, 1979) and high- er-level classifications largely based on these features (i.e., Taylor, 1966) are likely to produce polyphyletic and artificial groupings. In this paper we follow the subfam- ilial breakdown used by G. M. Davis et al. (1982) in which the Hydrobiinae, Litho- glyphinae (see also Thompson, 1984), Lit- toridininae (see also Hershler, 1985), and Nymphophilinae are recognized. Note, how- ever, that convergence may extend to key characters used in this classification: a sper- mathecal duct, diagnostic of the Littoridininae (G. M. Davis et al., 1983), has apparently arisen numerous times among various ris- soacean clades (Ponder, 1984). Further con- fusion has been added by the recent discov- ery of a genus having a hydrobiine-type fe- 132 HERSHLER 8 LONGLEY TABLE 4. Generalized cusp formulae for the four radular tooth types. Inner Outer Species Central Lateral marginal marginal Phreatodrobia micra 5-1-5 5-1-6 21-23 13-16 1-1 Phreatodrobia nugax 5(6-8)-1-5(6,7) 5-1-6(7,8) 24-34 19-26 1(2)-1 Phreatodrobia rotunda 6-1-5(6) 5-1-5(6) 20-23 2 1-1 Phreatodrobia conica 6(7,8)-1-6(7,8) 7(8)-1-10(11) 21-26 15-17 Phreatodrobia plana 6(8)-1-6(7,8) 8(9)-1-11(12) 23 24-28 Phreatodrobia imitata 7(8)-1-7(8) 6(7)-1-6(7) 21-23 14-17 Phreatodrobia punctata 7(9)-1-7(8) 7(8,9)-1-10(11,12) 23-26 22-24 Stygopyrgus bartonensus 4(5)-1-4(5) 4(5-7)-1-5(6) 2225 15-17 1-1 Balconorbis uvaldensis 5(6)-1-4(5) 6-1-6(7,8) 21-24 17-20 male reproductive system and a littoridinine- type репа! gland (Giusti & Bodon, 1984; fig. 2B, G). While a search for morphological characters that identify clades within the Hydrobiidae must obviously be continued, we feel that the above classification 1$ still the best available. Only character-states unique to or diagnos- tic of the taxa concerned are emphasized in this paper. Radular data are presented in Ta- ble 4. Shell and other morphological data are presented in Appendices 2 and 3, respec- tively. The phenogram shown in Fig. 28 was generated using the CLUSTAN software package (developed by David Wishart of the Universities of St. Andrews and London), with Ward's method (error sum of squares) of clus- tering selected. DESCRIPTION OF TAXA Family HYDROBIIDAE Subfamily Hydrobiinae Phreatodrobia Hershler & Longley, new genus Horatia Bourguignat, 1887 (in part): 47 Hauffenia Pollonera, 1898 (in part): 3 DIAGNOSIS. Shell (Figs. 3-6) minute (max- imum dimension, < 2.5 mm), with four or fewer whorls, colorless, transparent, varying in shape from planispiral to trochoid to coni- cal. Protoconch (Figs. ЗК, L, $, 6F, 7A-F) with pitted microsculpture; teleoconch sculpture variable (Figs. 3—6, 8). Aperture slightly to highly flared. Operculum (Figs. 9, 10) nucleus typically sub-central; ventral surface smooth or with a central, knob-like process. Animal unpigmented and without eyespots (Figs. 11, 12). Pallial cavity typically slightly longer than wide; ctenidium absent, incomplete, or fully formed (Fig. 13A-G). Stomach chambers poorly distinguishable externally (Fig. 18). In- testine with loop(s) in roof of posterior portion of pallial cavity (Figs. 12, 13A-G), and some- times on style sac (Fig. 18B, C). Central cusp of central and lateral teeth sometimes not enlarged relative to rest of cusp row (Figs. 16А-С, 17A-D); central teeth with (Figs. 14, 15А-Е) or without (Figs. 15F-H, 16, 17A—D) basal cusps (projecting from the lateral an- gles). Radular cusps dagger-like in shape, sometimes highly numerous on central and lateral teeth. Pallial portion of prostate typi- cally 50% of total prostate length (Figs. 19D, E). Penis simple, without lobes or specialized glands (Figs. 19A, C). Capsule gland usually large compared to albumen gland, with ante- rior end sometimes having a terminal bend or coil (Fig. 20A, B, H). Bursa copulatrix large relative to the seminal receptacle, with a straight or coiled duct, and largely or totally posterior to the albumen gland (Figs. 20, 21). Seminal receptacle and oviduct coil appres- sed to, or largely or totally posterior to albu- men gland. Oviduct opens into the posterior or anterior end of the albumen gland. TEXAS PHREATIC HYDROBIIDS 133 FIG. 3. Shells of Phreatodrobia nugax inclinata (A-E, |)(Locality 11) and P. micra (F-H, J-L, M-T). Shell widths are as follows: А, В (1.33 mm, holotype); С (1.05 mm), D (1.04 mm), Е (1.15 mm); F (1.10 mm), J (1.10 mm), N (1.16 mm) (syntypes, Locality 6); @ (1.14 тт), К (0.945 mm), O (1.23 mm)(Locality 5A); H (0.8 mm), L (1.80 mm), Р. (0.76 mm)(Locality 3); М (0.93 mm), О (0.75 mm)(Locality 9); R (1.07 тт), $ (0.89 mm), T (0.877 mm)(Locality 10). The scale bar next to | equals 0.1 mm. REMARKS. The following features are also typical of the genus: а) the anus is positioned close to the mantle edge (Fig. 13); b) the kidney opening (Ro, Fig. 13A) is simple; c) the penis 1$ three-four times as long as the snout (in preserved specimens), slender, with folds along the inner curvature, and coils on the right side of the “neck”; d) the gonads are simple and without lobes, and e) the anterior vas deferens exits from the mid-prostate just anterior to the end of the pallial cavity (Fig. 19D). Phreatodrobia is distinguished from Ho- ratia s.s., Hauffenia s.s., and other European phreatic hydrobiines by its minute, fragile shell and simple penis, which lacks lobes and glandular swellings (see below). The minute shell with pitted apical microsculpture, simple penis, and lack of eyespots and body pigment distinguish Phreatodrobia from all other de- scribed North American hydrobiids. Apart from these features, the following shared character-states unite the morphologically di- verse members of this genus: operculum nu- cleus central or near-central; pallial prostate relatively large; intestine with loop(s) in pallial roof; and bursa large relative to seminal re- ceptacle and positioned partly or totally pos- terior to the albumen gland. 134 HERSHLER 8 LONGLEY FIG. 4. Shells of Phreatodrobia nugax nugax. Shell widths are as follows: A, F, J (1.23 mm, holotype, Locality 6); B (1.71 mm), G (1.71 mm), K (1.6 mm)(Locality 4); C (1.55 mm), D (1.79 mm), E (1.79 mm), H (1.23 mm), L (1.53 mm)(Locality 3); | (1.07 mm, Locality 1); М (1.0 mm), Q (1.13 mm), U (1.47 mm), Y (1.09 mm), 2 (0.89 mm)(Locality 2); N (1.57 mm), R (1.51 mm), V (1.64 mm)(Locality 8); O (1.53 mm)(Locality 10); P (1.08 mm)(Locality 24); S (0.85 mm), W (0.85 mm)(Localtiy 9); T (0.85 mm)(Locality 12); X (0.945 mm)(Locality 14). TYPE-SPECIES. Phreatodrobia micra (Pilsbry 8 Ferriss, 1906). DISTRIBUTION. Found throughout the Edwards (Balcones Fault Zone) Aquifer. Some species also probably live in other aqui- fers in south-central Texas (see below). ETYMOLOGY. The generic name is derived from the Greek word phreatos, referring to the groundwater habitat shared by members of this taxon. Phreatodrobia micra (Pilsbry 4 Ferriss) Figs. 3F-H, J-T, 9F, С, 13B, 15A—C, 20D Valvata micra Pilsbry 8 Ferriss, 1906: 172—173. Horatia (Hauffenia) micra (Pilsbry & Ferriss). Pilsbry, 1916: 84 Hauffenia micra (Pilsbry & Ferriss). Burch, 1982: 30. MATERIAL EXAMINED. HAYS COUNTY: San TEXAS PHREATIC HYDROBIIDS 135 FIG. 5. Shells of Phreatodrobia plana (A—D, С, H, К, L), P. rotunda (E, F, |, J; Locality 3), and P. а Shell widths are as follows: А (0.822 mm, holotype), B (0.830 mm), C (0.77 mm), G (0.64 mm), mm)(Locality 3); D (1.14 mm), H (0.945 mm), L (1.12 mm)(Locality 8); E (2.26 mm, holotype, ANSP), mm), | (1.84 mm), J (2.06 mm); M (shell height, 1.61 mm, holotype, ANSP), N mm)(Locality 5B); P (0.945 тт; Locality 5A). Marcos Springs. COMAL COUNTY: Guadalupe River drift; Honey Creek Cave; Hueco Springs. KENDALL COUNTY: Century Cav- erns. DIAGNOSIS. A small-sized species (shell width about 1.00 mm) with a planispiral, or near-planispiral shell, and a circular aperture that abuts against the penultimate whorl (Figs. 3F-H, J-T). Operculum (Figs. 9F, G) circular, with well-developed knob-like pro- cess on ventral surface; nucleus central. Ctenidium incomplete; osphradium fills large (39%) fraction of pallial cavity length (Fig. 13B). Central tooth of radula with single pair of basal cusps (Fig. 15C). Stomach almost twice as long as style sac; intestine with tight, U- shaped loop in pallial roof; long axis of loop at on» 225 7] (shell height, 1.79 mm), oblique angle to pallial cavity length (Fig. 13B). Ovary and testis occupy large propor- tion (50%) of digestive gland length. Albumen and capsule glands about equal in length (Fig. 20D). Oviduct opens into anterior end of al- bumen gland. REMARKS. Distinctive features of this spe- cies include the typically planispiral shell with tubular whorls and simple aperture, circular operculum with well-developed process on the ventral surface, incomplete ctenidium, large-sized osphradium, and large-sized ovary and testis. DESCRIPTION. The shell has 2.2-2.5 tubu- lar whorls and averages 0.84-1.08 mm in width for the four populations studied. The protoconch has 1.25 whorls and has fairly 136 HERSHLER 8 LONGLEY FIG. 6. Shells of Phreatodrobia imitata (A-G) and Р. punctata (H, |; Locality 3). Shell heights are as follows: А (1.07 mm, holotype), В (1.05 тт), С (0.89 тт), D (1.05 mm), F (shell width, 0.75 mm), G (shell width, 0.75 mm)(Locality 20); E (1.0 тт; Locality 21); H (1.05 mm, holotype), | (1.15 mm). large and deep pits (Fig. 3K, L, S). Axial growth lines are weakly developed on the teleoconch. Only the sample from Century Caverns includes shells with a slight spire projecting above the coil of the body whorl (Fig. ЗВ). A slight flaring of the арепуге, par- ticularly the inner lip, is seen in some speci- mens. The aperture is never free from the body whorl. The description of operculum and anatomy is based on study of specimens from San Marcos Springs. The flat, thickened opercu- lum has 2.5 whorls and is dark amber in color. The elevated knob-like process occupies a small portion of the operculum area and 1$ composed of horny material (as is the rest of the operculum). All specimens dissected had an incom- plete ctenidium, consisting of an efferent branchial vessel (Ev) with a few stubby filaments at its anterior end (Fig. 13B). The filaments are much smaller, relative to pallial cavity width, than those of Phreatodrobia nugax (see below). А similar incomplete ctenidium was described for Paluccia Giusti & Pezzoli, a European hydrobiid (Giusti & Pezzoli, 1981). The large osphrad- ium is typically positioned towards the ante- rior end of the incompletectenidium (Fig. 13B). The central tooth of the radula is trapezoidal in shape. Note that the lateral angles are highly divergent (Fig. 15C). The stomach lacks a caecal appendix. The ovary and testis consist of a solid, non-lobed mass. The vas deferens exits from the anterior end of the testis and con- sists of a few thickened coils (as in Fig. 18C) on the posterior half of the stomach. The prostate overlies the entire style sac and its anteriormost 44% is pallial. The penis is three-four times as long as the snout. The capsule gland opening is wide and slightly subterminal. The anterior portion of the cap- sule gland lacks a twist or coil. The tight coil of the anterior oviduct is appressed to the al- bumen gland and the seminal receptacle opens into the left side of the coil (Fig. 20D). The oviduct enters the albumen gland just after receiving the short duct from the bursa copulatrix. HOLOTYPE. ANSP 91322 (cotypes) (Fig. SE; y, IN): TYPE-LOCALITY. Drift debris of Guadalupe River about four miles above New Braunfels, Comal County (Fig. 1, Locality 6). TEXAS PHREATIC HYDROBIIDS 137 > = == FIG. 7. Protoconchs of Phreatodrobia nugax nugax (A, B; Localities 3, 2), P. punctata (C, Locality 3), P. conica (D, Locality 5B), P. rotunda (E, Locality 3), P. plana (F, Locality 3), Balconorbis uvaldensis (G, Locality 30), “Horatia” sp. (H), and Наийета subpiscinalis (1). All scale bars equal 0.1 mm. DISTRIBUTION. Edwards (Balcones Fault Phreatodrobia nugax (Pilsbry 8 Ferriss) Zone) Aquifer, and (possibly) Cow Creek and Figs. 3A-E, |, 4, 7A, В, 9A-E, 10-12, 13A, Glen Rose Aquifers in Hays, Comal, and Ken- 14, 18A, 19A, B, D dall Counties (Fig. 1, Localities 3, 5, 6, 9, 10). Valvata micra nugax Pilsbry & Ferriss, 1906: 138 HERSHLER 8 LONGLEY FIG. 8. SEM photographs of shell sculpture on Phreatodrobia imitata (А, В, О; Locality 20), P. conica (С; Locality 5B), and P. punctata shells (E, Locality 3). All scale bars equal 0.1 mm. The distance between spiral lines in D is 0.01 mm. 173: Horatia (Hauffenia) micra nugax (Pilsbry & Ferriss). Pilsbry, 1916: 84. Horatia nugax (Pilsbry & Ferriss). Taylor, 1975: 131. MATERIAL EXAMINED. TRAVIS COUNTY: Salamander Cave; Barton Springs. HAYS COUNTY: San Marcos Springs; SWTSU Well. COMAL COUNTY: Guadalupe River drift; Nat- ural Bridge Caverns; Honey Creek Cave. KENDALL COUNTY: Century Caverns. BEXAR COUNTY: Longhorn Portland Cement Com- pany Well; Brackenridge Zoo Well; Union Stockyards Well. UVALDE COUNTY: W. C. Reagan Well. DIAGNOSIS. A moderately large species (shell width about 1.3 mm), typically with a low trochoid shell (Figs. 3A—D, 4), but varying from near planispiral (Fig. 4H, N, X, Z) to low conical (Fig. 4Y). Aperture often free from penultimate whorl and highly flared; last 12% of body whorl highly thickened, imparting a white, opaque appearance to this part of shell (Fig. 4D, E). Operculum can have a central elevated thickening (Fig. 9B, D) on ventral surface. Ctenidium complete, with eight to 18 low filaments; osphradium length typically 25% of that of the pallial cavity (Fig. 134). Central radular tooth with one or two basal cusps (Fig. 14). Stomach length almost twice that of the style sac (Fig. 18A); intestine with loose U-shaped loop in pallial roof; long axis of loop at oblique angle to pallial cavity length (Fig. 13A). Ovary and testis occupy large pro- portion (67%, 57%) of digestive gland length. Capsule gland with (Fig. 20A, B) or without terminal coil; oviduct opens into anterior end of albumen gland. REMARKS. This species is distinguished by its low trochoid shell with thickened end of body whorl, complete ctenidium, and unusu- ally large ovary. The capsule gland coil and TEXAS PHREATIC HYDROBIIDS 139 FIG. 9. Operculae of Phreatodrobia nugax nugax (A-E), P. micra (Е, G; Locality 3), P. rotunda (H, |; Locality 3), and P. imitata (J, K; Locality 20). Oper- culum lengths are as follows: A (0.63 mm), С (0.57 mm)(Locality 3); В (ventral aspect, left-right dis- tance, 0.29 mm), О (ventral aspect, 0.56 mm)(Lo- cality 4); E (ventral aspect, 0.536 mm; Locality 24); F (0.29 mm), G (ventral aspect, 0.29 mm), H (0.536 mm), | (ventral aspect, 0.609 mm), J (0.29 mm), K (ventral aspect,, 0.314 mm). second pair of basal cusps on the central radular tooth, seen in some populations of this species, are unique in the genus. Two subspecies are recognized on the basis of differences in shell morphology. Кате! (1978) described shells of this species (as Gastropod Genus No. 1) from Brackenridge Zoo Well. Longley (1975, 1981) incorrectly identified specimens of this species from the SWTSU Well as P. micra. A A ááá==> ===, 0.50 mm FIG. 10. Operculum of Phreatodrobia nugax nugax (Locality 3). DESCRIPTION. The shell has 2.5-3.2 whorls and averages 0.93-1.71 mm in width. The protoconch, sometimes tilted (Fig. 31), has 1.25-1.50 moderately pitted whorls (Fig. 7A, B). Axial growth lines are typically well developed on the teleoconch (Fig. 4), which may also have collabral costae (Fig. 4U). General shell form is highly variable in some populations. For the San Marcos Springs pop- ulation, most individuals have the typical low trochoid shell (Fig. 4C), yet occasional Tn 7 A А y FIG. 11. Dorsal aspect of head of Phreatodrobia nugax. Note the presence of granules in the base of the tentacles (Tn), and the ciliation on the tentacle tips and along the outside edge of the base of the left tentacle. The dashed lines on the snout (Sn) indicate the position of the buccal mass. The ante- rior end of the foot is shown beneath the snout. Sn = snout; Tn = tentacle. 140 HERSHLER 8 LONGLEY A 0.50 mm FIG. 12. Morphology of a female Phreatodrobia nugax (minus the head-foot), as seen from the right (and slightly dorsal) side. The kidney tissue is not shown. Ag = albumen gland; Bu = bursa copulatrix; Cg = capsule gland; Dg = digestive gland; Dgg = digestive gland granule; Edg = posterior end of digestive gland; In = intestine; Me = mantle edge; Oo = oocyte; Ov = oviduct; Ova = ovary. smaller individuals with near-planispiral shells (Fig. 4H) were also found. Shell form in the Barton Springs population varies from near- planispiral to low trochoid, to low trochoid with costae, to low conical (Fig. 4М, О, U, Y, 2). The aperture is moderately flared all around, with flaring most pronounced in large individ- uals. While typically wider than long and near- circular, aperture shape can be modified (es- pecially in large-sized individuals from San Marcos Springs and the SWTSU Well) by a slight adapical notch or a pronounced abaxial fluting of the ощег Пр. The aperture 1$ free from the body whorl in 0—70% of the samples from given populations, with populations with large-sized individuals having the highest in- cidence. The inner lip is especially thickened. The white thickening of the end of the body whorl was seen in all fresh specimens. The operculum has 2.5 whorls (Fig. 10), with the nucleus positioned at about 43% of the operculum length, and varies from near- circular to ellipsoidal in shape (Fig. 9A, С, Е). Only individuals from San Marcos Springs and the SWTSU Well typically have a well- developed thickening on the ventral opercular surface. In these populations the operculum has a low conical shape, with the process consisting of extra layers of material depos- ited at the apex (operculum nucleus) of the cone (which points into the foot). The process is less prominent and elevated than that of P. micra (compare Fig. 9B, D with Fig. 9G). In other populations the operculum is flatter TEXAS PHREATIC HYDROBIIDS 141 ESP MES i XOX - ) A / N \\ es | Wy) \ =. ae y) FE, Y E я © Я === 2 Km un ST Nan re — = = =| ee E > > ST za à =—— : | YA RN NN nn ) \ WIS © À ET A em \ j AS = \ + | EA =.) Г / | SÓN / ь rt fy | = / | NON / KEN Es E a AS (en) a ue = ae FIG. 13. Contents of the pallial cavity of Phreatodrobia nugax (A), P. micra (B), P. conica (C), P. imitata (D), P. plana (E), P. punctata (F), P. rotunda (G), Balconorbis uvaldensis (H), and Stygopyrgus bartonensus (I). The pallial roof has been slit along its length on the extreme right side and then folded over to the left and pinned. The pallial reproductive organs are not shown and the kidney 1$ figured only in A. All scale bars equal 0.25 mm. An = anus; Ct = ctenidium; Ev = efferent branchial vessel; In = intestine; Me = mantle edge; Os = osphradium; Ro = renal opening. and the process reduced or virtually absent (Fig. 9E). Anatomical description is based on exten- sive study of material from San Marcos Springs and lesser study of populations from Barton Springs and wells owned by SWTSU (including live material), Longhorn Portland Cement Company, Union Stockyards, Brack- enridge Zoo, and W. C. Reagan. The snout (Fig. 11) is longer than wide, with folds along its sides. The pink-colored (hae- moglobin) buccal mass 1$ readily seen through the snout. The tentacles are elon- gate, rounded at the tips, moderately thick- ened relative to snout width, and held at 60-100 degrees to one another. Stiffened, elongate cilia fringe the tentacle tips. Four to five hypertrophied ciliary tufts line the outer side of the left tentacle near its base. White granules are clustered at the base of the tentacles and clear crystalline granules ex- tend back along the “neck.” The foot is long and slender, broad anteriorly and tapered posteriorly. The pedal glands consist of a sin- gle massive central gland flanked by seven to eight smaller glands on either side. Crystalline granules are found on the sides of the foot. The shell is carried with the coiling axis tilted about 30 degrees to the plane of the sub- strate. The intestine, gonad (ovary, white; tes- tis, yellow), and pallial gonoduct (oviduct, white; prostate, green) are visible through the shell. The ctenidium typically occupies 75% ofthe 142 HERSHLER 8 LONGLEY FIG. 14. Radulae of P. nugax nugax (A, C-H) and P. nugax inclinata (B). The localities are as follows: A, F (Locality 2), B (Locality 11), C (Locality 3), D, E, G (Locality 4), H (Locality 14). The scale bars equal 0.1 mm (A), 0.01 mm (B-D, F-H), or 0.001 mm (Е). pallial cavity length (Fig. 13A). The ctenidial filaments are triangular in shape (when dis- sected out and laid flat), almost twice as long as wide, well-ciliated, and moderately thick- ened. The broadest part of the filament 1$ positioned at almost two-thirds of the filament length. The osphradium occupies 22-28% of the pallial cavity length in all populations ex- cept that from the Union Stockyards Well (41%), which consists of very small-sized in- dividuals. The osphradium is centered to- wards the posterior end of the ctenidium, with 20-30% of the ctenidium posterior to the end of the osphradium. The central radular tooth is trapezoidal in shape with well developed lateral angles. A „FIG. 15. Radulae of Phreatodrobia micra (A-C; Locality 3), P. rotunda (D, E; Locality 3) and P. imitata (F-H; Locality 20). Scale bars equal 0.01 mm (A, B, D-H) or 0.001 mm (C). second pair of basal cusps on the central tooth was seen only in individuals from San Marcos Springs and the SWTSU Well. The inner marginal tooth is noteworthy for the large number of cusps it possesses (24-34) relative to other congeners. The stomach has a small caecal appendix (Fig. 18A). Note that the anterior arm of the U-shaped intestine loop in the pallial roof bends back posteriorly before turning anteriorly (Fig. 13A) (compare to simpler U-shaped loop in P. micra, Fig. 13B). The loop usually abuts against the long- est gill filaments. The testis (Fig. 19D) consists of simple lobes joining a central basal mass. The vas deferens exits at a point 25-30% back along the testis length. The seminal vesicle (Sv) TEXAS PHREATIC HYDROBIIDS 143 FIG. 16. Radulae of Phreatodrobia plana (А-С; Locality 3) and P. conica (0-Е; Locality 5B). All scale bars equal 0.01 mm. 144 HERSHLER 8 LONGLEY FIG. 17. Radulae of Phreatodrobia punctata (A-D), Balconorbis uvaldensis (E-G), and Stygopyrgus bartonensus (H, |). All scale bars equal 0.01 mm. consists of a few, non-thickened coils largely hidden under the anterior portion of the testis. The anteriormost 52% of the prostate is pal- lial. The penis (Fig. 19A, B) is about three times as long as the snout and has pronounced lobes along much of the inner curvature. The short penial filament lacks folds and has co- lumnar epithelium (Co) along its sides. Long cilia are active along much of the length of the penial filament and extend onto the folds of the inner curvature. The penial folds have a glandular edge (not figured), and occasional glandular clusters (spherical bodies, Fig. 19A) are present along the penial length. The vas deferens undulates slightly in the penis. The large ovary (Ova) lacks pronounced lobation and typically contains six to ten оо- cytes (Oo) of various sizes (Fig. 12). The anterior end of the ovary abuts against the stomach. The oviduct exits near the anterior end of the ovary and disappears beneath the bursa. The capsule gland is twice as long as the albumen gland. The ventral channel 1$ narrow and has a pronounced lateral fold (not figured). The anterior end of the capsule gland terminates in a muscularized S-shaped coil (Fig. 20A) or simple twist (Fig. 20B) to the left side (populations from San Marcos and Bar- ton Springs, SWTSU Well, Longhorn Portland Cement Company Well), or lacks such mod- ifications and has a subterminal capsule gland opening (population from Union Stock- yards Well). The development of the coil cor- relates with body size: larger females from Barton Springs, for instance, have the S- shaped coil while smaller individuals have a simple twist. The oviduct coil lies partly pos- terior to the albumen gland (Fig. 20A). The seminal receptacle opens into the end of the oviduct coil (right side) just posterior to the point where the short bursal duct joins the oviduct (Fig. 20C). The seminal receptacle always has a pink sheen as does the rather swollen oviduct coil, suggesting that the latter may also serve to store sperm. The bursa 1$ flimsy in texture and was easily ruptured dur- ing dissection. HOLOTYPE. ANSP 77574 (Fig. 4А, Е, J). TYPE-LOCALITY. Drift debris of Guadalupe River about four miles above New Braunfels, Comal County (Fig. 1, Locality 6). DISTRIBUTION. Edwards (Balcones Fault Zone) Aquifer, and (possibly) Cow Creek and Glen Rose Aquifers in Travis, Hays, Comal, Kendall, Bexar, and Uvalde Counties (Fig. 1, Localities 1-4, 6, 8-12, 14, 26). VARIATION. The large shell form variation seen in the populations from San Marcos and Barton Springs requires comment. Living specimens of the smaller, flat “form” from San Marcos Springs were always found encrusted with epigean epibionts whereas the more typ- ical trochoid “form” was always found clean. A different habitat is suggested for these two “forms”, with the former perhaps dwelling at or very near to the groundwater outlet. We do not know whether this dimorphism is pheno- typic or genetic. The shells of living speci- mens of all Barton Springs “forms” were al- ways clean. We suspect that this population is highly variable because it is of hybrid origin, with incompletely formed species having been brought into sympatry. More work 1$ needed to sort out the systematics of this polytypic species. DISCUSSION. Phreatodrobia nugax was originally described as a subspecies of P. TEXAS PHREATIC HYDROBIIDS 145 0.50 mm FIG. 18. Stomachs of Phreatodrobia nugax (A), P. rotunda (B), and P. punctata (C). Note the lack of external differentiation of the stomach chambers (Ast, Pst), and the coiling of the intestine (In) on the style sac (Sts) in Band C. The thickened coils of the seminal vesicle (Sv) of P. punctata are also figured in C. Ast = anterior stomach chamber; Cae = caecal appendix; Dga = digestive gland opening; Es = oesophagus; In = intestine; Osv = opening of seminal vesicle into testis; Pst = posterior stomach chamber; Sts = style sac; Sv = seminal vesicle. micra and there has been some question as to whether these taxa are separate species (Fullington, 1978). We collected both species at three localities, two of which were caves from which only empty shells were obtained, the third of which was San Marcos Springs, from which living specimens of both species were obtained. As indicated above, there is more than sufficient anatomical evidence to merit separate species status for the San Marcos Springs populations. Note that at this locality the two species differ very significantly (p < .001, Mann-Whitney U Test) in size (shell width, data from Appendix 2). At the other two localities, Honey Creek Cave and Century Caverns, the size difference is somewhat less significant (p < .05). The shell shape difference between species is pronounced in samples from all three of these localities as well as for the type-specimens (which were also collected together at the same locality): compare Figs. 3F, J, N with 4A, F, J; 3R, S, T with 40; 3M, Q with 4S, W; and 3H, L, P with 4C, H, L. A shell height versus shell width plot for samples from given populations readily separates the two species (Fig. 22). Such pronounced differ- ences in shell size and shape in sympatric populations suggests that two species exist at these localities (as well as in San Marcos Springs). It should be noted, however, that P. nugax can apparently converge on P. micra when found alone. Individuals from Union Stockyards Well have the very low spire, small size, and relatively large osphradium typical of P. micra (but have a complete ctenidium). Phreatodrobia nugax nugax MATERIAL EXAMINED. As for the species, excluding Longhorn Portland Cement Com- pany Well. DIAGNOSIS. Shell variable in size and form. Protoconch without tilt. Aperture rounded, free from or touching the penultimate whorl; 146 HERSHLER 8 LONGLEY FIG. 19. Male reproductive morphology. A, B. Penis (dorsal aspect) of Phreatodrobia nugax. C. Penis (dorsal aspect) of P. rotunda. Note that the penis does not coil on the “neck” as in B. D. Testis (Te), Seminal vesicle (Sv), and prostate (Pr) of P. nugax. Note that about half of the prostate length 15 pallial. Е. Testis (posterior portion not shown), seminal vesicle, and prostate of P. plana. The dashed lines indicate the position of the stomach. Co = columnar epithelium; Emc = posterior end of pallial cavity; Gl = glandular units; Pr = prostate; Sv = seminal vesicle; Te = testis; Vd = vas deferens; Vd, = vas deferens from seminal vesicle to prostate; Vd2 = vas deferens from prostate to penis. inner lip thickened and slightly to moderately flared. Apertural plane with or without slight tilt relative to coiling axis. Phreatodrobia nugax inclinata Hershler & Longley, new subspecies MATERIAL EXAMINED. BEXAR COUNTY: Longhorn Portland Cement Company Well. DIAGNOSIS. Shell (Figs. 3A—E, |) only slightly wider than tall, of globose appear- ance, with protoconch tilted at 20° —30° rela- tive to the teleoconch (Fig. 31). Aperture fused to (not merely touching) the penultimate whorl; inner lip thin and flared only at fusion point. Aperture slightly angled anteriorly, apertural plane highly tilted (>30°) relative to shell axis. REMARKS. The protoconch tilting and apertural peculiarities distinguish this subspe- cies from P. n. nugax. No anatomical differ- ences were seen. HOLOTYPE. ANSP 359089 (Fig. 3A, B). PARATYPES. ANSP A10623D. TYPE-LOCALITY. Longhorn Portland Ce- ment Company Well (Fig. 1, Locality 11). DISTRIBUTION. Thus far known only from the type-locality. ETYMOLOGY. The subspecific epithet re- fers to the inclined or tilted position of the protoconch in this subspecies. TEXAS PHREATIC HYDROBIIDS 147 FIG. 20. Female reproductive morphology of Phreatodrobia nugax (A-C), P. micra (D), P. conica (E), P. plana (Е, С), and P. rotunda (H, |). The left aspect is shown in A, В, О, Е, Е, and H; the right aspect is shown in all other figures. Only the anterior end of the capsule gland (Cg) is shown in B. Note the variation in coiling of the anterior end of the capsule gland (Cg) in P. nugax (A, B). All scale bars equal 0.2 mm. A and B have the same scale. Ag = albumen gland; Bu = bursa copulatrix; Cg = capsule gland; Cga = capsule gland opening; Dbu = duct of bursa copulatrix; Dsr = duct of seminal receptacle; Emc = posterior end of pallial cavity; Oov = opening of oviduct into pallial oviduct; Ov = oviduct; Sr = seminal receptacle. Phreatodrobia rotunda Hershler & Longley, new species ROSE IDE ЭН ЗЕ, 150: E eB! 19С, 20Н, | MATERIAL EXAMINED. HAYS COUNTY: SWTSU Well; San Marcos Springs. DIAGNOSIS. A large-sized species (shell width 2 mm). Shell (Fig. 5E, Е, |, J) planispiral, with flattened base. Operculum (Fig. ЭН, |) multi-whorled and striated. Ctenidium absent (Fig. 13G); osphradium short relative to pallial cavity length (13%). Central tooth of radula with a single pair of basal cusps (Fig. 15D, E). Style sac length 70% of that of stomach (Fig. 18B); intestine with complex coil on style sac (In, Fig. 18B) and narrow, elongate U-shaped loop in pallial roof; long axis of loop parallel to length of pallial cavity (Fig. 13G). Ovary and testis fill small portion (22%, 21%) of digestive gland length. Penis enlarged and without tight coil (Fig. 19C). Bursa duct coiled (Dbu, Fig. 201); oviduct opens into anterior end of albu- men gland. REMARKS. This species is distinguished from other congeners by the following unique character-states: large planispiral shell, stri- ated operculum, complex intestine coil on style sac, and very small-sized ovary and testis. DESCRIPTION. Morphological description is based on study of material from San Mar- cos Springs. The shell has 3.0-3.8 whorls and varies from 1.83-2.16 mm in width. The pro- toconch, hidden (in apertural view) by rapidly expanding teleoconch whorls, has 1.25 148 HERSHLER 8 LONGLEY FIG. 21. Female reproductive morphology of Phreatodrobia imitata (A, B) and P. punctata (C, D). The left aspect is shown in A and C, and the right aspect in B and D. Note the muscularized capsule gland opening of P. imitata (A). All scale bars equal 0.2 mm. Ви = bursa copulatrix; Оби = duct of bursa copulatrix; Emc = posterior end of pallial cavity; Oov = opening of oviduct into pallial oviduct; Sr = seminal receptacle. whorls, the first whorl of which is marked by small pits, followed by a quarter whorl having strong, elevated growth lines (Fig. 7E). The teleoconch has fine growth lines. The aper- tural lip is relatively thin and appears broadly notched adapically when seen from above (Fig. 51). The aperture is rounded and flared above and below. The inner lip is fused to the penultimate whorl. The outer lip is advanced relative to the remaining peristome, often an- gled or twisted (Fig. 5E), and is not flared. Note that the aperture does not extend above the penultimate whorl (Fig. 5E, F). The operculum (Fig. ЭН, 1) is extremely thin, arched into a low cone, and has about five whorls. The ventral surface is smooth. The nucleus is positioned at about 44% of the operculum length. As for other Phreatodrobia that have a thin operculum, the operculum has a very light amber tint. Note the unusual operculum shape, with a sudden narrowing to the right in Fig. 9H. A large number of short, deep striations are arranged in rows which cross the growth lines at a high angle. The central tooth of the radula is quite broad, as the lateral angles diverge at a high angle. The caecal appendix is enlarged com- pared to that of P. nugax (Fig. 18A, B). The intestine coil on the style sac sometimes ex- tends onto the stomach. The U-shaped intes- tine loop in the pallial cavity roof is twice as long as wide and extends far into the anterior half of the pallial cavity (Fig. 13G). The ovary and testis consist of a single solid mass. The vas deferens exits from the ante- rior end of the testis and has a few coils posterior to the stomach. The anteriormost 60% of the prostate is pallial. The penis is about four times as long as the snout and does not coil on the neck, but extends ante- riorly (Fig. 19C). The capsule gland is more than twice as long as the albumen gland (Fig. 20H) and has a slight twist at its anterior end with a terminal opening. The section of ov- iduct anterior to the seminal receptacle is of- ten swollen and has a pink sheen. The coiled bursa duct is shown in Fig. 201. The seminal receptacle is positioned posterior to the bursa TEXAS PHREATIC HYDROBIIDS 149 SHELL HEIGHT (mm) 0.75 Е SHELL WIDTH (mm) FIG. 22. Plot of shell height versus shell width for sympatric populations of Phreatodrobia nugax and P. micra. The populations represented are from San Marcos Springs (open square, P. nugax; filled square, P. micra) and Century Caverns (open cir- се, P. пидах; filled circle, P. micra). Type speci- mens for the two species are indicated by an open (P. nugax) or filled (P. micra) star. Note the clear separation between the species in the plot. and albumen gland, and opens via a short duct anterior to the oviduct coil, and well pos- terior to the point where the bursa duct joins the oviduct (Fig. 20H, |). HOLOTYPE. ANSP 359090 (Fig. 5Е). PARATYPES. ANSP A10623E. TYPE-LOCALITY. San Marcos Springs (Fig. 1, Locality 3). DISTRIBUTION. Edwards (Balcones Fault Zone) Aquifer in Hays County (Fig. 1, Local- ities 3, 4). ETYMOLOGY. The epithet refers to the rounded outline of the shell of this species. Phreatodrobia conica Hershler 8 Longley, new species Figs. 5М-Р, 7С, 8C, 13C, 16D-F, 20Е MATERIAL EXAMINED. COMAL COUNTY: Hueco (A, B) Springs; Honey Creek Cave. BEXAR COUNTY: City Water Board Artesian Station, Well 4. DIAGNOSIS. Shell (Fig. 5M-P) large for ge- nus (shell height 1.7 mm), conical, with a simple aperture and usually with a varix near the end of the body whorl (Fig. 5P). Teleo- conch surface mottled with numerous short ridges. Ctenidium absent, osphradium filling 20% of the pallial cavity length (Fig. 13C). Central tooth of radula square-shaped and without basal cusps (Fig. 16D-F). Style sac length 79% of that of stomach; intestine loop in pallial roof U-shaped and with long axis parallel to pallial cavity length (Fig. 13C). Ovary and testis fill 30-40% of digestive gland length. Pallial portion of prostate relatively small; penis greatly elongate relative to snout. Oviduct enters anterior end of albumen gland. REMARKS. Distinguishing features for this species include the conical shell with unusual sculpture pattern of small ridges, square- shaped central radular tooth, and small pro- portion of pallial prostate. DESCRIPTION. The shell has 3.5-4.0 rounded whorls, a pronounced spire, and var- ies from 1.49-1.86 mm in height. The pits on the 1.5 protoconch whorls are well developed (Fig. 7C). Teleoconch sculpture consists of a large series of short ridges that are aligned at a slight angle to the whorl length. The ridges are occasionally elevated, particularly near the end of the body whorl, producing a mottled effect (Figs. 50, 7C, 8C). Thickened axial growth lines are also present on the last half of the body whorl. Note that the ridges often converge and may be largely or totally absent in worn sections of the teleoconch (Fig. 5N). Seventy-three percent of adult shells from Hueco Springs (A, B) have a varix behind the aperture. The aperture is non-circular, thick- ened all around, and usually slightly sepa- rated from the penultimate whorl in adult shells. Description of operculum and anatomy is based on study of specimens from Hueco (B) Springs. The operculum is flat, with about three whorls, and is slightly longer than wide. The nucleus is positioned at about 41% of the operculum length. Spiral growth lines are well developed on its surface. The lateral angles of the central radular teeth project downwards instead of outwards, resulting in the unusual square shape of the tooth (Fig. 16D—F). Note that the lateral teeth sometimes have 10 or more cusps on one side of the central tooth (Fig. 16E). The stom- ach has a small caecal appendix. The testis, unlobed as is the ovary, fills 42% of the digestive gland length. The seminal vesicle exits from the anterior end of the testis and coils posterior to the stomach. Only the anteriormost 33% of the prostate is pallial. The penis is seven-eight times as long as the 150 HERSHLER 8 LONGLEY snout (preserved specimens) and has a somewhat thickened filament (not figured). The capsule gland is more than twice as long as the albumen gland and its anterior end bends back towards the end of the pallial cavity (Fig. 20E). The capsule gland opening is terminal. The tight oviduct coil is appressed against the posterior portion of the pallial ov- iduct, with the seminal receptacle opening into the right (inner) side of the coil. The short bursa duct joins the oviduct just before the opening into the albumen gland (not figurea). HOLOTYPE. ANSP 359086 (Fig. 5M). PARATYPES. ANSP 359087/A10623B. TYPE-LOCALITY. Hueco (B) Springs, Comal County (Fig. 1, Locality 6). DISTRIBUTION. Edwards (Balcones Fault Zone) and possibly Cow Creek Aquifers in Comal and Bexar Counties (Fig. 1, Localities 6, 9, 13). ETYMOLOGY. The epithet refers to the con- ical shell shape of this species. Phreatodrobia plana Hershler & Longley, new species Figs. 5A-D, С, Н, К, L, 7F, 13E, 16A—C, 19E, 20F, G MATERIAL EXAMINED. HAYS COUNTY: SWTSU Well, San Marcos Springs. COMAL COUNTY: Comal Springs; Natural Bridge Caverns. DIAGNOSIS. A small-sized species with shell width ranging from 0.75—1.1 mm. Shell (Fig. 5A—D, G, H, K, L) planispiral, base flat- tened, aperture extends above penultimate whorl (Fig. 5А-С). Teleoconch sculpture con- sists of thickened, wrinkled collabral lines (Fig. 7F). Ctenidium absent, osphradium fill- ing 33% of pallial cavity length (Fig. 13E). Central tooth of radula without basal cusps; lateral tooth lacking enlarged central cusp (Fig. 16А-С). Stomach length more than twice that of style sac (Fig. 19E); intestine coil in pallial roof complex (Fig. 13E). Ovary and testis fill 50-60% of digestive gland length. Oviduct enters anterior end of albumen gland (Fig. 202). REMARKS. Distinguishing features include the minute planispiral shell with adapically extended apenture, teleoconch sculpture con- sisting of thickened, wrinkled collabral lines, and the complex intestine coil in the pallial roof. DESCRIPTION. The shell has 2.75-3.0 whorls and, while extremely small in the Comal and San Marcos Springs populations (shell width, <0.88 mm), often exceeds 1.00 mm in width in specimens from Natural Bridge Caverns. Note that the shells from Natural Bridge Caverns are considerably more flat- tened than those from San Marcos Springs. The whorls are rounded at the periphery and somewhat angled above and below. The pro- toconch is largely overlapped by the first teleoconch whorl (Fig. 7F). The first half whorl of the protoconch is covered with shallow pits which are joined by the thickened, wrinkled lines on the final protoconch whorl. Note how these thickened lines often converge on the teleoconch and appear plicate in places (Fig. 7F, upper left corner). These collabral lines are not well pronounced on the worn shells collected from Natural Bridge Caverns (Fig. 5D, H, |). The aperture is longer than wide, fused to the body whorl (Fig. 5B, C, D), and thickened all around. While extending above the penultimate whorl, the aperture does not extend below the flattened base. The lip 15 often fluted back ad- and abapically (appear- ing notched when seen from above and be- low), and the outer lip is bent adaxially in specimens from Natural Bridge Caverns (Fig. 5D) and Сота! Springs (not figured). Description of the operculum and anatomy is based on study of specimens from the SWTSU Well. The flat operculum is some- what longer than wide, with about three whorls. The nucleus is positioned at about 40% of the operculum length. The lateral an- gles of the central radular tooth diverge at a high angle, producing an elongate trapezoidal shape for the tooth (Fig. 16B, C). Note the large number of cusps on the lateral tooth (often exceeding 20 for the entire row) and lack of enlarged central cusp. The stomach lacks a caecal appendix. The simple U- shaped intestine loop in the pallial roof, typical of Phreatodrobia, is modified in this species by addition of an extra, final loop outside of the preceding loop (Fig. 13E). Note that the long axes of the loops are oriented perpen- dicular to the pallial cavity length. The testis and ovary are unlobed and fill 59% and 52% of the digestive gland length. The seminal vesicle exits from the anterior end of the testis and coils on the posterior part of the stomach (Fig. 19E). The anteriormost 47% of the prostate 15 pallial. The penis 1$ about three times as long as the snout. The capsule gland is about twice as long as the albumen gland and has a terminal opening TEXAS PHREATIC HYDROBIIDS 151 (Fig. 20F). The oviduct coil is appressed to the left side of the albumen gland: note that the coil is counter-clockwise, not clockwise as 1$ typical for the genus (Figs. 20А, D, E, 214). The seminal receptacle opens into the right side of the oviduct coil just posterior to where the bursa duct enters (Fig. 20G). HOLOTYPE. ANSP 359091 (Fig. 5А). PARATYPES. ANSP A10623F. TYPE-LOCALITY. San Marcos Springs, Hays County (Fig. 1, Locality 3). DISTRIBUTION. Edwards (Balcones Fault Zone) and possibly Glen Rose Aquifers in Hays and Comal Counties (Fig. 1, Localities 3, 4, 8). ETYMOLOGY. The epithet refers to the planispiral shell of this species. Phreatodrobia imitata Hershler 4 Longley, new species Figs. 6A-G, 8A, В, D, 9J, К, 13D, 15F-H, 21А, В MATERIAL EXAMINED. BEXAR COUNTY: Verstraeten Well; O. R. Mitchell Well. DIAGNOSIS. Shell (Fig. 6A-G, 8A, В, D) elongate-conical, height about 1 mm, with highly flared aperture. Teleoconch sculpture consisting of collabral costae and spiral lines. Ctenidium absent (Fig. 13D); osphradium fill- ing 26% of pallial cavity length. Central tooth of radula without basal cusps (Fig. 15Е-Н). Style sac length two-thirds that of stomach; intestine coil in pallial cavity complex (Fig. 13D). Ovary and testis filling 30-40% of di- gestive gland length. Bursa duct coiled (Fig. 21B); oviduct entering posterior end of albu- men gland; capsule gland opening with mus- cularized lip (Fig. 21A). REMARKS. Distinguishing features of this species include the shell sculpture, consisting of collabral costae and spiral lines, complex intestine coil in pallial roof, and muscularized lip Surrounding the capsule gland opening. Partial descriptions for this species were pro- vided by Fullington (1978; for Paludiscala sp.) and Karnei (1978; for Gastropod Genus No. 2, Species 1 and 2) (both unpublished theses). DESCRIPTION. The shell has 3.3-3.5 well- rounded whorls with deep sutures. Shell height averaged 1.01 mm for the Verstraeten Well sample and 1.03 mm for the O. R. Mitchell Well sample. The pits on the 1.0-1.25 protoconch whorls are well-developed (Fig. 6F). Spiral lines begin at the end of the pro- toconch and appear slightly wrinkled under high magnification (Fig. 8D). The lines cross the collabral costae (Figs. 6D, 8B). Low col- labral ridges run between the costae and join the spiral lines (Fig. 8B, D). Costae are some- times absent on the first 0.5-1.0 protoconch whorl (Fig. 6E). The costae are typically low (as in Figs. 6B, C, D, 8A, B): broad, lamelli- form costae were not seen in the sample from O. R. Mitchell Well (n = 14) and were seen in only 10% of a sample from the Verstraeten Well (n = 32). Note that the lamelliform cos- tae are not oriented perpendicular to the whorl surface, but curve (to the left in Fig. 6D). The aperture is rounded, moderately thickened, and highly flared, although flaring of the adapical lip is sometimes reduced (Fig. 6E). The aperture is sometimes loosened from the penultimate whorl. The umbilicus is open (Fig. 6G). Anatomical description is based on study of material from Verstraeten Well. The thin oper- culum (Fig. 9J, K) is about as long as it is wide, with the nucleus positioned at about 40% of the operculum length. The ventral opercular surface is near smooth, with only a very small, low process (Fig. 9K). The central tooth of the radula is trapezoidal in shape (Fig. 15F-H) and has especially nar- row and elongate cusps. The stomach lacks a caecal appendix. The coil of the intestine in the pallial cavity roof (Fig. 13D) is modified by addition of an extra loop to the inside of the preceeding one. Note that the long axes of the loops are oriented parallel to the pallial cavity length. The ovary and testis consist of a single solid mass and fill 34% and 28% of the digestive gland length. The vas deferens exits from the anterior end of the testis and coils posterior to the stomach. The anteriormost 45% of the prostate is pallial. The penis is only twice as long as the snout. The capsule gland is about one and a half times the length of the albumen gland (Fig. 21A). Note that a larger proportion of the capsule gland is pallial than typical for the genus. The narrow capsule gland opening is sub-terminal and surrounded by a muscular lip which measures 0.08 mm x 0.06 mm. The oviduct coil is partly posterior to the albumen gland. The short duct of the bursa loops back onto the right side of the bursa and joins the oviduct just before the opening into the pos- terior end of the albumen gland (Fig. 21B). Note that the bursa is oriented with its long axis parallel to the length of the pallial oviduct. HOLOTYPE. ANSP 359088 (Fig. 6A). PARATYPES. ANSP A10623C. 152 HERSHLER & LONGLEY TYPE-LOCALITY. Verstraeten Well, Bexar County (Fig. 1, Locality 20). DISTRIBUTION. Edwards Balcones Fault Zone) Aquifer in the Von Ormy Section of Bexar County (Fig. 1, Localities 20, 21). ETYMOLOGY. The epithet refers to the con- vergence in shell form between this species and Paludiscala Taylor, 1966. Phreatodrobia punctata Hershler & Longley, new species Figs 6H). 7С, ВЕ, 135 17A—B: 186, Zi. В MATERIAL EXAMINED. TRAVIS COUNTY: Barton Springs. HAYS COUNTY: San Marcos Springs. DIAGNOSIS. A small-sized species, aver- aging 1.13 mm in shell height, with broadly conical shell (Fig. 6H, I) and flaring aperture. Teleoconch surface punctate (Figs. 6l, 7C, 8E). Ctenidium absent, osphradium filling 19% of pallial cavity length (Fig. 13F). Central tooth of radula almost square-shaped, without basal cusps (Fig. 17A—D). Central and lateral teeth without enlarged central cusp (Fig. 17A-D). Style sac length 63% of stomach length; intestine with loop on style sac and complex coil in pallial roof (Fig. 13F). Ovary and testis fill about 30% of digestive gland length. Oviduct enters posterior end of albu- men gland (Fig. 21C). REMARKS. This species is distinguished by its broadly conical shell with punctate teleo- conch sculpture, and unusual morphology of the central and lateral radular teeth. DESCRIPTION. The shell is one and a third times as long as wide, and has four moder- ately rounded whorls. The protoconch and teleoconch surfaces are covered with a series of deep pits surrounded by slight elevations, with sculptural relief more pronounced in the teleoconch. Note that the teleoconch sculp- ture is sometimes arranged into spiral and/or collabral rows (Fig. 61, 7C). The aperture is moderately thickened, pyriform above, rounded below, and flared except where the inner lip fuses with the penultimate whorl (61). Umbilicus present. Description of operculum and anatomy is based on study of material from San Marcos Springs. The flat operculum is one and a half times as long as wide, with the nucleus posi- tioned at about 41% of the operculum length. Spiral growth lines are well developed on the operculum surface. The ventral surface lacks a process or thickening. Note that the lateral angles of the central tooth of the radula project down, imparting an almost square-shape to the tooth (Fig. 17C). The central and lateral radula teeth have as many as 18 and 22 cusps, respectively. The basal process of the central tooth is some- what thickened. The stomach has a small caecal appendix (Fig. 18C). The looping of the intestine onto the right side of the style sac is shown in Fig. 18C. The intestine coil in the pallial roof has an extra loop to the inside of the previous loop with the long axes of the loops oriented parallel to the pallial cavity length. Note that the coil closely resembles that of P. imitata (compare Fig. 13D and F). The ovary and testis, both without lobes, fill 32% and 35% of the pallial cavity length. The seminal vesicle exits from the anterior end of the testis and is composed of several thick- ened loops posterior to the stomach (Fig. 18C). The anteriormost 60% of the prostate is pallial. The capsule gland is over twice as long as the albumen gland, lacks an anterior twist or bend, and has a terminal opening. The oviduct coil is considerably posterior to the end of the albumen gland. The bursa is heart- shaped, and the short duct exits from the end of the shorter axis. The seminal receptacle enters the left side of the oviduct coil. Note that the oviduct widens greatly as it merges with the albumen gland. HOLOTYPE. ANSP 359092 (Fig. 6H). PARATYPES. ANSP A10623G. TYPE-LOCALITY. San Marcos Springs, Hays County (Fig. 1, Locality 3). DISTRIBUTION. Edwards (Balcones Fault Zone) Aquifer in Hays and Travis Counties (Fig. 1, Localities 2, 3). ETYMOLOGY. Theepithetrefers to the punc- tate teleoconch sculpture characteristic of this species. Subfamily Littoridininae Balconorbis Hershler & Longley, new genus DIAGNOSIS. Shell (Fig. 23A, В, Е-Н) minute (width about 1.0 mm), planispiral, transparent, colorless. Protoconch and teleo- conch sculpture consisting of spiral lines (Fig. 7G). Operculum paucispiral with sub-central nucleus. Animal unpigmented and without eyespots. Pallial cavity longer than wide, ctenidium absent, osphradium filling 24% of TEXAS PHREATIC HYDROBIIDS 153 FIG. 23. Shells of Balconorbis uvaldensis (A, В, E-H; Locality 30), Stygopyrgus bartonensis (С, D, 1; Locality 2), Horatia klecakiana (J), and “Horatia” sp. (K,L). Shell widths are as follows: A (1.02 mm, holotype), В (1.15 mm), C (shell height, 0.97 mm, holotype), D (shell height, 0.97 mm), E (0.932 mm), F (0.986 mm), G (0.803 mm), H (1.03 mm), | (0.513 mm), J (1.48 mm), K (1.6 mm), and L (1.83 mm). pallial cavity length (Fig. 13H). Central tooth of radula (Fig. 17E-G) with single pair of basal cusps arising from lateral angles. Intestine with U-shaped loop in posterior portion of pallial roof; long axis of loop almost perpen- dicular to pallial cavity length (Fig. 13H). An- teriormost 47% of prostate pallial. Penis with single spherical lobe on outer curvature bear- ing a large apocrine gland (Fig. 24B). Capsule gland (Cg) with two tissue sections and ter- minal opening (Fig. 25B); posterior end of albumen gland coiled. Sperm pouches ab- sent; sperm stored in anterior coil of oviduct. Spermathecal duct (Sd) issues from posterior end of pallial oviduct (where oviduct enters), joins capsule gland anteriorly. REMARKS. The minute planispiral shell with spiral sculpture and absence of sperm pouches in the female reproductive system distinguish this genus from other littoridinines. TYPE-SPECIES. Balconorbis uvaldensis Hershler 8 Longley, new species. DISTRIBUTION. The Edwards (Balcones Fault Zone) Aquifer in Uvalde County. ETYMOLOGY. The generic name is derived by combining Balcones, referring to the pres- 154 HERSHLER 8 LONGLEY 0.20 mm FIG. 24. Penes (dorsal aspect) of Stygopyrgus bartonensis (A) and Balconorbis uvaldensis (B). The dashed lines in the penial lobes (Plo) of Stygopyrgus bartonensis indicate the narrow distal ends of the ducts in the lobes. Note the lack of undulation of the vas deferens (Vd) in the penes. Both scale lines equal 0.20 mm. Oap = Opening of apocrine gland; Plo = penial lobe; Vd = vas deferens. ence of this genus in the Balcones Fault Zone Figs. 7G, 13H, 17E-G, 23A, В, E-H, 24B, region, with the Latin word orbis, referring to 25B the circular outline of the shell. Balconorbis uvaldensis Hershler & Longley, MATERIAL EXAMINED. UVALDE COUNTY: new species King Farms Well; R. Carnes Well; R. K. TEXAS PHREATIC HYDROBIIDS 155 Oov 0.20 mm FIG. 25. Female reproductive morphology of Stygopyrgus bartonensis (A) and Balconorbis uvaldensis (B). The left aspect is shown in A and the right aspect in B. Note that both species lack a bursa copulatrix and have two tissue sections in the capsule gland (Cg), as indicated by dashed lines. Ag = albumen gland; Cg = Capsule gland; Cga = capsule gland opening; Emc = posterior end of pallial cavity; Oov = opening of oviduct into раша! oviduct; Оу = oviduct; Sd = spermathecal duct; Sr = seminal receptacle. 156 HERSHLER 8 LONGLEY Dunbar Well; S. Moerbe Well; G. Ligocky Well; Uvalde National Fish Hatchery Well. DESCRIPTION. The shell has 2.75-3.0 tu- bular whorls and varies in width from 0.9 to 1.22 mm. Note that the spiral lines, which appear ragged on the protoconch (compared to the teleoconch lines), extend almost to the apex (Fig. 7G); and no other protoconch sculpture is present. The teleoconch also has numerous strong collabral growth lines which cross and often offset the spiral lines. The collabral lines are of highest relief near the aperture (Fig. 23G). The body whorl has about 50 spiral lines. The thin-lipped aperture is wider than long. The peristome is fairly straight while the rest of the lip is rounded. The description of operculum and anatomy is based on study of specimens from G. Ligocky Well. The flat, thin operculum is wider than long, with about 2.5 whorls. The nucleus is positioned at 41% of the operculum length. The central tooth of the radula 15 trapezoidal in shape with widely diverging lateral angles (Fig. 17E-G). Note that the cusps of the cen- tral teeth either come to a point (Fig. 17E, G) or are somewhat rounded (Fig. 17F). The anterior and posterior stomach chambers are well distinguished externally. The stomach lacks a caecal appendix. The anus, typically positioned near the mantle edge in hydrobiids, is about 30% back along the pallial cavity length (Fig. 13H). The ovary and testis are without lobes. The vas deferens exits from the anterior end of the testis and consists of a few non-thickened coils posterior to the stomach. The penis is attached to and coils on the right side of the “neck”, and is twice as long as the snout. Small folds extend for about two-thirds of the penis length from the base along the inner curvature. While attached to the right edge of the penis (about 0.08 mm from the tip), the penial lobe was folded under the penis (to the left side, as in Fig. 24B) in all but one speci- men dissected. In that specimen the penial lobe simply projected to the right of the narrow attachment area. It is possible that the typical condition resulted from muscle contraction during fixation. The near-circular penial lobe measures about 0.10 mm across. The glan- dular opening is terminal and wide (Oap). The glandular lumen is fairly large and the gland is apocrine in type. The vas deferens does not coil in the penis. The anterior end of the ovary abuts against the stomach. The coiled portion of the anterior oviduct has several swellings where sperm is presumably stored. Taking the coiled portion of the albumen gland into account, this gland is equal in length to the capsule gland. The anterior capsule gland section is about half the length of the posterior section and is clear, while the latter is white. Note the blunt anterior end of the capsule gland. The spermathecal duct (Sd) issues from the albumen gland just at the point where the former receives the oviduct. The spermathecal duct is largely ven- tral to the pallial oviduct and fairly thickened. It enters the capsule gland 0.04 mm from the terminal capsule gland opening. HOLOTYPE. ANSP 359084 (Fig. 23A). PARATYPES. ANSP 359085/A10623A. TYPE-LOCALITY. G. Ligocky Well, Uvalde County (Fig. 1, Locality 30). DISTRIBUTION. As for genus (Fig. 1, Local- ities 23, 27, 28, 29, 30, 31). ETYMOLOGY. The epithet refers to the dis- tribution of this species in Uvalde County. Stygopyrgus Hershler & Longley, new genus DIAGNOSIS. Shell (Fig. 23C, D, 1) minute (about 1.0 mm in height), elongate-conic, transparent, colorless. Protoconch sculpture pitted (Fig. 231); teleoconch sculpture consist- ing of spiral lines. Operculum paucispiral; nu- cleus positioned at 36% of operculum length. Pallial cavity about as wide as long, ctenidium absent, osphradium filling 25% of pallial cavity length (Fig. 131). Central tooth of radula with one pair of basal cusps arising from prominent lateral angles. Intestine with U-shaped loop in pallial roof; long axis of loop perpendicular to pallial cavity length (Fig. 131). Ovary and testis filling 33% and 60% of digestive gland length, with ovary covering posterior half of stomach. Half of prostate pallial; slender penis with two glandular lobes on inner curvature (Fig. 24А). Capsule gland with two tissue sections and subterminal, muscularized opening (Fig. 25A). Oviduct coil, spermathecal duct, and seminal receptacle appressed to left side of pallial oviduct; bursa absent. Oviduct and spermathecal duct open jointly into anterior end of albumen gland. Seminal receptacle opens into spermathecal duct; anterior end of spermathecal duct fused with capsule gland opening. REMARKS. The minute, elongate-conical shell and unique configuration of the female reproductive system distinguish this genus from other Littoridininae. TEXAS PHREATIC HYDROBIIDS 157 TYPE-SPECIES. Stygopyrgus bartonensis Hershler 8 Longley, new species. DISTRIBUTION. Thus far known only from Barton Springs, Travis County. ETYMOLOGY. The generic name 1$ derived from the Greek words Stygos, meaning lower world, and pyrgos, meaning tower, and refers to the phreatic habit and elongate shell of this taxon. Stygopyrgus bartonensis Hershler & Longley, new species Figs. 131, 17H, 1, 23С, D, 1, 24A, 25A MATERIAL EXAMINED: TRAVIS COUNTY: Barton Springs. DESCRIPTION. The shell has 4.0—4.6 well- rounded whorls. Shell height varies from 0.97-1.3 mm. Note that the pitted micro- sculpture is best developed on the first half whorl of the protoconch (Fig. 231). A total of about 20 fairly regularly spaced and pro- nounced spiral lines are found on the body whorl. Note that the lines are poorly devel- oped to absent on the adapical third of the whorl. Collabral growth lines are also well developed on the teleoconch, although of lower relief than the spiral lines. The aperture is longer than wide, somewhat angled above, and touches the penultimate whorl. The Пр 15 slightly thickened and does not flare. The um- bilicus is chink-like. The operculum is thin and flat. The central tooth of the radula 1$ trapezoidal in shape, with diverging lateral angles (Fig. 17H, |). The dagger-like cusps of all four tooth types are elongate. The stomach lacks a caecal appen- dix. The anterior and posterior stomach chambers are well-distinguished externally. Both gonads are without lobes. The semi- nal vesicle exits from the anterior end of the testis and consists of a few thickened coils posterior to the stomach. The penis is at- tached to and coils on the right side of the “neck,” and is five times as long as the snout. The posterior half of the penis has deep folds (Fig. 24A). The two penial lobes, of similar appearance, are elongate (three times as long as wide), without folds, with a narrowed tip, and with a terminal pore through which glandular products are secreted. The glandu- lar lumen is fairly large, filling about half to two-thirds of the penial lobe, and terminate distally in a narrow neck. The general appear- ance of the glandular lobe 1$ similar to that described for Mexipyrgus Taylor, 1966 (Taylor, 1966; Hershler, 1985). A few glan- dular clusters are scattered throughout the penis and the penial folds have glandular edges. The small section of penis between the lobes and tip is lined (at the edges) with ciliated columnar epithelia. The vas deferens does not coil in the penis. Small spherical epibionts were seen clinging to the distal ends of the penial lobes. The capsule gland is slightly larger than the albumen gland (Fig. 25A). The albumen gland is clear. The anterior section of the capsule gland narrows somewhat and 1$ clear, while the much larger posterior section is yellow. The capsule gland opening is very slightly posterior to the anterior end of the gland. The U-shaped oviduct loop is centered at the junc- tion between the albumen and capsule glands. The narrow posterior extension of the fused oviduct and spermathecal duct opens into the anterior end of the albumen gland on its left side. The seminal receptacle has a pink sheen and is less than 10% of the pallial oviduct length. Note that the seminal recep- tacle is positioned anterior to the end of the pallial cavity. The spermathecal duct joins the posterior side of the muscular capsule gland opening. HOLOTYPE. ANSP 359093 (Fig. 23C). PARATYPES. ANSP A10623H. TYPE-LOCALITY. Barton (“Concession”) Springs (Fig. 1, Locality 2). DISTRIBUTION. As for genus. ETYMOLOGY. The epithet refers to the type-locality. DISCUSSION Systematic relationships. À comparison among the nine species considered in this paper, as well as phreatic “Orygoceras” sp. (also known from south-central Texas), т- volving 40 binary-coded characters (72% from anatomy), is given in Appendix 4. A phenogram based on these data is shown in Fig. 26. As seen in the phenogram, the seven Phreatodrobia spp. form a cluster quite dis- tinct from the remaining three species, all of which are monotypic littoridinines. The de- scription of Balconorbis and Stygopyrgus brings the total of described phreatic littoridi- nine genera in the Western Hemisphere to four: Paludiscala and Coahuilix were previ- ously described from (and considered en- 158 HERSHLER 8 LONGLEY 0.8527 0.781+ а | 0.568 0.498 0.427 0.356 0.285 0.214 We 2 ey 4) GG 7 eh IE FIG. 26. Phenogram showing similarities among 10 species of phreatic Hydrobiidae from south-central Texas. The numbering of species is as in Appendix 4, which is also the source of the data used to generate the phenogram. demic to) the Cuatro Ciénegas Basin in north- ern Mexico (Taylor, 1966; Hershler, 1984, 1985). A comparison among these taxa in- volving 16 morphological features is given in Table 5 and the anterior portion of their female reproductive systems is schematically dia- grammed in Fig. 27. “Orygoceras” sp. is not included in this comparison as it has a number of unique character-states and probably rep- resents an offshoot within the Littoridininae separate from the lineage(s) to which these other four genera belong to. The most similar pair among these four genera is Paludiscala and Coahuilix. Their penes are nearly identi- cal as is the groundplan of their female repro- ductive systems, typified by the possession of a large-sized bursa, loss of the seminal re- ceptacle (although a presumably secondarily- derived one is found in Paludiscala), and joint opening of the oviduct and sperm duct into the posterior section of the albumen gland (Fig. 27A, B). We believe that these two genera are closely related and belong to the same lin- eage within the Littoridininae. The penis of Balconorbis is the same type as that of Paludiscala and Coahuilix, with a single spherical lobe on the outer curvature bearing an apocrine gland, and its female reproductive system (Fig. 27C) could repre- sent a modification of that of the latter two genera involving loss of the bursa and result- ing anterior extension of the sperm (now ef- fectively a spermathecal) duct to join the cap- sule gland. Other character-states found in Balconorbis, notably the spiral protoconch mi- crosculpture, could mitigate against a close phyletic relationship of this genus with those from Cuatro Ciénegas, although the value of this feature in separating higher taxa has re- cently been questioned (Hershler, 1985). The mammiform penial glands and aspects of the female reproductive system (presence of a seminal receptacle and muscularized capsule gland opening, position of oviduct coil on left side of pallial oviduct) of Stygopyrgus readily distinguish this genus from the other three. Stygopyrgus may belong to a separate (from the above) littoridinine invasion of the phreatic habitat, perhaps from an ancestor belonging to the Mexipyrgus-Durangonella group (Hershler, 1984). Note that Mexipyrgus and Durangonella also have a muscularized anterior portion of the capsule gland and elon- gate, glandular (mammiform in Mexipyrgus) penial lobes. In conclusion, it appears that Balconorbis and Stygopyrgus may have close relatives among phreatic or epigean hydrobiids of the southwestern United States and northern Mexico. Phreatodrobia nugax and P. micra were previously considered congeners of Horatia s.s. and Hauffenia s.s. from Europe. This is not an unrealistic possibility: such a situation exists for two Edwards Aquifer crustacean genera, Monodella Maguire and Palaemon- etes Heller, both of which are considered to be of Tethyan origin and of brackish-water or marine ancestry (Stock, 1976; Strenth, 1976). The Hydrobiinae of Europe and North Amer- ica also probably share a common, brackish- water ancestor (Johannson, 1956; С. М. Davis, 1979). A number of general similarities are seen between the above two sets of hydrobiid taxa. Note that Horatia and Hauffenia also have a low trochoid-planispiral shell (Binder, 1957, fig. 1; Pollonera, 1898, fig. 2); intestinal loop (simple) in the pallial roof (Boeters, 1974, fig. 3; Hershler, personal observations); bursa copulatrix positioned largely posterior to the albumen gland; oviduct coil located on the left TEXAS PHREATIC HYDROBIIDS 159 TABLE 5. Comparison of 16 morphological features among phreatic littoridinine genera of Texas and Mexico. Data regarding Paludiscala and Coahuilix are from Hershler (1985). Balconorbis Stygopyrgus Paludiscala Coahuilix 1. Maximum shell size 1.22 1:30 2.60 1.37 (height or width, mm) 2. Shell form planispiral elongate-conic elongate-conic planispiral 3. Protoconch sculpture spiral lines punctate punctate punctate 4. Teleoconch sculpture spiral lines spiral lines collabral absent costae 5. Ctenidium absent absent present present” 6. Intestine loop in pallial present present absent absent roof 7. Number of penial lobes 1 2 1 1 8. Position of penial lobe(s) outer curvature inner outer outer of penis 9. Penial gland type apocrine mammiform apocrine apocrine 10. Anterior oviduct coil ventral to pal- on left side of absent present lial oviduct pallial oviduct posterior tip of albumen gland 11. Oviduct opens into 12. Bursa copulatrix absent 13. Seminal receptacle absent 14. Openings of spermathecal fused duct and capsule gland 15. Number of capsule gland 2 tissue sections 16. Capsule gland opening simple *Ctenidium absent in С. hubbsi, present in С. landyei. **Seminal receptacle secondarily derived. side of the pallial oviduct, and seminal recep- tacle opening into the oviduct coil (Radoman, 1966, fig. 8; Hershler, personal observations). Such general similarities also extend to the radula (see Fig. 27C, D for European genera) and protoconch microsculpture (Fig. 71). The similarity of shell form, however, is superficial, the shell of the European genera (Fig. 23.) is larger, thicker, and more globose than that of any Phreatodrobia. Both Horatia and Hauffe- nia have penial lobes (Boeters, 1974, fig. 2; Giusti et al., 1981, fig. 1), and the penial sur- face has complex glandular swellings (Her- shler, personal observations), character- states not seen in any Phreatodrobia. No Eu- ropean hydrobiid has the thickened opercular process seen in the two Phreatodrobia spp. Other character-states seen in some Phreatodrobia, but not in any European hydrobiids, include a highly complex intestinal coil in the pallial roof, lack of basal cusps on anterior end of albumen gland posterior sec- tion of albumen gland posterior sec- tion of albumen gland absent present present present present** absent fused fused separate 2 3 2 muscularized simple muscularized the central tooth of the radula, and a coiled anterior end of the capsule gland. These dif- ferences suggest that Phreatodrobia repre- sents a separate adaptive radiation meriting generic distinction from Horatia, Hauffenia, and other European hydrobiines. Several other phreatic taxa are also known from North America, notably Antrobia Hubricht, Fontigens Pilsbry, Antroselates Hubricht, and “Horatia” (Hubricht, 1940). While none of these taxa has received de- tailed anatomical study, a limited comparison can be made with Phreatodrobia. Fontigens, while having a hydrobiine-type female repro- ductive system (Hershler, personal observa- tions), is clearly separated from all other North American Hydrobiidae by its unique penis, which has two accessory glands fed by thick- ened ducts which run through the penis base to end blindly in the nuchal cavity. Antrosel- ates, while lacking basal cusps on the central 160 Sl : / с y / ae À A wen À A Æ AS NA \ eas о === ра « Sd | Оу HERSHLER 8 LONGLEY Ov. FIG. 27. Schematic representation of the female reproductive systems of Paludiscala (A), Coahuilix (B), Balconorbis (C), and Stygopyrgus (D). The right aspect is shown in all cases. In D, the arrow indicates the opening of the oviduct (Ov) into the pallial oviduct. Ag = albumen gland; Bu = capsule gland; Cga = capsule gland opening; Ov = oviduct; Sd = spermathecal duct; Sdu = Sr = seminal receptacle. tooth of the radula (Hubricht, 1963), as do some Phreatodrobia, is much larger (shell height, > 5 mm) and has an elongate, high- spired shell (Hubricht, 1963, pl. 8). “Horatia”, collected from Manitou Cave in Alabama (Hubricht, 1940), has a near-planispiral shell (Figs. 23K, L) and an intestinal loop in the pallial roof (Hershler, personal observations), but differs from Phreatodrobia in having spiral lines on the protoconch (Fig. 7H). Note that the central teeth of the radula of both “Hora- tia” (Fig. 28A) and Fontigens (Fig. 28B) have short cusps, contrasting with the typically elongate cusps of Phreatodrobia. Antrobia has a hydrobiine female reproductive system, a slight intestinal loop in the pallial cavity, and a simple penis (Hubricht, 1971; Hershler, per- sonal observations), yet differs from Phreatodrobia in having a thickened, globose, amnicolid-like shell (Hubricht, 1971, figs. 4—6) and spiral lines on the protoconch bursa copulatrix; Cg = sperm duct; (Hershler, personal observations). On the ba- sis of the limited data available, we conclude that while Phreatodrobia is probably not closely related to either Fontingens or Antros- elates, it may belong to the same lineage as one or more of the remaining two taxa. It should also be pointed out that there are no known hydrobiines among epigean freshwa- ter hydrobiids of North America. As indicated in the phenogram, two pairs of Phreatodrobia spp. link closely: P. micra and P. nugax (0.175); and P. conica and P. punc- tata (0.200). Phreatodrobia micra and P. nugax share distinctive character-states that include presence of a ventral operculum pro- cess, an incomplete or complete ctenidium, and basal cusps on the central tooth of the radula. Note, however, that some populations of P. nugax have a smooth operculum. The shell similarity between these two species has been discussed above. TEXAS PHREATIC HYDROBIIDS 161 FIG. 28. Radulae of “Horatia” sp. (A), Fontigens nickliniana (B), Horatia klecakiana (C), and Hauffenia subpiscinalis (D). The scale bars in A and B equal 0.1 mm, while those in C and D equal 0.01 mm. Phreatodrobia conica and P. punctata are distinguished from the above species pair by the following features: teleoconch sculpture consisting of low swellings or ridges; central tooth of radula square-shaped; lateral tooth of radula with numerous cusps; intestinal loop in pallial roof complex; and oviduct opening into posterior tip of albumen gland. The remaining three species, however, have a mosaic of character-states from the above two sets. Phreatodrobia rotunda has basal cusps on the central tooth of the radula, but has a complex pallial intestinal loop. Phreatodrobia imitata has a trapezoidal-shaped central tooth, but lacks basal cusps. The radula and pallial intestinal loop of Phreatodrobia plana are similar to those of P. conica and P. punc- tata, yet the organization of the bursa copu- latrix complex in this species is near-identical to that of the other species pair. Morphological diversity is marked within the genus, and a large number of unusual char- acter-states are spread out among the vari- ous congeners. The morphological distinc- tiveness of the various species may explain the sympatry of five congeners in the aquifer beneath San Marcos Springs. Considering the mosaic pattern of character-state distribu- 162 HERSHLER 8 LONGLEY tion among congeners, we feel that subge- nera should not be recognized, although we suspect that P. micra and P. nugax are par- ticularly closely related. The problem of convergence. Several mor- phological features are probably particularly prone to convergence among phreatic hydrobiids. Blindness and lost body pigment are associated with invasion of the phreatic habitat (Culver, 1982) and have likely oc- curred in diverse hydrobiid lineages. Small body size, typical of phreatic hydrobiids, may also be associated with their purportedly food- poor environment (Culver, 1982). Several character-states are highly correlated with small body size, notably loss of the ctenidium, loss of sperm pouches, and looping of the intestine in the pallial roof. Again, shared pos- session of such character-states may not in- dicate phyletic affinity. If the systematist dis- regards all possible convergent character- states among these tiny snails, there may be rather few character-states from gross mor- phology remaining, as the anatomy of minute hydrobiids is rather simplified. Clarification of the systematic relationships of phreatic hydrobiids may only come when histological study of morphology is applied to the taxa in question. Not only does examination of tissue sections provide data on additional character- states, but it also can resolve whether given structures are homologous (a major concern when structures can be lost and regained), as convergence is unlikely to be precise at the cellular level. For examples of such studies applied to systematic relationships of ris- soaceans, see Ponder (1984). Distribution and habitat. The distribution of species (and subspecies) is shown in Fig. 29. Note that the fauna includes what may be locally endemic species as well as much more widespread species, a pattern also seen in the phreatic amphipod fauna of south-central Texas (Holsinger, 1967; Holsinger & Longley, 1980). While all species are found in the Edwards (Balcones Fault Zone) Aquifer, ma- terial collected from three localities in the drainage zone (Fig. 1, Localities 8-10) may not be from this aquifer. All three localities are wet caves, and all yielded only fresh shells. The source of water for the permanent streams in these caves is probably as follows: Natural Bridge Caverns (Glen Rose Forma- tion), Honey Creek Cave (Cow Creek Forma- tion), and Century Caverns (Lower Glen Rose Formation) (Knox, 1981; J. Knox, personal communication, 1984). Both the Glen Rose and Cow Creek are also Cretaceous lime- stone, but are members of the Trinity Group underlying the Edwards (Ashworth, 1983). While snail populations may not be living in the cave waters (the shells could have been washed in from elsewhere), it is still unlikely that the shells came from populations living in the Edwards (Balcones Fault Zone) Aquifer, given the sporadic occurrence of this aquifer in this region (Ashworth, 1983). It is therefore likely that the four species collected from these caves are found in aquifers other than the Edwards (Balcones Fault Zone). Note that Taylor (1974) collected living “Horatia” (pos- sibly P. micra or P. nugax) from a spring in Real County, also in the drainage zone of the Edwards (Balcones Fault Zone) Aquifer. Also note that other invertebrate species (or sister species thereof) are found in the Edwards (Balcones Fault Zone) Aquifer as well as other aquifers in the Hill Country or Edwards Pla- teau (Mitchell & Reddell, 1971). All of the hydrobiid species may have considerably wider ranges than outlined above as the aqui- fers of south-central Texas have not been well sampled (see below). The 14 artesian wells that yielded snails ranged in depth (beneath ground level) from 59-582 m (Table 2). All of these wells are tightly cased and there is no doubt that the snails were expelled from the deep artesian zone. Their habitat probably includes frac- tures, joints and caverns in the bedrock; and possibly, given the minute size of the snails, even interstices. Note, however, that snails were absent from eight wells in Bexar County (where three species occur), five of which yielded other troglobites (Table 2), a point which mitigates against common use of the interstitial habitat by the snails. It is likely that all or most of the species dwell in similar habitats in the recharge zone, where the aqui- fer is unconfined. Faunal diversity in the Edwards (Balcones Fault Zone) Aquifer. The groundwater fauna of Texas has traditionally been sampled by collecting in wet caves. A tremendous effort has gone into such collecting and the aquatic fauna of caves in several physiographic re- gions of the state is well known (Mitchell & Reddell, 1971). Yet caves only offer a very small fraction of the total phreatic habitat. The deep artesian zone, for instance, is probably not accessible from caves. The recent appli- cation of sampling techniques involving plac- TEXAS PHREATIC HYDROBIIDS 163 EDWARDS (Balcones Fault Zone) AQUIFER REGION AUSTIN SUBREGION SAN ANTONIO SUBREGION FIG. 29. Distribution of the 10 species and subspecies found in the Edwards (Balcones Fault Zone) Aquifer. 1 = Phreatodrobia nugax nugax; 2 = P. п. inclinata; 3 = Р. micra; 4 = P. сотса; 5 = P. plana; 6 = P. rotunda; 7 = P. punctata; 8 = P. imitata; 9 = Stygopyrgus bartonensis; 10 = Balconorbis uvaldensis. ing nets over artesian well (for depths, see Table 2) or spring outlets has yielded many new species (see above) and demonstrated that a deep phreatic fauna does exist in some areas. Note that while the SWTSU Well (59 m deep) has yielded 20 troglobitic species, in- tensive collecting in the pool at the bottom of nearby Ezzell’s Cave (in the same phreatic pool of the aquifer) only yielded nine species (J. D. Davis, 1979). With the description of the seven new species in this paper, the total described troglobitic fauna of the aquifer now totals 39 species. Collections from the locali- ties considered in this paper have also yielded, apart from the snails, an additional 10-15 undescribed invertebrate species. Given the paucity of such sampling in most parts of the aquifer, a much larger number of species may yet await discovery. There is a large potential for discovery of additional new taxa in other aquifers of south-central Texas. The huge Edwards (Plateau) Aquifer, for in- stance, may contain a large number of sister taxa of the Edwards (Balcones Fault Zone) species, given that the two aquifers formed a single unit until at least the Miocene. Contin- ued collecting in caves augmented by wide- spread application of the above sampling techniques will be necessary to more com- pletely sample these and other aquifers in south-central Texas. ACKNOWLEDGMENTS The staff of the EARDC is thanked for their help with numerous aspects of the project. The senior author thanks the junior author for providing funds and facilities during an ex- 164 HERSHLER 8 LONGLEY tended stay at the EARDC. We thank numer- ous individuals for allowing us to collect on their property. The following individuals lent material from either their personal or institu- tional collections: Drs. G. M. Davis, Academy of Natural Sciences of Philadelphia (Phreatodrobia micra, cotypes, ANSP 91322; Phreatodrobia nugax, holotype, ANSP 77574); A. Solem, Field Museum of Natural History (Antrobia culveri, holotype and paratypes, FMNH 164171, 164170/15); F. Giusti (Horatia klecakiana, Hauffenia subpis- cinalis); and Mr. Leslie Hubricht (“Horatia”, Fontigens nickliniana, Phreatodrobia nugax from Salamander Cave). Additional funding for the project came from grants to the senior author by the National Speleological Society and the United States Fish and Wildlife Ser- vice (Contract No. 14-16-0002-84-228, Amendment No. 1). LITERATURE CITED ASHWORTH, J. B., 1983, Ground-water availability of the Lower Cretaceous Formations in the Hill Country of south-central Texas. Texas Depart- ment of Water Resources Report, 273: 1-172. BINDER, E., 1957, Note sur le genre Horatia. Jour- nal de Conchyliologie, 97: 59-62. BOETERS, Н. D., 1974, Ногайа Bourguignat, Plag- igeyeria Tomlin und Litthabitella Boeters. Archiv für Molluskenkunde, 104: 85-92. BOURGUIGNAT, J. R., 1887, Etude sur les noms generiques des petits Paludinidees a opercule spirescent, suivie de la description d'un nouveau genre Horatia. Tremblay, Paris, 56 p. BOWMAN, Т. E. & LONGLEY, G., 1976. Redescr- iption and assignment to the new genus Lirceo- lus of the Texas troglobitic water slater, Asellus smithi (Ulrich) (Crustacea: Isopoda: Asellidae). Proceedings of the Biological Society of Wash- ington, 88: 489—496. BURCH, J. B., 1982, Freshwater snails (Mollusca, Gastropoda) of North America. Environmental Monitoring and Support Laboratory, Office of Re- search and Development, United States Environ- mental Protection Agency, 294 p. CLIMO, F. M., 1974, Description and affinities of the subterranean molluscan fauna of New Zealand. New Zealand Journal of Zoology, 1: 247-284. CLIMO, F. M., 1977, Notes on the New Zealand hydrobiid fauna (Mollusca: Gastropoda: Hydrobiidae). Journal of the Royal Society of New Zealand, 7: 67-77. CULVER, D. C., 1982, Cave life. Evolution and ecology. Harvard, University Press, Cambridge, Mass., 189 p. HERSHLER, R. & LONGLEY, G., in press, Hado- ceras taylori, a new genus and species of phreatic Hydrobiidae (Gastropoda: Rissoacea) from south-central Texas. Proceedings of the Biological Society of Washington. 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., MAZURKIEWICZ, M. & MANDRAC- CHIA, M., 1982, Spurwinkia: morphology, sys- tematics, and ecology of a new genus of North American marshland Hydrobiidae. Proceedings of the Academy of Natural Sciences of Philadel- phia, 134: 143-177. DAVIS, J. D., 1979, A faunal survey of Ezell’s Cave, Hays County, Texas. M. S. Thesis, Southwest Texas State University, San Marcos, 96 p. DAVIS, J. R., 1983, An additional record of living Orygoceras (Hydrobiidae) from Texas. Nautilus, 97: 112-113. FRETTER, V. & GRAHAM, A., 1962. British prosobranch molluscs. Ray Society, London, xvi + 755 p. FULLINGTON, R. W., 1978, The Recent and fossil freshwater gastropod fauna of Texas. Ph.D. Thesis, North Texas State University, Denton., 279 p. GIUSTI, Е. & PEZZOLI, E., 1981, Notulae malaco- logique XXV. Hydrobioidea nuove o poco conos- ciute dell'Italia appenninica (Gastropoda: Prosobranchia). Archiv fúr Molluskenkunde, 111: 207-222. GIUSTI, F., PEZZOLI, E. 8 BODON, M., 1981, Notulae malacologique, XXVII. Primo contributo alla revisione del genre /slamia (Radoman, 1973) in Italia. У Convegno Nazionale della Società Malacologica Italiana: 49-71. GIUSTI, F. & BODON, M., 1984, Notulae malaco- logique dell’ Italia nord-occidentale. Archiv für Molluskenkunde, 114: 157-181. HART, C. W., Jr., 1978, A new species of the genus Sphaeromicola (Ostracoda: Entocytheridae: Sphaeromicolinae) from Texas, with notes on relationships between European and North American species. Proceedings of the Biological Society of Washington, 91: 724—730. HERSHLER, R., 1984, The hydrobiid snails (Gas- tropode: Rissoacea) of the Cuatro Ciénegas Ba- sin: systematic relationships and ecology of a unique fauna. Journal of the Arizona-Nevada Academy of Science, 19: 61-76. HERSHLER, R., 1985, Systematic revision of the Hydrobiidae (Gastropoda: Rissoacea) of the Cuatro Ciénegas Basin, Coahuila, Mexico. Malacologia, 26: 31-123. HERSHLER, R. 8 LONGLEY, G., in press, Hado- ceras taylori, a new genus and species of phreatic Hydrobiidae (Gastropoda: Rissoacea) from south-central Texas. Proceedings of the Biological Society of Washington, 99. HOLSINGER, J. R., 1967, Systematics, speciation, TEXAS PHREATIC HYDROBIIDS 165 and distribution of the subterranean amphipod genus Stygonectes (Gammaridae). United States National Museum Bulletin, 259: 1-176. HOLSINGER, J. В. & LONGLEY, G., 1980, The subterranean amphipod crustacean fauna of an artesian well in Texas. Smithsonian Contribu- tions to Zoology, 308: 1-62. HUBRICHT, L., 1940, A subterranean snail from an artesian well. Nautilus, 54: 34-35. HUBRICHT, L., 1963, New species of Hydrobiidae. Nautilus. 76: 138-140. HUBRICHT, L., 1971, New Hydrobiidae from Ozark caves. Nautilus, 84: 93-96. JOHANNSON, J., 1956, Genital organs of two Alvania species and a comparison with related families (Moll. Pros.). Arkiv för Zoologi, 9: 377-387. KARNEI, H. S., 1978, A survey of the subterranean aquatic fauna of Bexar County, Texas. M.S. The- sis, Southwest Texas State University, San Mar- cos, 116 p. KLEMT, W. B., KNOWLES, T. R., ELDER, С. В. 8 SIEH, T. W., 1979, Ground-water resources and model applications for the Edwards (Balcones Fault Zone) Aquifer in the San Antonio Region, Texas. Texas Department of Water Resources Report 239: 1-88. KNOX, J., 1981, Natural Bridge Caverns. Texas Caver, 26: 84-87. KURODA, T. & HABE, T., 1958, Troglobiontic aquatic snails from Japan. Venus, 19: 183-196. LONGLEY, G., 1975, Environmental assessment Upper San Marcos River Watershed. Report to the Soil Conservation Service, Temple, Texas, 367 p. LONGLEY, G., 1978, Status of the Texas blind salamander. Endangered Species Report 2, United States Fish and Wildlife Service, p. 1-45. LONGLEY, G., 1981, The Edwards Aquifer: earth's most diverse groundwater ecosystem? Interna- tional Journal of Speleology, 11: 123-128. MACLAY, R. W. 8 SMALL, T. A., 1976, Progress report on geology of the Edwards Aquifer, San Antonio area, Texas, and preliminary interpreta- tion of borehole geophysical and laboratory data on carbonate rocks. United States Department of the Interior, Geological Survey, Open File Report 76-627, 65 p. MITCHELL, В. W. & REDELL, J. R., 1971, The invertebrate fauna of Texas caves. In: LUN- DELIU, E. |. & SLAUGHTER, В. H., (eds.), Nat- ига! History of Texas Caves, р. 35-90 Gulf Nat- ural History, Dallas. MORRISON, J. P. E., 1949, The cave snails of eastern North America. Annual Report of the American Malacological Union for 1948: 13-15. РЕ В: М, Jenks GEORGES \/ © 1956, Ground-water resources of the San Antonio area, Texas. Texas Board of Water Engineers, Bulletin No. 5608, 1-2: 1-859. PILSBRY, H. A., 1916, Note on Valvata micra. Nautilus, 30: 83-84. PILSBRY, H. A. 8 FERRISS, J. H., 1906, Mollusca of the southwestern states. Part Il. Proceedings of the Academy of Natural Sciences of Philadel- phia, 58: 123-175. POLLONERA, C., 1898, Intorno ad alcune Conch- iglie del Friuli. Bolletino Musei di Zoologia ed Anatomia comparata, 13: 1-4. PONDER, W. F., 1966, On a subterranean snail and a tornid from New Zealand. Journal of the Malacological Society of Australia, 10: 35-40. PONDER, W. F., 1984, Review of the genera of the Barleeidae (Mollusca: Gastropoda: Rissoacea). Records of the Australian Museum, 35: 231-281. RADOMAN, P., 1966, Die Gattungen Pseu- damnicola und Horatia. Archiv fúr Mollusk- enkunde, 95: 243-253. RADOMAN, P., 1973, New classification of fresh and brackish water Prosobranchia from the Balkans and Asia Minor. Prirodnjacki Musej u Beogradu Posebna Izdanja, 32: 1-30. REDDELL, J. R., 1965, A checklist of the cave fauna of Texas. |. The Invertebrata (exclusive of Insecta). Texas Journal of Science, 17: 143-187. REDDELL, J. R., 1967, A checklist of the cave fauna of Texas. Ill. Vertebrata. Texas Journal of Science, 19: 184-226. REDDELL, J. R., 1970, A checklist of the cave fauna of Texas. IV. Additional records of Inver- tebrata (exclusive of Insecta). Texas Journal of Science, 21: 389-415. REDDELL, J. В. & MITCHELL, В. W., 1969, A checklist and annotated bibliography of the sub- terranean aquatic fauna of Texas. Texas Tech- nological College, Water Resources Center (Lubbock), Special Report 24: 1-48. SMITH, A. R., 1971, Cave and karst regions of Texas. In LUNDELIUS, Е. L. & SLAUGHTER, В. H. (eds.), Natural history of Texas caves, p. 1-14. Gulf Natural History, Dallas. STOCK, J., 1976, A new genus and two new spe- cies of the crustacean order Thermosbaenacea from the West Indies. Bijdragen tot de Dierkunde, 46: 47-70. STRECKER, J. K., 1935, Land and freshwater snails of Texas. Transactions ofthe Texas Acad- emy of Science, 17: 5-44. STRENTH, М. E., 1976, A review of the systematics and zoogeography of the freshwater species of Palaemonetes Heller (Crustacea: Decapoda) of North America. Smithsonian Contributions to Zo- ology, 228: 1-27. TAYLOR, D. W., 1966, A remarkable snail fauna from Coahuila, Mexico. Veliger, 9: 152-228. TAYLOR, D. W., 1974, The Tertiary gastropod Orygoceras found living. Archiv für Mollusk- enkunde, 104: 93-96. TAYLOR, D. W., 1975, Index and bibliography of Late Cenozoic freshwater Mollusca of Western North America. Museum of Paleontology, Univer- 166 HERSHLER 8 LONGLEY sity of Michigan, Papers on Paleontology, 10: 1-384. THOMPSON, F. G., 1984, North American fresh- water snail genera of the hydrobiid subfamily Lithoglyphinae. Malacologia, 25: 109-141. VANDEL, A., 1965, Biospeleology. The biology of cavernicolous animals. International Series of Monographs in Pure and Applied Zoology, 22: 1-524. YOUNG, F. N. & LONGLEY, G., 1976, A new sub- terranean aquatic beetle from Texas (Coleop- tera: Dytiscidae-Hydroporinae). Annals of the Entomological Society of America, 69: 787-792. APPENDIX 1. Data for all Texas localities consid- ered in this paper. Numbers (in parentheses) refer to locations in Fig. 1. The collector and date(s) of collection are also given. TRAVIS COUNTY: Salamander Cave (1), J. Red- dell, 20 Ill 1966, 30°22'04" М, 97°45'12” W; Ваг- ton Springs (2), T. Spinelli, VI-VII 1982, В. Her- shler, VI-VIII 1984, 30°16’ N, 97°47' W. HAYS COUNTY: San Marcos Springs (3), J. Davis, VIII-IX 1979, 29°54' М, 97°56’ W; Artesian Well at Southwest Texas State University (4), G. Longley, IV-IX 1976, 29°53'24" N, 97°56'08" W. COMAL COUNTY: Hueco Springs (A, B) (5), M. Brzozowski, VI-VIl 1981, 29°46’ N, 98°08’ W; Guadalupe River drift (6), Pilsbry & Ferriss (1906), 29°45'30” М, 98°08'30"” W; Comal Springs (main spring) (7), M. Brzozowski, VII- МИ 1981, 29°42’ М, 98°08’ W; Natural Bridge Caverns (River Styx at bottom of cave) (8), R. Hershler, 7 IX 1984, 29°41'22” М, 98°20'30" W; Honey Creek Cave (seeps feeding main spring from cave) (9), В. Hershler, 10 IX 1984, 29°50’ М, 98°30’ W. KENDALL COUNTY: Century Caverns (stream at bottom of cave) (10), В. Hershler, 1 VIII 1984, 29°53'20" М, 98°37' W. BEXAR COUNTY: Longhorn Portland Cement Co. Well (#2) (11), М. Brzozowski, И-М 1981, 29°32'05” М, 98°24 W; Brackenridge Zoo Well (12), City Water Board Artesian Station, Well 4 (13), М. Brzozowski, X 1980—VIII 1981, 29°25'48” М, 98°26'14” W; Union Stock- yards Well (#3) (14), M. Brzozwski, 8-17 XIII 1980, 29°24'10” М, 98°30'55” W; City Water Board Mission Station Well (15), M. Brzozowski, 4-25 V 1981, 29°13'18" М, 98°29'49" W; J. H. Uptmore Well (#5) (16), М. Brzozowski, ХИ 1980—Ш 1981, 29°12'11” М, 98°43’40” W; Lackland City Water Co. Well (17), В. Rutland, VII-VI 1979, 29°21' М, 98°36'58" W; City Public Service Board Well (#1) (18), В. Rutland, VII-IX 1979, 29°20'52" М, 98°34'35” W; Rio Vista Farms Well (19), М. Brzozowski, ХИ 1980—IV 1981, 29°20'23" М, 98°44'38" W; Verstraeten Brothers Well (20), H. Karnei, IV 1977- 1978, 2919' М, 98°39' W; О. В. Mitchell Well (27), H. Karnei, 14-23 VI 1977, R. C. Weidenfeld, 1-3 V 1980, 29°18’ М, 98°38’ W; J. W. Watts Well (22), М. Brzozowski, 5, 7 XI 1980, 29°17'38" М, 98°41'20" W. UVALDE COUNTY: King Farms Well (23), M. Brzozowski, |-Il 1981, 29°17'15” М, 99°39’ W; О. С. Carnes Well (24), М. Brzozowski, I-Il 1981, 2916'37" М, 9940'51” W; С. Reagan Well (25), М. Brzozowski, ll- VIII 1981, 29°16'22” М, 9936'30" ММ; W. С. Reagan Well (26), В. С. Weidenfeld, 1-25 Ш 1980, 29°16’06” N, 99°34' М; В. Carnes Well (27), M. Brzo- zowski, 2-16 II 1981, 29°13'59" М, 99°51'52” W; В. К. Dunbar Well (28), М. Brzozowski, 2-16 Il 1981, 29°13'’32” М, 99°51'40” W; $. Moerbe Well (29), М. Brzozowski, П-Ш 1981, 29°13’06” N, 99°49'33" М; G. Ligocky Well (30), В. С. Weidenfeld, I-VIII 1980, 29°11'37" М, 99°48'38” W; Uvalde National Fish Hatchery Well (37), В. С. Weidenfeld, I-IV 1980, 29°11'26” N, 9950'53" W. TEXAS PHREATIC HYDROBIIDS 167 APPENDIX 2. Shell measurements. N = number of specimens, NW = number of whorls, PD = protoconch diameter, SH = shell height, SW = shell width, АН = aperture height, AW = aperture width, BW = height of body whorl. Locality numbers are those used in Fig. 1 and Appendix 1. Species Locality М ММ/ PD SH SW AH AW BW P. micra 6 (syntype) 1 2.5 — 0.419 1.02 0.419 0.434 0.419 3 6х 243 — 0.336 0.791 0.300 0.307 0.336 $ 0.103 — 0.023 0.050 0.020 0.015 0.023 5(В) 14 x 2.49 — 0.471 1.08 0.416 0.454 0.454 9 oS 2 Pi — 0.353 0.840 0.300 0.353 0.316 s 0.19 — 0.037 0.092 0.036 0.040 0.065 10 ПХ 612752 — 0.454 1.02 0.348 0.399 0.426 s 0.048 — 0.039 0.088 0.036 0.046 0.032 Р. пидах пидах 6 (holotype) 1 27 — 0.806 1.43 0.558 0.589 0.651 1 1 РТ 0.317 10.837 1:05 0.372 0.434 0.651 2 30 EUX (о ТМ 1.14 0.570 0.605 0.868 x 5 0:22 0015.10.12 0.07 0.044 0.04 0.083 2.94 0.355 1.03 1.53 0.575 0.727 0.832 3 20 x se 2.051 0.025 0.13 0.12 0.068 0.058 0.097 4 24 x 2.83 0.377 09351251 0.532 0.695 0.766 $ nOns 0.022 7040 0.13 0.07 0.057 0.07 8 6х 295 — 032111574 0.530 0.663 0.705 SONT — 0.094 0.108 0.040 0.039 0.075 9 их 25 — 0.613 0.933 0.360 0.425 0.493 $ 0.058 — 0.059 0.066 0.03 0.039 0.052 10 6209 — 0.953 1.25 0.530 0.558 0.744 510.2 — 0.13 0.20 0.059 0.068 0.098 14 3 X 2.47 0.342 0.440 0.01 0.350 0.456 0.394 s 0.06 0.011 0.041 0.026 0.023 0.023 0.012 26 Xe Pals) 0.302 0.698 1.24 0.465 0.512 0.605 $ — 0.007 0.022 — — 0.022 0.066 P. nugax inclinata 11 (holotype) 1 3.4 037618 1533 0.496 0.620 0.899 11 NOM Xan 13.017, 0:360 1.11 142: 0.456 0.606 0.828 se 10817. 0.017 0.092 0.11 0.059 0.057 0.079 P. rotunda 3 (holotype) 1 3.6 0.455 0.837 2.26 0.837 0.837 — 3 OXES42 0.392 0.769 2.01 0.769 0.741 — S 0.206 0.031 0.037 0.153 0.037 0.050 — P. conica 5(B) 1 3:75 0.36 1.61 1.21 0.744 0.729 1.22 (holotype) 5(B) 5 3.82 0.343 1.76 1.41 0.831 0778127 x s 0.26 0.01 0.057 0.041 0.01 0.021 0.015 5(А) 6х 392 0.345 1.64 133 0.760 0.707 1.14 s 012 0.01 0.061 0.057 0.015 0.021 0.037 Р. plana 3 (holotype) 1 80 — 0.455 0.822 0.455 0.297 0.277 3 8х 2.93 — 0.450 0.820 0.444 0.305 0.273 s 0113 — 0.041 0.037 0.042 0.028 0.030 1 275 — 0326 01752) 10356 005720277 Р. punctata 3 (holotype) 1 38 0297 105 0.791 052 0434 0744 3 15 X 39 0.319 1.13 0.871 0.564 0.474 0.784 s 015 0.015 0.081 0.055 0.050 0.035 0.056 P. imitata 20 (holotype) 135 0:86: 107 MOB 1043 10471 20713 20 18 Хх 35 03 1.01 0.747 0.388 0.403 0.663 S012 001 0038 0:42 0265 027 70040 21 4x 34 034 1.03 0.833 0.446 0.428 0.701 an xı Sa — № o о № © 0.020 0.08 0.034 0.03 0.015 168 HERSHLER 8 LONGLEY APPENDIX 2 (Continued) PD Species Locality N NW B. uvaldensis 30 (holotype) 1 2.75 30 Ir 2 Ss 0312 S. bartonensis 2 (holotype) 1 4.0 2 5х 4.39 s 0.26 SH 0.434 0.428 0.044 0.970 1.16 0.11 SW 1.02 1.08 0.088 0.495 0.546 0.023 AH 0.31 0.347 0.035 0.337 0.374 0.015 AW 0.372 0.397 0.029 0.287 0.329 0.021 BW 0.326 0.375 0.036 0.594 0.681 0.050 169 TEXAS PHREATIC HYDROBIIDS vL0 €6r'0 9091 si vv 0 ud] dd] S ÿco' 0 8610 | 651`0 | 0c'0 € c0'0 8c' 0 v E00 ¡AO v Sc0'0 £0€'0 | 96€'0 v ce 0 vtr 0 v 8210 dd] v ¿€0'0 SOt 0 G 160'0 1430 v 6700 $55'0 S SL LO eve 0 € LLO eS 0 € LEO 0€'0 € 910`0 ce 0 9 c80 0 ¿90 v cl 0 150 9097 ЛОЛ $ — — vS0'0 == = 917`0 == == S € 2100 r10°0 = L9¢°0 0€0'0 = v с a 2€0'0 >= cv0 0 £ce 0 650'0 601`0 S v v 920`0 E 9100 $1<'0 70`0 910 € С v 190'0 ae = £0£ 0 650`0 660'0 S с € $20`0 = g10°0 ger 0 650'0 8-0 € € a 100 c0'0 vlOO 68b'0 6/00 881'0 9 9 v OL'O 210'0 70`0 260 960`0 €c 0 v a v 8rr'0 1200 10'0 £ce O vS0'0 vllLO ЛОЛ YS] nal до S cl 0 v9s'0 € 2500 ees'0 1:53] ss] ESTAS MESA SS ФЕ LS) iS SAS Es) iS EPS (SIRS A = ES) {= (05) sisuapjean ‘а (г) sısusuoueg 'S (5) ejejound “y (02) ejeyuw y (5) виа “y (95) E9IU09 ‘а (5) epumoi ‘4 (e) xebnu “y (5) вош “y (Ау[еЭ0]) saidads ‘| xıpusddy pue | 'Biy ul pasn э$оц} o] 19,81 злэашпи Ащеэол ‘чбиз| SS9} = $17 ‘uBuel] ayeysoud jeıjjed = 447 “uBua] а}е}зо.а = нал ‘чбиа| puejB элцзабр = HAT ‘биз! Алело = ЛОЛ ‘щбиа| эюе}даээл 1ешшэ$ = ys7 ‘yBue] esınq = Na7 ‘yBue] Jonpıno jeijjed = Od7 ‘чбиз| des ajÁs = ss7 ‘чбиз| цэешо]$ = |S] '$элпопд$ pue зиебло |ешед-иои yo ззиэшэ.п$еэи\ “VE XIGNAddV 170 HERSHLER 8 LONGLEY APPENDIX ЗВ. Measurements of pallial structures. LPC = pallial cavity length, МС = intestine coil length, WIC = intestine coil width, LOS = osphradium length, LCT = ctenidium length, WCT = ctenidium width, NF = number of gill filaments, IMA = distance from anus to mantle edge. Locality numbers refer to those used in Fig. 1 and Appendix 1. PELOS Species (locality) LPC EIG Мо “EOS” СГ М ONE IMA. EPG ZERE Р. тега (3) x 0.366 701099 MOTEROS = — 0.064 0.273 0.385 $ 0.057 0.014 0.028 = == == — 0.019 0.033 0.064 п 4 6 6 5 = = — ‘| 5 5 P.n. inclinata (11) x 0.713 0.232 0.376 0.181 0.505 — 13.7 0.178 0.340 0.274 s 0.054 0.07 0.056 0.031 0.014 — 1.42 — 013 0.037 п 5 3 2 if 2 — 5 11 1 3 = P.n. nugax (3) x 0.7547 0.289 0.319 0.183 0.578 0.119 14.4 0.178 0.40 0.25 s 0.10 0.034 0.069 0.026 0.062 0.034 1.69 0.014 0.07 0.04 n 13 7 7 12 9 5 11 5 7 12 (4) x 0.792 0.218 0.356 0.174 0.590 0.178 13.0 0.198 0.31 0.222 5 0:118 0.028) 0:084 (01022 0:392° > — 0.71 — 0.042 0.036 n 5 2 2 5 4 1 5 1 2 5 (2) x 03/29 0:396— 4 0.158 0.560 — 12.4 0.139 0.54 0.235 Se 0.0887 == = 0.047 0.073 — 1.36 (0:05 002 0.07 MANS 1 5 5 ==. MM 4 1 2 (14) x 0.382 0.139 0.198 0.153 0.322 0.079 11.0 0.109 0.32 0.41 S=.0:.001200232 2 — 0.01 0.041 = 0.82 0.0124 — 0.056 п 3 3 1 4 3 1 E 2 1 4 (12) х 0455 0265 0.257 0.119 77 — == 8.5 0.119 0.44 0.26 в == = = = = = 0.071 = — — (ale 1 1 1 = = 2 1 1 1 (25) x 0.653 0.168 0.244 0191 0614 — 9.0 0.158 0.247 0.28 $ 0.052 0.026 0.01 0.037 20.0287 — — 0.028 0.032 0.014 n 3 4 4 3 3 = 1 3 3 2 P. rotunda (3) х 1108 0.644 0337 9.139 — = = 0.119052 0.13 $ 0.06 0.042 0.052 0.02 = = — 0.032 0.04 0.01 M5 2 3 3 = = = © 2 3 P. conica (5B) X 109 0.590’ 0.402. 0.224 — — = — 2044197050 0.20 s 0.073 0.082 0.042 0.063 — = — 0.028 0.07 0.05 n 4 4 3 4 = = = 8) 3 3 P. plana (3) ELA 01153) 021497 1041352 2 — = — 01099" 04154033 S 00635 10/03 0.017 0.025 — = = — OS 0.059 п 6 7 6 5 == = AS 4 4 P. imitata (20) X 01600’ 0.33 0.247 10155 — = — 012506 0.26 5 0.01 005%. 0.027 0.015 7 — = — 1003 00% 0.02 AS 6 6 6 = = — mu: 4 6 P. punctata (3) x 0.709 0.360’ 0.236 0.134 — = — 0.141 0:475 0185 s 0.068 0.03 0.053 0.01 = = — (0.012) 0.02 0.017 NAO 5 Uh 4 = = — 4 4 4 S. bartonensis (2) х 0.302 0.094 0.198 0.089 — = — 007902700226 s 0.043 0.03 0.039 0.01 = = — 0.052 08 0.018 he 5 4 5 5 = = NÉ 4 5 В. uvaldensis (30) x 0.671 0.172 0.232 0.160 — = — 10.210 10.25 0.24 5 0.103: 0.054 0.032— (0.022 — = — 0.06 01095 (0.029 n 8 6 6 8 = = ht 5 6 8 TEXAS PHREATIC HYDROBIIDS 171 APPENDIX 4. Comparison of 10 spp. of phreatic hydrobiids from south-central Texas involving 40 characters. OTU 1 = Phreatodrobia micra, 2 = P. nugax, 3 = P. rotunda, 4 = P. conica, 5 = P. plana, 6 = P. punctata, 7 = P. imitata, 8 = Balconorbis uvaldensis, 9 = Stygopyrgus bartonensis, 10 = “Orygoceras” sp. Data for “Orygoceras” sp. are from Hershler & Longley (in press). OTU Character 1 2 3 4 5 6 7 8 9 10 1. Maximum shell dimension >1.25 mm (0,1) 0 1 1 1 0 0 0 0 1 1 2. Shell form planispiral (0,1) 1 1 1 0 1 0 0 1 0 0 3. Shell form trochoid-low conic (0,1) 0 1 0 1 0 1 0 0 0 0 4. Shell form elongate-conic (0,1) 0 0 0 0 0 0 1 0 1 0 5. Shell uncoiled (0,1) 0 0 0 0 0 0 0 0 0 1 6. Protoconch microsculpture: spiral lines (0); 1 1 1 1 1 1 1 0 1 1 punctate (1) 7. Teleoconch typically with spiral lines (0,1) 0 0 0 0 0 0 0 1 1 8. Teleoconch typically with irregular ridges 0 0 0 1 1 1 0 0 0 0 (0,1) 9. Teleoconch typically with collabral costae 0 0 0 0 0 0 1 0 0 0 and spiral threads (0,1) 10. Operculum concentric (0,1) 0 11. Operculum with ventral process (0,1) 1 12. Ctenidium (or vestige) present (0,1) 1 13. L osphradium/L pallial cavity typically >30% 1 (0,1) 14. Number of intestinal loops in pallial cavity: 1 0 0 0 0 1 1 1 0 0 0 (0); 2 (1) 15. Long axis of loop(s): parallel to L pallial 1 1 0 1 0 0 0 1 1 1 cavity (0); perpendicular (1) 16. Central radular tooth with basal cusps (0,1) 1 1 1 0 0 0 0 1 1 1 17. Central tooth shape: trapezoidal (0); square 0 0 0 1 0 1 0 0 0 0 (1) 18. Lateral tooth with >20 cusps (0,1) 0 0 0 1 1 1 0 0 0 0 19. L style sac/L stomach >50% (0,1) 1 1 1 1 0 1 1 1 1 1 20. Intestine with loop on right side of style sac 0 0 1 0 0 1 0 0 0 1 (0,1) 21. Pallial gonoducts more than four times as 0 0 0 0 0 0 0 0 0 1 long as wide (0,1) Sey =) iS) OSOS 22. Penis lobed (0,1) 0 0 0 0 0 0 0 1 1 0 23. Penis with specialized glands (0,1) 0 0 0 0 0 0 0 1 1 0 24. L testis/L digestive gland >40% (0,1) 1 1 0 1 1 1 1 1 1 1 25. Seminal vesicle exits from anterior tip of 1 0 1 1 1 1 1 1 1 1 testis (0,1) 26. Seminal vesicle coils on stomach (0,1) 1 0 0 0 1 0 0 0 0 0 27. L pallial prostate/L prostate >40% (0,1) 1 1 1 0 1 1 1 1 1 1 28. Vas deferens exits from prostate tip (0,1) 0 0 0 0 0 0 0 0 0 1 29. Е ovary/L digestive gland typically >40% 0 1 1 1 0 0 0 0 1 1 (0,1) 30. Sperm travels in female via: spermathecal 1 1 1 1 1 1 1 0 0 0 duct (0); ventral gutter (1) 31. Capsule gland more than twice the length of 0 0 1 1 1 1 0 0 0 0 albumen gland (0,1) 32. Anterior end of capsule gland typically coiled 0 1 0 0 0 0 0 0 0 0 (0,1) 33. Capsule gland opening muscularized (0,1) 0 0 0 0 0 0 1 0 1 0 34. Capsule gland with >1 tissue section (0,1) 0 0 0 0 0 0 0 1 1 1 172 HERSHLER 8 LONGLEY APPENDIX 4 (Continued) OTU Character 1 2 3 4 5 6 7 8 9 10 35. Posterior end of albumen gland coiled (0,1) 0 0 0 0 0 0 0 1 0 1 36. Albumen gland sac-like (0,1) 0 0 0 0 0 0 0 0 0 1 37. Oviduct coil positioned: on left side of albu- 0 0 1 0 0 1 0 1 0 0 men gland (0); posterior or ventral to gland (1) 38. Oviduct opens into posterior end of albumen 0 0 0 0 0 1 1 1 0 1 gland (0,1) 39. Bursa present (0,1) 1 1 1 1 1 1 1 0 0 1 40. Seminal receptacle present (0,1) 1 1 1 1 1 1 1 0 1 0 — д3—8—«Д&дДдЫ nn — Revised Ms. accepted 1 May 1985. MALACOLOGIA, 1986, 27(1): 173-183 MODIFICATION AND EVALUATION OF BURCH AND CUADROS'S MEDIUM FOR THE MAINTENANCE OF THE TESTES OF A MARINE GASTROPOD' Thomas C. Cheng 4 Eric J. Pearson Marine Biomedical Research Program? and Department of Anatomy (Cell Biology), Medical University of South Carolina, P. O. Box 12559 (Fort Johnson), Charleston, South Carolina 29412, U.S.A. ABSTRACT Testes of llyanassa obsoleta were maintained in a modification of the medium originally devised by Burch 8 Cuadros (1965). The efficacy of the medium was assayed by monitoring oxygen utilization by testes and their ability to incorporate ®H-thymidine. Scintillation counting and autoradiography were employed to quantify the uptake. It was determined that the testes could be effectively maintained for two weeks, the duration of the experiment. Control testes in isosmotic saline were not maintained as well as in the modified medium. The intent of this study was to devise a maintenance medium so that molecule(s) suspected of being responsible for parasitic castration associated with larval trematodes could be tested т vitro. INTRODUCTION Although the phenomenon of parasitic cas- tration in molluscs due to larval digeneans has been recognized since the initial observation by McCrady (1873), the responsible mecha- nisms have not been studied except at the interpretive, descriptive level. In this labora- tory we have been conducting experiments aimed at elucidating the chemical basis for parasitic castration. As one approach, we have been exposing molluscan gonads in vi- tro to molecules associated with species of larval trematodes known to cause castration. This has required devising media for the maintenance of molluscan gonads. As one of our models for studying parasitic castration, we have been employing the inter- tidal gastropod llyanassa obsoleta (Say) which is known to be castrated by sporocysts of Zoogonus lasius (Olsson) (Cheng et al., 1973; Sullivan et al., 1985). Consequently, studies have been carried out to ascertain whether modifications of established media could be used to maintain the gonads of I. obsoleta for a satisfactory length of time. As a result of an earlier study (Cheng et al., 1984), we reported the efficacy of modifica- tions of two established media for molluscan tissues for the maintenance of /. obsoleta gonads; specifically, that devised by Chernin (1963) for maintaining the heart of the fresh- water pulmonate Biomphalaria glabrata (Say), and that devised by Tripp et al. (1966) for maintaining heart tissues and amoebo- cytes of the American oyster, Crassostrea virginica (Gmelin). Reported herein are our findings relative to the use of modifications of another medium, that of Burch & Cuadros (1965), originally designed for maintaining the gonads of a terrestrial gastropod, Helix poma- tía Linn. and two freshwater species, Biom- phalaria glabrata and Pomatiopsis lapidaria (Say). MATERIALS AND METHODS Testes In view of the objective of this study, we only attempted to maintain the testes of I. obsoleta in vitro because these male gonads include considerably more dividing gametic cells than ovaries (our unpublished work). All of the snails from which testes were removed were collected between January 1 and June 1, 1984, from the same intertidal mudflat in Clark Sound off Charleston Harbor, “This research was supported by a grant (820536) from the UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Disease, and a contract (DE-A509-83ER60132) from the U.S. Department of Energy. World Health Organization Collaborating Centre on Molluscicides. (173) 174 CHENG 8 PEARSON South Carolina. After ascertaining that these male snails were not infected with trema- todes, specimens, each measuring 18 + 3 mm in shell length, were sterilized externally by swabbing with 70% ethanol and their shells were gently crushed aseptically in a vice and the shell fragments were removed. Subse- quently, the bright orange testis was carefully severed by dissection and placed in isosmotic saline (675 т Osm) maintained at 4°С. Prior to being placed in the maintenance medium being tested, each testis was passed through a series of six 10-min washes in ster- ile Petri dishes. The initial two washes were in isosmotic saline plus a high concentration of antibiotics (i.e., 1.5 x normal concentration of penicillin G, streptomycin sulfate, and Fun- gizone listed under “maintenance medium”). The subsequent four washes were in isosmo- tic saline plus the normal concentration of antibiotics. Maintenance medium As stated, the medium that we tested was a modification of that devised by Burch & Cuadros (1965). Specifically, we omitted the addition of crude snail extracts so that possi- ble neurosecretory factors from the central nervous system or tentacles would not inter- fere with the results. The medium was ren- dered isosmotic (675 + 10 т Osm) by adding 22 g of NaCl/liter. This amount of salt was arrived at by basing our calculation on the osmolality of the pooled hemolymph from 25 snails. Bacterial and fungal growths were in- hibited by adding 800 IU of penicillin G/ml, 800 ug of streptomycin sulfate/ml, and 1% Fungizone. The final pH of the medium was adjusted to 7 with НС! or NaOH solutions. A single testis was placed aseptically in each sterile 15-ml screw cap culture tube con- taining either 3 ml of isosmotic saline (control) or 3 ml of the maintenance medium (experi- mental). All cultures were incubated at 25°С on a slowly rotating rack (1/5 rotation/min). Forty-five testes were maintained in isosmotic saline and 45 in our modification of Burch 4 Cuadros's medium for each of the following time intervals: 0, 24, 96 hr, and 1 and 2 weeks. The viability of the maintained testes was assayed at each of the time intervals by measuring oxygen utilization by tissues. Thirty, additional testes were maintained in isosmotic saline (control) and 30 in the mod- ified Burch 8 Cuadros's medium (experimen- tal) for each of the same five time intervals and the viability of these tissues were deter- mined by measuring the ability of the gametic cells to incorporate *H-thymidine by employ- ing scintillation counting. In addition, 10 testes were maintained in the modified maintenance medium and in isosmotic saline for each of the five time intervals and their ability to incorpo- rate °H-thymidine was ascertained by auto- radiography. The detailed protocols followed in each of the three assay procedures are presented below. Respirometry Testes of the control and experimental groups employed in the respirometric assays at each of the five time intervals were pooled in 15-ml reaction vessels, 15 testes from each group per vessel. Each vessel contained 5 ml of fresh maintenance medium or isosmotic saline. This was done because preliminary studies had revealed that at least 15 testes per vessel were required for a reliable mea- surement of oxygen consumption. A filter paper fan (Whatman No. 2) was placed in the center well of each reaction vessel. Also placed in the center well was 250 ul of 1M KOH to absorb the CO, given off by the respiring tissues. A model GRP-14 Gilson Differential Respirometer (Middleton, Wiscon- sin) was employed. After equilibration involv- ing shaking (80 oscillations/min) for 60 min at 25°C, respirometric readings were recorded at 20 min intervals for 4 hr. At the end of 4 hr, the 15 gonads compris- ing each group were placed on pre-weighed aluminum foil, dried for 2 days at 130°C, and weighed. Subsequently, the respirometric data were converted to Ll O, consumed/g dry weight of tissue/20 min. The respirometric determinations were made in triplicate for both the experimental and control groups, 15 in each group, hence a total of 45 testes per category. Exposure to *H-thymidine Following maintenance in the medium (ex- perimental) or saline (control) for O, 24, 96 hr, and 1 and 2 weeks, the testes comprising each group were pooled in groups of ten in 3 ml of the maintenance medium or isosmotic saline each containing 5 pCi/ml of 9H- thymidine (64 Ci/mmole, ICN, Irvine, Califor- nia) in sterile vials and rotated for 3 hr at 100 rpm at 25°С оп а rotating platform. Although ILYANASSA TESTES CULTURES AWS 3H-thymidine is taken up by cells within 15 min (Baserga & Malamud, 1969), we em- ployed 3 hr of exposure to ensure that uptake throughout each testis had occurred. Subse- quently, the testes were removed from the radioactive medium and washed six times in isosmotic saline before being processed for either scintillation counting or autoradio- graphy. Scintillation counting Following maintenance and exposure to $H-thymidine, 30 testes comprising each of the five experimental groups and 30 compris- ing each of the five control groups were indi- vidually placed in scintillation vials (Fisher Scientific, Pittsburgh, Pennsylvania) and dried for 24 hr at 75°C. Subsequently, the tissues were homogenized by employing a glass homogenizer and the homogenates were digested with Scintigest (Fisher). Ten ml of Scintiverse (Fisher) were added to each digested sample and the solution was deco- lorized using hydrogen peroxide (0.25 ml/ vial). Counts (decays detected by the pho- tomultiplier)/min were obtained using a 2000 Series Packard Tri-Crab liquid scintillation spectrometer (Downers Grove, Illinois). Autoradiography After maintenance of the five time intervals and exposure to ®H-thymidine, 10 testes of both the experimental and control groups at each of the five time intervals were individu- ally placed on glass slides and macerated with a scapel. The macerated gonads were then smeared and air dried. The smears were fixed with Carnoy’s fixative for 30 min and again were allowed to dry. Subsequently, the smears were rehydrated and dipped in liquid МТВ. (Eastman Kodak, Rochester, New York) three times in 5 sec in the dark. The smears were then permitted to dry for 1 hr prior to being stored in a light-proof box for 7 days at 4°C. The smears were developed in D-19 developer (Kodak) and fixed in Kodak rapid fixer (Gude, 1968). The smears were then stained with hematoxylin and eosin. The number of cells with labeled nuclei per 1000 cells and the number of silver grains associated with the nucleus of each labeled cell were counted under oil immersion. Pre- liminary studies had revealed that when fewer than 1000 cells were counted per slide, vari- ability in the number of labeled cells in re- peated counts was greater (+ 28% of 500 cells counted, + 14% of 1000 cells counted). To avoid biased sampling, all counts were made using a double blind protocol. Further- more, only those cells with three or more silver grains associated with the nucleus were counted to avoid including cells labeled due to background exposure. Statistical analysis Data obtained from employing respirome- try, scintillation counting and autoradiography at the stated time intervals were compared by employing analysis of variance. Specifically, the respirometric data were compared by us- ing a split plot analysis computer program (Winer, 1977) and the scintillation and auto- radiographic data were compared by use of a two-way factorial computer program (Winer, 1977). Comparison between the results ob- tained with the control (isosmotic saline) and experimental (modified Burch & Cuadros's) media at each time interval was performed with the least significant difference test (Winer, 1977). RESULTS Respirometry The results of our respirometric determina- tions on the experimental and control groups of testes at the five time intervals are pre- sented in Table 1. Note that the mean rate of O, consumption by the testes maintained in isosmotic saline declined with time, whereas the mean rate of O, consumption maintained in the modified medium remained relatively constant. After the second week the O, utili- zation for testes maintained in the modified medium was 6.494 x 10-* pl O,/gm dry weight/20 min, which is almost the same as that of O hr testes. On the other hand, the mean rate of O, utilization for testes main- tained in saline for two weeks was essentially zero. This indicates that the control testes were no longer viable after two weeks. A split-plot analysis of the respirometric data revealed that the difference in the means of testes maintained in the modified medium and saline at all of the time intervals was statistically significant (p >0.01), with that of the experimental testes being higher (Table 2). In addition, this analysis revealed that CHENG 8 PEARSON 176 juawnedxe эщие зпочбполц} sdnoiB |одиоэ pue ¡ejuawuadxa ay, jo ajes Алоелд$э1 UPS = g ‘6 шпраш $,50.репэ 3 y9mg payipow ul ayer Aoyelıdsaı ueaw = Og + “aues onousos! ul ayes Ало}е195эл URAL = S| € "uondumsuos uaBÁxo ON ‘г “¡enajul au 20} ayes Aioyeiidsai uesw = | ‘| 65520 = ¡SL G99€ 0 = PL 5769`0 = ‚EL p9ESO = ¡91 tL660 = '1 96p90 = X ,9/EL0— = 'X 151590 = EX 62120 ='X 60/0 = X 995870 = !X 0550 = “X 4041670 = !x 66290 = “X 6$95`0 = 'X 228b0=8 €909'0 - 99 (dnowd/gı = и) 185€ 0 +51 erıso 15900 8150 95620 5598`0 21090 92€9 0 S96+ 0 8S€+ 0 LSLE'O € lueuedxz 6cct 0 8 0875'0=99 (dno16/g, = и) 8/6 0 =$1 cv0r 0 ¿09180 16830 8e01'0 69/2 0 v8€s 0 cect 0 LLeG'O Liv8 0 91+9 0 2 Juauadx3 zS0S 0 ‹'а £869 0 -,99 (dnoi6/s, = и) LICE O — .SI 66201 23250 - 81190 cvSc 0 r19r 0 801€'0 5599`0 7919'0 29090 51190 | juawiadx3 unipaul aules шпраш auies unipau auljes шпраш auijes шпраш auljes payipow 91}OUISOS| payipoy 21}OUSOS| payipoy INOWSOSI paıyıpoyy 9404/50$| payipon 21}OWSOS| SH98M с Y99M | 24 96 44 ta 44 0 ‘SIPAJ8JUI эшц BAI} Je SUINES э40ц$0$! JO шгираш S.SOIPENI $ YING рашрош ul рэшеишеш sajsaj jo (un oz/1ybiam Aıp 116/29 ¡ti ui) sajes Aroyeııdsaı иеэи ‘| FIGVL ILYANASSA TESTES CULTURES 177. TABLE 2. Split plot analysis of respirometric data comparing testes maintained in modified Burch & Cuadros’s medium with testes maintained in isosmotic saline at 1, 24, 96 hr and 1 and 2 weeks. Degrees of Sum of squares freedom Mean square F Tail probability Mean 6.62982 1 6.62982 111.08 0.0000 Trial 0.63606 2 0.01803 0.30 0.7473 Time 0.55586 4 0.13897 2:39 0.1437 Error 0.47750 8 0.05969 Media 0.62583 1 0.62583 29.32 0.0006' Media x trial 0.02363 2 0.01182 0.55 0.5955 Media x time 0.52225 4 0.13056 6.12 0.0148? Error 0.17078 8 0.02135 1. p <0.01 20105 TABLE 3. Analysis of the mean respirometric rates (Ll 0>/дт dry weight/20 min) of testes maintained in modified Burch 8 Cuadros's medium and isosmotic saline. The differences between the rates in the modified medium and saline has been compared at each time interval by employing the least significant difference test. Time Mean respirometric Mean respirometric rate interval rate in isosmotic saline modified medium Difference LSD' Ohr 0.5649 0.6299 0.0650 < 0.2751 24 hr 0.4977 0.5750 0.0773 < 0.2751 96 hr 0.4855 0.7032 0.2177 < 0.2751 1 week 0.2179 0.5151 0.2972 >0.27512 2 меек 0.1376 0.6494 0.7873 >0.2751? 1. LSD = Least significant difference at which the null hypothesis can be accepted. The LSD is calculated from the equation x — y >t (0.025) / 2 MS error. : 3 2. The difference between the mean respirometric rates are statistically significant (р <0.05, two tailed). TABLE 4. Two-way factorial analysis of scintillation counting log-transformed data. The analysis was of the mean counts per minute of whole testes maintained in modified Burch & Cuadros’s medium or isosmotic saline for 2, 24, 96 hr, 1 and 2 weeks. followed by exposure to 5 uCi/ml of ®H-thymidine. Degrees of Sum of squares freedom Mean square F Tail probability Mean 32290.278 1 32290.278 59451.5 0.0000' Time 210.842 4 5257 lil 97.05 0.0000' Media 8.413 1 7.413 13.65 0.0003" Time x media 7.125 4 1.781 3.28 0.01192 Error 156.423 288 0.543 Us [20] 2 01= 0:05 there 15 а statistically significant interaction between the groups of testes and the time intervals (p >0.05). This indicated that the respirometric rates for the two groups of tes- tes needed to be examined within each time period. As a result of employing the least significant difference test (Winer, 1971) to an- alyze the mean respirometric rate (pooled for the three trials) of the two groups of testes at each of the five time intervals it was found that the mean rate of the experimental testes was significantly higher than that of the control CHENG 8 PEARSON 178 : (pares OM) ‘<0'0 > а) зиэлэшр Ayueoyiubis. ay u № gg (960 0)) .05) (Table 3). Scintillation counting The raw data collected were as counts per minute (cpm) per gonad. These were not con- verted to cpm/g dry weight because accurate weighing of individual dried testis was render- ed questionable because of the small weights of the tissue and its hygroscopic nature. The data were log-transformed for statisti- cal analysis in order to achieve homogeneity of variances of the mean cpm at the five time intervals. One assumption of the F-test (Wi- ner, 1971) that was performed as part of the two-way factorial analysis was that the stan- dard deviations were all estimates of the same “true” standard deviation. Thus, the data were log-transformed so that the stan- dard deviations were normalized. As a result of the two-way factorial analysis of the log-transformed data, the differences between the data collected at the five time intervals and the differences between the data associated with the experimental and control groups were ascertained to be statis- tically significant (p <0.01) (Table 4). In other words, the average cpm for testes maintained in the modified medium and in saline at all of the time intervals were also significantly dif- ferent. Furthermore, interaction between the media data and the time interval data was significant (р <0.05), thus indicating that a least significant difference test (Winer, 1971) was necessary to compare the mean cpm obtained from the two groups of testes at each time interval. As a result, the mean cpm for testes maintained in the modified medium was determined to be significantly higher than the mean cpm for testes maintained in saline at 24 hr, and 1 and 2 week time periods (p <0.05, two-tailed) (Table 5). This permits the conclusion that the testes maintained in the modified medium were incorporating more °H-thymidine. It is noted, however, that the difference between °H-thymidine uptake by testes maintained in the modified medium and saline was not significant at 96 hr. This, in our opinion, is an artifact. 180 CHENG 8 PEARSON 170 150 130 110 90 ‘ 70 50 MEAN NO. OF LABELED CELLS/1000 CELLS 30 10 30 50 70 MAINTENANCE PERIOD (hours) FIG. 2. Mean number of labeled cells/1000 cells determined by counting smears of testes maintained in either modified Burch & Cuadros’s medium (©) or isosmotic saline (@) for the time periods indicated. Vertical lines represent 1 standard deviation. Autoradiography Smears made from testes that were freshly dissected (0 hr) or maintained for 24 hr in- clude numerous cells that were heavily la- beled (Fig. 1). As depicted in Fig. 2, the num- ber of labeled cells/1000 cells decreased somewhat between the O and 24 hr mainte- nance periods in both the experimental and control groups. At 96 hr and 1 week, however, the labeling index for testes maintained in isosmotic saline fell while that for testes main- tained in the modified medium remained con- stant (Fig. 2). This is interpreted to mean that the experimental testes maintained for 96 hr and 1 week incorporated more °H-thymidine than the control testes at the same time inter- vals. By designating those cells with 10 or mor silver grains associated with their nuclei as being heavily labeled, the percent of heavily labeled cells in testes maintained in saline declined dramatically at 96 hr (as did the total number of labeled cells), whereas the per- centage in testes maintained in the modified medium remained at a fairly constant level (Fig. 3). A two-way factorial analysis (Winer, 1971) performed on the cell labeling index data (i.e., total number of labeled cells/1000 cells) for cells maintained in the modified medium and saline for 0, 24, 96 hr and 1 week revealed that the differences in the labeling indices (combining the values for both groups of tes- tes) for each time period, and the difference in indices between the two groups of testes (combined across all time periods) was sta- tistically significant (p<0.01) (Table 6). Also, there was a statistically significant interaction between the two groups of testes and the time ILYANASSA TESTES CULTURES 181 90 an 3 u 70 о o Wu y ш о E Se = $ | — > < т $ т 30 # ? T | 10 30 50 70 90 110 130 150 170 MAINTENANCE PERIOD (hours) FIG. 3. Mean number of heavily labeled cells (with 10 or more silver grains associated with the nucleus), determined by counting smears of testes maintained in either Burch & Caudros's medium (0) or isosmotic saline (e) for the time periods indicated. Vertical lines represent 1 standard deviation. periods, indicating that further analysis was necessary to compare the indices of the two groups of testes at each time interval. Application of the least significant differ- ence test revealed that the labeling indices for the experimental testes maintained for 96 hr and 1 week were significantly higher than those for the control testes maintained for the same time periods (p >0.05, two tailed) (Table 7). The labeling indices for experimental testes maintained for 96 hr and 1 week were almost identical; however, the standard deviation of the 1 week value was large in comparison to that of the 96 hr value. The results of our statistical analyses per- TABLE 6. Two-way factorial analysis of autoradiographic data. Shown are the mean number of labeled cells/1000 cells for testes maintained in isosmotic saline or modified Burch & Cuadros's medium for 0, 24, 96 hr and 1 week followed by exposure to 5 uCi/ml of ®H-thymidine. Degrees of Sum of squares freedom Mean square [Е Tail probability Mean 673259.37 1 673259.37 549.18 0.0000' Time 142220.00 3 47406.67 38.67 0.0000' Media 28492.71 1 28492.71 23.24 0.0000' Time x media 27536.18 3 9178.73 7.49 0.0002" Error 87041.96 71 1225.94 ' Statistically significant (р <0.001). 182 CHENG 8 PEARSON TABLE 7. Analysis of mean number of labeled cells/1000 cells in testes maintained in modified Burch & Cuadros's medium and isosmotic saline and exposed to ?H-thymidine. The difference between the experimental and control groups is compared at each time period using the least significant difference test (LSD). 96 hr 1 week Isosmotic Modified Isosmotic Modified Isosmotic Modified Isosmotic Modified O hr 24 hr saline medium saline Mean no. of labeled cells/1000 cells 163.2 159.5 90.3 Standard deviation 32.47 48.59 36.37 Sample size (n) 10 10 10 Comparison of difference of means to LSD statistic 37 = 31:22 5.3 ' Statistically significant (р < 0.05). mit the conclusion that testes maintained in the modified medium for 96 hr and 1 week incorporated more °H-thymidine than did tes- tes maintained in saline for the same time periods. DISCUSSION Several maintenance media have been de- vised for molluscan cells and/or tissues (see Malek & Cheng, 1974, for review). In this study, the testes of /. obsoleta were main- tained in vitro in a modification of the medium originally devised by Burch & Cuadros (1965). By employing three assay methods (re- spirometry, scintillation spectrometry, auto- radiography), we have demonstrated that our modification of Burch and Cuadros’s medium can be employed to maintain testes of /. obso- leta for 2 weeks, the duration of the study. Specifically, we have demonstrated that tes- tes maintained in the modified medium re- vealed significantly higher O, utilization than did testes maintained in isosmotic saline at 1 and 2 weeks. Furthermore, the mean re- spirometric rate for testes maintained in the modified medium remained fairly constant (ranging from 0.5151 to 0.7032 ul O,/gm dry weight/20 min) at all of the time intervals (0, 24, 96 hr, 1 and 2 weeks), whereas the mean respirometric rate of testes maintained in isosmotic saline fell precipitously after 96 hr (from 0.4855 to 0.1376 ul O,/gm dry weight/ 20 min). medium saline medium saline medium 95.6 22.2 94.7 17.8 95:7 25.43 17.28 23.41 12.41 57.10 10 10 10 10 10 — 31:22. 72:5 31.22 97.9 >32.08' Our quantitative autoradiographic data re- vealed that /. obsoleta testes maintained т the modified medium took up ®H-thymidine at about the same level as at 24 and 96 hr although there was a larger variance at one week. In contrast, uptake of ®H-thymidine by testes maintained in saline dropped dramati- cally after 24 hr. Finally, our scintillation counting studies re- vealed that testes maintained in the modified medium for 24 hr, 1 and 2 weeks took up greater amounts of ®H-thymidine than did tes- tes maintained in saline. The only discrep- ancy between our respirometric and autoradi- ographic results and our scintillation counting results rests with our findings that there was no difference in the uptake of 9H-thymidine by the experimental and control groups of testes at 96 hr. This is believed to be an artifact. In an earlier study, Cheng et al. (1984) compared the efficacy of modifications of the media devised by Chernin (1963) and Tripp et al. (1966) for maintaining gonads of I. obso- leta. lt was found that these two media could be employed to maintain ovaries and testes of l. obsoleta for up to 48 hr. In view of the results reported herein, our modification of Burch 8 Cuadros's medium is more efficacious for maintaining the testes of /. obsoleta and con- sequently is currently being employed in т vitro tests for the detection of a molecule from sporocysts of the trematode Zoogonus lasius responsible for parasitic castration of /. obso- leta. ILYANASSA TESTES CULTURES 183 ACKNOWLEDGEMENTS We thank Dr. Rebecca G. Knapp, Depart- ment of Biometry, Medical University of South Carolina, for her assistance in the statistical analysis of our data and J. Bradford Cheng for preparing the photomicrograph. REFERENCES CITED BASERGA, В. & MALAMUD, D., 1969, Modern methods in experimental pathology: autoradiogr- aphy: techniques and applications. Harper and Row, New York, 281 p. BURCH, J. B. & CUADROS, C., 1965, A culture medium for snail cells and tissues. Nature, 206: 637-638. CHENG, T. C., HOWLAND, K. H. & SULLIVAN, J. T., 1984, Comparison of the efficacy of two main- tenance media for gonads of //yanassa obsoleta (Mollusca: Gastropoda). Transactions of the American Microscopical Society, 103: 249-262. CHENG, T. C., SULLIVAN, J. T. & HARRIS, К. R., 1973, Parasitic castration of the marine prosobr- anch gastropod Nassarius obsoletus by sporo- cysts of Zoogonus rubellus (Trematoda): his- topathology. Journal of Invertebrate Pathology, 21: 183-190. CHERNIN, E., 1963, Observations on hearts explanted in vitro from the snail Australorbis gla- bratus. Journal of Parasitology, 49: 353-364. GUDE, W. D., 1968, Autoradiographic techniques: localization of radioisotopes in biological mate- rial. Prentice-Hall, Englewood Cliffs, New Jersey, 113 p. MALEK, E. A. & CHENG, Т. C., 1974, Medical and economic malacology. Academic Press, New York, 398 p. McCRADY, J., 1873, Observations on the food and the reproductive organs of Ostrea virginica, with some account of Bucephalus cuculus nov. spec. Proceedings of the Boston Society of Natural History, 16: 170-192. SULLIVAN, J. T., CHENG, Т. С. & HOWLAND, К. H., 1985, Studies on parasitic castration: castra- tion of llyanassa obsoleta (Mollusca: Gas- tropoda) by several marine trematodes. Trans- actions of the American Microscopical Society, 104: 154-171. TRIPP, М. R., BISIGNANI, L. A. & KENNY, М. T., Oyster amoebocytes in vitro. Journal of Inverte- brate Pathology, 8: 137-140. WINER, B. J., 1971, Statistical principles in exper- imental design. McGraw-Hill, New York, 907 р. " oR == / Ad } N E fj q 5 à. w q 7 y i | | Y Г 7 a E or | u ya FURIA RE ni : | р VAT i a oe , u ‘ © ¡ODA № E : | ns ct UE a E В à Wel роет миа a “Fito Ne os: y. af 0 Re Mai | xl = = М. © SN К. MALACOLOGIA, 1986, 27(1): 185-202 THE REPRODUCTIVE CYCLES AND GLOCHIDIA OF FRESH-WATER MUSSELS (BIVALVIA: HYRIIDAE) OF THE MACLEAY RIVER, NORTHERN NEW SOUTH WALES, AUSTRALIA H. A. Jones, R. D. Simpson 8 C. L. Humphrey Department of Zoology, University of New England, Armidale, NSW 2351, Australia ABSTRACT An investigation of the reproductive biology of five fresh-water mussel species, Cucumerunio novaehollandiae, Hyridella australis, H. depressa, Hyridella sp. and Alathyria profuga was undertaken in the Macleay River, northern New South Wales. Gametogenesis was studied in detail for C. novaehollandiae but only the cycle of larval production was described for the other species. In C. novaehollandiae gametogenesis occurred throughout the year. Ripe oocytes and spermatozoa were abundant in the ovaries and testes from January until August. The breeding season was highly synchronized and occurred in April, although it is possible that a second breeding season occurred during August in the upper reaches of the river. The reproductive cycle of the downstream populations lagged behind the cycle of the upstream populations of C. novaehollandiae. Spawning was associated with the occurrence of floods, and the resulting drop in water temperature might possibly be an important exogenous factor influencing spawning. The brooding period extended over nine weeks. Glochidial release proceeded from mid-May to the end of July in the upper reaches of the river and from June until August further downstream. The breeding season in Hyridella australis, a repetitive breeder, occurred from spring to autumn but was observed only during spring and summer in H. depressa and Hyridella sp. H. australis produced three broods during the breeding season. The brooding period was from eight to eleven weeks depending on water temperature. Glochidia were released throughout most of the year with peak release periods in November, February and May. Females of A. profuga were gravid in mid-summer when most individuals were collected. The glochidia of Cucumerunio novaehollandiae, Hyridella australis and H. depressa are described. Except for H. depressa, these are much smaller than the known glochidia of other Australian species and also differ markedly in shape. INTRODUCTION The fresh-water Unionacea have highly specialized life cycles. Among the Hyriidae, the eggs are moved into specialized portions of the inner gills (marsupia) where they de- velop into a hooked larval stage (the glochid- ium). Mature glochidia are released into the water where they spend some time attached to a vertebrate host. This is generally a fish, although tadpoles (Seshaiya, 1941; Walker, 1981) and a salamander (Howard, 1951) have also been shown to be host species. General reproductive patterns are well known for both North American unionaceans (Lefevre & Curtis, 1910, 1912; Coker et al., 1921; van der Schalie, 1938; Pennak, 1953; Clarke & Berg, 1959) and European union- aceans (Bloomer, 1935, 1946; Negus, 1966; Tudorancea, 1969, 1972; Wood, 1974; Haukioja 8 Hakala, 1978; Dartnall & Walkey, 1979). Detailed life histories providing infor- mation on gametogenesis, breeding seasons, periods of glochidial release, fish hosts and the duration of the parasitic period are un- available for most species. The reproductive biology of Margaritifera margaritifera, how- ever, is well known in both Europe and North America (Murphy, 1942; Roscoe & Redelings, 1964; Wood, 1974; Smith, 1976, 1979; Bauer, 1979). Trdan (1981) determined the breeding season, period of glochidial development and fish hosts for Lampsilis radiata siliquoidea but ignored gametogenesis. Reproductive cycles of unionaceans, including gametogenesis, have been determined in both temperate (Matteson, 1948; van der Schalie 8 van der Schalie, 1963; Stein, 1969; Yokley, 1972; Giusti et al., 1975; Heard, 1975; Zale & Neves, 1982) and tropical (Lomte 4 Nagab- hushanam, 1969; Ghosh & Ghose, 1972; (185) 186 JONES, SIMPSON 8 HUMPHREY Nagabhushanam 8 Lohgaonker, 1978; Dud- geon 8 Morton, 1983; Humphrey, 1984) re- gions. Despite the numerous morphological de- scriptions of glochidia in the literature (Surber, 1912; Coker et al., 1921; Clarke 8 Berg, 1959), few of these enable identification of glochidia at the species level (Rand 8 Wiles, 1982). The type of glochidium (ге. hooked, hookless and axehead types) is constant for the genus and in some cases the shape 1$ also characteristic (Lefevre & Curtis, 1910). Identification of glochidia at species level, es- pecially conspecifics, is often more difficult (Porter & Horn, 1980) but it has been achieved by using scanning electron micro- scopy (Rand 8 Wiles, 1982) and analysing glochidial morphometrics (Wiles, 1975). Little is known of the reproductive biology of the fresh-water mussels from the Australasian region. There 15, at present, only one compre- hensive study of the reproductive biology of an Australian mussel and this is a tropical species (Humphrey, 1984). Fish hosts have been found for several species (Percival, 1931; Hiscock, 1951; Atkins, 1979; Walker, 1981; Humphrey, 1984). The available data indicate that the glochidia of Australian fresh- water mussels are nonspecific parasites of fish (Atkins, 1979; Walker, 1981; Humphrey, 1984). The glochidia of less than half of Australia's 17 species of fresh-water mussels have been described (McMichael & Hiscock, 1958; Atkins, 1979; Walker, 1981) although there are several unpublished records (K. F. Walker, personal communication). The aim of the present study was to inves- tigate reproductive strategies of warm- temperate mussels in the Macleay River, New South Wales. (Other workers are currently studying reproduction of fresh-water mussels in the Murray River.) Five and possibly six species occur in the Macleay River system although one, Velesunio ambiguus, is found only in the tablelands section of the Apsley River (Fig. 1). Cucumerunio novaehollandiae (Gray) is ubiquitous throughout the river and for this reason was chosen for a detailed investigation of its reproductive cycle, includ- ing gametogenesis, breeding season and the period of glochidial release. Upstream and downstream populations were chosen for a comparison of the reproductive cycle in differ- ent parts of the river. Less detailed study was made of the reproductive cycles of the four other hyriid species occurring in the river— Hyridella australis (L.), Hyridella depressa (L.), Hyridella sp. and Alathyria profuga (Gould). MATERIALS AND METHODS The study area The Macleay River 15 situated in northern New South Wales, with its source in the New England Tablelands. Three major tributaries, the Apsley, Chandler and Muddy Rivers drain the central catchment area via a system of deep gorges from which the river emerges near its junction with the Georges River (Fig. 1). From here, the river flows more or less directly to the sea 220 km downstream. The Macleay is a bicarbonate river, with soft waters of low salinity and turbidity. The chem- ical characteristics of the river at Turner's Flat (mean values) were as follows: Calcium 9.48 mgl*, bicarbonate 55.51 mgl*, pH 7.7, hard- ness 46.0 mgl*, chlorine 12.40 mgl*, con- ductivity 143 jScm”* and salinity 62 таг" T.D.S. (N.S.W. Water Conservation and Irri- gation Commission; Australian Water Re- sources Council, 1976). Discharge is seasonal but variable and the highest discharge rates occur during the months from January to June with a minor peak in the spring (Fig. 2). During the study period water temperatures rose to a peak of 27°C in mid-summer and began falling during March, reaching a minimum of 11°C in June (Fig. 3). This was typical of previous years. Little difference in temperature was apparent between upstream and downstream parts of the river except in May when the water tem- perature of the lower reaches was 2°C higher than upstream. Collections and species identifications Sampling of the fresh-water mussel popu- lations was confined to stretches of the Macleay River below Georges River since the rugged terrain and inaccessible nature of the central gorge system precluded sampling above this point. The river was regularly sam- pled at two stations: at Honeymoon Bend (station 1), approximately 160 km above the tidal limit and at Toorooka (station 4), 50 km above the tidal limit. Infrequent sampling was carried out at two other stations (2 and 3) along the river (Fig. 1). REPRODUCTION IN AUSTRALIAN HYRIIDAE 187 Armidal Walcha г x + VA, MN 4 \ ° 4 № if <) 7 о Te b- Kempsey Z р o о 20 40 Lt |) Т =Turner's Flat FIG.1. The Macleay River, showing the locations of the sampling stations (1, 2, 3, 4). Hatching indicates waterfalls at the top of the central gorge system. Species identifications were based on McMichael 8 Hiscock (1958) and shell collec- tions held by the Australian Museum (Sydney). We are confident about the identi- fications made of four species that occur in the main stem Macleay River. These species are Cucumerunio novaehollandiae, Hyridella australis, H. depressa and Alathyria profuga. (The occurrence of A. profuga in the Macleay River drainage 1$ a new distributional record.) There is a likelihood, moreover, that a fifth species occurs although uncommonly; some shells collected during the present study match closely, published descriptions (McMichael & Hiscock, 1958) and museum identifications of H. drapeta. However, the glochidia (found in only one of the specimens collected from the Macleay River) do not match published descriptions of the glochidia of this species (Atkins, 1979). In morphology they are similar to the glochidia of H. de- pressa, although from the scant data avail- able they are smaller (see Table 1). More collections will need to be made to determine whether these shells are merely ecopheno- typic variants of H. depressa or whether in fact they represent individuals of an undescribed species of Hyridella. For now, individuals of this type are referred to as Hyridella sp. The abundances of the five species occur- ring in the main stem Macleay River were, in decreasing order, Cucumerunio novaehollan- diae, Hyridella australis, H. depressa, A. profuga and Hyridella sp. Specimens were collected monthly from July 1982 to July 1983 by snorkelling or by hand. Supplementary col- lections were made in January 1985. Care was taken to process mussels quickly after collection since gravid females are known to abort the larvae from the marsupia when un- der stress (Lefevre & Curtis, 1910, 1912; Hiscock, 1951). Animals were either packed 188 JONES, SIMPSON 8 HUMPHREY 100,000 _ 10,000 2 2 Lu 1,000 E < Е © D = 100 А = о 1982 D J FM А MJ 1983 FIG. 2. Daily discharge of the Macleay River during the study period. (Based on data supplied by the New South Wales Water Conservation and Irrigation Commission). in ice or fixed in 10% formalin made up from river water. Approximately 50 C. novaehollan- diae were collected monthly from each ofthese stations, but the other species were collected in similar numbers from station 1 only. Only C. novaehollandiae was abundant at station 4. Examination of gonads Adult C. novaehollandiae ranging from 100-160 mm in length were sectioned at 6 um through the central region of the visceral mass and stained with Masson's trichrome or Mayer's haematoxylin and counterstained with eosin. Quantitative data of the stages of spermatogenesis were obtained from micro- scopical examination of stained sections through the central region of the visceral mass. In each of five individuals from each month, ten acini were selected at random and the proportion of each cell type, along a line through each acinus, was calculated using an ocular micrometer. Oocyte sizes were measured with an ocular micrometer from visceral smears of 10 indi- viduals per month and seasonal changes in the mean oocyte size determined. 4500 oocytes were measured. The number of oo- cytes (sample size) required to give a repre- sentative mean oocyte size in each individual was determined by plotting the means against sample size (Fig. 4) until the mean value ceased to fluctuate (Elliott, 1977). The inner demibranchs of the gills of fe- males from all four species were examined to determine gravidity. Small portions from gravid gills were removed and examined un- der the microscope so that the stage of de- velopment of the larvae could be determined. Four stages were recognized: Stage I. Marsupium empty and undevel- oped. Stage Il. Eggs or embryos present in the marsupia. Embryos included all stages of de- velopment from zygotes to individuals in which the larval shell had not formed. Stage Ill. Glochidia present in the marsupia. Glochidia were characterized by the develop- ment of the adductor muscle and the larval shell. This included individuals in which hooks were unformed or rudimentary to fully devel- oped larvae, free of their vitelline membrane. REPRODUCTION IN AUSTRALIAN HYRIIDAE 189 TEMPERATURE (*C) J F M A M JA" ОМ D FIG. 3. Monthly water temperatures for the Macleay River, showing readings taken during the study period (broken line) and the mean monthly water temperatures for Georges River junction (closed circles) and Turner's Flat (open circles) since 1976. (Data supplied by the New South Wales Water Conservation Irrigation Commission) Stage IV. Post-glochidial release phase. Some glochidia may be retained in the mar- supium, which often has a bubbly appearance immediately after glochidial release. The water tubes are clearly visible in the inner demi- branchs of C. novaehollandiae that have recently released glochidia. The dorsal two- thirds of the marsupia in H. australis retains a mass of tissue after the glochidia have been released, which imparts a false impression of gravidity. Aborted glochidia could be distin- guished from those normally released at stage IV only if they were not mature glochidia. The stages of glochidial development were further subdivided into early glochidia, inter- mediate glochidia and mature glochidia after Heard (1975). Dimensions of length (maxi- mum valve diameter in the hinge plane), hinge length and height (depth from hinge to hook) of mature glochidia were measured. Refer- ences made to glochidial dimensions are in the order of height x length. Definition of some terms is needed here. The breeding season, for animals that fertilize externally, spans the coincident periods of spawning of eggs and spermatozoa (Giese, 1959; Simpson, 1977). For those animals that do not fertilize externally, spawning and fertil- ization can be two separate events. The breeding season of fresh-water mussels is when the spermatozoa are spawned by the males to fertilize ova being moved into the gill chambers. The gestation period then follows and refers to the time elapsing between the movement of the oocytes into the marsupia until the development of mature glochidia. The gestation period is part of the total brood- ing period, which spans the time of placement of oocytes in the marsupia to the release of glochidia. 190 JONES, SIMPSON 8 HUMPHREY TABLE 1. Morphometric data for glochidia of the five species of fresh-water mussels found in the Macleay River. Mean Mean Hinge Ht/Lth Hinge/Lth length height length ratio ratio Species n° + SD (um) + SD (um) + SD (um) (%) (%) C. novaehollandiae 50 52.2 =10:6 64.1 + 0.2 35 116 64 H. australis 50 7:3.9=3.0:5 94.7 + 0.3 40 128 68 Н. depressa' 50 259.5 244 +5 15216 97 60 Hyridella sp." 20 239 + 4 233 + 4 136 + 2 97 57 A. profuga' 20 239 5 204 + 2 165 + 4 85 69 A. profuga? = 245 200 = 82 — Н. drapeta? = 330 + 10 230 + 10 248 71 75 *From five individuals (C. novaehollandiae, H. australis and H. depressa) and one individual each (Hyridella sp. and A. profuga). ‘Measurements made from preserved material. “From McMichael & Hiscock (1958). 3From Atkins (1979). Hinge length is estimated from an illustration. RESULTS Spermatogenic cycle The pattern of spermatogenesis in male Cucumerunio novaehollandiae from the up- stream station between July 1982 and July 1983 was determined (Fig. 5). Spermato- genesis occurred throughout the year but at reduced tempo during the colder months of June and July as indicated by the reduced numbers of cells in the earlier spermatogenic stages (Table 2). The period between August and November was a recovery period in which 40 30 MEAN OOCYTE SIZE (pm) 10 unspawned gametes from the previous sea- son were either resorbed or released and during which a build-up of spermatogonia and sperm-morulae occurred (Fig. 6). Typical spermatogenesis was almost completely ab- sent during this time, most of the activity being directed towards the production of sperm- morulae. An increase in the tempo of sperma- togenesis occurred from late November until late April during which intensive production of spermatozoa and enlarging of the acini oc- curred (Figs. 7-8). This phase was also char- acterized by large clusters of spermatids that were absent in the months prior to November. 20 SAMPLE SIZE 30 40 FIG. 4. Relationship between number of oocytes measured (sample size) and mean oocyte size in Cucumerunio novaehollandiae from the Macleay River: 26 January 1983, Honeymoon Bend (triangles); 26 March 1983, Toorooka (crosses); 15 July 1983, Toorooka (closed circles). REPRODUCTION IN AUSTRALIAN HYRIIDAE 191 100 O O 14 АК RS FREQUENCY OF EACH SPERMATOGENIC STAGE (%) > € O on m = O о 4 2 SPERMATOZOA SPERM - MORULAE E SS ASS НИ NOV DEC JAN FE NOS SSS ИИ OO AAAS ALLA MOSSOS E ISSO] e. OMA œ < > aD > U >) > < < € г = = Г 1983 Be SPERMATIDS S S SPERMATOCYTES SPERMATOGONIA FIG. 5. Monthly changes in the proportion of the spermatogenic stages represented in the testes of Cucumerunio novaehollandiae from the upper reaches of the Macleay River. Spawning occurred in the period of late March and April but was incomplete and many spermatozoa remained in the acini. After spawning, spermatogenesis continued from the remaining spermatogenic cells and the acini were again packed with spermatozoa in June and July (Fig. 9). Hence, three phases of spermatogenesis were recognized: 1) a recovery phase characterized by nests of spermatogonia and a high proportion of sperm-morulae; 2) an active phase characterized by nests of primary spermatocytes, secondary sper- matocytes and spermatids. Spermatozoa were present in the lumen of the acini, and 3) a maturation phase in which sperma- tozoa were abundant but spermatogenic ac- tivity was reduced. Female cycle The following description of seasonal оо- cyte production is for females from the upper reaches of the river. Both developing oocytes, connected to the follicle walls by a stalk, and mature oocytes were absent from the ovary between August and November. Mature oo- cytes remaining from the previous reproduc- tive period were resorbed early during this period (Fig. 10). Nutritive granules were pro- lific along the follicle walls which, by late Oc- tober, were thickened with oogonia (Fig. 11). 192 JONES, SIMPSON 8 HUMPHREY TABLE 2. Monthly descriptions of spermatogenesis in Cucumerunio novaehollandiae. JULY: Both spent and mature individuals were present. The acini of mature individuals were large and closely spaced. The acini were filled with mature spermatozoa and very few of the earlier spermatogenic stages were present. The acini of spent individuals were small and widely spaced. Few spermatozoa were present in the lumen of the acini and spermatogonia and sperm-morulae were abundant. AUGUST: Acini were slightly reduced in size when compared with individuals from July and fewer spermatozoa were present. Acini walls were thicker and there were numerous clusters of spermatogonia. Sperm-morulae were common. SEPTEMBER: Acini were small, widely spaced and completely filled with sperm-morulae. Nests of spermatogonia were common and very few spermatozoa were present. OCTOBER: Very little change from the previous month. NOVEMBER: Clusters of primary spermatocytes appeared in some individuals and many of the sperm- morulae appeared to be metamorphosing into clumps of spermatozoa. DECEMBER: Spermatozoa began to build up in the acini lumina and clusters of spermatocytes and spermatids were abundant. Sperm-morulae still dominated the acini. JANUARY: In all individuals, spermatozoa were abundant and large clusters of spermatids in various stages of metamorphosis into spermatozoa were present. Spermatocytes were very common but there were fewer sperm-morulae. FEBRUARY: Acini were large and closely spaced. Spermatogenesis was slightly more advanced than the previous month. MARCH: Acini were densely packed with spermatozoa with peripheral bands of spermatocytes and spermatids. APRIL: Acini were reduced in size and contained fewer spermatozoa than in March. The incidence of sperm-morulae increased. MAY: Little change from April. JUNE: Acini were again filled with spermatozoa but very few earlier spermatogenic stages were present. JULY: Little change from the previous month except for the appearance of bands of spermatogonia and sperm-morulae around the periphery of acini. Rapid growth of primary oocytes owing to vitellogenesis occurred from November (Fig. 12) and continued until the end of March when the acini were packed with mature oocytes. A sharp decrease in mean oocyte diameter oc- curred between March and April, coinciding with the movement of eggs into the marsupia. Following spawning, there was a second rapid build-up in the numbers of mature оо- cytes and acini were again packed with oo- cytes by mid-July (Fig. 13). Oogenesis in the downstream population lagged behind the up- stream population and spawning did not occur until late-April or early-May (Fig. 14). Hence, three phases could be recognized in the oo- genic cycle: 1) a recovery period from the end of the previous breeding season until late October (upstream) or December (downstream). Ripe oocytes remaining from the previous repro- ductive period were resorbed or passed out of the ovaries and the follicle walls thickened owing to a build-up of oogonia and nutrient reserves; 2) a growth phase from late October until about the end of March (upstream) and from December until about May (downstream). During this phase vitellogenesis occurred and the acini became swollen as they filled with primary oocytes, and 3) a spawning phase from late March (up- stream) until early May (downstream). Description of the glochidia Mature glochidia of all species were col- lected, although these were found in the mar- supia ofonly one female each of A. profuga and Hyridella sp. The glochidial morphometrics of these species are summarized in Table 1. The glochidia of C. novaehollandiae (Figs. 15-17) are extremely small (64 x 55 um), globose and of suboval outline. Two short hooks are present on the inside of each shell valve. Scanning electron microscopy has re- vealed a convoluted surface structure of the shell valves (K. F. Walker, personal commu- nication). Glochidia of H. australis (Figs. 18-20) are small in comparison with other hyriids, measuring 95 x 74 um. The glochidia are slightly elongate, subtriangular and the shell valves are perforated by small pores. The larvae are double-hooked and the teeth are recurved and set on acommon base. The REPRODUCTION IN AUSTRALIAN HYRIIDAE 193 97 LMI E ao ap: q se и о a Ps Da A Bee A DR 4 tee; KT e & У: +) ter à * > Е x > BE 4. AS PS г РЕЗ ‘ ‘<> +5 h р 4 i + j 3 Ne я wie e, % ya: & “ AA Le E = — ’ >? ÉS > TH, nn ar 1 a m y в 5200 1060" 1 al TUT Sie n°60 O4Y RS Tre cas. 194 JONES, SIMPSON 8 HUMPHREY glochidia of H. depressa (Figs. 21-25) are large (244 x 253 um), subtriangular and pos- sess a single, bifurcated hook set slightly off- centre on the ventral surface of each valve. The glochidia of Hyridella sp. also possess a bifurcated hook; they are very similar in mor- phology to H. depressa but are slightly smaller (233 x 239 um). The glochidia of A. profuga compare well with published descriptions for this species (McMichael & Hiscock, 1958) (see Table 1). Brooding period Cucumerunio novaehollandiae C. novaehollandiae is a winter breeder with a highly seasonal cycle (Figs. 26-27). The brooding period extends from April to August. The reproductive cycle of downstream popu- lations, however, lags slightly behind that of the populations upstream (Table 3). When collected on 9 May 1983, immedi- ately after a major flood (Fig. 2), most of the females from the downstream population had just moved eggs into their marsupia. Five weeks later these were at an intermediate glochidial stage and mature glochidia were being released in mid-July, 9 weeks after spawning. Mussels from the upstream station were releasing glochidia during mid-May and had finished by mid-July. On the basis of a similar progression in development, females in the upstream population probably spawned in late March or early April. Hyridella australis The reproductive activity in this species ex- tended throughout most of the year except for the coldest month, July (Fig. 28). Most of the November collection from the upstream pop- ulation aborted their larvae. Consequently, it was not possible to accurately separate fe- males with mature glochidia in their gills (stage III) from those which had released their larvae (stage IV). There were at least two, probably three, brooding periods during 1982/1983, indicated by the high proportion of females carrying glochidia at different times of the year. The four week intervals between sampling pre- cluded accurate assessment of the time elapsing between fertilization and glochidial release. However, the gestation period ap- pears to have been about eight weeks during the summer months when the water temper- ature was in the vicinity of 27°C, and about 11 weeks during the autumn when the water temperature was lower (11°С). Breeding pe- riods must have occurred between 26 No- vember and 22 December 1982, and again between 28 February and 26 March in 1983, that is, approximately 13 to 14 weeks apart. One of us (C.L.H.) has observed the re- lease of glochidia of H. australis while diving in clean river conditions in January 1985. Glochidia are extruded from the exhalent si- phon of mature females in a wormlike conglu- tinate. The conglutinate, approximately 4 cm in length, is tan coloured and bears white, transversely striated bands along one side. Laboratory examination revealed that the tanned material is composed entirely of ma- ture glochidia bound together in a mucous matrix with the white striations comprising a loosely binding tissue. Distinct, rhythmical pumping actions of the exhalent siphon were noted that caused the wormlike mass to wave and fall about the siphon where it was posteriorly inserted or attached. Discharged conglutinates were also found lying free and intact on the sediments adjacent to adult females. Hyridella depressa, Hyridella sp. and Alathyria profuga Scant data were obtained for the brooding periods of these species (Table 4). However, H. depressa and Hyridella sp. were gravid during the spring and summer. It is possible that both species breed more than once per year, as in H. australis. Females of A. profuga bearing glochidia were present only in mid- MR ooo FIGS. 6-13. Stages of gonadal activity in Cucumerunio novaehollandiae collected from the upper reaches of the Macleay River. FIG. 6. Sperm-morulae in an inactive testis, 26 October 1982. FIG. 7. Maturing testis, 26 February 1982. FIG. 8. Mature testis immediately prior to spawning, 26 March 1983. FIG. 9. Mature spermatozoa in a testis, 15 July 1983. FIG. 10. Deteriorating oocytes (arrow) from an ovary, 27 July 1982. FIG. 11. Ovary in the resting phase and filled with nutrient matter, 26 October 1982. FIG. 12. Early oogenesis, 26 November 1982. FIG. 13. Mature ovary, 15 July 1983, N. nutrient granules; PO, primary oocytes; SG. spermatogonia; SM, sperm-morulae; SSP, secondary spermatocytes; ST, spermatids; SZ, spermatozoa. REPRODUCTION IN AUSTRALIAN HYRIIDAE 195 70 E = 60 x LJ > > < 50 a Li E o 40 O о < ш 30 = Upstream A Downstream Downstream o O ss A a o O O o ПЕЧИ ee el М A M J Jd UTA ES О м 1982 JE 1983 FIG. 14. Seasonal variation in primary oocyte size of Cucumerunio novaehollandiae. The bars are equal to 1 standard error of the mean. Brooding periods (stippled areas) are shown for upstream and downstream populations. -summer, although very few specimens were collected at other times of the year. DISCUSSION Gametogenesis occurred throughout most of the year in Cucumerunio novaehollandiae, with peak activity during the hottest months of the year. This is typical of unionaceans from temperate climates (van der Schalie & van der Schalie, 1963; Yokley, 1972; Giusti et al., 1975; Heard, 1975; Smith, 1979; Zale 8 Neves, 1982). Mature ova were produced during the summer and autumn, coinciding with the peak in spermiogenesis. A highly synchronized breeding season oc- curred during the autumn. However, males only partially spawned. This could reduce the risk of mistiming the release of gametes by the sexes. In other temperate fresh-water mussels both males and females spawn а the same time (Zale & Neves, 1982). Giusti et al. (1975), however, noted that not all male Anodonta cygnea spawned at the same time and Elliptio complanatus males release sper- matozoa over a time span of about one 196 JONES, SIMPSON & HUMPHREY 50um 50 um 20 FIGS. 15-17. Glochidium of Cucumerunio novaehollandiae. FIG. 15. Lateral view. FIG. 16. Ventral view. FIG. 17. Outline of the shell valves. FIGS. 18-20. Glochidium of Hyridella australis. FIG. 18. Oblique ventral view. FIG. 19. Lateral view. FIG. 20. Outline of the shell valves. month, which overlaps the period when fe- males are likely to be receptive to fertilization (Matteson, 1948). In contrast to C. novaehollandiae, other members of the southeastern Australian fresh-water mussel fauna may have much broader breeding seasons. Atkins (1979), in a study of a coastal Victorian stream, found Hyridella drapeta glochidia on fish throughout the year, with peak infections during the spring. Similarly, Hiscock (1951) found Ve- lesunio ambiguus glochidia on fish throughout the year except between May and Septem- ber, when no fish were examined. In agree- ment with Hiscock's data, Walker (1981) found that glochidia may be present in the marsupia of Velesunio ambiguus throughout the year, although two peaks in gravidity were recognized; one in spring and the other during late summer/early autumn. The results of the present study indicate that H. australis, H. depressa and Hyridella sp. breed throughout much of the warmer part of the year. This also appears to be the case for the New Zealand species, H. menziesi (Percival, 1931) and prolonged breeding seasons may well prove to be characteristic of the genus Hyridella. A series of broadly synchronized reproduc- REPRODUCTION IN AUSTRALIAN HYRIIDAE 197 24 100 um FIGS. 21-25. Glochidium of Hyridella depressa. FIG. 21. Oblique view. FIG. 22. Ventral view. FIG. 23. Lateral view of the forked teeth and the protuberances at their bases. FIG. 24. Structure of the teeth. FIG. 25. Outline of the shell valves. tive cycles, such as occurs in H. australis, have been infrequently reported for fresh- water mussels from temperate regions (Allen, 1924, cited by Heard, 1975) but asynchro- nous, repetitive breeding cycles occur throughout the year in some tropical union- aceans (Kenmuir, 1981b; Humphrey, 1984). More detailed investigations of the reproduc- tive biology of temperate zone unionaceans may find repetitive breeding to be more wide- spread than presently indicated; especially since some tachytictic species such as Unio spp. have the potential to produce many broods per year (Dudgeon & Morton, 1983). There are, however, several cases of fresh- water mussels breeding twice per year (Le- fevre & Curtis, 1910, 1912; Ortmann, 1912; Wood, 1974; Heard, 1975). Successive build- ups of large numbers of mature gametes in- dicated that C. novaehollandiae may breed 198 JONES, SIMPSON 8 HUMPHREY GLOCHIDIA RELEASED GLOCHIDIA ABORTED GLOCHIDIA a о ANNO EMBRYOS EGGS % OF FEMALES IN EACH STAGE JAS ONDJFMAM J J n= 0 0 10 13 15 9 25 20 19 29 6 2722 35 1982 1983 el 1 10) N : | ge q E Q о APA < ВИ uy RÓS = N BR un D 2 2 Ne: = Be LE [ra O x 1982 1983 % OF FEMALES IN EACH STAGE a о AS ОМ РЕМ AIM Jad 4 N= 4 0 33 12 18 16 14 10 10 15 14 1982 1983 FIGS. 26-28. Seasonal distribution of the reproduc- tive stages in the gills of freshwater mussels from the Macleay River. FIG. 26. Cucumerunio novaehol- landiae, upstream population. FIG. 27. C. novaehollandiae, downstream population. FIG. 28. Hyridella australis, upstream population. For indi- vidual animals, open circles = non-gravid females; closed circles = embryos in marsupia; closed trian- gles = glochidia in marsupia; open triangles = non-gravid but glochidia recently released. TABLE 3. Proportion of Cucumerunio novaehol- landiae with different stages of developing larvae in gills throughout the Macleay River, 9 May 1983. Station Stage of development 1 2 3 4 | 0.0) 53 5.08400 Il 0.0 0.0 0.0 50.0 Ш Early glochidia 0.0 94.7 0.0 0.0 Intermediate glochidia 100.0 0.0 95.0 50.0 Mature glochidia 0.0 00 0.0100 twice per year: once in autumn and again in late winter/early spring. This pattern only oc- curred in the upper reaches of the river. Sim- папу Porter & Horn (1980) detected area vari- ation in the reproductive cycle of a North American mussel in Lake Waccamaw. Also, variation in reproductive patterns within the one species of fresh-water mussel has been previously shown between rivers (Bauer, 1979) and at different latitudes (Matteson, 1948; Smith, 1976; Kenmuir, 1981a). Sperm-morulae are of widespread occur- rence in the Bivalvia (Bloomer, 1935, 1936, 1939; Coe 8 Turner, 1938; Ropes 8 Stickney, 1965; van der Schlie & Locke, 1941), includ- ing the Hyriidae (Heard, 1975; Peredo 8 Par- ada, 1984; Humphrey, 1984). Sperm-morulae occurred seasonally in C. novaehollandiae and were absent during the period of spermi- ogenesis. Seasonality of occurrence of sperm-morulae has also been described for other bivalves (Heard, 1975; Smith, 1979). Bloomer (1946) inferred that sperm-morulae metamorphosed into spermatozoa in Anodonta cygnea from the observation that sperm-morulae disappeared prior to the ap- pearance of spermatozoa. This was also the case for C. novaehollandiae in which the pro- cess of atypical spermatogenesis was ob- served to the stage where the sperm-morulae consisted of elongate spermatids immediately prior to active spermatogenesis. Coe & Turner (1938) thought that sperm-morulae in Mya arenaria underwent cytolysis but there was no evidence of this in C. novaeholland- гае. Loosanoff & Davis (1952) and Sastry (1963) have demonstrated from their experi- ments with marine bivalves the importance of temperature as an activator of spawning. En- vironmental cues responsible for initiating REPRODUCTION IN AUSTRALIAN HYRIIDAE 199 TABLE 4. Miscellaneous brooding records for individuals of the minor species found in the Macleay River. Embryos Hs* Ap* SE 2 Date 3 Aug. 1982 26 Oct. 1982 7 Nov. 1982 26 Nov. 1982 22 Dec. 1982 22 Jan. 1983 26 Feb. 1983 26 Mar. 1983 26 Apr. 1983 16 May 1983 1/2 Jan. 1985 16 Jan. 1985 4 —- 6 | mw | lol srl oo | | nods к = № State of marsupia Glochidia Empty На Hs Ар На Hs Ар 3 — — 5 — — == = 1 = == == — — — — — 1 — — — — — 2 == u = 2 = = — — — 1 — 1 = == an 1 = 2 6 = 3 — 1 9 5 — 2 6 3 6 “Hd = Hyridella depressa; Hs = Hyridella sp.; Ap = Alathyria profuga spawning in fresh-water mussels have not been experimentally identified but numerous authors have brought attention to the correla- tion between breeding season and water tem- perature (Harms, 1909; Tudorancea, 1969; Yokley, 1972; Kenmuir, 1981b; Zale & Neves, 1982). Reproductive activity in H. australis may well be limited by low temperatures since gravid females were absent during the coldest months of the year. Similarly, low water tem- peratures were found to reduce breeding ac- tivity in Velesunio angasi, a tropical northern Australian species (Humphrey, 1984). Con- versely, spawning in C. novaehollandiae ap- pears to be related to falling water tempera- tures. Spawning in this species took place immediately after a flood which resulted in a sudden drop in water temperature. Sudden temperature changes brought about by the Danube high floods were thought to initiate breeding in Unio tumidus (Tudorancea, 1969, 1972) and a similar mechanism may be re- sponsible for the highly synchronous breeding season in C. novaehollandiae. The duration of the brooding periods in C. novaehollandiae and H. australis are compa- rable with many temperate unionaceans. Organogenesis is completed in two weeks and glochidia are mature in one month in Elliptio complanatus (Matteson, 1948). Devel- opment took two months in Anodonta cygnea (Wood, 1974; Giusti et al., 1975), one month for Unio tumidus (Tudorancea, 1969) and Margaritifera margaritifera (Smith, 1976) and seven to eight weeks for four species of fresh- water mussels from the Upper Tennessee River drainage, U.S.A. (Zale & Neves, 1982). Yokley (1972) found that the brooding period in Pleurobema cordatum took four to six weeks depending on water temperature. Wa- ter temperature also affects the gestation pe- riod in М. margaritifera (Smith, 1976). The results of the present study suggest that the gestation periods of H. australis and C. novaehollandiae are similarly influenced by water temperature. Glochidial release by way of worm-like con- glutinates such as occurs in H. australis has not previously been reported in hyriid union- aceans, although this mode of release is com- mon to members of several genera of North American unionids (Kat, 1984). The larval conglutinates reported in unionids resemble various vermiform food items of the fish host and thereby enhance the likelihood of host contact (Chamberlain, 1934; Kat, 1984). At this early stage of investigation the appear- ance and rhythmical waving action of the con- glutinates of H. australis suggest a similar mimicry of host food items as is displayed by the North American species. Of the species examined in the present study, only the glochidia of H. australis and A. profuga have been previously described (McMichael 8 Hiscock, 1958). However, the descriptions by McMichael 8 Hiscock for the glochidia of H. australis match in size and general outline those for the glochidia of H. depressa, described here for the first time. п the present study, identifications of adults 200 JONES, SIMPSON 8 HUMPHREY were carefully checked and therefore it ap- pears that McMichael & Hiscock incorrectly assigned a description of H. depressa gloch- idia to H. australis. The glochidia of the Australian hyriids can no longer be viewed as a group which vary only slightly in size and shape (Atkins, 1979; Walker, 1981). C. novaehollandiae and H. australis produce much smaller glochidia than any other Australian fresh-water mussel. In- deed, in relation to other unionaceans the glochidia of C. novaehollandiae are amongst the smallest known, exceeded only by M. margaritifera (47 pm diameter) (Roscoe 4 Redelings, 1964) and Margaritana (= Marg- aritifera) monodonta (50 x 52 um) (Lefevre 8 Curtis, 1912). Moreover, the morphology of the hooks of H. australis and C. novaehollan- diae differ markedly from previous descrip- tions of glochidia of Australian fresh-water mussels. Typically, the Hyriidae possess a single, curved hook on each valve which may, or may not, have a forked point (Parodiz & Bonetto, 1963). These hooks, however, have been greatly modified in H. australis and C. novaehollandiae. The glochidia of H. australis bear a protruding double-hook on each valve. H. glenelgis also produces double-hooked glochidia (K. F. Walker, personal communica- tion). In C. novaehollandiae the glochidia pos- sess a pair of short, recurved hooks on each valve. Further, finer details of the hook mor- phology of H. depressa glochidia show clear differences between this species and its con- geners. H. australis, H. depressa and H. drapeta have almost identical geographical ranges and possess only slight conchological and anatomical differences (McMichael 8 His- cock, 1958). The distinctive characteristics of the glochidia of each (namely, size and shape of the shell and structure of the hooks) are strong evidence that these mussels are separate species and not ecophenotypic vari- ants of the one species as hinted at by Walker (1981). On the contrary, the limited evidence presented in this study suggests that the Hyridella complex may need to be subdivided even further. The hooks of H. de- pressa and H. australis appear to be modifi- cations of a single hook in which a forked point has evolved. In H. australis this is more advanced. Even in C. novaehollandiae it ap- pears as though the two hooks have evolved from a single hook through to the stage where they are now almost separate. ACKNOWLEDGEMENTS Thanks are due Dr. K. F. Walker who pro- vided information on fresh-water mussels of southeastern Australia. We are grateful also for the services of Ms. L. Bridges (illustra- tions), Mr. R. Hobbs and Ms. L. Keogh (pho- tography). REFERENCES CITED ATKINS, L. G., 1979, Observations of the glochidial stage of the fresh-water mussel Hyridella (Hyrid- ella) drapeta (lredale) (Mollusca: Pelecypoda). Australian Journal of Marine and Freshwater Re- search, 30: 411—416. AUSTRALIAN WATER RESOURCES COUNCIL, 1976 ['1975"], Review of Australia’s water re- sources. Department of Natural Resources, Aus- tralian Water Resources Council, Canberra: Aus- tralian Government Publishing Service, 170 p. BAUER, G., 1979, Untersuchungen zur Fort- pflanzungsbiologie der Flussperlmuschel (Margaritana margaritifera) im Fichtelgebirge. Ar- chiv für Hydrobiologie, 85: 152-165. BLOOMER, H. H., 1935, A further note on the sex of Anodonta cygnea L. Proceedings of the Malacological Society of London, 21: 304-321. BLOOMER, H. H., 1936, A note on the sex of Anodonta anatina. Proceedings of the Malacolo- gical Society of London, 22: 129-134. BLOOMER, H. H., 1939, A note on the sex of Pseudanodonta Bourguignat and Anodonta Lamarck. Proceedings of the Malacological So- ciety of London, 23: 285-297. BLOOMER, H. H., 1946, The seasonal production of spermatozoa and other notes on the biology of Anodonta cygnea (L.). Proceedings of the Malacological Society of London, 27: 62-68. CHAMBERLAIN, T. K., 1934, The glochidial con- glutinates of the Arkansas Fanshell, Cyprogenia aberti (Conrad). Biological Bulletin, 66: 55-61. CLARKE, A. H. 8 BERG, C. O., 1959, The fresh- water mussels of central New York. Memoirs of the Cornell University Experimental Station, 367: 1-79. COE, W. В. & TURNER, Н. J., 1938, Development of the gonads and gametes in the soft-shell clam (Mya arenaria). Journal of Morphology, 62: 91-111. COKER, В. E., SHIRA, A. F., CLARK, Н. W. € HOWARD, A. D., 1921, Natural history and prop- agation of freshwater mussels. Bulletin of the United States Bureau of Fisheries, 37: 77-181. DARTNALL, H. J. ©. 8 WALKEY, M., 1979, The distribution of glochidia of the swan mussel, Anodonta cygnea (Mollusca) on the three-spined REPRODUCTION IN AUSTRALIAN HYRIIDAE 201 stickleback Gasterosteus aculeatus (Pisces). Journal of Zoology (London), 189: 31-37. DUDGEON, D. & MORTON, B., 1983, The popu- lation dynamics and sexual strategy of Anodonta woodiana (Bivalvia: Unionacea) in Plover Cove Reservoir, Hong Kong. Journal of Zoology (Lon- don), 201: 161-183. ELLIOTT, J. M., 1977, Some methods for the sta- tistical analysis of samples of benthic inverte- brates. Freshwater Biological Association, Scien- tific Publications, No. 25: 1-144. GHOSH, C. 8 GHOSE, K. C., 1972, Reproductive system and gonadal activities in Lamellidens marginalis (Simpson, 1900). Veliger, 14: 283-288. GIESE, A. C., 1959, Comparative physiology: an- nual reproductive cycles of marine invertebrates. Annual Review of Physiology, 21: 547-576. GIUSTI, Е. L., CASTAGNOLO, L., FARINA, М. & RENZONI, A., 1975, The reproductive cycle and glochidium of Anodonta cygnea. L. from Lago Trasimeno (central Italy). Monitore Zoologico Italiano, 9: 99-118. HARMS, W., 1909, Postembryonale Entwicklun- gsgeschichte der Unioniden. Zoologische Jahrbücher, Abteilung für Anatomie und Ontoge- nie der Tiere, 28: 325-386. HAUKIOJA, E. & HAKALA, T., 1978, Life history evolution in Anodonta piscinalis (Mollusca, Pe- lecypoda). Oecologia, 35: 253-266. HEARD, W. H., 1975, Sexuality and other aspects of reproduction in Anodonta (Pelecypoda: Union- idae). Malacologia, 15: 81-103. HISCOCK, I. D., 1951, A note on the life history of the Australian freshwater mussel, Hyridella aus- tralis Lam. Transactions of the Royal Society of South Australia, 74: 146-148. HOWARD, A. D., 1951, A river mussel parasitic on a salamander. Natural History Miscellanea, Chi- cago, No. 77: 1-6. HUMPHREY, C. L., 1984, Biology and ecology of the freshwater mussel Velesunio angasi (Bivalvia: Hyriidae) in the Magela Creek, Alligator Rivers Region, Northern Territory. Ph.D. Thesis, University of New England, Armidale, 479 p. KAT, P. W., 1984, Parasitism and the Unionacea (Bivalvia). Biological Reviews, 59: 189-207. KENMUIR, D. H. S., 1981a, Seasonal breeding activity in freshwater mussels (Lamellibranchi- ata: Unionacea) in Lake Kariba and Lake Mcllwaine. Transactions of the Zimbabwe Sci- ence Association, 60(4): 18-23. KENMUIR, D. H. S., 1981b, Repetitive spawning behaviour in two species of freshwater mussels (Lamellibranchiata: Unionacea) In Lake Kariba. Transactions of the Zimbabwe Science Associa- tion, 60(4): 49-56. LEFEVRE, G. & CURTIS, W. C., 1910, Reproduc- tion and parasitism in the Unionidae. Journal of Experimental Zoology, 9: 79-115. LEFEVRE, С. 8 CURTIS, W. C., 1912, Studies on the reproduction and artificial propagation of freshwater mussels. Bulletin of the United States Bureau of Fisheries, 30: 105-201. LOMTE, V. $. & NAGABHUSHANAM, R., 1969, Reproductive cycle in the freshwater mussel Par- reysia corrugata. Marathwada University Journal of Science, 8: 113-118. LOOSANOFF, V. |. 8 DAVIS, H. C., 1952, Tem- perature requirements for maturation of gonads of northern oysters. Biological Bulletin (Woods Hole, Massachusetts), 103: 80-96. MATTESON, M. R., 1948, Life history of Elliptio complanatus (Dillwyn, 1817). American Midland Naturalist, 40: 690-723. McMICHAEL, D. F. & HISCOCK, I. D., 1958, A monograph of the freshwater mussels (Mollusca: Pelecypoda) of the Australian region. Australian Journal of Marine and Freshwater Research, 9: 372-508. MURPHY, G., 1942, Relationship of the freshwater mussel to trout in the Truckee River. California Fish and Game, 28: 89-102. NAGABHUSHANAM, В. & LOHGAONKER, A. L., 1978, Seasonal reproductive cycle in the mussel, Lamellidens corrianus. Hydrobiologia, 61: 9-14. NEGUS, С. L., 1966, A quantitative study of growth and production of unionid mussels in the River Thames at Reading. Journal of Animal Ecology, 35: 513-532. ORTMANN, A. E., 1912, Notes upon the families and genera of the naiades. Annals of the Carn- egie Museum, 8: 222-365. PARODIZ, J. J. & BONETTO, A. A., 1963, Taxon- omy and zoogeographic relationships of the South American naiades (Pelecypoda: Union- acea and Mutelacea). Malacologia, 1: 179-213. PENNAK, R. W., 1953, Freshwater invertebrates of the United States. Ronald Press, New York, p. 694-707. PERCIVAL, E., 1931, A note on the life history of Diplodon lutulentus Gould. Transactions and Proceedings of the New Zealand Institute, 62: 86-91. PEREDO, S. 8 PARADA, E., 1984, Gonadal orga- nization and gametogenesis in the fresh-water mussel Diplodon chilensis chilensis (Mollusca: Bivalvia). Veliger, 27: 126-133. PORTER, Н. J. & HORN, К. J., 1980, Freshwater mussel glochidia from Lake Waccamaw, Colum- bus County, North Carolina. American Malaco- logical Union Bulletin, 1980: 13-17. RAND, Т. С. 4 WILES, M., 1982, Species differen- tiation of the glochidia of Anodonta cataracta Say, 1817 and Anodonta implicata Say, 1829 (Mollusca: Unionidae) by scanning electron mi- croscopy. Canadian Journal of Zoology, 60: 1722-1727. ROPES, J. W. 8 STICKNEY, A. P., 1965, Repro- ductive cycle of Mya arenaria in New England. Biological Bulletin (Woods Hole, Massachu- setts), 128: 315-327. ROSCOE, E. J. & REDELINGS, S., 1964, The ecology of the freshwater pearl mussel, 202 JONES, SIMPSON 8 HUMPHREY Margaritifera margaritifera (L.). Sterkiana, 16: 19—32. SASTRY, А. N., 1963, Reproduction of the bay scallop. Aequipecten irradians Lamarck. Influ- ence of temperature on maturation and spawn- ing. Biological Bulletin (Woods Hole, Massachu- setts), 125: 146-153. SCHALIE, H. van der, 1938, The naiad fauna of the Huron River, in south-eastern Michigan. Univer- sity of Michigan Museum of Zoology Miscella- neous Publications, 40: 1-83, 12 pl., map. SCHALIE, H. van der & LOCKE, F., 1941, Her- maphroditism in Anodonta grandis, a freshwater mussel. Occasional Papers of the Museum of Zoology, University of Michigan, 432: 1—7. SCHALIE, H. van der & SCHALIE, van der, A., 1963, The distribution, ecology and life history of the mussel, Actinonaias ellipsiformis (Conrad), in Michigan. Occasional Papers of the Museum of Zoology, University of Michigan, 663:1-17. SESHAIYA, R. V., 1941, Tadpoles as hosts for the glochidia of the freshwater mussel. Current Sci- ence, 10: 535-536. SIMPSON, R. D., 1977, The reproduction of some littoral molluscs from Macquarie Island (sub- Antarctic). Marine Biology, 44: 125-142. SMITH, D. G., 1976, Notes on the biology of Margaritifera margaritifera (Lin.) in central Mas- sachussetts. American Midland Naturalist 96: 252-256. SMITH, D. G., 1979, Sexual characteristics of Margaritifera margaritifera (Linnaeus) popula- tions in central New England. Veliger, 21: 381-383. STEIN, C. B., 1969, Gonad development in the three-ridge naiad Amblema plicata (Say, 1817). Annual Report of the American Malacological Union, 1969: 30. SURBER, T., 1912, Identification of the glochidia of freshwater mussels. United States Bureau of Fisheries Document, No. 771: 1-10, 3 pl. TRDAN, R. J., 1981, Reproductive biology of Lamp- silis radiata siliquoidea (Pelecypoda: Unionidae). American Midland Naturalist, 106: 243-248. TUDORANCEA, C., 1969, Comparison of the pop- ulations of Unio tumidus Philipsson from the complex of Crapina-Jijila marshes. Ekologia Polska., A, 17: 185-204. TUDORANCEA, C., 1972, Studies on Unionidae populations from the Crapina-Jijila complex of pools (Danube zone) liable to inundation. Hydrobiologia, 39: 527-561. WALKER, K. F., 1981, The ecology of freshwater mussels in the River Murray. Australian Water Resources Council Technical Paper, No. 63. WILES, M., 1975, The glochidia of certain Union- idae (Mollusca) in Nova Scotia and their fish hosts. Canadian Journal of Zoology, 53: 33- 41. WOOD, E. M., 1974, Development and morphology of the glochidium larva of Anodonta cygnea (Mollusca: Bivalvia). Journal of Zoology (Lon- don), 173: 1-13. YOKLEY, P., 1972, Life history of Pleurobema cordatum (Rafinesque, 1820) (Bivalvia: Union- асеа). Malacologia, 11: 351-364. ZALE, А. V. & NEVES, В. J., 1982, Reproductive biology of four freshwater mussel species (Mollusca: Unionidae) in Virginia. Freshwater In- vertebrate Biology, 1: 17-28. WHY NOT SUBSCRIBE TO MALACOLOGIA? ORDER FORM Your name and address Send U.S. $17.00 for a personal subscription (one volume) or U.S. $27.00 for an institutional subscription. Make checks payable to “MALACOLOGIA.” Address: Malacologia, Academy of Natural Sciences Nineteenth and the Parkway, Philadelphia PA 19103, U.S.A. The Division of Mollusks, Department of Invertebrate Zoology, National Museums of Natural History, Smithsonian Institution announces the availability of two fellowships to be awarded to graduate students of systematic malacology: 1. Rosewater Fellow Award (up to $500) 2. Smithsonian—COA Fellow Award (up to $1,000) These awards are to help support students for short term research visits to the collections and libraries of the Division of Mollusks, National Museum of Natural History and are to be used for systematic studies of Mollusca. Funds can help cover travel, subsistence, and research costs [xerox, postage, etc.]. Interested students should submit a 1-page proposal, a budget with indication of matching funding, if available, and a supporting letter from their faculty advisors. Deadline for applications is March 1, 1986. Awards will be announced on April 1, 1986. 27(1) INSTRUCTIONS FOR AUTHORS 1. MALACOLOGIA publishes original re- search on the Mollusca that is of high quality and of broad international interest. Papers combining synthesis with innovation are par- ticularly desired. While publishing symposia from time to time, MALACOLOGIA encour- ages submission of single manuscripts on diverse topics. Papers of local geographical or systematic interest should be submitted elsewhere, as should papers whose primary thrust is physiology or biochemistry. Nearly all branches of malacology are represented on the pages of MALACOLOGIA. 2. Manuscripts submitted for publication are received with the tacit understanding that they have not been submitted or published elsewhere in whole or in part. 3. Manuscripts may be in English, French, German or Spanish. Papers in lan- guages other than English must include a translation of the Abstract in English. Authors desiring to have their abstracts published in other languages must provide the translations (complete with main titles). Include all foreign accents. Both American and British spellings are allowed. 4. Unless indicated otherwise below, con- tributors should follow the recommendations in the Council of Biology Editors (CBE) Style Manual (ed. 5, 1983) available for U.S. $24.00 from CBE, 9650 Rockville Pike, Be- thesda, MD 20814, U.S.A. 5. Be brief. 6. Manuscripts must be typed on one side of good quality white paper, double-spaced throughout (including the references, tables and figure captions), and with ample margins. Tables and figure captions should be typed on separate pages and put at the end of the manuscript. Make the hierarchy of headings within the text simple and consistent. Avoid internal page references (which have to be added in page proof). 7. Choose a running title (a shortened version of the main title) of fewer than 50 letters and spaces. MALACOLOGIA 1986 8. Provide a concise and informative Ab- stract summarizing not only contents but re- sults. A separate summary generally is superfluous. 9. Supply between five and eight key (topic) words to go at the end of the Ab- stract. 10. Use the metric system throughout. Mi- cron should be abbreviated рт. 11. Illustrations are printed either in one column or the full width of a page of the journal, so plan accordingly. The maximum size of a printed figure is 13.5 х 20.0 ст (preferably not as tall as this so that the caption does not have to be on the opposite page). 12. Drawings and lettering must be dark black on white, blue tracing, or blue-lined paper. Lines, stippling, letters and numbers should be thick enough to allow reduction by Y. or ‘3. Letters and numbers should be at least 3mm high after reduction. Several drawings or photographs may be grouped together to fit apage. Photographs are to be high contrast. High contrast is especially im- portant for histological photographs. 13. All illustrations are to be numbered se- quentially as figures (not grouped as plates or as lettered subseries), and are to be arranged 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 conven- ient lengths (e.g. “200 jm,” not “163 um”). Magnifications in captions are not acceptable. 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 dupli- cate information given in the text. Each let- tered abbreviation labeling an individual fea- ture in a figure must either be explained in each caption (listed alphabetically), or be grouped in one alphabetic sequence after the Methods section. Use the latter method if many abbreviations are repeated on different figures. 17. Tables are to be used sparingly, and vertical lines not at all. 18. References cited in the text must be in the Literature Cited section and vice versa. Follow a recent issue of MALACOLOGIA for bibliographic style, noting especially that se- rials are cited unabbreviated. Supply pagina- tion for books. Supply information on plates, etc., only if they are not included in the pagination. 19. In systematic papers, synonymies should not give complete citations but should relate by author, date and page to the Litera- ture Cited section. 20. For systematic papers, all new type- species must be deposited in museums where they may be studied by other scien- tists. Likewise MALACOLOGIA requires that voucher specimens upon which a paper is based be deposited in a museum where they may eventually be reidentified. 21. Submit each manuscript in triplicate. The second and third copies can be reproduc- tions. REPRINTS AND PAGE COSTS 22. When 100 or more reprints are or- dered, an author receives 25 additional сор- ies free. Reprints must be ordered at the time proof is returned to the Editorial Office. Later orders cannot be considered. For each au- thors' change in page proof, the cost is U.S. $3.00 or more. 23. When an article is 10 or more printed pages long, MALACOLOGIA requests that an author pay part of the publication costs. SUBSCRIPTION COSTS 24. For Vol. 27, personal subscriptions are U.S. $17.00 and institutional subscriptions are U.S. $27.00. For information on Vol. 28, address inquiries to the Subscription Office. VOL. 27, NO. 1 MALACOLOGIA CONTENTS AMERICAN MALACOLOGICAL UNION SYMPOSIUM PROCEEDINGS MOLLUSCAN EXTINCTIONS IN THE GEOLOGIC PAST AND AT THE PRESENT TIME Organized by Geerat J. Vermeij 9 August 1983. Seattle, Washington, U.S.A. С. J. VERMEIJ Molluscan extinction: introduction to а symposium ..................... P. WARD Cretaceous ammonite; shell shapes ds ES MES IN С. J. VERMEIJ 8 E. J. PETUCH Differential extinction in tropical American molluscs: endemism, architecture, and the ‘Panama land bridge... ey hota ee A OR SRE D. JABLONSKI & K. W. FLESSA The taxonomic structure of shallow-water marine faunas: implications for Bihharierozoic.eXtinelions. Hur kre Oo eg A clears ен RS ER hep M. G. HADFIELD Extinction in Hawaiian achatinelline snails ...........................- хжжжжх Е. J. GARCÍA, J. С. GARCÍA & J. L. CERVERA Estudio morfológico de las espículas de Doriopsilla areolata (Gastropoda: Nudißranehi) a... 2 ee en Rett peters аа MERE IRIS M. S. JOHNSON, J. MURRAY & B. CLARKE High genetic similarities and low heterozygosities in land snails of the genus Samoana from the: Society Islands ......1 1. COLE RE CR RER P. W. KAT Hybridization in a unionid faunal suture zone ......................... В. HERSHLER & G. LONGLEY ; Phreatic hydrobiids (Gastropoda: Prosobranchia) from the Edwards (Bal- cones Fault Zone) Aquifer region, south-central Texas................. T. C. CHENG & E. J. PEARSON Modification and evaluation of Burch and Cuadros’s medium for the mainte- nance of the testes of a marine gastropod............................ Н. A. JONES, В. D. SIMPSON & С. L. HUMPHREY The reproductive cycles and glochidia of fresh-water mussels (Bivalvia: Hyriidae) of the Macleay River, northern New South Wales, Australia ... 1986 29 43 67 83 97 107 127 173 185 М 0 llusks MOL. 27; NO. 2 MALACOLOGIA International Journal of Malacology Revista Internacional de Malacologia Journal International de Malacologie Международный Журнал Малакологии Internationale Malakologische Zeitschrift MALACOLOGIA Editors-in-Chief: GEORGE M. DAVIS Editorial and Subscription Offices: Department of Malacology The Academy of Natural Sciences of Philadelania Nineteenth Street and the Parkway Philadelphia, Pennsylvania 19103, U.S.A. Associate Editors: JOHN B. BURCH University of Michigan, Ann Arbor ANNE GISMANN Maadi, A. R. Egypt MALACOLOGIA is published by the INSTITUTE OF MALACOLOGY, the Sponsor M of which (also serving as editors) are: KENNETH J. BOSS, President-Elect Museum of Comparative Zoology Cambridge, Massachusetts JOHN B. BURCH MELBOURNE R. CARRIKER, President University of Delaware, Lewes _ GEORGE М. DAVIS Secretary and Treasurer PETER JUNG, Participating Member Naturhistorisches Museum, Basel, Switzerland OLIVER E. PAGET, Participating Member Naturhistorisches Museum, Wien, Austria ROBERT ROBERTSON | CLYDE E ROPER Environmental Protection Agency — Germantown, Maryland Copyright © 1986 by the Institute of Malacology —— ` ROBERT ROBER" Editorial Assistants; По JEAN M. CRABTREE 47. T3 MARY DUNN AR Assistant Managing Editor: Е CARYL HESTERMAN es Smithsonian Institution Washington, D.C. Syracuse University, New York NORMAN F. SOHL United States Geological Survey | Washington, D.C. SHI-KUEI WU University of Colorado Mec Bou J FRANCIS ALLEN, Emerita Washington, D.C. ELMER G. BERRY, Emeritus 1986 EDITORIAL BOARD J. A. ALLEN Marine Biological Station Millport, United Kingdom E. E. BINDER Muséum d'Histoire Naturelle Geneve, Switzerland А. J. САМ University of Liverpool United Kingdom P. CALOW University of Glasgow United Kingdom A. H. CLARKE, Jr. Portland, Texas, U.S.A. B. C. CLARKE University of Nottingham United Kingdom R. DILLON College of Charleston SC, U.S.A. C. J. DUNCAN University of Liverpool United Kingdom E. FISCHER-PIETTE Muséum National d'Histoire Naturelle Paris, France У ЕВЕТТЕВ University of Reading United Kingdom E. GITTENBERGER Rijksmuseum van Natuurlijke Historie Leiden, Netherlands Е. GIUSTI Universita di Siena, Italy A. N. GOLIKOV Zoological Institute Leningrad, U.S.S.R. S. J. GOULD Harvard University Cambridge, Mass., U.S.A. A. V. GROSSU Universitatea Bucuresti Romania T. HABE Tokai University Shimizu, Japan A. D. HARRISON University of Waterloo Ontario, Canada J. A. HENDRICKSON, Jr. Academy of Natural Sciences Philadelphia, PA, U.S.A. K. E. HOAGLAND Lehigh University Bethlehem, PA, U.S.A. B. HUBENDICK Naturhistoriska Museet Göteborg, Sweden S. HUNT University of Lancaster United Kingdom R. JANSSEN Forschungsinstitut Senckenberg, Frankfurt am Main, Germany (Federal Re- public) R. N. KILBURN Natal Museum Pietermaritzburg, South Africa M. A. KLAPPENBACH Museo Nacional de Historia Natural Montevideo, Uruguay J. KNUDSEN Zoologisk Institut & Museum Kobenhavn, Denmark A. J. KOHN University of Washington Seattle, U.S.A. Y. KONDO Bernice P. Bishop Museum Honolulu, Hawaii, U.S.A. J. LEVER Amsterdam, Netherlands A. LUCAS Faculté des Sciences Brest, France С. MEIER-BROOK Tropenmedizinisches Institut Túbingen, Germany (Federal Republic) H. K. MIENIS Hebrew University of Jerusalem Israel J. E. MORTON The University Auckland, New Zealand J. J. MURRAY, Jr. University of Virginia Charlottesville, U.S.A. R. NATARAJAN Marine Biological Station Porto Novo, India J. ОКЕАМО University of Oslo Norway Т. OKUTANI University of Fisheries Tokyo, Japan М. Е. PARAENSE Instituto Oswaldo Cruz, Rio de Janeiro Brazil J. J. PARODIZ Carnegie Museum Pittsburgh, U.S.A. W. F. PONDER Australian Museum Sydney A. W. B. POWELL Auckland Institute & Museum New Zealand В. О. PURCHON Chelsea College of Science & Technology London, United Kingdom QIPZMNE Academia Sinica Qingdao, People's Republic of China N. W. RUNHAM University College of North Wales Bangor, United Kingdom S. G. SEGERSTRÁLE Institute of Marine Research Helsinki, Finland G. A. SOLEM Field Museum of Natural History Chicago, U.S.A. F. STARMÜHLNER Zoologisches Institut der Universität Wien, Austria У. |. STAROBOGATOV Zoological Institute Leningrad, U.S.S.R. W. STREIFF Université de Caen France J. STUARDO Universidad de Chile Valparaiso T. E. THOMPSON University of Bristol United Kingdom S. TILLIER Muséum National d'Histoire Naturelle Paris, France FSTOFFROLEMIO Societa Italiana di Malacologia Milano R. D. TURNER Harvard University Cambridge, Mass., U.S.A W. S. S. VAN BENTHEM JUTTING Domburg, Netherlands J. A. VAN EEDEN Potchefstroom University South Africa М. Н. VERDONK Rijksuniversiteit Utrecht, Netherlands B. R. WILSON Nedlands, Western Australia H. ZEISSLER Leipzig, Germany (Democratic Republic) A. ZILCH Natur-museum und Forschungs- Institut Senckenberg Frankfurt-am-Main, Germany (Federal Republic) MALACOLOGIA, 1986, 27(2): 203-241 NEW CALEDONIAN CHAROPID LAND SNAILS. I. REVISION OF THE GENUS PARARHYTIDA (GASTROPODA: CHAROPIDAE) Peter Mordan' & Simon Tillier? ABSTRACT Six species of the charopid genus Pararhytida, three of them previously undescribed, are recognised in a taxonomic revision based on material from 72 sites. Pararhytida is endemic to New Caledonia, being found in most areas of primary forest on the mainland, as well as the Belep Islands. lt appears to be absent from the Loyalty Islands and earlier records from the Isle of Pines are not confirmed. Whereas the largest species, P. dictyodes, occurs throughout the mainland, the remaining species are more restricted in distribution. The occurrence of spermatophores in the Charopidae 1$ recorded for the first time. Key words: Charopidae; Pararhytida; taxonomy; New Caledonia. INTRODUCTION The endemic New Caledonian charopid genus Pararhytida, previously revised by Franc (1956) and Solem (1961) on a purely conchological basis, 15 remarkable in several respects: 1. One species reaches 37 mm in shell diam- eter, exceeding the size of any other known endodontoid. 2. Part of the dorsal surface of the tail is thickened to form a pseudo- operculum, a structure analogous to the operculum of prosobranchs and some lower pulmonates, and known elsewhere only in the related New Caledonian genus Rhytidopsis Ancey. 3. Sperm is exchanged in a horny spermatophore, which in some species is strikingly similar in morphology to that of helicarionid snails. Although previously unrecorded in endodontoids, the occurence of a horny spermatophore is common in New Caledonian charopids, but only in Pararhytida is it formed of a fusiform body prolonged as a thin, denticulate tail. Six species of Pararhytida are recognised in the present paper, three of them being described as new: Pararhytida dictyodes (Pfeiffer), the type species; P. mouensis (Crosse); Р. marteli (Dautzenberg); Р. phacoides n.sp., P. pyrosticta n.sp. and P. thyrophora n.sp. One of the four species recognised by both Franc (1956) and Solem (1961), P. dictyonina (Euthyme), is synony- mised with P. mouensis. Species belonging to Micromphalia Ancey, 1882, and Plesiopsis Ancey, 1888, treated by both Franc and Solem as subgenera of Pararhytida, are ex- cluded from the genus. Tropidotropis gudei Preston, 1907, considered by Solem (1961) to be a synonym of T. trichocoma (Crosse), is a juvenile Pararhytida mouensis. This study is based on material collected at 72 sites in New Caledonia (Table 1; Fig. 1). Except where otherwise stated, specimens are from the Muséum national d’Histoire naturelle, Paris (MNHN), but some are also from the Field Museum of Natural History, Chicago (FMNH). All the relevant type mate- rial is housed in either the MNHN or the British Museum (Natural History), London (BMNH) and has been examined. DISTRIBUTION AND ECOLOGY Pararhytida is found in almost all the main- land areas of New Caledonia where primary forest remains, as well as the Belep Islands, but is apparently absent from the Loyalty Is- lands. We do not record Pararhytida from the Isle of Pines, although both Crosse (1894) and Franc (1956) mention P. dictyodes from there; this may be due to insufficient collecting by us. ‘Department of Zoology, British Museum (Natural History), London SW7 5BD, England. ?Laboratoire de Biologie des Invertebres marins et Malacologie, Museum national d'Histoire naturelle, 55 rue Buffon, F-75005 Paris, France. (203) 204 MORDAN & TILLIER TABLE 1. List of sampling stations. 5. Le Cresson, 164° 18' 36” E; 20° 29° 00” $. 100 m, dry forest on calcareous outcrop. Rainfall 1200 mm. A. & S. Tillier coll, 30.vi.1979. P. dictyodes: 3a + 23s. P. Mordan, A. & $. Tillier coll., 29.1.1981. P. dictyodes: 1$ + 1 juv. a + 1 juv. $. Probably idem. FMNH 159259, L. Price coll. P. dictyodes: 17a + 1$. 6. Grottes de Koumac, 164” 20' 27" E; 20° 31' 52" $. 80 т, dry forest on calcareous outcrop. Rainfall 1200 mm. P. Bouchet coll., 14-15.vi.1978. P. dictyodes: 11$ + 1 juv. a + 5 juv. or broken. A. & 5. Tillier coll. 30.vi.1979. P. dictyodes: 3s + 3 juv. s + 2 juv. a. P. Mordan, A. 4 $. Tillier coll., 29.1.1981. P. dictyodes: 6s + 2 juv. $. 7. Mandjelia, 14° 30’ 06” E; 20° 22’ 29" $. 400 т, 5 km М of sawmill, rainforest. Rainfall 1900 mm. A 8 S. Tillier coll., 2.vii.1979. P. dictyodes: 2a. 9. Ruisseau de l'Etoile du Nord (Oue Paoulou), 164° 20' 27” E; 20° 34’ 48” S. 150 m, dry forest, probably on a calcareous outcrop. Rainfall 1200 тт. A & S. Tillier coll., 30.vi.1979. P. dictyodes: 2s. 12. Mt. Taom, 164° 34’ 45” E; 20° 46’ 55” S. 900 m, altitude rainforest in a thalweg, on ultrabasic rock. Rainfall 2500mm. A. & S. Tillier coll. 30.vi.1979. P. dictyodes: 5s + 1 juv. $. + 1 juv a. (SEM). 14. Momies de la Fatenaoué, S side Mt. Tende, 164° 43’ 22" E; 20° 52’ 36” $. 100-200 m, dry forest. Rainfall 1250 mm. A. & S. Tillier coll. 4 vii.1979. P. dictyodes: 2s. 16. Plateau de Tango, track to Bobeitio, 164° 00’ 27" E; 20° 58’ 29” $. 300-350 m, rainforest. Rainfall 1800 тт. Р. Bouchet coll. 24.xii.1978. P. dictyodes: 1s + 1 juv. $ + 2 broken + 1 juv. a. 18. Сори, 165° 16° 30’ E; 21° 13’ 19’ S. 50-150 т, rainforest. P. Bouchet coll. 6.v.1979. P. dictyodes: 5s. 19. Forêt Plate, 165° 06’ 23” E; 21° 08’ 57” S. 540 m, NE slope Mt. Paéoua, rainforest. Rainfall 1841 mm. P. Bouchet, A. & 5. Tillier coll. 15.vii.1979. P. dictyodes: 1s + 1 juv.s + 1 broken. 20. Mt. Paéoua, 165° 05’ 27" E; 21° 10’ 48” $. 950-1000 m, altitude rainforest. Rainfall 3000 mm. A. & $. Tillier coll. 5.vii.1979. P. dictyodes: 2a + 3 juv. а + 3 juv. $ + 1 broken. 25. Adio, vallée seche, 165° 14’ 46” E; 21° 14’ 44” S. 180 m, dry forest. P. Bouchet coll. 6.v.1979. P. dictyodes: 6$ + 2 juv. + 1 broken. L. Price coll. 7. xi.1967. FMNH 159309. P. dictyodes: 9a + 1 juv. a. 36. Mt. Vulcain, Gallieni mine, 166° 20’ 55” E; 21° 54' 33” S. 700-900 m, maquis. Rainfall 3500 mm. Р. Bouchet coll. 5.xii.1978. P. dictyodes: 1 juv. $. 37. Mt. Dzumac, 166° 27’ 19” E; 22° 02' 30" 5. 950-1000 т, NW of summit, rainforest. Раша! 3000 mm. P. Bouchet 4 $. Тег coll. 4.vi.1979. P. mouensis: 1a. P. dictyodes: 1 juv. $. 43. Rivière Bleue, 166° 39’ 25” E; 22° 05' 47" $. 160 m, right side of the river, rainforest. Rainfall 2700 mm. P. Bouchet coll. 6.1.1979. P. mouensis: 1a. P. Mordan & S. Tillier coll. 29.1.1981. P. mouensis: 3a + 3s. 47. Mt. Guemba, 166° 56’ 10" E; 22° 10' 22” 5. 450 m, rainforest. Rainfall 3200 mm. P. Bouchet coll. 10.vi.1978. P. mouensis: 1a. P. marteli: 1a. 48. Touaourou (St.-Gabriel), 166° 58’ 00” E; 22° 12' 00” S. 10-30 m, rainforest on uplifted coral reef. Rainfall 3000 mm. A. Waren coll. 30.vii.1979 (road to Ni mine, 200 m from main road). P. marteli: 1a + 7s + 6s juv. or broken; P. Bouchet coll. 29.v.1978, 8.xii.1978 and 19.vii.1979: P. marteli: 4a + 3 a juv. + 9s + 6s juv. or broken. 49. Right side of Kuebeni River, 167° 00' 07” E; 22” 16' 23” S. 50-80 m, rainforest on ultrabasic rocks. Rainfall 2500 тт. Р. Bouchet coll. 12.1.1979. P. marteli: За + 7 juv. a + 7s. 50. Goro, 167” 00’ 21” E; 22° 18' 57" S. 30 m, rainforest on ultrabasic rock. Rainfall 1900 mm. P. Bouchet coll. 3.ix.1978. P. marteli: 1a. 57. lle Pott (Belep Islands), S plateau, 163° 16' 00" E; 19° 35’ 27” $. 100-150 m, maquis. Rainfall 1250 mm. P. Bouchet & С. Cherel coll. 27.viii.1978. P. thyrophora: 3s. 58. lle Art (Веер Islands), М plateau, 163° 24' Е; 19” 42' S. 200-250 m, maquis. Rainfall 1250 mm. Р. Bouchet & A. Waren coll. 20.viii.1978. P. thyrophora: 13a + 50$ + 48 juv. (29a) + 1 broken. 65. Mt. Nindo (near Poum), 164” 10' 41” E; 20° 17' 47” $. 70 т, gallery forest. Rainfall 1250 mm. P. Bouchet coll. 20.viii.1978. P. dictyodes: 3a + 2 juv. a. 66. Col d'Amos, 164° 25’ 20” E; 20° 18’ 52” 5. 200 m, rainforest. Rainfall 1600 mm. P. Mordan, A. & S. Tillier coll. 31.1.1981. Р. dictyodes: 17a + 6s + 6s juv. or broken. 69. Mandjelia, 164° 30’ 06” E; 20° 22’ 29” $. 550m, below the sawmill, rainforest. Rainfall 1800 mm. A. & 5. Tillier coll. 30.vi.1979. P. dictyodes: 1$. 70. Station de Djavel, 164° 23’ 36” E; 20° 24’ 33” S. 50m, secondary dry vegetation. Rainfall 2000 mm. A. & S. Tillier coll. 30.vi.1978. P. dictyodes: 4s. 71. Nehoue valley, 164 °16' 00” E; 20° 26’ 18" 5. 50 m, dry forest. Rainfall 1300 mm. P. Mordan, A. 8 $. Tillier coll. 1.1.1981. P. dictyodes: 2a + 11s + 3 juv. (1a). 72. Mt. Ignambi, 164° 36’ E; 20° 27’ $. 850-950 m, rainforest. Rainfall 3000 mm. P. Bou- chet coll. 27.xii.1978. P. pyrosticta: 1a + 1 juv. a. 74. Colnett, 164° 44’ 24” E; 20° 29’ 34” $. 20 т, track to the cascade. Rainfall 4000 mm. P. Mordan & S. Tillier coll. ii.1981. P. dictyodes: 1 juv. $. 79. E. slope of Mt. Panié, 164° 48’ 43” E; 20° 35’ 27" 5. 280 т, rainforest. Rainfall 5100 mm. P. Bouchet & С. Cherel coll. 14.viii.1978. P. dictyodes: la +A uva + 15 + адм 5. 80. E. slope of Mt. Panié, 164° 48' 22” E; 20° 35' 54" 5. 580 m, rainforest. Rainfall 5700 mm. P. Bouchet & С. Cherel coll. 14.viii.1978. P. dictyodes: 1a + 1 juv. a + 1 juv. $. L. Price coll. 500-700 m, 4. xi.1967. FMNH 159344. P. dictyodes: 2a. 900-970 m, rainforest. Rainfall 6400 mm. P. Bouchet & C. Cherel coll. 14.vili.1978. P. dictyodes: 2a. 83. Thiem, 165° 06' 23" E; 20° 45’ 43” S. 10-50 m, rainforest. Rainfall 2500 mm. P. Bouchet TAXONOMY OF PARARHYTIDA 205 coll. 25.xii.1978. P. dictyodes: 2a + 12s + 6 juv. a. 84. 5. slope of Mt. Tchingou, 165” 00' 00” E; 20° 54’ 27" S. 900-1000 m, rainforest. Rainfall 3000 mm. P. Bouchet, A. 4 $. Tillier coll. vii.1979. P. pyrosticta: 1a + 1s. 1250 m, rainforest. Rainfall 3500 mm. P. Bouchet, A. & S. Tillier coll. vii. 1979. Р. pyrosticta: ба + 2s + 6 juv. a + 5 juv. $. 86. N side of Amoa River, 10 km up the valley, 165° 12’ 12” E; 20° 58’ 00” S. 20 m, rainforest. Rainfall 2500 mm. P. Bouchet, A. & S. Tillier coll. 12.vii.1978. P. dictyodes: 1a. P. Mordan, A. 4 S. Tillier coll. 18.1.1981. P. dictyodes: 1a. 88. S side Mt. Koniambo, 164° 49’ 25” E; 21° 02’ 00” $. 600m, maquis. Rainfall 1500 mm. P. Mordan, A. & S. Tillier coll. 28.1.1981. P. dictyodes: iS: 89. N side of Tchamba River, 5 km up the valley, 165° 19’ 52” Е; 21° 01’ 51" $. 100 m, forest. Rainfall 2500 mm. P. Mordan, A. 4 5. Tillier coll. 17.1.1981. Р. dictyodes: 2$. 91. Mt. Aoupinié, 165° 18’ 00” Е; 21° 10’ 09” $. 600 m, track above the sawmill, above Goa tribe, rainforest. Ваша! 2700 mm. P. Mordan, A. 4 5. Tillier coll. 16.1.1981. P. dictyodes: 5a + 1 juv. a + 1s + 1 juv. $. P. pyrosticta: 1a. 92. Mt. Aoupinié, 165° 15’ 42” E; 21° 10' 42” $. 1000 m, summit area, altitude rainforest. Rainfall 3500 mm. Р. Mordan, A. & 5. Tillier coll. 16.1.1981. Р. pyrosticta: 1 juv. a + 2 juv. $. 94. Moneo, 165° 29’ 31” E; 21° 09’ 36” S. 10-50 m, rainforest. Rainfall 2500 mm. P. Bouchet coll. 15.v.1978. P. dictyodes: 6$ + 5 juv. $. 97. Mt. Boulinda, 165° 08' 57” E; 21° 14' 44” $. 980-1020 m, between Petit and Grand Boulinda, altitude rainforest. Rainfall 3000 mm. A. & S. Tillier coll. 6.vii.1979. P. dictyodes: 1s + 1 juv. s. P. phacoides: 3a + 1s (broken). 98. Adio caves, 165° 14' 18” E; 21° 15’ 36” 5. 180 m, forest on a calcareous outcrop. Rainfall 1400 mm. P. Bouchet & С. Cherel coll. 20.viii.1978. P. dictyodes: 1a + 11$ + 1 juv. $. 110. $ slope of Mt. Table Unio, 165° 45’ 55” E; 21° 33’ 36” $. 850-950 m, rainforest. Rainfall 2400 mm. 5. Tillier coll. 7.v.1979. P. dictyodes: 4a aE USS 114. Mt. Rembai, 165° 50’ 13” E; 21° 34’ 54” S. 800-850 m, N crest, rainforest. Rainfall 2400 mm. S. Tillier coll. 8.vi.1979. P. dictyodes:1a + 2s + 1 juv. a + 1 juv. $. 115. Mt. Сапа, 165° 55’ 48” E; 21° 35’ 00” $. 900-1050 m, rainforest. Rainfall 2800 тт. P. Bouchet coll. 21.1.1979. P. dictyodes 2a. 116. N side Col d'Amieu, 165° 48’ 08” E; 21° 36’ 00” S. 400-500 m, W of the Maison Forestiere, rainforest. Rainfall 1800 mm. Р. Bouchet coll. 18.xi.1978; S. Tillier coll. 7.v.1979. P. dictyodes: 2a E SI UVAS: 117. Plateau de Dogny, 165° 52’ 33" E; 21° 36' 26” $. 950m, rainforest. Rainfall 2600 mm. P. Bouchet coll. 1.1.1979. P. dictyodes: 1s broken + 1 juv. $. 118. Mt. Nakada, 166° 02’ 26” E; 21° 38’ 37” 5. 850 m, rainforest. Rainfall 2600 mm. $. Tillier coll. 19.vi.1979. P. dictyodes: ба + 1s + 2 juv. a. P. phacoides: 1$ + 1 juv. a + 1 juv. s. 119. Mt. Nakada, 166° 03’ 08” E; 21° 38' 57" 5. 500 m, above the sawmill, rainforest. Rainfall 2000 mm. S. Tillier coll. vi.1979. P. dictyodes: 2a. 123 Mts Dor 165259 tdi! E2452 SOES: 950 т, rainforest. Rainfall 2600 mm. P. Bouchet coll. 16.iv.1979. P. dictyodes: 1a + 2s + 1 juv. $. 125. Mt. Humboldt, 166° 23’ 29" E; 21° 53’ 28” 5. 1150 т, crest leading to Mt. Vulcain, rainforest. Rainfall 4500 mm. $. Tillier coll. 22.1.1981. P. mouensis: 1 juv. a. 128. Col de la Ouinné, between Mt. Dzumac and Mt. Ouin, 166° 27’ 54” E; 22° 01' 18” $. 850 т, rainforest. Rainfall 3000 mm. P. Mordan, A. 4 5. Tillier coll. 25.1.1981. P. mouensis: 1a. 130=Mt. Mou 166? 20’ 34” Е; 227037 5528: 1200 m, altitude rainforest. Rainfall 3400 mm. P. Bouchet & С. Cherel coll. 9.viii.1978. P. mouensis: 1 juv. s. 131 Mt Mou 166? 19’ 467 Е; 2222047 2875: 370-450 m, E of sanatorium, rainforest. Rainfall 1800 mm. P. Bouchet & С. Cherel coll. 5.viii.1978. P. dictyodes: 1s + 1 juv. a. P. Mordan, A. & 5. Tillier coll. 23.1.1981. Р. dictyodes: 3$ + 2 broken. A. 8 В. Solem coll. 23.1.1962. FMNH 135440. P. dictyodes: 3s. 136. Montagne des Sources, 166° 35’ 56” E; 22° 07' 32” S. 875m, W slope, rainforest. Rainfall 3500 mm. P. Bouchet, S. Tillier & A. Warén coll. 3.v.1979. P. mouensis: 1a + 1$ + 2s broken. 140. Mt. Koghi, 166° 30’ 21” E; 22° 10' 35” 5. 480-520 m, rainforest. Rainfall 2000 mm. P. Mordan, A. & $. Tillier coll. 10.1.1981. P. mouensis: la. 142. Col de Mouirange, 166° 39’ 00” E; 22° 12’ 00” $. 180-250 m, rainforest. Rainfall 1800 mm. P. Mordan, A. & S. Tillier coll. 11.1.1981. P. dictyodes: 1$ broken + 1 juv. $. P. mouensis: За + 3$ + 1 juv. a. 143. Col. de Mouirange, 166° 40' 14” E; 22° 13' 19” S. 200 m, rainforest. Rainfall 1800 mm. 5. Tillier coll. 5.vi.1979. P. dictyodes: 2s broken. 1146: Гас еп! \, 166? 55” 42” Е: 222157 365: 250 m, maquis. Rainfall 3200 mm. P. Bouchet & 5. Тег coll. 25.vi.1979. P. dictyodes: 1 juv. a. 150. Prony, Вае Est, 166° 54' 11” E; 22° 22' 23" S. 150 m, rainforest. Rainfall 2700 mm. P. Bouchet coll. 10.vi.1978. P. mouensis: 1s. 179. Forét Nord, 166° 53’ 00” E; 22° 17' 00” $. 220-250 m, rainforest. Rainfall 3000 mm. P. Mordan, A. 4 $. Tillier coll. 24.1.1981. P. mouensis: 2a + 3s + 1 broken. 181. Mt. Ningua, 166° 09’ 25” E; 21° 45' 17” $. 950 m, rainforest. Rainfall 2800 mm. $. Tillier coll. vi. 1979. Р. dictyodes: 1 juv. a + 1$ + 1 juv. 5. 700 m, rainforest. P. Lespes coll. 10.viti.1979. P. dictyodes: 1s (broken). 183.6km E of Ouegoa, 140m. Rainfall 1600 mm. L. Price coll. 1.xi.1967. FMNH 159246 8 159228. P. dictyodes: 8a + 2s + 4 juv. a. 184. Bac de la Ouaieme, 0 m, rainforest. Rainfall 3000 mm. P. Bouchet & С. Cherel coll. 12.vii.1978. Р. dictyodes: 1a + 4s + 2 juv. $. 185. Near Thiem, 100 m, rainforest. Rainfall 2800 mm. L. Price coll. 15.x.1967. FMNH 159219. P. dictyodes: 1a. 206 MORDAN & TILLIER TABLE 1 (Continued) 185. Near Thiem, 100 m, rainforest. Rainfall 2800 mm. L. Price coll. 15.x.1967. FMNH 159219. P. dictyodes: 1a. 186.N side of Tiwaka River, near Ouagap, 100 m, rainforest. Rainfall 3200 mm. L. Price coll. 11.x.1967. FMNH 159247. P. dictyodes: 3a + 1 juv. a. 187. S side of Amoa River, 4 km up the valley, 20 m, rainforest. Rainfall 2800 mm. A. Waren coll. 8.viii.1978. P. dictyodes: 3$ + 2 Juv. $. 188. N side of Tiwaka River, 13 km up the valley, 100 m, forest. Rainfall 2500 mm. P. Mordan, A. & S. Tillier coll. 17.1.1981. P. dictyodes: 7s + 1 чу. $ + 2 broken. 189. E side Col des Roussettes, 400 m, rainfor- est. Rainfall 1700 mm. L. Price coll. FMNH 159325. P. dictyodes: 2a. 190. Col d'Amieu, 530 m, rainforest. Rainfall 1800 mm. L. Price coll. 1967. P. dictyodes: 10a + 2 juv. a. 191. Dothio-Nakey, 400 m, rainforest. Rainfall 2000 mm. L. Price coll. 21.x.1967. FMNH 159341. P. dictyodes: 2a. 192. Near Mt. Ouénarou, 200 m, rainforest. Rain- fall 2300 mm. L. Price coll. 15.xi.1967. FMNH 159239. P. dictyodes: 3a. P. mouensis: 3a. 193. Forét Cachée, 250 m, valley of the Creek Pernod, rainforest. Ваша! 2700 mm. McKee coll. 29.vi.1978. P. dictyodes: 1s. 194. Faux Bon Secours, 300 m, rainforest. Rain- fall 3000 mm. McKee coll. 17.1.1980. Р. dictyodes: 1$. 195. 8 Кт S of Yaté, 30 т, rainforest. Rainfall 3000 mm. L. Price coll. 19.xi.1967. ЕММН 159238 P. marteli: 10а + 7 juv. a. Pararhytida was collected from forest with rainfall ranging from 1200 mm to more than 6000 mm a year. lt is absent from very dry environments, and also from high-altitude rainforest (“foréts a mousses”). It appears that with the exception of P. dictyodes, which is found throughout the entire rainfall range of the genus, each species is restricted to a part of this range. Living Pararhytida is always found at ground level, resting in the leaf litter. Dead and rotting palm sheaths are a particularly favoured resting site; the snails are never found associated with logs, a site occupied by a number of other New Caledonian charopid genera. ANATOMY AND MORPHOLOGY Shell The shell of Pararhytida is very large for a charopid, ranging from 13.5 mm to 37 mm in diameter. No other endodontoid is known to reach such a large size (Solem, 1961; 1976; 1983). The shell is rather flat (H/D ranging from 0.44 to 0.65, with a mean of 0.56), and carinated. The umbilicus is open and rather small, from 0.055 to 0.15 (mean 0.088) the shell diameter. The adult shells have from 5.5 whorls in small species, to 6.9 whorls in the largest (P. dictyodes). The aperture of adult shells is slightly ex- panded, but not deflected as in many endodontoids. As the shell takes on the adult characters (in its last 0.2 of a whorl), the position of maximum apertural height is dis- placed outwards from the columellar extrem- ity of its basal border to the middle of the latter. Shell sculpture consists of oblique radial ridges, which are too dense and faint to be accurately counted. The apical whorls of large P. dictyodes have only coarse radial ridges, whereas the apical whorls of smaller species additionally show traces of very faint spiral sculpturing (Fig. 2A). In New Caledonian charopids the distinction between groups having radial apical sculpture and those hav- ing spiral apical sculpture is much less well defined than was stated by Solem (1961). The colour pattern consists primarily of reddish-brown flammules radiating outwards from the suture, on a light beige background. The flammules are generally well defined only near the suture; further away they are inter- rupted by pale zones and spotted by almost white, oval specks. Young shells of P. mouensis and P. pyrosticta are known to have periostracal hairs along the carina (Fig. 2B) which, as shown by Preston's original description of Tropidotropis gudei, resemble those of Tropidotropis. Foot and pseudo-operculum The foot of Pararhytida is aulacopod. Its most striking feature is the presence of a pseudo-operculum on the tail (Solem, Tillier & Mordan, 1984), formed from a dorsal epider- mal thickening. lt occupies the same position as the prosobranch operculum, and com- pletely occludes the shell aperture when the foot is retracted. It is not horny as in prosobranchs. Surprisingly, neither Fischer (1875) nor Starmühlner (1970), both of whom dissected P. dictyodes, mentioned this fea- ture. A similar structure is found in the New Caledonian charopid genus Rhytidopsis, but TAXONOMY OF PARARHYTIDA 207 79-80 b 48 195 49 50 130-131 136 179 12 {> 142-143 193 146 < 194 150 FIG. 1. Мар of collecting stations (listed in Table 1). FIG. 2. Shell of juvenile P. pyrosticta п. sp., Tchingou, sta. 84. A. Surface sculpture, scale 300 ym. B. Carinal hairs, scale 100 um. 208 MORDAN & TILLIER TABLE 2. Number and size of radular teeth in Pararhytida. Number of teeth Per side No. E Species specs. Perrow Marg. Lat. dictyodes 11 79-93 23-30 14-19 (1 juv.) mouensis 4 65—75 20-26 11-14 marteli 1 67 21 12 phacoides 2 69 24 10 (1 juv.) pyrosticta 2 67—71 22-23 11-12 thyrophora 1 75 22 15 is known from no other stylommatophoran. The supposed function of the pseudo- operculum in efficiently blocking the shell aperture implies a novel pattern of foot retrac- tion in the Stylommatophora: instead of being proximal to the mantle border inside the shell cavity when retracted, the tail of Pararhytida remains distal to the border. Furthermore, its tip has to be retracted before its dorsal side, such that only the pseudo-operculum remains exposed in the fully retracted animal. Jaw and radula The jaw is thin, arcuate and smooth. It is not very rigid and 1$ easily dissolved by so- dium hypochlorite solution. Its size appears to vary roughly in proportion to that of the ani- mal. The overall pattern of radular anatomy in Pararhytida is relatively uniform, and con- forms well to the standard charopid pattern described by Solem (1983: 34). The central tooth 15 tricuspid, and smaller and narrower than the adjacent laterals. The lateral teeth are also tricuspid, but asymmetrical in that the endocone is often slightly higher than the ectocone, and typically rather narrower. The transition from laterals to marginals is grad- ual, occuring over two to three teeth, thus making precise counting of the numbers of laterals and marginals impossible. The mar- ginal teeth are tricuspid and have a highly characteristic shape: the mesocone is nor- mally broad and blunt; the endocone is of equal height to, or slightly shorter than the mesocone, and strongly falciform, pointing in towards the mesocone; the ectocone 1$ Width of tooth in mm Central Lateral Mesocone Whole tooth Mesocone Whole tooth 12.5-22.5 20.5-38.5 15.5-28.0 23-45 12-16.5 26-32.5 16.5-22 30-39 13 26.5 20.5 30 12.5 26 13 30.5 9.5-11 20-20.5 12-13.5 22.5-23 8.5 25 16 30 sharp, conical, and with its tip normally well below the top of the tooth. The anterior surfaces of the central and lateral teeth bear a characteristic group of shallow pits and grooves (Fig. 14) which presumably form part of the inter-row support mechanism (Solem, 1972), and articulate with the basal plate of the tooth in front. Table 2 lists the ranges of tooth number and size for the material examined. It is clear from the data for P. dictyodes, where a relatively large number of specimens have been examined, that there can be great intra-specific variation. Also, the table demonstrates that there 1$ considerable size overlap between the vari- ous species. There were no obvious differ- ences seen in the radulae of sympatric spe- cies pairs: P. mouensis and P. marteli from sta. 47 had remarkably similar radulae, in both tooth number and shape, and sympatric populations of P. dictyodes and P. mouensis (sta. 192) had normal radulae for their spe- cies. Indeed, the most extreme forms of P. dictyodes were found in allopatric situations. From the limited information available, there would thus appear to be no evidence for any form of character displacement in the radular morphology of Pararhytida. Visceral mass In Pararhytida the total length of the vis- ceral mass varies from 3.5 to 5.5 whorls, but intraspecific variation never exceeds one whorl. Its length is not directly proportional to the length of the coiled shell since the species with the most whorls (P. dictyodes) has the shortest visceral mass (Table 3). The length of the lung varies from 0.5 to 1.2 TAXONOMY OF PARARHYTIDA 209 TABLE 3. Length in whorls of various parts of the visceral mass. Visc. Sta. mass Lung Pararhytida dictyodes Y .6 3 075 20 4 0.66 25 3.3 0.75 25 3.35 0.75 79 4.5 0.75 80 5.25 0.8 83 4 0.75 86 3.25 0.6 91 3.25 0.55 110 4.4 0.75 115 3.75 0.75 118 4.25 0.6 123 4.2 0.7 184 3.6 0.75 185 3.5 0.75 186 3.45 0.8 189 3.75 0.5 190 3.75 0.75 191 ? 0.75 192 4 0.8 Pararhytida mouensis 43 162 47 572 0.95 128 4.5 0.95 136 4.3 1.15 141 led 179 4.2 0.9 192 4 1 Pararhytida marteli 47 5.4? 1.05 48 5.6 1ES 48 1.2 195 515? 0.9 Pararhytida phacoides 97 4.2 0.9 118 0.6(juv.) Pararhytida pyrosticta re 4.75 0.65 84 4.9? 0.85 84 SR 0.75 91 4? 0.75 Pararhytida thyrophora 58 5.4? 0.6 Top of Upper stomach Stomach spire 1 2.33 1 1.55 Ты 0.9 Was 1.3 1(0.25) 2.4 1.25 1.1(0.25) 3.2 1 2.25 0.75 0.66 1.9 ee 1 125 1515 1 225 1.2 1(0.25) 1.9 115 1?(0.25) 2.5 162 1?(0.25) 2.3 Us) 1:55 1:3 1.45 0.9 75 1 2.25 1 0.75 2 1 2 1 2(0.3) 2.2 152 1625 3 Weal 2:55 И 2.05 1.3 2 12 1.8 0.9 1 33 lea st 3.5? 162 zul 1.1 1(0.2) 3 14115 2.9 1815 1 3.2 1.25 2? 1.4 1.3(0.2) 3.4? whorls above the pallial border. Within a spe- cies, maximum variation is 0.4 whorls. The stomach and crop, which generally occupy an entire whorl, have their distal extremity nor- mally between 0.9 and 1.2 whorls above the lung, although exceptionally it may lie further down (0.75 whorls in some P. dictyodes), or up (1.4 whorls in P. thyrophora). The upper part of the visceral mass, which includes the di- gestive gland and hermaphrodite gland, varies in length between 1.5 and 3.5 whorls, and almost as much variation may be found within a single species (Table 3). Internally the digestive tract shows only faint oesophageal ridges, and no crop ridges or typhlosole. 210 MORDAN & TILLIER FIG. 3. Central nervous system of Pararhytida. A. P. dictyodes, Col d'Amos, sta. 66; B. P. marteli, S of Yaté, sta. 195. Scale line both 1 mm. CG, cerebral ganglion; PaG, parietal ganglion; PG, pedal ganglion; PIG, pleural ganglion; VG, visceral ganglion. Central nervous system The central nervous system has been de- scribed by Fischer (1875) and Starmühlner (1970), but with so little accuracy that no useful comparisons can be made. In particu- lar, Fischer's figure misled Bargmann (1930) to erroneous conclusions concerning the pat- tern of compaction of the visceral chain: the visceral ganglion is not, in fact, fused with the left parietal ganglion. Within Pararhytida, the arrangement of the central nervous system seems to relate primarily to size: 1. In the four smallest species the cerebral commissure and the lateral connectives are relatively long (Fig. 3B). 2. In the large P. dictyodes, the cerebral commissure 1$ short, and the lateral connectives shorter, particu- larly on the right side (Fig. ЗА); the left parietal ganglion is attached to the left pleural gan- glion, and the visceral, right parietal, and right pleural ganglia form a single mass in which individual ganglia are barely distinguishable. 3. In species of intermediate size (P. mouensis), the arrangement of the central nervous system is also intermediate. In all species the ganglia of the visceral chain are adpressed, and displaced to the right to such an extent in small species that the left parietal ganglion is close to the me- dian plane; in the large P. dictyodes it actually lies in the median plane. Additionally, in small species, the right parietal ganglion has moved to a position underneath the right pleural (Fig. ЗА). Pulmonary complex The kidney is U-shaped (Fig. 4), and varies from almost one-half to a little less than TAXONOMY OF PARARHYTIDA 211 FIG. 4. Pallial complex of Pararhytida dictyodes, Le Cresson, sta. 5. Scale line 5 mm. A, auricle; K, kidney; O, anus and ureter opening; PB, pallial border; PV, pulmonary vein; R, rectum; U, ureter; V, ventricle. one-third the length of the lung. The two arms of the kidney are subequal in P. phacoides and P. pyrosticta; the rectal arm is slightly longer than the cardiac arm in P. thyrophora and P. dictyodes, and is about 1.5 times the length of the cardiac arm in P. mouensis and P. marteli. Only the principal pulmonary vein can be clearly distinguished on the lung roof, which does not have any other obvious venation. The rectum and ureter open contiguously on the dorsal side of the pneumostome, and the opening is protected ventrally by a small lappet of the mantle border. 212 MORDAN & TILLIER FIG. 5. A, genital apparatus of Pararhytida dictyodes, showing position of spermatophore, Mt. Table Unio, sta. 110. B, spermatophore. Scale lines both 5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; P, penis; PA, penial appendix; PR, penial retractor; SO, spermoviduct; SP, spermatheca; Sp, spermatophore; T, talon; V, vagina; VD, vas deferens. TAXONOMY OF PARARHYTIDA 213 Genital apparatus Hermaphrodite and female portion (Fig. 5A) The hermaphrodite gland is composed of between three and seven lobes, and their number can vary considerably within a spe- cies (from three to six in P. dictyodes, and from four to seven in P. mouensis). The two extremities of the hermaphrodite duct are thin and almost straight, but the median portion is thickened and convoluted, forming a seminal vesicle (sensu Bayne, 1973). lts distal end opens into the stalk of a spherical talon, whose intraspecific variation in size may be as great as that within the entire genus (12 or slightly more). The size of the albumen gland 1$ also variable, apparently depending primarily on the degree of maturity. When fully developed, itoccupies most of the space between the top of the kidney and the crop, as in most charopids (Solem, 1983). The length of the free oviduct varies from one-half to about twice the spermoviduct length. It is usually contorted, and maintained in this state by connective tissue attached to a branch of the retractor muscles, which inserts in the angle between the oviduct and spermathecal stalk. Its wall is thick and internally bears coarse, longitudinal ribs. The spermatheca is always short, its top never reaching as far as the carrefour region. It is adpressed to the columellar side of the spermoviduct. The head varies from pear- shaped to elongate, and shape appears fairly constant within species. The spermathecal stalk is thick-walled, with longitudinal internal ribbing. Generally it contains a curious structure resembling a small, hard, hemi- spherical knob, housed in a horseshoe- shaped ridge (e.g. Fig. 19D). The position of this structure varies, even within species, from just above the oviducal opening to the top of the spermathecal stalk. Its function 1$ unknown. The vagina is long, and in all but one species contains a large, transverse ridge. P. mouensis has only a vaginal constriction which, from its position, may well be homolo- gous with the ridge (Fig. 28B). This species additionally possesses one or two vaginal pouches or appendices close to the constric- tion. In all species the atrium is very short. Penial complex (Fig. 5A) This comprises a short, fusiform penis with a subapical (possibly glandular) appendix, and a long, thin epiphallus. The penis retrac- tor muscle 15 inserted at the penis/epiphallus junction, and originates about one-fifth the way up the inner lung wall. The internal ornamentation of the penis (Fig. 19) consists basically of two main longitudinal pilasters: one interrupted by the opening of the penial appendix, and often bifurcated at its upper part; the second situated on the opposite side, and extending further up the penis. The upper part of the second pilaster may be prominent, forming a kind of verge, and may also be prolonged as a transverse ridge running between the opening of the epiphallus and the opening of the penial appendix. The space between these two pilasters is often occupied by a number of less-prominent secondary pilas- ters. This pattern may be altered in various ways: all pilasters may be equally developed, or so reduced that only an apical verge may be distinguished within the penis. In some cases the epiphallic pore is prominent, or surrounded by a circular ridge, reminiscent of the ring pilaster of typical charopids (Solem, 1983). A few millimeters above the pore, the wall of the epiphallus bears numerous ridges; further up only three ridges are normally distinguishable, two of which are larger than the third. The junction of the epiphallus and vas deferens does not show any particular morphological differentiation, being marked only by the termination of the epiphallic ridges and by a diminution in the diameter of the duct. Spermatophore (Fig. 5B) Although previously undescribed in charo- pids, the occurrence of a horny spermato- phore appears to be the rule rather than the exception in New Caledonian members of this family (Tillier, unpublished). In Pararhytida the spermatophore is always elongate- fusiform in shape and bears a longitudinal ridge. Complete spermatophores have only been recorded from within the spermatheca, when the spermatophore pore has been di- rected towards the oviduct. In some species it is prolonged by a long tail which runs down the spermathecal stalk and then up the free oviduct, reaching as far as the spermoviduct 214 MORDAN & TILLIER (Figs. 5A, 27D, 37A). When such a tail is developed, the spermatophore ridge extends to its extremity, where it becomes finely denticulated. The function of this tail is most probably to transport the sperm from the spermathecal head to the spermoviduct. TAXONOMIC CRITERIA This section considers the rationale upon which the taxonomic decisions taken in this paper were made. The basic criterion used to decide whether two forms belong to different species was the absence of intermediate forms, which we suppose reflects the ab- sence of gene flow. Two situations may arise: 1. The two forms are sympatric. In such a situation specific distinction has been easy as there is normally a large morphological gap between taxa. Such situations have allowed us to gauge the degree of interspecific char- acter difference expected within the group. 2. The two forms are allopatric. If the mor- phological gap is smaller than that between sympatric (more strictly syntopic) species, then we have considered the forms con- specific. When the gap is larger, as has happened frequently, one can either separate the two forms as distinct species, or integrate them into a coherent pattern of intraspecific geographic variation. An example will illustrate this last case. There are greater character differences be- tween some northern and southern P. dictyodes than between sympatric P. dicty- odes and P. phacoides. However, we have considered the former to be conspecific be- cause in this case not only does each char- acter vary along a geographical cline, but also the morphoclines themselves vary indepen- dently, and we have interpreted this as evi- dence of continuous gene flow. The alterna- tive situation would be one in which there is congruence between morphoclinal disconti- nuities, suggesting areas where gene flow 1$ absent or severely restricted. Where any major doubt has remained, our decision has been to lump rather than split. Particularly in the cases of P. dictyodes and P. mouensis, however, we cannot exclude the possibility that our taxonomy has been too conservative; had there been less material of either species more taxa would probably have been recognised. The result is that in Pararhytida no single character seems to allow species recognition throughout the genus: species are defined in terms of character combinations, in relation to geographical position. In all cases intrapopu- lation variation is much lower than that be- tween allopatric populations; that 15, variation is mainly geographic. Multivariate analysis of clinal variation in both conchological and anatomical charac- ters of Pararhytida will form the subject of a further paper. SYSTEMATIC REVIEW Genus Pararhytida Ancey, 1882 Type species: Helix dictyodes Pfeiffer, 1847 (by subsequent designation of Pilsbry, 1894: 52). Diagnosis Pararhytida differs from all known charopid genera, other than Ahytidopsis, by its pseudo-operculum and by the presence of a transverse ridge, sometimes developed into a foliated appendage, within the vagina. It dif- fers from Rhytidopsis by: 1. Its mode of life in the leaf litter (Rhytidopsis is arboreal); 2. The simplicity of the transverse vaginal ridge, which is considerably more complex in Rhytidopsis; 3. The shape of the penis and the long, thin epiphallus; and 4. Its large, flattened carinated shell, and weak shell sculpture. (The anatomy and biology of Rhytidopsis will be the subject of a further paper.) Micromphalia Ancey, 1882 and Plesiopsis Ancey, 1888 were considered by both Franc (1956) and Solem (1961) to be subgenera of Pararhytida. As they possess neither a pseudo-operculum nor any kind of peculiar vaginal structure, they are here removed from Pararhytida. Pararhytida dictyodes (Pfeiffer, 1847) (Figs. 4-20) Helix dictyodes Pfeiffer, 1847: 111 (New Guinea). Gassies, 1863: 241, pl. 1, fig. 19; Fischer, 1875: 273. Helix dictyoides [sic] Pfeiffer. Reeve, 1852: pl. 80, species 423. Trochomorpha dictyodes (Pfeiffer). Crosse, 1894: 241, pl. 8, fig. 3. TAXONOMY OF PARARHYTIDA 215 Pararhytida dictyodes (Pfeiffer). Dautzen- berg, 1923: 140. Pararhytida (Pararhytida) dictyodes (Pfeiffer). Franc, 1956: 136; Solem, 1961: 467; Starmühlner, 1970: 302, figs. 14-18. Lectotype (here designated): New Guinea (in error), Lieutenant Ince, Cuming Collection. BMNH Reg. no. 1981262 (Fig. 6). Dimen- sions (mm): Shell height 16.1. Shell diameter 27.1. Aperture height 10.6. Aperture diameter 13.9. Umbilicus width 2.8. Whorls 6.25. Paralectotypes: 2 specimens from above lot. BMNH Reg. no. 1981263. Dimensions (mm): Shell heights 14.5, 15.2. Shell diame- ters 26.2, 26.8. Aperture heights 10.0, 10.5. Aperture diameters 13.3, 13.8. Umbilicus widths 2.4, 2.5. Whorls 6.4, 6.1. Other material: Sta. 7(2), sta. 9(2), sta. 12(7), sta. 14(2), sta. 16(5), sta. 18(5), sta. ), sta. 20(9), sta. 25(19), sta. 36(1), sta. 1), sta. 65(5), sta. 66(29), sta. 69(1), sta. 4), sta. 71(16), sta. 74(1), sta. 79(5), sta. 7), sta. 83(20), sta. 86(2), sta. 88(1), sta. 2), 91(8), sta. 94(11), sta. 97(2), sta. 13), sta. 110(5), sta. 114(5), sta. 115(2), sta. 116(4), sta. 117(2), sta. 118(9), sta. 119(2), sta. 123(4), sta. 131(10), sta. 142(2), sta. 143(2), sta. 146(1), sta. 181(4), sta. 183(14), sta. 184(7), sta. 185(1), sta. 186(4), sta. 187(5), sta. 188(10), sta. 189(2), sta. 190(12), sta. 191(2), sta. 192(3), 193(1), sta. 194(1). Preserved material: 5, 6, 7, 12, 16, 20, 25, 65, 66; 71, 79, 80, 83, 86, 91, 98, 110, 114, Ыб. 118119, 123, 131, 146, 181, 183, 184, 185, 186, 189, 190, 191, 192. Distribution P. dictyodes occurs principally on the main- land; it is absent from the Belep Islands, and its presence on the Isle of Pines (Crosse, 1894) has not been confirmed by recent col- lections. lt appears to be absent from the coastal areas along the W coast, and also from the extreme SE. It is, however, the most widely distributed species of Pararhytida, tol- erating the greatest range of altitude and rainfall (Table 1). Shell The shell of P. dictyodes is generally larger than in other species, ranging from 21 x 11.1 mm to 36.8 x 21.2 mm. There are from 5.6 to 6.9 whorls (mean 6.24, s.d. 0.19) in adult shells. lts dimensions overlap only with those of P. mouensis from the extreme SE (Mt. Guemba, sta. 47), where P. dictyodes does not occur. The only shell character allowing constant specific recognition, even in juve- niles, is the relative flatness of the shell apex. Geographic variation in shell dimensions is considerable and will form the subject of a further paper. The shell is relatively small in the extreme N (Fig. 7), and in some localities on the NW side of the mainland (Mt. Koniambo, sta. 88; Plateau de Tango, sta. 16). Its size increases southeastwards, and approaching the isolated massifs along the northwestern coast (eastern coast: stas. 83 (Fig. 8), 86, 89, 185, 186, 187, 188; north- western massifs: 12, 20, 97). The maximum size is reached SE of the Houailou valley (around stas. 110, 114 (Fig. 9), 189). In the SE plains (stas. 142, 192, 193, 194 (Fig. 10)) size again decreases slightly. Radula (Figs. 11-14) One juvenile and eleven adult radulae, from a wide geographical range of sites, were examined with a stereoscan electron micro- scope. The species shows considerable variation in tooth size, overlapping at the lower end of the range with the five other species (Table 2). The number of teeth per row, however, always exceeded that of other Pararhytida species. Overall tooth shape varies particularly in respect of the central tooth, which appeared to be especially narrow at stations on the NE coast (stas. 66, 80, 184); Fig. 11 shows the normal central and lateral dentition of P. dictyodes. There was little difference between the central and lateral teeth of immature and adult individuals at sta. 91, although the marginals of the former were markedly narrower (Fig. 12). A radula with quite exceptionally large teeth (Figs. 13, 14) was recorded from Mt. Paeoua, but other similarly isolated mountain sites nearby (stas. 91, 189) did not show any tendency towards size increase. Specimens of P. dictyodes occurring sympatrically with P. pyrosticta (sta. 91), P. phacoides (sta. 118), and P. mouensis (sta. 192) all appear to have quite normally sized teeth, and in the last case (sta. 192) the two co-occurring species have radular teeth of almost identical size. MORDAN £ TILLIER FIGS. 6-10. 6. Helix dictyodes Pfeiffer, lectotype, BMNH 1981262. New Guinea [sic], leg. Lieutenant Ince, Cuming Collection. 7. P. dictyodes, Grottes de Koumac, sta. 6. 8. P. dictyodes, Thiem, sta. 83. 9. P. dictyodes, Rembai, sta. 114. 10. P. dictyodes, Faux Bon Secours, sta. 194. Scale line all 10 mm. TAXONOMY OF PARARHYTIDA 217 FIGS. 11-14. 11. Central and lateral teeth, P. dictyodes, Tiwaka, sta. 186. 12. Marginal teeth, P. dictyodes, Aoupinié, sta. 91. 13. Central and lateral teeth, P. dictyodes, Mt. Paéoua, sta. 20. 14. Side view of lateral teeth, P. dictyodes, Mt. Paéoua, sta. 20. Scale divisions all 10 рт. Pulmonary complex The length of the lung (Fig. 4) is normally about 0.75 whorls. lt never exceeds 0.8 whorls, and is shorter (0.5—0.6 whorls) in the wet stations between the valleys of the Amoa and Houailou rivers (stas. 86, 91, 189). The rectal arm of the kidney is slightly longer than the cardiac arm, and its length is ca. уз the lung length. Genital apparatus (Figs. 5, 15-20) As with the shell and other organ systems, the genital apparatus of P. dictyodes shows considerable geographic variation. However, each part of the genital system shows inde- pendent clinal variation. Hermaphrodite gland (Figs. 5, 15-18): In the northern part of the mainland (stas. 5, 7, 65, 66, 71, 79, 80, 183, 184; Figs. 15A, B, C) 218 MORDAN & TILLIER FIG. 15. Genital apparatus of Pararhytida dictyodes. A, sta. 5; B. sta. 184; C. sta. 80; D. sta. 83. Scale line all 5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; P, penis; PA, penial appendix; PR, penial retractor; SO, spermoviduct; SP, spermatheca; T, talon; V, vagina; VD, vas deferens. TAXONOMY OF PARARHYTIDA 219 FIG. 16. Genital apparatus of Pararhytida dictyodes. A, sta. 20; В. sta. 91; С. sta. 25. Scale line all 5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; P, penis; PA, penial appendix; PR, penial retractor; SO, spermoviduct; SP, spermatheca; V, vagina; VD, vas deferens. 220 MORDAN & TILLIER FIG. 17. Genital apparatus of Pararhytida dictyodes. A, sta. 115; B. sta. 189. Scale line both 5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; P, penis; PA, penial appendix; PR, penial retractor; SO, spermoviduct; SP, spermatheca; V, vagina; VD, vas deferens. TAXONOMY OF PARARHYTIDA 221 SP FIG. 18. Genital apparatus of Pararhytida dictyodes, sta. 192. Scale line 5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; P, penis; PA, penial appendix; PR, penial retractor; SO, spermoviduct; SP, spermatheca; V, vagina; VD, vas deferens. it comprises five lobes, but along the E coast there is a progressive tendency for the second and third lobes (that is, from the top of the stomach) to fuse: at Thiem (sta. 83, Fig. 15D) they are almost fused and further southeastwards (stas. 25, 86, 91, 185, 186; Figs. 16 B, C) fusion is complete. In the central western massif of the Paéoua (sta. 20) the hermaphrodite gland has five (or possibly six) lobes; this may simply be an effect of increased size (the snails are larger than at sta. 25, the nearest from which we have preserved material), or it may be that the number of lobes remains constantly five along the NW coast (there is no preserved material between stas. 6 and 20). Further southeastwards, between the Col des Rous- settes, the Col d'Amieu, and Mt. Do, the number of lobes is reduced to three by the fusion of the upper two lobes (stas. 110, 123, 189, 190; Fig. 17B), but eastwards and southwards it becomes four again (stas. 115, 118, 191, 192; Figs. 17А, 18). Albumen gland and spermoviduct: size variation in the albumen gland, talon and spermoviduct was not analysed because this appears to depend principally on the state of maturity of the animal. 222 MORDAN & TILLIER Free oviduct: generally this is longer than the spermatheca when uncoiled; when coiled itis compacted along the spermatheca stalk. It does, however, show enormous geograph- ical variation in length, by up to about four times. It tends to be shorter in the NE valleys (stas. 86, 185, 186), and reaches its minimum on the central Mt. Paéoua (sta. 20, Fig. 16A). It seems that the reduction in length of the free oviduct from М to $ is gradual, whereas the increase in length southeastwards 1$ rather abrupt between stas. 20 (Fig. 16A) and 86 on the one hand (short oviduct) and sta. 91 on the other (long oviduct). The oviducal length 15 intermediate at sta. 25 (Fig. 16C). In the $ plains (sta. 192, Fig. 18) the oviduct is again shorter than further N. Spermatheca and vagina: the head of the spermatheca is thin-walled and smooth. It is always elongate, and almost constant in shape. Its size is more or less uniform within each of two large geographical regions: it is smaller north of the Col d’Amieu (sta. 190; circa 1 cm in length), and larger S and E of this station (circa 1.5 cm in length). In the S plains (sta. 192) it is again reduced to about the same size as S of the Col d’Amieu. The spermathecal stalk and vagina form a single functional unit, as shown by the ab- sence of any discontinuity in internal orna- mentation at the level of oviduct insertion (Figs. 19, 20). The total length of this unit is almost constant over the entire range of the species, being shorter only in the S plains (sta. 192, Fig. 20D), but the relative size of stalk and vagina exhibit considerable geo- graphic variation. The stalk is longer than the vagina in the NW region (stas. 65, 66, 70, 71, 184), and shorter on the E slope of Mt. Panié (stas. 79, 80). The stalk continues to reduce in relative length southeastwards (stas. 83, 185, 186), reaching a minimum at stas. 20 and 86. lts relative length again increases southeastwards of these stations, and from the Col d'Amieu southeastwards the stalk length 15 equal to or slightly greater than that of the vagina. The level of insertion of the transverse vaginal ridge is generally at the mid-point of the length of the vagina, but varies between the lower quarter at Mt. Paéoua (sta. 20, Fig. 20A) and the upper extremity of the vagina at Mt. Ouenarou (sta. 192, Fig. 20D). This vari- ation in the position of the ridge is probably also geographic, since it lies below the middle of the vagina in all stations around Mt. Paéoua (stas. 25, 86, 91, 115, 189, 190; Figs. 20B, С), that is, in central New Caledonia. The transverse ridge in the vagina 1$ never perfectly symmetrical, tending to be more expanded on the penial side (Fig. 19A). In Thiem (sta. 83), on Mt. Paéoua (sta. 20, Fig. 20A) and in the southern Mt. Ouenarou (sta. 192, Fig. 20D), this trend is developed to a point where the transverse ridge becomes a foliated appendage hanging in the vagina, attached only along a small portion of the vaginal circumference. Along the NW coast at least, this particular character-state 1$ devel- oped along a cline: an intermediate condition can be observed in stations along the north- western coast around Thiem (stas. 80, 86, 184, 185; Figs. 19 B, C). The position of the knob inside the spermathecal stalk varies independently from the length of the latter: in N populations (southwards to stas. 20 and 86), it is at about the same level or just above the oviducal opening; from stas. 25 and 91 southeast- wards it is at the upper extremity of the stalk (Figs. 19, 20). Penial complex: externally, the general trend is reduction in absolute and relative length of the penis proper from М (Fig. 15) to S (Fig. 18). In N New Caledonia, the eastern slope of Mt. Panié excepted, the penis 1$ longer than the vagina (stas. 5, 7, 65, 66, 71, 183, 184; Figs. 15A, В). Further $ the penis 1$ generally slightly shorter than the vagina (Fig. 16). On the eastern slope of Mt. Panié (stas. 79, 80; Fig. 15C), in the Amoa and Tiwaka valleys (stas. 86, 186), and on Mt. Paéoua (sta. 20; Fig. 16A), the penis becomes much shorter than the vagina, although this is due more to a relative increase in the length of the vagina than to a shortening of the penis. From stas. 91 and 189 southeastwards the abso- lute length of the penis regularly decreases in correlation with the vaginal length (Fig. 17B). Penial shape also varies in relation to internal characters. In N stations (stas. 79 and 80 excepted) the two principal pilasters are par- ticularly prominent, and the longer one api- cally inflated, even forming a verge in stas. 65, 66 and 183 (Fig. 19A). Correlatively the head of the penis becomes inflated (stas. 5, 7, 65, 66, 71, 83, 183, 184, 185; Fig. 15А, В, D). Further $, and on Mt. Рате, the apical part of the principal pilaster is weaker or even lacking and the two pilasters less prominent, resulting in a more fusiform penis (stas. 20, 25, 79, 80, 86, 91, 110, 186, 189; Figs. 15C and 19B, TAXONOMY OF PARARHYTIDA 223 FIG. 19. Genital apparatus of Pararhytida dictyodes. A, sta. 66: В, sta. 80; C, sta. 184, D, sta. 83. Scale lines all 5 mm. E, epiphallus; O, oviduct; OO, oviducal opening; РАО, penial appendix opening; PI, penial pilasters; PR, penial retractor; SK, spermathecal knob; SS, spermathecal stalk; VA, vaginal appendix; VE, verge; VR, vaginal ridge. 224 MORDAN & TILLIER FIG. 20. Genital apparatus of Pararhytida dictyodes. A, sta. 20; В, sta. 91; С, sta. 191; D, sta. 192. Scale lines all 5 mm. E, epiphallus; O, oviduct; OO, oviducal opening; РАО, penial appendix opening; PI, penial pilasters; PR, penial retractor; SK, spermathecal knob (cut in 20A); SS, spermathecal stalk; VA, vaginal appendage; VR, vaginal ridge. TAXONOMY OF PARARHYTIDA 225 16A and 20A, 6B and 208). $ and Е of Col d'Amieu (stas. 115, 118, 123, 190, 191; Fig. 20C) the pilasters are thicker, stouter, and more regular. The longer principal pilaster 1$ prolonged transversely between the opening of the penial appendix and the epiphallic pore, and one of the secondary pilasters 15 apically inflated. This internal morphology produces a penis which externally resembles that from the N stations, but which results from a differ- ent internal pilaster structure (Figs. 19A, 20C). In the southernmost station (192) the penis has only numerous more-or-less equal pilasters that abut apically onto a large trans- verse ridge (Fig. 20D). Spermatophore: In contrast to most other genital characters, the size and shape of the spermatophore of P. dictyodes seems re- markably constant. It is horny, and comprises a fusiform body, and a thin tail at least five times the body length. The distal part of the tail is thicker than the proximal part (Fig. 5B), becoming thinner again towards its extremity, which is perforated. lt bears a denticulate ridge which originates on the distal part of the spermatophore body. Discussion P. dictyodes is easily recognised by its large size and flat apical whorls. Anatomi- cally, it is the only Pararhytida with: 1. A lung never exceeding 0.8 whorls in length, and typically shorter. 2. A spermathecal head of the size and shape described above. Other characters are so variable that they cannot be considered diagnostic. The problem lies not so much in the recog- nition of P. dictyodes as defined here, but rather to be sure that all the populations studied belong to a single species. However, even if geographic forms could be recog- nised, the fact that characters appear to vary independently suggests that gene flow is т- deed taking place. Pararhytida mouensis (Crosse, 1868) (Figs. 21, 24, 25, 27-29) Helix mouensis Crosse, 1868: 152, pl. 8, fig. 5 (Mt. Mou). Helix dictyonina (Noumea). Helix dictyonina var. globulosa Euthyme, 1885: 256. Trochomorpha dictyonina (Euthyme). Crosse, Euthyme, 1885: 257 1894: 243, pl. 8, fig. 4; Dautzenberg, 1906: 258, pl. 8, figs. 4-6. Pararhytida dictyonina (Euthyme). enberg, 1923: 140. Charopa (Tropidotropis) gudei Preston, 1907: 220, fig. 7 (New Caledonia). Pararhytida (Pararhytida) mouensis (Crosse). Franc, 1956: 137; Solem, 1961: 467. Pararhytida (Pararhytida) dictyonina (Euthyme). Franc, 1956: 137; Solem, 1961: 467. Dautz- Lectotype of H. mouensis Crosse: MNHN. Mont Mou, Marie Colln. (Fig. 21). Dimensions (mm): Shell height 8.6. Shell diameter 17.3. Aperture height 6.4. Aperture diameter 8.2. Umbilicus width 2.7. Whorls 5.2. Other type material: Lectotype (BMNH Reg. no. 1907.5.20.106) and paralectotype (BMNH Reg. no. 1923.2.20.7) of C. gudei Preston, New Caledonia, ex Preston. Lectotype and 5 paralectotypes (MNHN) of H. dictyonina Euthyme, New Caledonia (Fig. 22). Other material: sta. 37(1), sta. 43(7), sta. 47(1), sta. 125(1), sta. 128(1), sta. 130(1), 136(4), sta. 140(1), sta. 142(7), sta. 150(1), sta. 179(6), sta. 192(3). Preserved material: 37, 43, 47, 125, 128, 136, 140, 142, 179, 192. Distribution P. mouensis is restricted to the SE mainland of New Caledonia, from Mt. Humboldt (sta. 125) southeastward. We did not collect it in the coastal lowlands NW of Noumea, nor between Yaté and Goro (stas. 48, 49, 50, 195). It was found at stations with rainfall ranging from 1800 mm (Col de Mouirange, sta. 142) to 4500 mm (Mt. Humboldt, sta. 125). It occurs sympatrically with P. dictyodes at low altitudes inthe S plains (stas. 142 and 192), and with P. marteli on Mt. Guemba (sta. 47). Shell (Figs. 21, 22) The shell varies in size from 18.9 mm x 10.5 mm to 24.9 mm x 15.1 mm. These extremes are found only at the edges of the range, the smallest in the W Mt. Koghi (sta. 140) and Mt. Mou (the type locality), and the largest on the eastern Mt. Guemba (sta. 47). The species is generally intermediate in size between the larger P. dictyodes and the re- maining four smaller species. In contrast to P. dictyodes, the suture is moderately im- 226 MORDAN & TILLIER FIGS. 21-23. 21. Helix mouensis Crosse, lectotype, MNHN. Mt. Mou, leg. Marie, 1868. 22. Helix dictyonina Euthyme, lectotype, MNHN. Noumea, Jousseaume Collection. 23. Trochomorpha (Videna) marteli Dautzenberg, lectotype, MNHN. New Caledonia. Scale line all 5 mm. pressed. The shell is dome-shaped, quite unlike that of any other Pararhytida. The adult whorl count is broadly related to shell diame- ter, from 5.6 whorls in Mt. Koghi (sta. 140, D = 18.9 mm) to 6.25 whorls in Mt. Guemba (sta. 47, D = 24.9 mm) and 6.4 in Col de Mouirange (sta. 142, D ca. 22 mm); mean whorl count is 6.08 (s.d. 0.26). Radula (Figs. 24, 25) The radulae of four specimens were exam- ined. The size and shape of individual central and lateral teeth are close to those of P. dictyodes, although appearing slightly shorter and broader, and show a considerable de- gree of variation (Table 2). The marginal teeth are in general shorter and narrower than in P. dictyodes, and both these and the laterals are typically fewer. At two of the four sites P. mouensis is sympatric with other species: with P. marteli at sta. 47, and with P. dictyodes at sta. 192. However there was as much radular variation between the two sympatric sites as between allopatric situa- tions, and at sta. 47 tooth size and number TAXONOMY OF PARARHYTIDA 22 were almost identical in P. marteli and P. mouensis. Pulmonary complex The pulmonary cavity occupies the last 0.9 to 1.2 whorls of the visceral coil, but we did not have enough preserved material to distin- guish any clear geographic pattern of varia- tion. The rectal arm of the kidney is only slightly longer than the cardiac arm in north- ern stations (near Mt. Dzumac, sta. 128), but about 1.5 times longer in the southern plains (stas. 43, 47, 142, 179, 192). Genital apparatus The number of lobes in the hermaphrodite gland increases from N to S, from four close to Mt. Dzumac (sta. 128), to five on the Montagne des Sources (sta. 136) and Mt. Ouénarou (sta. 192), six in the Riviere Bleue valley (sta. 43) and Mt. Guemba (sta. 47), and finally seven in the Col de Mouirange (sta. 142) (Fig. 27). The relative length of the free oviduct also increases from N to S: it is shorter than the spermathecal length in northern stations (stas. 43 and 128; Figs. 27A, B), about the same length at station 136, and distinctly shorter at the other stations (sta. 47, 142, 179, 192; Figs. 27C, D). The spermatheca has a highly characteris- tic shape, with a very long head of a constant diameter that is greater than the diameter of the short spermathecal stalk (Fig. 27). The vagina does not possess a well- developed transverse ridge or internal ap- pendage as do all other species (Fig. 28), but at most shows a thickening of the wall and an interruption in its longitudinal internal ridges. From their position, these ridges are consid- ered homologous with those of other species; when present they are located below the mid-point of the vagina, except at sta. 128 (Fig. 28A). Although an internal ridge is lacking, the vagina of P. mouensis does possess one or two outgrowths of the vaginal wall which form pouch-like structures. These are not visible externally except as a thicken- FIGS. 24-26. 24. Central and lateral teeth, P. mouensis, Mt. Ouénarou, sta. 192. 25. Central and lateral teeth, P. mouensis, Col de la Ouinné, ; sta. 128. 26. Central and lateral teeth, P. marteli, 26 Mt. Guemba, sta. 47. Scale divisions all 10 um. 228 MORDAN 4 TILLIER FIG. 27. Genital apparatus of Pararhytida mouensis. A, sta. 128. B, sta. 43. C, sta. 142. D, sta. 47, showing position of spermatophore. Scale line all 5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; P, penis; PA, penial appendix; PR, penial retractor; SO, spermoviduct; SP, spermatheca; Sp, spermatophore; V, vagina; VD, vas deferens. TAXONOMY OF PARARHYTIDA 229 FIG. 28. Internal genital morphology of Pararhytida mouensis. A, sta. 128. В, sta. 43. С, sta. 48, showing the interior of the two vaginal pouches, but the vagina unopened within the sheath that surrounds the pouches. D, sta. 136. E, sta. 142. Scale lines all 5 mm. E, epiphallus; O, oviduct; OO, oviducal opening; OVP1, OVP2, openings of the vaginal pouches; PA, penial appendix; SK, spermathecal knob; SS, spermathecal stalk; VP1, VP2, vaginal pouches; VR, vaginal ridge. 230 MORDAN & TILLIER ing of the vagina, being tightly bound in connective tissue, but are clearly visible when the vagina is opened (Figs. 28B, C). Their number and position vary geographically. In southern and central stations (stas. 43, 136, 179; Fig. 28B) the lower pouch lies just above the transverse ridge, and the upper pouch just beside the oviducal opening. In the NE (Mt. Guemba, sta. 47) only the upper pouch 1$ present; this pattern probably arose through the loss of the lower pouch, which 1$ less well developed in sta. 179 than in stas. 43 and 136. In the М (sta. 128, Fig. 28B) there 1$ also a single upper vaginal pouch, but here it probably results from the fusion of the two pouches, since at the intermediate stations (142, 192) the two pouches are contiguous beside the oviducal opening (Fig. 28E). The knob in the spermathecal stalk is always situated just above the oviducal opening. The penis 15 nearly constant in size, and 1$ of a regular fusiform shape (Fig. 27). The epiphallus 15 a little more than twice as long as the penis, except in the northernmost station (128) where it is shorter (Fig. 274). The internal ornamentation is typically formed of numerous short pilasters, and a verge originating from the upper end of the pilaster zone (Figs. 28A, C-E). This pattern may be modified, even within a population, by loss of the verge or weakening of the pilasters, the latter arrangement (Fig. 28B) permitting recognition of the homology of pilasters and verge between P. mouensis and P. dictyodes. Spermatophore: these or fragments were only found at three stations (47, 136, 142); here the shape is fusiform, with no clear delimitation between body and tail (Fig. 29). A ridge runs the entire length of the spermatophore, appearing smooth at low magnifications, but minutely serrate at high. At Mt. Guemba (sta. 47), the spermatophore is much larger and more elongate than at the other stations. The size is probably related to the larger size of the animals there, but the change in shape is more problematic. At sta. 47 a spermatophore was found in situ, with its body in the spermathecal head, and the tail lying within the free oviduct (Fig. 27D). Discussion P. mouensis is generally smaller than P. dictyodes and larger than the other four spe- cies of Pararhytida. lt overlaps in size with P. FIG. 29. Spermatophore of Pararhytida mouensis, sta. 47. Scale line 1 mm. dictyodes only at Mt. Guemba (sta. 47) where dictyodes has not been recorded, and with the other four species in the chain of Mt. Mou and Mt. Koghi (type locality and sta. 140), where similarly none of the small species 1$ found. Conchologically it can be distinguished from P. dictyodes by its more impressed suture (especially on the first whorl), and from the small species by its rounded profile. It is the only species without a vaginal ridge or appendage, and possessing vaginal pouches. Although we were unable to collect topotypic material of P. mouensis, we are in little doubt that it is only the westernmost form of what has previously been called Р. dictyonina, from which it differs in whorl num- ber but not in rate of whorl increase. Further- more, the shells of P. dictyonina become flatter at stations approaching those where typical P. mouensis occurs. The attribution of animals from Mt. Guemba to P. mouensis 1$ more questionable, since their shells more closely resemble those of P. dictyodes. How- ever, these animals are closer in anatomy to P. mouensis than to any other species, and we have no reason to preclude the possibility of a cline existing between Mt. Guemba and the remaining stations. Moreover, it seems TAXONOMY OF PARARHYTIDA 231 unlikely that the observed differences in spermatophore morphology between М. Guemba specimens and other P. mouensis are sufficient to prevent successful copula- tion. The anatomical change may result from sympatry with the smaller P. marteli at Mt. Guemba, and with the larger P. dictyodes over the rest of its range. Pararhytida marteli (Dautzenberg, 1906) (Figs. 23, 26, 30) Trochomorpha (Videna) marteli Dautzenberg, 1906: 257, pl. 8, figs. 7-9 (New Cale- donia). Pararhytida (Pararhytida) marteli (Dautz- enberg). Franc, 1956: 138; Solem, 1961: 467. Lectotype (here designated): New Caledonia, leg. Martel, MNHN. (Fig. 23). Di- mensions (mm): Shell height 9.8. Shell diam- eter 17.9. Aperture height 6.8. Aperture diam- eter 8.7. Umbilicus width 1.7. Whorls 5.9. Other material: sta. 47(1), sta. 48(36), sta. 49(20), sta. 50(1), sta. 195(17). Preserved material: 47, 48, 49, 50, 195. Distribution P. marteli has been recorded only from the extreme SE of New Caledonia, from Yate (close to sta. 47) to Goro (sta. 50). However, we did not collect it from the Kouakoue, NW of sta. 47, and cannot be sure that it does not occur there. It was typically collected from stations with high rainfall, from 1900 mm to 3000 mm a year. Shell (Fig. 23) The shell is sharply carinated, domed above and shallowly rounded below, ranging in size from 16.1 x 8.8 тт to 20 x 10.9 mm. The carina is situated about half way up the palatal wall. Adult shells have from 5.2 to 6.2 whorls (mean 5.59; s.d. 0.2). Radula Only one radula was examined, from a specimen collected at Mt. Guemba (sta. 47, Fig. 26). The teeth are generally similar in shape, size, and number to P. phacoides (Table 2), and also to P. mouensis with which it is sympatric at sta. 47, although in the former case the lateral mesocones of P. phacoides are significantly smaller. Pulmonary complex The length of the lung varies between 0.9 and 1.3 whorls, as in P. mouensis (the stom- ach and crop have the usual length of one whorl, but the upper part of the visceral coil is longer than usual, extending between 3.3 and 3.6 whorls). The rectal arm of the kidney is about 1.5 times longer than the cardiac arm, and occupies less than 1/3 of the lung length. Genital apparatus The hermaphrodite gland has five lobes at stas. 47, 49 and 195 (Figs. 30A, C), but only four at Touaourou (sta. 48). Externally the genitalia resemble, in reduced form, those of P. mouensis (Figs. 27, 30). From N (sta. 47) to S (sta. 48 southward) the relative length of the free oviduct varies from slightly longer than the spermatheca to about half the length of the latter. Internally the only constant dif- ferences from the arrangement found in P. mouensis are the presence of a vaginal ap- pendage and the absence of a vaginal pouch in P. marteli; the appendage is attached at the mid-point of the vagina (Figs. 30C, D). Spermatophore: One complete spermato- phore was found at Mt. Guemba (sta. 47; Fig. 30E). It has no tail and is smaller and stouter than in P. mouensis. Its longitudinal ridge is denticulate only at the open end. Discussion The shell of P. mouensis is thicker and more rounded than that of P. marteli. Al- though the anatomical differences between these two species are slight, their distinctness is confirmed at Mt. Guemba where they occur sympatrically. Distinguishing the shells from those of P. phacoides, P. pyrosticta and P. thyrophora is more difficult: those of P. marteli are interme- diate in shape between the more rounded P. thyrophora and the flatter P. phacoides and P. pyrosticta. In these last two species, the carina is situated higher on the whorl contour, 232 MORDAN & TILLIER FIG. 30. Genital apparatus of Pararhytida marteli. A and С, sta. 195. В, External morphology, sta. 49. О, internal morphology, sta. 48. E, spermatophore, sta. 47. Scale lines A and В both 5 mm. С-Е both 2.5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; OO, oviducal opening; P, penis; PA, penial appendix; PAO, penial appendix opening; Pl, penial pilasters; PR, penial retractor; SK, spermathecal knob; SO, spermoviduct; SP, spermatheca; V, vagina; VA, vaginal appendage; VD, vas deferens; VR, vaginal ridge. TAXONOMY OF PARARHYTIDA 233 and they more closely resemble juveniles of P. dictyodes. Pararhytida phacoides Mordan & Tillier, п. sp. (Figs. 31, 34, 37) Holotype: Mt. Boulinda, 980-1020 m, be- tween Petit and Grand Boulinda, altitude rainforest. Coll. A. 4 5. Tillier, 6.vii.1979 (sta. 97), MNHN (Fig. 31). Dimensions (mm): Shell height 10.1. Shell diameter 19.4. Aperture height 7.3. Aperture diameter 9.9. Umbilicus width 1.7. Whorls 5.6. Paratypes (preserved): 2, MNHN. Other material (preserved): sta. 118 (2 + 1 shell), 97 (1 broken shell). Etymology: lens-like. Greek: phacos, a lentil. as above, Distribution The two stations where this species was collected (stas. 97 and 118) probably repre- sent the N and S limits of its distribution: to the N it is replaced by P. pyrosticta and to the S by P. mouensis. At both stations rainfall is high (3000 mm at sta. 97 and 2600 mm at sta. 118), and the distributions of the small Pararhytida species suggest that this rela- tionship is not chance. P. phacoides 1$ probably endemic to very wet rainforests on the mountains of southern central New Caledonia. Shell (Fig. 31) Only three adult shells were collected, ranging from 19.7 to 21.5 mm in width, from 9.7 to 10.6 mm in height, with an umbilicus of circa 2 mm in diameter. They have from 5.3 to 5.7 (mean 5.5, s.d. 0.18) whorls, and are sharply carinated, with the carina on the upper part of the palatal wall. The aperture 1$ only slightly expanded. The upper surface of the initial whorls is convex, giving the shells a slightly more conical shape than in juvenile P. dictyodes. The border of the umbilicus of shells from Mt. Nakada 1$ slightly shouldered. Radula Two specimens were examined from the only known localities for this species (stas. 97 and 118); the radula from sta. 118 was juve- nile and was not measured. The radula from sta. 97 (Fig. 34) is similar to that of P. marteli, except that the marginal teeth tended to be broader, whilst the centrals and laterals are slightly narrower and more pointed. Pulmonary complex The lung occupies the final 0.9 whorl in the fully adult preserved specimen (sta. 97). The arms of the kidney are almost equal in length, which is about ‘3 of the total lung length. The pulmonary complex of P. phacoides is closer in anatomy to that of P. dictyodes than to either P. mouensis or Р. marteli. Genital apparatus In both internal and external anatomy, the genitalia of P. phacoides resemble a reduced version of the genital apparatus of P. dic- tyodes (Figs. 37A, B). The hermaphrodite gland of the only specimen in which it was observed had four lobes. The only significant difference is in the shape of the spermathecal head, which is triangular and relatively shorter than in P. dictyodes. However, this shape may simply have been due to the presence of a spermatophore. Spermatophore: this found at Mt. Boulinda is identical in shape and arrangement with those of P. dictyodes, and is only slightly smaller (Fig. 37C). Discussion If P. phacoides had not been found sympatrically with P. dictyodes at both stations where it was collected, separation from the latter would have been extremely difficult. It resembles closely a young P. dictyodes. There 1$, however, a slight differ- ence in convexity of the early whorls, and also in rate of whorl increase, and maturity 1$ reached at less than six whorls. In shell characters P. phacoides 1$ simlar to P. pyrosticta, but the latter is slightly smaller for the same whorl count. Anatomical differ- ences between P. phacoides, P. pyrosticta and P. thyrophora will be considered later. Pararhytida pyrosticta Mordan 4 Tillier, п. Sp. (Figs. 2, 32, 35, 38) Pararhytida marteli (Dautzenberg). Dautz- enberg, 1923: 140. 234 MORDAN & TILLIER FIGS. 31-33. 31. Pararhytida phacoides n. sp., holotype, MNHN. Mt. Boulinda, sta. 97. 32. Pararhytida pyrosticta n. sp., holotype, MNHN. Mt. Tchingou, sta. 84. 33. Pararhytida thyrophora n. sp., holotype, MNHN. lle Art, sta. 58. Scale line all 5 mm. Holotype: S slope of Mt. Tchingou, 1250 m, rainforest. Coll. P. Bouchet, A. & S. Тег, vii.1979 (sta. 84), MNHN (Fig. 32). Dimen- sions (тт): Shell height 8.0. Shell diameter 16.5. Aperture height 5.8. Aperture diameter 8.1. Umbilicus width 1.8. Whorls 5.1. Paratypes (preserved): 5 + 6 juveniles, as above, ММНМ (7 shells). Other material: sta. 72(2), sta. 91(1), sta. 92(3). Preserved material: 72, 84, 91, 92. Etymology: with flame-like dots. Greek: pyr, fire; stictus, spotted. Distribution P. pyrosticta is restricted to very wet rainforests on the mountains of mainland New Caledonia N of the Houailou valley. This area corresponds to the northernmost part of the central chain together with the eastern slopes and summit areas of the Panié massif. At the stations from which it was collected rainfall ranges from 2500 to 3500 mm per annum. Shell (Fig. 32) The shell is similar to that of P. phacoides, being only slightly smaller and with fewer TAXONOMY OF PARARHYTIDA 235 DE |: OS ee 36A B FIGS. 34-36. 34A, central and lateral teeth; В, marginals. Pararhytida phacoides п. sp., Boulinda, sta. 97. 35A, central and lateral teeth; B, marginals. Pararhytida pyrosticta n. sp., Mt. Ignambi, sta. 72. 36A, central and lateral teeth; В, marginals. Pararhytida thyrophora п. sp., lle Art, sta. 58. Scale divisions all 10 шт. 236 MORDAN & TILLIER rIG. 37. Genital apparatus of Pararhytida phacoides п. sp., sta. 97. A, External morphology, indicating position of spermatophore. Scale line 5 mm. B, internal morphology. Scale line 2.5 mm. C, spermatophore. Scale line 2.5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; OO, oviducal opening; P, penis; PA, penial appendix; PAO, penial appendix opening; Pl, penial pilasters; PR, penial retractor; SK, spermathecal knob; SO, spermoviduct; SP, spermatheca; Sp, spermatophore; SS, spermathecal stalk; V, vagina; VD, vas deferens; VR, vaginal ridge. TAXONOMY OF PARARHYTIDA 237 whorls when adult. Shells measure from 14.7 to 18 mm in diameter and from 7.3 to 8.4 mm in height, with an umbilicus from 1 to 1.9 mm in diameter. They have from 5 to 5.3 whorls on Mt. Ignambi (sta. 72) and Mt. Aoupinié (stas. 91, 92), but only between 4.7 and 5.1 whorls on Mt. Tchingou (sta. 84). Mean adult whorl count is 4.99 (s.d. 0.23). Radula Two radulae were examined (stas. 72 and 84, Fig. 35) and in both, the lateral teeth were distinctly narrower and sharper than in any other Pararhytida species. The mesocone of the central tooth also is narrow in comparison with species other than P. thyrophora. The number of teeth (Table 2) is about average for the smaller taxa. Pulmonary complex The lung cavity extends back 0.6 whorls on northern Mt. Ignambi (sta. 72), 0.75 whorls on the southern Aoupinié (stas. 91 and 92), and 0.8 to 0.9 whorls on the northwestern Mt. Tchingou (sta. 84). The rectal arm of the kidney 1$ slightly longer than the cardiac arm; its absolute length is nearly constant, and thus it occupies more than Y of the lung length in Mt. Ignambi, but less than Y in Mt. Tchingou. Genital apparatus Externally, the genital apparatus of P. pyrosticta is characterised by a penis that 1$ markedly shorter than the vagina, and by the size and shape of the spermatheca which has an ovoid head and a short, thick stalk (Fig. 38B). Internally the penis has at most equally weak pilasters, and a small subapical verge in Mt. Tchingou specimens. All dissected spec- imens possessed an elongated vaginal ap- pendage which is inserted above the mid- point of the vagina in Mt. Ignambi and Mt. Aoupinie specimens, and below it in speci- mens from Mt. Tchingou. The spermathecal knob is located just above the oviducal open- ing (Figs. 38C, D). In all specimens dissected the hermaphrodite gland has five lobes. The free oviduct is shorter at Mt. Tchingou than at the other two mountains. Spermatophore: One complete spermat- ophore and some fragments were found in the Mt. Tchingou specimens. The complete spermatophore is very elongate, but without a differentiated tail, as in P. mouensis (Fig. 38A). The ridge is finely denticulated and extends around the imperforate extremity. Discussion P. pyrosticta looks externally very like P. phacoides, being only slightly smaller. In terms of genital anatomy the two species differ principally in spermatophore morphol- ogy, spermathecal shape, oviducal length, and internal penial and vaginal ornamenta- tion. Pararhytida thyrophora Mordan & Tillier, п. sp. (Figs. 33, 36, 39) Holotype: lle Art, Belep Islands, N plateau. Coll. P. Bouchet and A. Warén, 7.vii.1979 (sta. 58). MNHN (Fig. 33). Dimensions (mm): Shell height 9.6. Shell diameter 18.3. Aper- ture height 6.7. Aperture diameter 9.1. Umbil- icus width 1.7. Whorls 5.2. Paratypes (preserved): 10 + 25 juveniles, as above, MNHN (+ 69 shells); 3 specs. BMNH Reg. no. 1984101; 3 specs., The Australian Museum, Sydney. Other material: sta. 57(3). Etymology: door-bearing. Greek: thyra, a door. Distribution P. thyrophora is known only from the Belep islands of Art and Pott. It was collected in dry, high maquis, where rainfall averages 1250 mm per year. Shell (Fig. 33) The shell ranges in size from 17 x 9 mm to 22.5 x 12.1 mm, with the umbilicus about 1 mm in diameter. The upper surface of the shell is more convex than in either P. phacoides or P. pyrosticta, but less than in some P. mouensis. The aperture is more rounded in adults than in juveniles, and the carina is situated about half way up the palatal wall. The number of whorls in the adult varies from 5.3 to 6.2 (mean 5.65; s.d. 0.19). 238 MORDAN & TILLIER PA PAO RNIT: wk FIG. 38. Genital apparatus of Pararhytida pyrosticta n. sp., stas. 84 and 91. A, spermatophore, sta. 84, scale line 1.25 mm. B, external morphology, sta. 84, scale line 5 mm. C (sta. 84) and D (sta. 91), internal morphology, scale lines both 2.5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; OO, oviducal opening; P, penis; PA, penial appendix; PAO, penial appendix opening; PR, penial retractor; SK, spermathecal knob; SO, spermoviduct; SP, spermatheca; SS, spermathecal stalk; V, vagina; VA, vaginal appendage VD, vas deferens. Radula A single radula was examined from lle Art (sta. 58, Fig. 36). It was characterised by the narrowness of the central tooth mesocone relative to the width of the entire tooth and relative to the mesocone of the laterals. The size and shape of the lateral teeth were normal for the small species of Pararhytida, but the number of teeth in each row (75) was higher. TAXONOMY OF PARARHYTIDA 239 FIG. 39. Genital apparatus of Pararhytida thyrophora n. sp., stas. 57 and 58. A (sta. 58) and C (sta. 57), external morphology, scale lines both 5 mm. B (sta. 58) and D (sta. 57), internal morphology, scale lines both 2.5 mm. AG, albumen gland; E, epiphallus; HD, hermaphrodite duct; HG, hermaphrodite gland; O, oviduct; ОО, oviducal opening; P, penis; PA, penial appendix; РАО, penial appendix opening; Pl, penial pilasters; PN, penial nerve; PR, penial retractor; SK, spermathecal knob: SO, spermoviduct; SP, spermatheca; SS, spermathecal stalk; V, vagina; VD, vas deferens; VR, vaginal ridge. 240 MORDAN & TILLIER Pulmonary complex The lung cavity extends about 0.6 whorls in the dissected specimens (sta. 58). The rectal arm of the kidney 1$ only slightly longer than the cardiac arm, and its length ap- proaches one half of the lung length; only P. dictyodes has such a short lung, and no other species of Pararhytida has such a proportion- ally long kidney. Genital apparatus The penis is nearly as long as the vagina, and the free oviduct longer than the spermatheca (Figs. 39A, C). The shape of the latter is characteristic, with a pear-shaped head and a relatively thin stalk which 1$ longer than the head. We were unable to count the lobes of the hermaphrodite gland accurately, but there were probably five or six. Internally the penis has weak pilasters and a subapical verge at sta. 58 but not at sta. 57 (Figs. 39B, D), although in both cases the animals were fully mature. The vagina has an irregular transverse ridge which 1$ almost developed into a full appendage. It is weaker than the appendage of P. pyrosticta, but better developed than in P. phacoides. It appears to be composed of two parts which partially interlock and occlude the vagina, and is situated at about the middle of the vaginal length. The spermathecal knob lies well above the oviducal opening, at about the mid-point of the spermathecal stalk. No spermatophore was found. Discussion The shell is more rounded in shape than any other small species, some P. mouensis excepted; in this last species, however, the shell is much thicker. In comparison with P. marteli the carination is less sharp and the apex more shallowly domed. Anatomically, the length and proportions of the pulmonary complex, and the shape of the spermatheca, allow easy species recognition. The insular situation of P. thyrophora makes the possibil- ity of any significant gene flow with the main- land, and thus with other species of Pararhytida, seem highly unlikely. ACKNOWLEDGEMENTS We are extremely grateful to Dr. Alan Solem of the Field Museum of Natural His- tory, Chicago, for the loan of material, and to Anne Thompson for assistance with the stereoscan photomicrography. REFERENCES ANCEY, C. F., 1882, Classification des formes helicoides de la Nouvelle-Calédonie. Le Naturaliste, 4: 85-87. ANCEY, C. F., 1888, Nouvelles contributions malacologiques. Bulletin de la Société Malacologique de France, 5: 341-376. BARGMANN, H. E., 1930, The morphology of the central nervous system in the Gastropoda Pulmonata. Journal of the Linnean Society of London, Zoology, 37: 1-59. BAYNE, C. J., 1973, Physiology of the pulmonate reproductive tract: location of spermatozoa in isolated, self-fertilising succinid snails (with a discussion of pulmonate tract terminology). Veliger, 16: 169-175. CROSSE, H., 1868, Description d'especes ter- restres inedites provenant de la Nouvelle Calédonie. Journal de Conchyliologie, 16: 146-164. CROSSE, H., 1894, Faune malacologique terrestre et fluviatile de la Nouvelle-Calédonie et ses dépendances. Journal de Conchyliologie, 42: 161-473. DAUTZENBERG, P., 1906, Description d'une nouvelle espece terrestre Néo-Calédonienne. Journal de Conchyliologie, 54: 257-259. DAUTZENBERG, P., 1923, Mollusques terrestres de la Nouvelle-Calédonie et des lles Loyalty. In: SARASIN, Е. & ROUX, J., Nova Caledonia. A. Zoologie, 3: 135-156. EUTHYME, 1885, Description de quelques mollusques exotiques nouveaux. Bulletin de la Société Malacologique de France, 2: 237-260. FISCHER, P., 1875, Note sur l'anatomie de l'Helix dictyodes, Pfeiffer. Journal de Conchyliologie, 23: 273-276, pl. 14, figs. 3-6. FRANC, A., 1956, Mollusques terrestres et fluviatiles de l’Archipel Neo-Caledonien. Mémoires du Museum National d'Histoire Naturelle, n.s., A, Zoologie, 13: 1-200, 24 pl. GASSIES, J.-B., 1863, Faune conchyliologique ter- restre et fluvio-lacustte de la Nouvelle- Calédonie. Actes de la Société Linnéenne de Bordeaux, 24: 211-330, 8 pl. PFEIFFER, L., 1847, Descriptions of 38 new spe- cies of land-shells in the collection of Hugh Cuming. Proceedings of the Zoological Society of London, “1846”: 109—116. PILSBRY, H. A., 1894, Guide to the study of Helices. Manual of Conchology. Ser. 2: Pulmonata, 9: 366 p., 71 pl. Academy of Natural Sciences, Philadelphia. PRESTON, H. B., 1907, Descriptions of nine new species of land-shells from New Caledonia. An- nals and Magazine of Natural History, ser. 7, 19: 217-221. TAXONOMY OF PARARHYTIDA 241 REEVE, L. A., 1851-4, Monograph of the genus Helix. Conchologia Iconica, vol. 7. Reeve, Lon- don. SOLEM, A., 1961, New Caledonian land and fresh- water snails. An annotated check list. Fieldiana Zoology, 41: 419-501. SOLEM, A., 1972, Malacological applications of scanning electron microscopy. Il. Radular struc- ture and functioning. Veliger, 14: 327-336. SOLEM, A., 1976, Endodontoid land snails from Pacific islands (Mollusca: Pulmonata: Sigmu- rethra). Part |. Family Endodontidae, 508 p. Field Museum of Natural History, Chicago. SOLEM, A., 1983, Endodontoid land snails from Pacific Islands (Mollusca: Pulmonata: Sigmu- rethra). Part Il. Families Punctidae and Charopidae, Zoogeography, 336 p. Field Mu- seum of Natural History, Chicago. SOLEM, A., TILLIER, S. & MORDAN, P., 1984, Pseudo-operculate pulmonate land snails from New Caledonia. Veliger, 27: 193-199. STARMUHLNER, F., 1970, Ergebnisse der óster- reichischen Neukaledonien-Expedition 1965. Terrestrische Gastropoda | (excl. Veronicellidae und Athoracophoridae). Annalen des Natur- historischen Museums in Wien, 74: 289-324. Revised Ms. accepted 4 October 1985 A IN м al pa dh ive ya N In 1 TCM у iv И Mavi ee 1 м ! RB, 1 nit In т à Pr | i 1 i и | т y } q i Kane! o ar E Te ; м (ll AN À | у у E т UNA Mh 14 Y Г À k dd | Ku * J À L у À e An L 1 | er ar 4 Le ia LO ar y у a à ) FR if | A ET) A ON N Е 4 р N i 14 у à Г вы о u) 1 i у Sau, en, т | ' \ i el o Sa р { N i ( bis 1 у à tel | | | у ily | L ns y tal (ROULE h en | я | но LEERE ET BER) en | | o я lcd o | duo, р И О ИР Ei . Nee nel un o ICH | : у А 107 i} aad er 1 ñ (Me ley . . a 5 р м т o | un t 1 у À LD un Al т m o | 1 i | ten, a Li [ou I f . 4 fi | | FU | и 7 } ал | as TN MALACOLOGIA, 1986, 27(2): 243-247 GENETIC STUDIES OF BIPHALLIC BIOMPHALARIA GLABRATA Charles $. Richards! ? & Dennis J. Minchella? ABSTRACT A biphallic Biomphalaria glabrata was observed with preputium, verge and vas deferens on the right as well as the left side. The snail was allowed to reproduce by self-fertilization and its offspring were followed for several generations. Despite selection, the biphallic condition continued to occur in about Y4 of the snails. The original mutant snail was mated in series to four normal albinos of different stocks of B. glabrata. Each cross resulted in some biphallic F,s, and by selection some lines were derived showing essentially 100% frequency of biphallic snails. A dominant factor with low penetrance 15 postulated to be involved in the biphallic trait. Key words: Biomphalaria glabrata; biphallic snails; genetics, molluscan; Schistosoma mansoni, selection. INTRODUCTION Genetic studies on Biomphalaria glabrata (Say) concerned with variations in patterns of susceptibility for infection with different strains of Schistosoma mansoni have in- volved snail crosses. Although pigmentation variations in B. glabrata have provided good genetic markers for crossing experiments (Richards, 1984) and competition studies (Minchella & LoVerde, 1983), other visible characters, particularly any demonstrating linkage with susceptibility factors, would be valuable. Modifications of the reproductive system of B. glabrata, visible in vivo, have been described (Richards, 1974). Eversion of the preputium even in isolated snails varies in frequency in different B. glabrata stocks, apparently involving heritable factors. Abnor- malities such as double, triple, or forked preputia also occur more frequently in some stocks. Eversion and abnormality of the preputium, associated with swollen tentacles, is inherited as a simple recessive character (Richards, 1974). The genetic basis of the biphallic trait in B. glabrata was studied using selection via self-fertilization and crossing experiments. Snails were also exposed to two strains of S. mansoni to determine if the biphallic trait was associated with susceptibility factors. MATERIALS AND METHODS B. glabrata stocks and S. mansoni strains used have been described (Fletcher et al., 1981; Richards, 1984). Snails isolated as juveniles were reared singly in 400 ml bea- kers and fed romaine lettuce. Juveniles were exposed individually to 5 miracidia per snail. Adults (determined by onset of egg-laying) were exposed to 25 miracidia per snail. Snails of various ages, both fresh and preserved, were dissected in order to study the morphology of the biphallic trait. However, during the genetic studies, the presence or absence of a developing preputium on the right side was determined in vivo by examina- tion with a binocular dissecting microscope. During selection for the biphallic trait, only those snails expressing the trait were allowed to contribute to the next generation via self- ing. In crossing experiments two snails were maintained together in a 400 ml beaker for one week, and then reisolated. RESULTS Morphology Snails expressing the biphallic trait were dissected. The male genitalia on the right side developed from the external opening inward. ‘Biomedical Research Institute, 12111 Parklawn Drive, Rockville, Maryland 20852, U.S.A. “Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland 20014, U.S.A. 3Department of Biological Sciences, Purdue University, West Lafayette, Indiana 47907, U.S.A. (243) 244 RICHARDS 8 MINCHELLA TABLE 1. Frequencies of biphallic snails in descendents of snail 831 following four serial crosses with normal albino stock snails (percentage in parentheses). The F, and F3 generations were products of selection and self-fertilization. Normal parent Е, 243432....24 9/65 (13.8) 4XR 7/69 (10.1) 21539 4/32 (12.5) 4132 6/14 (42.8) Ро Fa 76/188 (40.4) 85/312 (27.2) 15/75 (20) 41/100 (42) E». 1. showing everted preputia. Scale line = 1 mm. Photomicrograph of biphallic snail 831, In adult snails the preputium, verge, verge sac and vas deferens appeared normal, but no connection with the rest of the reproduc- tive system was observed in any of the dis- sected snails. The vas deferens terminated posteriad with a closed unattached end. Since the snail stock in which the biphallic condition occurred has the preputium (or preputia) everted most of the time, the biphal- lic condition was readily apparent in adult snails (Fig. 1). Occurrence of biphallic snails in descendents of snail #83 After mating with snail 13142-10-1-684117, parent 13141-121116 produced hybrid F, +8 which appeared normal as did its offspring 83 produced by selfing. (For genealogical details of these snail stocks see Richards, 1984.) Three of four offspring of 83 by selfing iso- lated as juveniles survived. One of these (831) was biphallic, the other two (832 and 834) normal. All three snails produced prog- enies by selfing including some biphallic snails. Descendents of snail 831 were fol- lowed for four generations by self-fertilization of isolated snails and with selection for the biphallic condition in each generation. Snail 831 produced 5/28 (17.9%) biphallic off- spring; the five biphallic snails produced 34/185 (18.4%) biphallic offspring; the third generation yielded 65/370 (17.6%) biphallic snails; the fourth generation 121/426 (28.4%). In succeeding generations there was no con- sistent indication of higher frequencies. Crossing experiments The original biphallic snail (831) after self- ing was mated in series to four albino snails of different stocks. Since snail 831 was black- eyed with coalesced mantle pigment, albino snails were used in the crossing experiments so that it could be determined whether the progenies were hybrids or the products of self-fertilization. The results are shown in Table 1. Each albino parent produced some F,s with the biphallic trait and these snails were selfed to produce future generations. In the first cross F, 41 produced 22/71 (31%) biphallic Fas by selfing. Fı #1 was then backcrossed to one of the offspring by pre- cross selfing of the albino parent (1221124-1) (Fig. 2). This albino produced 20/72 (27.8%) hybrid biphallic offspring. The offspring of Е, #1 by selfing were followed through several generations, including lines of three pigment phenotypes: black-eyed snails with coalesced mantle pigment (c°c°S?%S°), black-eyed snails with spotted mantle (c?c?SS) and albino snails (cc—) (Richards, 1984). In all three pigment phenotypes selection resulted in lines showing high biphallic frequencies. One line (1-19-56) was maintained since it dem- onstrated a pattern of S. mansoni suscep- tibility not previously observed in our snail stocks: juvenile susceptible/adult nonsuscep- tible to both PR-1 and PR-2 (Richards, 1984). GENETICS OF BIPHALLIC BIOMPHALARIA 245 831 243432 .....24 Oy By va 4 chobsasd > cc(SS) 5/28 (18 %) | FiS 9/65(13.8 %) SELFING сс > BACK CROSS oy EN orne So Y ma + SELFING" 227 11 (3195) 20/72 (20.8%) E. ©; Sp cbcbS dSd CC cbBcBSS 1/17(5.9%) 9/21 (42.8%) 10/31 (32.3 %) a A | JN 1725 | 12/14 5 ae | (44%) | (85.7 %) | (40%) cr = Oo nie = AL Е 10/12 (83.3 %) | (93.7%) (100-%) ES er А %) LÍA 16/17 (94.1%) 1715 (733%) FIG. 2. Diagram showing frequencies of biphallic snails; in three lines from Е, #1, by pre-backcross selfing, and from a backcross between F, #1 and a normal stock snail. 246 RICHARDS 8 MINCHELLA Susceptibility to S. mansoni The parental stocks from which snail 83 was derived differed in their susceptibilities for S. mansoni strains PR-2 and PR-T-13. Snail stock 13141 was susceptible to PR-T- 13 but not to PR-2, while stock 13142-10-1-6 was susceptible to PR-2 but not PR-T-13. Snail 831 was exposed as an adult to PR-T- 13 with negative results, snail 832 to PR-T-13 with positive results, and snail 834 to PR-2 with positive results. Each of these snails by selfing produced lines giving consistent sus- ceptibility results: lines 831 and 834 were PR-2 susceptible and PR-T-13 nonsuscep- tible, while line 832 was PR-2 nonsusceptible and PR-T-13 susceptible. DISCUSSION The results presented suggest that the biphallic condition is inherited. The first snail in which this was observed, 831, involved snail lines that had been followed as individ- uals reared in isolation and reproducing by selfing through many generations, except when controlled crosses were made. The biphallic condition had never been observed before in our snail stocks. A mutation is thus suggested. Three sibling snails (831, 832, 834) all produced progenies including biphal- lic snails. This could result either from a mutation in their parent snail (83), or in inter- breeding of the three snails before isolation. In studies on size of B. glabrata at maturity (Richards & Merritt, 1975) onset of egg-laying had been observed at snail diameters as small as 5 mm. Snail stocks involved т studies reported here were relatively small. If snails 831, 832, and 834 mated before isola- tion, a mutation in 831 might explain the results. 5. mansoni susceptibility results, however, suggested that the three snails had not interbred, since each gave rise to a snail line showing a consistent susceptibility pat- tern. Snails 831 and 832 differed in suscepti- bility to PR-T-13 and gave rise to lines show- ing reverse susceptibilities: 831 susceptible to PR-2, nonsusceptible to PR-T-13; 832 sus- ceptible to PR-T-13, nonsusceptible to PR-2. From these results it is concluded that a mutation for the biphallic condition occurred in the parent snail (83). The descendents of biphallic snail 831 were followed for several generations by iso- lating biphallic individuals and allowing them to reproduce by self-fertilization. Although fre- quencies of the biphallic condition increased from 17.9% to 28.4%, they continued to vary unpredictably. When biphallic snail 831 was mated with normal stock snails, some hybrid F;s of the normal parents were biphallic, suggesting in- volvement of a dominant factor with low penetrance. When descendents of the first cross were followed through several genera- tions, with selection, several lines were de- rived showing consistently high biphallic fre- quencies. One line had a stable frequency of 100% biphallic individuals indicating that the low penetrance may be improved by selec- tion. The genetics of the biphallic condition are not simple. The biphallic trait did not show an association with susceptibility to the two strains of 5. mansoni tested, and variation in expression of the trait makes it a relatively poor genetic marker. The biphallic condition may be regulated by a dominant factor with low penetrance which can be improved by selection. This study contributes to our under- standing of the genetics of Biomphalaria. However additional genetic studies will be required in order to investigate rationally the interactions between Biomphalaria and its schistosome parasites. ACKNOWLEDGMENTS The technical assistance of Mr. Paul C. Shade is gratefully acknowledged. These studies were funded in part by Office of Naval Research Contract N1.N00014-78-C-0081. LITERATURE CITED FLETCHER, M., LoVERDE, Р.Т. & RICHARDS, C.S., 1981, Schistosoma mansoni: electro- phoretic characterization of strains selected for different levels of infectivity to snails. Experimen- tal Parasitology, 52: 362-370. MINCHELLA, D.J. & LoVERDE, P.T., 1983, Labo- ratory comparison of the relative success of Biomphalaria glabrata stocks which are suscep- tible and insusceptible to infection with Schisto- soma mansoni. Parasitology, 86: 335-344. GENETICS OF BIPHALLIC BIOMPHALARIA 247 RICHARDS, C.S., 1974, Everted preputium and swollen tentacles in Biomphalaria glabrata: ge- netic studies. Journal of Invertebrate Pathology, 24: 159-164. RICHARDS, C.S., 1984, Influence of snail age on genetic Variations in susceptibility of Biomphalaria glabrata for infection with Schistosoma mansoni. 493-502. RICHARDS, C.S. & MERRITT, J.W., 1975, Varia- tion in size of Biomphalaria glabrata at maturity. Veliger, 17: 393-395. Malacologia, 25: Revised Ms. accepted 13 September 1984 И MALACOLOGIA, 1986, 27(2): 249-263 FACTORS REGULATING OVIPOSITION IN BULINUS TROPICUS IN SNAIL-CONDITIONED WATER М. Asghar Chaudhry 8 Elfed Morgan Department of Zoology and Comparative Physiology, University of Birmingham, P.O. Box 363, Birmingham, B15 2TT, United Kingdom ABSTRACT Ageing of the culture medium results in inhibition of oviposition in Bulinus tropicus, but the effect does not seem to be due to changes in the oxygen tension or ionic composition of the rearing water. Medium conditioned by closely related species also inhibits egg laying but that of taxonomically more remote snails has no such effect. Three week old snail-conditioned water retains its inhibitory property after dialysis but artificial media of similar inorganic ion concen- tration, with or without lettuce extract are not inhibitory. The results are consistent with the suggestion that some inhibitory compound is produced by the snails themselves but conspecific faecal homogenates do not have this inhibitory property. Bioassay of different fractions of the culture medium, separated by ultra-filtration, suggests that the inhibitory compound has a molecular weight of less than 500. Key words: Bulinus tropicus; oviposition; inhibition; closed cultures. INTRODUCTION A progressive decline in the oviposition of freshwater snails maintained in closed cul- tures, and its subsequent acceleration on replacement of the medium has been re- ported by various workers for a number of different snail species (Wright, 1960; van der Steen, 1967; Thomas, 1973; Kits & ter Maat, 1982). The onset of oviposition is often imme- diate and predictable, and it is likely that some change in the composition of the holding water is involved. Various factors have been implicated. For example, Mooij-Vogelaar & van der Steen (1973) have suggested that inhibited oviposition in fresh water pulmonate snails may be due to the depletion of dis- solved oxygen in unchanged culture water, a point of view shared by Kits & ter Maat (1982). Alternatively the depletion of calcium or the accumulation of excretory ammonia may limit the oviposition of Biomphalaria glabrata (Thomas, 1973; Thomas & Benjamin, 1974a, b and Thomas et al., 1974), and indirect evidence for the release of an oviposition- inhibiting substance appears frequently in the literature. For instance Wright (1960) showed that growth and oviposition in Bulinus forskalii was inhibited by conditioned water from a crowded conspecific culture but the inhibitory effect was lost if the same water was allowed to stand over activated charcoal for 24 hours. An inverse relationship between the densities of adult and young snails has also been found under field conditions in Lymnaea elodes (Eisenberg, 1966) and Wright (1960) sug- gested that under crowded conditions fecun- dity was inhibited by some chemical sub- stance released by snails. Essentially similar results have been reported by Berrie & Visser (1963) and Levy et al. (1973) and recent studies on a terrestrial pulmonate, Theba pisana have also revealed a negative feed- back effect on growth and oviposition consis- tent with the release of an inhibitory sub- stance under crowded conditions (Lazaridou- Dimitriadou & Daguzan, 1981). Berrie & Visser (1963) and Levy et al. (1973) were able to isolate inhibitory compounds from water containing crowded populations of Biomphalaria sudanica and Fossaria cuben- sis respectively, and Gazinelli et al. (1970) have suggested that snail faeces may be the source of a similar inhibitory substance in B. glabrata. However, as Thomas (1973) has pointed out, the snail-conditioned environment is a complex mixture of both organic and inor- ganic molecules originating from the snails themselves and from their decaying plant food and the source of such an inhibitory substance is difficult to predict. Indeed it may be argued that the case for the existence of an inhibitory compound per se is far from (249) 250 CHAUDHRY 8 MORGAN definitive (Thomas, 1973; Thomas, Lough & Lodge, 1975). Both promoting and inhibiting effects have been reported on the growth of undivided snails (Thomas, Goldsworthy & Aram, 1975; Thomas & Benjamin, 1974b) and bioassay of fractions separated by ultra- filtration techniques suggests that several growth factors may be involved (Thomas, Goldsworthy & Aram, 1975). The present work investigates the factors regulating oviposition in Bulinus tropicus (Krauss) maintained in different culture media in the laboratory, and the results of a prelim- inary attempt to isolate the inhibitor using ultrafiltration membranes are described. MATERIAL AND CULTURE METHODS Bulinus tropicus was selected from labora- tory stock in the Department of Zoology and Comparative Physiology at the University of Birmingham, derived from samples provided by Dr. C.A. Wright at the British Museum (Natural History). The snails were kept in shallow plastic trays (35 + 24 + 5 cm deep) containing fish-conditioned water, i.e. water taken from 15 litre tanks containing five juve- nile fishes of the species Cichlasoma nigro- fasciatum, the water in these holding tanks being replenished periodically with Birming- ham tap water, dechlorinated by standing. The snails were maintained under a LD 12:12 lighting regimen in a controlled environ- ment room at 24 + 1°C, the trays being covered with glass plates to reduce evapora- tion, and the animals fed daily on dried scalded lettuce given in excess. For experimental purposes the snails were transferred to enamelled pie dishes (24 + 14 + 5 cm deep) covered with perforated cling film, with 5 snails to 400 ml of fish-conditioned water unless otherwise stated. All experi- ments were carried out in the snail culture room where the animals had been kept for over 30 generations. Other experimental de- tails are indicated where relevant in the text. STUDIES ON THE BEHAVIOUR OF SNAILS IN UNCHANGED MEDIUM The locomotor activity, feeding and egg laying behaviour of B. tropicus maintained in closed cultures all decrease with time in un- changed media. To investigate these changes in behaviour 15 snails were trans- ferred to 900 ml of fish-conditioned water in a small perspex tank and their feeding and locomotor activity recorded by direct observa- tion for a period of 8 min each morning, afternoon and evening for three weeks, using the scoring method described by Chaudhry 8 Morgan (1983). The relationship between egg laying and water change was investigated in another experiment in which the oviposition rate of snails kept in an unchanged medium was compared with that of those for which the water was changed weekly. Six replicate dishes (5 snails per 400 ml) were divided into two equal groups and the numbers of eggs produced recorded over a five week period. In one group the water was changed just once, at the end of the third week of the experiment, while that of the other, control, snails was replaced each week with fish-conditioned water. Results and Discussion The mean numbers of animals feeding or actively crawling were higher during the first week of the experiment (Fig. 1) but both activities declined gradually during weeks two and three in unchanged cultures. By the end of this period nearly all the animals were found to be inactive on the sides or bottom of the dish. Control animals, for which the water was changed each week showed no such progressive decline in feeding behaviour and locomotor activity. Egg laying was similarly attenuated (Fig. 2). During the initial week of the experiment the rate of oviposition was the same in both experimental and control groups but where the water remained unchanged egg laying declined rapidly during the second and third weeks. Replacing the culture water resulted in a significant increase in fecundity which again declined when the water was unchanged. In contrast, changing the culture water each week during the five week period resulted in an almost constant rate of oviposition. These changes in behaviour are predict- able and presumably are precipitated by dif- ferences in some physical or chemical prop- erties of fish-conditioned water and the snail culture medium it replaces. Both water bodies had equilibrated to room temperature in the snail room where the animals were kept, and BULINUS OVIPOSITION REGULATION 251 uy ud = > U < oO =z FE < uw = 1 0 MEAN NO. FEEDING 12 18 DAYS FIG. 1. The locomotor and feeding activity (a and b respectively) of 15 B. tropicus snails recorded daily (Monday to Friday) for a period of three weeks in unchanged cultures. The points indicate mean values obtained from three experiments, and the vertical bars show the standard error values (98% confidence limits = 2 x S.E.). although the temperature was not measured during the experiments it is unlikely that changes in this parameter were responsible for the changes in behaviour recorded. The snails themselves will modify their me- dium in many ways. Oxygen and essential minerals will be removed from the water, while the products of metabolic waste will accumulate. Other compounds of organic and inorganic nature will be introduced into the medium in the food, and the results described above could equally be attributed to a deple- tion of resources or to the accumulation of some inhibitory factor in the medium. These possibilities have been investigated in a series of experiments in which changes in the rate of oviposition have been used as a bioassay. IS INHIBITION DUE TO REDUCED OXYGEN TENSION OR CHANGES IN INORGANIC ION CONCENTRATIONS? The possibility that the results described above (Fig. 2) could be due to changes in oxygen tension was investigated in an exper- iment in which 30 actively laying snails were divided into two groups in two lots of three dishes each containing five snails. In all ex- perimental dishes the water remained un- changed during the first three weeks, after which time it was replaced with fish- conditioned water, but in one group of three it was aerated by bubbling with com- pressed air throughout the experiment. The oxygen level of the culture water was deter- 252 CHAUDHRY 8 MORGAN 50 Ha 40 30 20 NO- EGGS/ SNAIL/ WEEK WEEKS FIG. 2. The oviposition of two groups of 15 B. tropicus snails. In (a) the water was replaced completely at the beginning of the 4th week as indicated by the arrow, while in (b) the water was changed weekly. Figures at the column heads indicate the total number of egg masses per snail, and the vertical bars indicate the standard error values derived from three experiments. mined weekly, 2 ml water samples being taken from each dish and analysed with a Radiometer (PHM71) electrode calibrated with water of known PO, at 24°С. The number of egg masses produced in each dish was recorded daily. In another series of experiments the possi- ble influence of changes in inorganic ions was investigated. The ionic composition of the culture media was ascertained by analysing 15 ml aliquots removed from the experimental dishes. The water sample was replaced with an equal volume of medium taken from an equivalent dish of similar age and snail den- sity. Preliminary analysis using conventional spectrophotometric and volumetric tech- niques (see A.O.A.C. handbook, 1975) showed the main cations present in snail- conditioned water to be Na*, K*, Mg**, Ca‘ *,and NH2*, other elemental ions being present only in trace quantities, while the major anions were found to be CI , NO; and HCO; . Changes in the concentration of one or more of these ions could, in theory, cause a decrease in the oviposition rate of B. tropicus, and this possibility has been inves- tigated in three different series of experi- ments. In the first of these three replicate dishes were set up, each containing 5 reproductive B. tropicus snails in 400 ml of fish-conditioned water, and the pH, ammonia and mineral content of the water were recorded at the beginning of the experiment, and each week BULINUS OVIPOSITION REGULATION 253 over a period of three weeks, the ammonia concentration being determined by the A.O.A.C. (1975) method while the metallic cations were measured by atomic absorption spectrophotometry. The pH was measured using a Radiometer (Copenhagen) pH meter. No attempt was made to monitor changes in anion content during the conditioning pe- riod. Instead the possible involvement of free anions in the regulation of oviposition was investigated in two series of experiments in- volving the bioassay of artificially prepared media, and of dialysed snail-conditioned wa- ter. п the first of these, artificial media were prepared by dissolving known weights of NaCl, MgO, Ca(NO3)> and NH4NO3 in deionised water to produce the ionic equiva- lent of fish-conditioned water or of three week old snail-conditioned water. A sample of dried, scalded lettuce was macerated and added to 400 ml of the latter medium which was kept for three weeks at 27°С to allow for the possible introduction of other ions into the culture through the decay of uneaten food. A further sample of lettuce infusion was made up using fish-conditioned water. Five groups of snails were set up, each with three dishes of 5 snails in 400 т! of fish-conditioned water. The water of all snails was replaced with new fish-conditioned water at the end of each of the first two weeks of the experiments. At the end of the third week the water in two groups of snails was replaced with artificial three week old snail conditioned water or with fish- conditioned water, in both of which partially homogenized lettuce had been allowed to stand for 3 weeks. Two further groups were transferred to artificial media containing the ionic equivalents of three week old snail- conditioned water alone, or of unused fish- conditioned water. Between weeks four and five the water of these four groups remained unchanged. The fifth group served as a con- trol and its medium was replaced with natural fish-conditioned water at the end of each week throughout the experiment. Oviposition was noted daily for each group. In the third series of experiments, 15 ma- ture snails were maintained in 900 ml of fish-conditioned water which was circulated (20 ml min *) through a dialysis tube т- mersed in a bath of fish-conditioned water. This was changed daily. Two similar popula- tions served as controls. In one of these the water was replaced weekly while in the other it remained unchanged throughout the exper- iment. Each group consisted of three repli- cates, and the concentrations of the major inorganic cations and the pH of the dialysed culture medium were monitored weekly to confirm the efficacy of the dialysis. Results and Discussion In the present investigation (Fig. 3a) the level of oxygen in the undisturbed medium was considerably reduced during the three week conditioning period. Nevertheless the depletion of oxygen per se does not seem to have limited the oviposition of B. tropicus in the above experiments. Egg laying was sim- ilarly inhibited in the parallel group of snails in which the oxygen tension of the culture water was maintained at a constant level, or even increased slightly by bubbling with com- pressed air (Fig. 3b). Replacing the culture water at the end of the third week resulted in a marked increase in oviposition rate in both groups. The ionic composition of the undisturbed media altered gradually during the condition- ing period (Table 1). The concentrations of Na’, K*, MG** and NH,* increased pro- gressively, presumably introduced into the system in food, or as the byproducts of me- tabolism, while the Ca* * content decreased significantly (P < 0.001, Anova) from 0.43 to 0.14 m mol 1 * over the three weeks. Similarly the pH of the snail-conditioned water decreased significantly during this period (Р < 0.01). It is unlikely that changes in these cations are responsible for the observed de- crease in oviposition. Transfer to artificial environments equivalent to three week old snail-conditioned water with or without lettuce extract did not inhibit oviposition in actively laying В. tropicus (Fig. 4), and the laying rates in these media were comparable to those in which the water was replaced at the end of week 3 with fish-conditioned water or an artificial medium of similar ionic composition. The decrease in oviposition seen during the fifth week of this experiment is clearly differ- ent from the almost complete inhibition which follows transfer to snail-conditioned water (e.g. compare Fig. 4 with Fig. 2a, Fig. 3c, d) and probably reflects a depletion of physio- logical resources. Furthermore, the oviposition of snails kept in dialysed culture media showed, over the first three weeks of the experiments, an ex- ponential decrease characteristic of snails kept in closed culture (Fig. 5). Changing the 254 CHAUDHRY 8 MORGAN NO. EGGS /SNAIL / WEEK WEEKS FIG. 3. Oxygen levels (a and b) and the oviposition rate (с and а) of В. tropicus cultures in non-aerated and aerated water respectively. The arrows indicate the time of replacement of both culture media with fish-conditioned water. Otherwise legend as for Fig. 2. TABLE 1. lonic concentration of unchanged В. tropicus culture media (5 snails per 400 ml of fish-conditioned water), measured at the start of the experiment, and at weekly intervals thereafter. Week М’ (m. mol/l) К’ (m. mol/) Са’ * (m. том) Mg* * (m. mol/l) NH, (ppm) pH 0 0:3. = 0:02 0.045 + 0.07 0.44 = 0.03 0.07 + 0.01 0109 = 0106) 73 1ЕЕ 005 1 0.48 + 0.02 0.47 = 0.14 0.40 = 0.06 ОИ == 0102 2.01 = 027777203€70:05 2 0.426 + 0.60 0.60 + 0.04 0.32 + 0.08 0.16 + 0.03 3:48 = 01367 706015 3 0.79. 320510 35 0130 0.14 = 0.09 0.19 + 0.04 4.14 + 0.27 6.8 = 0.20 water of the dialysed group at the end of the third week increased their oviposition rate, while control snails for which the culture water was replaced weekly remained equally fe- cund throughout. Unlike unchanged culture media the concentration of the major cations and the pH of the dialysed water remained constant during the experiment (Table 2). The anion composition of the water will also change during the conditioning of unchanged culture media but again the inhibition of egg laying cannot be attributable to these changes, at least in so far as the major inorganic anions are concerned. Like the cat- ions the anions equilibrate when dialysed against fish-conditioned water, but as may be seen from Fig. 5, oviposition still declined under these conditions. On the other hand transfer to artificial media enriched with chlo- ride and nitrate did not impair egg laying (Fig. 4e, d). It would appear therefore that inhibition is effected by the accumulation of some other substance, possibly of organic nature, in the culture medium. As plant food extract was found to be without effect it is likely that this compound is produced by the snails them- selves. BULINUS OVIPOSITION REGULATION 299 NO. EGGS / SNAIL / WEEK 30 20 0 ao) WEEKS FIG. 4. The oviposition rate of five groups, each of 15 B. tropicus snails in different culture regimens. Fig. 4a shows the results obtained with a control group of snails for which the water was replaced with new fish-conditioned water each week throughout the experiment. Other groups were transferred at the end of week three to artificial media equivalent to three week old snail-conditioned water, with or without lettuce extract (c and d respectively), or to artificial fish-conditioned water similarly lacking (b) or containing (e) lettuce extract. The times of transfer of the snails to the different experimental media are indicated by the arrows. Otherwise legend as for Fig. 2. SPECIES SPECIFICITY OF CONDITIONED MEDIA The specificity of inhibition with regard to egg laying in B. tropicus was investigated using water conditioned by mature specimens of three different fresh-water pulmonates, Bulinus globosus, Biomphalaria glabrata, Physa sp. and by the prosobranch Mela- noides tuberculata, each species being 256 CHAUDHRY 8 MORGAN 50 | 40 30 20 10 NO. EGGS / SNAIL/ WEEK 30 20 10 ¡A E eta WEEKS FIG. 5. The oviposition rate of 4 groups, each of 15 B. tropicus snails, after transfer to media conditioned by heterospecific snails. In each group the water was replaced weekly with fish-conditioned water for the first three weeks, after which the water was replaced with media conditioned by Bulinus globosus, Physa sp., Biomphalaria glabrata and Melanoides tuberculata (a to d respectively). The time of transfer is indicated by the arrows. Otherwise legend as for Fig. 2. BULINUS OVIPOSITION REGULATION 257 TABLE 2. lonic concentrations of dialysed culture media of B. tropicus (15 snails/900 ml of fish-conditioned water) recorded at the start of the experiment and at weekly intervals thereafter. Week Ма’ (т. mol/l) K* (m. mol/l) Са‘ * (m. том) Mg** (т. том) NH¿(ppm) pH 0 0.3 0.04 0.44 0.08 0.09 6.9 1 0.24 0.22 0.45 0.10 0.25 6.7 2 0.32 0.20 0.42 0.10 0.29 6.8 3 0.25 0.23 0.41 0.10 0.31 6.5 maintained at a density of five snails in 400 ml of water for a period of three weeks. Four groups of B. tropicus, selected randomly from laboratory stock were transferred, five to a dish containing 400 ml of fish-conditioned water, and their media replaced weekly for the first three weeks. At the beginning of the fourth week each group was transferred to one of the heterospecifically conditioned me- dia described above, the oviposition rates of each group being recorded weekly. All snails were fed dried, scalded lettuce ad libitum during the conditioning and experimental pe- riods, and each experiment was replicated three times. Results and Discussion The inhibitory properties of heterospe- cifically-conditioned media appear to be re- lated to the taxonomic affinities of the species involved. Media conditioned by Bulinus globosus and Physa sp. significantly reduced the oviposition rate of B. tropicus (Fig. 6), the latter being rather less effective (Anova, P < 0.01 and < 0.05 respectively). Media conditioned by Biomphalaria glabrata lowered the oviposition rate of B. tropicus but not significantly so, and water conditioned by Melanoides tuberculata also failed to inhibit. Inhibitory secretions have been postulated to account for the success of Helisoma duryi in competition with other helminthologically im- portant snails (see references in Madsen, 1982), and for other fresh-water pulmonates (e.g. Levy et al., 1973) and the results ob- tained here support this hypothesis. Madsen (1979a, b), investigating the competition be- tween Helisoma duryi and Biomphalaria sp. in the laboratory, found many unhatched egg masses in older aquaria, but as the effect showed no species specificity, he concluded that the factors responsible may have origi- nated from the food or metabolic waste. He subsequently attributed the success of Helisoma to direct interaction (Madsen, 1982), but did observe inhibitory effects in newly established aquaria with low densities of adult snails, conditions which may have approximated to those of the present experi- ments. FRACTIONATION OF SNAIL-CONDITIONED MEDIUM An attempt was made to separate the in- hibitory fraction by filtration and ultra-filtration techniques. In a preliminary experiment three groups of actively laying snails were main- tained, 5 per 400 ml of fish-conditioned water for three weeks, and the medium was changed weekly. The faeces were then re- moved by filtration, homogenized and made up to 400 ml with fish-conditioned water for bioassay. Three week snail-conditioned water from which the faeces had been filtered was similarly bioassayed while in another experi- ment 3 further groups of 5 snails were kept in unchanged culture for 5 weeks and the me- dium bubbled over activated charcoal for 5 min at the end of each week when oviposition was recorded. In a further series of experiments three- week conditioned medium was concentrated by evaporation at room temperature and fil- tered through Whatman No. 1 filter paper, and Gelman metrical autoclave and Millipore 0.22 um membranes to remove faecal material and bacteria respectively. Further fractionation was achieved by ultrafiltration using Millipore Pellicon (PT se- ries) and Amicon Diaflo membranes in mag- netically stirred cells pressurized with nitro- gen to 10 p.s.i. The active fraction was subjected to high pressure liquid chromatography (HPLC) when one litre of three-week snail- conditioned water was concentrated by freeze-drying after the initial filtration. The 258 CHAUDHRY 8 MORGAN NO. EGGS /SNAIL / WEEK WEEKS FIG. 6. The oviposition rate of B. tropicus in unchanged culture medium (a), in dialysed medium replaced with fish-conditioned water when indicated by the arrow (b), and replaced weekly with fish-conditioned water (c). Otherwise legend as for Fig. 2. residue was dissolved in 25 ml of deionized water before filtering through an Amicon UM- 05 membrane with a 500 MW cut off, the filtrate being again freeze-dried and redis- solved in 4 ml of distilled water to obtain a stock solution. Part of this solution was diluted with fish-conditioned water to obtain a stock solution and the remainder subjected to HPLC analysis. The chromatography column (25 + 4.5 mm) was packed with ‘Spherisorb 50 DS’ and the sample spun at 1.5 ml min ! at 3000 p.s.i. and at room temperature. The eluents were scanned with a uv detector (Spectromonitor Ill) at 210 nm and the peaks recorded at 500 mm/h. Polar and non-polar fractions were eluted with distilled water and 30% methyl cyanide (acetonitrile) respec- tively and the two fractions bioassayed as before. The solvent in the latter fraction was rotary-evaporated at 34°С and freeze-dried prior to assay. Results and Discussion Faecal homogenates are without effect on the oviposition of B. tropicus (Fig. 7). Faecal homogenates have been shown to both ac- celerate and retard the growth of Biom- phalaria glabrata, the effect being determined by the diet of the donor snail (Thomas, Lough 8 Lodge, 1975), and Gazinelli et al. (1970) have reported an active component in the faeces of this species which inhibited the uptake of °ЗЕе, and presumably growth. Oviposition in B. glabrata however, is unaf- fected by faecal material (Thomas, Lough 8 Lodge, 1975). The oviposition rate of snails transferred to faecal homogenate 1$ not significantly lower than in the first three weeks of the experiment when the water was replaced weekly with fish-conditioned water alone. In contrast the oviposition of snails transferred to three week BULINUS OVIPOSITION REGULATION 259 > 5 NO o NO. EGGS / SNAIL / WEEK WEEKS FIG. 7. The oviposition rate of B. tropicus following transfer of snails to experimental media containing faecal homogenates of conspecific snails (a) and three week conspecific conditioned water (b). The water was replaced with fish-conditioned water at the end of each week for the first two weeks of each experiment and with the experimental medium at the end of week three, as indicated by the arrows. In Fig. 7c the water remained unchanged but was filtered each week through a column of activated charcoal after bubbling with compressed air. Otherwise legend as for Fig. 2. old snail-conditioned water from which the faeces had been removed 15 considerably reduced. However, the inhibitor appears to be removed from the medium by passing over activated charcoal (Fig. 7c). Snails kept in medium bubbled each week over activated charcoal show a slight decrease in weekly oviposition rate, but this is not statistically significant (Anova P > 0.5) and the rapid attenuation of egg laying during the second and third weeks, which is characteristic of snails kept in unchanged media, is not evi- dent. Similar results have been obtained by Wright (1960) with В. forskalii. Bioassays performed on media fraction- ated by ultrafiltration indicate that the inhibitor is of low molecular weight. The patterns of oviposition in groups of five actively laying assay snails on being transferred to the vari- ous experimental media are seen in Fig. 8, together with the results of a further assay performed on snail-conditioned water filtered through a preliminary filter of 0.22 um роге diameter. The oviposition of assay snails was inhib- ited by each filtrate fraction down to and including that obtained with membranes hav- ing a5 x 10% MW cut off, whereas the resuspended retentates of these membranes were ineffective. On subfractionation of the 5 x 10? MW filtrate by HPLC analysis, two major peaks become evident (Fig. 9) representing the po- lar and non-polar fractions respectively. Each peak is a compound one, being made up of a number of sub-fractions, indicating the pres- ence of a number of different types of mole- cule, but detailed interpretation of the second peak is complicated by the presence of the methyl cyanide elutant which forms the pre- dominant component. The results of a preliminary bioassay of the polar and non-polar fractions were inconclu- sive. The inhibitory potency of the whole 5 x 260 CHAUDHRY 8 MORGAN 55 LS 35 25 un o > un МО: EGGS /SNAIL / WEEKS 4 urn uy un 55 | 45 35 25 15 WEEKS FIG. 8. The oviposition rate of B. tropicus after transfer to water containing different fractions of three week old snail-conditioned water, separated by ultra-filtration. Histograms a to d show the results obtained with 0.22 um filtrate; 100,000 MW retentate; 10,000 MW retentate, and 10,000 MW filtrate respectively. The results of similar experiments involving the 5,000 M retentate; 1,000 MW retentate; 500 MW retentate and 500 MW filtrate are shown in e to h respectively. Otherwise legend as for Fig. 2. 10° MW fraction was unaffected by the freeze- drying process, and the polar fraction of the 5 x 10° MW filtrate again appeared to have retained its inhibitory potency. The non-polar fraction however was found to be toxic to the assaying animals, all of which died during the first 24 hours after transfer to the test medium. GENERAL DISCUSSION The inhibitory effects of snail-conditioned media on the oviposition of B. tropicus de- scribed here are similar to those reported by previous authors for other fresh-water pulmonate species (Chernin & Michelson, BULINUS OVIPOSITION REGULATION 261 FIG. 9. HPLC spectrum of the 500 MW filtrate fraction. Two consecutive runs are shown, and the arrows indicate where the polar and non-polar fractions were separated prior to bioassay. 1957a, b; Wright, 1960; Berrie 8 Visser, 1963; Gazinelli et al, 1970; Levy et al, 1973; Madsen, 1979a, b). Although the ionic com- position of the medium altered gradually dur- ing the conditioning period it is unlikely that changes of an inorganic nature are responsi- ble. Egg laying was inhibited even when the oxygen tension was held constant (Fig. 3), and the transfer of snails to artificial media containing the major inorganic ions in the concentrations found in three week old snail conditioned water had no observed inhibitory effect (Fig. 4). Trace elements and the major anions were not measured routinely, but the specific nature of the inhibition (Fig. 6) and the inhibitory effects of dialysed media are difficult to explain in terms of changes in these factors. Instead a compound of organic na- ture is implicated, and it seems likely that this may be produced by the snails themselves as fish-conditioned water containing lettuce alone (Fig. 4) had no inhibitory effect. Similar inhibitory factors have been reported for ter- restrial pulmonates (e.g. Cameron & Carter, 1979; Dan & Bailey, 1982). Where snail pheromones or growth- modulating compounds have been isolated from snail-conditioned media they have been found to be made up of relatively small mol- ecules and the inhibitor of egg laying in B. tropicus appears to be typical in this respect. In Biomphalaria sudanica an inhibitor of so- matic growth has been identified as a dimethyl ester of formula C,gH3207, with a molecular weight of 360 (Berrie & Visser, 1963), whereas the alarm pheromones pro- duced by the opisthobranch Navanax have been shown to be methyl ketones of even smaller molecular weights (Sleeper & Fenical, 1977; Sleeper, Paul & Fenical, 1980). The growth-promoting substances present in B. glabrata-conditioned water are similarly small molecules, within the 5 x 10°-10* MW range, and Thomas, Goldsworthy & Aram (1975) have suggested that polypeptides and/ or glyco or lipo proteins may be implicated. Тре inhibitory activity recorded with В. tropicus-conditioned water after passing through а 5 x 10° ultrafiltration membrane indicates that the size of the inhibitor mole- cule in the present experiments may be of the same order of magnitude and as such might be expected to pass through the viscin dialy- sis membrane used in some of the above experiments. Snail-conditioned water thus dialysed retains its inhibitory properties how- ever, and there are a number of possible explanations for this apparent paradox. In the dialysis experiments the media were unpres- surized and it is conceivable that pressures comparable to those used in ultrafiltration would be required to force the larger organic molecules through the viscin tubing. An alter- native possibility is that the inhibitor may be present in a charged form, and is retained within the dialysis membrane by the osmotic gradient and indeed HPLC analysis of the 500 MW filtrate indicates that the inhibitor is present in the polarized fraction. As in Navanax, the crude extract contains more than one compound (Sleeper & Fenical, 1977; Sleeper, Paul & Fenical, 1980) but the relative potency of the different subfractions has yet to be investigated for B. tropicus. 262 CHAUDHRY 8 MORGAN REFERENCES CITED A.O.A.C., 1975, Official Methods of Analysis of the Association of Official Analytical Chemists. Ed. 12, 1904 p. Washington, D.C. BERRIE, А.О. & VISSER, S.A., 1963, Investigation of growth-inhibiting substance affecting a natural population of freshwater snails. Physiological Zoology, 36: 167-173. CAMERON, R.A.D. 8 CARTER, M.A., 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. CHAUDHRY, М.А. & MORGAN, E., 1983, Circad- ian variation in the behaviour and physiology of Bulinus tropicus (Gastropoda: Pulmonata). Ca- nadian Journal of Zoology, 61: 909-914. CHERNIN, E. 8 MICHELSON, E.H., 1957a, Stud- ies on the biological control of Schistosoma- bearing snails. Ш. The effects of population density on growth and fecundity in Australorbis glabratus. American Journal of Hygiene, 65: 57-70. CHERNIN, E. & MICHELSON, E.H., 1957b, Stud- ies on the biological control of Schistosoma- bearing snails. IV. Further observations on the effect of crowding on growth and fecundity of Australorbis glabratus. American Journal of Hy- giene, 65: 71-80. DAN, N.A. & BAILEY, S.E.R., 1982, Growth, mor- tality and feeding rates of the snail Helix aspersa at different population densities in the laboratory, and the depression of activity of helicid snails by other individuals or their mucus. Journal of Mol- luscan Studies, 48: 257-265. EISENBERG, R.M., 1966, The regulation of density in a natural population of the pond snail, Lymnaea elodes. Ecology, 47: 889-906. GAZINELLI, G., ROMALHO-PINOT, F.J., PELLE- GRINO, J. 8 GILBERT, G., 1970, Uptake of °°Fe as a tool for study of the crowding effect in Biomphalaria glabrata. American Journal of Tropical Medicine and Hygiene, 19: 1034—1037. KITS, K.S. & MAAT, A. ter, 1982, Neurophysiology of the peptidergic caudo-dorsal cells in Lymnaea stagnalis. In: Proceedings of the International Minisymposium on Molluscan Neuro- endocrinology. Free University Amsterdam. Netherlands, August 16-20, р. 60-68. LAZARIDOU-DIMITRIADOU, М. & DAGUZAN, J., 1981, Effects of crowding on growth, mortality rate and reproduction of Треба pisana (Gastropoda: Рштопаа). Malacologia, 20: 195-204. LEVY, M.G., TUNIS, М. & ISSERHOFF, H., 1973, Population control in snails by natural inhibitors. Nature, 241: 65-66. MADSEN, H., 1979a, Further laboratory studies on the interspecific competition between Helisoma duryi (Wetherby) and the intermediate hosts of Schistosoma mansoni Sambon: Biomphalaria alexandrina (Ehrenberg) and B. camerunensis (Boettger). Hydrobiologia, 66: 181-192. MADSEN, H., 1979b, Preliminary observations on the role of conditioning and mechanical interfer- ence with egg masses and juveniles in the competitive relationships between Helisoma duryi (Wetherby) and the intermediate host of Schistosoma mansoni Sambon: Biomphalaria camerunesis (Boettger). Hydrobiologia, 67: 207-214. MADSEN, H., 1982, Development of egg masses and growth of newly hatched snails of some species of intermediate hosts of Schistosomiasis in water conditioned by Helisoma duryi (Wetherby) (Pulmonata: Planorbidae). Mala- cologia, 22: 427—434. MOOIJ-VOGELAAR, J.W. 8 STEEN, W.J. van der, 1973, Effects of density on feeding and growth in pond snail Lymnaea stagnalis (L.). Proceedings Koninklijk Nederlandse Akademie van Wetenschappen, ser. C, 76: 47-60. SLEEPER, H.L. & FENICAL, W., 1977, Navenones A-C; trail-breaking alarm pheromones from the marine opisthobranch Navanax inermis. Journal of the American Chemical Society, 99: 2367-2368. SLEEPER, H.L., PAUL, V.J. & FENICAL, W., 1980, Alarm pheromones from the marine opistho- branch Navanax inermis. Journal of Chemical Ecology, 6: 57-70. STEEN, W.J. van der, 1967, The influence of environmental factors on the oviposition of Lymnaea stagnalis (L.) under the laboratory con- ditions. Archives Neerlandaise Zoologie, 17: 403-468. THOMAS, J.D., 1973, Schistosomiasis and the control of molluscan hosts of human schistosomes with particular reference to self- regulatory mechanisms. In DAWES, B., ed., Ad- vances in Parasitology. Academic Press, Lon- don, 11: 307-394. THOMAS, J.D. & BENJAMIN, M., 1974a, The ef- fects of population density on growth and repro- duction of Biomphalaria glabrata (Say). Journal of Animal Ecology, 43: 31-50. THOMAS, J.D. & BENJAMIN, M., 1974b, Effects of numbers, biomass and conditioning time on growth and natality rates of Biomphalaria glabrata (Say), the snail host of Schistosoma mansoni Sambon. Journal of Applied Ecology, 11: 832-840. THOMAS, J.D., BENJAMIN, M., LOUGH, A. & ARAM, R.H., 1974, The effect of calcium in the external environment on the growth and natality rates of Biomphalaria glabrata (Say). Journal of Animal Ecology, 43: 839-860. THOMAS, J.D., GOLDSWORTHY, G.J. & ARAM, R.H., 1975, Studies on the chemical ecology of snails. The effect of chemical conditioning by adult snails on the growth of juvenile snails. Journal of Animal Ecology, 44: 1-27. THOMAS, J.D., LOUGH, AS. & LODGE, R.W., 1975, The chemical ecology of Biomphalaria BULINUS OVIPOSITION REGULATION 263 glabrata (Say), the snail host of Schistosoma WRIGHT, C.A., 1960, The crowding phenomenon mansoni Sambon. The search for factors in in laboratory colonies of freshwater snails. An- media conditioned Бу snails which inhibit their nals of Tropical Medicine and Parasitology, 54: growth and reproduction. Journal of Applied 224-232. Ecology, 12: 421-436. Revised Ms. accepted 14 March 1985 | y iy y | | i | ; | i i | i o | i i di nu m iy i i | | i | | | | | | й TL i | | | i | it ia | i Mr i iy Mi m | | ik o | u | . Mm iy N Ta , 7 0 en NT | ni ul, 1L | | | - . 1 И у i L | el no A o | i | | i o un y ui i | | i | ds In i | MALACOLOGIA, 1986, 27(2): 265-269 EFFECTS OF LONG-TERM EXPOSURE TO LOW CONCENTRATIONS OF MOLLUSCICIDES ON A FRESH-WATER SNAIL, INDOPLANORBIS EXUSTUS, A VECTOR OF SCHISTOSOMIASIS B.D. Parashar & K.M. Rao Department of Entomology, Defence Research & Development Establishment, Tansen Road, Gwalior 474002, India. ABSTRACT This paper deals with the long-term effects of low concentrations (0.01, 0.05, 0.10, and 0.15 mg/l) of four molluscicides, namely Santobrite, copper sulphate, Yurimin, and Bayluscide on the fresh-water snail /ndoplanorbis exustus, vector of schistosomiasis. The results show that Santobrite is less effective at 0.15 mg/l, and ineffective at the other concentrations against immature, young mature, and adult stages of the snail. Yurimin is effective against immature and adult /. exustus at 0.05, 0.10 and 0.15 mg/l, and against the young mature stage at 0.15 mg/l. Copper sulphate is effective at all four concentrations against immature and adult /. exustus and at concentrations of 0.05, 0.10 and 0.15 mg/l against the young mature stage. Bayluscide is highly effective at all concentrations against all snail stages. Use of low concentrations of molluscicides for snail control, keeping in view their toxicity to non-target organisms, is discussed. INTRODUCTION Indoplanorbis exustus (Deshayes) is a fresh-water planorbid gastropod, the vector of Schistosoma indicum, S. nasale, and S. spindale, the causative agents of schistoso- miasis among horses, mules, sheep, goats, camels and cattle of economic and agricul- tural significance (Malek & Cheng, 1974). Synthetic molluscicides are used as one of the major components of integrated pest management for the control of fresh-water snails. Two methods of application of mollusci- cides are currently in use for snail control, i.e. use of high concentrations for the short-term, and low concentrations for long-term. If con- tinuous molluscicidal treatment of snail habi- tats can be satisfactorily implemented by use of low concentrations for long periods, multi- ple attack points in the trematode life cycle against both parasite and snail could be achieved. Whereas, short-term applications of high concentrations of molluscicides kill the existing snails with immediate lethal effects on biota of the environment, prolonged appli- cations of molluscicides at low concentrations would cause minimal ecological disturbance in addition to snail control. The present study has been undertaken so as to work out the susceptibility of immature, young mature and adult stages of the snail /. exustus to low concentrations of four mollus- cicides. This study has an important bearing on the control of snails by working out effec- tive concentrations of molluscicides that may be needed for the control of this species in nature. MATERIALS AND METHODS Snails were drawn from standard labora- tory cultures of I. exustus. At the start of the experiment, they were grouped into the fol- lowing categories: 1. Immature 3—6 mm diameter, 2. Young mature 9-12 mm diameter, 3. Adult Above 13 mm diameter. Four chemicals, namely Santobrite (sodium pentachloro-phenate), copper sulphate, Yuri- min (3,5-dibromo4-hydroxy-nitrobenzene) and Bayluscide (2-aminoethanol salt of 2',5- dichloro-nitrosalicylanilide) were chosen for the evaluation of their toxicity at low concen- trations against three stages of snails. The low concentrations employed for study were 0.01, 0.05, 0.10 and 0.15 mgflitre. For each concentration of molluscicide, 20 (265) 266 PARASHAR 8 RAO snails were exposed in a glass container (28 cm diameter) containing 5 litres of dechlorin- ated tap water for 24 hr. After this period, snails were transferred to another glass con- tainer having the same concentration of mol- luscicide in a similar amount of dechlorinated tap water. Five replications were employed so as to minimize variation in mortality rates of snails. For each stage of snail, different ex- perimental sets were conducted. Dead snails were removed from the container so as to prevent fouling of the environment. Spinach leaves were supplied as food for these snails ad libitum. This experiment was continued till a last snail survived. Controls were also kept under similar conditions. All these studies were carried out in a controlled environment room (temperature 30 + 2°С; В.Н. 70% + 5). The lighting regimen maintained during ex- perimentation was equal periods of light and dark (LD 12:12) in a 24 hr cycle. The data on exposure period and resultant percent mortality were subjected to probit analysis for the determination of LD (dura- tion for 90% mortality of snails) as per method of Finney (1971). In the present study, the concentration which brought about 50% or more reduction in the LD% value of experi- mental snails as compared to controls was considered an effective concentration. RESULTS The mean survival time of immature, young mature and adult stages of /. exustus at 0.01, 0.05, 0.10 and 0.15 mg/l concentrations of Santobrite, copper sulphate, Yurimin and Bayluscide are illustrated in Figs. 1, 2 and 3, respectively. The LD values for control snails of imma- ture, young mature, and adult categories were 188.9, 150.1 and 101.2 days, respec- tively. For each concentration of mollusci- cides, percent reduction in LD value of experimental snails as compared to control values was worked out. These values were shown as relative toxicity for various molluscicides in Table 1. Santobrite was not effective at all the con- centrations against three stages of the snail. Yurimin 15 effective against immature and adult /. exustus at 0.05, 0.10 and 0.15 mg/l concentrations, while against young mature snails only 0.15 mg/l concentration was effec- tive. Copper sulphate 1$ effective at all con- centrations ranging from 0.01 mg/I-0.15 mg/l against immature and adult stages, but in the case of young mature 0.05-0.15 mg/l concen- trations were effective. Bayluscide 1$ effective against all the stages of /. exustus at all concentrations used. Out of three stages of snail, the adult stage has been observed to be most susceptible and the young mature stage as the most resistant stage of the snail to molluscicides. It has been observed during experimenta- tion that none of the three stages of the snails tried to leave the container containing a molluscicidal solution. DISCUSSION The present study indicates the ineffective- ness of Santobrite at 0.01-0.15 mg/l and Yurimin at 0.01-0.10 mg/l concentrations for snail control. Copper sulphate appeared to be a promising compound at 0.05-0.15 mg/l concentrations against the most resistant stage of the snail, /.e. the young mature stage (Table 1, Figs. 1, 2, and 3), the Ооо being 44.5-60.0 days. Walker & Cardarelli (1975) estimated the LD, for Biomphalaria glabrata between 0.025—0.030 ppm/day when the ex- posure was extended for 60 days. Upatham & Christae (1975) noted that E-51, a slow re- lease formulation of copper sulphate at 1.25 ppm available copper, gave LT;5% of 19 days against St. Lucian strain of B. glabrata, while 0.63 ppm resulted ЁЕТ+оо of 11 days against Lymnaea cubensis. Cardarelli (1977) re- ported 58%, 94% and 100% mortality of B. glabrata in 60 days at 0.013, 0.065, and 0.13 ppm concentration of copper sulphate, respectively, with control mortality of 4%. The relative toxicity of Bayluscide was higher than other molluscicides (Table 1) against all stages of snails, as is also appar- ent from their mean survival time at different concentrations (Figs. 1, 2 and 3). Its lower value of LDgo (4.72-74.06 days) than the control (150 days) against the most resistant stage of snail shows its effectiveness at all concentrations (0.01-0.15 mg/l). Long-term exposure of B. glabrata to 0.05 ppm concen- tration of Bayluscide resulted in 32, 94, and 100% mortality in 30, 60, and 120 days, respectively, while exposure to 0.1 ppm ef- fected 62 and 100% mortality in 5 and 30 days, respectively (Cardarelli, 1977). This study reveals that out of four EFFECTS OF MOLLUSCICIDES ON INDOPLANORBIS 267 o = © © MEAN SURVIVAL TIME (DAYS) an о 0.01 0.05 040 0.15 1.0 CONCENTRATION (mg/L) FIG. 1. Mean survival time of immature stage of snail, Indoplanorbis exustus, at four concentrations of molluscicides, viz. Santobrite (0), Yurimin (e), copper sulphate (4) and Bayluscide (A). O © MEAN SURVIVAL TIME (DAYS) 0.01 005 010 045 1.0 CONCENTRATION (mg/l) FIG. 2. Mean survival time of young mature stage of snail, Indoplanorbis exustus, at four concentra- tions of molluscicides, viz. Santobrite (0), Yurimin (e), copper sulphate (A) and Bayluscide (A). al о JL MEAN SURVIVAL TIME (DAYS) 0.01 0.05 040 045 1.0 CONCENTRATION (mg/l) FIG. 3. Mean survival time of adult stage of snail, Indoplanorbis exustus, at four concentrations of molluscicides, viz. Santobrite (0), Yurimin (e), cop- per sulphate (A) and Bayluscide (4). molluscicides, Bayluscide proves to be the most toxic to all stages of /. exustus at low concentrations, being followed by copper sulphate and Yurimin. However, Santobrite is least toxic (Table 1, Figs. 1, 2 and 3). It has also been observed that out of three stages of the snail, the adult is the most susceptible and the young mature stage is most resistant. The less percent reduction in LDy values of the most resistant stage of /. exustus as com- pared to the controls, even at 0.15 mg/l of, Santobrite suggests that relatively higher con- centrations of this chemical would be required for achieving considerable effect on snail mor- tality. However, it may be harmful to other or- ganisms at higher concentrations. Besides, it has been reported to be strongly piscicidal (WHO, 1965) and herbicidal (WHO, 1980) in activity and potentially dangerous to opera- tors (Blair, 1961). Its acute oral toxicity against rats varies from 40-250 mg/kg (ЕОьо) (WHO, 1980). The effective low concentrations of Yurimin (0.10 and 0.15 mg/l) against the most resis- tant stage of /. exustus are enough lower to cause serious damage to the non-target biota as evident by its toxicity to animals and plants. The fish LC: varies from 0.16-0.83 mg/l. The acute oral mammalian toxicity of Yurimin is also low (LDs = 168 mg/kg) in mice). No apparent toxic hazards to human beings or vegetation were reported. It does not exert herbicidal activity (WHO, 1973). The low concentrations at which copper sulphate is effective against young mature snails (0.05-0.15 mg/l) are safer for non- target organisms. Copper sulphate is toxic to fishes but only somewhat at higher concen- trations (WHO, 1965). Its acute oral LD, for 268 PARASHAR 8 RAO TABLE 1. Relative toxicity of molluscicides to three stages of I. exustus based on percent reduction in LDgo value as compared to control. Concentration Molluscicide (mg/l) Santobrite 0.01 0.05 0.10 0.15 Yurimin 0.01 0.05 0.10 0.15 Copper sulphate 0.01 0.05 0.10 0.15 Bayluscide 0.01 0.05 0.10 0.15 Relative toxicity 3-6 mm 9-12 mm 13 + mm (dia.) (dia.) (dia.) 18.90 6.52 26.18 36.58 25.45 42.00 41.68 33.11 50.53 64.72 44.60 68.75 32.08 25.52 42.20 51.91 30.38 57.09 63.54 51.39 65.36 WOT 66.00 79.10 55.48 45.61 68.22 67.88 59.99 ИГ Ро - 76.39 62.17 85.56 81.46 70.33 88.84 59.89 50.66 83.39 92.78 88.85 95.31 96.34 94.30 97.36 97.96 96.86 98.80 rat is 300 mg/kg (WHO, 1980). It also exerts some herbicidal activity (WHO, 1980). Bayluscide has the advantage over other molluscicides of being extremely effective against all stages of snails at very low con- centrations (0.01-0.15 mg/l) and at the same time it is not toxic to humans and has limited biocidal effect. However, fish proved to be extremely susceptible to it (ЕСьо = 0.05-0.30 ppm) (Malek & Cheng, 1974). The oral LD: for rats has been reported to be over 5 g/kg of body weight (WHO, 1980). Human beings show no toxic symptoms at 30 mg/kg admin- istered as a single oral dose. Regarding phytotoxicity of Bayluscide, Abdallah & Nasr (1961) found that it is harmless to crops at dosages very much higher than those used for snail control. It also does not exert any herbicidal activity (Malek & Cheng, 1974). At higher concentrations snails detect ions of molluscicides and react in many instances by leaving the water, thus escaping the toxic effects of molluscicides (Etges, 1963). How- ever, this avoidance behaviour is not exhib- ited at low concentrations (Frick 8 De Jimenez, 1964). The concentrations used in the present study are too low to be detected by snails. It is also apparent from the behaviour of snails during this study, since they do not react to molluscicide exposure by leaving the experimental containers. It can be surmised from this study that 0.15, 0.05-0.15 and 0.01-0.15 mg/l concentrations of Yurimin, copper sulphate, and Bayluscide respectively may be used for the control of I. exustus since these are effective concentra- tions and also safer as regards their toxicity to non-target organisms. ACKNOWLEDGEMENT The authors are indebted to Dr. P.K. Ramachandran, Director, Defence Research & Development Establishment, Gwalior, In- dia, for his keen interest and constant encour- agement. REFERENCES CITED ABDALLAH, A. & NASR, T.S., 1961, Evaluation of a new molluscicide, Bayer, 73. Journal of Egyp- tian Medical Association; 44: 160-170. BLAIR, D.M., 1961, Dangers in using and handling sodium pentachlorophenate as a molluscicide. Bulletin of the World Health Organisation, 25: 597-601. CARDARELLI, N.F., 1977, Controlled release molluscicides. University of Akron, Akron, Ohio, 133" p: ETGES, F.J., 1963, Effects of some molluscicidal chemicals on chemokinesis in Australorbis EFFECTS OF MOLLUSCICIDES ON INDOPLANORBIS 269 glabratus. American Journal of Tropical Medicine and Hygiene, 12: 701-714, 1 fig. FINNEY, D.J., 1971, Probit analysis. Cambridge University Press, London, 333 p. FRICK, L.P. 8 JIMENEZ, W.A. de, 1964, Mollus- cicidal qualities of three organotin compounds revealed by 6 h and 24 h exposure against representative stages and sizes of Australorbis glabratus. Bulletin of the World Health Organisa- tion, 31: 420—431. MALEK, Е.А. & CHENG, T.C., 1974, Medical and economic malacology. Academic Press, New York and London, 398 p. UPATHAM, E.S. & CHRISTAE, J.D., 1975, Labo- ratory trials of copper and TBTO slow release compounds. Controlled Release of Molluscicides Newsletter. Univ. Akron, Akron, Ohio. WALKER, K.E. 8 CARDARELLI, N.F., 1975, Slow release copper toxicant compositions. U.S. Patent. App. Ser. No. 557, 051. WORLD HEALTH ORGANIZATION, 1965, Snail control in the prevention of bilharziasis. World Health Organization Monograph Series, 5: 255. WORLD HEALTH ORGANIZATION, 1973, Schis- tosomiasis control. Technical Report Series, World Health Organization, 515: 46 p. WORLD HEALTH ORGANIZATION, 1980, Epide- miology and control of schistosomiasis. Techni- cal Report Series, World Health Organization, 643: 63 p. Revised Ms. accepted 9 July 1985 MALACOLOGIA, 1986, 27(2): 271-280 THE TAXONOMIC STATUS OF PHILOMYCUS TOGATUS (PULMONATA: PHILOMYCIDAE): A MORPHOLOGICAL AND ELECTROPHORETIC COMPARISON WITH PHILOMYCUS CAROLINIANUS H. Lee Fairbanks Pennsylvania State University, Beaver Campus, Monaca, PA 15061, U.S.A. ABSTRACT The past and present taxonomic status of Philomycus togatus (Gould) is reviewed. Morpho- logical comparisons between P. togatus and P. carolinianus (Bosc) from three localities demonstrated clear and consistent differences in mantle pattern, color of the foot margin, and color of the mucus. Reproductive system differences include shape and size of the penis and length and thickness of the penial sheath. Electrophoretic comparisons of seven enzyme systems revealed 13 loci. Р. carolinianus was monomorphic at all loci. P. togatus was polymorphic at two loci; however, no heterozygotes were found. Between species differences existed at four loci, with each species fixed for alternative alleles, only one of eleven alleles was found in both species. The genetic distance (Nei’s (1978)D) between species was .432. Key words: Philomycidae; Philomycus; taxonomy; morphology; electrophoresis; genetic distance. INTRODUCTION Gould (1841) described Limax togata, not- ing that “It is very probable that the great development of the shield . . . may entitle this animal to be regarded as a new genus.” A. Binney (1842, 1851) placed L. togata in the synonymy of Tebennophorus caroliniensis A. Binney (1842). Later, Gould (1862: 182) stated “Limax togata is Tebennophorus caroliniensis Binney.” W. G. Binney (1878) also placed L. togata in the synonymy of 7. caroliniensis. No further mention of Limax togata was made until Pilsbry (1948) included it (as well as T. caroliniensis) in the synonymy of Philomycus carolinianus (Bosc, 1802), con- sidering both as рай of P. с. flexuolaris (Rafinesque, 1820). However, Hubricht (1951) elevated P. flexuolaris to specific rank and later (1956) identified slugs from Shenandoah National Park, Virginia as Philomycus carolinianus togatus. In addition, he placed P. c. collinus Hubricht, 1951 in the synonymy of P. c. togatus (1956). Hubricht (1968) identified slugs from Kentucky as Philomycus togatus, thus becoming the first author to return the taxon to the species level. There have been several additional identifica- tions of slugs as P. togatus (e.g. Grimm, 1971; Hubricht, 1971, 1973, 1977; Mac- Namara & Harman, 1975; Kearney & Gilbert, 1978), including fig. 405f in Pilsbry's 1948 monograph (Hubricht, 1951). Very little is known about the biology of the Philomycidae in the United States of America. Webb (1970), in a study of Philomycus carolinianus, noted that mating was recipro- cal, a preformed spermatophore was not in- volved (also noted by Kugler, 1965), and the dart was used to “ . form a definite wound. ...” Tompa (1980) also studied P. carolinianus and noted that the slug “ . .. does not lose or detach the dan during mating. ... Ikeda (1937) studied P. biline- atus (now Meghimatium bilineatus; see Pilsbry, 1948) in Japan. He noted that M. bilineatus reproductive activity is “... defi- nitely cyclic, . . . ”, that “ . . . such definite cycle of sexual activity is accompanied by certain definite changes in reproductive or- gans, ... ”, and that the ovotestis, albumen gland and penis seem to undergo the greatest cyclic changes in size. Despite the taxonomic changes no data have been published to substantiate the redesignation of Philomycus togatus as a distinct species. In light of this, a comparative morphological study of specimens of P. carolinianus and specimens identified as P. togatus was initiated. In addition, because it has been shown that electrophoresis of pro- teins can be useful in the detection of cryptic (271) 272 FAIRBANKS = rr mn — = 25 => = > FIG. 1. Collection sites in Beaver County, Pennsyl- vania. IN = Brush Creek; 2N = State Game Area 173; 3N = State Game Area 285. species of gastropods (e.g. Chambers, 1978; Murphy, 1978), an electrophoretic analysis of some tissue proteins was included in the study. MATERIALS AND METHODS Specimens of Philomycus carolinianus and P. togatus were collected from three localities in western Pennsylvania (Fig. 1). Site selec- tion was based upon availability of old growth forest on public land. Voucher specimens from each locality have been deposited in The Academy of Natural Sciences of Philadelphia (ANSP) as noted below. Locality descriptions are as follows: 1N—Brush Creek County Park; on slopes са. 10 т above creek; elevation ca. 290 т; Marion Township, Beaver County, Pa.; 40°47'50"N; 8014'14"W. Collection area = ca. 40 m?. Species collected P. carolinianus (18 specimens); ANSP A10655; HLF 374, 445. 2N—State Game Area 173; along banks of McLaughlin Run; elevation ca. 340 m; Ohio Township, Beaver County, Pa.; 40°40'19"N; 80°28'23"W. Collection area = са. 400 m?. Species collected P. carolinianus (10 speci- mens); ANSP A10651; HLF 441. P. togatus (11 specimens); ANSP A10652; HLF 438. 3N—State Game Area 285; along banks of small creek ca. 0.6 km E of Ohio State line; elevation ca. 295 m; Darlington Township, Beaver County, Pa.; 40°47'22"N; 80°30'15" W. Collection area = 400 m?. Species col- lected P. carolinianus (5 specimens, not found at the time of the electrophoretic tests); ANSP A10653; HLF 436. P. togatus (19 spec- imens); ANSP 358186, A10205, A10654; HLF 437. Because the holotype of Limax togata Gould, 1841, is not extant, the identi- fication of Philomycus togatus was corrobo- rated by L. Hubricht (personal communica- tion, 16 February 1983). Identification of P. carolinianus was accomplished using the de- scription given by Pilsbry (1948). For the morphological comparisons simi- larly-sized slugs were selected. Following the examination of external characteristics, the slugs were drowned in distilled water. The specimens were dissected in water, and the jaws, radulae, and reproductive systems were removed. The dissected bodies were pre- served in 70% ethanol. Some reproductive systems were stained, cleared and mounted on glass slides using Gregg's (1959) method with the following modifications: the tissues were left in hematoxylin for 3 min and the acid-alcohol destaining solution for 30 sec. Some radulae and jaws were mounted in Permount on glass slides. The tissue samples for electrophoretic analysis were obtained by cutting off the posterior 5 mm of each slug. The slugs were then killed by freezing and saved for morpho- logical comparisons. The tissue sample was prepared for electrophoresis using the proce- dures of Brussard & McCracken (1974) and modified as described by Selander & Hudson (1976). The samples were stored at —85°С until the tests were conducted. A Buchler Vertical Gel Apparatus (catalog number 3-1072) with a Gelman regulated power sup- ру (model 38520) was used for electrophoresis. All tests were conducted in a refrigerator at 4°С. Voltage was set at 250 volts, and current flow ranged from 47—49 milliamperes during each of the 5 hr tests. The gels (52 g of starch in 400 ml of buffer) were made using Electrostarch (Madison, Wis.), lot number 392. PHILOMYCUS TAXONOMY 273 The following enzyme systems were as- sayed: (1) Stain formulae from Siciliano & Shaw (1978) a. Glucose-6-phosphate dehydrogenase = G-6-PD b. Glutamate oxalacetate transaminase = GOT с. Isocitrate dehydrogenase = IDH d. Malic enzyme = ME e. 6-phosphogluconate dehydrogenase = 6-PGD +. Phosphoglucomutase = PGM (2) Stain formula from Selander et al. (1971) Leucine aminopeptidase =LAP The buffer system used for all stains was Tris-Versene-Borate pH 8.0 (Siciliano & Shaw, 1978). Gels were scored as described by Cham- bers (1978) except that the electromorph at a given locus most common among all speci- mens was designated with the superscript 100. Overlapping tests were run to ensure that the electromorphs on different gels were scored correctly. All presumed alleles were assumed to be autosomal and codominant. Genetic distance (D) was calculated using Nei’s (1978) method for small samples. RESULTS Morphology Specimens used for the external morpho- logical studies and in the electrophoretic tests were collected between 29 June 1982 and 28 September 1982. External examination of the bodies of the slugs (Fig. 2) revealed three obvious and consistent differences between species. All of the specimens of Philomycus carolinianus had a double row of black spots (one row on each side of a black-brown antero-posterior stripe) on the mid-dorsal surface of the mantle (Fig. 2c, а). All had a cream-white foot edge, and produced milky-white mucus when han- dled. The specimens of P. togatus had one broad or two narrow gray-black stripes, had an orange-red foot edge, and produced or- ange mucus when handled. There was an antero-posterior gray-black stripe on each side of the mantle in both species. Collections of specimens for dissection were made between 16 May and 21 May 1985, the dissections were conducted on 23 May 1985. Eight specimens of Philomycus carolinianus, three each from localities 1N and 2N and two from locality 3N (avg. wt. = 3.2 g, range = 2.4 94.1 д, s.d. = .62) and six specimens of P. togatus, three each from localities 2N and 3N (avg: wt. = 3.3 4, range = 2.5 g-4.1 9, s.d. .61) were dissected. The only obvious difference between the re- productive systems of these two species (Fig. 3) was in the appearance of the penis. Mea- surements (Table 1) show that the penis of P. carolinianus is shorter, larger in diameter (at the origin), and tapers more rapidly to the distal end than the penis of P. togatus. Lon- gitudinal sections of the two penises (Fig. 4) revealed obvious differences in penis shape and length of the penial sheath, i.e. the penial sheath in P. carolinianus was nearly as long as the penis whereas in P. togatus it was approximately 75% as long as the penis. The remaining organs of the reproductive systems were similar in the two species. Comparisons of the jaws provided no diag- nostic characters; they were all typical of the genus (Pilsbry, 1948: 750). The same was true of the radulae. Electrophoresis Thirteen loci were detected among the seven enzyme systems assayed. Eight loci, G-6-PD 1, G-6-PD 2, IDH 2, 6-PGD 1, GOT 2, PGM 1, ME 1, and ME 2, were monomorphic with the electromorphs for these loci the same in both species. The frequency distribution of the 11 alleles (electromorphs) detected for the пс polymorphic within species or with different alleles in each species is shown in Table 2. Only one of these alleles was found in both species. The Philomycus carolinianus popu- lations were monomorphic for these loci whereas each population of P. togatus had at least one polymorphic locus. No hetero- zygotes were detected. Table 3 shows the genetic distance esti- mates (D) between the various populations for both species. The interspecies values averaged .411. The interpopulation value for Philomycus togatus was .013. DISCUSSION Gould’s 1841 description of Limax togata: . and the shield extends quite back to the 274 FAIRBANKS FIG. 2. Mantle patterns: А-В, Philomycus togatus; C-D, Philomycus carolinianus. Scale bar 5 mm. А 5 mm piece of tail has been removed from C and D. PHILOMYCUS TAXONOMY 275 FIG. 3. Genitalia: A, Philomycus togatus; В, Philomycus carolinianus. Scale bars 5 тт. AG, albumen gland; AR, accessory retractor muscles; D, dart; DS, dart sac; G, gonad; HD, hermaphroditic duct; P, penis; PR, penial retractor muscle; S, spermatheca; UV, free oviduct; V, vagina; VD, vas deferens; Y, atrium. extremity of the animal, enveloping the whole animal except the head ...” indicates quite clearly that the slug was a philomycid. The description of the mantle color pattern, “ ... its margin is light fawn-color, the back is dark purplish slate color, and the sides are mottled with the two colors . . . ” implies a broad antero-posterior band on the dorsal surface without the double row of black spots as in Philomycus carolinianus. Examination of a 276 FAIRBANKS TABLE 1. Penial measurements. Measurements obtained using an ocular micrometer; all measurements in mm. Numbers represent mean, range, and standard deviation. L = length; A = diameter midpoint; С = diameter distal end. P. carolinianus diameter proximal end; B М 8 Penis [E 75 6.0-11.0 A 3.2 2.8- 3.5 B 1.3 0.8- 1.6 С 0.6 0.5- 1.0 total of 63 specimens of both species (33 P. carolinianus and 30 P. togatus) substantiates the constancy of this difference. Chichester 4 Getz (1973) noted that the color of the mucus should be recorded when one is gathering data concerning a species of slug. Unfortunately, few descriptions have included such data and therefore compari- sons are difficult. For Philomycus carolinianus these data have been recorded (ibid.) and they substantiate the difference between the above species and P. togatus as noted in this study. Among the species of Philomycus only P. rushi Clapp, 1920, a small (15-20 mm) slug, has been described as having orange or red in the sides of the foot. Pilsbry (1948) synon- ymized P. rushi with Pallifera ohioensis (Sterki). There are two large Pallifera (P. varia and P. ragsdalei) with red or orange in the foot margins. However, Philomycus togatus has a dart sac and dart and therefore must remain in the genus Philomycus. This makes it the only known species in the genus re- corded as having orange foot margins. Chichester & Getz (1968) noted that in most species of terrestrial slugs “ . . . the distal genitalia are specifically diagnostic.” The examination of specimens of Philomycus togatus and P. carolinianus, in the same stages of the reproductive cycle, demon- strated significant differences in the terminal genitalia (Fig. 4), /.e. differences in size and shape of the penis. The length and thickness of the penial sheath provided additional differ- ences. Although Pilsbry (1948) noted an ex- tension of the atrium covering the proximal portion of the penis in P. carolinianus, Kugler (1965) called it a penial sheath. No within species differences and no intergradation be- tween species was observed. Available biological data demonstrate that P. togatus 6 1.49 9.9 8.0-12.0 1.32 .26 2.5 2.0- 2.8 29 .26 1.6 1.2- 2.0 29 .30 0.6 0.5- 0.7 05 cross-fertilization (amphimixis) occurs within populations of Philomycus carolinianus (Kugler, 1965; Tompa, 1980; Webb, 1970). McCracken & Selander (1980) using electrophoretic data have noted “The agree- ment between observed and expected pro- portions of heterozygotes in our genetically variable populations of slugs indicates that they, too, are largely if not completely amphimictic. These species are E Philomycus carolinianus and three unidenti- fied species of Philomycus. ...” In this study both populations of P. carolinianus were monomorphic at all loci studied. Foltz et al. (1982) have noted that a possible explanation for monogenicity among cross-fertilizing spe- cies is that “ . they are outcrossers that have lost all heterozygosity as a conse- quence of founder effects and genetic drift in small populations.” In view of the lack of data supporting self-fertilization in this species this explanation seems most probable for the vari- ation observed. There are no published data available con- cerning the biology or genetic variation of Philomycus togatus. One might assume that the biology of P. togatus would be similar to that of P. carolinianus. However, this investi- gation detected two polymorphic loci in P. togatus, but no heterozygotes. In population ЗМ the probability of not detecting a heterozygote is highly significant (.0002). That is, it is highly unlikely, based upon the sample size, that no heterozygotes would be observed. The absence of heterozygotes in polymorphic populations implies some sort of automictic reproduction. Ikeda (1937) stated that Philomycus bilineatus “ . In isolation reproduces by self-fertilization . . . ” Nicklas and Hoffman (1981) concluded that Deroceras laeve reproduces by apomictic parthenogenesis. Foltz et al. (1982) con- PHILOMYCUS TAXONOMY PTIT VD FIG. 4. Longitudinal section of the penises: A, Philomycus carolinianus; B, Philomycus togatus. Scale bars 1 mm. DS, dart sac; P, penis; PR, penial retractor muscle; PS, penial sheath; V, vagina; VD, vas deferens; Y, atrium. 278 FAIRBANKS TABLE 2. Allelic frequency distribution at loci polymorphic within species and at loci with different alleles in each species. Philomycus Philomycus togatus carolinianus Site code 3N 2N 2N 1N Number of specimens 10 5 5 10 Locus Allele IDH 1 103 1.00 .40 — 100 — .60 1.00 1.00 6-PGD 2 107 .70 .60 — — 105 30 .40 — — 100 = — 1.00 1.00 GOT 1 100 1.00 1.00 — — 96 = = 1.00 1.00 LAP 1 103 — — 1.00 1.00 100 1.00 1.00 — — ГАР 2 100 1.00 1.00 — — 98 — — 1.00 1.00 Number ot loci polymorphic 1 2 0 0 Observed heterozygosity 0 0 — — Probability of not detecting a heterozygote .0002 .051 — — TABLE 3. Genetic distances (Nei, 1978) between populations. Site Philomycus Philomycus code togatus carolinianus Site code 3N 2N 2N 1N P. togatus 3N — .013 .433 .433 P. togatus 2N = 389 .389 P. carolinianus 2N = 000 P. carolinianus 1N — cluded that three species of arionid slugs were reproducing by self-fertilization. Thus, automixis does occur т slugs. Self- fertilization would seem to be the most likely explanation for the observed lack of heterozygotes in P. togatus. However, the small sample sizes necessitate only a tenta- tive conclusion. Electrophoretic comparisons of specimens from locality 2N, in which Philomycus togatus and P. carolinianus are sympatric, show that of eleven alleles only one is found in both species (Table 2). The genetic distance be- tween these species in this locality was .389. The interpopulation genetic distance estimate for P. togatus (.013) was comparable to intraspecific distances between populations in two species of Physa (= .012) (Buth & Suloway, 1983) and within each of three species of Crepidula (= .097) (Hoagland, 1984). Both of these authors used Мег$ co- efficient. The overall genetic distance esti- mate between P. togatus and P. carolinianus was .432, similar to that between two species of Physa (.45) (Buth & Suloway, 1983), and between several species of Sphincterochila (.372—.399) (Nevo et al., 1983). Davis (1983) has noted that for some Bivalvia there is high probability that two taxa are distinct species of О = .222, which is well below the estimates above. Philomycus togatus and P. carolinianus PHILOMYCUS TAXONOMY 279 were found in sympatry in two localities, 2N and 3N. In both of these areas these species can be collected from the same log. Both species have been collected grouped to- gether under the same piece of loose bark. In spite of the opportunity for cross-fertilization between these species, the data demonstrate that this does not occur. lt is clear that the morphological and electrophoretic data sup- port the earlier redesignation (Hubricht, 1968) of Philomycus togatus as a distinct species. LITERATURE CITED BINNEY, A., 1842, Binney on the naked air-breathing Mollusca. Boston Journal of Natural History, 4: 171. BINNEY, A., 1851, The terrestrial air-breathing mollusks of the United States. Gould, A.A., ed., 2320: BINNEY, W.G., 1878, The terrestrial air-breathing mollusks of the United States. 5: 182, Published in Bulletin of the Museum of Comparative Zool- ogy, Harvard, vol. 4. BOSC, L.A.G., 1802, Histoire naturelle des vers contenant leur description et leurs moeurs. 1: 80. BRUSSARD, Р.Р. & McCRACKEN, G.F., 1974, Allozymic variation in a North American colony of Cepaea nemoralis. Heredity, 33: 98-101. BUTH, D.G. 8 SULOWAY, J.J., 1983, Biochemical genetics of the snail genus Physa: a comparison of populations of two species. Malacologia, 23: 351-359. CHAMBERS, S.M., 1978, Ап electrophoretically detected sibling species of Goniobasis floridensis (Mesogastropoda: Pleuroceridae). Malacologia, 17: 157-162. CHICHESTER, L.F. 4 GETZ, L.L., 1968, Terrestrial slugs. The Biologist, 50: 148-166. CHICHESTER, L.F. 4 GETZ, L.L., 1973, The ter- restrial slugs of northeastern North America. Sterkiana, 51: 11-42. CLAPP, W.F., 1920, The shell of Philomycus carolinianus (Bosc). Nautilus, 33: 83-89. DAVIS, G.M., 1983, Relative roles of molecular genetics, anatomy, morphometrics and ecology in assessing relationships among North Ameri- can Unionidae (Bivalvia). In OXFORD, G.S. & ROLLINSON, D., ed., Systematics Association Special Volume 24: 193-222. РО DWE OCHMANS she ОМЕЗ. AUS" EVANGELISTI, S.M. & SELANDER, R.K., 1982, Genetic population structure and breeding sys- tems in arionid slugs (Mollusca: Pulmonata), Biological Journal of the Linnean Society, 17: 225-241. GOULD, A.A., 1841, Report on the invertebrates of Massachusetts, comprising, the Mollusca, Crustacea, Annelida and Radiata, vol. 3. GOULD, A.A., 1862, Otia Conchologica: Descrip- tions of shells and Mollusks from 1839 to 1862. Boston, 256 p. GREGG, W.O., 1959, A technique for preparing in toto mounts of Molluscan anatomical dissec- tions. The American Malacological Union Annual Report for 1958, 25: 39. GRIMM, R.W., 1971, Annotated checklist of the land snails of Maryland and the District of Co- lumbia. Sterkiana, 41: 51-57. HOAGLAND, K.E., 1984, Use of molecular genet- ics to distinguish species of the gastropod genus Crepidula (Prosobranchia: Calyptraeidae). Malacologia, 25: 607-628. HUBRICHT, L., 1951, The Limacidae and Philomycidae of Pittsylvania County, Virginia. Nautilus, 65: 20-22. HUBRICHT, L., 1956, Land snails of Shenandoah National Park. Nautilus, 70: 15-16. HUBRICHT, L., 1968. The land snails of Kentucky. Sterkiana, 32: 1-6. HUBRICHT, L., 1971, The land snails of Virginia. Sterkiana, 42: 41-45. HUBRICHT, L., 1973, The land snails of Tennes- see. Sterkiana, 49: 11-17. HUBRICHT, L., 1977, The land snails of Missis- sippi. Sterkiana, 68: 1—4. IKEDA, K., 1937, Cytogenetic studies on the self- fertilization of Philomycus bilineatus Benson. Journal of Science of Hiroshima University, ser. B, div. 1, 5: 67-123. KEARNEY, S.R. 8 GILBERT, F.F., 1978, Terres- trial gastropods from the Himsworth Game Pre- serve, Ontario, and their significance in Parelaphostrongylus tenius transmission. Cana- dian Journal of Zoology, 56: 688-694. KUGLER, О.Е., 1965, A morphological and histochemical study of the reproductive system of the slug, Philomycus carolinianus (Bosc). Journal of Morphology, 116: 117-132. MacNAMARA, М.С. & HARMAN, W.N., 1975, Fur- ther studies on the Mollusca of the Otsego Lake Area. Nautilus, 89: 87-90. McCRACKEN, С.Р. & SELANDER, R.K., 1980, Self-fertilization and monogenic strains in natural populations of terrestrial slugs. Proceedings of the National Academy of Science, 77: 684-688. MURPHY, P.G., 1978, Collisella austrodigitalis sp. nov.: a sibling species of limpet (Acmaeidae) discovered by electrophoresis. Biological Bulle- tin, 155: 193-206. NEVO, E., BAR-EL, C. 8 BAR, Z., 1983, Genetic diversity, climatic selection and speciation of Spincterochila landsnails in Israel. Biological Journal of the Linnean Society, 19: 339-373. NEI, M., 1978, Estimation of average heterozygos- ity and genetic distance from a small number of individuals. Genetics, 89: 585—590. NICKLAS, N.C. 8 HOFFMAN, R.J., 1981, Apomic- tic parthenogenesis in a hermaphroditic terres- trial slug, Deroceras laeve (Muller). Biological Bulletin, 36: 80-85. PILSBRY, Н.А., 1948, Land Mollusca of North America (North of Mexico). Vol. 2, part 2. Mono- 280 FAIRBANKS graph 3, Academy of Natural Sciences of Phila- дерма, р. 750-759. RAFINESQUE, C.S., 1820, Annals of nature or annual synopsis of new genera and species of animals and plants discovered in North America, pato: SELANDER, В.К. 8 HUDSON, R.O., 1976, Animal population structure under close inbreeding: the land snail Rumina in southern France. American Naturalist, 110: 697-718. SELANDER, R.K., SMITH, М.Н., YANG, S.Y., JOHNSON, W.E. & GENTRY, J.B., 1971, Bio- chemical polymorphism and systematics in the genus Peromyscus. |. Variation in the old-field mouse (Peromyscus polionotus), Studies in Ge- netics VI. University of Texas Publicaton 7102: 49-90. SICILIANO, M.J. & SHAW, C.B., 1978, Separation and visualization of enzymes on gels. In SMITH, |, ed., Chromatographic and electrophoretic techniques, 2, ed. 4, p. 187-216, Heineman Medical, London. TOMPA, A.S., 1980, The ultrastructure and miner- alogy of the dart from Philomycus carolinianus (Pulmonata: Gastropoda) with a brief survey of the occurrence of darts in land snails. Veliger 23: 35—42. WEBB, G.R., 1970, Observations on the sexology of Philomycus carolinianus (Bosc). Gastropodia, 1(7): 62-65. Revised Ms. accepted 22 August 1985 MALACOLOGIA, 1986, 27(2): 281-290 MODIFICATION OF PREDATORY SNAIL CHEMOTAXIS BY SUBSTANCES IN BIVALVE PREY ODORS Dan Rittschof! & Anne В. Brown? College of Marine Studies University of Delaware Lewes, DE 19958, U.S.A. ABSTRACT Effects of bivalve odors were tested on chemotactically stimulated movements in newly hatched predatory marine gastropods (Urosalpinx cinerea). Odors from Mytilus edulis (shown previously to contain chemosuppressant activity, Williams et al., 1983) go well as odors from Geukensia demissa, Mercenaria mercenaria, Mulinia lateralis, and Tagelus plebeius contained substances that suppress chemotactic responses of newly hatched snails to purified barnacle odors. In addition we: 1) demonstrated that mussel suppressants are molecules of less than 1000 Daltons, 2) showed suppression is by a non-competitive mechanism, 3) used laboratory chemicals identified as components of mussel odors and determined that ammonium ion enhances chemotactic responses to barnacle odor. The other components of mussel odor that could be identified with certainty did not affect snail responses to barnacle odor. INTRODUCTION Newly hatched predatory snails, Urosalpinx cinerea, are attracted specifically to barnacles (Rittschof et al., 1983) by peptides (Rittschof et al., 1984a) present in odors associated with intact prey. Prey odors are the assemblage of volatile and non-volatile substances that re- sult from soaking living prey in sea water. Snails move in a directed manner that is determined by a combination of flow and odor (Rittschof et al., 1983; Brown & Rittschof, 1984). Although odors from mussels, Mytilus edulis, and oysters, Crassostrea virginica, are not in themselves attractive to newly hatched snails (Rittschof et al., 1983), consumption of these prey by them results in development of a chemotactic response to their odors (Wood, 1968). Odor components involved in the later response are substances similar in chromato- graphic properties to barnacle odor peptides. In the laboratory, odors from M. edulis and C. virginica interfere with responses of newly hatched U. cinereato attractant from barnacles (Williams et al., 1983). Williams et al. (1983) demonstrated that chemotaxis to attractive prey odors is suppressed by both long- and short-term exposure of juvenile snails to oys- ter or mussel odor. Mixtures of relatively high concentrations of oyster or mussel odor with odor from barnacles evoke lower percentages of chemotaxis by the snails, than does barna- cle odor alone. Williams et al. (1983) ob- served that combinations of dilute attractant and suppressant occasionally result in en- hancement of the chemotactic response of the snails (an increase in the percentage of snails responding to mixtures when com- pared to the percentage of snails responding to just barnacle odor). The method used to generate suppressant odors in the Williams study was similar to techniques used with barnacle prey to obtain stimulants of U. cinerea and similar to the technique used with mussels to generate odors attractive to lob- sters (Derby & Atema, 1981). Subsequently, Rittschof et al. (1984b) found that newly hatched Urosalpinx cinerea incubated as encapsulated embryos in mus- sel or oyster odors until hatching are more sensitive to barnacle odor than are snails that developed either in control sea water or in the presence of barnacle odor. Increased sensi- Present address: Duke University Marine Laboratory, Pivers Island, Beaufort, NC 28516, U.S.A. “Present address: Aquatic Terrestrial Research Inc., 1230A W. Second Street, Los Angeles, CA 90026, U.S.A. (281) 282 RITTSCHOF 8 BROWN tivity of snails to barnacle odors after ex- posure to odors containing suppressant substances suggests that snails have mech- anisms for compensating to chemical interference. Further chemical characterization of sub- stances that suppress predatory snail chemo- taxis was performed in the present investiga- tion. Specifically, we: 1) confirmed that suppressant substances are released by bivalves, 2) demonstrated that mussel sup- pressants are molecules of fewer than 1000 Daltons, 3) used laboratory chemicals identi- fied as components of mussel odors and determined that ammonium ion enhances chemotactic responses to barnacle odor. The other components that could be identified with certainty did not affect snail responses to barnacle odor. MATERIALS AND METHODS Biological materials Capsules containing embryonic Urosalpinx cinerea (Say) were collected from both break- waters of Delaware Bay during July and Sep- tember and maintained in the laboratory (Rittschof et al, 1984a). Capsules were cleaned of debris and maintained in aerated sea water at 22 + 2°C at approximately 1000 capsules liter '. Snails were used in assays one to five days after hatching. Five species [Mytilus edulis Linné, Geuken- sia demissa (Dillwyn), Mercenaria mercenaria (Linne), Tagelus plebeius (Lightfoot) and Mulinia lateralis (Say)] of living, pumping bivalves were used to generate test odors. Blue mussels, M. edulis, and ribbed mussels, G. demissa, were collected from the mouth of the Broadkill River, Lewes, Delaware. Mus- sels were cleaned of most fouling organisms by brushing; however, mussels fouled with bryozoans or barnacles were not used. Hard clams, M. mercenaria, and stout razor clams, T. plebeius, were collected from Savages Ditch, Indian River Bay, Delaware. Razor clams were bound loosely with rubber bands to prevent gaping. Adult mud clams, M. lateralis, were raised in the laboratory from larvae (Brown, 1984). Prey odors were generated by exposing 50 g fresh weight of live, pumping bivalves per liter of sea water for four hours. This was done in 4.5 liter aquaria with aged sea water (Rittschof et al., 1983; Williams et al., 1983). After four hr of exposure with aeration, bivalves were removed and the sea water filtered through 1, 0.45, and 0.22 micron filters, and either used immediately, or frozen in 50 ml portions in a dry ice/methanol bath and stored at — 20°C until used. Biological assays Snail creeping assays (Rittschof et al., 1983) used homogeneous solutions contain- ing aged sea water (sea water control). Sea water plus an amount of fractionated barnacle odor sufficient to elicit response from approx- imately 50% of the snails (stimulus control) or the same amount of stimulus and a dilution of bivalve (or fractionated) bivalve odor. Using an amount of stimulus (barnacle odor) suffi- cient to elicit a 50% response enabled detec- tion of enhancement (significantly greater re- sponse) or suppression (significantly reduced response) with the same assay. In the assay, snails were exposed to a flow of homoge- neously mixed solution. Snails creeping up- stream 1 or more cm/10 min were scored as responding. Snails that did not creep 1 cm upstream were counted as not responding. Stimulus for snail creeping was pressure dia- lyzed barnacle odor <10,000 D and >1000 D that was purified and concentrated several thousand-fold (Rittschof et al., 1984a). Poten- tial suppressants were diluted between 1:1 and 1:50 with sea water. At least three dilu- tions of suppressant odor, a stimulus control and a sea water control were included in each assay series. Potential suppressants were also tested for stimulus activity in the absence of barnacle odor by dilution with aged sea water (Rittschof et al., 1983; Williams et al., 1983). In a separate series of assays with mussel suppressant, the level of suppressant was held constant in the presence of a 50-fold range of dilution of stimulus. The concentra- tion of suppressant used was that necessary to inhibit responses (Р <0.01) by snails to the 50% response level of attractant. Dilutions of attractant tested were x (standard stimulus level in all except this series of experiments), 4x,2x,0.30x, and 0.20x. Fractionation of mussel suppressant A series of experiments was conducted to determine basic molecular characteristics of suppressant(s) from mussels. In all experi- ments sea water was subjected to the same BIVALVE SUPPRESSANTS 283 treatment and tested in suppressant assays as a control. Molecular sizing In order to compare stimulus and suppres- sants on the basis of molecular size, mussel odor was fractionated in size ranges of <500, <1000, 1000-10,000 and 10,000-1000,000 Daltons by pressure dialysis (Amicon UMOS, UM2, YM10, and YM100 membranes). After dialysis each fraction was diluted with filtered sea water (<1000 Daltons) to the pre-dialysis volume. The same dilution of filtered sea water was used in stimulus and sea water controls. Adsorption chromatography As another basis of comparison with stim- ulus, mussel odor was fractionated by pas- sage of sea water containing suppressant through Amberlite XAD-7. Stimulus from bar- nacle odor can be extracted and concentrated from sea water with Amberlite XAD-7 resin. Other potential adsorbents were also tested. These included: Amberlite XAD-2 (Rohm and Haas Inc.), Sep-Pak C4 cartridges, and Sep- Pak silica cartridges (Waters Corp.). Amberlite resins were extensively washed as recommended by Jolley (1981) prior to use. Adsorbents were washed with ten volumes of distilled water. Adsorbed material was eluted with 100% Liquid Chromatography grade methanol. Methanol was removed under vac- uum and remaining material reconstituted with sea water to its pretreatment volume. In three series of experiments the pH of mussel odor was lowered to 2 with concen- trated reagent grade НС! to increase adsorp- tion of organic compounds (Jolley, 1981). The XAD-7 column was washed with ten volumes of pH 2 water. Elution of adsorbed material was with 90% methanol .01 М NaOH. After elution, the pH was adjusted back to the initial pH of 7.65—7.75 with concentrated NaOH. Final experiments tested commercially pur- chased primary amines [ammonium chloride (Fisher), tryptophane, and L-dopamine (Sigma)] that were detected in the size frac- tion (< 1000 Daltons) of mussel odor with suppressant activity. These compounds were tested in dilution series that bracketed the range that they occurred in suppressant. Con- centrations and identities of these com- pounds were determined by High Perfor- mance Liquid Chromatography. Amino acid analysis Mussel suppressant was tested for pres- ence of amino acids, other primary amines, and carbohydrate. Primary amines were derivitized with o-pthaldialdehyde, separated by high performance liquid chromatography on an Altex 25 cm, 5 um particle size, octadecylsilane column, detected fluoro- metrically and identified by co-migration with standards (Lindroth & Mopper, 1979; Mopper & Lindroth, 1982). Data analysis The Log Likelihood Statistic (G, Sokal & Rohlf, 1981) was used to test for suppressant activity (Williams et al., 1983). In addition, Probit Analysis (Finney, 1969) was used to determine concentration of test solutions that were effective in inhibiting responses by 50% (ECso). ЕСьоз were determined by a BASIC computer program (Lieberman, 1983) modi- fied from Applesoft BASIC to MBASIC. RESULTS Bivalve suppressants Prior study of bivalve odor potency has shown that pumping activity is important for odor production (Blake, 1960; Williams et al., 1983). Accordingly, tests comparing bivalve odor potency used only preparations where bivalves were actively ventilating. Odors from all five bivalve prey contained highly signifi- cant (p <<0.01 by G statistic) suppressant activity (Fig. 1A-E). Probit estimates (Lieber- man, 1983) of odor dilutions effective for reducing responses to 50% of control ranged from 1:50 for Mytilus edulis odor to 1:12 for Mulinia lateralis odor. There were no signifi- cant differences in potencies of suppressants produced by the 5 bivalves tested (p >.05). Mytilus edulis suppressant Suppressant from blue mussels (Fig. 1A) was chosen for further study. In addition to being the only common bivalve consumed by oyster drills at Delaware Bay breakwaters, blue mussels pumped predictably and could be obtained readily in quantity and in an unfouled condition. First, the effects of vari- ous stimulus concentrations were tested. Then, molecular size and adsorptive proper- 284 RITTSCHOF 8 BROWN ol = Mytilus eduls A N Mercenaria mercenaria В P о + я x uy Ww 5 п ul x O SEA STIMULUS + + + SEA CONTROL Mw мы nw WATER CONTROL MM MM MM WATER 1100 110 11 CONTROL 1 50 110 tt CONTROL 10 Tagelus plebeius С Mulinia lateralis 08 о q x с 06 el E n # 04 | 02 ее в = 150 110 (BI CONTROL 150 110 и CONTROL 10 Geukensia demıssa ‘| 08 | o = a [mau] Se WW é n 2 ost О J a FIG. 1. Suppressant activity in odor preparations from five bivalve species. Because of variability in response of different batches of snails, the ratio of the response to stimulus control to stimulus + dilutions of each bivalve odor has been presented. Dilutions tested, suppressant odor : sea water are presented below each bar. Numbers of snails tested per bar ranged from 110 to 60. Responses of snails to stimulus in sea water varied from 40 to 60%. BIVALVE SUPPRESSANTS 285 TABLE 1. Response of snails to attractant in the absence and in the presence of suppressant (1:15 dilution) from blue mussels. Without suppressant Number Dilution of attractant tested ASE 87 2х 53 1% 137 OSX 52 0.2 x 64 With suppressant Percentage Number Percentage responding tested responding 67 71 25 50 56 18 39 142 18 31 55 20 22 81 17 *x is the concentration of stimulant used in all other experiments. ties were examined, enabling comparison of suppressant with known stimulant sub- stances (Rittschof et al., 1984a). Effect of modifying stimulus concentration The mechanism of action (competitive vs. non-competitive) was tested by determin- ing the effect of varying stimulus concentra- tion. Attractant concentration was varied 20- fold and tested with the same concentration of suppressant from mussels. In the absence of suppressant there was a consistent, positive, dilution versus response relationship (Overall G = 37.2 4d.f. p <<0.001) (Table 1). In the presence of suppressant, the response to stimulus was reduced to the same level in all dilutions of stimulus (Table 1). Responses by snails in the presence of suppressant were low and statistically the same through a 20- fold range of stimulus concentration (Overall д = 2.1 4 d.f. p >0.05). Size fractionation of mussel inhibitor Cascade pressure dialysis was employed to estimate the molecular size of substances in mussel water with suppressant activity. Suppressant substances passed readily through membranes with nominal molecular exclusions of 100,000, 10,000, and 1,000 Daltons. The original suppressant activity was detected in the fraction of substances with molecular weights of less than 1000 Daltons. There was no evidence of substances greater than 1000 Daltons with suppressant activity (Fig. 2). Next, suppressant was subjected to pressure dialysis with a 500 Dalton cutoff membrane. Suppressant activity was retained above the membrane. The fraction containing substances passing through the membrane facilitated snails’ responses to attractant (Fig. 2). Affinity of suppressant for adsorbents Mussel suppressant at normal ionic strength and pH of sea water had little or no affinity for Amberlite XAD-2, XAD-4, Amberline XAD-7, Waters Sep-Pak Silica or Waters Sep-Pak reversed phase Silica. In each case the eluate (undiluted and di- luted 1:10) was as potent as the starting material. As observed occasionally in mo- lecular sizing experiments, low dilutions of suppressant containing eluate (1:50 to 1: 100) periodically facilitated the response to stimulus. Amberlite XAD-7 is effective at removing substances such as humic acids form sea water if the pH is lowered to 2.0 during the adsorption process (Jolley, 1981). A portion of the suppressant was removed by lowering the pH to 2.0 prior to passage through XAD-7 (Table 2). Attempts to recover suppressant adsorbed on the resin by elution with 90% methanol 0.01 N sodium hydroxide were unsuccessful. The basic methanol elu- ates of columns, whether they were control or experimental, were slightly inhibitory. This was the case even after removal of methanol and neutralization of pH. Information about size and adsorptive prop- erties was used to formulate additional chem- ical and biological tests of mussel odor and its components. The data indicate suppressants are relatively small hydrophilic compounds. Two common classes of small hydrophilic compounds are primary amines (amino acids, ammonium and other organic amines) and sugars. Chemical tests determined the types and concentrations of potential suppressant 286 RITTSCHOF 8 BROWN À UNTREATED MUSSEL WATER ра PRESSURE DIALYZED MUSSEL WATER RESPONSE RATIO O © [==] == Sa aaa E» = Sa fe] [===] [===] ESS ==] [===] [==] [===] [==] [===] [==] == == [===] [==] [==] =>] [===] ===] ES je] Sa = >=} ==] ==] ===] [===] [===] ===] ===] [===] = E» [===] [===>] [===>] [=== =] [===] Е STIMULUS + + + + CONTROL MW MW MW MW 1:10 1:50 IOKD 1:50 + Ш ^ ЗЕМ (LIT 2 + LNH м oy < 585+ LUN о SEA MW WATER >50° <500D CONTROL 1:50 ЕЮ 1:50 FIG. 2. Size fractionation of suppressant from Mytilus edulis. Stimulus was mixed in 1 part sea water <1000 Dalton and 49 parts unfiltered sea water. Dilution of inhibitor fractions was in a similar fashion. Inhibitory activity resided in the <1000 Dalton fraction and in the fraction <1000 Daltons and >500 Daltons. The fraction less than 500 Daltons enhanced response percentages. substances, and biological assays tested, the effectiveness of these compounds as sup- pressants at the concentrations at which they occurred in active suppressant preparations. Suppressant preparation for these experi- ments was first passed through XAD-7 resin and 10,000 and 1,000 Dalton membranes prior to chemical and biological determina- tions. HPLC analysis of primary amines in sea water and mussel suppressant indicated elevated levels of L-dopamine, tryptophane and ammonium in the suppressant prep- aration. Comparison of peak areas with standards of known concentration indicated that there were approximately 0.6 uM levels of L-Dopamine, 0.2 uM tryptophane and ap- proximately 3 uM ammonium. Bioassays were used to test the activity of 2 micro- molar to .2 micromolar tryptophane, 1.33 to .13 micromolar L-Dopamine and 30 to 3 micromolar ammonium. Analysis of reagent grade tryptophane showed 5 to 10% con- tamination with dopamine. After each assay, water from the most concentrated dilu- tion of each compound tested was anal- yzed by HPLC. In each case, the resultant peaks demonstrated that concentrations of the added compounds were approximately ten times greater than the mussel odor fraction with 100% suppressant activity. Bio- logical assays indicated that ammonium and L-DOPA had insignificant levels of in- hibitory activity. Tryptophane significantly in- hibited responses of snails at three of the four dilutions tested. However, com- pared to suppression of a comparable dilution of mussel water, inhibition was slight (Table 3). Additions of ammonium at levels approaching those found in suppress- ant enhanced responses significantly (Table 3). BIVALVE SUPPRESSANTS 287 TABLE 2. Removal of suppressant activity from sea water by adsorption onto Amberlite XAD-7 resin at pH 2.0. Group or Number Percent fraction* tested response Comparison G Sig Experiment One Stimulus control? 72 50 -— -— — + MW 1:15? 83 22 vs cont 12.5 <0.005 Effluent® 73 36 vs cont 2.5 ns vs +MW 4.1 <0.05 Experiment Two Stimulus control 74 51 — — = + MW 1:15 82 26 vs cont 9.9 <0.005 Effluent 83 Sil vs cont Sal ns vs +MW 2.6 ns Experiment Three Stimulus control 90 47 — — — + MW 1:15 85 24 vs cont 10.3 <0.005 Eluate? 81 37 vs cont 1.6 ns vs + MW 315 ns Experiment Four Stimulus control 99 38 — —- — Eluate al 37 vs cont 0.1 ns Eluate control® 103 РУ vs cont 3.4 ns vs eluate 2.1 ns aStimulus control: sea water whose pH was lowered to 2.0 with НС! and then returned to 7.7 with NaOH was diluted 1:15 with untreated sea water; then stimulus was added. b+ MW: The pH of mussel water was lowered to 2.0, then returned to 7.7. This was then diluted 1:15 with untreated sea water and stimulus was added. “Effluent: Mussel water at pH 2.0 was passed through an XAD-7 column. The pH was then returned to normal and the solution mixed with untreated sea water and stimulus as described above. dEluate: The XAD-7 column from above(*) was rinsed with pH 2.0 distilled water, then eluted with 90% methanol in 0.01 N NaOH. This material was rotary evaporated to near dryness, reconstituted with sea water to 0.5 the original volume, and tested for inhibitory activity. eEluate control: Identical to Eluate (7) except the material originally passed through the column (as in ©) was sea water, not mussel water. DISCUSSION Substances that suppress chemotactic re- sponses of newly hatched oyster drills to barnacle attractant occur in all five species of bivalve tested. Thus the suppressant phe- nomenon reported by Williams et al. (1983) appears common to odors of many bivalves. Within the limits of discrimination of the bio- logical assay, suppressant potencies in all odors were similar. Cascade dialysis experi- ments showed that suppressant activity present in blue mussel odor is associated with molecules of less than 1000 Daltons. Little confidence can be placed upon the size of the suppressant based upon its retention by the 500 Dalton membrance because charged substances of less than 500 Daltons are often retained. The passage of facilitory activity through the membrane 1$, however, strong evidence that facilitation and suppression are due to different substances. Experiments with organic adsorbents indicate that at least a fraction of the suppressant activity 1$ associ- ated with hydrophilic, possibly charged sub- stances. Ammonium, tryptophane and L-Dopa were present in relatively high con- centrations in fractionated mussel odor with potent suppressant activity. L-Dopa, a neurotransmitter in a variety of vertebrates and invertebrates, is also a major building block of mussel periostracum and byssus (Waite & Tanzer, 1981). Thus, it is not sur- prising to find L-Dopa in high concentration in mussel odor. When tested for suppressant activity, ammonium, L-Dopa and tryptophane were ineffective at concentrations similar to 288 RITTSCHOF 8 BROWN TABLE 3. Tests of primary amines for suppressant activity. Number Substance* Concentration tested Stimulus control Standard 69 + 1:10 MW — 78 + Ammonium 3.0 x 10-5 55 + Ammonium 1.5 x 10-5 87 + Ammonium 3.0 x 10-6 61 No stimulus — 54 + Ammonium 3 x 10-5m 67 + Ammonium 3 x 10-6M 40 Stimulus control Standard 52 + 1:10 MW — 89 + Tryptophane 2 x 10-6M , 56 + Tryptophane 1 x10-6M 46 + Tryptophane 2 x 10-7M 56 + Tryptophane 2 x 10-8M 59 No stimulus — 64 Stimulus control Standard 71 + 1:10 MW —- 81 + L-DOPA 1.3 x 10-6M 55 + L-DOPA 6.6 x 10-7 46 + L-DOPA 1.3 x 10-7 71 No stimulus — 77 + L-DOPA 1.3 x -6M 36 + L-DOPA 1.3 x —7 29 Percentage responding G vs. cont Sig. (p) 56 — — cs} 32.9 <<0.005 62 0.4 ns 65 Wee ns 77 6.1 <0.05 2 3 4 87 o — 17 69.6 <<0.005 52 15.6 <0.005 89 0.2 ns 66 6.2 <0.005 al 3.9 <0.05 2 52 — — 17 20.8 <0.005 44 0.9 ns 63 1.4 ns 52 0 ns 4 8 8 *For each substance the concentration approximately equivalent to the concentration of that substance in a 1:10 dilution of inhibitor is underlined. those in mussel suppressant. Ammonium at 3 x 10 6 M facilitated responses to attractant. The general occurrence of suppressant in odors of all bivalves tested suggests that interference with drill chemoreception may be due to products common to bivalve metabo- lism. Interference is not competitive with stim- ulus because inhibition cannot be eliminated by increasing attractant concentration. Sup- pressants are markedly different chemically from stimulus. Stimulants are proteinaceous (Blake, 1961; Rittschof et al., 1984a). Sup- pressants have a molecular size that corre- sponds at their largest size to tripeptides and disaccharides and at their smallest to inor- ganic ions. The resistance of suppressants to extraction procedures known to isolate stim- ulus (Rittschof et al., 1984a) supports the view that they are markedly different from stimulants. Sugars and amino sugars would be resistant to these extraction procedures and are thus additional candidates as sup- pressants. Although substances produced by bivalves interfere with chemotactic responses of newly hatched drills, larger drills that have con- sumed bivalve prey, including oysters, mus- sels, clams and scallops, can locate all of these prey from a distance (Haskin, 1950; Blake, 1961; Wood, 1968; Pratt, 1974; Ordzie 8 Garofalo, 1980). Work in progress demon- strates that the stimulants in at least Cras- sostrea virginica odor are similar chemically to stimulants from barnacles. Thus, condition- ing of drills to bivalve prey odors may include sensitization to stimulant substances and de- sensitization to suppressants. The occur- rence of desensitization to suppressants 1$ supported by the observation that exposure of snail embryos to bivalve odors results in increased responsiveness of the snails to barnacle attractant upon hatching (Rittschof et al., 1984b). Purification and identification of all active odor components will assist inter- pretation of these interactions. Facilitation (Williams et al., 1983; Rittschof et al., 1983) of drill responses to stimulus can be attributed to presence of ammonium. Am- monium was reported to be attractive to Urosalpinx cinerea by Blake (1961) and was suggested as a non-specific attractant by Wood (1968). However, Pratt (1974) showed that ammonium at levels corresponding to those of prey effluents was not attractive to U. BIVALVE SUPPRESSANTS 289 cinerea and noted that his results disagreed with the opinions of Blake and Wood. Our finding that ammonium is not attractive in itself but has the capacity to facilitate re- sponses in concert with low levels of specific stimulants resolves this conflict. The well-documented ability of drills to lo- cate prey with relatively high metabolic rates [Haskin, 1950; (either smaller individuals or higher oxygen consumption) Blake, 1962] may be explained by the facilitation of chemotaxis by ammonium. Drills may be di- rected to prey by the combination of stimulant and ammonium. If the mechanisms of gener- ation of ammonium and stimulus are different (for example, slow release and hydrolysis of attractant and metabolic production of ammo- nium) then the ratio of ammonium to stimulus would be higher in metabolically active prey. Ammonium ion concentrations may be used by drills to determine distances to prey. Brown & Rittschof (1984) demonstrated that drills respond to combinations of flow and stimulus concentrations predictably. Ammo- nium ion could serve to modulate responses by providing distance information. Diffusion would result in relatively rapid attenuation of the ammonium signal with little concomitant change in the more slowly diffusing peptide attractant signal. Thus, the presence of am- monium with attractant could facilitate a re- sponse because it indicates prey are rela- tively close-by. The possibility exists that ammonium could function in a similar fashion for other organisms, lobsters, for example (Derby & Atema, 1981). In the case of the latter, the existence of specific and sensitive ammonium receptors has been demonstrated (Derby & Atema, 19824). Bivalve odors contain a mixture of low molecular weight substances that can have both negative and positive effects on chemoreceptive responses of Urosalpinx cinerea. Negative effects of bivalve odors (specifically of M. edulis) on chemotaxis have been observed previously for U. cinerea (Wil- liams et al., 1983) and for Asterias forbesii (Davis, 1975). Many reports indicate positive chemotactic responses to these types of odors especially after ingestive experience. These reports include responses by gastro- pods (Haskin, 1950; Chew & Eisler, 1958; Wood, 1968), echinoderms (Castilla, 1972), flatworms (Ferrero et al., 1976) and crusta- ceans (Derby & Atema, 1980, 1981, 1982a, b). In the case of drills, it appears that the observed response is dependent upon the concentration of each effector. At high odor concentrations the noncompetitive nature of suppressant activity dominates. Suppressant effectiveness diminishes rapidly with dilution. Even our most potent suppressant solutions cannot be diluted more than 100 fold without total loss of activity. In comparison, attractant from barnacles generated from a similar biomass can easily be diluted 500 fold and retain detectable activity (Rittschof et al., 1983, 1984a). As suppressant activity 15 di- luted, facilitation of response to stimulus by ammonium is manifested. Although chemical detection of prey by snails is obviously com- plex, understanding the nature of the chemi- cals involved provides a basis for future hy- potheses. The ability to detect levels of stimulus in natural waters containing very high concen- trations of mussels and barnacles (Rittschof et al., 1984b) using newly hatched drills sug- gests that bivalve suppressants are not func- tioning as defenses at distances of several meters. They may be effective however over short distances (centimeters). Laboratory ev- idence of attractiveness of mixtures contain- ing bivalve odors (Rittschof et al., 1983), and the evidence presented here of general pro- duction of suppressant activity by bivalves, many of which experience little if any muricid gastropod predation, support the hypothesis that suppressants are not specific antipreda- tor substances. However, the interplay of these natural suppressants with stimulants and enhancing substances provides the basis for continued study and dissection of a com- plex rheotactic and chemotactic guidance system. ACKNOWLEDGEMENTS We thank R. Shepherd, K. Mopper, M.R. Carriker, C. Merrill, D. Keiber, C. Griffith, D. Einholf and J. Deschamps for critical advice and assistance. Sea Grant #NA83AA-D- 0017. LITERATURE CITED BLAKE, J.W., 1960, Oxygen consumption of bi- valve prey and their attractiveness to the gastro- pod, Urosalpinx cinerea. Limnology and Ocean- ography, 5: 273-280. BLAKE, J.W., 1961, Preliminary characterization of oyster metabolites attractive to the predatory 290 RITTSCHOF 8 BROWN gastropod, Urosalpinx cinerea. Ph.D. thesis, Uni- versity of North Carolina, Chapel Hill, 46 p. BROWN, A.B., 1984, Development of Mulinia lateralis (Say) in response to estrogen supple- ments. Masters thesis, University of Delaware, 114 p. BROWN, B.B. 8 RITTSCHOF, D., 1984, Integration of flow and chemical cue by predatory snails. Marine Behavior and Physiology, 11: 75-93. CASTILLA, J.C., 1972, Responses of Asterias rubens to bivalve prey in a Y-maze. Marine Biology, 12: 222-228. CHEW, К.К. & EISLER, E., 1958, A preliminary study of the feeding habits of the Japanese oyster drill, Ocinebra japonica. Journal of the Fisheries Research Board of Canada, 15: 529-535. DAVIS, S., 1975, Chemoreception in the starfish, Asterias forbesi. Masters thesis, University of Delaware, 95 p. DERBY, C.D. 8 ATEMA, J., 1980, Induced host odor attraction in the pea crab Pinnotheres maculatus. Biological Bulletin, 158: 26-33. DERBY, C.D. 8 ATEMA, J., 1981, Selective im- provement in responses to prey odors by the lobster, Homarus americanus following feeding experience. Journal of Chemical Ecology, 7: 1073-1080. DERBY, C.D. 8 ATEMA, J., 1982a, Chemosensitiv- ity of walking legs of the lobster Homarus americanus: neurophysiological response spec- trum and thresholds. Journal of Experimental Biology, 98: 303-315. DERBY, C.D. 8 ATEMA, J., 1982b, The function of chemo- and mechanoreceptors in lobster (Homarus americanus) feeding behaviour. Jour- nal of Experimental Biology, 98: 317-327. FERRERO, E., TONGIORGI, P., GALLENI, L., SALGHETTI, U. & SALVADEGO, P., 1976, Chemical attraction of Stylochus mediterraneus Galleni (Turbellaria: Pollycladia) toward its prey Mytilus galloprovincialis L. Marine Biology Let- ters, 1: 213-224. FINNEY, D.J., 1971, Probit analysis. Cambridge University Press, London, England, 333 p. HASKIN, H.H., 1950, The selection of food by the common oyster drill. Proceedings of the National Shellfisheries Association, 1950: 62-68. JOLLEY, R.L., 1981, Concentrating organics in water for biological testing. Environmental Sci- ence and Technology, 158: 874-880. LIEBERMAN, H.R., 1983, Estimating LD50 using the probit technique: A BASIC computer pro- gram. Drug and Chemical Toxicology, 6: 111-116. LINDROTH, P. & MOPPER, K., 1979, High perfor- mance liquid chromatographic determination of amino acids by precolumn derivitization with opthaldialdehyde. Analytical Chemistry, 51: 1667-1674. MOPPER, K. & LINDROTH, P., 1982, Diel and depth variations in dissolved free amino acids and ammonium in the Baltic Sea determined by shipboard HPLC analysis. Limnology and Oceanography, 27: 336-347. ORDZIE, C.J. & GAROFALO, G.C., 1980, Preda- tion, attack success and attraction to the bay scallop, Argopecten irradians (Lamarck) by the oyster drill, Urosalpinx cinerea (Say). Journal of Experimental Marine Biology and Ecology, 191: 199-209. PRATT, D.M., 1974, Attraction to prey and stimulus to attack in the predatory gastropod Urosalpinx cinerea. Marine Biology, 27: 37—45. RITTSCHOF, D., WILLIAMS, L.G., BROWN, B. 8 CARRIKER, M.R., 1983, Chemical attraction of newly hatched oyster drills. Biological Bulletin, 164: 493-505. RITTSCHOF, D., WILLIAMS, L.G. 4 SHEPHERD, R.G., 1984a, Concentration and preliminary characterization of a snail attractant from sea water. Journal of Chemical Ecology, 10: 63-79. RITTSCHOF, D., KEIBER, D. & MERRILL, C.L., 1984b, Modification of response thresholds of newly hatched snails by odor exposure during development. Chemical Senses, 9: 181-192. SOKAL, В.В. & ROHLF, F.J., 1981, Вютейу. Freeman, San Francisco, U.S.A., 829 p. WAITE, J.H. & TANZER, M.L., 1981, Polyphenolic substances of Mytilus edulis: Novel adhesive containing L.-DOPA and hydroxyproline. Sci- ence, 212: 1038-1040. WILLIAMS, L.G., RITTSCHOF, D., BROWN, B. 8 CARRIKER, M.R., 1983, Chemotaxis of oyster drills Urosalpinx cinerea to competing prey odors. Biological Bulletin, 164: 536-548. WOOD, L., 1968, Physiological and ecological as- pects of prey selection by the marine gastropod Urosalpinx cinerea (Prosobranchia: Muricidae). Malacologia, 6: 267-320. Revised Ms. accepted 18 March 1985 MALACOLOGIA, 1986, 27(2): 291-298 ANOMALIES IN NATICID PREDATORY BEHAVIOR: A CRITIQUE AND EXPERIMENTAL OBSERVATIONS Jennifer A. Kitchell', Christofer H. Boggs*, James A. Rice?, James Е. Kitchell{, Antoni Hoffman? & Jordi Martinell® ABSTRACT Reports of multiply bored prey of naticid gastropods, a rare but persistent occurrence in the fossil record dating from the Late Cretaceous, represent potentially serious problems for the study of naticid prey selection in the fossil record. We report controlled laboratory trials wherein multiple complete naticid boreholes were produced on single, live Terebra dislocata as a result of active escape behavior by the prey. These results are compared with a fossil assemblage characterized by a high frequency of multiple complete boreholes. Passive mechanisms such as interruption of predatory action and mechanical abrasion of incomplete boreholes can also result in apparent multiply bored prey in fossil assemblages. Generalizations that naticids frequently consume prey without drilling or that naticids frequently bore empty shells are unsubstantiated. Key words: predator-prey interactions; naticid gastropods; prey selection; phenomenon of multiple boreholes; laboratory trials; fossil assemblages. INTRODUCTION Multiply bored prey of drilling predators present a potential enigma. Previous experi- mental work (Kitchell et a/., 1981) demon- strated that a naticid gastropod predator, Polinices duplicatus (Say), behaviorally se- lects prey in accord with an energy maximization principle. These results are sig- nificant in that they permit predictions of predator-prey interactions on an evolutionary as well as an ecological time scale (Kitchell et al., 1981; Kitchell, 1982), and provide the necessary framework for studies of predator- prey coevolution (DeAngelis et al., 1984; Kitchell, 1983). Antithetical to these results are (1) reports of occasional multiply bored prey (e.g. Hoffman et al., 1974; Stanton & Nelson, 1980), (2) allegations that naticid predators Боге dead or empty shells (e.g. Hoffman et al., 1974; Stanton 8 Nelson, 1980), and (3) reports that naticids do not bore all of their prey (e.g. Taylor et al., 1980). Allegations (1) and (2) raise serious doubts about the application of foraging theory to the prey choice behavior of naticids; allegation (3) raises serious doubts about the importance of prey handling time (/.e., drilling time) in this predator-prey interaction. Our purpose in this paper is to assess these allegations. We present laboratory data on the production of multiply bored prey, comparison with fossil assemblages, and reassessment of several older studies whose results have been uncriti- cally perpetuated in the literature. In general, individual prey of naticid gastro- pods exhibit a single borehole. Single, com- pleted boreholes signify successful predation and are significantly more common than sin- gle, incomplete (i.e. imperforate) boreholes that evidence unsuccessful or interrupted pre- dation. Occasionally, a pair or series of boreholes are observed, consisting entirely of incomplete boreholes (Fig. 1A) or of a single complete borehole and one (Figure 1B) or several (Figure 1C) incomplete boreholes. In an earlier study (Kitchell et al., 1981), we used the term “multiple boreholes” in refer- ence to boreholes of this type. Such a pair or series evidences a chronological sequence of unsuccessful predation attempts, followed ul- timately by successful predation. These “mul- tiple boreholes” may be of equal diameter "Museum of Paleontology, University of Michigan, Ann Arbor, Michigan 48109, U.S.A. “National Marine Fisheries Service, Honolulu, Hawaii 96812, U.S.A. “Department of Zoology, North Carolina State University, Raleigh, North Carolina 27695, U.S.A. “Department of Zoology, University of Wisconsin, Madison, Wisconsin 53706, U.S.A. SPolish Academy of Sciences, Warsaw, Poland. Departamento de Paleontologia, Facultad de Geologia de la Universidad de Barcelona, Barcelona, Spain. (291) 292 KITCHELL ET AL. FIG. 1. A. Multiple incomplete boreholes resulting from observed active escape behavior by prey. Prey, Мегсепапа mercenaria; predator, Polinices duplicatus. Outer borehole diameters range from 3.2 mm for the most shallow borehole to 3.7 mm for the two deep boreholes. B. Single incomplete borehole, followed by observed complete borehole penetration (r:R ratio 0.47) and prey consumption. Prey, Mercenaria тегсепапа; predator, Polinices duplicatus. С. Multiple interruption of drilling by observed prey escape behavior, followed by successful drilling of complete borehole (r:R ratio 0.49). Prey, Mercenaria mercenaria; predator, Polinices duplicatus. D. Two naticid boreholes (one complete, r:R ratio 0.5), one nonfunctional (r:R ratio 0.28) in Venus multilamella; Pliocene, l'Emporda deposits, Spain. E. Two boreholes in Corbula gibba, Pliocene, l'Emporda deposits, Spain. Е. Typical naticid borehole in Corbula gibba illustrating pronounced conchiolin layer; Pliocene, l'Emporda deposits, Spain. G. Two complete boreholes in Venericardia granulata, Jackson Bluff Formation, Miocene, Leon Co., Florida (r:R ratios 0.51 and 0.62). H. Multiple naticid boreholes in the naticid Euspira rectilabrum, Late Cretaceous, Ripley Formation, Coon Creek, McNairy Co., Tennessee (USGS Loc. 16951); specimen maximum diameter 3.45 mm; outer borehole diameters 0.64 mm, 0.62 mm, 0.5 тт; r:R ratios 0.58, 0.83, and 0.55, respectively. 1. Multiple nonfunctional and incomplete boreholes in Strioterebrum monidum, Miocene (?), lower Gatun Formation; USNM 646047. Height 23.5 mm; see Woodring, 1970, pl. 64, figs. 3, 4. J, K, M, N. Observed nonfunctional boreholes resulting from multiple predation attempts and prey escapes; prey, Terebra dislocata; predator, Polinices duplicatus. L. Observed nonfunctional borehole and complete borehole; prey, Terebra dislocata; predator, Polinices duplicatus. NATICID PREDATORY BEHAVIOR 293 (Fig. 1A-C), indicating repeated attempts by the same or similarly-sized predator, or of unequal diameters evidencing repeated at- tempts by different predators. The “multiply bored prey” referred to in the present study are categorically distinct from those described above. In exceptional instances, a pair or series of naticid boreholes, each of which completely perforates the shell, is encoun- tered in a single prey. These exceptional cases are a major focus of this study. Multiply bored prey, in which more than one borehole is complete, are extremely rare, whereas the other category of multiple boreholes is merely uncommon. We intro- duce the term “nonfunctional borehole” to refer specifically to a borehole that has com- pletely perforated the prey's shell but is nonfunctional in that the opening of the inner borehole is not sufficiently large for insertion of the proboscis. A nonfunctional borehole evidences that boring of the prey is nearly complete, but feeding has not yet begun. Typical complete naticid boreholes, for exam- ple, have ratios of inner borehole diameter (r) to outer borehole diameter (R) of >0.5 (see Kitchell et a/., 1981). Nonfunctional boreholes typically have r:R ratios <0.5 (see below). The significance of this lack of functionality has not been appreciated. FOSSIL EVIDENCE OF MULTIPLY BORED PREY Carriker & Yochelson (1968) reviewed Re- cent predatory gastropod boreholes and pro- vided one example of multiply bored prey (their pl. 2, fig. 8). However, only one of the boreholes is functionally complete; the sec- ond borehole is perforate but appears nonfunctional (r:R = 0.2 in one dimension). Hoffman et al. (1974) reported examining 20,000 Middle Miocene specimens (Korytnica clays, Poland) of potential prey of both naticid and muricid drilling gastropods, and provide one unambiguous example of a multiply bored naticid in which three boreholes are complete (their pl. 2, fig. 3). One of us (Martinell), examining 4,200 Pliocene speci- mens (ГЕтрогаа, Spain), observed 6 gastro- pods each with two naticid boreholes, and two bivalves with multiple naticid boreholes. The bivalve examples are given in our Fig. 1D, E. One of the boreholes in Venus multilamella (Fig. 1D) is functionally incomplete. The se- nior author has obtained metric data on 14,000 specimens (assemblages range in age from Late Cretaceous-Recent) of poten- tial naticid prey, and has found 12 examples of multiply bored prey (e.g. Fig. 1G, H). It is of some interest that one of these multiply bored prey is naticid (Euspira rectilabrum) from the Late Cretaceous Ripley Formation (USNM #16951). The significance is two-fold: first, it evidences the relative antiquity of multiply bored prey. Until recently, the oldest docu- mented naticid boreholes of any type were of Albian age (Sohl, 1969; Taylor et al., 1980; but see Fursich 8 Jablonski, 1984, and Newton, 1983, for Triassic reports of naticid boreholes). Secondly, Vermeij 8 Dudley (1982) recently reported finding no evidence of any cannibalistic interactions in these same Late Cretaceous assemblages. This speci- men, however, documents that both cannibal- ism and multiply bored prey range at least from the Late Cretaceous to the Recent. ALTERNATIVE EXPLANATIONS OF MULTIPLE COMPLETE BOREHOLES Multiple complete boreholes are evidence of several possibilities regarding naticid predator-prey interactions: (1) The prey was simultaneously bored by more than one predator. Although this behav- ior is common among drilling muricids, none who has observed naticid predators in exper- imental or field settings has reported multiple, simultaneous predatory attacks, a finding consistent with the naticid behavior of envel- oping the prey in the mesopodium. We dis- count this possibility. (2) The shell of the bored prey was bored again after the prey had been consumed by a naticid predator. This is the current explana- tion provided in Stanton & Nelson (1980) who concluded that “the presence of some shells with two or three borings would suggest that naticids bored whatever shell they encoun- tered, whether it was alive or dead, and apparent prey may have been only dead empty shells” (p. 127). Similarly, Hoffman et al. (1974) reported that “... naticids do not distinguish between the alive and dead spec- imens and they bore each shell they find, sometimes those previously bored” (p. 252). The latter study concluded that “this indicates that the object of attack was selected by the naticids rather at random and without any preference.” Such a premise is not logically sound. Selection should rapidly discriminate against any naticid that expended time and energy, both of which are substantial in this 294 KITCHELL ET AL. TABLE 1. Borehole number and type on Strioterebrum monidum of the Bowden Formation, Jamaica, and Gatun Formation, Panama. USNM and USGS locality numbers. Height refers to actual specimen height; all specimens are not complete. Collection Height No. of locality # specimen (mm) boreholes 135283 17.0 8 135283 18.3 4 135283 15.5 2 135283 14.9 1 135283 16.1 3 135283 14.2 4 135283 19.8 11 646046 17.0 15 2580 13.8 5 2580 9.2 7 23741 18.3 4 23741 17.8 9 23741 15.6 6 23741 15.0 4 23741 15.5 9 23741 10.1 2 23741 13.4 6 23741 13.2 3 23741 11.6 1 23741 13.1 1 23741 8.3 2 23741 3.6 3 No. of incomplete Borehole type: nonfunctional Complete oOOoo-DOo--DOoNnDow@wooooo— Oo GO = © À GO © © À BR O1 B À O1 D © PB À © = © Or-0O-ONO-0>NO0000-0- system (see Kitchell et al., 1981), in boring empty shells. Secondly, such a premise has no empirical support. No researcher working in experimental settings (e.g. Edwards & Heubner, 1977; Kitchell et al., 1981) has reported the selection and boring of empty shells, despite the potential, ¡.e., naticids maintained in experimental chambers with both live prey and empty shells. Thirdly, if this were the case, one would expect many more multiply bored prey simply because empty shells are relatively common. (3) Another mechanism that would produce apparently multiply bored prey is abrasion of the remaining shell material in an incomplete borehole. In our laboratory studies, incom- plete boreholes are observed to result from active escape behaviors of the prey. In the field, stochastic interruptions of the predatory interaction during drilling would also result in an incomplete borehole. Incomplete boreholes may ‘weather out” producing a prey which is only apparently multiply bored. This is a logical possibility, particularly for species of Corbula (Fig. 1F) whose inner and conchiolin layers are frequently sloughed off. Incomplete boreholes may readily become “complete”. (4) A bored prey may escape consumption passively by accidental interruption of the predation process after boring but prior to feeding. We find this a satisfactory explana- tion for the rare occurrence of multiple com- plete boreholes in such sessile or extremely sluggish prey as corbulid bivalves (Ziegelmeier, 1954; Hoffman et al., 1974) or scaphopods (cf. Reyment, 1966; Hoffman et al., 1974). (5) Lastly, a bored prey may escape preda- tion actively after boring is completed but before feeding 15 initiated. We will now sub- stantiate this explanation for multiple com- plete boreholes in mobile prey. EXPERIMENTAL PRODUCTION OF MULTIPLE BOREHOLES Fossil Example. To our knowledge, there 1$ one exceptional species of naticid prey in the fossil record characterized by an extremely high frequency of multiple complete bore- holes. This is Strioterebrum monidum, first described by Woodring (1970) from collec- tions of the Miocene Bowden Formation, Jamaica, and Gatun Formation, Panama (Pliocene? N. Sohl, personal communication, 1983). Woodring remarked that “the extraor- NATICID PREDATORY BEHAVIOR 295 TABLE 2. Results of predation experiments on Terebra. Brackets indicate same predator individual-same prey individual interaction over time. Subscripts a, b, c refer to chronological series of predation events. Predator Prey Observation diam. (mm) height (mm) г.В ratio Borehole type Prey escaped 16.2 Sii 0.21 Nonfunctional Prey escaped 16.4 32.2 0.07 ¡ Nonfunctional Prey consumed 16.4 32.2 0.60 | Complete Prey escaped 16.9 21.0 0.35 Nonfunctional@ Prey escaped 16.9 21.0 0.45 | Nonfunctional® Prey consumed 16.9 21.0 0.58 Complete Prey escaped 16.9 29.7 0.18 Nonfunctional Prey consumed 16.9 32.7 0.59 Complete Prey escaped 18.5 223 0.12 Nonfunctional® Prey escaped 18.5 22.1 0.43 | Nonfunctional® Prey consumed 18.5 22.1 0.48 Complete Prey escaped 18.5 28.6 0.24 Nonfunctional Prey escaped 21.7 29.0 0.10 Nonfunctional Prey escaped Path 30.9 0.08 Nonfunctional Prey escaped 23.0 24.7 0.09 Nonfunctional@ Prey escaped 23.0 24.7 NA | Incomplete? Prey consumed 23.0 28.4 0.59 Complete Prey escaped 23.0 29.9 NA Incomplete? Prey escaped 23.0 29.9 0.45 | Nonfunctional® Prey escaped 23.0 31.2 0.21 Nonfunctional? Prey escaped 23.0 312 0.09 | Nonfunctional® Prey escaped 23.2 20.7 0.13 Nonfunctional Prey escaped 232 29.1 0.21 Nonfunctional@ Prey escaped 23.2 29.1 0.21 | Nonfunctional? Prey consumed 23.2 28.9 0.60 Complete Prey escaped 23.2 34.2 NA Incomplete Prey escaped 24.8 31.2 0.14 Nonfunctional Prey escaped 24.8 33:5 0.15 Nonfunctional Prey escaped 26.2 PET 0.22 Nonfunctional Prey escaped ZUR: 16.3 0.44 Nonfunctional Prey escaped 27.2 29.5 0.09 Nonfunctional Prey escaped 30.6 29.2 NA Incomplete® Prey escaped 30.6 29.2 0.26 Nonfunctional? Prey consumed 30.6 29.2 0.49 Complete dinary number of bore holes, made by an unknown predator, in some shells of this species from Jamaica and Panama, 1,000 kilometers apart, is noteworthy.” The senior author has more recently examined these exceptional assemblages (USNM #369347, #646047, USGS #135283). The number of boreholes per specimen, as well as the type of borehole and its diameter, are given in Table 1. The total number of boreholes per individual ranged from 1 to 15. Boreholes were of the complete, incomplete, and nonfunctional types (Fig. 1L). Outer borehole diameters, a measure of predator size (see Kitchell et al., 1981), ranged from 0.5 to 1.7 mm. Materials and Methods. In search of an explanation for these multiply bored gastro- pods, we attempted an experiment. Live Terebra (T. dislocata (Say); see Abbott, 1974; T. Bratcher, personal communication), a spe- cies similar in morphology to S. monidum, and live co-occurring Polinices duplicatus, a naticid, were collected in the Gulf of Mexico near Panacea, Florida. We maintained live prey and predators in plexiglass experimental chambers submerged in a 400 | Instant Ocean aquarium. Chambers contained 3-4 cm of fine sand. Temperature was maintained at 20°С. Photoperiod approximated a normal day-night cycle with seasonal variation (see Kitchell et al., 1981, and Boggs et al., 1984, for additional description of general laboratory conditions and observational methods). The Terebra ranged in size (height) from 16.3 to 48.8 mm and were placed in chambers with the predators. The predators ranged in size (maximum dimension) from 16.2 to 27.2 mm. 296 KITCHELL ET AL. Predators and prey were monitored regularly, and all predation attacks and prey escapes were recorded. At periodic intervals, prey were briefly removed from their chambers in order to assess the number, location, size, and type of boreholes. Consumed prey were removed from the system; live prey were returned to the system for continued expo- sure to predation. It is of interest that although only naticid predation occurred, not all subse- quent incomplete boreholes were character- ized by a central boss, a condition that closely resembles the fossil assemblage. Results. A number of live perforated Terebra were observed in the chambers (Ta- ble 2). In several cases, live Terebra had multiple perforations. These results conclu- sively demonstrate that Terebra is capable of escaping naticid predators, yet escape 1$ fre- quently not initiated until the borehole has perforated the shell (Fig. 1J, К, L, М, N). In all these cases, the prey successfully escaped before the predator could enlarge the inner diameter of the borehole for feeding. For example, a 18.5 mm predator was confined with a 22.1 mm prey. The first borehole was perforate but nonfunctional; the ratio of the inner borehole diameter (r) to the outer borehole diameter (R) was 0.12. The prey had escaped before the predator was able to complete the requisite inner borehole dimen- sion and was alive and active. We later ob- served a second functionally incomplete borehole in the same individual; the r:R ratio was 0.43. Again, the prey was alive, although the predator had achieved a greater degree of opening of the second inner borehole diame- ter. We then observed a third predation at- tempt, which was successful. The r:R ratio in this case was 0.48. In other examples, initial boreholes were incomplete, indicating prey escape occurred before perforation. The size range of bored (i.e. selected) prey, evidenced by a complete or incomplete borehole, ranged from 16.3 to 33.5 mm, indi- cating that predation attempts were confined to the smaller available prey sizes. Such size selection of prey would be predicted given the size range of predators (Kitchell et al., 1981). Discussion. Outer borehole diameter 1$ a function of predator size, and does not signif- icantly change dimensions over the duration of drilling. Inner borehole diameter, however, is obviously smallest at the time of initial perforation. Carriker & Van Zandt (1972) ob- served the formation of the inner borehole by muricid drilling gastropods: “As the diameter of the break at the bottom of the borehole approaches the diameter of the tip of the proboscis, the snail attempts to force the proboscis through the opening.” Such at- tempted forcing of the proboscis may be the stimulus which leads to Terebra's escape, so frequently coincident with perforation. Car- riker & Van Zandt continued: “This testing is repeated at the beginning of each rasping period, and sometimes at its termination, until the hole is large enough to admit the probos- cis. Hole boring is then discontinued, and the snail begins feeding.” As is evident from the number of prey successfully preyed upon by the predator, not all Terebra escape attempts are successful. It is noteworthy, however, that in all consumed prey, the final borehole was a typical complete borehole: the ratios of outer to inner borehole diameters were normal 055): These results indicate that multiple com- plete boreholes evidence successful prey es- cape, rather than “predation” on an empty shell. Prey such as Terebra can bear the scars of multiple predation attempts, including perforated boreholes, and yet be living; only complete boreholes with normal r:R ratios show mortality by predation. We conclude that multiple complete boreholes on single prey individuals are (i) in general, extremely rare; (ii) when frequent, usually associated with highly mobile prey. Mobility does not exclusively refer to gastro- pod prey. In laboratory trials using the bivalves Mya arenaria and Mercenaria mercenaria as prey, we have also observed active prey escape behavior that disrupted drilling as illustrated in Fig. 1A, В, С; (ili) in addition, we have suggested several passive mechanisms that may result in apparent mul- tiply bored prey. DO NATICIDS CONSUME PREY WITHOUT DRILLING? Although we cannot deny that some naticids consume some prey without drilling [e.g. Schneider’s (1981) report on Ensis pre- dation; also Vermeij, 1980], we question any such generalization. In particular, such gen- eralizations cannot be made reliably by refer- ence to either Medcof & Thurber (1958) or Edwards (1975), as has been the case. In the latter study of beach assemblages, for exam- ple, Edwards (1975) correctly reported the proportion of bored (naticid) prey to total or NATICID PREDATORY BEHAVIOR 297 available prey, a statement that was subse- quently misinterpreted (see below) to repre- sent the ratio of prey that naticids drill to those prey that naticids consume but do not drill. Edwards (1975), referring to this total beach assemblage, reported that “the overall value was 3/4ths bored”. This proportion refers to the percentage of total shells collected by Edwards that are bored, ¡.e. 3/4 of all shells are bored. This statement was misconstrued apparently to read that of all naticid prey, 3/4ths are bored and eaten and 1/4th are not bored but eaten. Taylor et al. (1980: 397), in reviewing naticid predation, reported, citing Edwards (1975), that “in some Recent spe- cies such as Polynices duplicatus ... about 25% of the prey are not bored.” Unfortu- nately, subsequent papers have cited this reference as the source for the statement that naticids do not drill all of their prey. The Medcof 8 Thurber (1958) situation 1$ somewhat different. Their Table V summa- rizes their results: 4,428 Mya arenaria were planted in 3 experimental plots with the naticid, Lunatia heros. Twelve days later, the plots were censused, and only 1,244 live Mya were recovered. One hundred and sixty eight empty, articulated Mya were also recovered. Of these, 88 were not drilled and 80 were drilled. This ratio of not drilled:drilled repre- sents 52.4%. Remarkable as it may seem, this observation led Medcof & Thurber to conclude that “drills kill more than half their prey without boring their shells.” Moreover, another 13 Mya had incomplete boreholes. These numbers (88 + 13/168) compute to 61.1%, representing the proportion of recov- ered, empty shells lacking complete bore- holes. Medcof & Thurber concluded that “an acceptable explanation is that about 60% of the time greater clam drills destroy soft-shell clams without perforating their shells” (р. 1366). The authors assigned all recovered but empty shells within the study plots to naticid predation. A more parsimonious ex- planation in the absence of any direct evi- dence 1$ that Mya experienced a high mortal- ity rate after being transported to the experimental plots. In our laboratory studies, for example, we routinely keep prey in holding chambers separate from the predators, yet prey die and rapidly decompose, leaving an unbored, empty articulated shell. Similarly, prey within predation chambers with Polinices duplicatus have been observed to die, gape, and decompose without the predator taking any part in the process. Thirdly, in chambers holding bivalves, naticids, and crabs, we have observed a crab successfully interrupt naticid drilling and consume the vulnerable bivalve. To summarize, in laboratory studies of nor- mal naticid predation, the evidence pre- ponderantly indicates that naticids drill shells of prey they consume (e.g. Edwards & Huebner, 1977; Kitchell et al., 1981, Ansell 1982; Boggs et al., 1984). Moreover, naticids drill live prey even when recently dead or moribund prey are available. We have me- chanically opened holes in M. mercenaria, for example, without harming the prey, and yet predators have not utilized these ready-made holes. We also have observed predators to initiate a new borehole even when an incom- plete borehole, produced by our deliberate interruptions of drilling, was available (Kitchell et al., 1981). CONCLUSIONS Multiple complete boreholes in mobile prey can be due to prey escape, as evidenced experimentally for Terebra, and can also re- sult incidentally from the weathering, abra- sion, or sloughing off of shell material as- sociated with incomplete boreholes. Nonfunc- tional boreholes evidence prey selection be- havior but do not evidence mortality by naticid predation. The incidence of mortality by pre- dation must be based on complete boreholes. Multiple complete boreholes do not pose a problem for models of prey selection. We conclude that naticid predators can readily distinguish live prey from empty shells, and that drilling is a highly stereotypic behavioral response of naticid gastropods to selected prey. REFERENCES ABBOTT, R.T., 1974, American seashells. Ed. 2. Van Nostrand Reinhold, New York, 663 p., 24 pl. ANSELL, A.D., 1982, Experimental studies of a benthic predator-prey relationship. Ш. Mala- cologia, 22: 367-375. BOGGS, C.H., RICE, J.A., KITCHELL, J.A. & KITCHELL, J.F., 1984, Predation at a snail's pace: what's time to a gastropod? Oecologia, 62: 13-17. CARRIKER, М.В. & VAN ZANDT, D., 1972, Pred- atory behavior of a shell-boring muricid gastro- pod. In WINN, H.E. 8 OLLA, B.L., ed., Behavior of marine animals, 1: 157-244. Plenum Press, New York. 298 KITCHELL ET AL. CARRIKER, М.В. & YOCHELSON, E.L., 1968, Recent gastropod boreholes and Ordovician cy- lindrical borings. United States Geological Sur- vey Professional Paper, 593-B: 1-23. DeANGELIS, D.L., KITCHELL, J.A., POST, W.M. & TRAVIS, C.C., 1984, A model of naticid gastro- pod predator-prey coevolution. Lecture Notes т Biomathematics, 54: 120-136. EDWARDS, D.C., 1975, Preferred prey of Polinices duplicatus in Cape Cod inlets. Bulletin of the American Malacological Union, 40: 17-20. EDWARDS, D.C. & HUEBNER, J.D., 1977. Feed- ing and growth rates of Polinices duplicatus preying on Mya arenaria at Barnstable Harbor, Massachusetts. Ecology, 58: 1218-1236. FURSICH, F.T. & JABLONSKI, D., 1984. Late Triassic naticid drillholes: carnivorous gastro- pods gain a major adaptation but fail to radiate. Science, 224: 78-80. HOFFMAN, A., PISERA, A. & RYSZKIEWICZ, M., 1974, Predation by muricid and naticid gastro- pods on the Lower Tortonian mollusks from the Korytnica clays. Acta Geologica Polonica, 24: 249-260. KITCHELL, J.A., 1982, Coevolution in a predator- prey system. Proceedings of Third North Ameri- can Paleontological Convention, 2: 301-305. KITCHELL, J.A., 1983, An evolutionary model of predator-mediated divergence and coexistence. Geological Society of America Abstracts with Programs, 15: 614. KITCHELL, J.A., BOGGS, C.H., KITCHELL, J.F. & RICE, J.A., 1981, Prey selection by naticid gas- tropods: experimental tests and application to the fossil record. Paleobiology, 7: 533-552. MEDCOF, J.C. & THURBER, L.W., 1958, Trial control of the greater clam drill (Lunatia heros) by manual collection. Journal of the Fisheries Re- search Board of Canada, 15: 1355-1369. NEWTON, C.R., 1983, Triassic origin of shell- boring gastropods. Geological Society of Amer- ica Abstracts with Programs, 15: 652-653. REYMENT, В.А., 1966, Preliminary observations on gastropod predation in the western Niger delta. Palaeogeography, Palaeoclimatology, Pa- laeoecology, 6: 45-59. SCHNEIDER, D., 1981, Escape response of an infaunal clam Ensis directus Conrad 1843, to a predatory snail, Polinices duplicatus Say 1822. Veliger, 24: 371-372. SOHL, N.F., 1969, The fossil record of shell boring by snails. American Zoologist, 9: 725-734. STANTON, R.J., Jr. 4 NELSON, P.C., 1980. Re- construction of the trophic web in paleuntology: community structure in the Stone City Formation (Middle Eocene, Texas). Journal of Paleontol- ogy, 54: 118-135. TAYLOR, J.D., MORRIS, N.J. & TAYLOR, C.N., 1980, Food specialization and the evolution of predatory prosobranch gastropods. Palaeon- tology, 23: 375—409. VERMElJ, G.J., 1980, Drilling predation on bivalves in Guam: some paleoecological implications. Malacologia, 19: 329-334. VERMEIJ, С... & DUDLEY, E.C., 1982, Shell repair and drilling in some gastropods from the Ripley Formation (Upper Cretaceous) of the southeastern U.S.A. Cretaceous Research, 3: 397-403. WOODRING, W.P., 1970, Geology and paleontol- ogy of Canal Zone and adjoining parts of Panama; description of Tertiary mollusks (Gas- tropods: Eulimidae, Marginellidae to Helmintho- glyptidae). [United States] Geological Survey Professional Paper 306-D: iii + 452 p., pl. 48—66. ZIEGELMEIER, E., 1954, Beobachtungen über der Nahrungswerwerb bei der Naticidae Lunatia nitida Donovan (Gastropoda: Prosobranchia) Helgoländer wissenschaftliche Meeresunter- suchungen 5: 1-33. Revised Ms. accepted 18 January 1985. MALACOLOGIA, 1986, 27(2): 299-305 ALGAL GARDENS AND HERBIVORY IN A SCAVENGING SANDY-BEACH NASSARIID WHELK S.A. Harris’, Е.М. da Silva', J.J. Bolton? & A.C. Brown" Departments of Zoology' and Botany”, University of Cape Town, South Africa 7700 ABSTRACT The shells of living Bullia digitalis (Dillwyn), a nassariid whelk common on the sandy beaches of the west and south coasts of South Africa, are colonised by a variety of algae. Some of these are typical of rocky shores but the most consistent invader is a green, filamentous, boring alga (Chlorophyta), the morphology of which corresponds to Eugomontia sacculata Kornm. Behavioural observations indicate that some or all of these algae supplement the predominantly carnivorous diet of the whelk. The gut contents and digestive glands of the animals are frequently green and it is demonstrated that this is due to the presence of chlorophyll a. Cellulolytic symbiotic bacteria, as well as a-amylase, cellulase and laminarinase activity, are shown to be present in the gut. It is concluded that В. digitalis ingests and utilises green algal material growing on its shell and thus plays a more complex role than previously thought in the sandy-beach food web. INTRODUCTION Виа digitalis (Dillwyn) is а nassariid whelk that is characteristic of high-energy sandy beaches along the west and south coasts of southern Africa. Our knowledge of this and other species of the genus has been reviewed by Brown (1982). The whelk is essentially a scavenger of washed-up animal matter, al- though it will turn predator on occasion. The supply of carrion to the beaches in question is highly erratic, however, and tends to be sea- sonal, while predation appears to be relatively uncommon. Thus, although the animal can consume food up to one third of its own tissue weight in a single meal (Brown, 1961), it seemed unlikely that carrion and prey could supply all its requirements. Colclough & Brown (1984) therefore investigated the pos- sibility of the animal making use of dissolved organic matter in the surrounding sea water to supplement its diet. The results proved posi- tive but it is clear that this source of nutrition could not by itself supply the needs of the whelk for an extended period. The possibility that the animals eat stranded plant material has never been completely rejected, as some members of the Nassariidae are known to graze plants (Kilburn & Rippey, 1982) and Nassarius obsoletus has been shown to be *From whom reprints may be obtained. an obligate omnivore (Curtis & Hurd, 1979). Nevertheless, this possibility has not been confirmed in the field, nor has algal material offered to captive whelks in the laboratory ever been eaten. Omnivory is not well documented among the Neogastropoda, with the exception of the Nassariidae. It has not been reported for the Bullia group. Of relevance to nutrition is the presence or absence of a crystalline style and gastric shield, structures which may also have systematic and evolutionary significance. Yonge (1930) considered that “the crystalline style of Mollusca and a carnivorous habit cannot normally co-exist” and indeed among Neogastropoda crystalline styles appear to be virtually confined to the omnivorous Nas- sariidae (Kato & Kubomura, 1955; Curtis € Hurd, 1979). Dissection of the gut of Bullia digitalis and of B. rhodostoma and serial sections of the former species (H. du Preez, S.A. Harris 4 A.C. Brown, unpubl.) have failed to reveal the presence of a style, although it is possible that a style is only transitorily present or shows daily cycling, as has been demon- strated in Nassarius obsoletus (Cuntis, 1980). Recently, da Silva 8 Brown (1984) have reported the consistent presence of algae associated with the shells of living B. digitalis and have described behaviour suggesting (299) 300 HARRIS ET AL. FIG. 1. Light micrograph of a shell fragment of Bullia digitalis ( x 200), showing branching filaments of the green alga Eugomontia embedded within it. that the animal periodically crops this “gar- den” with its long, mobile proboscis and in- gests the algal material. Contributory circum- stantial evidence is that the gut contents and digestive glands are frequently green. The aims of the present work were to establish the nature of the alga or algae associated with the shell, to discover whether the green colour of the digestive system 1$ due to ingested chlorophyll and to assess the digestive capabilities of the animal with re- gard to utilising algal material. THE ALGAL GARDEN Examination of intact shells and shell frag- ments of Bullia digitalis by light microscopy revealed the consistent presence of an exten- sive growth of a green filamentous, boring alga embedded within the outer layer of the shell (Fig. 1). Contact between the alga and the exterior was maintained through numer- ous small holes, some 7 ym in diameter (Fig. 2). lt was found possible to isolate the alga from the shell by treatment with dilute HCI. It consists of uniseriate filaments displaying an irregular but predominantly opposite pattern of branching. The diameter of the algal fila- ments is 5 to 7 um. The vegetative features thus correspond to those of Eugomontia sac- culata Kornmann, a boring species found in both living and dead shells in temperate re- gions in both hemispheres (Kornmann, 1960; Wilkinson 8 Burrows, 1972a, b; South 4 Adams, 1976). Scanning electron microscopy of shell frag- ments fixed in 2% gluteraldehyde, critical point dried after dehydration and coated with gold-palladium, showed that the algal fila- ments bore only into the outer prismatic layer and ramify throughout the crystalline matrix of this layer (Fig. 3). Culture of the alga was attempted by keep- ing shell fragments in an enriched seawater medium (ES of Provasoli, 1968) at 15°С and а photoperiod of 16 hr at a light intensity of 50 to 60 Em 2.5 *. This proved successful in that after some weeks isolated patches of the alga were observed growing on the bottom of the Petri dish, presumably resulting from re- leased spores. Sporangia were not observed HERBIVORY IN A SANDY-BEACH WHELK 301 FIG. 2. Scanning electron micrograph of the shell surface of B. digitalis, with holes caused by the boring alga (х 1670). The tips of the filaments are situated in the holes, just below the surface. on the shell surface due to dense growth of other algae which flourished in the culture medium. These algae, which were initially present on the shell in low numbers, were typically rocky-shore forms comprising spe- cies of Ectocarpus, Enteromorpha and Ulva. Their possible use as food cannot be ignored. DIGESTIVE GLAND CHLOROPHYLL ANALYSIS The gut contents and digestive glands of B. digitalis vary considerably in colour. They are frequently brown or grey and are bright blue after the animals have been feeding on the siphonophore Physalia. The green colour al- ready referred to is also common, particularly in individuals from the west coast during winter. Green digestive glands from four whelks from the west coast were each homogenised in 12 ml 90% acetone and the extracts ultrasonified for 30 min in the dark on ice. The samples were centrifuged at 9,000 rpm for 15 mins at 15°C and the supernatant analysed on a spectrophotometer linked to a Spectro- printer. The four samples gave very similar results, showing absorbance peaks in the 660—663 nm and 400—410 nm regions. These maxima correspond to those of chlorophyll a (Bogorad, 1962; Round, 1973) and we con- clude that the green colour observed is in- deed due to chlorophyll resulting from the ingestion of green algal material. CARBOHYDRASES Material and methods: To test for cellulolytic bacteria in the gut of B. digitalis, a modification of the method of Teather & Wood (1982) was employed. The whole gut, without digestive gland, was homogenised in 600 yl sea water and the crude supernatant sub- jected to a dilution series. 10 to 20 ul of the diluted bacterial suspension were then plated onto a growing agar medium (GAM) (2% 302 HARRIS ET AL. FIG. 3. Algal filaments ramifying through the crystalline matrix of the outer prismatic layer of the shell of B. digitalis (SEM x 1670). TABLE 1. The carbohydrase activities obtained from the Nelson-Somogyi enzyme assays (n = 10). Enzyme activity (mg glucose:mg protein *-hour?) Substrate Enzyme Х = SD: % Total activity Glycogen a-amylase 0.4256 + 0.0922 39.31 Starch a-amylase 0.4113 = 0.1016 37.99 СМС cellulase 0.1511 = 0.0634 13.96 Laminarin laminarinase 0.0946 + 0.0822 8.74 agar, 0.5% peptone and 0.1% yeast extract). Colonies of pure bacterial isolates were stabbed onto a final plate medium consisting of 2.5% w/v agar containing 0.1% w/v pep- tone and 0.01% carboxymethyl cellulose. The plates were incubated at 23°C for four days and then flooded with an aqueous solution of Congo Red for 15 mins. This stain reacts with B-D-glucans in the CmC substrate, providing a rapid and sensitive assay for bacterial strains possessing ß-D-glucanohydrolases (Teather & Wood, 1982). The Congo Red was then poured off and the visualised zones of ß-D-glucans hydrolysis stabilised by flooding with 1N HCL for 15 mins; this changes the dye colour to blue and inhibits further enzyme activity. Results were recorded photographi- cally. Scanning electron micrographs were prepared both of bacteria from the colonies growing in culture and of the bacteria in situ in the animal's gut. In a further series of experiments, homo- HERBIVORY IN A SANDY-BEACH WHELK 303 genised whole guts were centrifuged in phos- phate buffer (pH 7) and the supernatant subjected to cellulase, a-amylase and lam- inarinase determinations, using the Nel- son-Somogyi colorimetric method (Nelson, 1944; Somogyi, 1952). Attempts to first elim- inate enzyme-producing bacteria with В-тег- captoethanol or Ampicillin had to be aban- doned, as both interfered with the assay. In order to obtain comparative results, each sample was analysed for total protein content, using the Folin Ciocalteus (2N) reagent and following the method of Lowry et al. (1951). Crystalline bovine albumin was used as a standard. Results of the enzyme analysis were expressed in mg glucose evolved. mg protein !. hour !. Results: Isolated bacterial colonies ap- peared on the GAM plates within one to three days. About 10% of the bacteria were cellulolytic, as was clear from the visualised zones of ß-D-glucans. The bacteria were of the agarose-eating type, identifiable by de- pressions in the agar. SEM examination showed the pure bacterial isolates to be rod- like, while those photographed in the gut were predominantly coccoid. These could, how- ever, be the same bacteria under different conditions. Results of the carbohydrase assays are given in Table 1. All three activities tested were positive, at levels indicated by the amount of reducing sugar evolved from each substrate. Thus a-amylase activity (0.837 mg glucose.mg protein ‘hr ') accounted for some 77% of the total activity measured, while cellulase and laminarinase accounted for only 14% and 9% respectively. DISCUSSION Among marine invertebrates, the hydrolytic capacity of the carbohydrases is generally greater with respect to reserve carbohydrates (starch, glycogen, laminarin) than to structural carbohydrates (cellulose and chitin) (Elya- kova etal., 1981). The present study supports this with regard to a-amylase and cellulase but not with regard to laminarinase. It is difficult to compare our work quantitatively with other published data, because of the variety of techniques employed, but qualita- tive comparisons can be made. Stone & Morton (1958) studied the distribu- tion of carbohydrases in molluscs having a wide range of feeding habits. They found a-amylase activity to be present among both herbivores and carnivores, in accordance with the ability to digest starch and glycogen respectively. There was minimal cellulase ac- tivity but high levels of laminarinase in the carnivorous whelks, a finding confirmed for Nassarius reticulatus by Kristensen (1972). Slight hydrolysis of alginic acid and alginate was also detected in this species, although the animal apparently never feeds on brown algae. К was suggested by Stone & Morton (1958) that, far from being strictly functional, cellu- lases may form part of the basic digestive enzyme system in the Mollusca, and this concept has been supported by subsequent work. Yokoe 8 Yasumasu (1964) proposed that the distribution of cellulases in inverte- brates generally is more closely correlated with phylogeny than with feeding habits, while Gianfreda et al. (1979) were unable to explain the significance of the В-1, 4-glucanase (cellulase), found in most carnivorous mol- luscs examined, on functional grounds. They inclined to the view that it simply represents an evolutionary remnant from an originally herbivorous stock. This view is supported by Agnisola et al. (1981). The presence of these carbohydrases in Bullia digitalis is thus no proof, by itself, that the animal ingests plant material; on the other hand, it is a good indication that the whelk can utilise such material if it is ingested. Production of cellulase by symbiotic bacte- ria in the gut is well known among the Mollusca. In most cases it is not certain, however, whether the host animal also pro- duces such enzymes, although this is clearly the case in marine borers. Morton (1978) has suggested that in these animals the bacteria provide the enzymes for the initial breakdown and that this is followed by digestion within the cells of the animal's digestive diverticula, using enzymes produced by the animal. In the present work, the presence of cellulose-di- gesting bacteria in Bullia digitalis has been demonstrated but whether the animal itself also produces such enzymes is not clear, nor is it known to what extent digestion may be intra-cellular. However this may be, the presence of chlorophyll in the digestive system, together with enzyme activity appropriate to the digestion of algal material, leads to the firm conclusion that the whelk can and does sup- plement its predominantly carnivorous diet with algae. It thus plays a more complex role 304 HARRIS ET AL. than previously thought in the sandy-beach food web. Less certain is the source of this algal material. Macrophytic algae do not occur on high-energy sandy beaches (Brown, 1964) and there is no hint, after nearly three de- cades of observation, that Bullia eats stranded plant material. The only alternative would appear to be an algal “garden” growing on the shell itself. Cropping of such a garden has been witnessed (da Silva & Brown, 1984) and subsequently confirmed but whether it is the embedded Еидотопйа or other green algae growing on the shell which are being ingested 15 not certain. The boring alga grew only very slowly in culture and according to Kornmann (1960) only the reproductive struc- tures of Eugomontia sacculata protrude from the shell. It is thus questionable whether the small amount of material made available could be of any importance in the diet of the whelk or account for the large amounts of chlorophyll found in the digestive system. The same applies to the non-boring algae discov- ered on the shell, for these were very sparse until they were cultured. It may, however, be that field conditions, such as the buffetting of the shell by waves and sand, stimulate algal growth or even that the whelk itself encourages the growth of the algae in some way. Since the whelk com- monly buries itself just below the sand surface (Brown, 1982) burial is unlikely to adversely effect algal growth, as light is known to pen- etrate several centimeters through sand, many algae flourishing while buried in sand for long periods (Bally et al., 1984). ACKNOWLEDGEMENTS Mr. Klaus Schultes and Miss Jean Harris assisted with the preparation of scanning electron micrographs, while Mr. Neville Eden prepared the photographs for publication. Miss Cathy Roberts gave advice on the bac- terial section of the work. We are indebted also to Dr. Robert Robertson for notes on the extent of herbivory and the occurrence of crystalline styles т neogastropods. The project was supported by a postgraduate CSIR grant made available to the first author. REFERENCES CITED AGNISOLA, C., SALVADORE, S. & SCARDI, V., 1981, On the occurrence of cellulolytic activity in the digestive gland of some marine carnivorous molluscs. Comparative Biochemistry and Phys- iology, 70B: 521-526. BALLY, R., MCQUAID, C.D. & BROWN, A.C., 1984, Shores of mixed sand and rock: an unexplored marine ecosystem. South African Journal of Sci- ence, 80: 500-503. BOGORAD, L., 1962, Chlorophylls. In LEWIN, R.A. (ed.), Physiology and biochemistry of Algae, Academic Press, р. 395—408. BROWN, А.С., 1961, Physiological-ecological studies on two sandy-beach Gastropoda from South Africa: Bullia digitalis Meuschen and Виа laevissima (Gmelin). Zeitschrift fúr Morphologie und Okologie der Tiere, 49: 629—657. BROWN, A.C., 1964, Food relationships on the intertidal sandy beaches of the Cape Peninsula. South African Journal of Science, 60: 35—41. BROWN, A.C., 1982, The biology of sandy-beach whelks of the genus Bullia (Nassariidae). Ocean- ography and Marine Biology Annual Review, 20: 309-361. COLCLOUGH, J.H. & BROWN, A.C., 1984, Uptake of dissolved organic matter by a marine whelk. Transactions of the Royal Society of South Africa, 45: 169-176. CURTIS, L.A., 1980, Daily cycling of the crystalline style in the omnivorous, deposit-feeding estua- rine snail //yanassa obsoleta. Marine Biology, 59: 137-140. CURTIS, L.A. 8 HURD, L.E., 1979, On the broad nutritional requirements of the mud snail, llyanassa (Nassarius) obsoleta (Say), and its polytrophic role in the food web. Journal of Experimental Marine Biology and Ecology, 41: 289-297. DA SILVA, Р.М. & BROWN, A.C., 1984, The gar- dens of the sandy-beach whelk Bullia digitalis (Dillwyn). Journal of Molluscan Studies, 50: 64—65. ELYAKOVA, L.A. SHEVCHENKO, NM. & AVAEVA, S.M., 1981, A comparative study of carbohydrase activities in marine invertebrates. Comparative Biochemistry and Physiology, 69B: 905-908. GIANFREDA, L., IMPERATO, A., PALESCAN- DOLO, R. & SCARDI, V., 1979, Distribution of B-1,4-glucanase and B-glucosidase activities among marine molluscs with different feeding habits. Comparative Biochemistry and Physiol- ogy, 63B: 345-348. KATO, K. & KUBOMURA, K., 1955, Uric acid and guanine excretion from the crystalline style sac in some gastropods. Science Reports of the Saitama University, ser. B, 2: 21-34. KILBURN, R. & RIPPEY, E., 1982, Sea shells of southern Africa. Macmillan, South Africa, 247 p. KORNMANN, V.P., 1960, Die heterogene Gattung Gomontia. Il: Der fädige Anteil, Eugomontia sacculata (nov. gen., nov. spec.). Helgoländer wissenschaftliche Meeresuntersuchungen, 6: 229-238. KRISTENSEN, J.H., 1972, Carbohydrases of some HERBIVORY IN A SANDY-BEACH WHELK 305 marine invertebrates with notes on their food and on the natural occurrence of the carbohydrates studied. Marine Biology, 14: 130-142. LOWRY, O.H., ROSEBROUGH, N.J., FARR, A.L. 8 RANDALL, R.J., 1951, Protein measurement with the Folin phenol reagent. Journal of Biolog- ical Chemistry, 193: 265-275. MORTON, B., 1978, Feeding and digestion in shipworms. Oceanography and Marine Biology Annual Review, 16: 107-144. NELSON, N., 1944, A photometric adaptation ofthe Somogyi method for the determination of glu- cose. Journal of Biological Chemistry, 169: 375-380. PROVASOLI, L., 1968, Media and prospects for culturation of marine algae. WATONABE, A. 4 HATTORI, A. (ed.), In Cultures and collections of Algae. Japanese Society of Plant Physiology, р. 47-49. ROUND, F.E., 1973, The biology of the Algae. Arnold, London, 278 p. SOMOGYI, M., 1952, Notes on sugar determina- tion. Journal of Biological Chemistry, 195: 19-23. SOUTH, G.R. & ADAMS, N., 1976, The marine algae of the Kaikaura coast. National Museum of New Zealand Miscellaneous Series, 1: 67 p. STONE, В.А. & MORTON, J.E., 1958, The distri- bution of cellulases and related enzymes in Mollusca. Proceedings of the Malacological So- ciety of London, 33: 127-141. TEATHER, В.М. & WOOD, P.J., 1982, Use of Congo Red-polysaccharide interactions in enu- meration and characterization of cellulolytic bac- teria from the bovine rumen. Applied Environ- mental Microbiology, 43: 777-780. WILKINSON, М. & BURROWS, E.M., 1972a, The distribution of marine shell-boring green algae. Journal of the Marine Biological Association of the United Kingdom, 52: 59-65. WILKINSON, М. & BURROWS, E.M., 1972b, An experimental taxonomic study of the Algae con- fused under the name Gomontia polyrhiza. Jour- nal of the Marine Biological Association of the United Kingdom, 52: 49-57. YOKOE, Y. & YASUMASU, 1., 1964, The distribu- tion of cellulase in invertebrates. Comparative Biochemistry and Physiology, 13: 323-338. YONGE, C.M., 1930, The crystalline style of the Mollusca and a carnivorous habit cannot nor- mally co-exist. Nature, 125: 444-445. Revised Ms. accepted 18 March 1985 MALACOLOGIA, 1986, 27(2): 307-311 THE AUTOTOMY ESCAPE RESPONSE OF THE TERRESTRIAL SLUG PROPHYSAON FOLIOLATUM (PULMONATA: ARIONIDAE) |. Deyrup-Olsen, А. W. Martin 8 В.Т. Paine Department of Zoology, University of Washington, Seattle, Washington 98195, U.S.A. ABSTRACT The autotomy escape response of the terrestrial slug Prophysaon foliolatum (Gould, 1851) is mediated by a peripheral reflex mechanism, with sensory components restricted to the region directly involved in autotomy (autotomy zone and autotomy section of the foot). The response has two major components: (1) Muscle contraction severs the posterior autotomy section of the foot from the rest of the body. (2) Mucus is secreted differentially, with sticky, yellow тисиз over the anterior body, and relatively non-viscous, colorless mucus over the autotomy section. The autotomy section is packed with glycogen cells. If food is lacking, the stored nutrients are utilized by the slug. The autotomy section offers a significant reward to a natural predator, such as the carabid beetle Scaphinotus angusticollis. Key words: Prophysaon foliolatum; Pulmonata; autotomy; escape response; mucus; glycogen cells. INTRODUCTION Since the early observations of Raymond (1890), Pilsbry & Vanatta (1898), and others it has been known that terrestrial slugs of the genus Prophysaon can autotomize the “tail” (posterior section of the foot). Observations of Prophysaon andersoni were reported by Hand & Ingram (1950), who noted that there is considerable variation between individuals in responsiveness to stimulation of the tail by cutting or puncture, and that the tail section can be regenerated. They demonstrated that attack by predators normally occurring with Prophysaon (the carabid beetle Scaphinotus sp.; the snail Haplotrema minimum) could initiate autotomy. They suggested that auto- tomy in Prophysaon is a mechanism of de- fense. Indeed, a wide variety of molluscs are known to effect defensive autotomy (literature reviewed by Stasek, 1967), and the process is known to occur in some arthropods, verte- brates, and other animals as well. Autotomy of the tail in lizards in response to attack by predators has been studied fairly extensively in relation to energetics and adaptive strate- gies (Vitt et al., 1977; Ballinger, 1981). In contrast, little is known of the physiology of the process in molluscs. Accordingly, we have studied structural and functional aspects of autotomy in Prophysaon foliolatum (Gould, 1851; Arionidae), a slug in which the re- sponse is readily and reproducibly elicited. METHODS AND MATERIALS Slugs were collected on Tatoosh Island, off the northwest coast of Washington State, in summer and autumn. The animals were maintained in the laboratory at 10°C, in plastic boxes on moist paper towels with food (let- tuce, potatoes, carrots, or yams) present at all times. Eggs laid in September or October were allowed to hatch and the young were reared to maturity. Both animals collected in the field and those reared in the laboratory were used in the investigation; no systematic differences were noted between these two groups. Experiments were carried out with two ma- jor objectives: (1) Investigation of the nature and localization of stimuli which elicited the autotomy response, and the general site of integration of the response (whether periph- eral or central); (2) Description of the struc- ture and composition, with respect to stored nutrients, of the autotomized part of the foot. Observations of autotomy were made both on intact slugs and on ап in vitro preparation of the posterior body wall (Deyrup-Olsen & Mar- tin, 1982). The latter was prepared as follows. The body was rapidly severed about halfway between the anterior and posterior margins of the mantle and the cerebral ganglia were destroyed. The viscera were then withdrawn from the posterior portion of the body, leaving a sac of epidermis and underlying muscle (307) 308 DEYRUP-OLSEN, MARTIN 8 PAINE terminated by the autotomy section (AS; this part of the body was termed a “tail” by earlier workers). The sac, or posterior chamber, was then attached to a glass manometer tube, while a fine plastic tube, threaded through the manometer tube to the lower end of the chamber, served as a portal of entry of in- jected materials. The chamber was filled with Ringer solution (slightly modified from Roach, 1963, in accord with our measurements of the composition of Prophysaon blood; the os- motic pressure was increased by 10%, and the Ca concentration raised from 3 mM to 5mM). The preparation could then be stimu- lated mechanically or electrically, or sub- jected to the action of a variety of agents, especially neurotropic agents known to func- tion in gastropod physiology. Stimuli tested were as follows: (1) Electri- cal pulses delivered from a Grass Instru- ments Stimulator Model $4, at varied strength and frequency up to 10 V and 20 Hz. (2) Injections into the intact slug or the posterior chamber of Ringer solution containing 1 to 10 mol of one of the gastropod neurotransmit- ters: acetylcholine, 5-hydroxytryptamine, dopamine, gamma-aminobutyric acid, nor- adrenaline, FMRFamide (L-phenylalanyl-L- methionyl - L - arginyl - L- phenylalaninamide), octopamine, or histamine; and neural block- ing agents: atropine sulfate and hexamethon- ium bromide (block acetylcholine receptors), propanolol and phentolamine (block ad- renergic receptors), and cyproheptidine (blocks 5-hydroxytryptamine receptors in some systems). All test substances were ob- tained from Sigma Chemical Company, St. Louis. (3) Mechanical stimuli: In applying this type of stimulation, we attempted to simulate the attack on Prophysaon foliolatum by the carabid beetle, Scaphinotus angusticollis. The beetles were collected in the same area as the Prophysaon individuals, and were placed in closed containers with individual slugs for observation of the interaction be- tween them. The acts of predation by the beetle were highly variable, although a beetle generally attacked a slug within seconds to an hour of access to it. The attacks consisted of several to scores of mandibular bites deliv- ered at first at apparently random sites on the slug’s body. The bites were shallow and were never observed to draw blood. Although in half of the 10 trials of slug-beetle pairs the beetle failed to cause autotomy within a 12 to 24 hour observation period, in 5 cases the slug autotomized. In each test in which the beetle succeeded in causing the response, it was biting and often tugging on the autotomy section or at the autotomy zone. The beetle seized the AS and held it against its mouth for several hours; at the end of this time the beetle had ingested the core tissue and only a limp sac of muscle and epidermis remained. Meantime, the slug withdrew from the field of action, or was removed by the experimenter. Basing our technique on these observations, we stimulated the body wall with repeated shallow pinches with a pair of iris forceps, a type of stimulation referred to as “beetle” in this paper. Because Hand & Ingram (1950) used puncture and cutting of the AS in Prophysaon andersoni, we also tested the effect of application of stimulation with a sur- gical towel clamp with sharp points, penetrat- ing the body wall and causing loss of a small amount of blood. This type of stimulation will be referred to as “penetrating.” The signifi- cance of differences between the effects of different stimuli was assessed with the f test. Direct and microscopic observations were made of the mucus output of the body wall resulting from effective stimulation. The ob- servations were supplemented with study of fresh sections of body tissue stained for gly- cogen (Lugol's reagent) and of histological preparations (Bouin's fixation, paraffin sec- tions, hematoxylin-eosin staining). The mass and water content of the AS, relative to the body as a whole, were measured (Mettler analytical balance; drying at 105°C to con- stant weight). Chemical analyses of the AS, following freeze-thawing, included tests for glucose (glucose kit, Boeringer Mannheim; based on glucose oxidase oxidation of glu- cose, with peroxidase conversion of the chromogen 2,2'-azino-di(3-ethylbenzthiazo- line)-6-sulfonate to the form absorbing at 575 nm); glycogen (Murat & Serfaty, 1974); galactogen (hydrolysis followed by tests for galactose with a kit supplied by Boeringer Mannheim (based on B-galactose dehydro- genase oxidation, and measurement of NADH absorbance at 356 nm); and soluble protein (Lowry et al., 1951). RESULTS In a typical autotomy response, induced by Scaphinotus or by our technique of beetle stimulation, the sequence of events was fairly stereotyped. If the attack was anterior to the AUTOTOMY IN THE SLUG PROPHYSAON FOLIOLATUM 309 Foot | | Autotomy Section | - | 1 Stimuli causing autotomy of whole animal (W.) or chamber (P.C.) Bester | Penetrating | Penetrating : autotomy autotomy 11+4 (8) W. 43+25 (n= 14) м. 43+25 (n=7) p.c. 9+5 (n=6) w. 2+1 (n=6) p.c. FIG. 1. Number of stimuli, of varying type and site of application, resulting in autotomy in Prophysaon foliolatum. Autotomy occurred only when stimuli were applied to the autotomy zone or autotomy section; only the whole animal, with beetle type stimulation, was tested at the autotomy zone. Stimuli were more effective at the autotomy zone than the autotomy section (P < 0.05). Penetrating stimuli were more effective than beetle stimulation in the whole animal (P < 0.05) and the posterior chamber (P < 0.01). The whole animal and posterior chamber did not differ in sensitivity of the autotomy section to beetle stimuli, but penetrating stimuli were significantly less effective (P < 0.05) in the whole animal than in the posterior chamber. AS, a golden (or yellow-green), sticky mucus was secreted at the site of attack. Attack on the AS itself elicited locally a colorless, thin secretion. We noted that a beetle, if biting the anterior part of the body, frequently withdrew and wiped its mandibles vigorously on the substrate, clearing them of adherent mucus. After a highly variable number of pinches, if these were delivered at or posterior to the autotomy zone (junction between the anterior body and the AS), the following events took place: the autotomy zone contracted sharply; the AS appeared to swell slightly, as if en- gorged with blood; and the anterior body secreted copiously its characteristic yellow mucus. Autotomy was then completed rap- idly, in less than 2 to about 5 sec. Following autotomy, the AS remained in place; we never observed the crawling movement of the AS described by Hand & Ingram in Prophysaon andersoni. However, in our ex- periments a slight swaying motion of the AS was noted and its surface displayed shallow, puckering movements. Among the stimuli tested, only mechanical stimuli resulted in autotomy. Electrical and chemical stimuli, although applied in widely ranging intensity and at varying sites, on or in the body, never caused autotomy. The т vitro preparation responded to mechanical stimu- lation in a manner indistinguishable from the responses of the intact animal, indicating that the head ganglia were not involved directly in the reflex. Furthermore, in no case did autotomy result from stimulation of the ante- rior region of the body. The most sensitive region for triggering the response was the autotomy zone, while repeated stimulation of the AS itself also induced the response. Mechanical stimulation involving penetration of the body wall was significantly more effec- tive than the superficial stimulation character- istic of beetle bites. These results are sum- marized in Fig. 1. Although stimulation anterior to the autotomy zone did not induce autotomy, in some instances it appeared to accelerate the response when stimuli were shifted to the AS. In addition, with penetrating stimulation of the AS, more stimuli were re- quired in the intact animal than in the in vitro system. Thus, the anterior part of the body does exert some control over the AS. We used the in vitro posterior chamber preparation to investigate whether adminis- tration of neurotropic agents (neurotransmitt- ers and blocking agents) could accelerate or otherwise alter the autotomy response. In no case was acceleration seen, and the only 310 DEYRUP-OLSEN, MARTIN 8 PAINE agent which appeared to have any effect, 5-hydroxytryptamine (1 to 10 jmol/test), gave inconsistent results, ranging from no signifi- cant effect (8/13 trials) to pronounced inhibi- tion (5/13 trials) of the response. Injections of agents into the anterior hemocoel or AS of intact slugs also were, generally, without ef- fect. However, a single agent—atropine sul- fate—infiltrated into the AS (2.5 mol in 0.25 ml slug Ringer solution) totally suppressed the autotomy response. The results indicate that the reflex involves a cholinergic mecha- nism (muscarinic; hexamethonium, a nicotinic blocking agent, was without effect). In addi- tion, it could be concluded that the neuro- muscular mechanism for autotomy 1$ not readily accessible to chemical influences from the anterior body nor, in general, from the AS itself. The differences between the mucus pro- duced by the AS and that produced by the rest of the body were associated with differ- ences between the epithelial mucus cells in the two regions. On the mantle and back to the autotomy zone very large yellow cells were present; their dense distribution around the mantle margin gave this region a golden color. On effective stimulation they released elliptical vesicles which burst rapidly and re- leased a dense yellow mucus characterized by many granules. The AS and foot sole also had mucus cells but their secretions were released in relatively stable vesicles and ap- peared thin, non-granular, and colorless. This mucus was similar to that of the arionid slug Ariolimax columbianus (Deyrup-Olsen et al., 1983). The AS, ranging from 200 to 500 mg wet weight, comprised 11% (sd + 2, п = 11) of the total body weight of adult (reproductively capable) slugs. It is made up chiefly of spher- ical cells (diameter 50 to 70 um); these stain deep red with iodine-potassium iodide solu- tion in the characteristic reaction for glycogen. We conclude that they are glycogen cells, widely distributed storage elements of gastro- pod connective tissue (Joosse 4 Geraerts, 1983). The cells are embedded in a mesh- work of muscle cells and connective tissue. In animals starved for 13 days the glycogen cells appeared to be depleted. The tissue making up the AS had relatively high contents of water (89.4%, sd + 2.3, п = 6, of total tissue weight), soluble protein (13.5%, sd + 2.5, n = 6, of dry weight) and glycogen (17.5%, sd + 3.6, n = 5, of dry weight), and no traces were found of galactogen, a storage carbohydrate used by gastropods in reproduction. Thus the central core of the AS provides a good source of water, protein, and carbohydrate to a predator such as Scaphinotus angusticollis (the weights of these beetles varied, averaging about 100 mg). Paired blood vessels run longitudinally through the AS and injections of colored materials (e.g., India Ink) into the anterior hemocoel resulted in rapid coloration of the AS. Thus, there is effective blood circulation between the AS and the anterior body—of obvious importance in the supply of nutrients for storage in the AS. However, autotomy did not result in significant loss of blood from the anterior body, which was rapidly and effec- tively sealed off at the autotomy zone by contraction of its sphincter muscle (Hand & Ingram, 1950). In accord with the report by Hand & Ingram (1950), the slugs proved able to regenerate the AS in a few weeks. In our laboratory conditions the process of autotomy and re- generation was repeated 3 successive times in 8 slugs, and perhaps could have occurred further had it appeared useful to extend the observations. Young Prophysaon foliolatum individuals, within a day or two of hatching (body size less than 10 mg, compared with the adult range of about 2 to 10 g) showed fully coordinated autotomy responses with temporal character- istics similar to those of the adult. DISCUSSION The escape reaction of Prophysaon foli- olatum 1$ complex, involving diverse struc- tural and physiological mechanisms. Mainte- nance of stored water and nutrients in the AS, which offers a significant reward to predators such as carabid beetles, must depend on coordination of metabolism and circulation. The neuromuscular reflex severing the AS from the body as a whole is entirely periph- eral, and 1$ triggered specifically by receptors located within the autotomy zone and the AS. This confirms the work of others, indicating that the sensory zone for autotomy in a vari- ety of molluscan species 1$ localized in the autotomized part (Stasek, 1967). Information from these sensitive areas in Prophysaon is also transmitted anteriorly, since mucus se- cretion over the body as a whole acompanies autotomy. The yellow color and dense, sticky AUTOTOMY IN THE SLUG PROPHYSAON FOLIOLATUM 311 quality of the anterior mucus may tend to divert the attack by predators from the ante- rior body towards the dispensable AS. Whereas the slug's behavior—moving away when attacked—indicates that the an- terior ganglia (brain) register events occuring in the AS, the brain appears to be unneces- sary for autotomy. Such decentralization of important neuromuscular functions is well known in molluscan physiology, as in the highly coordinated movements of the isolated foot in slugs (Deyrup-Olsen & Martin, 1982) and of severed tentacles in cephalopods (Lucas, cited by Stasek, 1967). The response depends on cholinergic neurons. The quality of stimulation—penetrating or shallow—af- fected the speed of the response, but our work with neurotropic agents offered no clues as to the nature of this differentiation. Indeed, it was surprising that the autotomy response appeared to be so refractory to such sub- stances which, in our tests, frequently caused strong stimulation of the general body mus- culature and mucus cells. Overall, it may be concluded that the autotomy response is a highly organized and stable mechanism of defense. Its persistence in the genus Prophysaon, despite its presumed high cost, suggests that the response contributes signif- icantly to survival of slugs in nature. ACKNOWLEDGEMENTS We thank colleagues in the Department of Zoology, University of Washington, for the following assistance: J. S. Edwards, for iden- tification of the carabid beetle; E. Plisetskaya, for glycogen analyses; P. M. Brunner, for histology. LITERATURE CITED BALLINGER, R. E., 1981, Can predator defense be tributive or toxins non-toxic? American Natural- ist, 117: 794-795. DEYRUP-OLSEN, 1., LUCHTEL, D. L. & MARTIN, A. W., 1963, Components of mucus of terrestrial slugs (Gastropoda). American Journal of Physi- ology, 235: R448-452. DEYRUP-OLSEN, I. & MARTIN, A. W., 1982, Sur- face exudations in terrestrial slugs. Compara- tive Biochemistry and Physiology, C, Compara- tive Pharmacology, 72: 45-51. HAND, С. & INGRAM, W. M., 1950, Natural history observations on Prophysaon andersoni (J.G. Cooper), with special reference to amputation. Bulletin of the Southern California Academy of Sciences, 49: 15-28. JOOSSE, J. & GERAERTS, W. P. M., 1983, in SALEUDDIN, A. S. M. & WILBUR, K. M., ed., The Mollusca, 4: 317-406. Academic Press. LOWRY, O. H., ROSEBOUGH, N. R., FARR, L. A. 8 RANDALL, В. J., 1951, Protein measurement with the Folin phenol reagent. Journal of Biolog- ical Chemistry, 193: 265-275. MURAT, J. С. 8 SERFATY, A., 1974, Simple enzymatic determination of polysaccharide (gly- cogen) content of animal tissues. Clinical Chem- istry, 20: 1576-1577. PILSBRY,H. А. 8 VANATTA, E. G., 1898, Revision of the North American slugs: Binneya, Hemphil- lia, Hesperarion, Prophysaon and Anadenulus. Proceedings of the Academy of Natural Sciences of Philadelphia, 50: 219-248. RAYMOND, W. J., 1890, Why does Prophysaon shed its tail? Nautilus, 4: 1. ROACH, D. K., 1963, Analysis of haemolymph of Arion ater L. Journal of Experimental Biology, 40: 612-623. STASEK, R., 1967, Autotomy in the Mollusca. Occasional Papers of the California Academy of Sciences, 61: 1-44. МТТ, L. J., CONGDON, J. D. & DICKSON, N. A., 1977, Adaptive strategies and energetics of tail autotomy in lizards. Ecology, 58: 326-337. Revised Ms. accepted 15 July 1985. Ne MALACOLOGIA, 1986, 27(2): 313-321 POLYMORPHISM IN A LABORATORY POPULATION OF BIOMPHALARIA GLABRATA FROM A SEASONALLY DRYING HABITAT IN NORTH-EAST BRAZIL O.S. Pieri’ & J.D. Thomas School of Biological Sciences, University of Sussex, Falmer, Brighton BN1 9QG, United Kingdom ABSTRACT Quantitative studies of a laboratory population of Biomphalaria glabrata (Say), originating from a seasonally drying habitat in NE Brazil, revealed that the snails were polymorphic with respect to morphology, behavior and diapause. The various morphs can be placed in a series on the basis of discontinuous variates, such as apertural lamellae, and continuous variates, Such as shell weight and surface area of the shell aperture. Statistical analysis revealed that lamellate snails (both diapausing and resident) tend to have relatively heavier, flatter shells, with smaller, flatter apertures, which are more deflected to the left than is the case with resident non-lamellate snails. Snails with lamellae, particularly those with six lamellae, were more prone to emigrate from the water than those without lamellae. All the various sets of lamellae tend to occur at constant relative distances from the aperture, possibly because post-lamellate growth is programmed to stop, either after a fixed time, or after a certain amount of growth has occurred. It is postulated that the polymorphisms exhibited by the snails are adaptive, and should be taken into account when designing control measures against these snails, which act as hosts of Schistosoma mansoni. INTRODUCTION It is known that populations of Biomphalaria glabrata (Say) from seasonally drying habi- tats, such as those occurring in NE Brazil, may be polymorphic with respect to morphol- ogy, behaviour and dormancy (Paraense, 1957; Richards, 1963, 1964, 1967; Etges & Gilbertson, 1966; Michelson & Mota, 1982). Although, according to Richards (1968), the morphological features in the polymorphic series are controlled by multi-factorial inherit- ance, there is also evidence that environmen- tal factors may be involved in triggering gene expression. However, as these have not been identified a research programme has been initiated in this laboratory to rectify this defi- ciency in our knowledge. As a prerequisite for such a study it is necessary to provide a sound, quantitative description of the various morphological types and elucidate the rela- tionships between these and the propensity to emigrate from the water or undergo diapause. The work described in the present paper aims to provide more precise information re- garding the following questions: (i) to what extent are morphological features such as deflection of the shell aperture to the left, thickening of the shell and similar features correlated with the presence of apertural lamellae? (ii) to what extent are morphologi- cal features, such as the presence of varying numbers of apertural lamellae or their ab- sence correlated with behavioural and physi- ological responses such as emigration from the water or diapause and (ili) what are the costs and benefits of these polymorphic fea- tures to the snails? MATERIALS AND METHODS I—Pre-treatment The experimental snail colony, which was initiated by seeding an aquarium containing 30 litres of aerated, filtered tapwater (Thomas, 1973) with four lamellate (4 + 1 тт shell diameter) B. glabrata from Touros, NE “Present address: Departamento de Biologia, Instituto Oswaldo Cruz, CP-926 Rio de Janeiro, Brazil. (313) 314 PIERI & THOMAS rs Is = ER = т ER Qu = м N B == = a b с FIG. 1. Schematic representation of the measurements taken of the shell of Biomphalaria glabrata. The shells were drawn in three different positions using a dissecting microscope at 12x with a camera lucida. See text for detailed explanation. (a). Shell drawn from the left side allowing the counting of the number of whorls and measurements of the shell diameter (line AB). (b). Shell drawn with a front view of the aperture allowing measurements of its surface area as well as height and width (lines AB and CD, respectively). (c). Shell drawn at a vertical position allowing measurements of the shell aperture deflection (angle ABC) and of the shell height (line MN). Notes: (i) On (a) above the radial lines (—-—) are placed at 45 degrees from each other; the horizontal radius indicates the limits of each whorl, numbered anticlockwise 1, 2 and 3. See Mandahl-Barth (1962). (ii) The axis of volution, not seen on (a), is represented as (-.:—) on (b) and (с). (iii) The position of one of the palatal lamellae (lamella III) is indicated by la on (a). (iv) Оп (с) the right side of the shell is indicated by rs and the left one, by Is. Brazil, was maintained at a temperature of 26 + 1°С and a constant photoperiod of 12 hr light, 12 hr dark. Fresh lettuce was provided as food, care being taken to ensure that some always remained uneaten. Tygan mesh bar- riers were used to prevent snails from leaving the aquarium. Sampling of the population began approxi- mately seven months after the founder snails were introduced. By this time it had been observed that considerable numbers of snails from a population of approximately 600-800 were beginning to emigrate from the water and enter into diapause on the sides of the tank. Preliminary observations revealed that most of these were in the 2.1 to 6.0 mm shell diameter range. For example the result of one census showed that the average number of emigrants in the 2.1—4.0 тт, 4.1-6.0 mm and 6.1-8.0 mm shell diameter range were 105.6, 89.8 and 13.7 per month respectively; none of the emigrants was smaller than 2.1 mm shell diameter. It was, therefore, decided to sample from within the following size cate- gories, 3.1-3.5, 3.64.0, 4.1-4.5 and 4.6-5.0 mm shell diameter, for the reasons given below. Firstly, most of the lamellate snails were in this range. Secondly, it has been observed by Paraense (1957) and Richards (1963) that snails larger than 5 mm shell diameter may resorb the lamellae. Approxi- mately 20 snails were taken as representa- tives of each size category from both emigrat- ing, diapausing and resident, non-diapausing snails at three day intervals over a period of five weeks. This meant that all the diapausing snails were roughly in the same phase and had not suffered much loss of organic re- serves (von Brand et al., 1957). The following criteria were used in selecting representative snails. Firstly, only snails outside the water, with their bodies retracted at least one eighth of the way into the body whorl of the shell, were selected as examples of diapausing snails. Secondly, only snails that were actu- ally moving or feeding below the water sur- face were selected as being representative resident snails. Approximately equal numbers of both categories were selected at the same time. Immediately after collection the resident snails were immersed in water at a tempera- ture of approximately 70°С for 15-30 sec, with the shell aperture slightly above the POLYMORPHISM IN BIOMPHALARIA GLABRATA 315 water surface and then completely immersed for 1-2 sec. This treatment made it possible to withdraw the entire body with the aid of a gentle, steady pull from a fine forceps (Paraense, personal communication). Dia- pausing snails were treated in the same way after immersion in water at 26 + 1°С had forced them to re-emerge from their shells. I—Measurements Both the shells and bodies of the snails were placed in individually numbered vials and dried at 110°C until a constant weight had been achieved, using a Sartorius analytical balance with a readability of 0.01 g. Shell measurements were made from drawings (12 x magnification) on 2 mm graph paper with the aid of a camera lucida attached to a dissecting microscope (Brown, 1980). The conventions used for shell measure- ments are illustrated in Fig. 1 a-c and de- scribed below: (i) Shell diameter. This was measured on the left side of the shell, along a straight line running from the extreme outer edge of the aperture through the center of the spire. (i) Number of whorls. These were counted as described by Mandahl-Barth (1957, 1962) (Fig. 1a). Accurate measure- ments, to the nearest eighth of a whorl, were obtained with the help of tracing paper divided into eight equal seg- ments by radii superimposed on the left side of the shell. For example, in Fig. 1a the aperture is estimated at three whorls from the base of the embryonic shell and the lamellate set, at 2/8 of a whorl from the aperture. (iii) | Aperture surface area. As the aperture margin (peristome) is in one plane its surface area was calculated as follows. The shell was placed under the micro- scope so that the image of the peristome was in the same plane as the 2x2 mm graph paper. Accurate posi- tioning was facilitated by using the larg- est magnification (50 x ). The measure- ments were taken as described by Batchelet (1975) using the grid formed by the 2x 2 mm graph paper (Fig. 1b). By counting the squares whose centres fell within the limits of the peristome at 12x magnification it was possible to estimate the area. For example, in Fig. 1b a total of 66 squares was counted giving an area of 66 x 0.03 = 2.0 mm‘. (iv) Maximum height and width of shell ap- erture. A straight line was first drawn through the intersections between the body whorl and the left and right lips. The maximum length was a line drawn parallel to it to give the maximum dis- tance between the right and left lips (AB in Fig. 1a). The aperture width given by line CD (Fig. 1b) was the maximum distance between the outer lip and the junction of the aperture plane with the body whorl. This junction was delin- eated with the aid of a small glass slide placed at the aperture to delimit the peristome plane. (у) Deflection of the aperture. This was measured with the shell axis and aper- ture plane positioned horizontally and vertically respectively (Fig. 1c). One side of the angle is given by the line AB, drawn over the peristome contour and the other by the line BC drawn parallel to the shell axis. In the example given the aperture is deflected 33° to the left. (vi) Shell height. This was the maximum height along a line parallel to the shell axis (MN in Fig. 1c). RESULTS The forms of the shell apertures of lamell- ate and non-lamellate snails are illustrated in Fig. 2. The convention used for identifying the lamellae shows that, although the numbers may vary, their distribution conforms to a basic pattern. By far the commonest catego- ries of lamellate shell were those with four, five or six lamellae (Figs. 2, c-f). Other pat- terns such as those with only one large parietal lamella (Fig. 2b) or two lamellae (e.g., lamellae | and V) were found in only 10.7% and 13.3% of diapausing and resident, lamell- ate snails respectively. Table 1, based on data from 77 diapausing and 82 resident snails (3.1-5.0 mm shell diameter), gives the number of snails with non-lamellate shells and also those with lamellate shells of various kinds. These data were used as a basis for testing several null hypotheses by means of contingency tables. The results of these statistical analyses in 316 РЕВ! 8 THOMAS FIG. 2. Apentural lamellae of Biomphalaria glabrata with shell diameters ranging from 4.5 to 5.0 mm. Individual lamellae were numbered according to Paraense 8 Deslandes (1956). (a). Non-lamellate shell. ). Shell aperture with one parietal lamella (I). ). Shell aperture with one parietal (1) and three palatal (Ш, IV and V) lamellae. ). Shell aperture with two parietal (I and Il), and three palatal (Ш, IV and V) lamellae. ). Shell aperture with two parietal (I and Il), and four palatal (Ш, IV, V and VI) lamellae. f). As in (e) but with lamellae relatively smaller. TABLE 1. Number of diapausing and resident B. glabrata of different size categories ranging from 3.1 mm to 5.0 mm of shell diameter with or without apertural lamellae. Different types of lamellae set are schematically represented in Fig. 2. A, 3.1 тт-3.5 mm shell diameter; В, 3.6 mm—4.0 mm shell diameter; С, 4.1 mm-4.5 mm shell diameter; D, 4.6 mm-5.0 mm shell diameter. DIAPAUSING RESIDENT LAMELLAE FORMATION A B С D TOTAL A B C D TOTAL Six lamellae set 12 10 14 7 43 2 3 3 4 12 Five lamellae set 5 6 4 2 17 2 2 1 1 6 Four lamellae set 0 2 1 4 7 2 0 0 6 8 Others 2 2 3 1 8 0 0 2 2 4 Total lamellates 19 20 22 14 75 6 5 6 13 30 Nonlamellates 2 0 0 0 2 19 10 11 12 52 Total 21 20 22 14 UU 25 15 И 25 82 Table 2 таке it possible to make the following evidence, therefore, that the formation statements. of lamellae is correlated with the ten- (i) | The results in Table 2a reveal a signif- dency to diapause. Thus, 97.4% of icant association between the two clas- diapausing snails were lamellate com- sifications (lamellae formation and the pared with 34.1% of resident snails. tendency to diapause). There is strong (ii) Table 2b shows that there is no signif- POLYMORPHISM IN BIOMPHALARIA GLABRATA 317 TABLE 2. Chi-square evaluation of various aspects of the data in Table 1. Numbers in parentheses indicate the expected values according to the null hypothesis. р, probability; x*, Chi-square; v, degrees of freedom. A. Lamellae formation vs. tendency for diapause Tendency for diapause Lamellae Chi-square formation Diapausing Resident evaluation Lamellates 75 (50.8) 30 (54.2) yo = 65.50 v= À Nonlamellates 2 (26.2) 52 (27.8) p < 0.05 B. Lamellae formation vs. shell diameter Lamellae formation Shell Chi-square diameter Lamellates Nonlamellates evaluation 3.1-3.5 mm 25 (30.4) 21 (15.6) 2 - 401 3.6-4.0 mm 25 (23.1) 10 (11.9) E à 4.1-4.5 mm 28 (25.8) 11 (13.2) у Е 4.6-5.0 mm 27 (24.4) 12 (12.6) р : C. Type of lamellae vs. tendency for diapause Tendency for diapause : Type of Chi-square lamellae Diapausing Resident evaluation Six lamellae set 43 (39.3) 12D) 2 _ 5.94 Five lamellae set 17 (16.4) 6 (6.6) oS 5 Four lamellae set 7 (10.7) 8 (4.3) S y 0.05 Others 8 (8.6) 4 (3.4) PATES D. Predominance of a given type of lamellae among diapausing snails Six lamellae Five lamellae Four lamellae Chi-square set set set Others evaluation 43 1174 7 8 res dpi DNS (18.7) (18.7) (18.7) (18.7) PESADOS E. Predominance of a given type of lamellae among resident snails Six lamellae Five lamellae Four lamellae Chi-square set set set Others evaluation 112 6 8 4 x? = 4.90 y. =38 (7.5) (7.5) (725) (7.5) p > 0.05 icant association between the two clas- size within the 3.1 to 5.0 mm shell sifications (lamellae formation and shell diameter range. diameter in the 3.1-5.0 mm range). It (iii) Neither was there a significant associ- can be concluded that the ability to form lamellae may be shown by snails at any ation between the classification in Ta- ble 2c (type of lamellae and tendency to 318 PIERI & THOMAS diapause). It can be concluded, there- fore, that there are no differences be- tween diapausing and resident snails in the relative proportion of the various sets of lamellae. (iv) Table 2d shows that, among diapaus- ing snails, the six-lamellate type 1$ sig- nificantly predominant over others. (v) In contrast, there was no significant difference between the proportion of the various lamellate types among res- ident snails (Table 2e). Morphometric studies were carried out on the 30 resident non-lamellate snails as well as on 30 randomly selected snails among both the 75 resident lamellates and the 50 diapausing lamellates. The mean values of the various measurements and their derived ratios are given in Table 3. Possible differ- ences between the three categories were evaluated statistically by means of a one-way analysis of variance. Tukey's Multiple Com- parison Test was used for assessing pairwise differences in cases where the F-ratios obtained were higher than expected accord- ing to the null hypothesis (Meyers & Grossen, 1974). The results can be summarized as follows: () There were no significant differences between the mean shell diameter, the mean ratios of number of whorls/shell diameter or the mean ratios of body dry weight/shell diameter of snails in the three categories studied. (1i) The shell dry weight/shell diameter ra- tios were significantly greater in both diapausing and resident lamellate snails than in resident non-lamellate snails. However, there were no signifi- cant differences between this ratio in diapausing and resident lamellate snails. (iii) The following ratios were significantly less in both diapausing and resident lamellate snails than in resident non- lamellate snails: apertural surface area/ shell diameter, apertural surface area/ body weight, height/width of shell aperture, shell height/shell diameter. However, there were no significant dif- ferences between any of these ratios in diapausing and resident lamellate snails. (iv) The deflection of the shell aperture to the left was significantly greater in both diapausing and resident lamellate snails than in resident, non-lamellate snails but there were no significant dif- ferences between this measurement in the case of snails in the first two cate- gories. The values of coefficients of variation (stan- dard deviation as a percentage of the mean) in Table 3 indicate considerable variation in the measurements. This is particularly the case with the ratios involving dry weight mea- surements. However, there is nothing to sug- gest that any category of snails is more variable than another. Measurements of the distance between the lamellae and the shell aperture revealed that the lamellae were situated 1/8, 2/8 or some other fraction of a whorl from the aperture in 50 (66.7%), 17 (22.7%) and 8 (10.6%), re- spectively, of the 75 diapausing lamellate snails examined. Resident lamellate snails resembled them in this respect. Thus, of the 30 such snails examined 22 (73.3%), 5 (16.7%) and 3 (10.0%) had the lamellae situ- ated 1/8, 2/8 or some other fraction of a whorl, respectively, from the aperture. Chi-square tests (one-way classification) showed that the location of the lamellae 1/8 of a whorl from the aperture was significantly predominant over the other two categories for both diapausing lamellates (x? : 8.33; P<0.01) and resident lamellates (x? : 6.53; P<0.05). During the course of this study other obser- vations were made on the behaviour of the snails which are relevant to the polymorph- isms. Firstly, after the snails have started to diapause, their bodies are generally retracted an appreciable distance into the body whorl of the shell. Secondly, after snails have entered diapause it proved difficult to reverse the process in the short term. Thus, even after being returned to the water repeatedly they persisted in leaving the water to diapause on the sides of the aquarium. DISCUSSION The results show that the B. glabrata pop- ulation studied was polymorphic, although it had started from four lamellate individuals originating from a seasonally drying habitat in Touros, NE Brazil. The various morphs can be placed in a series, on the basis of discontin- uous variates, such as apertural lamellae and continuous variates, such as shell weight, surface area of aperture, etc. At the end of the range are the morphs with a full set of six lamellae and relatively heavier, flatter shells POLYMORPHISM IN BIOMPHALARIA GLABRATA 319 TABLE 3. Shell measurements and derived ratios (mean and standard deviation) of diapausing lamellate (DL), resident lamellate (RL) and resident non-lamellate (RN) B. glabrata. F-ratios were obtained through One-way Analysis of Variance comparing differences in each of the various shell measurements and ratios among the three groups of snails (two degrees of freedom for the between-group variance and 87 degrees of freedom for the within-group variance). Significance of pairwise differences was evaluated through Tukey's Multiple Comparison Test (Meyers & Grossen, 1974). Values for coefficient of variation are given in parenthesis. HSD, Tukey's range statistics. Resident Shell measurements Diapausing Resident non- F-ratio Tukey's Significant pairwise and ratios lamellates lamellates lamellates (Fo 87) HSD differences (p < 0.05) Shell diameter (mm) 4.1! = 015 ASEOS 410 = 07 1.38 u — (12.1) (11.9) (17.5) Number of whorls up to 0.86 + 0.08 0.84 + 0.08 0.87 + 0.09 0.54 — — the aperture/shell (8.3) (9.5) (10.3) diameter ratio (mm ') Shell dry weight 1.29 == 0:30) 1.19 = 0/31 0:84. = 0131) 17:33** 10.195. Divs; RN andlRE vs: RN shell diameter ratio (23.1) (26.1) (36.9) (mg х mm !) Body dry weight 0'30==0.10’ 0.32 = 0A) 0.28 =007 127 — — shell dry weight ratio (33.3) (34.4) (25.0) (mg x mm !) Aperture surface area 076 = 0:12 0/76 210108 0:85 = 0516 4.717 0.08 DL vs. АМ and RL vs. RN shell diameter ratio (15.8) (10.5) (18.8) (mm) Aperture surface area 2162 = 0:55. 2:56) 10197, Salat 0:72. 4371 0.48 DL vs. АМ and RL vs. АМ body dry weight ratio (20.9) (37.4) (23.0) (mm? x mg ') Aperture height 1.293020 36 10524 G70 5028742012 DL vs. RN and RL vs. RN aperture width ratio (15.5) (17.6) (8.9) Shell height/shell 0.43 + 0.03 0.44 + 0.04 0.49 + 0.04 28.03*** 0.02 DL vs. АМ and RL vs. АМ diameter ratio (6.9) (9.1) (8.1) Aperture deflection 316’ = 5104: 29:5 = 4.56 25:3. =4.36 Sor... 2:94 DL vs. АМ and RL vs. АМ (degrees) (15.9) (15.4) (17.2) (*) Significant at the a = 0.05 level; (***) significant at the a = 0.001 level. with smaller, flatter apertures which tend to be more deflected to the left than in other morphs. There was a definite tendency for this morph and others with smaller number of lamellae to emigrate from the water and enter into diapause. The emigrating, diapausing, lamellate snails differed from the resident lamellate in having proportionately more indi- viduals with a fully developed set of six lamellae. At the other end of the scale are the snails with non-lamellate, relatively lighter, higher shells with larger, higher apertures which are less deflected to the left than in other forms. These results provide quantita- tive support for the observations made by Paraense (1957) and Richards (1963, 1964, 1967, 1968) on lamellate B. glabrata collected from various areas. Intermediate morphs can be distinguished by the number of lamellae. Thus, some snails have only one lamella to a set whereas others may have two, four or five. lt 15 likely that these are distinct morphs rather than transi- tional stages in the development of a full six lamellate set for the following reasons. Firstly, each set of lamellae occurred in approxi- mately the same proportion in all the size categories of snails examined. Secondly, all the lamellae observed in sets of five or less were as thick and well formed as those in sets of six. In fact only three out of 105 lamellate shells examined had rudimentary or incom- pletely developed lamellae. Two of these were six lamellate sets and one had four lamellae. The appearence of lamellae 1$ only one of several changes that occur during the growth and development of lamellate snails. Certain of these, such as shell thickening, a reduction in the size of the aperture and a change in its shape appear to be synchronized with the appearence of the lamellae, and it would be of interest to ascertain whether the greater rela- tive weight of lamellate shells is due mainly to 320 РЕН! & THOMAS their post-lamellate growth. lt was surprising to find that all the sets of lamellae occurred at an almost constant relative distance of 1/8 to 2/8 of a whorl from the aperture. A likely explanation is that somatic post-lamellate growth of the shell is programmed to stop either after a fixed time interval or after a certain amount of growth has occurred. lt can be postulated that lamellate snails emerge from the water and undergo diapause only after growth has slowed down or stopped. However, some lamellate snails appear to remain in the water. Two explanations can be advanced to account for this phenomenon. First, they may be transient stages which have not yet emigrated because growth has not yet stopped. Alternatively, they may enter a state of diapause or reduced activity while still in the water thus representing another survival strategy. These hypotheses could be tested by monitoring individual snails during the transition from the non-lamellate to the lamellate, diapausing stage. Growth stop- page has also been observed in B. glabrata following a period of unusual shell growth leading to the formation of circular thicken- ings, called ribs, around the aperture (Richards, 1963). Snails entering diapause may deposit an epiphragm of mucous mate- rial near the aperture as reported by Paraense (1957) and Richards (1963, 1964, 1967). It follows that these morphological and behavioural changes must be accompanied by changes in the physiology and biochemis- try of the snail. However, practically nothing is known about the molecular nature of the latter although they are manifested by changes in the colour of the hepatopancreas, the rate of shell deposition and growth as the snails enter the lamellate phase (Richards, 1963). As might be expected the rate of oxygen consumption declines as snails enter the diapausing or aestivating phase (von Brand et al., 1948; Magalhaes-Neto, 1953; von Brand et al., 1957; Heeg, 1977). The sequential series of events that lead to different morphs (or phases), dispersion and a state of dormancy described for lamellate B. glabrata have also been noted in other mol- luscan species, including Biomphalaria schrammi by Paraense & Deslandes (1956) in Brazil, Biomphalaria pfeifferi gaudi by Mc- Cullogh (1958) in Africa, four varieties of Biomphalaria т Africa by Mandahl-Barth (1957), 21 species of freshwater molluscs including members of the subfamilies Plan- orbinae, Helisomatinae, Segmentininae, Planorbulinae, Plesiophysinae by Richards (1963) and Ferrissia wautieri by Richardot (1977a, b). The behavioural responses shown by the lamellate snails such as emi- gration from the water and dormancy are also characteristic of a large number of non-la- mellate pulmonates including those which serve as hosts for schistosomiasis. Various terms have been used to describe the state of dormancy which these snails enter into, such as anhydrobiosis (Stiglingh 4 van Eeden, 1977) or aestivation (Brown, 1980). As pointed out by the latter author, this strategy is central to the ecology of many freshwater snails including snail hosts of schisto- somiasis. The seasonally drying aquatic habitats en- countered in many areas of NE Brazil (Barbosa & Olivier, 1958; Barbosa, 1962) appear to favour the mixed strategies involv- ing the formation of lamellate, emigrating, diapausing morphs as well as others. They, therefore, appear to resemble plants which also live in seasonal environments with un- predictable levels of future adversity, in their strategy for survival. Thus, seed and dor- mancy polymorphisms are common among fugitive or weedy species such as Rumex crispus and Xanthium spp. (Cavers & Harper, 1966; Harper, 1977), while leaf polymorphism is frequently encountered in desert plants. As pointed out by Harper (1977) polymorphism is advantageous because it gives a degree of buffering against sudden selective forces that favour one or other morph. This helps to explain why snails, like weeds, are often difficult to control or eradicate by the applica- tion of chemicals and biological methods. ACKNOWLEDGEMENTS We express our gratitude to the following: UNDP/World Bank/WHO Special Programme for Research and Training in Tropical Dis- eases for providing financial support, to Dr. W.L. Paraense for providing snail material, to Professor J. Maynard-Smith for providing fa- cilities in the School of Biology, University of Sussex, to Mr. M.J. Stenning for technical assistance, to Dr Р.А. Sterry and Dr R.L. Patience for reading the manuscript critically. REFERENCES BARBOSA, F.S., 1962, Aspects of the ecology of the intermediate hosts of Schistosoma mansoni POLYMORPHISM IN BIOMPHALARIA GLABRATA 321 interfering with the transmission of bilharziasis in North-Eastern Brazil. /n WOLSTENHOLME, G.E.W. & O'CONNOR, М. (ed.), CIBA Founda- tion Symposium on Bilharziasis, Churchill, Lon- don, p. 23-35. BARBOSA, F.S. & OLIVIER, L., 1958, Studies on the snail vectors of bilharziasis mansoni in North- Eastern Brazil. Bulletin of the World Health Or- ganization, 18: 895-908. BATSCHELET, E., 1975, Introduction to mathemat- ics for life scientists. Springer-Verlag, Berlin, Sip: BRAND, T. von, NOLAN, М.О. & MANN, E.R., 1948, Observations on the respiration of Australorbis glabratus and some other aquatic snails. Biological Bulletin, 95: 199-213. BRAND, T. von, MCMAHON, P. & NOLAN, M.O., 1957, Physiological observations on starvation and dessication of the snail Australorbis glabratus. Biological Bulletin, 113: 89-102. BROWN, D.S., 1980, Freshwater snails of Africa and their medical importance. Taylor & Francis, London, 487 p. CAVERS, P.B. & HARPER, J.L., 1966, Germina- tion polymorphism in Rumex crispus and Rumex obtusifolius. Journal of Ecology, 54: 367-382. ETGES, F.J. & GILBERTSON, D.E., 1966, Repel- lent action of some chemical molluscicides on schistosome vector snails. American Journal of Tropical Medicine and Hygiene, 15: 618-624. HARPER, J.L., 1977, Population biology of plants. Academic Press, London, 892 p. HEEG, J., 1977, Oxygen consumption and the use of metabolic reserves during starvation and aestivation in Bulinus (Physopsis) africanus (Pulmonata: Planorbidae). Malacologia, 16: 549-560. MAGALHAES-NETO, B. 1953, Acao da dessecacao e do jejum sobre a respiracao do Australorbis glabratus. Publicacoes Avulsas do Instituto Aggeu Magalhaes, 2: 5-10. MANDAHL-BARTH, G., 1957, Intermediate hosts of Schistosoma, African Biomphalaria and Bulinus: |. Bulletin of the World Health Organiza- tion, 16: 1103-1163. MANDAHL-BARTH, G., 1962, Key to the identifica- tion of East and Central African freshwater snails of medical and veterinary importance. Bulletin of the World Health Organization, 27: 135-150. McCULLOUGH, F.S., 1958, The internal lamellae in the shell of Biomphalaria pfeifferi gaudi (Ranson) from Ghana, West Africa. Journal de Conchyliologie, 97: 171-179. MEYERS, L.S. & GROSSEN, N.E., 1974, Behavior research: theory, procedure and design. Free- man, San Francisco, 355 p. MICHELSON, Е.Н. 8 MOTA, E., 1982, Malacologi- cal observations bearing on the epidemiology of schistosomiasis in a rural Bahian community. Revista do Instituto de Medicina Tropical de Sáo Paulo, 24: 75-82. PARAENSE, W.L., 1957, Apertural lamellae in Australorbis glabratus. Proceedings of the Malacological Society of London, 32: 175-179. PARAENSE, W.L. & DESLANDES, N., 1956, Ob- servations on Australorbis janeirensis (Clessin, 1884). Revista Brasileira de Biologia, 16: 81-102. RICHARDOT, M., 1977a, Ecological factors induc- ing aestivation in the freshwater limpet Ferrissia wautieri (Basommatophora: Ancylidae). I—Oxy- gen content, organic matter content and pH of the water. Malacological Review, 10: 7-13. RICHARDOT, M., 1977b, Ecological factors induc- ing aestivation in the freshwater limpet Ferrissia wautieri (Basommatophora: Ancylidae), II— Photoperiod, light intensity and water tempera- ture. Malacological Review, 10: 15-30. RICHARDS, C.S., 1963, Apertural lamellae, epiphragms and aestivation of planorbid mol- lusks. American Journal of Tropical Medicine and Hygiene, 12: 254-263. RICHARDS, C.S., 1964, Apertural lamellae as sup- porting structures in Australorbis glabratus. Nautilus, 78: 57-60. RICHARDS, C.S., 1967, Estivation of Biomphalaria glabrata (Basommatophora : Planorbidae), as- sociated characteristics and relation to in- fection with Schistosoma mansoni. American Journal of Tropical Medicine and Hygiene, 16: 797-802. RICHARDS, C.S., 1968, Aestivation of Bio- mphalaria glabrata (Basommatophora: Planorbidae). Genetic Studies. Malacologia, 7: 109-116. STIGLINGH, 1. & EEDEN, J.A. van, 1977, Popula- tion fluctuations and ecology of Bulinus tropicus (Mollusca: Basommatophora). Wetenskaplike Bydraes van die PU vir CHO Reeks В: Natuurwetenskappe, 87:1-37. THOMAS, J.D., 1973, Schistosomiasis and the control of mulluscan host of human schisto- somes with particular reference to possible self- regulatory mechanisms. In DAWES, B. (ed.), Advances in Parasitology. Academic Press, Lon- don, 11: 307-399. MALACOLOGIA, 1986, 27(2): 323-339 GENETIC VARIATION IN SEVEN WOOD-BORING TEREDINID AND PHOLADID BIVALVES WITH DIFFERENT PATTERNS OF LIFE HISTORY AND DISPERSAL K. Elaine Hoagland Center for Marine and Environmental Studies, Lehigh University, Bethlehem, PA 18015, U.S.A. ABSTRACT Intrapopulation genetic variation and genetic distance between populations were investigated for ten populations of seven teredinid and pholadid species in the bivalve superfamily Pholadacea. Genetic variation as measured by allozyme heterozygosity, percent polymorphic loci, and number of alleles per locus tended to be higher in species with planktotrophic larvae, lower in species with brooded larvae. Two introduced populations were genetically impover- ished, perhaps due to founder effects and bottlenecks. Most of the 20 loci resolved for all species showed heterozygote deficiency, as has been reported for many marine bivalves. Wahlund effects and inbreeding are not sufficient to explain heterozygote deficiency; natural selection and null alleles may also be causes. Species both with and without planktonic larvae show heterozygote deficiences. Substantial genetic differentiation exists between populations of one species with planktonic larvae. Patterns of genetic variation could not be correlated with latitude or local environmental factors. The data suggest that breeding structure and larval dispersal influence intra- and interpopulation allozyme variation. Key words: shipworm; Teredinidae; Pholadidae; Pholadacea; allozymes; heterozygosity; genetic variation; brooding; planktotrophy; dispersal; heterozygote deficiency. INTRODUCTION Since the advent of electrophoresis as a tool in population genetics, many researchers have sought to describe patterns in genetic variation at the population and species levels. Some have related genetic variation to heter- ogeneity or uncertainty of the environment (Levinton, 1973; Selander & Kaufman, 1973), although Johnson (1976) demonstrated the ambiguity of many such studies. High genetic variability has been correlated with geograph- ical range and colonization ability (Nevo, 1978). Species with dispersive larvae are assumed to be outbreeders and good colo- nizers, hence more variable genetically at the population level, but less genetically differen- tiated between populations, than species that brood their young or otherwise lack a disper- sive larval stage (Crisp, 1978; Ament, 1979; Hamrick et al., 1979). Most of the data sup- porting these ideas involve several popula- tions of one or two species (Koehn et al., 1976; Ayala 8 Valentine, 1974; Lavie 8 Nevo, 1986). Most molluscan data sets include only a few enzyme systems that are usually highly polymorphic (Levinton, 1975; Berger, 1977; Wilkins and O'Regan, 1980; Beaumont et al. 1980; Badino & Sella, 1980). The type of larval development in bivalves is related to the ability of the young to dis- perse; hence it bears a relationship to geo- graphic range, population size, and genetic structure of populations. The positive correla- tion of genetic variation with these three fac- tors has been supported theoretically and empirically (Soulé, 1976; Snyder & Gooch, 1973), but contradictory evidence was cited by Valentine (1976). Few data are available that delineate genetic patterns in similar spe- cies with different types of larval develop- ment and different levels of fecundity (Lavie 8 Nevo, 1986). This paper reports the electrophoretic anal- ysis of a broad range of enzyme systems for ten populations of seven species of wood- boring teredinid and pholadid bivalves within one superfamily, the Pholadacea. Two popu- lations are introduced; the effects of introduc- tion on population genetic structure was in- vestigated. The genetic distance between populations was compared for two species and related to typical interspecific values. The other purpose of the work was to compare heterozygosity in populations of species with (323) 324 GENETIC VARIATION IN WOOD-BORING PHOLADACEA different life histories and modes of dispersal. Each species has a characteristic develop- ment pattern, ranging from completely plank- tonic to completely brooded development with the release of pediveliger young. Some spe- cies follow an intermediate path, with a brooded period followed by a planktonic pe- riod. Levels of fecundity are proportional to the amount of time spent in the plankton. The species are otherwise ecologically similar, spending their adult lives burrowing in wood. Some of the species chosen for analysis are congeners. Several species were taken from the same geographical area, so that the de- gree of environmental variation (e.g., sea- sonal temperatures and salinities; variation in food and exposure to predators) would not be a major factor explaining the observed differ- ences in genetic variation among taxa. METHODS Populations Studied The populations of Pholadacea that were examined are described in Table 1. The den- sity of each species in the white pine collect- ing panels indicates the strength of the age- class settlement over the time period of the submergence of the panels. The accuracy of the estimate is lower for species lacking a planktonic larval stage, such as Teredo bartschi. Such species are expected to have highly clumped distributions (Hoagland et al., 1981). The salinity at all sites is seasonally vari- able. The yearly salinity ranges are based on my observations in the case of the New Jersey sites, and data from laboratories at localities cited in Table 1 for the other sites. All sites are estuarine except Millstone, Con- necticut, and Virginia Key, Florida, which have approximately full oceanic salinity for most of the year. Teredo bartschi is a tropical to subtropical species that was accidentally introduced into the thermal effluents of two nuclear power plants in the cold temperate zone (Hoagland & Turner, 1980). The Fort Pierce, Florida, population is within the natural distribution of the species. The New Jersey and Connecticut populations had been established for seven and five years, respectively. Life History Data on the life history of the teredinids from New Jersey were obtained between 1976 and 1981 by submerging wood panels for one to twelve month periods, then retriev- ing them and submerging new panels monthly at 20 stations in Barnegat Bay (Hoagland, 1983a). In the laboratory, the shipworms were dissected from the wood, identified, measured, and examined for larvae in the gills. Other life historical and ecological data were available in the literature (Turner & Johnson, 1971; Hoagland & Turner, 1981). Temperature and salinity tolerances were de- termined in the laboratory; details of these experiments are reported elsewhere (Hoag- land, 1983b, 1986). Electrophoresis Specimens were dissected from the wood while alive, identified, and either electro- phoresed immediately or frozen in Tris tissue buffer until they were subjected to electrophoretic analysis. Voucher specimens are on deposit at The Academy of Natural Sciences of Philadelphia (ANSP); catalogue numbers are in Table. 1. Horizontal starch-gel electrophoresis followed by staining for spe- cific enzymes was employed to study 23 enzyme systems. The general methods of Ayala et al. (1973) were applied to mollusks as amended by Dillon & Davis (1980) and Davis et al. (1981). Starch gels (13%) were prepared using 33.5 g of Electrostarch and 250 ml of one of four gel buffers: 1) tris citrate, pH 6.0; 2) tris NaOH borate (Poulik), tray buffer pH 7.6 and gel buffer pH 8.9; 3) tris- EDTA-borate (TEB) pH 8.0; and 4) TEB, pH 9.1. Four enzyme systems were run on TEB of pH 9.1 but with tray buffer of pH 8. En- zymes assayed and electrophoretic condi- tions are reported in Table 2. Five wicks of No. 3 Whatman filter paper were Saturated with homogenized tissue, blotted, and applied, one wick from each individual, to each gel. Five gels were run concurrently; each was then analyzed for three enzyme systems. Each individual was investigated for all enzyme systems over a 2-day period. Six individuals from a reference population and 25 individuals from the population being tested were run on a single gel. Runs were repeated if the relative position of the refer- ence population was unclear or if results were HOAGLAND 325 TABLE 1. Populations of Pholadacea used in this study. НЕЕ ЕВЕ ЕН E ee E SA Da А No. of Yearly genomes Approx. salinity Species Population sampled density‘ range (%o) Comments Teredo Blue Hole, intracoastal 98 10'—102 10-39 Panels submerged May 2, bartschi waterway near Harbor 1980 removed Nov. 20, Branch Laboratory, Ft. 1980 in mangrove area. Pierce, Florida ANSP A9100. Teredo Oyster Creek nuclear 200 102—103 7-30 Panels submerged from bartschi generating station, Oyster docks and removed in Creek and Forked River, 1979 and 1980. ANSP New Jersey A8717A-E. Teredo Millstone nuclear 40 10°-10' 28-33 Panels submerged from bartschi generating station, docks June 17 and Millstone Point, Niantic removed Nov. 7, 1980. Bay, Waterford, Conn. ANSP A8693, A8726C. Teredo Oyster Creek nuclear 180 10' 7-30 Panels submerged from navalis generating station, Oyster docks and removed in Creek and Forked River, 1980. ANSP A8362. New Jersey Teredo Millstone nuclear 102 10' 28-33 Panels submerged from navalis generating station, docks June 17 and Millstone Point, Niantic removed Nov. 7, 1980. Bay, Waterford, Conn. ANSP A8726A & B. Bankia Oyster Creek and Forked 214 10' 7-30 Panels submerged from gouldi River, New Jersey docks and removed in 1980. ANSP A7927E. Bankia Rocky Point, 1 mi. S. of 100 10°-10' — 10-30 Dead red mangrove fimbriatula “Crossroads” where collected Oct. 22, 1979. Intracoastal Waterway ANSP А8680А. meets St. Lucie River, near Hobe Sound, Florida Lyrodus Rosenstiel School, U. 104 10' —31-35 From pine panels floridanus Miami, Bear Cut, Virginia submerged for 6 months Key, Florida from a dock. Collected Oct. 29, 1980. ANSP A8680B & С. Lyrodus Little Jim Creek, Ft. Pierce 130 101-122 10-39 Panels submerged May 2 bipartitus Inlet, near Ft. Pierce, and removed Nov. 20, Florida 1980, exposed in red mangrove area. ANSP A8726D. Martesia Rocky Point, 1 mi. S. of 124 102 —10—30 From dead red mangrove striata “Crossroads” where collected Oct. 22, 1980. Intracoastal Waterway meets St. Lucie River, near Hobe Sound, Florida. ANSP 353444. I[gq q qu O A A ee “Order of magnitude estimate per wood volume equivalent to a test panel 2 x 9 x 21 cm. otherwise ambiguous. Although a sample size of at least 50 individuals per population was sought for each enzyme locus, it was not always reached because a few species were rare. The sample size also varied among enzyme systems for a single population when the zymograms could not be resolved for all individuals. After preliminary results showed no differences in allelic patterns due to tissue type, entire animals minus brooded larvae were homogenized. All specimens were 4 to 12 months old and most were female. Gels were scored as described in Ayala et al. (1973). The alleles of each locus were 326 GENETIC VARIATION IN WOOD-BORING PHOLADACEA TABLE 2. Enzymes assayed, buffers, current, voltage, and duration of electrophoresis. No. Gel & tray Current/ Run Enzyme loci buffer voltage time (hr) Acid phosphatase (AcPh) 1 TC6 35 MA 3.5 Adenylate kinase (Adkin) fe Poulik 35 MA 3.0 Aldehyde oxidase (AO) 2% TEB 9 350 V 4.5 Aspartate amino transferase (AAT) 1 TEB 9 350 V 4.5 Esterase NA (EST NA) 32 TEB 9/8 35 MA 2.0 Glucose-6-phosphate dehydrogenase (G6PDH) de TC6 35 MA 2.5 Glucose-phosphate isomerase (GPI = PGI) 23 TC 6 35 MA 2.0 Glutamate dehydrogenase (GDH) ihr Poulik 35 MA 3.0 Glyceraldehyde-3-phosphate dehydrogenase (G3PDH) il TEB 8 35 MA 915 a-Glycerophosphate dehydrogenase (a GPDH) ye Poulik 35 MA 3.0 Hexokinase (HEX) 1 Poulik 35 MA 3.0 Isocitrate dehydrogenase (IsDH) 2 TEB 8 35 MA 3:5 Lactate dehydrogenase (LDH) 1 TEB 8 35 MA 95 Leucine amino peptidase (LAP) 1 TC6 35 MA 2.0 Mannose-6-phosphate isomerase (МР!) 1 TEB 9/8 35 MA 2.0 NAD-dependent malate dehydrogenase (NAD-MDH) 22 TC 6 35 MA 3.5 Peptidase С (Pep С) SE TEB 8 35 MA 3.5 Phosphoglucomutase (PGM) 1 TC 6 35 MA 2.0 MEB19 350 V 4.5 6-phosphogluconate dehydrogenase (6-PGDH) 1 Poulik 35 MA 3.0 Sorbitol dehydrogenase (SoDH) 1 Poulik 35 MA 3.0 Superoxide dismutase (SOD) 2 TEB 9/8 35 MA 2.0 Triosephosphate isomerase (ТР!) Ae TEB 8 35 MA 3.5 Xanthine dehydrogenase (XDH) 1 Poulik 35 MA 3.0 *Locus with missing data. (If there are multiple loci, the upper locus has missing data, except for Pep G, where the lowest locus has missing data.) identified by the distance, in mm, that they migrated with respect to the most common allele of a reference population. Teredo bartschi from Oyster Creek, New Jersey, was the reference population because it was abundant and nearly monomorphic. Assign- ment of electrophoretic patterns to loci and consequent interpretations were made with the aid of data collected on the same enzyme systems for mollusks (Davis et a/., 1981) and other organisms (Lewontin, 1974). Calculations were made of the allele fre- quencies at each locus, the average number of alleles per locus (A), the percent polymorphic loci (P), and Neïs genetic dis- tance (D) between populations. A locus was scored as polymorphic if the frequency of the most common allele was <.99. Heterozygos- ity of each enzyme locus for each population (h) was predicted from the Hardy-Weinberg equilibrium. Heterozygosity was also calcu- lated directly (hops). A heterozygote defi- ciency index was calculated as (hobs.-Aprea.)/ Пргеа. The average individual heterozygosity per locus was calculated, again using both the Hardy-Weinberg assumptions (Hprea.) and the direct method (Hops.), by averaging the respective h values over all loci. RESULTS Genetic variation The allele frequencies for each population are in the Appendix. Of the 32 loci resolved, 20 yielded consistent results for all popula- tions and were used in the analysis of genetic variation. Fourteen loci were polymorphic in at least one species. At 5 loci (AcPh, SOD, LDH, 6-PGDH and IsDH), one allele was fixed in each of the seven species. Only for XDH HOAGLAND 327 TABLE 3. Pholadacea. Average heterozygote deficiencies. Mean + S.E. het. def. Т.Б. Conn. —1.00 T.b. NJ. — 0.50 T.b. Fla. —0.54 + .18 Lora: — 51-11 Т.п. Сопп. —0.36 = .08 Т.п. N.J. —0.67 + .09 Lag TAY —0.42 + .15 BEI —0.38 + .12 B.g. М... —0.48 + .08 M.s. Fla. 0:43:12 was the same allele fixed in all the species. Regardless of the species, genetic variation was consistently higher at some loci (eg., GPI, РСМ, МР!) than others. The heterozygote deficiencies (Table 3) were large at most loci. Lack of heterozygotes occurred at three loci in three populations, but allthree populations also exhibit loci in Hardy- Weinberg equilibrium (d = O). Loci lacking heterozygotes are not the same in each pop- ulation. Heterozygote deficiencies ranged widely among populations of one species. The introduced population of Teredo bartschi from Millstone, Connecticut, had only two polymorphic loci, and both had complete lack of heterozygotes. Table 4 summarizes commonly-used indi- ces of genetic variation for the 10 populations. The values of P and A for the Millstone population of Teredo bartschi may be under- estimated because only 20 specimens were obtained over a 2-year period. Means of the various indices for the Pholadacea were cal- culated with and without the two introduced populations. In comparing Teredo bartschi with other taxa, the latter values should be used. The values for the recently-introduced T. bartschi are probably altered by founder effects and are not representative of natural populations (Hoagland, 1983b). Interpopulation variation is assessed using Nei’s genetic distance (D), Table 5. The rela- tionships do not change using other indices. Differences between populations are consid- erably smaller than those between species. The genetic distance between populations of Teredo navalis with planktonic larvae 1$ sub- stantial at 0.29. In fact, the genetic distance between the populations of Т. bartschi 1$ much less (0.08 to 0.11). No. loci Total no. polymorphic loci lacking heterozygotes 2 2 2 0 he 3 10 1 10 0 11 1 9 3 1 1 14 0 13 3 Ecology Table 6 reviews life historical and ecologi- cal information based on field and laboratory work (Turner, 1966; Turner & Johnson, 1971; Hoagland, 1983b, 1986). The most important difference among the species is the type of larval development. Two species, Teredo bartschi and Lyrodus bipartitus, brood the young in the gill to the pediveliger stage; they produce relatively few yolky eggs. Teredo navalis and L. floridanus also brood the young, but release them in the straight hinge veliger stage, after which they undergo sev- eral more weeks of planktotrophic develop- ment. The other three species are completely planktonic; fertilization of up to a million eggs per reproductive event is external, or could involve pseudocopulation, transfer of sperm using the siphons. Despite these differences, all shipworms can be classified as opportunistic (Turner, 1973), because they colonize wood, an ephemeral substrate and food resource that they themselves destroy. Although the peaks in settlement activity of sympatric species do not coincide, most natural wood collected in Florida contained numerous co-existing spe- cies. Lyrodus massa, L. floridanus, L. bi- partitus, Teredo furcifera, T. bartschi, and Martesia striata were found together in man- grove wood from the Ft. Pierce Inlet, Florida, although L. massa and T. furcifera were rare. Most predators on the shipworms that | ob- served, including nereid polychaetes, flat- worms, and protozoa, appeared to be non- selective. All 7 species of shipworms have the same trophic position, eating both wood and algae. | have reared adults and larvae of 4 of the 7 species on the same diet. 328 TABLE 4. Genetic variation in 10 populations of Pholadacea, 20 loci. GENETIC VARIATION IN WOOD-BORING PHOLADACEA Percent „Average no. Avg. individual Avg. individual poly- of alleles observed predicted morphic per locus heterozygosity heterozygosity Population loci, P* A (S.E.) (SIE?) (SIE) Brooders T. bartschi Conn. 10 VA (а 0 (- 0.04 (0.03) T. bartschi N.J. 10 aa (al) 0.004 (0.003) 0.01 (0.01) T. bartschi Fla. 35 1.85 (all) 0.008 (0.004) 0.05 (0.03) L. bipartitus 45 1.95 (.26) 0.082 (0.025) 0.18 (0.05) Mixed development T. navalis Conn. 55 2.20 (.34) 0.150 (0.042) 0.24 (0.06) T. navalis N.J. 55 2.0132) 0.069 (0.024) 0.21 (0.05) L. floridanus 45 11558115) 0.042 (0.020) 0.09 (0.04) Planktonic larvae B. fimbriatula 55 1.90 (.25) 0.138 (0.038) 0.24 (0.05) B. gouldi 70 2.40 (.28) 0.133 (0.033) 0.27 (0.05) M. striata 65 2.30 (.29) 0.146 (0.049) 0.25 (0.06) Mean (S.E.), 10 pops. 44 (07) 1.81 (0.15) 0.077 (0.020) 0.16 (0.03) Mean (S.E.), excluding introduced populations 53 (04) 1.97 (0.13) 0.096 (0.019) 0.19 (0.03) A AA A A A —— *A locus is considered polymorphic if the frequency of the most common allele is less than .99. TABLE 5. Pholadacea. Nei's Distance Matrix. ТБ: Т.Б. Ten: Ten: Bf: L.b. B.g. В.Е М.5. Fla. Conn. М... Сопп. Fla. Fla. N.J. Fla. Fla. T.b. N.J. 0.11 0.10 0.67 0.88 0.60 0.94 0.95 0.71 1.23 Tbarla: 0.08 0.63 0.85 0.57 0.99 1.09 0.89 E27 T.b. Conn. 0.69 0.82 0.58 1.04 1.07 0.86 1.24 fen М 0.29 0.82 0.92 0.82 0.85 1.23 Т.п. Сопп. 0.98 0.95 0.81 1.09 1.06 ЛЕТА: 0.84 0.90 0.75 1.23 Eo AE 1.07 0.78 1.62 B.g. М... 0.48 0.98 В.Е Fla. 1.53 All the species of this study have been found as adults in driftwood. Brooding fe- males are able to reach new sites while carrying larvae. Species such as Teredo bartschi and Lyrodus bipartitus that brood young during all months of the year are very likely to colonize as adults with larvae. Teredo bartschi living in a New Jersey heated effluent retained larvae in the gills through the winter months, although successful settlement did not occur between December and April. Lar- vae of T. bartschi have survived for at least two weeks in the laboratory without growth when either food or temperature was inade- quate for metamorphosis. This phenomenon has been reported for many teredinids (Turner & Johnson, 1971). All populations used in this study lived in water 0.5 to 2 m deep and were exposed to seasonal changes in temperature and salinity (Tables 1, 6). The environment is seasonally variable with re- gard to temperature in Connecticut, salinity in Florida, and both in New Jersey. Field observations suggest that in New Jersey, adult Teredo bartschi and T. navalis rarely live more than 18 months. Bankia gouldi often lives a full 2 years or longer. Specimens of Т. bartschi have been found brooding larvae only 4 weeks after they have come in contact with wood, while the other two New Jersey species live in the wood for a minimum of 5 weeks before reaching maturity 329 HOAGLAND ‘0.0 0} doip ued элпуелэдше} ayy ‘бицелэдо jou si 49819 JajsÁO ui juejd 1emod ay) чечме 'SÁBp OL Jaye sjenpinipui 001 JO %06 UY) slow JO jeninung, Le-p1 5—0 Lep1 Le91 56-21 5—0 21-е (9.) saunjesadua) pie14 umouyxun 08-0 umouyun umouyun umouyun O£—0 98-11 (9.) ‚эЭие.э|0} элпуелэдше1 umouyun + GHZ umouyun umouyun umouyun +Gp-Z 09-2 (°%) ¡a0uesajo] Алинес ajeJadula] J8JEM WIEM еэцэшу 'S Y JOJEM WIEM JOJEM шлем J9JEM WIEM лэуем pjoo J8JEM шием IPIMPLIOM "N 15209 ‘3 SPIMPLIOM ONUBN Y ‘М ONUBNY ‘M SPIMPLIOM SPIMPLIOM эбиел зэюэ4$ umouy umouyun 201 umouyun umouyun 201 201 OL jaued }S9} sad Ajisuap uolejndod и! чоцецел 1294 о} JeaA 91 jnoge umouyun umouyun umouyun umouyun umouyun vl-0 (sAep) эбе}$ 1abijanipag y ¡noqe y ¡noqe + ¡noqe г Inoge 0 г node 0 (syaam) эбе]$ [елле| эшозиеа 0 0 0 с jnoge 7 с jnoge 7 (syaem) эбе]з jee] pepooig 910 JO „OL 21OW JO OL Blow JO „OL cOL-,Ol »OL-201 cOl-,O1 501-501 juana eanonpoidei 1э4 $663 umouyun 'Bny=unf umouyun JeaÁ || JeaA ||V ‘0—eunp ‘AON-AEN uoseas BuipaaJg "SR (ely) (MN) (el) (ely) (ey) (CTN) (rn) BJEINS W Ipınob :g вт еиЧшу ‘9 snuepuoy ‘1 snuedig *7 SIIENEU | Iyosueg | "tae pejoyd JO Salseds usnes jo Або|оээа pue Auoısıy ау! enNeledwong ‘9 FIGVL 330 GENETIC VARIATION IN WOOD-BORING PHOLADACEA as females. Some populations of Teredo bartschi and Lyrodus bipartitus are highly skewed toward individuals brooding young (as high as 90%). Sex ratios are close to 60% female in B. gouldi, B. fimbriatula, and T. navalis. Although most males are smaller than most females, the largest individuals of all species tend to be male. This fact coupled with the sex ratio bias towards females sug- gest that pholadaceans undergo alternating sexuality. The life history characters of the subtropical larviparous species Teredo bartschi and Lyrodus bipartitus, including shorter genera- tion time, more generations per year, and shorter lifespan, imply a higher population turnover rate than in the species with plank- tonic larval development. The potential for inbreeding is also greater in the larviparous species because the pediveliger young from one brood often settle together (and with the parents) on the same substratum (Turner 8 Johnson, 1971; Hoagland 8 Turner, 1981). The histories of the two introduced popula- tions of Teredo bartschi are quite distinct. The New Jersey population, founded in late 1973 or 1974, underwent a severe bottleneck in 1976 when the thermal effluent ceased tem- porarily during winter (Hoagland & Turner, 1980). After rebounding in 1978, another bot- tleneck and recovery occurred т 1980, when the thermal effluent was absent from January 5 to July 17. Fluctuations of the native 7. navalis and Bankia gouldi were less severe, but the population of В. gouldi has been low (about 10 per panel) since the introduction of T. bartschi, compared with the 100 to 1,000 per panel in 1971-1973. The Connecticut population of T. bartschi has not suffered dramatic bottlenecks but has remained small (fewer than 10 per panel). The native Florida population of T. bartschi is patchy and unpre- dictable in time and space. A few specimens were collected in panels submerged in Au- gust, 1979, and retrieved in October, 1979, near Ft. Pierce, while panels submerged at the same locality in May, 1980, and retrieved in November, 1980, contained hundreds of specimens. Panels submerged less than a mile away contained no T. bartschi, and none were found in the Miami area. The distribution of Lyrodus bipartitus is more extensive and predictable; the species was found in abun- dance (200/panel) at both localities and in both years. Bankia fimbriatula was rare where it was collected, while Martesia striata was very common. DISCUSSION Patterns of genetic variation at the various enzyme loci It is reasonable to ask whether methodol- оду 15 relevant to the interpretation of genetic variation. Sequential electrophoresis and isoelectric focusing have revealed more ge- netic variation at some loci than simple starch-gel electrophoresis (Ramshaw et al., 1979). However, there is no relationship be- tween the amount of genetic variation uncov- ered by these newer techniques and the species (Hamrick et al., 1979) or the bio- chemical function of the enzymes (Jones, 1980). No systematic bias is expected from using the starch-gel method. The P values in Table 4 are similar to those summarized by Selander (1976) for marine invertebrates (0.587) and for all invertebrates (0.467). Selander's heterozygosity of 0.147 for marine invertebrates and 0.083 for marine snails compare well with the 0.096 for the 7 pholadacean species, excluding the 2 intro- duced populations. Electrophoretic results obtained by Cole & Turner (1978) for pholadaceans are not di- rectly comparable with the data reported here. Their H values were higher, and their P values inconsistent with mine. They analyzed a different set of enzymes for each species and obtained results across all 5 species for only 6 highly polymorphic loci of the 22 exam- ined, accounting for their high H values. A similar explanation might account for the high average H value (0.367) calculated by Wilkins (1975) for 12 marine bivalves. Simon 4 Archie (1985) demonstrated the necessity of using the same enzymes in any comparison of H values across taxa. However, Archie (1985) also pointed out that certain statistical tests to detect differences in heterozygosity between samples are invalid when the same set of (non-random) loci are used. Generally, low values of H and loci fewer than 40 make statistical comparisons difficult. The amount of genetic variation in the pholadaceans was found to differ in a char- acteristic way for particular genetic loci. The esterases, peptidases, GPI, PGM, LAP, and МР! were highly polymorphic, as has been found for many organisms (Hamrick et al., 1979; Badino € Sella, 1980). Sarich (1977) claimed that the highly polymorphic enzymes accumulated substitutions at a rate ten times that of the less polymorphic enzymes. He HOAGLAND 331 proposed that rapidly-substituting loci should be more polymorphic than slowly-substituting loci at any instant even if polymorphism is due to neutral allele substitutions. But natural se- lection on individual enzymes could be re- sponsible for maintaining polymorphism in enzymes whose substrates are variable (Gil- lespie & Langley, 1974), or in enzymes that participate in regulatory reactions (Johnson, 1976). The heterozygote deficiencies in Table 3 are large. They show no relationship with species-specific life history patterns or the metabolic functions of the enzymes. The lack of heterozygotes at two loci in the Millstone population of Teredo bartschi could be due to separate introductions, or simply to small sample size. No species shows a pattern of complete absence of heterozygotes, decreas- ing the possibility that obligate self-fertilization or parthenogenesis occurs (Selander & Hud- son, 1976). No marine bivalves are known to undergo parthenogenesis. In shipworms, plenty of male gametes are present in all populations studied. Circumstantial evidence for self-fertilization has been reported in shipworms (Eckelbarger & Reish, 1972). Large heterozygote deficiences have been reported for numerous marine invertebrates, including bivalves. Zouros & Foltz (1984) re- viewed the literature and mathematically an- alyzed the relationship of differential selection in different parts of the life cycle and differen- tial reproduction of heterozygote individuals to heterozygote deficiency. They concluded that, to achieve levels of heterozygote defi- ciency reported in the literature, selection differentials would have to be very high. Beaumont (1982) and others have suggested a combination of factors could be operating, including inbreeding, selection on juvenile stages, and a Wahlund Effect, although in lab experiments Beaumont et al. (1983) impli- cated selection alone. Inbreeding alone can- not explain the heterozygote deficiencies seen for teredinids, because within a popula- tion some loci lack heterozygotes while others are in Hardy-Weinberg equilibrium. Population subdivision models of hetero- zygote deficiency have been suggested by Tracey et al. (1975). Because many marine populations with large heterozygote defi- ciency are founded by pelagic larvae, they proposed temporary subpopulations and fine- scale patchiness to explain the genetic struc- ture. Johnson & Black (1984), studying a limpet with planktonic development, sug- gested that heterozygote deficiency at all 7 loci was due not to temporal variation, but to binomial sampling variance among small lo- cal breeding groups plus mixing of larvae ona local scale. In teredinids, species both with and without planktonic larval dispersal show heterozygote deficiency. Temporal and small-scale local genetic variation could be important for all species of teredinids. Shipworm larvae can travel within an estuary or along a coastline before settlement. It is likely that even pediveligers move far enough to encounter temperature and salinity regimes, with con- commitant changes in food and predators, different from those of the parents. But adults also disperse in small groups within driftwood or boats. The extent of the Wahlund Effect may depend upon the life history characteris- tics of the species, but it could operate in all species. The presence of null alleles is a possible explanation of heterozygote deficiency, one that is difficult to verify without breeding ex- periments. Thus, natural selection, null al- leles, inbreeding, sampling error, and Wahlund Effects can all contribute to heterozygote deficiency; the values in Table 3 suggest that more than one factor is involved. Mis-identification of specimens can lead to artificially high heterozygote deficiencies if there are loci with fixed alternate alleles in the confounded species. The only species that was difficult to identify was Lyrodus flori- danus, which is easily confused with L. pedicellatus when not brooding young. Since many animals were brooding, and no speci- mens of L. pedicellatus were identified in the material, it is unlikely that those species were confused. Morphologically similar species have been identified electrophoretically; in fact | have found undescribed species of Crepidula via electrophoretic analysis (Hoag- land, 1984a). In such cases, certain speci- mens show a consistent pattern; they are fixed for alternate alleles at one or more loci. This situation has not been found in the data set reported upon here. Correlation of genetic variation with life history and other possible factors Comparing the genetic variation (Table 4) with life history (Table 6), an imperfect but suggestive trend of greater genetic variation with longer larval life in the plankton can be seen (Fig. 1). If means are calculated sepa- 332 GENETIC VARIATION IN WOOD-BORING PHOLADACEA FIG. 1. The values of predicted heterozygosity (Ные«), observed heterozygosity (Hops) and percent polymorphic loci (P) for populations of species with the three types of larval development. А = brooded development, release of pediveliger. e = planktonic development, release of eggs and sperm. № = mixed brooded and planktonic development, release of straight-hinge larvae. The original data are in Table 4. Averages of data for each larval type are indicated by the symbols with bars above. rately for the introduced populations, the nat- ural populations of larviparous species, the populations with mixed development, and those with pure planktonic development, these life history categories representing in- creasing opportunity for dispersal and outbreeding fall in the order of increasing heterozygosity (Table 7). The data suggest that the level of heterozygosity and genetic polymorphism in general may be related to life history characteristics. If so, this could be an example of population structure influenc- ing the level of polymorphism. Battaglia et al. (1978), working with six species of crustaceans, also found that some oviparous species had higher heterozygosi- ties than a brooded species. Unfortunately, information on the type of larval development was not given for all six species. Nevo et al. (1984) and Snyder 8 Gooch (1973) reported a correlation between species heterozygosity and dispersal. Hamrick et al. (1979) noticed higher variation in widespread outcrossing and wind-pollinated plants, relative to regionally-distributed plants. They also found that fecundity was positively correlated with genetic variation. Fecundity is positively cor- related with both the length of planktonic life and heterozygosity in Pholadacea (Table 6). Lavie & Nevo (1986) attributed higher heterozygosity in Cerithium scabridum than in C. rupestre to greater niche-width in the former species due to its high intertidal posi- tion. However, C. scabridum also has higher fecundity, correlated with planktonic larvae lacking in C. rupestre. Dispersal rather than niche-width at any one locality might be re- sponsible for the high heterozygosity of pop- ulations of C. scabridum. Soulé (1976) suggested that life cycle het- erogeneity should be reflected in high genetic variation. If so, planktonic developers should have higher genetic variability than brooders, and this is so. But we might also expect the species with mixed brooded-planktonic devel- opment like Teredo navalis to have the high- est genetic variation; this is not so. Animals with pelagic larvae are expected to have similar gene frequencies throughout their ranges (Soulé, 1976), yet | found con- siderable genetic distance between two pop- ulations of T. navalis (Table 5). Perhaps it is HOAGLAND 333 TABLE 7. Pholadacea. Heterozygosity values averaged over populations with the same type of larval development. Number of populations in parentheses. Data from Table 6. H observed H predicted E A Introduced brooders (2) .002 .025 10 15 Brooders, natural populations (2) .045 115 40 2.08 Mixed strategy (3) .087 .180 52 1.95 Planktotrophic species (3) .139 253 63 2.20 due to the patchy distribution of wood sub- stratum in the Northwestern Atlantic, which effectively isolates populations despite the pelagic larvae. Burton (1983) also made the point that substantial genetic differentiation has been observed between populations of marine invertebrates supposed to have high dispersal capabilities. The proximal cause of greater intrapopula- tion genetic diversity in populations of species with planktonic larvae has been cited as the larger gene pool (Selander, 1976). The data for pholadaceans are in agreement. The smaller population size of Bankia fimbriatula may explain its lower genetic variation com- pared with B. gouldi. Fuller 8 Lester (1980) showed that heterozygosity was proportional to effective population size as well as to the potential for immigration; Motro & Thomson (1982) showed a large effect of repeated bottlenecks on level of heterozygosity. The high percentage of animals brooding eggs in some populations suggests that Teredo bartschi might be a simultaneous hermaphro- dite, which would influence the effective pop- ulation size. Richards et al. (1980) found hermaphrodite gonads in a large percentage of T. bartschi, unlike Bankia gouldi and Т. navalis, which are usually either male or female. Even if selfing is rare and alternating sexuality is the rule as in T. navalis and В. gouldi (Hoagland, 1984b), the population structure of T. bartschi in which offspring (males) settle on the same wood as parents (females) leads to inbreeding. Dispersal of adults in wood does occur regularly, but large groups of closely-related individuals move together in the same piece of wood. Pedi- veligers of T. bartschi are capable of swim- ming and crawling before metamorphosis, but field studies in Barnegat Bay indicate that most settle near the parents. The pholadacean data show that species with brooded development such as Teredo bartschi can be widely distributed and have low genetic variability within and between populations yet have isolation of populations and high potential for inbreeding, compared with the planktonic developers. Similarly, Davis et al. (1981) found that some wide- spread, phenotypically uniform species of freshwater mussels (Unionidae) have low ge- netic variation within and between isolated populations. Selander & Kaufman (1975) and Selander 8 Hudson (1976) found mono- morphism yet good colonization ability in the widespread introduced self-fertilizing land snail Rumina decollata; McCracken & Selander (1980) found the same pattern in selfing populations of introduced slugs. Teredo bartschi fits the “general purpose genotype” model of Selander & Hudson (1976) for species with broad physiological tolerances despite reduced genetic variability. There are no differences in breadth of toler- ance in the 10 populations of shipworms that can be related to levels of genetic variation. 7. bartschi of New Jersey has as broad a toler- ance for temperature and salinity as the Flor- ida population (Hoagland, 1986). There 15 no detectable relationship be- tween the level of genetic variation and latitude of the pholadacean populations. The three sympatric species of New Jersey differed much more in levels of genetic variation than did the congeners Bankia gouldi and В. fimbriatula, which are temper- ate and tropical respectively, but which have similar patterns of larval development and dispersal. Redfield et al. (1980) also found no correlation between latitude and genetic variation. Nelson and Hedgecock (1980) implicated a heterogeneous diet in levels of enzyme heterozygosity. No differences in diet or par- asite infection are known that correlate with enzyme polymorphism in the Pholadacea. The anatomical data for the six Teredinidae of this study (Turner, 1966) reveal a similarity in the size of the gill versus the caecum com- pared with some other teredinid genera, sug- gesting a similar diet. 334 GENETIC VARIATION IN WOOD-BORING PHOLADACEA Populations introduced in thermal effluents Nevo et al. (1977) found lower hetero- zygosity for barnacles in a thermal effluent than at a control station. Such a trend 1$ not evident in my data except as due to founder effects and bottlenecking (Table 7). Teredo navalis from a thermal effluent actually had a higher heterozygosity than a population from just outside such an effluent (Table 4). There is some evidence that enzymes such as aGPDH in invertebrates have greater vari- ability in thermally variable environments (Johnson, 1976). The thermal effluent of a nuclear power plant is actually a variable environment because of frequent plant out- ages. Genetic similarity of the introduced pop- ulations of T. bartschi (Table 5) could confirm the identity о the species but not the pathway of introduction. It is quite likely that the intro- ductions were independently derived from the Floridian or other natural populations. CONCLUSION The ecological requirements of the pholadaceans of this study are similar. All face a temporally unstable environment over more than one generation. Some species show high polymorphism within populations, while others are nearly monomorphic as de- termined by electrophoresis of enzymes. The sympatry of some of the species with differing levels of population heterozygosity is due in part to factors other than the physical envi- ronment. The degree of genetic variability is related to the breeding structure and type of larval development and dispersal of the spe- cies. Species of Pholadacea with completely planktonic larval development have higher genetic variability, higher fecundity, more constant population size, and probably more outbreeding than do larviparous species. Larviparous species are more likely to be monomorphic through inbreeding and local bottlenecks; their distribution is patchy. | hy- pothesize that constraints of life history and population structure affect the genetic struc- ture (i.e., level of polymorphism) of popula- tions. ACKNOWLEDGMENTS This work was supported by U.S. Nuclear Regulatory Commission contract AT-04-76- 347 and a Fleischmann Foundation grant to the Wetlands Institute, Stone Harbor, New Jersey. Laboratory facilities for electro- phoretic work were made available Бу С. М. Davis at The Academy of Natural Sciences of Philadelphia. D. Mook of the Harbor Branch Laboratory, Fort Pierce, Florida, and D. Moore of the Rosenstiel School of Atmo- spheric and Planetary Science, University of Miami, helped collect specimens. M. Kessen and D. Morgan of the Millstone Environmental Laboratory, Northeast Utilities, made it possi- ble for me to collect specimens near their facility. L. Crocket and J. Selman assisted in laboratory and field work. R. Turner assisted in field work and in identification of speci- mens; without the foundation of knowledge she has built for the Teredinidae, this work would not have been possible. The manu- script was improved by the criticisms of A. J. Cain, G. M. Davis, M. Itzkowitz, and J. J. Murray. LITERATURE CITED AMENT, A., 1979, Geographic variation in relation to life history in three species of the marine gastropod genus Crepidula: growth rates of newly hatched larvae and juveniles. In STANCYK, S., ed., Reproductive Ecology of Marine Invertebrates, p. 61-76. University of South Carolina Press, Columbia, S.C. ARCHIE, J. W., 1985, Statistical analysis of heterozygosity: data-independent sample com- parisons. Evolution, 39: 623-637. AYALA, Е. J., HEDGECOCK, D., ZUMWALT, С. 5. 8 VALENTINE, J. W., 1973, Genetic variation in Tridacna maxima, an ecological analog of some unsuccessful evolutionary lineages. Evolution, 27: 177-191. AYALA, Е. J. & VALENTINE, J. W., 1974, Genetic variability in the cosmopolitan deep-water ophiuran Ophiomusium lymani. Marine Biology, 27: 51-57. BADINO, С. & SELLA, G., 1980, Phosphoglucose isomerase variability in sympatric populations of Mediterranean species of Patella (Gastropoda, Prosobranchia). Marine Ecology—Progress Se- ries, 2: 315-320. BATTAGLIA, B., BISOL, P.M. & FAVA, G., 1978, Genetic variability in relation to the environment in some marine invertebrates. In BATTAGLIA, В. 8 BEARDMORE, J., ed., Marine organisms: genetics, ecology and evolution, р. 53-70. Ple- num, N.Y. BEAUMONT, A. R., 1982, Variations in hetero- zygosity at two loci between year classes of a population of Chlamys opercularis (L.) from a HOAGLAND 335 Scottish sea-loch. Marine Biology Letters, 3: 25-34. BEAUMONT, A. R., BEVERIDGE, C. M. 8 BUDD, M. D., 1983, Selection and heterozygosity within single families of the mussel Mytilus edulis (L.). Marine Biology Letters, 4: 151-162. BEAUMONT, А. R., DAY, T. R. & GADE, G., 1980, Genetic variation at the octopine dehydrogenase locus in the adductor muscle of Cerastoderma edulis (L.) and 6 other bivalve species. Marine Biology Letters, 1: 137-148. BERGER, E. M., 1977, Gene-enzyme variation in three sympatric species of Littorina. Il. The Roscoff population, with a note on the origin of North American L. littorea. Biological Bulletin, 153: 255-264. BURTON, В. S., 1983, Protein polymorphisms and genetic differentiation of marine invertebrate populations. Marine Biology Letters, 4: 193- 206. COLE, T. & TURNER, R. D., 1978, Genetic rela- tions of deep-sea wood-borers. American Malacological Union Bulletin for 1977: 19-25. CRISP, D. J., 1978, Genetic consequences of different reproductive strategies in marine inver- tebrates. InBATTAGLIA, B. & BEARDMORE, J., ed., Marine organisms: genetics, ecology, and evolution, p. 257-274. Plenum, N.Y. DAVIS, G. M., HEARD, W., FULLER, 5. L. H. 8 HESTERMANN, C., 1981, Molecular genetics and speciation in Elliptio, and its relationship to other taxa of North American Unionidae (Bivalvia). Biological Journal Linnean Society, 15: 131-150. DILLON, В.Т. 8 DAVIS, G. M., 1980, The Goniobasis of southern Virginia and northwest- ern North Carolina: electrophoretic and shell morphometric relationships. Malacologia, 20: 83-98. ECKELBARGER, К. J. 8 REISH, D. J., 1972, A first report of self-fertilization in the wood-boring fam- ily Teredinidae (Mollusca; Bivalvia). Bulletin of the Southern California Academy of Sciences, 71: 48-50. FULLER, В. & LESTER, L. J., 1980, Correlations of allozymic variation with habitat parameters using the grass shrimp, Palaemonetes pugio. Evolu- tion, 34: 1099-1104. GILLESPIE, J. H. & LANGLEY, С. H., 1974, A general model to account for enzyme variation in natural populations. Genetics, 76: 837-848. HAMRICK, J. L., LINHART, Y. В. 4 MITTON, J. B., 1979, Relationships between life history charac- teristics and electrophoretically detectable ge- netic variation in plants. Annual Reviews of Ecol- ogy and Systematics, 10: 173-200. HOAGLAND, K. E., 1983a, Ecological studies of wood-boring bivalves and fouling organisms in the vicinity of the Oyster Creek Nuclear Gener- ating Station. Final Report, Sept. 1, 1976— Dec. 31, 1982. NTIS # NUREG/CR-3446, 173 р. HOAGLAND, К. E., 1983b, Life history character- istics and physiological tolerances of Teredo bartschi, a shipworm introduced into two temper- ate zone nuclear power plant effluents. т SENGUPTA, S. 8 LEE, 5. S., ed., Waste heat management and utilization, p. 609-622. Hemi- sphere Publishing, N.Y. HOAGLAND, K. E., 1984a, Use of molecular ge- netics to distinguish species of the gastropod genus Crepidula (Prosobranchia: Calyptrae- idae). Malacologia, 25: 607-628. HOAGLAND, К. E., 1984b, Use of the terms protandry, protogyny, and hermaphroditism in malacology. American Malacological Bulletin, 3: 85-88. HOAGLAND, K. E., 1986, Effects of temperature, salinity, and substratum on larvae of the shipworms Teredo bartschi Clapp and Т. navalis Linnaeus (Bivalvia: Teredinidae). American Malacological Bulletin, 4: 89-99. HOAGLAND, K. E., CROCKET, L. 8 TURNER, R. D., 1981, Ecological studies of wood-boring bivalves in the vicinity of Oyster Creek Nuclear Generating Station, Sept. 1, 1979—Feb. 28, 1980. Report to the U.S. Nuclear Regulatory Commission. NTIS 4 NUREG/CR-1517, 65 р. HOAGLAND, K. E. 4 TURNER, R. D., 1980, Range extensions of teredinids (shipworms) and polychaetes in the vicinity of a temperate-zone nuclear generating station. Marine Biology, 58: 55-64. HOAGLAND, K. E. & TURNER, R. D., 1981, Evo- lution and adaptive radiation of wood-boring bivalves (Pholadacea). Malacologia, 21: 111-148. JOHNSON, G. B., 1976, Genetic polymorphism and enzyme function. In AYALA, Е. J., ed. Evolution, p. 46-59. Sinauer, Sunderland, Mass. JOHNSON, M. S. & BLACK, R., 1984, The Wahlund effect and the geographical scale of variation in the intertidal limpet Siphonaria sp. Marine Biology, 79: 295-302. JONES, J. S., 1980, How much genetic variation? Nature, 288: 10-11. KOEHN, R. K., MILKMAN, R. & MITTON, J. B., 1976, Population genetics of marine pelecypods. IV. Selection, migration, and genetic differentia- tion in the blue mussel Mytilus edulis. Evolution, 30: 2-32. LAVIE, B. & NEVO, E., 1986, Genetic diversity of marine gastropods: contrasting strategies of Cerithium rupestre and C. scabridum in the Mediterranean Sea. Marine Ecology—Progress Series, 28: 99-103. LEVINTON, J., 1973, Genetic variation in a gradi- ent of environmental variability: marine Bivalvia. Science, 180: 75-76. LEVINTON, J. S., 1975, Levels of genetic polymorphism at two enzyme encoding loci in eight species of the genus Macoma (Mollusca: Bivalvia). Marine Biology, 33: 41-47. LEWONTIN, R. C., 1974, The genetic basis of evolutionary change. Columbia University Press, N.Y., 346 p. 336 GENETIC VARIATION IN WOOD-BORING PHOLADACEA McCRACKEN, С. Е. & SELANDER, В. K., 1980, Self-fertilization and monogenic strains in natural populations of terrestrial slugs. Proceedings of the National Academy of Sciences, U.S.A., 77: 684-688. MOTRO, U. 8 THOMSON, G., 1982, On heterozygosity and the effective size of popula- tions subject to size changes. Evolution, 36: 1059-1066. NELSON, K. 8 HEDGECOCK, D., 1980, Enzyme polymorphism and adaptive stragegy in the decapod Crustacea. American Naturalist, 116: 238-280. NEVO, E., 1978, Genetic variation in natural popu- lations: patterns and theory. Theoretical Popula- tion Biology, 13: 121-177. NEVO, E., BEILS, A. & BEN-SHLOMO, R., 1984, The evolutionary significance of genetic diver- sity: ecological, demographic and life history correlates. In MANI, G. S., ed., Evolutionary dynamics of genetic diversity, lecture notes in biomathematics, 53: 13-213. Springer-Verlag, Berlin. NEVO, E., SHIMONY, T. 8 LIBNI, M., 1977, Ther- mal selection of allozyme polymorphism in bar- nacles. Nature, 267: 699—701. RAMSHAW, J. А. M., COYNE, J. A. 8 LEWONTIN, R.C., 1979, The sensitivity of gel electrophoresis as a detector of genetic variation. Genetics, 93: 1019-1037. REDFIELD, J. А., HEDGECOCK, D., NELSON, К. 8 SALINI, J. P., 1980, Low heterozygosity in tropical marine crustaceans of Australia and the trophic stability hypothesis. Marine Biology Let- ters, 1: 303-313. RICHARDS, В. R., BELMORE, С. 1., HILLMAN, В. E. & MACIOLEK, N. J., 1980, Woodborer study associated with the Oyster Creek generating station. Annual Report for the period Dec. 1, 1978— Nov. 30, 1979, to Jersey Central Power and Light Co., Feb. 29, 1980. 18 p. and 3 appendices. SARICH, V. M., 1977, Rates, sample sizes, and the neutrality hypothesis for electrophoresis in evo- lutionary studies. Nature, 265: 24-28. SELANDER, R. K., 1976, Genic variation in natural population. In AYALA, F. J., ed., Molecular evo- lutions, p. 21-45. Sinauer, Sunderland, Mass. SELANDER, Fi. K. & HUDSON, В. O., 1976, Ani- mal population structure under close inbreeding: the land snail Rumina in southern France. Amer- ican Naturalist, 110: 695-718. SELANDER, R. K. & KAUFMAN, D. W., 1973, Genic variability and strategies of adaptation in animals. Proceedings of the National Academy of Sciences, U.S.A., 70: 1875-1877. SELANDER, R. K. & KAUFMAN, D. W., 1975, Genetic population structure and breeding sys- tems. In MARKERT, С. L., ed., /sozymes IV: genetics and evolution, p. 27-48. Academic Press, N.Y. SIMON, C. & ARCHIE, J., 1985, An empirical demonstration of the lability of heterozygosity estimates. Evolution, 39: 463-466. SNYDER, J. P. & GOOCH, J. L., 1973, Genetic differentiation in Littorina saxatilis (Gastropoda). Marine Biology, 22: 177-182. SOULE, M., 1976, Allozyme variation: its determi- nants in space and time. п AYALA, Е. J., ed., Molecular evolution, р. 60-77. Sinauer, Sunderland, Mass. TRACEY, М. L., NELSON, K., HEDGECOCK, D., SHLESER, В. A. & PRESSICK, М. L., 1975, Biochemical genetics of lobsters: genetic varia- tion and the structure of American lobster (Homarus americanus) populations. Journal of the Fisheries Research Board of Canada, 32: 2091-2101. TURNER, R. D., 1966, A survey and illustrated catalogue of the Teredinidae. Museum of Com- parative Zoology, Harvard University, Cam- bridge, Mass., 265 p. TURNER, R. D., 1973, Wood-boring bivalves, op- portunistic species in the deep sea. Science, 180: 1377-1379. TURNER, R. D. 8 JOHNSON, A. C., 1971, Biology of marine wood-boring molluscs. In JONES, E.B.G. & ELTRINGHAM, S.K., ed., Marine bor- ers, Fungi, and fouling organisms of wood, p. 259-301. Organization for Economic Coopera- tion and Development, Paris. VALENTINE, J. W., 1976, Genetic strategies of adaptation. In AYALA, F. J., ed., Molecular evo- lution, р. 78-94. Sinauer, Sunderland, Mass. WILKINS, N. P., 1975, Genic variability in marine Bivalvia: implications and applications in mollus- can mariculture. Proceedings of the 10th Euro- pean Symposium on Marine Biology, 1: 549-563. WILKINS, М. P. & O'REGAN, D., 1980, Genetic variation in sympatric sibling species of Littorina. Veliger, 22: 355-359. ZOUROS, E. & FOLTZ, D. W., 1984, Possible explanations of heterozygote deficiency in bi- valve molluscs. Malacologia, 5: 583-591. HOAGLAND 337 APPENDIX. Allele frequencies for ten populations of Pholadacea. Т.Б. navalis; L.f. = Lyrodus floridanus; L.b. = Lyrodus bipartitus; B.g. fimbriatula; M.s. = Martesia striata. Localities as in Table 1. Teredo bartschi; T.n. = Teredo Bankia gouldi; B.f. = Bankia и Populations ТБ, TED: lsh. ЕП: T.n. LT Edo]. B.g. Bf. Ms. Locus Allele NJ. Conn. Fla. NJ. Conn. Miami Ft. Pierce М Fla: Fla. AcPh 98 0.00 0.00 0.00 1.00 0.00 0.00 0.00 0.00 0.00 0.00 100 1.00 1.00 1.00 0.00 1.00 0.00 0.09 100 0.00 1.00 103’ 0100" 000 40100" 000 000 1.00 0.91 0.00 1.00 0.00 АО | 85 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 97 0:00 042 000 000 073 0.43 0.00 0.29 0.00 0.00 100 1.00 0.58 1100’ ‘1:00’ 0.27 0.57 0.82 (07120: 85 50:00 102 0.00 0.00 0.00 0.00 0.00 0.00 0.18 0.00 0.15 0.00 ААТ 78 000 000 000 000 0100 0.00 0.00 0.00 0.00 0.08 80 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.22 82 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.70 93 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.86 0.05 0.00 95 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.14 0.69 0.00 97 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.26 0.00 99 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 100 1.00 1.00 1.00 1.00 1.00 0.00 0.00 0.00 0.00 0.00 EST МА | 937 0005 000, 000 70:00 0100 0.00 0.00 0.00 0.00 0.98 95 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 98 0.007 000” (0/027 0.007 0:00 0.00 0.00 0.00 0.00 0.00 100 0.96 100 0.98 0.96 0.50 0 96 1.00 0.16 0.00 0.00 102 0.02 0.00 0.00 0.00 0.50 0.00 0.00 0.00 0.00 0.00 103 0.00 0.00 0.00 0.04 0.00 0.04 0.00 0.84 1.00 0.00 104 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 EST NA II 98 0.00 0.00 0.00 0.00 0.00 0.00 0.05 0.00 0.00 0.68 100 0.96 0.13 0.00 0.00 0.00 0.00 0.90 0.74 100 DIS? 101 0.00 0.08 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 102 0.4 0.79 1.00 0.84 0.45 0.04 0.05 0.26 0.00 0.00 1032.0.002=0:0022 0:00’ 0108 0100 0.96 0.00 0.00 0.00 0.00 105 0.00 0.00 0.00 0.08 0,55 0.00 0.00 0.00 0.00 0.00 GPI | 100 1.00 100 047 0.00 0.00 0.00 0.00 0.00 0.00 0.00 102 0.00 0.00 0.00 0.00 0.00 0.00 0.00 OOOMNOOF 10105 103 0.00 0.00 0.53 0.00 0.00 0.00 0.63 0.00 0.00 0.00 1052 (0:00) 0.00) 000 95 0.19 0.20 0.37 0.57 S070 10:50 107 0.00 0.00 0.00 0.34 0.74 0.76 0.00 0.02 0.00 0.35 109 0.00 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.07 111 0.00 0.00 0.00 0.06 0.07 0.00 0.00 0.36 0.00 0.03 11157 70.002 40:00) 0.00’ 009 0.00 0.00 0.00 0.05 0.00 0.00 НЕХ 91 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.90 94 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.10 95 0.00 0.00 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 97 000110100’ (0102 0:73. 20:94 0.04 1.00 0.68 0.35 0.00 100 1.00 100’ 098 0:27 20:02 0.96 0.00 O-25 065000 102 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.07 0.00 0.00 IsDH | 94 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 100 1.00 1.00 1.00 1.00 1.00 1.00 0.00 00:00:00 IsDH II 95 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 97 0000100 000 000 000 0.03 0.73 0.00 0.00 0.00 100 1.00 1.00 1.00 0.92 0.00 0.97 0.27 000’ 0.5411 10'00 102 0.00 0.00 0.00 0.08 1.00 0.00 0.00 0.00 0.00 0.00 103 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.46 0.00 null 000 000 000 000 000 000 0.00 000 000 100 338 GENETIC VARIATION IN WOOD-BORING PHOLADACEA APPENDIX (Continued) Populations T.b. T.b. T.b. T.n. Ten! [PE L.b. B.g. BE IMLS. Locus Allele NJ. Conn. Fla. N.J. Conn. Miami Ft. Pierce NJ. Fla. Fla. LDH 96 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.00 0.00 0.00 97 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 100 1.00 1.00 1.00 0.00 0.00 1.00 0.00 0.00 0.25 0.00 101 0.00 0.00 0.00 1.00 1.00 0.00 0.00 1:00. 0750100 LAP 96 0.00 0.00 0.00 0.07 0.00 0.00 0.00 0.07 0.00 0.00 99 0.00 0.00 0.00 0.03 0.01 0.00 0.00 0.00 0.00 0.00 100 1.00 0.00 0.09 0.64 0.34 0.00 0.00 0.51 0.56 0.00 101 0.00 0.00 0.00 0.00 0.00 0.00 0.13 0.00 0.00 0.00 102 0.00 OOOO MOMO 10715 0.02 0.28 0.26 ПОЗ 103 0.00 0.00 0.91 0.00 0.01 0.00 0.00 0.00 0.00 0.00 104 0.00 0.00 0.00 0.12 0.48 0.98 0.50 0.16 0.00 0.00 105 0.00 0.00 0.00 0.04 0.01 0.00 0.09 0.00 0.00 0.28 106 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.09 108 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.50 МР! 92 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.09 0.15 0.00 94 0.00 0.00 0.00 0.19 0.02 0.70 0.36 0.66 0.65 0.00 96 0.00 0.00 0.00 0.25 0.73 0.00 0.08 0.14 0.00 0.00 98 0.00 0.00 0.00 0.45 0.25 0.23 0.38 9 029 ©7171 100 1.00 1.00 0.98 0.11 0.00 0.07 0.18 0.00 0.00 0.16 102 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.00 0.00 0.73 МОН | 95 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.98 0.00 0.00 97 0.00 0.00 0.00 1.00 1.00 0.00 0.03 0.00 0.00 0.00 100 1.00 1.00 1.00 0.00 0.00 1.00 0.00 0.02 1.00 0.00 101 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 103 0.00 0.00 0.00 0.00 0.00 0.00 0.91 0.00 0.00 0.00 104 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.98 105 0.00 0.00 0.00 0.00 0.00 0.00 0.06 0.00 0.00 0.00 109 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 Pep G II 98 0.00 0.00 0.04 0.00 0.00 0.00 0.00 0.00 0.00 0.00 100 1.00 1.00 0.96 0.25 0.36 0.00 0.00 0.0372. 0212=.0:00 103 0.00 0.00 0.00 0.00 0.02 0.00 0.08 0.50 0.69 0.54 105 0.00 0.00 0.00 0.00 0.58 1.00 0.82 0.16 0.06 0.28 106 0.00 0.00 0.00 0.00 0.00 0.00 0.08 0.00 0.00 0.00 107 0.00 0.00 000’ 2.0775 0.04 0.00 0.00 031 7 0:04220412 109 0.00 0.00 000) 0.00’ (0700 0.00 0.02 0.00 0.00 0.06 Pep G Ш 90 000’ 0:00’ 0100) 0:00 0:00 0.00 0.00 0.09 0.00 0.00 93 0.00 0.00 0.00 0.00 0.00 0.00 1.00 0.11 0.00 0.00 95 0.00 0.00 0.00 0.00 0.28 0.00 0.00 0540005050 98 0.00 0.00 0.00 0.24 0.20 0.00 0.00 0.66 0.00 0.45 100 1.00 1.00 100 0.76’ 038 1.00 0.00 0.00 0.00 0.37 102 0.00 0.00 0.00 0.00 0.12 0.00 0.00 0.00 0.00 0.00 103 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 0.88 0.00 104 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.08 105 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.007 04122=0:00 PGM 95 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.02 97 000 000 000 005 70:08 1.00 0.00 0.09 0.08 0.11 100 1.00 1.00 12000 Oli 0.32 0.00 0.85 0.09 0.50 0.09 102 0.00 0.00 0.00 0.69 0.56 0.00 0.13 0.75 0.08 0.78 103 000 0.00 0.00 0.00 0.00 0.00 0.02 0.00 0.00 0.00 104 000 0.00 0.00 0.15 0.04 0.00 0.00 0.07 0.30 0.00 106 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.04 0.00 HOAGLAND 339 APPENDIX (Continued) Populations Ted: Ile: ТВ: T.n. El} т. L.b. B.g. Bit M.s. Locus Allele М... Conn. На. N.J. Conn. Miami Ft. Pierce NJ. Fla. Fla. 6-PGD 100 1.00 1.00 1:00) 1.00 1.00 1.00 1.00 0.00 0.00 0.92 101 0.00 0.00 0.00 0.00 0.00 0.00 0.00 1.00 1.00 0.00 105 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.08 SoDH 94 0.00 0.00 0.00 0.86 1 00 0.06 1.00 0.03 1.00 0.00 98 0.00 0.00 0.00 0.14 0.00 0.94 0.00 0.83 0.00 0.80 100 1.00 1.00 1.00 0.00 0.00 0.00 0.00 0.14 0.00 0.20 SOD | 95 0.00 0.00 0.00 1.00 1.00 0.00 0.00 0.00 0.00 0.00 100 1.00 1.00 0.98 0.00 0.00 1.00 1.00 0.96 1.00 0.00 102 0.00 0.00 0.02 0.00 0.00 0.00 0.00 0.04 0.00 1.00 106 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 XDH 100 1.00 1.00 1.00 1.00 1.00 1.00 1.00 0025002251200 Revised Ms. accepted 10 April 1986 I I 1 N р f ans \ р \ | et i \ i y ar ® Г i | | LS ie — ii Inn I i y 1 | Г i i | . "IN DE j i i \ / i 1 | TD, rl : р И t | u | ' | i ) fh | | i я 1 i i u Hi TA il i | MALACOLOGIA, 1986, 27(2): 341-355 ENVIRONMENT AND DIVERSITIES OF FOREST SNAIL FAUNAS FROM COASTAL BRITISH COLUMBIA R. A. D. Cameron Department of Extramural Studies, University of Birmingham, P. O. Box 363, Birmingham B15 2TT United Kingdom ABSTRACT The snail faunas of 38 lowland forest sites from Vancouver Island, the lower Frazer Valley and the Chilliwack Valley, British Columbia, are described and related to environmental variation between sites. Characters of the litter and soil, and the associated vegetation appear to explain most of the variation in diversity and abundance of snails between sites. Sites with mor litter, usually dominated by Douglas Fir, and with dry soils, have impoverished faunas in which only five species are found frequently. Sites with mull litter and damper soils, in which Western Red Cedar and various hardwoods are important, are much richer, with 14 species occurring frequently, and 26 species recorded overall. Successional status and disturbance by logging or burning have lesser effects, involving only a few species, except where disturbance 1$ very recent. The faunas are discussed in a regional context, and compared with those from temperate forests elsewhere. Diversity is greater than in most other coniferous forests, but less than in the richest deciduous forests. These differences are accompanied by differences in the range of morphology of snails shown, a study on which is forthcoming. INTRODUCTION Studies on the land molluscs of the Pacific North-West of North America have been con- cerned with taxonomy and geographical dis- tribution (Henderson, 1929, 1936; Pilsbry, 1939-1948; Branson, 1977) and ecological information is scanty, except for some intro- duced species (Rollo & Wellington, 1975; Roth 8 Pearce, 1984). The forests of the coastal Pacific North- West occur in an area with a mild oceanic climate favourable to snails. This climate is generally similar to that of areas originally occupied by temperate deciduous forests in western Europe, except that rainfall is more seasonal, summers being drier (Waring & Franklin, 1979). This pattern of precipitation may explain the predominance of conifers. The climate, the infrequency of natural fires, and the quality of litter produced by some of the dominant species may account for the unusually high botanical diversity for conifer- ous forest (Franklin & Dyrness, 1973). This study surveys snail faunas from for- ests on a variety of soils and in different successional states. It examines environmen- tal correlations with diversity, and compares the diversity of snails found with that found in other temperate forest regions. A later paper (Cameron, in preparation) will discuss the range of morphological types within these faunas, and compare it with the ranges found elsewhere, as an indication of the extent to which similar niches are occupied. METHODS Sampling of molluscs Sample sites were confined to areas of relatively uniform vegetation within a 30 x 30 m square. Within this, molluscs were hand- collected for one hour, paying particular atten- tion to logs, rocks and tree-trunks. Litter was collected in small amounts from all over the site, to a total volume of about 5 litres. This litter was bagged, returned to the laboratory, and passed through a coarse mesh sieve to remove large debris, any snails seen being removed. The residue was oven-dried for 24 hr and immersed in water. Floating material was removed, re-dried and passed through a graded series of sieves. Material passing through the smallest mesh (0.5 mm), was (341) 342 CAMERON discarded, and the remaining material searched for snails under bright lights. This method reliably extracts intact shells (Cameron & Morgan-Huws, 1975; Cameron, 1982). Snails alive at the time of sampling dry out, and air in all intact shells causes them to float. Slugs are lost in this process, and since searching in dry weather is also inadequate (Wareborn, 1969), slug records are not analysed in detail. Site descriptions Records were made at each site of the presence or absence of all canopy tree spe- cies, and of shrub and field-layer species chosen for conspicuousness, frequent occur- rence and ease of identification. Since each site was not searched exhaustively, recorded absence may include rarity or unusual incon- spicuousness. For presentation and analysis all species recorded in less than one fifth of the sites for which full records were made have been omitted. Dominance in the canopy was also noted. At some sites, two or more species were recorded as co-dominant. Litter was recorded as mor, mull or inter- mediate, the latter category including moders, and cases where the litter showed great vari- ation within the site (Klinka et a/., 1981). Soil moisture was ranked subjectively by feel ona five point scale, later condensed to three categories for analysis. There was virtually no precipitation during the sample period (Au- gust 14-27, 1984). Timing of major disturbance, such as log- ging or crown fires, was estimated on the sizes of the dominant trees. For analysis, disturbance has been reduced to three cate- gories: disturbed in last 40 years, disturbed between 40 and 100 years ago, any distur- bance more than 100 years ago. Many sites in the last category had trees 300 or more years old. Lesser disturbances, such as ground fires, could have escaped detection. NOMENCLATURE AND IDENTIFICATION OF MOLLUSCS With the few exceptions noted below, no- menclature follows Branson (1977). The great majority of specimens presented no difficulty in identification; Pilsbry (1948) and Pearce (in preparation) were used as aids. Retinella electrina and R. binneyana oc- cidentalis are recorded in the region (Henderson, 1929; Pilsbry, 1948). Their sta- tus seems somewhat unclear, and in the past they have been ascribed to European species now placed in the genus Nesovitrea (Walden, 1966). All specimens found in this study have been ascribed to binneyana occidentalis, and | have followed Walden in using the genus Nesovitrea. Branson (1975) refers specimens from the Olympic Peninsula, Washington, which re- semble Striatura pugetensis to Radiodiscus hubrichti. This ascription is rejected by Solem (1977) on the basis of shell microsculpture, and all shells of this type have been identified as S. pugetensis. A large species of Vertigo was found at sites 27 and 28 near Lake Horne, Vancouver Island. The specimens, which are clearly dis- tinct from the V. columbiana and V. an- drusiana found elsewhere, have been provi- sionally identified as V. rowelli, which is, however, not recorded north of Oregon (Pilsbry, 1948). Haplotrema sportella was generally easily distinguished from H. vancouverense by small size (11-16 mm diameter), the marked convex depression of the upper surface of the last whorl at the mouth, and by pronounced radial ribbing. At site 7, where H. vancou- verense was not found, it retained the shell characters, but was much larger (19-22 mm diameter), overlapping the range of H. vancouverense. Specimens of all species collected are held by the author. SAMPLE SITES AND THEIR CHARACTERISTICS Fig. 1 shows the distribution of sample sites, which range from the west coast of Vancouver Island to the Chilliwack Valley. All sites are below 350 m above sea level and lie in the Coastal Western Hemlock and Coastal Douglas Fir zones of Krajina (1969), although those in the extreme W are in the Sitka Spruce zone differentiated by Franklin & Dyr- ness (1973). Precipitation ranges from 1200 to 3000 mm annually (Anon., 1982); most lie in the 1600-2000 mm range characteristic of the transition between the Hemlock and Douglas Fir zones (Krajina, 1969). The sites are in no sense a random selec- tion of stands. They were chosen to give a range of successional and edaphic status, ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS 343 BRITISH 19,20 21,22, 23 Lake Horne . 27, 2873 Cameron Lake 22300 - Ma, 35,36 — ah PORT ALBERNI Sproat Lake Cowichan Lake 7 37, 38 COLUMBIA a. A Г | 7 EOS CHILLIWACK | ASSET “12,13 1 @-victor lA FIG. 1. Map of the lower Frazer valley and south Vancouver Island, showing the distribution of sample sites. and in particular to include examples of bottomlands and limestone substrate on which mull litter might be found. Four sites out of the 38 sampled had either suffered gross recent disturbances or were without full canopy cover. Two, sites 6 and 13, were on limestone crags with scrub by road- sides in the Chilliwack Valley, one, site 23, was on a grassy bluff partly overhung by Douglas Fir, and the last, site 24, was in a stand of large Douglas Fir under which there had been an intense ground fire in the last twelve months. These sites are omitted from the vegetation analysis, and their snail faunas are considered separately in some analyses. Full environmental data for the remaining 34 sites are given in Appendix A. Given the limited and partly subjective na- ture of the environmental data, sophisticated statistical analysis has not been attempted. Table 1 shows strong associations between mor litter, dry soils and dominance of Douglas Fir, between intermediate litter and soil mois- ture and dominance by Red Alder (sometimes with Douglas Fir or Sitka Spruce), and be- tween mull litter, damp soil and dominance by a number of combinations of canopy species, always including Western Red Cedar, Large- leaved Maple or Cottonwoods. Associations between recorded species have been estimated using Fager’s Index (Southwood, 1966). Fig. 2 shows these asso- ciations as a dendrogram. Two groups of loosely associated species can be identified. Fig. 3 shows the association of each species (and of canopy dominants) with litter type. Members of one group (A) associate with mor litter, those in the other (C) with mull litter. In all but two cases, associations with interme- diate litter are intermediate between those with mor and mull. The two exceptions, oc- currence of Western Hemlock and dominance by Red Alder, relate to disturbance (Table 2). Red Alder is a characteristic pioneer, while Western Hemlock is the presumed climax dominant (Krajina, 1969; Franklin & Dyrness, 1973). Their associations are concordant with their successional status. These patterns of site characteristics are concordant with the pattern of plant associa- tions in the region (Krajina, 1969; Franklin & Dyrness, 1973). Moisture and nutrient gradi- ents are the prime determinants of community composition, with Douglas Fir and Gaultheria shallon characterizing dry, oligotrophic sites, and Cedar and Polystichum munitum the damper and more enriched ones. Most sites investigated have some degree of disturbance and even those not obviously logged are likely to have been subject to occasional wildfires (Franklin & Hemstrom, 1981). Mueller-Dombois (1965) shows that the understory and field layer species compo- sitions associated with particular edaphic conditions are not destroyed by logging, and the association of Cedar, Large Leaved Ma- 344 CAMERON TABLE 1. Numbers of sites with particular combinations of soil, litter and canopy dominants. For Dominants, A = Douglas Fir, Sitka Spruce and any combination of those with Red Alder. Alder = Dominated by Red Alder alone. B = any combination of Dominants which includes Western Red Cedar, Large-leaved Maple or Cottonwoods. Litter Litter Soil Mor Int Mull Dominants Mor Dry 6 0 0 A 8 Int. 1 5 2 Alder 0 Damp 1 4 15 B 0 ple and Cottonwoods with favourable edaphic conditions is widespread (Franklin & Dyrness, 1973; Franklin & Hemstrom, 1981). On this basis the sites have been allocated to categories by litter type for analysis of snail faunas but some attention is given to distur- bance and moisture regime in particular in- stances. THE MOLLUSCAN FAUNA Appendix B gives the numbers of each snail species found at each site and the presence or absence of slugs. Twenty-six species of snail and 7 of slug were recorded in the study. Numbers are based on speci- mens alive at time of sampling, and obviously fresh shells. Old, badly worn shells (usually broken) are excluded. They constituted only a small proportion of finds. Tables 3 and 4 show the relationships of snail diversity and numbers to litter type for the 34 sites in the main series. Since neither numbers of species nor of individuals are normally distributed, medians and ranges are used in preference to means and standard errors. Sites with mor litter have fewest species, and fewest individuals overall and species by species, while those with mull litter have the most. Five species occur in more than half the mor sites, while 14 do so in the mull. No species 15 characteristic of either mor or inter- mediate litter, but 10 species are found only in ти! sites. Differences in median number of species in each litter type are highly signifi- cant, both overall (Kruskall-Wallis test, p < 0.001) and in paired comparisons (Mann- Whitney test, p<0.001 in each case). Similarities between snail faunas at each site are estimated using Sorensen's Index (Southwood, 1966). Fig. 4 shows relation- ships between sites as a dendrogram, and Soil Int. Mull Dominants Dry Int. Damp 5 0 A 6 5 2 4 1 Alder 0 2 3 0 16 B 0 1 15 Fig. 5 orders similarities with a typical mor and a typical mull site. Associations of sites related closely to litter type, and by inference, to the associated vegetation. Sites closely associated with those of a different litter type have either the highest (mor and intermedi- ate) or lowest (mull) diversities for their type. Figs. 3 and 4 include the four aberrant sites. The sites from limestone crags (6 and 13) associate completely with the ти! sites, showing no reduction in diversity. The other two (23 and 24) are more isolated. Site 7 also stands apart, though clearly closest to the mull series to which it belongs. A few species show signs of association with environmental factors other than litter Douglas Fir Salal Oregon Grape Huckleberry Hemlock Alder Thimbleberry Sword Fern L.L. Maple Cedar Dogwood Devil's Club Skunk Cabbage 90 80 7707526055550 Index of Affinity FIG. 2. Dendrogram of associations between plant species, links made with closest member of each group. Scientific names of plants are given in Appendix A. ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS 345 +1 0 +1 | Dominants 'A' Western | Hemlock | 0 | | ı B _1 | +1 C Alder Sword Fern = Thimbleberry Cedar de Oregon Salal Grape A [en Huckleberry Douglas Fir — —— — — — — — — — — — =— — — —- Dominants ee Dominant Skunk Cabbage L.L. Maple Ty Devil's Dogwood FIG. 3. Frequency distributions of plant species occurrence and dominance as a proportion of maximum deviation from random expectation. Section A: occurrences associated with mor litter: Section B: Canopy dominance by ‘A,’ Douglas Fir, Sitka Spruce or either with Alder; Alder: Red Alder alone. 'B,' any combination including Cedar, Maple or Cottonwood. Section C: occurrences associated with mull litter. The left column for each diagram indicates the association with mor litter, the centre column the association with intermediate litter, and the right association with mull litter. type. Triodopsis germana shows a stronger association with damp soil than with mull litter, while Carychium occidentale and to a lesser extent Monadenia fidelis occur less often in the more recently disturbed sites, even within the mull litter series. Microphysula cookei was found only at one site in the main series, but at three of the disturbed sites. These sites have in common the presence of exposed limestone rock, which, however, also occurs at site 28 in which M. cookei was not found. A number of other species occur at only a few sites. More samples would be needed to determine any geographical or habitat pat- terns, although the restriciton of Allogona townsendiana to the vicinity of the Chilliwack Valley is in accord with its previously known 346 CAMERON TABLE 2. Numbers of sites containing Alder and Hemlock, classified by time since last major disturbance. Disturbance occurred A B C Sites with less than 40 yr ago 40-100 yr ago More than 100 yr ago x? A+B:C Alder 4 10 8 4.63 P <0.05 Hemlock 0 1 8 7.40 P <0.01 Total sites 12 Ue TABLE 3. Medians and ranges for numbers of species and individuals of snails found in sites with different litter types, and the number of species occurring in given frequency ranges. No. spp. occurring in No. spp. per site No. individs. per site n Median Range Median Range 100% 50%-99% <50% Total Mor 8 5:5 ЭЙ 28 17—55 3 2 4 9 Intermediate 9 10.0 7-11 57 22-170 4 6 6 16 Mull 17 15.0 11-17 151 67-226 6 8 12 26 distribution in British Columbia (Pilsbry, disturbance, and may have been spread lo- 1948). Amongst the slugs, Ariolimax columbianus was recorded at nearly all sites, and was usually abundant. The introduced Arion ater was found in the Chilliwack Valley, and at sites close to Vancouver. Deroceras laeve shows a pronounced association with damp soil. Other species were recorded too infre- quently to interpret. DISCUSSION The regional context None of the species recorded here, with the exception of Vertigo rowelli (see above), are outside their previously known range. Many other species are known from the region. Many of these are characteristic of higher altitudes, of wetlands, or of open habitats. (Henderson, 1929, 1936; Pilsbry, 1948; Branson, 1977). In open and disturbed sites there are a number of introduced species, of which only Arion ater appears to have pene- trated the more natural sites studied here, although Vallonia pulchella is also recorded from one site (Branson, 1977; Rollo & Wellington, 1975; Roth & Pearce, 1984). Cionella lubrica was found at four sites. Al- though native, it is frequently associated with cally by human activity (Roth & Pearce, 1984). The faunas described here are very similar to those recorded by Myers (1972) in a small but detailed study in the Western Cascades of Washington, and to those recorded by Branson (1977) on the Olympic Peninsula, if comparisons are restricted to forested sites below 500 m above sea level. Pearce (in preparation) records a similar set of species from Thurston County, Puget Sound, Wash- ington. Some species found here have wide geo- graphical ranges, with some being Holarctic (Pilsbry, 1948; Kerney & Cameron, 1979). There are some affinities with faunas from the mountain belt, and with those from Albertan prairie parkland to the east of the continental divide (Karlin, 1961; Platt, 1980; Boag & Wishart, 1982; Van Es & Boag, 1981). Both in Alberta and further south, there are some species, especially Discus cronkhitei, Zonitoides arboreus and Vitrina alaskana which are both abundant and widespread, but which are rare and restricted in the Pacific North-West, at least at lower altitudes (this study; Branson, 1977). Diversity, abundance and the environment The results of this study are in agreement with the generally observed pattern of high ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS 347 TABLE 4. Frequency of occurrence and median numbers (where present) for each species of snail in sites classified by litter type. A = species occurring in all or nearly all sites; B = species occurring in all litter types, but least frequent in mor; С = species absent from sites with mor litter; D = species found only in sites with mull litter. Mor Intermediate Mull n 8 п —9 = м % Median % Median % Median occur- no. where occur- no. where occur- no. where rence present rence present rence present Comments A Haplotrema vancouverense 100 3 100 3 94 5 Not at Popkum Pristiloma lansingi 100 4 100 6 100 1174 Striatura pugetensis 100 6 100 11 100 25 Punctum randolphii 87 6 89 10 100 ual B Vespericola columbiana 50 2:5 100 3 100 6 Haplotrema sportella 37 2 89 3 88 6 Nesovitrea binneyana 37 2 67 2.5 100 4 Planogyra clappi 37 2 56 6 100 11 Columella edentula 25 1 56 2 94 4 C Vertigo columbiana 0 = 89 3 88 4 Triodopsis germana 0 — 33 2 71 3 Euconulus fulvus 0 — 22 1 76 3 Carychium occidentale 0 — 22 26 59 22 Monadenia fidelis 0 — 11 1 82 3 Zonitoides arboreus 0 — 11 1 29 1 Pristiloma stearnsi 0 = 11 2 41 6 Vancouver Island only D Punctum conspectum 0 — 0 — 35 4 Discus cronkhitei 0 — 0 — 18 6 Cameron Lake V.I. only Cionella lubrica 0 = 0 — 18 2 Allogona townsendiana 0 = 0 = 12 2 Chilliwack only Pristiloma johnsoni 0 = 0 = 12 2 Vancouver Island only Vertigo andrusiana 0 = 0 — 12 725 Vancouver Island only Vertigo cf. rowelli 0 — 0 = 12 2.5 Lake Horne, \.1. only Microphysula cookei 0 — 0 = 6 10 Also in disturbed sites Vitrina alaskana 0 = 0 = 6 5 Popkum Vallonia pulchella 0 — 0 = 6 3 Popkum diversity and abundance associating with damp and nutrient-rich substrates in temper- ate forests (Burch, 1955; Wareborn, 1969; Cameron, 1973, Coney et al., 1982). The faunas on mor and intermediate litter are merely impoverished versions of those found on mull. In some other studies of temperate forest, where moisture and litter type are not so closely associated, a few species are characteristic of more oligotrophic sites (Cameron, 1973; Coney et al., 1982). Soil, litter and vegetational characters are usually strongly associated, and even sophis- ticated multivariate analysis cannot certainly determine causal relationships with the snail fauna. Soil moisture and calcium availability can affect snails directly (Boycott, 1934; Wareborn, 1969, 1982). Vegetation may have less direct effects, through the creation of shelter, through variation in the quantity and availability of calcium salts in the litter, and through the effects of litter chemistry on soil structure and content (Wäreborn, 1969, 1979; Coney et al., 1982). No chemical analyses of litter were carried outin this study. Myers (1972) gives some data on litter under pure stands of Cedar, Red Alder and Douglas Fir. As in this study, snail faunas were most diverse and abundant un- der Cedar, and least under Douglas Fir, and this correlates with moisture content and chemical composition. Coniferous forests are often reported as having very impoverished and sparse mollus- 32 35 348 CAMERON au lu er 949000000000 8OOL86 OG 100 90 = E Е on o o 00 о 70 FIG. 4. Dendrogram of affinities between the snail faunas at each site, links made with the closest member of each group. Open circles = sites with mor litter; stippled circles = sites with intermediate litter; black circles = sites with mull litter; crossed circles = disturbed or open sites. can faunas. Many coniferous forests have mor litter and podsolized soils, and often occur in climatic regimes less favourable to snails than temperate hardwood forests. In both European and North American studies, conifer stands, especially of Pinus or Picea on podsols or peat, are very impoverished, and may hold only one or two species (Bless, 1977; Matzke, 1965; Favre, 1927; Burch, 1956; Karlin, 1961), but where soil and litter conditions are more favourable, rich faunas are found, often far exceeding those of oligotrophic hardwood stands (Favre, 1927; Frómming, 1958; Burch, 1956). Many coniferous forests studied are also unstable, being successional regrowth follow- ing logging or fire; young stands have a particularly dense canopy, which may tempo- rarily extinguish the ground flora. Second growth conifer stands in the central Russian plain have much lower densities and diversi- ties of snails than mature old-growth forests with a more open canopy (Shikov, 1984). Some caution is necessary in interpreting diversity. In nearly all the studies quoted above, low diversity is associated with low density, and many species have highly aggre- gated distributions even within the confines of a single site. The chances of failing to detect a species clearly increase as densities fall. Karlin (1961) noted that although coniferous forests in the mountains of Colorado and New Mexico had very few species, successional stands of Aspen within them had diverse faunas. He suggests that the complete suite of species is present throughout, but in- creases dramatically in density during an Aspen-dominated successional episode. Boag & Wishart (1982) found that a pure Spruce stand supported about the same num- ber of species as did deciduous and mixed stands, but that densities were lower. Each site was visited on several occasions, reduc- ing the effect of sample-size or density on records of occurrence. In the sites studied here, canopy was never ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS 349 90 O 70 Affinity with Site 22 30 30 90 10 90 Affinity with Site 34 FIG. 5. The scatter of snail faunas from each site ordered on their affinities with that of site 22 (mor) and site 34 (mull). Open circles = mor litter, stippled circles = intermediate litter, black circles = mull litter, circles with cross = open and disturbed sites. so dense as to eliminate the ground flora, and Douglas Fir litter is less acidic and nutrient deficient than that of some Pines and Spruces. The total array of species recorded in mor litter is less than that recorded for any single undisturbed mull site, and there seems no reason to doubt a genuine difference in diversity between the two types. Most of the sites in this study are disturbed to some degree, and some, dominated by hardwoods, are at quite early stages in the protracted succession to dominance by Hem- lock (Franklin & Dyrness, 1973). As is the case with plants (Mueller-Dombois, 1965), only a few snail species appear to be affected by successional state independently of soil and litter conditions. European temperate for- ests (nearly all of which have been heavily managed for many centuries) also support large and diverse snail faunas, and forests with appropriate soil and litter characters only have a depauperate snail fauna if they have Originated in isolated plantations or from suc- cession on previously open, agricultural land (Boycott, 1934; Paul, 1978; Cameron, 1973; Cameron, Down & Pannett, 1980). A large number of species survive routine forestry operations. 350 CAMERON The upper limit of diversity Allowing a margin of error for species missed in sampling, the upper limit on snail species in a single site in this study could be put at around 20-22 species (observed max- imum 17) and for the forest fauna in this biogeoclimatic zone of about 30 (26 ob- served). Inspection of Branson (1977) and Pearce (in preparation) suggests that Triodopsis devia, Megomphix hemphilli, Nesovitrea electrina, Hawalia miniscula and Vertigo modesta might be found in forests of coastal British Columbia, since they have been found in similar habitats further south. This level of diversity is high for coniferous forest, and perhaps reflects the peculiarly favourable climatic regime of the Pacific North-West compared with other areas where conifers are naturally dominant (Waring & Franklin, 1979). More southerly hardwood and mixed for- ests with favourable soil conditions in N. America have richer faunas, at least in the Appalachian ranges. Coney et al. (1982) record 57 species from 37 forest sites in Tennessee, and Branson & Batch (1970) 45 species in a forest series in Kentucky, with 30 species at the richest site. Other North Amer- ican studies reporting lower diversities are generally from sub-optimal habitats or have used sampling methods inappropriate for pro- ducing a full faunal list. European forest faunas at comparable or more northerly latitudes are also generally more diverse where soil and rock conditions are good. Even the rather dry, rock-free and disturbed woods of the English South Downs, where the dominant Beech Fagus sylvatica produces a rather harsh and slowly decom- posing litter, have a recorded total of 34 species, and a maximum of 24 at any one site (Cameron, 1973). Individual woodland sites with 30—45 species of snail are known from a number of European countries (Cameron, 1978; Long, 1969; Kórnig, 1966; Favre, 1927; Walden, 1981). None of these temperate forest faunas reach the levels of diversity recorded in cer- tain gully woodlands in the North Island of New Zealand (Solem, Climo & Roscoe, 1981; Solem & Climo, 1985; Solem, 1984) where the total forest fauna is in excess of 80 species, and the richest individual sites may contain 60 or more species. Solem (1984) gives a survey of diversity levels in land snails. The causes of such differences in diversity, however, can only be elucidated by detailed studies of niches available and occu- pied in different regions. The snail fauna of the forests of the Pacific North-West differ from those of other temperate forests not only in diversity, but also in the range of morphol- ogies present; a discussion of these differ- ences is in preparation. ACKNOWLEDGEMENTS This study was made possible by the award of a British Ecological Society Travelling Fel- lowship, and by support from the University of Birmingham. | thank Professor G. G. E. Scudder for generously making facilities available at the Department of Zoology, Uni- versity of British Columbia, Dr. Tom Carefoot and Dr. Sandra Millen for advice and assis- tance, Dr. Judith Myers and Dr. Jamie Smith for help and hospitality, and Professor A. J. Cain for access to literature and for his critical reading of the manuscript. Mr T. Pearce helped with the identification of difficult spec- imens, and with discussion of the results. LITERATURE CITED ANONYMOUS, 1982, Canadian Climatic Normals Volume 3: Precipitation 1951-1980. Environ- ment Canada: Ottawa. BLESS, R., 1977, Die Schneckenfauna des Kottenforstes bei Bonn (Mollusca: Gastropoda). Decheniana (Bonn), 130: 77-100. BOAG, D.A. & WISHART, W.D., 1982, Distribution and abundance of terrestrial gastropods on a winter range of bighorn sheep in south western Alberta. Canadian Journal of Zoology, 60: 2633-2640. BOYCOTT, A.E., 1934, The habitats of land Mollusca in Britain. Journal of Ecology, 22: 1-38. BRANSON, В.А., 1975, Radiodiscus hubrichti (Pulmonata: Endodontidae) new species from the Olympic Peninsula, Washington. Nautilus, 89: 47-48. BRANSON, B.A., 1977, Freshwater and terrestrial Mollusca of the Olympic Peninsula, Washington. Veliger, 19: 310-330. BRANSON, В.А. & BATCH, D.L., 1970. An ecolog- ical study of valley forest gastropods in a mixed mesophytic situation т northern Kentucky. Veliger, 12: 333-350. BURCH, J.B., 1955, Some ecological factors of the soil affecting the distribution and abundance of land snails in eastern Virginia. Nautilus, 69: 62-69. BURCH, J.B., 1956, Distribution of land snails in ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS 351 plant associations in eastern Virginia. Nautilus, 70: 60-64. CAMERON, R.A.D., 1973, Some woodland mollusc faunas from southern England. Malacologia, 14: 355-370. CAMERON, R.A.D., 1978, Terrestrial snail faunas of the Malham area. Field Studies, 4: 715- 128: CAMERON, R.A.D., 1982, Life histories, density and biomass in a woodland snail community. Journal of Molluscan Studies, 48: 159-166. CAMERON, R.A.D., DOWN, K. & PANNETT, D.J., 1980. Historical and environmental influences on hedgerow snail faunas. Biological Journal of the Linnean Society, 13: 75-88. CAMERON, R.A.D. & MORGAN-HUWS, D.l., 1975, Snail faunas in the early stages of a chalk grassland succession. Biological Journal of the Linnean Society, 7: 215-229. CONEY, C.C., TARPLEY, W.A., WARDEN, J.C. 8 NAGEL, J.W., 1982, Ecological studies of land snails in the Hiwassee River Basin of Tennes- see, USA. Malacological Review, 15: 69- 106. FAVRE, J., 1927, Les mollusques post-glaciaires et actuels du Bassin du Geneve. Memoires de la Société de Physique et d'Histoire Naturelle de Geneve, 40: 171—434. FRANKLIN, J.F. & DYRNESS, C.T., 1973, Natural vegetation of Oregon and Washington. U.S. Forest Service General Technical Report PNW-8. FRANKLIN, J.F. & HEMSTROM, M.A., 1981, As- pects of succession in the coniferous forests of the Pacific North West. т WEST, D.C., SHUGART, Н.Н. & BOTKIN, D.B., (ed.), Forest Succession, Concepts and Application. Springer Verlag, New York. FROMMING, E., 1958, Schnecken im Nadel- holzwald. Biologisches Zentrallblatt, 1: 54-63. HENDERSON, J., 1929, The non-marine Mollusca of Oregon and Washington. University of Colo- rado Studies, 17: 47-190. HENDERSON, J., 1936, The non-marine Mollusca of Oregon and Washington—supplement. Uni- versity of Colorado Studies, 23: 251-280. KARLIN, E.J., 1961, Ecological relationships be- tween vegetation and the distribution of land snails in Montana, Colorado and New Mexico. American Midland Naturalist, 65: 60—66. KERNEY, M.P. & CAMERON, R.A.D., 1979, Field- Guide to the land snails of Britain and north-west Europe. London: Collins. KLINKA, K., GREEN, R.N., TROWBRIDGE, R.L. & LOWE, L.E., 1981, Taxonomic classification of humus forms in ecosystems of British Columbia. Land Management Report 8, Ministry of Forests, Province of British Columbia. KORNIG, G., 1966. Die Molluskengesellschaften des mitteldeutschen Húgellandes. Malakolog- ische Abhandlungen Staatliches Museum für Tierkunde in Dresden, 2: 1-112. KRAJINA, V.J., 1969, Ecology of forest trees in British Columbia. Ecology of Western North America, 2: 1-146. LONG, D.C., 1969, A small bog in Cranham Wood. Conchologist's Newsletter, 30: 111-113. MATZKE, M., 1965, Die Molluskenfauna in den Forsten und Waldern bei Lichtenstein am Fusse des Erzgebirges. Malakologische Abhandlungen Staatliches Museum fúr Tierkunde in Dresden, 1: 139-157. MUELLER-DOMBOIS, D., 1965, Initial stages of secondary succession in the coastal Douglas-Fir and Western Hemlock zones. Ecology of west- ет North America, 1: 38—41. MYERS, L.D., 1972, Primary and secondary influ- encing agents on gastropod populations of three habitats in Washington State. Sterkiana, 47: 39-45. - PAUL, C.R.C., 1978, The ecology of Mollusca in ancient woodland. 3: Frequency of occurrence in West Cambridgeshire woods. Journal of Conchology, 29: 295-300. PILSBRY, H.A., 1939, Land Mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia Monographs, 3, 1(1): 573 p. PILSBRY, H.A., 1940, Land Mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia Monographs, 3, 1(2): 418 p. PILSBRY, H.A., 1946, Land Mollusca of North America (north of Mexico). Academy of Natural Sciences of Philadelphia Monographs, 3, 2(1): 520 p. PILSBRY, H.A., 1948, Land Mollusca of North America (North of Mexico). Academy of Natural Sciences of Philadelphia Monographs, 3, 2(2): 593 p. PLATT, T.R., 1980, Observations on the terrestrial gastropods in the vicinity of Jasper, Alberta (Canada). Nautilus, 94: 18-21. ROLLO, C.D. & WELLINGTON, W., 1975, Terres- trial slugs in the vicinity of Vancouver, British Columbia. Nautilus, 89: 107-115. ROTH, В. & PEARCE, T.A., 1984, Уйгеа contracta (Westerlund) and other introduced land mollusks in Lynnwood, Washington. Veliger, 27: 90-92. SHIKOV, E.V., 1984, Effects of land use changes on the land mollusc fauna in the central portion of the Russian plain. In: SOLEM, A. & VAN BRUG- GEN, A.C. (ed.), World-wide snails. Brill, Leiden. SOLEM, A., 1977, Radiodiscus hubrichti Branson, 1975, a synonym of Striatura (S.) pugetensis (Dall, 1895) (Pulmonata: Zonitidae). Nautilus, 91: 146-148. SOLEM, A., 1984, A world model of land snail diversity and abundance. /n SOLEM, A. & VAN BRUGGEN, A.C., (ed.) World-wide snails. Brill, Leiden. SOLEM, A. & CLIMO, F.M., 1985, Structure and habitat correlations of sympatric New Zealand land snail species. Malacologia, 26: 1-30. SOLEM, A., CLIMO, Е.М. & ROSCOE, D.J., 1981, Sympatric species diversity of New Zealand land 352 CAMERON snails. New Zealand Journal of Zoology, 8: 453-485. SOUTHWOOD, T.R.E., 1966, Ecological methods. Methuen, London. VAN ES, D.A. 8 BOAG, J., 1981, Terrestrial mol- luscs of Central Alberta. Canadian Field Natural- ist, 95: 75-79. WALDEN, H., 1966, Zur Frage der Taxonomie, Nomenklatur und Okologie von Nesovitrea ham- monis (Strom) und petronella (L. Pfeiffer). Archiv für Molluskenkunde, 95: 161-195. WALDEN, H., 1981, Communities and diversities of land molluscs in Scandinavian woodlands, |. High diversity communities in taluses and boul- der slopes in S.W. Sweden. Journal of Conchology, 30: 351-372. WAREBORN, !., 1969, Land molluscs and their environments in an oligotrophic area in southern Sweden. Oikos, 20: 461-479. WAREBORN, 1., 1979, Reproduction of two spe- cies of land snails in relation to calcium salts in the foerna layer. Malacologia, 18: 177-180. WAREBORN, I., 1982, Environments and molluscs in a non-calcareous forest area in Southern Sweden. Ph.D. Thesis. Lund University. WARING, R.H. & FRANKLIN, J.F., 1979, Ever- green coniferous forests of the Pacific North- west. Science, 204: 1380-1386. 353 ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS at + + + за 001 > Ag JON 8€ + + + Je + + 3 М 001=> ul ju] Ze 9487 UBYDIMOD + ++ + + IM POEN + 001 dweg INN 9€ + + + + + + + + + ‘po +001 dweq INN Ge + + + + Je de + + + OPD + 001 dweg INN ve aye] ио1эшеЭ A du i + IV or > dueq yul 55 + + + Hr 2 + ‘IWS + 001 dweq jul Ze + IV AS Ob > dweq JON LE рэцшея + + + + + + + + 19'EW + 001 dweq IMA 0€ + + + + à ya 001 > ul м 62 aye7 120.45 зе + + as + a) IV IW ew 001 > ашеа INN 83 + + de + + + "EN 'PO'JO 00! > dweg INN 12 auJOH aye7 + + 5 + + Ja + 00} Ag JON 92 Е + г: + + за +001 Aig JON SzZ 1X9} 89S ‘P10981 о} paqinjsip 00] за + Аа x pa aye] ana 1X98} 88S ‘P10981 о} paqunjsip 001 Е Ag 7 92 + + + + за + 001 Aig JOW гг ях ae al За + 001 ju] yu] Le al + + + + + + a EN IA PO + 001 dweg IMAN 02 Janıy |egdweg 3 + + a + + I + DE JO PO + 001 dweq INN 61 ‘SIA 413 + + in + а +001 ul w Bh a + 2 > за + 001 Aa on Zt JOA + + + + WIM OOl> dueg INN 91 еа9шеэ ‘Ава saizua + + + + + + + eno + 001 dweq HIN SL je + + + + + 4 END + 001 dweq INN pl 1x3] ээ$ 'p10931 о} paqun]sip 00] 4 dweq IMAN £L + + + + E + sE JO EN Ov > jul IMA Zl A as ag за Ov > ju] jul LL + + + + ir + eN PO 001> dweg IMA OL de + 4 + + m + 19'P9 + 001 dweq IMAN 6 + + + + + + = eN PO + 001 ашеа INN 8 + qe + + + MIO Ov > dweq IMA 7 Ха} 89S ‘p10981 O] padiNJSIP OO] 4 dweq IMAN 9 эАаиел ADEMINUO sE “ E + + + IV 00L> yul INN G pue 18Z814 и a as IV 001 > dweq jul y + т + + 10 001> Aa on € ie + 6 ste IV ‘Jd 00l> ju] JOW a + 7 is + + + IV 001 > dweg ju] | spue7] шэшмориз JAN A A A A A | AA ARO E CIA O IEA IN > PULP US SOPAS PO 5a en IV ‚a WH 4S ul эа MS AH 10 es SOURUILWOG a9ueqInisig 2INSIOW 18h17 ‘ON uol2907 US 29U211n990 salad ‘dds xıyes моим IM шпиуодллеа штизэвд ÁlaqamanH pay AH 5плолаеа sngny Auagaıqwiyl UL ısaızuaw e6nsjopnesd ‘114 зебпоа а шпуипш шпуэц$ оч U184 PIOMS MS E13,1U0J0JS ‘D pue jelnu $пило) ‘зроомбоа Ba SISU9YOJ/IS EINIG BONIdS PAYS. MS snpioy Xxeuedo¡do ‘9п|Э s mag eq uojleys виэципео eres ES edıesoyaı snindoy ‘POOMUOOD 19 ESOABU eluoyeyy adels u0baiQ 10 веоа e/ny, ‘лерэЭ pay ulaisam РЭ шпуАчаолви J99y ‘aide pares abe] en ısaızuaWw sninqiy ‘зтпам м ejlÄydossjay ебп$| '490|way илэвэм WHY BIQns SNuUJy ‘лэрм pay IV :saldads juejd jo зашеи эцциеюз pue зиоцелалаа\ ‘обе sieaÁ 001 чец} элош “¡e ye Ji paqunisip = +001 ‘обе sieaÁ 001 pue оф чаэмеа paqunisip = 001 > 'sıeaA Op jse| ul paqunysip = Opy> “aus uado 10 ээцедитер зиэоэл 35046 = , ‘sayis бицое|оэ UO вер |езиэшиоллие Jo зивеа ‘у XIGNAddV 354 CAMERON APPENDIX B. Numbers of snails, and presence or absence of slugs at each site. Sites arranged by litter type. Snails Allogona townsendiana Carychium occidentale Cionella lubrica Columella edentula Discus cronkhitei Euconulus fulvus Haplotrema sportella Haplotrema vancouverense Monadenia fidelis Microphysula cookel Nesovitrea binneyana Planogyra clappi Punctum conspectum Punctum randolphii Pristiloma johnsoni Pristiloma lansingi Pristiloma stearnsi Striatura pugetensis Triodopsis germana Vallonia pulchella Vespericola columbiana Vertigo andrusiana Vertigo columbiana Vertigo rowelli Vitrina alaskana Zonitroides arboreus No. of species No. of individuals Slugs Ariolimax columbianus Arion ater Deroceras laeve Prophysaon andersoni Prophysaon foliolatum Prophysaon vanattae Hemphilla glandulosa Wh 24 10 Mor litter sites Nite w 2209 25126 Sit 38 Intermediate litter sites ES = № 25 120 3 4 4 W223 № GY 7 5155 517222478833 32 + + + + + 31 MA ake ee 2 2 1 1 1 2 2 4 3 5 4 7 1 2 SAM 6) 2 2 3 6 2 Co (CaaS 6 32 21 5 35 25 10 4 1 1 4 2 We 3 4 3 1 SJ aie ala lal MATEO 370033 + + + + + + 32 50 45 11 170 33 22 37 112 ENVIRONMENT AND DIVERSITIES OF SNAIL FAUNAS 355 Recently disturbed and Mull litter sites open sites 5 Us Shao ee O2 AS O 19) 202728) 3000 S34. 135.136 6 13 23 24 2 2 29 ¡OS 37 64 154 MO 18 2 2 6 4 AAA SA 5 2 3 9 9 e Ч 3 A SS 8 7 6 28 16 Ser ei) Gr ait 3 2 4 1 2 4 4 4 3 ¡OZ 6 14 SSA. 9 4 3 2 210 3 ASS Uf 6 3 ay ge 3 5 2 9 5 CN 6 2 = 2 8 м :6 1 3 1 3 CA 2 3 3 2 3 3 Shee 1 1 10 6 2 10 4 36 3 16 ЗО 8 3 4 1 3 9 10 4 UA A: 3 2A 4 4 36 3 16 Syne lO 8 3 4 1 3 9 10 4 7 4 3 2 il 4 4 4 5 1 1 4 A SMS MA RR 185125 19 8 8 1 10’ 16 ONO 1 1 2 2 es hil В NA CSS 40) 55 A STA NA a O Я 1 4 SG 8 5 726 Zoe O A CN CN AS 135 A572 250 15237 A SA A E 187 AE 127125528 1 4 3 5 7 3 3 a9 2 3 2 3 3 7% 6 6 Si Y/ 6 8 3 7 6 2 8 + 9 6 4 16 Dez 9 10 5 3 Dre TAs 532 5 3 6 4 TÍ 4 10 2 GEIST 8 Te 1 4 5 2 1 2 1 1 2 1 11 WG ae ae al lal (AS 14 NS ON O AS и 86 121-52. 226167. 20972167 158 149) 1457 STI a O 167 92.168 154 АЛ 62 > — + - + + + + + + + + + + + > u + “ + + + + + + ar + + + + + + + - + + Revised Ms. accepted 10 April 1986 1 R | fi у i = SE > nn т x ! i ï i 7 MALACOLOGIA, 1986, 27(2): 357-374 MICROGEOGRAPHIC VARIATION IN THE BANDED SPRING SNAIL (HYDROBIIDAE: MEXIPYRGUS) FROM THE CUATRO CIENEGAS BASIN, COAHUILA, MEXICO Robert Hershler Department of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, DC 20560, U.S.A. W.L. Minckley Department of Zoology, Arizona State University, Tempe, Arizona 85287, U.S.A. ABSTRACT Microgeographic variation is documented among 16 populations of the aquatic snail Mexipyrgus churinceanus Taylor from El Mojarral spring system in Cuatro Ciénegas, northern Mexico. Patterns of variation among these snails include a steep cline linking two nominal species, and a gentler one tending to link one of these with a third nominal species from an adjacent spring system. The form of these clines is apparently explained by the presence or absence of gene flow barriers in the spring system, although alternative explanations are discussed. A review of ideas on the origin of aquatic habitats in the basin suitable for Mexipyrgus suggests that the clines represent secondary intergradation, with divergence having occurred in allopatric populations isolated by desiccation of a barrial lake(s) or fragmentation of a massive, continuous spring, and secondary contact made possible by reintegration of habitats. The results, suggesting genetic exchange among distinctive populations having restricted distributions, support an earlier contention that these populations comprise a single, polytypic species. Key words: Hydrobiidae; Cuatro Ciénegas; springs; morphology; variation; clines. INTRODUCTION Aquatic prosobranch snails of the genus Mexipyrgus Taylor, 1966 (Gastropoda: Hydro- biidae), endemic to the desert valley of Cuatro Ciénegas, northeastern México, provide out- standing examples of localized morphological differentiation. Distinctive populations are dis- tributed among spring-fed pools and stream outflows of springs in this small, 1200 km?, endorheic, intermontane basin (Taylor, 1966; Taylor & Minckley, 1966; Hershler, 1984, 1985). Two contrasting opinions have been put forward regarding their taxonomic status: Taylor (1966) described six species from seven collecting sites; and Hershler (1985) reduced the genus to monotypy (Mexipyrgus churinceanus Taylor), based on principal component analysis of data from more than 30 localities. Systematic problems associated with allopatric populations (Mayr, 1963; Wiley, 1981) are thus epitomized by these animals. While spring-fed pools in the basin are insular in nature, some have outflows inter- connecting with other outflows or spring pools, providing possible contact zones be- tween populations. The snail is habitat- specific, occurring only where there are am- ple and invariable sources of water, gentle or negligible currents, warm and constant tem- perature, and substrate consisting of an ac- cumulation of soft, flocculent sediments (Taylor, 1966; Hershler, 1984, 1985). Physi- cally connected places that provide continu- ous habitat of this type are rare. Discovery of a zone of syntopy would be of great value in providing a measure of evolutionary diver- gence among distinctive stocks. One interconnected series of large springs and outflows in an area locally known as El Mojarral contains type localities of two nomi- nal species of Mexipyrgus (sensu Taylor, 1966). Evidence was sought of syntopy, al- lotopy, and/or intergradation involving these as well as a third nominal species from an adjacent spring system. We here describe (357) 358 HERSHLER 8 MINCKLEY microgeographic variation of snails in the Mojarral area, comparing patterns of mor- phology in a relatively continuous habitat with those in one with apparent barriers to gene flow. Opportunity also is taken to report addi- tional hydrographic descriptions and interpre- tations which pertain to origins, age, and evolutionary history of the basin and its unique biota. STUDY AREA, MATERIALS, AND METHODS Description of the area. Average annual rainfall of less than 200 mm (Vivo Escoto, 1964), coupled with high temperatures and evaporation rates, disallows other than storm- induced runoff, and permanent aquatic habi- tats in the Cuatro Ciénegas Basin are fed exclusively by springs (Minckley, 1969). Wa- ter rises near ends of bajadas with consider- able force, its artesian nature indicating an origin at high elevation such as from precipi- tation on the eroded, north-plunging Sierra de San Marcos anticline. Massive Cretaceous limestones of surrounding mountains (Baker, 1971) are characterized by permeable strata and solution channels through which water may pass until forced upward by obstructing faults of basin margins. Minckley (1969) presented a simplistic ex- planation of origin and succession in most present aquatic habitats of the basin, begin- ning with development of solution channels. Such channels presumably foundered as the basin was dewatered to form pits (pozos) that expanded by lateral collapse to form lake- springs or limnocrenes (lagunas). Further slumping of banks produced increased sur- face area and heterogeneity, followed by veg- etation invasion and development of marshlands (cienegas). Surface streams were further proposed to represent foundered subterranean waterways. Downflow, dune- impounded or otherwise formed lentic habi- tats developed into terminal, shallow, variably mineralized barrial (basin floor) lakes. With climatic changes toward greater aridity (Van Devender, 1976, 1977; Wells, 1978; Axelrod, 1979) the successional sequence ends in formation of extensive playas, wet only after periods of rainfall. El Mojarral. A series of springs and marshlands comprising El Mojarral is about 11.5 km SW of the town of Cuatro Ciénegas (Minckley, 1969; Fig. 2). The drainage (Fig. 1) includes three major spring pools: Mojarral West Laguna; “Middle spring;” and Mojarral East Laguna. Extensive ciénegas lie between and adjacent to pools, and a number of smaller springs, streams, and marshes are upslope (south) of the major part of the sys- tem. Drainage of El Mojarral is N and E toward an eastern sump. A surface outflow of Mojarral West enters “Middle spring,” whose surface outflow in turn (and т рай) enters Mojarral East. There also are underwater outflows (black dots in Fig. 1); one from Mojarral West Laguna is via a large, tubular vent in the eastern end. Underwater inflow (open circles in Fig. 1) and outflows in “Middle spring” are similarly large and tube-like, and a number of smaller springs rise from the floor of the western end of Mojarral East. The surface outflow of Mojarral East receives sev- eral small, surface distributaries of the Río Mesquites (arrows in Fig. 1), and ultimately joins that river, the largest in the basin (not included in Fig. 1). Spring inflows to the Mojarral system are thermal at 30 to 35” C and characterized by hard water of crystal clarity. Depths of pools range from less than 1.0 to 4.7 m. They are unshaded except locally by banks; sunlight penetrates to the bottoms. All may have com- mon groundwater sources or are linked by subterranean conduits, given their linear alignment, proximity, and similarities in water quality. Slight temperature decreases downflow may thus represent cooling of groundwater in travel from its source, or may indicate multiple sources. Reductions in EDTA water hardness downflow from Mojar- ral West Laguna (to Mojarral East Laguna) indicated by Minckley & Cole (1968) (1254 vs. 1208 mg/l) were probably not significant. Val- ues in the two springs were actually or essen- tially the same in 1966 and 1968 (1234 mg/l in each and 1236 vs. 1249, respectively; Arnold, 1972; Minckley, unpublished data). Attempts to trace subterranean flow from Mojarral West to “Middle spring” by copious application of fluorescein dye in 1968 and 1970 failed (Arnold, 1972; Minckley, unpublished data). Dye was not detected by eye or through use of activated charcoal collectors examined un- der strong ultraviolet light at various, presum- ably appropriate times following repeated dye application. Discharge volumes are great and dye dilution was either too high or subterra- nean distances and complexities far greater than anticipated. Spring pools have little substrate diversity, VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 359 16 < Mojarral East Laguna 100m 8 A < "Middle spring” 5 3 Mojarral West Laguna > 1 O FIG. 1. Map of El Mojarral drainage showing the location of 16 sampling points. Open and filled circles indicate spring orifices and underwater outflows, respectively. The large arrows to the E indicate inflowing branches of the Río Mesquites. consisting predominantly of flocculent, or- ganic, copropelic sediment less than 1.0 cm to greater than 0.5 m deep that overlies firmer layers of bits of travertine and snail shell. Exposed shell and travertine fragments armor areas where currents remove copropelic ma- terials. Local stands of waterlily (Nymphaea ampla [Salisb.]) vegetate some bottoms, sparse beds of sedges are in shallows, and stony travertine deposits comprise shorelines and bottoms adjacent to inflows. Nominal taxa. Three nominal species of Mexipyrgus are in the Mojarral area, capsule descriptions for which are as follows (modi- fied from Taylor, 1966). 1) Mexipyrgus mojarralis Taylor. Shell (Fig. 3A-F) small, 4.5-5.0 mm high; spiral sculpture well-developed on body whorl; periostracal bands few (two to four) in number, with thickened subsutural band usually HERSHLER 8 MINCKLEY 360 "N ou} шод pamela ‘еипбел 1523 jeueloyy ‘а ‘рие}з Апиэуем элвиаха ay) эои шеб\у ya] JAMO] BU} и! SI мощпо asejns ay] “au judas e Aq penowau /|эбле| uaaq seu puno16810} eu} и! uonejaban uenediy ‘$ ay) шо pamaın Buds ajppiW, ‘9 ‘Buuds eu} Buisso19 pues Аниэуем eu} эюм „“Buuds sjppıy,, JO JEU UISULON ‘а ‘модпо aoejns эц} sajeo1pul мое eu №. зуби ay) о} эап} мощпо 1ajemapun pue рипо/бэло;} ay) ul PUEJS Аплеуем ay] SJON ‘гипбел Isa ¡esefoyy jo uoiuod 3$ “y '$1009 Buds |елеГоу\ 13 "2 ‘914 VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 361 FIG. 3. Photographs of shells of Mexipyrgus churinceanus from El Mojarral. The shells are from localities 2 (A-C), 6 (D-F), 9 (G-l), 12 (J-L), 14 (M-O), and 16 (P-R). Shell “A” is 4.89 mm in height and the other photos are printed to the same enlargement. present; penis with a single (not two 8 as implied in Taylor, 1966) lobe on ES the inner curvature (Fig. 4B). Type PRE Le locality, Mojarral West Laguna. A 2) M. multilineatus Taylor. Shell (Fig. | 3J-O) intermediate in size, 5.0-5.1 Ses mm high; spiral sculpture poorly de- fined on body whorl; up to 20 bands UN x present, with thickened subsutural = 1.0 mm band absent; penis with one or two lobes on inner curvature (Fig. 4). 1 NE: » Type locality, Mojarral East Laguna. FIG 4. Penial variation seen among Mexipyrgus р : Е churinceanus from El Mojarral. Note the variation in >) M. aoe и. т 19. aa A) number of lobes on the inner curvature (A, B) and arge, 7.3 mm high; spiral scu pture occasional presence (C) of an additional lobe on on body whorl poorly developed; the outer curvature. 25-35 bands present, with thick- 362 HERSHLER 8 MINCKLEY TABLE 1. Collection sites for this study, numbered as in Fig. 1, with collection dates and USNM (National Museum of Natural History) catalog num- bers for samples in parentheses. 1. Mojarral West Laguna, 16 т $ of N end, right offshore of E side, 3cm deep (3/11/81) (850287). 2. Mojarral West Laguna, 4 т $ of NW corner, right offshore, 1 m deep (3/19/81) (850285). 3. Stream from Mojarral West Laguna, 20m downstream from spring (12/12/81) (850281). 4. Stream from Mojarral West Laguna, 80m downstream from spring (12/12/81) (850288). 5. Stream from Mojarral West Laguna, 42m above entrance to “Middle spring” (12/12/81) (850276). 6. “Middle spring,” 1 т $ of М corner, 5 т W of E side, 2 m deep (12/14/81) (850277). 7. “Middle spring,” 8 т $ of М tip, 6 mW of E side, in Nymphaea “reef,” 7cm deep (12/14/81) (850286). 8. “Middle spring,” 1 м N of S tip, 3 mE of W side, 1 m deep (12/14/81) (850282). 9. Mojarral East Laguna, western lobe, at inflow from “Middle spring” (12/14/81) (850279). 10. Mojarral East Laguna, western lobe, 13 m NW of connection to eastern lobe, 0.7m deep (12/14/81) (850284). 11. Mojarral East Laguna, eastern lobe, 53 m SE of connection to western lobe, 1 m deep (3/18/81) (850274). 12. Mojarral East Laguna, eastern lobe, 11 т $ of NE corner, 7 m offshore (2/14/81) (850278). 13. Mojarral East Laguna, eastern lobe, 40 m N of SW corner, right offshore (3/17/81) (850273). 14. Stream from Mojarral East Laguna, 30 m down- stream from spring, east side, near inflow from first arm of the Río Mesquites (12/14/81) (850283). 15. Stream from Mojarral East Laguna, at inflow of the second arm of the Rio Mesquites (12/14/81) (850275). 16. Stream from the Mojarral East Laguna, shallow pooled area, N side of stream, just above inflow of third arm of the Rio Mesquites (12/14/81) (850280). ened subsutural band absent; penis with two lobes on inner curvature (Fig. 4A). Type locality, Rio Mes- quites, one to two km upflow from confluence with outflow of Mojarral East Laguna. Specimens examined and methodology. Specimens of Mexipyrgus were collected during March and December 1981 from 16 sampling points along a transect through the El Mojarral system (Fig. 1). Precise locality data are in Table 1. At each locale 100-300 individuals were secured by repeated sweeps of a fine-meshed hand sieve through soft sediments within a randomly chosen area not exceeding 4.0 m?. Snails were relaxed in the field using menthol crystals, fixed in dilute formalin, and preserved in 70% ethanol. A series of adults, recognized by comple- tion and thickening of the inner shell lip, was chosen from each sample; empty shells were excluded. Relaxed snails were readily sexed by noting presence or absence of the penis. The following features, including all those used in diagnoses of nominal species (ex- cepting shell sculpturai pattern), were scored, counted or measured (numbers per sample in parentheses) as follows: presence or ab- sence of banding on the outer shell lip (100); presence or absence of a thickened sub- sutural band (relative to other shell bands) on the outer lip (50 banded shells); number of bands on the outer lip (50 banded shells); shell length (15 females); and number of lobes on inner and outer curvature of the penis, expressed as an “inner-outer” formula (25 males). In addition, the following shell parameters were measured or counted for 15 individuals of each sex from five of the 16 sampling points (2, 8, 12, 14, and 16; Fig. 1): shell height (SH), shell width (SW), length of body whorl (LBW), aperture height (AH), ap- erture width (AW), and number of whorls (WLS). Measurements were made at 25x using a Wild M-5 dissecting microscope fitted with an ocular micrometer. All shell bands distinctive at 25x were counted. Descriptive statistics were generated using the computer-mediated SAS program, while ANOVA and Tukey HSD multiple range test (р = .05) were performed using SYSTAT (Wilkinson, 1984). CLUSTAN (Wishart, 1978) was used to extract principal components from the correlation matrix of shell morphometry data, with separate analyses for males and females. RESULTS Summary statistics are in Tables 2, 4, and 5, with frequencies or mean values of several characters plotted by collecting station in Fig. 5. Results of Tukey HSD Test for multiple comparisons among means of shell heights are in Table 3. Several kinds of variation were evident VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 363 TABLE 2. Summary data for morphologic features scored or measured for samples from 16 localities. The numbers of individuals used are indicated in parentheses. Locality 0OJONADN— Mean adult shell length (15) Frequency of banded shells (n = 100) 83 (n = 113) Mean number of bands on the shell (n = 50) — © © = | TABLE 3. Results of the Тикеу HSD multiple com- parison test among means for shell height. The means (station numbers in parentheses) are ranked by magnitude on the left, and groups of stations containing means that do not differ signif- icantly (p = .05) from one another are indicated to the right. (Fig. 5). Size (SH), although variable and even differing significantly among sampling points within the relatively small Mojarral West (Table 3), showed a general pattern of gradual increase downflow, with overlap es- pecially common among adjacent, upflow sampling points. Note that significant breaks in size occurred at sampling point 9 and others downflow (Table 3). Numbers of shell F f о Frequency of males with the shells with : 2 thickenad following penial formulae subsutural band (We ee! (n’= 50) 1-1 21 Other 78 96 — 4(0-1) (п = 26) 96 96 — 4( (п = 26) 86 100 — = 88 100 — = 94 100 — — 86 100 — — 90 96 — 4(0-1) 60 96 4 — (п = 26) 56 94 3 3(1-2) п — 33) 0 91 3 3(1-2); 3(2-2) (п = 34) 2 85 6 6(0—1); 3(1-2) (п = 33) 0 8 Aliz7/ — (п = 30) 5 (п = 42) 61 36 3(0-1) 2 60 32 4(0-0); 4(1-2) 3 (п = 38) 32 52 4(0-0); 4(2-0); 8(2-2) 1 (п = 35) 16 84 — bands increased downflow in a similar fash- ion. Overlap was again pronounced upflow, with, for instance, no significant differences among any pairs of samples from points 1-8 (Tukey HSD Test, p > 0.05). Variation in subsutural banding involved a steep cline, with high frequencies of thickened banding upflow declining sharply to near-zero frequen- cies downflow (Table 2). Frequency of thick- ened banding for sampling points 8 and 9 (58%, pooled data) differed significantly with either pooled data from upflow points 1-7 (88%) or downflow sampling points 10-16 (3%) (separate Chi-square tests with continu- ity corrections, p < .001 for both compari- sons). Snails from not only Mojarral West (sam- pling (points 1 and 2; Fig. 3A-C), but also its outflow points 3-5) and the two northernmost points in “Middle spring” (points 6, 7; Fig. 3D-F) are clearly referable to Mexipyrgus mojarralis. Small size (SH less than 5.1 тт), few shell bands (fewer than 7), high fre- quency of thickened subsutural banding (greater than 78%), well-developed spiral sculpture on the body whorl, and a “1-1” penial type are characteristic (Table 2). Snails from the next two downflow sampling points (8, 9; Fig. 3G-l) were intermediate between М. mojarralis and М. multilineatus. Non- thickened subsutural banding, rarely seen in upflow M. mojarralis, was common, and a few individuals had the “2-1” penial type not seen 364 HERSHLER 8 MINCKLEY TABLE 4. Shell measurements for females from five localities. For all samples, n = 15. For explanation of abbreviations see p. 362. Parameter Locality SH SW LBW AH AW WLS 2 x 4.80 2.65 3.22 1.79 1.60 6.13 $ 0.167 0.101 0.124 0.087 0.061 0.160 8 X 5.16 2.67 387 1.93 1.63 6.43 $ 0.456 0.221 0.261 0.178 0.102 0.506 12 x 5.97 3.00 3.90 PRA 1.87 6.42 5 0.464 0.143 0.220 0.123 0.101 0.376 14 x 6.38 3.05 3.93 2.19 1.87 6.87 $ 0.346 0.133 0.209 0.125 0.067 0.229 16 x 7.29 3.67 4.64 2.65 2.25 6.95 $ 0.361 0.190 0.230 0.172 0.101 0.254 TABLE 5. Shell measurements for males from five localities. For all samples, п = 15. For explanation of abbreviations see p. 362. Locality SH SW 2 x 4.14 2.19 Ss 0.210 0.094 8 x 4.61 2.42 5 0.207 0.191 12 x 5.67 РТ] 5 0.243 0.061 14 x 5.82 2.78 $ 0.262 0.118 16 x 6.28 3.17 5 0.276 0.092 Parameter LBW AH AW WLS 2.81 1.63 1.41 6.00 0.108 0.057 0.042 0.267 3.10 1.85 1.56 6.17 0.152 0.119 0.106 0.244 3.82 2.19 1.83 6.43 0.098 0.102 0.056 0.200 3.82 235 1.86 6.63 0.156 0.128 0.078 0.160 4.11 2:52 2.09 6.54 0.180 0.128 0.091 0.229 upflow. Widespread character discordance occurred (i.e., individuals with M. mojarralis sculpture, yet M. multilineatus banding). Sep- aration of snails from these points on the basis of subsutural banding type yielded groups (thickened banding, mean SH = 5.35 mm, N = 21; non-thickened, 5.47 and 20 respectively) that did not differ in size (t-test, p > 0.2), further indicating that syntopic taxa were not present. Snails from sampling points 10-13 resem- bled Mexipyrgus multilineatus, but trends among these points and those further downflow involving increasing size, number of bands, and frequencies of “2-1” penial type, indicated apparent gradation toward a more М. lugor-like snail (Fig. 3J-R). Note that snails from point 16 remained transitional in mor- phology in that 100% occurrence of the “2-1” penial type (typical of M. lugoi, fide Taylor, 1966, and Hershler, 1985) was not present. Again, syntopic taxa were not recognizable at any point. Principal components analysis was used to gauge distinctiveness of populations refer- able to the three nominal species (sampling points 2 and 12 as respective topotypes of Mexipyrgus mojarralis and M. multilineatus and snails from sampling point 16 as a pop- ulation trending strongly toward M. lugoi) as well as two populations from intermediate geographic positions (points 8 and 14). Scores for the first two principal components are in Fig. 6 (females) and Fig. 7 (males). Topotypes of Mexipyrgus mojarralis (sam- pling point 2) and M. multilineatus (point 12) are broadly separated, especially males (Fig. 7), while an intermediate state of specimens VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 365 ООО а © р. 772 6.75 5.75 4.75 % with thickened subsutural banding (©) Mean adult shell length (?,mm)(@) 3.75 SAA A A 970217 PA 1a lo) Station FIG. 5. Variation among localities of mean value of shell length, frequency (%) of occurrence of a thickened subsutural band, and penial lobation (circles above top of plot). Variation in penial lobation is expressed as relative frequency of “1-1” (light) versus “2-1” (dark) types. 6.6% FIG. 6. Plot of scores of first two principal components extracted from the female shell data set. from point 8 15 clearly defined. Variability т (males, 5.74; females, 7.51), also 15 elevated the last sample, expressed as mean coeffi- relative to that seen in animals from sampling cient of variation for the six shell parameters points 2 (4.03, 3.74) and 12 (3.31, 5.86). 366 HERSHLER 8 MINCKLEY > 09 ° 6 д. 16 a Po e | bd sa — O 2 E e . > 2 aig la . e , Oe ее, ‚ D & : Г %, | v % о de et 12 + © Oo O 89% FIG. 7. Plot of scores of first two principal components extracted from the male shell data set. While topotypes of Mexipyrgus multiline- atus are less separated from the population trending toward M. lugo! (sampling point 16), neither sex from sampling point 14 occupied a morphologically intermediate position, nor did they exhibit elevated variation relative to the other two (locality 14, males 4.22, females 4.62; locality 16, males 4.10, females 4.96; data for locality 12 given above). Separation is largely along the first principal component, which accounts for 88-89% of total variation and is clearly related to size given high posi- tive coefficients of all shell parameters (Table 6). The major role of size or size-dependent shape also is indicated by significant correla- tions (p < .05) between shell height and the other five parameters (males, 78% of possible comparisons significant; females, 92%). Dis- tinctiveness of the М. /ugoi-like population in this light is largely attributable to the signifi- cant increases in size among downflow pop- ulations (Table 3). DISCUSSION AND CONCLUSIONS We conclude that upstream to downstream genetic exchange among populations in the Mojarral system 1$ evident: 1) we were unable to demonstrate syntopy between forms; 2) there was an absence of absolute barriers that would promote allotopy and Mexipyrgus was almost continuously distributed along a transect of 16 sampling points; and 3) there seems to be ample evidence for clinal intergradation of morphological characters along geographic and habitat gradients. Clinal variation. Clinal variation among interfertile populations is not surprising (Endler, 1977), especially in a sedentary, ovoviviparous animal that lacks broadcast gametes or larvae. However, the two patterns that exist, a steep cline linking Mexipyrgus mojarralis with M. multilineatus and a more gentle one that tends to connect M. multiline- atus and М. lugoi, require further explanation. Steep clines predictably occur in areas having reduced (Endler, 1977) or punctuated gene flow, rather than consistent temporal and/or spatial passage of genetic material. The zone of steepness in the Mojarral sys- tem, at the southern end of “Middle spring” and entrance into Mojarral East (sampling points 8, 9), occurs coincident with a major habitat discontinuity. Soft substrate 1$ essen- tially absent in the lowermost 60 m of surface stream from Mojarral West Laguna as well as from the virtual entirety of stream connecting “Middle spring” and Mojarral East. In both cases water flows swiftly over bare travertine. Only unidirectional and perhaps infrequent gene flow is therefore likely, with snails occa- sionally displaced downstream by currents or drifted with algal mats (at least as young), floating due to accumulation of photosynthetic gasses (Arnold, 1972). Any possible continu- ity of soft bottom type 15 further interrupted within “Middle spring” by steep (near vertical) slope of a travertine or travertine-armored, Nymphaea-covered reef, that largely sepa- VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 367 TABLE 6. Principal components analysis of shell parameters. Females Parameter PCI Shell height 0.428 Shell width 0.420 Length of body whorl 0.395 Aperture height 0.421 Aperture width 0.419 Number of whorls 0.362 rates the spring into two halves (Fig. 8). This may either represent part of a collapsed roof of the spring or an accumulated travertine postdating such an event. Transport may also occur through subsurface channels possibly connecting at least Mojarral West with “Mid- dle spring” and perhaps Mojarral East. Sub- strate and other conditions in underground conduits are unknown. Contrasting lack of evident habitat discon- tinuity provides an apparent explanation for the uniform cline trending from Mexipyrgus multilineatus toward М. lugoi. Although much ofthe western portion of Mojarral East Laguna is floored by travertine, pockets of copropelic sediments are present and bottoms of the eastern two-thirds and the laguna outflow are almost continuously of copropel. The laguna is relatively large, and most currents are from wave action despite a net linear flow from west to east. Such should allow multidirectional ac- tive and passive dispersal, and if gene flow occurs it should produce a gradual (net) pat- tern of influence in the same direction, as was observed in morphology. Although these explanations appear rea- sonable, other alternatives exist that merit some speculation. One viable option relates to changes in selection pressure, which can also effect clinal variation (Endler, 1977). In Cuatro Ciénegas, differential predation pres- sure by а molluscivorous form of the polymorphic fish Cichlasoma minckleyi Kornfield & Taylor! which feeds heavily on hydrobiids (Taylor & Minckley, 1966; Sage & Selander, 1975; Kornfield et al., 1982), has already been proposed as a major evolution- ary force (Vermeij & Covich, 1978). Cichlasoma minckleyi is a visual predator, fanning aside soft substrates to expose and eat Mexipyrgus, and foraging in vegetation Males РСП PCI РСН 0.047 0.425 0.038 —0.308 0.417 —0.242 0.076 0.424 —0.089 —0.231 0.418 — 0.263 — 0.322 0.425 —0.166 0.860 0.331 0.914 and over hard bottoms in search of other molluscan prey (e.g. Mexithauma quadri- paludium Taylor and Nymphophilus minckleyi Taylor). Decreased incidence of periostracal banding on Mexipyrgus in Mojarral East (Ta- ble 2, 15-18%) compared to upstream sam- pling points (77-100%) may indicate that unbanded snails appear more cryptic on or in the light-colored sediments in shallow water. Sediments in Mojarral West are comparably light, but the spring is smaller and deeper, with more shading from banks and travertine ledges that may impart a selective advantage to a more heavily banded shell. A significant increase in shell size at the northern end of Mojarral East Laguna could also reflect an increased selective pressure, with increased size reflecting adaptation affording resistance to a crushing predator (Vermeij 8 Covich, 1978). Predation by fishes could also influence population sizes of snails, which in turn might be reflected in temporal changes such as stunting or other density dependent factors. Possibly density dependent influences on sculpture, periostracal banding, etc., are not apparent. No quantitative data on Mexipyrgus populations other than a maximum density of 49,000 individuals/m? (Hershler, 1985) are available. Differential abundance of the molluscivorous form of cichlid relative to an- other (detritivorous) form, or changes in ab- solute abundance (and thus influence) of the predator due to variations in year class strength, might also be significant factors in such a system. As for snails, no adequately quantitive data exist on population size of the basin's fishes (see, however, Minckley, 1984). Hershler's (1985) comment that tem- poral variations in snail size occur at a given locality was based on casual observation and 'Minckley (1984) perceived the Cuatro Ciénegas cichlids to comprise a flock of distinct species rather than a single, polytypic form; the history of the discovery of, and research on, this fascinating problem was reviewed and further discussed by Williams et al. (1984). 368 HERSHLER 8 MINCKLEY N en A 10m FIG. 8. Map of the “Middle spring.” The location of sampling points 6-8 is shown, as are the positions of the large spring orifice (open circle), underwater outflow tube (closed circle), surface inflow and outflow streams (arrows), and extent of the elevated Nymphaea reef (stippled). The area in between the outline of the spring and inner line continuing around most of the spring circumference 1$ bare travertine. The М edge of the reef, as well as the area just E of the stream inflow (indicated by a dashed line) are elevated above the bottom of the spring. has not yet been tested. Another possibility 15 that sizes of Mexipyrgus in Cuatro Ciénegas relate to habitat conditions. The unusually small shell of M. mojarralis, for example, could reflect warm water, and the larger shell of multilineatus a phenotypic response to cooler or slightly varying water temperatures. Retention of juvenile characters, small size relative to other populations, and various anomalies have been demonstrated in warm- spring fishes (Hubbs, 1959; Miller, 1961; Dea- con & Minckley, 1974; Hubbs et al., 1974). However, there seems little correlation among water temperatures and shell size within the Cuatro Ciénegas basin, e.g. Mexipyrgus from Laguna Escobeda, one of the warmest large springs, are not exception- ally small relative to animals from cooler habitats such as Laguna Tio Candido, and are larger than specimens from slightly cooler Mojarral East (Taylor, 1966; Hershler, 1985, fig. 37, table 48). Further, Nymphophilus minckleyi from three diverse habitats do not differ significantly in size (warm, Mojarral East; cool, Laguna Tio Candido; variable, Rio Mesquites; Hershler, 1985, table 3). Primary or secondary intergradation? An answer to the question of primary vs. second- ary intergradation necessitates reexamination of ideas on origins of aquatic habitats now VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 369 occupied by Mexipyrgus. The following sce- nario was presented for evolution of large laguna systems: “Development of these complex lake-springs begins with sinkhole for- mation. Subsequently, in a actively- flowing aquifer-system, foundering and possible dissolution of the banks produce alinear, tortuous channel. п systems on the barrial, far from the mountain fronts, continuing headwa- ter foundering probably produces elongate channels similar to that oc- cupied by the Río Mesquites” (Minckley, 1969: 18-19). Minckley (unpublished data) is now con- vinced that aggradation may have played as great an alternative role in formation of present-day habitats as do processes of chemical dissolution, foundering, and ero- sion. Travertine deposits, often masked by accumulation of evaporites or overgrowth of halophytes in moist areas, are substantially more extensive throughout parts of the basin fed by mineralized water than before realized. Included are linings for waterways that grow to enclose flowing streams, broad cones that elevate springs above surrounding terrain, and travertine shields downslope from out- flows. These structures are most readily iden- tified in places desiccated by lowering of water table due to canalization (Minckley, 1969, 1978; $. Contreras-Balderas, 1984) or natural processes. Similar fossil to modern deposits have been described for Miocene to Recent springs of the Verde Formation, Arizona (summarized by Donchin, 1983). Cole & Batchelder (1969) documented incipient roof- ing of a spring outflow by travertine. Minckley (1973) described isolation of an Arizona spring by formation of a travertine dam, as illustrated by Hendrickson & Minckley (1985, fig. 20). Travertine deposition may be chem- ical due to changes in pH, physical through release of pressure or evaporation, biogenic through algal activity, or a combination of all three processes (Bathurst, 1975; Hardie et al., 1978). In Cuatro Ciénegas and elsewhere, migra- tion of spring sources and shifts in outflows have obviously occurred through irregular travertine impoundment. As discharge vol- umes vary, so do spatial relations of deposi- tion. Springs break out to flow from bases of travertine tubes, domes, and shields so that source migrations may occur in essentially any direction. Downslope movement of spring sources that might be expected with dis- charge decrease, for example, might be countered by travertine accumulation, so that sources remain near mountain fronts despite lowering of the basin floor due to dewatering (see below). Origin of the molluscan fauna of such a system is problematical. Did the progenitor(s) of Mexipyrgus populations achieve a wide distribution through active, upflow dispersal over hard bottoms? Was an ancestral form widely distributed in an intermontane lake, then left as relicts as inflowing springs were isolated by dropping water level? Does the present spring complex represent remnants from a formerly massive, single outflow, frag- mented by travertine deposition or other fac- tors? Or, did an ancestor(s) arrive and be- come distributed within the area by passive means, such as in mud on the feet of waterbirds? Mexipyrgus is so remarkably restricted to soft sediments that the first query seems readily rejected. Further, other hydrobiids in the basin that commonly occur on hard sur- faces, e.g. Mexithauma quadripaludium and Nymphophilus minckleyi (Taylor, 1966; Hersh- ler, 1984, 1985), show little intraspecific vari- ation attributable to isolation. They presum- ably disperse (or dispersed in the past, assuming interconnection of now isolated habitats) at rates adequate to maintain panmixia. Intermontane lakes were common in north- ern México during wetter periods of the Pleistocene (Miller, 1981). Some filled to top their basin walls or were captured by headward erosion of adjacent streams (Strain, 1966, 1971), while others had maxi- mum levels controlled by climatic factors since they occupy basins that remain closed today. Highest stages of such lakes corre- sponded to an exceedingly wet period 22,500-15,000 years before present (ybp) (Wendorf, 1961). Minckley (1969) reported no evidence in the Cuatro Ciénegas basin for high level lacustrine conditions, e.g. wave-cut terraces or beachlines. He favored presence of lower level lakes on the barrial due to persistent (or periodic) drainage through deep, antecedent channels breaching sur- rounding mountains. Permanent shoreline marks are not necessarily formed in lakes of short duration or when water onlaps fine- grained sediments of bajadas or basin floors. 370 HERSHLER 8 MINCKLEY Perhaps stabilized dunes in the far western and eastern parts of the Cuatro Ciénegas basin (Minckley, 1969) are mute testimony to presence of such a lake(s). Barrial lake con- ditions were directly indicated in cores drilled by Mexican government workers (verbally re- ported to Minckley [unpublished data] as т- tended for petroleum exploration) in the late 1960s. Sediments in three cores drawn from the basin floor 2-3 km southwest of the village of Cuatro Ciénegas were uniformly saturated with water, malodorous, and vari- ably light in color, with only minor interbed- ding of dark, apparently organic material. Soft, light-colored sediments appeared as al- ternating crystalline evaporites and dune sands. Other harder, stony materials were apparent marls, showing varve-like banding as in lacustrine beds deposited below depth of wave action. Thicknesses and depths of various layers below surface could not be ascertained, no conglomerate or travertine was seen, and the drillers did not encounter bedrock at maximum depths of 400-600 т. In contrast, only minor indications of lake sediments appeared in the areas of present springs on or adjacent to bajadas of surround- ing mountains. Four holes drilled by Mexican workers within 1.2 km of Sierra de San Marcos yielded fine-grained (calcareous) spring sediments, organic materials, and travertines, interbedded with angular fan- glomerates typical of bajada surfaces (Minckley, unpublished data). Again, no depths or thicknesses of sediment layers could be determined, but drillers encountered limestone bedrock at 75 to 250 m. In that same area, Meyer (1972, 1973) documented a sequence of springs, ciénegas, travertine deposits, and adjacent grasslands similar to those existing today in sediments and pollens of two cores (6.2 and 13.9 m long). Varve-like banding was noted in short segments of his cores, but no extensive lake deposits were penetrated. The same evidences of little or no change in aquatic and/or terrestrial habitats and an absence of lake beds were in four additional cores 2.7 to 7.9 m long, taken at the same time at nearby places (Minckley, un- published data). Radiocarbon dates near the bottom of Meyer’s (1973) longest core indi- cated an age of ca. 30,000 ybp, providing an estimate of minimum age for springs to have been undisturbed by lacustrine inundation in the Cuatro Ciénegas basin. Water chemistry of an intermontane lake might have been amenable for Mexipyrgus or its ancestor(s) at a time of high inflow volume or external drainage. Present lakes on the floor of the Cuatro Ciénegas basin are almost cer- tainly too saline and fluctuant in chemistry (Minckley & Cole, 1968; Arnold, 1972) to sup- port the taxon. Climatic conditions might also have been more moderate, allowing for at least seasonal dispersal, yet the region has been at a temperate latitude (Dickinson, 1981) and presumably experienced thermal varia- tions for millenia. Soft bottoms would have been available, or springs may have entered the bottom without or with minimal travertine formation due to hypolimnetic conditions. Evidence of hydrobiid snails associated with lacustrine habitats of western North America is not uncommon. A possibility thus exists that the progenitor(s) of Mexipyrgus attained Cuatro Ciénegas springs through intralacustrine dispersal. In one example, however, an abundance of fossil and subfos- sil Tryonia protea (Gould) and lesser numbers of Fontelicella longinqua (Gould) around the Salton Sea, California (Gregg & Taylor, 1965; Taylor, 1981), most probably result from ero- sional reworking of the Mio-Pliocene Bouse Formation (J.J. Landye, unpublished data). Populations of the former persist in only a few thermal springs (Taylor, 1981). Spring- inhabiting Fontelicella spp. (determined by Landye, in Donchin, 1983) were apparently restricted to the vicinity of groundwater in- flows in the Mio-Pleistocene Verde Lake, Ar- izona (Donchin, 1983). Hydrobiids of the gen- era Durangonella Morrison, 1945 and Tryonia Stimpson, 1865 nonetheless inhabit lakes, springs, marshlands in and near the Cuatro Ciénegas basin. The former was thought re- lated to Mexipyrgus by Hershler (1985), an opinion modified by Hershler & Thompson (1986), who demonstrated a nearer relation- ship between Mexipyrgus and Tryonia. The third alternative, a massive spring that now exists as a series of isolated sources, is perhaps more tenable than the presence of major lake(s) and requires less rationalization of habitats. Cretaceous limestones of Texas of the same origins and cavernous qualities of those in northern México (Baker, 1971) pro- vide aquifers for massive springs (Smith, 1971; Brune, 1981), individuals of which now or in the recent past discharge(d) water vol- umes surely equivalent to total output of the Cuatro Ciénegas basin. Further, discharges of Cuatro Ciénegas springs must have been greater in times of more precipitation and less evapotranspiration. Higher volumes would VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 371 tend to move zones of travertine deposition downflow and resist sealing, diversion, and impoundment of outflows, further maintaining thermal and chemical constancy requisite to Mexipyrgus evolution. Fragmentation of such a spring could be piecemeal or systematic, and if the latter could help explain trends in morphology indicated by Hershler (1985: 104) for extant Mexipyrgus stocks. Lastly, passive dispersal of Mexipyrgus or its ancestor(s) to and/or within the Cuatro Ciénegas basin cannot be precluded. Waterbirds are abundantly attracted to aquatic habitats in an otherwise hostile desert (Urban, 1959; A. Contreras-Balderas, 1984), and bird movements are evident from place to place within the basin. Founding of new populations by a few individual snails in this manner, if such were documentable, could explain confusing distributions of some morphological types (Hershler, 1985) in the basin. Transport by movements of waterbirds could further explain occurrence of Mexi- pyrgus and other snails in isolated habitats where no access is obvious. On the other hand, if passive dispersal resulted in sub- stantial gene flow, differentiation т Mexipyrgus should be suppressed or ne- gated. We cannot further assess this mecha- nism as a factor in the origin and evolution of the present fauna. The unlikelihood that Mexipyrgus disperses upflow over hard bottoms, as evidenced by its ecology (Taylor, 1966; Hershler, 1984, 1985), seems to preclude possibilities that inter- gradation in the Mojarral system 1$ primary in nature. Divergence of allopatric populations isolated by desiccation of a barrial lake(s) or progressive fragmentation of a large, contin- uous spring, seem a viable alternative hy- pothesis for the origin of differentiation, with reintegration of habitats resulting in second- ary contact and intergradation. This does not preclude the possibility for passive dispersal as a contributing factor. Systematics. Taylor's (1966: 189) suspi- cion that “divergence of the various popula- tions of Mexipyrgus may have had nothing to do with reproductive isolation, except through geographic separation” is confirmed, at least in the case of the Mojarral area. Our results clearly demonstrate genetic exchange be- tween two nominal species (M. mojarralis and M. multilineatus) in the Mojarral system, and provide strong indications of intergradation between M. multilineatus of Mojarral East and М. lugoi of the Río Mesquites. When one applies the biological species concept, synonymization of nominal Mexipyrgus to a single, polytypic species, М. churinceanus, is supported (Hershler, 1985). The three nominal taxa do, however, con- form to the concept of subspecies in that they present arrays of populations with distinctive features and restricted distributions. We stress that differences among these snails are not mere correlates of a downstream trend toward increased size. Number of shell bands, for instance, does not significantly correlate with shell height at any of the 16 sampling points, although an apparently spu- rious correlation (r = .91) occurs when com- paring means for these characters among points. As mentioned above, subsutural banding does not appear directly related to size, nor does penial lobation, as seen by a lack of size differences among groups of males having “1-1” (mean shell height, 5.71 mm, n = 21) or “2-1” (5.89, 9) penial types (pooled data from sampling points 13 and 14; t-test, р > .05). Similar morphological characters involving shell sizes, sculpture, and banding pattern have apparently been independently derived in separated populations of Mexipyrgus (Hershler, 1985). Or, do characters in com- mon among populations reflect historical pro- cesses and events that we do not yet under- stand? We have not yet devised a way to separate historic vs. derived conditions or states. Even more enigmatic is the question as to why Mexipyrgus shows such differenti- ation while populations of other genera are almost monotypic in the basin. The amount of time required for (or available for) differentia- tion of the kind seen in Cuatro Ciénegas Mexipyrgus also is poorly understood and in debate. Taylor (1966) suggested serious con- sideration that habitats of the area may have existed since the middle or early Tertiary, and pointed out that Mexipyrgus shared charac- ters with Tryonia (late Oligocene or early Miocene to Recent) and Pyrgophorus Ancey, 1888 (early Pliocene to Recent). Hershler (1985) considered the possibility that differen- tiation could have been far more rapid, per- haps within the later Pleistocene. There 1$ little doubt that basin and range topography formed in northern México in early Middle Tertiary (Minckley et al., 1986), and that po- tentials for hydrobiid snail habitat date to that time and before. Documentation that hybridization occurs between distinct phenotypes of Cuatro 372 HERSHLER 8 MINCKLEY Ciénegas Mexipyrgus presents the further possibility of a complex history of differentia- tion involving genetic exchange as means of disseminating distinctive variation among populations. Perhaps characters or character sets assort independently in hybrids where they are fixed by selection or by chance. If repeated separation and reintegration of aquatic habitats has in fact occurred, as im- plied by either the presence of a single, large, travertine-mediated spring outflow frag- mented into subsystems or by presence of numerous, temporally-separated barrial lakes amenable for dispersal of Mexipyrgus, such may be the case. We cannot answer these questions with the current data set or within the framework of the present paper. The “natural laboratory” of Cuatro Ciénegas springs (Taylor & Minckley, 1966) seems to provide an inexhaustible supply of problems to be explored. ACKNOWLEDGEMENTS We thank the Mexican government for pro- viding the necessary permits for collecting freshwater snails in their country. Computing facilities were provided by the University of Florida. Mr. Victor Krantz photographed the shells and Mrs. Molly Ryan (USNM) helped prepare the illustrations. Mr. J. Landye shared with us his ideas concerning snail evolution in Cuatro Ciénegas. Landye as well as Drs. J.A. Endler, M.G. Harasewych and two anony- mous reviewers provided useful criticism of the manuscript. LITERATURE CITED ARNOLD, E.T., 1972, Behavioral ecology of pupfishes (Cyprinodontidae, genus Cyprinodon) from northern Mexico. Unpublished Ph.D. disser- tation, Arizona State University, Tempe, AZ, U.S.A., x + 128 pp. AXELROD, 0.1., 1979, Age and origin of Sonoran Desert vegetation. Occasional Papers of the California Academy of Science, 132: 1-74. BAKER, C.L., 1971, Geologic reconnaissance in the eastern cordillera of Mexico. Special Paper of the Geological Society of America, 131: ix, 1-83, 20 pls. BATHURST, R., 1975, Carbonate sedimentology, carbonate sediments, and their diagenesis. De- velopments in Sedimentology, 12: 1-658. Elsevier, Amsterdam, Netherlands. BRUNE, G., 1981, Springs of Texas, vol. 1. Branch- Smith, Inc., Fort Worth, TX, U.S.A., 584 pp. COLE, С.А. 8 BATCHELDER, G.L., 1969, Dynam- ics of an Arizona travertine-forming stream. Jour- nal of the Arizona Academy of Science, 5: 271-283. CONTRERAS-BALDERAS, A.J., 1984, Birds from Cuatro Cienegas, Coahuila, Mexico. Journal of the Arizona-Nevada Academy of Science, 19: 77-80. CONTRERAS-BALDERAS, S., 1984, Environmen- tal impacts in Cuatro Cienegas, Coahuila, Mex- ico: a commentary. Journal of the Arizona- Nevada Academy of Science, 19: 85-88. DEACON, J.E. & MINCKLEY, W.L., 1974, Desert fishes. In BROWN, G.W., Jr. ed., Desert biology, 11: 385—488. Academic Press, New York, NY, U.S.A. DICKINSON, W.R., 1981, Plate tectonics and the continental margin of California. т ERNST, W.R., ed., The geotectonic development of Cali- fornia, pp. 1-28. Prentice Hall, Inc., Englewood Cliffs, NJ, U.S.A. DONCHIN, J.H., 1983, Stratigraphy and sedimen- tary environments of the Miocene-Pliocene Verde Formation in the southeastern Verde Val- ley, Yavapai County, Arizona. Unpublished M.S. thesis, Northern Arizona University, Flagstaff, AZ, U.S.A., xvi + 182 pp., 2 pls. ENDLER, J.A., 1977, Geographic variation, speciation, and clines. Princeton University Press, Princeton, NJ, U.S.A., ix + 246 pp. GREGG, WO: & TAYLOR м Т965: Fontelicella (Prosobranchia: Hydrobiidae), a new genus of West American freshwater snails. Malacologia, 3: 103-110. HARDIE, L.A., SMOOT, J.P. 4 EUGSTER, H.P., 1978, Saline lakes and their deposits: a sedimentological approach. In MATTER, A. 4 TUCKER, M.E. eds., Modern and ancient lake sediments. International Association of Sedimentology, Spec. Bull. No. 2, pp. 7-41. Blackwell Science Publication, London, England. HENDRICKSON, О.А. 8 MINCKLEY, W.L., 1985, Ciénegas, vanishing climax communities of the American Southwest. Desert Plants (Special Is- sue), 6(“1984”): 131-175, front and back cover pl. HERSHLER, R., 1984, The hydrobiid snails (Gastropoda: Rissoacea) of the Cuatro Ciénegas Basin: systematic relationships and ecology of a unique fauna. Journal of the Arizona-Nevada Academy of Science, 19: 61-76. HERSHLER, R., 1985, Systematic revision of the Hydrobiidae (Gastropoda: Rissoacea) of the Cuatro Cienegas Basin, Coahuila, Mexico. Malacologia, 26: 31-123. HERSHLER, R. & THOMPSON, F.G., 1986, North American Hydrobiidae (Gastropoda: Rissoacea): redescription and systematic relationships of Tryonia Stimpson, 1865, and Pyrgulopsis Call and Pilsbry, 1886. Submitted to Nautilus. HUBBS, C., 1959, High incidence of vertebral de- VARIATION IN A CUATRO CIÉNEGAS HYDROBIID 373 formities in two natural populations of fishes inhabiting warm springs. Ecology, 40: 154—155. HUBBS, C.L., MILLER, R.R. 8 HUBBS, L.C., 1974, Hydrographic history and relict fishes of the north-central Great Basin. Memoirs of the Cali- fornia Academy of Science, 7: 1-259. KORNFIELD, 1., SMITH, D.C., GAGNON, P.S. 8 TAYLOR, J.N., 1982, The cichlid fish of Cuatro Ciénegas, México: direct evidence of con- specificity among distinct trophic morphs. Evolu- tion, 36: 658—664. MAYR, E., 1963, Animal species and evolution. Belknap Press, Harvard University Press, Cambridge, MA, U.S.A., xiv + 797 pp. MEYER, ER. 1972, Late-Quaternary paleo- ecology of the Cuatro Cienegas basin, Coahuila, México. Unpublished Ph.D. dissertation, Arizona State University, Tempe, AZ, U.S.A., x + 74 pp. MEYER, E.R., 1973, Late-Quaternary paleo- ecology of the Cuatro Ciénegas basin, Coahuila, México. Ecology, 54: 982-995. MILLER, R.R., 1961, Speciation rates in some freshwater fishes of western North America. /n BLAIR, W.F., ed., Vertebrate speciation, pp. 537-560. University of Texas Press, Austin, TX, U.S.A. MILLER, R.R., 1981, Coevolution of desert and pupfishes (genus Cyprinodon) in the American Southwest. п NAIMAN, RJ. & SOLTZ D.L., eds., Fishes in North American deserts. Wiley, New York, NY, U.S.A. MINCKLEY, W.L., 1969, Environments of the Bolson of Cuatro Ciénegas, Coahuila, México. University of Texas at El Paso Science Series, 2: 1-65. MINCKLEY, W.L., 1973, Fishes of Arizona. Arizona Game Fish Department, Phoenix, AZ, U.S.A., xvi + 293 pp. MINCKLEY, W.L., 1978, Endemic fishes of the Cuatro Ciénegas basin, northern Coahuila, Méx- ico. In WAUER, R.H. & RISKIND, D.H., eds., Symposium on the biological resources of the Chihuahuan Desert region, United States and Mexico. U.S. National Park Service Transactions and Proceedings Series, No. 3 (1977), U.S. Government Printing Office, Washington, DC, U.S.A. MINCKLEY, W.L., 1984, Cuatro Ciénegas fishes: research review and a local test of diversity versus habitat size. Journal of the Arizona-Nevada Academy of Science, 19: 13-21. MINCKLEY, W.L. 8 COLE, G.A., 1968, Preliminary limnologic information on waters of the Cuatro Ciénegas Basin, Coahuila, México. Southwest- ет Naturalist, 13: 421—431. MINCKLEY, W.L., HENDRICKSON, D.A. 8 BOND, C.E., 1986, Geography of western North Ameri- can freshwater fishes: description and relation- ships to intracontinental tectonism. /n HOCUTT, C.H. 8 WILEY, E.O., eds., Zoogeography of North American freshwater fishes. Wiley, New York, NY, U.S.A. SAGE, В.О. & SELANDER, R.K., 1975, Trophic radiation through polymorphism in cichlid fishes. Proceedings of the National Academy of Sci- ences, 72: 4669-4773. SMITH, A.R., 1971, Cave and karst regions of Texas. т LUNDELIUS, E.L. & SLAUGHTER, B.H., eds., Natural history of Texas caves, pp. 1-14 Gulf Natural History, Dallas, TX, U.S.A. STRAIN, W.S., 1966, Blancan mammalian fauna and Pleistocene formations, Hudspeth County, Texas. Bulletin of the Texas Memorial Museum, 10: 1-55. STRAIN, W.S., 1971, Late Cenozoic bolson inte- gration in the Chihuahua Tectonic Belt. In Sym- posium in honor of Professor Ronald K. Deford. West Texas Geological Society Publication 71-59, рр. 167-173. Midland, TX, U.S.A. TAYLOR, D.W., 1966, A remarkable snail fauna from Coahuila, Mexico. Veliger, 9: 152-228. TAYLOR, D.W., 1981, Freshwater mollusks of Cal- ifornia: a distributional checklist. California Fish and Game, 67: 140-163. TAYLOR, D.W. 8 MINCKLEY, W.L., 1966, New world for biologists. Pacific Discovery, 19: 18-22. URBAN, E.K., 1959, Birds from Coahuila, Mexico. Publications of the Museum of Natural History, University of Kansas, 11: 443-516. VAN DEVENDER, T.R., 1976, The biota of the hot deserts of North America during the last glacia- tion: the packrat midden record. American Qua- ternary Association Abstracts for 1976 meeting, pp. 62-67. VAN DEVENDER, T.R., 1977, Holocene wood- lands in Southwestern deserts. Science, 198: 189-192. VERMEIJ, С... & COVICH, A.P., 1978. Coevolu- tion of freshwater gastropods and their preda- tors. American Naturalist, 112: 833-843. VIVO ESCOTO, J.A., 1964, Weather and climate of Mexico and Central America. In WEST, R.C., ed., Handbook of Middle American Indians, Vol- ume |, Natural Environments and Early Cultures, pp. 187-215. University of Texas Press, Austin, TX, U.S.A. WELLS, P.V., 1978, Post-glacial origin of the present Chihuahuan Desert less than 11,500 years ago. In WAUER, R.H. & RISKIND, D.H., eds., Symposium on the biological resources of the Chihuahuan Desert region, United States and Mexico. U.S. National Park Service Trans- actions and Proceedings Series, No. 3 (1977), U.S. Government Printing Office, Washington, DC, U.S.A. WENDORF, F., 1961, An interpretation of late Pleistocene environments of the Llano Estacado. In Paleoecology of the Llano Estacado, pp. 115-133. Fort Burgwin Research Center, Mu- seum of New Mexico Press, Albuquerque, NM, U.S.A. WILEY, E.O., 1981, Phylogenetics, the theory and practice of phylogenetic systematics. Wiley, New York, NY, U.S.A., xiv + 439 pp. 374 HERSHLER 8 MINCKLEY WILKINSON, L., 1984, SYSTAT, the system for statistics (version 2). Systat, Inc., Evanston, ILL, U.S.A. WILLIAMS, J.E., BOWMAN, D.B., BROOKS, J.E., ECHELLE, AA, EDWARDS, RJ., HEN- DRICKSON, О.А. 8 LANDYE, J.J., 1986, Endangered aquatic ecosystems in North American deserts, with a list of vanishing fishes of the region. Journal of the Arizona-Nevada Academy of Science, 20(“1985”): 1-62, frontis- piece. WISHART, D., 1978, Clustan user manual (third edition). University College London, London, Great Britain, 175 pp. Revised Ms. accepted 9 June 1986 MALACOLOGIA, 1986, 27(2): 375-402 A. MYRA KEEN (1905-1986): A brief biography and malacological evaluation. В. Robertson ........................... 376 Hishommoiluscanhtaxan ENV A COAN PRE RE TR eue ann ee: 383 Malacologicalbibliogaphy EVE CON taa Le De Marae ee 388 Index of specific key words in titles and contents of molluscan papers ................... 398 @menliterature:citedior:consulled sex ne ee La Te E ee 401 BiSWorAOHaxa.namediin:hONOL о A en 402 A. M. KEEN (posthumous) Some important sources for molluscan generic type designations...................... 403 (375) 376 ROBERTSON 8 COAN A. MYRA KEEN (1905-1986): A BRIEF BIOGRAPHY AND MALACOLOGICAL EVALUATION! Robert Robertson Department of Malacology, Academy of Natural Sciences, Nineteenth and the Parkway, Philadelphia, PA 19103, U.S.A. Introduction Professor (Angeline) Myra Keen, who has been called “one of the great giants of Amer- ican malacology” (Abbott, 1986), died at the age of 80 on January 4, 1986, after a long and courageous bout with cancer. During most of her career she considered herself an inverte- brate paleontologist first and a malacologist second. She contributed greatly to both fields. Professor Keen was successful as a teacher, advisor, compiler, researcher, author, scholar, editorial referee, expert on zoological nomen- clature, artist, photographer, curator, public speaker, and creator of exhibits. She had an excellent international reputation among her peers, and because several of her books have strong appeal to shell collectors she was fa- mous among them too. Youth Professor Keen, who in later years preferred to be called A. Myra Keen or simply Myra Keen (hereafter shortened to “Myra”), was born on May 23, 1905, in Colorado Springs, Colorado, U.S.A. She was an only child, and her up- bringing was rural: her parents farmed and raised cattle on a ranch about twenty miles south of Colorado Springs. When only four years old she rode a burro, and when she was thirteen she bought her own horse; she fan- cied being a rodeo queen! lt was on the ranch that the young Myra became interested in the natural sciences. First she showed an interest in birds, then insects?. Myra first went to school in a one-room building housing eight grades. After graduating, her liking for birds took her to Colorado College. Initially she intended to be- come a naturalist, but she ended up majoring in psychology because she did not like dis- secting cats, seeing insects in their death throes in cyanide jars, or the smell of formalin. Myra received her A.B. from Colorado College in 1930. Graduate Schools Myra then moved to California, where she first saw the ocean that was later to become of such interest to her. She went to Stanford University with a scholarship and received her M.A in psychology in 1931. Thence she went to the University of California, at Berkeley, where she was awarded a Ph.D. in psychol- ogy in 1934. Her thesis was on “Children’s reasoning in psychological and physical cau- sation.” She had some biology and geology courses in college, but was largely self-taught in malacology, the field to which she would devote much of her life. The training in statis- tics that she had as an adjunct to psychology was, though, of later use. Early years at Stanford University Myra graduated during the Depression, and could not find a job in psychology. She and her mother moved to the coast at Monterey, where living was inexpensive. Myra made a little money by sewing, and her father kept sending some of the proceeds from a Colorado chicken farm. By chance, Myra one day saw and bought some seashells in a Berkeley curio shop. She became fascinated by them. Shortly 'There are extensive papers and other memorabilia from Dr. Keen in the Smithsonian Institution Archives, Washington, D.C. These were not consulted during preparation of this paper. Any more extensive biography should make use of these archives. 2The quotations in the remainder of this paper either are from references cited on p. 401, or are Myra's sayings recalled by this author. 3Keen, Angeline, 1928, Photographing insects afield. Photo-Era Magazine, 61(2): 76-78 (August). In 1936, she also published a photograph of cliff swallow nests in National Geographic Magazine, 69(4): 522 (April). Her early photos no doubt were published elsewhere too. 4In 1930, when her affiliation was still “Colorado College,” Myra published “Growth curves and IQ's, as determined by testing large families” (School and Society, 32(831): 737-742 [29 Nov.)). A. MYRA KEEN 377 afterwards, she learned that Ida Shepard (Mrs. Tom Shaw) Oldroyd (1856-1940) was work- ing on shells at Stanford University and wanted a helper. (Mrs. Oldroyd is best remem- bered for her books on the marine shells of the Puget Sound area (1924) and of the entire west coast of North America (1924-1927)). Myra hastened to volunteer, and in the same year that she received her Ph.D. (1934) she began working at Stanford University, where she was to remain for the rest of her long and active career. Mrs. Oldroyd, who proved to be no teacher, first had Myra identify and curate land shells. Myra, however, soon came under the influ- ence of Dr. Hubert Gregory Schenck (1897—1960), a Stanford paleontologist whom she found “very stimulating” (Myra used the word “very” sparingly); “from then on | had somebody who could give me the academic guidance that | needed.” Schenck steered her interests into paleontology, but first they worked together on a project quantifying ma- rine faunal provinces on the western coast of North America, using as a basis Recent ma- rine mollusks. Myra was appointed Curator of Paleontology in 1936 (a post created just for her and that she retained until 1957). She had become Mrs. Oldroyd's successor. The year 1937 was a banner one: Myra received her first stipend from Stanford! Many malacologists influenced Myra in her formative years. Among these were the great collectors Emery Perkins Chace (1882-1980) and Elsie Margaret (Herbst) Chace (1885-1975). Junius Henderson (1865-1937) was a visiting scientist (and amateur malacolo- gist) at Stanford in 1934, and taught Myra the use of some of the books in her field. John Quincy Burch (1894-1974), a book and spec- imen shell dealer, was editor of the mimeo- graphed Minutes of the Conchological Club of Southern California—which Myra used as an outlet for some of her early work. Dr. Fred Baker (1854-1938), whom she visited in San Diego in 1936, was another important influ- ence, as was Dr. Paul Bartsch (1871-1960), whom she came to know when she spent a month at the United States National Museum (Smithsonian Institution), Washington, D.C., in 1940. These insights proved useful to Myra as she worked and corresponded with the indi- viduals involved. World War Il Myra Keen and Dr. Eliot Blackwelder, Chairman of the Department of Geology, FIG. 1. Prof. A. Myra Keen. Pacific Grove, Califor- nia. June 20, 1969. Photo by Robert Robertson. were the only ones left to teach geology during one war year. Myra was the first woman to teach in the department. Professorship Belatedly, Myra was appointed Assistant Professor of Paleontology and Research As- sociate т Geology in 1954. In 1955, a formal course on the Mollusca taught by Myra was introduced as рай of the advanced pale- ontology program at Stanford. She became Curator of Malacology in 1957. In 1960, she became Associate Professor with tenure, and in 1965, at age 60, full Professor. Myra was then one of the three women professors in the sciences at Stanford. She also taught popular for-credit courses in advanced paleontology, biological oceanography, and curatorial meth- ods. Students It was Myra who was the main force in train- ing students in malacology at Stanford. She advised more than a dozen advanced degree candidates in geology and biology. Her stu- 378 ROBERTSON 8 COAN FIG. 2. Prof. A. Myra Keen with four of her students. From left to right: Judith T. Smith, Eugene V. Coan, A. Myra Keen, Robert Robertson, and James H. McLean. Pacific Grove, California. June 20, 1969. Photo by Rudolf Stohler. dents who have gone on to publish on mol- lusks are (with their main research interests): Eugene Victor Coan (1943-) [systematics of Recent mollusks, especially bivalves, in the eastern Pacific; zoogeography; nomencla- ture; history of malacology]. Carole Stentz Hickman (1942-) [Tertiary mollusk faunas; systematics; archaeo- gastropods; radulae; functional morphology]. Cortez William Hoskins [Recent Cuban molluscan biofacies; petroleum geology]. James Hamilton McLean (1936-) [system- atics of Recent eastern Pacific marine mol- lusks; Archaeogastropoda; hydrothermal vent limpets; Recent Monoplacophora]. David Nicol (1915-) [bivalves, etc.; evolu- tion; ecology; paleontology; systematics; ma- rine zoogeography]. Robert Robertson (1934-) [systematics and natural history of marine gastropods]. Judith Terry Smith (1940—) [systematics, zoogeography and evolution of Cymatiidae; giant pectens; Cenozoic marine mollusks of the eastern Pacific]. Lee Anderson Smith (no kin of Judith) [paleontology and systematics of Clavagel- lidae; Gulf of Mexico Pleistocene]. Frances Joan Estelle Wagner [Canadian Quaternary marine Mollusca]. Myra always tried to teach her students to think and write clearly, to describe shells well, and to be meticulous in bibliographic and nomenclatural matters. In her course on Curatorial methods in paleontology, Myra would give each student a mixture of heavy and fragile junk shells to pack as if for mailing. She then proceeded to climb on a chair and to hurl the packages one by one onto the floor. Few students passed this test! Field work In 1935, in connection with her zoogeo- graphic work with Schenck, Myra and her parents drove to Neah Bay, northwestern Washington, where she began to collect shells at about every degree of latitude south to northern California. Myra collected in southern California in 1936 (Mission Bay at San Diego), and in the Gulf of California (twice south to Jalisco), Mexico, in 1941, 1956, 1960 (twice), and 1965. On one of these trips she noted “very makeshift conditions” (another “very”!). She was also impressed by the great rise and fall of tides at Puerto Peñasco. A. MYRA KEEN 379 Research interests Myra's research interests ranged widely: from marine molluscan Cenozoic paleontol- ogy, neontology, and zoogeography of west- ern North America, to marine molluscan (es- pecially bivalve) systematics, to zoological nomenclature, to the Recent marine mollusk fauna of tropical West America (the Panamic Province). Above all else, Myra liked un- scrambling difficult (“thorny”) nomenclatural problems, and writing good synonymies. Myra was also taxon-oriented: she was particularly interested in the systematics of the Cardiidae, Vermetidae, Muricidae—espe- cially the subfamily Typhinae, and Berthelinia (a bivalved gastropod). Myra’s first paper (1936a) concerned a new genus of cardiids (Clinocardium). One of her last papers (1980f) reviewed the systematics of the whole family. The Vermetidae were a challenge; Myra’s systematics helped to put them in order (1960h, 1961c, 1982c, d, 1983c; Keen & Hadfield, 1985; Keen & Morton, 1960). Myra’s interest in the Typhinae derived from discovery of a Neogene Typhis in Cali- fornia—a new stratigraphic and geographic record (1944a; Keen & Campbell, 1964). In a small boat, looking through a clump of seaweed that a diver had brought up, Myra discovered living Berthelinia on its algal host Caulerpa in the Gulf of California, Mexico (1960g; Keen & A.G. Smith, 1961). Previ- ously, this bivalved gastropod had been known only from European fossils and live- collected specimens from the western and central Pacific and Australia. Publications Myra was the author of 14 books (including different editions), and 64 malacological pa- pers three or more pages long, and 128 shorter papers and abstracts (see the accom- panying bibliography); unrevised reprints and book reviews are not counted. Myra also published three papers on Foraminifera and one on Brachiopoda. Myra’s first book was An abridged check list and bibliography of [Recent] west North American marine Mollusca (1937g), which supplemented William Healey Dall’s (1845-1927) Summary of the marine shell- bearing mollusks of the northwest coast of America. . . . (1921). Myra attempted to ascertain the latitudinal range end points of each species, information needed for her zoogeographic work with Schenck. An Illustrated [dichotomous] key to west North American [marine] pelecypod genera then received Myra's attention. In 1939, she published this in collaboration with Donald Leslie Frizzell (1906-1972); there was a 1953 revision. A companion booklet on gastropods, co-authored with John C. Pearson, appeared in 1952 and was revised in 1958. The two works were revised and combined in Marine molluscan genera of western North America: an illustrated key (Keen, 1963b). This was further revised т a second edition, co-authored with Eugene V. Coan, published in 1974. Schenck 8 Keen's California fossils for the field geologist was issued in preliminary form in 1940(b) and published by Stanford in 1950. Myra co-authored with Herdis Bentson a Check list of California Tertiary marine Mol- lusca (1944). The beginnings of Myra's greatest book are of interest, told here mainly in her own words. In the middle 1950's, Harry J. Bauer, “a wealthy man in southern California, kept clamoring for a book [on Recent Panamic marine mollusks], and John [Q.] Burch told him that | could do [such] a book. . . . So [Bauer] wrote to me and wanted me to do the book, and | said ‘Nothing doing. I'm not inter- ested in doing books.’ So he kept after me, until finally I said, ‘Well, ГИ get a group of students together as a committee, and ГИ supervise them. That's all | can do. So [Bauer] sent money to start the project ..., and | got my committee together and got them started. One was going to do the cowries, and another was going to do another group. It looked as though we were going fine. Then the committee all fell apart. One man died and another moved away, and there was nobody left but me. We had accepted the money for it, so | was stuck with doing the book. ... The actual writing . . . took about six months. Of course it took a little longer than that to get the information together [beforehand], and a year or so to get the manuscript into shape for the [Stanford] press to publish it.” The result was Sea shells of tropical west America; marine mollusks from Lower California to Colombia (ed. 1, 1958d). The second edition (1971c), subsidized by Dwight Willard Taylor (1932-), was greatly revised and enlarged; the south- ern limit was extended to Peru, and nudibranchs and cephalopods were included; in all, some 3,325 species were treated. This 380 ROBERTSON 8 COAN is Myra's largest and undoubtedly greatest work. It remains the prime reference on east- ern Pacific mollusks, and it has had the beneficial effect of stimulating further re- search. Myra devoted much time and effort to re- vising Cenozoic mollusk families for the en- cyclopedic Treatise on Invertebrate Paleon- tology. Only her archaeogastropod (1960j) and bivalve (1969d) family treatments have been published. It is most regrettable that her work on certain mesogastropod, neogas- tropod, and opisthobranch families remains unpublished (some of this work has been circulated in manuscript form). One additional useful nomenclature paper by Myra is published posthumously here (p. 403). Myra named 3 families, 7 subfamilies, 7 genera, 5 subgenera, 69 species, and 2 subspecies, a total of 93 taxa (see the accom- panying alphabetical list), all of them marine mollusks. Curation Myra spent many years at Stanford Univer- sity adding to, cataloging, and systematically arranging the Cenozoic mollusk collection. She enjoyed “trying to make sense out of the things that had been described and putting them in an orderly fashion.” As curator, Myra “was proud that the ... specimens [are] not only well arranged but also identified. She acquired shells from all over the world in return for identifying duplicate lots.” In 1961, Myra believed that the Stanford mollusk collections “rank[ed] among the half- dozen largest university collections in the [United States].” According to Solem (1975), the Recent collection ranked fourteenth among all museum and university collections in the country (judged by the numbers of cataloged lots). Judith T. Smith (1978) has published on the Primary types т the Stanford paleontological [and neontological] type collection, a very useful compilation and key to Stanford malacological publications. Myra planned, and with amateur help put together, Stanford’s “Conchology Museum” (one room in the Geology Department), one of the nicest such exhibits that | have seen anywhere. Memberships and elected positions Myra belonged to numerous paleontologi- cal and malacological organizations, only a few of which are mentioned here. She was President of the American Malacological Union [AMU] in 1948. Myra presided over the fourteenth national annual meeting, which was held in Pittsburgh, Pennsylvania; there were 41 attendees. Also in 1948, Myra was one of the chief organizers and Acting Chair- man of the Pacific Division of AMU; she was Chairman in 1964. In 1949, Myra was Chair- man of the Pacific Coast Section and Fellow of the Paleontological Society. In 1970, she was President of the Western Society of Malacologists [WSM], the organization re- placing the Pacific Division of AMU. She was an Honorary Life Member of both AMU and WSM. Myra was also Chairman of the Com- mittee on Nomenclature of the Society of Systematic Zoology. Honors and other responsibilities Myra received many awards, and only a few of them are mentioned here. In 1929, as an undergraduate, she was elected to Phi Beta Kappa. Myra was on the Editorial Boards of the Veliger since 1960, and of Malacologia since its beginning in 1962. In 1964, she was awarded a prestigious John Simon Guggenheim Fellowship (see below under European travels). Myra was honored publicly by Emperor Hirohito of Japan when he requested a meet- ing with her during his state visit to the United States in 1975 (Keen, 1985). She talked with him about the molluscan faunas of Japan and northwestern North America, and they ex- changed gifts. In 1979, Myra was the first woman to receive the Fellows Medal from the California Academy of Sciences. In 1984, she received a citation from the Board of Trustees of Col- orado College, her alma mater, in recognition of her personal and scholarly accomplish- ments. European travels Myra spent nearly all her adult life in Cali- fornia. She did, though, visit Europe four times, searching for, studying, and photo- graphing type specimens. In 1958, she worked for three weeks at the British Museum (Natural History) [BMNH], London, after which she wanted to make a return visit. Her wish was fulfilled. During parts of 1964 and 1965, when she was on sabbatical leave from Stanford, her Guggenheim Fellowship en- abled her to visit Europe for the second and third times. She saw some of the marine A. MYRA KEEN 381 laboratories and worked further at BMNH and some other museums. On the continent, she ranged from Copenhagen to Amsterdam. In 1967 she visited BMNH for another three weeks. Myra published four papers on West American mollusk types at BMNH (1966a, c, e, 19680). Professional shortcomings Myra was more а compiler than a re- searcher, and she considered nomenclature to be all important. On occasions, she would even decide the outcome of a taxonomic problem from nomenclatural considerations alone. In distinguishing taxa, she often used single characters, and she named some new species distinguished zoogeographically rather than morphologically (such as cognate taxa on the eastern and western coasts of tropical America). Myra never studied radulae or anatomy, her reasons being that she was a paleontologist with a primary interest in bivalves. In her paper on refereeing manu- scripts (1978b) she hardly mentioned “value as a contribution to science,” i.e. the vital matter of whether a manuscript has sub- stance. Personality, beliefs and quotations Myra was the gentlest, most serene person | have known. She also had a marvelous faculty to put at ease anyone with her, and to bring out the best in a person. Myra was shy, but sure of her convictions. She was a paci- fist, a nonmilitant feminist, and a conserva- tionist. In 1964, she joined the Religious So- ciety of Friends (“Quakers”). Some quotations that may help to reveal more of Myra the person are: Values: “People are more important than shells any day.” To a proud amateur with an uninteresting shell: “My, that is a shell!” On marriage: “| was never particularly ad- verse to the idea of marriage. | expected to marry some day if the right person came along, but | was never out searching. | was too interested in what | was doing.” On excesses: “One should not have an appetite one cannot control which is likely to lead to excesses. One should avoid excesses of any kind.” On the United States space program: “There are many greater needs here on earth.” Classical music “feeds the spirit. Music continues to be one of the joys of my life.” (She particularly liked Brahms. She also liked poetry referring to seashells and to the sea.) То an A+ student (jokingly): “You haven't left any room for improvement.” Last years and legacy Myra became Professor of Paleontology Emeritus and Curator of Malacology Emeritus in 1970, and she continued to teach at Stanford until 1972. Then she went into retire- ment, first in her modest home in nearby Palo Alto, and then in Friends House, a retirement community in Santa Rosa, California, where she spent her last two years. Myra was one of the world's foremost ex- perts on the systematics of Cenozoic marine mollusks, a well-deserved reputation gained during her 38-year career at Stanford. She attracted and welcomed numerous visitors (providing they did not smoke). She also corresponded extensively, keeping in close touch with ex-students and other friends. Myra did all she could to help amateur shell collectors. Last but not least, she made Stanford a malacological center. Among the legacies left by versatile Myra are many publications notable for their no- menclatural precision, a series of malacologi- cal books found very useful by professionals and amateurs alike, and a dozen or more devoted students—some of whom presently are privileged to try following and going be- yond her scholarly footsteps. Myra also left an excellently curated research collection. Unfor- tunately, she could not prevail upon the Stanford administration to have a continuing professorship or curatorship of malacology at Stanford. Sadly, following Myra's retirement, the entire Stanford collection of fossil and Recent mollusks and other groups was sent on permanent loan to the California Academy of Sciences, San Francisco. Myra was greatly admired by her students and the host of associates and colleagues who came to know her gentle and gracious ways. Myra's closest relatives are some lov- ing cousins. Her students and associates are her “children.” Acknowledgments The following colleagues helped greatly by providing information and by criticizing drafts of the manuscript: Dr. Arthur E. Bogan, Dr. 382 ROBERTSON 8 COAN Eugene V. Coan, Jean M. Crabtree, Dr. McLean, Ellen J. Moore, and Dr. Judith Terry George M. Davis, Helen DuShane, Dr. Smith. | alone, though, am responsible for the Kenneth С. Emberton, Jr., Sandra M. selection and interpretation of facts. Gardner, Virginia Orr Maes, Dr. James H. A. MYRA KEEN 383 A. MYRA KEEN (1905-1986): LIST OF MOLLUSCAN TAXA Eugene Coan Research Associate, Department of Invertebrate Zoology, California Academy of Sciences, Golden Gate Park, San Francisco, CA 94118, U.S.A. Full references are given in the Bibliogra- phy beginning on p. 388. Citations to type species of the new generic units, to the genera that are the basis of the new family- level taxa, and to homonyms that she re- named and their preoccupying taxa are not provided.” ABBREVIATIONS: AMNH—American Mu- seum of Natural History [New York] CAS— California Academy of Sciences [San Francisco]; CASPTC—California Academy of Sciences Paleontological Type Collection; LACM—Los Angeles County Museum of Natural History; LSJU—{Leland] Stanford [Junior] University; OD—original designation; SBMNH—Santa Barbara Museum of Natural History; SUPTC-Stanford University Paleon- tological [and neontological] Type Collection; UCMP-—-University of California Museum of Paleontology [Berkeley]. LIST OF TAXA americana, Leptomya—Keen, 1958c: 246, 254255, в! 30.165. 9, 10, pl. 311, figs. 3,5, 6: East side of Punta Alegre, San Miguel Bay, Panama; Robert Van Vleck Anderson, 1913. Holotype—SUPTC 8504 [at CAS]. Re- marks—Synonym of L. ecuadoriana Soot- Ryen, 1957, according to Keen (1971c: 259). amictoideum, Сутайит (Gutturnium)— Keen, 1971c: 505, 506, fig. 954. 27 to 55 m off the northwestern end of San José Island, Panama Bay; R. G. Shaver. Holotype— SUPTC 10043 [at CAS]. anchuela, Mitrella (Mitrella—-Keen, 1943b: 48, 57, pl. 4, fig. 12. About twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7539 [at CAS]. Remarks—See Addicott (1970: 87, pl. 9, figs. 9, 10, 21, 22). anomioides, Plicatula—Keen, 1958c: 241— 242, 255, pl. 31, figs. 4, 7, 8. Guaymas, Sonora, Mexico, on breakwater in front of Miramar Hotel. Holotype—SUPTC 8500 fat CAS]. (Arctopratulum), Nemocardium— Keen, 1954d: 11-14. Type species (OD)—N. (A.) griphus Keen, 1954 [which see]. Aspellinae—Keen, 1971a: 296, a subfamily based on Aspella Mórch, 1877. Remarks— The validity of this subfamily remains in dis- pute (Vokes, 1975: 122-123). Axinopsida—Keen 8 Chavan, in Chavan, 1951: 211, new name for Axinopsis Sars, 1878, non Tate, 1868. bakhanstranum, Epitonium (Nitidiscala)— Keen, 1962f: 179. Salt Works, Carmen Island, Gulf of California. Holotype—CASPTC 4763. Remarks—Synonym of Nitidiscala hindsii (Carpenter, 1857), according to DuShane (1974: 34), who later elevated Nitidiscala to full generic status. belvederica, Berthelinia (Edenttellina) chloris—Keen & Smith, 1961: 51, 53—61, pl. 5, lower fig., text figs. 18, 19, 21-24, 27-32. Puerto Ballandra Bay, about 10 miles north- east of La Paz, Baja California [Sur]; A. G. Smith, 4 October 1960. Holotype—CASPTC 12317. Remarks—Synonym of B. (E.) chloris (Dall, 1918), according to Keen (1971c: 817). Bernardinidae—Keen, 1963b: 91, a new family based on Bernardina Dall, 1910. Re- marks—This family may belong in the Cyamiacea rather than in the Arcticacea, where it was originally placed (Coan, 1984: 228). berryana, Grippina—Keen, 1971c: 269, 270, fig. 693. Bahía Salinas, Isla Carmen, Gulf of California, in 5 to 9 m. Holotype— SUPTC 10040 [at CAS]. “Type localities are given more or less as in the original publications. It is beyond the scope of this paper to correct geological age data, etc. 384 ROBERTSON 8 COAN Бегу! Pitar (РПаг)—Кееп, 1971c: 167, 168, fig. 397. Off La Cruz, Banderas Bay, Jalisco, Mexico, depth 18 to 37 m. Holotype— SUPTC 10038 [at CAS]. birchi, Nucula (Ennucula)—Keen, 1943b: 41, 55, pl. 3, figs. 9-12. About twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7527 [at CAS]. bravoensis, Turbonilla (Pyrgiscus)—Keen, 1943b: 51-52, 57, 58, pl. 4, figs. 20, 26, 27. About twelve miles northeast of Bakersfield, Kern County, California; lower part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype— SUPTC 7546 [at CAS]. Remarks—See Addicott (1970: 146, pl. 21, figs. 34-36). Cabralista—Keen, 1969d: 651, new name for Cabralia Bóhm, 1899, non Moore, 1886. caulerpae, Mitrella—Keen, 1971c: 589, 590, fig. 1232. Puerto Ballandra, about 10 miles northeast of La Paz, [Baja California Sur], in sand among Caulerpa holdfasts; А. С. Smith, 1960. Holotype—CAS 13632 [miss- ing]. Chionista—Keen, 1958c: 242-243. Type species (OD)— Venus fluctifraga Sowerby, 1853: cistula, Lasaea—Keen, 1938а: 22, 24-26, 32, pl. 2, figs. 7-9. Moss Beach, Half Moon Bay, San Mateo County, California. Holo- type—SUPTC 6048 [at CAS]. clarki, Typhis (TyphisopsisM—Keen 4 Campbell, 1964: 48-50, pl. 9, figs. 15, 19, 23. Venado Island, Panama Bay, Walter D. Clark, March 1946. Holotype—SUPTC 9724 [at CAS]. Clinocardium—Keen, 1936a: 119-120. Type species (OD) —Cardium nuttalli Conrad, 1837. coani, Tellina (AngulusH-Keen, 1971c: 211, 212, fig. 512. Candelero Bay, near La Paz, Baja California [Sur]. Holotype—SUPTC 10039 [at CAS]. Remarks—See Gemmell et al. (1983). conchita, Balcis—Keen, 1943b: 43, 57, pl. 4, fig. 5. About twelve miles northeast of Bakersfield, Kern County, California; lower- most part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7538 [at CAS]. Re- marks—See Addicott (1970: 57-58, pl. 20, figs: 1, 32). cultrata, Adrana—Keen, 1958с: 240-241. Seven miles west of Champerico, Guatemala, in 14 fathoms CASPTC 9155. cultrata, Amerycina—Keen, 1971c: 135, 136, fig. 310. Off Isla Partida, Espíritu Santo Island, near La Paz, Baja California [Sur], in 5 to 33 m. Holotype—SUPTC 10037 [at CAS]. decoris, Phyllonotus peratus—Keen, 1960d: 107-108, pl. 10, figs. 4, 5, 7. West Mexican coast near the Guatemalan border; depth about 15 fathoms. Shrimp boat. Holotype—SUPTC 8753 [at CAS]. Re- marks—Synonym of P. peratus, according to Keen (1971c: 517). devexa, Episcynia—Keen, 1946: 9—11, pl. 1, figs. 1-4. Scorpion Harbor, Santa Cruz Island, Santa Barbara County, California, in 2 to 3 fathoms. Holotype—SUPTC 7907 [at CAS]. Distichotyphis—Keen & Campbell, 1964: 56. Type species (OD)—D. vemae Keen & Campbell, 1964 [which see]. durhami, Ferminoscala—Keen, 1943b: 46, 58, pl. 4, fig. 31. About twelve miles northeast of Bakersfield, Kern County, California; lower- most part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7534 [at CAS]. Re- marks—Scalina durhami (Keen), according to Addicott (1970: 56-57, pl. 3, figs. 21, 24). electilis, Moniliopsis—Keen, 1943b: 49, 57, pl. 4, fig. 15. About twelve miles northeast of Bakersfield, Kern County, California; lower- most part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7540 [at CAS]. Re- marks—Ophiodermella electilis (Keen), ac- cording to Addicott (1970: 135). elytrum, Macoma (Psammacoma)—Keen, 1958c: 244, 254, pl. 30, fig. 14. South- southwest of Maldonado Point, [Oaxaca,] Mexico. Holotype—CASPTC 10503. erythrostigma, Siphonochelus (Siphono- chelusH-Keen 8 Campbell, 1964: 51-52, pl. 10, figs. 27, 31, 35. Moreton Bay area off Brisbane, Queensland, Australia by Mr. Wicks. Holotype—SUPTC 9732 [at CAS]. (Eualetes), Tripsycha—Keen, 1971a: 296. (25 meters). Holotype— Type species (OD)—Vermetus centi- quadratus Valenciennes, 1846. fastigata, Nuculana (Saccella)—Keen, 1958c: 240, 255, pl. 31, figs. 1, 2. Off Ballenas Bay, [Puntarenas Prov.,] Gulf of Nicoya, Costa Rica. . . . 35 fathoms (64 meters). Holotype—CASPTC 9149. fayae, Anachis (?Costoanachis)—Keen, 1971c: 578, 579, fig. 1178. Guaymas, A. MYRA KEEN 385 Sonora, Mexico. Holotype—SBMNH 33316 [not SBMNH 12658 as originally published]. fayae, Pterotyphis (Tripterotyphis)—Keen & Campbell, 1964: 54-56, pl. 11, figs. 39, 40, 43, 44, text fig. 1. Barra de Navidad, Jalisco, Mexico; Faye Howard & Gale Sphon. January 7-11, 1962. Holotype—SBMNH 15999. ghanaense, Dendropoma—Keen & Mor- ton, 1960: 44-45, 48-51, pl. 4, figs. 7, 8, text figs. 14-19, 33. Dixcove, 10 to 15 miles west of Takoradi, Gold Coast (i.e. Ghana), West Africa. R. Bassindale, 1953. Holotype— SUPTC 8751 [at CAS] [in a cluster] [not SUPTC 8754 as originally published]. gluma, Volvulella—Keen, 1943b: 54-55, 57, pl. 4, fig. 10. About twelve miles northeast of Bakersfield, Kern County, California; lower- most part of Round Mountain silt, Temblor formation; lower to middle Miocene; Robert T. White; LSJU loc. 2641. Holotype—SUPTC 7550 [at CAS]. Remarks—See Addicott (1970: 141, pl. 20, figs. 3, 4). gnomon, Hastula—Keen, 1943b: 47, 57, pl. 4, fig. 11. About twelve miles northeast of Bakersfield, Kern County, California; lower- most part of Round Mountain silt, Temblor formation; lower to middle Miocene; Donald Birch; LSJU loc. 2121. Holotype—SUPTC 7536 [at CAS]. Remarks—See Addicott (1970: 127, pl. 17, figs. 23-26). griphus, Nemocardium (Arctopratulum)— Keen, 1954d: 12-14, 24, pl. 1 figs. 12, 14, 17, text figs. 3—4. Middle fork of Wishkah River, 14 mi. N. of Aberdeen, Grays Harbor Co., Washington; Astoria formation; middle [to late] Miocene; H. Hannibal, 1912; LSJU loc. NP-243. Holotype—SUPTC 8295 [at CAS]. Halistylinae—Keen, 1958d: 260, new sub- family [not indicated as new] based on Halistylus Dall, 1890. hannibali, Clinocardium—Keen, 19544: 18=19, 21, 24, pl. 1, tig: 16, text fig. 9: Chehalis and Summit Sts., Aberdeen, Washington. ... Montesano formation, upper Miocene-lower Pliocene; Harold Hannibal, 1912; LSJU loc. NP-235. Holotype—SUPTC 8302 [at CAS]. helenae, Nassarina (Cigclirina)—Keen, 1971c: 592, 594-595, fig. 1247. Guaymas, [Sonora, Mexico;] 45 m. Holotype—SUPTC 10047 [at CAS]. Homalopomatinae—Keen, 1960j: 270, new subfamily based on Homalopoma Carpenter, 1864. imperialis, Typhis (Typhina)—-Keen 4 Campbell, 1964: 46-48, pl. 8, figs. 1-4. Off Tosa, Japan in approximately 200 m. Holo- type—Teramachi Collection, Kyoto, Japan. (Indotyphis), Laevityphis—Keen, 1944: 55, 59-63. Type species (OD) —Typhis ban- tamensis Oostingh, 1933. insolida, Similivenus—Keen, 1954a: 218, new name for Venus solida Deshayes, 1825, non V. solida Schroeter, 1802. Acy [Oise], France, Paris Basin Eocene. Holotype— Ecole des Mines, Paris. ischnon, Olivella—Keen, 1943b: 50, 57, pl. 4, figs. 3, 4. About twelve miles northeast of Bakersfield, Kern County, California; lower- most part of Round Mountain silt, Temblor formation; lower to middle Miocene; Robert T. White; LSJU loc. 2641. Holotype—SUPTC 7542 [at CAS]. Remarks—O. (Olivella) ischnon Keen, according to Addicott (1970: 121-122, pl. 17, figs. 9, 13). Isoarcidae—Keen, 1969d: 241, new family based on /soarca Münster, 1842. (Isorobitella), Orobitella—Keen, 1962k: 323-326. Type species (OD)—O. (1) singularis Keen, 1962 [which see]. jayana, Cancellaria (Narona)—Keen, 1958c: 249-250, 254, pl. 30, fig. 5. Panama Bay, about 1 mile off entrance to Panama Canal, depth about 10 fathoms. Walter D. Clark, 1944. Holotype—SUPTC 8502 [at CAS]. Remarks—C. (Bivetia) jayana, accord- ing to Keen (1971c: 651). judithae, Liocerithium—Keen, 1971с: 410— 412, fig. 517. Angel de la Guarda Island, Gulf of California. Holotype—SUPTC 10042 [at CAS]. Laevicardiinae—Keen, 1936b: 367, a new subfamily [not indicated as new] based on Laevicardium Swainson, 1840. lampada, Typhis (TalityphisH—Keen, 1943b: 53-54, 56, pl. 3, figs. 14, 19, 23. About twelve miles northeast of Bakersfield, Kern County, California; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; Donald Birch; LSJU loc. 2121. Holotype—SUPTC 7548 [at CAS]. Re- marks—See also Keen (1939b) and Addicott (1970: 83, pl. 8, fig. 17, pl. 9, figs. 5, 18). lens, Teinostoma (Teinostoma?)—Keen, 1943b: 51, 57, pl. 4, figs. 7-9. About twelve miles northeast of Bakersfield, Kern County, California; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7545 [at CAS]. Remarks—Vitrinella (Vitrin- ellops) lens (Keen), according to Addicott (1970: 46, pl. 2, figs. 4, 5, 10). loismartinae, Cylichna?—Keen, 1943b: 44, 386 ROBERTSON 8 COAN 57, pl. 4, figs. 16, 18. About twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7532 [at CAS]. Remarks—See Addicott (1970: 139- 140; 220.105. 7. 1, 26): ludbrookae, Laevityphis (Laevityphis)— Keen & Campbell, 1964: 52-53, pl. 10, figs. 33, 34, 36, a new name for Typhis tripterus Tate, 1888, non T. tripterus Grateloup, 1833. Adelaide Bore, S[outh] Australia]; clayey green sand, 62 to 63 m level; upper Eocene. Holotype—University of Adelaide, Geology Т 453B. macleani, Decipifus—Keen, 1971c: 586, 588, fig. 1224. Puertecitos, Baja Calif. Norte; J. H. McLean, 1962. Holotype—LACM 1266. mamillatum, Stephopoma—Morton & Keen, 1960: 28-35, pl. 1, figs. 1, 2, text figs. 1, 2, 5-7, 13, 14. Off Gorée, Senegal; 50 m. Holotype—Paris Museum [in a cluster]. marchadi, Dendropoma—Keen & Morton, 1960: 37-41, 46-49, 51, pl. 2, figs. 1-3, text figs. 1-5, 23-25, 33. Goree, Senegal, in lower intertidal zone. Holotype—Paris Museum [in a cluster]. mariposa, Turbonilla (Pyrgolampros)— Keen, 1943b: 52, 57, 58, pl. 4, figs. 19, 25. About twelve miles northeast of Bakersfield, Kern County, California; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7547 [at CAS]. Re- marks—See Addicott (1970: 146-147, pl. 21, figs. 17, 26, 40, 42). medialis, Episcynia—Keen, 1971c: 381, 383, fig. 352. Off Cabo Haro, Guaymas, [Sonora,] Mexico, in 18 m. Holotype—SUPTC 10041 [at CAS]. menuda, Lucinisca—Keen, 1943b: 40—41, 56, pl. 3, figs. 15, 16. About twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Robert T. White. Holotype— SUPTC 7526 [at CAS]. nipponensis, Siphonochelus (Siphonoche- lusH-Keen & Campbell, 1964: 50-52, pl. 10, figs. 25, 29. Off Tosa, Japan, in excess of 200 m. Holotype—Teramachi Collection, Kyoto, Japan. nipponica, Lasaea—Keen, 1938a: 22, 24, 26-28, text fig. 14. Watonoha, Rikuzen, north-east Matsusima, Japan. Holotype— SUPTC 6049 [at CAS]. olssoni, Typhis (Talityphis—Keen, 1943b: 54, new name for Typhis (Talityphis) costaricensis Olsson, 1942, non T. linguiferus costaricensis Olsson, 1922. Quebrada Peñitas, Burica Peninsula, [Puntarenas Prov.,] Costa Rica; Pliocene. Holotype—Pa- leontological Research Institution 4064. oregonensis, Crassinella—Keen, 1938a: 31-32, pl. 2, figs. 11, 12. See also Keen (1939a: 252). South Slough, near highway bridge, Coos Bay, Oregon; 1 to 2 fathoms... M. Keen. Holotype—SUPTC 6052 [at CAS]. Remarks—This was based on a stray valve of C. lunulata (Conrad, 1834) brought to Coos Bay with a shipment of oysters for mariculture (Coan, 1979: 4-5). perata, Nassarina (Cigclirina)—Keen, 1971c: 592, 594, 595, fig. 1248. Puerto Videra, Chiapas, Mexico, 37 to 45 m. Holotype—LACM [1464]. peratus, Phyllonotus—Keen, 1960d: 105— 107, pl. 10, fig. 6. 14 mi. SE of Judas Point, [Puntarenas Prov.,] Costa Rica; depth 42 fathoms; March 1, 1938, mud and shell bot- tom. CAS loc 17974. Holotype—CASPTC 7780. perplexa, Азре!а (Dermomurex)—Keen, 1958c: 248-249, 254, pl. 30, figs. 11-13. Perlas Islands, Panama. Walter D. Clark, 1943. Holotype—SUPTC 8496 [at CAS]. Re- marks—Synonym of A. (D.) indentata (Car- penter, 1857), according to Keen (1971c: 527). personatum, Crucibulum—Keen, 1958c: 247-248, 254, pl. 30, figs. 6-8. Panama, James Zetek. Holotype—SUPTC 8498 [at CAS]. Remarks—C. (Crucibulum) persona- tum Keen, according to Keen (1971c: 462, 463, fig. 824). pomeyroli, Granocardium (Ethmocardi- um)—Keen, 1954d: 8-9, 24, pl. 1, figs. 24, text figs. 1, 2. Coal-bearing beds in area of Moméa tribe, New Caledonia. Upper Creta- ceous. R. Pomeyrol, 1951. Holotype— SUPTC 8287 [at CAS]. praeblandum, Clinocardium—Keen, 1954d: 15-16, 21, 24, pl. 1, figs. 1, 6, text figs. 5, 6. West end of Las Trampas Ridge near Walnut Creek, Contra Costa Co., California. Briones formation; lower upper Miocene. Bruce L. Clark. Holotype—UCMP 14836. precursor, Typhis (Talityphis-Keen 4 Campbell, 1964: 49-50, pl. 9, figs. 14, 18, 21, 22. 6 km west of Puerto Colombia, Depto. Atlántico, Colombia; Las Perdices shales, up- per Oligocene (possibly lower Miocene); Max Steineke; UC loc. S-8012. Holotype—UCMP 15083. A. MYRA KEEN 387 pristinum, Clinocardium—Keen, 1954d: 16-18, 21, 24, pl. 1, figs. 9, 15, text figs. 7, 8. Southwest part of Shell Ridge, near Walnut Creek, Contra Costa Co., California; San Pablo group, probably Neroly formation; up- per Miocene. Holotype—UCMP 14838. Protocardiinae—Keen, 1951а: 7, new subfamily based on Protocardia Beyrich, 1845. Pteropsellinae—Keen, 1969d: 605-606, new subfamily name based on Pteropsella Vokes, 1956 [replaces Pteropsinae Dall, 1894, which was based on a junior hom- опут]. Ptychomyidae—Keen, 19694: 655, new family based on Ptychomya Agassiz, 1842. rotundomontana, Chrysallida—Keen, 1943b: 43-44, 58, pl. 4, fig. 28. About twelve miles northeast of Bakersfield, Kern County, California; lowermost part of Round Mountain silt, Temblor formation; lower to middle Mio- cene; LSJU loc. 2121. Holotype—SUPTC 7531 [at CAS]. Remarks—Odostomia (Besla) rotundomontana (Keen), according to Addicott (1970: 143, pl. 21, figs. 29-31). Samarangiinae—Keen, 1969d: 679, new subfamily based on Samarangia Dall, 1902. scandix, Syrnola—Keen, 1943b: 50-51, 58, pl. 4, figs. 24, 29, 30. Twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7544 [at CAS]. Remarks—Synonym of Pyramidella (Syrnola) ochsneri (Anderson & Martin, 1914), according to Addicott (1970: 142, pl. 2, figs. 4-6). schencki, Laevityphis (Laevityphis)—Keen & Campbell, 1964: 50, 53-54, pl. 9, figs. 16, 20. Puerto Colombia, Dept. Atläntico, Colom- bia; Las Perdices shales; upper Oligocene (possibly lower Miocene); Hubert G. Schenck, ca. 1933. Holotype—SUPTC 9723 [at CAS]. singularis, Orobitella (Isorobitella)—Keen, 1962k: 323-326, figs. 4, 5. Bahía de San Quintín, Baja California Norte. . . , Mexico, mud flats on northeast part of bay; J. L. Barnard and P. T. Beaudette, April, 1961. Holotype—SUPTC 9518 [at CAS]. (Temblornia), Bornia—Keen, 1943b: 38-39, 55, pl. 3, figs. 6, 7. Type species (OD) —Donax triangulata Anderson & Martin, 1914. temblorensis, Cylichna—Keen, 1943b: 44— 45, 57, pl. 4, figs. 13, 14. About twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7533 [at CAS]. Remarks—See Addicott (1970: 140, pl. 20. 105. 10,11, 15, 25). teramachii, Typhis (Typhina)—Keen & Campbell, 1964: 48, pl. 8, figs. 9-11. Off Ки, Japan, in more than 100 т. Holotype— Teramachi Collection, Kyoto, Japan. tholia, Dendropoma—Keen & Morton, 1960: 41-48, 51, pl. 3, figs. 4-6, text figs. 6-13, 20-22, 33. Inhaca Island, off Lorenco Marques, Mozambique . . . in lower balanoid zone. William Macnae, May, 1953. Holotype— SUPTC 8750 [at CAS] [in a cluster] [not SUPTC 8753, as originally published]. Tripsycha—Keen, 1961c: 196. Type spe- cies (OD) —Vermetus tripsycha Pilsbry 4 Lowe, 1932. vemae, Distichotyphis—Keen 8 Campbell, 1964: 54, 56-57; pl. 11, figs. 45—47. Off the Panama-Costa Rica coast, in 1016 fathoms (uncorrected) = 1892 meters depth; Nov. 30, 1958; Vema Station. V-15-60. Holotype— AMNH 110459 Ventricolaria—Keen, 1954a: 218. Type species (OD)— Venus rigida Dillwyn, 1817. verruculastra, Semele—Keen, 1966d: 32-33. Hannibal Bank, Panama, in 64-73 meters; March, 1938; Sta. 224; CAS loc. 17996. Holotype—CAS 036679, ex CASPTC 9256. Remarks—Synonym of S. formosa (Sowerby, 1833), according to Coan (1983). vespera, Nassarina (Nassarina)—Keen, 1971c: 592, 594; fig. 1246. Port Parker, [Bahía Santa Elena, Guanacaste Prov.,] Costa Rica, 27 m. Holotype—CASPTC 13635 [missing]. watsonae, Anachis—Keen, 1943b: 42—43, 57, pl. 4, figs. 1, 2. About twelve miles north- east of Bakersfield, Kern County, California; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; LSJU loc. 2121. Holotype—SUPTC 7530 [at CAS]. Remarks—A. (Costoanachis) wat- sonae Keen, according to Addicott (1970: 86, pl: 9; figs: 11: 12). whitei, Ferminoscala—Keen, 1943b: 46, 58, pl. 4, figs. 32, 33. About twelve miles northeast of Bakersfield, Kern County, Cali- fornia; lowermost part of Round Mountain silt, Temblor formation; lower to middle Miocene; Robert T. White; LSJU loc. 2121. Holotype— SUPTC 7535 [at CAS]. Remarks—Scalina whitei (Keen), according to Addicott (1970: 57, pl. 3, figs. 20, 25-28). 388 ROBERTSON 8 COAN А. MYRA KEEN (1905-1986): MALACOLOGICAL BIBLIOGRAPHY? 2:3 Eugene Coan This Bibliography contains references to works by Dr. A. Myra Keen on the Mollusca and closely related topics, including papers cited in the preceding List of Molluscan Taxa. References to her publications on such other subjects as history, ornithology, photography, psychology, and religion are not included, nor are her published photographs of insects and birds. Her full-scale book reviews are in- cluded, but not short publication notices. When Myra Keen first began publishing photographs and articles about photography in the 1920s, she went under the name Angeline Keen. Later, as she came to prefer her middle name, a transition began—from Angeline М. Keen, through A. Myra Keen, and finally to simply Myra Keen. In addition to the published sections on bivalves and archaeogastropods that she pre- pared for the Treatise on Invertebrate Pale- ontology, Dr. Keen revised several meso- gastropod, neogastropod and opisthobranch groups that were not published. However, many of the manuscripts have been circu- lated among interested workers. Still unpublished is Dr. Keen's treatment of the Vermetidae of the Miocene of the Domin- ican Republic, co-authored with Peter Jung (see Keen, 1982c, abstract). In the following citations, volume, bulletin, monograph, memoir, minutes, and special pa- per numbers are listed in bold face; series numbers (in parentheses) precede volume numbers; issue numbers (in parentheses) fol- low volume numbers; supplementary informa- tion, such as secondary methods of listing vol- umes, part numbers, and parenthetical state- ments, are given in brackets. Plates are listed, but not text figures, maps, charts, or tables. Exact dates of publication are given when possible. The dates of works published by Stanford University Press have been verified through press records. In some cases, these differ from those given by other authors. ICZN-International Commission on Zoologi- al а. Opinions are rulings by | CHAVAN, André [Axinopsida Keen & Chavan in] 1951, Dénominations superspécifiques de mollusques modifiées ou nouvelles. Bulle- tin de la Société Géologique de France, (6) 1 (Compte Rendu Sommaire des Sé- ances, 11-12): 210-212 (July). KEEN, Angeline Myra 1936a, A new pelecypod genus of the family Cardiidae [Clinocardium]. Transactions of the San Diego Society of Natural History, 8(17): 119-120 (12 March). 1936b, Revision of cardiid pelecypods [abst.]. Proceedings of the Geological Society of America, 1935: 367 (June). 1936c, Edgar Allan Poe's conchological text. Nautilus, 50(2): 42-44 (29 Oct.). 1937a, [Review of] “Pyramidellidae from Siogama Bay, northeast Honsyu, Japan,” by S. Nomura. Nautilus, 50(3): 108 (29 Jan.). 1937b, Nomenclatural units of the pelecypod family Cardiidae. Bulletin du Musée royal d’Histoire naturelle de Belgique, 13(7): 22 pp. (March). 1937c, Status of west American molluscan names [abst.]. Proceedings of the Geolog- ical Society of America, 1936: 385 (July, as “June”). 1937d, Percentage method of correlation [abst.]. Proceedings of the Geological So- ciety of America, 1936: 390-391 (July, as “June”). 1937e, Statistical methods applied to paleon- tology [abst.] Proceedings of the Geolog- ical Society of America, 1936: 396 (July, as “June”). 1937f, Abridged check-list of western North American marine Mollusca [abst.]. Pro- ceedings of the Geological Society of America, 1936: 397 (July, as “June”). 193749, An abridged check list and bibliogra- phy of west North American marine Mollusca. Stanford, Calif. (Stanford Uni- versity Press) & London (Oxford Univer- sity Press), 87 pp. (28 Sept.) [Supplement: Keen, 1956b and 1982e]. 1938a, New pelecypod species of the genera Lasaea and Crassinella. Proceedings of the Malacological Society of London, 23(1): 18-32, pl. 2 (16 March) [see Keen (1939а)]. ‘Many untitled extracts of letters from Myra Keen concerning nomenclature were published in the mimeographed Minutes of the Conchological Club of Southern California. Only formally titled papers are listed here. All are well enumerated in Minutes 169: 18-20 (1957) and 199: 7 (1960). ED. 2Муга Keen published numerous notes and comments in the Bulletin of Zoological Nomenclature; only those mentioning mollusks are listed here. ED. 3The bibliographic style is more or less that preferred by Dr. Keen. A. MYRA KEEN 389 1938b, West American Cardiidae [abst.]. Pro- ceedings of the Geological Society of America, 1937: 295 (June). 1939a, New pelecypod species of the genera Lasaea and Crassinella [corrections]. Pro- ceedings of the Malacological Society of London, 23(4): 252 (15 March). 19396, New Typhis from the California Miocene [abst.]. Bulletin of the Geological Society of America, 50(12)[2]: 1972 (1 Dec.). 1940a, The percentage method of stra- tigraphic dating. Proceedings of the Sixth Pacific Science Congress (1939) of the Pacific Science Association, 2: 659-663 (Fall). 1940b, Molluscan species common to west- ern North America and Japan. Proceed- ings of the Sixth Pacific Science Congress (1939) of the Pacific Science Association, 3: 479-483 (Fall). 1942a, А statistical analysis of the percentage of living species of mollusks in the Vaque- ros formation. Pp. 35-37 [59—61], т: Hubert G. SCHENCK & T. CHILDS, “Sig- nificance of Lepidocyclina (Lepidocyclina) californica, new species, in the Vaqueros formation (Tertiary), California.” Stanford University Publications, University Series, Geological Sciences, 3(2): 59 pp. [25-83], 4 pls. (7 July). 1942b, Viability of a marine snail [Nerita scabricosta]. Nautilus 56(1): 34-35 (23 July). 1943a, A report on the Stanford University conchological collection. Minutes of the Conchological Club of Southern Califor- nia, 24: 5-8 (June). 1943b, New mollusks from the Round Moun- tain silt (Temblor) Miocene of California. Transactions of the San Diego Society of Natural History, 10(2): 25-58, pls. 3, 4 (30 Dec.). 1944, Catalogue and revision of the gastropod subfamily Typhinae. Journal of Paleontol- оду, 18(1): 50-72 (28 Jan.). 1945a, Information about the International Zoological Commission [sic]. Minutes of the Conchological Club of Southern Cali- fornia, 46: 5-6 (March). 1945b, List of shells collected in vicinity of Oro Bay, New Guinea, by Lt. Col. Hubert G. Schenck and associates. Minutes of the Conchological Club of Southern Califor- nia, 49: 36-38 (June). 1946, A new gastropod of the genus Episcynia Mórch. Nautilus, 60(1): 8-11, pl. 1, figs. 14 (30 Aug.). 1947a, Purpura and Thais. Minutes of the Conchological Club of Southern Califor- nia, 70: 1-2 (June). 1947b, Exhibit of rare shells received at Stan- ford University during the war [abst]. Amer- ican Malacological Union News Bulletin and Annual Report, 1947: 16-17 (Dec.). 1949a, History of marine malacology on the Pacific coast of North America [abst.]. American Malacological Union News Bul- letin and Annual Report, 1948: 9-10 (March). 1949b, Some techniques for mounting speci- mens in the museum [abstr.]. American Malacological Union News Bulletin and Annual Report, 1948: 17-18 (March). 1949c, The Pacific Division of the American Malacological Union. American Mala- cological Union News Bulletin and Annual Report, 1948: 21 (March). 1949d, Note on Axinopsis. Minutes of the Conchological Club of Southern Califor- nia, 94: 1 (Oct.). 1949e, Phylogeny of the genus Nemocardium [abst.]. American Malacological Union News Bulletin and Annual Report, 1949: 20 (Nov.). 1949f, The Japanese pearl culture industry [abst.] American Malacological Union News Bulletin and Annual Report, 1949: 21 (Nov.) [see also Keen 19639)]. 194949, Notes on west American species of “Vermetidae” [abst]. American Mala- cological Union News Bulletin and Annual Report, 1949: 23 (Nov.). 1950a, Notes on the history of Nemocardium (family Cardiidae). Journal de Conchy- liologie 90[(4)43] (1): 23-29 (15 Jan.). 19506, Species of the cardid genus Clinocardium [abst.]. American Mala- cological Union News Bulletin and Annual Report, 1950: 22 (Dec.). 1951a, Outline of a proposed classification of the pelecypod family Cardiidae. Minutes of the Conchological Club of Southern California, 111: 6-8 (July) [correction in Minutes 112: 12 (Aug.)]. 1951b, Veneridae of western North America. Minutes of the Conchological Club of Southern California, 111: 9 (July) [un- signed]. 1951c, The molluscan names in Renier's “Tavole.” Nautilus, 65(1): 8-15 (27 Aug.) [see also Keen 1954c)]. 1951d, Outline of a proposed classification of the pelecypod family Veneridae. Minutes of the Conchological Club of Southern California, 113: 2-11 [1-10] (Sept.). 1952a, Support for Dr. L.R. Cox’s proposal for the rejection of the “Prodromo” and “Prospetto della classe dei vermi” of Renier as not having been duly “pub- lished” as required by the “Règles.” Z.N.(S.) 432. Bulletin of Zoological No- menclature, 6(10): 312 (29 Aug.). 1952b, West American Tertiary molluscan faunas [abst.]. American Malacological Union Annual Report, 1952: 31 (Dec.). 390 ROBERTSON 8 COAN 1953a, Three regional check lists of interest to paleontologists. Journal of Paleontology, 27(1): 159 (16 Jan.). 1953b, Errata for Keen & Pearson. Minutes of the Conchological Club of Southern Cali- fornia, 126: 1 (Feb.) [see Keen & Pearson (1952)]. 1953c, A progress report on some revisions for the ‘Treatise on Invertebrate Paleontol- ogy [abst.]. American Malacological Union Annual Report, 1953: 8-9 (31 Dec.). 1953d, A critique of Dr. J. Brookes Knight's paper, ‘Primitive fossil gastropods and their bearing on gastropod classification’ [abst.]. American Malacological Union Annual Report, 1953: 24 (31 Dec.). 1954a, Ventricolaria and Similivenus insolida, new names in the pelecypod family Veneridae. Journal of Paleontology, 28(2): 217-218 (29 April) [reprinted in Keen, 19546]. 19546, Nomenclatural notes on the pelecypod family Veneridae. Minutes of the Con- chological Club of Southern California, 139: 50-55 (June) [partly a reprint of Keen, 1954a]. 1954c, Application for a ruling that works cred- ited to S. A. Renier as of the dates 1804 and 1807 were not published within the meaning of Article 25 of the “Regles.” Z.N.(S.) 688. Bulletin of Zoological No- menclature 9(9): 257-262 (22 Oct.) [see Opinions 316, 17 Dec. 1954, and 427, 26 Oct. 1956]. 1954d, Five new species and a new subgenus in the pelecypod family Cardiidae. Bulle- tins of American Paleontology, 35(153): 307-330 [1-24], pl. 29 [pl. 1] (20 Dec.). 1954e, The brackish-water Cardiacea [abst.]. American Malacological Union Annual Re- port, 1954: 26-27 (Dec.). 1955a, Request for the suppression of the generic name “Jumala” Friele, 1882 (class Gastropoda) as a name calculated to give offence on religious grounds. Z.N.(S.) 307. Bulletin of Zoological Nomenclature, 11(2): 61-65 (31 Jan.) [see Opinion 469, 31 May 1957]. 1955b, A few minor pelecypod groups revised for the “Treatise on Invertebrate Paleon- tology” [abst.]. American Malacological Union Annual Reports, 1955 [Bulletin 22]: 28 (31 Dec.). 1955c, Some rare books in the Stanford conchological library [abst.]. American Malacological Union Annual Reports, 1955 [Bulletin 22]: 29 (31 Dec.). 1956a, Comment on a paper by David Nicol [Arctic bivalves]. Nautilus 69(4): 139-140 (10 May). 1956b, Supplement to “An abridged check list . . . ‘; papers on west American marine Mollusca, published during the years 1937 to 1956. 13 pp. Los Angeles Calif. (J. О. Burch) (Aug.) [mimeo- graphed; reprinted: Keen (1982e)]. 1956c, Nomenclatural problems ш the Archaeogastropoda [abst.]. American Malacological Union . . . Annual Report, 1956 [Bulletin 31]: 6-7 (31 Dec.). 1958a, [Comment concerning specific names ending in “-i” and “-й.”]. Bulletin of Zoo- logical Nomenclature, 15(20-21): 684 (18 April). 1958b, Mode of citation to be adopted for citing an author's name subsequent to the original proposal of a zoological name. Z.N.(S.) 1331. Bulletin of Zoological No- menclature 15(23-24): 749—750 (9 May). 1958c, New mollusks from tropical west America. Bulletins of American Paleontol- ogy, 38(172): 235-255, pls. 30, 31 (23 May) [correction sheet supplied Sept. 1958]. 1958d, Sea shells of tropical west America; marine mollusks from Lower California to Columbia, ed. 1, Stanford, Calif. (Stanford University Press), xii + 624 pp.; 10 pls. (5 Dec.) [pp. 624-626, with additional errata, included in March 1959 binding; reprinted in 1960 with errata corrected in text]. 1959a, Cleaning shells with a vibratool [abst.]. American Malacological Union... Annual Reports, 1958 [Bulletin 25]: 34 (1 Jan.). 1959b, The new draft of the International Rules of Zoological Nomenclature [abst.]. American Malacological Union. . . Annual Reports, 1958 [Bulletin 25]: 40 (1 Jan.). 1959c, [Review of] “Manuel de Paléontologie Animale,” by L. Moret. Science, 129(3357): 1217-1218 (1 May). 1959d, Some side notes on “Seashells of Tropical West America” [Anomiidae; Muricidae]. Veliger, 2(1): 1-3, pl. 1 (1 July). 1959e, Some type designations in Anomi- acea. Minutes of the Conchological Club of Southern California, 191: 22 (Sept.). 1960a, [Review of] “Molluscs,” by John E. Morton. Veliger, 2(3): 68—69 (1 Jan.). 1960b, The Molluscan Section, British Mu- seum of Natural History [abst.]. American Malacological Union . . . Annual Reports, 1959 [Bulletin 26]: 36-37 (1 Jan.). 1960c, The London Colloquium on Taxon- omy, and its consequences [abst.]. Amer- ican Malacological Union . . . Annual Re- ports, 1959 [Bulletin 26]: 40—41 (1 Jan.). 1960d, New Phyllonotus from the eastern Pa- cific. Nautilus, 73(3): 103-109, pl. 10 (25 Jan.). 1960e, Some range extensions for Panamic province Mollusca. Minutes of the Conchological Club of Southern Califor- nia, 194: 19-19A (Jan.). A. MYRA KEEN 391 1960f, [Review of] “Pectinidae of the eastern Pacific,” by Gilbert Grau. Veliger, 2(4): 100-101 (1 April). 19609, A bivalve gastropod. Nature, 186 (4722): 406-407 (30 April). 1960h, Vermetid gastropods and marine inter- tidal zonation. Veliger, 3(1): 1-2 (1 July). 1960i, The riddle of the bivalved gastropod. Veliger, 3(1): 28-30 (1 July). 1960j, [Discussions of gastropods with Cenozoic type species]. т J. Brookes KNIGHT, L. В. COX, А. Myra KEEN et al. eds., “Part 1 [Gastropoda], Mollusca 1”: 351 pp., т Raymond С. MOORE, ed., Treatise on Invertebrate Paleontology. Lawrence, Kansas (Geological Society of America & University of Kansas) (about 15 Aug.) [Keen contributions: pp. 226-231 [Fissurellacea]; 233-236 [Patellacea, Cocculinacea]; 246-247, 249-262, 263-267, 268-274 [Trochacea]; 275-289 [Neritacea], some groups in cooperation with other workers; see p. 171 for expla- nation]. 1960k, [Review of] “Seashore life of Japan,” by K. Baba. Veliger, 3(2): 53-54 (1 Oct.). 19601, Pelecypoda fossils. McGraw-Hill Ency- clopedia of Science and Technology, 9: 613-614 (Oct.). 1961a, High-lights of a collecting trip [Gulf of California]. Veliger, 3(3): 79 (1 Jan.). 1961b, [Review of] “Gastropoda Euthyneura,” by A. Zilch. Veliger, 3(3): 88 (1 Jan.). 1961c, A proposed reclassification of the gas- tropod family Vermetidae. British Museum (Natural History), Bulletin (Zoology), 7(3): 181-214 pls. 54, 55 (24 Feb.). 1961d, [Reviews of] “Indo-Pacific Mollusca,” ed. by R. T. Abbott; “Symposium on edible mollusks,” by G. D. Waugh et al. Veliger 3(4): 115-116 (1 April). 1961e, What is Anatina anatina? Veliger, 4(1): 9-12 (1 July). 1961f, [Reviews of] “Colored illustrations of the shells of Japan,” by T. Habe; and “The molluscan shells,” by K. Oyama. Veliger, 4(2): 117-118 (1 Oct.). 19619, А molluscan paradox: the bivalved gastropod. The Shelletter of Shells and Their Neighbors, 1(7): 1-2 (Oct.). 1961h, Comments on the proposal to place the generic name Gari Schumacher, 1817, on the Official List unemended. Z.N.(S.) 1461. Bulletin of Zoological No- menclature, 18(5): 300 (10 Nov.) [see Opinion 910, 5 June 1970]. 1961i, Comments on the proposed use of the Plenary Powers to suppress the generic name Cerastes Laurenti, 1768. Z.N.(S.) 724. Bulletin of Zoological Nomenclature, 18(6): 354 (17 Nov.) [see Opinion 661, 26 April 1963]. 1961], Comment оп the proposed validation of Panopea Ménard de la Groye, 1807. Z.N.(S.) 1049. Bulletin of Zoological No- menclature, 18(6): 376 (17 Nov.) [petition revived by the ICZN Secretary, Bulletin of Zoological Nomenclature, 40(2): 179-183, 1983]. 1961k, What is Anatina апайпа? [abst.]. Amer- ican Malacological Union . . . Annual Re- ports, 1961 [Bulletin 28]: 36 (1 Dec). [see Keen (1961е)]. 19611, Oddities among the Pelecypoda [abst.]. American Malacological Union ... Annual Reports, 1961 [Bulletin 28]: 38-39 (1 Dec.) 1961m, Malacology at Stanford. School of Mineral Sciences Newsletter, Stanford University, December 1961: 1 (Dec.). 1962a, Reinstatement of the specific name Macoma inquinata (Deshayes). Veliger, 4(3): 161 (1 Jan.). 1962b, On the systematic place of Cypraea mus. Veliger, 4(3): 161 (1 Jan.). 1962c, [Review of] “The giant African snail,” by Albert R. Mead. Science, 135(3502): 427 (9 Feb.). 1962d, Shallow-water marine research, Gulf of California. Proceedings of the First Na- tional Coastal and Shallow Water Re- search Conference (Oct. 1961): 668—669 (Feb.). 1962e, Comment on the proposed designa- tion of a type-species for Clathurella Car- penter, 1857. Z.N.(S.) 518. Bulletin of Zoological Nomenclature, 19(2): 99 (23 March) [see Opinion 666, 12 July 1963]. 1962f, Nomenclatural notes on some west American mollusks, with proposal of a new species name [Apolymetis, Psam- motreta, Tresus, Schizothaerus, Epi- toniidae, Oken]. Veliger, 4(4): 178-180 (1 April). 1962g, [Review of] “Fossils: An introduction to prehistoric life,” by William H. Matthew, Ш. Veliger, 5(1): 59 (1 July). 1962h, [Review of] “Sea shells of the world,” by R. T. Abbott. Veliger, 5(1): 61 (1 July). 1962i, Comments on the proposed use of the Plenary Powers to suppress the generic name Pupa Röding, 1798. Z.N.(S.) 581. Bulletin of Zoological Nomenclature, 19(5): 260 (10 Sept.). 1962j, Comments on a paper by R. T. Abbott [terms for type specimens]. Veliger, 5(2): 95 (1 Oct.). 1962k, A new west Mexican subgenus and new species of Montacutidae (Mollusca: Pelecypoda), with a list of Mollusca from Bahia de San Quintin. Pacific Naturalist 3(9): 321-328 (16 Oct.). 19621, Small pelecypods: How to identify them [abst.]. American Malacological Union ... Annual Reports, 1962 [Bulletin 29]: 29 (1 Dec.). 392 ROBERTSON 8 COAN 1962m, Before Linnaeus [abst.]. American Malacological Union . . . Annual Reports, 1962 [Bulletin 29]: 30 (1 Dec.) [see also Keen, 1983a, d)]. 1963a, [Review of] “British prosobranch mol- luscs,” by V. Fretter 8 A. Graham. Sci- ence, 139(3550): 102 (11 Jan.). 1963b, Marine molluscan genera of western North America: An illustrated key. Stanford, Calif. (Stanford University Press) [vi +] 126 pp. (14 Feb.) [“with the assistance of Eugene Coan”]. 1963c, Comments on the proposed validation of Mórch's 1852-1853 catalogue. Bulletin of Zoological Nomenclature, 20(3): 164 (26 April) [see Opinion 714, 26 Nov. 1964]. 1963d, Comment on the name of the type- species of Xenophora. Z.N.(S.) 1483. Bul- letin of Zoological Nomenclature, 20(3): 164 (26 April) [see Opinion 715, 31 Dec. 1964]. 1963e, [Review of] “Fossils: A guide to prehis- toric life,” by F. H. T. Rhodes, H. Zim 4 Р. В. Shaffer. Veliger, 6(1): 54 (1 July). 1963f, Paleontological hoaxes [abst.]. Ameri- can Malacological Union ... Annual Re- ports, 1963 [Bulletin 30]: 36 (1 Dec.) [see Keen (1977с)]. 19634, Japanese pearl culture [abst.]. Ameri- can Malacological Union . . . Annual Re- ports, 1963 [Bulletin 30]: 36-37 (1 Dec.) [see also Keen (1949f)]. 1964a, Conus gloriamaris. Veliger, 6(3): 172 (1 January). 1964b, Species tentatively identified [Neri- topsis radula]. Hawaiian Shell News, 12(7) [ser. 53]: 6 (May). 1964c, A quantitative analysis of molluscan collections from Isla Espíritu Santo, Baja California, Mexico. Proceedings of the California Academy of Sciences, (4)30(9): 175—206 (1 July). 1964d, [Review of] “Fossils in America,” by J. E. Rawson. Veliger, 7(1): 58-59 (1 July). 1964e, Purpura, Ocenebra, and Muricanthus (Gastropoda): Request for clarification of status. Z.N.(S.) 1621. Bulletin of Zoologi- cal Nomenclature, 21(3): 235-239 (7 Aug.) [see Opinion 886, 24 Oct. 1969]. 1964f, Nana Schumacher, 1817 (Gastro- poda): Proposed suppression under the Plenary Powers. Z.N.(S.) 1622. Bulletin of Zoological Nomenclature, 21(4): 303-304 (16 Oct.) [see Opinion 793, 5 June 1970]. 1964g, Some nomenclatural problems [abst.]. American Malacological Union .. . Annual Reports, 1964 [Bulletin 31]: 50. (1 Dec.) [concerning Keen (1964e, f)]. 1964h, Comment on the proposed emenda- tion under the Plenary Powers to Cavolinia of Cavolina Abildgaard, 1791. Z.N.(S.) 1103. Bulletin of Zoological No- menclature, 21(6): 414 (31 Dec.) [see Opinion 883, 12 May 1969]. 1964i, Six misidentified type-species in the superfamily Muricacea (Gastropoda). Z.N.(S.) 1623. Bulletin of Zoological No- menclature, 21(6): 422-428 (31 Dec.) [see Opinion 911, 5 June 1970]. 1965a, [Review of] “Van Nostrand's standard catalog of shells,” by J. L. Wagner & В. T. Abbott. Veliger, 7(4): 255 (1 April). 1965b, [Review of] “Shelling in the Sea of Cortez,” by Paul E. Violette. Veliger, 8(1): 43 (1 July). 1965c, Search for tropical west American mol- luscan types in some European museums [abst.]. American Malacological Union ... Annual Reports, 1965 [Bulletin 32]: 49-50 (1 Dec.). 1965d, Some contrasting seashores, Pacific and Atlantic [abst.]. American Malacologi- cal Union .. . Annual Reports, 1965 [Bul- letin 32]: 53 (1 Dec.). 1966a, West American mollusk types at the British Museum (Natural History); 1. T. A. Conrad and the Nuttall Collection. Veliger, 8(3): 167-172 (1 Jan.). 1966b, [Review of] “A study on the Olividae of the China coast,” by Lou Tze-Kong. Veliger, 8(3): 205 (1 Jan.). 1966c, West American mollusk types in the British Museum (Natural History); Il. Spe- cies described by R. B. Hinds. Veliger, 8(4): 265-275, pls. 46, 47 (1 April) [correc- tions by editor: Veliger, 9(1): 87 (1 July)]. 1966d, Mórch's west Central American mol- luscan types with proposal of a new name for a species of Semele. Occasional Pa- pers of the California Academy of Sci- ences, 59: 33 pp. (30 June). 1966e, West American mollusk types at the British Museum (Natural History); Il. Alcide d'Orbigny's South American collec- tion. Veliger, 9(1): 1-7, pl. 1 (1 Juiy). 1966f, [Reviews of] “Illustrations to ‘Catalogue of the collection of Mazatlan shells’ by Philip P. Carpenter,” by D. C. Brann; “Neogene mollusks from northwestern Ecuador,” by A. A. Olsson; “A survey and illustrated catalogue of the Teredini- dae...,” by R. D. Turner; and “Catalogue of the Paleocene and Eocene Mollusca of the southern and eastern United States,” by K. v. W. Palmer & D. C. Brann. Veliger, 9(1): 88-89 (1 July). 1966g, [Reviews of] ‘British bivalve sea shells,” by N. Tebble; and “Mattheva, a proposed new class of Mollusks,” by E. L. Yochelson. Veliger, 9(2): 253 (1 Oct.). 1966h, Comment on the request for action on the name Voluta mitra Linnaeus, 1758 (Gastropoda). Z.N.(S.) 1728. Bulletin of Zoological Nomenclature, 23(4): 146 (14 A. MYRA KEEN 393 Oct.) [see Keen, 1967b; Opinion 885, 24 Oct. 1969]. 1966i, Tectarius (Mollusca: Gastropoda): Re- quest for validation in its accustomed sense. Z.N.(S.) 1754. Bulletin of Zoologi- cal Nomenclature, 23(4): 179-180 (14 Oct.) [see Opinion 871, 28 Feb. 1969). 1966], Hippella Moerch (Moliusca: Pele- cypoda): Request for suppression under the plenary powers. Z.N.(S.) 1755. Bulle- tin of Zoological Nomenclature, 23(4): 181-182 (14 Oct.) [see Opinion 872, 28 Feb. 1969]. 1966k, Some notes on the molluscan collec- tions at the University of Copenhagen [abst.]. American Malacological Union ... Annual Reports, 1966 [Bulletin 33]: 73 (1 Dec.). 1967a, [Review of] “Shell collecting. An illus- trated history,” by S. P. Dance. Veliger, 9(3): 357 (1 Jan.). 1967b, Comment on the proposed validation of Voluta episcopalis Linnaeus, 1758. Z.N.(S.) 1728. Bulletin of Zoological No- menclature, 24(1): 9 (6 March) [see Opin- ion 885, 24 Oct. 1969]. 1967c, Published in synonymy—published as a synonym. Veliger, 9(4): 444 (1 April). 1967d, [Reviews of] “The molluscan families Speightiidae and Turridae . . . , by A. W. B. Powell; “Shell structure of patelloid and bellerophontoid gastro- pods,” by C. MacClintock; and “How to clean sea shells,” by E. Bergeron. Veliger 10(1): 90 (1 July). 1967e, Support for the proposed addition to the Official List of Biradiolites d'Orbigny, 1850, and Durania Douvillé, 1908. Z.N.(S.) 1765. Bulletin of Zoological No- menclature, 24(4): 209 (20 Sept.) [see Opinion 891, 24 Oct. 1969]. 1968a, [Reviews of] “Van Nostrand's stan- dard catalog of shells” [ed. 2], by J. L. Wagner & В. T. Abbott; and “Chitons and gastropods . . . from the western Pacific Islands,” by H. S. Ladd. Veliger, 10(3): 295 (1 Jan.). 1968b, Rediscovery of a lost species, Columbella procera Sowerby, 1832 [abst.]. American Malacological Union. . . Annual Reports, 1967 [Bulletin 34]: 68 (20 March). 1968c, West American mollusk types at the British Museum (Natural History); IV. Car- penters Mazatlan collection. Veliger, 10(4): 389-439, pls. 55-59 (1 April). 1968d, [Review of] “Marine botany: An intro- duction,” by E. Y. Dawson. Veliger, 11(1): 83 (1 July). 1968e, Cenozoic invertebrate paleontology, western United States. P. 1334, in: Raymond С. MOORE et al., “Develop- ments, trends, and outlooks in paleontol- ” ogy.” Journal of Paleontology, 42(6): 1327-1377 (16 Dec.). 1969a, Problems and pitfalls in searching type specimens [abst.]. Echo (Western Society of Malacologists) 1: 9 (20 March). 1969b, An overlooked subgenus and species from Panama [Alora gouldii]. Veliger, 11(4): 439 (1 April). 1969c, Laternula living on the Pacific coast? Veliger, 11(4): 439 (1 April). 1969d, [Discussions of various families of bivalves]. In: L. В. COX et al., eds., “Part N [Bivalvia], Mollusca 6,” Vols. 1 & 2: xxxviii + 952 pp. т: В. С. MOORE, ed., Treatise on Invertebrate Paleontology. Lawrence, Kansas (Geological Society of America & University of Kansas) (Nov.). [Keen con- tributions: 1: 230-231 [Nuculidae], 241 [Isoarcidae], 269-270 [Philobryidae] 2: 537 [Chlamydoconchacea], 583-639 [Car- diacea, Tridacnacea, Mactracea, Solena- cea, Tellinacea], 643-658 [Dreissen- acea, Arcticacea, Glossacea], 664-702 [Corbiculacea, Veneracea, Myacea, Gastrochaenacea, Hiatellacea], 843-859 [Pandoracea, Poromyacea, Clavagel- lacea], some groups in cooperation with other workers]. 1970a, Memorial to Joseph John Graham (1909-1967). Proceedings of the Geolog- ical Society of America 1967: 211-213 (Jan.). 1970b, New processes in scientific illustration [abst.]. Echo (Western Society of Malacologists), 2: 29 (9 March). 1970c, Future work needed [abst.] [fauna of the Gulf of California]. Echo (Western Society of Malacologists), 2:39 (9 March). 1970d, Comments on the proposed ruling on works on New Zealand Mollusca by R. S. Allan and H. J. Finlay. Z.N.(S.) 1868. Bulletin of Zoological Nomenclature, 26(5/6): 184 (7 April). 1970e, [Review of] “Stratigraphy and paleon- tology of the marine Neogene formations of the Coalinga region, California,” by O. S. Adegoke. Journal of Paleontology, 44(4): 793-794 (23 July). 1970f, Comment on a nomenclatural matter in Mitridae. Veliger, 13(2): 202 (1 Oct.). 1970g, Western marine mollusks. Pp. 51-53, in: A. H. CLARKE, ed., “Proceedings of the American Malacological Union Sym- posium on the Rare and Endangered Mol- lusks of North America.” Malacologia, 10(1): 1-56, 2 pls. (14 Nov.). 1971a, Two new supraspecific taxa in the Gastropoda [Eualetes;, Aspellinae]. Veliger, 13(3): 296 (1 Jan.). 1971b, [Reviews of] “The systematics and bi- ology of abyssal and hadal Bivalvia,” by J. Knudsen; and “Coastal Brazilian sea 394 ROBERTSON 8 COAN shells,” by E. C. Rios. Veliger, 14(1): 135 (1 July). 1971c, Sea shells of tropical west America; marine mollusks from Baja California to Peru, ed. 2. Stanford, Calif. (Stanford Uni- versity Press), xiv + 1064 pp., 22 pls. (1 Sept.) [reprinted in April 1984 with only 12 pls.] [“with the assistance of James Н. McLean”; additions and corrections: Keen 8 Coan, 1975а]. 1971d, Sea shells of tropical West America, a revised edition [abst.]. Echo (Western So- ciety of Malacologists), 4: 23-24 (27 Dec.). 1971e, A review of the Muricacea [abst.]. Echo (Western Society of Malacologists), 4: 35-36 (27 Dec.). 1972a, A taxonomic note [Art. 23-b of the ICZN Code]. Veliger, 14(4): 440—441 (1 April). 1972b, [Review of] “Fossil mollusks of coastal Oregon,” by E. J. Moore. Veliger, 14(4): 445 (1 April). 1972c, [Review of] “The sea shells of Sagami Вау... ,” by T. Kuroda, T. Habe 4 К. Oyama. Veliger, 15(2): 163 (1 Oct.). 1973a, Suggested generic allocations for some Japanese molluscan species [Cardiidae, Vermetidae, Marginellidae]. Tohoku University, Science Reports (2) (Geology), special vol. 6: 1—6 (25 Feb.) [Marginellidae by Eugene V. Coan & Barry Roth]. 1973b, Geologic history of the Pelecypoda (Bivalvia) [abst.]. Echo (Western Society of Malacologists), 5: 31-32 (5 March). 1973c, Some nomenclatural problems in Sacoglossa. Veliger, 16(2): 238 (1 Oct.). 1973d, Comments on the problem of the type species of Lucina (Mollusca: Pelecypoda). Z.N.(S.) 2001. Bulletin of Zoological No- menclature, 30(2): 75-76 (10 Oct.). [see also Keen & Abbott (1972) and Opinion 1095, 1 Nov. 1977]. 1974a, Re Laura Trinchese, 1872 (Gas- tropoda: Opisthobranchia). Veliger, 16(4): 426 (1 April). 1974b, Taxonomic problems in the Saco- glossa [abst.]. Echo (Western Society of Malacologists), 6: 20-23 (3 April). 1974c, Memorial to John Quincy) Burch (1894-1974). Annual Report of the West- ern Society of Malacologists, 7: 8-9 (12 Nov.) [unsigned]. 1974d, Excerpts from and comments on: “Stanford contributions to malacology— an evaluation and appreciation” by Dr. S. Stillman Berry (originally presented to the Stanford meeting of the American Malacological Union, Pacific Division, July 15, 1955). Annual Report of the Western Society of Malacologists, 7: 18-19 (12 Nov.). 1974e, Sidelights on some malacologists, [Tryon, Kossuth, R. v V. Anderson, Willett, Rafinesque]. Annual Report of the West- ern Society of Malacologists, 7: 37—40 (12 Nov.). 1975a, [Review of] “Revision of Matajiro Yokoyama's type Mollusca from the Ter- tiary and Quaternary of the Kanto area,” by K. Oyama. Veliger, 17(3): 322 (1 Jan.). 1975b, On some west American species of Calliostoma. Veliger, 17(4): 413—414 (1 April). 1975c, [Review of] “Molluscan phylogeny: The paleontological viewpoint,” by B. Runnegar 8 J. Pojeta, Jr. Veliger, 17(4): 421 (1 April). 1976a, Pacific outposts of the Tertiary Carib- bean Province [abst.]. Bulletin of the American Malacological Union, 42 [for 1975]: 46 (30 Jan.). 1976b, Another check-list plan? [abst.]. Bulle- tin of the American Malacological Union, 42 [for 1975]: 66 (30 Jan.). 1976c, The letters a curator receives! [abst.]. Annual Report of the Western Society of Malacologists, 9: 48-49 (12 Oct.). 1977a, Comment on “A review of the Eratoidae” by Crawford N. Cate. Veliger, 19(4): 446-448 (1 April). 1977b, [Reviews of] “Bathyal gastropods of the family Turridae ...,” by С. S. Hick- man; “Brazilian marine mollusk iconogra- phy,” by E. C. Rios; and “Murex shells of the world,” by G. E. Radwin 8 A. D'Attilio. Veliger, 19(4): 455—456 (1 April). 1977c, Paleontological hoaxes. Natural His- tory, 86(5): 24, 26, 30 (May). 1977d, A new sea-floor oasis. Veliger, 20(2): 179-180 (1 Oct.). 1977e, Telegraphic style versus normal style. Veliger 20(2): 187 (1 Oct.). 1977f, A deep-water paradox [abst.]. Annual Report of the Western Society of Malacologists, 10: 10 (14 Dec.). 19774, Guidelines for writers and readers: А workshop [abst.]. Annual Report of the Western Society of Malacologists, 10: 11-12 (14 Dec.). 1978a, [Review of] “Shells of New Zealand,” by A. W. B. Powell. Veliger, 20(3): 306 (1 Jan.). 1978b, The role of the editorial referee. Veliger, 20(4): 387-390 (1 April). 1978c, [List of errata for plate legends]. 2 pp. inserted in copies of reprint edition of |. $. Oldroyd, The marine shells of the west coast of North America,” 4 vols. Stanford, Calif. (Stanford University Press) (19 April). 1978d, [Review of] “Marine shells of southern California,” by J. H. McLean. Veliger 21(1): 151 (1 July). 1979a, Phylogeny of the pelecypod family Cardiidae [abst.]. Annual Report of the A. MYRA KEEN 395 Western Society of Malacologists, 11: 11 (9 Jan.). 1979b, [Review of] “A new monoplacophoran limpet from the Continental Shelf off southern California,” by J. H. McLean. Veliger, 22(2): 212 (1 Oct.). 1980a, [Review of] “Bivalve mollusks of the western Beaufort Sea,” by F. R. Bernard. Veliger, 22(3): 208 (1 Jan.). 1980b, Spiroglyphus and Stoa, taxonomic problems in the Vermetidae. Veliger, 22(4): 388-391 (1 April). 1980c, Note on a Panamic province cone. Veliger, 22(4): 404 (1 April). 1980d, [Notice of] Bulletin of the Institute of Malacology, Tokyo. Veliger, 23(1): 107-108 (1 July). 1980e, Memorial to Hubert Gregory Schenck, 1897-1960. Memorials of the Geological Society of America, 10: 5 pp. (July) [preprint: March 1980]. 1980f, The pelecypod family Cardiidae: A tax- onomic summary. Tulane Studies in Ge- ology and Paleontology, 16(1-2): 1-40, 13 pls. (17 Sept.). 19809, Pseudocardia Conrad, 1866, a disre- garded name in Carditidae. Tulane Stud- ies in Geology and Paleontology, 16(1-2): 41-44 (17 Sept.). 1980h, Siphonium, an over-used name in Mollusca. Festivus (San Diego Shell Club), 12(10): 125-126 (Oct.). 1981a, [Review of] “Catalogo dei molluschi conchiferi viventi nel Mediterraneo,” by P. Piani; and “An outline of classification of living shelled marine mollusks,” by K. C. Vaught. Veliger, 23(3): 287-288 (1 Jan.). 1981b, [Review of] “Intertidal invertebrates of California,” by R. H. Morris, D. P. Abbott 8 E. C. Haderlie. Veliger, 23(4): 381 (1 April). 1981c, Siphonium, an over-used name т Mollusca [abst.]. Annual Report of the Western Society of Malacologists, 13: 10 (29 June) [see Keen (1980h)]. 1981d, In memoriam: Emery P. Chace (1882-1980). Annual Report of the West- ет Society of Malacologists, 13[1980]: 19-20 (29 June). 1982a, A footnote to the history of malacology in the United States [American Associa- tion of Conchologists]. Bulletin of the American Malacological Union, 1981: iv-v (Feb.). 1982b, [Review of] “James Graham Cooper— Pioneer western naturalist,’ by E. V. Coan. Veliger, 25(1): 90 (1 July). 1982c, Fossil Vermetidae from the Miocene of the Dominican Republic. Annual Report of the Western Society of Malacologists, 14: 15 (13 July). 1982d, Vermetidae of the Gulf of California, Mexico. Annual Report of the Western Society of Malcologists, 14: 16 (13 July). 1982e, Supplement on “An abridged check- list”: Papers on West American marine Mollusca, published during the years 1937 to 1956. Festivus (San Diego Shell Club), 14(9): 107-116 (9 Sept.) [reprint of Keen (1956b)]. 1983a, On Linnaeus’ bookshelf. Festivus (San Diego Shell Club), 15(1): 5-15 (Jan.). 1983b, Dr. Rudolf Stohler and The Veliger. Hawaiian Shell News, 31(7)[(n.s.)283]: 9-10 (July). 1983c, Vermetidae of the tropical eastern Pa- cific. Annual Report of the Western Soci- ety of Malacologists, 15: 10 (30 Aug.). 1983d, On Linnaeus' bockshelf [abst.]. Annual Report of the Western Society of Malacologists 15: 15 (30 Aug.) [see Keen (1983a)]. 1984, [Review of] “Illustration of the types named by $. Stillman Berry in his 'Leaflets in Malacology,”” by C. M. Hertz. Veliger, 27(2): 244 (5 Oct.). 1985, Myra Keen and the Emperor of Japan. Hawaiian Shell News, 33(1) [(n.s.)301]: 1, 8 (Jan.). 1986, see pp. 403—404 herein. KEEN, Angeline Myra; & Robert Tucker ABBOTT 1972, Problem of the type species of Lucina (Mollusca: Pelecypoda). Z.N.(S.) 2001. Bulletin of Zoological Nomenclature, 29(3): 158-161 (30 Nov.) [see Keen (1973d) and Opinion 1095, 1 Nov. 1977]. KEEN, Angeline Myra; & Herdis BENTSON 1940, Check list of California Tertiary marine Mollusca [abst.]. Bulletin of the Geological Society of America, 51(12)[2]: 1972-1973 (1 Dec.). 1944, Check list of California Tertiary marine Mollusca. Special Papers of the Geologi- cal Society of America, 56: viii + 280 pp. (30 Aug.). KEEN, Angeline Myra; & G. Bruce CAMPBELL 1964, Ten new species of Typhinae (Gas- tropoda: Muricidae). Veliger, 7(1): 46-57, pls. 8-11 (1 July). KEEN, Angeline Myra; 4 Eugene Victor COAN 1963, See Keen (1963b). 1969, Realia Baird, 1850 (Gastropoda): Re- quest for suppression under the Plenary Powers. Z.N.(S.) 1878. Bulletin of Zoolog- ical Nomenclature, 26(2): 99-104 (8 Aug.) [see Opinion 973, 31 Dec. 1971]. 1974, Marine molluscan genera of western North America; an illustrated key, ed. 2. Stanford, Calif. (Stanford University Press) vii + 208 pp. (2 May). 1975a, “Sea shells of tropical West America”: Additions and corrections to 1975. Occa- sional Paper of the Western Society of Malacologists, 1: 66 pp. (22 June). 1975b. Notice concerning Occasional Paper 1 of the Western Society of Malacologists. 396 ROBERTSON 8 COAN Annual Report of the Western Society of Malacologists 8: 5—6 (1 Nov.) [unsigned; more ‘additions and corrections’). KEEN, Angeline Myra; Eugene Victor COAN; & Barry ROTH 1973, See Keen (1973a). KEEN, Angeline Myra; 8 Charlotte L. DOTY 1942, An annotated check list of the gastro- pods of Cape Arago, Oregon. Oregon State Monographs, Studies in Zoology, 3: 16 pp. (15 May). KEEN, Angeline Myra; 8 Donald Leslie FRIZZELL 1939, Illustrated key to West North American pelecypod genera. Stanford, Сай. (Stanford University Press) 28 pp. (20 Feb.) [reprinted in 1946]. 1953, Illustrated key to west North American pelecypod genera, revised ed. Stanford, Calif. (Stanford University Press), 32 pp. (2 Feb.) [superseded by Keen (1963b) and Keen & Coan (1974)]. KEEN, Angeline Myra; 4 Michael G. HADFIELD 1985. Spiroglyphus Daudin, 1800 and Stoa de Serres, 1855 (Mollusca, Gastropoda, Ver- metidae): Proposed suppression of two equivocal generic names. Z.N.(S.) 2340. Bulletin of Zoological Nomenclature, 42(1): 46—49 (2 April) [по Opinion pub- lished yet]. KEEN, Angeline Myra; & James H. MCLEAN 1971, See Keen (19710). KEEN, Angeline Myra; & John Edward MORTON 1960, Some new African species of Dendropoma (Vermetidae: Mesogas- tropoda). Proceedings of the Malacologi- cal Society of London, 34(1): 36-51, pls. 2—4 (April). KEEN, Angeline Myra; 8 Siemon William MULLER 1948, See Schenk 8 McMasters (1948). 1951, Objection to the proposed acceptance of “Gryphaea angulata” Lamarck, 1819, as the type species of the genus “Gryphaea” Lamarck, 1819 (Class Pelecypoda): Comment on proposal sub- mitted by M. Gilbert Ranson. Z.N.(S.) 365. Bulletin of Zoological Nomenclature, 2(11): 332 (28 Sept.) [see Opinion 338, 17 March 1955]. 1952, Proposed use of the Plenary Powers to suppress the generic name “Астеа” Hartmann, 1821, and to validate the ge- neric names “Acmaea” Eschscholtz, 1833, and “Truncatella” Risso, 1826 (class Gastropoda). Z.N.(S.) 27. Bulletin of Zoological Nomenclature, 9(4—5): 130 (30 Dec.). 1953, Comment on the question of the spe- cies to be accepted as the type species of a nominal genus, the name of which was published prior to 1st January, 1931, in the synonymy of a genus. Bulletin of Zoolog- ical Nomenclature, 10(10-11): 342 (24 July). 1956, See Schenk 8 McMasters (1956). 1959, See Schenk 8 McMasters (1959). KEEN, Angeline Myra; 8 John C. PEARSON 1952, Illustrated key to west North American gastropod genera. Stanford, Calif. (Stanford University Press), 39 pp. (19 June) [offset errata sheet issued 11 Feb. 1953; see Keen (1953b); volume reprinted in Oct. 1958] [superseded by Keen (1963b) and Keen 8 Coan (1974)]. KEEN, Angeline Myra; Norman John SILBERLING; 8 Benjamin Markham PAGE 1971, [Memorial to] Siemon William Muller (1900—1970). Bulletin of the American As- sociation of Petroleum Geologists, 55(1): 133-134 (Jan.). KEEN, Angeline Myra; & Allyn Goodwin SMITH 1961, West American species of the bivalved gastropod genus Berthelinia. Proceedings of the California Academy of Sciences, (4)30(2): 47—66, pl. 5 (20 March). KEEN, Angeline Myra; & Thomas Е. THOMPSON 1946, Notes on the Gatun formation (Miocene), Panama Canal Zone [abst.]. Bulletin of the Geological Society of Amer- ica, 57(12)[2]: 1260 (Dec.). MORTON, John Edward; & Angeline Myra KEEN 1960, A new species of Stephopoma (Siliquariidae: Mesogastropoda) from the eastern Atlantic Ocean. Proceedings of the Malacological Society of London, 34(1): 27-35, 1 pl. (April). PAGE, Benjamin Markham; Norman John SILBERLING; & Angeline Myra KEEN 1975, Memorial to Siemon W. Muller, 1900-1970. Memorials of the Geological Society of America, 4: 142-146. SCHENCK, Hubert Gregory; & Angeline Myra KEEN 1936a, Bathymetric distribution of marine Pelecypoda [abst.]. Proceedings of the Geological Society of America, 1935: 367-368 (June). 1936b, West American marine molluscan provinces [abst.]. Proceedings of the Geological Society of America, 1935: 413-414 (June). 1936c, Marine molluscan provinces of west- ern North America. Proceedings of the American Philosophical Society, 76(6): 921-938 (Dec.). 1937, An index-method for comparing mollus- can faunules. Proceedings of the Ameri- can Philosophical Society, 77(2): 161-182 (26 Feb.). 1940a, Biometrical analysis of molluscan as- semblages. Pp. 379-392, 2 pls., in “Con- tribution a l'étude de la répartition actuelle et passée des organismes dans la zone néritique.” Mémoires de la Société de Biogéographie, 7: 436 pp., 9 pls. (25 May). 19406, California fossiis for the field geologist (preliminary edition). Stanford, Сай. A. MYRA KEEN 397 (Stanford University) 86 pp., 56 pls. (25 June) [not by Stanford University Press]. 1941, Renaming primary homonyms after ge- neric reallocation [abst.]. Bulletin of the Geological Society of America, 52(12) [2]: 1983 (1 Dec.). 1942, Renaming primary homonyms after ge- neric reallocation. Journal of Paleontol- ogy, 16(6): 779-780 (9 Nov.). 1950, California fossils for the field geologist, ed. 1. Stanford, Calif. (Stanford University Press), 88 pp., 56 pls. (12 Sept.) [reprinted June 1955]. SCHENK, Edward Theodore; 8 John Herbert McMASTERS 1948, Procedure in Taxonomy, revised ed. Stanford, Calif. (Stanford University Press), vii + 93 pp. (apparently early 1948) [revised by Angeline Myra Keen & Siemon William Muller]. 1956, Procedure in taxonomy, ed. 3. Stanford, Calif. (Stanford University Press), vii + 119 pp. (20 Sept.) [revised by Angeline Myra Keen & Siemon William Muller]. 1959, Procedure in taxonomy, ed. 3. 2nd printing ("with substantial additions”). Stanford, Calif. (Stanford University Press), vii + 149 pp. (5 Nov.) [revised by Angeline Myra Keen & Siemon William Muller]. ACKNOWLEDGMENTS | thank the following persons for providing information, advice, materials, or assistance: Bill Carver, Robert Hollywood, Ellen J. Moore, Robert Robertson, Suzie Ruggles, and Bob and Marie Schutz. 398 ROBERTSON 8 COAN INDEX OF SPECIFIC KEY WORDS IN TITLES AND CONTENTS OF MOLLUSCAN PAPERS Singly-authored papers are indicated by years and letters only; co-authored papers are indicated by “K” and the last name(s) of the other author(s), years, and letters (where these are needed). Book reviews are omitted. Abbott, R.T., 1962] Acmaea, K 8 Muller, 1952 Acmea, K 8 Muller, 1952 Africa, К 8 Morton, 1960 Allan, R.S., 1970d Alora gouldii, 1969b American Association of Conchologists, 1982a American Malacological Union, 1949c Anatina anatina, 1961e, К Anderson, R.v.V., 1974e Anomiacea, Anomiidae, 1959d, e Apolymetis, 1962f Arago, Саре, К & Doty, 1942 Archaeogastropoda, 1956c, 1960] Arctic, 1956a Arcticacea, 1969d Article 23-b, ICZN, 1972a Aspellinae, 1971a Assemblages, molluscan, Schenck 8 K, 1940a Atlantic Ocean, eastern, Morton & К, 1960 Atlantic seashores, 1965d Axinopsida, Chavan & К, 1951 Axinopsis, 1949d Bahía de San Quintin, 1962k Baja California, 1958d, 1971c [see also West American . . . (tropical)] Bathymetry, Schenck & К, 1936a Berry, S.S., 1984 Berthelinia, К & Smith, 1961 [see also bivalved gastropods] Biometry, Schenck & К, 1940a Biradiolites, 1967e Bivalved gastropods, 1960g, i, 1961g, К & Smith, 1961 Bivalvia, 1956a, 1969d, Schenck & K, 1936b Bivalvia, fossil, 19601, 1973b Bivalvia, how to identify small, 19621 Bivalvia, oddities among, 19611 British Museum (Natural History), Mollusca Sec- tion, 1960b, 1966a, c, 1966e, 1968c Burch, J.Q., 1974c California, 1943b, K 8 Bentson, 1940, 1944, Schenck & К, 1940b, 1950 Calliostoma, 1975b Cape Arago, К & Doty, 1942 Cardiacea, 1969d Cardiacea, brackish water, 1954e Cardiidae, 1936a, b, 1937b, 1938b, 1949e, 1950a, b, 1951a, 1954d, 1973a, 1979a, 1980f Carditidae, 1980g Carpenter, P.P., 1968c Caribbean Province, Pacific outposts in Tertiary, 1976a Cate, C.N., 1977a Cavolina, 1964h Cavolinia, 1964h Cenozoic bivalves, 1969d Cenozoic gastropods, 1960] Cenozoic invertebrates, 1968e Central America, west, see west American. . . (tropical) Cerastes, 1961i Chace, E.P., 1981d Check-lists, 1937g, 1953a, 1956b, 1976b, К & Bentson, 1940, 1944 Chlamydoconchacea, 1969d Citation, 1958b Clathurella, 1962e Clavagellacea, 1969d Cleaning shells, 1959a Clinocardium, 1936a Cocculinacea, 1960] Colombia, 1958d Columbella procera, 1968b Conrad, T.A., 1966a Conus, 1980c Conus gloriamaris, 1964a Copenhagen, collections, 1966k Corbiculacea, 1969d Correlation, percentage method, 1937d, 1940a, 1942a Crassinella, 1938a, 1939a Cypraea mus, 1962b Deep-water paradox, 1977f Dendropoma, К & Morton, 1960 Dominican Republic, 1982c Dreissenacea, 1969d Durania, 1967e Eastern Pacific, see West American. ... Episcynia, 1946 Epitoniidae, 1962f, 1969b Eratoidae, 1977a Eualetes, 1971a European museums, 1965c Faunules, molluscan, Schenck & К, 1937 Finlay, H.J., 1970d Fissurellacea, 1960] Fossils, California, Schenck & К, 1940b, 1950 Gari, 1961h A. MYRA KEEN 399 Gastrochaenacea, 1969d Gastropoda, К & Pearson, 1952 Gastropoda, Cenozoic, 1960] Gastropoda, classification, 1953d Gatun formation, K 8 Thompson, 1946 Glossacea, 1969d Graham, J.J., 1970a Gryphaea, К & Muller, 1951 Guidelines for writers and readers, 19774 Gulf of California, 1961a, 1962d, 1964c, 1970c, 1971c, 1982d Hiatellacea, 1969d Hinds, R.B., 1966c Hippella, 1966) History, 1949a, 1982a Hoaxes, 1963f, 1977с Homonyms, renaming primary, Schenck & К, 1941, 1942 “-P” and “-i?’, 1958a Illustration, scientific, 1970b Index-method, Schenck 8 K, 1937 International Commission on [and Rules of] Zoolog- ical Nomenclature, 1945a, 1959b, 1960c Intertidal zonation, 1960h Isla Espíritu Santo, 1964c Isoarcidae, 1969d Japan, 1940b, 1949f, 19634, 1973a, 1985 Japan, Emperor of, 1985 Jumala, 1955a Keys, 1963b, К 8 Coan, 1974, К & Frizzell, 1939, 1953, К & Pearson, 1952 Knight, J.B., 1953d Kossuth, 1974d Lasaea, 1938a, 1939a Laternula, 1969c Laura, 1974a “Leaflets in Malacology”, 1984 Letters a curator receives, 1976c Linnaeus, 1962m, 1983a, 1983d London Colloquium on Taxonomy, 1960c Longevity, 1942b Lucina, 1973c, К & Abbott, 1972 Macoma inquinata, 1962a Mactracea, 1969d Marginellidae, 1973a Mazatlán, 1968c Mexico, west, 1958а, 1962k, 1964c, 1971c Miocene, 1943b, 1982c, К & Thompson, 1946 Mitridae, 1970f Montacutidae, 1962k Mörch, 1963c, 1966d Mounting specimens, 1949b Muller, S.W., К, Silberling & Page, 1971; Page, Silberling & K, 1975 Мипсасеа, 19641, 1971e Muricanthus, 1964e Muricidae, 1959d, 1964e, 1971a, К & Campbell, 1964 Myacea, 1969d Nana, 1964f Nemocardium, 1949e, 1950a Nerita scabricostata, 1942b Neritacea, 1960) Neritopsis, 1964b New Guinea, 1945b New Zealand, 1970d Nicol, D., 1956a Nomenclature, 1956c, 1962f, 1970f Nuculidae, 1969d Nuttall collection, 1966a Ocenebra, 1964e Oken, 1962f Oldroyd, 1.S., errata, 1978c Orbigny, A. d', 1966e Oregon, K 8 Doty, 1942 Pacific Division, American Malacological Union, 1949c Paleontological hoaxes, 1963f, 1977c Paleontology, 1968e Panama, 1969b, K 4 Thompson, 1946 Panamic Province, 1958d, 1960e, 1971c, 1980c Pandoracea, 1969d Panopea, 1961] Patellacea, 1960] Pearl culture, 1949f, 1963g Pelecypoda, see Bivalvia Peru, 1971c Philobryidae, 1969d Phyllonotus, 1960d Poe, E.A., 1936c Poromyacea, 1969d “Procedure in taxonomy”, Schenk & McMasters, 1948, 1956, 1959 Provinces, marine molluscan, Schenck & К, 1936b, c Psammotreta, 1962f Pseudocardia, 19809 Pupa, 19621 Purpura, 1964e Rafinesque, 1974e Rare and endangered mollusks, 1970g Realia, К & Coan, 1969 Referee, role of editorial, 1978b Renier, 1951c, 1952a, 1954c Round Mountain silt, 1943b Sacoglossa, 1973c, 1974a, b Schenck, H.G., 1945b, 1980e Schizothaerus, 1962f Sea-floor oasis, 1977d “Sea shells of tropical west America”, 1971d Seashores, 1965d Semele, 1966d Siliguariidae, Morton & К, 1960 Similivenus insolida, 1954a rev. ed., 400 ROBERTSON 8 COAN Siphonium, 1980h, 1981c Solenacea, 1969d South America, 1966e Spiroglyphus, 1980b, K 8 Hadfield, 1985 Stanford University, 1943a, 1947, 1955c, 1961m, 1974d Statistics, 1937e Stephopoma, Morton & К, 1960 Stoa, 1980, К & Hadfield, 1985 Stohler, R., 1983b Synonymy, 1967c, 1953 “Taxonomy, procedure in”, Schenk & McMasters, 1948, 1956, 1959 Tectarius, 19661 Telegraphic writing style, 1977e Tellinacea, 1969d Temblor formation, 1943b Tertiary, 1952b, 1940, K 8 Bentson, 1940, 1944 “Treatise on Invertebrate Paleontology”, 1953c, 1955b, 1960}, 1969d Tresus, 1962f Tridacnacea, 1969d Trochacea, 1960} Truncatella, K & Muller, 1952 Tryon, 1974e Type specimens, 1962j, 1965c, 1966a, 1966e, 1968c, 1969a Typhinae, 1939b, 1944, K & Campbell, 1964 Typhis, 1939b United States, western, 1968e Vaqueros formation, 1942a “Veliger”, 1983b Veneracea, 1969d Veneridae, 1951b, 1951d, 1954a, b Ventricolaria, 1954a Vermetidae, 1949g, 1960h, 1961c, 1971a, 1973a, 1980b, 1982c, 4, 1983c, К & Hadfield, 1985, К & Morton, 1960 Vibratool, 1959a Voluta mitra episcopalis, 1966h, 1967b West American marine mollusks (temperate to arc- tic), 1937с, +, g, 1940b, 1951b, 1956b, 1963b, 1966a, 19704, 1975b, 1982e, K & Coan, 1974, К & Frizzell, 1939, 1953, Schenck € К, 1936a, с West American marine mollusks (tropical), 1958c, d, 19594, 19604, e, 1965c, а, 1966d, e, 1968c, 1971c, 1983c, К & Coan, 1975a, b, К & Smith, 1961 Willett, 1974e Writers and readers, 19779 Xenophora, 1963d A. MYRA KEEN 401 OTHER LITERATURE CITED OR CONSULTED [ABBOTT, R.T.], 1986, A. Myra Keen— 1905-1986. Nautilus, 100(1): 38 (31 Jan.). ABBOTT, R.T. 8 YOUNG, M.E., eds., 1973, Amer- ican malacologists . .., ed. 1 [only ed.]. Ameri- can Malacologists, Falls Church, Virginia, iv + 494 pp. ADDICOTT, W.O., 1970, Miocene gastropods and biostratigraphy of the Kern River area, California. [United States] Geological Survey Professional Paper, 642: 174 pp., 21 pls. AMERICAN MEN AND WOMEN OF SCIENCE, 1976. Bowker, New York and London, ed. 13, 3: 2267. ANONYMOUS, 1958, Sea shells of tropical west America—a new book by Myra Keen for the amateur sea shell collector. [Stanford University] Mineral Sciences Newsletter, p. 9 (Dec.). ANONYMOUS, 1986a, A. Myra Keen, Stanford professor, 79, authority on seashells, mollusks. San Jose Mercury News, 135(8): 7-B (8 Jan.). ANONYMOUS, 1986b, Dr. A. Myra Keen 1906 [sic]-1986. Hawalian Shell News, 33(3): 12 (March). COAN, E.V., 1979, Recent eastern Pacific species of the crassatellid bivalve genus Crassinella. Veliger, 22(1): 1-11, 4 pls. (1 July). COAN [E.V.], 1983a, Transcript of oral history of Myra Keen. Tapes in Smithsonian Institution Archives; copies of tape and transcript in Amer- ican Malacological Union Archives at Academy of Natural Sciences of Philadelphia (taped Sept.). COAN, E.V., 1983b, A Semele story (Bivalvia: Semelidae). Nautilus, 97(4): 132-134 (28 Oct.). COAN, E.V., 1984, The Bernardinidae of the east- ern Pacific (Mollusca: Bivalvia). Veliger, 27(2): 227-237, 3 pls. (5 Oct.). COAN, E.V., 1986a, Myra Keen, 1905-1986: an appreciation. The Festivus (San Diego Shell Club), 18(2): 14-15 (14 Feb.). COAN, E. [V.], 1986b, A. Myra Keen (1905-1986). Veliger, 29(1): 2 (1 July). DuSHANE, H., 1974, The Panamic-Galapagan Epitoniidae. Veliger, 16 (Suppl.): 84 pp., 15 pls. (31 May). EVITT, W.R., INGLE, J.C., Jr. & KRAUSKOPF, W.B., 1986, Memorial resolution: A. Myra Keen 1905-1986. Stanford University Campus Report, May 21, pp. 14-15. GEMMELL, J., MYERS, B.W. & HERTZ, C.M., 1983, Observations on Tellina coani Keen, 1971. The Festivus (San Diego Shell Club), 15(10): 103-104 (13 Oct.). GIBBONS, A., 1979, Stanford professor receives prestigious science award. Peninsula Times Tri- bune, 1(200): B-1 (20 Nov.). GREEN, L., 1975a, Shell expert to meet Hirohito. Palo Alto Times, 83(242): 13 (9 Oct.). GREEN, L., 1975b, Meeting an emperor, seated like a queen.' Palo Alto Times, 83(243): 1-2 (10 Oct.). GREEN, L., 1984, Myra Keen pulls up roots after a life devoted to science. Peninsula Times Tri- bune, 6(44): C-1, C-6 (13 Feb.). HOWARD, F.B., 1970, Myra Keen and the singing snails. The Tabulata (Santa Barbara Shell Club), 3(3): 3-8 (1 July). JOHNSTON, T., 1986, Myra Keen, world-renowned expert on seashells, dies at 80. Stanford Univer- sity Campus Report, 18(14): 4 (8 Jan.). KING, C.L., 1983, Stanford geologist became first lady. Stanford Earth Scientist Section of the Stanford Observer, October: 7-8. KREISLER-BOMBEN, K. von, 1984, Myra Keen, Pp. 44-45 in: So, who's retired? Stanford Mag- azine, 12(4): 42-47 (Winter). MOORE, E.J., 1986, [untitled obituary]. Shells and Sea Life, 18(1): 5 (Jan.). MOORE, G.W., 1986, Afngeline] Myra Keen. Geotimes, 31(4): 26. PALO ALTO FRIENDS MEETING, 1986, Memorial minute for Myra Keen, First month 1986, 2 pp. SCHUTZ, R. & RUTH, S., 1983, “There is a Beyond we're living in . . . ” Interviews with Myra Keen. . . . Friends Bulletin, 51(7): 111-115 (April). SMITH, J.T., 1978, Primary types in the Stanford paleontological [and neontological] type collec- tion Bulletins of American Paleontology, 72(300): 313-552 (14 March). SMITH, J.T., 1986, A. Myra Keen (1905-1986). Veliger, 28(4): 463-464 (1 April). SOLEM, A., 1975, The Recent mollusk collection resources of North America. A report to the Association of Systematics Collections. Veliger, 18(2): 222-236 (1 Oct.). STANFORD [UNIVERSITY] FACULTY, 1980 ed., р: 13: TIMES TRIBUNE STAFF, 1986, A. Муга Keen, seashell expert, former Stanford professor, dies. Peninsula Times Tribune, 8(8): C-6 (8 Jan.). VOKES, Е.Н., 1975, Cenozoic Muricidae of the western Atlantic region. Part Vi—Aspella and Dermomurex. Tulane Studies in Geology and Paleontology, 11(3): 121-162, 7 pls. (5 Feb.). 402 ROBERTSON 8 COAN LIST OF 40 TAXA NAMED IN HONOR OF DR. A. MYRA KEEN Mainly after Howard (1970) Unless otherwise indicated, the taxa are molluscan. Trachycardium keeni Glibert, 1936 Septifer keeni Nomura, 1936 Chattonia trigonata keeni Chavan, 1939 Alvania (Willettia) keenae Gordon, 1939 Anomalina keenae Martin, 1943 (Foraminifera) Glycymeris keenae Willett, 1944 Ocenebra keenae Bormann, 1946 Permopora keenae Elias, 1947 (Bryozoa) Rissoina keenae Smith & Gordon, 1948 Schizothaerus keenae Kuroda & Habe, 1950 Bornia (Temblornia) keenae Marks, 1951 Keenaea Habe, 1951 Teinostoma myrae Pilsbry & Olsson, 1952 Venericardia keenae Verastegui, 1953 Ensis myrae Berry, 1953 Pyrina keenae Hamilton, 1956 (Echinoidea) Angaria keenae Von der Osten, 1957 Chione keenae Soot-Ryen, 1957 Stephopoma myrakeenae Olsson & McGinty, 1958 Nomaeopelta myrae Berry, 1959 Eocypraea (Apiocypraea?) keenae Woodring, 1959 Trivia myrae Campbell, 1961 Periploma keenae Rogers, 1962 Glyphostoma myrakeenae Olsson, 1964 Transenpitar keenae Fischer-Piette & Testud, 1968 Typhis (Rugotyphis) keenae Gutmann, 1969 Cinclidotyphis myrae DuShane, 1969 Mitrolumna keenae Emerson 4 D'Attilio, 1969 Clinocardium myrae Adegoke, 1969 Scissurella keenae McLean, 1969 Murexiella keenae E. Vokes, 1970 Aspella myrakenae Emerson 8 D'Attilio, 1970 Calliostoma keenae McLean, 1970 Glyphostoma myrae Shasky, 1971 Callucina keenae Chavan in Moore, 1971 Primovula myrakeenae Azuma & Cate, 1971 Littorina keenae Rosewater, 1978 Myrakeenini Harry, 1985 (new tribe) Myrakeena Harry, 1985 (new genus) Tritonia myrakeenae Bertsch & Mozqueira, 1986 MALACOLOGIA, 1986, 27(2): 403—404 SOME IMPORTANT SOURCES FOR SUBSEQUENT DESIGNATIONS OF THE TYPE SPECIES OF MOLLUSCAN GENERA! A. Myra Keen (posthumous) Stanford University MONTFORT, D. de, 1810, Conchyliologie sys- tématique. Paris, 2: 676 pp. SCHMIDT, F.C., 1818, Versuch úber die beste Einrichtung zur Aufstellung, Behandlung und Aufbewahrung der verschieden Naturkórper und Geganstánde der Kunst vorzüglich der Conchylien-Sammlungen. ... Gotha, 8 + 252 pp. See WINCKWORTH, R., 1944, Proceedings of the Malacological Society of London, 26(1): 23-24. FLEMING, J., 1818, 1822, Conchology. Mollusca. Supplement to eds. 4-6 of Encyclopaedia Britan- nica, 3: 284-314; 5: 567-577. See WINCK- WORTH, R., 1929, Proceedings of the Malacological Society of London, 18(5-6): 224-228, 263. CHILDREN, J.G., 1822-1824?, Lamarck's genera of shells. Quarterly Journal of Science, 14: 64-86; 14: 298-322; 15: 23-52; 15: 216-258; 16: 49-79; 16: 241-264. Also reprint, 177 pp. See KENNARD, A.S., SALISBURY, A.E. 8 WOODWARD, B.B., 1931, Smithsonian Miscel- laneous Collections, 82(17): 1—0. ANTON, H.E., 1838, Verzeichniss der Con- chylien. . . . Halle, xvi + 110 pp. Date corrected from 1839 following statement on p. 110 and CERNOHORSKY, W.O., 1978, Veliger, 20(3): 299. HERRMANNSEN, A.N., 1846, 1849, 1852, Indicis generum malacozoorum primordia. . . . Kassel, 2 vols., xxvii + 637 + xxixoxlil + 717 + v + 140 pp. GRAY, J.E., 1847, A list of the genera of Recent Mollusca, their synonyma and types. Proceed- ings of the Zoological Society of London, 15: 129-219. WOODWARD, S.P., 1851-1856, A manual of the Mollusca. . . . [ed. 1]. London, pp. хуй + 486 + 24 pp. STOLICZKA, F., 1867-1868, Cretaceous fauna of southern India. The Gastropoda. Memoirs of the Geological Survey of India; Palaeontologia Indica, [ser. 1], 2: 498 pp., 28 pl. TATE, R., 1868, Appendix to the Manual of Mollusca of S.P. Woodward. . . . [in ed. 2], London, 86 pp. STOLICZKA, F., 1870-1871, Cretaceous fauna of southern India. The Pelecypoda, with a review of all known genera of this class, fossil and Recent. Memoirs of the Geological Survey of India: Paleontología Indica, [ser. 1], 3: xxii + 1-537 pp. KOBELT, W., 1876-1881, Illustrirttes Con- chylienbuch. Nürnberg, 2 vols., хм + 392 pp. 112 pl. Dates from REHDER, H.A., 1952, Nautilus, 66(2): 59-60. BUCQUOY, E., DAUTZENBERG, P. & DOLLFUS, G., 1882-1898, Les Mollusques marins du Rous- sillon. Paris, 2 vols., 570 + 884 pp. COSSMANN, [A.E.M.], 1886-1891, Catalogue [il- lustré] des coquilles fossiles de ГЕосёпе des environs de Paris. . . . [fasc. 1-5]. Annales de la Société Royale Malacologique de Belgique, 21: 17-186; 22: 3-214; 23: 3-324; 24: 3-381; 26: 3—163. PILSBRY, Н.А. et al., 1888-1898, Manual of Conchology, ser. 1, 10(2): 161 to 17: 348. Conchological Section, Academy of Natural Sci- ences of Philadelphia. DALL, W.H., 1890-1903, Contributions to the Ter- tiary fauna of Florida. . . . Transactions of the Wagner Free Institute of Science of Philadelphia, 3(1—6): 1654 pp. ЗАССО, F., 1890-1904, I molluschi dei terreni terziarii del Piemonte e della Liguria. . . . [contin- uation of work started by L. Bellardi]. Turin, parts 6-30. Later parts only. Also published in Memorie della Reale Accademia delle Scienze di Torino. COSSMANN, [A.E.M.], 1895-1925, Essais de paléoconchologie comparée. Paris, livraisons AS 159 AMIA Aone 2S} Se lo 261 + 248 + 215 + 292 + 388 + 348 + 345 рр. DALL, W.H., 1899-1903, [Synopses of Lep- tonacea, Solenidae, Tellinidae, Cardiidae, Lucinacea, Veneridae, Astartidae]. Proceedings of the United States National Museum, 21: 873-897; 22: 107-112; 23: 285-326; 23: 381-392; 23: 779—833; 26: 335—412; 26: 933-951. SUTER, H., 1913, Manual of the New Zealand Mollusca. Wellington, xxiii + 1120 pp. WOODRING, W.P., 1925, 1928, Miocene mollusks from Bowden, Jamaica [Part I], Pelecypods and scaphopods. Part Il, Gastropods and discussion of results. Carnegie Institution of Washington ¡Based on a work sheet given to students in Dr. Keen's malacology course in the mid-1950s. lt deserves wider dissemination. ED. (403) 404 A. MYRA KEEN Publication 366: v + 222 pp.; 385: vii + poda and some (mainly Paleozoic) Caenogastro- 564 pp. poda and Opisthobranchia. Geological Society of KNIGHT, J.B., 1941, Paleozoic gastropod geno- America and University of Kansas Press, хх + types. Geological Society of America Special 351 pp. Papers 32: vi + 510 pp. MOORE, R.C., ed., 1969-1971. Treatise on inver- MOORE, R.C., ed., 1960, Treatise on invertebrate tebrate paleontology. Part N. Mollusca 6. Bival- paleontology. Part |. Mollusca 1. Mollusca—gen- via. Geological Society of America and University eral features; Scaphopoda; Monoplacophora; of Kansas, 3 vols., xxxvili + ii + iv + 1224 pp. Gastropoda—general features; Archaegastro- INDEX TO TAXA IN VOLUME 27 An asterisk (+) denotes a new taxon. The articles on Dr. A. Myra Keen (pp. 379—408) are not indexed here (see index of key words in the titles and contents of her molluscan papers, рр. 402—404). Acanthoceras, 22, 25 angustatum, Hauericeras, 27 Acanthoceratidae, 25 angusticollis, Saphinotus, 307-311 Acanthodiscus, 19 Anisoceras, 24 Acanthohoplites, 20, 25 Anisoceratidae, 24 Acantholabia, 39 annectens, Samoana, 97-106 Acanthoscaphites, 24 Anodonta, 107-125, 195, 198, 199 Acanthotrophon, 39 Anodontinae, 125 Acer, 353 Antillophos, 39 Achatina, 72, 73, 75, 76 Antrobia, 159, 160 Achatinella, 67-81 antrorum, Palaemonetes, 129 Achatinellidae, 68, 72 Antroselates, 160 Achatinellinae, 67-81 apexfulva, Achatinella, 67 Acochlidiacea, 83 Aphera, 32, 33 Acompsoceras, 22 Apiocardia, 41 Acrioceras, 11, 24, 26 arboreus, Zonitoides, 346, 347, 354 Acrosterigma, 41 arbucklensis, Patagiosites, 25 Adkinsia, 22 Arbutus, 353 Aegocrioceras, 24 Arcidae, 30, 32 affinis, Partula, 97-106 Arcthoplites, 21 Agaronia, 31, 39 arenaria, Mya, 198, 296, 297 Agassitula, 39 areolata, Doriopsilla, 83-96 agnessense, Speetoniceras, 28 Aresoceras, 22 aguila, Hertleinites, 28 Argentiniceras, 18 Aioloceras, 21 Argonauticeras, 17 alaskana, Vitrina, 346, 347, 354 Arguethites, 26 Alathyria, 185-202 arguta, Partula, 97 Alcira, 39 Ariolimax, 310, 346, 354 aldersona, Anahamulina, 26 Arion, 346, 354 aldersona, Phylloceras, 27 Arionidae, 307—311 Allocrioceras, 24 arrhaphoceras, Pleurohoplites, 21 Allogona, 345, 354 Artesia, 129 Allotexiweckelia, 129 Artesiidae, 129 Alnus, 353 Arthropoda, 129 Alyxia, 68 Asellidae, 129 ambiguus, Velesunio, 186, 196 Asellus, 129 Ambleminae, 125 Aspella, 39 Ambystomidae, 129 Aspidostephanus, 18 americana, Parabogidiella, 129 Aspinoceras, 24 Americardia, 41 Asterias, 289 Ammonoceratites, 17 Asteroidea, 47, 48 Ammonoidea, 3-28 Astiericeras, 20 ampla, Nymphaea, 359 Astieridiscus, 19 Anachis, 39 Astyris, 39 Anagaudryceras, 13, 17, 26 ater, Arion, 346, 354 Anahamulina, 12, 24, 26 Atrimitra, 40 Anahoplites, 21 attenuata, Samoana, 97-106 Anapachydiscus, 20, 25 aurarium, Eogaudryceras, 26 Anapuzosia, 19 Auriculella, 80 Ancilla, 31-33, 39 Auriculellinae, 80 Ancyloceras, 15, 24, 26 Aurinia, 40 Ancyloceratidae, 10, 24, 26 Australiceras, 24 andersoni, Prophysaon, 307-311, 354 australis, Hyridella, 185-202 andrusiana, Vertigo, 342, 347, 354 angasi, Velesunio, 199 Bactrospira, 38 Angiostrongylus, 75 Baculites, 10, 13-15 angulata, Sarasinella, 27 Baculitidae, 13, 16 angulatum, Gabbioceras, 26 Bailya, 31, 39 (405) 406 INDEX balconis, Stygobromus, 129 californicum, Plesiovascoceras, 25 *Balconorbis, 129, 132, 137, 141, 144, *152-160, californicum, Thurmanniceras, 27 163, 168-171 californicus, Anapachydiscus, 25 Balearites, 24 Callianax, 39 Bankia, 325-339 Calliotectum, 40 Barremites, 19, 27 Calliptylloceras, 18, 27 Barroisiceras, 23 Calliptychoceras, 18 *bartonensis, Stygopyrgus, 129, 132, 141, 144, Callizoniceras, 19 153—*157, 163, 168—171 Calophus, 39 bartschi, Teredo, 324—339 Calotrophon, 39 batesi, Lytoceras, 27 Calycoceras, 22, 25 bauchioceras, Pseudotissotia, 23 camara, Lantana, 68 Bayleites, 20 Canadocerus, 14, 20, 25 Beaudanticeras, 27 Cancellariidae, 30-32 bellula, Achatinella, 70 cantabrigites, Mortoniceras, 22 Benueites, 22 cantonensis, Angiostrongylus, 75 Berriasella, 18 Cardiidae, 30, 32, 41 Berriasellidae, 11, 18, 27 Carella, 79 besairiei, Kilianella, 27 carolinianus, Philomycus, 271-280 Beudanticeras, 19 caroliniensis, Tebennophorus, 271 Bevahites, 23 Carychium, 345, 347, 354 Bhimaites, 19 Casmaria, 38 bifurcatus, Stygobromus, 129 Cassis, 38 bilineatus, Meghimatium, 271 Casuarina, 79 bilineatus, Philomycus, 271, 276 cataracta, Anodonta, 108-117, 119, 121-123, 125 binneyana, Nesovitrea, 247, 354 caudata, Eupleura, 39 binneyana, Retinella, 342 cavernarum, Cyclops, 129 Binneyites, 22 Cavilinga, 41 Binneyitidae, 22 Celatoconus, 39 Biomphalaria, 173, 243-247, 249, 255, 257, 258, Cephalopoda, 3-28 261, 266, 313-321 Cerithium, 332 bipartitus, Lyrodus, 325, 327-339 Charonia, 31 blandi, Toxoceras, 26 Charopidae, 203-241 Bochianites, 24 Cheloniceras, 12, 20, 21, 25, 27 Bochianitidae, 24 chicoense, Submortoniceras, 9, 25 Bogidiellidae, 129 Chicoreus, 39 Borrissjakoceras, 22 Chlorophyta, 299 Bostrychoceras, 9, 15, 24, 25 Choffaticeras, 23 boulei, Calycoceras, 25 Chordata, 129 boulei, Pseudoschloenbachia, 25 churinceanus, Mexipyrgus, 357-378 Brachiopoda, 55 Cichlasoma, 250, 367 Bradybaena, 76 Cichlidae, 367 Brahamites, 20 Cigclirina, 39 Brancoceras, 21 Cinctura, 39 branneri, Subprionocyclus, 25 cinerea, Urosalpinx, 281-290 Branoceratidae, 21, 25 Cionella, 346, 354 brevifrons, Chicoreus, 39 clappi, Planogyra, 347, 354 breweri, Beaudanticeras, 27 Clenchina, 40 broadi, Simbirskites, 28 Clioscaphites, 24 Broderiptella, 32, 33, 38 Coahuilix, 157-160 Buccinidae, 30-32, 39 Codakia, 41 Buchiceras, 23 Coelenterata, 46 buckhami, Pachydiscus, 25 Coilopoceras, 23 Budaiceras, 22 Coilopoceratidae, 23 bulimoides, Achatinella, 70 Colchidites, 24 Bulinus, 249-263 Collignoniceras, 23, 25 Bullata, 32, 33, 40 Collignoniceratidae, 23, 25 Bullia, 299-305 collinus, Philomycus, 271 burchi, Samoana, 97-106 Colombiceras, 20 Columbella, 39 caesia, Achatinella, 72 Columbellidae, 30-32, 39 Cainoceras, 22 Columbellopsis, 39 californica, Hypophylloceras, 27 columbiana, Vertigo, 342, 347, 354 INDEX 407 columbiana, Vespericola, 347, 354 dawsoni, Desmoceras, 27 columbianus, Ariolimax, 310, 346, 354 decollata, Rumina, 333 Columella, 347 Delphinites, 19 communis, Ficus, 38 demissa, Geukensia, 281-290 complanata, Elliptio, 107-125, 195, 199 Dendrodoris, 95 Conella, 39 denmanense, Gaudryceras, 26 confusa, Partulina, 68, 69, 71 denseplicatum, Gaudryceras, 9, 26 *сотса, Phreatodrobia, 129, 132-*149-172 densicostata, Sarasinella, 27 conica, Samoana, 97-106 depressa, Hyridella, 185-202 Conidae, 31, 32 Dermomurex, 39 conspectum, Punctum, 354 Deroceras, 276, 346, 354 cookei, Microphysula, 345, 347, 354 Deshayesites, 20 Coprosma, 68 Deshayesitidae, 20 corae, Toxoceratoides, 26 Desmoceras, 14, 19, 27 Corbula, 292, 294 Desmoceratidae, 10, 13, 15, 16, 19, 27 Cordatum, Pleurobema, 199 Desmophyllites, 9, 19, 27 Cornus, 353 deveroide, Romaniceras, 25 Coronites, 20 devia, Triodopsis, 350 Costoanachis, 39 Diadochoceras, 20 Cotonopsis, 39 diaphana, Samoana, 97-106 Cottreauties, 22 Diaziceras, 23 Crangonyctidae, 129 Dibaphimitra, 40 Craspeditidae, 11, 18 Dibaphus, 33, 40 Craspedodiscus, 18 Dichotomites, 18 crassiplicata, Kilianella, 27 dictyodes, Helix, 214, 216 Crassostrea, 173, 281, 288 dictyodes, Pararhytida, 203-241 crebrisulcatus, Protegragnoites, 26 dictyodes, Trochomorpha, 214 Crepidula, 278, 331 dictyonina, Helix, 225 Crioceratidae, 10, 26 dictyonina, Pararhytida, 203-241 Crioceratites, 11, 24, 26 Didymoceras, 24, 26 crispus, Rumex, 320 dieradoceras, Mortoniceras, 22 cronkhitei, Discus, 346, 354 digitalis, Bullia, 299-305 Crustacea, 3-28 dilatata, Elliptio, 109, 110, 114, 115, 119, 121, Ctena, 41 125 cubensis, Fossaria, 249 Dimorphoplites, 21 cubensis, Lymnaea, 266 Dinocardium, 33, 41 Cucumerunio, 185-202 Dioscorea, 70 culveri, Antrobia, 164 diphyloides, Desmophyllites, 9, 27 Cuyaniceras, 18 Diplacmoceras, 21 Cyamhoplites, 21 Diplomoceras, 24 Cyclopidae, 129 Diplomoceratidae, 10, 16, 24, 25 Cyclops, 129 Dipoloceras, 21 cygnea, Anodonta, 195, 198, 199 Discohoplites, 21 Cymatiidae, 31 Discoscaphites, 24 Cymatophos, 39 Discus, 346, 354 Cymbula, 38 dislocata, Terebra, 291-298 Cyphoma, 38 Dissimilites, 24, 26 Cypraeacea, 31, 32, 38 Distoloceras, 19 Cypraecassis, 38 Divalinga, 41 Cypraeidae, 31 Dodonaea, 68 Cypraeolina, 40 Dolicholatirus, 39 Cypraeovula, 33, 38 Dolomena, 33, 38 Cypridae, 129 Doridacea, 83 Cypridopsis, 129 Doriopsilla, 83, 96 Douvilleiceras, 20, 25 Dactylidella, 39 Douvilleiceratidae, 20, 25 Dactylidia, 39 drapeta, Hyridella, 190, 196 Dalium, 38 Dubautia, 68 dallasi, Paraphoplites, 25 Dufrenoyia, 20 Dallocardia, 41 duncanense, Hoplocrioceras, 26 damesi, Damesites, 27 Dunveganoceras, 22 Damesites, 13, 19, 27 duplicatus, Polinices, 291, 292, 295, 297 Daradiceras, 23 Durangonella, 158, 370 408 INDEX durnovarites, Mortoniceras, 22 Fagesia, 22 duryi, Helisoma, 257 Fagus, 350 Dytiscidae, 129 Farnhamia, 21 Fasciolaria, 39 Eburna, 31-33, 40 Fasciolariidae, 31, 32, 39 Eburnospira, 40 Favartia, 39 Echinoderma, 43 Favrella, 18 Echinoidea, 43-66 Ferrissia, 320 Echinophoria, 38 Ficheuria, 22 Ectocarpus, 301 Ficidae, 31 edentula, Columella, 347 fidelis, Monadenia, 345, 347, 354 edulis, Mytilus, 281-290 fimbriatula, Bankia, 325-339 egertoni, Pachydiscus, 25 flagellatus, Stygobromus, 129 Egouena, 40 flexuolaris, Philomycus, 271 Eichwaldiella, 38 Flickia, 22 electrina, Nesovitrea, 350 Flickiidae, 22 electrina, Retinella, 342 floridanus, Lyrodus, 325, 327-339 elephans, Ancyloceras, 26 florifer, Chicoreus, 39 Elliptio, 107-125, 195, 199 Floritula, 39 elodes, Lymnaea, 249 foliolatum, Prophysaon, 307-311, 354 elongatum, Bostrychoceras, 9, 25 Fontelicella, 370 Enaeta, 40 Fontigens, 159—161 Endodontidae, 72 Forbesiceras, 21 Engina, 31, 39 forbesi, Asterias, 289 Engonoceras, 20 Forbesiceratidae, 21 Engonoceratidae, 20 Forresteria, 23 Ensis, 296 forskalii, Bulinus, 249, 259 Enteromorpha, 301 Fossaria, 249 Entocytheridae, 129 fragilis, Anodonta, 108-114, 117, 119, 122, 123, Eocypraea, 38 125 Eodesmoceras, 19 Frenguelliceras, 18 Eogaudryceras, 17, 26 fulgens, Achatinella, 78 Eopachydiscus, 20 fulica, Achatina, 72, 73, 75, 76 Eotetragonites, 17, 26 fulvus, Euconulus, 327 Epengonoceras, 21 furcifera, Teredo, 327 epicheloniceras, Cheloniceras, 20 Fusimitra, 40 Epigoniceras, 9, 18, 26 Fusinosteira, 39 epigonum, Epigoniceras, 9, 26 Fusinus, 39 Epihoplites, 21 Epileymeriella, 21 Gabbioceras, 17, 26 Erato, 38 gabbi, Prochelaea, 41 Eratoidae, 31 gainesi, Eotetragonites, 26 Eratoidea, 40 gallus, Tricornis, 38 Erosia, 38 gardeni, Hauericeras, 9 Eubrancoceras, 21 дагапеп, Acanthoplites, 25 Eucalycoceras, 22, 25 Gargasiceras, 20 Euconulus, 347, 354 Gastropolites, 21 Euglandina, 75 gaudi, Biomphalaria, 320 Eugomontia, 299-305 gaudichaudiana, Scaevola, 68 Euhystrichoceras, 21 Gaudryceras, 9, 13, 14, 17, 26 Eulophoceras, 23 Gaultheria, 343, 353 Eulopia, 41 Gauthiericeras, 23 Eulytoceras, 17, 27 Gemophos, 39 Euomphaloceras, 22 Geoplana, 72, 73 Eupachydiscus, 9, 20, 25 germana, Triodopsis, 345, 347, 354 Eupleura, 31 Germaniceras, 23 Euptychoceras, 26 Geukensia, 281-290 Euryprene, 39 gibba, Corbula, 292 Euryptychites, 18 gibba, Partula, 97-106 eurystomus, Satan, 129 Gibberula, 40 Euspira, 292, 293 gibbosa, Anodonta, 110, 114, 125 Exiteloceras, 24 giganteus, Neocraspedites, 28 exustus, Indoplanorbis, 265-269 gigas, Tricornis, 38 glabrata, Biomphalaria, 173, 243-247, 249, 255-257, 258, 261, 266, 313-321 glabrus, Tetragonites, 26 glandulosa, Hemphilla, 354 Glebosoceras, 23 glenelgis, Hyridella, 200 globosus, Bulinus, 255—257 Glyptoxoceras, 9, 15, 24, 25 Gombeoceras, 22 Gordanops, 39 gorgasiana, Fasciolaria, 39 gouldi, Bankia, 325-339 grandis, Anodonta, 109-119, 121, 122, 125 granulata, Venericardia, 292 Graysonites, 25 greeni, Toxoceratoides, 26 Grevillea, 68 Groebericeras, 18 Grossouvreites, 20 guajava, Psidium, 68 gudei, Tropidotropis, 203, 206 Gunnarites, 20 Hadeoporus, 129 Hadziidae, 129 Hamites, 12, 14, 24, 26 Hamiticeras, 24 Hamitidae, 10, 24, 26 hamlini, Acrioceras, 26 Hamulina, 12, 24 Hamulitidae, 26 Haplotrema, 307, 342, 347, 354 haradai, Eupachydiscus, 9, 25 Haresiceras, 21 harleites, Forresteria, 23 Harpeola, 33, 40 Hatchericeras, 19 Hauericeras, 5, 9, 19, 27 Hauffenia, 127, 129, 132-152, 158, 159, 161 Hawaiarca, 33 Hawalia, 350 Hectoroceras, 18 Heilprinia, 39 Helisoma, 257 Helisomatinae, 320 Helix, 173, 214, 216, 226 Hemibaculites, 26 Hemihoplites, 24 Hemihoplitidae, 24 Hemiptychoceras, 24 hemisonneratia, Protohoplites, 21 Hemitetragonites, 17 Hemitissotia, 23 Hemphilla, 354 hemphilli, Megomphix, 350 Here, 41 heros, Lunatia, 297 hertleini, Eogaudryceras, 26 Heteroceras, 15, 24, 26 Heteroceratidae, 24 heterophylla, Tsuga, 353 Heterotissotia, 23 hetonaiense, Neophylloceras, 27 INDEX 409 hetonaiensis, Damesites, 27 hirsuta, Allotexiweckelia, 129 hofmanni, Puzosia, 27 Holcodiscidae, 19, 27 Holcodiscoides, 19 Holcodiscus, 19 Holcophylloceras, 18 Hollisites, 28 holthuisi, Palaemonetes, 129 Homo, 73 Homolsomites, 27, 28 Hoplites, 21 Hoplitidae, 21 Hoplitoides, 23 Hoplitoplacenticeras, 14, 15, 21, 25 Hoplocrioceras, 24, 26 Hoploscaphites, 24 Horatia, 127, 129, 132-159, 161 horridus, Oplopanax, 353 hubrichti, Radiodiscus, 342 Hulenites, 19 hyatti, Sarasinella, 27 Hydrobiidae, 127-172, 357-378 Hydrobiinae, 129, 132, 158 Hypacanthoplites, 20 Hyphantoceras, 9, 15, 26 Hyphoplites, 21 Hypophylloceras, 18, 27 Hypoturrilites, 24 Hyridella, 185-202 Hyriidae, 185-202 Hysteroceras, 21 Ictaluridae, 129 Idiohamites, 24 llyanassa, 173-183 *imitata, Phreatodrobia, 129, 132-*151-172 implicata, Anodonta, 122 “inclinata, Phreatodrobia, 129, 133-"146-172 inda, Pseudophyllites, 26 indicum, Glyptoxoceras, 25 indicum, Schistosoma, 265 indigenes, Melchiorites, 27 Indoceras, 23 indopacifica, Mesopuzosia, 27 Indoplanorbis, 265-269 inflatus, Hollisites, 28 inflatus, Zelandites, 26 insolita, Texiweckelia, 129 intermedia, Puzosia, 27 Isara, 40 jackieburchi, Samoana, 97-106 Jacobites, 20 jamaicensis, Stachytarpheta, 68 japonicum, Desmoceras, 27 Jaspidella, 40 jeletzkyi, Heteroceras, 26 Jenneria, 38 Jimboiceras, 19 johnsoni, Pristiloma, 354 jordanense, Euptychoceras, 26 Juddiceras, 24 410 INDEX Kamerunoceras, 22 Lucinisca, 41 Kanabiceras, 22, 25 Lucinoma, 41 kawasakii, Texanites, 25 lugoi, Mexipyrgus, 361, 364, 366, 367, 371 kayei, Vertebrites, 26 Lunatia, 297 Kenkiidae, 129 Luria, 38 Kilianella, 18, 27 Lyelliceras, 22 Kilianiceras, 11 Lyelliceratidae, 22 Kitchinites, 19 Lymeriellidae, 21 klecakiana, Horatia, 153, 161 Lymnaea, 249, 266 Knemiceras, 20 Lyria, 40 Kossmatella, 17 Lyriinae, 41 Kossmaticeras, 20 Lyrodus, 325, 327-339 Kossmaticeratidae, 19, 27 Lyticoceras, 18 kossmati, Desmoceras, 27 Lytoceras, 14, 17, 27 Lytoceratidae, 15, 17, 27 Labiostrombus, 33, 38 Lytodiscoides, 19 laeve, Deroceras, 276, 346, 354 Laevicardium, 41 Macgintiella, 40 Laevityphis, 39 Macgintopsis, 39 Lampsilini, 125 Macrocypraea, 38 Lampsilis, 107-125, 185 macrophyllum, Acer, 353 lansingi, Pristiloma, 347, 354 Mahonia, 353 Lantana, 68 Malea, 38 lanza, Ficus, 38 Mammilla, 31 lapidaria, Pomatiopsis, 173 Mammites, 22 lasius, Zoogonus, 173, 183 Manambolites, 23 lateralis, Mulinia, 282-284 Mansfieldella, 40 latus, Crioceratites, 26 mansoni, Schistosoma, 243-247, 313-321 learii, Cyclops, 129 Mantelliceras, 22 lecontei, Leconteites, 27 Maorites, 20 lecontei, Simbirskites, 28 Margaritifera, 185, 199 Leconteites, 27 margaritifera, Margaritifera, 185, 199 Lemintina, 38 Marginella, 40 Lenticeras, 23 Marginellidae, 31, 32, 40 Lentigo, 38 Marginocypraea, 38 leoniceras, Choffaticeras, 23 Marshallites, 27 Leopoldia, 19 marteli, Pararhytida, 203-241 Lepilucina, 41 marteli, Trochomorpha, 226, 231 Leptegouana, 40 marteli, Videna, 226, 231 Lepthoplites, 21 Martesia, 325, 327-339 Leptopartula, 97 massa, Lyrodus, 327 Leptotetragonites, 17 Mazatlania, 39 Leucozonia, 39 Megalytoceras, 17 Levimyrtea, 41 Meghimatium, 271 Lewesiceras, 20 Megomphix, 350 Leymeriella, 21 Melanoides, 255-257 Libycoceras, 5, 17, 23 Melchiorites, 19, 27 Limaites, 18 Menabites, 23 Limax, 271, 273 Menuites, 20, 25 lineatum, Pseudoxybeloceras, 26 Menuthiocrioceras, 24 Liochlamys, 39 menziesi, Hyridella, 196 Lirceolus, 129 menziesii, Arbutus, 353 Lissonia, 18 menziesii, Pseudotsuga, 353 Litotrema, 39 Mercenaria, 281-290, 292, 296, 297 littoralis, Achatinella, 72 mercenaria, Mercenaria, 281-290, 292, 296, 297 Littoridininae, 127, 129, 152-158 merriami, Desmoceras, 27 longinqua, Fontelicella, 370 Mesogaudryceras, 17 Lophocardium, 41 Mesopuzosia, 13, 19, 27 Lopholobites, 20 Metahoplites, 19 lubrica, Cionella, 346, 354 Metalytoceras, 17 lucasi, Hollisites, 28 Metaphos, 39 Lucina, 41 Metaplacenticeras, 15, 21, 25 Lucinidae, 30, 32 Metasigaloceras, 22 INDEX Metatissotia, 23 Metengonoceras, 21 Metoicoceras, 22 Metrolytoceras, 17 Metrosideros, 68 Metula, 39 Mexicardia, 41 Mexipyrgus, 157, 158, 357-378 Mexithauma, 369 micra, Hauffenia, 127, 129, 134, 138 micra, Horatia, 138 micra, Phreatodrobia, 129, 132-172 micra, Valvata, 134 Microcardium, 41 Micromphalia, 203, 214 Microphysula, 345, 347, 354 Microrhytis, 39 Microspira, 40 mihoensis, Scalarites, 25 mikobokense, Anagaudryceras, 26 Miltha, 32, 33, 41 minckleyi, Cichlasoma, 367 minckleyi, Nymphophilus, 368, 369 miniata, Dendrodoris, 95 minimum, Haplotrema, 307 Minioliva, 40 miniscula, Hawalia, 350 Minitula, 39 Miocenebra, 39 Miodesmoceras, 19 Miogalea, 38 Mitrella, 31, 39 Mitridae, 30-32, 40 modesta, Vertigo, 350 mohri, Sphalloplana, 129 mojarralis, Mexipyrgus, 359, 363, 364, 366, 368, 371 Mojsisoviczia, 21 Monadenia, 345, 347, 354 monidum, Strioterebrum, 292, 294 Monodella, 129, 158 Monodellidae, 129 monodonta, Margaritana, 200 Monostiolum, 39 moria, Sphaeromicola, 129 Mortoniceras, 22, 25 Morum, 38 mouensis, Helix, 225, 226 mouensis, Pararhytida, 203-241 Mulinia, 282-284 multilamella, Venus, 292, 293 multilineatus, Mexipyrgus, 361, 363, 364, 366-368, 371 multizonata, Achatinella, 70 Muniericeras, 20 Muniericeratidae, 20 munitum, Polystichum, 343, 353 Мигасургаеа, 32, 38 Murex, 39 Murexiella, 39 Murexsul, 39 Muricanthus, 39 Muricidae, 30-32, 39 Muricopsis, 39 mustelina, Achatinella, 68, 70, 73, 76-78 mutabilis, Homolsomites, 28 Mutulella, 33 Mya, 198, 296, 297 Myrtea, 41 Mysterostropha, 40 Mytilus, 281-290 nanaimoense, Pseudoxybeloceras, 5 nasale, Schistosoma, 265 Nassariidae, 299-305 Nassarina, 39 Nassarius, 299, 303 Natalites, 20 Naticidae, 31 Nautiloidea, 3-28 Nautilus, 3, 15 navalis, Teredo, 325-339 Navanax, 261 Nebularia, 40 Negreliceras, 18 Nemocardium, 41 Neocomites, 18, 27 Neocosmoceras, 11, 18 Neocraspedites, 18, 28 Neocrioceras, 24, 26 Neocylindrus, 33, 40 Neogastropoda, 30, 299 Neograhamites, 20 Neoharpoceras, 22 Neohoploceras, 19 Neokentroceras, 22 Neolobites, 21 Neophlycticeras, 22 Neophylloceras, 27 Neoptychites, 22 Neopuzosia, 19 neosaynella, Cleoniceras, 21 Neosaynoceras, 22 Neosconsia, 38 Neosilesites, 19 neptuni, Subprionocyclus, 25 Nerva, 39 nervosa, Mahonia, 353 Nesovitrea, 342, 347, 354 Neurarhytis, 39 newberryanum, Canadoceras, 25 Newcombia, 67 Niceforoceras, 23 Nicema, 39 Nicklesia, 20 nickliniana, Fontigens, 161 nigrofasciatum, Cichlasoma, 250 Nipponites, 24, 26 Niteoliva, 40 Nitidella, 39 normalis, Subprionocyclus, 25 Northia, 31, 32, 39 Nostoceras, 24 Nostoceratidae, 10, 15, 16, 24, 25 novaehollandiae, Cucumerunio, 185-202 Nowakites, 20 411 412 Nudibranchia, 83-96 nugax, Hauffenia, 138 nugax, Horatia, 127, 129, 138 nugax, Phreatodrobia, 129, 132-172 nugax, Valvata, 137 nuttalli, Cornus, 353 Nymphaea, 359, 362, 366, 368 Nymphophilus, 368, 369 obsoleta, llyanassa, 173-183 obsoletus, Nassarius, 299 occidentale, Carychium, 345, 347, 354 occidentale, Partschiceras, 27 occidentalis, Nipponites, 26 occidentalis, Retinella, 342 octona, Subulina, 76 Odontodiscoceras, 18 ohioensis, Pallifera, 276 Olcostephanidae, 11, 18, 27 Olcostephanus, 28 olivaeformis, Alyxia Olivella, 40 Olividae, 31, 39, 41 Отодутпа, 33, 39 Oniscoidea, 38 onoense, Hypophylloceras, 27 Oosterella, 19 Oosterellidae, 19 Ophiuroidea, 47, 48 Opisthobranchia, 31, 46, 83 Oplopanax, 353 oregonensis, Spitidiscus, 27, 28 oregonensis, Wellsia, 28 orientale, Calycoceras, 25 Orygoceras, 157, 158, 171 Osmanthus, 68, 69 Ostlingoceras, 24 otaheitana, Partula, 97-106 Otohoplites, 21 otsukai, Bostrychoceras, 25 Ovulidae, 31, 32 Oxytropidoceras, 21, 25 Pachydesmoceras, 19, 27 Pachydiscidae, 20, 25 pachydiscoide, Pachydesmoceras, 27 Pachydiscoides, 20 Pachydiscus, 5, 20, 25 Pachyolivia, 40 pacifica, Mesopuzosia, 27 pacificum, Metaplacenticeras, 25 packardi, Oxytropidiceras, 25 packardi, Wellsia, 28 Palaemonetes, 129, 158 Palaemonidae, 129 Pallifera, 276 Paluccia, 136 Paludiscala, 157-160 Panamurex, 32, 33, 39 Papyridea, 41 Paquiericeras, 18 parabevahites, Paratexanites, 23 Parabogidiella, 129 INDEX parabrancoceras, Eubrancoceras, 21 Paracalycoceras, 22 Paracanthoplites, 20 Paracraspedites, 18 Paracrioceras, 24 Parahoplites, 20 Parajaubertella, 17 Paralenticeras, 23 Paramammites, 22 Parametaria, 39 Parancyloceras, 24 Parandiceras, 18 Paraphoplites, 25 Parapuzosia, 19 Pararhytida, 203-241 Paraspiticeras, 20 Parastieria, 18 Paratexanites, 23 Paravascoceras, 22 Parengonoceras, 20 Partschiceras, 18, 27 Partula, 97-106 Partulina, 67-71, 76-79 Parvanachis, 39 parviflorus, Rubus, 353 parvifolium, Vaccinium, 353 Parvilucina, 41 Patagiosites, 20, 25 patricki, Shasticrioceras, 26 pattersoni, Trogloglanis, 129 pecki, Olcostephanus, 28 pecki, Stygobromus, 129 pedicellatus, Lyrodus, 331 Pedioceras, 24 Peroniceras, 23, 25 Persicula, 40 pfeifferi, Biomphalaria, 320 Phacoides, 41 *phacoides, Pararhytida, 203-*233-241 phestum, Eulytoceras, 27 Philomycidae, 271-280 Philomycus, 271-280 Phlogocardia, 41 Phlycticrioceras, 24 Pholadacea, 323-339 Pholadidae, 323-339 *Phreatodrobia, 129, *132-152, 157-172 Phylloceras, 18, 27 Phylloceratidae, 10, 15, 16, 18, 27 Phyllonotus, 39 Phyllopachyceras, 18, 27 Physa, 255, 257, 278 Physalia, 301 Picea, 348, 353 Pictetia, 17 pilus, Asellus, 129 Pinus, 348 pisana, Theba, 249 Pisania, 39 Placenticeras, 10, 21 Placenticeratidae, 15, 16, 21 "plana, Phreatodrobia, 129, 132-*150-172 Planogyra, 347, 354 Planorbidae, 265-269 Planorbinae, 320 Planorbulinea, 320 Platyhelminthes, 129 Platylenticeras, 18 plebeius, Tagelus, 281-290 Pleioptygma, 33, 40 Plesiophysinae, 320 Plesiopsis, 203, 214 Plesiospitidiscus, 19 Plesiotissotia, 23 Plesiovascoceras, 25 Pleurobema, 199 Pleurobemini, 125 Pleurobranchacea, 83 Pleurohoplites, 21 Pleurolucina, 41 Pleuroploca, 39 plicata, Thuja, 353 Plicoliva, 31, 32, 40, 41 Polinices, 291, 292, 295, 297 Polygona, 39 polymorpha, Metrosideros, 68 Polyptychites, 18, 28 Polyptychoceras, 24, 25 Polystichum, 343, 353 pomatia, Helix, 173 Pomatiopsis, 173 pontiente, Shasticrioceras, 26 popenoi, Hemihoplites, 26 popenoi, Pulchellia, 25 popetensis, Tetragonites, 9, 26 Populus, 353 Praetollia, 18 Prionocycloides, 21 Prionocyclus, 23 Pristiloma, 347, 354 Prochelaea, 40, 41 Procheloniceras, 20 profuga, Alathyria, 185-202 Prohauericeras, 21 Prohysteroceras, 22 Proleopoldia, 18 Proleymeriella, 21 Prolyelliceras, 22 Prophysaon, 307-311, 354 Proplacenticeras, 21 Propustularia, 38 Prosobranchia, 46, 48, 127, 172 Protacanthoceras, 22 Protancyloceras, 24 Protanisoceras, 24 protea, Tryonia, 370 Protegragnoites, 26 Protengonoceras, 20 Protetragonites, 17 Protetragonitidae, 17 Protexanites, 14, 23, 25 Protohoplites, 21 Proturrilitoides, 24 Prunum, 40 Pseudargentiniceras, 18 Pseudaspidoceras, 22 INDEX Pseudocyphoma, 38 pseudodeverianum, Romaniceras, 25 Pseudohaploceras, 19 Pseudohelicoceras, 24, 26 Pseudojacobites, 20 Pseudokossmaticeras, 20 Pseudomenuites, 20 Pseudoosterella, 19 Pseudophyllites, 18, 26 Pseudosaynella, 19 Pseudoschloenbachia, 23, 25 Pseudosonneratia, 21 Pseudothurmannia, 24 Pseudotissotia, 23 Pseudotsuga, 353 Pseudouhligella, 13, 19 Pseudoxybeloceras, 5, 9, 24, 26 Pseudozonaria, 38 Psidium, 68 Psilotissotia, 20 Pterolytoceras, 17 Pteropurpura, 39 Pterynotus, 39 Ptychoceras, 24 Ptychoceratidae, 24 Ptychosalpinx, 39 pugetensis, Striatura, 342, 347, 354 pulchella, Vallonia, 346, 347, 354 Pulchellia, 12, 20, 25 Pulchelliidae, 20, 25 Pulmonata, 271-280, 307-311 "punctata, Phreatodrobia, 129, 132-*152-172 Punctum, 347, 354 Purpurellus, 39 pusilla, Doriopsilla, 94 Pustularia, 33, 38 Pusula, 32, 38 Puzosia, 14, 19, 27 Puzosigella, 21 Pyrgophorus, 371 "pyrosticta, Pararhytida, 203-"233-241 quadripaludium, Mexithauma, 369 413 radiata, Lampsilis, 110-117, 120-122, 125, 185 Radiodiscus, 342 ragsdalei, Pallifera, 276 ramosum, Neophylloceras, 27 randolphii, Punctum, 347, 354 Raspailiceras, 19 rathbuni, Typhlomolge, 129 Rattus, 72 rattus, Rattus, 72 rebellus, Cyclops, 129 rectilabrum, Euspira, 292, 293 rectoris, Hertleinites, 28 redelli, Asellus, 129 redfieldii, Partulina, 68-71, 76-79 redmondi, Hoplocrioceras, 26 reesideoceras, Forresteria, 23 relicta, Seborgia, 129 rembda, Hauericeras, 27 reticulatus, Nassarius, 303 414 Retinella, 342 Rhipophos, 39 rhodostoma, Bullia, 299 Rhytidopsis, 203, 206, 214 Rissoacea, 127 robusta, Grevillea, 68 robusta, Typhlomolge, 129 Rogersites, 18 Roloboceras, 20 Romaniceras, 22, 25 rosea, Achatinella, 72 rosea, Euglandina, 75 “rotunda, Phreatodrobia, 129, 132-*147-172 rowelli, Vertigo, 342, 346, 347, 354 royerianus, Toxoceratoides, 26 rubescens, Partula, 97-106 rubra, Alnus, 353 Rubus, 353 Rugotyphis, 39 Rumex, 320 Rumina, 333 rupestre, Cerithium, 332 rushi, Philomycus, 276 russeli, Stygobromus, 129 Ryugusella, 25 ryugasensis, Ryugusella, 25 sacculata, Eugomontia, 299-305 Salix, 353 samacos, Texiweckelia, 129 Samoana, 97-106 sandwichensis, Osmanthus, 68, 69 sapiens, Homo, 73 Sarasinella, 18, 27 Satan, 129 saturnale, Lytoceras, 27 saulae, Toxoceratoides, 26 Saynella, 19 Saynoceras, 18 Scabricola, 40 scabridum, Cerithium, 332 Scaevola, 68 Scalarites, 25 Scaphella, 40 Scaphinotus, 307-311 Scaphitidae, 24 Schistosoma, 243-247, 265-269, 313-321 Schloenbachia, 21 Schloenbachiidae, 21 schrammi, Biomphalaria, 320 Scleractinia, 43, 47, 48 Sconsia, 38 Sebidae, 129 Seborgia, 129 Segmentininae, 320 selwynoceras, Collignoniceras, 23 Semicassis, 38 septemlineata, Geoplana, 72, 73 septemseriatum, Kanabiceras, 25 serititense, Neophylloceras, 27 shallon, Gaultheria, 343, 353 Sharpeiceras, 22 shastense, Eucalycoceras, 25 INDEX Shasticrioceras, 11, 24, 26 shirleyae, Chicoreus, 39 shoupi, Eotetragonites, 26 Silesites, 19 Silesitidae, 19 Silesitoides, 19 siliquoidea, Lampsilis, 109-115, 117, 120-122, 125, 185 Simbirskites, 18, 28 similaris, Bradybaena, 76 Simnia, 38 Sincola, 32, 33, 39 Siphocypraea, 38 Siphonochelus, 39 Siratus, 39 sitchensis, Picea, 353 smithi, Lirceolus, 129 Solenoceras, 24 Solenosteira, 39 Solgerites, 23 Sonneratia, 21 Spathiceras, 22 Spathites, 22 Speetoniceras, 18, 28 Sphaeromicola, 129 Spalloplana, 129 Spenodiscidae, 16, 23 Sphenodiscus, 5, 23 Sphincterochila, 278 spindale, Schistosoma, 265 spiniferum, Douvilleiceras, 25 spinosum, Calycoceras, 25 Spiticeras, 18 Spitidiscus, 19, 27, 28 splendida, Lampsilis, 112, 122, 125 sportella, Haplotrema, 342, 354 Springvaleia, 38 Stachytarpheta, 68 stantoni, Homolsomites, 27 Stantonoceras, 21 starkingi, Toxoceratoides, 26 stearnsi, Pristiloma, 347, 354 Stewartia, 33, 41 stippi, Thurmanniceras, 27 Stoliczkaia, 22 stoliczkai, Calycoceras, 25 stolonifera, Cornus, 353 Stomohamites, 12, 24, 26 Strephona, 33, 40 Strephonella, 40 Streptorygma, 39 striata, Martesia, 325, 327-339 Striatura, 342, 347, 354 Strigatella, 40 Strioterebrum, 292, 294 Strombidae, 30-32, 38 Strombina, 33, 39 Strombinella, 39 Strombinophos, 39 Strombus, 38 Stygobromus, 129 *Stygopyrgus, 129, 132, 141, 144, 153-*156-160, 163, 168-171 INDEX 415 Stylommatophora, 68 Torculoidella, 38 Subalpinites, 18 Tornatellinae, 79 Subastieria, 18 Toroliva, 40 Subcancilla, 32, 33, 40 townsendiana, Allogona, 345, 354 subcompressum, Glyptoxoceras, 9, 25 Toxoceras, 15, 24, 26 subcompressus, Pachydiscus, 25 Toxoceratoides, 26 Subcraspedites, 18 Trachycardium, 41 Submortoniceras, 9, 23, 25 Trachypollia, 39 Suboosterella, 19 Tragodesmoceras, 19 subpiscinalis, Hauffenia, 137, 161 Tragodesmoceroides, 19 Subprionocyclus, 23, 25 Trajana, 39 Subprionotropis, 23 trichocarpa, Populus, 353 Subpterynotus, 33, 39 trichocoma, Tropidotropis, 203 Subpulchellia, 20 trichotonus, Polyptychites, 28 subquadrata, Puzosia, 27 Trigoniocardia, 41 Subsaynella, 19 Trilobita, 55 subterranea, Artesia, 129 trinitense, Phyllopachyceras, 27 Subthurmannia, 18 Triodopsis, 345, 347, 350, 354 Subulina, 76 Tripterotyphis, 39 Succinea, 79 Trivia, 38 sudanica, Biomphalaria, 261 Trochleiceras, 20 susuki, Cleoniceras, 27 Trochleiceratidae, 20 sylvatica, Fagus, 350 Trochomorpha, 214, 226, 231 Trogloglanis, 129 Tagelus, 281-290 Tropaeum, 24 Talityphis, 39 tropicus, Bulinus, 249-263 Tebennophorus, 271 Tropidotropis, 203, 206 Tegoceras, 22 Tropitoides, 21 tehamense, Peroniceras, 25 truncatus, Neocomites, 27 Temnoptychites, 18 Tryonia, 370, 371 Terebra, 291-298 Tsuga, 353 Teredinidae. 323-339 tuberculata, Melanoides, 255-257 Teredo, 324-339 tulipa, Fasciolaria, 39 teshioensis, Eupachydiscus, 25 tumidus, Unio, 199 Tetragonites, 9, 13, 17, 26 turgida, Partula, 97 Tetragonitidae, 10, 15-17, 26 Turrilites, 12, 24 Tetrahoplites, 21 Turrilitidae, 24, 26 Tetrahoplitoides, 21 Turrilitoides, 24 texana, Monodella, 129 Turritella, 38 Texanites, 23, 25 Turritellidae, 32, 38 texanus, Hadeoporus, 129 Typhinellus, 39 texensis, Texiweckelia, 129 Typhlomolge, 129 Texiweckelia, 129 Theba, 249 Uhligella, 19 thetys, Phylloceras, 27 Uhligia, 24 Thiarinella, 39 Ulva, 301 Thomasites, 22 umpquanum, Phyllopachyceras, 27 thomeli, Ancyloceras, 26 Unio, 199 thompsoni, Eupleura, 39 Unionidae, 107-125 thompsoni, Protexanites, 25 Urosalpinx, 39, 281-290 Thuja, 353 Utaturiceras, 22 Thurmanniceras, 11, 14, 18, 27 “uvaldensis, Balconorbis, 129, 132, 137, 141, 144, *thyrophora, Pararhytida, 203-*233-241 152—154—156, 168-171 Tiara, 40 Tissotia, 23 Vaccinium, 353 Tissotiidae, 23 Valanginites, 18 togata, Limax, 271, 273 Valdedorsella, 19 togatus, Philomycus, 271-280 Vallonia, 346, 347, 354 Tollia, 18 Valvata, 134, 137 Tolypeceras, 18 vanattae, Prophysaon, 354 Tonna, 31, 38 vancouverense, Haplotrema, 342, 347, 354 Tonnacea, 30-32, 38 vancouverense, Hoplitoplacaticeras, 25 Torcula, 38 vancouverense, Polyptychoceras, 25 416 vancouverensis, Didymoceras, 26 varia, Pallifera, 276 varicans, Cyclops, 129 Vascoceras, 22 Vascoceratidae, 22, 25 Velesunio, 186, 196, 199 Venericardia, 292 Venezoliceras, 21 ventricosa, Ficus, 38 Venus, 292, 293 venustum, Hyphantoceras, 9, 26 Vermicularia, 38 Vertebrites, 17, 26 Vertigo, 342, 346, 347, 350, 354 Vespericola, 347, 354 vespertinum, Acrioceras, 26 Videna, 226, 231 vidua, Cypridopsis, 129 vigorosa, Wellsia, 28 virginica, Crassostrea, 173, 281, 288 viridans, Achatinella, 70 Vitrina, 346, 347, 354 Vitularia, 31 Voluta, 40 Volutidae, 31, 32, 40, 41 voyanum, Acrioceras, 26 Watinoceras, 22 INDEX wautieri, Ferrissia, 320 Wellsia, 28 whitei, Acanthoceras, 25 whiteneyi, Shasticrioceras, 26 whitneyi, Anagaudryceras, 26 Wichmanniceras, 19 Wikstroemia, 68 wilcoxensis, Anahamulina, 26 wilcoxi, Thurmanniceras, 27 wintunium, Hoplocrioceras, 26 wintunius, Eotetragonites, 26 wooldridgei, Graysonites, 25 woollgari, Collignoniceras, 25 Wrightoceras, 23 Xanthium, 320 Yabieceras, 23 Yokoyamaoceras, 19 yokoyami, Canadoceras, 25 Zafrona, 39 Zanassarina, 39 Zelandites, 17, 26 zelindae, Plicoliva, 41 Zonitoides, 346, 347, 354 Zoogonus, 173, 183 Zurcherella, 19 WHY NOT SUBSCRIBE TO MALACOLOGIA? ORDER FORM Your name and address Send U.S. $17.00 for a personal subscription (one volume) or U.S. $27.00 for an institutional subscription. Make checks payable to “MALACOLOGIA.” Address: Malacologia, Academy of Natural Sciences Nineteenth and the Parkway, Philadelphia PA 19103, U.S.A. AWARDS FOR STUDY AT The Academy of Natural Sciences of Philadelphia The Academy of Natural Sciences of Philadelphia, through its Jessup and МсНепгу funds, makes available each year a limited number of awards to support students pursuing natural history studies at the Academy. These awards are primarily intended to assist predoctoral and immediate postdoctoral students. Awards usually include a stipend to help defray living expenses, and support for travel to and from the Academy. Application deadlines are 1 March and 1 October each year. Further information may be obtained by writing to: Chairman, Jessup- McHenry Award Committee, Academy of Natural Sciences of Philadelphia, 19th and the Parkway, Philadelphia, Pennsylvania 19103, U.S.A. =) 27(2) INSTRUCTIONS FOR AUTHORS 1. MALACOLOGIA publishes original re- search оп the Mollusca that is of high quality and of broad international interest. Papers combining synthesis with innovation are par- ticularly desired. While publishing symposia from time to time, MALACOLOGIA encour- ages submission of single manuscripts on diverse topics. Papers of local geographical or systematic interest should be submitted elsewhere, as should papers whose primary thrust is physiology or biochemistry. Nearly all branches of malacology are represented on the pages of MALACOLOGIA. 2. Manuscripts submitted for publication are received with the tacit understanding that they have not been submitted or published elsewhere in whole or in part. 3. Manuscripts may be in English, French, German or Spanish. Papers in lan- guages other than English must include a translation of the Abstract in English. Authors desiring to have their abstracts published in other languages must provide the translations (complete with main titles). Include all foreign accents. Both American and British spellings are allowed. 4. Unless indicated otherwise below, con- tributors should follow the recommendations in the Council of Biology Editors (CBE) Style Manual (ed. 5, 1983) available for U.S. $24.00 from CBE, 9650 Rockville Pike, Bethesda, MD 20814, U.S.A. 5. Be brief. 6. Manuscripts must be typed on one side of good quality white paper, double-spaced throughout (including the references, tables and figure captions), and with ample margins. Tables and figure captions should be typed on separate pages and put at the end of the manuscript. Make the hierarchy of headings within the text simple and consistent. Avoid internal page references (which have to be added in page proof). 7. Choose a running title (a shortened version of the main title) of fewer than 50 letters and spaces. MALACOLOGIA 1986 8. Provide a concise and informative Ab- stract summarizing not only contents but re- sults. A separate summary generally is super- fluous. 9. Supply between five and eight key (topic) words to go at the end of the Abstract. 10. Use the metric system throughout. Mi- cron should be abbreviated рт. 11. Illustrations are printed either in one column or the full width of a page of the journal, so plan accordingly. The maximum size of a printed figure is 13.5 x 20.0cm (preferably not as tall as this so that the caption 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 ‘3. Letters and numbers should be at least 3mm high after reduction. Several drawings or photographs may be grouped together to fit a page. Photographs are to be high contrast. High contrast is especially im- portant for histological photographs. 13. All illustrations are to be numbered sequentially as figures (not grouped as plates or as lettered subseries), and are to be ar- ranged as closely as possible to the order in which they are first cited in the text. Each figure must be cited in the text. 14. Scale lines are required for all nondi- agrammatic figures, and should be conve- nient lengths (e.g., “200 um,” not “163 вт”). Magnifications in captions are not accept- able. 15. All illustrations should be mounted, numbered, labeled or lettered, i.e. ready for the printer. 16. A caption should summarize what is shown in an illustration, and should not dupli- cate information given in the text. Each let- tered abbreviation labeling an individual fea- ture in a figure must either be explained in each caption (listed alphabetically), or be grouped in one alphabetic sequence after the Methods section. Use the latter method if many abbreviations are repeated on different figures. | 17. Tables are to be used sparingly, and vertical lines not at all. 18. References cited in the text must be in the Literature Cited section and vice versa. Follow a recent issue of MALACOLOGIA for bibliographic style, noting especially that se- rials are cited unabbreviated. Supply pagina- tion for books. Supply information on plates, etc., only if they are not included in the pagination. 19. In systematic papers, synonymies should not give complete citations but should relate by author, date and page to the Litera- ture Cited section. | 20. For systematic papers, all new type- specimens must be deposited in museums where they may be studied by other scien- tists. Likewise MALACOLOGIA requires that voucher specimens upon which a paper 1$ based be deposited in a museum where they may eventually be reidentified. 21. Submit each manuscript in triplicate. The second and third copies can be reproduc- tions. REPRINTS AND PAGE COSTS 22. When 100 or more reprints are or- dered, an author receives 25 additional cop- ies free. Reprints must be ordered at the time proof is returned to the Editorial Office. Later orders cannot be considered. For each au- thors' change in page proof, the cost is U.S. $3.00 or more. 23. When an article is 10 or more printed pages long, MALACOLOGIA requests that an author pay part of the publication costs. SUBSCRIPTION COSTS 24. For Vol. 28, personal subscriptions are U.S. $17.00 and institutional subscriptions are U.S. $27.00. Address inquiries to the Sub- scription Office. VOL. 27, NO. 2 MALACOLOGIA CONTENTS P. MORDAN & $. TILLIER New Caledonian charopid land snails. |. Revision of the genus Pararhytida (Gastropoda: Charopidae) ео led SOL PRESS C. S. RICHARDS 4 D. J. MINCHELLA En Genetic studies of biphallic Biomphalaria glabrata ........................ 24 M. А. CHAUDHRY & E. MORGAN Factors regulating oviposition in Bulinus tropicus in snail-conditioned Мет A DIA EA A ВОИ ao В. О. PARASHAR & К. М. ВАО ie Effects of long-term exposure to low concentrations of molluscicides on а: fresh-water snail, Indoplanorbis exustus, a vector of ni le RE H. L. FAIRBANKS The taxonomic status of Philomycus togatus (Pulmonata: Philomycidae): a morphological and electrophoretic comparison with РИПОту NE О ое и ES a | D. RITTSCHOF & А. В. BROWN N Modification of predatory snail chemotaxis by substances in bivalve prey a RR A ИВ inet AL PT E о J. A. KITCHELL, C. H. BOGGS, J. is RICE, J. F. KITCHELL, A. HOFFMAN & J. MARTINELL ER Anomalies in naticid predatory behavior: a critique and experimental ODSEIVALIONS ae a N ae a ОВ à S. A. HARRIS, F. M. da SILVA, J. J. BOLTON & A. C. BROWN _О. S. PIERI & J. D. THOMAS Polymorphism in a laboratory population of Biomphalaria glabrata from a ane seasonally drying habitat in north-east Brazil ...... ral a eats = el Sod ed С - 1 3 3 | GE MA R. A. D. CAMERON y Environment and ANS of forest snail ari from coastal Briti A RA A A De a Re ae Be R. HERSHLER & W. L. MINCKLEY Microgeographic variation in the banded spring snail (Hydrobiidae: Mexipyrgus) from the Cuatro Ciénegas basin, Coahuila, PONES AA В. ROBERTSON & Е. V. COAN | a A. Myra Keen (1905-1986) ................... A à do Е A. M. KEEN (posthumous) jee Some important sources for molluscan generic In designations ........ INDEX TO VOL 27 NO a dci rr N SW KR о a МВА 4 072 160 500 UD А Er mrad ity t ee SRE: mA e ann К ии г pales heen y {AN eee НАТ CO é ES He ob à O NANTES С E MALO Dd миа EEN у АА ds >. A UA ER} ate A denne 4 RARE и t wa Pret MOTOS que a ! TP UE Nele! А АХ * uM ETA NANA] so Er у h ; Je мо ми ИЯ Qui HA $ ‘ Yun, t HS 4 rer 5 4 | sy bah À nee FON nt MAL at E + h thed'ed Leta a CRAN) Ut AA TEE ME р hf COTTON ET ECC + ryt 2 2 И ен SA PISTA MATO $74 Je pros + OP DT ANT EU] Id Add dre Mtra tr IA ot BI Rh of AA OLIS ES CRA] AAA fem М КАН a Чи м вом o rh + ¿e oth ate Me NAS MIE to het ивы м O Me Aya a ou hele de ET ANDA у We a ANE or Dike nie УИ un PEN « NES УХ BOM Mook ot A речи Е O Dan age me г Fam te ares НА y Pq PERLES nee RIEF EL Coon ee ee PA тва о tat a pe oar Fi epee) ui Minh Ph dde 11 id RPTE TIEN em o or RECENSE AN Dee ed кии НН НИИ Re e Dare qu ai НАУ plebe ét de nor , fig фени PARENT име ЛАМА LATIN ETIENNE OU A de A PARTIES ИРАК Meet ah A MH TES 4 due mo A O erh aay he MAS woe desde rn ik re >. ТЫ НИ АИ we À DRE « far qe ve NE aden pra Fa 12207922 Lie ho ae ОН af LE be т RRS in LU AAA E АК Es Ma AS it) PUES Sy Е SEAN MER POE A } VG Ke 1% EX AA N 6 Apu base hey [y Ко » ЗО 47 KICK: EE Ne Fe ‘On OAK, оса А Fat “6 И EEN ENS И TA AYNA a : finit АА N и м e A >: АН, LEE UNA NN E НЫ у BER HRS rer nz 4 ee ил: О de (4 .. { К 4 АА SST ASI EN АДА AAA Qi HAS 143 AA, A x KERN и НАХ NR EEE REN NN Е one RER FU AE dt А Pero A УЗНАЛА ан ИА v bis 5: uu RAT ip E ATU E * АА Ny MAN SPP eye are veus FX A vr KIN As RE, ое elite gy Ne UN о О vr BRENNEN Kr EINE NS HE OPEN RAGE CNN КН N hr ae ARAU Niet ea at ia he eau ren Ч f OS À F4 LS AAA A A . LE RECRUE A EN Ge, I AS AO ye aig) KENN AR Qin hy CORNE EEE be UA EL TEIL I A EA A A ALU A o A E A NA RIA AMIA DATES т AAA eo ATEN ARANA AO bate HR re IDO о ON PATTES НЙ AT eos AO ON EE wee mae Bi AA rite АМ ОСИ НН wre a DOTE aval UI AAA TA AA IA Y И Mur Tas AS tt tt) it LANCER eur Е Зее ссы рим НА i! ADs A tats vr ATEOS ws N AOS ae Nagy Ze ga A Иа DU ER ja Fu} О AA PUN A a A Ума, AN A AU to Re Е. ИА чер VO Pan VIO POO URL EN tlt N iat Voie УЛ takes o yrs 7) w fs ATA бу =: Wty HORA A vr vu ba vr ho i м fu ne LEA DER IT Eje dy a Ни НКО A ave, AA! а Я ( y y А а И А мА PULMONAR UV tit a wy AAC an? APCE WEINEN, yy LIRE a Li OA OL TIBIA LI ATAR UNI AA AS o CE e wore, wi, UV AG ne CALE RAIN Whey 4 ‘ vil, Ат. Али x HAT ‘ DIT vie’ A ATOUT EAN HELEN UHR vw уму vs RENNES ATP TEE TON DORA AMO Evi м 1 yo) VA ИУ: Es h este, CH ER Rn ei a APTA DA EE LHMA TAN Е: tits ae An kin [ya TEL wer uh + 4 м , ER: RAN LE As i EME LH a 4 у ОИ, АЛАН рой ПИ HE Вс ЖА ++ A e san ЕН ROLE И ti т о Be i are pad, ur PARUS A KAMEO ONE Ах bo A